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) .