LIBRA R.Y
OF THE.
U N IVERSITY
Of 1LLI NOIS
54-7
95Z/53
Return this book on or before the Latest Date stamped below.
University of Illinois Library
L161-— H41
Digitized by the Internet Archive
in 2012 with funding from
University of Illinois Urbana-Champaign
http://archive.org/details/organicsemi195253univ
SEMINAR TOPICS CHEMISTRY 435 I SEMESTER 1952-53
Some Recent Developments In the Field of Ellm3jia'tioi>^te&ctions
Ellas J. Corey? September 2S.«..,- • 1
The Meerwein Reaction
Edward. C. Taylor, Jr., Sep t e mb e r*"36. 6
The Structure of Terramycln
Charles King, October 3 11
Iron Bis-Cyclo'pentadienyl
Benjamin L. Van Duuren, October 3 14
The Vicinal Addition of Certain Reagents to Aromatic Systems
William S8 Friedlander, October 10. , 17
The Synthesis and Properties of Cyclob'lef ins Containing Nine and Ten Carbons
Elliott E. Ryder, October 10 22
The Alkoxylation of Simple Furpns and Related Reactions
' Paul Lu Cook, October 17 25
Attempted Syntheses of Simple Pentalenes
John R. Demuth, October 17 29
Asymmetric Citric Acid
Richard F. Heitmiller, October 24 35
Azo Nitriles
Barbara H. Weil, October 24 39
The Structure of Ketone Dimer
William S. Anderson, October 31 43
The Synthesis and Properties of Some Simple Amino and Hydroxy Pteridines
William R. Sherman, October 31 46
Hydrocarbons with Intercyclic Double Bonds
Michael J. Fletcher, November 7 52
New Reactions of Pyrroles
Robert E „ Putnam, November 7 57
'.y
The Skeleton of Piorotoxlnln
R. Thomas otiehl, November 7 62
Pinacol-Pinacolone Rearrangements
Ruth J. Adams, November 14 67
Formazans
Nikodems E. Bo jars, November 14 72
Di- and Polyacetylenes
Aldo J. Crovettl, Jr., November 14 78
Thenoylbenzoic Acids and Thiophanthraquinones
John A. MacDonald, November 21 • 84
New Methods f6r Spontaneous Resolution of Racemic Modifications
Harry J. Neumiller, November 21 89
The Reactions of Halogen (i) Salts of Carboxylic Acids
George W, Parshall, November 21 . .• 93
The Reaction of cc-Kaloketones with Dinltrophenylhydrazlne
Fabian T. Fang, December 5 . .. 96
Lanostadienol
David M. Locke, December 5 . 100
Recent Studies in the Chemistry of Indanthrones
William H, Lowden, Decembers •'• 105
Acyl O^N Migrations
Howard J. Burke, December 12 110
Some Chromic Acid Oxidations
Y. G-ust Hendrickson, December 12 t 115
A New Synthetic Route to Cyclopropane s
S. Lawrence Jacobs, December 12................ 120
Sulfonation of Acid- Sensitive Compounds
Clayton T. Elston, December 19... 124
Synthesis of Substituted Silanes
C. W. Hinman, Dooerrfber 19............ ,.,..... 129
Ring Contraction Reactions of Tropolones
Harry W. Johnsbn, Jr., December 19... 132
Concerted Reactions: %Polyfunctional Catalysts
Richard L. Johnson, January 9.. 137
Some Methods of Stepwise Peptide Degradation
N, W. Kalenda, January 9.7 142
Phosphate Esters of Nucleosides
James C. Kauer, January 9 146
» ■ - » t
—3
Trie Iky 1 Oxonium Spits
Robert J. Lokken, January 16 151
Aminations with Alkali Amides
Thomas R. Moore, January 16.. 156
G-riseofulvin
Paul D. Thomas, January 16 159
-I-
SOME RECENT DEVELOPMENTS IN THE FIELD OF ELIMINATION REACTIONS Reported by E. J. Corey September 26, 195?
Duality of Mechanism for E2 Processes. — The stereochemistry of the olefins produced by E2 elimination reactions, e.g. base catalyzed dehydrohalogenation, for many years has been mterpretec on the basis of preferential trans elimination. Thus, while
d l-a,aT-dibromosuccinic acid (IA) upon treatment with base yield? bromofumaric acid (Ha), meso-a,a'-dibromosuccinic acid {IB) affords bromomalelc acid (IIB).3 There are numerous other example
in which ' ation.4"8"
;rans elimination is heavily favored over els elimin-
HOOC
COOH
HOOC-— - — %
COOH
IA
IIA
C00w
COOH
COOH
13
IIB
Much of the recent work in the field of elimination reactior has been undertaken in order to determine (a) the circumstances under which els elimination can occur, (b) the reasons for the relative ease with which trans elimination usually takes place ar (c) whether .cis and trans eliminations proceed by different mechanisms*
Cristol and his coworkers have studied the kinetics of the dehydrohalogenation of the five known isomers of benzene hexa- chloride, a,P,^,cfand£ .9_11 In the case of the cc,p, * and £ isome the rate-determining step in the formation of trichlorobenzenes the elimination of the first hydrogen and chlorine and, consequer the kinetics of dehydrochlorination of these substances provide information concerning the first elimination only. The p-isomer (Hip), which initially c*n undergo only cis elimination, reacts with' hydroxide ion at a rate which is 7000 to °4,000 times slowei than the rate of reaction of the a, X and € Isomers (ill o , X , £ )
-2-
In which initial trans elimination lis ndsMSle.
TUP
TTIoc
TTlT
The second-order rate constants, experimental (Arrhenius) activation energies and entrooies of activation for the alkaline dehydrochlorination of III p, a, * and 6. ' are1 fisted in Table I.
Table I
1 i
I |
somer |
|
III |
(3 |
|
III |
a |
|
III |
y |
|
III |
6 |
k.^O.OO,
l./mole sec.
2.11. (l0)~'5.
0.500 0.151 0.182
£- exp, kca l./mole
31.0
18.5 20„6 21.4
cal./mole deg. 20.2
-1.0 3.6 6.5
It has been suggested that the large difference between the activation energies for cis and tmns elimination might be due to a difference in mechanism.9*11 As a working hypothesis it has been postulated that _trans elimination oroceeds by a concerted process of rather stringent steric re quire me ires and low activation energy (A) 9? xl > X3 and that cis elimination cannot be concerted and proceeds via a earba.'iion intermediate by a two stage mechanism of relatively high activation energy and low steric requirement (B).9'1*
A (concerted):
X
f R3C— CR3
H
B B (two stap*e):
X
R=,C~
R,C = CR.
RftC
■CR.
I
H 4
RpC
-CR3
e +H3
R3C = CR3 + X
Rough calculations by Cristol11 (which neglect the effect of solvent) indicate that the activation energy for the concerted
■ 3-
process should be considerably less (^7-14 kcal./mole) than that for the two- stage process.
At the present time Plausible, but not compelling theoretical reasons have been adduced to explain why concerted cis elimin- ation, if it occurs at all, should be so much slower than con- certed trans elimination.9'12 Contrary to earlier belief5 electrostatic repulsion between the nucleophilic reagent and the departing anion, while larger for cis than for trans elimination, has been shown to be an insignificant consideration in dehydro- halogenation reactions.11 It should be emphasized that at present there is no rigorous theoretical evidence to indicate that con- certed ^is elimination is not possible.
In order to determine experimentally whether cis elimination actually proceeds via a carbanion intermediate, the reaction of Hip' with", base in deuteroethanol (C2H50D) was studied.13 In- troduction of deuterium into undehydrochlorinated III p during the course of reaction would be an indication of a carbanion inter- mediate capable of removing a de.uteron from the solvent. The III £ recovered after one half-life of elimination contained only a small amount of deuterium and, hence, the existence of a carbanion intermediate was not demonstrated (though also not disproved) by this experiment.
The stringent steric requirements for facile tr^ns elimin- ation have been demout' tr.ited quite clearly by data for the dehydrochlorinption cf els - and trans - 11, iS-dlchloro-9,3.0- dihydro-9,10-ethanoanthracene (IVA and IVB).14 Here the cis
IVA
isomer (IVA), which can undergo trans elimination, reacts about seven times more slowly than the trans isomer (IVB), which can only undergo cis elimination. Although the difference in rate is due mainly to a favorable entropy of activation for the cis process (Table II), the energy of activation for the trans process is, none the less, considerably higher than usual.
Table II
Isomer |
lCTkllOj l./mole sec. |
IVA (cis) IV3 (trans) |
6.38 49.9 |
&Eexp, AS*
kcal./mole cal./mole deg
26.5 -11.2
30.6 3.2
_4-
The abnormally high energy of activation for trans elimin- ation in IVA supports the hypothesis13' 15 that the atoms involved in bond making and bond breaking must be coplanar for facile trans elimination, since the requisite coplanarity of C1X, C1S and vicinal hydrogen and chlorine is absent from IVA. Additional evidence has been brought to bear on this point by Barton and Miller in their study of the iodine-catalyzed debromination of the cholesteryl 5,6-dlbromides. 1S
Comparison of the energies of activation for cis elimination in Hip and IVB indicates that cis elimination does not demand a very specific spa't'ial arrangement of the atoms Involved in bond making and bond breaking. This finding lends some support to the two- stage mechanism for cis elimination.
Recently Noyes and Miller16 h«ve studied the kinetics of dehydrohalogenation of the cis- and trans-dlhalo ethylenes. In each case the cis isomer (trans elimination) reacts more rapidly than the trans isomer (cis elimination). It was found, however, that the superiority of the trans "orocess is sometimes due to a more favorable energy of activation (viz. with the dibromo-^nd diiodoethylenes) and sometimes due to a more favorable entropy of activation (e,g. with the dichloroethylenes) .
Steric Effe ct in Elimination Reactions « — Hughe s , Ingold et al.ls have stated that all SI reactions and E2 reactions with uncharged structure a (e^, halides) lead to the olefin with the moat highly substituted ethylenic linkage (Saytzeff rule) and they have attributed this result to a greater degree of stabilization by hyperconjugative resonance of the transition state leading to the Saytzeff oroduct. E2 reactions of ammonium and sulfonium salts, on the other hand, lead to the olefin with the least substituted ethylenic linkage (Hoffman rule) and it was oroposed1£ that the direction of elimination in these cases is controlled by the (reaction retarding) inductive effect of the p-alkyl groups.
C. H. Schram 7 and, more recently, H. C. Brown and I. Maritani18 have proDosed that steric effects alone account for the occurrence of Hoffman elimination and that in the absence of appreciable steric effect E2 reactions always Proceed according to the Saytzeff rule. The basis for this argument is that if the group being eliminated is large, e.g. (CH3)3N or (CH3)S3 , the transition state leading to the Hoffman product VA is much less strained than that leading to the Saytzeff oroduct VB.
H,C^
K3° VCH3
VA VB
-5-
Table III summarizes some of the findings of Brown and ' v Maritani for reactions of the tyrce:
R» ' R*
RCH2-C-CH3 -» RCH * Zx + RCH3-C * CH3
t OH,
X
R* = alkyl or H
Table III
Reaction Effect of increase in sterlc
type Group requirements of group
El R Ssytzef f -> Hoffman
El X No effect
E2 R Say tz ef f — ► Hoffman
E2 X Saytzef f — > Hoffman
Bibliography
1. A. Michael, J. prakt. Chem. , 52, 289 (1895).
2. P. F. Frankland, J. Chem. Soc, 654 (1912).
3. R. Flttlg and C. Petri, Ann., 195, 56 (1879).
4. J. Wlslicenus, ibid., 248, 28l"lT888).
5. W. Huckel, W, Tappe and G. Les-utke, ibid., 543, 191 (1940) .
6. M. C. Hoff, K. V. Greenlee «nd C. E. Boord, J. Am, Chem. Soc, 73, 3329 (1951) .
7. D. J. Cram, Ibid.. 74, 2149 (1952).
8. D. Y. Curtin and D. B. Kellom, Abstracts, 122nd Meeting of the American Chemical Society, Atlantic City, N. J. , Sept. 1952, P. "23M.
9. S. J. Cristol, J. Am. Chem. Soc, .69. 338 (1947).
10. S. J. Cristol, ibid., 71, 1894 (1949).
11. S. J. Cristol, "NTT. House and J. 3. Meek, Ibid., _73, 674(l95l)^
12. M. L. Dhar, E. D. Hughes, C. K. Inerold, A. M, M. Mandour, G. A. Maw and L. I. Wolf, J. Chem/ Soc, 2093 (1948).
13. S. J. Cristol, unpublished results.
14. S. J. Cristol and N. L. House, J. Am. Chem. Soc, 74, 7193
1952).
15. D. H„ R. Barton and E. Miller, ibid . , 72, 1066 (1950) .
16. S. I. Miller and R. M. Noyes, ibid.. 74, 629 (1952).
17. C. H. Scbrnm, Science, llg. 367 (195077
18. H. C. BroT"n *nd I. Maritani, Abstracts, l£7nd Meeting of the American Chemical Society, Atlantic City, N. J., Sept. 1952, p. 2M.
THE MEERWEIN "REACTION Reported by E. C. Taylor, Jr. September 26, 1952
General: In 19-39, Meerwein (1) reported that, under special conditions, aromatic diazonium halides will couple with un- saturated carbonyl compounds with evolution of nitrogen to form a new carbon- carbon bond., The reaction has since been extended to include the reaction of an aromatic diazonium halide with conjugated olefins, styrenes and acetylenes, with accompanying loss of nitrogen, and is known as the Meerwein reaction". The product of the reac-r'ion may be saturated or unsaturated, depending on whether the elements of EX have been los"C from the initial reaction product,,
ArN2+X~ + R-CR-CHR' -N.a. , Ar~CH CH-X ._ ArC=CH
R R? ft R'
+ HX
Conditions: A cold solution of the diazonium halide (usually chloride) is added to a solution of the unsaturated compound in aqueous acetone containing the salt of a weak acid (usually sodium acetate) and cupric chloride. Although some cases have >>een found where the presence of acetone has a deleterious effect on the yield. (15 ^-^): most workers have found that acetone is essential for the reaction, The role of the sodium acetate and euprio chloride is not clearly understood (see the section under Mecjiajiism) , although in most instances the presence of "both is essential., The temperature must generally ^e raised to about 20° before evolution of nitrogen begins.
Scope and Limitations: The yields from the Meerwein reaction are usually low (5 - 80%) and the products are sometimes diffi- cult to purify because of the simultaneous formation of tars, azo resins, Sandmeyer reaction products s chloroacetone and aromatic hydrocarbons * Ring substituents in the aromatic di- azonium halide influence the yield greatly; for the same sub- stituent in the o, m, and p_ positions, the yield, of coupled product increases in the order o ' m ,'p_. In many instances, no product at all is obtained with o- substituted diazonium halides , {^-Naphthalene diazonium chlorid.es give better yields of coupled product than >Onaphthalene diazonium chlorides, probably because of a steric effect. Negative substitution (-N03, -303Na, -Hal, -COoH, - COOR) generally leads to higher yields c The failure of positively substituted phenyl diazon- ium chlorides to undergo the Meerwein reaction in some instances has been reported (2,15),
Synthetic Applications: The Meerwein Reaction has been used for the preparation of the following types of compounds.
-?-
1. Aryl- substituted coumarins
(1)
■V \
!l !
+ ArN2 CI
*-RT
+ N* + HC1
•x o
=o
! 'I
V
2, Cinnamic acids
+„,-
(15) ArN8 CI + HOOC-CH=CH~COCH y Ar-CH=CH-COCH + C02
+ Na + HC1
(IS) ArN2+Cl"~ + CH2=CH-C00H
Ar-CH=CH~C00H + N2 + HC1
+
(18,23,24) ArN2 CI + CH2=CH-X
i. Ar-CH2-CH-X + N2
(X= -CN,-C0CC2H5) CI
hydrolysis j-HCl
Ar-CH=CH-C00H
a. X -Aryl cinnamic acids
-f.
(1) S- '•--•. -CH=CH-C00C2H 5 + ArN2 CI
3, Stil^enes
• %
•-■ » • ■- *
•\
9
vN-CH-CH-COEt
CI Ar
+ N2
j
| -HC1
,-CH=C-C00H
Ar
(13), (f ^,-CH=CH2 + ArN2+Cl~
» — .-»
\\
>-CH=CH-Ar + N2 + HC1
(2,3,4,5,8,7,9,1?) X — A \^ CH=CH-C00H + ArNo+Cl~
X-^'-- V'.-.CH=CH-Ar
.. . »
+C02 + HC1 + N2
-3-
a. X. -Cyano stil^enes (a) X-W ^_CH=CH-CN + ArN2+Cl"
• — ■ ••
CN
X -.-^-- >._CH=C-Ar + N2 + HC1
•- — •
h. "--Acetyl stil"benes
.— v 0 . _
(4) .s'-' %_CH=CH-C-CH3 + ArN2 CI
> •
•■— — •
00 CH
,_CH=C-Ar + N2 + HC1
4. Aryl~l,3-butadienes
. ,. + _
(20,21) 02N-.^ ^>-N2 CI + CH2=CH-CH=CH2
02N_ // ^.>_CH2-CH=CH-CH2C1
| -HC1 0aN-^ "; .„CH=CH-CH=CH2
* - m
(3,7,8,9,10,19) X-^ ._CH=CH-CH=CH-C0OH + ArN2+Cl >
X y-' >-CH=CH-CH=CH~Ar + C02 + N3 + HC1
\ — /
t — ,
5. Aryl raalelc acids (1,24) CH300C-CH=CH-C00CH3 + ArN2+Cl~ N
CH300C-CH=C-C00CH3
Ar
+ N2 + HC1 6. Styrenes
-4-
(22,23) C1CH=CC1S + ArN2+Cl~ y ArCHClCCl3 >-
ArCCl=CCl2
(22,23) HC^CH + ArN2+Cl~ y ArCH-CHCl + N2 + HC1
7, Indirect Syntheses; By appropriate treatment of the initial Meerwein reaction product, 4,4' -di substituted biphenyls (from
CN-CH -CH3-<^~ V-,,-* V^CHjb-CH-CN) CI ._. / x- -■..■' CI
bibenzyls (from styrenes "by reduction), ?f-arylpropylamines (from Ar-CH=CH-CN by reduction), etc., may be prepared. Notable among syntheses utilizing the Meerwein reaction as a key step are the phenanthridine synthesis of Braude and Fawcett (25) and the lin-quaterphenyl synthesis of Bergmann and Weizman (10).
Mechanism: Both free radical and ionic mechanisms have been proposed for the Meerwein reaction.
1. Free radical Koelsch and Boekelheide (19) have proposed the following scheme to account for the products and orienta- tion of the reaction,
(1.) ArN2+ + CH3C00~ * ArN=N-0C0CH3 . Ar • + N2 + CH3C00 *
■% -
(2.) Ar* + R-CH=CH-R' „ Ar-CHR-CHR' (A)
(3.) Ar-CHR~CHR» + Cu++ > Ar-CHR-CHR1 + Cu+
(3a.) Cu+ + CHa'COO' * Cu++ + CH3C00~
(4.) Ar-CHR-CHR* + Cl" > Ar-CHR-CH-R'
+ Cl
(4a.) Ar-CHR-CHR* y H+ + Ar-C=CHR'
+
R
(/. -Coupling with cinnamic acid derivatives was explained on the assumption that the free radical formed initially (Equation 2, A) would be more stable than the alternative structure because of resonance of the free electron into the phenyl group; and similarly, 6-ooupling with acrylic acid was explained on the assumption that the free radical formed initially could be stabilized by resonance of the free electron on the X. -carbon (ArCH2-CH-C00R) into the carboxyl group. These views were shared by Dhingra and Mathur (14).
-5-
2. Ionic The proponents for the ionic mechanism (11, 12, 13) point out that the orientation observed in the Meerwein reaction in many instances parallels that of ionic addition of HBr rather than the peroxide-catalyzed, free radical addition. Brunner and Perger (12) have proposed that the cupric chloride functions simply as a halogen carrier and have successfully substituted pyridine for cupric chloride in one instance. The beneficial effect of acetone has been ascribed to its function as a halogen carrier through the initial formation of chloro- acetone.
The arguments for each theory fail to exclude the possi- bility of the alternate mechanism, and it seems probable that, in view of the number of side-reactions which invariably accomp- any the Meerwein reaction, the reaction is exceedingly complex and may involve both radical and ionic mechanisms.
References
"1. H, Meerwein, E, Bucbner and K. van Emster, J. Prakt. Chem., 152, 237 (1939)
2. G. A. R. Eon, J. Chem. Soc . , 224 (1948)
3. D, M. Brown and G. A. R. Kon, ibid., 2147 (1948)
4. P. L'Ecuyer and C. A. Olivier, Can. J. Research, 27B, 689 (1949)
5. P. L'Ecuyer, P. Turcotte, J. Giguere, C. A. Olivier and P. Roberge, ibid., 26B, 70 (1948)
6. P. L'Ecuyer and P. Turcotte, ibid., 25B, 575 (19^7)
7. P. Bergmann and Z. Weinberg, J. Org. Chem., 6, 134 (1941)
8. G. B. Bachman and R. I. Hoaglin, ibid., 8, 300 (1943)
9. P. Bergman, J. Weizman and D. Schapiro, ibid,, 9, ^08 (19^-4)
10. P. Bergmann and J. Weizman, ibid., 9, 415 (1944)
11. P. Bergmann and D. Schapiro, iMd., 12, 57 (1947)
12. W. H. Brunner and H. Perger, Monatsh., 29, 187 (1948)
13. W. H. Brunner, ibid., 82, 100 (1951)
14. D. R. Dhingra and K. B. L. Mathur, Ind. Chem, Soc. J., 24, 123 (1947)
15. J. Rai and K. B. L. Mathur, ibid., 2£, 383 (1947)
16. J. Rai and F. B. L. Mathur, ibid., 24, 413 (1947)
17. R. C. Fusori and H. G. Cooke, Jr., J. Am. Chem, Soc, 62, 1180 (19/10)
18. C. F. Koelsch, ibid., 6J5, 57 (1943)
19. C. P. Koelsch and V. Boekelheide, ibid., 66, 412 (1944)
20. E. C. Coyner and G. A. Ropp, ibid., 70, 2283 (19^-8)
21. G. A. Ropp and E. C. Coyner, Org. Syn. 31, 80 (1951)
22. E. Miller, Angew. Chem., 61, 179 (1949)
23. E. Muller, Ueber die Einwirkung von aromatischen Diazo- verbindungen auf aliphatische ungesattigte Verbindungen,
PB 737, Office of Technical Services, Department of Commerce, Washington, D. C.
24. E. C. Taylor, Jr. and E. J. Strojny, Unpublished Work
25. E. A. Braude and J. S. Pawcett, J. Chem. Soc, 3113 (1951)
THE STRUCTURE OF TERRAMYCIN Reported by Charles King
October 3, 1952
Terramycin, C22H24IJ209, a new broad-spectrum antibiotic, has recently been assigned the structural formula I:
CH3 OH OH II
>-CH
I
\X
*r
OKf
^
0
v
i
OH 0 OH 0
HHS
Aromatization of terramycin yields naphthacene s end demonstrates the presence of this ring system in the molecule. Structure I is consistent with certain products identified from both acid and alka- line degradation of terramycin.2"8
Treatment of terramycin with 1.5 N aqueous hydrochloric acid yields, with dehydration and rearrangement, a- ?nd p-apoterramycin which are regarded as stereoisomers of structure II:
More vigorous treatment with dilute hydrochloric acid yields terri- nolide, III:
OH CH3 l OH
OH OH 0
III
-2-
The infrared absorption spectrum of terramycin shows no absorption between 5 and 6^, and indicates that the phthalide carbonyl in II and III is derived from a highly conjugated or enoiized carbonyl group. Moreover this carbonyl must be incorporated in an actual or potential p-dl carbonyl system. The alternative formula IV is ruled out on the basis that the pKa of the dime thy lam in o group is not appreciably altered in the transformation to the apoterramycins .
OH
OH
NH
0 0
VNH3
IV
Alkaline degradation of terramycin yields products V-IX, which appear to be consistent with I.
CHa-C
*
0
OH
/\ S
OK
YY
)H 0
%y — v
CH,
V
VI
VII
V
OH
<
0
OH
c=o
NCH3
CHpOH
VIII
IX
V
-3-
Bibliography
1. Finlay, A. C., Hobby, G. L., P'an, 3. Y., Regna, P. P., Routiexr, J. B., Seeley, D. B., Shull, G. II., Sobin, B. A., Solomons, I. '.. Vinson, J. J., and Kane, J, H., Science, 111, #5 (1950) .
2. Hochstein, F. A., Stephens, C. R., Conover, L. H. , Regna, ?. P.. Pasternack, R. , Brunings , K. J., and 'Joo&ward, R. B., J. Am. Chem. Soc, Jit* 3702 (1952).
3- Hochstein, F. A., Stephens, C. R., Gordon, P. II., Regna, P. P., Pilgrim, J. J., Brunings, K. J., and liooduard, R. B., J. Am. Chem. Soc., ]4, 3707 (1952).
k-. Hochstein, F. A., Regna., P. P., Brunings K. J., and Woodward, R. B., J. Am. Chem. Soc, ]k, 37Q6 (1952).
5. Pasternack, R., Regna, P., Wagner, R., Bavley, A., Hochstein, F. A., Gordon, P., and Brunings, K., J. Am. Chem. Soc 0 , 73, 2*1-00 (1951) .
6. Pasternack, R., Conover, L, H., Bavley, A., Hochstein, F. A., Hess, G. B., and Brunings, K. J., J. Am. Chem. Soc, f1!-, 1929 (1952) .
7. Hochstein, F. A., and Pasternack, ?., J. Am. Chem. Soc, 73, 5006 (1951) .
g. Kuhn, R. . and Dury, IC. , Ber . , S1!-, g*J-g (1951) .
I*+
IRON BIS- CYCLOPS NTADIENYL Reported by B. L. Van Duuren October 3, 1952
In an attempt to orepare organoehromium compounds from phenylmagnesium bromide and chromic chloride Bennett and Turners- obtained an almost quantitative yield of diphenyl formed by the couoling reaction:
2C8H8MgX + MXa -* C6HBCSHS + 2MgXa + M (a)
Later workers s,s have shown that unstable orgonometallic compounds are probably intermediates in the coupling reaction:
gR-Mg-X + M3C3 -> R'M'R + SMgXs (b)
R-M-R -* R'R + M (c)
Numerous attempts to prepare and isolate these organometallic compounds have been made, without success.
In 1951 Kealy and P^uson4 attempted the preparation of the hydrocarbon fulvalene6* I* from cyclopentadienylmagnesium bromide and anhydrous ferric chloride.
=1
Instead of the expected coupling product they obtained a new organo-iron compound which analysed for Ci0HioFs. They considered this compound to be iron b_is--cyclooentadienyl formed by reaction b, above.
Less than a month before this discovery was reported Miller and others5 reported that a yellow crystalline compound. C10Hi0Fe . is obtained by passing cyclopentadiene over reduced iron in the form of synthetic ammonia catalyst at i£O0° and atmospheric pressure in the presence of nitrogen.
The authors of both paoers realised that this compound was exceptional: it is stable to heat, water, alkali and concentrated hydrochloric acid. It is volatile in steam or ethanol and could be readily sublimed.
Kealy and Pauson4 wrote structure II for the compound and suggested that it acquires a negative charge, becomes aromatic and resonance forms such as III participate.
II III
-2-
Miller and coworkers6 also suggested structure IT by- analogy to potassium cyclopentadienyl7.
According to Fischer and Pfab8 iron bis-cyclopentadienyl is a penetration comolex. These pre stable complexes in which the valence electrons of the central atom form common shells with the electron Pairs binding the groups eg. the PtCl63 and Fe(CN)64~ ions?. In the compound under discussion the effective atomic number of the iron atom is 36 i.e. a krypton configuration as In the ferrocyanlde ion. These authors cited the diamagnetic properties of the compound as proof for this type of structure. They obtained evidence for an iron bis- cyclopentadienyl cation which they formulated as TV.
rw
(C5HB) =t Fe ?= (C5H5)' 1
rv
An important contribution as to the nature of the iron com- pound was made by Woodward and coworkers10' J S These workers also obtained evidence for a cation, [(C5HB)3Fe] ', and attributed the blue color, observed by Kealy and Pauson4, which accompanied solution of the compound in sulphuric or nitric acid to this cation. The cation was isolated as a crystalline tetrahydro- gallate, C10H10FeGaCl4. They noted also that the substance was diamagnetic and that the infrared absorption spectrum indicated a single sharp band at 3.85/u. From this result it was concluded that there is only one type of C-H bond in the molecule and structure V was suggested. They also proposed the name ferrocene for this compound.
/?
■ f
V
v-
This molecule consists of two rings each containing five equivalent C-H erroups so that the compound might be expected to behave as an aromatic substance. The idea of aroma ticity was probably also in the minds of Kealy and Pauson4 although they did not elaborate on it.
Woodward and coworkers11 showed that the compound does not exhibit any properties typical of polyolef lnic substances. With acetyl chloride a diacetyl derivative was obtained and with p-chloropropionyl chloride bis-p-chloropronionylf errocene and bis-acryloylferrocene were obtained. Oxidation of the diacetyl
-3-
derivative afforded a dicarboxylic acid. Substitution reactions typical of aromatic systems eg» the preparation of nitro- and bromo-derivatives could not be carried out in view of the ready oxidation to the cation by such reagents.
The infrared absorption sneetra of the ferrocene derivatives showed a marked resemblance to those from benzene. From the fact that pK^ for ferrocene dicarboxylic acid is very similar to pK for benzoic acid, Woodward concluded that the ring carbon atoms an< thus also the iron atom in ferrocene are substantially electricall; neutral.
BIBLIOGRAPHY.
1. G. M. Bennett and E. E. Turner, J. Chem. Soc, 105, 1057 (1914
2. E. Krause and B. Wendt, Ber., _56, 2064 (1923).
3. J. Krizei\rski and E. E, Turner, J. Chem. Soc, 115, 559 (1919).
4. T. J. Xealy and P. L. Pauson, Nature, 168, 1039 (1951).
5. R. D. Brown, Nature, 165, 566 (1950).
6. S. A. Miller, J. A. Tebboth and J. F, Tremaine, J. Chem. Soc, 1952. 632.
7. J. Thiele, Ber., 34, 68 (1901).
8. E. Q. Fischer and W. Pfab, Zeits, Nature., 7, 377 (1952).
9. W. Huckel, Structural Chemistry of Inorganic Compounds, Elsevier Publishing Co., Inc., New York, 1950, Vol. I., p. 58,
10. G. F. Wilkinson, M. Rosenblum, M. C, Whiting and R. B. Woodward J. Am. Chem. Soc, .74, 2125 (1952).
11. R. B. Woodward, M. Rosenblum and M. C. Whiting, J. Am. Chem. Soc, 74, 3458 (1952).
THE VICINAL ADDITION OF CERTAIN REAGENTS TO AROMATIC SYSTEMS Reported by William S. Friedlander October 10, 1952
In 1885 Buchner and Curtlus in searching for a suitable solvent for diazoacetic ester found that it attacked such aromatic compounds as benzene and toluene to yield the corresponding hepta- triene carboxyllc acids 1 . Further work has shown that diazo— acetic ester will add to benzene derivatives with unsubstitued ortb.o positions to yield norcaradiene dic»rboTylic acids, which rearrange with beat and alkali to cyclohentatriene carboyylic acids, the corresponding p^enylacetic acid, or a substituted phenylorooionlc acid. This is shoT<m with rv-vylene (I).
N3CH3C00ET
"7
OHgCOOH k?
2& CHaCHjfJOQH
HaC00H X ^
CH3
Doering and co-workers5'6 have used di^zomethane under ohoto- chemical conditions to produce trooolone (II) from benzene and tropone from anisole.
CHoN
3«a
^6 6
->
Nk
or
KMhO,
HO-
3\ f
or
II
*N |
^ |
T |
\c=o |
v* |
^H |
lis |
-8-
This method does not exclude Ila as a possible structure. Th$ yield of II, isolated as the copper salt, is 1% * Bartels-Kelth and Johnson (7) h^ve used ethyl diazo^cetate to synthesize tropolone carboxylic acid from veratrole.
In the case of benzene derivatives "which have no o-un-r substituted positions such as mesitylene (III) or durene, only rearrangement products are Isolated.
CH3
A
H Sj i
4T
JH.
NsCH3C00E?
HSC
COOET
Hs H3C_1]
CH3CH3COOET
Plattner4 has used the reaction of ethyl diazoacetate to oroduce azulene carbovylic acids (JV,V) from indane.
NSCH3CC0E!
(l)hydrolysis
COOH
->
(2)dehydrogenation
>
and
IV
COOH
These facts led Buchner to formulate a rule3: Condensation of ethyl diazoanetate with an aromatic hydrocarbon always involves addition to g non- substituted carbon atom. If the nature of the hydrocarbon nreouldes this tvn>e of addition, a rearrangement oro- duct of the h^nvcllo ester is obtained rather than the bicyclic ester itself. '
Reaction of dlazoacetic ester with condensed ring systems produces very stable norcaranes. For erprrrole, the 9, 10-dihydo- phenathr-9,10-ylene*cetlc acid (VI) produced from ohenanthrene iV; can be heated for 6 hours with sodium hydroxide in ethylene
-3-
glycol at 170° c and then can be recovered unchanged2. The structure of Vi was Droved by degradation to the known 1-(2T- carboxyohenyl)-9,<>-cyolopropanedicarboxylIc acid (VII).
.An
NaCHoCOOET
■s^s
.ckcooh -*«
/ to J
^
CHCCCH
CHCOOH
V
VI
VII
Other workers have shown that diazoacetio ester will add to the 1,2-bond of naphthalene1, the 1, S-bond of anthracene3, and the 4,5 (or 9,10) bond of pyrene3. In all cases the product (usually In low yield) is a norcarane carboxylic acid as illustrated with ohenanthrene (V).
The stability of the norcaranes formed from condensed ring systems is of some interest. It may be that the cyclopropane ring is conjugated with the remaining aromatic bonds of the v?.*? system8.
Anothe manner is o between thi addition to has used it present in tetroxide i colored, cy Hydrolysis to o-auinon shown with
r reagent which attacks aromatic systems In a vicinal smium tetroxide. Criegee9 has shown that the reaction s reagent and double bonds is quantitative and that aromatic bonds of high order also occurs. Badger to determine the amount of "double-bond character" carcinogenic agents1 x* When the addition of osmium s carried out in pyridine _„ the first product is a cllc osmate ester (complexed with pyridine) (VIII). of this produces cis 1,2-dIols, which can be oxidized es or dehydrated to "give Phenols. This reaction is 1,2-benzanthracene (IX).
^^
_4~
1 |
P.V |
|
C Ck 1 |
>0s'<S |
^0 |
c — o- I |
Py |
"^0 |
VIII
A
-H30
(l) OgO
su*
(2) Hydrolysis
>
OH H
Ox
Cook and Sohoentsl " have extended this reaction to many carcinogenic hydrocarbons such as the various rnethvl substituted 1,2-benzanthraeenes, ehrysene and pyrene. These ovidations to diols represent the first successful chemical qxidations of benzanthracene tyoe hydrocarbons in positions other than the reactive meso/*** position of the anthracene unit which is present.
Wibaut13 has done extensive work with the addition of ozone to aromatic systems. As with osmium tetroxide and diazoacetic ester the initial attack comes at the position having the highest bond order. Generally, the reagent has been most useful in stru- cture determination. However, Vollman has used it to prepare 4-formyl-5-carboxyphenanthrene from pyrene13, and Newman1*5 has
tJi
-5-
used it (going via Vollmanrs compound) to prepare 4,5-dimethyl- phenathrene.
The theoretical considerations relating to the addition to aromatic systems of these three reagents are quite interesting. As has been pointed out in the case of osmium tetrovide and ozone, the oxidation occurs at the bond which has the lowest "localization energy" for the -tT electrons. This position is almost never the same for normal electrophillc attack particularly for the larger condensed aromatic systems. The case of pyrene is perhaps the most striking of all. It normally undergoes electroohilic substitution at the 1,3,6, and 8 positions, but ozonolysis goes first at the 4,5 then 9,10 positions14 and reaction with osmium tetroxile goes at either the 4,5 or 9,10 position10.
REFERENCES
1. E. Buchner and S. Rediger, Ber. 36, 3502 (190?) .
2. N. L. Drake and T. R. Sweeney, J. Org. Chem. , 11, 67 (1946).
3. G-. M, Badger, J. W. Cook and A. R. Gibb, J. Chem. Soc, 1951, 3456. n n
4. P. A. Plattner, A. Furst, A. Muller and A. R. Somerville, Helv. Chim. Acta, 34, 971 (l95l).
5c W, von E. Doering and L. H. Knox, J. Am. Chem. Soc, 72, 2305 (1950).
6. W. von E. Doering and F. L. Detert, ibid, 21, 876 (1951).
7. J. R. Bartels-Keith and A. W. Johnson, Chem, and Ind., 1950, 677.
8. R. V. Volkenburgh, K. W, G-reenlee, J. M. Derfer, and C. E. Boord, J. Am. Chem. Soc, 71, 3595 (1949).
9. R. Criesee, B. Marchand and K. Wannowius, Ann. 550, 99 (1942).
10. G-. M. Badger, J. Chem0 Soc, 1949, 456.
11. J. W. Cook and R. Schoental, J. Chem. Soc, 1948, 170.
12. J. P. Wibaut, Comotes Rendus de la Quinzieme Conference, Union International de Chimie Pure et Appllquee 1949, p. 79. Vollman, et al, Ann. 551. 1 (1937). G-. M. Badgei" Rec Trav. Chim., 71, 468 (1952). Mo S. H&m&'y a :id H. S. Whitehouse, J. Am. Chem. Soc, 71, 3664 (l949)o
13. 14, 15.
("k
THE SYNTHESIS AND PROPERTIES OF CYCLOOLEFINS CONTAINING NINE AND TEN CARBONS
Reported by Elliott E. Ryder
October 10, 1952
The report of the synthesis and properties of eight membered carbocycles containing oleflnic and acetylenic linkages1 ' 2' 3 brought about an increased interest in similar compounds con- taining large rings.
Cyclononyne and cyclodecyne ^ere prepared in the following manner in about ten oercent yield4'5.
~N
Na
C=0
C3H503C{CH3)nC03C3H5 » (CH3)n iQ ^% (CKs)n k_ Q
xylene G~° CH,C03H V y
V_
f~
NH,NH
3lm3
, ' . C=NNH3 TTp-0 ^=NN ' 4
iM* (f^ * ^ (f^n Um -±» «*,),
V
V — J
\
c. c
The synthesis of cis-and trans-cyclononene was carried out by various methods.
C-H
Pd
^ (CH3)~ ""H2°
(CH3)7 I C
Na
C-H
S
8
-H30
H
X
/
/
/
/ CHOH
(CF3)
3/ 7
CH
>*.
3
/
\
CH;
NTHg \ H-C v
°~E /__..M^_ (CH3)7 i* .0
37 CHN(CH,),I
Cis-and trans- cyolodecene were prepared similarly
-2-
(CH3)B | _«»_» (CHS)8 fH ^__ (CH8). ^
/
CHOH (CHg) g ;
pH3
\
phthalic
anhydride
CHN(0H3)aI / Q-H CH2)8 cHa A^(CHS)8 J ^Zn
It is interesting1 to compare the properties of the eight, nine, and ten membered cycloolef ins. In strainless acyclic com- pounds containing multiple bonds t^ere is a decrease in refractive index in the order alkyne, cis-alkene, trans-slkenc. In the eight membered cyclic system the Positions of the cis-and trans- olefins are reversed, due presumably to the relatively great strain of the trans modification. In the nine membered series the cis and trans forms have very nearly the same refractive index, indicating the Presence of a small amount of strain in the trans isomer.. Cis-and trans-eyclodecene have values which vary considerably in the order which would be predicted from a con- sideration of acyclic compounds, thus implying that essentially no strain exists in the ten membered system.
Another comparison of the relative amounts of strain in these carbocycles is given by a study of the dehydration of the corres- ponding alcohols. Cyclooctanol gives only cis-cyclooctene while cyclononanol gives a mixture of the els and trans isomers, Cyclodecanol yields only trans-cyclodecene.
A final criterion by which one may Judge the relative sta- bility of isomers is in the reaction with phenyl azide1. Trans- cyclobctene reacts with this reagent within a few moments, considerably faster than the cis-form, Trans-cyclononene gives a crystalline adduct within a few hours while the cis isomer fails to react. Neither modification of cyclodecene gives any reaction,
3-
as might be expected.
It is of interest to note that by a study of structural models it is seen that enantlomorphs of trans— cyclononene should exist.
w
BIBLIOGRAPHY
1. K. Ziegler and H. Wilms, Ann., 167, 1 (1950).
2. L. E. Craig, Chem, Revs., 49, 103 (l95l).
3. N. A. Domnin, J. Gen. Ohem, (U. S. S.R.), 8, 851 (1938); C.A. , 33, 1282 (1939).
4. A. T. Blomouist, R. E. Surge, Jr., A. C. Sucsy, J. Am. Chem. Soc, 74, 3636 (1952).
5. A. T„ BlomQuist, L. H. Liu, J. C. Bohrer, J. Am. Chem. Soc, 74, 3643 (1952)1
.--..''
lb
THE ALKOXYLATION OF SIMPLE FURANS AND RELATED REACTIONS Reported by Paul L. Oook October 17, 1955
Introduction
The addition of alkoxy groups to the cc-oarbons of furans results in the formation of stable 3?5-dialkoxy-2,5-dihydrofurans The importance of these addition products is illustrated by the fact that they can be easily hydrolyzed to the corresponding un- saturated 1,4-dicarbonyl compounds whioh often are accessible only with difficulty by other methods. This alkoxylation pro- cedure is demonstrated in the following equation:1
• — , — « -» ±5 1* 2 |
m. .. .* 1 ^ , |
H30 ... > |
1 |
CH ) |
;l I CH30H |
CH„Oi J0CH3 + [ ?H3r] |
? |
CFO |
CHO |
Alkoxylation |
Alkoxylation of furans was accomplished in 19,37 by Meinel3, and by Clauson- -"Kaas and his associates3 T,iith bromine in alcohol at low temperature iu the presence of potassium acetate which neutralised the hydrogen bromide formed by the reaction- However, there were certain limitations to this reaction.. This method could not be used to alkoxylate furans which had pn electronega- tive substitusnt in the a Position- such as furoic acid, ethyl furoate, ethyl fury la cry late, acetyl fur an and *>,9 5-dlbromofuran. One exception was furfural3, but in this cape dimethoxydihydro- furfural dimethyl acetal wag formed, so that "0 rob ably a Totali- zation or semi-ace talization had taken Place prior to methoxy— lsticn, Methyl °-furoate was also methoxylated4, but the product was obtained in an impure state and in a yield of only 1?. per cent.. Another limitation of this reaction was the fact that the dialkoxydihydrofuran was contaminated with a small amount of some halogen containing impurity from which hydrogen bromide could be generat^do This impuj'j ty had an adverse effect upon the stability of the acid— sensitive dialkoxyfuran.
Recently Clpuson-Kaas has developed a methoxy la t Ion method which is simpler ana cheaper than thi one above and which gives « halogen-fres product6-" b< Furan is mixed with a methanolic solution of ammonium bromide and the mixture is electrolysed. At the cathode hydrogen and ammonia are formed, and at the anode bromine is produced. The bromine reacts immediately with furan and methanol to give dimcthoxydihydrof uran and hydrogen bromide. Ammonium bromide is regenerated from the ammonia at the cathode and the hydrogen bromide. The net equation for the dpocosp is:
-2-
NH^Br + 2CH30H >
+ H.
'0
s
CH„0; ;OCK,
The yield of dimethoxydihydrofuran Is 73f, about 10^ higher than that "by the other method.
This new electrolytic process has also made possible the methoxylation of a furan with f1 negatije a substituent. A 68^ yield of analytically Pure dimethoxydihyc'ro-S-furoic acid methyl ester Has obtained by electrolyzlng a mixture of methyl furoate, methanol and sulfuric acid.
-O
0H„OH
T~£ 4
^ICOOCH, S nA CH,0i
XC"C00CH.
— >
CH = CH
F~0 i
GHO C0C00CF.
GH.Ol
..OCH.
NKV
C00GH.
It is possible that this method may be extended to other furans with a negative substituent.
The electrolytic method of alkoxylation has also been tried on three p— substituted furans, namely, p-isopropylfuran, 4- isopropyl-2-furaldehyde dimethyl acetal, and methyl 4^isor>rooyl- 2-furoate. The product from the alkoxylation of P-isoDrooylfuran was hydrolyzed and the hydrolysis product treated with hydrazine to give 4-isor>ror)ylpyridazine.
...0H(CH3)
x.
o-
~ l
'CH3)3
F,0
CHgOl ;ocHq x0^
GH = C-CF(CH3)3 CHO CHO
NPH
2' 4
__. CH CF3)3
^N— fir
-.**.
Acetoxylation
The addition of two pcetovy grouos to furan was reported in 1947 ^y Cl^uson-Ka^s3' 7. The reaction was carried out hoth ^Tith lead tetraacetate *=nd with bromine in acetic acid, the better yield being obtained with the lead salt. In a recent oaper8 the same author reoorts an improved method of acetoxylation, again using lead tetraacetate. He also succeeded in isolating the cis and trans isomers of °, 5-diaeetoxy-2, 5-*dihydrofuran.
^
Pb(OAc)4 f
\
-V'
//
CAc
_ «
CAc
0.A,
HoO
CHOHO
i! + SACOK
CHGHO
TAC;:0
• <
2, 5-Diproprionoxy- and 2, 5-dibutyroxy-2, 5-dihydrofurans were also prepared from furan and the corresponding lead tetracyloxalates.
The discovery wag also made that nyrolysis of 2, 5-diacetoxy- dihydrofuran lead to 2-scetoxyfuran9, hitherto unknown. The yield of this reaction is low (35f), however.
2,5
-AC OH
AcO i .OHC
I i i
II
1 1
:oao
o
35*
2-Acetoxyfuran has been used »s an intermediate in the preparation of certain 5-substituted 2-oxo-2, 5-dihydrofurans
l !
^oy
L OAC
Br:
!ri . = 0 X0X
AcB:
■J' >-0Ac + Pb'OAc)
N>"
■AoOl .=0 + Ac30 + ?b(0AC)3 0'
The acetoxylation of furans with Pb(0Ac)4 is not nearly as
<~<J
-4-
general for furans as is alkoxy lation. Clauson-Kaas, Limborg and Fakstorp10 failed to acetoxylate .furoic acid and ethyl furoate by this method,, Attempt? to acetoxylate other a- substituted furans such as silvan, furfuryl acetate, furfural diacetate and 2-acetylfuran have also met with failure11. However, p-isopropyl- furan ^as acetoxylated with lead tetraacetate in fair (54*0 yield,
CH'CH3)3 . . . CH'GH3)3
|i || P^iOAck_ —
I1 ;1 AcOl LOAc
Summary
Improved methods of alkoxylation and acetoxy lation of furans have recently been developed „ Although a Ikoxy lation methods pre quite general for both a- and f3- substituted furans, acetoxy- lation has been successful only with furan itself and p-isopropyl- furan.
REFERENCES
1. N, ^.ia^^QA^Kaas, F. Limborg, and J. Fakstorp, Acta Chem. Scand., 2, 109 (1948).
2. K. Meinel, Ann. ^16, 231 (1935).
3. N. Clauson-Kaas, Kgi, Danske Videnskab Selskab, Mat.-fys. Midd., 24, 6 (1947;.
4. D. G-. Jones and Imperial Chemical Industries Ltd, Brit, patent 595041; C A., 4?, 2992 (1946).
5. N. Clauson-Kaas"~Tto Kimisk Vaerk Eo^e AIS) . Belg. Patent 500,356 (1951).
6. N, Clauson-Kaas, F. Limborg and K. G-lens, Acta Chem. Scand., 6, 531 (195?!).
7. N. Clauson-Kaas, Acta Chem. Scand., 1, 379 '1947),
8. N0 Elming and N. Clauson-Kaas, Acta "Chem. Scand., 6, 535 (1952).
9. N,, Clauson-Kaas and N. Elming, Acta Chem. Scand., 6, 560 (1952).
10. N. Clauson-Kaas, F. Limbore; and J. Fakstor^, Acta Chem. Scand. 2, 109 (1948) ,
11. N. Elming, Acta Chem. Scand., 6, 578 (1958).
kTJ
ATTEMPTED SYNTHESES OF SIMPLE PENTALENES Reported by John H. Demuth October 17, 195?
Introduction: Pentalene is ap yet an unknown hydrocarbon com- posed or two fused cyclooentadiene rjngs. Its structure, the numbering system used in naming pentalene derivatives, and the carbon to carbon distances calculated on the basis of theoretical considerations17'18 are shown in formula I. Armit and Robinson first postulated pentalene asp possible aromatic type, and with the recent surge of Interest in non-ben?enoid aromatic compounds, oentalene and its derivatives have been the subject of a number of investigations.
Whether oentalene will show aromatic or olefinlc behavior in its reactions is a matter of controversy. In a theoretical Paoer in which he used the molecular orbital theory to calculate pi electron densities, bond orders and bond lengths, Brown17 predicts that once formed, pentalene should be a reasonably stable molecule. Electrophilic attack should occur at carbon number two, and nucleophillc and free-radical attack at carbon one, since position one is the ooint of lowest electron density and of highest free valency. Craig and iiaccoll1 9; 30, on the other hand, using the valence bond method have come to the con- clusion that "pentalene should show marked unsaturation and un- equal carbon-carbon distances." The meager experimental data available indicate that perhaps the latter conclusions are the correct ones.
Types of Pentalene Compounds:
A. Pentalene, Bicy cl r [ 7. 7.0] o c t atetraene. The first attempt to synthesize pentalene was made by Barrett and Linstead7 who sought unsuccessfully to dehydrogenate bicyc 1 o [3.3.0. ] - octane over both Platinum and selenium. In their attempts to prepare the compound, they noted several interesting differences between the bicyclooctane ring system and the decalin ring system, notably: fa) trans bicyclooctane has a higher heat of combustion than the els isomer, while in the decalin series the reverse is true; <b) the cis bicycle- compound is unchanged by a Platinum catalyst which converts cis decalin into naphthalene; and (c) cis bicyclooctane upon standing over aluminum chloride rearranges to blcyclo[ 3.2.1] octane ill) — a change from a strain-free to a strained configuration — -whereas cis decalin under the same conditions is converted irreversibly to the ,trans configuration-^- a change from a strained to a strain-free con- figuration.
- fi-
ll
Blood pnd Linste^d have recently reported n ne^t attempt to synthesize pentalene sccording' to the equations shown below8*
EtOOC-GH3-CH3-CH~CCOSt 3r
+
Na EtOOG-CHg-CKg-G-COOEt
6n
0 II
EtOOC-CH3-CH3-0H-COOEt Et000-CH2-CH2-9-G00Et ON
H3 Ni
si
Ha0
1. CHgOH
HOOC-CH3-CH3-CK-COOH
2. Dieckmann H00C-0H-CH3-CH3-C0(
1. CHsJfeI
2. H30
njj oxs
<«.
/
\
V
CHs-.OH
Br.
\V
OH
\
P-CH30SO3C1
OH
OS03j6CT^:
cr^
0303j6CH,
-3-
Attempts to decompose (IV) thermally led to the formation of hydrogen bromide and an unsaturated product which at once polymerized to a dark non-volatile material, treatment of (IV) with silver acetate produced an unsaturated ketonic oil which was so unstable that further investigation of it was abandoned.
It wps thought that oerhaps (V) could be disprooortionated and then dehydropenated to 1,4-dimethylpentalene . Accordingly, a series of reactions wpg carried out under increasingly severe conditions, but they all failed to produce any change in the starting material.
B. Benzooentalene Derivatives: Benzopentalene, 1,2,3,8,9, 10-hexahydrocyclooentaindene is not yet known, although its isomer biphenylene has been synthesized31. It was thought for a time that biohenylene might actually be benzopentalene. Baker and his collaborators have "attempted the synthesis of benzo- pentalene by the following routes3* 4' 6 .
(A)
0 |
HO 6 |
.» » .» |
K y |
,-." \ / \/ \ |
|
».-■ » > |
i t |
[ f i ! ^MgX |
«L_ |
' v -* -vo • |
OH
I III
! PPA I Anh. CuS04
V *
t>
■■'-■■. / \
i
-o
ii iv
It will be noted that attempted ring closure of °-phenyl- 3-ketocyclopentane-l-carboxylic acid (I) with polyphosohoric acid, a new cyclodehydrating agent which was discovered in this laboratory as being useful in such ring closures34, failed to give the desired diketone (III). Instead, the elements of formic acid were lost giving rise to 2-ohenylcyclopentene-2-one— 1, The desired ring closure was accomplished by the use of fluoro- sulfonic acid, another new cyclodehydrating agent5.
_4~
ComPound TV could not be further dehydrogenated.
ci-c y
CRH
6n6
Aid,
->
I
\ y
.s.
v"
C«H
SJ16
"stct
o
M
\ /
/•-.
V
Clemmenson reduction of (V) yielded 1, ?, 3, 3,9, 10-hexahydro- cyclopentaindene which could not be dehydrogenated catalytically in either the liquid or vaPor phase, Bromination of (v) gave and a -bromoketone which wag stable toward silver oxide, potassium acetate and pyridine at 120° or toward quinollne at 170°, but was reconverted to (V) by alcoholic potassium hydroxide
A proposed new synthesis of benzopentalene is shown below6.
0 /»
<■' V V > 0BKX1ONO , •-•'"/ I .; | i .
.>. U — I.-., ..4=0 k.. ■;.
^y NOH
u
NHs0H, ^V'X^X^OH [H]
0
•■V
„=N0H
_L . ;=o
NH3 I .• '"' .v" ' v^^X- NK.
.i=N0H
NV
\j. „._L_Lnh3
VII
Exhaustive.
Me thy la t ion
■x
V
Steps from (VII ) have not yet been fully investigated.
C. Dlbenzopentalene Derivatives: Dibenzooentalene (i) is the only oentalene to have been synthesized. Its precursor, r, 6- diketodibenzopentalene was prepared by Roser33 In 1888. Brand and his cc— workers have reported the synthesis of several more or less complex diben70oentalene_derivatives of the general types shown below, (II) and fill)10 X6.
-5-
/V\7-V
R
i
4 8
f 1
111 » :- ' -
v V
R II R = CI or Ar
■ ii
III
R = X or Af
Treatment of (II) (R = Gl) ^ith 7,inc dust and acetic pcid yielded 2,6-dihydrodibenzopentalene15' sS. This compound adds "bromine but rabidly loses hydrogen "bromide and polymerizes, unles some device is employed to remove the HBr as it is formed. By carrying out the reaction in the presence of ammonia, Blood and
Linstead9 were able theoretical amount.
to isolate dibenzopentalene in QO°f of the
Dibenzopental lets which dissolv pound shows olum-c has no definite me chars at higher te acid, but dissolve solution the color It polymerizes in Chemical reduction
ne is reported as crystalizing in bronze leaf- to give orange—colored solutions. The com- nn-pR sr»rmr»f=> nnfler ultraviolet li.
I-
olored fluorescence under ultraviolet light. Iting point, but softens between 275-2805, and mperatures. It is insoluble in orthophosphorlc s in concentrated sulfuric acid to give a green
of which is destroyed by addition of ice water the presence of traces of mineral acids.
leads to 1,4-addition of hydrogen.
entalene is clearly t, no chemical evi- the oentalene system which can be digni- the long- wave absorp- ene and its dichloro— rom that of linear and indicates some ed state of the
Summary: "The general chemistry of dibenzop that of a conjugated diene. There is, as ye dence either from formation or reaction that has any soecial stability, certainly nothing fied by the term 'aromatic'. Nevertheless, tion, at about 400-420 m./^, of dibenzooental derivative shows a pig nlfipant difference f diene 3 with the same number of oi electrons degree of resonance interaction in the excit molecule9."
BIBLIOGRAPHY
W. and Robinson, R., J. Chem. Soc. 121, 8?8 (1922).
ibid., 1945, 258. and Leeds, W, C-., ibid. . 1948, 974. , and Jones, P. G-., ibid., 1951, 787. ., Coates, G-. E. and Slocking, F. ibid. , 1951, 1376. ., G-locklne-, F. and McOmie, J. F. W., ibid., 1951, 335' L. W. and Llnstead, R. P. ibid. , 1936, 611. T. pnd Llnstead, R. P., ibid., 1952, 2255. T. and Llnstead, R. P., ibid. , 1952, ?263. Ber. 45, 3071 (1912).
and Ludwig, H., ibid. , 53, 809 (l920). and Hofmann, F. W. , ibid. , _53, 815 (1920). and Nuller, K. 0. ibid., 55, 601 (1922) f and Ott, H., ibld.y 69, 2514 (1936).
1. |
Armit, |
J. |
2. |
Baker, |
W • > |
3. |
Baker, |
w. |
4. |
Baker, |
W-, |
5. |
B^ker, |
w-, |
6. |
Baker, |
w., |
7. |
Barret |
t. L |
8. |
Blood. |
6. |
9. |
Blood, |
c. |
10. |
Brand, |
K-.t |
11. |
Brand, |
K., |
12. |
Brand, |
K. |
13. |
Brand, |
K. |
14. |
Brand, |
K., |
-6~
15.. Brand, K., and Oft, H. , Ibid. 69, 2504 (1936).
16. Brand, K., and Hemming, W. ibid., 81, 382 (1948).
17. Brown, R. D., Trans. Faradsy"3oc. 45, 296 (1949).
18. Brown, R. D., ibido. 46, 146 (1949).
19. Craig, D- p- and Maecoll, A., J. Chem. Soc. 1949, 964.
20. Craig^ D. P., ibid. . 1951, 3175.
21. Lothru.o, W. C., J. Am. Chem. Soc j33, 1187 (l94l).
22. Roberts, J. Ca and G-orham, W. F. ibid., 74, 2278, (1952).
23. Roser, W., Ann. 247, 129 (1888)-
24. Snyder, H. R. and Berber, F. X., J. Am. Chem. Soc, 72, 2963 (1950).
25. Vogel, S., Ber. B5, 25 (1952).
26. Wgwzonei, S., J. Am. Chem. Soc , 62, 745 (1940) .
ASYMMETRIC! CITRIC ACID Reported by: Richard F. Heitmiller
October 24, 195?
Introduction:
Citric acid is constantly being formed by the condensation of oxalacetic acid and completed acetate in one stage of the reaction sequence whereby fats and carbohydrates are completely oxidized in all living cells. Once the citric acid is formed, it is converted by the enzyme aconitase to isocitric acid which in turn is oxidize to a-ketoglutaric acid, Tbis comolete reaction sequence is shown below.1
CH3-C02H
I
HO-C-C03K
I
CFgCOgH
K-C-COsH
CH3C03H
HO-CH-COgH I K-C-C02H - I CH«~CO*H
CC-C03K
I
CFs-C03H
Specificity of Enzyme Attack:
iX 4
It has been observed in many laboratories that when cc-C - carboxyl labled oxalacetic acid is used, the a-ketoglutaric acid formed from the above sequence of reactions contains virtually all of the C in the carboxyl group nevt^o the carbonyl group in a-ketoglutaric acid.8 If, however, C -carboxyl labled acetic acid is used, virtually all of the C14 is found in the carboxyl group more remote from the carbonyl group in a-ketoglutaric acid.3*' Similar experiments were carried out with C11 and Ci3 with analogous results, these are summarized below:5*6
CHg-CCgrA CO-CC;
2-1
COCOgH
+
-CHgCOgH
CH,
CHg-rCOgH
CHs-COgH
C0-C03R
+ *
-CHg-COgH;
COCOgH
CH3
f *
\j ri g— 00 gii
,14
A C carboxyl labled citric acid has been isolated by Wilcox.
-8- Heidelberger and Potter by the following series of reactions.7
OH OH
?-LchVC02H Resolve Via^ £ C1CH2-C-CH3-C03H V**** y
\ Brucine Salt f
C03H C03H
C1-CH3-C-CH3-C03I
OH
H03C14CHs-C-CH3-C03H
I
C03H
The apparent difference in reactivity of the two -CH3C03H grouns in citric acid has been demonstrated even more completely by Martlus and Schorre who synthesized a,a-dideuterocitric acid and resolved each of the isomers. The levo isomer was converted to a-ketoglutaric acid which contained nearly all the deuterium, and the dextro isomer was converted to oc-ketoglutaric acid which contained no deuterium. This reaction sequence is shown below. *
. — 0 i
0=C~C-CH3-C-CH3-C03H He so lv. 3 7ia> d- Lao tone ; ^-Lactone
{/ | Brucine Salt
0 C03H
* D303 ^D303
OH 9H
H02G-.CD3-C-OH3-C03H H03-C-CH3-C-CDs-C(fc \ \
C03H C03H
I
J,
C0C03H COC03H
I f
CH3 CH3
I I
CH3-C03H CD3-C03H
Asymmetric Citric Acid:
A theoretical explanation for the conversion of a compound C(AABD) where one of the like ptoups Is isotoolcally labled, to a Product in which the isotope is asymmetrically distributed has been Presented by Ogsten.10 His theory is based on a pair of assumptions which are mutually dependent, they are,
1. Both like s:rouos of the molecule must each be com- pleted with a separate reaction center on the enzyme surface which
-3-
contains three active centers.
2. The two combining: sites of the like grouos must be "catalytically different"
Wilcox, Heidelberger and Potter have modified Ogsten' s hypotheses as follows:7
1. There must be three distinct and specific points of interaction between enzyme and substrate.
2. Some other condition (steric hindrance, directed forces, or fourth point of interaction).
A much simpler approach to the problem can be made by focus- ing attention on the interaction between the enzyme and the two unlike groups on the central carbon of the citric acid (the carboxyl and the hydroxyl). These t^o functions are both reactive but reactive toward different tynes of reaction centers, they would, therefore, tend to interact non- interchangeably with different centers on the enzyme surface. This would serve to fix the relative orientation of the citric *cid molecule with respect to the en7yme. Since the enzyme surface is highly asymmetric, the two like prouos will, most probably, lie on areas of the enzyme which differ greatly in their ability in extracting a methylene hydrogen from the -CKsCOsH. Thus, the reactivity of one of the -CH2C6SH groups, i.e. the specificity of the enzyme,- is a direct function of its' special orientation with respect to the unlike groups as opposed to the dissimilar entantiamorphic orientation to these functions of the similar -CKsC03H groups. X1
The Asymmetric Synthesis of Citrlr> Acid:
In order to account for its subsequent asymmetric degradation an asymmetric addition must be Postulated in the biosynthesis of citric acid. In oxalacetic acid, esch of the ketonic carbonyl bonds is diametrically oooosite the other. If we consider the two unlike groups in this ketone to be interacting non-inter- changeably with the two active sites on the enzyme surface, then one of the ketonic carbonyl bonds will be oriented toward the enzyme surface, and the other away from it; hence the two bonds will differ in their chemical environment, and it would be ex- pected for them to differ in reactivity toward a carbonyl reagent. Thus, in the biosynthesis of citric acid, if only one of the bonds of the ketonic carbonyl in oxalacetic acid is available for reaction with eorrolexed acetate the formation of asymmetric citric acid is explained.
-4-
3IBLI0GRAPHY
1. H. A. Krebs, Advances In Enrmolo^y, _3, 191 (1943). 9. V. Lorber. M. F. Rudney, and M. J. Cook, J. Biol. Chem. , 185, 689 (1950).
3. A. B. Pardee, C. Heidelbererer, and V. R. Potter, ibid, 186, 695 (1950).
4. R. G. Gould, A. B. Hastings, C. B. Afinson, I. N. Rosenberg, A. K. Solomon, and Y. J. Torroer, Ibid, 177, 797, (1949).
5. E. A. Evans Jr., and L. J. Slotin, Ibid. 136. 301 (1940).
6. H. G. Wood, C. H. Werkmnn, A. Hemingway, *nd A.O. Neir, Ibid, 142, 31 (1949).
7. P. H. Wilcox, C. Heidelberger, and V. R, Potter, 3. Am. Chem. Soc, 79, 5019 (1950).
8. C. Martius, and G. Schorre, Ann., 570, 143 (1950).
9. C. Martius, and G. Schorre, Z. Naturforsch, .56, 170 (1950).
10. A. G. Ogsten, Nature, 169, 693 (1948).
11. P. Schwartz, Thesis, University of Illinois (195?).
AZO NITHILES Reported by Barbara H. Weil October 24, 1952
The earliest work on azo nitriles was that of Thiele and Heuser.1 The procedures they worked out have been modified to apply to o large number of azo-bls-alkyl nitrlles
<T 16
The general method of preparation follows?
j:C=0 + CN" + NH3NH3-H3304 20$ >C— NHNH-X! -EI^.iIhLz — =>
CH, CH3 VCH3 or NaN03,HCl
R. ^CN NC_ .R
\c— n=n- C^
Comocunds have bsen orepared where R=meth\'l; ethyl, iso- Dropyl. jfr-prb,nyl, cyolopropyl, n-butyl, Isobutyl, benzyl and p- substituted benzyl,.- Other azo nitriles have been Prepared from cyolopentanone and cyolohexanone . The hydra 7. o derivatives of some cike tones have been made of the general formula II / ° but when attempts were made to oxidize to the a?o compounds, only the substituted cyciobutane or cyclooentane was isolated, (ill)
NC_ /CH3
R-CH I , B^8TRqi ^
f
0H-> — r |
CN 1 |
/ / R-CH -'J \k3— C |
( two stereoismere CN obtained) |
R=H,CH3 |
NH
NC^ NCH3
II III
An alternative method of preparation is one described in a recent patent-11 The ketone is heated with hydrazine hydrate fur several hours to "oreoare the azine which is separated and distilled then heated under oressure with hydrogen cyanide.
One of the properties of azo nitriles which have made them of importance today is their ready decomposition to "oroduce nitrogen and free radicals. Some of the m have been found to be quite valuable as polymerization initiators,0' "* s?u,"u ri'ne decomposi- tion of the azo nitriles is a convenient synthetic method for
-P-*
obtaining tetra substituted succinonltriles and succinic acids. 1,b,4's
Kinetic measurements have been made by several workers6""* 19,1B>: using different methods. The results obtained by all the workers agree substantially: namely, the decomposition reactions of the various a2;o nitriles avpeer to be strictly first order and the rate constant is nearly independent of solvent type e The facts that the products of the decomposition of these comoounds are tetraalkyisuccinonitriles, that the decomposition rate is little affected' by change of solvent molarity and that these compounds initiate vinyl Polymerization supoort the postulate of a primary dissociation into free radicals. The final oroducts in solution depend on the manner and extent of reaction between the primary radical and the solvent. From studies on vinyl polymerization, with radioactive aliphatic azo nitriles as initiators,13'13 it has been concluded that both types of radicals A° and A-N=N* are formed in the decomposition and are capable of Initiating polymer- ization.
Rate constants have been determined at various temperatures for compounds of the general formula R4CH3-) (CN)-C-N=) 3. The rate constants are about the same for R=methyl, ethyl, n-propyl, isopropyl and n-butyl, but when R=isobutyl, there is a fivefold augmentation of rate. When the azo compound from eyclohexanone was used, there was a twenty-fold diminution of rate as comoared with the main group. There does not appear to be any Dlausible reason for these major differences on the basis of reasonance due to hypercon.iugation or inductive effects. Overberger has studied the group of azo nitriles where R=benzyl, p~chlorobenzyl and p- nitrobenzyl. He found that tViere is little or no effect of the group in the oara position of the benzene ring on the rate of decomposition. He also showed t^at the steric effect of the benzyl group is comparable to that of the methyl group in AIBN.
A study of Fisher-Hirschfelder models suggests a steric effec for the different rates of decomposition. These models indicate that only the trans- configuration of the azo compound is oossible. Little difference in rates of der»omr>osltion is observed with different stereoisomers of the azo nitriles. However, there is considerable interference of Pairs of croups, R and methyl, at the two ends of the molecule with each other. The interference is of comparable magnitude for compounds whose rates of decomposition fall in the main group; it is considerably more serious for the isobutyl corrmound and much less so for the cyclohexyl derivative as compared with the others. In the construction of the isobutyl comoound, it is impossible to arrange the R and methyl groups in a way which avoids contact of alkyl groups across the C-N=N-C linkage „ The repulsive forces arising from this crowding of group? may be expected to strain the C-N bond, displacing its ootential surface and decreasing its energy of association. The Parallelism between rates of decomposition and degree of strain revealed by the models is striking.
-3-
Overberfrer has investigated extensively the products of de- composition of various azo nitriles. 8~10 Some of the compounds he identified are as follows:
'H3C
CH-CH3-C N= H3C CH3
H3C! nc ON ^-ch3
^CH-.CH3~C-C-CH3CH H3C^ H3C 6h3 \CH3
IV dl or me so
(same product in both cases)
No products from the addition of the tertiary radical to the disproportionated products (CHy ;3-GH-CH=C CH.J (GW; or (CH3)3CH-CHS C(CN;(=CH3) followed 'oy abstraction of a. hydrogen atom by the adduo'c were found.
CN NC
CH3-C— N=N--C-CH3
CK3 ^3^
CNNC
> -
\ vCNNC /
?<■* = *\
CN CN
CH3-C C-
ch, ch
un3 3
CN CN -.
cT^-g V
CK3 X.
VI
PV^<C~
VII
VIII
The isolation of a mjxed eouoled product (VIII) is indicative of a decomposition mechanism by Which relatively free radicals are produced.
The dipstereomeric 5,Sf--azo--bls-S,3,3-trImethylbutyronitrlle was allowed to decompose in benzene solution for three days. Products identified by analysis, reactions and unequivocal syntheses by other methods were the following:
CPU CN [CHo-C — C— N=]
CH.
!H.
9H3
CK3-O— ■
6Ha
CH3 CN
CH3-C CH-CF.
CH3
£N 3 • X3H<
CH,-C
gh3 cn cn ch.
CH
9.
•CH.
3 gh3 6h3 ch3
IX
CK3 CN CH3-6 — C=CH3
ch.
XI
-4-
No evidence of rearranged products XII and XIII were obtained
CN CH9 ON
CH3-CH— 6-CH3 CH3=6 6-CH3
CH3 CH3 CH3
XII XIII
Work now in orogress by Overberger17 involves a study of azo nitriles from cyclic ketones from C=5 to G=10. The rate of de- composition is an accurate measure of differences in ring strain. The results of this investigation have not yet been published.
Bibliography
1. J. Thiele and K. Heuser, Ann., £90, 1 (1896).
2. Steff, Diss. Techn. Hochsch. Munchen (1914) 6; Beilstein IV (1st Sup.), 565.
3. A. W. Dox, J. Am, Ohem, Soc, 47, 1471 (1925).
4. H. Kartman, Rec, TravQ Chim. , .46, 150 (1927).
5. F. M. Lewis, M. S. Matheson, J. Am. Chem. Soc, 71, 747 (1949)
6. C. G. Overberger, M. T. 0' Shaughnessy and H. Shalit, ibid.,, 71, 2661 (1949).
7. C. G. Overberger, P. Fram and T. Alfrey, Jr., J. Polymer Sci., 6, 539 (1951).
8. C. G. Overberger and M. B. Berenbaum, J. Am. Chem. Soc., 73, 2618, 4883 (1951); 74, 3293 (1952).
9. C. G. Overberger and H. Biletch, ibid. , 4880 (1951).
10. 0. G. Overberger. T. B. Gibb, Jr., S. Chibnik, Pac-tung Huang and J. J, Monagle, ibid., .74, ^29° (1952).
11. W. L. Alderson and J, A. Robertson, 2,469,358, May 10, 1949.
12. L. M Arnett, J. Am. Chem. Soc, 74, 2027 (1952).
13. L. M. Arnett and J. H. Peterson, ibid., 74, 2031 (1952).
14. M. Hunt, P 2,471,959; May 31, 1949.
15. C. E. H. B awn and S, F. Mellish, Trans. Faraday Soc, 47, 1216 (1951).
16. K. Ziegler, W. Deoarade and W. Meye, Ann., J567, 141 (1950). 17.. C, Gr. Overberger et _al, Abs. 122nd Meeting, AGS, 53M (Seot.,
1952).
THE STRUCTURE OF KETEKE DIMER Reported by W. S. Anderson October 31, 1952
Although diketene has been known for many years, the problem of its structure has never been completely solved. A total of six formulas (I-VI) have been suggested for it since its isolation in 1908 by Chick and Wiiamore as a lachrymatory liquid, b.o. 127°.
CK3C0CH=C=0 |
C - CH3 ! 1 H3C - C II |
H0s ] C - CH2 11 (I £ - cv III |
HO XC - CH 1! M HC - C X0H IV |
CK3 = |
C I 0 |
- CH2 i - C = 0 V |
|
CH3 - C = CH |
|||||||
i | 0 - c = VI |
0 |
Mixtures and resonance hybrids of certain Dairs of these molecules have also been postulated in an attempt to erolsin its behavior. The substance is widely used both in industry and in the laboratory, but the structure controversy continues still.
Structure Investigations by Physical Methods
(a) Raman effect, Several workers have compared the Raman shifts^ of diketene with those of other compounds possibly related t0 it*, A comparison with 2,2,4,4-tetramethyl 1, 3-cyclobutanedione rules (VII J out the dione structure (II). Structures III and VI
do not exolain lines in the double bond region of the soectrum. Vinylaceto-p-lactone (V) or acetylketene (I) could exolain this Part of the spectrum. V ¥&hm:y*Tt&gl -■ the sample in o o 4- trimethylpentane produces no significant change in the spectrum, a fact which suggests that no rearrangement takes place when the substance is dissolved.
(b) Dioole moment. Diketene has a dipole moment of 3.18 D.3 The symmetric dione structure is totally incompatible with this value. The dimer of dimethylketene, on the other hand, has zero dipole moment; consequently formula (VII) is assigned to it.
C - C - CK
u"3
CH3 - C - C, CF» 0
VII
(c) Infra-red absorption. Absorption in the 2 -14/^ range Indicates that I is not the structure, since no bands are ^resent
-CU
which could be ascribed to C=C=0. The absence of the OH bond- stretching frequency is taken to mean that no enols are ore sent. Five strong bands in the double bond region suggest that diketene may be a mixture, since no single structure among those remaining contains more than two double bonds. The spectrum is probably best explained as that of the lactone mixture (V + VI) or possibly of V alone. 3
The spectrum of the vaoor as measured by Miller and Koch4 shows considerable change in form when the vaoor temoerature is varied. These authors attribute the change to shifts in the positir of the equilibrium V^> VI.
(d) Potentiometric and conductometric data. Wassermann!s measurements5 Indicate that a proton is dissociated from diketene when in dry acetone solution. Structure V most adequately explains this acidity.
(e) Ultraviolet absorption.6 By a comparison of the UV ab*~ sorption of diketene with that of cyclobutanone, the cyclobutane- dione structures of diketene are eliminated. The spectrum strongly suggests the presence of the group 0=0 — C:=*0. I and VI have this feature; however, the acetyl grouo of I would be expected to move the absorption of ketene into the visible range to make diketene
a colored corrmound, which diketene if not. Structure VI is left. An aiDoroximate calculation of the free energy change for the trans- formation I— yVl(^F^-l to -5 -f^pO indicates that the transforma- tion to acetylketene may be very easy.
(f) Electron diffraction. Workers at Cornell7 have applied the method of electron diffraction to diketene vaoor. They conclude that structure ?'V'"nnd. VI pre compatible with the oottern obtained.
(g) X-ray diffraction. The electron density mao and bond lengths determined by this method indicate that V is correct for the crystal molecule.
Structure Investigations "by Chemical Methods
(a) The formation of dehydracetic acid, a tetramer of ketene, may be conveniently explained as a Diels-Alder reaction of ecetyl-
Votono 9
- f C=0 CH.^/^yO
Y
(b) The or.onolysis of diketene10, previously thought to
ketene.9
CH3
+ H CH^ CHC0CH3 ^ iHCOCHj
-3-
substantiate the acetylketene structure, has recently been repeated. In the new work no pyruvaldebyde is found in the Product; instead, formaldehyde and malonic acid are isolated. These products could be formed from V.
(c) N-bromosuccinimide in chloroform at room temperature brominates in the fashion s^o^n:11
Diketene NB3 ) Unstable C^CH 0
bromo-derivstive CH3-C-CHBr-C00C3KB
sole orodu^.t If the structure were VI, bromoacetoacetic ester would be the product expected.
(d) Allene and carbon dioxide, previously believed absent in the pyrolysis oroduct, no^ have been found there.12 The formation of these Products from V is understandable.
(e) Acetoacetylation reactions are explained by the lactone formulas. A new example of this type of transformation is that of a recent patent:13
dry HC0N(CH3)3
polyvinyl alcohol + diketene -CHg-CTT-CHs-CH- 19$ of
0 OH OH groups -CH3-CH-CH3-CH- 1.5 hrs. C=0 acetoacetylatec
I j 1200 CH3
OK OH 9=0
CH
.•3
References
(l) Taufen and Murray, JACS, j57, 7^4 (1945).
(?) Ane-us, Leekie, LeFevre, Le^'eVre, pnd i'J'assermann, JC3 1935. 1751.
(3) %iffen «nd Thompson, JC3 1946. 1005.
(4) Miller and Koch, JACS 70, 1890 (1948).
(5) w^ssermann, JC5, 1948, 1323.
(6) 0«3vin, Magel, and Hurd, JACS 6^, 2174 (l94l).
(7) Bauer, Brcernan, and Wright son, Abstracts of Papers, 199th Meeting of' ACS, April, 1946 Page 15?.
(8) Katz and Lioscomb, J. Org. Chen. .17, 515 (195S),
(9) Whitmore, Organic Chemistry, ^nd ^d. p. 232. (lOfr Hurd and Blanchard, JACS 7?, 1461 (1950). (ll) Biomquist and Baldwin, JA53 70, 29 (1948) .
(12) Fitzpa trick, JACS 69, 22-^6 (1947).
(13) Jones, U. S. Patent 2,536,
980 [CA*46, 2572 (1952)]
THE SYNTHESIS AND PROPERTIES OF SOME SIMPLE AMINO AND HYDROXY
PTERIDINES
Reported by William R. Sherman October 31, 1952
Between 1891 and 1895, Rowland Hookins4 isolated several pig- ments from the wings of butterflies, and reported their extreme infusibility and very slight solubility. At this same time 0. Kuhling5 prepared what is known today as 2,4-dihydroxypteridine, which also exhibited the refractory and insolubility characteristic of Hopkins' compounds. It was not until 1940e> 9 that it was realiz that these tT*ro workers had isolated compounds of the same chemical family. Since 1940, the field of Pteridine chemistry has drawn an ever increasing number of workers into it, due to the highly important physiological role played by compounds containing the oteridine nucleus.
This paper is limited to a survey of the syntheses and proper- ties of some of the simple mono- and di- amino- and hydroxyoteri- dines. These compounds occur as the nucleus of Hopkins' pigments and of many other naturally occurring compounds, many of which are of extreme biological importance (e.g. xanthopterin, folic acid, rhizopterin, folinic acid, etc.). A great deal of work is now being carried on with simple amino- and hydroxyPteridines for the Purpose of elucidating their chemical and Physical properties, with the hope of throwing iight on the role of these substances in animal metabolism.
To this date the apparently anomalous behavior of these simple pteridines toward p variety of reagents is, for the most Part, without explanation. Many of these properties seem to be unique to derivatives of this single heterocyclic nucleus.
SYNTHESIS:
\
The only published synthesis of Pteridine (I) itself makes use of the condensation between 4, 5-diaminopyrimidine (II) and glyoxal. (1,2)
4 JL JS N
HC=0 HSN s# >s 7r^^V^ ^> 3
HC=0 ° ^V
S -4
II I
This fundamental reaction is employed in the preparation of the following pteridines:
2- hydroxy (2) 2,4-dihydroxy (2)
2-amino (2) 2,4-diamino (lO)
4-hydroxy (2) 2-amino, 4-hydroxy (10)
4-amino (2) 2-hydroxy,4-amino (lO)
-2-
b 11 erf *r"hitf;"fi a :re Prepared by condensing the appropriately sub- stituted 4,5-diaminopyrimidine with glyoxalT
Other routes to 2, 4-dihydroxypteridine Include Kuhling's original synthesis using tolualloxaBine (ill)5, and G-abriel ^nd Sonnfs later synthesis from pyrazine-S, 3-dicarboxamide (IV).6
OH
N
N
KMn04
III
,N
c
0 H
XNV OH V
OH
^S.-~ CNHS i CNHo
6
IV
>
KOBr
Albert3 used the following as an alternpte route to the 4- hydroxy compound:
,N
^C(OEt)
B .J? J l^er 10BVer'7 Albert Prepared the two monohydroxyoteridines substituted in the pyrazine ring!
OH tt6 - OEt
I
0=C - OEt
+ H
HaN
««+u ThI 6^?-^;ihydroxy compounds may be prepared by the oxidation of either 6- or 7-hydroxypteridine7.
HNO3 20s
^0
XN N
H303 Boiling HOAc
N-^N^-N
-3-
PHYSICAL PROPERTIES: Solubility (2):
Introduction of even one hydroxy or orimary amino group into the oteridine nucleus ereatly lowers solubility in all neutral solvents. A similar effect is observed to a lesser decree in other heteroaromatlc bases.
TABLE I Solubilities in water at ^0-25° C (?.)
Pteridines Solubility R^tio
Pteridine (unsubstituted) 1:7.2
2-amino- 1:1350
2-dimethylamino- 1:2.5
4-amlno- 1:1400
2- hydroxy- 1:600
4- hydroxy- 1:200
2- amino-4- hydroxy- 1:57,000
2-amino-4,6-dihydroxy- 1:40,000 (xanthopterin)
2-amino-4,6,7-trihydroxy- 1:750,000 (leucopterin)
The amino and hydroxy substituents undoubtedly play their more common role as solublizing groups in water; however, the presence of the strongly negative ring nitrogens of neighboring molecules brings powerful hydrogen bonding into play. When a primary amino group is replaced by a dimethylamino group on the pteridine nucleus, the solubility is increased more than five hundred fold (see table I In a similar manner, the hydroxy- and aminopteridines are virtually insoluble in ethanol, benzene or pyridine, while dlmethylamino- pteridine is extremely soluble in these solvents.
The fact that all known hydroxy- and orimary aminopteridines decompose above 240°0CJ without melting, is another indication of the strengthening these croups give to the crystal lattice. In contrast Pteridine and ^-dlmethylamlnopteridine meit at 140° and 126° C. without decomposition.
Ionizing Properties (2) : (see tnble II )
When a concentrated (colorless) solution of 4-p>minopteridlne if added to an evcess of O.^N-NpOF, * yellow color aPoeprs. Slow hydrolysis to 4-hydroxyPteridine t^kes Place under these conditions; however, the ultrpviolet absorption soectrum showa the oredominant species to be 4~am!inopteridine. Thus, 4-aminooteridine forms an anion; this ohen'omenon has not been observed with the amino quin- olines nor with their analogs containing more ring nitrogen atoms.
As would be expected, 4-hydroxyPteridine is a stronger acid than the lower order nitrogen heterocycles 4-hydroxyquinoline and 4-hydroxyquinszoline. 2-Hydroxyptcridlne is a* weaker acid than its 4-hydroxy isomeride (no data on the quinoline or quinazoline com- pounds are available for correlation). The 6- and 7- hydroxy- pteridines are stronger acids than their 2- and 4- isomeridcs.
_4-
The absence of a true hydroxy 1 function in the four position, and the probable existence of the cyclic amide form, most likely accounts for the lowering of the base strength.
N-H
TAB IE IT t>Ka in water at ?0°C. (?,?)
Pterldlncs Pka , concentration
TOteridine (nation) 4.19 M/gO
P-aminooteridine (cation) 4.^9 IV 100
4-amino- (cation) 3-56 M/pOO
g-dime thy 1 amino- (cation) 3.03 M/100
2- Hydroxy- (anion) 11.13 M/100 4- hydroxy- (anion) 7.89 M/lOO
6-hydroxy- (anion) 6.7 M/500
6- hydroxy- (anion) 6.41 M/pOO 6, 7-dlhydroxy7 (mono anion) 6.87 M/500
(di anion) 10.00 M/5OO
4-hydroxyquinoline IP. 4 (in 50^ ethanol)
4-hydroxyquinazoline 10.0 (in 50# ethanol)
CHEMICAL PROPERTIES:
Acid and base hydrolysis:
Pteridine, S -amino- snd 2-hydroxypteridine are all destroyed b; cold lOH-EOl or boil' r\g N-NaOH. Even absorption on alumina is enough to convert; ?L:/* of a 1^ benzene solution of pteridine to a rec oil of unkij-j-wn composition3.
Stability of the pteridine system is increased by substituting in the 4 position, 4- hydroxypterldine is recovered unchanged from a solution of, boiling 6N-NaOH, In acid or basic solutions the 4-amlno group is smoothly hydrolyzed to a 4~ hydroxy 1 group, with slight concurrent decomposition of the 4-hydroxypteridine in the acid solution.*"3
6'-AminoPteridine is hydrolyed to 6-hydroxyoteridine in acid solution. The 6-=hydroxy compound is stable in iON-HCi at P0° snd if re crystallized i'_-om n hot I5# HC'i solution.7 7—HydroxyPteridine (the 7—amino compound could not be "ore-oared due to the unstability of the T-riydrovy oortroound toward chlorinating and aminating agents) shows p similar behavior but is destroyed by prolonged boiling in K'~HfeSG<iw 6"Hydroxyuteridine I a unstable to base, decomposing to an unchar^cterlzed uompou.nd in 0.iM.~NaCH. The 7-hydroxy compound is unde composed by boiling N-NaOH. ~
-5-
TABLE III (14) Hydrolysis by boiling 6N-HC1
Pter^dine Time Product
4-hydroxy7 2-smino 0.5 hrs. no reaction
30 hrs. 2,4-dihydroxyoteridine 2-hydroxy-4-amino 0.33hrs. 2,4-dlhydroxypteridlne
Deamination by HONO
2-hydroxy-4-amino no reaction
4-hydroxy-2-amino 2, 4-dihydrovypteridine
Since mineral acid T-rill hydrolyze an imino group but not an amino grouo, and the converse is true for nitrous acid, the above information (table III) would indicate that an amino group in the four position exists oredominately in the imino rather than in the amino form.
The following consideration might also have some bearing on th< difference in the ease of hydrolysis of the amino groups in the 2- and in the 4- positions of r^teridine.. An amino group in the 2- position nartakes of the guanidine structure, while one in the 4- position is of an amidine tyoe . It is recognized that guanldines are more stable to acid hydrolysis than amldines..
If the base- stable 4-hydroxypterldlne is alkylated in the op- position, it undergoes a profound change in character. As a direct result of the substitution the ^-alkylated, 4-(keto)-pterldine undergoes ring cleavage in 0.1N-K0H3.
OlP No Hydrolysis r>
o:-r ,r \,/ m-
o
CNHR
-.y be due in the first case to the formation of a simple an/urn retiirttf: 'further base attack, and in the second
Co^:; '-"hero no :•■•,:.. *le anion can be f or^-u. lo a hyurolytic attack en C-£. 6r:* rii.Dsj^cju.ent ring opening.
BIBLIOGRAPHY
1. w, G. M, Jones, Nature, 162, 5S4, (1948).
2. A, Albert, D. J. Brown and G-. Cheeseman, J. Chem. S0c, 10?, 47' (1951).
-6-
3. E. C. Taylor, Jr., J. Am. Chem. Soc., 74, 2380, (1952) .
4. F. G-. Hopkins, Nature, 40, ,335, (1889); 45, 197, 581, (1892); Trans., Roy. Soc, 3186. 561, (1895).
5. 0. Kuhling, Eer., 28, 1968, (1895).
6. S. Gabriel and A. "Bonn, Ber., 40, 4857, (1907) .
7. A. Albert, &. J. Brown and G-. Cheeseman, J. Chem. Soc., 298 r 1620, (1952).
8. R. Rurrmann, Ann,, .544, 162, (1940); 546, 98. (1940).
9. H. Wieland and R. Purrmann, Ibid, 165 , (1940).
10. E. C. Taylor, Jr. and C. K. Cain, J. Am. Cham. Soc, 71, 2538, (1949).
Hydrocarbons with Intercyclic Double Bonds Re -ported by II. J. Fletcher November 7, 1952
A double bond may be regarded as intercyclic if it lies between two rings, as for example, in the following compounds:
D ib i phe ny 1 e n e e thy lene
II 1 |
f ^ |
M 1 |
\ ^ |
II |
|
0 |
|
Clan throne |
ou3
Bis-cyclohexylidene 2,2t-sulfone
These compounds, however, are special cases; the first two have intercyclic double bonds which are conjugated at both ends with aromatic systems, while in the last two the intercyclic double bonds are also intracyclic. On the other hand, bis-cyclohexylidene itself has an intercyclic double bond with no modifying factors.
<o=<z>
Bis-cyclohexylidene has been reported several times in the literature, but, until it had been prepared by Criegee1 and coworkers only one accurate description of it was extant, and in this case the compound was not recognized for what it was,
2
Sabatier and Kaihle thought they had obtained I by dehydrating 1-cyclohexylcyclohexanol (II) with zinc,, chloride, or distilling it over thorium oxide, but later work by Huckel and Teunhoffer3 showed that it was the isomeric hydrocarbon III.
<Cx-0
\_^
o
II
III
•2-
Senderens and Aboulenc reported that they had obtained I ?s a byproduct in the dehydration of cyclohexr.nol with concentrated sulfuric acid at 130°, although they gave no analytical data or structure proof. A consideration of its properties, however, shows it to be dicyclohe::yl ether.
5 Finally, Zelinsky and ochuilcin reported that they had pre- pared I by the ./olff-Fischner reduction of a-cyclohexylidene-cyclo- he-anone (IV) . (IV) has long been assumed to be the product of
the alkaline sell condensation of cyelohexanone. Reese6 has shown, however, that this is not the case. The reactions he used to prove this are as follows:
A
O
HOI
/v!LA
Na 0 Me not 7^0°
Pyridine, high temperature
1) K202 0H~
2) H+
y\
v-v
0
CrO<
/N
sy
CHOH I (CH3)4 f
CO OH
C=0 I
(CH3)4 1 COOH
CrO.
E,
3
V
fHOH
(9H3)4
COOH
-3-
Reese has not proved the structure of IV, but he has proved that this structure is correct for V, which is the ketone obtained by the alkaline self-condensation of cyclohexanone .
obtained a very small amount of a hydrocarbon which from the vapor phase nitration of cyclone xane . Al-
G-rundman seems to be I
though he considered the structure I for this compound, he rejected it on the
■ i
bra is of literature information now known to be false.
At first Corigee1 and coworkers attempted to prepare I by de- hydration of II under milder conditions than those Sabatier and
Maihle had used, but even the bon III.
Chugaev reaction led to the hydrocar-
Next the action of zinc on the dibromide obtained from cyclo-. hexanone pinacol, to which the structure 1,1* -drbromo-fais -cyclo hexyl
had been assigned8, was tried, but this, compound turn ed out to be very stable.
oxane
not only toward zinc in acetic acid, but also such reagents as magnesium in ether, metallic lithium, and sodium in boiling di yielding, when it reacted at all, unsaturated bromo compounds. Ivith copper-plated zinc in di oxane, however, a saturated hydrocarbon of unknown structure was obtained.
9
According to MereshkowshiTs rule , di tertiary dibromides, when
in acetic acid, lose all their bro-
treated with potassium acetate
mine to form dienes, while di secondary and secondary -tertiary dibromides lose only part of their bromine to form unsaturated bromo- compounds. This indicates strongly that the structure assignment to this dibromide is incorrect.
Besides, by the action of hydrob romic acid on cyclohexanone pinacol at -10°C, a dibromide was obtained which was easily converter" Into I in &5% yield by the action of zinc in acetic acid at 15-20° for l/2 hour. Similarly, bis-cyclopentylidene (VI) and bis-cyclo- heptylidene (VII ) may be prepared in yields of 90^ and 85% respective
<3,-<Z>^ O-iO
0 c
-10*
Br Br
Zn, CH3COOH 15-20°
0-0
-4-
oa
VI
VII
The necessary intermediates for the preparation of the four and eight membered compounds have not yet been obtainable.
The structures of these hydrocarbons were proved by the fact that, on oxidation with osmium tetroxide, they were all reconverted to the corresponding pinacols.
The stability, as well ?s the ease of formation of these com- pounds seems to be strongly dependent on the size of the ring. This may be illustrated by the following reactions:
1)
2) H20
3) HBr
:igBr
-^s.
Pyridine
<z>-<z
I Pyridine 1 s\ N
Br
It is of interest to note here thpt almost all elimination reac- tions not involving bulky groups, or in which the molecule does not
contain a positive charge
to start with
or example, a tetra- alkyl ammonium salt, give Saytseff elimination, yielding the most highly branched olefin possible. For instance, diethylisopro-oyl- chloromethane yields, on treatment with alcoholic potassium hydroxide l,L-.dimethyl-2 ,2 -die thyl ethylene .
(ci-:3)3ch-c-(csh5)3
CI
KOH
CyH5OE
(CK
3/2C-C (C3ri5 j 3
However, in all cases Trhere a competing elimination is possible in the attempts to prepare bis-cyclohexilidene , the isomeric hydro- carbon III was obtained. This was not the case with bis-cyclo- pentylidene. These facts may be explicable on steric grounds.
■-R-
Pibliography
1. E. Criegee, E. Vo-gel and H. Herder, Ber., §5, lkM- (1952).
2. P. Sabatier and A. Maihle, Oompt. rend., 138, 1323 (1903) ; Cf. Ibid, , 15^, 1392 (1912); Bull. soc. chin., ["3], 33, 7^ (19°5) •
1. 17. Hucl-el and 0. Neunhoffer, Ann, )£7J_, 106 (193077
4. B. Sender ens and I. Aboulenc, Comr>t. rend., 1§3, $31 (1925);
137, 110^ (1927} .
5. II. Zelinsky and N. 3chuikin, Chem. Journ. Cer. A. Journ. allg. Chen., 64, 671 (1932) [Chen Zentr., 1033, II , 16733.
6. J. Reese, Ber., J5., 3^ &9^2) .
7. Ch. G-rundmann, Angen. Chen., 62, 556 (I95O) .
3. 0. ./allaeh and F. Pauly, Ann, 3&L, H"3 (1911 ) . 9. K. B. Mereshfcowslsi, Ann, *!-31, 235 (1923 ).
NEW REACTIONS OF PYRROLS S Reported by Robert E . Putnam
November 7 , 1°52
In troduction
Pyrroles have often been compared to phenols because of their reactivity towards electrophilic substitution. Like phenols they can be nitrated, halogenated, alkylated, acylated and coupled with diazonium salts. In general, substitution tslr.es place at an ex- position. However, when both a-positions are blocked substitution at a p— position occurs readily. A different orientation is noted with the potassium salts of pyrrole and its derivatives. These compounds react with such reagents as HX, RCOX, ArCOX and ClC03Et to give Il-substituted pyrrole.-. Recently Treibs, I-llohl and Ott have reported the reactions of pyrrole cr.6. aZkylpyrroles with ben- zoyl chloride, isocyanates and diketene. These reactions will be discussed in the present seminar.
Benzoyl Chloride
pyrroles is treatment of t acid chloride (1, 2) . Pyr acyl pyrrole by heating wi dride (l) . Treibs and hie pyrroles with benzoyl chlo the conditions of the Scho derivatives. Pyrroles sub react. Pyrrole itself giv less, high melting solids
Isocvanates
of preparation of lT-acyl and N-aroyl potassium salt of the pyrrole with an Le itself can be converted to an N- th the appropriate aliphatic acid anhy- hl have now reported the reaction of rice and p-nitrobenzoyl chloride under tten-3aumann reaction to give U-benzoyl stituted with negative groups do not es an oil while a.lkylpyrroles give color- suitable for characterization.
There is only one report in the literature of the reaction of a pyrrole with an isocy-nate. Fischer, Sus and ,'eilguny (]4-) added phenyl isocyanate to hryptopyrrole , I, and obtained a solid which they formulated as II. Treibs and Ott (5) have shown that this
OH,
CnH
i
3^5
\,A
CE.
Ok;
0
S.AcH
COMI-IC6H5 II
Coli
3iJ-5
GK-
OH^j.t^OONHOsHs
1
H
III
reaction is quite general for a.lkylpyrroles but that the products are of type III rather than of type II. The structures were proved by comparison of the products with pyrrolecarboxanilides prepared
-2-
by treatment of the pyrrole with phosgene and reaction of the acid chloride formed with aniline. The anilides obtained by the two methods were identical.
R1
R
/
R
A../
i!
0
w
C1CC1
R>
R»
R
..A
C0C1
n u "TT-T
R1
N
V
E»
A
ccnhcs:l
The reaction with isocyanrtes is limited to those pyrroles which do not have electron withdrawing substituents. Compounds such as 2 tk— dimethyl-3-carbethoxypyrroie , 2-methyl-J-carbethoxy- pyrrole and 2,^4— diphenylpyrroie fail to react. 'ith pyrrole and alkylpyrroles addition occurs without a catalyst. This is remark- able in view of the fact that carbon alley lati on with phenyl iso- cyanate usually requires a catalyst. Thus phenyl isocyanate attacks benzene in the presence of aluminum chloride to give benzanilide (6) and also attacks active methylene compounds in the presence of aikoxides to give aliphatic anilides (7) • ^^-e e^se of reaction of pyrroles is given by the following sequence "here R and Rf are alhyl and R" is alkyl or hydrogen. In addition to phenyl isocyanate,
RT
R
A
N I
H
R
\y
•pt!
X t
> n
substituted phenyl isocya.na.tes and benzyl isocyanate t ory react & nt s .
ire satisfac-
The products are extremely stable, crystalline solids. They are unaffected by long boiling with concentrated hydro chloric acid or concentrated sodium hydroxide. Heating with concentrated sul- furic acid or fusion with potassium hydroxide causes some hydrolysis, the original pyrrole being isolated in each case. The presence of the carboxanilide group deactivates the ring somewhat. T-Ialogena- tion of the pyrrole nucleus is still possible but introduction of an aldehyde group by the S-atterniann method is very difficult. The anilides do not couple with dirzonium salts. However, they do con- dense with formaldehyde yielding dipyrrylmethanes , a reaction typical of most pyrroles.
_^_
Dike tene
Alkylpyrroles add to Slketene in the same way as to isocyanates (3). Again substitution takes place at a nuclear carbon atom. ™his was shown as follows. The product of the reaction between dik^tene and 2 fH— dimethylpyrrole, IV, was heated with concentrated sodium hydroxide to ^ive 2 ,^-dirnethyl-5-acetylpyrrole, V, which was identi- cal with the known acetylpyrrole prepared by a Ilouben-hoesch syn- thesis .
CH3 Crl3
Y
[CH3=C=0]2
CM
A /
3 TI
T
H
^--3 .1
H
IV
OH,
01
CH/V/
CH3CN HOI
-/
CO0H,
CH3ANAC-
I I
H OH-
IH-H01
It is notable that the position of an alkyl group on the pyrrole nucleus dees net influence the ease of reaction of the pyrrole with diketene. Moreover, the addition to diketene is strongly influenced by catalysts. Sodium acetate and other bases increase the rate of rea.ction but not the yields, since they also catalyze the polymeri- zation of dihetene. As In the case of addition to isocyanates, negatively substituted pyrroles do not react.
The products have proved to be very useful In the synthesis of more complica/ced heterocycles . Inasmuch as they are p-dike tones the Knorr synthesis is applicable to the preparation of dipyrryl- ketones . By suitable choice of reactants a variety of products may be obtained:
L5t
E CH3-C=0
OC-CH3 Zn
C J?„
s, \ naC
HON C03Et
ChT.
CH^j.j^COjjEt
I
H
-4-
R»
\
y\
H! CH3
C=0 Gliz-COzEt Zn
I + I -*
xHOH 0' NCHa
R1 CH3 C03Et
V /
HAc IT NH i
V^-oA^SH,
N
0 H
Hydrazine, phenyl hydrazine and hydroxylamine react at room tempera- ture to give, respectively, pyrazylpyrrolee and isoxazylpyrroies .
R'»
R'
C8HBNHNHS
H^jT^OOOHaOOGHa
H
^
* Or
V
^
Mechanism
R"
R»
V
H
R»
R'
n^y
<
XX
R" V
R'
R^
t
H
OH,
X /
i!
I
H
N
CHj
N N' I
OH,
V
f
/l-\
Treibs and Michl (6) explain the reaction of -oyrroles with isocyanptes end diketene ss en attack by the pyrrole on the polarized form of the isocypnate or diketene.
* — =>.
H
t
4 >
/
KJ~
-5-
N
i +
H
0
I0|
0 — Cxljj —0—0x^2 +
0
lo|
Proton
N O-OHs-O^OHa transfer
i +
H
Product
V -
I +
H
0 I! -
C-N-R +
Proton ;ransfer
Product
C-N-R
+
H
This mechanism is quite similar to that generally accepted for reac- tions of pyrrole C-rignard reagents.
^y
N
I
H
+ HMgX
N
f — - T
j | MgX + RH
X
J
N
\N
-I
RX
^ X.
Bib 11 ography
Pro ion transfer
l H
1. 2.
7.
Fischer and Orth, "Bie Chemie des Pyrrols," Vol. 1, p. 27, 193^
Rainey andAdkins, J. Am. Chem. Soc, 6l, 110^ (1939).
Treibs and„Nichl, Ann., 577, 115 (1952J7
Fischer, ous and ./eilguny, Ann., U-ol 159 (1930) .
Treibs and Ott, Ann., 577, 119 (X952) •
LeuJsart, Eer., lg, g73(lgg5) .
Petersen, Ann., ^o"2a 20o (1949) .
Treibs and Michl, Ann., 577, 129 C1952) ,
THE SKELETON OF PICROTOXININ Reported by R. Thomas Stiehl
November 7, 1952
Introduction
In 1812 a naturally occurring material, picrotoxin, was isolated and found to be physiologically active. Almost seventy years elapsec before attempts were made to elucidate its structure. Investigation revealed that picrotoxin is composed of two substances: picrotoxinin Ci5Hls06, and picrotin, Ci5H1Q07.
Conroy1""4 has recently reported structure studies on picrotoxini and has synthesized dl-Picrotoxadiene . Ke also formulated a skeleton for picrotoxinin and ventured a tenable structure.
Evidence for Assigned Structure
Infrared suggests that picrotoxinin is composed of two five— mem- bered lactone rings (1777 cm."-1 and 1798 cm.""1).
Results of bromination, hydroge nation, ozonolysis, nnd infrared (weak band at 1657 cm.""1) indicate~only one double bond.
No carbonyl derivatives are formed.
A Zerevitinov determination and infrared (3450 cm."1) show the presence of only one hydroxy 1.
Brominati bromides, 015K zinc to picrot compounds have absence of the mination. Thu been involved support to the and is not Pre
on yields two sparingly soluble^ stereo isomeric mono- 1506Br, which can Quantitatively be debrominated
rhich can oxlnln. Infrared
neither a double
latter is further confirmed s, both the hydroxyl and the in the bromination reaction.
Quantitatively be studies indicate that the bond nor a hydroxyl group. by a Zerevitinov double bond must This also lends
assumption that the sent as hydroxyl.
sixth oxygen is linked as
by monobromo The
deter- have some an ether
Conroy arrives at a skeletal structure (i); he deduces a com- plete structure (il) which is to be substantiated in later papers.
COO
COO
II
-?-
Several reactions of picro toxin In and its derivatives are em- ployed in assigning structure I. Although structures can be written "for most of the comoounds involved in these reactions, proof of structure has been published only through picrotovinide (III), which is related to a oicrotovinin derivative and to picrotoxadiene (VIII) through the following transformations.
D ihy dr o-cc- p i cr o- toxininic Acid
27^00
C03 + H30
III
H.
Pt03 CH3OH
III Plcrotoxinide
IV
Dihydrooicro- toxinide
IV
HSCHaCHp.SH
HC1 CHCI3
Nl(R)
^>
CK3CHsOH R.T.
V
VI
Tetrahydrodesoxy- picroto^inide
-3-
0 il
Pyr.
VII
^
+ j6C03H + C03
Synthesis of Plorotoxadlene NCCHaCOaEt
Et02C
-^
NHaOAc,HOAc
•H/ ^^
EtOgC
H.
Pd-C
EtOH
Mg
BrCH2C03Et jk-H
I
EtO.
r^
Pto3
HOAc
Ey NC03Et
CN'UNC03Et
HoO
HOI
10 hrs.
■>
EtCgC
EtOgC
C10C
/
1 equiv,
NaOH
EtOH
NO J 'C03Et
n
HO,C
H03C
Et03C EtOgC
HCl
Ha0
W
W
E£OH
(cont. on next page)
NaH
CH2(C03Et)s
)6-H — >
EtOgC
0 (Et02C)3CHC
HaO
H
.+
■>
GH3C
NaH tr. EtOH
;toH
>
OEt
OEt
Cold,
Dil. Acid
KaCHLi
T
Mixture
u
^,4-DNPH Chroma to^raphy on Alumina
NNHAr
Other Workers
Slater, by mepns of a conductlometric titration study , pro- vides additional evidence for a dilpctone structure of nicrotoxinln. Earlier, on the basis of infrared studies, he opposed such a structure, but later he reversed his st^nd as a result of further infrared studies.
On the basis of the failure of r>lcrotoxinin to exhibit behavior of an ethylene oxide he also questions the ether linkage in structure II.
BIBLIOGRAPHY
1. H. Conroy, J. Am. Chen. Soc, 74. 491 (l95S)«
2. K. Conroy, ibid., 74, ?046 (1952J.
3. H. Conroy, Ibid., 73, 1889 (l95l). ) Sources for
4. H. Conroy, Ibid. t 7j5, 1889 (l95l). ) other reference
5. S. N. Slater et al., J. Chem. Soc, 1952. 1042.
6. S. N. Slater, ibid. . 1949. 806.
PINACOL-PINACOLONE REARRANGEMENTS Reported by Ruth J. Adams November 14, 1952
Keeping' in mind the conditions under which carbonium ions may be assumed to be generated, the pinacol rearrangements might orooeriU be thought to belong to a family of pinacol-like rearrangements composed of the following reactions.
OR OH 0
1. R2C— C— R3 scld — ^ R3GCR
OH x C}
2. R30— GR3 Ag ,n R3CCR (X=halogen)
0 + 0
3. R3C— CR3 — S ^ R3CGR
OF NHS 0
4. R3C— 0R3 J&MQ ^ R3GCR
P
5. R3CCHO — pcld N RSC— C— R
H
Consequently, a great deal of what is known concerning one mem- ber of the above series can, with due restriction, be applied to the understanding of another related reation.
Migration Aptitudes,- The oinacol rearrangement was the subject of a group of eroeriments by Bachmann and co-workers, Given a
symmetrical Pinacol, that is, Ri=R3^R3=R4, and basing their con- clusions on the amount of eacb ketone isolated from the reaction
Ott OH J I R i — G — G — R 4
R3 R3
mixture, they found it possible to arrive at a set of values for the migration aptitudes of groups.
Some Migration Aptitudes Found by Bachmann
Migraticr Plnacol Groups Migration % Pinacol _Groups %
P-CH3C6H9 o-Tolyl 94 j P-CH30-CsH4 jAnisyl 98.6
"C-i Phenyl 6 | ^G-Hp^enyl 1.4
* OH \ C«H/
L J.
-6 4 OH;
-J2
(continued on next Page)
Pinacol
Group a
-2-
Migra-
tion % Pinacol
Group s
Migra- tion %
P-Tolyl 57 p-Biphenyl 4.3
1
p~CH3-0C6H4 ; Anisvl 96.7
^■£B-*ol*l 3.3
p_-CH3C6H4 ' qH|
L
-4 2
m-CH3C6H4
C6FS
^C
OH
m-Tolyl Phenyl
66 34
r^-c
CH3C-C«K
L
D~C6HrC6H4
^c
OH1
Anisyl .9 6.r
1 o-Biphenyl" 3.'
Quite recently, McEwen and Mehta3 plotted the log of the mi- gratory aptitudes obtained by Bachmann against Hammet's4' 5 sigma values [log of the ionization constant of the substituted benzoic acid minus the log of the ionization constant of benzoic acid] and have found good correlation. In other Words, the ability of a substituent on the benzene ring to release or withdraw electrons from the ring is independent of the reaction which is being studied or the pinacol of which it is a oart; and it is this characteristic which governs the outcome of the competition in migration of two groups in a symmetrical pinacol. As the table above shows, the Phenyl with the more electron-releasing substituent is usually the one which migrates.
Kinetic vs. Thermodynamic Control.- Prediction of Products of oinacol rearrangements is complicated by the fact that the carbonyl compound formed initially is sometimes unstable in the reaction medium. This easily could be resolved by Preparing the carbonyl com- pounds expected and subjectinp* them to the conditions under which the rearrangement is carrid. out. If both were stable under the conditions imposed, then obviously the experiment had in reality measured the migratory aotitudes. However, if one ketone is converted to the other, then no conclusion could be drawn con- cerning the migratory aptitudes of the two grouos. The work of Danllov and V. Danilova8 is of volue in this connection.
OH
I
C—
OH 1 x
dil.HsS04
i 4
0
■+
/
OH
I
•CHj6
conc.H3SO+ j63CHC-0 conc.H3304
j63CCH0
When the above rearrangement is carried out in conc.H3S04, we have no assurance whatsoever, without further investigation, that the aldehyde is not formed initially and it, in turn, rearranges, due its thermodynamic instability in comparison to that of the ketone in the more drastic conditions.
Competing Oxide Formation.— Some interesting work by Lane and Walters6 has been done on the oinacolio rearrangement of halohydrins
-:*-
the reaction of the bromohydrin t?>es NaOH Is used In r>lace of silver nitrate gnd
On treatment with silver nitrate, ?-bromo-l, 1, 2- trlnhenyletha nol goe-
to the. pln^colone. However,
a different course when
the eooxide is the product. _0n the basis of V/instein1 s7 neighboring
group theory, we see that ~0~ has a greater ability to dlsolace
bromine from the cc-carbon than nhenyl which, in turn, is better thai
-OH.
On the other hand, In a slightly more complex case, steric factors may cause deviations from what one expects from electric considerations only. K. Adams9 has shown that tetraphenylethylene glycol is converted not only directly to the phenyl trityl ketone bu- also to tetraphenylethylene oxide. In this case, the hydroxyl can successfully compete with phenyl. Presumably this complication is caused by the strain arising from the crowding of three phenyls on one carbon as is the case when phenyl migrates. Therefore -OH is in a more favorable situation to compete with phenyl and some epoxide is formed.
Competing Diene Formation.- The synthesis of the estrogen, 3, 4- bis (p_-hydroxy ohenyl J-'*, 4-hexadlene is another remarkable ex- ample of the striking Phenomenon of a change in the products of a reaction due to a change the medium in which it is conducted8.
CH3 CK,
CH:
ArC- i
OH
CH3
■C— I
OH
Ar
/
H
AcCl
Ar-^-
Only in acetylchlorlde does the dehydration-, take place. It has been postulated that the effect of acetyl^is°Suee to initial esteri- fication after which the potent neighboring group, $ dlsDlacer the remaining hydroxyl (as the conluerate acid) CH3C00~
CH3
HO °^e
M j
to give ! . On loss of a oroton,1*" » results
StJ |..Et ' Ar
Ar Ar which then splits off HOAc to give the diene.
V*3
0^ ^0
Another case in which the use of acetyl choride as a solvent alters the products of a reaction^ is illustrated by the work of Lyle and Lyle13.
(97#)
C=0
A V
/-.
-OH
AcCl
ZnCl.
H3_S04 v
(845?)
_4-
The mechanism of this preferential arrangement has not been com- pletely worked out.
in-
Demonstration of Intermediate Ion.— The concept, not of a termediate compound but of an intermediate ion, can clarify some of the experimental results that Criegee10 and Bartlett and Brown11'13 obtained. The rearrangement they observed was that of cis- and trans— 7, 8-diphenylacenaPhthene-7, R-diol. These structures excludec both ring contraction and simple dehydration. In anhydrous acetic acid, the cis- Isomer rearranged to the pinacolone 3-6 times as fast as the trans-. The limiting rates of both isomers in considerable
suggested a common intermediate.
VvThen
water were identical. This
the reaction of the trans-diol wag halted „ it was found to be a mixture of starting material, cts-diol and Pinacolone. The indica** tion is that water reacts with an intermediate ion to go back to cifl-dlol. Direct displacement of a protonated hydroxy 1 from the back by water seems unlikely for steric reasons. Brown has shown that alcohols can adopt the role of water with decreasing effective- ness as the bulk of the alkyl group is increased from CK3 to Et- to i-Pr- to t-Bu.
HO
trans
HO
OH
H90
H
=r
cis
•J
0
A
,~X
r
^ <>
HO
XT0H
S/
HO
OCH3
trans
roo
+
w
Vs
BIBLIOGRAPHY
1. Bachmann and Moser, J. Am. Ohem. Soc, _54, 1124 (193?); Bachrann
and Ferguson, ibid., 56, 2115? (1934). 2* Daniloff and V. Danilova, Ber. 59, 377 (1986). 3. McKwen and Mehta, ibid., 74, 526 (1952).
-5-
4. Hammett, L. Physical Organic Chemistry, p. 198, listed. (1940).
5. Organic Seminar Abstracts, University of Illinois, Dec. 7, 1951
6. Lane and Welters, Ibid., 7£, 4234, 4238 (l95l).
7. Winstein and Gr unpaid, ibid. . 70, 828 (1948) .
8. Lane and Shelter, ibid., 7?, 440P, 4411 ''195l).
9. Adorns, K., Abstracts of Paoers, 122nd Meeting of the American Chemical Society, *>4M (1952).
10. Criegee and Plate, Ber. 72, 178 (1939).
11. Bartlett and Brown, J. Am. Chem. Soc. 68, 2927 (1940).
12. Brown, ibid., 74, 428, 432 (1952).
13. Lyle and Lyle, ibid., 74, 5059 (1952).
■
■ ■ '
FORMAZANS Reported by N. E. Bojars November 14, 1955
Introduction
A special cIpss of the azo compounds consists of the formazans or formazyl derivatives (i).
\ |
||
N = |
=N \ |
|
H |
G- |
-R» |
X- |
-J |
|
/ R» |
6HS |
|
\= |
=rN |
V |
|
H |
0 |
\, |
,/' |
N— |
— N |
/ |
I II
The name "formazyl" was originally introduced ;* for the radical (II). Later the name "formazyl" was proposed for the un- substituted radical (ill).
N=N N=rN
H^ \ / \
C — H CH
H2N N H2N N
III IV
However, in the modern literature still another basis of the nomenclature is employed3'*4'5 whereby the name "formazyl" is al- together discarded. All the compounds of this class are derived from the hypothetic Parent compound formazan (IV).
Preparation of Formazans
The method of preparation, which hp s been most frequently em- ployed, uses as the starting materials aldehyde phenylhydrazones and aromatic diazonium compounds. The attack of the diazonium group occurs upon the aldehyde carbon atom carrying the hydrazone group, with the elimination of a molecule of a halogen acid which reacts with alkali. An example is the reaction (l), producing6 N,N*-di- phenyl-C-methyl-formazan (V).
-2-
H
(l) C6HB N3+ CI + H "C — CHS + NaOH
in
ethanol
>
CsH5
N — N
CgHs
N=K
\
H C — CH, + NsCl + HsO
/
N N
,/
/
V
Several different methods of preparation are known 7"r1-7.
Properties and Reactions
Formazans are colored (mostly red) crystalline compounds. Most of them are stable at room temperature, ^hey have the properties of dyesj some of them could be used to color wool and silk18.
Because of the limited space only a few reactions of formazans can be mentioned here. Concentrated sulfuric acid dissolves the formazans with blue-green or erreen color3' lx* x 3> 1 8> l9 which usually becomes yellow or brown upon standing.
A characteristic reaction of the formazans is the oxidation to the tetrarollum salts3-" 5> 1 3* s0. An examples© is the reaction (2), whereby K. N'-diphenyl-C-methylf ormazan is transformed into 2,3-di-
Phenyl-S^methyltetrazolium chloride .
(2) CSH,
\
N^=*N
H
X
N-
/
Ce"^5
\
CH.
.N
-CH3 + HC1 + CH-CH3-CH2-ONO // CH3"
CeH5 +
\
N = N
N-
CSH5
-N
■CH.
CI
-3-
The N-hydrogen atom of the hydrazone group can be acylated; thi reaction can be reversed under certain conditions13.
N
\
H CH
\ //
N N
C6HS \
07
A
Of?
v © ^
%
v0, /
\
CH
N N
♦-^/xCOCH,
nv
(3)
CH,
r
VI
boiling in ^
/
re so.
! ch3- y
^
^
CH
rtJ
VII
CH.
Tautomerism
In the reaction (3) the formulas VI and Via represent one and the same compound13. Upon boiling- with a little zinc chloride in acetic anhydride, two different compounds are formed in approximate]; equal yields (VII and VIIl).
Generally, compounds IX and IXa are identical21' 33.
R»
\
N=rN
\
H
\
NN N
\
C
N — N
■R"
<r
H
\
■ R"
\
N-
'/
R»
/
R
/
IX
N
IXa
These are the mesomeric limiting states; perhaps the best repre- sentation of the average electronic density is the formulation IXb; there a dotted line represents a "half bond" (average in time).
-4-
R |
||
\ |
||
,Nr: * |
"% |
|
H" |
>c~ |
-R" |
V |
-// |
|
/ |
<-> 1\
Ka " IX
R»
BCb
This formulation basically excludes the existence of isomers re- sulting by the exchange of R and R1 . However, a few workers have announced that they have found isomeric f ormazans1 e> 33~"37. This introduosd a seeming contradiction in the question of the structure of the formazans.
In the most cases no isomers were f«und2 3. This contradict- ion was solved33 only quite recently. It was Droved33 that the isomers, l8> 33""ss supposedly resulting by exchanging R and R1 in the formula I, are Isomers resulting by the nttack of the diazonium group either uoon the aldehyde carbon atom, as expected, or upon the aryl group of the aryl hydrazone Part of the molecule. Ex- amples of such isomers are the compounds X and XI. Only X la a formazau33, while XI is an azo compound Isomeric with it.
<^>
C6H5-CH=N-NH
XI
Thus the structure of formazans invrflvlflg the hydrogen bond seems to be now definitely proved.
!•
-5-
The Red and Yellow Forms of the Formazana
The solutions of formazans usually have a red color In benzene and similar organic solvents. The red solution of the formazan "becomes yellow by the action of the visible light. Recently attemo have been made33'34 to clarify the reason for this reversible changt of the color. Classically -possible are four geometrical Isomers (XII, XI1±, XIV, and XV ) .
/ \ / \ /
— C' — C — C
% \ ^
I cis-cis els- .trans trans-cis
XII XIII XIV
/
C
\
l
trans-trans XV
Since the yellow form is stable only in certain solvents or only under continuous illumination, it is supposed34 that the yellov form is a geometric isomer richer in energy, ^nd, therefore, less stable ther-modynamicalIy<> The question has not been yet decided as to which of the cis-trans isomers is the yellow form. The Problem is complicated by the question of whether the hydrogen bond is broken or not at the yellow r± red transitions.34
Bibliography
1. v. Pechmann, Ber. 25f 3177 (189?,).
2. Bamberger, .ibid. , J35, 3?07 (189?) «
3. v. Pechmann. ibid,, 97. 1683 (1894) .
4. WedekinCL Ibid., 31, 474 (l°98)c
5. Wedeklnd, JL^id., 30, 444, 446 (1897).
6. Bamberger and Pemsel, ibid,, 36, 54, 87 (1903) .
7. v. Pechmann, lblda, ^5T31Q6 Tl89<:)).
8. Claisen, Ann, 987, 368 (1895).
9. Walther, J. Drnkt'. Chem. [J] J53, 4?5 (1896).
10. Dains, Ber. .35, ?50? (190*7 .
11. Bamberger *nd Wheelwright, ibid., 95. 3*04 (1899).
-6-
12* v. Pechmann and Runge, lb Id . . 27, 2927 (1894).
13. v. Pechmann and Runge, ib id . . 87, 1698 (1894).
14. Bamberger and Billeter, Helv. chim. Acta 14, 219 (l93l).
15. Pinner, Ber. 17, 183 (1884).
16. Bamberger, lb id. . , 27, 162 (1894).
17. v. Pechmann, ibid.. 27, 322 (1894).
18. Fichter and Schless, ibid.. 33, 747 (1900) .
19. Fichter and Schless, lbld.t 33. 749 (1900).
20. Wedekind and Stauwe, Ibid., 31, 1756 (1898).
21. v. Pechmann, Ibid., 27, 1682~Tl894) .
22. Lapworth, J. Chem. 3oc. 83, 1119 (1903) .
23. Fichter and Froehlich, Chem. Zetr. 1903 II, 427.
24. Ragno and Oreste, G-azz. chim. ltal. 78, 2?8 (1948).
25. Fichter and Froehlich, Ztschr. Farb. Text. Chem. 2, 251 (1903)
26. Busch and Schmidt, J. Drakt. Chem. [2] 131, 182 Cl93l).
27. Ragno and Bruno, G-azz. chim. ltal. 67, 485 (1946 ).
28. v. Pechmann, Ber. _27, 1679 (1894).
29. v. Pechmann, lb id „ , 2B, 876 (1895) .
30. Marckwald and Wolff, ibid.. 25, 3116 (1892).
31. Kuhn and Jerchel, ibid., 74, 941 (l94l).
32. Hunter and Roberts, J. Chem. Soc. 1941. 820.
33. Hausser, Jerchel, and Kuhn, Ber. J34, 651 (l95l).
34. Hausser, Jerchel, and Kuhn, ib_ld., 82, 515 (1949).
10
DT- AND POLY ACETYLENES Reported by Aldo J. Crovetti, Jr.
November 14, 1953
Polyacetylenes have created Interest in different fields of chemistry. Because of their natural occurrence in essential oils from soecies of composites1, in some Basidiomycetes, and the oossi- bility of correlating light absorption properties with structure, the polyacetylenes have aroused the interest of the biochemist. The theoretical chemist has been Intrigued because of their linear structure and consequent simple geometry. From the standpoint of organic chemistry they have been recognized as potential sources of many compounds.
DIACETYLENES: Diacetylene itself ^as been known for many years, but its use has been limited because of the difficulty of its preparation in quantity. The most useful laboratory method used involves oxidative coupling of monosodio acetylide8 to give di- acetylene (^5^). The most oronising route to diacetylene and higher diacetylene s seems to be from 1, 4-dichloro-2-butyne3 which is now commercially available from the cheap sources, acetylene and formaldehyde .
HC = CH
CH,0
">
HOCHoC—CCKpOH
S0C1
ClCH3C=rCCH3Cl
fir
NpNHjj
liq.NH3
-70°C
2N*4C1
EC^C-C = CH
NpC= CO ECNa (II)
RX
RaCO
RC 3 0 " G S OR (RsAlkyl)
HORaCC = C-C~CCRsOH (R=Alkyl or aryl)
Until recently, contrasted %$ the many symmetrically substitute
ence of methyl, ethyl and vinyl diaeetylene has been detected
the high boiling residues from the Kills acetylene synthesis process8.
By using three molecular prwoortions of sodium amide, the monosodio diacetylide is presumed formed which upon alkylation with an alkyl halide gives rise to monosfckyl diacety'ienes (IV).
-2-
ClCH2C=CCHaCl
3N3NH3 liqNH3
tf
RX
NaC=C-C=CH
RC=C-C=CH
\ l.RCl
~CHO R3C0 |9.NH4Cl
NH401
v
r'oCO
I . ECHO
/ RCSC.C=CCR3OH
HOR3CC=C.C=CH HORCHC=C05CH RC=CC=CCHROH
VIII VI V VII
The compounds (IV; R=Me, Et, Bu, CH^OHCH^) have been made in fair (45^) yields7.
The reaction of carbonyl compounds such as acetaldehyde, butyraldehyde, acetone, benzophenone with the monosodlo compound (III) gives compounds of the type (V) and (VI ) respectively. Similarily, the monoalkyl diacetylene (IV) reacts to give compounds of the type (VII) and (VIIl).
A number of 4-alkyl— 1-lodo diacetylenes have been made by an extention of Vaughn1 s8method, which involves the iodination of a monalkyl acetylene in anhydrous liquid ammonia.
RC=CC=CH + I3 + NH3 IV
RC=CI + NH.I
TRIACETYLENES: Substances containing more than two conjugated acetylenic linkages have been almost unknown until recently. Di- phenyl triacetylene snd the glycol & have been described.
The analogy between the behavior of 1, 4-dichloro-^-butyne and vicinal dih^lides towards sodium amide in liauid ammonia has been extended11 to 1, 6-dichloro-bexa-2, 4-diyne. The analogous reactions have been realized.
NaC=CC=C.Na
II ClCH3C=OC=CCE"3Cl
-* 3 H0CF3C£OC£C.CH30H -X 'J±
?N*NH3 RX
4NaNH3 liqNH3
RC=OC£v>C=CH XII
3 C1CH3C=C.C=CCH3C1
2N114.CI . _ — __. _„_
HC^C.C=C-C=GH
R0=O'0feO*CfeC*R
XI
HOR3C.C=C-C=C.C=CCR3OH X
-.3-
The compounds (X; R=Me,Ph), (XI; R=Me, Et), and (XIII; R=Me) have been prepared.
TETR ACETYLENES: Until recently the only recorded examnle of a conjugated tetracetylenic compound was the highly unstable di- carboxylic described by Baeyer13.
RC=C-C=C'C=C.(^C-R HOR3C*(£C.CeC-C£C'C^C«CR3OH
XIII XIV
The previous method discussed has been found to be unsatisfact ory for the preparation of compounds of tyoe (XIIl) and (XIV) . The coupling action of ootassium ferricyanide and potassium permanganat on "the preformed copper di acetylide has proved unsuccessful^. However, the action of oxygen, in the presence of cuprous and ammonium chloride, cuprous bromide or iodine, on the G-rignard derivative has given good yields (66^) of the crystalline deca-2, 4,6, S-tetra.yne^
GH3C=C-C=CH~>t^gX CH3C=CC=CMgX -i? CH3C=C*C£C- C£C* C^O CH3
EtMgX t
RC=C«C=CH -* "" RC=CC=C.KgX-*2 RC=C 0=0 C=C.Cr=C.R
XV
Analogous methods enabled the compounds (XV; R=Et, Bu) to be made. In this series there is a decline in the melting point as the series is ascended. This has been attributed to an abrupt decrease in symmetry from the rigidly linear molecule (XV; R=Me) to the Z shaped molecule of (XV; R=Bu) . There have been Indications that the latter also possesses the ability of rotation in the solid state13.
The use of the G-rignard derivative to make tetra-ncetylenic alcohols in unsuceessf uL The compounds (XV; RurCHaUH) and (XV; R-=CM9-i.0E) were obtained in 74^ and P,9<f yields respectively by & catalytic oxygenation coupling of the respective monosubstitulE diacetylenlo alcohols.
HIGHER POLYACETYLENES: Attempts have been made to develop routes o: general apolicability for compounds ™ith more than four conjugated acetylene linkages and some progress has been made*4
H0CH3(C=C)nCH30H ClCH3(C£C)nCH3Cl HtflSfi^R CH3 (CsC)nCH3
XVI XVII XVIII XIX
a) The conversion of (XVI; n=l,?, 1*0 into (XVII; n=l,2,3) indicates a possible route to higher ooly-ynes by the scheme:
HOCH3(C=C) CH8OH Jt°Cl3 C1CK3 tC=C)nCH2Cl ilaNHs s K(C=C) +]H
— 77°
-4-
The glycol (XVI; n=3) treated in this manner, followed by extraction with oentane gave a solution which when examined spectroscooically gave evidence for octa-1, 3, 5, 7-tetrayne (XVIII; n=4) at an e stig- ma ted yield of lftf. In a similar way the glycol (XVI; n=4) gave a solution which when examined soectroscoolcally g^ve "bands exoected for deca-1,.^,5,7, 9-oentayne in an estimated 1^' yield14. As (n) Increases the yields in the process XVI — ► XVIX-* XVIII fall decided^ from nearly Quantitative (n=2) through about dOfc (n=3) , and about lgf for (n=4) to about 3f for (n=5) . Modification of reaction conditions seems necessary before this general method can be ex- tended.
b) The complimentary route indicated below thus far has proved fruitful only in the case where PUPh and Me when (n=3).
RCH(OH) (C=C)n_1CH{0H)R-5°Cls RCHCl (C=C)- ^CHClR ilaNHs R(c=C)nR
XX
In the case (XX; n=3, R=Me) a 56^ yield was obtained in contrast to that previously obtained by alkylation (28#),
c) Of potential value is the alternative method of obtaining monosubstituted polyaoetylenic hydrocarbons of Primary- secondary glycols. The latter are oreDared by condensation of a carbonyl compound with a compound of the type XMg(0^0)nC!I2OMgX
HX 1. SOCls
RCHO + MgX(C£C) CH3OMgX-> RCH(OH)((3SO)nCH2OH -+ R(C=C) H
?.NaNH3 n+1
3.NH4C1
In the case of hexaynes a s'hort step orocess must be used e.g«
NaNH, l)EtMgX
OlOH8OfeC-CfeC-CHaCl-^ ■ CK3C=OC£C!=C-H -^ CH3(C=C)6CH3
CHgl 8)la
Diethyl hexa -acetylene has been obtained in this way. Because of the photo- in stability of these comnounds their isolation is diffi- cult. The diohenyl hexayne, however, has been orepared in good yield by a similar route15.
1 . S0C1 CuCl
PhCJfcCCH04X%C=CCH30llgX-* PhCH=C- CH (OH)C£CCH3CH -V 2Ph ( C= C ) 3H ->
3r Br 2.NaNTrT3 * °s
3.NH4Cl"Ph(C£C)6Ph,
FROPilRTIES: The monoalkyl polyacetylenes are unstable compounds tending to volatilize easily and explode. The dimethyl poly- acetylenes are all crystalline solids which decompose
<^/fv
-5-
(except n<4) on heating. As larger R groups are attached, the com- pounds tend to become more easily volitali^ed. They are not dangerously explosive (n<5) . Those members in which n=>3 are very easily decomposed in the presence of light but in the dark they are much more stable. They are usually kept at -70°C, but all can be recrystalli7.ed form warm solvents. The diohenyl polyacetylenes are much more stable than the a.lkyl derivatives.
The glycols are all crystalline solids and appreciably more stable than the dimethyl Poly-ynes. The primary and secondary glycols are very much less stable than the tertiary glycols.
The crystals which a ments, usually a vivid re respect is the dichloride becoming dark blue on exp The process aooe^rs to be octa-3, 5-dlyne-l,8-diol i in solution, above its me liquid, yet a deep red co solid unless light is eye irradiation products are They are amorphous films ible with the fused Daren
re photolabile give intensely colored pig- d or blue. The most sensitive in this
C1CHS(C^G)4CH3C1, which surpasses AgBr, osure to diffuse daylight for 10-?0 seconds
associated with the crystal lattice e.g. s stable to light in liquid state, while
Iting point, or even as a super cooled lor rapidly develops on the surface of the luded15. The general nature of these similar amongst di- and polyacetylenes. insoluble in organic solvents and immisc- t acetylenlc compound.
LIGHT ABSORPTION: In all cases observed the spectra of these com- pounds consists of a medium intensity region and a high intensity region as seen in figure I for the case of dimethyl poly-ynes (n=5,6).
3500 3000 3500 4000
Wave 1 enpth A
Figure I
-6-
The maxima show n spacing of 5,000-2,300 cm." The very high intensity in the case of the dimethyl poly-ynes where n=6, the£max= 445,000 at 2840A, is the second largest extinction coefficient yet
BIBLIOGRAPHY
1. N. A. Sorensen and K. Stavholt, Acta. Chem. Scand., 4, 1567, 157? (1950).
2. H. Schlubach and V. Wolf, Ann., 568, 141 (1950),
3. J. B. Armitaere, E. R. H. Jones, and M. C. Whiting, J. Chem. Soc 44 (1951).
4. Yu. S. Zglkmd and M. A. Aiz.ikovich, J. Gen. Chem. (U.S.S.R.) 9, 961 (1939); 0, A. 33, 38695 (1939). .
5. H. Schlubaok *nd V. Franzen, Ann., J573, 105, 115 (1950).
6. J. W, Copenhaver and M. H. Bigelow, "Acetylene and Carbon Monoxide Chemistry," Reinhold Publishing Company, New York, 1949 p. 3, 121, 302.
7. J. B. Armitage, E. R. H. Jones and M. C. Whiting, J. Chem. Soc, 1993 (1952).
8. T. H. Vaushn and J. A. Neuland, J. Am. Chem. Soc, 55, 2150 (1933 J.
9. H. Schlubaok and V. Franzen, Ann., 572, 116 (l95l).
10. F. Bohlmann, Ange*% Chem., _63, 218~Tl95l); R. Kuhn, ibid. 173
(1951).
11. J. B. Arnitage, 0. L. Cook, E. R. H. Jones, and M. C. Whiting, J. Chem. Soc 2010 (1958).
12. A. Baeyer, Ber., 18, 2272 (1885),
13. J. B. Armitage, E, R. H. Jones, and M. C. Whiting, J. Chem. Soc, 2014 (1952),
14. C. L. Cook, E. R. H. Jones, and M. C. Whiting, ibid., 2883 (1952).
15. E. R. H. Jones, If. C. Whiting, C. L. Cook, and N. Entwlstle, Nature, 168, 900 (l95l).
THENOYLBENZOIC ACIDS AND THI0PHANTHRA<4UIN0NES Reported by J. A. MacDonald November 21, 1952
It would appear possible to prepare from thiophanthraquinones a series of dyes analagous to the anthraquinone dyes. This possibili- ty is responsible for at least a Dart of the interest recently shown in the synthesis of thioohanthraqulnones. The general method em- ployed for the synthesis of these compounds consists of the prepara- tion of 2-(2-thenoyl)-benzoic acids and subsequent ring closure.
2-(2-Thenoyl)~benzolc Acids
2-(2-Thenoyl)-benzoic acid was first Prepared by Stelnkopf1, who employed the reaction between ohthalic anhydride and thioohene in the presence of aluminum chloride:
0
COOH
\
x0
— c
*0
s-
■>
Buu-Hoi and co-workerss, starting from 2-methyl-, 2-chloro-, and 2-bromothiophene, used the Same method for the Preparation of 2-(2- thenoyl)-benzoic acids substituted in the 5 positon of the thiophene ring,
A different method, consisting of the action of 2-thienyl- magnesium iodide on phthalic anhydride, was used by G-oncalves and Brown3:
0 *v /' ^ COOH
XMe:
->
Ns'
1 Lc n
0
They investigated various solvents and temperatures for the reaction, and found that the use of anisole ps solvent, at a temperature below 27° gave the best results. A 90^ yield was obtained. The G-rignard method was also applied to the preparation of S- (S-thenoyl)-benzoio acids with methyl, ethyl, chloro ?>nd bromo substituents in the 5 Position of the thiophene ring.
Working independently, Lee ^nd Weinmayr4 employed the G-rignard method for the preparation of 2- (2-thenoyl)-benzoic acids with haloger atoms or nitro groups substituted in the benzene ring. They found that the action of thienylmagnesium hnlides on unsymmetrically sub- stituted pbthalic anhydrides yielded both of the possible isomeric products. Bro^Tn pn^ co-workers5 also investigated the synthesis of nitro-2-(2-thenoyl)~benzoic acids by the Grlgnard method, but found *
-o-
was
obtained from
at first that only one of the possible isomers
each of the mono nitro phthallo anhydrides. A closer investigation showed, however, that both products were produced in the reaction, and that in the purification by recrystallization from acetic acid one of the products was converted into the other. The conversion could also be effected by the use of concentrated sulfuric acid.
NO-
0
N0;
/V-c
*0 R%X . >
COOH
— Cx
\r~%
Vc-R
3
COOH
V.
Acid /
f;NO
\0 RMgX S '
*0
OOH
Acid
NO.
J*
1?
s X^COOK
Rearrangements of this tyoe have been observed and investigated in S-benzoylbenzoic acids6'7, and it has been suggested that the lactol form of the acid (i) is an intermediate.
0 II
r^\4\
0---H
I ■ II
The structures of the nitro-2~(<Vthenoyl)-benzoic acids were deter- mined by decarboxylation and comparison of the resulting ketones' with the compounds obtained from the reactions of nitrobenzoyl chlorides with benzene,
Weinmayr8 reinvestigated the Friedel- Crafts synthesis of 3-(3-» thenoyl)~benzoic acids, and obtained satisfactory yields when phthallc anhydride was condensed with thloohene or various substituted thio- phenes. The reaction 'was, however, not satisfactory for the conden- sation of thlophene with chloro- or nitroDhthalic anhydrides. In these cases excellent results were obtained by use of the reaction of 2-thienylmagnesium bromide with the anhydrides. The constitutions of the chloro-2-{2~thenoyl)~benzoic acids were determined by relating them %o the nitro-3fg-thenoyl)-benzoic acids through the conversion of the '*er to the former by reduction and use of the Sandmeyer reaction.
. ■ ,- . ')
-3-
ThioPhanthraquinone s
Steinkopf1 found that treatment of 2- (2-thenoyl)~benzoic acid with phosphorus pentoxide or concentrated sulfuric acid yielded thiophanthraquinone (II). When sulfuric acid was used there was ob- tained, in addition to this compound, a water soluble acid presumed t be a sulfonic acid derivative of the auinone. When this acid was fused with alkali an orange-red product, possibly the thiophene analo of alizarin, was obtained.
Buu-Hoi3 carried out the cyclization using benzoyl chloride as the dehydrating agent. Brown3'5 and Weinmayr8'9 have prepared many substituted thiophanthraquinone s by ring closure of the corresponding 2-(2-thenoyl)-benzoic acid chlorides, and also by the direct ring closure of the acids through the use of Phosphorus pentoxide, sulfuri acid and aluminum chloride.
From the riner closure of the four isomeric 2-(2-thenoyl)-benzoic acids mono substituted in the benzene ring, four thiophanthraquinone s would be expected. However, in the cases investigated9, only two were obtained, indicating that rearrangement occurred in at least some of the ring closures. In order to establish the identity of the thio- phanthraquinone s obtained, the four monochloro benzene substituted thiophanthraquinone s were prepared by the following method, known to yield unrearranged products in the case of 2-aroylbenzoic acids1^.
CI
^\-C^
V**
CI.
COOH
CrO.
$
Vv^"
OAc
8
The chloro thiophanthraquinone s were used as reference compounds in thf identification of the other thiophanthraquinone s.
When the products of cyclization had been identified, it was apparent that mono substituted thenoylbenzoic acids with nltro or chloro groups meta to the thenoyl group cycllzed normally, while those with either of these groups in an ortho or para position rearranged. Conversely, when an amino group was ortho or para to the thenoyl group cyollzation was normal, while the meta derivatives rearranged:
-4-
A = NO, or CI
NHa °.
0
A -. |l S V^N G *s
I
^V^COOH
X3
It would appear that the products , obtained could be accounted for by rearrangements of the type mentioned previously, in which the thenoyl and carboxyl groups of thenoyFbei&zoic acids are interchanged. How- ever, at least in the case of 3-nltro-?-(2-thenoyl)-benzoic acid, some evidence that the rearrangement involved is a different one was provided by the fact that on treating this acid with 100^ sulfuric acid, a rapid rearrangement to another acid, different from ^nitre- s'-(2~thenoyl)-benzoic acid, occurred. It has been suggested that the rearrangement involves a shift of the nitrophthaloyl radical from the 2- to the 3- pos-ition of the thioohene to yield 3-nitro-2~ (3- thenoyl )- benzoic acid.
Benzthiophanthrones
Thiophanthraquinone can form two benzthiophanthrones (ill and IV, Scholl and Seer11 fused 1- (2-thenoyl)-naohthalene with aluminum chloride, and obtained what they assumed to be 4, 5-benzthiophanthrone (IV)* Recently Weinmayr and co-workers18 repeated this synthesis, and also prepared a different benzthiophanthrone from the reaction of thiophanthraauinone with glycerol «nd iron in sulfuric acid. At firs" they believed this new Product to be 8,9-benzthiophanthrone (ill).
-5-
However, on oxidation it gave a thiophanthraquinone carboxylic acid the infrared snectrum of which was identical with the spectrum shown by thiopb.anthraquinone-5-carboxylic acid, and very different from the spectrum of thiophanthraquinone-8-carboxylic acid. The new benzthio- ohanthrone must therefore be 4,5-benzthiophanthrone, and it must be assumed that a rearrangement occurred during the aluminun chloride fusion of 1- ( 2- the noyljU naphthalene. The rearrangement shown below, similar to the one proposed in the case of 3-nitro-2- (2-thenoyl)- benzoic acid, has been suggested:
N^
Bibliography
1. 2.
3. 4.
5.
6.
7.
8.
9. 10. 11. 12.
W. Steinkopf, Ann. 407, 94 (1914).
Ng. Ph. Buu-Hoi, N. G-. Hoan, and N. G-, D. Xuong, Rec. trav. chim. 69, 1087 (1950).
R. Goncalves and E. V. Brown, J. Org. Chem. 17, 698 (1952 ). H. R. Lee and V. Weinmayr, U.S. Patent 2513573 (0. A. 45,665 (1951).); U.S. Patent 251357? (C, A. 45, 664 (l95l).). R. G-oncalves, M. R. Kegelman, and E. V. Brown, J. Org. Chem. 17, 705 (1952).
M. Hayashi,, J. Chem. Soc. 2516 (1927); 1513, 1520, 1524 (1930); M. Hayashi et al., Bull. Chem. Soc. JaPan JL1, 184 (1936). J. W. Cook, J. Chem. Soc. 1472 (1932). V. Weinmayr, J. Am. Chem. Soc. 74, 4353 (1952). H. E. Schroeder and V. Weinmayr, ibid. 74. 4357 (1952). L. F. Fieser and E. B. Hershberg, ibid. 59, 1028 (1937). R. Scholl and C. Seer, Ann. 594^ 131, 175 (1912). V. Weinmayr, F. S. Palmer, and A. A. Ebert, J. Am. Chem. Soc. 74, 4361 (1945).
NE.I METHODS FOR SPONTANEOUS RESOLUTION OF RACE! II C MODIFICATIONS Reported by Harry J. Neumiller November 21, 1952
I. INTRODUCTION
Spontaneous resolution may perhaps best be defined as ?ny process of resolution in which either no optically active agent is introduced at all, or in which the amount of optically active agent int^* duced is extremely small in comparison with the amount of suostance being resolved, the active agent serving only to initiate the pro- cess. The purpose of this re;oort is to review briefly the earlier known methods of spontaneous resolution, and to discuss in some detail two recently discovered methods.
II. HISTORICAL
1,3
Pasteur discovered in 1S42S, upon microscopic examination of sodium ammonium d5l- tartrate , that this substance consisted of two types of crystals which were non-superposable mirror images of each other. Upon separation of these two types of crystals by mechanical means, he obtained sodium ammonium (+)- and (-) -tartrates . In order to be resolved by this method, a racemic modification must
A slightly more general procedure involves selective crystalli- zation of one enantiomorph from a supersaturated solution of a race- mic modification. Host of these Drocedures involve inoculation of the solution with a crystal of the enantiomorph to be crystallized. However, sodium ammonium (+) -tartrate ha,s been precipitated from a. solution of the enantiomorph by the addition of a crystal of (-)- asparagine,4 the crystals of these tiro compounds being isomorphous. Resolution of atropine sulfa.te has been caused by microsco-oic crys- tals which were present in the atmosphere.5
in. ne;; letmcds
A. Tri thyme-tide
Tri-o-thymotide (l), a cyclic tries ter, has been shown to exist, due to hinderance between adjacent carbonyl and i-propyl groups, in a non-planar, "propellor"=lil:e configuration,6*7 in which the aromatic rings p.re a.rranged on the sides of a trigonal pyramid (Fig. l) . This gives rise to two enantiomorphous forms, which can be resolved by inocula.ting a solution of tri-o_-thymotide with a crystal of the desired enantiomorph. However, since trithymoticle possesses the property of crystallizing from a wide variety of solvents to give molecular complexes6 of the form 2 C33Ii36Os -M (M=X molecule of solvent), selective crystallization yields not the pure enantiomorph, but a complex of enantiomorph and solvent.8
-2-
/\oi
^3
(OH3)2HC^ A <°
\
f^ GTI(OH3);
0Ns /° °
<* /
*>
i
HaC
/ \Y u CW
II CE3 0
v //Orl(CH3)
w s Carbon x = Oxygen y r -CH(CH3)3
B~
•GH
3 3
hese complexes appear to be clathrate oom pounds, Clathrate are formed by the inclusion of one substance in cavities
which, due
cc a more a form in
compounds
existing in the crystals of a second substanc favorable potential energy value, does not crystallize in which the molecules are packed in the closest possible manner. It might be expected that if an enatiomorphous substance were to crys- tallize in this way, the cavities would be dissymmetric, and that as a result one ant ip ode of a solvent which was itself a. racemic modification would bs included in the cavities more readily than the other. Tri thy mo tide has been shown to be capable of resolving a solvent in this manner, a partial resolution o? 2-bromobutane having been attained in preliminary experiments.
B . Urea, Inclusion Compounds
It was discovered in 19^-0 that urea forms crystalline addition compounds with a large variety of straight-chain, saturated aliphatic hydrocarbons and their derivatives.10*11 The combining ratios of moles of urea to moles of hydrocarbon in these compounds are not, in general, quotients of small intergers.12 Recent investigation of the structure of the crystal lattice of the compounds has shown that the centers of the oxygen atoms of the urea molecules lie in the edges of regular hexagona.l prisms, arranged, in honey-comb fashior with the hydrocarbon chains situated a.long the axes of the prisms. The planes determined by the carbon and nitrogen atoms of the urea.
12 . 13
This
molecules lie In the faces of the prisms (Fig. 2).12>
arrangement allows no centers of symmetry to exist in the crystals,
and, as with crystalline quartz, the crysta.ls have screw-axes as
-3-
their elements of symmetry. The formation of a right- or left- handed lattice is thereby allowed, giving rise to optical activity in the crystal. It has been shown to be possible, by selective crystallization, to ca.use only one of these lattices to form.14
If an inclusion compound were made from urea and a. racemic modification, and if crystals of only one sense with respect to the screw-axis formed, the result would be two diastereomers. It would be expected that one of these would be more soluble than the other, and that a. partial crystallization process would give more of the less soluble isomer. Decomposition of the crystals by redis solving
then would give an optically active solution. In preliminary
F'VX
/ (Explanation on following page)
- (Axis of cell and approximate position (of hydrocarbon chain.
•^ (Urea molecule [positions of atoms pre: / (1-oxygen, 2-carbon, 3,3T-NH2], Molecules
1 o— -/2 = ^n ^i0*1 balls representing oxygen are "
(completely in black, together with (tion of hydrocarbon lying along ind
?or- 31 \ oxuii ui nyarocaroon xymg along ctica- (ted axis, comprise a unit cell^
Fig. 2. (Explanation)
BIBLIOGRAPHY
1. L. Pasteur, Ann. ohim. et phys. [3], 24, 442 (1343); 22, 56 (IS50;
2. L. Pasteur, "Researches on the Molecular Asymnp try of Natural Organic Products," Alembic Club Reprint Up, lH-9 University of Chicago Press, Chicago, 1902.
3. F. Ebel, in K. Freudenberg, "Stereochemie," Franz Deutlicke, Leipzig rnd Vienna, 1933, p. ^>6k.
4. I. Ostronisslensky, Ber~. 4l, 3035 (I9OCI).
5. L. Anderson and D. J. Hill, J. Chen. Soc., 192 3, 993.
6. './, Baker, B. Gilbert, and "i. D. Ollis, J. Chen. Soc . , 1952, 144-3.
7. P. G. Edgerley and L. 2. Sutton, J. Chen. Soc., 1951, 1069. 3. H. M. Powell, Nature, lJO, 155 (I952). *~
9. H. M. Powell, J. Chem. Soc, 1943, 6l.
10. F. Ben-en end W. Schlenk Jr., Ex^erientia R, 200 ' 1949) .
11. F. Ben-en, Angew. Chem., 63, 207 (1951).
12. \J. Schlenk Jr., Ann. 565, 204 (1949).
13. A. E. Smith, J. Chem. Phys. 13, 150 (1950) .
14. ./. Schlenk Jr., Experientia BJ 337 (1952).
ADDITIONAL REFERENCES Olathr ace Compounds :
15. D. E. Palin and H. M. Powell, J. Chem. Soc., 19*1-7 , 203; 1942,
16. II. II. Powell, J. Chem. Soc., 1950, 293, 300, 463.
General Review of Organic Inclusion Comuounds :
17. '•/. Schlenk Jr., Fortschr. chem. Forsch. 2, 92 (I95I) .
THE REACTIONS OF HALOGEN (i) SALTS OF CARBOXYLIC ACIDS Reported "by G-eorsre W. Parshall November 21, 1952
I. Preparation and Nature
The reaction of a metallic salt of a carboxylic acid with iodine, brom3ne or chlorine leads to the formation of the corresponding halogen (l) salt of the acid1. For example, the silver salt, which is usually the most convenient for preoarative purposes3, reacts wit!' an equimolar quantity of iodine as in step A. However, if less tha:.1 a.i equiraolar quantity of iodine is used, the excess silver salt form* a complex with the iodine salt as in step B3.
(A) RC03Ag + I3 ^ RC03I + Agl
(B) RC03Ag + RC03I > (RC03)3AgI
Although both the halogen salt and the complex are very sens- itive to heat and moisture, Prevost has reported the isolation of th<r ^cv^lex resulting from the reaction of equivalent auantities of i^.'i'j-.? s rA silver benzoate4, Blrckenbach and his co-workers have cif-f:: ;.;i i,Vr.;;r.d the existence of iodine (l) acetate by treating cyclo- hexsn? with a. silver- -free solution of this salt and isolating the I-C.C*.". :';v>:.r.y-- ^-iodocy -lohexane which was formed-3,
i'he positive nature of the halogen in these salts was demon- strate, by BoekemrJ. ler arid Hoffmann who found that a silver-free solution of bromine (i) butyrate has an oxidizing power eaual to that of a bromine solution containing twice as much of the halogen3.
II. Halogenation of Aromatic Compounds
The iodine salts of strong carboxylic acids, such as trifluoro- acetic aoid- apparently dissociate to some extent to give carboxylate and ioGonium ions., Since the latter are electrorhilic, these salts may act as halogenating agents for aromatic compounds. Iodine tri- fluoroacet^te has be en found to sive sood yields of mono- iodina ted products with a wide variety of aromatic compounds, Compounds con- taining electron-donating groups are iodinated almost exclusively at the para position., a fact which seems to confirm that this reaction proceeds by way of an ionic mechanism6.
Although iodine trifluoroacetate has received the most intensive study, iodine acetate, bromine acetate and bromine trifluoroacetate have been shown to react similarly7* 8. In the bromination of toluene vita the latter reagent, the product may be either j>bromotoluene or b^-v.;y;;. bromide depending on the temperature at which the reaction is curled out".
-8-
aC
Many examples of self-halogenation of bromine salts of Ids have been reoorted7' gt 10.
aromatic
til. Addition to Olefinic Double Bonds
These halogen salts add to olefinic double bonds to form esters Of ttnhalo alcohols as shown in the example below. This reaction pro- ceeds so readily that it has been proposed as a means of preparing solid derivatives of simple olefins11. The orientation of the sub- stituenta is such as to suggest an ionic mechanism in which the attack is initiated by a positive halogen ion.
'U
CHg=CH— CHg— CHg
0aN
COp-CIr^-CHo-CH.
V
0aN
\An
NO,
This reaction has been applied to a large number of ethylenic compounds3* &3 l3' ls and Prevost has reported similar additions to acety.Lenic compounds4 B The addition of iodine benzoate to butadiene was found to proceed in a 1,2 manner to give a product which could readily be hydrolyzed to 3, 4-dihydroxy-l-butene14.
CH^CH-CKr.-GHs + C6HsC0yI
CeHBC03CH-CH=C!Hs CH3I
I
CH3-CH-CH=<"m3 I { CH OH
IV*. Decarboxylation
The best known reaction involving the halogen salts of carboxylJ lc acids is the Hunsdieeker decarboxylation. In general this re-* action is carried out by heating the silver salt of the acid with an equimolar quantity of bromine. As is shown in the case of cycle-* butanecarboxylic acid, the carboxyl grouo is replaced by a bromine atom with the attendant formation of carbon dioxide and silver bro- mide15. However, if only an equivalent quantity of the halogen is used, the product is an ester as is shown with caproic acidls>2?t
C03Ag
Br
m "« i ii
■»
+ C03 + AgBr
Q
2C6HxlC03Ag + I3 ^ C5HxxC03C5Hxx + C03 + Agl
K ' V
-3-
The mechanism of this reaction has been the subject of much controversy, especially with regard to the decarboxylation of aro- matic acids. Although it is generally agreed that the halogen salts are intermediates, strong evidence has been presented for both the ionic and free radical mechanisms7'9.
This decarboxylation has been reported to be stereochemical^ specific in the cases of (+) 2-ohenylproplonlc acid, (+) 2-benzyl- butyric acid and (+) 2-ethylhexanolc acid1 7> l B> 3s . Unfortunately, the oroducts are usually racemized by the silver bromide which is formed in the reaction mixture.
^-C08Ag (CH3)n
l
COpCH.
Br
Br:
(GH2)n
\
•7
Br-(CH2)n C03H
!— • COoCK.
One of the most common uses of this reaction is for the pre- paration of O'-bromo acids from the mono-esters of dicarboxylic acids19'30. It has also proven very useful for introducing chlorine, bromine or iodine into perfluoroalkyl compounds31** 25.
1. 2.
3. 4, 5.
6. 7.
8.
9.
10.
11.
12. 13.
14. 15. 16. 17. 18. 19. 20.
21. 22.
23. 24. 25. 26. 27.
R. W. A. R. K. B. J. C. D. C. J.
165 (1935).
Bibliography
J. Kleinberg, Chem. Rev., _40, 381 (1947).
R. N. Haszeldlne, J. Chem. Soc, 584 (l95l).
W» Bockemiiller and F. W. Hoffmann, Ann., 519,
C. Prevost, Compt. rend., 196. 1129 (1933).
L. Birckenbach, J. G-oubeau and E. Berninger, Ber., .65, 1339 (1932)
N. Haszeldlne and A. G, Sharpe, J. Chem. Soc, 993 (1952).
G-. Dauben and H. Tilles, J. Am. Chem. Soc, 72, 3185 (1950).
L. Henne and ¥. F. Zimmer, Ibid., 73, 1362 (l95l).
A, Barnes and R. J. Prochaska, .ibid., 72, 3188 (1950).
Birnbnum and H. Reinherz, Ber., 15, 456 (1882.).
I. Halpe rin, H. B. Donahoe, J. Klelnbers and C. A. VanderWerf,
Org. Chem., 17, 623 (195?).
Prevost, Compt. rend., 197. 1661 (1934).
C. Abbott and C. L. Arcus, J. Chem. 80c, 1515 (1952).
Prevost and R. Lutz, Comot. rend., 198, 2264 (1934).
Cason and R. L. Way, J. Org. Chem., 14, 31 (1949).
A. Simoninl, Monatsh., 'l3, 320 (1P92).
D. C. Abbott and C. L. Arcus, J. Chem. Soc, 3195 (1952).
L. Arcus, A. Campbell and J. Kenyon, ibid. . 1510 (1949).
Lyttrlnghaus and D. Schade, Ber., 74B, 1565 (l94l).
riunsdiecker and C, Hunsdiecker, Ber., 75B 291 (1942).
N. Haszeldlne, J. Chem. Soc, 3490 (19527.
C A. H.
R.
M. HauPtschein
(1951).
M. M. M. F.
and A. V. Grosse, J. Am. Chem. Soc, 73, 2461
al. , al. , el. ,
ibid.
ibid.
24,
2±,
J.
Hauptschein, et
Hauptscheln, ejb
Hauptschein, et al.'. ib id . \ 74",
Bell and I. F. B. Smyth, J. Chem. Soc, 2372 (1949).
W. H. Oldham, ibid.. 100 (1950).
848 (1952). 1347 (1952).
849 (1952).
THE REACTION OF ct-HALOKETONES WITH DINITROPHENYLHYDRAZINE Reported by Fabian T. Fang December 5, 1952
Introduction
ct-Haloke tones in general present some interesting features in their behavior toward the usu^l carbonyl reagents. Hantzsch and Wild1 reported in 1896 that compounds of the type Ri-CHX-CO-R3 formed osa- zones and 1,2-dioximes with l-j moles of phenylhydrazine and hydroxy- lamine, respectively. Curtin pnd Tristram3 furnished evidence in favo of a tetrahydropyridazine structure for the product of the reaction be* tween cc-haloacetophenones and Phenylhydrazine. With hydroxylamine, o-bromoacetoPhenone is reported to form the dioxime of phenylglyoxal3. An unsuccessful attempt to prepare the carboxyphenylhydrazone of 2- chlorocyclohexanone has been recorded4 and this is in agreement with the isolation of a dinitrophenylosazone5 and of a l,2r-dioxlme6 on treatment of the same ketone with the corresponding carbonyl reagent.
In 1948, Mattox and Kendall7 recorded in a preliminary communis cation the interesting observation that when certain hormone inter- mediates containing the 3-ketc-4-bromo grouping (i) were treated in acetic acid solution with 1.2 moles of dinitrophenylhydrazine in the absence of molecular oxygen with or without the addition of sodium acetate, the dinitrophenylhydrazone (il) of the corresponding^4— 3- ketosteroid was obtained in excellent yield. Furthermore, on cleavage with pyruvic acid in the presence of hydrogen bromide, the unsaturated ketone (ill) could be regenerated almost quantitatively.
rt
rt
RNHN-
Br I
/,
II
^
/N-
0 ^1
V
:n
R = 2,4-(N0B)3C6K3-
Further studies8'9'10 have been Prompted on the course of this and useful dehydrohalogenation which is sometimes known as the Kendall reaction.
smooth Matt ox-
Scope and Limitation
Djerassi8 extended the reactio He found that the reaction of the s moles of dinitrophenylhydrazine was to five minutes. In addition to th apolicable to the dehydrobrominatio 3-ketqallosterolds (v) as well as 2 ketosteroids (VIII). The last two of the £^s-dien-3-one. The result 2, 4-dibromo-3-ketoallo steroids (VI)
n to other 3-ke teroidal bromok completed afte e 4-bromoketone n of 2-brorno— ( :-bromo- (VII ) a compounds both s are not concl
to-a-bromo steroids, etone with 1.0-1.1 r heating for three s (i), the method is IV) and 2,2-dibromo- nd 6-bromo- «cv4-3- yield the hydrazone usive in the case of
-3-
RNHN
->
Er
-*
&**y
<y
\
•N/
■>
o^\A/
RNHN ^\^\^
^V
VIII
The regeneration of the unsaturated carbonyl compounds from the dinitrophenylhydrazones was found to be feasible from e preparative standpoint only in the case of the^CaJ- and^4-3-ketones, thus imposing somewhat of a limitation on this reaction.
Phenylhydrazine, a-( ?,4-dinitroPhenyl)-a-methylhydrazine, hydroxy! amine and semicarbazide produced essentially the same results as di- nitrophenylhydrazine . Dinitrophenylhydrazine surpasses all other rea- gents because of the ease with T>Thich the dinitrophenylhydrazones crystallize in the steroid series.
Ramirez and Kirby10 investigated this dehydrohalogenation pro- cedure with simple cc-haloke tones of varied structures other than steroids. The a-halo din itroDhenylhydra zones of the following ketones were prepared in good yields by means of an aqueous methanolic solutioi of 2,4-dinitrophenylhycirazine sulfate containing excess sulfuric acid (Brady's reaerent*0.
-3-
CK3
.CH3
Y c
Br
IX
X
XI
XII
XIII
In general these hydrazones Proved to be quite stable when pure or in solutions of non- hydroxy lie solvents. When solutions of the ct- halo hydrazones of IX X or XI in acetic acid were kept at their boil-
ing points for five minutes, 3mooth dehydrohalogenation took place Lth formation of the corresponding cc, £~ unsaturated dinitrophenyl-
h;vdrazones. eolations of
The same results were the a-h^loke cones IX, 2,4-dinitroohenylhydrazlne; the rated hydrazones. The elimination
obtained on similar treatment of X or XI in acetic acid with one mole products isolated were the unsatu- of hydrogen halide appeared to
proceed slowly if at all at room temperature.
In the two cases (XII nnd XIII) in which an aromatic ring was conjugated to the dinitrophenylhydrazone group, no hydrogen bromide wac eliminated under conditions comparable to or even more drastic than those described above. As shown by the behavior of XI, the degree of substitution on a position adjacent to the hydrazone group seems to play no important role in the elimination.
The lability of the cc-halogen atoms in the hydrazones of IX, X and XI is apparent in their behavior toward methanol. In this solvent formation of the corresponding a-methoxy hydrazone was essentially complete on ^arming for a few minutes. The same treatment applied to 2-bromo-l-tetralone (XII) also led to an a-methoxy hydraxone. The a-halo hydrazones and the a~methoxy hydrazones obtained from IX, X and XI were all converted into the corresponding osazones by an excess of Brady's reagent.
Mechanism
The most probable mechanism of the reaction of a-haloke tones with dinltrophenylhydrazine is that suggested by Mattox and Kendall9.
-4~
0'
RNHNH,
Br I
xrv
-Br®
3>
R-N=N
H
">
FUIWCfc H
R-rN=\^
H
II
CH-, CH
XV
R-N-N= H
OCR.
XVI
The cc-bromo hydrazone XIV is initially formed. The loss of a bromide ion by solvolysis leads to the resonance-stabilized ion XV which then reacts either through loss of a proton and formation of a double bond to give the unsaturated hydrazone II or through the addi- tion of a negative grouo to give the substituent XVI.
The observations of Ramirez and Kirby10 are consistent *rith the view that in this reaction the formation of the a-halo hydrazone pre- cedes the dehydrohalogenation step.
BI3LI0GRAPHY
1. 2. 3. 4. 5. 6.
7. 8.
9. 10.
11.
Hantzsch *nd Wild, Ann., ?89, 285 (1896).
Curtin End Tristram, J. Am. Chem. Soc, 72, 5238 (1950).
Scholl and Matthaioooulos, Ber., 29, 1550 (1896).
Jenkins, J. Am. Pharm. Assoc, _32, 83 (1943).
Murphy and
Loftfield,
Tokura and
(1943).
Mattox and Kendall,
Djerassi, Ibid., 72
Mattox and
J. Am. Chem. ' 3oc Oda, Bull. Inst.
Phy s .
Chem. Soc
4707 (1951).
Chem. Research (Tokyo),
850
J. Am. ., 1003 (1949). Kendall,* ibid.. 72, 2290 (i960) Ramirez and Kirby, ibid., 74, 4331 (1952). Brady, J. Chem. Soc, 757*Tl9ol).
70, 882 (1948).
LANOSTADIENOL
Reported by David M. Locke
December 5, 1952
Lanostadienol Is one of a group of tetracyclic trlterpenee known as "isochole sterol" found in wool fat1 and In the mother liquor from the preparation of ergosterol from yeast3. Its structure has been of con- siderable interest since it exhibits reactions characteristic of both sterols and amyrins. The molecular formula is found to be C30H500; the compound contains two double bonds, one readily hydrogenated and one resistant to hydrogenation, four rings, and a secondary hydroxy 1 group1' s«
The most available looint of attack for degradation studies is the reactive double bond. By ozonolysis and osmium tetroxide-hydrogen peroxide this double bond may be placed in the -CH=C(CH3)2 moiety3'4.
The secondary hydroxy! group also provides a point of ready attack. Phosphorus pentachloride leads to a rearrangement exactly analagous to that in the pentacyclic triterpene series5.
JL
PCI
l)QgQ, 2)Pb(OAc)
>
acetone
Dehydrogenation of lanostadienol or of lanostene (side-chain hydrogenated and hydroxyl groun reduced) with selenium yields 1,2,8-tri- methylohenanthrene6' 7> a> . Thus structure I is indicated as a partial formula for lanostadienol.
HO \S
CH9-
CH,
/
V^CRs-j
-CH = C
^N
CH,
/
-CH = S
CH,
v
VNffl.-
CHg
II
In addition an angular methyl group might now be tentatively placec at the A:B ring juncture since one appears in this position in the othe: di- and trlterpenes (structure II).
Infrared evidence suggests that the nuclear double bond Is tetra- substituted, and the following seouence of reactions suggests that it occurs at a rlne luncture^.
2-
lanostenyl acetate Cr0 acetoxylanostendione
80o
S
cT
'G = C
V
(side-chain hydrogenated)
8
0
J
5eOqj
C = 0 G P = *
10 :
3e0
i
I
C = 0
c = c
acetic
anhydride / \
G =r
dioxane 180°
It may he shown that the o^rticular ring juncture at which the double bond occurs is the B:C ring juncture7' 10.
lanoste&dione
«n
CK- HOAc
lanostandione
GH-T
Ah?
CH3-
In each case the extension of the conjugated system by the in- troduction of the additional double bond in ring A indicates that the original point of un saturation did indeed lie at the B:C ring Juncture,
It should be noted that there is no extension of the conjugated system to the D ring. Furthermore, there is no evidence for any un- saturation extending to the carbons at the C:D ring juncture11'13*7. This evidence together with the dehydrogenatlon to l,°,8-trlmethyl- Dhenanthrene suggests two angular methyl groups at this ring juncture. The partial formula may now be represented by III.
Ill
CHr
CHr
CH.
- CH = CH
N3H.
• > n , ■
-3-
Oxidation of lanostenyl acetate (and of dike tola no stenyl acetate) yields a small amount of acetone and 6-me thy lhepta none— ?, identified as the 2,4-dlnitrophenylhydrazones and semioarbazones13' 14. This in- dicates that the side chain may be represented by the Dartial formula IV.
IV
^CH3 J3H3
CK CH = (T
CHgGHg CH3
A modified Barbler-Wieland degradation of the side chajln has also been 'iacfcomplished, indicating the same partial formula8'15 18.
Direct evidence for the size of ring D has been obtained from the tetracyclic degradation oroduct from the Barbier-Wieland series1 e'19.
\
C - CH
.CH,
"GHp -* GH>
.CH = C
CH.
OH,
'C - c
CH.
:c = o
Infrared indicates a ketone function in a f ive-membered ring and thus suggests formula V for lanostadionol.
The only remaining uncertainty is the r>oint of attachment of the side chain to the D ring. Two recent papers which discus? this point have appeared.
D. H. R. Barton and coworkers80 have obtained a ring D ketone (Vl) to which they have assigned the 15-keto structure rather than the 16-keto (not considering the 17-keto because it violates the isoprene rule) by comparing it with t^o other ketones, 3p— acetoxyandrostanone ;(VIl) and A-norcholestanone (VIIl)«
AGO
ACQ
0^
VII
1 7
VIII
! !
- .*»
'-. * '.
%% :
< :
<v
-4-
VI shoved significant steric hindrance being 1000 times slower than VII or VIII in reaction with °,4~dinitrooholy3hydrazine. In quantitative bromination VI and VII took uv ca. 2 moles of bromine whILf VIII took wo more than 3 under the same conditions. The intensity of the infrared absorption band at 1410 cm."1 (indicating methylene alpha to a carbonyl in a five-membered ring) was twice as large for VIII as for the other two ketones. This evidence indicates that of the struct- ures 15-keto or 16-keto the former is far more likely for VI.
Swiss workers31 have indicated in an "addendum in proof" to a re- cent article that they have carried out the following series of reactions:
Side-chain AcO degradation — > product
o*N/t
Xss
sA
HOOC
OAc
The decarboxylation attending the oxidation Indicates that a 0-ketc acid is produced, and this could occur only if the side chain had been attached at carbon 17. This work, however, can not be properly eval- uated until the experimental details are published.
Bibliography
1. 2. 3. 4, 5.
6.
7.
8. 9.
Jfo. 11.
12,
13.
|14. 15.
A. Windaus and R. Tschesche, Z, Physiol. Chenu, 190, 51 (1930). H. Wieland, H. Posedach, and A. Ballauf, Ann., 529. 68 (1937). H. Wieland and W. Benend, Z. Physiol. Chem., 274, 215 (1942). H. Wieland and E, Joost, Ann., 546, 103 (1941) . L. Ruzlcka, M. Montavon, and 0. Jeger, Helv* Chim. Acta, j31, 818
(1948).
H. Sehulze, Z. physlol. Chem., 238, 35 (1936).
D. H. R. Barton, J. Fawcett, and B. R. Thomas, J. Chem. Soc, 3147
(1951).
W. Voser, M„ Mijovic, 0. Jege r, and L. Ruzlcka, Helv. Chim. Acta,
34, 1585 (1951). „
W. Voser, M. Montavon, Hs. Grunt hard, 0. Jeger, and L> Ruzlcka, lblc
33, 1893 (i960).
J. Cavalla, J. MeGhie, and M. Pradhan, J. Chem. Soc, 3142 (l95l).
R. Marker, E„ Wlttle, and L. Mlxon, J. Am. Chem. Soc, 59, 1368
(1937).
L. Ruzlcka, Ed. Rey, and A. Muhr, Helv. Chim. Acta, 27, 472 (1944).
C. Barnes, D. Barton, J. Fawcett, S. Knight, J. McGhie, M. Pradhan,
and B, Thomas, Chem. and Ind., 1067 (l95l).
C, Barnes, D. Barton, J. Fawcett, and B. Thomas, J. Chem. Soc,
2339 (1952).
J. McOhie, M. Pradhan, J. Cavalla, and S. Knight. Chem. and Ind.,
1165 (1951).
. *>.
i. I I
'ft
• 1 r
..- ... ,J.
" .' ■♦ 11
3 to"?* .: ..-!
-5-
16. R. Curtis and H. Silberman, J. Chem. Soc, 1187 (1952).
17. W. Voser, 0. Jeger, and L. Ruzlcka, Helv. Chim. Acta, 35, 497 (1952).
18. W. Voser, 0. Jeger, and L, Ruzlcka, lb Id . . 55, 503 (1952).
19. W. Voser, Hs. Gunthard, 0. Je^er, and L. Ruzlcka, Ibid. 55. SQt
(1952).
20. C. Barnes, D# Barton, A. Cole, J. Favcett, and B. Thomas, Chern. and Ind., 426 (j.952).
21. W. Voser, Hs. Gunthard, H. Heusser, 0. Jeger, and L. Ruzlcka, Helv Chlm. Acta, 35, 2065 (1952).
RECENT STUDIES IN TIE CHEMISTRY OF INDANTHRONE S Reported by William H. Lowden December 5> 1952
Owing to their utility as vat dyes, the chemistry of the indan- thrones has been the subject of considerable research. The first of these dyes, Indanthrene Blue R, the trivial name for indanthrone, was investigated in 1901.
The synthesis of indanthrone (III, R=R,=K) by heating 2-amino- anthraquinone (I) with potassium hydroxide has initialed considerable controversy concerning the mechanism of the reaction. Although the originally assigned structure has been accepted the psstul^ted mechanism has lonp; since been proved erroneous. ' The formation of the intermediate compound I'll) was disproved and a new mechanism which involved sym-di-2-^antfaraquinonylhydraziiie (IV) as a precursor to indanthrone was postulated. However, it was soon noticed that this interpretation could be vrlid only in acidic media.5 It was also suggested that the enolic form of compound with ur.oth.ir- molecule of i
ough t to c omb i ne
amine (V, R=F:'
indanthrone by eye ligation and 0
proved quite popular and several
itself yielding 2-amino-l • 2 * -dianthraquinonj e nolic form of this adduct could then form n and oxidation. This inline addition orocer
[idation. This imine addition proces
.nvestigatore suggested slight
y^y (III)
0
0
aAa
_2-
modifications . ' ' Two other view-points were also proposed, a quinonoid-ion-radical hypothesis10 and a nuclear hydrogen substitu- tion by the anion of (i).11 Recently a series of papers has ap- 13_1, peared in which this problem has been more thoroughly investigated.
0
aAa
^/yN^
-3-
2-Amino-l:2'--dianthraquinonylamine was prepared and cyclized to indanthrone in acid, neutral and alkaline media.12 Indan throne results when the free amine is heated at 2^0° , or warmed in glacial acetic acid. Boiling pyridine does not cyclize the free amine, although potassium hydroxide in cold pyridine does. The N-metnyl derivatives, however, require an alkaline medium.
Upon reduction of 2-nitro-l:2f-dianthraquinonyIamine (VII) with alkaline dithionite, 2--aminoanthraquinone and 2-aminoanthraquinol result. This observation is interpreted by the ;oostulation of the elimination of the 2-anthraquinonyiemine-suDstituent from the 1- position of the 2-aminoanthraquinol nucleus.
H
0
0.
nVi
pi
c
HN
A^
Ss
Vv-NHa H
(I)
0
H
The instability of the intermediate is evidence in sumort of the
.3.. J jr> l\T T \ _.£» f-,T T> T> t TT\ _ ,- _ _ -S _ -t- ... ~ -T j; _ J J — ' i.1 jr> ~ J- i _
duced form (VI; of (V, R=RT=H) as an intermediate in' the indanthrone. It is evidence for the weakness of the bond linking the secondary nitrogen to the I-position of the aminated nucleus, also suggests that the bond formation is a reversible process.
re-
formation of It
Enolization of (V) would be a plausible mechanism for the ring closure to indanthrone; however, observations with the N-methyl derivative (V, R=Me, R'=H) , which cannot occur in an analogous enolic form, tend to discount this mechanism. Ring closure occurred when this compound was hea.ted with potassium hydroxide in pyridine, yield- ing (III, R=Me, Rf=H) . However, with the dimethyl derivative (V,
-4-
in
It has been concluded that the same line of reasoning is valid the union of tiro molecules of anthraquinene . 12
The substituting ability of 2-aminoanthraquinone is predictable on the basis of the rule that only the anions of the weakest bases are able to replace nuclear hydrogen in aromatic nitro or carbonyl compounds. Evidence in support of this rule is the fact that 2-amin
anthraquinone will condense with nitrobenzene in t. strong base to give 2-p-nitroanilinoanthraquinone,
presence of
A hydrogen atom adjacent to a car-bony!
an amino group without great difficulty fact that 2-aminoanthraquinone position by hydroxy 1 and
anilinium
will undergo
8,9
group can be replaced by .is is illustrated by the substitution at the 1-
ions
It has been suggested that the stability of indanthrone (or som< intermediate) is the reason why other products do not occur in abund- ance from the alkali fusion of 2-aminoanthraciuincne . 13 It has long been recognized that the unuaual stability of indanthrone was due to the four carbonyi groups.2 'iith this stability factor in mind, it if reasonable to assume that the ease of formation of indanthrone may b determined by the ease of formation of fflll) . As a result of the methanol-potassium hydroxide color test, it is quite possible that the methyl derivatives of (V) pass through a di hydro intermediate (IX R=Me) . Powdered sodium hydroxide converted (V, R=H) to (IX, E=HJ , which could be explained by the shift of two protons.
(VIII)
0
(t)-C4K
0 (XII)
It was noticed that flavanthr one was also a product of the alkali fusion of 2-aminoanthraquinone . -wo inadequate mechanisms were postu- lated for the formation of this product.3 *xo The more recent investi- gators have accounted for its formation in a similar manner as the indanthrone derivative.16
1 2,
(V, E=H»=H)
(III, R=Rf=H)
indanthrone
B ibliography
flavanthrone
Bonn. Gr. P. 129,^5,. Feb. 6, 1901. Scholl, Ber. 35, 3^10 (I903J. Scholl. Berbiinger and Mans ri eld,
li
11
Scholl 'and Eberl, lionatsh. , yd Kopetschni. Chen. Zentr., II, g^uu v B arn e 1 1 , "An tbrae e n e and Anthrac u i n on e Maki, J. Soc. Chen. Ind. Japan. Sutrol.. Maki., ibid.^ 37, jKZ (19^4-). TanakaT^JT & '
Ber., KO 320, 169I (1907).
ondon
. 1921. . <?>3 (1929).
Schwenk
heir., ooc Cheni.-Ztg. ,
"^pei, 56 ..192 (1935). 4-p [1928)
Bradley and Rooms on, »T, Chem. Bradlpv and Leete, ibid., 212Q md Leete. TSod:. J
4?
sod. 125^ (1932).
(19^1) . Bradley and Leete' ToTd*. \ 211!? (Wl) * Bradley, Leete and Stephens, ibid., 21RS (I951J . Bradley, Leete and Stephens, ibra. 21&3 (1951/1 • Bradley and Hursten, ibid., 217O"" '(19^1) . Bradley and Nursten, lblc. , 2177 (I951) . Bradley and Nursten, ibid., 3027 (1952).
ACYL Q^N MIGRATIONS
Reported by Howard J. Burke December 1?, 1952
INTRODUCTION
The first recorded 0 to N aoyl migration seems to "be that noted in 1883 during the reduction of o-nltrophenyl benzoate (i).1 Europe- an workers, until recently, have concerned themselves with analogous migrations which occur during the formation of nhenylhydra.zones of oc-acyloxyketones (II) and a—acyloryaldehy&es (ill),2 and durinsr the formation of o-hydrovybenzyl anilides (iv).3
0-C0-C6H5 ^ n ^ OH
VNN08 EtOH V^N^ ^V^NH-CO-CeH
Zn,HCl |
r \f Vo3r5 |
EtOH |
VVN/' |
0 )l R-CK-CH 1 O-CO-R |
fr
6-^5
0
R-CHC-R R-CK-CH //
0-CO-R* Uo-R* 4n>> ™*-m-Cs*s
ii in ln IV
Since, however, these migrations are not reversible by a shift in the pH, we shall not discuss therr. further.
REVERSIBLE MIGRATIONS
Reversible aoyl migrations between 0 and N were noted in the p-aminopropanol series in 1935. 4 The subject was pursued somewhat further until the outbreak of war,5""7 but really useful applications of such migrations were not brought out until later.
In 1947, in the course of a research to determine the relative
configurations of acetyl-ephedrine (v) and acetyl~^-ephedrine (VI),8
it was shown that under acid catalysis (v) and (VI) underwent an N
to 0 acyl migration, the former largely inverting its configuration
at Ci to give 0-acetyl-y>-er>hedrine (VIl) ; while the latter retained
its configuration to ffive the same product.
OH HO CHa 4 — Acp CF3
C6H5-C— C-H S£I^2| c6H5-^-C^H
H N(CH3)Ac *<s\. H NHCRVHC1
V
H CH»s J i a
CgHsC — -C— H
HO N(CH,)Ac Ac 6" NHCH3«HC1
VI
During the same investigation it was shown that in both cases the base-catalyzed 0 to N migration went with retention of configura- tion at a rate dependent upon the pH, being practically instantaneous
at a sufficiently high pH.
In 1949 mechanisms were proposed for the reactions T-rith retent- ion of configuration (R) and with inversion (l), as follows:9
— C j— — — G 3—
R v0 H
1 1
K-bm ft-CRV
R S-0H
WRlf
-9- -c
_Ci— -9 s-
R-' v0 H
H pK
-Cx C3-
Ox
C"P-H
>
R
N-CH3
HOH /
~x* C3-
0 .K-CH3
C rfe^F
ti jn
■C^ — Cg"*
0. <&n-ch-
R^ N0K
kH
/ 1
0 N-CR3
R^ ^0
These mechanisms were supported by work done on aroyl migration in which the aryl group was substituted in the ortho Position.9 It was found that the ortho substltuent markedly increased the amount .0 inversion, independent of the electron-displacement characteristics o the group, but roughly proportional to its size. This can be laid t the group's hindering the approach of the hydroxy 1-0 to the carbonyl- C in the R mechanism much more than the approach of carbonyl-0 to Ci in the I mechanism. Also, the kinetics of migration of N-benzoyl-j*'- eDhedrine (VIII) were of the second order.
About the same time it was shown that diastereoisomerlc amino- alcohols such as N-benzoyl-^-ephedrine (VIII) and N-benzoyl-ephedrlni (IX) could be separated by use of the N to 0 acyl migration; the former giving a water-soluble amine hydrochloride, the latter remain- ing unchanged.10
? ?H9
C0H5— C — (J— H
HO N(CH3)C0-C6H5 VIII
HO 0H3 1 1 C6H5-C — C— H
H N(CH3)C0-C6H5 IX
The reasons for course of reaction of the transition states and R mechanisms. It reacting by R (via A) energetically favorab (via B) the two group conformation. Conver conformation is prese and the unfavored cis
the difference, mentioned earlier, in rate and the two Isomers can be perceived by a look at required for t^em to react according to the I will be seen that in the case of the tp- Isomer the methyl and phenyl groups will be trans, an le situation, while to go by the I mechanism s must be forced into a sterlcally unfavored cjL sely, in the case of the other isomer, the t ran nt when it reacts by the I mechanism (via C), conformation when it reacts by R (via D).9
A
o=ef
B
H
H
/ c=o
0
CH,
D
•. :
It was desirable and hydroxy 1 groups _c with retention, so wo 2-aminocyclohexanols firmed the theory, bu the ring. This objee lopentanol ring, wh.os pure , 2 3 " 3 B The re su 1 the cis acyl migrated
to verify that the configuration with the amin^ is was indeed the one which reacted fastest and rk of this nature was performed on benzoylated of known configuration.11'13 The results con- t not conclusively, due to the flexibility of tion was overcome by going to the 2-aminocyc- e cis and trans isomers can also be obtained ts confirmed the earlier work completely, i.e.
readily and the trans did not.
^LiC<
ieomei' r- fee <t.iM co
times
.'. t:
" SA"
OH IS*
three to four
confirmation came from the work done on the " SA" and "SB" of tonofcomlne., in which the rate of N to 0 acyl migration war by determining the rate at which the amino grouo was 11b- 'bA-1 iscM.or liberated its amino group as fast as the r'SB" isomer.
OHwFo 0H
"OH
w
"SB"
e
steric course of
Acyl migration has been used to check the reactions, 18-*19> and in a confirmation of the configuration of chlor- amphenicol (X) , "u Ootical rotation data had suggested the config- uration to be related to^« ep'hecrine I'VIIl) rather than to eohedrine (l;C), Accordingly, (XI)-, which had been sterically related to (X), was converted to (XIl) and (XIIl), and these two compounds subjected to the action of absolute-alcoholic KCl. Both underwent instant migration with retention to give the 0x-03 diacyls, while their di- asterecisomers gave no rearrangement at all. Therefore (XIl) and (XIII ) -r.rere behaving in the same manner as W-enhedrine, and could be regarded as having the U> conf ierur^tion.
H CHsOH
D-.NOs-0eH4-^.C-H
HO ftH-CO-CHCl3 X
H CH2OH l ' 0 6^5-0- GI-
HO XI
■ H
H CH2OAc
C«HK-C-
J-9-H
HO NHAc XII
n tr
H CHs0Ac
5-9~9-h
HO NHBz XIII
Nor~£/-tropine (XIV ) and nor-troplne (XV) have been differenti- ated by this method,21 as have nor-ecgonine (VI) and nor-t^-ecgonine
(XVII). ss HO H H OH H OH un
HO H
COOH XIV XV "xvi
rn, ~~~ XVII
ihere are other instances of reversible acyl migration known, such as the ethanolamide (XVUl) — amlnoethylester "(XIX) inter- conversion, 32 and the iactone (XX) — lp^tam (XXl) interconversion, 33 and there seems to be no doubt that others will be noted in the future.
-4-
e
EtOH,H
R-CO-NH-CH3CH3OH ~
XVIII OK
R-C0-0CH3CH3NH3 XIX
CH3CH3NH3.HCl m 0 C CH3 OH
xx vor
GH3CK3OH
^e^s""V CH3
O^C GH3 XXI NH
ACYL EXCHANGE: Acyl migration in the 0-N-diacyl-o-aminophenol serier is probably actually acyl exchange. Much early ™ork in the field seemed to give evidence that when, for example, acetyl and benzoyl groups were introduced" into the o— emlnophenol (XXII) molecule in different orders, only one isomer was formed (XXIIl) in both cases, instead of different orders of acylation giving different isomers (XXIII) and {XXIV) as would be exoected. 3n"se
XXII
0— CO— Og H5
NH-C0-CH3 0-CO-CH,
xNH-C0-C6H5
XXIV
XXIII
Although the correct explanation of this phenomenon (i.e. a reversible equilibrium between the isomers, so that recrystalllzation would recover only the predominant one) was suggested early,85 it was discarded for lack of evidence. It was, however, brought out again in 1931, at which time it was shown that different isomers actually were produced, although they could not be obtained pure by re crystall- ization.39 It was not until 1948 that the isomers were separated and identified, and the eauilibrium definitely established as being cata- lyzed by acids such as water and alcohol, bases such as pyridine, and heat.30 The following mechanisms have been proposed for acid34 and base31 catalysis:
Jffl
By analogous stet>s In the -presence of an excess of the acetyl- pyridinium Ion the formation of the di- and tri-acetyl derivatives from (XXV) can be explained; the formation of these compounds having been a stumbling-block for the Previously suggested mechanisms.
Regarding the acid-catalyzed mechanism, it was found that the migration did not occur in mixed diacyl derivatives of o-alkylamino- ohenols, so it ™bq oostulated that the attainment of ohase II pro- bably required simultaneous elimination of a proton. It may, hovevei merely be sterlcally Impossible to get an o-hydroxypbenyl-, t^o carbonyls and an alkyl on one nitrogen- This is supported by the fsc that when one of the acyls Is a sulfonyl no migration is observed, even If a hydrogen is present on the nitrogen.
BIBLIOGRAPHY
1. W. Bottcher, Ber., JL6, 629-34 (1883).
2. X. Auwers, Ann., 365^ 278-90 (1909). et seq.
3. K. Auwers, Ber., 35, 1923-9 (1900).
4. V. Bruckner, Ann., J518, 226-44 (1935).
5* V. Bruckner and A. Kramli, J. orakt. Chem., 143, 287-97 (1935).
6. A. Kramli and V. Bruckner, ibid., 148. 117-25~~Tl937) .
7. E. Vinkler and V. Bruckner, ibid.. 151, 17-24 (1938). .8. L.H. Welsh, J. Am. Chem. Soc, 69, 128-36 (l947).
9. L..H. Welsh, ibid., TL> 3500-6 (1949) .
10. G. Fodor and J. Kls^, Nature, JL63, 287 (1949); J. Org-. Chem., JL4, 337-45 (1949)*
11. G. Fodor and J. Kiss, Nature, JL64, 91? (1949).
12. G. Fodor and J. Kiss, J. Am. Chem. Soc, 22s 3495-7" (1950) .
13. G.E. KcCasland and D.A. Smith, ibid., *72, ?190-5 (1950) .
14. G. Fodor and J. Kiss, Research, 4, 389-3 (lP5l)*
15. G. Fodor and J. Kias, J, Chem. Soc, 1°52. 1589-92.
16. L. Anderson and K.A. Lardy, J. Am. Chem. Soc, 72, 3141-7 (1952).
17. G.E. McC^sland, lb id . . 73, 2295 (l95l).
18. G. Fodor and K. Koczka, <J. Chem. Soc, 1952. 840-4.
19. G. Fodor et al.,. J. Ore:. Chem., 15, 227-32 (1950).
20. G. Fodor et al., Nature, ^67, 6ooTl95l); J. Chem. Soc, 1951 f 1858,
21. G. Fodor and K. Nsdor, Nature, 169, 462-3 (1952K
22. G. Fodor, ibid., JL70r 278-9 (i960.) .
23. A. Einhorn and B. Pfyl, Ann., 311, 34-73 (1900).
24. J.H. Ransom, Am. Chem. J., j?3, 1-50 (1900).
25. J.H. Ransom and R.E. Nelson, J. Am. Chem. Soc, 36, 390-3 (1914).
26. L.C. Raiford et al-, ibid.. 41, 2068-80 (1919); Ibid.. 44, 1792-8 (1922); Ibid. ^ 45, 469-75 (192^); ibid,. 46. 4"0-7, 2246-55, 2305- 18 (l924TT~lbld. . 47, 1111-23, 1454-8 (195^); ibid.. 48, 483-9 (1926); lbld.f bOT 1201-4 (1928); ibid., 56, 1586-90 Tl934) ; J. Org. Chem.., 4, ^07-19 (1939) ; Ibid.. 5, 300-12 (1940); J. Am. Chem, Soc, 65, 2048-51 (1934); J. Org:. Chem., JLO, 419-28 (1945); J. Am. Chem. Soc, 67, 2163-5 (1945).
27. R.E. Nelson et al., J. Am. Chem. Soc, 48, 1677-9, 16P0-3 (1926); ibid.. 49, 3129-31 (1927); 50, 919-23 (192P); Ibid.. 51, 2761-4 (1929); lb id . . 53, 996-1001~7l9.^1 ) .
28. F. Bell, J. Chem. Soc, 1930. 1981-7.
29. F. Bell, ib id r , 1931., 2962-7.
30. A.L. LeRosen and E.D. Smith, J. Am. Cbem. Soc, 70, 2705-9 (1948). ?1. A.L. LeRosen and E.D. Smith, Ibid.. 71, 2«15-18 TI940)..
32. A. P. Phillios and R. Baltzly, ibid. . 69, 900-A (1947).
33. E. Walton and M.F. Greenr J. Chem. Soc, 1945 1 315-19-
34. G.W. Anderson and F. Bell, Ibid. , 1949. 2663-71.
* .
SOME CHROMXC ACID OXIDATIONS Reported by Y. Gust Hendrickson December 1?, 195
Oxidation of Alcohols.- Primary ^nd secondary alcohols, oxidize by chronic? acid in aqueous sulfuric acid, give good yields of normal oxidation products* Since isoorooyl alcohol yields acetone quantitatively at a rate which can easily be measured, ^estheimer1 chose this system for a study of the mechanism. This excellent de- tailed study' revealed the following facts about the reaction.
3CH3CH0HCH3 + SHCrO"; + 8H -* 3CH3C0CH3 + 2Cr + 8H20
1. At constant low pH, using excess alcohol, the reaction is firs- order in isoorooyl alcohol, acid chroma te ion (HCrO"^), and second order in hydrogen ion concentrations; the rate expression being3
-d(Cr03)/dt = k(CH3CHOKCH3)(H0rCl) (H+)3
2. Mangae^us ion (Mn ) added to the reaction is oxidized to -■ manganese dioxide, the competition yielding a. limiting induction factor (the mole ratio of manganese dioxide producted to alcohol oxidized) ©f l/2. s' 3
£, Added manganese dioxide inhibits the rate of oxidation of alcohol by a factor approaching 50^ ss a limit.3
4e- The rate of oxidation of 2-deutero-2-Propanol is only 1/7 the rate of isopropyl alcohol,4'5 The rates for 1,1, 1, 3, 3,3-hexa- deutero-2-prooanol and i3opropyl alcohol are about the same.5
From these facts the following conclusions can be made:
A. The active oxidizing species is acid chromate ion. Constant rate constants were not obtained with an expression containing CrO. in Place of HCrO^*. However, by considering the equilibrium,
K80 + Cr307= ^_2HCrO~ ; K * 0.023 Mole/liter
assuming only HCrO"£ as the active SDecies, constant kT s were ob- tained.8
B. An intermediate species of Or IV Participates in the reaction. Since Mn is not oxidized by chromic acid under these conditions, a more active oxidizing agent must be formed during the reaction. The induction factor requires an entity of Cr IV.3
C. The secondary carbon-hydrogen bond must be cleaved in the rate determining step,4'5
From <jver 45 mechanisms considered, four reasonable mechanisms which explain all of the experimental facts emerge if one will assume that JL.- onlv these species Are possible participants* HCrO"^ H , CH3CH0HCH3, (CH3) 3CH0 -,^0- , CR3C0CH3,Cr V, Cr IV, Cr III, Cr II and 2.- reactions between two unstable soecies are negligible. Since gut ©catalysis is not observed many schemes can be discarded. Common to the four remaining mechanisms is the first, rate-deter- mining step1'3 Q
2H+ + HCrO"; +CH3CH0HCH3 -> Cr IV + CTT3C-CH3 + 2H+
which can be pictured as a concerted one- step reaction
OH I HO-Cr-0 I
OH
J
CH3 + H-C-OH CH3
H
6
I
H-O-Cr-OH
0
H
CH3
i
CHa
or a two-step process involving a chroma te ester. + fast _+
HCrO,
+ 2H + CHaCHOKCH.
CK3 / "
OH
OH
+ H3C
++
slow
[(CH3)2CH0Cr03K] + H30
0 CH3-C-CH3 + H3CrOa + H30
In the absence of Mn the following steps describe how Cr IV is converted to Cr III with the oxidation of two more molecules of alcohol and the reduction of one more Cr VI. One of the schemes is as follows:
Cr IV + CHgCHOHCF, Cr VI + Cr II -
0
Cr II + CHaC-CH.
!r III + Cr V
0 1/
Cr V + CK3CH0HCK3 -> Cr III + CH3C-CH3
i I, In the presence of Mn , only the following scheme will explain both the Inhibition and the induction factor.
Mn + Cr IV
Mn III + Cr III ++
2Mn III + SH30 -± Mn03 + Mn + 4H
The esterif ication mechanism is further substantiated by recent observations. Dilute benzene and toluene solutions of di-t-butyl and diisopropyl chromates have been "prepared.6* 7 Since the com- pounds are very unstable they have not yet been isolated; however, analysis of the solutions indicate a compound containing two mole- cules of alcohol per atom of chromium. These compounds can be ex- tracted into benzene but cannot be removed from benzene solution by extraction with aqueous bicarbonate or carbonate, indicating that they are neutral esters. Both hydrolysis and internal oxi- dation-reduction of diisopropyl chromate in benzene are catalysed by bases like pyridine, qulnoline and dimethyleD-iine.6' 7
hydrolysis R0Cr0s0R + H30^1CrO^ + 2ROH
Cr03 + H30 + C5H5N-^_C5H5NHHCr04
oxidation- (CH3) sCHOCrOsOCR(CHa') 3 reduction
(CH3)3C0 + (CH3)3C"OH+CrQ
The oxidation of iso^ropyl alcohol is also strongly catalysed by ■pyridine though the concentration of the free base is very small in this acidic medium.6
-3~
Oxidation of Olefins.- On vigorous oxidation with chromic acl in aqueous sulfuric acid, olefins generally give acids and ketones rising from the cleavage of the double bond. With some systems however, under less vigorous conditions "products of oxidation at the allyl position have been found. By slowly dropping the hydc- carbon dissolved in carbon tetrachloride into a solution of chromi acid in acetic anhydride at 0°, Treibs and Schmidt obtained small amounts of allylic oxidation.8
50 g,
o
50 ff)
■o
+ 20 g. starting
material
2 g.
^\^
n
+ 22 g. starting
material
2 £.
50 g.
pome
7~Keto~chole steryl acetate in somewhat better yield was obtained b| a similar oxidation of cholestervl acetate In glacial acetic •-••". acid.9'10
In addition to acetone, 2, 2-dime thy 1-4-penta none and tri- methyl-acetie acid, Byers and Hickinbottom11 obtained some 2,4,4- trimethylpentanoie acid and 2, 2,3, 3-tetramethylbutanoic acid from the chromic acid oxidation of a mixture of 2, 4; 4- t rime thy 1-1- pentene and 2,4,4-trimethyl«2~T)entene .
_4~
CH3 CH3
CH3-C-CH3-C=CH3 CH3
CH,
CH.
CKg-O-CHg-CH-COgH
+
CHg
CK3-C-GH=CV 3 J > CH*
+
CHa
'"rrT
+ normal TWoducti
By controlled oxidation in acetic anhydride the authors were able to isolate the e^orides of the two olefins. On hydrolysis with aqueous sulfuric acid, they gave 2,4,4~trimethyl-l, 2-r)entanediol, 2,4,4-trimethyloentanal, 2,4,4-trimethyl-2, 3-oentanediol, and 2,2, 3,3-tetramethylbutanal, which would be oxidized to the products ob- tained from the olefins by chromic acid. These epoxides may be intermediates in the aqueous oxidation of the olefins.
Paraffin Side Chains.- A striking difference in the behavior of chromic acid compared with other oxidizing agents was discovere< by Fieser.13'13 While permanganate, hydrogen peroxide, and hypo- chlorite oxidize the nucleus of hydroxyalkylnaphthoquinones, chromic acid, like enzymes in the human body, attack the side chain. The noint of attack and the oroducts vary with the side chain, but some generalizations can be made: 1.- No attack occurs at the a or .p carbon atoms near the quinone ring. 2.- Tertiary carbon atoms are most easily oxidized while methyl groups remain unchanged. 3.- Attack usually occurs at or ne^r the ^ carbon atom. 4.- After the introduction of a carbonyl group 6 oxidation can
These side chains
occur, producing degradation of t^e side chain.
yield:
CH.
I i un 3
,(CK2)n-Cf . +
0HCH3
OAc
PAc
n = 2,4,6,7 -CxoHgi-n
n = 2,185?; 4, 38/; n = 4,34/; 6, 21#. 6, 36/ yield 7, 7
I
-(CH3)6C-(CF3)3-CH3 + -(CH3)n-C02H 22^
+
other ketones
n = 2,3,4,5,6,7 totsl 24/
-CH3-CH-(CH3)5_GK3
CF3 o CFa
~CHs-CH-CCHs)4-C-CHa + -CF3-CH- fCHs) -C03H
n
-(CH3)B-C6H
0 M
7/
6n5
— (CH3; 4— C-C6HB 90/
n = 4,3,1
total 22/
-5-
BIBLIOGRAPHY
1. F. H. Westheimer, Chem. Revs., 45, 419 (1949).
2. F. H. Westheimer, J. Chem. Phys., 11, 506 (1943).
3. W. Watannbe and F. H. Westheimer, lb id . . 17, 61 (1949).
4. F. H. Westheimer and N. Nicolaides, J. Am. Chem. Soc, 71, 25
(1949).
5. M. Cohen and F. H. Westheimer, ibid. , 74, 4386 (195?).
6. F. Holloway, M. Cohen and F. H. Westheimer, Ibid., .73, 65
(1951).
7. A. Leo and F. H. Westheimer, ibid. . 74, 4383 (1952).
8. W. Treibs and H. Schmidt, Ber., 61, 459 (1928).
9. A. Windaus, H. Lett re' and F. Schenk, Ann.. 520, 98 (1935).
10. W. Buser, Helv. Chim. Acta, _3_0, 1379 (1947) .
11. A. Byers and W. J. Hickinbottom, J. Chem. Soc, 1948, 1334.
12. L. F. Fieser, J. Am. Chem. Soc, 70 323? ''1948).
13. L. F. Fieser, ibid. . 74, 3910 (l95?5.
A NEW SYNTHETIC ROUTE TO CYCLOPROPANE S Reported by S. L. Jacobs December 12, 1952
The following discussion is based primarily on the work re- cently done by Linstead and co-workers of the Imperial College of Science and Technology of London.1
The reaction of ethyl sodiomalonate with l,4-dibromobutene-2 in ethyl alcohol was employed in an attempt to prenare 3-hexene- 1,6-dlcarboxylic acid. The success of this reaction was antici- pated on the grounds that primary allyl halides had been shown to condense with sodiomalonate by a normal S g mechanism.2 The main product of the reaction, however, was found to be (l)«
CH3=CK-CH C (C03Et) 3
(I)
A minor fraction was obtained consisting mainly of (II) and (ill)
(Et03C)s-CH— CK3-CH=CH^CH2-CH(C02Et)
CH3=CH— CH~CH(C03St) 3 CK3-CK(C03Et)3
(II) (III)
The structure of (I) was proved by U.V. absorption and ozonolysis, Catalytic hydrogenation of (i) gave ethyl n-butylmalonate, CH3-CK3-CH3-CH3-CB"(C02Et)2 , with the up-take of two moles of hydrogen. Such ring fission usually requires more drastic condi- tions than were used here (Adams' catalyst). This ring fission may be thought of a s 1,4-addition to a system comprised of a
The following
double bond conjugated with a three-membered ring.
are other examples where cleavage of a cyclopropyl ring occurs
on hydrogens t ion :-
H2-Pd
R=H or OAc
=-*
iPr
Gyclopropylftlkeaee such as 2-ryclopropyloenten<?-2, 2- cyclopropyl-. propene, and vinylcyclopropane have been hydrogenated under suit- able conditions to give mixture p of cyclopropyl alkanes and straight- or brancbed-chain paraff ins'. 5' s> 7 "in some cases, similar conditions of hydrogenation do not cleave the cyclopropane ring as 1 s the case with many 2-cyclopro■nyl-l-^alkenesE, 8' 9 the trans- chrysanthemum mono- and dicarboYy'lic acldslc and certain tcrpcn^-s?'11
-2-
The ability of a cyclopropane ring to conjugate with various chromophores is' shown by physical evidence7'18 and by the reductive fission of vinylcyclopropanes mentioned above. The electronic interaction with other chromophores which is involved here is due to the fact that the electrons in cyclopropane rings are more weakly bound than the usual (^electrons and exhibit characteristics usually associated with unlocalized TT-eleetrons.
Molecular refractivlty data indicate that the exaltation ob- served for the new compound (l) is due mainly to interaction of the cyclopropane ring with the adjacent vinyl grout).
Moleculai (IV) |
? Ref ractivitie s |
|||
C03Et "^>^C033t |
Rt) ( ob s . ) |
Rt) (calc. ) |
Exaltation |
|
45.60 |
45.54 |
0.06 |
||
Crip=Crl_— i , |
(v) |
23.85 |
23.37 |
0,48 |
(i) |
54.96 |
54.36 54.41 ** 54.84 |
0.60 0.55 0.12 |
|
* Calcul ** Calcul » „_ , .,„ , , |
a ted from HD(obs. ) for (IV) a ted from RD(obs.) for ( V) |
.... |
The structure of (II ) was "oroved by conversion to suberic acid, and that of (ill) by oyolize tion with NaOEt to ethyl 2-keto-4-vlnylr- cyclopentane-l,3-dicarboxylate (VI ) ,
CH3=C^-CH-CH-C03Et I \1 = 0 1 f CKg-CK-CfOgEt
(VI)
Absence of solvolysis Products in the malonate-dibromobutene reaction indicates the absence of a carbonium ion mechanism, :and a migration of bromine from an a- to a Y —carbon before replacement is considered unlikely.13 Also, 3, 4-dibromobutene-l (to which the 1,4- compound may isomerize14) does not give the same reaction with ethyl sodiomalonate . The reaction is therefore said to proceed without initial rearrangement by a bimolecular nucleophillc attack and, after substitution of the first bromine, may be represented as
Br-CH3-CH=CH-CH3-CH(C03Et)3 -> Na ~CBr^CHs~CH=CH ^C(C03Et)a] -* (i). (VII) a (3 **NCH3
The high proportion of intramolecular V-attack (rather than a- at- tack) is probably due to the trans configuration of the double bond. Intramolecular rearrangements of allyl derivatives at C(tf) have previously been suggested to account for the rearrangements of allyl derivatives15'16, as
■ ■ r
-3-
N9 > ^ V ^<
>F3 -3H .^XpRa w3-
j6'ft
CH,=CH CHa=CIT uF-.,H3
Since there is no obvious reason why bimoleoular a-attack would not Predominate to give (ll), the formation of (ill) in greater Proportion than (ll) cannot be attributed to an inter- molecular ^-attack of (VII) . It Is more Probable that (II) and (ill) are formed by further attack of (I) . Similar reactions are known?-7 However, the main product (60^) of the reaction of ethyl sodiomalonate with (I) Is (VI), formed with the elimination of the elements of ethyl carbonate. Suc^ reactions for the conversion of a three- to a f ive— membered ring are known, as that of ethyl l-cyanocycloPropene-l-carbovylate T-rith ethyl cyanoacetate In the presence of some of the sodio derivative of the latter to give the imino compound. ls
The reaction of ethyl sodiomalonate with (i) gives, in addi- tion to (VI), a smaller fraction consisting of a mixture of (II) and (III). (Ill) can be cyclized to (VI) with NaOEt.
Therefore it was shown that the addition of ethyl sodio- malonate to (i) occurs mainly by attack at C/gi (see formula below) to give (ill) which then may undergo cycllzation to (VI). Some O(^) attack occurs to provide chemical confirmation for the existence of electronic Interaction between the double bond and the three-carbon ring in (I): ,'^q ^PNa
„> C \0Et
Na [(Et03C)3CH] CF2=CH-CH-C' -* (Et03C) 3-CF-CH3-CH=CH >NC0Et
(id
CH3
The experimental conditions for the reaction of ethyl sodio- malonate ^lth 1,4-dibromobutene-^ do not favor cycllzation of (ill) since no appreciable alkovide concentration Is built up. Alkoxide Is, however, Produce d in the condensation of malonate with (i). Here, therefore, evtenslve cycllzation occurs.
The foregoing discussion provides further evidence for the conjugation of the three-membered ring with the double bond through elucidation of the malonate condensation with l,4-dibromobutene-2, and a convenient new route to cyclopropane derivatives has been realized.
BIBLIOGRAPHY.
1. R. V»r. Klerstead, R. P. Linstead and B. C. L. VJeedon, J. Chem. Soc. 1952, 3610, 3616.
2. R. E. Kepner, 5. Winstein and W. G-. Young, J. Am. Chem. Soc. 71, 115 (1949).
3. A. G-. Short and J. Repd, J. Chem. Soc. 1959. 1040.
4. F. Richter, W* Wolff and W, Prestine;, Ber. 64, 871 (l93l).
5. V. A. Slabey and P. H. Wise, J. Am. Chem. 3oc. .74, 3887 (l952h
6. R. Van Volkenburgb, K. W. G-reenlee, J* M. Derfer and C. E. Boord, J. Am. Chem. Soc. 71, 172 (1949).
7* Ibid. 71, 3595 (1949) .
8. V. A. Slabey and P. H« Wise, Nat'l, Advisory Comm. Aeronautics, Tech* Note 2258-9 (l95l); 0. A. 4_5, 7531 (l95l),
9. V. A. Slabey *nd PfcH. Wise, J. Am. Chem. Soc. 71, 1518 (194p).
10. H. Staudine:er and L. Ruzicka, Helv. CMm. Acta. 7, 901 (l9S4)«
11. L. Tschueraev and W. Fomin, Compt. Rend. 151, 1058 (1910).
12. L. T. Smith pnd E, R. Rosier, J. Am. Chem. Soc. 73, 3840 (l95l)
13. A. (J. CptchTDole snd E. D*. Hughe p, J. Chem. Soc. 194°. 4.
14. E. H. Farmer, C. D. Laurence and J. F. T^oroe, J. Cbem. Soc. 1928. 729.
15. S. Winstein, Bull. Soc. Chim. 18, C43 (l95l).
16. A. G-. Catohpole, E. D. Hughes and C. K. Ingold, J. Chem. Soc. 1948. 8.
17. W. A. Bone pnd W. H. Perkin, J. Cbem. Soc. 67, 108 (1895).
18. S. R. Best and J. F. Thorpe, J. Chem. Soc. 1909^685.
SULFONATION OF ACID- SENSITIVE COMPOUNDS Reported by Clayton T. Elston December 19, 1952
With compounds that decompose or polymerize in the presence of strong mineral aoids the common sulfonating agents pre of very limited usefulness. Complexes of S03 with various organic bases have proved to be quite effective in the sulfonatlon of many such acid-sensitive materials. In 1926 Baumgarten1 Prepared a complex of pyridine and sulfur trloxide and observed that this complex de- composed to regenerate its components. Thus, a recent was now at hand which under controlled conditions could release Its S03 to nucleophilic compounds and effect sulfonatlon. Baumgarten then proceeded to test the effect of the reagent on a series of organic compounds. He found, for instance, that ohenol could be sulfated by this reagent without any nuclear sulfonatlon as occurs with concentrated sulfuric acid. Other workers, realizing- the advant- ages of the method used it in the sulfonatlon of proteins, amines, amides, oolysaccharide s and Polyvinyl alcohol.
Other S03 complexes are known, as for example those with di- methylaniline3, trimethylamine3 and dioxane4. Dimethylaniline sulfotrioxide is an extremely unstable substance which readily transforms into p-dimethy] anilinesulf onic acid. Trimethylamine sulfotrioxide on the other hand is very stable. However, it is this stability which limits its usefulness since it gives up its S03 only under rather drastic conditions. Dioxane sulfotrioxide, first prepared by Suter4 in 1938, is a rather unstable material and in contrast with pyridine sulfotrioxide decomposes rapidly in water, forming sulfuric acid. Suter hafl made extensive investigations on the use of dioxane sulfotrioxide in the sulfonatlon of unsaturated compounds.6
Beginning in 1946 Terentyev and his co-workers have conducted an extended series of researches on the use of pyridine sulfotri- oxide as a sulfonating agent for acid-sensitive compounds. The method involved heatingth© reactants together in a sealed tube at a temperature of 100-110°C. for a oeriod of eight to ten hours. Usually a threefold excess of pyridine sulfotrioxide gave the best results. Modifications of this general procedure involved reaction at slightly lower or higher temperatures and the use of an inert solvent such as ethylene chloride. In all cases the sulfonated compounds were isolated as their barium salts. This seminar will deal briefly with the sulfonatlon of: (i) furans and coumarone (ii) pyrroles (ill) indoles (iv) unsaturated compounds.
Furans
Mineral acids readily promote ring opening and polymerization reactions with furan and Its homologs and prior to Terentyev' s work very little w?g known about furansulf onlc acid derivatives. Using the general method outlined above he successfully sulfonated furan, 2- me thy If ur an, 2,5-dimethylfuran, 2-acetylfuran and cou- marone, obtaining furan- 2- sulfonic acid, 2-methylf uran-3, 5-di- sulfonic acid, 2,5-dimethylfuran-?- sulfonic acid, 2-aeetylfuran-5- sulfonic acid and coumarone-2- sulfonic acid respectively6' 8; 10' 1X ^f
-2-
At lower temperatures 2-methylfuran yielded 2-methylfuran- 5- sulfonic acid. High yields (55-90#) of the crystalline salts were obtained in these reactions. Furfural did not react under these conditions and the methyl ether of furfuryl alcohol yielded only a resinous product. 2-Furoic acid reacted at 140° with displacement of the carboxyl group to give furan- 2- sulfonic acid. The salts were stable in the presence of hot alkali but were hydrolyzed rapidly with dilute HOI. This hydrolysis, yielding S0S, proceeded with both the a- and {3-sulfo compounds but was much more rapid in the case of the former. However, coumarone-2- sulfonic acid yielded sulfuric acid and coumarone. Compounds in which the sulfonic acid group was a to the heterocyclic atom were readily oxidized by bromine water giving a precipitate of barium sulfate. The method of structure proof may be illustrated with 2- me thy If uran- 5- sulfonic acid.
. . CH=^CH
I] Bromine x | I
CHsy iS03Ba/2 water ~7 BaS04 + CH3C COOH
V
Pyrroles
I*
0
HOI (lSSOv boil " < SO.
Several examples of the sulfonation of pyrrole derivatives are known. In 1885 C^amician and Silber35 sulfonated 2-acetylpyrrole by treating it with sulfuric acid vaoor. By this method they separated and analyzed the potassium salt of the monosulfonic acid. However, they did not determine the Position of the sulfo grouo in the molecule. In 1935 Pratesi36 obtained an excellent yield of 2, 4-dimethyl-5-carbethoxyDyrrole-;3- sulfonic acid by treating the pyrrole with a chloroform solution of chloro sulfonic acid. With this reagent 2,4-dimethyl-3-carbethoxyoyrrole re sin if led, as was the case with pyrrole itself and its other homologs which were not stabilized by electron-withdrawing substituent s. With pyridine sulfotrioxide, sulfonation of the latter type of compounds was possible1 3' 15' 18' so> 22. Terentyev obtained pyrrole-2- sulfonic acid, 1-me thy lpyrrole- 2- sulfonic acid, 2-me thy loyrrole-5- sulfonic acid and 2, 4-dime thy lpyrrole-5- sulfonic acid by the sulfonation of the corresponding pyrrole derivatives. Yields were generally high, p-sulfonic acids were obtained from 1, 5-disubstituted pyrroles by varying the experimental conditions. Ether or benzene was used as solvent and the reaction mixture was heated to 100° for six hours. For example, a mixture of mono- and dlsulfonic acids was obtained from 2,5-dimethyloyrrole . Some of the reactions of these materials are given below.
-3-
• ' «
CHga '.CH3
H
S03Ba/3 C =CH
.S03Ba/.
II
> CH,!1 MCH
CrO
OrC
NH;
G=0 OBa/2
Ba/303S S03Ba/s
+ II II
H
Cr03
/»
C =-=^C-S03Ba/
(decolor- ized but no Ba304)
0=C
NH3
C=0 OB a/ 3
CrO.
water
CH-CH
I I
0=C G=0
H
XN''
BaS04
2-Acetyloyrrole can be sulfonated with fuming sulfuric acid to yiel 75# of a monosulfonic. acid. Contrary to expectat i ons, oxidation of this material with a dichromate sulfuric acid r.-'ixture yielded a sulf omaleamic acid, indicating that sulfonation had occurred in the
P*poGitio:na Sulf: oxide gave a mix I' a ce t y 1-0 yrx-o le- - 5 ., 5«
■nation of 2-aeetylryrrole wlch pyridine sulfotri- ire of 2- acetyloyrrole-4-csulf o: i 3 r.cld and 2- diaulfonic acid. Even the highly unstable 2-
chlor: xryrrole c^n be sulfonated with nyridine sulf otrioxide to yield the corresponding 5- sulfonic acid.
Indoles
In its behavior toward pyridine- SO 3 indole resembles pyrroles,
but it it; somewhat leas aoid-sensitive 7> 9; x 3> s'-. A sulfonation temperature of i^0° was found necessary since at lower temperatures the reaction apparently stops at the N-sulfo derivative. It is noteworthy that sulfonation occurs at the 2-oobition, whereas substitution reoc'cio:i3 with indole generally occur at the 3-posi- tion. Indole itself gave an almost quantitative yield of Indole- 2- sulfonic acid and 3-methylindole gave 3- methyl indole- 2- sulfonic acid in a yield of 55f . 2-Methylindole failed to react under the conditions employed. However, sulfonation in the 3-posltlon was effected with 2--bhenylindole. The structure of indole- 2- sulfonic acid was confirmed by fusion of the salt with potassium hydroxide. Oxindole was the oroduct of fusion. Oxidation of 2-phenyllndole-3- sulfenic acid with Potassium permanganate yielded bcnzoylanthranillc acid.
-4-
Unsaturated Hydrocarbons
Cyclopentadiene polymerizes very easily. This sensitivity to acids is so marked that even a minute trace of acid produces rapid tarring. Sulfonation with pyridine- S03 yielded 42^ of the mono- sulfonic acidsx. The fact that sulfonation had occurred at the methylene group T»ras shown by oxidation to sulfoacetic acid.
\ «
S03Ba/;
KMnQ4
COOK COOH
! + i
COOH CHS03Ba/ !
COOH
C-CH3-S03Ba/s + 3C03 Ba/2
Indene likewise yielded a monosulfonic acid16. The authors assumed that the sulfo grourc had entered the 2-position but offered no definite nroof.
Pyridine- 30 3 did not react with Paraffins, cycloparaff ins, benzene homologs or olefins with a non-terminal double bond. Good yields of sulfonic acids were obtained from cyclohexene, methylene— cyclohexane, camphene, styrene and conjugated dienes such as 1,3- butadiene and lsoprene16' 33. The formation of sulfonic acid deri- vatives was assumed to take place through addition of two moles of S03 at the double bond. In order to obtain the sulfonic acid salts the sulfonated mass was treated with barium carbonate. Then, de- pending upon the stability of the intermediate product, either the barium salt of the lsethionic acid derivative was obtained, or the splitting off of a molecule of sulfuric acid took place and the barium salt of the unsaturated sulfonic acid was formed. The lattei was observed In the case of camnhene, styrene, butadiene and isoprene.
Vinyl ethers react in a similar manner. For example, n-butyl vinyl ether yielded a barium salt having- the emoerical formula CeHi30eBa. The salt did not bleach bromine water and upon hydro- lysis yielded butyl alcohol, barium sulfate and the barium salt of sulfoacetaldehyde. Upon the basis of these results the following structure was assigned.
C4H90-CH-CH3-S03> 6-303-0-Ba'
'0
BIBLIOGRAPHY
1- P. Baumearten, Ber., 59, 1166, 197? (1926).
2. F. Beilstein and E. Wiegand, Ber., 16, 1267 (1883).
3. 0. ¥. Willcox, Am. Chem. J., 32, 450 (1904).
4. C. M. Suter, P. B. Evans and J. M. Kiefer, J. Am. Chem. Soc, 60, 538 (1938).
~5-
5. 6.
7.
8.
9.
10.
11. 12. 13.
14. 15. 16. 17. 18. 19.
20.
21. 22. 23. 24. 25. 26.
F. G-. Bordwell, 0. M. Suter and. A. J. Webber, J. Am. Chem. Soc, 67, 827 (1945).
A. P. Terentyev and L. A. Kazitslna, Comptj rend. acad. sci. U.R.S.S., 51, 603 (1946).
A. P. Terentyev and S. K. Golubeva, lb id . , 51, 689 (1946). A. P. Terentyev and L. A. Kazitslna, ibid.. 55, 625 (1947), A. P. Terentyev and L. V. Tsymbal, lb id . , 55, 833 (1947). A. P. Terentyev and L. A. Kazitslna,— J. Oen. Chem. U. S.S.R., 12, 723 (1948) (Engl, translation).
A. P. Terentyev mid L. A. Kazitslna, ibid. . 19. 481 (1949). ev and L. A. Y^novski, Ibid., J19, 487 ('1949).
K. G-olybeva and L. V. Tsymbal, ibid. . 19,
P. Terent:
rr
A.
A. P. Terentyev,
753 (1949).
A, P. Terentyev Terentyev Teren eye v Terentyev Terentyev To rent /ev
Sc
N. P. Volynsky, ibid.. 19, 767 (1949).
L. A. Yanovskaya, ibid. , _19, 1*67 (1949).
A. V. Dorribrovsky, "ibid. , 19. 1469 (1949).
L. A. Kazitslna, ibid., JL9, a 491 (1949).
L. A. Ya n ov ska y a . ib Id . . 19. a591 (1949).
Ae Kazitslna ^nd A. M. Turovskaya, lb id . , 187 (1900).'
P. Terentyev, L. A. Yanovskaya and V. G-. Ya shun sky, lb id . ,
P. P. P. P. P.
A, A, A, A. A.
£'2,
A.
j^O, 539 (1950)
A. P. Terentyev
A. P. Terentyev
A. P. Terentyev
A. P. Terentyev
G-. Ciarnician and
and
a i.i.d. a rid a nd
L.
and
and
and.
a nd P
A. V. Dombrovsky L. A. Yanovskaya
A , V . Do mbr ov sky L . A . Ya nov skay a Silber, Ber, , LP
P. Prate si, Gazz. chlm. ital., 65, 43 (1935).
21, 30? (1951).
21, 307 (1951).
21, 775 (1951). ibid,. £1 1415 (1951). 879 (1885).
ibid o , JLbld . ,
ibid. ,
SYNTHESIS OF SUBSTITUTED SI LANES Reported by C. W. Hinman December 19, 195?
Introduction
Organic compounds of silicon have been known for more than a hundred years, many of them havlna: been oreoared with the expecta- tion that they would be analogous to those of carbon. It was found, however, that silicon differs from carbon in many respects. One of these is that the silicon-oxygen bond is exceedingly strong as compared to silicon and any other element,, Another is that chains having mere than five consecutive silicon atoms are highly unstable and most readily subject to hydrolysis.1 Ore-anosllicon compounds are solely products of t^e laboratory, as none have been found in nature.
Synthesis
The basic starting materials for the Production of organo- silicon compounds is either elemental silicon or silicon tetra- halide. Elemental silicon is produced by reacting magnesium with silicon dioxide, or by reaction of an nlkall metal on silicon tetrahalide. Silicon tetrachloride is produced from ferrosilicon (FeSi) and chlorine, or from free silicon and chlorine.3
Frledel- Craft Method
The earliest method of forming carbon-silicon bonds, the so- called Friedel-Craf t Method, involved the use of zinc alkyls as indicated by the equation:
2ZnR3 + SiCl4 — ► SiR4 + 2ZnCl3
sealed tube
where R is either alkyl or aryl.3 From one to four positions can be filled by controlling the molar quantities of the reagents, but even though one product predominates a mixture alwayg results. Obvious c! isadvantafes of the method are the sealed tube conditions, the preparation and handling of the highly flammable and toxic zirc alkyls and the separation of products. Crgano disilanes have been prepared by this method also.
SisI6 + 3Zn(C3H5)2 > (CsHjsSi-SifCsKjs),, + 3ZnI 4
Wurtz Method
Generally, the Wurtz method is more versatile and gives more easily separated products.
S1C14 + 4RC1 + 8Na ► SiR4 + 8NaCl
The method finds its greatest use in the preparation of tetraalkyl- and -tetraaryl-silanes, or mixtures of these, two.5
-3-
(C6H5)SS1-C1 * H3CC1 -2Ua > (C6H5)3S1-CH3 + 2Na01
6
(C3H50)2S1C13 + 2NaC=CH > (C3H50) 3-Si- (C=CF)
It cannot be allied in the esse of the silane halides which con- tain hydrogen, "because "unsaturated" non-volatile hydrides with formulas varying from SiHn to (SlKi.8)n are formed.8
Direct Method
A method which ±g used commercially is the direct union of alkyl or aryl halides with metallic silicon.
2RC1 + Si --> RgSiClg
150- 3000 3 3
This reaction is carried out in the presence of finely divided copper or silver. It has "been found, that if R is alkyl, copper works best, if R is aryl silver is the most effective.9
Saturated Hydri de Synthesis
Saturated hydride silpnes are produced by allowing magnesium silicide (>ig3Si) to drop into a liquid ammonia solution of ammonium bromide, A mixture of gases consisting of hydrogen, silane (SiK4), disilane (Si3H6), and small amounts of~trisilane (Si3He) is pro- duced.10 .
Grlgnard Synthesis
This method has found the widest use of all, especially in the laboratory, for this method provides an easy means to most of the desired organosilicon compounds in good yields.
3RMgX + SlCl4 ► R9S1C1 + 3MgX3
3RMgX + SiHClg > R3SIH + 3MgXa
It is difficult to prepare tetraalkyl- or tetraaryl silane s by this method v however., lu oases where the Grrlgnard method gives poor yields or- fsile to reset it has been found that the orga no lithium compounds often Ccsii be used.,
S1014 + 4RLi ► SIR 4 + 4LiCl
This usually gives tetra substituted silanes, but in some cases of sterlcally hindered lithiv.m compounds only the di- or trisubstitut- ed compounds pre isolated. It should also be noted that trialkyl- and triaryl- silanes react with organolithium compounds.11
RaSlH + RLi > R4S1 + LIH
The reactions of organosilicon G-rignard reagents are very general, but for the sake of clarity only a few of the more simple ones will be given here to illustrate some of these reactions.
-3-
HoCj-a blT"GriaC±
H3C H3C H3C H3C H3C H3C HSC
H3C
Ms
— > (CH3)3Sl-CH3MgCl
3~Si"0HaMgCl + C03
3SiCH3MgCl + H3CCK
SiCH3MgCl + (H3CC0)20
3SiCH3MgCl + H3CCOCH3
0 3SiCH3MgCl + H3C- -CH3
3SiCH3MgCl + (CH3)3SiCl
3SiCH3MgCl + (CH3)3SiCl3
H
3SiCHMgBr + (C"3)3SiC3r
^6^5 C6HS
3SiCH3MgBr + BrCH3CH=CH3
-> (K30)9SiOH3COsH
-> (H3C)3SiCH3CHOHCH3
-> (H3C)3SiCH3COCH3
-> (H3C)3SiCK3COH(CH3)3
-> (HsC)3SiCH3CH3CHsOH
1 s
-> (H3C)3SiCH35i(CK3)
CF3
-> (H3C)3SiCH3-Si-CH3Si(CH3)3 CH3
Tf ?6FB -> (H„C)3Si-C 6-SifCH,) I *
13
-> (H3C)33iCH3CH3CH=CH3
1 4
H3C(CH3)3C=CMgBr + (CH3)3SiCl > H3C (CH3) 3C^CSi (CH3)
BrMgC=CMgBr + 2(CH3)3SiCl
-v (H3C)3SlC=CSi(CH3)3 ls
1.
2.
3.
4.
5.
6,
7.
8.
9. 10. 11. 12. 13. 14. 15.
BIBLIOGRAPHY
H. Hausman, J. Chem. Ed. _23, 16 (1946).
T. Alfrey, F. Honn, and H. Mark, J. Polymer Chem. 1, 102 (1946 J
Aldrich, Organic Seminar Abstracts, January 23, 1948;
Friedel, Compt. Rend. 68, 923 (1869).
G-ilman and Clark, J. Am. Chem. Soc. 68, 1675 (1946).
Bygen, Ber. 48, 1236 (1915).
Frlsch and Youne:, J. Am. Chem. Soc. 74, 4853 (1952).
A. Stock et.al. Ber. 54, 524 (l92l); 56, 1698 (1923).
E. Q. Rockow, J. Am. Chem. Soc. 67, 963 (1945).
Johnson and Hogness, J. Am. Chem. Soc. 56, 1252 (1934).
W. H. Hill, Jr., Organic Seminar Abstracts, November 19, 1948.
Hauser and Hance, J. Am. Chem. Soc. 74, 5091 (1952).
Whitmore et.al* J. Am. Chem. Soc. 70, 4184 (1948).
Hauser and Hance, J. Am. Chem. Soc. 74, 5091 (1952).
Frisch and Young, J. Am. Chem. Soc. 74, 4853 (1952).
RING- CONTRACTION REACTIONS OF TROPOLONES Reported by Harry W, Johnson, Jr. December 19, 195?
Among the more interesting: reactions which the trooolones and substituted trooones undergo are those which lead to the formation of benzenoid products. Four such reactions will be discussed here: A) the base induced reactions of trooolones, tropclcne ethers and 2-halctrooones; B) the reaction of trooolone with hyoohalite; C) the ring contraction encountered ™ith 3~ and 7-diazotrooolones; and D) the contraction of polynitrotrooolones under acidic or neutral conditions.
A« The base induced reaction
Tropolone, when heated to 230-235° in the presence of KOH, undergoes rearrangement to yield benzoic acid (17^) . Many such reactions of trooolone s are kno^n, and they have been used in the
tropolones.1' s' 3 mechanism for such
establishment of tbe structures of substituted Equation (i) illustrates the commonly asserted reactions.
e
(i)
-/
vr->
CO-H'
<o
<r
Included in the shove reaction sequence are intermediates of the norcaradiene type; and, while no substance has been shown to have the norcaradiene skeleton (eg. A), there is evidence that cyclo- heptatriene is in equilibrium with norcaradiene as sho^n by the formation of a Diels Alder addu^t derived from B.4
H H
(B)
The 2-halotropones also undergo ring contraction with hydro- xide ion, but the conditions required are much milder. 2-cMoro- tropone may be rearranged by heating it under reflux with 3 N
-2-
sodium hydroxide for 2 hours (70? vlplfl) 5'6 Tr*,ii« -pu„ o i. ^«^t3?^1aaeUnfaerS° — ranLLnt^n heals" t^iooltl?^ ^ after pu«f?™??™°Tr K T^ ,ln yle:Ld8 of 42 gnd 5^> respectively,
below [11) ig consistent with the above facts.
%—**
C0aH
r ^
Ri here
KJ^wgSlaS nethSx^e'lnVef^0?010116,^^^ e^V^ " benzoate (46/ on basis £? ^L? ref^xl£5 ?e*hano3]ryieldd T methyl cation) l8»aSn!J ° benzoic acid obtained on saoonifi- cation; as do the substituted ethers (see however ttxZ n^vt
paragraph on halotropolone ethers) In thi7JB« IS ', ! ?!
carbon bearing the methoxvl Soup Is frJtlLo f attaCk &t the q0 o-i--t-Q«v „4- .4-1- , ,-u sI"up is irultless for rearrangement
so attach at the carbonyl is Postulated as shown in equa^on (III)
@
(III)
OCH,
~>
OGH.
^__^
II ^
trichlofo benzoa' e^an! iS^etSxv^^l^r6^7^ f^1 8'4'6" this case the compound may leZ*. either 5-"f ^obenzoic acid. In giving displacement! ™ - * _ _ I either as a halotropone (normally rearrlneement? ?* ,- trooolone methyl ether (normally giving
arrangement). It is apparent that the compound reacts as an
-3-
ether, but since the following species (IV) is in the reaction sequence, either r'hWide or methoxide may be eliminated.
€>
-OCHg^ OCH* — f
-CI
->
0CH3
B. Ring contraction ^ith hypohallte
If tropolone is allowed to stand at room temperature with 2 N sodium hydroxide containg 2 molar equivalents of iodine or bromine^ triiodo- or tribromoohenol is obtained in yields of 20 or ZOf,1'5' respectively. The mildness of the conditions, as comoared with those reauired when sodium hydroxide alone is used, led to the
oostulation of a different mechanism an essential part, for the reaction, in equation (VI) .
0
{
in T'hich the halogen plays The mechanism, is illustrated
(VI)
OH i
-I
C. Formation of salicylic acids via diazonium salts
In attempts to make 3- or 7-halo or cyano tropolones, several authors tried to use the Sandmeyer reaction on the 3- or 7-amino- tropolones. In such cases one usually obtains a mixture of the Sandmeyer product and salicylic acid.3'3'9 For example, 7-amino- 4-isopropyl-troPolone^, when diazotized and subjected to the Sandmeyer reaction^ yields 3- and 7-isoprooyltropolone together with 25-30$ of p_-isor>ropyl- salicylic acid.9 The salicylic acid is obtained in yields of 50-60^ if the diazonium salt is' heated with dilute sulfuric acid.9'10 The following mechanism has been oostulated to account for the reaction. It should, perhaps, be noted tb^t the diazonium salts do not undergo immediate rearrangement, since Haworth has demonstrated that it is Possible to couple the dia- zonium salt from 2-aminc— 6-methyltrooolone with the sodium salt of p-napthol. s
(VII)
OH N
-4- 4>. OH OH
C(OH)s j^/V^OH
V
D. Acid catalyzed re a rrs nge me n t s of trononones
The nitration of 6-methyltror)olone yields a complex mixture o! mono-, di-, and trinitrotropolones, together *rith some ''yields not stated) 4,6~dlnitro~m-toluic acid.3 The route suggested for its formation is shown belo^.
OH OH
(VIII)
CH
COPH
S*\
CH3
Ml
NO.
NO-
In another no] yriitrotroPolone (3, 5- dinitro-6-isoprooyl~ tropolon.f- ) it "«r?s noted that rearrangement to the dinitro benzoate occur-ec; whfvh heated for a. few moments in methyl or ethyl alcohol. The suggested course of the reaction is shown in equation (IX).3,1C
OH
0
(IX)
OH
-Y\=^yKm
NO.
<3y C-OR
0
BIBLIOGRAPHY
1. W. von E. Doerinfy and L. K, Knox, J. Am. Chem. Soc, 73, 828
(l95l)0
2. R. D. Haworth and P, R„ Jeffries, J, Chem, Soc, ?067 (l95l).
3. A. J. birch, Ann. Rots., X U'lII . 185. (l95l).
4. E„ P. Kohler, M, Ti shier," H„ Potter and H. T. Thompson, ". . J. Am., Chem Soc, 61, 1057 (1939).
5. W. von E. Doering and L. H. Knox, J. Am. Chem. Soc, 74, 5663
(1952).
-5-
7. W. von B. Doerlog and L, H. Knox, J. Am. Chem. Soc, 74, 5688
8. J. W Cook, R. M. Glbb *Od R. A/ Raphael, J. Chem. Soc. 2244
9# ^J??^' Y* Eit*hsra, sisd K. Dol, J. Am* Chem. Soc, 73, lflQ5 11951;, ' — '
10, T. Nozoe, Nature, 1&7, 1055 (l95l).
CONCERTED REACTIONS: POLYFUNCTIONS CATALYSTS Reported by Richard L. Johnson January 9, 1952
In organic chemistry there are two main classes of reactions, ionic and free radical reactions. The polar displacements »re generally regarded as falling between two extremes: 8^1 and S 2 reactions, according to their apparent kinetic order in aqueous or other polar solvents. Swain1'8 has shown that both extreme cases display third order kinetics in nonoolar solvents such as benzene.
The principle that both electropMlic and nucleophillc reagent; are necessary for a "oolar chemical reaction is supported by the observations that no polar reactions have been found to occur in the gaseous phase, all reaction taking nlace on the ccnta Iner or catalyst provided, Swain1' s direct evidence was Procured through experiments in benzene solution T-rith pyridine and methyl bromide3
(for SAj2 type) and with trlp^enylmethyl halides and methanol4
(for S^l type).
In aqueous solution the kinetics of the enolization of acetone and the mutarotatlon of glucose (apparently first order in reactanl and hydronium ion in acidic solution and first order in reactant and hydroxy! ioi* in basic solution) were shown to be explainable on the basis oiy a termolecular reaction. b The nucleophillc reagent may be Off" or RCC3" in basic solution and H20 in acidic solution* The electrcohiiic reagent may be Hs0 in basic solution and H30+ or RC03H in acidic solution. Since water is nresent in the same constant hieh concentration with resoect to the other reagents, the t^ird order term is not experimentally detectable in the mutarotation of glucose in aqueous solution.
In ordinary reactions the nucleochyllc, electroohilic. and reacting groucs are in separate molecules 'Fig,. 1). If the nucleo- phillc group is in the same molecule as the reacting species, the neighboring group reactions, studied by Winste in, occur (Fig. 2). The unusually high reactivity of tetramethylene glycol toward HBr in phenol may be explained as an example in which the electronhilic and reacting groups are in tbe same molecule (Fig. 3). If the nucleophillc and ele.^tronhilic groucs are present in the same catalyst molecule, polyfunctional catalysis can occur (Fig. 4).
(N> -^ (r; -> (e) (NL^tJR) — ) ©
Figure 1 Figure 2
(g) ~> (SX^^JS) (SI —» ® ^X>3
Figure 3 Figure 4
When two functional groups are available in the same molecule, only a bimolecular collision is necessary to effect a reaction (Figs. 2? 3, 4)» therefore the rate is increased. This situation is
especially advantageous in dilute solutions where termoleeular collisions are rare in relation to "bimolecular ones.
The mutarotation of tetramethyl glucose in non-aqueous solutions provides an interesting example of acid-base catalysis. Prior to 1927, T. M. Lowry6' 7* ^had shown that the mutarotation reaction re- quires both an acidic (electrophilic) and a basic (nucleoohilic) catalyst for the formation of the free aldehyde from the hemiacetaL The recyclization of the aldehyde occurs at a faster rpte than the opening, hence does not effect the overall rate- In Inert 'aprotic) solvents, such as chloroform, the reaction -oroceeded at a slow rate* In the dryest benzene that Lowry could prepare, the rate was also slow, and it increased more than a hundredfold when a trace of water was added, addition of p trace
water was more tightly bound to the ester than to the benzene, re- ducing the catalytic activity. Pyridine alone was a Door catalyst, but a mixture of pyridine and water in the ratio t*ro to one was twenty times as effective* Dry ere sol likewise gave no appreciable catalysis alone, but one to two mixtures of pyridine and ere sol had a rate twenty times as fast. That the polarity or the dielectric constant of the aprotic solvent was not an important factor was shown by the demonstration that ethyl acetate and acetone retarded,
In ethyl acetate, where the rate was again slow, the of water w*>3 not so effective, because the
rather than speeded, the reaction
The salts of strong acids were
found not to catalyze the reaction, although undissociated weak acids and bases did catalyze it.
Lowry1 s work definitely established that mutarotation reauired both acidic and basic catalysts. Until Swain*s work, it was thought t^at there were two different mechanisms for the reaction, one operating in acidic media, the other in basic media. Both mechanisms led to a common intermediate, through the action of first one species of catalyst, then the other, and no reaction step involved more than a bimolecular collision. Other, more com- plex, mechanisms of this type have been considered, but none in- volves a step of more than second order.
The concerted mechanism9 shows the reaction proceeding by a simultaneous attack of acidic ^nd basic catalysts in a termolecular reaction as shown in Figure 5.
^/'(>lHC + B:
HC
7
:0: + _HA
A
V
0 n HC |
+ |
H:B |
||
:0:H ! |
+ |
A~ |
Figure 5
Swain and Brown9 repeated the work of Lowry, with the difference that they ran the reactions in benzene solutions instead of running them without solvent. Because the concentration of the catalysts
-7*-
did not change during the reaction, the kinetic order of each run was first order. The variation of the rate of reaction caused "by varying the amounts of the catalysts ws? used to determine the true order of the reaction. The total rate expression for mixtures of phenol and pyridine was found to he:
k = 0.0000013 +" 0.0081 C^OH)3 + 0.00048 (Py) (S + S» )
+ O.gl (Py) (j60H) + 0.84 (Vy) (j6oh)S
where the first term is the "blank" (rate in pure benzene) term, the second concerns catalysis "by phenol and a phenol dimer, the^ third shows action of pyridine as base and sugar as aeid,[(S + Sv representing total sugar concentration, both a and $ forms] , the fourth term represents the action of phenol as acid and pyridine as base, and the last, which is appreciable only in concentrated solutions, shows the action of a phenol dimer as acid with pyridine as base. The need to include the sugar in an "acid catalysis" term shows that the sugar itself may act as an acid in this re- action, but not as a base.
All except the "blank" term are at least third order when sugar- concentration is considered. The results of these experiments in- dicate that the mutarotation of tetramethyl glucose is indeed a termolecular reaction requiring both acidic and basic catalysts. The equation is in substantial agreement with the data of Lowry and Falkner.6 2-4-Dinitrophenol and p-nitroohenol gave much faster rates than Phenol when pyridine was added, but not enough data were gather©:" to show the kinetic order of the reaction. The catalytic effect of 3-hydroxyquinoline was tested in acetone rather than benzene because of the low solubility of this catalyst in benzene. Acetone alone caused no catalysis, and the reaction constant was first order in sugar and second order in catalyst, indicating that one molecule of the catalyst acted as a "^ase, another as an acid.
In the latest Paper in this series10, Swain and Brown have shown the existence of bifunctional catalysts for the mutarotation reaction. Theee catalysts have both acidic and basic groups so situated that one molecule of the catalyst can simultaneously donate one proton and remove another from the glucose molecule. When this situation occurs, it is necessary for only one molecule of catalyst to become associated with the reacting molecule to make the reaction possible; hence in non-polar solvents the reaction will obey second order kinetics.
The oolyifunctional catalyst most studied was 2~hydroxyovr idine. It is only 1/1000 as strong a. base as pyridine, and only 1/100 as strong an acid as Phenol, but it is found to be a very much stronger catalyst than both, as shown in the table below:
Cone. 2-OH Py Cone, each of Relative effectiveness
Py and j60H of 2- OK Pyridine
0.05 M 0.05 M 50 times better
0.001 M 0.01 M (Calculated 7CC0 times better
Rate)
Despite the fact that it is a nearly neutral molecule, the 2- hydroxypyridine is over ten times as effective in benzene as hydro- nium ion is in water. With ^.OOl M catalyst, the rate is not significantly changed "by the addition of either 0.1 M ohenol or 0.1 M pyridine, showing that the oolyfunctional catalyst is self- contained.
3- and 4-Hydroxypyridine are at least as reactive as the 2- hydroxyoyridine in ordinary reactions. The two functional groups are, however , too far apart to react simultaneously with the sugar molecule. These substances are le3s than l/lOOO as effective as the 2-hydroxypyridine, and the kinetic order of the rate deter- mining step is third order, showing that two molecules of catalyst are needed oer sugar molecule.
The 2-hydroxypyridlne and the sugar form a. complex immediately upon mixing, as evidenced by an increase in ootical rotation of the solution. Neither of these substances complexe s with either phenol or pyridine. The complex formed is probably a chelate, shown in Figures 6 and 7. Which form of ^the catalyst predominates ■ in benzene solution is unknown. The catalyst is found to complex also with 2-tetrahydr°pyranol, (Fig. 3), which thereby inhibits the mutarota- tion of the sugar.
<:
"V
S
X
HO H N_, \ CH — 0 ^
Figure 6 Figure 7 Figure P
From the kinetic data it appear^ that other catalysts for the reaction ^re: (in benzene solution}
Folyfunctional Acidic only Basic only
2-hydroxy-4-methylquinoline p-nitroPhenol pyridine
Benzoic Acid phenol 2-methoxypyridine
Picric Acid N- methyl 2-aminooyridine a-pyridone
In benzene solution the rate constant for 2-hydroxypyridine catalysis is half order in catalyst in concentrated solution, in- creasing to first order in very dilute solutions. In such solutions the sugar aporoaches zero order kinetlcally.
Chlorobenzene and acetone solutions gave like results to benzene solutions. In water solutions, glucose itself was used as the reactant since its rate of mutarotation is about the same as that of tetramethyl glucose o 2-Hydroxyoyridine was four to five times as effective as the calculations predicted on the basis of its acid- base constants as a monofunctional catalyst. This rate is not near- ly so spectacular as that in non-aqueous media.
The work of Swgin and Brown shows that a oolyfunctional catalyst must have both acidic and basic groups so arranged that they win
-5-
possessa pattern of molarities opposite to that of the reacting species in the transition state. The resemblance between poly- functional catalysts and enzymes becomes apparent at once. They have these characteristics in common:
1. They possess no extremely reactive functional groups.
2. They have high activity in dilute concentrations at mild temperatures and in nearly neutral solutions.
3. They are specific.
4. They form comole-sres with the reacting molecule before reaction occurs.
5. They react by molar, rather than free radical, reactions.
The Probability that enzymes react as polyfunctional cata- lysts in concerted c5 isT3lacement'a is succor ted by the observations that enzyme-catalyzed reaction-13 have lo^r activation energies, as could be achieved through polyfunctional catalysts. It is interesting to note that &n enzyme has been discovered which catalyzes the mutarotation of glucose at a rate 20 times that of hydroxy 1 ion.l*
REFERENCES
1. See: A. Kresge, Organic Seminars, U. of 111., I Semester, 1950
2. G. G. Swain, Record of Chemical Progress, JL2, 21, (1951)
3. C. C-. 3wain,et_al., J. Am. Chem. Soc^., 22/ 1119, (l94R)
4. C. G. Swain, et al ibid. 70, 2989, (1348)
5. C. G-. Swain, et al.. ibid. 72, 4573, (1950)
6. T. M. Lowry, J. Chem. 3oc, 127, 1.783, (1925)
7. T. M. Lowry, et_al., Jbid, 127. 2883, (1925)
8 . T . !i . Lorry, et al ., ibid . , 2539, ( 1 927 )
9. C. G-. Swain and J. F. Brown, J. Am. Chem. Soc . , 74, 2534, (1952)
10. C. G-. S^ain and J. F. Brown, J. Am. Chem. 3oc . 74, °538 (1952)
11. D. Keilln and E. F. Hartree, Biochem. J., 50, 341, (1952)
ItA
SOME METHODS OF STEPWISE PEPTIDE DEGRADATION Reported by N. W. Kalenda January S, 195
p, %
Peptides are relatively low molecular weight oolyamides formet from a- amino acids* Proteins, the most important of all organic compounds, are essentially large peptides. In order to analyse th< structures of compounds belonging to this latter class of sub- stances, methods must be used which will permit the determination of the exact sequence of the a— amino acids*
Two general procedures for determining the structure of peptides sre available. One procedure1 Involves the cleavage of the peptide chain into smaller fragments^, the separation and id- entification of these fragments., ana the reconstruction of the ch chain from the Information obtained. This method has been e to lo I "ti- ed with striking success by Sanger In his work on insulin »3 The other procedure Involves the stepwise removal of amino acids from the peptide chain. Most of the methods In this procedure make use of the driving force of a ring closure to eliminate the terminal amino acid-
For the stepwise procedure., the degradation_may be directed at either the free a- amino or the free car'boxyl3 7 end of the molecule* Methods involving the former will be considered in this seminar.
The methods to be considered have as their ultimate goal the degradation of naturally occurring proteins. At ore sent, however, the methods are being tested on simpler compounds — synthetic peptides.
The earliest method developed8' 9 Involves the treatment of the peptide with phenyliaocyanate to form a phenylureide (I) and the cleavage of the phenylureide to a hydantoin (II) and a peotide residue. The hydantoins are easily separated and identified by elementary analysis and by comparison with authentic samples. 5?he yields are good; the trioeptide alanylglycylleucine coupled in an almost quantitative yield and gave a hydantoin in a yield of 96%»
9 C6H8NC0 2 i? NPOH H3N-CH-C-NH- > C6H5NHC-NHCH-C1NH— " ^ > H3N
R R : I +
I reoeat
0 ' '
C6H5-N
II
The general utility of this method is limited by the fact that the bierorous hydrolysis procedure splits the peptide bonds to a small but definite extent.
/**•
-2-
In order to overcome the chief obstacle to the ohenyliso- cyanate method - a small amount of hydrolysis of other peptide linkages during the formation of (II) - Edman employed phenyl- isothiocyanate.10' ^ The thioureido derivative (III) forms a hydantoin (TV) under milder conditions. The reaction is extremely rapid, even at room temperature, and is unaccompanied by the cleavage of other peptide linkages. The thiohydantoin is cleaved by alkaline hydrolysis and the resulting amino acid is determined by paper-strip chromatography. The method has been made micro- analytical and requires only ^bout 10 mg„ of amino acid per peptide bond.
$ Ce.K5N0S % ? //° CH3N02.
H2N-CH-C-NH---r-— -~^ CeHBNHC-NHCH-CfNH ~7 H3N —
R
S
tJeHs~3fi. {'
i N& OH
III I repeat N
NH
y RCH-C03F
K R IV
Deviations occur in the hydrolysis of the hydantoins formed from arginine, asoaragaine, and tryotoohane . Arginine gives rise to two nin^.ydr in- "Positive compounds, one being ornithine and the other being unidentified; asparagine gives asoartic acid; tryptophane gives two soots, the more intense of >Thich is tryptophane. Pre- liminary attempts to prepare ohenylthiohydantoins from the amino acids serine, threonine, and cystine show that comoounds are ob- tained which correspond to the obenylthiohydantoins minus the elements of w^ter in 1he cases of serine and threonine and hydrogen sulfide in the case of cystine;13 these cases are being investigat- ed further.
A method investigated by Levy1 3 is the treatment of a peotide with carbon disulfide and barium hydroxide and the cleavage of the product (v) to a 2-thio-2, 5-thiazolidinedione (VI) . Levy has tried this method on a few di- and trioeptides. The possibility of using this method on higher polypeptides is being investigated.
S Ba(0K)3, CS3
RCH- C- NH - — — ~>
L3 N3fltn%> ir&-
,9
R-CH-CONH 0-0-
NH-C-3- S v
++ HC1
Ba -> H3N —
[ repea^
H
0
v
VI
H
-3-
Khorana14'15 has found that a peotide can be treated with methyl ethylxanthate to form an N(thionocarbethoxy ) peptide (VIl) which then is extracted from the reaction mixture and cleaved to form a 2, 5-thiazolidinedione (VIIl). The tMazolidinedione is cleaved to sive the amino acid which' is identified by oaper- strip chromatography. Experiments with some di- and trioeo tides have proved promising and the yields of (VTIl) obtained from them were quantitative .
SO SO HI
C2H50C-SCH3 + K3N-CH-C-NH- ^'l' > CH33H + C2H50C-NKCH-C-NH ~ 3 ' 24-48 hr: 4
VII
HCl'NHs + HN (-R l) alkali.
2) acid '/ RCK-C02H + COS repeat. A i. lu
The method aooears to offer some advantages over those developed by Levy and by Edman. Levy carried out the formation of intermed- iate (V) and cleavage to (VI) in the same solution, thus risking contamination of the degraded oeotide with the original one, while Khorana extracted his intermediate (VII ) from the reaction mixture before cleaving it. In Erdman's method, carefully controlled conditions are necessary.
A method which has oroved very promising has been developed by Holley and Holley.16 The reagent 4-carbomethoxy-2-nitrof luoro- benzene (l%) reacts with peptides to form an N(4-oarbomethoxy-2- nitroPhenyl) Peptide (X); the nltro group of (X) is reduced catalytically and the end amino acid splits off with rine closure to form a dihydroauinoxalone (XI) . The average yield of (XI) oer amino acid residue is 84^. The dihydroauinoxalone s are crystalline compounds ana" are identified by comparison«with samples prepared independently from (IX) and authentic amino acids.
o o o t o H
CH3od^r ~^>~F + H3N-CH-C-NH- -» OHgOO^^" ""^NHCH-G-NH- -> \^=^< r X— .< r Pt oxide
N03 N03
IX X
S . . jj 5 hr. at 25° $ / .
CH30-C^<^ ^wNH-CH-C-NH <-— : "Tze) CHaO-0-V^ x\_NH
^y — ./^ or 15 min. at 70o/ 3 ^X s \ R
NH2 XI HN->H
0 + H3N —
reoea
%
-4-
Problems which still remain to be investigated are (i) modifi- cation of the method for peptides containing cysteine, cystine, or methionine to take care of catalyst poisoning and (?) side re- acti ns occurring "between (IX) and functional grouos other than the terminal a-amino grouo.
BIBLIOGRAPHY
W. Fox, Adv. Protein Chem. 2, 155 (1945).
Sanger and H. Tupo, Biochem. J. 49, 46-3 (l95l).
Schisok and W. Kumpf, Z. Physiol. Chem. 154, 125 (1926).
Watson and 3. G-. Waley, J. Chem. Soc, 2394 (l95l).
Bergmann, Science 79, 439 (1934).
Beremann and L. Zervas, J. Biol. Chem. 113. ,341 (1936).
G-. Khorana, J. Chem. Soc, 2081 (1952; .
Bergmann, A. Miekeley, and E. Kann, Ann. 458, 56 (1927).
Abderhalden and H. Brockmann, Biochem. Z. 225, 386 (1930).
Edman, Arch. Biochem. 22, 475 (1949).
Edman, Acta Chem. Scand. 4. 283 (1950).
Edman, ibid., 4, 277 (1950J .
L. Levy, J. Chem. Soc, 404 (1950).
G-. Khorana, Chemistry and Industry, 129 (l95l).
W. Kenner and H. G-. Khorana, J. Chem. Soc, 2076 (1952).
W. Holley and A. D. Holley, J. Am. Chem. Soc 22, 5445 (1SG0.
1. |
S. |
2. |
F. |
3. |
P. |
4. |
J. |
5. |
M. |
6. |
M. |
7. |
H. |
8. |
M. |
9. |
E. |
10. |
P. |
11. |
P. |
12. |
P. |
13. |
A. |
14. |
H. |
15. |
G. |
16. |
R. |
PHOSPHATE ESTERS OF NUCLEOSIDES Reported by James 0. Kauer January 9,, 1953
Nucleosides are molecules composed of a monosaccharide linked "by a glyeosidic bond to a nitrogen atom of a nitrogenous base (generally a purine or pyrimidine derivative) „ The phosphate esters of these compounds are frequently called nucleotides* This seminar will deal primarily with the recent work of A. R0 Todd and coworkers at Cambridge University.
Nucleotides ere found in ell living cells* They form complex polymeric structures called nucleic acids in which the individual nucleosides are linked by esterif Ication *Tith phosohoric ecid. Nucleotides p]so form part of the structure of many coenzymes whic1 play an Important Part in cell metabolism*
A typical nucleotide is deoicted beloT-r. ^"2 adenylic acid (i)
„a or
9-pdenine-5T-nhosoho-p-D-ribo-
1 rV- f-
! 11
(i furanoslde
I' A* y) . i — o — i
N ^NQ
a j 9 OH OH
0 adenosine~-5f-phosphate
C -4 (- C-CH3-0-P-0H
H H H H OH
Early work by Todd and others was directed toward synthesis of the nitrogenous bases and the nucleosides.. Recent T-rork has dealt with phosphorylation of the nucleosides and linking of the resulting e seer a through their ohosphate groups.
Phosphorus Pxy-acld Chem.1 str.y
Phosphorus forms two oxy- acids whose esters are of importance in the synthesis of nucleotides.1
RO^ -0 RO^ RO^ J3
P^ >0H {= >*
RO^ OR RO RO H
Phosphates Phosphites
The most useful syntheses of these esters are reaction of phosphste salts T'?ith organic halldes ^nd reaction of phosphorus halides with alcohols.3
Acyl halides reart with Phosphate salts to form mixed an- hydrides. Pyrophosphates (diphosphates or phosphoric anhydrides) can be formed by reacting a phosphorous oxyhalide with a silver salt of phosphoric acid.
/^
0
* -
R_O-F-0
i
OR
-2r
As
+
0
R- 0-3- CI OR
>
chlorophosphonate
2 o
r_0-P-0-P-C-R | I
OR OR
py ropho spha te
Halogenatlon of the phosphites yields phosohonates which Todd found to be very valuable ohosphorylating agents.3'3
R0^ J> RO' VH
S0gCl3
?
or
N-chloro- succlnimide
R<V
RO
0
i
N51
RT0H
tert base
■>
RO <D |
|
RO^ |
V0R» |
Although S03Cl3 readily chlorinates the phosphite, the hydrogen chloride produced tends to rupture the nucleoside residue. For this reason, N-chlorosuccinimide givea much better results.4
To prevent the formation of byproducts Todd found it desirabl' to protect all but one of the free phosphate hydroxyls with benzyl groups which could be removed by hydrolysis or hyd rose no lysis. Hydroa-enolysis is preferable because nucleotides are readily de- stroyed by hydrolysis. He also found that tertiary bsses or lithium chloride would selectively remove only one benzyl group. An S 2 mechanism was proposed.
RO
\i
.0
0CH3G
»0-CF3j6
CI Li
+
cello^olve
-)
j6C^T30
RV
+ j6CH3Cl
0 Li
Structure and Synthesis of Nucleotides
All three of the possible monophosphate esters of adenosine have been isolated from natural sources. The 5T-phosphate can be differentiated from the others by degradative studies or by periodate oxidation.7 This latter technique, developed by Todd was also useful in verifying the fact that ribose nucleosides are furanosides and not pyra.nosides. 8
R
3' -
3»_.0H 4' OH
5' __
R
JOH I
0
2 Moles. NaIO-4 '
HQ-
CKO
C?H0 i
+HC00H
h3c i
nl
■•OH •OH
1 Mole
OlTaTO-:
>
pyranoslde
CH30H
furanoside
6ho
CHO HC
0
CHoOH
The 21- and ^'-monophosphates are stable to periodate while the 5!-phosPhate is oxidized.
first
Adenylic acid and adenosine diphosphate (ADP) were among the nucleotides to be synthesized in reasonable yield.9'10'11
-5-
NH.
0
acetone ZnCl2
OH OH I I
C — C—C — C— CH3OH
H H H H adenosine (II)
>
(III)
= R
C-C — C— C ~CH£-
H H H H
2J 3-lsooropylidene adenosine
^H20^p^0
j6CH20'
+ ROH CI (III)
dibenzyl chlorophosphonate
l) LiC;
/>CHaOx ^OR
(IV)
H0>
H2/PdO x EtOH aq. H6'
dil H+
'OR
(IV) celtosolvS AgV
2) AgN03
^CH30 0 ^ +- >^ 1) dibenzyl
OR chlorophosphonate,
2) Ha/PdO
P 0
^ R_C-P-0-P-0H
OH OH
SjS^lsopropylidene. adenosine— 5-pyrophosphate
The 21 and 3d- hydroxy la of adenosine (II) are protected "by reaction with acetone. The 5-1 hydroxy 1 Is ohosnhorylated with di- benzyl chloron^osohonate, Hydro^enolysis removes the benzyl ' groups, and hydrolysis removes the 2) S^isooronylidene residue to produce adenosine— 5* -Phosphate (l). Monodebenzylation of (IV) followed by phosphorylation produces the tribenzyl pyrophosphate. Removal of the Protecting e-rouns selves adenosine-o-pyroPbosohate (ADP).
Because of some inaccurate laboratory work,14 it was believed for some time that the 2-, ,V-, nnd 5-1 Phosphates had been un- ambiguously synthesized. 7» 13 It was believed that benzaldehyde formed a cyclic aoetal with the <5& and 5-1 hydroxyls of rlbose nucleosides. (Benzaldehyde is known to produce 1,3-cyclic acetals with many carbohydrates.)15 The product of phosphorylation of this supposed 3] o^-benzylidene derivative was assumed to be the 2J-phosphate . Later work showed that the 2} o^-benzylidene nucleo- side is produced, and that Phosphorylation gives the o^-phosohate.
Carter and Cohn16 isolated two Isomeric adenosine Phosphates (a) and (b) from nucleic acid hydrolysates which were shown to equilibriate in acid solution.7 Both were oxidized by periodate. It was proposed that these two acids were the 2X and 3*1 phosphates, and that the interconversion was analogous to that observed in the monophosphates of glycerol*
.4-
/ is
OH ,O^P-OH
0
*n
~7
— 0
0 OH
^v /
P
A
o b
O OH
r_C —| 1 C-CHsOH
H H K H ;R-H j~
i H H H
<v) £fc=r-
C- OH 2 OH H
R-C H
__. 0
OH i
HO-P-O
I
OH 0
i i —
H H
R=adenine
~0-CH3OH
The cyclic intermediate (V) was later synthesized and was shown to "be hydrolyzed into the observed mixture of (a) and (b). To date no reliable general method for determining which isomer is the 2- or (^phosphate has been worked out. It has recently been reported that by a study of the solution densities and dissociatior constants of the cy tidy lie acids (a) and (b) the structure cytidine-3-phosohate can be assigned to the (b) isomer. These methods are dependent on the variations in physical constants which take place as the distance between charged grouos of a zwitterion increases.19
Poly nuc le o t ide s
Todd has recently developed a synthetic route which should lead to a general method of synthesizing nucleotides.17'20
0
R_0_P_C1 +
0
Aff 0-P-OR'
0CK3j6
*
0CH3j6
(VI)
R and Rf are nucleoside residues.
0
• 0— P-
OR'
OCHa/6 0CP3j6
Until very recently no nucleoside chloroohosphonatep had been synthesized. A synthesis for nucleoside benzyl phosphites and a chlorination process were required. It was found that mixed phosohoric-Dhosohorous acid anhydrides would react with nucleosides to produce the phosphite esters. Chlorination with N-chloro- succinimide4 led to the desired nucleoside benzyl chloroohosohonate (VI).
PCI 3 + /)CH30H
RO ^0 0CH3O VC1
VI
II
<r
tfOH.q^O
j6CH30y XH
0
4
v
/ /
N-Cl
0
R=nucleoside residue
^tgN-HClx
j6CHa0
RO JD XP* H
P^ 00' N
+ 0
OH
ROH
£CH30 0
it
H^ OH
\
j6o^
o
P-Cl
base
pCH30 ^ ft Op
XP__0— P' h' N0j6
Hixed Anhydride
-5-
Todd and coworkers have suggested a mechanism for the hydro- lysis of nucleic acids which is based on a cyclic Intermediate similar to that involved in the ecmilibriation of the (a) and (b) forms of adenylic acid.31>ss
s' C-OH
C-0V _0 I P*
C OH 0-
C-0N J3
I >»'
C-0 - Q
\
C-
I
C-
I
-C
c-
i
n w
I
■OK
0 5.O
>
OH x0-
C— OH I C— 0
t
c
\
•o.
.0-
^
0
C-0
c— o^-
OH C-O-P-
C-0H^H
b' C-OH
4,
J
C— OH TT I ?H
C—O-P -
I OH
C-OH
3T,5' linked poly nucleotide
cyclic
intermediate
0
+ •-
Mixture of 2- and 31 Phosphates
If rupture of only one bond attached to ohos^horus in the inter- mediate is assumed, it can be seen that the rupture of either the 2-1 or 3- C-O-P bond would lead only to starting material or an isomer. Det>olymerizatlon would take place only by breakage of the 5-1 bond.
BIBLIOGRAPHY
1. 2. 3. 4. 5. 6. 7. 8.
9.
10.
11. 12. 13. 14. 15. 16. 17. 18.
19. 20.
21.
22.
Atherton, Quart. Revs. London, 3, 146 (1949).
Atherton, Onenshaw, and
Todd, ibid. . 647 (1946).
Kenne^, Todd, and Weymouth,
Baddiley, Clark, Miohalski, and
Clark and Todd, ibid.. 2030 (i960) .
:oda/j/Chem. Soc. , 38? (l945)
ibid., 3675 (1952) .
Todd, ibid. t 815 (1949)
Brown, Haynes, and Todd, Lyth,ioe and Todd, ibid. . B add 1 ley and Todd, ibid. Baddiley, Michel son, and
ibid. , 3299 (1950) 592 (1944).
648 ^1947). Todd, ibid.
582 (1949). Michelson and Todd. 'ibid.. 2487 (1949). Levene and Tipson, J. Biol. Chem. 121, 131 (1937). Michelson and Todd, J. Chem. Soc, 2476 (1949). Gulland and Overend, Ibid.. 1380 (1948). Haworth and Hirst, Ann. Rev. Biochem. 5, 82 (1936). Carter and Conn, Fed. Proc. 8, 190 (1949) . Corby, Kenner, and Todd, J. Chem. Soc, 1234 (1952). Brown, Magrath, and Todd, lb id . . 2708 (1952). Cavalieri, J. Am. Chem. Soc, 74, 5804 (1952) . Corby, Kenner, and Todd, J. Chem. Soc,. 3669 (1952). Brown and Todd, ibid., 52 (1952). Elmore and Todd. ibid.. 3681 (1952).
/ -'Z
TRIALKYL OXONIUM SALTS Reported by Robert J. Lokken
January 16, 195 3
Synthesis
R30+X"
Trialkyl oxonium salts of the type R3CfX such as (II ) are such strong alkylating agents that all attempts to prepare them by the alky lat ion of ethers with alkyl halides or dialkyl sulfates have failed. In the alkylation of the cyclic ether, 2, 6-dimethyl- pyrone (i), with methyl iodide a compound was obtained which was "thought to be a trialkyl oxonium salt with structure (II). How- ever, Baeyer showed that it was not a true trialkyl oxonium salt and that its structure was (III).1 On treating the salt with aqueous ammonium carbonate solution he obtained 4-methoxylutidine (IV) . This showed that the methyl group was not attached to the ring oxygen atom, but to the carbonyl oxygen.
CH.
i 0 /CH3
OH,
CH3
I® '
0. CH.
•<5>
\
+ CHaI
^/ v-
CH.
0 1
\y
0
11
0
V
v
0CH3
<=>
CH,
III
I<3
CH3
OH
+ (NH4)3C0,
Fa0
>
0CH<
t;
CH-
OCH,
IV
In 1937 Meerwein discovered oxonium salts while studying the etherates with epoxides. Earli alcohol-boron trifluoride comple ethers can be considered to be a that the reactions of boron trif gous to those of acid anhydrides, with epoxides to form esters of trifluoride etherates would be e
a method for synthesizing trialkyl reaction of boron trifluoride er he and Pannwitz had shown that xes are strongly acidic. Since Icohol anhydrides it was thought luorlde etherates should be analo —
Then just as anhydrides react the corresponding glycols, boron xpected to produce diethers.
When epichlorohydrin was added to an ether solution of boron trifluoride two products were formed, f^ne was an ether-insoluble crystalline solid which was assigned structure (V). The other, an ether- soluble solid, was shown to be (Vl).
,p_
C1CF3 - CH.
X
.0+4
OH,
(A)
CaH.fr* <->
"0 - BFa + 2
CH
2aB
^ CH-oi
ClCHg~ CH-O-fe B ! j C3HB-0-GH3
VI
s
C2H5
3 (C3H5)30 BF^
V
The structure of (V) was assigned on the basis of its carbon- hydrogen analysis, salt-like properties, and chemical behavior, decomposed on heating to 100° to give ether behaved as a strong ethylating agent.
(CaHJsO^BF? *=£ CH3-CH3F +
and ethyl fluoride , and
C3H5
C3H5
f+> H
0 - BF3
Meerwein found that the decomposition of the salt was re- versible and prepared trie thy 1 oxonium fluoroborate by letting boron trifluoride etherate, ethyl fluoride, and ether stand in a sealed tube at room temperature for three to four weeks. He also prepared dimethylethyloxonium fluoroborate from boron trifluoride dimethyl etherate and ethyl fluoride.
The assignment of structure (VI ) was based on its analysis and on its hydrolysis to (VII).
r
1
C1CH3 - CH
I CaHB-0- CH3
-.4-
B
Ha0
CH3 ■ ™ CH— — CH3
CI
OH 0-C3Hs
is - VT
H3B03 VII
Recently a new synthesis of trialkyl oxonium salts was attempt? ed by Klaees and Meuresch in G-ermany.4 Meerwein had used the strong formation tendency of the fluoroborate ion to overcome the resistance of ethers to alkylation. Klages and Meuresch thought that direct alkylation might be accomplished by a reaction the mechanism of which was different from that of the conventional type of alkylation. It was their theory that aliphatic diazo com- pounds should be capable of alkylating any compound which was sufficiently acidic. Accordingly, they tried to alkylate dialkyl oxonium salts (addition products of ethers and acids) with ali- phatic di^zo compounds.
Instead of the expected trialkyl oxonium halide, the alkyl halide (VIII) corresponding to the diazo compound was obtained. There are two possible Paths to this alkyl halide as shown in the equation.
-3-
HC1 + (CsH5)20t?[(C3F5)3#ia OlG QH3-CH t£^[ (C3H5) 3-0-gHs-CH3]cf
CK3-CH~N3 CK3-CW3C1 > Intramolecular 1 ^ * JTky lotion
VIII
These results indicate that the dialkyl oxonium salt which is used must be one of a complex acid such «s a hexachloroantimonate or a fluoroborate, since these ether^tes have pr^ccically no tend- ency to dissociate into ether and an acid, HX„ In addition, these complex ions are much poorer nucleophilic agents than is chloride ion and will not attack the oxonium ion to give intramolecular alkylation.
(C3H4)30- SbOls —1521^ C(0SHs)SCTH] SbCl^
_CHj:0H-N8 L(C3H5)3^] SbCli3
Mechanism
The mechanism of the reaction involved in Meerwein*s synthesis has not been elucidated. The following scheme is proposed as a likely possibility. The key step is apparently the opening of the epoxide >ring.
i-A
C3H5 _. 01CHS - OHH 4-) ClGHs-CH-O-BFg
^ 01 + | >0 - DF3 -}C3HS j
C3HS G^3 ^Oj CF;
M
This is followed by the equilibrium below.
H ©
Cl CH3-CH-0-BF3 ClCH2-CH-0-BE
C3HS I v C3H5e + I
C3He + ^0 — CH3 ^=± ^0-C3H5 C3H5-0-CH3
^01 08HB/+» C3H6
C3H5
This equilibrium is driven to the right because of the excess of diethyl ether and because of the subsequent removal of the tri- ethyloxonium ion as its insoluble fluoroborate. A further series of equilibrations redistributes the alkoxide and fluoride ions around bx>ron.
C1CH3 - CH - 0 hi BF3 v
I 7— >
C3H5-0-CH3 + BF3
+ BF4V C3H5 - 0 - CHS
C1CH3 - CH - &\— B ^
3
The removal of'BF^as its insoluble triethyloxonium salt shifts this equilibrium to the right also.
■
; ■ ,. «•'
The reaction of diethyl ether with BF3 shown "below (which is known not to proceed) is formally analogous to that shown in equation (a).
C3HS _ C3H5(+*-> v CSH5© e
^01 + ^0-BF3 ' ^0-C3HB + C3H5-0-BF3
CsHr 03Kf GSUS^
The success of the reaction in equation (a) is due to the driving force provided by the opening of the strained epoxide ring.
Chemical Reactions
Tertiary oxonium salts are the strongest alkylating agents known. That is to Say, it is very easy for even a weakly nucleo- Dhilic agent to attack one of the alkyl groups replacing an ether molecule. They undergo a replacement reaction with water to form an ether and an alcohol.
® m
Hs0 + CSHS - OfC3Fs)3 ^ H3crlC3H5 + 0(C3HB)
!f\ H0-C3H5
7
Alcohols, and even such weak "bases as phenols, are readily alkylated by trialkyl oxonium salts to form ethers; and esters are formed from organic acids.
R-OH + R'30"BF4 } R-0-R» + R' 30
R-(r + R«30 BF4 ^ R-C' + Rr30
X0H ^0-R*
Alkali phenolates and alkali salts of acids are alkylated in aqueous alkaline solution even more readily than are the free phenols or acids.
Sulfur-, nitrogen-, or oxygen-containing bases are alkylated to form sulfonium, ammonium, or new oxonium salts. Thus pyridine is converted to ethyl pyridinium fluoroborate and diethyl sulfide to trlethyl sulfonium fluoroborate in practically quantitative yield.
-5-
#M©
(CsH5)30^ BF
->
i^\ ®
BF.
+ (CSH5)
3n5 / 3
C3H5
(C2H5)3(^BfP + (C3H5)3S > (03Ee)3PBFp + (C2H5)2°
The tremendous alkylating power of the trialkyl oxonium salts is demonstrated by the fact that they can be used to alkylate coumarin and saturated and unsaturated ketones, while attempts to alkylate these compounds with the conventional alkylating agents have been unsuccessful.
^A
+ (C3H5)30 BF^
~>
^ ^UO-C3HB * BFp
m _ &
+ (C3H5)3Ov- BF
*
BIBLIOGRAPHY
1. A. Baeyer, Ber., 43, 2337 (1910).
2. A. Baeyer and V. Villiger, Ber., 34, 2681 (l90l).
3. Collie and Tickle, J. Chem. Soc, 75, 710 (1899).
4. Klages and Meuresch, Chem. Ber., 85, 863 (1952) .
5. H. Meerweln and coworkers, J. orakt. Chem., 147, 257 (1937)
AMINATIONS WITH ALKALI AMIDES Reported by Thomas R. Moore January 16, 1953
The react i potassium amid for some time. when allowed t anilines1 and placing an ami the ring,3 Bu 4,6-diaminodib they came aero
on of monohalobenzenes with sodium amide or
e in liquid ammonia to give anilines has been known
1 It has also been known that the monohalobenzenes
react with alkali dialkyl amides give N- substitute:' that pyridine and quinoline react with alkali amides, no group on the carbon adjacent to the nitrogen in L" Whin G-ilrnan and coworkers attempted to prepare snaofuran from the corresponding diiodo compound the following apparently anomolous reaction.:3
63
) S\
^V—^>
NaNHgx
"TO '
+ a diaminodibenzofuran
The diamino compound was proven not to be the 3,7 or 3,8, and was believed to be the 3,6.
It was then found that 4-iodo- or 4-bromodibenzofuran gave 3-aminodibenzofuran, while the 2-bromo compound gave the 2-amino- dibenzofuran. It was soon found that 4-iododibenzothiophene showed a similar rearrangement.4
Other compounds were then studied in an attempt to determine the scope of this rearrangement. This study unearthed the follow- ing reactions:
NaNH5
NH,
*
CH3 0
III)
I
X
CH3 0
LiH(GsK5)3 TTthe-r 7
X = I, Br, CI, F
•3-
IV)
H 0
^
V
Br
LiN(C3H5)
3n5/ 3
Ether
*
N(C3H5)S
KNK;
NH,
>
L1N(C3HS)3 "Ether
^
N(C2H5)
X = I, Br, 01, F
In reaction V it was noticed that ct-fluoronaphthalene gave un- rearranged product, unlike the other halogens. In reaction VI fluorine behaved as the other halogens did. It was also found that the p-balon^Dhthalenes gave f?-naohthylamine with only small traces of a product.7 a~ Ha loquino lines show no rearrangement.8
The same rearrangement was found to take place when the halogen is ortho to an N- substituted amino group9 or a trifluoromethyl group.10 That rearrangement from the Para as well as from the ortho position occurs was proven by the following exoeri— ments.: lx> ^
VII)
L1N(03H5)5 Ether '
(C6H5 ) 3 Si
LiN[CH3) ITther '
+ (some para-) N(C3H5)2
(C6H5)3 Si
N(CH3)
fcriflp • M
,r. i
-3-
The order of reactivity of halogens in such reactions apoears to be Br>I"^Cl.13 Oilman and co-workers have found that many alkali diamides will react, lithium di-n-butyl amide giving better yields than lithium diethyl amide. Lithium oloeridide and morohollde have also been used.6
Benkeser and Buting have attempted to discover the mode of formation of the products by a study of bromoanisoles containing a third substituent on the ring. Their conclusions are as follows:
1) The fact that 3-methyl-2-bromoanisole gives no product with sodium amide shows that the amino group goes to the position ortho to that Ox the halogen, not Para. This is confirmed by the fact that 6- me thy I- 2-bromoanl sole gives a 30^ yield of l-methyl-4- aminoanisole but no l-methyl-2-aminoanisole.
2) The fact that 4-trif luoromethyl-2-bromoanlsole gives un- rearranged product shows that the trif luoromethyl is stronger in orienting power than the methoxyl is in rearrangement-producing power,
3) The fact that 2-bromo~4-methylanisole gives over 50^ yield of 3-amino-4-me:;bylanisole shows that the entrance of the amino group is not hindered by the presence of an ortho methyl group.
REFERENCES
1. |
F. |
2. |
M. |
3. |
H. |
4. |
H. |
5. |
H. |
6. |
H. |
7. |
R. |
8. |
N. |
9. |
H. |
10. |
R. |
11. |
H. |
12. |
H. |
13. |
F. |
14. |
R. |
15. |
C. |
16. |
J. |
Bergstrom and W. Fernellus, Chem. Rev., 20, 437 (1937),
T. Leffler, Organic Reactions, Vol. I, 91 (1942).
Gil-nan and S. Avakian, J. Am. Chem. Soce, 67, 349 (1945).
Oilman and J. Nobis, Ibid., 67, 1479 (194577
Oilman et si., ibid.. 69, 2106 (1945).
Oilman and R. Kyle, Ibid., 74, 3027 (1952) .
Urner and F. Bergstrom, ibid. , 67, 2108 (1945).
Luthy, F. Bergstrom, H, Mosher, ibid.. 71, 1109 (1949).
Oilman, R. Kyle, R. Benkeser, ib id . , 68. 143 (1946).
Benkeser and R. Severson, ibid., 71, 3838 (1949).
Oilman and R. Kyle, ibid.. 70, 3945 (1948).
Oilman and H. Melvin, ibid., 72, 995 (1950).
Bergstrom and C. Homing, J. Org . Chem., 38, 254 (1946).
Benkeser and W. Buting, J. Am. Chem. Soc, 74, 3011 (1952)
Horning and F. Bergstrom, ibid. , 67, 2110 0l945).
Bunnett and R. Zahler, Chem. Rev., 49, 273 (1951).
CrRISEOFtJLVIN Reported by P. D. Thomas January 16, 1953
In 1938, Oxford and co-workers1 Isolated a metabolic product present In the mycelium of Penlollllum G-riseofulvum Dlerckx which they named griseofulvin. It was found to be a colorless neutral compound, C17H170QC1, m.p. 220°, [ d J 5790+ 354°, containing three methoxyl grouos. Subsequently, it was isolated from P. janczewskli [= P. nigricans] and its unique biological activity on moulds noted by Brian2'3 and McGowan4 who originally called it "curling factor" before the identity with griseofulvin was established.5'6
Oxford et al.1 noted that griseofulvin gave no color with FeCl3 and contained no free -OH or -COOK grouos. The ore se nee of a carbonyl grouo was established since a crystalline mono-oxime was readily obtained. They also noted that griseofulvin on acid hydrolysis afforded griseofulvlc acid, C16His06Cl, [a J 5461 + 50§ a monobasic acid containing two methoxyl groups and giving a feeble color with FeCl3. Hydrolysis of griseofulvin or further hydrolysis of griseofulvlc acid in 0.5 N NaOH yielded norgriseo- fulvic acid, C1SH1306C1, [a] 5461 + 609°, a dibasic acid contalntag only one methoxyl group, together with decarboxygriseofulvlc acid C15H15O4CI, [al 5461 - 31°, an insoluble neutral compound con- taining two methoxyl groups, giving no color with FeCl3, and derived from griseofulvlc acid by loss of one mole of C03. De- carboxygriseofulvlc acid was found to be stable to acid hydrolysis, hence they concluded griseofulvin contained only one -C00CH3 group and that the second acidic grouo in norgriseofulvic acid was a phenolic -OH group.
On oxidation of griseofulvin with KMn04 in acetone at room temperature two degradation products (i) and (il) were obtained.1
C17H170SC1 - KMn°4 ^ 1 COOW + C14H1507C1
CHa0
II
II, Ci4Hi507Cl, [a] 5790 -24°, w»8 a monobasic acid which gf^ve no color with FeCl3. It was shown to contain two -0CH3 groups, a C=0 group which cannot be a methyl ketone since it gave no iodo- form with alkaline iodine, and a tertiary OH group. On treatment with acetic anhydride in pyridine it did not give an acetate but the elements of water were eliminated to give a neutral substance C14Hl306Cl. The original substance, C14Hls07Cl, is therefore probably a y -OH acid of the form:
R
3
-C-C-C-COOH Ri,Rs^.H
-2-
Since II on further oxidation gave I, the structure XI- a was as- signed to II.
G-riseofulvin, on KOH fusion gave orcinol (Hi) which was con- cluded to be formed from a different Dart of the molecule than I.
Oxford and co-workers1 suggested the tentative structure IV which explained many of the experimental facts. This structure was at first supported, with slight modification (v) , by Grove and McGowan5 but it was later7 rejected by them as incompatible with the ultraviolet and infrared absorption spectra of griseofulvin and its derivatives; they7 concluded from the available spectro- scopic and chemical evidence that: (a) griseofulvin contained the partial structure VI and (b) -C00CF3 and -COOH groupings were ab- sent in griseofulvin and grlseofulvic acid respectively, and the acidity of the latter compound was attributed to the enolic group- ing VII-a.
HO
J\*
CH3
-J<VA- oh ch3 o-'^y^-'k/
ci cm*
0
OHaC
III
rv
CH3
COOCH3
'f"V CK30 COOCH3
8~14
Very recently G-rove, MacMillan, Mulholland and Rogers' collaborated to revise the structure of griseofulvin and its deriv- atives. They repeated the work of Oxford1 and agreed with the major portion of it. However, further work was required before the structure could be definitely established. The ultraviolet absorption of griseofulvin showed that the a, ^-unsaturated ketone system postulated by Oxford et jLL*1 could not be conjugated with the aromatic nucleus in griseofulvin as in IV, Neither did the absorption of griseofulvin agree particularly well with that pre- dicted for V. Rather, in agreement with partial Structure VI, it was typical of a compound in which ohloroglucinol and carbonyl chromophores were conjugated. ls' x 7
These workers provided further chemical evidence for the in- adequacy of formulas IV and V. Thus, while griseofulvin on mild hydrolysis with aquous alcoholic alkali gives griseofulvic acid, one of the reduction products, tetrahydrodeoxygriseofulvin, is very resistant to hydrolysis and this stability could not be dismissed on grounds of steric hindrance . 1 » 9> 14 Further the fact that on slightly more vigorous hydrolysis with aqueous alkali griseofulvic acid gives as one of the products decarboxygriseofulvlc acid, can- not be taken as proving the presence of the carboxylic acid group. They also found that griseofulvic acid possessed marked stability toward acid hydrolysis and suggested that the C03 lost in alkaline hydrolysis is derived from a carboxyl group which appears as a re- sult of a molecular rearrangement under alkaline conditions.13
-3-
The difference in behavior of griseofulvin and tetrahydro— deoxygriseofulvin towards hydrolysis caused Grove and McGowan7 to suggest that griseofulvin contained the grouping Vll-b, a methyl enol ether of a 1,3 diketone; griseofulvic acid, VII-a. In tetra- hydrodeoxygriseofulvin this would become VIII- a in which the methoxyl group is attached to a saturated carbon atom and would then resist hydrolysis.
OR 0 R R' 0 0CH3
X-C=CK~C-Y X-CH CH3-CH-Y X-C CH=C Y
VII (a) R=H VIII (a) R=0CH3; R« = H IX
(b) R=CH3 (b) R=R» = OH
R=K; R« = OH R=R»=H
Oxford et al.1 observed that an isomer, m.p. 200-201 , [a] 5790 + 2 23° , of griseofulvin was formed mixed with an approximatel equal amount of griseofulvin, when griseofulvic acid or norgrisec— fulvic acid was methylated with dia^ome thane . It appeared likely on the above hypothesis that isogriseofulvln would Drove to be the Isomeric methyl ether (IX), although no evidence was offered at that time. Grove jet aJL.^ in their most recent work have shown that isogriseofulvin can be easily made and in good yield by the action of methanollchydrochlorlc acid on griseofulvin. Isogriseo- fulvin is easily hydrolysed by dilute aqueous-alcoholic alkali to griseofulvic acid, which has the same optical rotation and con- figuration as that prepared from griseofulvin; the isomerism is thus not connected with asymmetry around a particular carbon atom, but arises from the presence of a tautomeric system (XII) in griseofulvic acid. Isogriseofulvln also differs from griseofulvin in that it does not readily form derivates with ke tonic reagents.®
Reduction of griseofulvic acid13 provided further evidence against the presence of a carboxylic acid group. When AdamT s platinum oxide catalyst in acetic acid was used, two neutral non— lactonic alcohols, C16Hi906Cl (A) and Ci6Hi905Cl (b) were isolated together with small amounts of a neutral non-lactonic compound, Gi6Hig04Cl (C). If the acidic component of griseofulvic acid is VII-a, these compounds can be written with the partial structures VIII-b, VIII-c and VHI-d respectively. The reduction product B was oxidized by chromic acid to the corresponding ketone, Ci6Hi705Cl, indicating that the -OH is secondary. The ultraviolet absorption curve of A shows that the main chromophoric system, VI, of griseofulvin is unaffected by reduction and since the chemically unreactive carbonyl group of griseofulvin can be identified in A by its infrared spectrum (band at 1685 cm""1) it follows that this carbonyl group must be the one in the partial structure VI. This unreactive carbonyl group can also be identified in the spectra of reduction products B (band at 1682 cm"1) and C (band at 1695 cm""1); A and B show typical alcoholic -OH absorption (at 3401 and 3425 cm"1 respectively), which is absent in the spectrum of C.
From oxidative degradation10'11 it was concluded that griseo- fulvin possesses a benzenoid ring (A) and a hydroaromatic slx- membered ring (C) thereby confirming the views of Oxford ejt al.1*
— »^J—
The nature and orientation of ring A follows from the formation of 3-chloro-2-hydroxy-4, 6-dimethoxybenzolc acid (i) from griseofulvi' by oxidation with zinc permanganate in acetone — conditions which are considered to preclude rearrangement. The structure of I has been established by two unambiguous syntheses.11
The presence of a second slx-membered ring (C) in is indicated by oxidation with chromic oxide to 3-methoxy-2, toluquinone (Xj R = CH^,) &nd by formation of orcinol (ill) by fusion. Since the three methoxyl groups in griseofulvin apoear in
grrlseofulvir 5- (III) bv KOH
the oxidation products (l and X; R = CH3)
griseofulvin is not the methyl ester OR PCH3
Of a
- 0
it is evident that carboxylic acid.
0 0
1 !
X^-C-C^-C-Y
XII
XI
(a (b
R=OH R=H
All the carbon atoms in griseofulvin are accounted for in the oxidation products (l) and (X; R = CH3). That none of them is common to both rings A and C has been demonstrated by cleavage of griseofulvin,18 in significant yield, into the acid I and orcinol monomethyl ether by 2 N sodium ethoxide. A similary fission14 has been encountered in the alkaline hydrolysis of dihydrogriseo-
salicyllc acid (i) (derived from ring A) C). Moreover, formation of (i) under provides convincing chemical proof that there is a carbonyl group directly attached to the benzenoid ring (A) as suggested by spectroscopic evidence.
fulvin which yields the and m-cresol (from ring hydro lytic conditions18
The hydroaromatic ring (C) must contain the C-methyl group and the olefinlc double bond known to be present in griseofulvin. Moreover, if it is assumed that rearrangement does not occur in the formation of orcinol (ill) and its monomethyl ether two potent- ial hydroxyl groups must be located in the ?,5 Dositions with respect to the C-methyl group. Finally, ring C must contain the methyl ether sroup of the system VII-b; the most likely skeleton of ring C therefore appears to be XIII,
OCH.
OCH -°
X=
0
3 II
C OCH3
CH,0
XIII
XIV
0
OCHg J!
CH30
V
CI
\
0 I
CFa
OCR.
XV
XVI
The manner In which the two partial structures VI and XIII
in griseof .ilvin, the authors oropose, is unequivocably "by the forr at: on of the acids C14H1506C1 and Ci4H1507Cl shown to be Xi-t) and Xl-a respectively. Final proof the oxidation with periodic acid, in which 1 mol. of consumed and the acid XI- a is split quantitatively methylsuccinlc acid, identical with an authentic
are linked determined which were comes from periods to is into I and (+)
specimen. These acids, derived from griseofulvic acid by alkaline peroxide and by permanganate oxidation respectively, are sub- stituted coumaran-3-ones in which the carbonyl and oxygen ether bridges from partial structure VI are linked to the same carbon atom, adjacent to a C-methyl group. Union of partial structures VI and XIII in a like manner leads to the two spiran structures XIV and XV which are isomeric methyl ethers of the tautomeric enol XVI and are therefore considered to represent griseofulvin and isogriseofulvin although not necessarily respectively. It is con- sidered that XIV is more likely to give rise to 3-methoxy-S, 5-
on chromic oxide oxidation and therefore
Isogriseofulvin, on the other hand, does Cr03 and is therefore considered to be XV: structure might be expected to yield has
toluquinone (X; R = CH3) represents griseofulvin. not yield a quinone with the 0- quinone which this
not been isolated; it probably does not survive the oxidation.13
BIBLIOGRAPHY
1.
2.
3. 4. 5. 6.
7.
8.
9.
10.
11.
12.
13. 14. 15.
IS:
Soc, P. W.
J. c.
J. F.
P. W.
Soc. ,
J. F.
J. F.
M. A.
J. F.
Rogers,
J.~F.
M. A.
J. F. Grove~
H. Raistrick and P. Simonart,
H. G. Hemming, Trans. Brit.
and
29, 188 (1946).
A. E. Oxford, H. Raistrick and P. Simonart, Biochem. J.,
240 (1939).
P. W. Brian, P. J. Curtis
29, 173 (1946).
Brian, Ann. Bot., 13, 59 (1949).
MoGbwnn, Trans. Brit. My col. Soc,
Grove and J. C. McGowan, Nature, 160. 574 (1947).
Brian, P. J. Curtis and H. G. Hemming, Trans. Brit.
32, 30 (1949).
Grove and J. C. McGowan, Chem. and Ind. 647 (1949).
Grove, D. Ismay, J. MacMlllan, T. P. C. Mulholland
T. Rogers, Chem. and Ind. 219 (l95l).
Grove, J. MacMillan, T. P. C. Mulholland, J. Chem. Soc, 3949 (1952).
Grove, D. Ismay, J. MacMillan,
T. Rogers, ibid.. 3958 (1952).
J. MacMillan, T. P. C. Mulholland ibid., 3967 (1952).
F* CTr9Y?* J- MacMillan. T. P. C. Mulholland •ogers, ibid., 3977 (19525.
I* d" 2- gulholland, ibid., 3987
T. P. C. Mulholland TW. 3994
P- Haas ibid. 89. lB7^tl906 .
ilri&S£%SF' AcT? C^emi Scand. 4. 772o(l950). ,
A. Morton and Z. oawires. J. -Chnm. Snn. lA Afip (iqao)
33, My col
MycoL
&
S* . naas • IJind . A. Mi
and
and M. A. T.
T. P. C. Mulholland and
and J. Zeallcy, pnd M. A. T.
(1952). (1952).
CHEMISTRY 435 II SEMESTER 1952-53
Strained Homomorphs
Seymour Pomerantz, February 13 1
The Structure of Cytlsine
Blaine 0. Schoepfle, February 13 5
Photochemical Reactions of Diazomethane
David B . Kellom, February 20 8
Steric Control of Asymmetric Induction
Moses Passer, February 20 12
Synthesis of Phenanthrenes
C . W . Hinrnan, February 27 • 15
The Synthesis of Cantharidin
Elliott E. Ryder, February 27 20
Humulene
W. S. Anderson, March 6 23
/ New Reactions of p-Propiolactone
William S . Fr iedlander , March 6 28
11- Oxygenation of the Ring-C-Unsubstituted Steroid Nucleus
Howard J . Burke , March 13 «, 33
Syntheses of Long Chain Fatty Acids
John R . Demuth, March 13 38
Abnormal Reactions of Heterocyclic Grignard Reagents
G. W. Par shall, March 13 43
The Willgerodt Reaction
S. L. Jacobs, March 20 46
Recent Studies on the Decomposition of Benzoyl Peroxide
James C. Eauer, March 20... tm 51
The Reaction of ortho-Halobenzoic Acids with Nucleophilic Reagents
Harry J. Neumiller, March 20 55
Some Base Catalyzed Rearrangements
Y. Gust Hendrickson, March 27 59
Migration in the VJagner Rearrangement
Thomas R . Moore, March 27 64
Configuration Studies by Asymmetric Synthesis
Edwin J. Strojny, March 27 69
Some Polypheny 1 Derivatives of Nonmet all Id' Elements in Their Higher Valence States
M. J. Fletcher, April 10 73
Rearrangements of 9-Substituted Fluorenes
Richard L. Johnson, April 10 77
A -New Synthetic Approach to o-Hydroxy I Phenol Derivatives
William H. Lowden, April 10 81
Developments in Azulene Chemistry
Aldo J. Crovetti, April 17 85
Alkaline Decomposition of Hydrazine Derivatives
David M. Locke, April 17 90
New Syntheses of Pyrlmldines
Paul D . Thomas K April 17 . . 95
Oxidation of Indoles
Allan S . Hay , April 24 100
The Structure of the Aminopyridines
Norman W . Kalenda, April 24 104
Reactions of 1,1-Diarylethylenes
Robert J. Lokken, April 24.... 108
Participation of Neighboring Groups in Addition Reactions
Fabian T . Pang, May 1 ....... • 112
Basicity of Aromatic Hydrocarbons and the Isomerization of the Methyl Benzenes
Harry W . Johnson , Jr . , May 1 117
The Neber Rearrangement
Lewis I. Krimen, May 1 *.... 121
Photochemical Reactions8'14
Ruth J . Adams , May 8 125
Condensations Involving Esters
Leroy Whi taker, May 8 129
The Lederer-Manasse Reaction
P. Wiegert, May 8 132
A New Mechanism for the Oxidation of Glycols by Lead Tetraacetate
Joanne G. Arnheim, May 15 137
2 , 3-Pyrrolidinediones
Clayton T . Elston, May 15 142
Products of 0,-Phenylenediamines and Alloxan in Neutral Solution
Harold H . Hughart , May 15 147
Recent Syntheses of Thiazoles and Thiazolines Prom Aminonitriles
N. E. Bo jars, May 22 151
The Mechanism of the Sandmeyer Reaction
A. B. Galun, May 22 154
The Alleged Rupe Rearrangement
William P> Samuels, May 22 157
STRAINED HOMOMORPHS Reported by Seymour Pomerantz February 13, 1953
In 1942 Brown, Schleslnger, and Cardon11 first proposed that the study of molecular addition compounds should furnish a con- venient tool for the estimation of steric strains in related carbon compounds. It was believed that steric effects in ethane derivatives, for example, should reveal themselves in other ways than by restricted rotation. The repulsion of the two parts of the molecule should result in a weakening of the bond joining them. Measurement of this weakness in simole hydrocarbons cannot generally be effected, but a study of compounds with similar mole- cular dimensions (called homomorphs) can lead to an estimation of this strain.
The geometry of the bortn- nitrogen bond is almost Identical with that of the carbon- carbon bond. «The bond distances11 are 1.54 1 for carbon- to- carbon and 1.5R A for boron-to-nltrogen. Strains should be duplicated, but the effect of such strains should be considerably magnified by the comparative weakness of the donor-acceptor bond.
Spitzer and Pitzer14 have discussed this relationship be- tween strains in addition compounds and strains in hydrocarbons. It is apparent, however, that the concent need not be restricted to hydrocarbons. Replacement of one or more atoms or groups in the hydrocarbon by other atoms or groups of closely similar dimensions should result in the formation of molecules with widely different functional groups, but with closely related strains.
Examples of five series of homomorphs are given on page two (2).
Homomorphs of Di-t-butylmethane . Examples of these are shown in Fig. 1, An estimate of the strain is provided by com- paring the heats of dissociation for _t-butylamine-trimethylboron (l3. 0 kcalO , and for the corresponding n-but diamine derivative (18.4 kcalJ. The difference (5.4 kcal.) is attributed to steric strain in the t-butylamine-trimethylboron. It follows from the proposed thesis that a steric strain of this magnitude should be present in homomorphs of di-jt-butylme thane. The thesis is supported by the value of the steric strain in dl-_t-butylme thane, estimated from heats of combuslon to be 5.2 kcal. Cther con- firmatory evidence is found in t^e difficulty in preparing di-t- butylether (IX), the ease of solvolysis of neooentyldimethyl- carbinylchloride (VIII)1'5'7, the slow rate of reaction of' neopentyldlmethylamlne with methyl iodide to form VIIs, and the instability of the addlon compound of neopentyldlmethylamlne with trimethylboron .
-%
Homomorphs of di-_t-butylme thane
o/vc
o cA?
/v
C C
VI
Homomorohs of 2,6-dimethyl- t;-butylbenzene
C Nv C
Ok/0
a
XI
<V
c Nc cr ^c
VII
c
'X
CI xc
VIII
•K
c'S CAC (A? (A*
IX
Figr. 1
Homomorphs of o-t-butyltoluene X7III
BH3 ^C
a-
XIX
XX CI
XXII
Fig, 5
XXIII
XII
c
X
N-
vy
XIV
Fig. 2
XIII
Homomorphs of _o-di-_t-butyl- benzene
C^ £
?
C ^C ^N Lc
^c
XVI XVII
Fig. 3
Homomorphs of hemlmellitene
^y |
c |
XXJV NH3 |
XXV BH3 |
xy |
|
XXVI Fig. |
XXVII 4 |
-5-
Homomorphs of 2,6-Dlmethyl-t-butylbenzene. Trimethylboron forms a stable compound with pryldine, with a heat of dissociation of 17.0 kcal.3; but with 2,6-lutidine, a stronger base lnaqueous solution, trimethylboron does not react to form XII. This points to a strain of at least 17 kcal. in these homomorphs. The parent hydrocarbon (XI) is not known and could not be prepared.8 Over a period of several months no significant reaction of 2,6,N, N-tetra- methylanlline to form XIII was observed8. 2,6-Di-methylphenyl- carbinol and mesltyldimethylcarblnol were prepared, but they could not be converted to the corresponding chlorides .
Homomorphs of o-fli-t-butylbenzene . The beat of reactlor
of boron trlfluorlde with pyridine (and with trimethylamine) is about 25 kcal, but boron trlfluorlde fails to add to o-t-butyl-N, N-dlmethylaniline# This points to a strain of about 25 kcal. in this series of homomorphs9. It follows that such homomorphs should be exceedingly difficult if not impossible to prepare.
Homomorphs of Hemlmellitene (l. 2. 6-trimethvlbenzene ) . From the heats of combustion of the trlmethvlbenzenes13, hemimellltene (XXIV) appears to be about 1.2 kcal. less stable than its Isomers. The energies of activation of the reactions of pyridine and 2,6- lutidine with methyl iodide (XXV) are 13,9 and 14.9 kcal/mole, respectively. It may be significant that the 1.0 kcal. difference in energy of activation is in fair agreement with the strain predicted from combustion data. m-2-Xylidinium ion (XXVI) is the conjugate acid of m-2-xylidine, which has a. P.K of 3.42 vs. 4.25 for aniline. But the operation of both the inductive effect and steric inhibition of resonance should tend to increase the strength of m-2-xylidine.
Homomorphs of o-t-Butyltoluene . From the observation that the heats of reaction of boron trlfluorlde with pyridine and 2-.t- butylpyridine are 25.0 and 14.8 kcal., respectively, an upper limit to the strain of 10 kcal. can be placed. But since the steric requirements of the borine group (BH3) are considerably smaller than for boron trlfluorlde, the actual strain must be considerably lower. The energies of activation for the reaction methyl iodide with pyridine and 2-_t-butylpyridine (XX) in nitro- benzene solution are 13.9 and 17.5 kcal,, respectively. The value of 3.6 kcal., then, can be tentatively adopted as a lower limit to the strain.
SUMMARY
Model Homomorph Strain Ene rgy (kcal. /mole)
Hemlmellitene 1-2
o-jfc- Butyl toluene 4-6
Di-t-butylme thane 5.4
2,6-Dlmethyl-t-butylbenzene > 17
o-Di-j-butylbenzene ^ 25
_4-
REFERENCES
1. |
H. C. |
2. |
H. C. |
3. |
H. C. |
4. |
H. C. |
5. |
H. C. |
6. |
H. C. |
7. |
H. C. |
8. |
H. C. |
9. |
H. C. |
10. |
H. C. |
11. |
H. C. |
325 ( |
|
12. |
D. P. |
1345, |
|
13. |
W. H. |
Natl. |
|
14. |
R. Sp |
Science. 103. 385 (1946).
et.al., J. Am. Chem. Sqc. 75,1 ( 1953) ,
1137 (1947).
Brown,
Brown,
Brown and G-. K. Barbaras, lb Id . . 69.
Brown, Ibid. . 75, 6 (1953) .
Brown and H. L. Berne is, Ibid.. 75 10 (1953).
H. Bonner, ibid.. 75, 14(1953).
S. Fletcher, ibid.. 71, 1845 (1949).
Grayson, ibid.. 75, 20 (1953).
B. Johannesen, ibid. . 75. 16 (1953).
L. Nelson, ibid.. 75, 24 (1953).
Brown Brown Brown
Brown Brown Brown,
1942).
and and and and and H.
W. R. M.
R. K.
I.
Schlesinger, and S. Z. Cardon, Ibid. , 64. Watson, and R. Williams, J. Chem. Sqc.
Evans,. H. B. 1348 (1939).
Johnson, E. J. Prosen, and F. D. Rossini. J. Research Bureau of Standards. 35. 141 (1945 ). ltzer and K. S. Pitzer, J. Am. Chem. Soc. 70, 1261 (1948)
The Structure of Cytisine Reported by Blaine 0. Schoepfle February 13, 1953
The alkaloid cytisine was first isolated in IS651, how- ever, the correct emperical formula, C11H140N2, was not deter- mined until 1590. 3
A Zerevitinov determination indicated that the alkaloid contained one active hydrogen while the formation of a mono- N-acyl derivative fixed it as being attached to a nitrogen rather than to the oxygen.
The reduction of cytisini with Hl/P yielded several 6,3-dimethyl ouinolines3*4 and subseoue,ntly led to the structural proposals of Freund (I), Spath (II), and Ewins (III).
Nst^OH
CH.
v\\
vN
NT *0
H~N J
(II)
(III)
Structures (I) and (II) were eventually discarded6 since they do not contain a potential 6, 2-&i methyl ouinoline nucleus and since the corresponding N-methyl ouinolines were shown incapable of rearranging to the reauired 6,3-configuration. 6
Ing soon Questioned the presence of a ouinoline nucleus in cytisine and theorized upon the rearrangement of (IV) and (V) tyDe structures to ouinolines under the influence of Hl/P4. CH3 OH;
IV
ed a
The first exhaustive methylation studies of cytisine yield- dimer, C33H3303N36. This result Implied the absence of a condition which is satisfied only by a formula of type of which there are two, (VI) and (VII).
-2-
(VI)
(VII)
Subseouent exhaustive methylatidns have shown that the dimerization can be avoided8, thus, structures (VI) and (VII) were rejected in favor of a type (IV) structure. Five structures (VIII - XII) of this type must be considered.
H-N CHp N
V
1 1 \\
VIII
CH3"*
CH3 HN I |i
>"
N
If 0
XI
XII
The oxidation of N-methy ley ti sine yielded two isomeric compounds, C12H1403Ns9. These compounds Trere shown to be lactams in which the CHg-N(CH3)- groun in me thy ley ti sine had been oxidized to -CO-N(CH3)-. Thus, structures (IX), (X), and (XII ) can be eliminated since there is no chance of isolating isomers from the oxidation of there ^-methylated derivatives.
A choice between (VIII ) and (XI) was made on the basis that N-benzenesulphonyl-N-methyl-f-cytisami o acid looses C03 at its melting point, whereas the ct-acid derivative melts without decomposition.9 One of the isomeric N-benzenesulphonyl deriva- tives of (VIII) would be expected to loose C02 in the same manner as demonstrated in the case of 6-hydroxy-14--methylpyrldyl-2- acetic acid.10 This difference between the a and p-derivatives would be difficult to explain in the case of structure (XI).
by:
Subseouent proof &r structure (VIII) has been demonstrated
a) Ozonolysis of the exhaustive nethylation product, hydrolysis, and oxidation to a,at-dimethylglutaric acid.8
-3-
CH-
1 if
Oytisine->-fiH2 N *
CH:
'3
0
^ l)hydrol.
NH -^
CH3r— CH3 CHo-U— J 2)oxidn.
CHg-j-COOH CH3-J-COOH
b) Exhaustive methylation of N-acetyl-tetrrhydrodioxy- cytisine (XIII) and the subsequent conversion of its product to p-methylnicctenic acid11.
0 J f—
C%C-N CH2 ft
1 «j ■ -*
?, 1 — rCsHi1
CH3-C-N CH2
CH
Vv^11^
>V CDOI
1 '1
V
c) The degradation of cyt&sine to (XIV) 12 and its sub- senuent synthetical conformation. 13
CH3 CH?
COOEt
CH3^V
a) ethyl succinate _- >_
b) redn.
GUZ%S COOEt
CHg^v/ dOOEt GRaNs/V^
XIV
7,
9. lo( 11, 12, 13.
Bibliography
Hausemann and Marmer, Z. Chen., 1, l6l (1665).
Partheil, Ber. 23, 32OI (1290).
Ewins JCS 103, 97 (1913).
Spath, Monatsh. 40, 93 (1919).
Ing;, JCS, 2195, (1931).
Spath,. Mo natsh. kO 15, 93 (1919).
Infold and Jessop, JCS, 2357 (1929).
Spath and Galinovsky, Ber. £5, 1526 (1932).
Ing, JCS, 277& (1932).
Collie JCS 11, 299 (1^97).
Spath and Galinovsky, Ber. 66, 133^ (1933).
Indem, ibid., 63, 761 (1936).
Indem, ibid., 21, 721 (1932).
X
PHOTOCHEMICAL REACTIONS OF DIAZOMETHANE Reported by D*vld B. Kellom February 20, 1953
Although diazomethane has been frequently employed as a methylating reagent for phenols and other compounds contalng acidic hydroxyl groups, Its light-catalyzed reactions with organic com- pounds have only recently attracted attention. Therefore, it is the purpose of this seminar to review some of the recent work on the photochemical reactions of diazomethane with a number of organ! compounds.
Photochemical Reactions with Ethers
In 1901 Hantzsch and Lehman1 reported that a solution of diazomethane in ether slowly lost its yellow color on exposure to light. This reaction gave according to Curtius8 a viscous residue together with an evolution of gas, presumably nitrogen and ethylene However, it was not until forty years later that this reaction received further study.
From an examination of the evolved gases and the vaseline- like residue, Meerwein and co-workers3 could account for only about 22# of the diazomethane that had been irradiated in diethyl ether. Careful fractionation of the solvent established the presence of ethyl n-propyl ether (i) and ethyl J.-oropyl ether fll). When tetrahydrofuran was used as the solvent, a- and p-methyltetra- hydrofuran (III and IV) vere isolated.
CH3CH3OCH3CH3CHs CH3CH30CHCH3
CH3 I II
III IV
Similarly the Photochemical decomposition of diazomethane in i-propyl aclohol gave products corresponding to the addition of a fragment CH3 to the solvent. Indeed these results do suggest the presence of a highly reactive, high energy fragment.
CH3N3 > N3 + :CH3
Structures containing divalent carbon were frequently suggest- ed by early chemists.4 8 But only in the thermal or photochemical decompositions of diazomethane or ketene have such intermediates been definitely established.7
Photochemical Reactions with Hydrocarbons
From the Irradiation of a solution of diazomethane in benzene, Uoering and Knox8 reported the isolation of a small amount of a hydrocarbon, C7H8 (V or VI), identical with a sample of "cyclo- heptatriene" prepared by KohlerS from cycloheptanone. On oxidizat- ion with potassium permanganate the hydrocarbon gave tropolone (VXl).
-2-
OH
+ CH2Na * N3
, KMn04
VII
VI
The structure of the hydrocarbon has not been definitely established. But it is interesting to note In this connection, the recent case of valence tautomerism Cope10 has found for 1,3,5- cyclooctatrienc (VIII) and bicyclo [4,2,0] octa-2,4-diene (IX).
V
VIII
IX
11
Extending their studies, Doerlng and Knox found that the photochemical or thermal decomposition of ethyl dlazoaeetate in hydrocarbon solvents gave saturated esters in yields of about 40^. For example,
o*
^:CH-CK
N.CHCOOEt
O
CH2C00Et
+ N,
CH
+ N.CHCOOEt
<
^CH-CH CH3 ^CH3
.CHsCH2C00Et
+ N.
They concluded that the reaction involved a divalent carbon intermediate which they called a "carbene".
N3CHC00Et
■> N:
CHCOOEt
RH
RCHgCOOEt.
As evidence that the intermediate was not a diradical, the authors noted that it showed no preference for a tertiary hydrogen as do free radicals. Also they observed that while ethyl diazoacetate gave ethyl cyclohexaneacetate when heated in cyclohexane, the same reaction in the presence of copper powder gave only diethyl fumarate.
/'
Photochemical Reactions with Poly halome thanes
Recently Urry and Eiszner13 have studied the light- Induced reactions of diazomethane with polyhalomethanes. These reactions were found to yield Dolyhaloneonentane derivatives, a methylene group being interposed between each halogen atom in the organic halide and the carbon atom to which it is attached. For example, carbon tetrachloride gave a 60^ yield of totrachloroneopentane together with nitrogen and a small amount of poly methylene. Similar results were obtained with bromotrichlorome thane, chloro- form and methyl trichloroacetate.
CC14 + 4 CH3N3 _> 4 N2 + <3(CH3Cl)4
For these reactions a free radical mechanism is suggested by the following observations:
1. It is light- induced.
2. It is inhibited by dir)henylamine.
3. It is observed only with organic halides known to undergo free radical addition to olefins.' J
4. Equal volume 3 of nitrogen are evolved per unit time at constant light intensity and temperature and hence the reaction is of zero order.
Furthermore a reaction sequence involving stable intermediates as
CC14 -> C1CH2CC13 -> (C1CH3)3 CC13 etc.
is most unlikely for the intermediates would have to be far more reactive with diazomethane than carbon tetrachloride which is present in great excess. However, the first intermediate 1,1,1,2- letrachloroe thane did not give this reaction with diazomethane. Also 1,1,1-trichloroe thane, the possible first intermediate in the reaction with chloroform, was equally unreactive.
Therefore, the accumulated evidence favors a free radical, chain reaction mechanism involving only unstable intermediates. The following reaction scheme which was proposed for carbon tetra- chloride fulfills these conditions by postulating successive free radical rearrangements Involving 1,?- shifts of chlorine alternating with reactions with diazomethane. Reactions 1 and 2 are the chain- initiating and reactions 3 to 10 are the chain-propagating steps. No evidence is at Present available regarding chain termination.
1- CH3N3 > N3 + :CH3
2. :CH3 + CC14 — ^.CH3C1 + .CCI3
3. .C01a + CH3N3 — >N3 + .CH3CC13
4. •CH3CC13 ^ 01CH-CC1,
— 4—
5. ClCHgCCls + CH3N3 — ^ClCH3CCl3
• CH3
6. C1CH3CC13— ) ClCH3CCl
• CH3 CH3C1
7. (C1CH3)3CC1 + CH3N3 > (C1CH3)2CC1
• CH3
8. (ClCH3)3(jJCl > (C1CH2)3C.
• 0H3 CH3C1
9. (C1CH3)3C- + CH3N2— > N3 + (C1CH3) 3CCHg-
10. (ClCH3)aCCH3 + CC14 > <C1CH3)3CCH3C1 + -CC13
The possible competing reactions of the Intermediate free radicals with either diazomethane or organic halide impose stringent requirements if the self-sustaining chain reaction is to occur. Thus, the reactions of methyl chloroscetate and methyl dichloro- acetate with diazomethane to give oroducts having a wide boiling range suggest the failure of these compounds to react readily in step 10.
A similar reaction scheme seems generally applicable to the other photochemical reactions of diazomethane which have been discussed.
REFERENCES
1. A. Hantzsch and M. Lehman, Ber., 34, 2522 (l90l).
2. T. Curtius, A. Darapsky and E. Muller, ibid.. 41, 3168 (1909).
3. H. Meerwein, H. Rathjen and H. Werner, Ibid. , 75, 1610 (1942).
4. A. G-uenther, Ann., 123, 121 (1862).
5. J. U. Nef, ibid., 298, 367 (1897) .
6. J. Thiele and F. Dent, ibid . . 302, 273 (1898).
7. R. G-. W. Norrish and G-. Porter, Discussions of the Faraday Soc, No. 2, 97 (1947).
8. W. von E. Doering and L. H. Knox, J. Am. Chem. Soc, 72, 2305
(1950).
9. E. P. Kohler, M. Tishler, H. Potter and H. T. Thompson, ibid., £1, 1057 (1939).
10. A. C. Cope, A. C. Haven, Jr., F. L. Ramp and E. R. Trumbull, ibid, T 74, 4867 (1952) .
11. W. von Doering and L. H. Knox, Abstracts of Papers, 119th Meeting, American Chemical Society, Boston, Mpss., April 2, 1951, p. 2M.
12. W. H. Urry nnd J. R. Eiszner, J. Am. Chem. Soc, 74, 5829 (1952).
13. M. S. Kharasch, E. V. Jensen and W. H. Urry, Ibid.. 69, 1100 (1947); M. S. Kharasch, 0. Relnmuth and W. H. Urry, ibid.. 69, 1105 (1947).
STERIO CONTROL OF ASYMMETRIC INDUCTION Reoorted by Moses Passer February 20, 1953
The laboratory synthesis of a new asymmetric center In a molecule already containing (at least) one asymmetric center, with the resulting diastereomers formed In unequal proportions, was first achieved nearly half a century ago,3" and many Instances are now known of such intramolecular asymmetric syntheses. s' 3 An example of Intcrmolecular asymmetric synthesis is afforded by the reaction of benzaldehyde with hydrogen cyanide,3 which in the presence of quinine gives excess of (+) - mandelonitrlle, and in the presence of quinidine excess of the other enantlomorph. The use of an enzyme in the reaction medium3 is an elegant way to achieve asymmetric induction, and recently asymmetric syntheses have been accomplished through the use of ootically-active reducing agents to reduce ketones to active alcohols. 4y B' 6' 7
The steric factors involved in asymmetric synthesis have been studied by several investigators. Fleser5 has proposed certain generalizations that obtain in asymmetric syntheses in steroids, while Hassel,fe Pitzer,10 and Barton11 present evidence suggesting that both in simole eyclohexane systems9'10 and in cyclohexane- incorporating steriods x the preferred configuration on any given asymmetric carbon atom is that in which the more bulky substituent occupies the equatorial position in the chair conformation. This generalization is supported by electron-diffraction studies,9 thermodynamic calculations,10 and studies of the relative stabillt, and propensity toward formation of the isomers,11
Most recently, and nearly simultaneously, Curtin s and Cram have each proposed for the acyclic series a generalization to correlate the relative bulk of substltuents on an asymmetric carbor atom alpha to a carbonyl group with the observed stereospecif icity of reactions in which the carbonyl undergoes transformation to an asymmetric alcohol. The two versions of the rule essentially state that non-catalytic reactions of the type shown in equation (l) will give rise preponderantly to the Indicated diastereomer.
reagent v
(i)
In Cram's version, the more generpl of the two, L and S, represent- ing the larger and smaller groups, may be alkyl, aryl, amino or
-2-
substltuted amino, or hydroxy or substituted hydroxy; R may he hydrogea, alkyl, or aryl; and the reagent may he a G-rignard reagent or a reducing agent such as lithium aluminum hydride, aluminum alkoxide, sodium-alcohol, or sodium amalgam. In Curtin'p (more limited) version L is methyl or phenyl; S is amino, hydroxy, or methoxy; R is p- substituted Phenyl or cc-naphthyl; and the reagent is ]>- substituted phenyl- or a-naphthylmagneeium halide.
Curt in1 s statement is based on his work with twelve reactions of this general type, in each of which the relative cosf lguratlonr for both reaotants and products are known. Cram!s broader general lzation is based on a study from the literature of twenty- seven such reactions, essentially similar to Curtin*s in that in nearly every case group 3 on the ct-carbon is either hydroxy or amino, and on his own work with Bine reactions in which the o-substituertf are hydrocarbon groupings. I» addition, each author calls attention to a number of reactions in which configurations, as yet unknown, Can be predicted by the rule asd which when established by independent methods will provide tests for the rule.
Certain of the concepts inherent in the rule are applied by Oram (a) to explain the variations in ratio of structural isomers obtained i» the reactions of a pair of diastereomers with 3-nitro- phthalic aahydride14, and (b) to account for the difference in rates of the Sj.2 reactions of halide loa with the brosylates of a pair of diastereomers,
Tw© problems arise when a reaction is studied as to its consistency with the rule. Cne is the matter of yield. Clearly, if only one isomer is isolated and in less than 50f yield, there exists the possibility of its being the less abundant of the two possible products. In Curtln's work the yields in all cases but two are 5fl£ or greater. Cram points out that in most of his examples in which the yields are low the other isomer was obtained in approximately equal yield by re-versing the order in which the substituents were introduced. Although not rigorous, this pro- vides reasonable assurance that the Isomers isolated, even though in low yield, are indeed predominant. The second, more serious difficulty has to do with the question of relative bulk of groups1,6 especially as the concept is utilized in this particular situation. For example, Cram considers a Phenyl effectively more bulky than a dodecylamino because of the greater volume of the former in the immediate vicinity of the reaction center. However, situations are conceivable in which the planarity of a Phenyl group would allow it to offer less hindrance to an approaching reagent than would a smaller, but non-planar group.
Cram suggests two explanations for the observed stereo- specificity. First, coordination of either the G-rignard reagent or of the reducing agent with carbonyl oxygen would tend to in- crease the effective bulk of the latter and orient it as shown in equation (l), situated between the two least bulky groups of the adjacent carbon. A second explanation is analagous to the cyclic intermediate Droposed by Doering5 for his steroospecif lc Meerweln-Ponndorf-Verley reduction of a ketone by optically-active 2-butanol.
-3-
REFERENCES
1. A. McKenzle, J. Chem. Soc, 85, 1249 (1904),
2. E. E. Turner and M. M. Harris, Quarterly Reviews, 1, 299 (1947>
3. R. L. Shriner, R. Adams, and C. S. Marvel, In "Organic Chemistry," H. Oilman, editor, John Wiley and Sons, Inc., New York, second edition, 1943, volume I, pp. 308-315.
4. W. von E. Doering and T. C. Aschner, J. Am. Chem. Soc, 71, 838 (1949).
5. W. von E. Doering and R. W. Young, lbld.t 72, 631 (1950).
6. H. S. Mosher and E. La Combre, ibid.. 72, 3994, 4991 (1950).
7. A. Bothner-By, Ibid.. 73, 846 (1951).
8. L. F. Fieser, Exoerientla, 6, 312 (1950) .
9. (a) 0. Hassel and H. Vlervoll, Acta. chem. Soand., 1, 149 (1947); (b) 0. Hassel and B. Ottar, Ibid,. 1, 929 U947); (c) 0. Baftlansen, 0. Ellerson, and 0. Hassel, ibid., 3, 918 (1949).
10. C. W. Beckett, K. S. Pitzer, and R. Soitzer, J. Am. Chem. Soc, 69, 2488 (1947).
11. D. H. R. Barton, Experlentia, 6, 316 (1950).
12. D. Y. Curtdn, E. E. Harris, and E. K. Meislich, J. Am. Chem. Soc, 74, 2901 (1952).
13. D. J. Cram and F. A. Abd Elhafez, jLbld., 74, 5828 (1952).
14. Idem, JJbld., 74, 5846 (1952).
15. Idem, ibid., 74, 5851 (1952). u
16. (a) L. Pauling, "The Nature of the Chemical Bond, Cornell University Press, Ithaca, New York, second edition, 1940, p. 164; (b) reference (3), first edition, 1938, volume I, p. 268.
/ w>
SYNTHESIS OF PHENANTHRENES
Reported by C. W. Hlnman
February 27, 1953
In 1935 Haworth developed a synthesis many of the substituted phenanthrenes in go Involves initially the reaction of an acid chloride with naphthalene in the presence o The treatment thereafter depends largely up As an illustration of its use the reaction duction of 1 methyl phenanthrene is given.
AA.
HpC — C
0
W
H2C C^
\ A101, > /
0
H3CMgBr
which has provided od yields. The method anhydride or acyl f aluminum trichloride. on the derivative desin sequence for the pro—
0 C-(CH2)3C02H
Clemmensen
/\
cyclization
w
Se
340-350
7ZT$
Much of the recent work done on phenanthrenes has been based on the method of synthesis first used by Bardham and Sengupta . 8 This synthesis involves the alkylation of a carbethoxycyclohexanone followed by hydrolysis and cyclization.
CHsCH2Br
C-0C3H6
-2-
0 0 powdered j< c-0
K
toluene
CH2-CH3
Sn5
CH3- <^_^»
vj/
hydrolysis
CHg— ■ C H g
-o>.
■ Na C3HsOH J-
CHg— CHg
-•<z>
The product obtained proved to be 1,2, 3, 4,4a, 9, 10, 10a octa- hydrophenanthrene and by aromatizatlon with selenium at 280-340° C, yielded phenanthrene. By proper substitution of either the cyclo- hexane ring, or the benzene ring many other octahydrophenanthrenes can be obtained. By treating the 2-p phenyl ethyl cyclohexanone with the G-rignard reagent an angular methyl group has been intro- duced into the 4a position.3
The chief objection to this synthesis of octahydroohenanthrene is that the cyclization step produces a soirane to the extent of 50 - 60 per cent.
Pa0
2^5
CH--CH
runs
Bogert found that by reacting certain carbinols with concentrated sulfuric acid octahydrophenanthrenes could be produced which were Identical with those made by Bardham and Sengupta.4
He was able to overcome the soirane formation, however, by using beta-P^enylethyl-2 methyl cyclohexanol. He obtained almost
/ /
-3-
exclusively the 4a-methyl-l,2,3,4,4a,9, 10, 10a, octahydrophenanth- rene.
Cook has shown conclusively that the octahydroohenanthrene obtained by either the Bardham-Sengupta or the Bogert method is a mixture of two isomers.5
Barnes has shown that if an angular methyl group is intro- duced the resulting 4a methyl, 1,2,3,4, 4a, 9, 10, 10a octahydro- phenanthrene has but one form, which he believes to be the trans isomer.6
Introduction of an angular methyl g-rouo actually facilitates the cyclization step in these syntheses, but introduction of an angular carboxyl .group gives rise to new difficulties. When the cyclization step is attempted with 85^ sulfuric acid only de- hydration takes place. Only after hydrolysis of the ester and treatment with 90^ sulfuric acid is any of the desired cyclic product obtained, and even then lactone formation is the favored reaction. 7
H30-0
H3C-0 CH3
C0SH
VS
//
H,C0
85^ , A/V ' 3 I i
on..
H3S04
CHp — urij
90^
!r H3S04
hydro!.
C02Et N^ Br
T
CHg — CHg
H-C-0
Br
+ s<?
v^
.0=0
>
0CH?
S\
y^
Br
0H3 - CH3
The usual procedure for synthesizing compounds having a cyclopentanophenanthrene nucleus has been carried out by building up ring C and D starting with a substituted benzene or naphthalene A new approach, most used by Barnes, starts with a hydrindene and adds to it the A and B rings.6
V^Cl
0 CHo-CHp-C^OH
•4-
0
cyclize y LJ,y
V
CI
vy
Redn.
H3C-CHs-MgBr
OH
1)
^
Li
0
HpC-CJ
3v^-oH3 S) PBr3 3) Mg
CHo -
90/ H2S04 >
aromatization
The products of cyclizatlon were proved to be cyclizatlon products by oxidation with dll.HN03 which yielded 1,2,3,4 tetra carboxybenzene*
/J
-5-
BIBLIOCrRAPHY
1. R. D. Hnworth, J. Chem. Soc, 1125 (1932).
8. J. G. Bardham and S. C. Sene:upta, J. Chem. Soc, 2520 (1932).
3. J. C. Bardham and S. 0. Sengirota, J. Chem. Soc, 2798 (1932).
4. D. Perlman, D. Bavldson, and M. T. Bogert, J. Org. Chem. 1, 288 (1936).
5. J. W. Cook, C. L. Hevett, and A. M. Robinson, J. Chem. 168 (1939).
6. R. A. Barnes and R. T. G-ottesman, J. Am. Chem. Soc, 74, 35
(1952).
7. R. A. Barnes, H. P. Hirschler, B. R. Bluestein, J. Am. Chem. Soc, 74, 32 (1952).
8. R. A. Barnes, L. Gordon, J. Am. Chem. Soc., 71, 2644 (1949).
THE SYNTHESIS OF CANTHARIDIN
Reported by Elliott E. Ryder
February 27, 195;
Cantharidin was first obtained as a crystalline substance in 1810 by Pcbiquet, ard in 1914 the following structure was suggested by G-adamer and co-workers.
CO
A number of atterrrots have been made throughout the years to synthesize this compound, however, an acceptable method was not developed until recently.
In 1928 von Bruchhausen and Bersch attempted the following condensation,
&*,
cu.
Off,
0
however, it was found that dime thy lmaleic anhydride would not enter into a Diels-Aider reaction with this dier.e.
Attempts to methylate the hydrogens ted addition product of maleic anhydride and furan failed to produce the desired compound.8
Pyrolysis of hydrobromocantharic acid. I, was shown to reform canthsridlr. in good yield3, so Ziegler and co-workers attempted to synthesize this bromoacid.3 As is shown, their synthesis pro- duced the eD Irrier of hydrobromocantharic acid; though pyrolysis of this compound produced cantharidin in small yield, it may be looked upon as its first total synthesis.
0,
L
>
C^ow,
CF3
OH, COOH
CO
>6o
NBr
7>
S^CK* >■ Ag80
Br,//
As err,"*
"/ C00CHn CPOCH,
-2-
The synthesis which is the subject of this seminar has as its initial step the Diels-Alder condensation of dimethyl acetylene- dicarboxylate with furan.4
COOCH3
C *
HI
c
_COOCH,, 00nf7W3
Cone".
COOCH.
Butadiene was then added to the latter compound, and the ester groups were transformed into the methyl groups of the desired Pro- duct by the following method.
_COOCHaqtH(
GOOCF.
nHa3QaGL
^
IOOCF-,
K52aHs
r c^pO sJpOh
GF3OS03CH3
•>
CT^3SCHPCH. CHoSCHpCH.
0.q0
S^4
*
I CK330HSCH3
CH3 i>CH3 G ~t 3
NiR
C3HB0H
^
This glycol was then oxidized to t^e dialdehyde which was caused to undergo an intramolecular aldol condensation to a cyclo- pentenealdehyde; treatment of this compound with phenyl lithium followed by nryolysls of the stearate of the rearranged alcohol produced a diene which was then ozonized to yield cantharidln.
~3~
HIO*
>
pyridine acetate
C6HgLi
OHO
^
^Hd6C *
x o3
^HCSH5 2. F,0
^
2^3
CHOH
6^5
J)00017^,5
'^CHC6H5
BIBLIOGRAPHY
1. F. von Bruchhausen and H. ¥. Bersch, Arch. Pharm., 266. 697
(1928).
2. K. Ziegler, G. Schenck, E. W. Krockow, A. Siebert, A. Wenz, and H. Weber, Ann., 55JL, 1 (1942).
3. S. G-adamer, Arch. Pharm., 252. 636 (1914).
4. G. Stork, E. E. van Tamelen, L. J. Friedman, A. W. Burgstahler, J. Am. Chem. Soc, 75, 384 (1953).
HUMULENE Reported "by W. S. Anderson
March 6, 195.?
Two sesquiterpenes from clove oil/ftand^earyophyllene, hav* been assigned the structures given below.1
-,<ar
nA
Ts/
p-caryophyllene
^f-caryophyllene
Until, recently
5 little was known of the structure of-x-caryophyl- so as humulene), a third sesquiterpene (Ci5H34) in clove oil and a major constituent of hep oil.3,s
lene (known
a
Evidence for the presence of only one ring
The unriaturation in any comoound Ci5Ka4 can be accounted for by l) four double bonds s) three double bonds and one ring 3) two double bends and two rin^s 4) one double bond and three
rings 4) four rings 5) combinations with fewer than three double bonds. acid titration with analysis of the the presence of three double bonds, must be present in humulene.
one involving triple bonds
Hydro^e nation, perbenzoic epoxide, and infrared show
A single ring, therefore,
Ozonolysls of tetrahydrohumulene, which according to the infrared data has an endocyclic double bond (967 cm. x), yields a C16 dicarboxylic acid. This ozonolysis product, which con- tains the same number of carbon atoms as the starting material, confirms the presence of a ring.
The hydrohumulenes
The hydrogenated derivatives of humulene have been prepared in an effort to determine the humulene structure.7 The methods of preparation are outlined in Table 1.
The gem- dimethyl group
Oxidation of humulene ozonlde gives cc,fx-dimethyl succinic acid? From the oxidation of p-dihydrohumulene, a,a-dimethyl succinic acid and p,p-dimethyl adipic acid can be isolated, leaving no doubt that a structure of the following type is ore sent:
CH3— C— CH3 CH3
J
-c
6 CD bD
^ C £
fd CD
>srH
£ 3
05 £
CD 4J
+3 rH O CC
£ O
OK WO
KO
oo
^
£ I
£
o
c
0)
£ CD H
£ bD 3 £ .£ -H O E U U t3 O >-.<♦_, ,£ I CC 4J ^ rH +3 CC CD CD 4Jw
tJ |
|
<r- |
|
Ph rH |
|
O O |
|
d o |
|
.£ s |
|
c ■* |
|
o « |
|
CO ffi |
|
O ifl |
|
^ H |
|
+^o |
|
«H |
|
£ |
|
/ |
k |
o |
|
s |
|
o |
rH |
10 |
O |
tc |
K |
re |
|
o |
+ |
CD |
|
£ |
|
CD <* |
|
H « |
|
3ffi |
|
E u) |
|
3 * |
|
.£ |
:o |
4-3 CO
^
1 |
CD |
||||||
O CD |
£ |
£ |
|||||
£ £ |
£ |
CD |
|||||
•H CD |
cc |
rH |
- |
||||
EH |
£ |
3 |
|||||
CC 3 |
<•-» |
E |
|||||
rH E |
o |
3 |
|||||
>s 3 |
K |
X! |
|||||
.£,£ |
V |
•g |
|||||
4-3 O |
? |
||||||
CD U |
t3 |
||||||
E -d |
>5 |
||||||
•H >i |
.£ |
||||||
d ,c |
•h |
||||||
CC |
•d |
||||||
/ ^ |
! |
||||||
4-3 |
CO. |
||||||
3><D |
|||||||
4J |
CD £ |
£ |
|||||
| |
CD |
CC |
CD |
||||
O CD u £ |
+ w |
t r-1 O 3 |
E |
£ |
CO O |
CD £ |
|
fd <d |
o o |
£ E |
O |
rH |
4J to |
CD |
|
>;rH |
coo |
•H ^ |
ffi |
3 |
f^ K |
rH |
|
.c 3 |
K O |
ex: |
E |
3 |
|||
CC E |
o K |
cc o |
3 |
E |
|||
K 3 |
rH U |
X! |
3 |
||||
CD .£ |
N- |
>i<& |
\ |
O <d |
^ |
.£ o |
|
X! — O |
7 |
,£ >i — -P .£ |
y |
||||
£ |
CD CC |
>. |
*d |
||||
•H |
E X |
.£ |
>-. |
||||
E |
«H CD |
CC |
.£ |
||||
CD |
<d .£ |
Jh |
CC |
||||
1 |
f |
4-3 |
K |
||||
d |
u |
CD 4J |
CD .£ |
£ |
|||
CD |
£ |
||
£ |
cc |
||
CD |
£ |
||
r-\ |
<M |
CD |
|
> 2 |
O |
£ |
|
o £ |
K |
CD |
|
£ ja |
\ |
H £ |
|
E O |
/ |
||
a ?h |
3 |
||
rH rrj |
.£ |
||
>. >-. |
o |
||
X £ |
u |
||
+J CC |
<d |
||
CD ?h |
>-. |
||
£ 4J |
A |
||
■H CD |
•H |
||
Tj += |
1 |
||
1 3 |
1 |
-3-
The size of the ring
If one assumes that the gem-dimethyl grout) Is located on a ring carbon atom, then that ring must be at least six-membered to explain the 0,0- dimethyl adlpic acid oxidation product. The Cis tricarboxylic acid obtained by ozonolysis forms an anhydride when treated with acetic anhydride, but does not form a ketone when its thorium salt is heated. A six-membered ring should give on ozonolysis a ketonizable acld^ therefore, the gem- dimethyl grout) , if on a ring, 1 e on a large ring. Infrared also gives no support to the six-membered ring structure; 1000 cm ~1 and 1055 cm x peaks are missing.
Levulinlc aldehyde is another humulene ozonolysis product. If one now assumes that humulene, like other cyclic sesquiter- penes, has its isoprene units linked head-to-tail as in the farnesene chain, that it is a single substance or a mixture of substances differing only in the oosition of the double bonds, and that the gem-di methyl grouo is on a cyclic carbon atom, then only carbon skeletons I through VI can be written for humulene . 7
c_c-c-c~c — 1 J c-c-c-c-1 1^
C-C-C-
II
III
-fyv*
s?
» JS.
TV
V
VI
I has been synthesized;4 it is not hexahydrohumulene, a fact which supports the absence of a siv-membered ring. Skeletons II- V cannot afford both levulinlc aldehyde and a,a-dimethyl succinic acid as degradation oroducts. Structure VI is left as the carbon skeleton for humulene.
The double bonds
Indications that the double bonds in humulene are not conjugated are l) failure of the Diels- Alder reaction of ac- dlhydrohumulene with maleic anhydride 2) failure of sodium and alcohol reduction 3) ultraviolet absorption only below 2350 it 4) absence of an exaltation of molecular refraction. The evidence is not conclusive, however, since conjugated double bonds in large rings do not behave normally. 1,3-cyclooctadlene,
-4-
for example, does not form a simple adduot with maleic anhydride.
Ozonolysls yields formaldehyde as well as levulinic aldehyde an observation indicating a terminal methylene group.
Infrared data are given in Table II.
Table II
Frequency Assignment
Where present Humulene 0-Dihydro Tetrahydro Hexahydrr
1360 |
CH3 |
X |
X |
X |
X |
1450 |
-CHr |
X |
X |
X |
X |
967 |
t trans RCH=CHR |
X |
X |
X |
- |
840 ublet 831 |
rch=crr" (two types) |
X |
X |
- |
- |
885 |
CHS=CRR' |
X |
X |
— |
- |
The presence of four double bond frequencies in humulene and in 0-dihydrohumulene is explained if these substances are mixtures of isomers having the double bonds in different positions. A 1,5,8 arrangement of double bonds in humulene for example, would account for obtaining levulinic aldehyde and a,a-dimethylsuccinic acid. However, an exocyclic double bond is required to obtain formaldehyde, which has been obtained in 96^ of the yield expected from one exocyclic double bond. The existence of more than one isomer is, therefore, a strong possibility, and the best representation which can be made for humulene is
This large ring can easily accommodate the endocycllc trans double bond indicated by infrared; the model is compact and strainless) and finally, the observed (probably spurious) optical activity [-a ] 3^
:1. Do is compatible with this
structure.
-5-
REFERENCES
1. John Walker, Organic Chem. Seminars, University of Illinois, March 21, 1952.^
2. V. Herout, M. Streibl, J. Mleziva, and F. Sorm, Coll. Czech. Chem. Comm., 14, 716 11949). ,
3. F. Sorm, J. Mleziva, Z. Arnold, J. Pliva, ibid. t 14, 699
(1949).
4. F. gorm and L. Dolejs, Ibid.. 15, 96 (1950).
5. F. Sorm, M. Streibl, J. Pliva, V. Herout, Chem. Id sty, 45, 308 (l95l); [CA46, 449? (1952)] .
6. G-. R. Clemo and J. 0. Harris, J. Chem. Soc, 1951, 22.
7. G. R. Clemo and J. 0. Harris, ibid.. 1952, 665-
8. Gr. R. Clemo and J. 0. Harris, Chem. and Ind., 1951, 50.
9. G. R. Clemo and J. 0. Harris, ibid., 1951. 799.
10. M. L. Wolfrom and A. Mishkin, J. Am. Chem. Soc, .72, 5350 (1950).
11. Atsushi Fullta and Yosho Kiroae, J. Pharm. Soc. Japan 71, 176 (1951)} [CA 45, 6804 (l95l)j.
oCi
SEW REACTIONS OF $~PROPIOLACTONE Reported by William S. Friedlander March 6, 1953
Prop io la a tone, well established as a synthetic tool In organi chemistry, owes its utility to the fact that it is a (3 -lactone. While y and f lactones usually undergo normal ester hydrolysis, it is not difficult with propiolactone, by using appropriate con- ditions, to obtain cleavage of the g -carbon- oxygen bond. Some authors ' 8 have attributed this reaction path to the strain in this bond due to the size of the ring.
A summary of the main reactions of propiolactone follows: l) Reaction with alcohols:3
CH5- CHg
II
0 C=0
ROH
acid
■>
base
■>
s) Reaction with Phenols:4
CH3-CH3 — C=0
no cat.
ArOH
or base
sot
(cat)
->
ROCH3CH3COOH HCCH3CK3COOR
ArOCH3C*H3COOH H0CH3CH3C00Ar
3) Reaction with carboxylic acids, carboxylic acid salts, and acid chlorides.5
II
0 (UO
!
RCOONa
>
RCOOCH3CH3COONa
BC0.QK ^ RCOOCH3CH3COOH
RG0C1
>
RC00CH3CH3C0C1
4) Active methylene compounds.
CHp— CHp
II
0 — c=o
RHa.
>
R— CHn-CHs
I
K0-C=0
5) Reaction with sodium nitrite, sodium dithionite, sodium cyanide, sodium thiocyanate, sodium succlnimide and aryl sulfinic acid salts: 13
CHp- CH«5 t t
0 c=o
NaNO-
aq .
^ 02NCH3CH3C00Na
H
^> 03N0H3CH3C00H
-s-
+
CH3CH3COONa
b) 2(1) + Na3S304 v | H"
Sa04 'x > S03(CHsCH3COOH)
I
CH3CH3C00Na + S03
+
c) I + NaCN ^ NCCH3C"3COCNa H— > NCCH3CH3COOH
(cone, aa . s oln . )
0
d) I + NaSCN — 2£^> NC3CH3CH3000Na > H3N-C-S-CH3CH3C00H
+
e) I + C6H5S03Na -J aq« v> C6HBS03GH3CH3C00N9 -JL^ G6H5S03CH3C00:
f) I + CH3-C0 CH3-C0 H+
| ^>NNa > > I ^> NCH3CH3C00Na
CH3-C0 CH3CO /
CH3-C0 ^
! J^> NCH3CH2COOH CH3-C0
In all the reactions mentioned above it is to be noted that time, temperature, order of mixing, p_H, and type of reactant ©re important factors in determining the course of the reaction and the final product. In some of the reactions, such as with alcohols for exanrole, some generalizations can be made, but the behavior of the amines, particularly of the allohatic amines, is very unpre- dictable.
Recently, in order to test the generality of a mechanism9 for the reactions of alcohols in basic solution with prooiolactone Hurd and Hayaox have studied the reactions of some substituted anilines with pror»iolactone. Their findings as well as those In G-re sham's earlier work with amines and rsrooiolactone10 are summarized below.
General reaction:
CH3-CH3 ^ R3NCH3CH3C00H R« s=H, alkyl, Arvl
' I R3N /~ (exceot tertiary)
0 C=0 \ ^ H0GH3CH3C0NR3
1.) There is no correlation between the basic strength of
primary and secondary amines and the amount of amino acid formation. ■' ■ Thus ammonia, dimethylamine ,
and ethylamine give mostly amino acid, while methylamine, diethylamine and propylamine give mostly amides.
-3-
2.) Aromatic or cyclohexylamines give amino acids more con- sistently than alkylamines.
3.) Usually water is the best solvent for amide formation; and scetonitrile is best for promoting amino acid for- mation.
4.) The order of mixing is important here. Thus when dl- methylamine is added to the lactone in ether the amino acid is formed. But when the lactone is added to the amine, the amide is the main product; and when the two are added simultaneously to ether the amino acid and amide are produced in about equal proportions.
5.) When substituted anilines are used, the only products formed are B- (anilino)-oropionic acids.1
I *"| YCSH4NH3 .YC6H4NHCH3CH3COOH
o — c= o ' 7
Y=COOH, COOEt, S03H, 303NH3, CI, Br, N03
a.) The course of these reactions is unaffected by sulfuric acid, sodium ethoxide, or whether the solvent is water, acetone or acetonitrile .
6.) Primary amines will react with two moles of lactone to form tertiary amines:
CH3-CH3
YC6H4NH3 0 0=0^ YC6H4N(CH3CH3C00H)3
Y=m-CO0Et, p-COOEt, p-Cl
7.) In general the reactions of amines with proplolactone are almost quantitative in contrast to -reactions with alcohols or acids ^here there is a good deal of polyester formation observed.
Mechanism:
In general the hydrolysis of a lactone can proceed by three possible mechanisms separately or simultaneously, depending upon the joH of the medium. 11,ls
1.) Neutral w nw. H CH.
H CH3 H CH3
H-O-H + 0 C H HO C-OH
0=C CH3 lH ^ 0=C CH3
HOH
*■■' -'3 " |; '
S*Jkf
-4-
2.) Acidic. |
|
© * V HOC 1 1 |
|
0=C GH3 |
|
H H |
|
3.) Basic |
|
H CH3 |
|
H-O-H |
+ 0 c |
0=C CH;
+ HO"
->
■>
HO-
o=c- I
H 0H3 \/ — C
I
■CH.
OH
0=C
H
©
HOH
When alcohols or -phenols are present, the reaction must also take one of these courses. Bartlett and Ry lander have shown that the methanolysis of nropiolactone In basic media takes the follow- ing course:
GHp— CHp
II
o — c=o
CH,0H
oh"
(8 min) CH30CH3CH3C00CH3
CHa0Na
H20
<■
->
H0CH3CH3C00CH3
II 17%
■H30 /
I
QHflOH
V/
CH3=CHC00CH3
CH,OCH9CHoCOONa
If the hydroxy ester (H)is substituted for proDiolactone, the same product is formed.
The initial attack by methoxide is in agreement with mechanisrr 3 above. Since phenoxldes are less basic than alkoxidea the mechanism of their reaction is of type 1, and the nucleophlllc attack by the Phenoxlde is at the B-carbon.
Because the reaction with amines Involves basic conditions, one would exoect type 3 mechanism" to prevail when they react with Propiolactone and the same type of Intermediates as with the methanol reaction should be pre sent i However, none of these could be Isolated and the course of the reaction appears to be different from type 3.
-5~
BIBLIOGRAPHY
1. CD. Hurd and Shin Hayao, J. Am. Chem. Soc, 74, 5889 (1952).
2. W. E. Smith, Organic Seminar, Spring Semester, 1951.
3. T. L. Gresham, J. E. Jansen, F. W. Shaver, J. T. Gregory, and W. L. Beearg, J. Am. Ohem. Soc, 70, 1004 (1948).
4. T. L. Gresham, J. E. Jansen, F. W. Shnver, R. A. Bankert, W. L Beears and Marie G. Prendersast, ibid . , 71. 661 (1949).
5. T. L. Gresham, J. E. Jansen and F. W. Shaver, ibid., 75, 72 (1950).
6. T. L. Gresham, J. E. Jansen, F. W. Shaver and J. T. Gregory ibid.. 70, 999 (1948).
7. T. L. Gresham, J. E. Jansen, F. W. Shaver, M. R. Fredrick and W. L. Beears, ibid.. 75, 9345 (l95l).
8. P. D. Bartlett and G. Small, Jr., .ibid., 22,4867 (1950).
9. P. D. Bartlett and P. N. Ry lander, ibid., 73, 4273 (1951).
10. T. L. Gresham, J. E. Jansen, F. W. Shave r,~~H. A. Bankert and F. T. Fiedorek, lbid.r 73, 3168 (l95l).
11. A. R. Olson and R. J. Miller, ibid.. 60, 2687 (1938).
12. A. R. Olson and J. F. Hyde, ibid.. 63, 2459 (l94l).
13. T. L. Gresham, J. E. Jansen, F. W. Shaver, M. R. Frederick,
F. T. Fiedorek, R. A. Bankert, J. T. Gregory and W. L. Beears, ibid., 74, 1323 (1952).
11- OXYGENATION OF THE RING- C- UN SUBSTITUTED STEROID NUCLEUS Reported by Howard J. Burke March 13, 1953
ISHtt OJ"3 1 7
......
A.
HOBr-CrOa
In recent years many wpyg; both chemical and microbiological, of oxygenating the 11- position of ring- O-un subs i;itu ted steroid nuclei have boon developed. The chemical methods all have in common the feet that they use as starting material a steroid having un- saturation Forne*rhere in ring 0„ These method* are .'
Methyl & -c^olenate (II; yields methyl 11-keto- cholanate (III)1:
Br
Br
i
i
0
COpCH
^x^
HO
Mi^ >
x^
In the original paper the soatial configurations at CX1 and C13 were just the reverse^, but the conclusion was reached some years late:**0 that the assignment of the ,6- orientation to the Ci2 sub- stituent of the parent compound was in error. Accordingly, the orientatior.s have been corrected, bringing the results into line with more reoenc work*
B_. PeracMsl>4"',3^lGrjLS^14;?ls?19,S3- The reaction of an 11,12 double
bond is bhoix'n with methyl ZS - lithocholenate acetate (IV)2'
OH
0..
'yNp -gtOa> Ha
^ -monoleflns give the <t-eooxide4' °"2 which, upon oxidation may yield an eooxyketone4' 10 or a keto-hemlacetal3' sl , as shown with methyl A9 (xl )-Iithocholenate acetate (Va) sl and methyl ^ &(11)- lithocholenate (Vu)10 (R=-CH(CH3)CH2CK3COOCH3) :
-2-
R
HpO-HQAc-
] 0CO,H
V (s)R=Ac (b)R=H
N?OMe CrO^H30
WO
J Np8Cra07-H0Ao 1
2}Zn-H0Ac
Figures I and II describe, respectively the reactions of /\ ©(n)-dienes with performic and perbenzoic, and with monoper- ohthalic (MP A) acids. Not all of the 7react ions have been shown to^ to occur with the model compound — Z± ' 9^lx >i 22-ergostatriene (VII; but all have been demonstrated on one or more compounds from the sterol, bile acid, or steroidal saoogenin series. The reactions cannot always be transferred from one class to another, e.g. with performic acid VII yields epoxyketones of the type VIII from sterol and steroidal sapogenins8, but unsaturated ketones of the type XI from members of the bile-acid series9. References to some work in which these types of reactions have been used are given in paren— ■ theses near the appropriate arrows.
Figure I
(g,lg,23)
<
20CO3H (22)
I
(6,125,22)
VIII
/ r
"te
(g;ig,23)
H30, SWli
strong
alkali (IS) ( (19?
0
.. MeOB-KOH CH3T (6)
(22)
xi °Jtf Vxi
Hydro], ytier Rearrsnge- ment ( j
HO
rL_ Zn-HOAc 111 ^777)"
'o ".\s vo
Further transformations of VII, X, XI, and XIII are shown in Fig. II.
XIV
(7,11,12,13)1
VMPA
(7,H,12) y. (l^h]2' °
ail, H3S04 J,.-0 JL BF3-Et20
XVI
dioxane
(7,11,12)
Cr03
HO Ac
(16)1?
C«H
Sn6
XI IsoprapBnyl Acetatje, CsHs
I (19)
MPA
XXI Et30 \OAc (19,22)
(S,l^)
"Jo Iff-
Kischner
Cr03 (3,14)
Wolff- ^>
Kischner XIII (5,6,7,22)1 " JXVI
o /r^x
(7,12,17,22) XTV-'
— » XIII
1) HS-CH20Hs-SH
_ > XIX
2) R-Ni
9,24
-4-
(xi)
C. NBS- t-BuOH" '. A ^-7,s XAi '-diene such as methyl 3a-
acetoxy- -^^ 7 9 (** )> -choladlenate yields up to three products of
types IX, X, and XVII, depending on conditions; this particular reaction was run in t-3uOH with dil. H2S04 at 0°.
D. KMnO.-HOAc
20, zs>
When treated with 5% KMnO. in HOAc at
10 , Va gives the p-epoxide, as contrasted to the cc-product from the action of perbenzoic acid.
E. Na2Cr207'2H20 - gl.HOAc®'22 ,23. Oxidation of such dienes as methyl 3a- acetoxy- ^ 7 9 (ll ),-choladienate can give at least two products of the types IX and XV.
F. Fe++ - H20222,2a. From Z^ 7 ,9 (l1 ^-cholestadiene benzoate (XXIV) the 9a,lla-oxido-7-ketone is formed3 3, while from methyl 3oc-hydroxy- & 7 9 (n )rcholadienate (XXV) the ^s^U1)- and ^ s-7- ketones are formed22 .
It is of interest to note that while reductions of 11-keto steroids with LiAlH4, LiBH4, NaBK4 , or catalytic hydrogenation yield the llp-hydroxy product, reduction with Na and boiling pro- panol give the lla-hydroxy isomer13.
Microbiological ll-Oxidatlon30"9 .
Only recently have processes been reported for 11-oxygenation of steroids by various microorganisms, namely: 1) Aspergillus nlger, 2) Streptomyces fradlae, 3) Rhizopus nigricans, 4) Rhizopus arrhizus. The method is characterized by its simplicity, ease of workup, speed, and good yields. Some of the transformations which have been made are:
Compound Treated
Progesterone
Reichstein's Cpd. S
Desoxycorticosterone
Reichstein's Cpd. S
Desoxycorticosterone
Progesterone
Reichsteln' s Cpd. S
17^-Hydroxyprogesterone
6-Dehydroprogesterone
Hydroxy 1 |
||||
Introduced |
Fungus |
Yield |
Ref. |
|
Hoc; 11a, 6 j |
A. |
niger |
— |
30 |
11a |
»t |
— . |
30 |
|
11a |
r? |
— ; |
30 |
|
IIP |
S. |
fradiae |
— |
33 |
11a |
R. |
nig. |
50-60$ |
35 |
11a |
R. |
nig. |
nearly quant. |
35 |
11a |
R. |
nig. |
60-80$ |
36 |
6P |
R. |
arrh. |
good |
36 |
lla |
R. |
nig. |
70-75$ |
37 |
6p |
R. |
arrh. |
45^ |
37 |
lla |
R. |
nig. |
50-60$ |
38 |
XXVI
The greatest stimulus to this work has been the desire for better routes to cortisone (XXVI), and several articles record the use of these methods to supply new routes to the compound25""9136.
-5-
BIBLIOGRAPHY
1) H. Reich and T. Reichstein, Helv. Chim. Acta, 26, 562-85 (1943).
2) E. Berner and T. Reichstein, ibid.. 29, 1374-81 (1946).
3) L. F. Fieser, H. Heymann and S. Rajagopalan, J. Am. Chem. Soc, 72, 2306 (1950).
4) L. F. Fieser and S. Rajagopalan, ibid., 73, 118-22 (1951).
5) E. M. Chamberlin, et al., ibid.. 73, 2396-7 (1951).
6) L. F. Fieser, J. E. Herz and W, Y. Huang, ibid., 73, 2397 (1951).
7) H. Heumann, et al. , Helv. Chim. Acta, 34, 2106-32 (1951).
8) G. Stork, et al., J. Am, Chem, Soc, 73, 3546-7 (1951).
9) L. F, Fieser, et al., ibid a , 73, 4053-4 (1951).
10) H. Heymann and L, Fu Fieser, ibid., 73, 5252-65 (1951).
11) H. Heusser, et al., Helv, Chim. Acta, 35. 295-307 (1952).
12) H. Ileusser, et al., ibid » , 35, 936-50 Tl952) „ H. Heusser, R. Anliker and 0., Jeger, ibid., 35, 1537-41 (1952). C. Djerassi, et al., J. Am, Chem. Soc, 74, 1712-15 (1952).
E. Schoenewaldtj et al., ibid., lit S6SS Cl952).
F. Sondheimer, et al,, ibide, 2±t 2696-7 (1952). J. Romo, et al., ibid. , 7£,' 2918-20 (1952). R. Budziarek, et al. , J. Chem. Soc, 195:?., 2892-2900.
C. Bjerassi, et al., J. Am. Chem. Soc, 2i> 3321-3 (1952). J. II. Constantin and L« K. Sarett, ibid., 74, 3908-10 (1952). H. Heymann and L. F, Fieser, ibid., 74, 5938-41 (1952). L. F„ Fieser, W. Y. Huang and J. C. Babcock, ibid. , 75. 116-21
(1953).
L. F. Fieser and J. E. Herz, ibid., 75, 121-4 (1953).
L. F. Fieser, W. P. Schneider and W. Y. Huang, ibid., 75, 124-7
(1953).
H. Heymann and L. F. Fieser, ibid., 73, 4054-5 (1951).
J. M„ Chemerda, et al., ibid., 73, 4052-3 (1951).
G. Rosenkranz, J. Pataki and C. Djerassi, ibid., 73, 4055-6 (1951) R. E„ Woodward, F. Sondheimer and D. Taub, ibid., 73, 4057 (1951). 0. Mane era, et al., ibid,, 74, 3711-12 (1952). J. Fried, et al., ibid., 74, 3962-3 (1952).
D. H. Peterson, et al., ibid., 74, 5933-6 (1952). D, H„ Peterson and H. C. Murray, ibid-, 2£, 1871-2 (1952). D. R. Golingsworth, M. P. Brunner and W. J. Haines, ibid., 74. 2381-2 (1952),.
P. D. Meister, et al., ibid., 75, 55-7 (1953). S. H. Eppstein, et al., ibid a , 75, 408-12 (1953). D. H. Peterson, et al . , ibid., J75, 412-15 (1953). P. D, Meister, et al., Ibid. , 75, 416-18 (1953). D. H. Peterson, et al., ibid., 75, 419-21 (1953). S. H. Eppstein, et al., ibid.. 75, 421-2 (1953). M. Sorkin and T. Reichstein, Helv. Chim. Acta, 29, 1218 (1946).
SYNTHESES OF LONG- CHAIN FATTY ACIDS Reported by John R. Denuth March 13, 1953
iith the renewed interest in synthetic methods of producing long-chain fatty acids, there have been developed several new genera procedures for synthesizing them. It will be "the object of this seminar to describe some of these new methods.
Stetter and Dierichs have developed a method of producing fatty acids which uses as starting material l,3-cyclohexanedione, produced by the hydrogenation of resorcinol over Raney Nickel.13 The success of this method depends upon the occurrence of carbon-alky lati on of the dione. According to the investigations of G-. Schwarzenba.cn,11 l,l~dimethyl-3 ,5-cyclohexanedione exists in the enol form in aqueous solution to the extent of 95*3 Per cent. It seems likely that a similar equilibrium, lying much in favor of the enol form also would exist for dihydroresorcinol. Therefore, it is not surprising that until the appearance of the current series of papers by Stetter and Dierichs, only two reports of _C-alkylation of this compound were to be found in the literature."
8,9
An outline of the synthesis developed by Stetter and Dierichs is shown below.
0
H
8
K
+ RX
CH3OH
"Oil'
■R
\A
0
HgC CH2-R
I H2C. xCOONa CH2
;iolff- Kishner >.
Reduction
R-(CHs)s-COONa
In order to find the optimal conditions for carbon-alky lati on of 1,3-cyclohexanedione, the conditions of alkylation were varied systematically. Because of its intermediate position between the lower halid.es which would be expected to be more reactive, and the higher alkyl halides, the reaction of which would, be more interest- ing, n-butyl bromide was chosen as the halide for these experiments.
The reaction was studied with respect to its dependence upon solvent, the alkali metal used, concentration of 1,3-cyclohexanedlon and type of alkyl halide employed.
The ratio of C_ to O-alkylation was found to be independent of the alcohol chosen as solvent, although the use of methanol for this purpose led to a significant increase in over-all yield of alkylated product. Of the three alkali metals employed; lithium,
-2-
sodium, and potassium, the last named was found to give the highest 0- to O-alkylation ratio, tilth methanol as solvent, it was found that the more concentrated the dione, the more favorable was the 0/C alkyl at Ion ratio. When n-butyl iodide was used in place of the corresponding bromide, the percentage of C -alkylated product in- creased.
Under the conditions described above, a series of alkylated 1,3-cy clone zianediones was prepared. The results of these reactions are summarized in Table I.
Table I
Reaction of l,3~Cyclohexanedione with Various Alkyl Iodides
Alkyl Iodide |
Reaction Time |
C-comr). ef |
0- com id . of " |
Total |
c/o Ratio ^ |
Methyl |
k-5 min. |
51.5 |
' ri. |
||
Ethyl |
3 hrs . |
27.2 |
4-3.° |
70.2 |
1:1.6 |
n-Propyl |
3 hrs . |
26.0 |
32.5 |
53.5 |
1:1.25 |
n-Butyl |
3 hrs . |
2SA |
36.1 |
Q\.& |
1:1.3 |
n-Cetyl |
2K hrs. |
27.0 |
51.0 |
73. 0 |
1:1.9 |
From this table, it can be seen that the size of the alkyl iodide employed has little effect on the amount of (/-alkylated product obtainable or upon the ratio of C- to O-alkylation product, except that when methyl iodide was used, none of the enol ether was formed.
The carbon substituted l,3-c3rclohexanediones are colorless crystalline compounds which must be used soon a,fter preparation for
they show signs of decomposition - discoloration and development of a disagreaable odor - after a day or two. The rate of decomposition increases with increasing length of the alkyl radical introduced.
Hydrolysis of the alkylated dihydroresorcinol with baryta water yielded the 5-ketoacid which was easily converted to the saturated acid by the Huang-Minion modification of the Jolff-Kishner reduction
In subsequent experiments directed toward extending the useful- ness of this reaction, Stetter and Dierichs condensed two moles of the dike tone with one mole of formaldehyde to obtain the compound shown as (III) below.13 Under very mild conditions, (III) was con- verted to the expected monocarboxylic acid (IV). However, opening
-3-
of the second ring was accompanied by immediate recycllzation by the loss of a molecule of water to give rise to compound (V) .
0
0
s s
o 0
M
CH3-CH3-C- (CH3 ) 3-CCOH
A/1
in
iv
CH3-CH3-C00H
ch3-ch3-ch3-coq
V
An ingenious though simple method for avoiding this difficulty was devised. By carrying out the ring opening with base in the presence of hydrazine under the conditions of the ./olff-Kishner reduction, not only was the desired 1,11-undeoanedicarboxylic acid obtained in quantitative amounts, but also one step of the already short synthetic route was eliminated. Further experiments have show: that this shortened procedure is generally applicable to alkylated 1,3-cyclohexanediones.
Compounds having an active halogen were found to alkylate di- hydroresorcinol rather readily. By working in aqueous rather than in methanolic solution, and using the simplified method of ring- opening and reduction, relatively high yields, 6^,-^0% , of the acids derived from alkylation of dihydroresorcinol by the following com- pounds were obtained: bromoacetic acid, allyl bromide, l-bromo-2- " cyclohexene, benzyl chloride (in the presence of Kl) ajnd 19H~ dibromo-2- butene (reac ted with 2-moles of 1,3-cyclohexanedione) .
Work is now in progress to see if the production of branched chain fatty acids is feasible by the replacement of the second active hydrogen of a l-alkyl-2,6-cyciohexanedione by another alkyl group.14 At the present time, one such a-cid, 6-methyl-7-phenyl-heptanoic acid, has been prepared.
The second synthetic route to be described is much more elabor- ate than the previous one, but it seems to be rather versatile and tc be applicable to the production of very long chains.4
An outline of the method is shown below.
Na
(HsC3OOC)3CH-(CH3)n-COOC3H5 -> (PhCKa03C)3C-(CHs)n--CGOCK3Ph
3?h0H20H I
I Na II
•*•
RC0C1 (III)
(H02C)8C-(CPI2)n-COOH
I
COR V
I
~co3
R_C-fCH2)n+1-COOH 0 VI
H=
«-*■
Pd-C
(PhCH202C)2C— (CH2)n-COOCH2Ph J COR IV
Jolff-Kishner
> R- (CH2 ) n+2-COOH
VII
use
The advantages of this debenzylation synthesis are said to be several. First, the yields are usually "J0% or higher: secondly, the "chain extender'1 is a malonic ester which ra^y be obtained in several ways and; thirdly, the intermediate reactant (IV ) is rendered solu- ble by three benzyl groups so that subsequent reaction can be car- ried out in not too dilute solution.
This method has been applied to the synthesis of straight chain acids containing l'4, lo, 23, 3&» on^- 5& carbons.
Ames and Bowman *5 have shown that long-chain unsaturated acids may be synthesized by condensing (II) with an a-alkcxy acid chloride followed by debenzylation, reduction to the glycol by us of aluminum ijso-propoxide, conversion to the dibromide, and final introduction of the double bond by the use of zinc in ethanol.
By a suitable choice of starting materials and using the route outlined above, the authors were able to synthesize 9-~methyloctad.ec- 9-enoic acid, the 12 -methyl and the 5 j7jl3,17-^e"kr-?Ine'-;hyl analogues of the same acid,3
The third method of producing long-chain fatty acids is not a new one, but is simply a modification of the well-known Kolbe elec- trolysis. In the ordinary Kolbe electrolysis, the salt of an acid is electrolyzed to produce the hydrocarbon formed by coupling two hydrocarbon radicals.
e~ 2EC00" -5- R-R + 2C02
By using the half ester of a dicarboxylic acid, the ester of another dibasic acid having 2n-2 carbons is formed.
e~ 2K00C-(CHa)n-C00CaHs -* HsC2OOC- (CHa )n— (CH2)n-COOC2H5
-5-
7 10
In the procedure advocated by Greaves et al. ' a mixture of a saturated carboxylic acid with the half ester of an cc,u;-dicar- boxylic acid is electrolyzed.
RCOOH + HOOC-(CH3)n-0OOMe -> R-R + R- (CH3)n-C0GMe
I II
+
Me00C-(CH3)3n-C00Me + CH3^CH-(CH3)n o-COOMe
To be sure, all the expected products of such an electrolysis are formed, but the use of absolute methanol as solvent and a high concentration of (i) favors the formation of (II) .
A wide difference between the sizes of the coupling units can be tolerated. Stearic acid has been made by coupling a C5 to a C13f a C9 to a C9, and a C17 to a Gx residue. Therefore by a careful choice of reactants, a product is formed which is uncontaminated by substances of the same or very similar molecular weight, so isola- tion of the desired compound in a pure state is simplified.
BIBLIOGRAPHY
1. Ames, D. E., Bowman, R. S. end Mason. R. G., J. Chem. Soc., 1950.
17^.
2. Ames, D. E. and Bowman, R. E., ibid. , 1951, 1079 . "5. Ames, D. E. and Bowman, R. E., ibid . , 1§51 , 10&7.
4. Ames, D. S. and Bowman, R. E., ibid., 1952, 677.
5. Bowman, R. E., ibid., 1952, 177.
6. Bowman, R. E. and Mason, R. G., ibid., 1951, 27^3.
7. Greaves, i. 8., Linstead, R. P., Shenhard, b. R., Thomas, S. L. S. and V/eedon, B. C . L. , ibid., 1950, 3326.
g. Howett, C. L., ibid. , TJJF, 50.
9. Klingenfuss, M., Festschrift Emil Barell, Basel, 1936 , 217.
10. Linstead, r. P., Lunt, J. C, and ieedon, B. C. L., J. Chem. Soc.
i252, 3331-
11 . Schw
12. Si
13. Ste-
14. SJ
ABNORMAL REACTIONS OF "HETEROCYCLIC GRI GUARD REAGENTS
Reported by G. \'tm Par shall
March 15, 1953
Since the development of the "cyclic reactor" which facil- itates the preparation of Grignard reagents from extremely re- active alk^l halides, 1 the preparation of several heterocyclic
In marc! re-
gnificance since endent
analogues of benzylmagne ■-dum chloride has been reported, their reactions with carbonyl compounds, these new Grign_. agents have been found to undergo ally lie rearrangements similar to these which have been observed in the benzyl series.3 These "abnormal" reactions have acauired particular significanc- it was observed that the extent of rearrangement is deper of the aromaticity of the system being studied.3'4
Grignard Reagents Trith the Thioohene Nucleus Both 2-
and 3-(chloromethyl)-thiana;ohthene yield stable C-rignard reagents in the cyclic reactor, but when these ere allowed to react T-rith carbonyl compounds, the products are predominantly abnormal ?* The reactions of 2-thianaphthenylmethylmagnesium chloride (I) with carbon dioxide, acetyl chloride, formaldehyde and benzoyl- durene are shown below.
j\
V^
'/
Dur
I II
COsH -0H3
I jpr -
ccrcr
OH-
3-Thianaphthenylmethylmagnesium chloride (II) behaves sim- ilarly in that rearranged products are obtained from its reactions with ethyl chloro carbonate and formaldehyde and the unrearranged product is obtained with benzoyldurene. However it differs in that a mixture of acids is obtained when it is carbonated, ratio of the rearrangement product (3-methyl-2-thianaphthenoic acid) to the normal product (3-thianaphthenylacetic acid) is 3.5 to 1.
The
One mechanism which; has been Postulated for this type of
is illustrated in the carbonation of 3-thianaph" thenylmethylmagnecium chloride (II).
rearrangement6
-2-
II
-CHa!teCl
CO.
CH2 p-
III
In an attempt to isolate an "igoaromatic" product corres- ponding to the intermediate (III), G-aertner treated 2-chloro- methyl-3-methylthianaphthene (V) with magnesium in the cyclic reactor and carbonated, the resulting solution. However only a trace of an organic acid could be isolated from the reaction mix- ture. The main product was 2,3-dimetrylthianaphthene (VII) which aooarently resulted from a cleavage reaction similar to that pre-* viously observed with 2-(chloromethyl)-benzofuran.7 The inter- mediate p.-(a-rnethylallenyl)~thicphenol (VI) could not be isolated but the corresponding thiolacetate was obtained by treating the reaction mixture with acetyl chloride.8
ng_
/-CH8C1
OH,
^
<eaaC=CH3 TT _
-^•3e MgCl1
CH3 C=C=CH3
3H
V
VI
VII
The reactions of 2— thenyl magnesium chloride are very similar to those of 2-thianaphthenylmethylmagnesium chloride (I) except that a mixture of normal and rearranged acids is obtained. TThen it is carbonated. The normal product, 2-thienylacetic acid, ore- do m i na t e s in th e mi xt vr e • 9
When the Grignard. reagent prepared from 5-me thy 1-2- thenyl bromide is carbonated, only the rearranged product, 2, ^"dimet'-iyl- 3-thenoic acid, is obtained.10
Grimard. Reagent s wit.^ the Fur an Hue has succeeded in preparing a. Grignard rea halide since t^e^e f?~halo ethers undergo treated with magne sium.7 > 1 1 In contrast a Grignard reagent in Jlft yield by conven 3-furfurylmagnesium chloride is carbonate acetic acid and 3-methyl-2— furoic acid is the rearranged product, constitutes appro tur e . 3
Zeus To date, no one
,gent from an a-furfuryl cleavage when they are 3-furfuryl chloride forms tional methods. When d, a mixture of J-furyl-
produced. The latter, ximately JOf of the mix-
An "i so aromatic" product is obtained when the Grignard reagent prepared from 3~chloromethyl-2-methylbenzofuran (VIII ) is treated, Trith ethyl chlorocarbonate. This product, 2-methyl- 3-methylene-2,3-d.ihydro-2-benzofuroic acid. ( IX) , is also obtained when the Grignard. reagent is carbonated, but carbonation yields in addition a trace of the normal product, 2-methyl~3-benzofuryl- acetic acid.12
0HSC1
T.T
IS.
*
CHsMgCl
CO
^
Relationship to Aromatic. Character It has been observed
that aromatic systems posserjainj? high resonance energies have little tnedency to undergo the type of rearrangement ^escribed in this muer. This tendency Hag been quantitatively expressed in terms of the proportion of "abnormal" acid produced upon car- bonation of the Grignard reagent. The table below indicates the possible relationship involved. The a-picolyl Grignard reagent is placed above benzylnagnesium chloride because the latter under- goes rearrangement in its reaction with acetyl chloride while the former does not.13
Grignard RGagent
a-Picolyl
Benzyl
2-Thenyl
3-Thianaphthenylmethyl
3-Furfuryl
2-Thianaphthenylmethyl
Propor ti |
.on |
or |
jrv.e so nance |
abnormal |
acid |
energy |
|
0 |
J+3 kcal./mol |
||
0 |
39 |
||
33 |
31 |
||
7S |
|||
90 |
23 |
||
100 |
— |
BIBLIOGRAPHY
1. |
D. |
2. |
117 R. |
3. 4. |
E. (19 R. |
7. 2. |
R. J. " R. R. |
9. 10. |
R. J. |
11. |
H. |
12. |
R. ' |
13. |
H. |
C. Rowlands, K. !f. Greenlee and C. E. th A. C. S. Meeting, Philadelphia, Pa
Boord
, Org. and E
0. Kerr
Sherman
50).
Gaertner, ibid. t jK,
Gaertner, ibid., 55,
R. Johnson, ibid.
Gaertner, ibid. ,
G-aertner, ibid. ,
Gaertner, ibid.
» -
r>
hi
Seminar, Univ. of Illinois
D. Amstsutz, J. Am. Cbem.
(Nov. Soc.,
Abstracts, (April 1950)
2, 1951). 12, 2195
>
Z2,
Buu
21^5 (1952).
766 (1952).
sr 3029 (1933). ij4oo (1951).
29°l (1952). 393I1 (io51)#
Hoi, Compt. rend.
Lecocq and. N. P.
Normant, Bui. soc. chim. 'France, G-aertner, J. Am. Chem. Soc., jk] G-ilman and J. L. Towle, Rec. trav.
22^
(5) 12
(1952). chim., o2,
5319
6Rg (19^-7). (19^5).
^23 (1950).
THE T,/ILLGERODT REACTION Reported by S. L. Jacobs March ?0, 1953
A reaction (I, see below) In which a ketone may be transformec into an amide with the Same number of carbon atoms was first de- scribed by Wiligerodt in 1887. x The reagent generally used for this purpose Is ammonium poly sulfide, prepared by saturating con- centrated aqueous ammonia with hydrogen sulfide and dissolving in the solution 1C# by weight of sulfur. A modification (II) of the reaction was discussed by Kindler in 1941s which Involved substltu^ tlon of a mixture of a dry amine and sulfur for the aqueous (NH*)g v§x to ob tain th leram Ides. A further and widely used modification ©f Kindler' s procedure was developed by Schwenk and Bloch5 who utilized morpholine as the amine. The Wlllgerodt and Kindler re- actions have been extended to the aryl- substituted acetylenes and olefins to yield carbonamides and thloamides (ill).4
(I) j6C0(CH3)nCH3 (NH4)8SX . jtaH2(CH2)nC0NH3
H30 '
(II) j6C0(CHs)nCH3 HNH^ L__ ^ ^CH3(CH3)nCSNR3
(III) j6CH:CHCH3 ( -MH^m^l ^ 0CH3CH3CONH
Kindler >> jj5CH30H8CSNR
Addition of pyridine4 or dloxane5 as solvent allows the reaction to proceed at a lower temperature so that side reactions are minimized.
Several reviews of the Wnigrerodt reaction are available which cover the literature up to 1948.'e>7'8 This seminar will review some of this work and summarize that which has been done since.
The reaction aDt>lies both to aryl-alkyl ketones and to com- pletely aliphatic ketones where there is a tendency to preferent- ially produce the amide group at the end of the shorter chain.9
CH3CH3CH3COC^H3 -J^±l^\ C14H3 (CH3)3CONH3 + OH, (CH3)3C140NH2
H30 Total yield = ZOf0 (3$*) (66f)
It has been observed that in the case of certain ketones, the yield drops drastically with reaction temoeratures above 16Q°C. This has been shown in some cases to be due to instability of the product (e.g. 2-thienylacetamide).10
It is currently the opinion of all workers that these react- ions, whether Wlllgerodt or Kindler, and whether starting with ketones, olefins, acetylenes, or even alcohols, halides, amines or thiols, all proceed according to the same mechanism Involving the preliminary formation of a labile intermediate with an unsaturated C-C bond in the chain.4' lx 'ls> »■»» l« In the case of ketones, this linkage is pictured as originally being located adjacent to the carbonyl group of the ketone through enolization of the alpha- hydrogen. A shift of this bond towards the terminal carbon occurs
-2-
through successive additions *nd eliminations of an un symmetrical reagent. According to Carmack, et al., the intermediate is acetyl- enic; this cannot be considered likely since hranOiRd-cVi^in com- pounds as pC0CHsCH(CK3)2 are known to glre ths ejected grmi&e, j6CH3CH2CH(CH3)0ONH2, with no loss of carbon. Also, /&'!OCHsCDaCH8CHs has been shown to retain some of its deuterium after having under- gone the reaction.11 All the deuterium would have been lost were the Intermediate ace tylenic It may be noted here that the re- action will not proceed with a chain containing a quaternary carbon atom. McMillan and Kins,13'14 however, are of the opinion that the intermediate is olefinic. They believe that hydrogen sulfide is the specific unsymmetrical reaerent that causes migration of the olefinic bond to a terminal position and finally "the formation of a primary thiol which is irreversible oxidized by sulfur (the amine is also involved here) to the thioamide which remains as such in the Kindler modification or is converted to the carbonamide in the reactions which are run in aqueous media. Some retention of deuterium by the ketone j6C0CH2CD2CH2CH3 is found as would be ex- pected from an olefinic intermediate. It has further been shown that there is no rearrangement of the carbon skeleton during the Willgerodt reaction,18 contrary to results previously reported.15
The reaction might best be described according to the following scheme in the light of the information available to date:
#C0CH(CH9)3S -* j6CH0HCH(CH3)s
U^ )6C0H:C(CH3)S ^^ j6CH:C(CH3)
^^^ $CH2CSH(C._ J6CSCH(CH3) 3 ~ ± j6CHSHCH(CH3)s -£r j6CH2C (CH3) : CH
j6CH2CSH(CH3)
j6CH2CH(CH3)CH2SH~ Then, if R = j6CH2CH(CH3 )-, as in the example above —
2 RCH2SH + S— ^ RCH2SSCH2R + HSS (well established)
RSNH + S->R'2NH^S *->R'2NSH (Rr2NH= Moroholine)
Both the amine and the sulfur are necessary for the next sten —
R'8NSH + RCH2SSCH2R ^r1 RCHSSCH2R + H2S
NRfs RCHSSCH2R + R'2NSH > H2S + RCHSSCHR 2 R>sTO > 2 rch(nr,s)s
RCH(NR»2)3 + S >RCSNR»2 + HNR'2
The overall equation for this irreversible oxidation would be:
RCH2SH + R'2NH + 3 S ^ RCSNR' 2 + 2 H2S
It has been shown that compounds other than ketones, olefins, and acetylenes will give the predicted product in the Wnigerodt Reaction. In the following table, the indicated yields of phenyl- acetamide were obtained using yellow ammonium poly sulfide in dioxane in sealed tubes at 1V0°C. for seven hours.16
-3-
1-Phenylethylamine 61%
1-Phenylethyldimethylamine 31 1-Phenylethyl- (monoethanol)-amine 63 1-Phenylethyl- (diethanol)-amine 66 Phenaoylpyridinium Iodide 53 £y*Morpholinoacetophenone 72
2-Phenylethylamine 1-Phenylethyl bromide 2-Phenylethyl bromide Styrene oxide P-Bromostyrene
32?
40
66
87
80
These compounds are all very similar to postulated intermediates o. either Carmack or McMillan and King, or they may very easily be converted to these intermediates.
The Wnigerodt reaction has been apolied to a series of mercaptans, primary and secondary alcohols.17 The following table illustrates the results obtained using five Darts by weight of aqueous ammonium poly sulfide solution at 210°C. for 5-16 hours.
Starting Material
Ylelff
Product
EtSH PrSH BuSH C q Hi *f SH
Cio^si SH CsHsCHgSH
CgHg^HgOHgSH
Me3CHCH3SH
Me2CHSH or CH3:CHCH3SH
C6H5CH(CH3)SH
Me3CHSH
CH3:CHCH30H Me3C0H or Me3CSH C6HECH(CH3)0H Me3CHCH(C6H5)0H EtaCHOH
100< 53 95
44 34
48
AcNH3
EtC0NH3
PrC0NH3
caprylamlde
capramlde
C6H50ONH3
C6H5CH3C0NH3
Me3CH0ONH3
EtOONH3
C6H5CH30ONH3
EtC0NH3
EtC0NH3
Me3CHC0NH3
C6HsCH3C0NH3
C6H5CH3CH(Mc)C0Nn3
BuCONHg
The action of 3,5 grams of sulfur and 25 grams of yellow ammonium poly sulfide in 25 ml,- of dioxane on 5 grams of various thiophene derivatives in sealed tubes at 150- 160*0. gave the following re- sults:10'18'19
Substrd Thienyl Thlenyl Compound Amide Obtained Yield
2, 5- Me 2- 3- thienyl Me Ketone 5-Et-2-thienyl Me ketone 5- Me- 2- thienyl Me ketone 3 , 4- *Ie 3- 2- thienyl Me ketone 3- Me- 2- thienyl Me ketone 2, 3- Me a-5- thienyl Me ketone 3-thienyl Me ketone 2- thienyl Me ketone 2- th leaiy 1 a c e t one 2-vinylthiop'hene 2-thienylc=«roo-xoldehyde 2-thienylmetnylo.arbi-o.l
i k—
3-Thienylacetamide 95^
2-t?iienylacetamide 55
2-thlenylacetamide 54
2-thienylacetamide 30
2-thienyla.cetamlde 26
5-thienylacetamide 55
3-thienylacet«mide 13
2-thienylacetamlde 45
2-t/.ienylpro"Dionamlde 28
2-tliiei.y J ace tamide 30
2-thienylearboxamidc 70
2-thienylacetamide 35
-4- S
u
Thloamides such as ArGH3CNBls (NR'3 = moroholine) were prepared from the following ketones containing an aryl radical substituted with hydroxyl, nitro, amino, or acylamino groups:30
o-(OH)C6H4COCH3 p-(OH)C6H4COCH3
3,4- (0H)sc6H3C00H3 m- (H3N) CsH4COCH3
p- (H3N)C6H4C0CH3 m- (CH3C03) C6H4COCH3
m- (CH3CONH) C6HaGOCH3 o- (CH3CONH) C6H40OCH3
p-T0H3CONH)C6H4COCH3
Some completely aliphatic ketones that have undergone the WiHgercdr Reaction with (NH4)33X are:
Ketone Amide Ref .
CH3CH3COCH3 CH3CH3CH3CONH3 (9)
CH3CH2CK3GOCH3 CH3CH3CK3CH2C0NH3 (9)
GBHiiGOCHg CH3{CK3)5GONH3 (9)
(CH3)3CHCH3COCH3 (CH3) 3CHCH2CH3CONH3 (2l)
CH3CH3GOCH3CH3 CH3(CH3) BGONH3 (2l)
Piperazine and sulfur have been used for the reaction and give a product of the form
R-CS-N/~\-CS-R ,3S
o
Some thloamides which are formed in the Willgerodt-Kindler reaction are unstable to acids or alkali which are used for hydrolysis to the acid. For such compounds a method has been developed for thiomorphollde breakdown without disruption of the entire molecule:83
r/"""*V onVnr^ flU.T A~„ ~ ../"""V
)6CH3CSN/ 0 anhyd. GH3I . j6cH:C-n' D ..
W -u~ 7T^ 7 Y \— / tit reflux v.
v-r heat ' ftm* -HI -Sfim wh:>
SCH3 ' n± with NRV
heat
j6CH3CONIL
H30
$CH3C-SCH3 li 0
BIBLIOGRAPHY
1. Willgerodt, Ber. 20, 2467 (1887); 21, 534 (1888).
2. Kindler and Li, Ber. 74, 321 (l94lJ7
3. Schwenk and Bloch, J. Am. Chem. Soc. 64, 3051 (1942).
4. Carmack and DeTar, ibid.. 68, 2025, 2029 (1946) .
5. Fieser and Kilmer, ibid., 62, 1354 (1940) .
6. Carmack and Spielman, "Organic Reactions" Vol. Ill, 1946.
7. Leubner, Organic Seminar, Nov. 8, 1946.
8. Caesar, Organic Seminar, March 18, 1949.
-5-
9. Cerwonka, Anderson and Brown, J. Am. Chem. Soc. .75, 28 (1953).
10. Blanchette and Brown, Ibid.. 74, 1066 (1952).
11. Cerwonka, Anderson and Brown, ibid. . 75. 30 (1953).
12. Brown, Cerwonka and Anderson, lb Id . , 73. 3735 (l95l).
13. King and McMillan, Ibid.. 68, 63? (1946) .
14. McMillan and Kiner, ibid.. 70, 4143 (1948).
15. Dauben, Reid, Ynnkwich and Calvin, ibid., 72, 121 (1950) .
16. Gerry and Brown, ibid.. 75, 740 (195377
17. King (to WintbroD-Stearns, Inc.) U.S. 2,459,706, Jan. 18, 1949 C.A. _43, 3028b (1949) .
18. Blanchette and Brown, J. Am. Cbem. Soc. 73, 2779 (l95l).
19. Brown and Blanchette, ibid.. 72, 3414 (l950).
20. King and McMillan (to Win throo- Stearns Inc.) U.S. 2, 568, Oil, Sept. 18, 1951; C.A. 46, 3081b (1952).
21. King (to Winthroo- Stearns Inc.) U.S. 2,456,785, Dec. 21, 1948; C.A. 43, 30271 (1949).
22. Chabrier and Renard, Comot. Rend. 228. 650 (1949) .
23. Rogers, J. Chem. Soc. 1950. 3350.
RECENT STUDIES ON THE DECOMPOSITION OF BENZOYL PEROXIDE Reported by James C. Kauer March 20, 1953
Benzoyl and related peroxides have been widely used as source of free radicals for the initiation of chain reactions. These peroxides have been recently studied for possible synthetic appli- cations.
The kinetics of the thermal decomposition of benzoyl peroxide in various solvents was systematically studied in 1946. 1 It was found that the rate of decomposition of the peroxide could be re- presented by
-{©- V4^1"
where k was the first order rate constant due to the spontaneous decomposition of the peroxide, and ks was a higher order rate constant representing the induced decomposition of peroxide by secondary radicals.53'3
l) (j6C00)s &i ^ 2 ficoo*
S) )6C00« ^ p* + C03 RH= Solvent
3) j6C00* + RH ^ j6C00H + R*
These secondary radicals could attack the peroxide to initiate a chain decomposition.
4) t>* (or R*) + (/>C00)3 3*3 ^ j6C00/3 + ^COO- It was believed that a primary decarboxylation reaction might
also take place.
5) (jOCOO)3 £t ^ j6C00<*> + C03 + />•
Recent work has tended to disprove this.4
The Decomposition of Substituted Benzoyl Peroxide sls* 5
Recently kinetic studies of the decomposition of symmetrically substituted dibenzoyl peroxides have been run on dilute peroxide solutions in acetophenone. Under these conditions the Induced decomposition was inhibited; the reaction was first order in peroxide.
It was found that the ortho- substituted peroxides decomposed at a much higher rate than the meta- or para- substituted. This effect was probably due to the electrical repulsion of the sub- stitutent groups.
Electron releasing groups in the meta and para positions were found to increase the rate of decomposition. This was attributed to an increase in the repulsion between the carboxyl diooles. Electron withdrawing groups had an opposite effect, although a minimum was reached. Very strongly electronegative groups seemed
-2-
to reverse the trend. The bls-p-nltrobenzoyl peroxide, for in- stance, decomposes at a rate very close to that of the unsub- stituted peroxide. It has been suggested that this enhanced reactivity may be due to a reversal in the dipole direction re- sulting in increased repulsion between the carboxyl groups.
Decomposition of Benzoyl Peroxide in the Presence of Iodine
In 1945 it was reported that benzoyl peroxide reacted with olefins in the presence of iodine to form dibenzoates.
6) fi
H H
P n n
c = cr + 06coo)s + i2 ,vC1..4 — ^ $— c c— jb
$C00 OOGp
azfc
With cyclohexene not only was the dibenzoate isolated but also the iodobenzoate . This latter reacted further to produce the di- benzoate. The reaction suggested that benzoyl hypoiodite (iodine (i) benzoate) might be the reactive species. nrjOO 00C&
7) ^=y ^co01 > <> — <> ^ooi>
Hammond has recently observed that benzoyl peroxide will react with iodine In the presence of carbon tetrachloride to oroduce high yields of iodobenzene.
8) (j6COO)3 + I a 0Cl4 x /?~ ~^>I + j^COOH + $C00)6 + C0S
79° ' X — r/
A. If 3 . 1#
The iodine effectively inhibits the induced reaction by removing the initially formed radicals from solution. The reaction is first order in peroxide and independent of iodine concentration. The utilization of iodine seems to be abnormally high at first. An intermediate capable of reducing thiosulfate seems to build up in the reaction. The following mechanism is proposed:
9) (0COO)3 ^ 2 /taoo*
10) j6C00- + Is ^ $C00I + I«
li) j6cooi > j6i + co2
12) 81 • > I2
It is known that benzoyl hypoiodite is readily hydrolyzed by water. The following reactions were run:
-3-
15) (j6C00)s + I2 + 8so
CCl.
16) (j6C00)
79'
+ HoO
CCl
-^ 0COOH
(Quantitative)
^ #(31 + C13CCC13 (No Benzoic 790 ^ + co3 Acid/
The observed results would be eroected on a basis of benzoyl hypolo&lfce as an Intermediate. The quantitative yield of benzoic acid in reaction 15 also indicates that decarboxylation does not occur In tbe Primary orocess of thermal decomposition (see reaction 5) but is strictly a secondary reaction.
When reaction 8 was run in benzene, 10fc decarboxylation of the benzoate radical took place.
Decomposition of Benzoyl Peroxide in the Presence of the Trlphenyl-r methyl Radical
In 1937 Wieland reported that dlbenzoyl oeroxide decomposed in the presence of triphenylmethyl. e' 9
17) j6sO + (j6C00)
<=>
R
fi3G
-<=>
0 R + p3C0C-p
+ 0COOH
This reaction took place rapidly at room temperature. One of the aryl groups of the product tetraarylme thane was derived from the solvent. Hammond repeated this work and found that the yields of tetraphenylmethane ranged from SOf to 30f on a basis of tri- phenylmethyl.10'11 Since the reaction proceeds at room temperature, it seems unlikely that the spontaneous decomposition of oeroxide occurs to a significant extent. The reaction is attributed to the induced decomposition of oeroxide by trityl radical.
18) T* + (j&COO)s
19) /6C0O« + T> -
-^ j6C00T + $COO<
20) )6C0O + ArH
21) Ar- + T. -
-> 0COOT
ArH = Aromatic Solvent
T
>
$COOH + Ar* ArT /
v
-4-
This reaction differs from other reactions in which benzoate radicals are postulated as intermediates in that no decarboxylase seems to occur. Even when the reaction is carried out at elevated temperatures no carbon dioxide is oroduced. If benzoate radicals are Involved in both mechanisms, the only significant difference in their environment is the nature of the other radicals in solution. If radicals influence these reactions,, they must do so in concerted reactions in which two radicals attack the solvent simultaneously. One suggested explanation is that reactions 20 and 21 occur as a concerted reantion:
22) T- + ArH + /6C00* > j^COOH + ArT
An argument based on a study of the ratio of ester to acid producec* appears to exclude this possibility.
Another explanation has been advanced. It seems possible thai when benzoate radicals are formed in nairs in the thermal decom- position reaction, these radicals may make concerted attacks on a solvent molecule while they are held in close proximity in the "solvent cage". The benzoate radicals produced by the attack of trltyl radicals on r>eroxide are single entitles, however, and in solution rarely last long enough to make a close enough approach t< each other to make a concerted attack on a solvent molecule.
This may also explain the differences observed in the reactiv- ity of certain other radicals which are apparently identical in nature but differ in the method of generation.
BIBLIOGRAPHY
1. Nozaki and Bartlett, J. Am. Chem. Soc, 68, 1686 (1946) .
2. Hartman, Sellers, and Turnbull, JJbid., 69, 2416 (1947).
3. Barnett and Vaughan, J. Phys. Coll. Chem., 51, 926 (1947).
4. Hammond and Soffer, J. Am. Chem. Soc, 72,, 4711 (1950).
5. Blomquist and Buselli, ibid.. 73, 3883 Tl95l).
6. Perret and Perrot, Helv. Chim. Acta. 28, 558 (1945).
7. Hammond, J. Am. Chem. Soc. 72, 3737 (l950) .
8. Wieland, Ploetz, and Indest, Ann. 532, 166 (1937) .
9. Wieland and Meyer, Ibid., 551, 249~TL942) .
10. Hammond and Raave, J. Am. Chem. Soc. 73, 1891 (1951 ).
11. Hammond, Rudeslll, and Modic, ibid.. 73, 3929 (l95l).
12. Swain, Stockmayer, and Clark, ibid., 72, 5426 (1950).
THE REACTION OF ortho-HALOBENZOIC ACIDS WITH NUCLEOPHILIC REAGENTS Reported by Harry J. Neumlller March 20, 1953
I. HISTORICAL INTRODUCTION -
The high reactivity of the halogen substituent in ortho-chlorr benzoic acid in the presence of a copper catalyst was discovered by Ullmann6' 7 in 1903. This has since been extended to include the copper catalyzed reactions of orthc—halobenzolc acids and a variety of substituted ortho-halobenzoic acids with a large number of nucleophillc reagents. The general formal reaction is given by
(i)
MA
■>
COOH
V
MX
where X in most cases is Cl, Br, (rarely l); A is -NHAr, -OAr, -NRAr, - NRH, -NR8, -OR, -NHS, -OH; and M is generally H, Na, K,
II. SYNTHETIC VALUE
The products obtained with aryl amines can be cyclized to yield acridone derivatives (l)„ reduction of which gives acridlne derivatives (II), or the Products can be cyclized directly to acridines. Important derivatives of seridine Include dyes, bactericides, and the antimalarial drug atabrine, the use of this reaction in the synthesis of atabrine11 being perhaps its most important commercial application to date.
Cyclization of the products obtained with phenols leads to xanthone derivatives (ill). Substituted anthranllic acids result from treating ortho-chlorobenzoic acid derivatives with ammonia. 3 An early method9 for hydrolysis of substituted ortho-chlorobenzoic acids to salicylic acid derivatives by treatment under pressure with water, lime, and copper powder has experimental disadvantage sV This hydrolysis is now accomplished in a recently described3 im- proved general procedure by treatment with aqueous K3C03 in the
Cul and copper powder at 150-180u and 70-130 p.s.l.
H
kV
Acridone I
Acridlne II
Xanthone III
-2-
III. CATALYST
The reaction (l) requires the presence of a metallic catalyst The most effective and generally used are metallic copper, cuprous or cupric salts, or some combination of these. Ullmann8 found tha salts of iron, nickel, zinc, lead, and platinum (listed in order of decreasing effectiveness) would also catalyze the reaction, but not so effectively ss copper or copper salts. Salts of manganese and tin were found not to catalyze the reaction.
The amount of catalyst required is exceedingly small, 8 X 10 g. of copper in the form of copper sulfate being sufficient to glv< a 97^ yield of product in the reaction of 1.6 g. of ortho- chloro- benzoic acid with aniline. Careful purification of starting materials, however, showed that this same reaction would not occur without the presence of a metallic catalyst.8
IV. MECHANISM
In addition to the need for the presence of a metallic catalyst, other facts which must be considered in proposing a mechanism for this reaction are the stability of the ortho- chlorine substituent in ortho- ohlorobenzoio acids in the presence of high hydroxy 1 concentration, a and the fact that the ortho- chlorine sub- stituent in 2,4-dichlorobenzoic acid reacts with nucleophllic reagents to the complete exclusion of the p_ara- chlorine sub- stituent.3'11' 1>3. In evaluating the latter information, however, 1* should be observed that apparently anomalous "ortho effects" also occur in reactions of other negatively substituted aryl halides with nucleophllic reagents, an example being the reaction of 2- aminoethanol with 2,4-dichloro-l-nitrobenzene to give an 88^ yield of 2-(5-chloro-2-nitroanilino)-ethanol.5
Bunnett and Zahler1 have proposed an ionic mechanism in which the copper, reacting in the cuprous oxidation state, coordinates with the halogen substituent, converting it to an onium state, It is postulated that this increases the reactivity of the halogen substituent, by analogy with the enhanced reactivity of stable onium compounds, such as ammonium compounds, with nucleophllic reagents. The complete scheme is given by
X X-Cu \ X-Cu Y
(2)
Cu
:±>
*
CuX
where X is halogen and Y is a nucleonhilic reagent.
This representation does not account for any of the previously mentioned peculiarities of the reaction. In view of this, G-oldberg has proposed that the reaction proceeds by way of a non-ionized six-membered copper chelate complex (Fig. 1.). It is postulated
-3-
that by thereby Including participation of the carboxyl group, the greater reactivity of an ortho-halogen substituent over a para- halogen substituent is explained. In this connection it is also of interest to observe that ethyl .o-bromobenzoate, P_-bromobenzoic acid, and jo-bromonltrobenzene will not react with the sodium derivatives of certain active methylene compounds, in the presence of a copper catalyst, under the same conditions that cause o-bromo- benzoic acid to react,4
C
I
Fig. 1.
0
L
rv
An attempt is made to correlate this proposed mechanism with the previously described effect of high hydroxy 1 concentration, by comparing the yield of ?-carbory -4*- methyldiohenylamine from the copper catalyzed reaction of o_-chlorobenzoic acid and p_-to- luidine in amyl alcohol, with the stability of an amyl alcohol solution of the chelate copper complex of ac e tylaoetone3 (rv). The effects of adding equivalent amounts of dry K2C03 (insoluble ir amyl alcohol), equivalent amounts of dry KOH (soluble in amyl alcohol), and excess aqueous K2.C03 were measured. The results ob- served are summarized in Table 1. The results lend some support to the proposed mechanism, but cannot be accepted as a complete proof of it.
Table 1.
Reagent Added
Yield of Reaction
Equiv. amt. of dry K2C03
Large amt. of aqueous K3C03
Equiv. amt. of dry KOH
85#
| Effect on Stability | of Acetylacetone
:-_r__rr
Complex
Completely stable, ' even on prolonged heating. ______
Yield decreases as amt. of aq. KsC03 is increased
Slow decomposition.
No product obtain- ed;' 92f recovery of v starting acid.
Immediate, complete; decomposition.
-J
_4~
BIBLIOGRAPHY
1. J. F. Bunnett and R. E. Zahler, Chem. Revs. 49, 392 (l95l).
2. H. Diehl, Chem. Revs. 21, 63 (1937) .
3. A. A. Goldberg, J. Chem. Soc . 1952. 4360.
4. W. R. H. Hurt ley, J. Chem. Soc. 1929. 1870.
5. C. B. Kremer and A. Bendich, J. Am. Chem. Soc. j51, 2658 (1939)
6. F. Ullmann, Ber. 36, 2382 Cl903) .
7. F. Ullmann, Ber. 37, 853 (1904).
8. F. Ullmann, Ann. 355, 312 (1907).
9. F. Ullmann and C. Wagner, Ann. 355, 359 (1907) .
10. E. Wenia and T. S. Gardner, J. Am. Pharm. Assoc. 38, 9 (1949).
U. British Patent 353, 537, Apr. 30, 1930, [C.A. 26, 5311 (1932)]
12. German Patent 244,207, Mar, 2, 1910, [C.A. 6, 2293 (1912)] I
SOME BASE CATALYZED REARRANGEMENTS Reported by Y. Gust Hendrlckson March 27, 19F3
I. Chlorohydrlns. The treatment of chlorohydrlns _with> base^
usually gives epoxides. Thus, from trans-g-chlorocycloocntanol, prepared by the addition of hypochlorous acrid to 1- me thy level o- pentene; l-methylrvy.;looentene ovide is obtained. However, the els isomer, obtained by adding methylmagnesiun bromide to ^-c^lor; cyclopentanone, rearranges to 2-methylcyclopentanone1 .
Hn
Y
r*v
33* NaOH
in Hs0
—5^,
H
y
^n3
N
0
Similar reactions have been observed with the cis isomers of g-chloro-1-methyicyclohexanol s, ^-chlorocyo'lorex^nol* and 2-chloro-
1-indanol4; yielding respectively, methyl cyolopeiityl ketone, cyclohexanone and i-lndanone. The expected epoxides are obtained with each of the trans isomers. On treatment with sodium methoxite in methanol, the monotosylate of j£i_e-l,2-cyclopentanediol drives cyclopentanone, while the trans isomer gives oyclooentene oxide5.
These reactions probably proceed by hydroxyl proton by the base; followed by e:roup or hydrogen, with its pair of elec simultaneous exoulsion of chloride or to all reaction is very similar to the acid rearrangements. The significant differe base catalysed reaction requires a negat state while the acid catalyzed reaction charged transition state. The two react expected to show differences in the effe in migration altitudes.
the abstraction of the the migration of an alkyl tron^; with subsequent or sylatc ion6'7- The over- cat a ly zed 1;«ragne r- Me e rwe in nee, however, is that the ively charged transition involves a positively ions might; therefore, be cts of substituents and
!!• Halides. Early workers attempting to clarify the isomerism
and structures of teroenes came across the rearrangement of bornyl and isobornyl chlorides to camphene, under comparatively mild conditions, with Potassium and calcium hydroxides andnnillne8' 9' 10 ,
rV1
»
i i
A similar rearrangement wag recently observed by Gone and Fenton .
a-"M collne
^
-2-
III. Benzlllc Acid and Related Rearrangements. Benzil yields
benzilic acid by a process, the
rate of which is first order in lonls. Under conditions which give I. Roberts and H.CG. Urey1^ have found
pi 8
both benzil and hydroxide
negligible rearrangement,
that benzil does exchange oxygen with solvent enriched in Hs<
(in neutral aqueous methanol," 43±6^ is exchanged in four minutes;
100*6$ is exchanged when this solvent is 0.02 N in sodium hydroxide
The experiment rules out a mechanism which involves the addition
of hydroxide ion to a carbonyl grcuo in the rate determining step.
In anhydrous ether, eouimolar ratios of benzil and potassium
hydroxide yield fll^ Potassium benzilate14. This fact, along with
the base catalysis of oxygen exchange noted above, lends supoort
to the mechanism given below.
©
n IT V 6 -5
0-
0 li
6H5
0U»H
C6K500C0C6H5 + qi er^J53X} i| f«et
06F5 0 HO C C-CSH5
© 018
Slow
> iC6H5)3C0HC03H
A recent study of p-methovyben,7ll using 014 has determined the approximate relative migratory aptitudes of Phenyl and p- methoxyphenyl groups15.
p_-CH30C6H4
,14
\9 = o jsi~m~
Phenyl >^
' shift ^
N^1 4
/
!r0.
C6HE
OHCO.H
-^
il 4,
p-0H3OC6H4
^C=?0 t CO,
p-CH30CsH4
,C— 0
0-H
6n5
CsH5
il 4,
anisyl v
shift ^ )P0HC**08H p-CK30C6H4
CrO
3 \
~7
C6H5 NC=0
+ o14o
p-CHa0C6H4
In experiments carried out at 25, 70 and 100°, the ratio of Phenyl to p-methoxyPhenyl migration was found to be 1.90, 1.72 and 2.17 respectively; as compared with 0.014 found in the acid catalyzed pinacol rearrangement3-6. A sizable isotope effect17, however, has been noted with C14.
^6% i4 ^CO
o6tt5
^OH
o'^o 1 II
OH
0 0<=> ,8u6~0aHs
OH
-\> (CeHB)aCOH0i*OaH I A)
-> (CsH5)3Cl40HC03K (3)
' J5-"
The ratio of A to B obtained was 1.11-0.01.
By heating with sodium t-butoxide in benzene solution, some aliphatic dike tones (RCOCOR where R is isooropyl, t-butyl and neonentyl) have been rearranged to the corresponding acids16, formation of the acid rather than the ester suggests the operation of a different mechanism.
The rearrangement of phenylerlyoxal to mpndelic acid, which takes place by an internal hydrogen shift, resembles the rearrange- ment of benzil19' 30 > 3l .
An cr-diketone, isolated from the reaction mixture by Baker and Robinson23, is thought to be an intermediate in the^alkaline rearrangement of benzylideneacetoDhenone oxide. C 4 has been used to determine benzyl migration351 .
14,
c6Hscyo
CH
l>0
CsH5CH
C6H5C^0
!
c=o
OH
e
C«H
Sn5
-> CL40
OeHgCHa C6H^CH3— C-
0
G)
OH
-^ 06H50H3
CeH5 • C^OH
!
c=o
I
OH
In a dilute solution of sodium hydroxide in aqueous ethanol, diphenyl triketone rearranges, decarbonylate s and cleaves to give, on acidification, benzoic acid, mpndelic acid (triketone cleavage products), ben7oin and carbon dioxide (rearrangement nroduct s) 34. Benzoyl migration and the loss of the center carbonyl have been demonstrated by J. D. Roberts and coworkers15 with the use of C14. The carbon dioxide obtained was found to be inactive T'rhen the triketone wfls labeled as shown below.
CeH5C = 0
?
:0
1 4,
06HBC^O
NaOH
0
h
n tj y 6 : : 5
4 c«hk c14 a'1 — oh !
'6"S
CO,
Na
1. H
+ C6H5Cx*0C
l*nn^J
H0HC*HP
^
?. -co;
CO.
IV. q- Hydroxy Ketones. Attempt s to saponify 17-acetoxy-°0-keto
steroids lead to the discovery of a rearrangement which expands the D ring. An examole studied by 3hopr>ee and Prlns35 is given below.
Ac0
OA,
M
OCH.
KOH
•>
H
^4-
Products obtained from the alkaline cleavage of a- substituted benzoins show that a similar rearrangement to isomeric a- hydroxy ketones can take place before splitting occurs. The reaction as pictured by Sharp and Millerse follows:
OH
I OH
C6H5-C-COC6H5 ^
R H*°
(C)
Q
R
0
II
■ C— C6H5
C6H5
0 0 (i f
£>
R-C-C fC6H5)s'
Hs0
0
-i
OH
,0
R-C-COHfCsHs);
(D^
(C)
OH
C«H,
^CHOH + C6H5C03H j D -^-> (C6HB)sCH0H + RC03K
Benzoin gives Products that Indicate about 27? rearrangement; the methyl and benzyl derivatives show respectively 47 and 40^ re- arrangement. Although the p-tolyl compound shows only 13f and the J2J-tolyl, 60-70^ rearrangement; only the rearrangement' products, in 98? yield, are obtained from cr-Phenyl and a-o-tolyl benzoins. The aryl compounds were cleaved in refluxinsr lOfmetb* nolle potassium hydroxide (20^ water); methyl and benzyl derivatives required 160°, hence diethylene glycol was used in place of methanol.
v» cc-Hqlo Ketones. — «• When a-chlorotetralones are treated with sodium methoxide, a ring contraction occurs37.
COoOH
3^x13
V"
Varying the conditions greatly alters the products formed when ct-halocyclohexyl phenyl ketone is treated with base3*.
C~ 0 6 Hs
base v s — > < S
C03H /
^6^5
<:k
0
l!
C— C5H5
In refluxine- dioxane, with no added based, 8^ acid and 60^ a, ^-unsaturated ketone were isolated. With finely divided
sodium
-5-
hydroxide vigorously stirred in ref luring xylene as much ?s 53/ rearrangement product and 95? P-hydroxy ketone were obtained. Sodium methoxide in boiling methanol yields the e^oyy ether.
BIBLIOGRAPHY
1. P. D. Bartlett and R. V. White, J. Am. Ohem. Son., 56, 2785
(1934).
2. P. D. Bartlett and R. H. Rosens Id, ibid., 56, 1990 (1934).
3. P. D. Bartlett, ibid.. 57, 224 (19357.
4. C. M. Suter and G-. A. Lutz, ibid.. 60, 1361 (1938).
5. L. N. Owen and P. N. Smith, J. Chem. Soc, 195?, 4026.
6. C. K. Infold, Ann. Repts., 25, 124 (1928).
7. C. R. Hauser, J. Am. Chem. Soc, 62, 933 (1940) .
8. 0. Walls oh, Ann., 23p, 233 (1885) .
9. 0. As chan, .ibid., 410, 222 (l915)«
10. A. Reychler, Bar., 29, 696 (1P96).
11. A. C. Cope and" S. W. Fenton, J. Am. Chem. Soc., 73, 1673 (1951,.
12 . F . K „ We s the Ime r , ibid . , 58, 2209 ( 1936 ) .
13. I, Roberts and H, C. Urey, ibid. , 60, 880 (1938).
14. T, Evans and W. Dehn, ibid.. 52, 252 (1930) .
15. J, L. Roberts, D. K. Smith and C. C. Lee, ibid.. 73, 618
(1951) .
16. R. J. Adams. Organic Seminar, Fall Semester 1952, p. 67.
17. W. K0 Stevens and R. W. Atree, J. Chem. Phys., 18, 574 (1950).
18. T. S„ Garwood, A, Pphland and J. L. Burhans, Abstracts, 105th Meeting of the American Chemical Society, Detroit, Mich., April 1943, P. 27M.
19. E. R. Alexander, J. Am. Chem. Soc, 69, 289 (1947).
20. W. von S. Doerine, T. I. Taylor and E. F. Schoen^alt, ibid. . 70, 455 (1948).
21. 0. K. Neville, ibid.. 70, 3499 (l°48).
22. W. Baker and R. Robinson, J. Chem. Soc, 1952. 1798.
23. C. J. Collins and 0. K. Neville, J. Am. Chem. Soc, 22, P471 (1951).
24. R. de Neufville and H. von Pechmann, Ber., 23, 3375 (lflPO) .
25. C. W. Shoppee and D. A. Prins, Helv. Chlm. Acta, 26, 185 (1943).
26. D. B. Sharp and E. L. Miller, J. Am. Chem. Soc, 74, 5643
(1952).
27. M. Mousseron and N. Phuoc Du, ComPt. rend., 218, 281 (1944).
28. C. L, Stevens and E. Parks s, J. Am. Chem. Soc, 74, 5352 (1952).
MIGRATION IN THE WAGNER REARRANGEMENT Reported by Thomas R. Moore March 27, 1953
In recent ye«rs C. J. Collins and others at the Oak Ridge National Laboratory have been interested in oroducing C 4-labeled poly nuclear hydrocarbons. This has been accomplished by means of the' Wagner rearrangement. The first synthesis developed was that of Dhenanl-hr-ene-9-C14. The steps in this procedure v^ave been usee as models for- the syntheses of more complicated molecules and are as follow 3*. x s fe
Na
I)
v
*S
*GHoOH
1) LiAlH4 ^ T-
(CflFs)aCNa
">
2) K
C03CK3 CHftOH
R
T"
H
\/
Pa0,
3U5
xylene
<S\
v
N^
The next compound synthesized in this series was 1,2-benz- anthracene,3 T»rhich was made from ?, 3-ben7,ofluorene . In this synthesis two different oositions could become labeled, depending on the way the Wagner rearrangement "Proceeded.
II)
CFsOH
P305
xylene
•>
V
-2-
The actual position of the label wan determined as follows
N^
5.55 J4 c.
+ C*02 6.05^C/ o.
This degradation shows that in the original Wagner rearrangement the ratio of migration of p-naohthyl to migration of phenyl is 522 48* (it is to he noted here that these groups are not really "P-nvohthyls" or ""Phenyls11 because of the Presence of the blPhenyl bond, but such terns can he used to distinguish the groups as ™ell as the more cumbersome correct terms.)
Chrysene-5. A- r.x 4 was ne^t synthesized4 from 1, ?-ben/of luorene by methods similar to those previously described. Here aerain the label could appear in two places, and the degradations used to determine the actual position of the label showed that the ratio of active carbon in Position 5 to that in position 6 is 76: ?4.
-3-
*CH.pH
III)
<**V
v;
•>
That the tendency of a-nar>hthyl to migrate is definitely greater than that of Phenyl might have been predicted because of the great er reactivity of the a-Dosition of naphthalene than of the P-oosition with resoect to aromatic substitution, which was theoretically justified by Wheland.6
In an attempt to learn more about the nature of these migra- tions the following reactions were studied:6
IV)
V
^\
V)
*
/^J-
V
Here it was found that in reaction IV the ratio of f-naPhthyl migration to Phenyl migration was 56:44. In reaction V the ratio of ct-naphthyl migration to Phenyl migration was 5°:48. Comparison of these results with those of reactions II and III shows that the presence of the biphenyl bond enhances the ability of the ct- naphthyl grout) to migrate in preference to the phenyl group. How- ever, the chance of migration of the P-naphthyl is lessened slightly when the biphenyl bond is present. These results have not yet been satisfactorily explained.
Work wps then undertaken which it was hoped would give in- formation on the steric effect of an ortho group.6 The following reactions were studied:
v>- -
-4-
VI)
*CH3OH CH3
CK.
>
VTI)
*CH30H CH3
^v^
V"
-is
V
■>
Degradative studies show that the ratio of Dhenyl migration to that of o-tolyl in reaction VI is 55:45, while in reaction VII it is 50:50. Since this represents a real difference and is not within the range of evDerimental error, it would seem th^t the relative migratory aotitudes vary when different systems undergo the same reaction.
Burr and Clere szko9' 10 have studied reactions of the type
*CH,0H I VIII) Ar-CH-C6H5
and
H
^ Ar-C*H=C*H~C6H5
IX) Ar-CH-C6H5
HONO
/
1
mixed
earblnols
r
PoOc
~> Ar-C*H=C*H-C6H5 KMn04 I 0H~
IE5^ Hooc-<^y.c*cai
-5-
For these reactions the results are tabulated below,
Ar
PER GENT OF MIGRATION OF Ar Reaction VIII Reaction IX
T)-Dix>henylyl
m-Tolyl
p- (2-Propyl)-phenyl
3 , 4-D lme th y loh e ny 1
p-Tolyl
p-Ethylphenyl
p- (t-Butyl)-TDhenyl
p- Me thovyobe ny 1
57 61 65
66 66 69 76 96
50 48
47 59
This table makes it evident that the relative rates of migration differ in different reactions.
Thus it seems that the relative migratory aptitudes of various groups in carbonium ion reactions are functions of both the type of system used and the reaction involved.
REFERENCES
1.
2. 3. 4.
5. 6.
7.
8. 9.
10.
C. J. Collins, J. Am. Chem. Soc, 70, 2418 (lP4ft) .
W. G-. Brown «nd B. Blue stein, ibid. , .62, 3256 (1940).
C. J. Collins, J. G-. Burr, D. N. Hesp, ibid.. 73, 5176 (l95l).
C. J. Collins, D. N. Hess, R. H. Mpyor, G-. M. Toff el, A. R.
Jones, ibid. , 75, 397 (1953).
B. M. Benjamin, 0. J. Collins, ibld.? 75, 402 ^1953).
75, 405
L. S. Ciere^ko, J. G-. Burr, ibid C. G-. LeFevre, R. J. W. LeFevre, J. Chem. Soc,
C. J. Collins,
(1953).
E. D. Hue-he s,
202 (1937).
G-. W. Wheland, J. Am. Chem. Soc
J. GS-. Burr and L. S. Clereszko,
L. S. Cieres^ko and J. C-. Burr,
64
900 (l942) .
ibid'., 74, 5426 (1952).
ibid., 74, 5431 (1952).
CONFIGURATION STUDIES BY ASYMMETRIC SYNTHESIS Reported by Edwin J. Strojny March 27, 1P5?
Recently, a method for the determination of the absolute configuration of an asymmetric carbon containing a hydroxy 1 group has been developed which is based on the asymmetric course of the reaction between a G-rignard reagent and an alpha-ketoacid ester^3'J The rules used in this procedure, propounded by Prelog, were de- rived from the asymmetric syntheses studies of MoKenzie and co- workers1 and are analogous to those of Curt In and Cram T-rhich were discussed in the recent seminar by Passer*4 The object of this seminar is the presentation of this method as it is applied to the configuration studies of natural products.
The reaction shown here:
sequence used for the configuration studies is
R3R4R5COH + RiCOCOCl I II
R3R4R5COH + R1RsC(OH)C08H 4
P/rieine^ R^OCOgCRaR^Rs
III
Ha OH
Ha0
RiRgC (OH) CO30R3R4Rs IV
RgMgX
By this scheme alpha-- hydroxy a but rather the solution of th alcohol (l) is hydroxy acid is specific rotat pure enantiomo are so chosen configuration active alcohol appreciably in sterlc, course
he enantiomorphs of the antiomorphs are not separated, cal rotation of the etna nolle e mixture is noted and the configuration of the deduced-. The Percentage of excess of the alpha-
, an excess of one of old is formed. The en? direction of the opti<
readily determined by ion of the mi/ lure to i'Ph by 100. '/'he ketoa that an alphn-hydroxya is known. The groups, are hydrocarbon radio their spacial require of the C-rignard react!
multiplying the ratio of the the specific rotation of the cid and the G-rignard reagent cid is obtained whose absolute
R3,R-i, and R5, of the optical^ als or hydrogen and must diffei ments in order to alter the on sufficiently.
the
(a)
The absolute configuration of the reaction tvpes illustrated below:
+ R2MffX
alcohol is derived from
Rvx OH 0
i
OH
R.
(b)R
+ R2MgX
->
Ra < H4 < R-
-2-
The structures of the alpha-hydroxyacld Droduced In excess are shown, ©nly the conf lguration of the asymmetric carbon which contains the hydroxyl group Is deduced in this manner.
In his configuration studies by asymmetric synthesis, Prelog used phenylglyoxalic acid and methyl magnesium iodide. These reagents yielded atrolactlc acid whose absolute configuration and rotatory power are known. Since his studies involved only secondary alcohols, the reaction tyoes given above can be simplified to the following scheme:
L (+)
C03H \ HO-C-CH3
a
CqHb
atrolactlc acid in excess
R4
H-C-OH
i
R5
R,
<R,
C03H I CH,-C-OH I C6KS
">
HO-C-H I R*
D (-) atrolactlc acid in excess
R4 < R5
Prelog and his co-workers first tested the method by applying it to (-7 menthol, (+) neomenthol, (+) borneol and (-) isoborneol. These compounds gave the expected results which are summarized in the table on t^e next page. The authors then used this procedure for absolute configuration studies on the triterpene aloha-amyrln (Vl) and on the steroids dlhydrolanopterft/t Cvil) and euphw»ol (VIII) . The Phenylglyoxalic ^>cid esters of these three substances induced an asymmetric reaction with the methyl magnesium iodide which yielded, on saoonif ication, a dextrorotatory atrolactlc acid, in excess. This means that these alcohols belong to the reaction type a and that the asymmetric carbon containing the hydroxy 1 group would have the configuration corresponding to this type. Such a configuration is in opposition to that arbitrarily assumed for the first two compounds, but is in agreement with the structure assumed for the latter steroid. Other studies with 17 cc-androstanol (IX), 7 a- and 7 p-cholestanol (X and XI resp.) showed that the configurations agreed with the previously arbitrar- ily assumed ones. 3 J3-cholestanol (XII ) gave only a slight excess of levorotatory atrolactlc acid so that no conclusion can be reached with certainty on its configuration by this method.
The authors gave certain precautions which should be observed in the application of this Procedure. First, complete saponifi- cation of the resultant cc-hydroxyacid ester must be assured. Prelog has shown that the saponification can proceed asymmetrically so that when only partial hydrolysis is achieved, »n excess of the acid opposite in configuration to the ester obtained in excess can
-3-
be obtained. Another possible source of error may arise from the fact that the G-rignard referent repots further with the a-hydroTy~ acid ester to "Produce the glycol. This reaction cnn also proceed asymmetrically and may impair the results, especially if the yield of the glycol is relatively large. For this reason, little faith Is placed on experiments where the ootioal yield is low. Other phenomena, such as the occurrence of precipitation during the reaction, which may influence the steric course must be pvoided.
Some Data Obtained by Asymmetric Syntheses
Atrolactlc acid
CalD = 37. 7C
Alcohol
Type
(-) menthol
(+)neomenthol
(+)borneol
(-)isoborneol
a-arnyrin
dihydrolanoster^i
euph«nol
3£-cholestanol
lTOr-androstanol
7op-cholestanol
7£~cholestanol
Yield
la}
!>-*
in alcohol (excess)
#
\y <^y
VI
VII
VIII
IX
XI
. *
_4~
XII
BIBLIOGRAPHY
1. Preloe:, Helv. Ohlm. <\ota, 36, .^08 (1953) .
2. Prelog;, nnd Meier, Ibid.,, *?0 (1953).
3. Dauben, Dickel, Jep-er, and Preloe:, Ibid., 325 (lP5?) .
4. Seminar Abstracts, University of Illinois, February 20, 1953,
SOME POLYPHENYL DERIVATIVES OF NONMETALLIC ELEMENTS IN THEIR VJ.QVER VALENCE STATES
Reported by M. J. Fletcher April 10, 1953
Although compounds such as trlobenylohosphorus have been known for a long tine, it is only in the ia st few years that polypheny lateo compounds of these elements in their higher valence states have been made.
Compounds of the Type (C6H5)5 Z (l) (g)(3)
Preparation
(C6HB)& ? = 0 + C6H5 Li -> (C6Hb)4 P - OH (CgHj* y - OH + H I -* (OsIIb/4 ? I
(aaH6)4 J? I + C6HS Li -> (CaHs)B P + Li I
where Z - P As, or Sbo The yields from the last step are 60f,
65^ and 7?# re ?oe - Lively ,
(CeHsh As 0la * SCgHyLi JgttfO -^ (CeH5)5 As 47jf
(C6H5)3 Bi 01B + 2CsHyLi J^i^~>(CeH5)5 Bi 81*
(CeHe)3 Sb Cls + 3GaHsLi -I^f~>[ (C6H5)S 3b] Li
H30
(GSHK)'5 Sb
Stability
The stability of these confounds T-rith respect to heat and most acidic reagents increases in the following order:
Bi <P <( As \ Sb
Tlie preparation of pentaPhenyl bismuth must be carried out at -75c o It precipitates from the ether solution pis a yellow solid which changes on wgrulijg to a violet powder. Tine other compounds in tnis series are colorCass. On being heated to 105°, pentaphenyTbismuth decomposes vigorously.. All the compounds in this series decompose at their melting points.
(C6H6)S pJ^AL-.> C6H5 + Resin o
(CeHs)5 As -15— > C3gHs + (06HS)3 As '81*) + C6H5 - 0eH6
(C6HS)5 Sb i££L> C0SH5)3 Sb (96$0 + C6HB-C6HB (98.5*)
(CSH5)5 Bi , 10P°x CaHs (g5#) + C6HB~06HB (44#) + (CSH5)3 Bi (81*)
This decomposition seems to pro by a free radical mechanism. Pentaphenyl phosphorus is a good catalyst for the polymerisation of styrene.
These compounds are all insoluble in water and easily soluble
-2-
in organic solvents, indicating that their bond linkages are covalent .
Reactions:
1) PentaDhenylantimony renots very readily with one mole of bhenyllithium to e-lve n coffiDlex salt which is unstable toward woter.
(C6H5)5 Sb + CeHB Li --.-} [(C6H5)S Sb] Li ..f?»^ (C6HB)BSb + C6He+LiCF
2) These compounds react with halogens to yield, in most cases, tetraphenyl metal perhalides.
(C6H5)5 P + Br3 } (C6HS)4 P Br3 + C6H5Br
(CeH5)5 AS + I3 — > (06HB)4 As I3 + C6H5I
(C6H5)5 Bl + P3r;
-70°
^ (C6HS)4 Bi 3r3 + CSH5 B:
\!/
warming to -30*
(C6H5)5 Sb + Br3
(CSHB)4 Sb Br isolable
(C6H5}3 Bi Br3 + C6H5 Br ■ Brs> (C6H5)4 Sb Br3
\y
130
(C6H5)3 Sb Br3
+ C6H5 Br
However: fC6H5)5 Sb + 0ls \ (CSH5)4 Sb CI
Ola
boiling CsH6
>r
(C6H5)3 Sb Cls + C6H5CT1 (C6HB)4 Sb 013 has not yet been prepared. 3; Reactions with halogen acids
(0SHB)B Z + HX -* (CSH5)4 Z X + CSH6 where Z is P or as
(C6HB)5 Sb + HBr-> (06Hs)4 Sb Br iEl^ (C6H5)3 Sb Br3 + C6H5Br
isolable
, 9 i
-3-
(C6H5)5 Bi + HC1
dry ether -70°
^ (C6H5)4 Bi + C6HS
"pr miner to -20°
>/CsH5)3 Bi + C6H5 CI These reactions aDT>ear to he ionic.
These conrnounds also react with Lewis acids such as triohenyl
boron
(CSH5)S Z + (CgHg)3 B
^ Q
k)
v [ fCeH5)4 Z ] ' C(C6K5)4B]
ether
Tetra/ohenyltellurium (4)
The only tetraohenyl derivative of an element in the fourth group which has yet been prepared is tetranhenyltellur.ium
(C6H5)3 Te 01S + CSH5 Li -> [ (CSH5 ) 3 Te CXl]
N/
C6H5 Li
(C6H5)4 Te + Li Cl
Tetraphenyl tellurium, is less stable than penta^henyl- phosphorus. It is also decomposed by water
(C6H5)4Te ^ (C6HB)3 Te + C6HB-C6H5 + C6H6
(C6H5)4 Te + H30 ^ fC6HB)a Te OH + CSH6
It forms a stable comoley with trinhenyl boron.
(C6H5)4 Te + fC6Hs)3 B__>[^sw5)3 Te] [(CSHB)4 B]
In its reactions it behaves very much like the G-rignard reagent.
(C6H5)4 Te + CH3Cl3 -> (CeHB)3 Te Cl + CsHBCHgCl
(C6HB)4 Te + CHC13 — ^ (CSHB)3 Te Cl + C6HBCHC13
(CSH5)4 Te + C6HBCH0*-^ (after hyfl -oly sis) (CsH6)aT©01 + (C6HB)(SI0H
Triphenyliodine (l), (4)
Triohenyl iodine is the only nolyohenyl derivative of an element in the seventh erouo which has been found isolable.
-4-
C(C6HB)aI] I + C6H5Li or C6H5I Ola + 2CfeHB Li
-80<
-801
r> (C6H5)3 I + Li I ^ (C6H5)3 I + 2 Li CI
This compound is very unstable at temperatures as low as -10 After drying in a vacuum, however, it can be keot awhile at 2 * It explodes on being brought to room temperature.
It forms the expected complex with tripbenyl boron, which is stable at room temperature.
(CSH5)3 I + (C6HB)S B-4[i'C6H5)3 I ] [(C6H5)4 B ]
BIBLIOGRAPHY
n!/
1. G. Wittle: and M, Rieber, Ann., J552, 187 (1949).
2. Or. Wittie: and Kf Clause, Ann., 577. 26 (l95g).
3. Cr. Wittie and K. Clause, Ann., J578, 136 (1952).
4. Gr. Vittig and H. Fritz, Ann., 577, 39 (1952) .
REARRANGEMENTS OF 9- SIB STITUTSD F LUCRE NES Reported by Richard L. Johnson April 10, 1955
The first report of a rearrangement of a 9- substituted fluorene compound was by Kilbert and Pinck1 in 1938. They reoorte that dimethyl-9-fluorenyl sulfonium bromide (i) was transformed in the presence of alkali to an equilibrium mixture of ylid-like compounds3. One of these (il) vps stable, while the ether (III; rearranged to form methyl- (l~f luorenylmethyl)- sulfide (IV) . reaction differs from all subsequent rearrangements of in that the alkyl group migrates to the 1- rather than 9-fluorenyl position. This difference is explained by tion that compound II Is stable, and that compound III reactive species.
.©
this type to the the as sump- is the
Br
CH,
it q nu v- — '
^yY^
ii
in
Wittig and Felletschin3 found that 9-fluo ammonium bromide (V) when treated with Phenyl 1 stable fluorenylio* (VI ) . When this ylid was h to form 9- me thy 1-9- dime thy la mi nof luorene (VII) of 9-f luorenylbenzyldimethylammonium bromide wa in the same manner. The product, formed in ne yields, was 9~benzyl-9-dimethylamlno£ luorene . strates cnat the benzyl group migrates in pref group in this reaction.
N-(CH3)3
IV
renyl trimethyl- ithium produced a eated it rearranged
The reaction s also carried out arly quantitative This fact demon- erence to the methyl
CH3 N-(CH3)3
V
v
^~ ^
V
VI
VII
Wittig and Felletschin also studied the rearrangement of 9-fluorenyl methyl ether. This reaction had been suggested by 4 similar reactions of benzyl ethers which had been studied earlier , When the ether (VIII) was treated with phenyllithium, intermediate IX formed and rearranged to X which could be hydrolyzed to 9-methyl- 9-fluorenol. Rearrangement of other 9-fluorenyl ethers wp q also studied5. Phenyllithium was found to be the most effective catalyst in these rearrpngements, and tetrahydrof uran was found to be a suitable solvent6. The ethyl and allyl ethers were made by the Same method as the methyl ether: fluorene wag brominated bv
N-bromosuccinimlde, and the product wflp treated with silver nifra4 and the appropriate alcohol. The benzyl ether was prepared from 9-bromofluorene and benzyl alcohol with no catalyst. The phenyl ether could not be prepared from 9-bromofluorene and either phenol or potassium. phenoxide* These regents produced 9- (p-hydrovy- phenyl)-fluorene (XI) and bidlph.enyleneethylene (XIV) respectively. When 9-hromof luorene, phenol, and potassium phenoxide were re- fluxed in tetrahydrofuran the desired ether was obtained. In like fashion the p-me'thylPhenyl-, P_-chloroohenyl-, p-iodophenyl-, p_-nitrophenyl-, and p-trimethylammonlurnohenyl~ethers were prepared
VIII
® LleO-CH3
%/ — \^
IX
S \W OH
Li®
XII
XIII
XIV
The rearrangement of these eth nitrogen atmosphere. The lithium d prepared by the addition of ohenyll ethers produced lithium derivatives When these derivatives ™ere heated rearranged to the corresponding 9-s and benzyl ethers produced lithium at once to the expected fluorenols. di-9-f luorenyl ether rearranged at fluorenol. When phenyl f luorenyl e lithium, a lithium derivative forme temperature but which rearranged at ethylene (XIV) and. Potassium phenox through Intermediates XII and XIII,
ers was carried out under a erivatlves of the ethers were 1th ium. The ethyl and methyl
stable at room temperature, at 100°C. for four hours, they lkyl-9-f luorenols. The allyl derivatives x^blch rearranged
The lithium derivative of once, forming 9-f luorenyl- 9- ther was treated with phenyl d which was stable at room
100° C. to form bidlohenylene- ide . This reaction occurs
which have been isolated.
■o-
The para- substituted phenyl ethers, Kith the exception of the p.-nitro compound, also formed bidlohenyleneethylene . The p-nitro- phenyl fluorenyl ether, hearing an electron-withdrawing sub- stituent, rearranges, as the alfcyl ethers do, when it is treated with lithium methoxldo at 100eC*. . Lithium methoxide w«s used instead of phenylllthium to prevent reduction of the nltro group.
In another scries of experiments Wittig, Keintzler, and Wetterling7 oroduoed "organometal"! ie tetramethylammonlum compounds' from organometallio compounds and tetramethylammonium halides. Among the carbanions so formed was that from 9-bromof luorene (XV) . These compounds were ionic salts, The °-fluorenyl compound re- arranged when Seated, forming 9- me thy If luorene . * The salt reacted as an organometallio compound, forming 9-fluorenyldlPhenyl carbinol (XVI) frocr benzophenone .
©
N-(CH-)
3^ 4
Den^oD^enone
>
XV
XVI
Bahn and Solms have reported the rearrangement of tertiary amines. They were attempting to synthesize 9-f luorenylnaphtheyl- methylm.e thy] amines (XX a. and p.). The route of synthesis is shown (XVII through XX). The reactions proceeded as' shown when diethyl ether was used as solvent, but when tetrahydrofuran was used^as solvent for the last reaction, an isomeric secondary amine (XXI; was formed in 56^ yield. Hydrogenolysis of the secondary amines produced the corresponding 9-f luorenylnaohthylme thanes which were also prepared by independent synthesis. It was found that lithium aluminum hydride in tetrahydrofuran isomerized XX to XXI by acting as a strong base. That the secondary amines were not XXII was not proved, but these structures -ire unlikely in view of the higher acidity of the 9-fluorenyl hydrogen as compared to the napthpylrrethyl hydrogens. This rearrangement could not occur through an ylid intermediate since the amine was secondary. Thus, it more nearly approximates the rearrangements of the ethers.
-4-
H 0 N-CH
CH3 I
H N-CO— Naphtt
Na^hthoyl Chloride ?
XVII
XVIII
CK3
/
H N-CH3-Naohth.
Li
CH3-Naohth,
XIX
cc.rNaPhth. = a Naohthyl p.rNaphth. = p Nnphthyl
H
H N-CH3 \S .C-NaT5hth .
XX (a. and p.) XXI (a. and p.)
* — v
XXII (a, and p.)
Still another rearrangement of a fluorene derivative has been reported "by Arnold, Parham, and Dodson9. The allyl ether of 9-f luorenecarboxyllc acid (XXIII ) wsg inomerized to 9-allyl- fluorene-9-carboxylic acid (XXXV) by lithium amide. This type of reaction had been found to occur with dipheny lace tic acid esters10, with a resulting allyllc shift. Further publication was promised concerning the rearrangements of benzyl esters, which seemed to give mixtures of products, but none has yet appeared.
H
0 ii
C_0-CHoCH:CH.
0 if
HOG CH2CH:CH3
'/
v
LiNH.
Ss — v^
->
XXIII
xxrv
l.
2.
3. 4. 5. 6. 7. 8. 9. 10.
BIBLIOGRAPHY
Hilbert and Pinck, J. Am. Chem. Soc, 60, 494 (1938).
Love joy, Organic Seminars, U. of I., I Semester 1951.
Wittig and Fej.letschin, Ann., J555, 133 (1944) .
Wittig and Lohmann, Ann., 550 T ?60 (l~42).
Wittig, Doser and Lorenr, Ann., 56?, 19° (1949).
Wittig and Haooe, Ann., J557, °05~Tl947) .
Wittig, He in trier and Wetterllns", Ann., J557, 901 (1947).
Bahn and Solms, Helv., 34, ?0& (l95l) .
Arnold, Parham, and Dodson, J. An. Chem. Soc, 71, 2439 (1949)
Arnold et. al., ibid.. 71, 1150 (1949) .
J
A NEW SYNTHETIC APPROACH TO o - HYDROXY PHENOL DERIVATIVES
Reported by William H. Low&en
April 10, 195
C'z;
INTRODUCTION
A recently developed method for the synthesis of catechol and pyrogallol derivatives has ooened an attractive new route to com- pounds which were previously obtained with great difficulty. The reaction, in addition to its synthetic v«lue, can also be used advantageously in structure proofs.
SYNTHESIS
2-Chloro~5-nltrobenzrcohenonc and p phenol repct in the pres- ence of a base to form a 5~nitro— ^-flryloxybenzoobenone pf would be predicted on the basis of the Williamson ether synthesis. Cycliz- ation to the corresponding xanthylium pplt is conducted in the presence of sulfuric acid. Analogous cycll^tions have been re- Ported of substituted diaryl ethers and corresponding thio com- pounds.1'3 Upon extreme dilution, the vanthhydrol is formed. Thi> material is dissolved in acetic and sulfuric acids with slow addi- tion of hydrogen peroxide 0 The resulting ^-nltro-2- (S'-hydroxy )- aryloxybenzoPhenone is then cleaved by piperidine. Similar cleav- ages have been known since 1927. 3> 4 The products Isolated are the 2-piperidino-5-nitrobenzoPhenone and an _o-hydroxy phenol correspond- ing to the original Phenol.
03N . C0j6
yV
|AJ + V^Cl HO
KOH
OaN
> I
r
cold cone.
H3S04
>
0,N
0 OH Ri
X
H30
(II) R4
9-phenyl xanthylium sulfate
0*N y\ X A
(III)
R.
Rs
R,
H30s
0SN ^
HAc-HoSdl
^N> I
o
-2-
DISCUSSION AND LIMITATIONS
The activity of the chlorine atom in an ortho oosition to a strong electron accentor is enhanced by the presence of electron withdrawing substituents on the nucleus.5 Hence, the aromatic chlorine is sufficiently active to give good yields of the ether In the reaction with a phenol.
The final concentration of sulfuric acid in the cycllzatlon step has proved to be the crucial factor in the synthesis. Similar work on tMoxantbene and derivatives has shewn that this cycli^.ation is deDendent unon the nucleophllic reactivity of ring
The essential contribution in this series of nan choice of the oxidising medium. Hydroeren neroxide, l and acetic acid mixture, has the ability to introduce hydroxy 1 groun in the 2! nositlon under these condltl resulting compound, (IV), is cleaved with pineridlne desired phenol. However, in many instances, a rearr* dehydration to a fluorone occurs. Apparently, the de mainly occurs during the removal of t^e niperidlne by Methyl, methoxy or p- toluene sulfonic acid erroups (loc and R3) prevented the dehydration whereas the bromo d cyclized readily. An example of this cycllzation fol
ers is the n a sulfuric
a free ons. The yielding the ngement or a hydration
acid, ated at Ri erivatlves lows:
OaN
<^S
OnN
Hfl0
2^3
H3304-HAc
fi
(VI)
OpN
^
CH
The rearrangement which takes nlace In an alkaline solution Is similar to the tyne first observed by Smiles. He has shown that an important factor in the rearrangement is the different electron donor abilities of the atoms Y and Z,
OH
G
<-
^v
■v
_3-
This rearrangement occurs In this series of compounds in the following manner:
03N . CO0
yV
OH
<■
O 0SN
Conclusive evidence for the rearrangement is obtained upon treatment of (VII) with alkaline methyl sulfate. Rearrangement must, therefore, occur before methylatlon. Cleavage T«rith piperidine gives rise to the corresponding phenol.
(VII)
(CH3)3304°2^ ^ .C0j6
OH^'
CH30
CH.
HO
s\r
Elazomethane, unlike methyl sulfate pnd alkali, methylates the compound without rearrange men t.
02N . C0j6 (VII) CH9N2 \/V CH.O-*
/,
OH 3
CH30
V^>
HO
A^
CH.
The condensation product of gualacol and 2-chloro-5-nltro- benzophenone "as identical with the methylation product of (IV) (Hi , R3, R3, R4 - H). Hence, there is no doubt as to the nature of the ring opening.
Repeated hydroxylation of a phenol can be accomplished as Is shown by the synthesis of te tr a me thoyy benzene.
C?j
t> OH
OCR,
-4-
Other than the aforementioned limitations, the reaction apoears to be quite general, as is exemplified by the following list of compounds prepared in this manner.
catechol
pyrogallol
3, 5-dime thy lea t echo 1
3,6- »
4, 5- »
3-phenylcateehol 4- »
2-me thoxy- 3, 5-dime thy lphenol
2- » 2- " 2- "
3,6- » 4,6- » 5-phenylPhenol
3, 4-dihydroxy toluene 2-hy dr oxy- 3- me th oxy t olue ne 3- " 2- "
4,6-dimethyloyroeallol 4,6-dibromoT)yrogallol-l,3-dimethy
ether 4,6-dibromo-5-methylpyrogallol-l
3-dimethyl ether 4-hydroxy-7-nitrc-9-phenyl-
fluorone 3-bromo-4-hydroxy-7-nitro-9-
phenylf luorone
BIBLIOGRAPHY
1. Decker, et.nl., Annalen 348, PJ51, 238 (1906).
2. Reilly, J. and Drumm, P. J., J. Ghem. Son,., IP 30, 455.
3. Le Fevre, J. W,, Saunders, L. M. and Turner, E. E., J. Chem. Soc, 19?7. 1168.
4. Groves, L. G-., Turner, E. E. and Sharo, G-. I., J. Chem. Soc, 1929, 512.
5. Holmes, C. W, N. and Loudon, J. D., J. Chem. Soc, 1940, 1521.
6. Camobell, C. V. T., Dick, A., Fereruson, J., and Loudon, J. D., IF. Chem. Soc, 1941. 747.
7. Quint, F. and Dilthey, W., Ber. 64, 2082 (l93l).
8. Loudon, J. D., Robertson, J. R., T'*atson, J. N. and Alton, S. D. J. Chem. Soc, 1950, 55.
9. Loudon, J. D. and Scott, J. A., J. Chem. Soc, 1953. 265. 10. Loudon, J. D. and Scott, J. A., J. Chem. Soc., 1953. 269.
DEVELOPMENTS IN AZULENE CHEMISTRY Reported by Aldo J. Crovettl April 17, 1953
Azulene chemistry has been the subject of many reviews. Most recently it has been extensively reviewed through January 1951 by M. Gordon.1 Also a recent seminar deals with recent work through 1949. 3 This seminar will deal mainly with work since 1951.
The classical method for azulene synthesis employing ring ex- pansion with diazoacetlc ester has recently been exploited by Herz3'4'5, who has used this method for the preparation of tri- alkylated azulene s. The synthesis of 2,4,8-trimethylazulene (i) is outlined below.
OH rCH3Cl H-i-^C03Et)3
Na, xylene reflux
<o
C-CH3
/ 7
(C08Et):
OH '
-co"
,Hg/Zn
3 x HCl
C=0
0 H l,SOa0la 2,A1C13
1,NSCHC03E- 3,?d-C,360
The product did not appear to be indent ical with pyrethazulene whose structure was suggested as (i) by Schechter in 1941.
In a similar manner Herz4 has synthesized 1,4, 8-trimethyl- azulene (il) and l,4-dimethyl-3-isopropylazulene (ill), previously not described.
-CH,
Hg/z
HCl
ft
l,N3CHC03Et 2.0Hr^>
3,Pd-C
CH3MgI 50^ XS reflux
1,KHS04,A
2,Pt02/H
-, l,Ns0HCOd3t
2, OH© 3,Pd-C
CH
CH CH3
chC vch^
III
-2-
W. Treibs6 has employed this procedure using 4-azaf luorene an< obtained an azaazulene (iv) or (V) .
C03Et
]C03Et
The free acids obtained by saponification at room temperature undergo facile decarboxylation to give products which are less stable to light and air than the carbethoxy derivatives.
Treibs7 has also applied this procedure to the synthesis of azulene esters (Vila) or Vllb) of tetrahydrofluoranthene (VI).
<<=:V |
f^ |
i |
N\ // |
II |
R.Ni |
II |
i 150 atm.'l20° 3 hrs. |
N3CHC02Et
Soont . dehydrog.
\i
Vila
7cosst
w\
Regarding the position of the ester group it is interesting to note that Plattner8, using indsne and 2- me thy lindane, was able to separate from the reaction mixture the 5- and 6-carboxylic acid esters, but found no trace of the 4-isomer.
Recently several azulene syntheses have been announced which do not involve ring expansion. H. Pommer9'10 has developed a promising method for the synthesis of 1, 5, 8-, 1,4, 7-, 1, 7,8-, and 1, 4, 5-tri substituted azulene s which heretofore have not been prepared in Pure form (if reported at all) due to the complications of ring expansion procedures. The procedure appears to be general:
■ "... ,' ■
-3-
R
R l,CHaC03Et R
R^\(TCHO Qo -7 N}^ <0°,4 hrs. X0^
NaOH,5 hrs. 3, warm to RT
VIII 8C% IX 70% X
R.N1
R.T.
120%~l00^fcm Rt 4^ LcH3-CH-CK3-CH3-C03Et -ggy- autoclave "■ EtOH
87^ 50-6C^
Rf-CH-CH3CH3-CH-CH3-CH-CH3CH3COsSt
*r B'r XII
R
2Na0H^0OaEt)a
80#
>
CH3-CH-CH3CF3C03Et
C03Et C03Et
OH
iz>
\
Rf XIII
975?
CHo-GH- CKsr- CKoCOOH
R<
L1A1H.
CO OH XIV
KH30
Fe
^
Bp(0H)3, /^
70-75^ R
->
/ Y \ *> / V
R«H
XVI
R'
droooed on
15f ?d-C
36'Oo distilled.
XVIII
KH304,A
15^Fd-C
360° distill
^
-4r-
Using this general scheme, (yields based on the dehydrogena tion steps XVII— >XVTII and XX-*XXl) Pommer synthesized azulene (VIIt-» ^ XVIII; R'=R=H) ^4. 550 (necessarily indicating a poor overall yield;;
5-methvlazulene (VIII-^VIII; R=H, R=Me)(l8#); 1,5-dlmethylazulene (VIIMCVIII; R?=R=Mc) (l8?0; 4- methyl R«=Me) (15^); 5,8-dimethylazulene
arulene fVIH-XWOCT; R'=R=H, ^•■=rLuj M.vjrj; o,n-aimetnyxaau.xene (VIII--XV-^CXI; R»=H, R=R» = Me) (13$) and 5-methvl-8-isoorooylazulene (VIII--XV--XXI; R»=H, R=Me, R»=isopr.,
In an attempt to prepare 1, 5,8-trimethylazulene through the same sequence a mixture of the 1,5,8- and 2, 5, 8- isomers w^s ob- tained, indicating a migration of a methyl group during or after dehydrogenation. Migrations of an isopropyl5 and a phenyl11'18 group have been reported.
A. G-. Anderson13 reported the synthesis of what appears to be the first heteroazulene of known structure.
£x
->
l-Azabenz-[b]-azulene
An improved synthesis of azulene itself not involving a ring enlargement step has been developed.14
OK
P90
3^5
H3P04
N 40fCH3QC0H
bfo NaH003
~>
» 4-Azulol»
XXIV
Presence of the hydroxy 1 group In (XXII) has been demonstrated and its position inferred from the visible light absorption spectrum, which was characteristic of that given by 4-alkyl derivatives. In this regard Treibs1 s has shown that 1-methoxy and ethoxy derivati\ef have identical soeetra with those of 1-alkyl derivatives. Eromination of (XXIII) gave three bromoazulenes of unknown struct- ure,
Azulenes with other functional errouos have been reported. Anderson has reported the preparation of l-azulenea^obenzene from azulene and benzenediazoninm chloride in the oresence of sodium acetate. In acid solution a red compound is formed which gives the azo compound in brown-black needles on addition of base. Treibs1 s has also reported the preparation of 6~carbethoxy-l, 2- benzazuleneazobenzene by the action of ohenyld::.a.zonium chloride on 1, 2-benzazulene .
Action of an excess of acetic anhydride on azulene, in the presence of aluminum trichloride and carbon disulfide, gave 1,3- diaoetylazulene (62^0 as red needles.1'17
A nitroazulene (probably the 1-nitroazulene) was obtained in red needles (51#) on treating azulene with cuprjc nitrate and acetic anhydride. Alkylation of azulene with methyl chloride or iodide by the Frledel-Crafts reaction gave minute quantities of 1-methylazulene (?).
In Germany synthetic azulene is being produced commercially16. It is being used in preparations of chamomile (which is a natural plant oil containing chamazulene) to increase the efficiency of the chamazulene present in checking skin inflammations. Cther suggested uses include; addition to perfume compounds for ointments and creams, addition to therapeutic preparations, and to medicinal hair lotions, mouth washes, and tooth-pastes.
REFERENCES
1. M. Gordon, Chem. Rev., 50, 127 (1952).
2. R. Roeske, Ore. Chem. Seminars 1949-50.
3. W. Herz, J, Am. Chem. Soc, 73. 492? 11951).
4. W, Herz, ibid. . 74, 1350 (195°) .
5. W. Herz, Ibid. . .75, 73 (1953).
6. W. Treibs, H. Barchet, G. 3pch, W. Kerchhof, Ann., _574, 54
(1951) «
7. W. Treibs, ibid., 574n 60 (lQ5l) ,
8. Pi. A. Plattner, A. Furst and A.R. Somerville, Helv. Chlm. Acta., J34, 971 (l95l).
9. H. Pommer, Naturwisgenschaf ten, 39. 44 (1952) .
10. H. Pommer, Ann., 579, 47 (l95?5 .
11. Pi. A. Plattner, R. Sandrin, J. Wye a, Helv. Chlm. Acta ., 29, 1804
(1946). „
12. Pi. A. Plattner, A. Furst, M. Cordon, K. Zimmermann, ibid. . 53 1910 (1950).
13. A. G. Anderson, Jr. and James Tazuma, J.Am. Chem. Soo. , 74. 3455 (1952).
}!• A'&m Anderson, Jr. and J. A. Nelson, Ibid., 73, 232 (l95l).
15. W, Treibs and' A. Stein, Ann., 572^ 161 (195177
16. W. Treibs and A. Stein, ibid., 57*, 165 Tl95l).
17. A.G. Anderson, Jr. and J. A. Nelson, J. Am. Chem. Soc, 72, 3824&950J
18. H. Janlstyn, Perfumery and Essential Oil Record, 42,~TB 5, 236(195 1).
ALKALINE DECOMPOSITION OF HYDRAZINE DERIVATIVES Reported by David M. Locke April 17, 1953
Introduction
As early as 1885 the action of aqueous sodium hydroride on benzenesulfonylphenylhydrazone was recorded by Escales.1
^9 Vy^NH-NK-SOg^V- NJv._T_^ S/ >S> + Ns + <f NS-S03N«
Many different types of substituted hydrazines (arylsulf onyl- hydrazines, hydrazidcs, and hydrazones) have since been reacted with basic reagents, usually at high temperatures, and their characteristic behavior under these conditions has led to the development of several useful synthetic reactions which exhibit a certain mechanistic similarity.
The Wolff-Klshner Reaction
The Wolff-Klshner reaction is a well-known method for re- ducing a carbonyl group to a methylene group.2 It is perhaps most conveniently used as the Huang-Minion modification: the carbonyl compound in ethylene glycol is treated with hydrazine hydrate; after formation of the hydrazone is complete the low-boiling material is distilled off, and the hydrazone is decomposed at 200° with alkali.3
Kinetic studies by Balandin nnd V^skevitch have indicated the formation of an intermediate in the reaction, »nd these workers have suggested the diimine, which is isomeric *rith the hydrazone, as the reactive intermediate.4 Todd has suggested that the diimine then undergoes decomposition in a fashion analogous to that of dialkylazo compounds, which are known to decompose to free radicals.5
The base-catalysed lsomerization of hydrazones to diimines has been formulated by Seibert as involving an intermediate anion.
R • Rv R
^0=NNH3 > [ C'=NNH $ > ^C-N=NH]
K R V,
In the wolff-Kishner reaction this hybrid ion may itself decompose giving the observed reaction product, or it may serve only as the precursor of the diimine. In either case there are three possible modes of rupture for the C-N bond:
1. ^C-N=NH — >^Ct + NS-N + H
H 'H
2. )c-N^NH_v )c- + N=N + H- XH H
3. Simultaneous rupture of the C-N bond and formation of the new C-H bond.6
-v
-2-
The McFady en- Stevens Reaction
The alkaline decomposition of srylsulf onacylhydrazides has proved to be a useful method of converting acids to aldehydes. The most convenient method of "DreDaring the arylsulfonacyl- hydrazide is by the action of the arylsulfonyl chloride on the acid hydrazidc'. 7 And it has been found that the reaction itself is best carried out with Dotassium carbonate in glycerol at 200° e. It is believed to take the following course:
RCNHNHSO n 0
K„C0.
VS— S02K +
[RCN=NH] II
0
RCH0+N2
The reaction fails completely with aliphatic aldehydes ' 9 and is also unsuccessful in some cases in the aromatic series. McF9dyen and Stevens found that while m-nitrobenzoic acid under- went the reaction, jD-nitrobenzolc acid failed to do so.7 With carboxylic acids on heterocyclic nuclei the same sensitivity to electronic effects may be observed; picollnlc and nicotinic acids, for example, react satisfactorily while lsonicotinic acid does not.10
The Method of Albert and Royer
McFadyen and Stevens indicated that it was also possible to replace active halogen on the benzene ring by hydrogen in the same manner.
0aN
«
NO,
N0a
Cl
NHajNHg-HpO >
0,N— ^
01 so.
->
'/ \>
OpN
NHNHSO.
K-.C0.
4
OoN-^V
The analogy of picryl chloride with acid chlorides is well known. It is therefore, not surprising to find that it undergoes a reaction analogous to the McFadyen- Stevens reaction.
The first attempt to apply this reaction to the replacement by hydrogen of a halogen atom on a heterocyclic nucleus by Dewar met with little success.11 An experimental procedure was, however, later worked out by Albert and Royer. It was found that sodium carbonate and ethylene glycol at temperatures above 100° performed the replacement smoothly.13
-3-
+ NHoNHSO,
o
CH.
AAA
/,
The reaction has since nroved extremely valuable with heterocyclic compounds, particularly those containing a readily reducible group such as the nltro or cyano group. Catalytic reduction, another method of accomplishing the same conversion, cannot be used in these cases.
Again electronic influences are important in the reaction. For example, it was found Possible to reduce by this method I, II, and III, but not IV1 a
NCb CI
0*N
II
0ftN
IV
Alkaline Decomposition of p- Toluene su If onylhydrazones
Bamford and Stevens have found that p_-toluenesulf onyl- hydrazones are decomposed in a similar fashion by the action of sodium in hot ethylene glycol.14 A variety of products wag obtained, the structure of the Product depending on the typo of carbonyl comoound used.
In general the j>- toluene sulf onylhydrazones of aliphatic aldehydes and ketones yielded olefins.
CH.
CH.
N C= NNH- 30 0 yy ^v-CH
CH.
A
< CH,'
CH
In some cases rearrangements were observed.
-4-
C(CH3)3
!
OsNNHSQjb^ 'V-CH3
I
CH3
\ (CH3)SC=G(CH3)
v
3/ 2
Most aromatic aldehydes and ketones gave diazo compounds or their further reaction products.
-x Vy-CH3OCH3CH3OH
^r~^S^CH=N-NH-SOa_^~~^V-CHn-> ;^"~ ~^s_CHN3
7* HOCH3CH3OH
\
ilOH.
.O-
so.
<^~^>-GH3S03<^^ ^CIJ
Alpha diketones were found to yield sulf onamidotria7,oles.
CH3-C=NNHS03<^ ^-CHg
0H,C3=NNHSO-^
~>
CH\
CH3-C-.N"
* ( /
CK3-C-N
,NHS03<^ >^VCK3
Bamford and Stevens have outlined the following mechanism for the reaction:
R
\
+ R -H vl
R + - NC-C=N=N
C-C=NNHS03Ar ■> C-C=N-N-S03Ar
S' '
I
/
I
^C=C + NsN / i
R
The triazole from the dike tone may arise from some such Intermediate secies as the following:
M=N=C-C=NN302Ar * f R R
-5-
BIBLIOGRAPHY
1. Escales, Ber., 18, 893 (1885).
2. Todd, Orer. Reactions, 4, 378 (1948) .
3. Huang-Minion, J. Am. Chem. Soc, 68, 2487 (1946).
4. Balandin and Vaskevich, J. Gen. Chem. (USSR), 6, 1878 (1936;, [CA, 31, 4575 (1937) J.
5. Todd, J. Am. Chem. Soc, 71, !355 (1949).
6. Seibert, Ber., 80, 494 (1947), 81, 266 (l948).
7. McFsdyen and Stevens, J. Chem. Soc, 1936, 584.
8. Natelson end G-ottfried, J. Am. Chem. Soc, 63, 487 (l94l).
9. Price, Mpy and Pickel, J. Am. Chem. Soc, 62, 2818 (1940) .
10. Niemann, Lewis and Hays, J. Am. Chem. Soc, 64, 1678 (1942).
11. Dewar, J. Chem. Soc, 1944, 619.
12. Albert and Royer, J. Chem. Soc, 1949, 1148.
13. Morley, J. Chem. Soc, 1951, 1971. Alford and Schofield, J. Chem. Soc, 1953, 609.
14. Bamford and Stevens, J. Chem. Soc, 1952, 4735.
NSW SYNTHESES OF PYRIMIDINES Reported by Paul I). Thomas
April 17, 1953
The most common method available for formation of the pyrlmidine nucleus involves the base catalyzed condensation of urea. 3 ; thioureas, guanidines and amidines with active methylene compounds. u Most of these syntheses and others, less frequently used, yield poljy substituted pyrlmidine s. 1'he substltuents are often transformed into those presonx in th is illustrated below by the synthesis i
H
desired comoound. cy to sine ,
Thli
Et-S-C
//
NH
0-0
\
CH.
'NH2 EtO-C ii
0
/
i^/%/\
Y
\
OH
#t3 N
n i v n ' y
NH.
7
*~^\
N
y
NH8
dll. HOI "
->
cy to sine
Some prominent features of pyrimidine chemistry are briefly:
(l) In simple derl or halogen atoms, b aroma tic ch ara c te r (%) Nuclear substi no sit ions which the e;roup can be loose J. normally has when a Bf 4 and 6 marked a" sub s t i t ue n t s i n the aromatic character are introduced into (3) Simple pyrimld substitutions.
the a
vftivefc, containing alkyl, u.t no »0H or -NH3 groups, and behaves like that of o tuents vary m ■''heir behav y occupy > At "Position 5 t y described as similar 1 ttached to an aromatic nuc! e\ lations from normal beha1 se 00 si t ions are mure or 1< diminishes progressively a positions 2- 4 and 6c ines appear to be very resistant to electrophillr
aryl, nitro groups the nucleus has yridine <,
ior according to he properties of those which it leus. At positions vlor are observed, ess labile c The s -OH or NH3 groups
Very recently a new synthesis of pyrlmidine s, based on the condensation of orthoesters with ureas, has been developed by Whitehead.3 This may be illustrated by the following reaction sequence:
X XX
, v II s K I J
(1) 2 RN-C-NHS + CH(O02H5)3 > RNH^C-N=CH-NH-0-NHR+?C2H5OH
1 1J
(R=H, alkyl, or aryl, X=0 or R= alkyl, X=S)
-2- XX XX
. . H |l II II
(2) RNH-C-N=CH-NH-C-NHR + YCHgZ — * RNH-C-NH-CH-NHC-NHR }
III Y-CH-Z
X r IV
RNHC-NH-CH=C-Z
V (Y and Z = CN, CH3CO, C03H, C0303H5, CONH3 or COCOsC3Hs)
(3) Cycllzation of the ureidoethylenes, for example:
R R
I j Np0£,t > Ij I
t ^C-C03Et N C-G02Et
NH
CH
In the reaction of ethvl orthoformate with ureas (Equation l) scyl ureas (I, R=CH3CO, X=0; Pnd N,N'-dialkyl ureas failed to react under the same conditions. Aromatic ureas (I, R-aryl, X=0) caused cleavage of the N-C bond of the urea; thus, the desired N;N»-di- carbamylf ormamidine (II) was not obtained as the major product.
The methylene compounds (ill) undergo addition to the nitrogen- carbon double bond of the formamidines (II ) producing the unstable adducts (IV ) which, by loss of a molecule of the original urea, yield the ureidoethylenes (V) . The rate of addition was shown to be dependent Upon the activating grouos Y and Z. The order of activity is -C03K>-CN)> -C0CH3^ )-C03C3H5. Malonic acid reacts in 3-4 minutes while malonic ester requires approximately 60 hours.
Whitehead3 was also able to obtain the ureidoethylene inter- mediates (V, R^alkyl, X-0, Y=Z=G0303H5) by the reaction of ureas with diethylethovyrcethylenemalonate. These intermediates could be cyclized to 5-carbethoxyuracils, the amides of which showed diuretic activity.
Simple Pyrlmidlnes
Recent attempts have been made to correlate the structure of Polysubstltuted pyrlmidlnes with their infrared4 and ultraviolet5 absorption soectra, but these attemnts have been handicapped by th lack of suitable model compounds. Several English worker se_1 3 have undertaken the preparation of monosubstltuted pyrlmidlnes for fundamental "Physical, chemical and biological studies; as a result, new pyrlmidlnes have been reported, and some older syntheses im- proved.
In order to nreoare the parent compound, Whlttaker13 has re- Ported 92/ yields of pyrimidine by use of palladium-charcoal with MgO on 2r-chlorooyrimidlne. Other less successful methods are also reported.
-3-
5-aminopyrlmidine has "been prepared from uracil by the follow- ing method:12
OH OH
ho_<</ <N HN°3 s m-4. ^\no3 fo"13 > CI -^
Fe(OH)
[H]
?d/C, %0
^
N .
NH:
A new synthesis of mercaPtopyrimldiness, 8 is illustrated by the example below:
V
^
N
S II
NH,-C-NH.
%
V
Eton
^
N
N
V
The yields are from 50-90^, and with the monochloro compounds no intermediate products were isolated. This method is also successful with dlchloropyrimidines, it was not successful T"rith the corresponding bromo derivatives. This Is p considerable improve- ment, in many cases, over the "orevious methods. The mercspto com- pounds are useful intermediates for obtaining other pyrimidines, for example:
% ,N 3H
N
-J
CH3NH2
I! -> N
[H]
N ^NH0H3
Ni(R)
N
V
In considering the structure of hydroxy; merca'oto or amino- pyrimidines, it is important to note that in solution an equilibrium will exist between the possible tautomeric forms.
_^
H.
sM
N
N
By comparison of the ultraviolet and infrared spectra of the hydroxyoyrlmidines with those of the corresponding methoxy deri- vatives it has been shoTTn that 4-hydroxyoyrlmidlne11 and 2-hydroxy- Pyrimidine6 exist mainly in the laotam forms in aqueous solution. Similar comparison of aminopyrimidines and the corresponding dimethylamlmo compounds have shown that the amino form is ore4. dominant.
_4-
New snytheses based on 5-aminopyr Imldlne s .
Todd15'16 has earlier shoT-m that condensation of eruanidines with ©zo-l,?-dicprbonyl compounds followed by catalytic reduction of the resulting intermediate 5-azopyrlmidines is a useful route to Doly substituted 5-amlnopyrimidlnes.
CH H8N-C' + OH-N=N.
^NHg 0=C
^>
3
[H] v N
VI
This work has been extended by Rose.17 The -properties of the pyrimldine VI were studied. It wqs shown to be freely water soluble and its solutions exhibited strong fluorescence. Acetylation and benzoylation give the 5-acyl derivatives. These did not react with nitrous acid. The 5-amino compound could be diazotlzed to form a rather stable dlazonium salt which could be easily cyclized in good yield to the corresponding tetraazaindene.
H,N N
This prompted a study of other 5-aminopyrimidine s of the type VII in this reaction.
HONO
VII VIII IX
The nature of the substituent X markedly influenced the ease of formation of the pyrazole ring. With X=NH, RN, 0 or S the diazonium salt cyclized preferentially to give triazole, oxadiazole and thiadlazole rings respectively. Compounds of the type IX were
3
successfully prepared directly from dlazotized VII when R was alkyl, phenyl, dlalkylamino, alkylthio and carboxymethylthlo. The substituents R1 were' NH3-,CK3S-, CH3NH-, (CH3)SN-, EtNH-, nC3K7NH-, 1C3H7NH-, nC4H9NH-, iC4HgNH-, n05H^iNH-, piperidino-, guanidino, p-chlorophenylguanldino and p-anlsldino. Yields In some instances were low.
Other heteroblcyclic systems were prepared from compounds of the type VII as shown below.
H2N N CI
YY __,
N A HC1
Y^NHa
CH,
H3N N S HaN ■ \ #aC°9H8N N Sv
N
/*
N
A
NH;
HOI
CH.
N
C=0
BIBLIOGRAPHY
1. 2. 3. 4.
5. 6. 7. 8. 9.
10. 11. 12. 13. 14. 15.
16.
17.
B. Lythsoe, Quart. Revs., 3, 181 (1949).
C. W. Whitehead, J.A.C.S., 75, 671 (1953) . C. ¥. Whitehead, ibid. . 74, 4267 (l95g) . I. A. Brownlie, J. Chem. Soc, 3052 (1950) .
Cavalier! and A. Bendich, J.A.C.S., 72
L. M. M. M. M.
F. P. P. P.
9587 (1950).
W. McOmie, W. MoOnie, W. McOmie, McOmie and
J". Chem. Soc, 1218 ibid.. ,3715 (1952) . 3722
W. McOmie, ibid., 331
ibid.,
R. N. Tlmms,
ibid., 4942 T1953).
(1952). ibid.,
(1952).
V. Boarland and J. F.
V. Boarland and J. F.
V. Boarland and J. F. P. V. Boarland, J. F. W. (1952).
M. P. V. Boarland and J. F. B. J. Brown and L. N. Short, N. Wittaker, ibid. , 1565 (l95l77
J. Braddlley, B. Lythgoe and A. R. Todd, Ibid., 571 (1943). J. R. Marshall and J. "Walker, ibid. , 1004 R. Hull, B. J. Lovell, H. T. Openshaw, L. Todd, Ibid., 357 (1945). R. Hull, 3. J. Lovell, H. (1947). F. L. Rose, ibid., 3448 (1952);
(1951)
4691
T>
(1951).
C, Pay man and A. R.
Openshaw and A. R. Todd, ibid., 41
OXIDATION OF INDOLES Reported "by Allan S. Hay
Aoril 24, 1953
In 1881 Jackson treated 2- methyl indole with alkaline permanganate and by oxidative fission of the 2,3-double bond, acetylanthranilic acid wa3 obtained. The same product was obt^inc
tht
H
CH.
COOH NHCOCH3
by Baudisch and Hoschek in 1916 by ©u to -nidation of 2-methyl- indole in the presence of sunlight.
More recently interest has been revived in this type of re- action because it has provided routes to various 2-«minoaryl ketones. Chromic acid in glacial acetic acid has commonly been used, notably by Schofield^ and his coworkers. Excellent yields were obtained by oxidation of some nltro-2, 3-diphenylindoles to the corresponding benz.oohenones (Ri = R3 = 0) . When the sub- stituent in the 2-oosition (Rs = fllkyl) is alkyl good yields may
NO.
. COR!
■>
NO.
n^Anh
also be obtained, however, when both groups are alkyl (P-i = R2 = alkyl) vigorous oxidation occurs resulting in appreciably lower yields.
Nitration with mixed acids of 2, ?-dialkylindoles yields the expected 5-nitrolndoles. However, when the indole nitrogen is acylated, different results are obtained4. For example, with
A
I !
w
R3
Ri
-R3
or
<S\
NO.
VnA
.Ri -Rs
OH
II
tetrahydrocarbazole (Rx + R3 = _(CH3)4-) when R3 is acetyl, carbethoxyl or ohenylacetyl, a oroduct of tyoe I is obtained. When R3 is benzoyl, tyoe II Is obtained which may be converted to type I by boiling in ethanol. By boiling II in aauecus potassium
_o_
hydroxide the 2,3 "bond may be broken to give III.
-COCCH3)4C03H
i
"V^
.NHCO/3
III
Glycols of type I have more recently been obtained by oxidation with osmium tetroxlde6. The N-acetyl derivatives of tetrahydrocarbazole, and 2, 3-dimethyl-, 2,3-diphenyl-, and 2,3,5- tr ime thy 1 indole with osmium tetroxide in pyridine - benzene gave, quantitatively, highly crystalline osmic esters, which were hydrolyzed to the glycols in moderate yield. Crystalline products were in no case obtained from N-unsubstltuted indoles; in such experiments colloidal osmium was liberated.
Ozone has been used by various authors for oxidative fission of the 2,3-double bond. Schofield et al5 have treated a number of indoles with this reagent in various solvents. The yields of ketones obtained range from zero to about sixty per cent, sur- passing those from chromic acid oxidation only when indoles with alkyl groups in the 2-, and 3-T)ositions are oxidized.
In some instances the isolation of relatively stable, crystalline or.onides has been reported. Witkor) and Patrick6 have studied extensively the ozonide (V) obtained from t>henylskatole (IV). It is unusually stable and may be recry stalllzed from boiling ethanol and stored in any nuantity. Molecular weight determinations have shown that it is a monomer. Benzoylamlno- acetophenone (VII) is obtained from it by treatment with acid, heat, or hydrogenation with palladium in ethyl acetate7.
<7
CH,
y
0.
H
">
1. LiAlH* or
2. NaBH4 or
3. Pd,H2
V
1. H or
2. ^ or
3. H3,Pd
C0CH3
or
^
■*
i
VI
VII
-3-
The hydroperoxide (*Vl) obtained from phenylskatole by autoxidation in hydrocarbon solvents undergoes an acid - catalyzed rearrangement to give the same product6.
Ozonide s of this type have been formulated as ring-chain tautomers6 by Criegee. Treatment of the crystalline sulfate of the ozonide (V) with acetic anhydride gives N-ben^oyl-O-Pcetyl-^- aminophenol (VIIl). The following mechanism has been put forth.
V 4
H or OH
neutral
H>
VIII
OCOCH.
NHC 0t>
<~
Ss—/-
CK3
I
0
0 I
Vh^
\
H
V°H ln Ac o
->
t
- u^q- <■
H
»
CH,
^N
0
OHr
H
0
*J0
0
H #
By the reaction of the ozonide T-rith metallic ootassium in boiling benzene and subseauent methylation, l-methyl-2-'Phenyl-4- quinolone (IX) was obtained6.
The catalytic oxidation of tetrahydrocarbazole with platinum catalyst and oxygen has been studied recently8. The hydro- peroxide (X) obtained is fairly stable in the dry state but changes rapidly in the presence of polar solvents to the cyclic lactam(Xl).
IX
^\y
XII
o3
Pt
-->
00H
V^N^V
\y
H
r^\ OO^X
k/^'
C0<v
H
XI
-4-
The same lactam is obtained in excellent yield from tetra- hydrocarbazole "by treatment with ozone in methanol at -79°, or by oxidation of 11-hydroxy-tetrahydrocarbazolenlne (XII) with perbenzoic acid8.
BIBLIOGRAPHY
0. R. Jackson, Ber. 14, 885 (l88l) .
Bsudisch and A. B. Hoschek, ibid., 49, 9579 f1916) .
Schofield and R. S. Theobald, J. Chem. Soc, 1505 (1950).
Schofield, Quart. Rev., 4, 391 (1950) .
W. OcKenden and K. Schofield, J. Chem. Soc, 61° (1953).
Witkot) and J. B. Patrick, J. Am. Chem. Soc, 74. 3855 (195?
WitkoD and J. B. Patrick, ibid. . 74, 3861 ClP5?) .
>ritkoT) and J. B. Patrick, ibid.. 73, 2197 (1951 ) .
1. |
0. |
2. |
0. |
3. |
K. |
4. |
K. |
5. |
D. |
6. |
B. |
7. |
B. |
8. |
B. |
-2-
!
%Anh3
l) OH3I
2) ag2o
^
alkali
=.NH
III
/
y\
IV
=0 + NH3
1
H
6^5
NH +
0=C-C«H I
BrCH3
(5) Behavior on catalytic reduction. Reduction of 3-amino- pyridine with platinum oxide as a catalyst yields 3-amino- piperidine.0 However, the reduction of the 2- isomer has been reported to proceed differently upon reduction with either platinum or platinum oxide.10 The reduction of (i) with platinum as a catalyst gpve a quantitative yield of 2-iminopiperidine while re- duction with Platinum oxide as a catalyst gsve plperidine. Attempts to prepare 2-aminopiioerldine from 2-iminoPiperidine by hydrogenatlon with a platinum oxide catalyst were unsuccessful; only starting material and piperidine could be isolated. This behavior has been explained by assuming the existence of 2-amino- pyridine in the imino form (lb).
+ NH.
EVIDENCE FOR THE AMINO FORMS.
( (I) Resonance energy. It has been well established that 2- and 4-hydroxypyridine exist mainly in the pyrldone form.11 From
this fact an analogy has been drawn between the hydroxypyridlnes and their amino counterparts. This analogy is not as .well grounded as it may appear. In the tautomeric equilibrium -NH-C=0^-N=C-0H,
-3-
the former tautomer is more stable by about 10 kcal/mole. 5 Thus, any energy gained from resonance contributions due to the enol form is counterbalanced by the gain in bond energy of the lie to form, and the keto form might ^ell be exoected to predominate. On the other hand, in the amidine system -NH-C=NH ^ -N=C-NH3 there is no change in bond energies15 and , therefore, there is nothing to compensate for the loss in resonance energy resulting from a change from the aromatic amino to the non-aromatic imino form. The aminopyridines should, therefore, be expected to exist in the amino form.
(%) Absorption spectra. A study has been ma
the aminopyridines and some
violet absorption of atives.13'14 The
atives.13'14 The soectra of l-methyl-2-pyridone~ (dimethylamlno)-oyridine (V) (definitely in the a 8-amlnopyridlne (I) were obtained and compared. 2-aminopyridine (i) corresponded to that of (V) a of (ill), thus, Indicating that (i) is in the ami The spectra of l-methyl-4-pyridone-imine (VI), 4- pyridine (VII), and 4-aminooyridine (II) produced No evidence was obtained for the ore se nee of the and lib).
de of the ultra- of their deriv- imine (ill), 2- mino form) , and The spectrum of nd not to that no form (la) . (dime thy lamino)- the same results, imino forms (lb;
The infra-red spectra of the aminopyridines and some related compounds were obtained and compared.14'16 The spectra of 2- and 4-amln ©pyridine closely resembled those of aniline, p_-nitroaniline, cc-naphthylamine, and 3-aminopyridine. On the other hand, the spectra of 2- and 4-aminopyridine differed from the spectra of l-methyl-2-pyridone-imine (ill) and l-methyl-4-oyridone~imine (Vl) respectively. These studies indicate that 2- and 4-aminopyridine exist in the amino forms rather than in the imino forms.
(5) Determination of the tautomeric equilibrium constant .
Since, by the addition of a "proton, the same resonating cation is
obtained from both the amino and the imino forms, the following
equilibria coexist, where Ka( amino) and Ka f imino 5 ^re the acid
dissociation constants of the option as the conjugate aold of the
amino and imino forms respectively, and K-^aut ^-s ^^-e tautomeric
equilibrium constant:
+
+
Kafpmlno) = [amino] [H ]/[ cation] Ka (imino) = [ imino] [H ]/[cation]
K
taut
J I Ka (amino)
= [amino] / [imino] = Ka (amino)/Ka (imino)
(imino)
NH.
+
H
•v
*V
;aut
u\
+
-4-
The difficulty of determining the dissociation constants of both of the tautomers limits the Applicability of this procedure. However, significant results have been obtained by the use of appropriate approximations* The Ka of the corresponding 1- methyl pyridone-imine (Ka(Me)) was substituted for Ka(imlno). Likewise, the Ka of the amlnopyridines themselves was substituted for ^&(Qmlac with the assumption that the concentration of the iinino form is very small in comparison with the concentration of the amino form,
pKa pKa(Me) KtflUt -^-F (kcal/mole)
2-aminopyridine 6.86 12. 2C 2xl05 7.3 4-aminopyridlne 9.17 12.5 2xl03 4.5
The large values for K-fcaut obtained in the experiment indicates that the predominant specie s present in both 2- and 4-anlnooyridine is the amino form rather than the lmlno form.
The validity of the calculations holds only for the dilute aqueous solutions in which the ^Ka values were determined. On the other hand, the large ^ F values indicate that a change in solvent is not likely to change the position of the equilibrium.
CONCLUSION.
The data presented support the assignment of the amino rather than the imino structure to the 2- and 4-amlnopyridlnes.
BIBLIOGRAPHY
1. S. J. Angyal and C. L. Angyal, J. Chem. 3oc., 1461 (1952) .
2. W. Marckwald, Ber., 27, 1317 ("1894).
3. L. 0. Craig, J. Am. Chem. Soc, .56, 231 (1934).
4. N. V. Sidgwick, The Organic Chemistry of Nitrogen. Rev. by T. W. J. Taylor and W. Baker, Ovford, 1942, P. 529.
5. A. E. Tschitschibabin, R. A. Kcnowalowa, Ber., 54, 814 (l92l) .
6. A. E. Tschitschibabin and E. D. Ossetrowa. ibid., 58, 1708 (l (1925). ~~
7. A. E. Tschitschibabin, ibid.. 59B . 2048 (1926).
8. A. E. Tschitschibabin and M. Plpsohenkowa. ibid., 64, 2842
(1931).
9. H. Nienburg, Ibid.. 70S, 6-^5 (1935) .
10. T. B. C-rave, J. Am. Chem. Soc, 46, 1460 (1924).
11. R. C. Elderfield, Heterocyclic Compounds. I, John Wiley and Sons, Inc., New York, N. Y., 1950, p. 435.
12. R. C. Elderfield, Ibid., p. 444.
13. L. C. Anderson and N. V. Seeger, J. Am. Chem. Soc, 71, 340
/ \ -7 7 — — — 7
(1949).
14. J. D. 3. Coulden, J. Chem. Soc, 2939 (l°5°).
15. G-. E. K. Bmnch and M. Calvin, The Theory of Organic Chemistry. Prentice-Hall, Inc., New York, N. Y. , 1941, td . 289.
16. C. L. Angyal and R. L. Werner, J. Chem. Soc, 2911 (1959).
REACTIONS OF 1, 1-DIARYLETHYLENES Reported by Robert J. Lokken April 24, 1953
Introduction
Some interesting reactions of 1, 1-diarylethylenes have been ob- served in which the ethylenic double bond behaves not as an aliphatic, but as an aromatic double bond. These compounds react with acid chlorides and with thionyl chloride in a manner analogous to that of Friedol-Craf ts reactions with the exception that no Friedel-Crafts catalyst is employed. As in the Frledel-Craf ts reaction, substitution rather than addition to the double bond takes place.
Reaction with Acid Chlorides
Recently, Bergmann has investigated the reaction of 1, 1-diaryl- ethylenes with acyl chlorides in the absence of a Friedel-Crafts catalyst. Aluminum chloride could not be used because it promotes dlmerizatior. and nuclear acylation. However, since aromatic hydrocarbons have been acylated at high temperatures without a catalyst,6 it is not too much to exoect that these ethylenic double bonds, which are so susceptible to polarization, could be acylated in the absence of a catalyst.
Indeed, Berermann found that 1, 1-diP^enylethylene was acylated by benzoyl-, cinnamoyl-, and fumaryl chlorides to g-lve the corresponding ketones, fumaryl chloride reacting with only one mole of olefin. When phenacetyl chloride was used, what at first appeared to be a different reaction occurred. Two moles of acid chloride were used Per molo of olefin and the product was not the expected ketone 'l) <■ However, treatment with alcoholic potassium hydroxide converted the product to (l) , Since the ultra-violet absorption spectrum of the original product resembled very closely that of 1. I, 4-tripheayl butadiene, Bergmann decided that the re- action proceeded as exoected to give the ketone fl), but that the enol form of the ketone was benzoylated to give the product (il) which was isolated.
oo
C6H5 /, CSH5 H
^;C=CH2 + CSHB-CH2-C ^ ^0=CH-C-CH2-C6H5
C6V "CI / Csh5.
I
0 0
x C h ?H C6H5-CH2-C-Cl n v 0-C-C6H5
* C=CH-C=CH-CSH5 £ ' ^C^CH-C^CH-CqHb
CsH5 ale. KOH CeH5
II
Of several saturated aliphatic acyl chlorides which were tried, none reacted with 1, 1-diarylethylenes because they decomposed at a temperature below that necessarv to Promote this tvoe of re- action (l90-?00°).
-2-
ReagtiQB with Thlonyl Chloride
1, 1-Dianlsylethylene, according to Petal and Bergmann, is so active that it reacts with phosgene to produce the corresponding acid chlordie and hydrochloric acid.1'7
0 if
f
0=CHa + Cl-C-Cl — 6-,S .^/CH30
R0° ?V
S
\
0
N
Vvic^CH-a-c:
yj
+ HOI
Following this observation they experimented with thionyl chloride. a formal analog of phosgene, in the hope that it too might react with diarylethylenes. On treating the reaction mixtures with water, they obtained the sulfinic acids corresponding to the di- arylethylenes which were used. However, the major product was the diary lvinyl chloride, which could be formed from the sulflnyl chloride by loss of sulfur monoxide. This is the most convenient method for making diary lvinyl halides.
Ar,
Ar"
C=CH3 + S0C1S
-»
Ar-
Ar
0 C=CH-3-Cl
Ha0
Ar.
Ar
C=CKC1
:!
C=CH-S-OH
Reaction with Oxalyl Chloride
The reaction of 1, 1-diarylethylenes with oxalyl chloride is probably the most attractive of all from a synthetic T^oint of view. The primary product is a p,|3-di substituted acrylyl chloride, which can be hydrolyzed to the nnid, ^hir>,h Crn in turn be- v»ydroeen«ted catalytloally to the p, p-disubstituted Propionic acid.
f . . ) (COCl)3 / s ;v N m
m3o-<> >\4_c=ch3 :..\/ch3o^' ^\J-c=ch-c-ci
-S)
J
H30^
fcHgO-^" "^>1-.0-CH-COOH
H3 / \
Ni
COOH
-3-
Reaction with Quinones
The condensation of quinones with 1,1-diarylethylenes has been studied by Gates. The mechanism of the reaction is essential!: the same as that of the other reactions discussed. There is a slight difference in that the attacking species is not an ion,
ii
but another polarized molecule. (-)
(?)
Ar -'
\
A>
->
o
0 II
/NAs
OH
V
C=CH.
s\
^s
/j
OH
Apparently this type of condensation is not as facile as that of acid chlorides T-Tith diarylethylenes. For the quinone con- densations, compounds with strong electron-donating groups on the aromatic nuclei are necessary. For example, 1, 1-dlanisylethylene condensed with a- naphthoquinone but not with J3 -naphthoquinone, while 1, 1-bis- (p-dimethylaminophenyl)-ethylene reacted readily with both a - and 6 - naphthoquinone.
Mechanism
The surprisingly high degree of aromaticity of the ethylenic double bond is due to its conjue,ation,>rith the aromatic nuclei. At the approach of a Polar reagent, A^-, polarization of the double bond is enhanced because the Positive charge which is developed on the 1-carbon atom can be dispersed over the two aromatic rings. The reagent attacks the center of induced negative charge, and the resulting highly resonance-stabilized carbonium ion can then lose a proton to form the substituted diary lethylene. This mechanism is exactly analogous to that of aromatic substitution, as is shown in the following scheme:
_4-
<z>ytjw <S\© ?
^ ^
C=CH
C-GH-A
VC=CHA
<zy —4 <I>
/
-H
®
*
Aw
V'-1
Resonance Structures
There is much exnerimental evidence to substantiate the proposal of an Ionic rather than a free radical mechanism for these reactions. For example, Kharasch has found that the reactions of oxalyl chloride with unsaturated hydrocarbons are neither catalyzed by free radical catalysts such as light and peroxides nor inhibited by free radical inhibitors. Kharasch has found further that com- pounds which readily add reagents of the type HX by a polar mechanism react with oxalyl chloride, while others do not.5 Crates has proposed that the driving force behind the reactions of quinones with 1, 1-diarylethylenes is the electron deficiency of the quinine and the electron-donatine Power of the polarized ethylene molecule.4
The effect of substltuents on the aryl nuclei also supports the ionic mechanism. It has been found that electron-donating substltuents increase the rate of reaction, while electron-with- drawing groups decrease it.1 This would be expected in view of an ionic mechanism, because the more negative aryl nuclei would be better able to distribute positive charge.
BIBLIOGRAPHY
1. 2. 3. 4. 5. 6. 7.
Ghem. Soe 195?, 25 66, 124
70, 1612 (1948).
Bergmann, e_t al, J. Ai
Bergmann, J. Ghem. Sou.,
Gates, J. Am. Ghem. Soc, .66, 124 (1944).
Gates, C. A., 42, 4609f.
Kharasch, J. Am. Chem. Soc, 64, 333 (194?) .
Nenitzescu, Isacescu, and Ionescu, Ann., 491, 210
Patai and Bersmann, J. Am. Chem. Soc, 72, 10^4 (1950)
(1931).
PARTICIPATION OF NEIGHBORING GROUPS IN ADDITION REACTIONS Reported by Fabian T. Fang May 1, 1953
Neighboring groups which participate in nucleoohilic replace- ment processes with relatively large driving forces1 can also be expected to participate in addition reactions to the olefinic linkage initiated by electrophilic reagents3. The Participation takes place cither by a concerted mechanism or by a two-step pro- cess which involves a non-classical intermediate3.
A A
^•S — . . ^S-v,
(-4/ I -*
C-C=C C-C-C
Br - Br f Brs-'
A classical analogue of the latter phenomenon can be found in the isomer ization of 1, 1, 3-triphenyl-?, 3-epoxy-l-nrooanol into l,3,3-triphenyl-2,3-epoxy-l-oroPanol in the presence of cold, dilute methanolic potassium hydroxide4.
O^P 0
CCTSH5 ) 3C-CH-CHCeH5 \ ( C6H5 ) 3C-CH-C"CSH5
Neighboring Hydroxyl Group
Dime thy lvinylcarbitlbl ™a s found to give 1-bromo-?, 3-epoxy-3~ methylbutane on treatment with hyoobromite, whereas allyl alcohol failed to yield the corresponding eoo-j.y compound.5-
OH {TTj 0
t x ' CEr , N />
(CH3)3C-CH=CH3 > (CH3)3C-CH-CH3Br
Neighboring Acy lam i no Groups
The acyl-mino groups are examples of so-called complex neigh- boring group a with ra'Ciher lavut driving f or^ss? s* 7 and turn out to participate iu a^cnt/.cn re cot lend in a very useful ^manner. For example. Tfi—y. -n,ie.tl'.-.oxy'c»&nz.ovl£ulyiar.iiric gives the bromooxazoline in high yield when tre&tbcl in acetic acid with N-bromosuccinimlde3.
CGH40CH&-p_
H-K ' > 0 ^ >
I V_) -H
CH3 CB=CH*
^ <^
Br-X
C6H50CH3- 1 |
-B |
|
i |
1 |
|
! CH3- |
I CH-CH |
2Br |
■2-
This type of reaction constitutes a way to set up three functional groups with a definite stereochemical relationship.
Neighboring CgrboTyl Group
In an early study of the mechanism of addition reactions, Tarbell and Bartlett8' reported the formation of hplcgenated p- lactones from the sodium salts of dimcthylmalelc and dimethyl- fumaric acid3 by treatment ^ith chlorine water or bromine water. The mechanism of a two-ster> addition nrocess in fast succesion was proposed and the enhancement of ring closure by methyl sub- stitution was noted.
CH3 ^CH3 CH. /5H3 CK3 CH3
/C^^ ^G03^ ' CI
C (-> 0X
C— 0
+ 01®
It has long been recognized that Y,& -unsaturated acids fre- quently react with electrophlllc reagents to form /. -substituted- V -r>entanolactones Instead of the simple addition Product. Fittig and Hjelt9 reported in 1883 the laotonlzation of diallylacetic, allylmalonic, and diallylmalonic acids by treatment with bromine or hydrobromic acid. The interesting formation of nonodilactones was observed in the last case.
CH3=CH-CH3 CH2-CH=CH2 BrCH3-CH-CK3 CH2-CH-CH2Br
H02C^ ^C02H ^—> 0— 00--^" ^CO-^O
Craig and vltt10 found that lactonlzation occurred when 2,2- diphenyl-4-oentenolc and 2,?-dlpbenyl-4~methyl-4-r>entenoic acids were treated with sulfuric acid or phosphoric acid to give the valerolactones in excellent yields. The same acids reacted with bromine to give good yields of the corresponding bromovalero- lactones. The reaction scheme orooosed by Craig11 might involve a cyclic bromonlum ion. /—
(CSH5)2C B^ > (C6H5)2Cr | ^ (C6H5)2C \o + H
H2C^ H2C >V H3C /
C=CH3 ^-C-CH2 xC-CH2Br
(i
r
K = H or CH3
Br
-7\
-3-
The fact that 2,2-diphenyl-4-methyl-4-pentenoic acid lactonizes more readily with bromine or sulfuric acid than does 2,2-diphenyl-4-pentenoic acid suggests that the methyl group facilitates attack at the V-carbon atom to form the 5-membered ring. Analogous substituent effect was found in the acid hydrolyeer of ethyl methallylacetamidomalonate and of ethyl allylaeetamido- malonatels.
Similar Participation of neighboring carboxyl groups in addi- tion reactions has been observed, with oleanolic acid13 and 2- phthalimldo-4-pentenoic acid14. Lactonizations have been demon- strated by the use of acids, bromine, acetyl hypobromite18, and mercuric salts16 as electroPhilic reagents. Even iodine (both alone and catalyzed by mercuric chloride) and cyanogen iodide have been shown to possess sufficient cationold reactivity to bring about lactonization of"/ , £ -unsaturated acids, the product in each case being a^ -iodo-Y -pentanolactone17. 2, 2-Dlphenyl-4- pentenolc acid, 9-allyl-9-f luorenecarboxyllc acid, and 4-pentenolc acid all give the expected iodolactone when treated with cyanogen iodide. This was claimed as the first observed spontaneous re- action of cyanogen iodide with carbon-carbon double bond18.
Neighboring Carboxamido Groups
The participation of neighboring carboxamido groups in addi- tion reactions has been investigated by Oraig11 with a number of amides derived from 2,2-diphenyl-4-pentenoic acid. When the amides are treated with bromine in carbon tetrachloride, a facile ring closure occurs in each case to give the corresponding 3,3-dlphenyl- 5— bromomethyl-2-imlnotetrahydrofuran hydrobromide in good yield.
Ri £± Ri *}Ri
.C^N' C-N' C=n'
(CsiUsC '^0NR3 (C6H5)2C/^R3 (C6HB)gCf V R
HflC
H3C &/ H3C /
XC H= C H 3 ^CH- OH 3 ^C H- CHgB r
V^ Bl£> Br°
Ri = H CH3 C3K5 CK3CH3 R3 = H CH3 C3H5 CH3CH2^
Neighboring Carbalkoxyl Groups
The esters of V , q -unsaturated acids also lactonize readily with the elimination of the alcoholic alkyl group11'15'18. The stereochemical proximity of the carbonyl oxygen atom to the Y - carbon atom allows it to participate in the formation of a resonance stabilized, cyclic oxonium salt.
_4-
,^R
R3Cr *0
HaC.
I \>
X
CH=CH3 Br-Br
*
C-OR
L3V
H2C
xCH~CH3Br
C=OR R3(T \
I 0
H3CN X
xCH-CH3Br
Br
&
Among the poss (l) the formation o S 2 type substitutl or the brornolactone or Ex mechanism, re lactone and an olef competing dibromide of the oxonium salt of the reaction med
ible reaction f the bromola on with comr>l
and an alkyl spectively; ( in via an E3
formation.
depends uoon ium.
s of the cyclic in ctone and an alkyl ete inversion; (2)
bromide or an ole ?) the formation o tyDG elimination; The actual type of
the nature of R
termed iate are: bromide via an the formation
fin via an S 1
f the bromo-N
and (4) the decomposition
and the nature
Ethyl bromide was isolated in considerable quantities from the bromination without solvent of diethyl allylbenzylmalonate; but hydrogen bromide was formed in 84/ yield when the bromination was carried out in chloroform solution. The same ester reacted with acetyl hyoobromite to give 50-65/ of ethyl acetate and 63/ of the brornolactone. .
The d- and JL-2-octyl esters of 2, 2-diphenyl-4-oentenoic sold reacted separately with bromine in chloroform solution to give 39/ of the 2-bromooct^nes with 97.5/ and 98.5/ inversion of configura- tion, respectively. However, the neooentyl ester of the same acid gave on bromination 41/ of the brornolactone and 59/ of hydrogen bromide. The steric hindrance of the neooentyl group largely eliminates bimolecular attack and favors solvolytic reactions with rearrangement of the carbon skeleton.
Substitution exerts remarkable bromide formation Methyl 2,2-dioheny quantitative yield ly similar methyl dibromide in addit obvious that the c exhibit a smaller phenyl compound.
, d
in the a-oosition of "> influence on the relative e competes with brornolactone 1-4-oentenoate was found to s of the bromolactone1<:), 11„ 9-allyl-9-fluorenecarboxyla ion to a 48/ yield of the b oplanar benzene rings of th steric effect than that exh
-unsaturated esters xtent to which di- formation9'15.
give virtually
whereas the apparent- te gave 33^ of the romolactone15 . It is e fluorene derivative ibited by the di-
C03CH3 CH,-CH=CH,
.C02CH3
^CHo-CH^CH.
-5-
BI3LI0GRAPHY
1. Winstein and G-runvald, J. Am. Chem. 3oc, 70, 828 (1948).
2. Winstein, Goodman and Boechan, Ibid., 72, 2311 (lQ50).
3. Winstein, Abstracts of Papers, Eleventh National Organic Chemistry Symposium of the American Chemical Society (June 1949), p. 72.
4. Kohler, Richtmyer and Hester, J. Am. Chem. Soc, 53, 205 (1931;
5. Winstein and Goodman, unpublished work.
6. Winstein, Hanson and Crun^ald, ST. Am. Chem. Soc, 70, 812
(1948).
7. Acker, Organic Seminar, University of Illinois, March 9, 1951.
8. Tarbell and Bartlett, J. Am. Chem. Soc, 59, 407 (1937).
9. Fittig and Hjelt, Ann., 216, 52 (1883).
10. Craig and Witt, J. Am. Chem. Soc, 72, 4925 (1950).
11. Craig, Ibid., 74, 129 (1952 ).
12. Coerlng, Cristol and Dittmer, Abstracts of Papers, 113th Meeting of the American Chemical Society (April 1948), P. 69L.
13. Wintersteln and Hammerle, Z. Physiol. Chem., 199., 56 (l93l).
14. G-audry and G-odin, Abstracts of Papers. 123rd Meeting of the American Chemical Society (March 1953;, p. 14M.
15. Arnold, Campos *md Lindsay, J. Am. Chem. Soc, 75, 1°44 (1953).
16. Rowland, Perry and Friedman, ibid., 73, 1040 (l95l) .
17. Arnold and Lindsay, ibid.. 75, 1043 71953).
18. Arnold and Lindsay, Abstracts of Papers, 122nd Meeting of the American Chemical Society (September 1952), p. 21M.
BASICITY OF AROMATIC HYDROCARBONS AND THE ISOMERIZATION
OF THE METHYL BENZENES
Reported by Harry W. Johnson, Jr.
May 1, 1953
The basicity of aromatic hydrocarbons has been the subject of several investigations in recent years. Frey1 has reoorted the earlier work in the field.
Klatt3 noted that aromatic hydrocarbons were soluble in liquid HF in the order benzene ^> toluene > m-xylene. Brown3, who corrected Klatt1 8 data for the vanor pressure of the hydrocarbon, obtained the onDoslte order. H*mmett4 has suggested that the re- sults obtained by Klatt were explicable on the basis of the equatta
HF + Ar
ArH
McCaulay and Lien5 studied the reactions of the methylated benzene? with HF-3F3 and obtained values for the relative basicities of the compounds, and Brown3'6 studied the relative babicities of the same series toward HCl (by determination of the Henry's law con- stant for the solubility of HCl in the hydrocarbon). Kilpatrick7 has measured the relative basicities of the methylated benzenes toward HF through measurement of the conductance of a solution of the hydrocarbon In HF. The results of the studies mentioned are summarized in Table I.
Table I Relative Base Strengths of Hydrocarbons Toward Various Acids
Hydrocarbon
Benzene Toluene p_- Xylene jD-Xylene m-Xylene
Relat Basic • HCl 0.61 0.9? 1.00 1.1 1.56
ive ity
Pseudocumene (l,£,4) Hemimellitine (l,2,3 Mesitylene (l,3.5) Durene (l,2,4,5; Prehnitine (l, 2,3,4) Isodurene (l,2,3,5; Pentamethylbenzene Hexamethylbenzene Ethylbenzene i-Propylbenzene t-Butyibenzene
1.36
)1.46
1.59
1.63
1.67
1.06 1.54 1.36
Relative Ba sicity HF~BF33'
0.01 1.0
2
50 40 ca 40 2800 120 170 5600 8700 89000
Relative |
Relative |
Basicity |
Activity |
HF7 |
(Halogenation) |
0.09 |
O". 0005 |
0.63 |
0.157 |
1.00 |
1.00 |
1.1 |
2.1 |
26 |
200 |
63 |
340 |
69 |
400 |
13000 |
80000 |
— |
1400 |
— |
2000 |
16000 |
240000 |
29000 |
360000 |
97000 |
— |
— |
0.13 |
— |
0.080 |
— — |
0.050 |
Bnowns in the re la HCl as comp than durene HCl; and m- toward HF-B that in the of halogena
noted than inversions in the order of basicity occurred tive basicities of some members of the series toward ared with HF-BF3 (for example, mesitylene is more basic
toward HF-BF3, while the order is the reverse toward xylene, mesitylene and isodurene are far more basic F3 than their basicity toward HCl would indicate), and
cases where the inversions occurred the relative ease tlon followed the basicity toward HF-BF3 rather than HCl.
-2-
Further, the complexes formed with the two reagents had different properties. The complexes formed with HOl were colorless, non- conducting solutions which did not exchange nuclear hydrogen for deuterium with DC1, while the comolexef with HF-BF3 are highly colored, conducting: solutions which do exchange nuclear hydrogen for dueterlum with DCl. 3' Q' 9 (Brown9 and others10 have discovered that although AlCl3 and AlBr3 do not react with the corresponding HX in the dry state or in saturated hydrocarbons, they do react in the presence of an aromatic hydrocarbon to give colored, highly conducting systems which are good solvents for the aluminum halides Thus, these systems T-Tould seem to be in the same class as the HF-BF3-Ar complexes discussed above.)
Browns accounts for the differences noted above with the assumption that two different types of Interactions are involved. For the comDlex between KOI and the hydrocarbons he suggests that a 7T -complex Is Involved. The molecular orbitr.l Picture of benzene has the 7T -electron cloud in two rings above and below the nucleus, and it Is suggested by Brown that the HC1 interacts with the cloud without seriously deforming it and without the proton becoming attached to any particular carbon atom of the nucleus. Tnis is the picture of Dewar13 for a "77" -complex, and might be re- presented as follows:
CH.
V
+ HC1
^
\^
— , ^ HC1
For the interaction Postulates the forma which the proton is by a (T -bond (hence In this complex the seriously distorted, distortion is postul be represented as fo
of the aromatic compound with HF-BF3, Brown2 tion of a carbonium ion (or sigma complex) in attached to a single carbon atom of the nucleus the name) which is stabilized by resonance. IT -electron cloud of the nucleus has been
compared with the tt- complex, where little ated. The complex of HF-BF3 with toluene might Hows: h +
v
HF + BF,
<r
V
CH.
V
CHo 1
f — *
A
H
H
e
BF.
J
(The ion in which th ortho to the methyl
e proton has been added to the carbon atom srouo is eaually likely.)
The stability of the HCl complexes is relatively insensitive to the number or position of added methyl grout) s (note the range of basicities of the methylated benzenes toward HCl). The range of basicities of the methylated benzenes toward HF-3F3 is much greater, and the relative basicity of isomers having more than one
_.^
methyl group In a oosition to stabilize the ion by hyoercon jugation is much greater than those In which only one methyl group is in a position to stabilize the ion through hyperconjugation. This is
below with
m- xylenes.
BF.
e
3F,
In the case of the .o- xylene, only one methyl grout) contributes through hyoer con^uerat ion, while in m- xylene both may contribute.
Evidence3 seems to be accumulating which indicates that the full electron donating ability of the methyl group is not realized in systems in which no electron deficiency occurs. For example, the presence of a methyl grout) does not greatly affect the acidity of benzoic acids or the basicity of anilines, but does greatly in- fluence the rate of solvolysls of phenylcarbinyl halides. In the systems at band, the HOI complex has a TT-electron cloud which is not seriously distorted, and, therefore, the influence of the methyl groups be smaller (operating predominantly through an in- ductive effect) than In the case of the HF-3F3 complex In which the methyl group can stabilize the carbonium ion through hyper- conjugation as illustrated above.
In postulating two types of complexes between electrophilic reagents and aromatic nuclei. Brown differs from Dewar3, who would prefer to regard all such complexes as being of the /T' -variety. Since two types are apparently observed experimentally, and since the substitution reactions seem to Parallel the ^--complex stability Brown postulates that the stability of the c~ -complex is the im- portant factor in aromatic substitution rather than the stability of the T^-comDlex as postulated by Dewsr,
APPLICATION TO T^S ISOMERIZATION OF THE METHYLBENZENES
McCaulay and Lien 2 have studied the Isomerlzation of the xylenes and trimethylbenzenes with HF-BF3 at various temperatures. They found that at 80 and 1?0° the xylenes rearrange to mixtures of xylenes, toluene and trimethylbenzenes, of which the xylene fraction has the eouilibrium concentration of isomers (as calculated by Taylor) if the amount of BF3 present w*fl small (0.13 mole BF3/ mole hydrocarbon or less), but that ™ith Increasing amounts of BF3
the amount of m-xylene Increased until a value of 100^ was reached with 3 moles BF3/mole hydrocarbon. At 30° it was noted that no disproportionation occurred although isomerization was complete, which indicates that the energy of activation is less than the energy of activation of disproportionation. The kinetics of re- arrangement of o- xylene were studied at 3 and 30°, and It was found that the reaction wag first order in xylene with an activation energy of 12.7 kcal./mole.
In the isomerization of the trimethylbenzenes it was found that a plot of the mesitylene content of the product vs. the BF3 concentration yielded a straight line, and in all cases in which the BF3 content wag greater than 1 mole/mole hydrocarbon only mesitylene was obtained.
In both cases it will be noted that when sufficient BF3 was present to complex the product ps the Ar-HF-BF3 complex, the only isomer obtained T^s that corresponding to the strongest base toward the isomerizlng mixture . The mechanism suggested for the rearrange- ment is shown below for the case of p- xylene.
CH3
Y
^N
H
®
CH3
V\
&
H
L.CH;
/;
CK,
H
V
v
^
<w.
CF
CH.
Y
CH.
©
(A)
(AB)
m- Xylene is the strongest base and since its salt is the most m-isomer is favored.
(B)
(C)
(toward HF-BF3) of the xylene Isomers, stable, isomerization toward the
On the basis of this work, it is oossible that the formation of m-dialkylbenzenes or 1,3, 5-trialkylbenzenes in the Friedel-Crafts reaction proceeds after the formation of the "normal" products, but the occurrence of such a sequence is not required.
1* 2. 3.
4.
5. 6. 7. 8. 9. 10.
11.
12. 13.
■Hill Book Co, 2013 (1951).
BIBLIOGRAPHY
S.E. Frey, Univ. of 111. Seminar Abstracts, 38 (1951-1952).
H. C. Brown and J.D. Brady, J. Am. Chem. 3oc, 74, 3570 (1952).
W. Klatt, Z.anorg.allgem.Chem. , 234. 189 (1937).
L.P. Hammett, "Physical Organic Chemistry", McG-raw-
Inc, New York, N. Y., 1940, pp. 293-294*.
D.A. McCaulay and A. P. Lien, J. Am. Chem. Soc, 73,
H.C. Brown and J.D. Brady, ib id . . 71. 3573 (ln49T7J
M. Kilpatrick and F.E. Luborsky, Ibid., 75, 577 (1953) .
A. Klit and A. Langseth, Z.Phy sik.Chem, , 176, 65 (1936).
H.C. Brown and H. W. Pearsall, J. Am. Chem. Soc, 74, 191 (1952).
D.D. Eley and P.J. King, J. Chem. Soc, 497° (lP5?77
M.J.S. Dewar, "Electronic Theory of Organic Chemistry", Oxford
University Press, New York, N. Y. , 1949.
D.A. McCaulay and A. P. Lien, J. Am. Chem. Soc, 74, 6746 (1952),
C. K . Ingoia, C. C. Raisin *nn" C. L. Wilson, J. Chem. Soc, 1637
(1936). ' '
THE NEBER REARRANGEMENT Reported by Lewis I. Krimen
May 1, 1953
Introduction. About 25 years ago Neber discovered that cer- tain oximes in the presence of p- toluene gulf onic aclo1 did not undergo a normal Becfcmann rearrangement to give the amide but in- stead yielded a Ipha-emlnoketone s . * Neber and his coworkers con- ducted' a series of investigations1' 3> *' 4' 5 in an attempt to
~CH3-C- I)
N-OT.oe
1. NaOEt, EtOF F. HOI, H30
-CH-
0 M
C-
NH, + 01"
determine the course of the rearrangement and succeeded in two easees-»4 in isolating unstable intermediate -j whose assigned structures (XA, IB) were considered to Provide a satisfactory ex- planation for thn entire reaction mechanism, The presumed sub- stance, 2- (2, 4 diniti'ophenyl)-3"-meth,yl-S-&zirine (lA) was well characterized through analysis and molecular weight determinations.
Until recently this novel reaction has received very little attention and Neber7 e suggestion that the unstable intermediates contain an azacyulopropene ring (azirine ring system)6 warrants critical examination.
Cram and Hatch7 undertook the present investigation in order to substantiate or disprove the original proposal of the azirine structure of the intermediates and to elaborate in more detail the mechanism and scope of the entire rearrangement.
Reactions of the Azirine Ring System. The reactions of Cram's investigation were conducted on compound IA, which was prepared from 2,4 dinitrophenyiacetoneoxime-p- toluene sulfonate as follows:
NO.
0aN
>X
/>
-CHg— C— R
NOTos
1. pyridine
2. ice water
NO.
3. e+"her extrac Oo
£
0nN-
v\
/>
.CH-C-R
IA, R = CH3 IB, R = C6H5
The structural evidence is most slmnly explained on the basis of the azacyclopropene ring.
Ar CH C CH3
.• NNX \ a' b
The ring has been opened at "a" by catalytic hydrogenation using Raney nickel to erlve a semi-dark solid which when chroma tograohed on alumina produced 2,4 dinintrooheny lace tone. An lmlno compound was probably produced which hydro ly zed during the chromatographic
-2-
stage. Hydrogenatlon over a oa Had lum- carbon catalyst in the presence of acetic anhydride produced two isomeric vinyl acetamide compounds. The susceptibility of the carbon- nitrogen single bond of the three member ring to hydrogeno lysis is explained as a conseauence of the strain associated with the azirine ring and with the stabilizing: effect of the nitro grouos uoon the transition state of the reduction reaction.
Treatment of the azirine with lithium aluminum hydride reduced the double bond at "b" to give an ethylenlmlne which was character- ised through its tosyl derivative.
Attemots to produce optically active azirine (IA) both through resolution and by asymmetric synthesis failed.
The Structure of the Intermediate in the Neber Rearrangement.7 While Neber postulated structure I for the intermediate, several others have been oroposed. These structures are:
Ar CH C CH3 Ar
/
3 rtr \J ^ : — ^"3
V
I £ II
Ar C CH CH3 Ar CH C=CHS
y
V V
III H IV
Of these four tautomerically related structures III Is inconsistent with both the hydrolytic behavior of the intermediate as noted by Neber, as well as with the reduction reactions reported by Cram. Structures II and IV contain an N-H linkage and since this band8 is missing in the Infrared spectrum of the intermediate, these two structures become unlikely.
In contrast, structure I is consistent with both the chemical and spectral data. The infrared spectrum of the intermediate has a band occurring at 5. 55y^( which is absent in that of the derived ethylenimine and can be attributed to the "> C=N- stretching vi- bration present in the intermediate and absent in the imine .
The Neber Rearrangement In the Desoyybenzoin System.9 Although Neber converted the p- toluene sulfonate of desoxybenzoin- oxime ?V) to desylamine hydrochloride fVTIl) no4 Intermediates were isolated. Cram and Hatch repeated this investigation with slight modifications to give jcis-2, 3-d Id henyle thy len imine fVIl)
C6H5 — CH2 — C — C6H5 II
Tos — 0'"
V
1, KOEt, EtOH
2, LiAlH4,EtOEt
EtOH
KOEt
■>
CeH
6n5"
UA1K4
CsH5 — CH — CH — ^6^5
\ /
N 1
is
VII
CRH
6n5'
OEt
.CH C-
H
NHoCl 0
■C«K
Sn5
VI
I HC1
I
1
i
4
H20
■C«H
6^5
VIII
Compound VI wss demonstrated to be 2,3-diphenyl-2-ethoxy- ethylenimine and apparently represent a the first known example cf its structural class.10
Evidence for the assigned structure is as follows:
(1) The molecule possesses a molecular formula of CiSHi?NO,
(2) The substance contains one ethoxyl group. (3) When hydrolyzed by aqueous acid, the compound save desylamine hydro- chloride. (4) When reduced with lithium aluminum hydride, the compound gave cls-2,3-diphenylethylenlmine (VII). (5) The ultra- violet absorption spectrum closely resembles that of cls-2,3-di- Phenylethylen-imine. (6) The band in the infrared spectrum that occurs at 2,9/u is evidence for an N-H bond in the molecule. The same band appears in the spectrum of cis-2. 3-diphenylethylenlmlme (VII ) and is undoubtedly due to an N-H stretching frequency.
In order to obtain further evidence regarding the intermediates in the Neber rearrangement, the jd- toluene sulfonate of P_,P'-di- chlorodesoxybenzoinoxime was prepared and submitted to the usual reaction conditions. Although no ethoxyethylenimine was isolated good evidence for its existence was obtained.
Structural Features Necessary Since all the systems that have be arrangement contain a methyl or me function, the question arises as t ed to compounds of such structural nitriles resulted when oxime tosyl to reaction conditions of the Nebe any extension of the rearrangement methinylketoxime tosylates.
for the Neber Rearrangement. 9 en submitted to the Neber re- thylene group aloha to a ketone o whether the reaction is limit- types. An E3 reaction to form ates of aldehydes were submitted r rearrangement, and, therefore, could involve onlv alpha-
The Mechanism of the Neber Rearrangement.9 The first step in the over-all reaction can be generalized in terms of a base in- duced 1,3 elimination reaction (with ring closure) upon which has been superimposed a 1,2 addition reaction. This oicture is essentially the same as the one suggested by Neber.4
-4-
-OTos
-CH = C - 4 >
I
k:
-CH-C-
e n:
®
_CH C-
:n:
5|
A formulation which satisfies all the facts known about the lithium aluminum hydride reduction of the Intermediate is pictured b e 1 o'W c
mt
■ CH-u
\ /
N
H
k LiAlH4
-H.
7? 0&t
_CH C-
Vr
H-
i/-
H
!
H
-OEt
Q
*
-CH-CH
\/
N I
A1H2
H-,0
■>
CH-CH-
\ /
N ( H
The reduction of the ethoxyethylenlmines with lithium plumlnum hydride possibly goes as shown in the formulation.
>C - 0
\
>C - N H
><v
1, UA1H4
2, H20
->
>C - OH
>c
N - C <
H H
The preferential cleavage of the 0-C bond is analogous to the well-known reduction of amides to amines, and the recently reported reduction of oxazolidines to N-substituted alpha- amino alcohols.11
BIBLIOGRAPHY
1. P. W. Neber and A. Friedolsheim, Ann., 449, 109 (1926).
2. P. W. Neber, A. Bursard, pnd W. Thler, ibid., 526 277 (1936) .
3. P. W. Neber and A. Burgard, ibid. , 49?. 281 (1932).
4. P. W. Neber and G-. Huh," ibid.. 515, 283 (1935).
5. P. W. Neber and A. Uber, ibid. . 467, 52 (1928).
6. A. M. Patterson and L. T. Cape 11, "Ring Index", Reinhold Publishing Corp., New York, N. Y., 1940, p. 3. D. Cram and M. Hatch, J. Am. Chem. Soc, 75, 3? (1953) . H. M. Randall, R. Fowler, N. Fuson, and J. R. Dangl, "Infrared Determination of Organic Structures", D. Van Nostrand Co., Inc., New York, N. Y., 1949, p. 6.
9. M. Hatch and D. Cram, J. Am. Chem. Soc, 75, 38 (195?) .
10. J. Fruton in Slderf ield' s, "Heterocyclic Compounds," Vol. I, John Wiley snd Sons, Inc., New York", N. Y., 1050, p. 61„ and F. King, J. Chem. Soc, 1318 (1949).
11. E. Bergmann, D. Lavie, and S. Pinchas, J. Am. Chem. Soc, 73, 5662 (1951).
7.
8.
PHOTOCHEMICAL REACTIONS Reported by Ruth J. Adams
8, 1 4
May 8, 1953
Although in solution photochemical reactions usually oroneed by paths involving free radicals and radical- ions, it is in some instances Possible to use light in promoting, under mild condition chemical reactions which are known to hpve ionic mechanisms. For example, esterif ication of trans-hevahydroterephthalic acid15 has been found to proceed in the absence of strong acids or bases to give a fifty percent yield of the trans-dimethyl ester simply by the irradiation for two weeks of a methanolic solution of the acid
PHOTO-ISOMERIZATION
In oerhpos no other class of reactions is lle-ht a more useful synthetic tool th^n it is in the isomerization of or^anin com- pounds. Its utility in the interconversion of ois- -and trans- form f is veil k nowm. In many cases, photoisomerization is the most convenient wpy to accomplish euch conversions, in others, it is the only way, e.g., the preparation of cls-azobenzene from the normally obtained trans-azobenzene . Supposedly, be cause-fT bonds are weaker thpn (f bonds, the change from els to trans and con- versely proceeds by means of a dlradica] intermediate which is capable of free rotation around the central bond.
R
3^c=c/Rl
VR;
h-V
R3
Rx
,Ri "R3
.S*
Rr
Rs
u=c"
Rix ^Ri
PHOTO- ADD IT ION
The light catalyzed production of free radicals or atoms and their subseauent addition to unsaturated compounds is exemplified by the photo-addition of the halogens, halogen acids, alcohols, mercaptans, thioacids, etc. to olefins. The monomeric addition of phenanthraquinones and phenanthraquinonimines to unsaturated compounds11 * x 3 is of some "theoretical interest since it presumably occurs as the result of the formation of a diradical.
/>S
hVv,
#hc=ch/>
T
rr
Similarly, the addition of benza^dehyde to Phenanthranuinones probably Involves the Initial Production of the same type of radical. Acetaldehyde, p-aniealdehyde, ^nd benzaldehyde add very rapidly, but the reaction is considerably "lower T,rith 2-methoT.y-l-naPhth- aldehyde, possibly because of sterie hindrance (ortho effect).13
_o_
Air is excluded in these reactions to prevent side reactions due to oxidation.
w
v
y"^o
V
b)
d)
,) j6c« +
o /5c
(quantitative)
PHOTO-REARRANGEMENTS
Photochemical intramolecular rearrangements provide the best synthetic route to some molecular species. Two such evamples are given below.
|^VN02 I^JLCHO
h-\X
^N-1
NO
NX^COCH (quantitative)
(reference 4)
CoHx
h~v"
.^
+ 8 other isomers {reference ?)
Ergosterol
Vitamin D2
In a number of cases, g-lycidic ketones have been observed to isomerize under the influence of light to (3-diketones. 3
0
G^fi
h~V
^
0 0 /)_C-CH2C-j6
PHOTO- REDUCTION snd-OXIDATION
The classic example of photo-reduction is the formation of benzooinacol from benzoPhenone in the Pre sense of IsonroPanol in nearly theoretical yield. Ot^er readily oxidizable compounds may take the role of isooropyl alcohol, e.g.5
R
H
\
ROOC
CH
tA\
.COOR CH*
0
OH
w
I
H
h"v< - .^\~COOR
^3-C-
CH I
•O-fia
Pyrrole is changed completely by the action of sunlight into a variety of products, among which succinimide is formed in low yield.4
n
H
h~\^
air
HO
OH
H
CH:
o=a
■CH3
!
6=o
/
Anthraquinone3 in xylene solution can be quantitatively oxidized to diphenlc acid. Although 3-methoxy-4-hydroxyben'7aldehyde remains unchanged when irradiated with red light, blue lierht is instru- mental in causing a curious dehydrogenation. The Product is 2,2*- dif ormyl-4,4,-dimethoxy-5,5-d.ihydroyydiPhenyl. 8
OH OH
CH,0
OCH,
CHO
CHO
-.4-
CONCERTED PHOTOCHEMICAL REACTIONS
In recent times, there has been some relatively intensive study of Photolysis reactions in the gaseous phase using known wavelengths and" intensities of light at specified temperatures and pressures. Using mass-spectrograPhic and chemical methods, Quantitative isolation and identification of virtually all the products arising from some photolysis reactions have been accorrro- lisned. Careful analysis of the data has shown that some products are the result of intramolecular concerted decompositions which involve no free radical intermediates. An example of this is found in the photochemistry of propionaldehyde .x At ,3130 A, nearly all the molecules that decompose, dissociate into ethyl and formyl radicals; however, the absorption of more energetic quanta at shorter wavelengths favors an intramolecular dissociation which yields ethane and carbon monoxide directly.
1) CH3CH2CH0 * CHaCHs« + -CHO
2) CH3CH2CH0 ^ CH3CH3 + CO- Addition of iodine to the reaction mixture results in the inter- ception of the radicals formed in reaction (l) to give ethyl iodide as the Product rather than ethane. Reaction (2) is unaffected by iodine.
Even though this field is just beginning to be explored with quantitative methods, it holds important implications for synthetic organic chemistry. By varying the wavelength, it may be possible to change the products of a photochemical reaction.
BIBLIOGRAPHY
1. Blacet and Pitts, J. Am. C*ern.__Soe., 2±, 3382 (195?) .
2. Benrath and Mever, Ber., 45, 2707 (1912) .
3. Bedforss, Ber., 51, 214 fl918) .
4. Ciamician and Silber, ComPt. Rend., [5], 10, I, 228.
5. Ciamician and Silber, Ber.", 44, 1558 (l91l).
6. Ciamician and Silber, Ber., 45, 1842 (l912)#
7. Fieser and Fieser, Natural Products Related to Ftienantbrene, p. 168, TReinhold Publishing Uo., New York, w. Y., 1949.
8. Houben, "Die Methoden <Ser oSc^snlSnfien themie^" Vol. II, p. 1221- 1326, G-. rhieme, Leipzig, u-ermany, ±92b.
9. Moore and Waters, J. Chem. Soo., 1953, 238.
10. Organic Syntheses. Collective Vol. II, p. 71, John Wiley and Sons, Inc., New York, N. Y. , 1943.
11. Schonberg and Awad, J. Chem. Son., 1945, 197.
12. Schonberg, Aw»d, Latif and Moubasber, "J. Chem. Soc. 1950, 374.
13. Schonberg and Moubasb.er, J. Chem. Spc. , 1939, 1430.
14. Steacie, Atomic and Free Radical Reactions J Reinhold Publishing Corp., New York, N. Y., 1946.
15. Stoerraer and Ladewig, Ber. . 47. 1803 (1914) .
CONDENSATIONS INVOLVING- ESTERS Reported "by Leroy Whltaker May 8, 1953
There pre four Possible positions of reaction when a "base attacks a carboxylic ester.1 The carbonyl carbon and the ct- hydroo:en in the aeyl portion of the ester generally enter into reaction with bases, but the a-carbon and the f-hydroeren in the alkoxy portion are also capable of reacting. These four possible positions of attack are marked by asterisks in the following general formula.
0
*i «* '*' *
H-C-C-O-C-C-H l I I
Reactions of the type involving attack at the a-carbon of the alkoxy portion are illustrated by the reaction of methyl benzoate and sodium met^oxide to form dimethyl ether. s Reaction of the £- hydrogen is exhibited by the r6ar*tion of p-phenyl-ethyl mesitoate with potassim amide to yield mesitoic acid and styrene.1 The other two Possible positions of attack will be discussed in more detail.
In the presence of sodium amide there are two types of re- actions exhibited by carboxylic esters.3 One type involves re- action at tbe carbonyl carbon to form the corresponding amide, wUi while the other consists in the removal of the a-hydrogen to form the ester anion. This latter type reaction is involved in the formation of half malonic acid esters (equation l).
R-OHs-COgR* NpN;H3 \ Na /RCH-C03R )" + NH3
CO — ^ R-
C03H C03CH3
(1)
— } R-CH-C03R — ° g s) R-CR-COR
Either isoorooyl magnesium bromide or sodium amide in ether will bring about the self-condensation of t-butyl acetate to give 40-50^ yields of t-butyl acetoacetate . 4 Mixed condensations of esters with esters to form p-keto esters have also been observed.6 These condensations have generally been done using sodium ethoxlde or sodium or notassim triphenylmethide , but Hauser, et al., have used sodium amide for this type condensation.5 These condensations proceed through the ester anion.
Esters will also react with active methylene compounds in the presence of a base. PhenoxyacetylacetoPhenone can be prepared by the action of ethyl ohenoxyacetate on acetonhenone in the presence of sodium ethoxlde.6 o-Xylyiene cyanide reacts with ethyl oxalate in the presence of sodium ethoxide to give a quantitative yield of l,4-dicyano-2, 3-d ihydroxynaoht^alene . * Early work on the condensa- tion of esters with ketones showed that in the presence of sodium ethoxide the products obtained were P-diketones. s In all t^e
-2-
above reactions the sodio derivative of the motive methylene com- pound is first formed, and this reacts with the carbonyl carbon of the ester. The reaction is illustrated by eauation 2.
+ - R'COOR" /
CH3COR + NaNH2 > Na (CK3COR) > R» COCH3COR + NaOR" (2/
The influence of the structure of the ester on the two courses of reaction may be summarized by the following generalizations.3 (a) Substitution of an a- hydrogen of an ester by a phenyl group favors the formation of the ester anion, whereas the substitution of an a- hydrogen by an alkyl grouos favors the formation of the amide, the effect being especially marked by the introduction of a second alkyl group, (b) Substitution of alkyl groups in the alkoxy portion of sn ester favors the formation of the ester anion, es- pecially if the alkoxy Portion becomes t-butoxy.
Hamell and Levine8 found that the Pize and basic strength of the base used have a great effect upon the Position of attack. They used three different lithium bases with ethyl isobutyrate and found that the ^esker »nd less comolex the base, the more likely it is that the attack will come at the carbonyl carbon.
In 1951 Hpuser and Puterbaugh9 simulated the Reformat sky type of reaction using t-buty lace t ate instead of an cc-halo ester. The acetate was first converted to its sodio derivative by the use of sodium amide; zinc chloride ™as added at -70°, and then acetophencns was added. A ?1# yield of t-butyl p-hydroxy-p-phenylbutyrate was obtained. The reaction is shown by equation 3.
1. NaNH3 C6H5-C0-CH3 CH3C00C(CH,)3 ^ ClZnCH3C00C(CH3)3 _
2. ZnCla
OH
* (3)
CsH5-C-CH3C00C (CH3) 3 CH3
They then discovered that the same reaction could be brought about by using lithium amide without zinc chloride. The yields were higher by this method. The use of sodium amide alone failed.
On further investigation of the use of lithium amide to pre- pare p-hydroxy esters, Hauser and Puterbaugh10 found that in order to minimize self-condensation of the ester and to ensure preferent- ial metalation of the a- hydrogen, the t-butyl ester should be used. These Possible side reactions and the main reaction in this method of preparation of P-hydroxy esters are illustrated by the following equations.
-3-
CH3COOC(CH3)a + LiNH:
CHs0ONHg ) Ll'.CK3C00C(CH3)3
(4)
LiCHsCOOC(CH3)
CH3COOC(CH3)
L
0eHB0OCHs
-> CH,COCHPCOOC(CH3)
3/3
(5)
-? C6H5C(CF3)CH3C00C(CH3)3 OLi
C6H5C(CH3)CHsCOOC(CHa)3 I OLi
HOH
-^ C6H5C(CH3)CH2C00C(CH3)3 (6) i
ON
In general the yields of the p- hydroxy ester? obtained by this method are comparable to those obtained In the Reformatsky reaction, and this method seems to be more convenient. One serious dis- advantage of this reaction is that it is entirely satisfactory only with t-butyl esters.
REFERENCES
1. C. R. Hauser, J. C. Shivers, and P. 3. Shell, J. Am. Chem. Soc, 67, 40Q (1945).
2. A. Magnanl and S. M. McElvain, ibid., 60, 813 (1938).
3. C R. Hauser, R. Levine, and R. F. Kibler, ibid. , 68, 26 (1946).
4. J. C. Shivers, B. E. Hudson, Jr., and C. R. Hauser, Ibid. t 65, 2051 (1943).
5. J. 0. Shivers, M. L. Dillon, and C. R. Hauser, Ibid.. 69, 119 (1947).
6. R. von Wslther, J. Prakt. Chem., 83, 171-82 (l91l) .
7. W. Wislicenus and W. Silbersteln, Ber., 43, 1837 (1910) .
8. M. Hamell ana R. Levine, J. Org. Chem., 15, 16? (1950).
9. C. R. Hpuser and W. H. Pauterbaugh, J. Am. Chem. Soc, 73, 2972 (1951) .
10. C. R. Hauser and W..H. Puterbaugh, ibid., 25, 1068 (195?).
THE LEDERER-2-1ANAS3E REACTION Reported by P. Tfiegert Hay 8, 1953
Methylolphenols were first generally prepared by reduction of the corresponding aldehydes, acids, and amides with sodium amalgam However, the yields and availability of the starting materials mad a direct synthesis of the alcohols desirable.
The course to be followed in such a synthesis was first indie ed by Green"1" in lggO when he announced the preparation of o-hy- droxybenzyl alcohol by the condensation of phenol and methylene chloride in the presence of sodium hydroxide.
OH OH
Various workers immediately tried to replace the methylene chloride with formaldehyde, but since acid.ic media were employed^: ly resins and. dirxhenylme thane derivatives were obtained.. Baeyer had already noticed t'-is in 1372 when he reported that phenols re- act Tritb aldehydes in the presence of acid, although the products rrere not identified. Lederer3 nnd Manasse , working independently, then tried alkaline systems and accomplished the condensation of various phenols with formaldehydie. Under alkaline conditions the condensation of the alcohols is slower than is their rate of form- ation.
CH3 CH3
\> _ /^SCK2OH
I 4- CH20 B*se , [J J
H0\ Af HOV/^
CH(CH3)3 CH(CH3)3
Lederer used weakly alkaline systems and effected completion of reaction "ov heating. Resin formation was minimized by keeping the heating period as short as possible. Manasse, on the other hand, preferred strong alkali and permitted, the reaction to proceed at room tempers ture until the odor of phenol had disappeared. Several days were usually reciuired. In spite of this difference in techniaue the claims of the two workers very largely coincided, as to results, and their methods were combined in a single Patent. However, the procedure cf Manasse has come to be more or less stan- dard.
The original papers ^ere almost devoid of experimental detail; indeed, the patent literature was about the only source of such in- formation. Detailed, experimental methods Trere then devised by Auwers^ and associates. These workers also established definitely the identity of the products, most of these experiments being carried out with xylenols. Finally, in 1907 Auwers published a general review of the subject. The original statements of Lederer
(
-2-
and Manasse, that the condensation occurs exclusively at the positions ortho and para to the hydroxy 1 group, were confirmed. Many bases effect the condensation, such as sodium and ootassium hydroxide, calcium carbonate, zinc oxide, lead oxide, and sodium acetate .
Strong alkali was found to favor formation of the oara isomer in many instances. o-Xylenol, sodium hydroxide and f ormalclehyde gave an almost quanitative yield of the corresponding alcohol:
CH,
/*■
i
CH3 «/*\0H30H
HO V
CH.
HO^
However, the above statement about the oreoonderance of the para isomer may have been influenced by the fact that Its solubility is less than that of the ortho isomer and is thus easier to isolate pure.
Formation of diohenylmethane derivatives accompanies many of these reaction, but because of their lower solubility in most solvents they are quite easy to remove. Formation of this oroduct is favored by increasing the strength of the base. For example, as. m-xylenol when condensed with formaldehyde in the presence of calcium carbonate gives the alcohol (I) in good yield but if sodium hydroxide is used (no matter hoy dilute) the main oroduct is the diphenylme thane derivative, 3, 3, 5, 5-tetramethyl-2, 2-dlhydroxy- diphenylme thane (il).
CH3 CH3 CH3
CH3l JCH3OH
OH I
CH
3~
OH
CH.
II
Base strength is not the only governing factor, however, for some ohenols, e.g., p-naohthol or vie, m-xyienol e;ive only the diDhenyl- methane derivatives no matter what base is used.
OH
CH,
/V.CH
CH,0
V
-3-
The presence of halogens or nitro groups may cause the reactio to fail completely. For example, it had been hoped to Prepare o-homosalierenin (ill) by reduction of the alcohol obtained by con- densing formaldehyde with 2-methyl~4-bromophenol (IV) .
Br
CH,
IV
OH
The first step of the process failed completely, however. Examples are known nontheless, in which bromine-containing phenols react almost quantitatively.
CH^OH
Br j/^Br
CH3L y\ GH3
OH
CH3OH Br.-^^SBr CH3|JcH3
OH
Sometimes the m.ethylolohenol Produced undergoes condensation more easily than does the original phenol.12
OH
+ CH-0
Equimolecular Proportions
OH
HOCHg/^CH. 1/2 I
Cl
,0H
+ 1/2
o Most of the methylolphenols melt In the range 75-115 and on
treatment with ferric chloride solution give the characteristic
test. The tendency is toward a blue coloration.
Early workers had noted the formation of 'polyalcohols by re- peated condensations, but in 19?^ Granerer7 showed that such behavior was by no means as unusual as >->ad been supposed. He demonstrated poly alcohol formation with Phenol ^nd o-oresol and indeed, showed that such behavior wag to be expected with any Phenol which has more than one ortho or para position open. Thus, for phenol the number of Possible isomers is five (two mono-, two di-, and one tri-functional alcohol). Using the method sketched by Manasse, Granger found that when phenol was treated with an equimolecular quantity of formaldehyde, all t^e aldehyde underwent condensation with but two- thirds of the Phenol. He was unable to isolate the higher derivatives, however.
An attenrot was made by Sprenfflins1 and Freeman8 to determine which methylol derivatives form when Phenol is treated with a small excess of formaldehyde (ratio 1:1.4). The method employed was methylation of the reaction mixture followed by oxidation. Results
-4-
were as follows:
OH OH
/^> f^NCHsOH
5-l0#
OH
OH
OH
OH
^^^S //^0HaOHHXJV/^|CEaCH H0C%^\,CH30f
10-15^
V
CHsOH CH3OH
/,
Of
CHgOH
4-8#
The first practical method for the separation of mixtures of polymethylolPhenols wag developed by Martin.9 The Phenol alcohols were separated as their trimethylsilyl derivatives by fractional distillation, the derivatives being obtained by treatment of the Phenol-aldehyde reaction products with trimethylch1 orosilane in the oresence of pyridine.
ROH + (CH3)3 Si 01
Pyridine
R0-Si(CH3)
3/ 3
[HOI]
The trimethylsilyl derivatives are then hydrolyzed to the phenol alcohols. Martin lists the following- properties of the trimethyl- silyl derivative a which makes them especially well suited to the separation at hand. They are easily prepared, are thermally stable, resistant to air oxidation and are easily hydrolyzed under neutral conditions. This last is especially important as a number of the polymethylolohenols are very sensitive to acids and bases.
By this method Martin was able to prepare and separate p-methy- lolphenol, 2,4-dimethylolohenol and 2,4, 6-trlmethylolohenol from the reaction of two moles of formaldehyde and one of Phenol.
Perhaps the best procedure presently available for the prepara- tion of methylolphenols is given by Ruderman.10 He emphasizes that experimental conditions are often very critical due to the fact that the reaction may proceed beyond the desired hydroxy! stage to produce condensed products such as dihydroxydiohenylmethane and higher polymers. It is also pointed out that when an oil is ob- tained upon acidification of the reaction mixture, the best pro- cedure is to subject the oil to intense refrigeration to induce crystallization .in situ rather than to extract the oil with ether and attempt to crystallize the ether extract.
The synthesis of some PolynethylolPhenols by other means has been completed recently.11'13 The method used was the reduction of the corresponding esters with lithium aluminum hydride. It was by this process that 2,4,6-trimethylolohenol was first synthesized by Carpenter and Hunter.11
A number of ring-methylated phenols have been synthesized by converting methylolphenols to the corresponding halomethylphenols, followed by hydrogenoly sis of the halo methyl groups.13
-5-
OH
i
OCH
+ 2CH30
OH CflC0H)3 H0CF3f^//VtCH30H l) HBr
2) H3,Pt on 0.
75^
CH.
7
OCH.
Unsymmetrical diary lmethanes may be prepared "by condensation of a Lederer-Manasse methylol product *Tith some other phenol.14
OH
C(CH3)3
OH
Glacial (CH*) rf« f^| CH2
HAc JC
OH
+ H3r
BIBLIOGRAPHY
1. 2. 3. 4. 5.
6. 7. 8.
9. 10. 11.
12. 13.
14.
Greene, Compt. Rend., 90, 40- Am. Chem. J., 2, 19 (lR80) .
A. Baeyer, Ber., 5, 25~Tl872) ; 5, 280 (lR72) .
0. Manas se, ibid . . 27, 240Q (1004).
L. Lederer, J. nrakt. Chem., J50, 22.^ (1894).
K. Augers and Anselmino, Ber., 35, 137 (1902); Augers and van
de Rovaart, Ann., 50%. 105 ('l89"8~7; Augers and Erklentz, ib id . .
302, 115 (lROR); Manasse, Ber., 35, 3R44 ^loO"); Bamberger,
ibid., 36, ^0,^6 (l90s) .
K. Anwers, ibid. . 40, 2524 (190?).
F. S. Grander, In3 . Ens.. Chem., 24 f
442 ^1932). J. Am. Chem.
Soc, 72, 1982
G. R. Snrene-liner and J. H. Freeman, (1950).
R. W. Martin, ibid.. 74, 30^4 ^1952). Ruderman, ibid. , 7_0, 1662 (1948).
Carpenter and Hunter, Plastics, 287 (l°50); J. Apolied Chem., 1, 217 (1951).
J. H. Freeman, J. Am. Chem. Soc, 74, 6257 f 1952) . W. J. Moran, E. C. Schreiber, . E. Enp-el, D. C. Behn, and J. L. Yamins, ibid., 74, 127 (1952). H. E. Faith, .ibid., 72, 837 (1950).
A NEW MECHANISM FOR THE OXIDATION OF GLYCOLS BY LEAD TETRAACETATE Reported by Joanne G-. Arnheim May 15, 1953
In 1931 Criegee1, working with cyolopentadiene oxidation by lead tetraacetate, found that an imourity, cyclopentanediol, was reacting in the following manner:
^-C-OH ^C = 0
I + Pb(OAc)4 K + ^(0Ac)3 + 2H0Ac
-C - OH ^C = 0
i
He found the reaction to be general for all glycols which possess at least two hydroxy 1 groups on two neighboring carbon atoms. The oroducts obtained were aldehydes and ketones, according to the glycol used, by a fission between t^e two hydroxy 1 groups.8
Mechanism Proposed by Criegee
Criegee became interested in this reaction and in 1931 pro- posed that the reaction followed this scheme: » / OF
~C - OH -C ^C = 0
^ OCOCHg
-C - OH ""* >
I OCOCH3 ^C = 0
-<
< OH
which involves an initial attack on the carbon atom attached to the hydroxy 1 e-rouo. This mechanism doesn't explain the fact, how- ever, that only glycols «n& not their ethers or esters undergo this cleavage.
Therefore, after further study on the chemistry and kinetics of the reaction, he oroposed the following mechanism.3
1 -c - 1 |
OH + Pb(OAc) OH |
4 |
A |
\ |
I -C - 0 Pb(OAc) i |
|
1 |
t |
|||||
,c - |
v |
- C - OH |
||||
1 |
J I |
|||||
! -C - 1 |
0 Pb(0Ac)3 |
B |
\ |
> 1 > |
- 0. vPb(0Ac)2 II |
|
~ C - 1 |
OH |
• |
||||
» -C - 1 |
^PbfOAn)3 |
C |
\ |
}c = 0 4- |
||
~c - |
? |
^0 = 0 |
HOAc
HOAc
Pb(OAc)a
Criegee has shown that HOAc retards the rate of the reaction, where- as non-Polar solvents like benzene and nitrobenzene accelerate the reaction. Traces of hydroxylic solvents present in the HOAc are
-9-
found to accelerate the reaction also.* This fact suoports the evidence for stage A- an equilibrium - before the rate determining steo. Kinetic evidence pointed to a second order reaction, corresponding to B ng the slow staere. He ha* also proven the existence of" compounds like I since Pb (OAc) a(0H) (0CH3) wPs obtaine< as the product of the reaction of Pb(OAc)4 with methyl alcohol.
In the decomposition of II the quadrivalent lead becomes di- valent. Criepee has said it may occur in one of three Kays.
u - u ^ - *~~*~' ~^OAc )C -
III IV V
;c .
0 0
)o»
0-
vi wis
0 0©
Such radicals as III are known to be unstable and break between th two carbon atoms thus leading to the ejected products. IV is mor' likely, as it is similar to the intermediate in the periodic acid type of oxidation, involving the shift of six electrons. V is Jus as likely a s HI and would group glycol cleavage with numerous othe: reactions in which a cationic oxygen atom affords the driving fore for a cleavage of a C-C bond.
As further evidence for the existence of the cyclic inter- mediate, Criegee has shown that cls-glycols, esoecially those of five membered rings, which have rigid cis valences, react much morf rapidly than the corresponding: trans comoounds. x > 5 For example, the rate constant, k3, for cis-eyclo^entanediol at 20° is greater than 40,000 while the corresponding constant for the trans com- pound is 19.8.
A surprise behavior of trans-9, 10-decalin was noted by Criegee.5 Ring formation is, for snatial reasons, entirely ex- cluded, yet the diol reacts smoothly and not especially slowly with Pb(0Ac)4. To explain this, he has recently Proposed a different reaction mechanism in this case.1
/C
•l£
■ 0'-
HO-C
HO-cC
H 4/
Ac
oi- Pb^
Pb(OAc)
OAc OAc
OAc
<£>
*s
C = 0
0 = c
Ac H-0 +
r
,0Ac
Pb-OAc . N0Acj
©
+ HO-C.
HOAc
Pb(OAc)
Mechanism Proposed by Waters
Waters, e'7 however, seemed to prefer a free radical reaction. He proposed the following:
-?-
Pb(OAc)4 ' P ^ Pb(QAc)3 + 2 ' OAc
-C - 0-
' + 2H0Ac - C - 0*
I
- C - OH 1 |
+ |
2 |
•OAc |
E s " |
|
-C - OH i |
P |
||||
i -C - 0' I - c - o» i |
— ' — |
F |
-4 |
>C = 0 • C = 0 |
reaction (D) being the slow steo. Kis reasons for such a mechanism are that Fieser and Kharasch have shown (D) to occur and that the acetoxyl radical should be stabilized by resonance.
Bell, Sturrock, and Whitehead8 studied the kinetics of the fission of ethylene glycol and showed it to be first order with respect to both the glycol and the, Pb(0Ac)4. By determining the A in the Arrhenius equation, k=Ae~E/RT, theoretically and experi- mentally, they concluded that the steric factor was aonrovimately unity. This seemed to be evidence for a radical rather than an ionic mechanism.
Evidence Against A Free Radical Mechanism
Kharasch,9'10 in an attempt to Drove whether Pb(0Ac)4 de- composed in the following manner:
1. Pb(0Ac)4 >Pb(0Ac)3 + 2CH3C00- -^ 2CH3. + C03
or
T
2. Pb(0Ac)4 ^ Pb(0Ac)s + (CH3C00)3
compared the products of the reaction of butanediol and of hydro- benzoin with diacetyl peroxide and with Pb(0Ac)4. In the case of the diacetyl peroxide, the products consisted mainly of hydroxy aldehydes and diketone3 instead of the normal glycol fission pro- ducts. He noted, too, that the reaction of Pb(0Ac)4 and diacetyl peroxide with acetic acid lead to different oroducts, to aceto- acetic acid and to succinic acid respectively. He concluded, therefore, that the reactions of Pb(0Ac)4 do not proceed by an initial decomposition into lead diacetate and acetoxyl radicals.
New Mechanism
Recently Cordner and Pausacker,11 after redetermining the order of the reaction and the constants in the Arrhenius equation for ethylene glycol, oropylene glycol, glycerol ct-monochlorohydrin, isobutyl tartrate, nlnacol, oinacol hydrate, and a number of sub- stituted benzopinacols, at various temperatures, proposed the following mechanism:
-4-
-C - OH /
-C - OH
I
+ Pb(OAc)
(1)
HOAc
Pb(OAc)
rapid
_..-\
~C - OPb(OAc)
I ^C - OH
I
Ac OH
I
(2) rapid
N,<
-0-0* (3) -C - 0.
+ I x~^ — '
_C - 0* slow-c - OH 1 I
Pb'OAc)
(4)
Fast
,0=0 ic = 0
The following scheme was proposed for stage (l):
- CH.
iii
Ji
(AC0)3 - Pb
T
\
+ |
0 = |
= C-CI |
|
0 |
0 |
||
! |
! |
||
R |
H |
0 - (Ao0)3 - Pt/-^
i R
The above scheme is supported by the fact that of the substituted benzopinacols used, electron releasing groups (CH3, 0CH3) accelerat ed the reaction, whereas electron attracting groups (Cl) retarded it. Thus, when the electron availability on the ovygen atom was increased, the equilibrium was shifted further to the right. Work on the aril iodosoaoetates,13 which are capable of the same action
as Pb(0Ac)4, has supported this theory.
Ac -
Ar
0
I
1<
0
- 0 -
CH,
ri)
Ac - Ac -
0
0 =
0
I R
•-!?
0
»
R
When electron attracting groups were ore sent in the equilibrium was shown to be farther to the right by the rate constant.
C - CH.
\
0 j
H
aryl group, the an increase in
Stage 2 represents a normal dissociation of quadrivalent compounds, postulated by Kharasch and co-workers. *°
lead
The products resulting from stage 3 and their subsequent de- composition are the same as those that have already been Postulated by Waters for this reaction. This homolytic fission is preferred for several reasons over Criegee's cyclic intermediate:
1. It would mpke this radical action of Pb(0Ac)4 consistent with nearly all of the other reactions of this compound.
2. The ovldation of ohenol* by Pb(OAc)4 can be erDlained by this type of mechanism. Phenolf are orldized to derivatives of
-5-
cyclohexadlenone, that is, o- ore sols yield 2-acetoxy-2-methylhexa-? 5-diene-l-one .
OPb(OAc)3
Me
+ Pb(OAc)4 ^
V
+ HOAo *
\y
Ac0Ae
•^ + Pb(OAe)
Pb(OAo) ~
\?
+ Pb(0Ac)3'
3. The oxidation of trans-9,1 O-decalin does not have to be accomodated by a different mechanism from that of the oxidation of other glycols.
4. Even the orucial fact that ois-glycols are oxidized more readily than the trans compounds can be explained. It is known that glycols with free rotation (that is, cis-glycols) display in- tramolecular hydroeren bonding, whereas trans-glycols would be bond- ed intermolecularly . Thus, whereas only one of the hydrogen atoms of the glycol will participate in hydrogen bonding in cis-glycols, both hydrogen atoms .may participate in trans-glycols thus de- creasing the rate of the reaction.
Therefore, it may be seen that all of the facts can be as satisfactorily explained by this mechanism as by Criegee' s.
BIBLIOGRAPHY
1.
2. 3. 4. 5. 6. 7. 8.
9. 10.
11. 12.
Criegee, R., Organic Chemistry Seminars, Massachusetts Institute
of Technology, September 26, 1951.
Criegee, R., Ber. 64, 1931, 260.
Criegee, R., Kraft L. , and Rank 0., Ann., 507, 1933,
Criegee, R. and Buchner, Ber., 73, 1940, 563.
Criegee, _R., JBuchner, _E., and Walther, W., ^er.} 73,
Waters,
Waters,
Bell, R.
Soc, 1940,
Kharasch, M.
Chem., 14, 949,
Kharasch, M. S.
P.
159.
1940, 571. J. Chem.' Soc'., 1Q39 , 1805 '. "Chemistry of the Free Radical s" Ovford, 1949, 228. , Sturrook, J. G-. R., and Whitehead, R. L., J. Chem. 82.
N. and. Urry, W, H., J. Ore:.
S.
Friedlander, 91. Friedlander, Chem., 16, 1951, 533. Cordner, J. P. and Pausacker,
H. H.
N., and Urry, W.
J. Ore:.
K. H., J. Chem. Soc, 1953. 102.
Pausacker, K. H., J. Chem. Soc, 1953. 107.
2,3-Pyrrolidinediones Reported by Clayton T. Elston
May 15, 1953
In 1887 Doebner1 discovered that cinchonic acids could be synthesized by the condensation of an aromatic amine with a pyruvi acid and an aldehyde. He noticed also that a neutral Product was often formed in the reaction mixture. The neutral product pre- dominated if the reaction was carried out at room temperature while higher temperatures (l00°) favored the formation of the cinchonic acid. Sohiff and Bertini3 Postulated the 2,3 oyrrol- idinedione structure for these compounds and the following reactio* scheme was later proposed by Borsche3. cc-Ketobutyric acids, of
/,
*N
NH,
V
R
CH— CH.
/f\m
NH
V
c=o
COOH
R
^ j
i
V
CH— GH.
I C— — — & It ii
0
0
R-CH
0 0
// >/
CH,C-C-OH
->
,<^\n=CH-R
i
ii
COOH
III
the type I, are known to be formed in such reaction mixtures but their reluctance to undergo rine1 closure casts some doubt uoon their role a q intermediates.4*5 The oyrrolidinedione structure III has been verified by independent synthesis.6
CqH5-CH — CH3OOC— 0H3
•N-06H5
3=0
C00CH3
igHg— CH N—C6H5
i
CH3 C=0
\/
II
o
Infrared absorption soectra of such compounds show p hydroxyl band
at 2.93X/ y indicating that they exist at least partially in the enolic form.
7
Bodforss found that when benzylidenepyruvic acid was treated
with aniline it yielded the anil, c-phenylimino-p-benzylidene-
propionic acid IV, which on heating in acetic acid produced the
-2-
0 //
Ar-CH=CH-C-COOH + Ar-NH3 > Ar-CH=CH-C-GOOH
N-Ar
IV yCH2-C=0 Ar-CH t
I
Ar
correspond ing pyrrol idinediones. However, only pyrrolidinediones were produced in this reaction. Work recently reported by Vaughan and Peters ffives some evidence that o-iminoprooionlc acids are als intermediates in the synthesis involving the anil and the pyruvic acid.6 Thus, the cinohonio acid synthesis and the pyrrol id inedion- synthesis may involve different intermediates.
Johnson and Adams9 were the first to observe the unusual de- carboxylation reaction of 1, F-diaryl-2, o-pyrrolidinediones . They found that the product obtained by the condensation of arsanilic acid, benzaldehyde and pyruvic acid evolved carbon dioxide when heated to its melting point. Although such a reaction was consis- tent with the cinchonic acid structure other reactions of the com- pound, such as the occurrence of aniline in the sodium hydroxide fusion products indicated the pyrrolidine structure. Recently Vaughan Pnd Peters6 have identified t^e decarboxylation products of such compounds as anils of cinnamaldehydes and have also studle the decarboxylation reaction.
The thermal decomposition of the type exhibited by 1,5-diaryl- 2, 3-oyrrolidinediones is not a prener.nl reaction of N-substituted a-ketoamides. Ben^ylidenepyruvanilide (ArCH=CH-CO-00-NH-Ar) T-ras found to be stable under the conditions which led to carbon dic-Tld* evolution in the pyrrolidine compounds. Pyruvanilide shows a similar stability. The reaction appears to depend also on the position of the substituents in the pyrrolidine rins*. Thus,, 1,4- diPhenyl-2, 3-pyrrolidinedione and 1,4, 5-triphenyl-2, .^-pyrrolidine- dione are stable under the reaction conditions. It may be noted that distillation of the latter compound yielded stllbene as well as unidentified products.3
Vaughan and Peters9 prepared eight 1, 5-diaryl-2, 3-pyrrolidine- diones and determined the rate constants of the decomposition re- action at various temperatures. Dilute solutions of the pyrrol- idinedione (0.2 - 0.4#) in o-dichlorobenzene were used in the rate studies, and in all cases the reaction was found to follow first order kinetics. In comparing the rate constants for the sub- stituted pyrrolidinediones they found that an electron- re lea sing substituent (CH30-) in Position Ri or R3 increased the rate of decomposition while an electron-withdrawing group (-N03) had the reverse effect. In general, the effect of a substituent at R3 on the rate is less than the effect of the same substituent at Ri .
-3-
■x.
CH3 „C=0
U
0
The rate constants were found to "be dependent on the solvent em- ployed in the reaction. Quinoline showed a marked accelerating effect on the decomposition of 1, 5-diphenyl-2, ^pyrrol id in ed ione . The first order rate constant varies almost linearly T<rith the amount of quinoline added to the _o-c!3 chlorobenrene solution of the pyrrolidine. The initial concentration of the oyrrolidinedione was also found to affect the rate constants to some extent. This might be expected in view of the acceleration observed with quinoli* and since the product of the decomposition is basic.
The authors Propose that the thermal decomposition of the 1, 5-diaryl-2, .^-oyrrolidinedione Proceed** through the isomeric a- arylimino-P-ben^ylideneDromionic acids.10 Evidence that such an equilibrium does indeed exist is supported by numerous lines of evidence. p-Anisylldene-a-anisyliminoPropionic acid V and 1,5-di- anisyl~2,3-pyrrolidinedione VI undergo decomposition with elimina- tion of carbon dioxide to yield N- (4-methoxycinnamylidene )-4t- anisidine VII. These compounds were selected for study because the former could be prepared and Purified without extensive re- arrangement to the isomeric 2, .^-pyrrol idinedione. A plot of log K vs 1/T (where K= the first order rate constant, T= the absolute temperature) for the decomposition reaction of the two compounds shows complete congruence. Compounds of type V can be converted
\>~ C H= CH- CW= N- <^~ NNOCr
CH-0 /f vN,-CH=CH-CH=N
VII
+ C!02
CHsO — ^ *"^S_CH==C!H-C--COOH
^v-ocu
V
to VI by recrystallization from a larre volume of acetic acid-ethanoL This conversion proceeds so rapidly that Purification of the a- iminopropionic acids by crystallisation is not usually possible.
_4-
The reverse transformation can be effected by warming VI with a small volue of methanol. However, l,5-diansyl-2, 3-pyrrolidinedione was the only pyrrol idlnedione which could be isomerlzed to the cc- Iminoacid upon' warming with alcohol. The structure of the cc-imino- acid V was determined by reduction to cr-anisylamino- X -anisyl- butyric acid IX with hydrogen over platinum. Compound IX was also synthesized from benzylidene methyl pyruvate VIII.
0 0
a //
Ar-CH=CH-C-C-0CH3 VIII
1. 2. 3. 4.
H3,Pt PBr3 H30, H ArNH3
* Ar-GH2OH3CH-COOH
i
NH i
Ar
IX
Ar-CH=CH-C-COOH it
N
i
Ar
V
A freshly Prepared methanolic solution of either V or VI exhibits a changing1 ultraviolet absorption spectrum. The two spectra become identical after several hours. Both solutions show absorption maxima at 324 mM and 230 m<M . The intensity of the 324 m_.M band decreases with time for solutions of V while the same band shows increased intensity with time for solutions of VI. After standing several weeks the solutions show a new absorption band at 269 myU whrj.ch is different from the decarboxylation Product and that of methyl anisylidenepyruvate . The nature of the secondary reaction is not known. At room temperature solutions of VI in di-n- butyl ether show & constant spectrum but at elevated temoeratures the spectrum changes in a manner which indicates that decarboxylation is occurring.
A suspension of V in methanol-dioxane can be titrated with sodium hydroxide to give a normal titration curve. Similar curves were obtained for the titration of VI. The electrical conductivity of freshly prepared methanol solutions of VI increases on standing and gradually reaches an equilibrium v^lue. By assuming that the conductance of the non-ionic species VI is negligible p<nd that the molar conductance of V is equal to the molar conductance of the similar ex~anlsylami no- "T •'■-anisylbutyric acid the equilibrium con- stant for thy reaction VI ^ V was found to he 0', 298 at 25°. The average rate constant for the conversion of VI — » V in methanol at
25° was calculated to be 1,4 v 10~3 mm
-l
The rate constant for
the decomposition of VI in q-dicblorobenzene at 100° was shown to be 1.18 v 10~2 mir.fi. Assuming a doubling of rate for eacr 10° rise in temperature, the conversion of VI — > V would be eighteen times as fast as decarboxylation at the same temperature.
VI
fast
V
Slow
-> CO.
-5-
bibliography
1. 0. Doebner, Ann . , 242, 265 (1887).
2. R. Schlff and C. Bertini, Ber. , 30; 601 (1897).
3. W. Borsche, lbid.f 48, 4072 (lPCgJ.
4. H. T. Bucherer and R. Russisohwlli J. prakt. Chem., 128 t 59 (1930).
5. F. Misani and M. T. Bo$?ert, J. Or*. Chem.,, 10, 458 (1945).
6. W. R. Vau^han and L. R. Peter?, lb Id . , JL8, 382 (1953).
7. S. Bodforss, Ann., 455, 41 (1927TT*
8. J. R. Johnson and R. Adams, J. Am. Chem. Son., 45, 1307 (1923) .
9. W. R. Vauechan and L. R. Peters, J. Org. Chem., 18, 393 (1953/. 10. W. R. Vpu^han and L. R. Peters, Ibid.. 18, 405 Tl953).
PRODUCTS OF o-PHENYLENEDIAI!INS3 AND ALLOXAN IN NEUTRAL SOLUTION Reported by Harold H. Hughart May 15, 1953
It is well known that the hydrochlorides of diorimary and ■primary- secondary jo-ohenylenediamines react with alloxan, forming alloxazines and lsoalloxazines. The condensation of alloxan with free o-phenylenediamines, however, follows a different course, and forms products which have generally been formulated as alloxan anils. The acceptance of this type of structure hinges largely on the work1 of Rudy and Cramer, who allowed alloxan to react with o-dimethylamino-aniiine. The product is typical, and seemingly required the anil structure.
But these compounds fail to undergo the expected acid hydroly- sis, nor are they cyolized in acid media to alloxazines or iso- alloxazines. Further, their visible and u.v. absorption spectra differ markedly from that of the previously described2 5-p_-di- methylamino anil. King and Clark-Lewis, accordingly, undertook the reinvestigation3* 4* B of these compounds.
Primary. Tertiary o-Dlamines3
The product of the reaction of alloxan with o-d imethylamino- aniline was shown by a Herzlg-Meyer determination to contain only one N-methyl group. The other methyl group, as well as the pri- mary amino nitrogen must have reacted with the alloxan, and probably at the 5 position, to give a six-membered ring of the hydroquinoxaline system, I.
cw_
H C — N
0=C
/
N
-J
It
0
/
0=0
s?\/
H
\
*
0^0 h
c/" nc=o
'^C_N' N0 H
fragment
Methylation of this product with d. iazomet^ane replaced two hydrogens and left a compound with one replaceable hydrogen (Zerewitinof f method), thus corresponding to t^e above formula. By 30^ aqueous sodium hydroxide, I is transformed into a compound having a similar u.v. absorption spectra but one ?, H less. That this has the structure II (R=H) has -C-N- been shown by synthesis.
CF„ I
II
-2-
0 H 0 Gl.CHg-C.N.O-OEt
s
<s
K>
CH3. I
N-S03C7H7
x N ^C=0
\
+ 0=0- H
/
M
H
CH,
N.
II (R=H) ^Sn
HC1
I 8 *
0=CkJ
NH
Na
^
Si/
I 3
N-H
HC- OEt
\ c=o
This type of alkaline degradation (I — » II) has been observed previously. The structure of II thus, serves to fix the structure of I, especially since the u.v. absorption spectra support the preservation of the hydroquinoxaline system.
Bv-Product Formation6
Accompanying* the Shiran in the condensation of alloxan with o-dimethylamino aniline is an et^er. This was considered by Rudy and Cramer to contain the ben/ imidazole structure III, since vigorous oxidation with hydrogen Peroxide converts it into 1— methyl benzimidazole .
Ill
CH.
N
II ^a
^V^N
/
0
f C— N
cr Nc=o
SC — n' II
0
H
CH.
N
j CH
T^0:
~7
%A
4
N
However, since the ether can molecular oxygen or alloxan sumably structurally similar have formulated the ether as this alcohol can be obtained by oxidation of the spiran I water. This latter synthesi of pseudo strychnine from st hydrolysis to produce V and
be obtained merely by the action of on the so Iran, I, the two are pre-
On this basis, King and Clark-Lewis
an anhydride of the alcohol IV (R=H);
by boiling the ether with water, or
with molecular oxygen in boiling s is comparable with the Preparation rychnine.8 IV (R=H) reacts on alkaline formic acid, substantiating structure
-3-
CH3 H
I
N
t)H
/£^V'XCHOH0 H H mTH
IV
>N NsOH
/y
CH,
N-H
v^
H
H
C=0 X0H
Compound IV ia resistant to acid hydrolysis; this behavior would not be expected in a structure with a carboxyureide chain in place of the barbiturate ring. However, IV does not form ethers with simple alcohols, This has been attributed to its low solubility in the se re agen t s .
Prims ry and Secondary o- Pi a mines 4
Diprimary and Drimary*- secondary jch-Phenylenediamines react with alloxan to form products showing the characteristic quinoxaljne u.v. absorption maxima and minima.9 These have been demonstrated to have the structure VI.
<^V
R I N-H
V^N-H,
0 >Bs
0=0.
I
y
NH
\
VI (R=H, alkyl, aryl)
If VI (R=H) is methylated by dla-omethane , the product is the O-methyl ether, as shown by a Zeisel determination. This can be
-4-
hydrolyzed "by cold, aqueous sodium hydroxide to 3-methoxy- quinoxkline-2-earboxylIc acid, VII, Identified by comparison with a synthetic soecimen.
N
fc— 0-CH3
C— OH
VII
The constitution of the product Vl(R=CH3) was also determined by acid hydrolysis to the corresponding" quinoxaline-2-carboxylic acid. An identical acid can be synthesized from N-methyl-o_- phenylenediamine and ethyl mesoxalate.
Compounds V (R=H, R=CH3) react with methyl iodide-ootasslum carbonate to form the same trimethyl derivative. Dlazomethane treatment of VI (R=H) replaces only one hydrogen, giving VI, (F^CHa).
o_-Amino-diphenylamine condenses with alloxan to form a typical quinoxaline, VI (R=pbenyl) . It is less stable than the N-alkyl derivatives, however, and slowly deposits o_-ami.no-diohenylamine from cold 1 N alkali.
The 2-alkylamlno-3-aminopyridines form two series of products with alloxan. One is yellow and unstable, readily changing to the second, colourless, stable form. The structures are not known •
BIBLIOGRAPHY
1. Rudy and Cramer, Ber., 21, !934 (1938) .
2. Piloty and Finckh, Ann., 355. 57 (1904).
5. King and Clark- Lewis, J. Chem. 3oc, 5080 (l95l) .
4. King and Clark-Lewis, ibid.. 5579 (l95l) .
5. King and Clark- Lewis, ibid.. 179 (1955).
6. Frerichs and Breustedt, J. Prakt. Chem., 66, S3l (190?) .
7. King and Clark- Lewis, J. Chem. Son., 5077H'l95l).
8. Leuchs, Ber., 70, 1545 (1957) .
9. Kuhn and Bar, Ber., 67, 898 (l°34).
RECENT SYNTHESES OP THIAZOLES AND THIAZOLINES PROM AMINONITRILES Reported by N. E. Bojars May 22, 1953
HISTORICAL. Thiazole (I) and thiazoline (II) compounds can be readily obtained by various methods, which have been worked out largely by Hantzsch.
r N 5 2'! |
(II) |
N
,15 2'
(I)
Several preparative methods are known for the compounds (I) and 01) • A useful synthesis of (I) and some of its derivatives has been worked out by Traumaon,1 Naf,aand Popp; 3 thiourea serves as a reagent.
(1) ClCHa-CEClOC2Hs -:- H20 — > HC1 + C3H5OH + [ClCH2CHOl
(2) [C1CHSCH0] + (NK2)2CS -*Ha0 +
■N
1* Lnh;
N^
■HCl
(3) $ N
II \l
1» lUNKa°KCl + ONOCaH ^S^
2^5
in ethanol
N
Vq/
Another synthesis of (I) employs thioformamide and chloro- acetaldehyde hydrate.4
(4) NH
M
H^
HO
H
+
NSH
VC
Cl'' XH
.N
HpO
!■
+ 2HP0
HCl
N*<
Thioformamide is also useful in the preparation of thiazoline.'
(5)
CH2NH2
I
CH3Br
HN.
.N
y
•CH
Vc>
+ NH4Br
The general reaction (6) has been developed by Hantzsch.
(6) R-CO
i
CH2C1
HN
R
,.N
+
H3'
C—R*
I
R'
6 }7
Of several other methods the one originated by Gabriel can be mentioned. Acylated aminoaldehydes, aminoketones, and amino- acid esters react with phosphorus (V) sulfide to produce thiazolesf
-2-
(7) NH- j
R-CO
-CH2 COR»
^3^5
->
Formamidoalkylmercaptans yield thiazolines.
P S (8) R-CH-CH3NH-CHO Z 5 >
! in
SH
N
in benzene
At
Thiazole and thiazoline themselves also can be made by this general method involving the use of phosphorus (V) sulfide,6*7
The Properties of Thiazoles and Thiazolines, Thiazoles are quite stable compounds.5 They show little tendency to react with nitric acid, and are not affected by the usual reducing agents. They form stable salts with acids,5 which have an acid reaction, while the aqueous solutions of free thiazoles are neutral. The odor of thiazoles is similar to that of pyridine compounds,7 and the two types show similarities of the chemical behavior and of the physical constants.
Thiazolines and thiazoles behave similarly* however, the former are stronger bases.
The ring structure (III) can be found in the nuclei of peni- cillins, Thiazolinium salts, of which vitamin Bi is an example, have the general formula (IV) .
C — N
C £
(in)
R:
N.
HaN.
SO?NH
(V)
"V
Sulfathiazole (V) is among the most useful of the sulfa drugs.
The Reaction of Amlnonitriles with Carbon Oxy sulfide* A. H. Cook, Sir I. Heilbron and coworkers have published recently a series of articles "Studies in the azole series." A part of this series8 discloses a new method of synthesis of thiazoles from a.-amino- nitriles and carbon oxy sulfide.
(9)
R-CH-CN
I
NHS
+ COS
R
H2N ^Ns
N (VI)
The structure proof of the compounds (VI) (R = C6H5 carried out by several reactions,8 among others, by
or C03Et) was the preparation of a benzylidene derivative with benzaldehyde, which confirmed the
-3-
exlstence of the amino group in (VI) , The authors8 also in- vestigated the rearrangement of 5-amino-2-hydroxythlazoles (VI) into thiohydantoins (VII) .
(10)
HaN
Raney Ni
or aaueous alkali
(VI)
Aminoacetonitrile reacts with carbon oxy sulfide to produce in- tractable tars; however, the former can be used in the synthesis of 2-mercapto--5-aminothiazole. 8
(11)
HpNCHpCN + CS.
HftN
While 2-phenyl-2-aminoacetonitrile and 2-carbethoxy-2-aminoaceto- nitrile react according to the equation (9) , it is remarkable that 2-alkyl-2-aminoacetonitriles yield instead iminothlazolines.9
(12) R-CH-CN + COS + CH30Na
i
NH2#HC1
R
in CH30H H j
HN^SX
(VIII)
N + NaCl + CH3OH OH
The compounds of the- general formula (VIII), where R is a methyl, ethyl, n-propyl, or n-hexyl group, are yellow crystals, insoluble in most organic solvents except the amines like pyridine and methylmorpholine.9 They are soluble in dilute alkali and ammonia, and are repreclpitated by acidification. A sodium salt has been isolated in one case. The imine structure (VIII) , in contrast to the amine structure (VI) , is proved by the absence of amine reactions; these alkyl derivatives do not react with benzaldehyde, for instance. On the contrary, they show typical imine reactions.9
BIBLIOGRAPHY
1. V. Trauraann, Ann. 249, 36 (1888).
2. E. Naf, Ann. 265, 110 (1891).
3. G. Popp, Ann. 250, 275 (1889).
4. R. Willstatter and T. Wirth, Ber. 42, 1908 (1909).
5. A. Hatzsch, Ann. 250, 257 (1839).
6. S. Gabriel and M. Bachstez, Ber. 47, 3170 (1914).
7. S. Gabriel, Ber. 49, 1112 (1916),
8. A. H. Cook, Sir I. Heilbron, and G. D. Hunter, J. Chem. Soc . 1949, 1443.
9. J. Parrod and L. Van Huyen, Compt. rend. 236. 933 (1953).
THE MECHANISM OP THE SAND MEYER REACTION Reported by A. B. Galun May 22, 1953
In 1884 Sandmeyer1 tried to obtain phenylacetylene from benzene diazonium chloride and copper (I) acetylide, but found that he obtained chlorbenzene. Later he used copper (I) salts instead of acetylides2 discovering thereby a method for in- troducing halogens into aromatic nuclei!, which proved to be a very useful synthetic tool.
The first kinetic studies of this reaction were carried out by Waentig and Thomas3 in 1913. They reported that the reaction was first order in diazonium ion, was accelerated by an increase in total copper (I) chloride and retarded by hydrogen chloride. They also isolated complexes of the type X»C6H4 •NaClCu3Cl2e
Some ten years ago a radical mechanism was proposed by Waters and an ionic mechanism by Hodgson (an interesting pictorial presentation was given earlier by Hantzsch and Blagden4 . )
Water's radical mechanism5 can be represented as follows:
1. Cu.+ + Ar-N=N ^ Cu++ + Ar- + ^N^N:
Waters postulated that the essential role of the copper (I) ion in the Sandmeyer reaction is its ability to participate in steps involving transfer of a single electron, and that in the gatter- raann modification (metallic copper catalyst) an electron is first donated by metallic copper. The formation of the side products, Ar-Ar and Ar-N=N-Ar, is easily accounted for: 2Ar- ^ Ar-Ar
and Ar. + Ar-N=N ^ .Ar-N=N-Arj ;
JAr-N=N-Ar I + + e £ Ar-N=N-Ar
The main objections to this mechanism are6: 1) the absence of extensive reactions between Ar° and water to yield phenols and hydrocarbon 2) the entire absence of unsymmetrical diary Is of the type ArC6H4Cl 3) the fact that an increase in diazonium ion concentration does not increase the yield of azo compound 4) the mechanism necessitates the assumption that during nitrile synthesis the radical always reacts preferentially with the copper (I) cyanide even in solutions containing an excess of chloride ions.
Hodgson's ionic mechanism7-11 is essentially a nucleophilic displacement:
+ •• (") Ar-N3 XI: Ar ; •• i -
-i
r ? :ci:
ci c\\- > ^ „.cr.:
""Cu \ :ci: ,. Cu' '.' } + ns
»ci-"" \ci; " |:ci-x ^cir:
-2-
The complex [CuCl4]"Is regarded as a halogen carrier. In order to prove that the oxidation of copper (I) ions is not the important stage (as Waters claimed), he showed that several metallic salts In their highest state of oxidation, such as CuCl2 or SnCl4 can also catalyze the reaction9. The formation of biphenyl and azo compounds is explained by Hodgson as in- volving the radical mechanism proposed by Waters,
A kinetic study by Cowdry and Davies6 proved that the reaction is first order with respect to both diazonium ion and dissolved copper (I) chloride; the rate is, however, inversely proportional to the square of the total chloride ion con- centration. , They inferred that the primary reaction is a collision between ArN2 and CuCl'i ions. At higher chloride ion concentration CuCl2 is converted Into unreactive [CuCl4]=, so that CuCl2 +
2Cl^*rCuCl^l*is aoutally the retarding reaction. The following
mechanism was suggested:
a) a slow coordination of zn^ terminal nitrogen atom of ArN2 to the copper in CuClg giving [ArN2CuCl2] b) decomposition of this complex to ArCl or c) further fast addition to it of ArN2 to give [ (ArN2) 2CuCl2]+, which then either d) decomposes to ArCl or e) reacts with CuCl2 to give ArN=NAr.
This mechanism is consistent with the effect of electron withdrawing groups such as N03 which increase the rate of the reaction.
Hodgson's displacement mechanism does not explain the fact that increased chloride ion concentration retards the reaction, and it necessitates a completely separate formulation of the side reactions.
"Recently Pfeil and Velten12 >13 pointed out that the ion [CuCl4]- does not exist in detectable amounts under the con- ditions of the Sandmeyer reaction. Since [CuCl3]~2 exists in solution in appreciable concentrations and since the rate of the reaction is inversely proportional to the square of the chloride ions they assume that copper (I) chloride itself is the catalyst. They further assume that while the Sandmeyer reaction is first order with respect to copper (I) chloride, the two side reactions
2(ArNj) + 2CuCl i ArN=NAr + N2 + 2(CuCl) + and
2(ArN2") + 2CuCl ~— ^ Ar-Ar + 2N2 + 2(CuCl)
are second order with respect to copper (I) chloride. Hence, the observation that the side products become predominant if the concentration of copper (I) ions is increased, is explicable. By increasing the chloride ion concentration the formation of a copper complex is favored, thereby decreasing the copper (I) ion concentration and suppressing the side reactions.
-3-
The authors postulate the following mechanism:
1. rCuCl3]---: CuCl + 2Cl~ (this step controls the con- centration of the catalyst and therefore the rate and yield)
+ +
2. (R-Na) + CuCl — ? [R-N=N CuCl]
3. TH-N=N CuCl]+ a [R. + CuIJCl+~\ + N2
4. [R« + Cu1 Cl+] 1 RC1 + Cu+
5. Cu+ + 3C1 . > [CuCl3l =
The by-products are formed by following bimolecular reactions (which consume catalyst) :
2fR-N2: Cu^'Cll -. .-> R-N=N-R + N3 + ZCuL1Cl+
2[R-N2: Cu' CI] — £ R-R + BN2 + 2Cu Cl+
The authors showed that copper (II) ions cannot act as catalyst, and were reduced in Hodgsons* experiments by free amine (present through a reversal of the diazotation) ** On the other hand, copper (II) ions form a complex with copper (I) ions16 thereby suppressing the side reaction and increasing the yield, though retarding the reaction rate, Increase of chloride ion con- centration increases also the yield, but may retard the reaction to such an extent that heating becomes necessary. Leonards ob- tained in some cases a 100$ yield by working according to these considerations „
BIBLIOGRAPHY
1. T. Sandmeyer, Ber., 17, 1633, 2650 (1884).
2. T. Sandmeyer, Ber., 23, 1880 (1390).
3. P. Waentig and J, Thomas, Ber., 46, 3923 (1913).
4. A, Hantzsch and J. W. Blagden, Ber., 33, 2545 (1900).
5. W. A. Waters, J, Chem, So©,, 1942 . 266,
6. W. A. Cowdrey and D. S. Davies, J. Chem. Socq Suopl., 1949, 48-59,
7. H. H. Hodgson, S. Birtwell and J. Walker, J. Chem. Soc, 1941. 770.
8. H. H. Hodgson, S. Birtwell and J. Walker, J. Chem. Soc, 1942, 376, 720.
9. H. H. Hodgson, S. Birtwell and J. Walker, J. Chem. Soc, 1944, 18,
10. H. H. Hodgson and Sibbald, J. Chem. Soc, 1944, 393.
11. H. H. Hodgson and Sibbald, J. Chem. Soc, 1945. 819.
12. E. Pfeil and 0. Velten ,~ Ann. Chem. Justus Liebigs, 562 , 163 (1949) .
13. E. Pfeil and 0. Velten, Ann. Chem. Justus Liebigs, 565. 183 (1949) .
14. E. Pfeil and 0. Velten, Angew. Chem., 65, 155 (1953). 15 o Leonards, Dissertation Marburg (Germany) (1952')'.
16. Kohlschiitter, Ber., 37, 1170 (1904).
THE ALLEGED RUPE REARRANGEMENT Reported by William P. Samuels May 22, 1953
The product resulting from the Meyer-Schuster rearrangement of acetylenlc carb.iziols containing a free ethynyl would be ex- pected to be an a,5~unsaturated aldehyde, and since these com- pounds are not very easily accessible it appeared that this type of rearrangement would offer a convenient method of synthesis.
In 1926 Rupe and Karabli reported the rearrangement of acetylenic carbinols in the presence of 80^ formic acid to un- saturated aldehydes according to the equation:
OH
R I HC03H R>^
^C -CSC II x > ^C-CH-CIIO
Rf^ P-i
The product obtained from 3-methyl 1- ethynyl 1-cyclohexanol by this method was reported to be 5-methyl-cyclohexylici ene acetalde- hyde in 80.^ yield.
H,C yv °H H3C
'*»/"< —
CHCHO C=CH — >
V
Rupe proposed a raecshanism analagous to that of Meyer and Schuster involving the addition of water and then subsequent loss of water followed by rearrangement to the unsaturated aldehyde :
H *H R n R>v -F 0 R ^ R-v
^C-C=CH— ilsH^ >C-CH=CH0H — --> J>C=C=CH0H — > yC=CHCH0
Ri Ri R'i Ri
The method was then extended to include the rearrangement of a considerable number of tertiary acetylenic carbinols. Some yielded aldehydes and some ketones. The aldehyde forming carbinols included the acetylenic carbinols synthesized from fenchone3, tetrahydrocarvone3 , cyclohexanone2 , methyl isohexyl ketone4, p-phenylethyl methyl ketone5 >6, acetone1, ethyl methyl ketone1, acetophenone4 , and the acetylenic carbinol resulting from a mixture of d-isomenthone and 1-menthone7 . The products were all reported as a, B— unsaturated aldehydes. The acetylenic carbinols synthesized from 4-meth3'lcyclohexanone8, B-phenylethyl methyl ketone5 >e, and 3-me-chylcyclohexanone9 were reported to yield a-B-unsaturated ketones. With the last mentioned compound a mixture of isomers is produced in the ratio of 3:1 respectively:
C0CH3 COCH3
OH I ^
C=CH HCOaH y <\ + /%
KJ ^ CH3A/^ X^CH3
CH
3
-2-
In attempting to rearrange the acetylenic carbinol of methyl- heptenone with formic acid Rupe and Lang10 obtained a tetra- hydropyran derivative:
H3C pH3 C '/ OH
HaC.
f v
C-C=CH CHo
->
^
CH-
L C=CH
CH-
Kilby and Kipping11 have reported a similar rearrangement with the acetylenic carbinol of dimethylheptenone .
The validity of Rule's results were first questioned by
of their projected synthesis lupefs work and found that the
-12
Fischer and LowenbergJ-"s as a of phytol, They reinvestigated products were invariably unsatu
>ated ketones. Other workers who
were unable to obtain aldehydes were Davies, Heilbron, Jones, and
Lowe
1 3
and Dimroth14 .
In view of this Hurd and Christ15 reexamined the reaction in some detail* They found that the product obtained from 1- ethynylcyclohexanol was l-acetylcyclohexane0 Ethynyl phenyl methyl carbine! gave a small amount of acetophenone, and not ,6-phenylcrotonaldehyde as reported by Rupe and Giepler4-. The main product obtained here was a tar probably resulting from the
polymerization of 9sHs '°
CH2 = C - C - CH3. of camphor, ethynylbornyl alcohol (I) camphane (II). The conversion of (I) rearrangements .
CH-
OH
HaC-C-CH,
PC^CH
*^_
The acetylenic carbinol
yielded 2-acetyl-6-hydroxy- to (II) involves two Wagner
COCH3
-^X"'
II
Hurd and McPhee16 found that dimethyl ethvnyl carbinol gave CH2 = C(CH3)-C s CH resulting from dehydration along with a small amount of CH2 = C (CH3)-COCH3, dimethylacrolein as the product 3
Rure and Kamble1 reported
Chanley17 has found aldehydes as minor products in the com- pounds he investigated 0 This is in agreement with the faint aldehyde tests obtained by Rupe and Hurd.
C=CH
-3-
r>coCH:
+
j/*\=CHCHO
50$
H3C CH3 \/ OH
VC=CH
H'aC
j^>
V
H,C
/N.rrCH
0.8% CH3 CKCHO
6.0%
Recently Hennion et al.18 studied the action of formic acid on dialkvl ethvnylcarbinols and found the reaction to be best r eor e s ent ed by ?
R1
R-CKfc-C-C=CH
*
R-CH-C-COCH.
OH
and not unsaturated aldehydes as proposed by Rupe. Aldehydes would be expected if 'che reaction followed the course of the Meyer-Schuster rearrangement which involves an anionotropic migration similar to the allylic rearrangement:
? + r r1 a\
+H ' * tV
R-C-C=CH — - > R-C-C=CH — > R-C=C-CH
i ~H20 /r\
0H 1 *'
Rx R
1
-H
+
J^
+ H20
R-C=C--=CHGH
■>
R-C=CH-CH0
The Rupe reaction is thus an apparent 1,2 shift of the hydroxyl while the Meyer-Schuster is a 1-3 or allylic shift.
In studying (la) Hennion found the product to be (I la) and not "S-butylidene acetaldehyde" as reported by Rupe « They proved that (Ila) was formed by the dehydration of the carbinol (la) to -3-raethyl 3-penten-l-yne (III) and subsequent hydration of the triple bond. This conclusion emerged from the observation that the carbinol (la), the corresponding vinyl acetylene (III), the chloride (IV), and the acetate ester (V) yielded the same product (Ila) upon treatment with hot formic «cid„
OH
CH3-CH2-C-C=CH l
CH-
- 0
CH3~CH=C-C~CH3 I CH3
CH -a— C H — u"~C —OH
CH.
la
Ila
III
CH3~GH2— G— 0 ^^H IV CH3
0C0CH3 0H3~0 H2 — 0— G=C H V CH3
-4-
That hydration of the triple bond did not precede the dehydration was evident from the fact that the acyloin (VI) did not react with hot formic acid.
OCHO
OH I
CH3CH2C-CC)CH3
CH3 VI
CHqCK?C-C=CH
I
CH.
VII
The alternate explanation involving thermal decomposition of the formate ester (VII) was considered untenable since (VII) was not decomposed by heating above its boiling point.
Russian workers19'20'21 attempted to extend the rearrangement to vinylethynyl carbinols but found that treatment with formic acid according to Rupe or acetic acid and sulfuric acid according to • Meyer and Schuster led to dehydration:
OH
^
R-CH2-C-C=C-CH=CH2
R In one case they obtained a dimer: CH3
R-CH=C~C=C-CH=CH2 i
R
2 CH3-C-CSC-CH=CH2 OH
■>
BIBLIOGRAPHY
1.
2.
3.
4.
5.
6.
7.
8.
9. 10. 11. 12. 13. 14. 15, 16. 17. 18. 19.
20. 21.
Rupe and Kambli, Helv. chim. Acta., 9, 672 (1926) Rupe, Messner, and Kambli, ibid. . 11, 449 (1928). RuTDe and Kuenzy, ibid., 14, 708 (1931). Rupe and Giesler, ibid., 11, 656 (1928) Rupe and Herschmann,
ibid
14, 637 (1931) . Rupe and Werdenberg, ibid. . 13, 542 (1935). RuDe and Gassmann, ibid., 17, 283 (1934). Rupe and Kuenzy, ibid., 14, 701 (1931). Rupe, Haecher, Kamble, and Wassieleff, ibid. Rupe and Lang, ibid., 12, 1133 (1929). Kilby and Kipping, J. C. S0, 1939, 435. Fischer and Lowenberg, Ann., 475, 183 (1929) Davies, Heilbron, Jones, and Lowe, J. C. 3., Dimroth, Ber., 71, 1933 (1933).
16, 685 (1933)
1935, 586.
Hurd and Chris Hurd and Chan ley, Hennion, Nazarov,
113 (1937) '(194 9) .
J. A. C. S., 59, McPhee, ibid., 71, 399 ibid., 70, 244 (1943). Davis, and Maloney , J. A. C. S., Nngibina, and Zaretskayce, Bull. S., Classe sci. chim,, 1940, 447;
71, 2813 (1949) acad. sci . ,
35 , 5092 (1941%
U. R. S.
Nazarov and Slizarova, ibid. , 1940, 223; ibid. , 36, 746 (1942) Nazarov and Verkholetova, ibid. , 1941, 556; ibid. , 37, 2343 (1943) .