Bawl
<Hmhi
HHHK llf
ifffi
'LIBRA RY
OF THE
U N I VERS ITY
Of 1LLI NOIS
G.547
U<6s l94l/42
REFERENCE
Return this book on or before the Latest Date stamped below.
University of Illinois Library
Ll<. 1—1141
Digitized by the Internet Archive
in 2012 with funding from
University of Illinois Urbana-Champaign
http://archive.org/details/organicsemi4142univ
SEMINAR REPORTS
I Semester 1941-42
Pae;e Alkaloids of Crotalaria, Senecio, Heliotropium Trichodesma and Erechtites 1
Roger Adams
Progesterone and Related ConiDOunds 6
R. L. Frank
Koelsch's Work on Cyclic Diketones 12
Stanley Wawzonek
The Rearrangement of Allyl G-roups in Three-Carbon Systems 18
Clay Weaver
Syntheses of Aldehydes through Grignard Reagents 23
R. B. Carlin
Synthetic Estrogenic Compounds 27
C. F. Jelinek
Derivatives of ^%thallyl Chloride 33
R. S. Voris
Pinacol Rearrangements 38
J. 0. Corner
New Syntheses of ^exatrienes and Squalene 44
P. F. ^ar field
Attempts to Determine the Structure of Phthioic Acid 48
R. G. Chase
The Optical Isomers of Cis-9-Methyl-l-Dec alone 52
J. D. Garber
Structures of Pyrethrins I and II 56
D. ff. He in
The Michael Condensation: Some Recent Investigations 60
R. E. Foster
Reaction of Alkyl Benzoates with Sodium Alkoxides 64
G. L. Schertz
The Structure of Lignin 68
B. C. McKusick
Syntheses in the Triphenylene Series 72
P. L. Southwick
iq/| /a. 7 |
|
Sigh Octane Aviation Fuels |
Q. F. Soper |
Page 78
The Mechanism for the Coupling of Diazonium Salts with Aromatic 82 Amines and Phenols
W. E. Blackburn
A Review of the Organic Chemistry of Arsenic 90
C. W. Theobald
Addition Products between Ketenes and Unsaturated Hydrocarbons 99
W. H. Kaplan
Solvolytic Reaction Mechanism 104
0. D. Jones
Aromatization of Aliphatic HydrocarDons 108
S. P. Rowland
o
<r
Thiazoles: Some Syntheses and Reactions 112
G. W. Cannon
High Pressure Hydrogenations over Nickel and Copper Chromite 118
S. M. Himel and R. C, Gunther
The Chemistry of Organoboron Oomp0unds 130
R. M. Roberts
Cardiac Aglycones of the Strophanthidin Group 135
R. F. Phillips
Preparation of Nitriles by the se of Cuprous Cyanide 139
A, V. Mcintosh, Jr.
The ^eduction of Multiple Carbon-Carbon Bonds 142
J. C, ^obinson, Jr.
Polyenes and Cumulenes 145
S. B. Speck
Reactions By Pyridine 149
F. J. ffolf
ALKALOIDS OF CROTALARIA, SENECIO, HELIOTROPIUM TRICHODESMA, AND ERECHTITES
i
Crotalaria is a genus of leguminous plants, many species of which are commonly used in the southern part of the United States as soil-enriching legumes. All contain smaller or larger quantities of alkaloids. At present Crotalaria spectabilis is the most widely planted. Its alkaloid, extracted by alcohol, has the for- mula C16Hs306N. Through its chemical reaction it has teen shown to be closely related to alkaloids extracted from various species of Senecio, Heliotropium, Trichodesma, and Erechtites. In view of the fact that Senecio alone has over 1200 species, the alkaloids of this general type promise to be the largest single class known. Of the 20-30 thus far characterized, not a single one has had its structure clarified. These alkaloids are esters, made up of an acid containing eight or ten carbons (typical formulas are C8H1205, C8H1004, C10H14t04, C10H1606) and a bicyclic nitrogen base con- taining two hydroxyls, at least one of which is esterified with a carboxyl of the acid. All the bases appear to be derivatives of methyl pyrrolizidine.
This report will be confined to the experiments on mono- crotaline, the alkaloid of Crotalaria spectabilis. The two most important reactions which cleave the molecule into parts more suitable for structural study are (l) hydrolysis with alkali; (2) hydrogenolysis with platinum of nickel catalyst and hydrogen.
Monocrot aline
2Ha C16H33p6N AgNaOH
(CTH1i03)C00H + C8H150N (CgHnOjCOOH + C8H1302N
monocrotalic acid retronecanol monocrotic retronecine
acid
I NaOH ^
Monocrotalic acid (optically active) loses carbon dioxide upon treatment with alkali to give monocrotic acid (i). Its properties follow: (l) optically inactive; (2) monomethyl ester with diazome thane; (3) forms a dinitrophenylhydrazone; (4) gives iodoform with iodine and alkali; (5) oxidizes with sodium hypobromite to a mixture of meso and racemic a,a'-dimethylsuccinic acids; (6) on heating loses water to give an unsaturated neutral substance reconverted to the acid by hydrolysis. Monocrotic acid thus appears to be a, ^-dimethyl levulinic acid which was established by synthesis.
JuAtv.
2 -
CH3-CH-COCH3
I CH3-CH-COOCH3
Monocrotic Acid
CK3-CH-COCH3
I CH3-CH-COOH
4r
CK3-C=C-CH3
CK3-CH-CO II
CH3-CH-COOH CK3-CH-COOK
CH3-CH-CH-CH3
1 >
CH3-CH-CO
Konocrotalic acid is obviously closely related to Jimettoyl levulinic acid. Monocrotalic acid was shown to have the x clicking properties? (l) back titration from alkali indicates a earboxyl ilT^llone linkage; (2) monomethyl ester with_ thiazomethane which shows one active hydrogen; (5) heat decomposition gives a, P> Y- "^methylangelicalactone (II ; (4) heat decomposition of methyl monocrotalate gives an unsaturated ester by loss of a molecule of ?r the unsaturated ester can be reduced to a saturated ester which'is hydrolyzed to a stable, crystalline ^tonic acid; the unsaturated ester can be hydrolyzed with acia to the -actone oh tained by heat decomposition of monocrotalic acid. These facte may be explained by any of three formulas which decarboxylase in
OH
OH I CH3-CH-C-CH3
xo /
CH3-C— CO CO OH
III
COOH/OK CK3-C— C^-CHa 0
Cn3-CH-CH
IV
CH,-CH-C-CHaC00H
CK3-CH-CO
-lkpline solution to the tautomeric form of dimethyl levulinic acid and whose esters dehydrate to unsaturated lactone. They are the lactone forms of ^-keto acids.
The observation that monocrotalic acid esterifies with great difficulty with acid and methanol led to the elimination, perhaps uncorrectly, of V. The necessity of high pressure reduction made unlikely the unsaturated ester corresponding to IV. Orxgin^liy III was accepted as the most likely. However, the results of tne study"of compounds VI and VII led to different conclusions.
COOR CH0-CH-COCH3 I
3 l 3 CH3-C-COCH3
CH3-C-C00R I
3 | CH3-CH-COQR
COOR
VI
VII
■* ,~1
f. • - '-V • »
•;;•
Acid hydrolysis of compound VI resulted in formation of di- methyl levulinic acid (I) which indicates this configuration is un- stable to acid while monocrotalic acid is stable to acid. Alkali saponification gave a dibasic acid and no decarboxylation took place. Acid and alkaline hydrolysis of VII also gave dimethyl levulinic acid. Compound V now seems the most likely formula but no proof is yet available.
The basic part of the molecule obtained by saponification is retronecine which has been shown by previous investigators to be a di-hydroxy methyl pyrrolizidine containing one double bond. By hydrogenolysis of retronecanol a oonohydroxy methyl pyrrolizidine results*
Menshikov dehydrated retronecanol to the corresponding un- saturated compound and reduced this product to the saturated base. His proof that this substance, called heliotridane, C8H15N, is 1- methyl pyrrolizidine was as follows: it had the following properties — (l) a tertiary base and no N-alkyl groups, thus it must be a bicyclic molecule with a nitrogen atomj common zo both li.igSs Exhaustive methylation foil* pyrollidine derivative shown by its pyrrol. Subsequent reduction gave l_ dine, a synthesis for which by Menshikov failed. He prepared all of
smooth dehydrogenation to a in optically inactive pyrolli— ' tiled. He prepared
iy
except the desired one and drew his conclusions by elimination. Thus the three possible methyl pyrrolizidines would yield an ex- haustive methylation depending on the point of attack on the molecule the following compounds.
ssible pyrrolidines which could be fori.ed theoretic
or
CH3
•CH2CH2Cri2CH3
VIII
•N" CH3
■CH2CH2CR3
IX
N
VV
-CH-
"1 _ ,
L v— CH3CH(CH3 ) 3
in*
CH,
or
X
V ) — CH2CH2CH3 N
CH,
XI
N
CH,
w
T
CH.,
CH3
CKCH2CH3
XII
or
XIII
_ 4 -
All the pyrrolidines except XIII were synthesized and the solid picrates compared in m.p. with the picrate of the optically inactive product obtained by degradation of heliotridane. None were identical. Menshlkov degraded further the natural pyrrolidine base and obtained an open chain compound which gave a picrate identical with that obtained by a similar reaction with XII.
\/.,jr CH2CH3CH3
W*
* CH3CH3-CH — CH~CH3CH2CH3 fea N(CH3)3
4-
CHr
XIII
CH3
•CH3
■ CrICn3uR3
XII
This reaction establishes the structure except for one point. Con- pound XII can exist in two stereochemical forms and Menshlkov may have isolated the one not identical with the compound from the natural oroduct.
In this laboratory compound XIII was synthesized and shewn to give an identical picrate (m.p. 116°).
Oil 3 C6H5OCH2-CH2-C-CN
COOC^H
CK3
CcHcOCHpCHp— Cn— l»w
3ll-5
CH3 C6H5OCH2CH2-CH-COCH2CK2CH:
HBr
CK3 3rCH3CH2-CK-CH-C3H7
NHCH3
PtO.
GH3NH3 |
^6^5^'-' |
|
Ea |
||
■CH3 |
||
NaOH * |
||
V CHj |
C3H7 j |
CH3 Gg n5OCri2 Cn 3 — C n— C nC 3 rl 7
NHCH3
Use of copper chromite, dry methyl amine and hydrogen gave a dia- stereoiscmeric pyrrolidine picrate (m,p* 126°).
Placement of the hydroxyls and double bond has not yet been completed. On the basis of the exhaustive methylation of retronec= nol, Menshikov claimed he obtained a tertiary alcohol. Accepting this as correct, retronecanol would have the structure
<
H3
\
s/V
,-•
- 5
the
A ■
other
The following experimental facts were used to place hydroxyl and double bond in retronecine. (l) One hydroxy! is readily replaced by hydrogen; (2) after reducing the double bond both hydroxyls are stable to reduction; (3) hydroxyl s in the saturated molecule esterify at different rates; (4; no enol group; (5) saturated dihydroxy compound readily forms an internal ether. The only formula satisfactorily fulfilling these conditions is shown in XIII, assuring retronecanol to be XIV.
HO
CK3 — K^,
N
OH
■w
hC
CHa
N
OH
W
"Tl
VV
— CH3OH
XIII
XIV
XV
Compound XV represents another structure on the assumption of no tertiary hydroxyl. It satisfies all the other facts and also the stability which might not be true of a vinyl amine type, (XIII )
Bibliography
Neal, Rusoff,
Ahmann , J .
Adams and
Rogers,
ibid.
Adams, Rogers and Sprules Adams, Rogers and Long, Adams and Adams and
Long, ibid. . 62 Rogers, ibid.
Am. Chem. Soc. 57, 61, 2815 (1939).
ibid., 61, 2819 (1939), ibid.. 61. 2822 (1939).
2560, (1935).
2289 U940). 63, 228 (1941).
Adams and Rogers, ibid., 63, 537 (1941).
Reported b} Roger Adams September 24, 1941
PROGESTERONE AND RELATED COMPOUNDS
Progesterone is the female sex hormone produced in the corpus luteum, a small yellow body in the ovary. Its function is to condition the uterus for fertilization.
The isolation of pure crystalline progesterone from sows1 ovaries was first announced by Butenandt and his coworkers at the Kaiser-Wilhelm Institute in Berlin in 1934 (only shortly before similar announcements by Slotta, Allen and Wintersteiner, and Hartmann and Wettstein). Its structure was first suggested by Slotta but was confirmed largely through the synthetic work of Butenandt and his collaborators, who synthesized the hormone from pregnanediol and from stigmasterol.
31CK3
sol
Progesterone
yn3
to
Progesterone is the most highly specific of the sex hormones. Very few of the compounds related to it show any of its biological activity. Some of those which do are the 3-enol acetate of progesterone (100$) , testosterone (low), 17-methyltestosterone (low), 17-ethinyltestosterone (good), 6-dehydroprogesterone (b0%) t 6- hydroxyprogesterone acetate, 21~hydroxyprogesterone, 21-methyl- progesterone, and 20-norpregnanolcne,
Cole has very recently announced the preparation of a number of compounds of the type II:
CH3
R = Me, Et, or phenyl
II
These were prepared by the addition of dialkylcadmium or zinc com- pounds to the acid chloride of A5-3-acetoxybisncrcholenic acid (from stigmasterol),. The pure products were inactive, but the crude materials showed progestational activity.
Considerable research has been carried out since 1934 in an effort to find practical syntheses for progesterone from the cheaper sterols such as cholesterol, stigmasterol, the bile acids, and the
- .-
\ \
V
J - . (K-
- 2 -
saponins. At attempt has been made to summarize this synthetic work in the following pages.
Syntheses of Progesterone from Stigmasterol:
HC02H
ozone
AlA/ — ,kXJ m
acetate di
bromide
Stigmasterol
3-Hydroxybis- norcholenic acid
COCH3
cyclohexanone al-isopropcxid
Br2 in AcOH
Pyridinium salt
<A>-1, 2-isomer
1 .1
a
'. .. Ui t.: - , •» -
/\
8
-3~
Syntheses of Progesterone from Cholesterol
V
/KV"
CrO,
Zn
KOH
VIII acetate difcromi
Cholestero
CK2CH3
L-OH
Acetate
t, Jij. . -nr\m NaOH
Pyridine + P0C13 MeQK 1 CHCH3
CH,
^ C-C02Et
X
V\
0s04
4 CKOKGH3 LoH
CH.
Ac,0
/
pyridine
CHOCOCH3 ,4- OH
Zn
4/
COCHgf
cyclohexanone al-isopropoxide
■ /
17-isoprogesterone
HC1
'NaOH
hA/V
/ Dehydroandrosterone chlero- \ e (/KCCH KCN\AcOH H ^1 CN
OH - i_OH
6
x— — j
Kg acetamide
4,
pyridine + 4/ POCI3 CN
Free acid -C02
cyclohexanone al-ieopropoxide
Progesterone
-4-
Spielman and Meyer have developed a method of oxidizing di- bromochole sterol directly to progesterone by means of acid per- manganate, The yield is 0„2$ and the product is not isolated in a pure state, but the concentrates have been found to be satisfactory in biolog ).c al w ox k ,
Convex 'sion p f 3h £ ogenin s to Pregnane Derivatives
/
"N
! - / C^3 A 0
CH--C > CH-CR3 -a-*
X 6 CH2-CH2 200
HO Sarsasapogenin
XI
CK3
CHoOH
Pr egnane di one-3 , 20
/
CH=C»CHSCHSCH
! \
\^°
CH,
Pseudo sarsasapogenin
KOAc
Na
* Pregnane-
EtCK diol-5,20
*«. Tiogenin (differing from sarsasapogenin in having an allo-H at C5) can be similarly transformed.
Marker has recently applied this series of reactions to diosgenin, which has a double bond at C5 6, and obtained progesterone in good yields (See Organic Seminar by EiH. Riddle, April 30, 1941),
Synthesis of Progesterone from Bile Acids
HO
XII
OH
C02H
Barbier- Wieland
NiArV
COpH
Barbier- Wieland
Hyodesoxycholic acid
\
-5-
/o
v^
<
C03H
/•
Barbier- Wieland
COCH3
KHSO4 ^
-K20 (3-acetate )
HO
A
ISO1
Progesterone
Hoehn and Mason have prepared lithocholic acid from desoxy- cholic acid (from bile) in 50 per cent yield by preferential'- _ esterif ication of the 3-OK and oxidation of the hydroxy! at position 12 to a keto group. This was then removed by Wolff- Kishner reduction. Lithocholic acid can be converted to progesterone by known steps.
OH
I
/\jL/X~\AcOaH
/M^V
* *
HO
Desoxycholic Acid
CO^H
Lithocholic Acid
1367
04tu ^-LbKtu;. m.t IX, 331 (1940); X, 303 (1S41). )hal, and Hohlweg, Zeit. Physiol£hem. ,
Bibliography
G-ilaan, Organic Chemistry, p
Ann. Repts., 37, 540 (1940).
Ann. Rev. Block e
Bute.nandt, Westph„.
Slotta, Ruschig, and Pels, Ber. , 67, 1624 (1934).
Allen and Winter steiner, Science, 80, 190 (1934).
Kartmann and Wettstein, Helv. Chim. Acta, 17, 878
Bufcenandt and Schmidt, Ber., 67, 1893, 1901
Butenandt, V/estphal, and Cooler, Ber., 67.
Kahlbaum a~G. Fr. 822, 551 (Jan. 4 1938). rfestphal, Naturwiss., 24, 696 (1936); Ruzicka, Hcfmann,
227, 84 (1934)
1365 (1934 ). (1934). 1611 (193 4)^- Schering-
Eelv. Chim.
388, 1367
318 (1940
Soc. Exp tfamoli, Ber., 71. Ehrhart, Ruschig,
nd Meldahl, Acta, 21, 371 (1938); Wett stein* et al. . 'ibid. . 25.
7 1371 (1940); Ehrenstein and Stevens, J. Org. Chem., 5, 0); Van Keuverswyn, Collins, Williams, and Gardner, Proc. . Biol, Med., 41, 552 (1939).
2701 (1938).
and Aumuller,
Z. Angew. Chem., 52, 365 (1939),
Butenandt and Marnoli, Ber., 67, 1897 (1934); 68, 1847, 1850 (1935).
. _
V
« -. t
«
t--« ■ <
r4s
t ^
1 1
-6-
Marker, Wittle, and Plambeck, J. Am..Chem. Soc, £1, 1332 (1939).
Butenandt, Dannenbaum, Hanisch, and Kudszus, Zeit. Physiol Chem.,
257, 57 (1935). Butenandt, Schmidt-Thome, and Paul, Ber., 72,1112 (1939). Miescher and Kagi, Helv. Chim. Acta, 22, 184 (1959). Goldberg and Aeschbacher, ibid. . 22, 1185 (1939) „ Butenandt and Schmidt-Thome, Ber., 71, 1487 (1938), Spielman and Meyer, J. Am. Chem. Soc, 61, 893 (1939). Marker and Rohrmann, ibid. r 62, 518, 898 (1940). Marker and Krueger, ibid., 62, 3349 (1940), Marker and Krueger, ibid. . 62, 79 (1940). Hoehn and Mason, ibid. , 62, 569 (1940).
Reported by Robert L. Frank October 1, 1941
12
KOELSCH'S WORK ON CYCLIC DIKETONES
Koelsch's work on diketones derived from cyclopentane may be divided into two parts: (X) Enolizability; (2) Reactivity of the carbonyl groups when adjacent to each other B
A survey of the literature indicated that the following generalities could be made concerning the enolization of poly- ketocyclopentanes ;
(1) Cyclopent&nes containing more than one ketone group and having the ability to enolize twice, exist as mono enols.
(2) If the ketone is a derivative of cyclopentene, enolization is hindered.
Under the influence of reagents which produce enolization such as alkali, the above statements do no hold. Treatment with base, depending upon the amount used, forms a yellow or red sodium salt of the compound if only one double bond is present in the ring, or a purple one if a second double bond is introduced. Necessary conditions for the fulfillment of these effects are the presence of the following chromophonic groupings:
0-
One example of this behavior in the literature is oxalyldibenzyl- ketone.
0 0
HO
0
* ».
=0
NaO 0
— 0
NaO 0
ONa 0
0 purple
yellow yellow
Exceptions to these generalities were the iollowing:
0 0
0
\s
,\ — C6H5
\y
H =0
(A)
(B)
(C)
r:
V
St-X V v>
- 2 -
13
l,2-Diketo-3-phenylhydrindene(A) is completely enolic while l,3-diketo~3-phenylhydrindene (B) is predominantly enolic in polar solvents, l,3-Diketo-p_-iodophenylhydrindene (C) gave two crystalline forms, depending upon the solvents used. From acetic acid red violet prisms were obtained, while from non-polar solvents colorless needles were obtained.
The colors of the salts of the compounds in this series were likewise peculiar, Compounds of type (D) are red and dissolve easily in alkali to give deep blue or green salts. l,5-Diketo-2- phenylhydrindene (B) dissolves in aqueous alkali to give a deep red
H ^0
\y
JsrO
R
>v c
c=o
(D)
(E)
i OM
(F)
(G)
sodium salt the structure of which is written as E. von Braun, who worked with compounds of type D primarily on the basis of the difference in color, assigned the ortho-ouinoid structure (F) to these compounds. Koelsch, using 1, 2-diketo-3-phenylhydrindene as a representative of group D, investigated the enolate structure of this compound. He found that the deep blue alkali salt when methylated with (CK3)sS04 gave a red 0-methyl derivative which had an absorption spectra similar to the enol itself and to 2,3-diphenyrindone* The compound added one mole of C6H5MgBr to give G-, since on oxidation O-dibenzoylbenzene was obtained. This behavior indicated that the ortho-ouinoid structure was not present,
.Turning now to the enolization of the diketones themselves, two explanations are possible for their abnormal behavior.
1. The double bond common to both rings can migrate totally or martially into the benzene ring, depending upon the environment.
4
-^s
6
-OH 0
0
2. Since 1, 2-diketohydrindene and 1,3-diketohydrindene are both ke tonic, the grouping H influences the enolization in
such systems.
— '■?«-
. ~<s
-V
f
!
•
, • A ' A •
• . y ■
V ' V
A /\
I
"!
I
A* )»
_,•
-
..—*■
. • -
- 3 -
14
To distinguish between these two possibilities, l,"2-diketo- 3,4,5-triphenylcyclopentene-3, 1, 2-diketo-3,4-diphenylcyclopentene-3 and lr3-diket0"2,4,5~triphenylcyclopentene-4, analogs to the above in which the double bond is fixed, were synthesized and their enolizations compared with corresponding hydrindenes.
Methods of synthesis used were as follows:
0-C=O
I
0-C=O
GH,
ctCr
c=o
0_C=CR \
HI
C=0
/
0C-?
OH H3
>-0
NOH
0
*
0ChsCOOH
— >
\
j
0-C 0-C
CH
' \ 0
X
0
0 0
H
NaOCH.
-0
n\
0CN /
II
0
HNO,
,0
A comparison of reactions is given in Table I (page 4).
Results point to a ketonic character for the cyclopentene analogs and, therefore, indicate that the first explanation was th< correct one. Reduction oroductfc ~in this series are enolic.
In the second part of the work the purpose was to find out which of the groups in the cyclic a-dike tones was more reactive and how this reactivity compared with that of carbonyl groups in an acyclic cc-diketone.
Activation of carbonyl groups is usually characterized by a tendency to add alcohols or water. Acyclic a-dike tones are apparently not active enough to do this but a-ke toaldehydes and triketones (2, 3, 4-triketopentane ) apparently are.
To study this activation cyclic dike tones were chosen which showed no tendency to enolize. A summary of the observations made : given in Sable II (page 5),
■J ..
■ ■
jl
': V
■
<
y;
•
.: .•
• ■
Reagent
0 H
_4_
TABLE I
0-C-C
.0
0-^V
H 0
C=0
Coffipounds
//
/V'\ J
/XH
0 0
0
/,
V
I*
*0
15
color
color in
alkali
Br:
0COC1 in alkali
/C0C1 in pyridine
0MgBr
red
blue-green
in cold CCI4
1 mole addition
orange yellow
white
blue-green red in alco- holic alka- li
Reacts in hot HOac
0-benzoate
2 moles addition
in cold alco- hol but not in ether or CHC13
0-benzoate
1 mole addition
yellow
blue in alco- holic alkali
Reacts in hot
HOAc
C-benzoate
C-benzoate + 0-benzoate
Reduction
-
- .- .
■ -. 1 r ..
Compound /V"\=0
-W-
-U
-5-
TABLE II RCH 1 eq= NHSCH
UnsteMe J car-boj't.yi-2. oximated aeetal
16
1 eq. 0MgBr
0 OCH3
/\X=0
v
^0
carbonyl-2- gives aeetal
carbonyl-2 gives aeetal
no rx
carbonyl-2 gives aeetal
carbonyl-2 oximated
carbonyl-2
oximated
carbonyl-1 reacts
=0 0
H
no rx
no rx
carbonyl-1
oximated
■
• •
»•'' " V
"*
--■ U'i
■
V
. ^.-A
17
- 6 -
From the table one can conclude that in compounds bearing no phenyl group on carbon— 3, carbonyl-2 shows Tore activation than a earbon yl group in acyclic co-diketone. One phenyJ group diminishes this acti ration while two removes it entirel^c
One explanation for the greater activity cf the 2-carbonyl group is that 'Che double bond conjugated with the 1-carbonyl tends to lessen the polarize t ion of this group. Therefore in the rev or £.icle reactions of oxime or hemiacetal formation, any reagent attacking the more exposed. l-oarbon-yl is eventually given up to the L-earbonyl. Phenylmagnesiun bromide in its addition being irreversible adds to the l-carbonyl •
The effect of the phenyl groups is to deactivate the 2-carbonyl through electron donation. If the effect were one of steric hindrance alone only the rate of reaction would be diminished.
Bibliography
Eoelsch, J. Am, Chem. Soc. . 53, 1321 (1936). Keel 3 en' and Hochman, J. Org- ~Chem. , 3. 503 (1938). Koelech and G-eissman_. ibid. , 3, 480 TiS38-V. G-e i s sman and Koe 1 s ch '. i oi d „', 3 , 489 ( 1938 ) « Koelsch and LeClaire"- ~ibTd*. , .6, 516 (1941). Wawzonek, Thesis, University of Minnesota, 1939.
Reported by S. Wawzonek October 1, 1941
18
THE REARRANGEMENT? OF ALLYL GROUPS IN THREE-CaRBON SYSTEMS Cope, et al
while carrying out an investigation of the alkylation of cyanoacetic ester, Cope and Hardy found that when the sodium derivative of ethyl (1-methylpropylidene )-cyanoacetate was treated with aliyl halides the. expected product I was obtained.
CH3 CK3
CH3CH2C=C(CN)COOC2H5 NaQC3H5 , [CH3CH=C-C(CN )COOC2H5]~Na+
CH2=CHCH2X
CH3 CH3 CH=C- C ( CN } COOC 2H5
CH2 CR=l>R2 I
Although the ester I was sufficiently stable to permit purification by means of distillation under reduced pressure, repeated dis- tillations caused isomerization with the formation of a new com- pound having a higher boiling point and a higher index of refrac- tion. A good yield of the isomer could be obtained by refluxing I at atmospheric pressure. That the isomerization involved a shift of the allyl group from the aloha- to the yam ma —position was shown by establishing the structure of the product as ethyl (1,2- dimethyl-4-pentenylidene)-cyanoacetate (ll}»
CH3
pAfj0 I
I > CH3CH-C=C(CN)COOC2Hs
i «
C3H5
II
The structure of II was established by cleavage with aqueous ammonia to unsym-methyl allyl acetone and cyanoacetamide., and verified by synthesis from the same ketone and ethyl cyanoacetate.
CH3
II ^H4OH > CH3CH-C=0 + CH2(CN)CONH2
C3H5
CH3CR C=C + CH2(CN)COOC2H5 -* II
C3K5 CK3
Y
- 2 -
19
Recently the investigation has been extended to include several additional cyanoacetic esters, as well as malonic esters and malononit riles. All three classes of compounds were found to undergo the rearrangement involving a shift of the allyl group . from the a- to the ^-carbon atom with an accompanying shift of the double bond from the p, ^f- to the a, ^-position.
X
i ( i
-C=C-C-Y i
C3H5
x^
X i ; i -C-C=C-Y i
C3H5
X and I = CN or COOCPH
2115
The isomerizations were brought about by refluxing the pure compounds in a partial vacuum, t]
produce the desired boiling temp all the cases investigated the r ment product was higher than tha pletion of the reaction was evid steady increase. The structure either by cleavage with ammonia or by reduction followed" by cond with urea to give a derivative o
;he pressure being regulated to erature of the liquid. Since in efractive index of the rearrange- t of the starting material, com- enced by a constant value after a of the products was established , to ketones, as illustrated for II, ensation of the reduced compound f barbituric acid.
H
R3C-CH=C(COCC3H5) -£->• R3C-CH2CH(COOC3K5)2
urea
<5»H
3r-5
O^jtl
NaOCpH
3ju7
2iA5
0=C-
R3G- 1
CaH
•CH3-CK l 0=G-
-NH
C=0
I
■NH
3ri7 v.
In some instances the structure was verified by means of an independent', synthesis.
Cyanoacetic Ester Series
Six cyanoacetic esters, disubstituted by an alkyl vinyl and an allyl group, were rearranged and the products identified by cleavage with ammonium hydroxide.
C(CN)COOC3H
3A16
CaH
3^5
III
CH3 CH3
1 \
C3H5CH=C~C(CN)CCCC3K5 n-C4H9CK=C-C(CN)COOCsI%
CaH
3aB
IV
C3H
3ix5
CH«
CRH
6ix5
C3H5
i-C3H„CH=C-C ( CN) COCC 3H5 CH3=C-C ( CN ) COOC3K5 CH3CH«C-C ( CN ) CCCC3H5
C3n5 CnHi
f
C3H5 VI
VII
'3n5
VIII
- 3 -
20
Mslononitrile Series
Two repr*eentative£ of this class, l-cyclohexenyl-allylmalono- nitrile (IX) and 1-ethylpropenyl-allylmalononitrile (X) were found to isomerize very readily. The structures of the rearrangement products were established by cleavage with aqueous ammonia
/
■C(CN)a
C3H5
IX
C3H5 CH3CH=C-C(CN)2
C3H5^ X
Malonic Ester Series
"GWO
Of the four disubstituted malonic esters investigated, only ethyl propenylallylmalonate (XI) s.nd ethyl (1-butenyl)- allylmalonate (XII), could be rearranged with the formation of pure products. Side reactions occurring during the rearrangement of the higher molecular weight esters (XIII and XIV) produced mixtures of indefinite boiling points*
CH3CH=CH-C(COOC2H5)
C-aHt
XI
C3H5CH=CH-C ( CCOC2H5 ) a C3H5
XII
( CH3 ) 3CHCH=GH-C ( COCC2H5)
C3H1
XIII
0
-C(C00C2K5)
1 rii
XIV
The structures of the isomers produced by the rearrangement of XI and XII were established by reduction followed by condensation with urea. The barbituric acid derivative thus obtained was com- pared with an authentic sample.
The rates of the rearrangements were studied for the ten com- pounds above which gave clean reactions. The progress of the re- actions was followed by means of the increasing refractive index and rate constants were calculated from the following equations:
k, =
t 2 - tl
1 - xx *i and x2 = fraction rearranged at 2,305 log - times t1 and t3
1 - x
per cent rearranged at time t =
"t - n1 nt - ni
x 100 of starting mat- erial
nt = refractive index of product.
- f ,
■
X
•■
- fc
- 4 ~
Good first order rate constants were obtained in all cases except the malonic ester XII, the rearrangement of which was accompanied by side reactions affecting the refractive index. Correspondence of the rates to first order kinetics was interpreted as strong evidence that the rearrangement is intra- rather than inter- molecular. This view receivea support when it was found that there was no interchange of groups when mixtures were rearranged. For example, rearrangement of a mixture of ethyl isopropenyl crotylmalonate (XV) and ethyl (l-methyl 1-hexenyl )--allylcyano- acetate (V) yielded no crotylcyanoacetate or allylmalonaie derivatives.
QH3
JHs
XV CH2=C-C(COOC2H5)2 + V A ■ > CH2-C=C(C00C2H5) 2 + CH3CH=CHCH3 'v CH3CHCH=CH2
pH3 C4H9CK-C=C(CN)COOC2H5
A great difference in ease of isomerization of the three classes of compounds was found. The most striking difference in rate exists between the malononitriles, which rearrange much faster than the cyanoacetic esters, which in turn isomerize faster than the malohic esters*
In order to determine whether the allyl group undergoes in- version during the rearrangement, the crotyl derivatives XV and
XVI were isomerized*
CH3 CH3
CH3CH=C-C(CN)COOCaHs ■ ■" ■> CH3CH-C=C(CN)COOC2H5
CH3CH=CHCH2 CH3CHCH=CH2
XVI
In both cases, the only product was the one resulting from an in- version of the crotyl group.
The authors conclude that the rearrangement of allyl groups in acyclic or alicyclic three-carbon systems is a general reaction when the a-carbon atom is attached to two nitrile or carbethoxyl groups, or to one of each of these groups. The isomerization is similar in type to the Claisen rearrangement of allyl ethers of phenols and enols in which the allyl group shifts from the electron attracting oxygen atom to a carbon atom.
_C=c-0-C3H5 -+* ► C3H5-C-C=0
In the isomerization involving a three-carbon system, the allyl group becomes detached from the a-carbon atom, electron attracting because of the two negative groups attached to it, and recombines
OC 22
- 5 -
with the 0- carbon atom which is lees electron attracting. It is generally accepted that one motivating force responsible for the Claisen rearrangement is the unequal sharing of the electron pair binding the allyl group to the a-atonic The fact that the nitrile group is more electron attracting than the carbethoxyl group offers a ready explanation for the greater ease with which a malononitrile isomerizes as compared with the cyanoacetic and malonic esters.
Bibliography
Ccpe and Hardy, J. Am.' Chem. Soc, 62, 4<±1 (1940). Cope. Hoyle and Heyl, ibid. . 65, 1843 (1941), Gope^ Hofmann and Hardy, ibid., 65 1852 (1941).. Tarbell, Chem. Rev.", 27," 495 (1940).
Reported by Clay leaver October 8, 1941
23
SYNTHESES OF ALDEHYDES THROUGH GRIGNARD REAGENTS
Formic Esters
The first preparation of aldehydes by means of Grignard re- agents was reported by Gattermann and Maffezzoli, who treated various Grignard reagents with a three-molar excess of ethyl formate at -50°C> Aldehydes were formed in yields varying between ten and fifty per cent.
/)MgX F+ RMgX + HCOOEt -» RCH — » RCHO + Mg(OEt)X
OEt
Cther esters of formic acid, namely methyl and amyl formates, have also been used in this reaction. At the low temperature prevailing throughout, secondary alcohol formation was minimized.
Ethyl Orthoformate
That Grignard reagents could be caused to react with ethyl orthoformate to yield aldehyde acetais was simultaneously recorded by Tschitschibabin and by Bodroux.
RMgX + HC(OEt)3 -* RCH(OEt)2 + Mg(OEt)X
Tschitschibabin added the orthoformate to an ether solution of the Grignard reagent, refiuxed, and then evaporated most of the ether. After most of the ether had been removed, a very vigorous reaction occurred and considerable heat was evolved. The pasty re- action mixture was acidified and the acetal isolated by distillation. The yields ranged between fifteen and eighty per cent*
Tschitschibabin and his coworkers were able to show that large excesses of Grignard reagent and increasingly strenuous conditions resulted in two, or even three, of the ortho ester ethoxyl groups being replaced by the organic portion of the Grignard reagent .
In general, it has been shown, aliphatic and arylaliphatic Grignard reagents give better yields of aldehydes than do the aromatic
compounds.
Formic Acid
Zelinsky found that aldehydes were formed when Grignard re- agents were treated with formic acid.
H + KCOOH + RMgX -> HCR(OMgX)a > RCHO + Mg(OH)X
Houben investigated this reaction and reported that the yields were never greater than thirty per cent. Use of copper and other salts of formic acid in place of the acid itself failed to improve the reaction.
i
CI*
- 2 -
Disubstitut ed Forme.mid.es
~jouveau.it showed that disubstituted formamidee reacted with Grignarc reagents to give products which formed aldehydes on hydrolysis,,
M2 K+
IICONHg + R'MgX -> RCR\ ► R;CHO + R2NH + Mg(OH)X
OMgX
He reported no yields, although he pointed out that a secondary re- action id prominent in some cases.
HCON(CH3)2 + gCBHnMgCi — > HC(CsKi J aN(CHsJ a + MgO + MgCl2
A number of G-rignard reagents were used, and a variety of secondary am?i,es w&.'s- studied^ In addition to the simultaneous reactions in- dicated above, ether side reactions occurred in special cases, and tne reaction was generally undefendable. Yields of aldehyde ranged from twenty to fifty per cent, In all cases twenty to forty per cent of the substituted amide was recovered unchanged.
I soo.vanid.es
Sachs and Loevy reported that benzaldehyde was formed when phenylmagnesium bromide was treated with methyl isocyanide.
C£H5 CPHBMgX + CH3N=C -» CH,K=C( -> CH3N=CHC6HB -> CsHBCHO + CH3NHa
MgX
Gilman and Heckert later investigated this reaction and dis- covered uhat the particular case studied by Sachs and Loevy was the only one of the general type which could be made to occur. Even this reaction produced benzaldehyde in negligibly small yield.
Et ho xy methyl ene An 11 in e
Monier-Williams showed that aldehydes are produced when G-rig- nard reagents are treated with ethoxymethylene aniline.-
RMgX + CsH-EN-CHOEt -* EtOMgX + RCH=NC6HB -> RCHO + C6HBNHa
Ke reported yields of thirty to sixty per cent.
G arbodit hio Acids
Wuyts prepared aldehydes from Grignard reagents by means of a series of reactions, with the carbodithio acid as the predominant
intermediate.
RMsX -— > RCS3H
'O'
ROSSH + K^KCOKKNHg * RCSNHCQNHNH2 + H2S
RCSNHCONHNH2 > RCK=NCONHNH2 + S
RC=NCONHNHa ^ —■ ► RCHO
'
oc
- 3 -
Wuyts and his coworkers found that semicarbazide was superior to other well-known nitrogenous aldehyde-ketone reagents in this re- action. Side reactions are prominent, however, in this preparation. Most troublesome of these is nitrile formation. Aromatic aldehydes are formed in better yield than aliphatic aldehydes, since reactions leading to the latter frequently stop after the second step*
Chi-;, al
Savariau reported yields of aldehydes of about fifty per cent from the reaction of Grignard reagents with chloral
RMgX + C13CCH0 -> C13CCH(0H)R NaQH ■ > RCHCOOH *?°Pa» > RCHO
l OH
Aldehyde Syntheses in which the Group *r.MgX is replaced by a Formyl homologue
Spath prepared substituted acetaldehydes by means of the re- action between Grignard reagents and ethoxyacetal.
EtOCH3CH(OEt)3 RMfiX> EtOMgX + RCHCH2OEt HsS°4 i RCH3CKO
OEt
25
* EtOMgX + EtOH +RCH=CHOEt 1 H23°4
I
It was demonstrated that aldehyde was formed from both intermediates.
A clever preparation of substituted acetaldehydes from Grignard reagents was devised by Herschberg, who coupled Grignard reagents with allyl bromide.
RMgX + BrCH3CHCH3 -> MgXBr + RCH3CH=CK3 — ► Dibromide
4/ RCH3CHO 4 Glycol 4r Diacetate
Either ethoxyacetic ester or its phenoxy analogue was used as a starting material for a Grignard synthesis of disubstituted acetaldehydes by Behal and Sommelet.
ROCH3COOEt + R'MgX -* R0CH3C{0H)R3, -» R3 » C=CHOR -> R3 ' CHCHO EtOCH3CQR + R»MgX ■-* RRf C(0H)CHs0Et >RR'CHCHO
Comparative Studies by Smith and Nichols
Of the Grignard reactions leading to aldehydes in which the -MgX group is replaced by the formyl group, Smith and Nichols selected those involving ethyl orthof ormate, ethoxymetbylene aniline, and carbcdithio acids as the ones showing the greatest promise. Grignard reagents were prepared from a series of bromomethyl- and bromopolymethylbenzenes which ranged from p_- and £-bromotoluene to bromopentamethylbenzene and compared the yields of aldehydes obtained when each Grignard reagent was converted to the corresponding
- 4 - 26
aldehyde by each of the three different methods. Best of the three methods proved to be that involving ethoxymethylene aniline, by means of which the aldehydes were produced in yields of sixty to eighty per cent. The yields obtained in this synthesis were four to seven- teen per cent better than those obtained by means of ethyl ortho- formate, which ranged between forty-five per cent and seventy-five per cent. Only the Grignard reagent from p_-bro mo toluene produced tA respectable yield of aldehyde by means of the carbodithio acid "vnthesis, p_-tolualdehyde being formed in sixty per cent yield. In '-'~.l other cases the reaction was useless a-s a preparative method.
An outgrowth of this work which is worthy of mention is the - ; ., . v- of the formation of bisulfite addition products of these lr-r,'-?ere3. aldehydes. It has long been recognized that benzaldehydes he • -.'.v:g at least one ortho group unsubstituteu formed bisulfite addition products rapidly and in good yield. Smith and Nichols were able to show that aldehydes having both ortho positions occupied by hindering groups but its para position unsubstituted formed bisulfite addition products in excellent yield, although slowly. If, however, both ortho positions and the para position were sub- stituted, only very poor yields of bisulfite product could be obtained, even after a long time.
Bibliography
Smith and Bayliss, J, Org. Chem-.., 6, 437 (1941).. Smith and Nichols, ibid., 6, 489 Tl94l).
All previous references are cited in the above two papers
Reported by R. B. Carlin October 8, 1941
SYNTHETIC ESTROGENIC COMPOUNDS
Dodds, Lawson, et al, Courtauld Institute of Biochemistry, London
Robinson, et al, Oxford University
27
Estrone and estradiol, the female sex hormones secreted by the Graafian follicle, possess the physiological function of preparing the vagina and uterus for fertilization.
These hormones are very expensive because (l ) they do not occur naturally in great quantities, (2) they cannot be prepared cheaply from related natural compounds such as cholesterol, sligmasterol, or equilenin, and (3) great difficulty has been encountered in synthe- sizing them from cheap starting materials.. Consequently, much work has been done in recent years to synthesize more accessible com- pounds which would possess estrogenic activity.
Estradiol
Dodds and Lawson in 1934 started a study to determine the molecular structure essential for a compound to exhibit estroegnic activity. It was found that a cyclo-pentano-phenanthrene structure was not essential when they discovered' that 1-k.eto-l :2:3 :4-tetrahydro- phenanthrene (III) displayed marked estrogenic activity. They syn- thesized this by the method used by R.D. Haworth for its preparation;
X
\x
y\?
CH,
CH.
-C
0
\
/
,4>
0
<0
0
A-i-C 1 ^
nitrobenzene
s0
Chg— CHg— CH3— C02H
8b% H2S04
100'
C-CH2-CH3-C-0H
Clemmensen
III
They next found that a compound need not contain a phenanthrene- type structure in order to be estrogenically active. As an example,
. ;
./,;
- 2 - 28
diphenyl-a-naphthyl~carbinol i\?as moderately active. However, the corresponding £>-naphthyl-carbinol was not effective. Of much greater importance was the fact that while 4:4?-dihydroxybiphenyl was moderately active, 4;4*-dihydroxv stilbene displayed a much greater activity.
On observing these facts, Robinson noted that both hydroxyl groups might well be phenolic, although only one is phenolic in estradiol, and that a stilbene type of compound seemed to have an enhanced activity. Consequently, he synthesized a series of C- alkylated derivatives of 4:4,-dihydroxystilbene by the following method which used anisole as the starting material:
i£°L „ „ /-=\ 8 S /=v OOH Sn-H3-
CK3Q-< ><H — CH3Q-< >C^-<"
KC1
CHaO^ >~CH2-C~<, />-OCH
RI
^3
C3H5ONa
S— -N, ? 9 /^=r*\ RMgBr
CH30-< > C-C-<v ,,VoCH3 —
X , , £ N. V
(90$ yield)
PBr.
\ xA-0CH3
R OH (90$ yield)
CHC1.
^N ? ^ N K°K, C2H50H
ch3o-<; ^>-?"c"< />-och3 : >
X / h >s V 200°
(83$ yield)
R
H0-<^ ^>-C=C-^ ^Xqh
R (85$ yield)
Biological tests showed that a remarkable peak in estrogenic activity was found when R=Ethyl, and that this compound is two to
■•■'■ ■.■-•■
'
.. I
•
•
•.'
- 3 r-
n
three times as active as estrone itself, The reason for this activity probably is that it is closely related stereochemical^ to estradiol;
29
CH
CH,
/%*/
OH
0^
H
P JS
I
.CH2
CH,
CHa
CH8 C=
j
OH
Trans-Die thylst lib estrol "Stilbestrol"
cia-Di ethyls tilbestrol
One difficulty connected with this synthesis is that in addition to the highly active diethylstilbestrol (m.p,, 171°), a less active geometrical isomer (mfp., 141°) is obtained. By stereochemical analogy, it is apparent that the trans-isomer is more closely related to estradiols and,, therefore, that the trans-isomer should be the more active isomer. - Wessely showed that this is true by hydrogenating the two isomers of dime thy Is tilbestrol, the structures of which are known,, Hydrogenation of els-dimethyl stilbestrol gives the me so- compound, while hydrogenation of trans-dimethyls tilbestrol gives a racemic compound 9 Since hydrogenation of the more active diethyl- Sbilbestrol gives a racemic compound^ it must have the trans- configuration0 CI s~di ethyl stilbestrol can be converted completely into trnns-diethyls tilbestrol by exposure to sunlight, so that none of the less active cis-isomer need be obtained*
Another difficulty connected with this synthesis is that in the dehydration of the carbinol intermediate by means of phosphorus tribromide, the dehydration can give an unsaturated side~chain as
we-Lx
suit this that them
as the stilbene compound, so that two racemic mixtures could re- in addition to diethylstilbestrol. Wessely showed that really happens by isolating the two racemic mixtures. He found these could be converted into diethyl stilbestrol by warming in a chloroform solution of iodine.
In the course of their work, Dodds and Robinson hydrogenated stilbestrol and obtained a compound which is even slightly more active than stilbestrol. This compound is called "hexestrol, "
Diethyl stilbestrol
/CH3 OH
CHS f/S/
Hexestrol
IV
.I-. .. • *.',
•
•
•f <■ \
-•
' : :
.
"2 } ■ - • - . . ■
* -i. ' . ' ,' ■"," ■ ■ ■ j .{ ; : * - "i i • '■ o i
;.u:
: •' .' . ;
!••:,/
t
*
'r. v'^
\ '/
- 4-
30
cis-3tilbestrol on hydrogenation in the presence of platinum oxide yields a compound melting at 184°, while the trans-isom-aTon hydrogena- tion in the presence of a palladium catalyst- yields a compound melting at 128°. The estrogenic activity of the lower-melting form is much less than that of the higher-melting form. As shown by Figure IV, the compound analagous stereochemically to estradiol is the meso compound, so it was postulated that the active, higher- melting compound was the meso compound, and that the less active, lower-melting form was a racemic mixture, Wessely has shown this to be true by resolving the lower-melting product by means of its a- bromo-TT - camphor- sulfonat e.
Dodds and Robinson also prepared hexestrol by the following method:
0" H
^\
/
CpH,
Al.Hg
moist ether
^V"
ACpO
Di- acetate
^2
0
OAc
Pd
Hexestrol
btlll another method of preparation of hexestrol is that developed simultaneously by Dodds and by Bret Schneider whereby ethyl feiagnesiun bromide is added, to anisaldazine with simultaneous loss of nitrogen:
.0
2
//
/^
J OH,
C*
/
NH2-NH:
/Vc=N-N=C CH3
■OCH.
p8HRMgBr -Nft
OCH,
KOH
Hexestrol
200
It was possible to isolate both the racemic dimethoxyhexestrol and the meso-dimethoxyhexestrcl, and it was found that the racemic mixture could be converted into the highly active me so-compound by heating in the presence of a palladium- charcoal catalyst.
a.-ril
■■■.,'•■ » ■ ■
.
' i
.-»••
V
. '. ■•
h
'■ '' ■ .... (--•■'
r ;'■;: H£'t-
-5-
31
Since the over-all yield in the latter synthesis employing the aldazine intermediate is low, and since the hydrogenation of stil- bestrol to hexestrol is very difficult, various investigators have been investigating possible new methods of synthesis of hexestrol in order to develop a cheap way of preparing this compound.
Two syntheses which are promising necessitate the preparation of the Intermediate, a-p^-metnoxyphenyla^-bromopane,
H
C— CHp^CH;
The method employed by Bernstein and Wallis is represented below by Method I, while that of Docken and Spielmann is represented by
Method II.
Method I
o-
H
0 II C-C2H5
NaOH
(CH3)2SC4
/
0 CH3
0
C-C2H5 Na
— ■ — >
CpHk0H
dry HBr 0°
Br
i
C-C2H5 H Na
CaH
2xi5
CH-CH
11 ^A
HI
-*• Hexestrol
0CH?
Overall yield, based on ,p_-hydroxypropiophenone, about 7%
Method II
CH=CH-CH.
V
CH3
Ane thole
HBr
V V-
oA
/
H
OHs
CH-C3H5 Br
a-p_~methoxyphenyl-a-bromopropane
Docken and Spielmann obtained hexestrol from the a-jD-methoxyphenyl-a<- bromopropane by coupling the propane derivative with magnesium, followed by demethylation at high temperatures by means of potassium hydroxide. The over-all yield, based on anethole, is 10-15$.
I I
I :
.1
1 ' :
> ■
- 6 - 32
Bibliography:
Cook, Dodds, Hewett, Law son, Proo. Roy. Soc, 114B. 272 (1934).
Haworth, J. Chem. Soc, 1932, 1125.
Dodds and Lawson, Proo. Roy, Soc, 125B. 222 (1938); Nature, 159,
627, (l&r.?Y-; ibid., 159, 1068 (193777 ibid. , 140, 772 (1937). Dodds, Go:.'';crg, Lawson and Robinson, ibid. , 141, 247 (1938); ibid. ,
.V;,2. 34 v-"3); ibid., 142, 211 (193877 Proc, Roy. Soc, 127B.
140' Ur33.}6 Rename and iloblnson, J. Chem. Soc, 1955 , 607. Wease'iy ar.c. ivVilebsr, Naturw, , 28, 780 (1940), jifes36ly, Kerschbaum, Kleedorfer, Prillinger, and Zajic, Monatsh.,
i o , -!.;, . v ■'■-' ±^' J «
Oaiv.pbell, Socles, and Lawson, Proo. Roy. Soc, 128B. 255 (1940). BretsnhneJldsr, dsJonge-Bretschneider, and Ajtai,, Ber.# 74, 571 (1940). flerr.steir: ana \vallis, J. Am. Chem. Soc, 62_, 2871 (194077 Doclr.en and Gpielmann, ibid. t 62 , 2163 (1940).
Reported by C. F. Jelinek October 15, 1941
.(
« % 't
/ J- «. .
• 1 • '
•
» .
...... * §
/ ■
33
DERIVATIVES OF METHALLYL CHLORIDE Shell Development Company, Emeryville, California
As a result of the recent advances In the chlorine substitution of isobutylene _ on a commercial scale methallyl chloride (^-methyl- allyl chloride) has assumed importance as a synthetic intermediatec The presence of a reactive chlorine atom and an olefin linkage makes possible a large number of useful reactions,,
Commercial methallyl chloride containing 4$ of £, p-dimethylvinyl chloride is prepared by allowing a mixture of chlorine and liquid iso- tu'jylene to react for 0,006 second at such a temperature (C-1500 0) thai; a liquid film on the walls will catalyze substitution, and then passing the reaction products into water tc reduce secondary re- actions of HC1 on the olefins present, The impurity of vinylic chloride, which is extremely inert, is ol little importance in re- placement reactions; this is not true for reactions involving the clouble bond.
Metathesis Reactions of Methallyl Chloride
The reactivity of the chlorine atom in this molecule has been measured by means of the reaction velocity with potassium iodide; U.ns a comparison shows that allyl < methallyl <crotyI <M-chloro-2- ine ieh.yl- 2-but ene .
CH3=CHCH3C1 <;'CH2=CCHSC1 <T CH3CH=CHCHSC1 < CH3CH=CCH3C1
CHa CH3
The chloride is readily hydrolyzed by alkali, the product de- pending upon the temperature, alkalinity, and agitation. Inefficient mixing may cause local points of acidity and rearrangement to iso- butyraldehyde; high alkalinity favors formation of dimethallyl tther from the methallyl alcohol and unreacted methallyl chloride. The necessity for a large excess of water and careful pH control indicates that allylic chlorides are capable of reacting in two ways; i.e* by a unimolecular reaction with the solvent (alcohol or water) and by a bimclecular replacement of the chlorine atom by a hydroxyl or alkoxyl group. It is interesting to note that methallyl chloride is structurally incapable of eliminating HC1. Careful control will give the alcohol in 90$ yield.
Methallyl chloride is so reactive that many of its ethers can be made by heating with the alcohol and concentrated aqueous sodium hydroxide* Thus, formation of dime tidily 1 ether in "90$ yield is maintained by the exothermic reaction cf the chloride and alcohol in 50% alkali. The ethers from the more highly dissociated alcohols - methyl, ethyl, isopropyl, and phenol - which furnish a more favorable ratio of alkoxyl to hydroxyl ions give good yields of the correspond- ing ether even in the presence of water. Acid-catalyzed dehydration of the alcohol fails because methallyl alcohol rearranges to iso- butyraldehyde in acid medium. The isobutyl group can be introduced into phenols without formation of tertiary phenols by rearranging methallyl phenyl ether to isobutyl phenol (Ciaisen Rearrangement), followed by hydrogenaticn.
I
' i - ■
\i i.. &
. ;■< : •:••::;;■ • ■■'.'
■. ti •»■■■;'■■•' ;> '' ': . .!.■;■ '.;•; ■ /; I .:•* :■■■/'■ ''■■'. i- •
;■;•:■ ! ...:■' ■ ■ . . !: f .'..-.. . ; .. ■ •: i ■ .. :.. 5 '■• f : ; ■ f ■.: ;
34
Treatment of met allyl chloride with aqueous ammonia in an auto- clave for two minutes yields 56$ primary, 26$ secondary, 8$ tertiary, and 5$ quaternary amines; the high percentage of secondary amine shows that primary methallyl amine is more reactive than ammonia. By including ammonium chloride in the reaction mixture the amount of secondary amine is reduced, since the primary amine forms a hydro- chloride salt as produced*
Methallyl chloride reacts with many metal salts, such as KBr, KI, Na3S, NaSH, NaSCN, to form the bromide, iodide, sulfide, mercaptan, and thiocyanate respectively. Cuprous cyanide must he used in pre- paring the cyanide since the alkali cyanides cause a shift of the double bond. Substituted barbituric acids can be prepared from meth- allyl chloride, a monosubstituted barbituric acid, and caustic or by the chloride with malonic ester and sodium, followed by urea.
Magnesium gives first the Grignard reagent, which immediately couples to form dimethallyl. However, 90$ G-rignard reagent may be obtained by using a large excess of magnesium and ether. The Barbier reaction has been applied to utilize the Grignard as formed - for example, acetone yields 2,4-dimethyl-4-penten-2-ol.
Reactions Involving the Double Bond
o Stirring methallyl chloride with 80$ sulfuric acid at 10 , then decomposing the sulfuric ester with ice yields 65$ of isobutylene chlorohydrin (chloro-tert. butyl alcohol), the remainder of the methallyl chloride being rearranged to £, p-dimethylvinyl chloride. At 40° the sulfuric acid treatment causes 85$ rearrangement; the re- moval of unreacted methallyl chloride from the vinylic chloride is effected by saponifying with KOH.
Concentrated HC1 adds to the double bond| it has been reported that HF catalyzes the addition of methallyl chloride to benzene to form (chloro-tert. butyl) benzene,
CH3=C-CH3C1 + C6H6. ijfffru C6H5-C(CH3)3CH3C1
CHo Chlorination of liquid methallyl chloride or j3, p-dimethylvinyl chloride results in ?0$ of the unsaturated dichlorides; bromine gives 93$ addition* A solution of chlorine in xvater gives chiefly dichloro- tert. butyl alcohol, along with some dichloro-isobutylenes, tri- chlor-alcohols and trichloropropane. A reasonable mechanism postu- lates that chlorine substitutes to form a positive charge on the tertiary carbon atom, to which water or chloride ion may add or from which a proton may be expelled to cause unsaturation at different points in the chain,
Un
CH3=CCH3C1 fil*-* CH3C1-C+-CH3C1 "■2-° > CK3C1-C~CH3C1 + H+ CH3 CH3 ^
C1- +C1- > CH3C1-CC1-CH3C1
CH
3
CH2C1-C-CH3C1 or CH3C1-C=CHCL U H-
CH2 h+ CH-
■•: : : -;. ■.:'.
\
■(•
. y
f .-,••« <«
.... - . ......V -mi
' ..-_..:
» 81*
In comparing isobutylene and methallyl chloride, it is seen that the former is more reactive, as evidenc-ed in polymerization, addition of HC1, and chlorine substitution.
Reactions of Methallyl Alcohol
Methallyl alcohol is rearranged by 12$ sulfuric acid in almost quantitative yield to isobutyraldehyde; any impurity of dimethallyl ether is also hydrolyzed and rearranged.
CH8=CCH8OH -> CH3GCHO
CH3 CH3
This aldehyde can be oxidized to isobutyric acid; the overall conversion from isobutylene to isobutyric acid is 75$.
Refluxing methallyl alcohol and isobutyraldehyde in 2b% sulfuric acid gives isobutyleneglycol-J-spbutyracetal^ which is decomposed to the glycol and aldehyde on heating with dilute acid.
35
9 ^H* . 9H
CH3CHCH-0~(jjCH3 -* CH3CHCHO + CH3CCH3OH
Esterification proceeds by distilling the organic acid with methallyl alcohol. Again, use of mineral acid must be avoided be- cause of rearrangement to the aldehyde* Methallyl alcohol, its esters, and ethers may be hydrogenated below 200° over nickel to the corresponding isobutyl compound.
Methacroiein is prepared either by dehydrogenation or oxidation of methallyl alcohol; the oxidation process is employed because of less rearrangement to isobutyraldehyde. The process consists of passing alcohol and air over silver gauze at 500°. Unless inhibited, methacroiein polymerizes on standing at room temperature to a white granular solid.
Methallyl xanthate, ueed in ore-flotation, can be made from methallyl alcohol, alkali, and carbon disulfide.
6-methvl glycerol and its Derivatives
Methallyl chloride has a series of derivatives similar to those of allyl chloride, hence the "£-methylg3.ycerol" homologs*
CI OH CI , . 0
CH3=CCH3C1 3°£k^ 6H8dCHa J£i2££8 CH3-C-CH3
CH3 CH3 01 6h3
dichlorc-tert .butyl p-methylepichlorohydrin alcohol
N/
H30 ,H3S04
'. .1
C l '.
.!
\'
on j
lo U .-v i ; ,.; '• .
. .1
1 s,
«'■
rt .2 1 u {
: ; . '. ' .•-..•
* '
x u
• -I
£*<*« •-'• i: <■ -n*
l >, T f
Ka0
OH OH OH
chsc — ch2 4-
fy H2S04
p-methylglycerol
OH
0
- 4 m NaOH
CI OH CH3-C— CH3
OH I
OH-
p-methylglycidol
£~methylglycerol mono- chlorohydrin
The advantage of this series of reactions over the direct hydrolysis of di chl or o-tert .butyl alcohol or jis-methylglycerol monochlorohydrin is that the salt need not be removed from the glycerol produced.
Derivatives of Methallyl Chloride
isgbiitY]*-] Isobut^racetalHoSQ Isobutylme aiaehyQ.3 iof iSobutylene Glycol
RQH
NaOH Ks0
Methallyl Ethers
OH-
Methallyl Alcohol
H8S0j Glycol
o'xidi. £ e
Isobutyr- jjwgange with. ikgpj aldehyde
Alcohol NaOH
Metal
Methallyl Chloride
MRthally.l ,
Dimethallyl
Ether
dehyi-r
Salts
JSSi
Methallyl Bromide, Iodide j Sulfide |Thiocyanate, jMercaptan, etc*
dil.
"Sethacrofein
oxidize
RCOMMethallyl I Esters ,
Isobutyric
Acid
^sfl so butyl j S ilAlcohol j
Grignard Metais
Methallyl Amine
Methallyl
Chloride
Dimethallyl Amine
etc*
Methallyl Subst. Hydrocarbons. [Dimethallyl
H0C1
I Dichlor o-tert* (butyl alcohol
pH
distil
£-methyl-
epichloro-
hydrin
Ha0
H2S04
KoSO,
HaS04
NH* j OH-
£-methyl Glycerol Monochlor- hydrin
Rearrange; p-dimethylvinyl
chloride
H20 JQLl
Isobutylene Chlorohydrin
OH
1,3-diamino-
2-methyl-
propen-2-ol
NapH jp-methyl | glycidol — i
ilstil
i
Isobutylene Oxide
H2C H2£04
^-methyl glycerol
Dichloro- Isobutylenes
p-methyl glycerol differs from glycerol in its easy conversion to the unsaturated aldehyde by atmospheric distillation from 12$ sulfuric acid. These same conditions give methacrolein from any member of the above series* Trichloro-tert .butyl alcohol, a minor
•
. ■
u
.
•
•
! — >,L. .
;:
'■
—
■
»
' |
i ' ' ■-' |
,.- ivvr.J |
|
Ji. |
•-f-i-v ■ |
• - '.. |
■ «■..-.... .. |
"~* |
|
<5 : |
" t ,
...
I
jr.
i
-■• -
-> • ■*.- *
- 5 -
by-product from the chlorohydrination, , gives an analagous series containing a chloromethyl group in. place of the methyl.
Compounds Containing Conjugate Unsaturatlon from Methallvl Chloride.,
Methallyl chloride affords a feasible approach to the synthesis of substituted isoprenedicarboxylic acids through ^-methyl glut a conic ester.
CH3-C
CH3
W H0C1
CHa-Cl
CH2C1 I CH3-C-.0H / CH3C1
KCN
CHoCN
CHa-C«-OH
EtOH
I dry HC1
GH*GN k°So4
CH3 I 0-CH=C~C=CH-COOH
COOH
a-benzal-p-methylglutaconic acid
/ uGl
. 0CHO <:
KOH
CH2COOEt i
CH3-C-OH
CH2C00Et
U
distil..
CH-COOEt /I
CH3-C I
CH2COOEt
p-methylglutaconic ester
37
By use of £>-cyclocitral in place of benzaldehyde, this gives a new route to the carat enoids and Vitamin A..
Bibliography
Burgin, Engs, Groll,tand Hearne, Ind. Eng* Chem., 31, 1413 (1939).-
Tamele, Ott, Marple, ' and Kearne/ ibid., 33, 115 tl94l),
Burgin, Hearne, and Rust, ibid..f 33 '. 385"Tl94l)«
Hearne,. Tamele, and Converse, ibid. f 35r, 805 (1941),
Hearne and DeJong, ibid,, 33, 941 (1941 5.
Hurd and Abernathy, J, Am* Chem. Soc, 63, 976 (1941).,
Reported by R. S. Voris October 15, 1941
li, -aoGlq ft; i* :
Li !
•
"i
■ . - - ■ t .
• ; &:•:. < »• » »...-,. ........
I •-• • •
,'.,;;:.•■ i .. • -" - ' . •
• i - • i ' %-U.J. t • ' > -■ ; * ■ ' • ■ \ • ■■> '■ ■
i .
38
PINACOL REARRANGEMENTS
The rearrangements of tri~ and tetrasubstituted ethylene glycols and a, £-amino alcohols are summarized in this report.
I. Plnacol-oinacolone rearrangement
When tetrasubstituted ethylene glycols (pinacols) are de- hydrated, they rearrange to form pinacolones*
R R R3£C-c£-R
;>c~c< _
R 0 0 R
H H
Sulfuric acid of various concentrations, hydrochloric acid, acetyl chloride, and acetic acid and iodine are the most common reagents used to bring about the transformation* The mechanism usually given is the following, using pinacol itself as an example.
CH3 2
CH3 :
CH3 :
CH3 :
C : 0 * • • •
C : 0 • »
H H
hV
H
CH;
CH,
CH3 S p
ch3 :
l3 * H
ch3 ; u
» a
• • 0 |
: H |
+. |
0 9 |
||
• » |
||
0 |
\ H |
|
• * |
CH3 : CH3 :
c^
CH
3 •
C : 0: H
ran
3 Li.
X
CH3 CH3
(CH3)3 >C~C~CH3 + HX + H.O
There are four types of pinacols, which can undergo the types of reaction shown.
xc— c^
R^O 0XR H H
Type I
HO— C-—R VR
I
(CH2)
X
c— -c
0
H H
R
0NR
Type II
(OH.)
n
\R °
/ //
C— C-R or
(CH.)
n
~V
0
R R
i'l'. . [fiA3/; .::••.•.::
■
:
:
.1 1.
I . ■ i •
-
i
■
• .'
\ i_
I
-,.;-*.
•'•..
•</■■
» «
i . :
t- ■
:l ', ■■
r.—~ 'I. ■-■■{- \
t »•
» *
« . ■» ♦
.A-
**-*-*4^« .
- 2
39
r —
(CH3)n
\ /
c— c
/O 0\
Type III
<?H2)m
(CH2)
n
, C
(CHa)
^
0
tn
or
(CH2)
n
(CH3)n
nS f n
//
0
(CHS)
R .C-OH
(CHa)
n
.C-OH R
R
I
C-R
.0=0
or
(CK2)
n
R
I
.C
'1
0=0
I
R
Type IV
The course of the rearrangement depends, first, on which carbon atom loses an hydroxyl group, and, secondly, on which radical mi- grates to stabilize the molecule. In type I, if the radicals were all different, four different products could result.
Ro~ ■— C— 0— R^
#°/R»
R i — C— C~ — R3
R.
\,
Ra^ 0
H
XR3 <
0 RA
Rs — C— C--R3 R$
R2> C£C— 0f R3 NR*
Since the hydroxyl group takes with it a pair of electrons, it will be lost from the carbon atom having the greater capacity for electron release* The second consideration is the migratory ability of the groups on the other carbon atom. The migrating group carries a pair of electrons so that its electron-attracting power, of affinity cap- acity, is important. In the aliphatic series there is an inverse relationship between migratory power and affinity capacity while in the aromatic series the two go hand-in-hand, showing that in one case or the other there is another factor under observation.
The pinacol rearrangement has been used to compare the affinity capacities and migratory power of various groups. Bachmann has studied the rearrangement of symmetrical pinacols of the type ArAr' C ( OH)C ( OH ) Ar Ar ' , The question of which hydroxyl group is lost is eliminated and the product depends solely on the relative mi- gratory tendencies of Ar and Ar' . In all cases both products were
( :.
. .;
i • :■
*'■»' 1.
n-nm
i : - *
40
- 3 -
obtained, but the amounts of each gave an indication of the migratory power of the groups. If the relative migratory power of A and B and that of «A and C were known, it was found that the behavior of B and C together could be predicted. Thus the tendency to migrate was shown to be a property of the group and not of the molecule. The following series gives the relative migratory power of various aryl radicals (phenyl = 1; anisyl, 500; phenetyl, 500; p_-tolyl, 15,7; £-biphenyl, 11.5 ; p_-isopropylphenyl, 9; p_~ethylphenyl, 5; m-tolyl, 2; m-metnoxyphenyl, 1.6; phenyl and p_-iodophenyl, 1; p_-bromophenyl, 7; p_-chlorophenyl, .66; o-methoxyphenyl; .3; m-bromophenyl, o- and &- chlorophenyl, nearly zero. Except for steric factors this series is comparable to the series for affinity capacities of these groups.
Bachmann has also studied the rearrangement of unsymmetrical pinacols of the type ArArC(CH)C(OH)ArfArf . The migrating tendencies found here were greatly different from those found in the symmetrical type. They bore no simple relationship to the other values and could not be predicted. This is because the ease of loss of hydroxyl group rather than the migratory power was the factor under observa- tion. Apparent migratory capacities in this series are as follows: p-biphenyl> phenyl JN m-tolyl^ p_-tolyl\ p_- chlorophenyl J> phenetyl anisyl,)' p_~fluorophenyle
The rearrangement of pinacols of types II, III, and governed not only by the two factors mentioned above, bu size of the ring. Thus, in type II, if n is four, a rin will take place, as is shown by the following example:
C=0
IV is also
; enla
by the
rgement
i.
If n is five, a mixture of the possible products is form of the type represented below rearrange to both fluorene Dhenanthrenes.
and
ed, Pinacols s and
Bachmann investigated pinacols of type IV Only one reaction was found to take place.
.n the phenanthrene series.
■' ■
. , ,■ . ... , . j ■» - 1 • .
■■-■■■ r, , ' ■ '•!'««
■
■
1 . :. J
! \ *
1
- .,
- 4
4J
A
Ar -OH
Ar OH
— -*
Ar 9 anisyl, p_-tolyl,
m-tolyl, p_-chloro- phenyl, p_-fluoro- phenyl, jo-biphenyl, and a-naphthyl
In the allcyclic series, type IV gives a ring contraction in one case.
CH3 yt^ GH3
-OK
CH3 OH
Recently it has been suggested that the pinacol rearrangement in- volves a Walden invej&sion. Bartlett has studied the rearrangement of the els- and trans- forms of 7,8-diphenylacenaphthenediol-7,8.
0
OH
OH
0_C C~0
CUt-t-C — 0
Both geometrical isomers gave the same product, but the kinetics of l-he reaction showed that the rearrangement of the trans- form was much slower, corresponding to conversion to the ci&r "form before re- arrangement. Some e^s~ pinacol was actually isolated from a partially rearranged portion of _t:j..ans'- pinacol. Since the oi.s- form Is the one which can rearrange by ftalden inversion, that procedure seems most likely in the light of these facts.
II
Semipinacol Rearrangement
The dehydration of trieubstituted ethylene glycols can take p. lace in a number of ways.
1.. vinyl dehydration H R« r ft:
R-C— C-R;i i R-C^C-R"
0
H
Q 0 K H
R-i?"
R? 1
C-R"
H
I'..
. ir.
■ ■
. '
'■>■■
' '"• . '■ .
! H
; II
■
■ >
■ !
i ,-■■.' -::
^....." . ., r -
^.•'
..'*
r ...
,/!,
■ • • • . ' •■.•'. i "- •; :lv.r^: .J-
•
-«. ■'* '- - .■--—' —
- 5 -
H R»
i I
r_C— C-R"
6 6
H H
2. Semihydrobenzoin rearrangement
rpNr '
|r| _C— -C-R" 6
42
i
H-C-C-R'
11 ^R»
3. Semipinacol rearrangement
H R' i I
r_C— .C-R"
I I
0 0
H H
-*->
H ri
R-C— C-R"
I I
0
R'-C-C-R" R/ (,
riCc-C-R'
In the first two reactions, the tertiary hydroxyl group is eliminated, but in the last one the secondary hydroxyl group is eliminated. The removal of one or the other is controlled by the nature of the groups attached to the respective carbon atoms as well as by the nature of the reagent. Concentrated acid promotes reaction 3, while milder conditions favor reaction 2, Dehydration of aryl dialkyl glycols, ArCH(OK)C(OH)RR, with oxalic acid or dilute sulfuric acid gives aldehydes (reaction 2) and with concentrated sulfuric acid gives ketones (reaction 3). As the radicals become larger the tertiary hydroxyl group is stabilized and the semipinacol rearrangement takes place regardless of the reagent used.
Alkyl hydrobenzoins C6H5CK(0H)C(0H)RC6K5, give desoxybenzoins by loss of tertiary hydroxyl when R is methyl, isobutyl, or phenyl. When R is ethyl, propyl, isopropyl, butyl, Isoamyl, and cyclohexyl, of desoxybenzoins and benzhydryl alkyl ketones, (CSH5)3CH-C=0 are
R produced. Evidently the size of the radical is not the determining factor in the rearrangement of alkyl hydrobenzoins*
It has been shown that the presence of an aryl group on the carbon atom carrying the secondary hydroxyl group is necessary before a semipinacol rearrangement can occur.
III. Retrooinacol r ear t a n gement
The dehydration of alcohols derived from pinacolones is accompanied by the reverse pinacol rearrangement.
H
R3^C-C-R
i
fi-
RS>C=C <R.
Mixtures result when the radicals are different, Whitmore studied the dehydration of two isomeric pinacolyl alcohols and found that they gave the same three products.
■
.;■ O:,, . :,. |
•
■
■■ ■: .
,
'4 .:
f '
* - ■
. ;
■» -
■ ■ / . • ■ .
'. ' ■ •
x : ' ' •
.' . .
■
:
....
I
! 0 ' .;
"' I ■ , '■ • ' ';..-•
w 6 *•
43
C4H9 H / 9Ha 9H3 C4H9*(k~- C-CH3 ( CaH»-CH=C- C-C4H
9
I
CHa 0 / H
H _ — M / CH3 CH3
CH3 JJ ^ C4H-9— C ■ '■G—C4H9
J I
C4H9O — C— C4H9
ch3 6 I f3 F*H*
jj \ CH3-C=== C--C4H9
IV, Deamination reactions
Pinacol and semipinacol rearrangements may take place if one hydro xyl group of a pinacol or trisubstituted ethylene glycol is re' placed by an amino group. In this case there is no question of the course of the reaction. The amino group is always eliminated and migration takes place away from the carbon atom carrying the hydroxyl group.
1. Pinacolic deamination
C6H5x /CgHs hono jP /G6H5
>C— C ► C6H5-C~ C-Me
C6H5X^ I XMe XC6H5
?
H
NH
a
2, Semipinacol ic deamination
CsH5x?_C^e _HONO^ C6H5_cl0c^e
C6H5/ 0. NH3 NSeH5
H
The relative ease of migration of aryl groups may be determined by making the groups different.
Whitmore has shown that semipina colic deamination involves a Walden inversion. The optically active amino alcohol was used in reaction 2 above and an optically active ketone resulted. By re- lating the configuration of starting material and product to com- pounds of known configuration, it was proved that a Walden inversion has occurred.
Bibliography
Ann. Reports, 25, 134 (1928); 27, 116 (1930)* 30, 181 (1933), Complete references given.
Bachmann et al«, J.Am, Chem. Soci, 54, 1124, 1969, 2112 (1932); 55,
3819 (1933); 56, 170, 2081 (1934); 57, 1095 (1935); 58,~Tll8 (1936), ' —
Whitmore, ibid., 55, 1528 (1933); 61, 1324 (1939).
Bartlett, jjbid., 6g, 2927 (1940).
Tiffeneau, "Glycols", Masson and Co,, Paris (1940),
Reported by J. 0. Corner, October 22, 1941
': - . -.,
•' £.M - . .,
' ■'■'■'■■ '■'■■'■ ♦ ;
1 ••..-• , J-1,
> ■ ~
i ' :
* ■•■•■■ '"
NEW SYNTHESES OF HEXATRIENES AND SQUALENE
The basis of these syntheses is the use of the Grignard compound of 1,4-dibromobutane in reactions with carbonyl compounds to form 1,6-glycols by regular (1,2) addition, subsequent dehydration and dehydrogenation giving the desired hexatriene.
The 1,4-dibromobutane is obtained by treating N-benzoyl pyrrolidine with PBr5 according to the von Braun method?
">
NCO0
PBr,
•7"
y
Br C0
Br
— CHgBr \N=CBr0
BrCH3CH3CH3CH3Br + 0CN
When the G-rigna mine, and been shown It is the C6H5MgBr r dissolved almost qua the purifi with Se03, lization o pale green
BrMgCH2CH3
an ether solution of ben rd compound of 1,4-dibro the reaction product sep
that this product is 1, same product that Bouvet eact with ethyl adipate, in water, a spontaneous ntitative yields of 1,1, ed hexadiene is dissolve
one obtains 1,1,6,6-te f the product from aceti
plates - the color aris
0
CH3CH3MgBr + C6H54-C6H5
zophenone is added to a solution of mobutane, the solution turns car- arates as a viscous mass. It has 1,6, 6~tetraphenyl-l,6~hexanediol. obtained when he let excess When this impure alcohol is dehydration takes place giving 6, 6-tetraphenyl-l, 5-hexadiene. If d in glacial acetic acid and heated traphenyl hexatriene upon recrystal- c anhydride. It is in the form of ing from the conjugation.
OH OH
} i
— > CgHs— C— CH3CH3CH3CH 3— C—CgHs — > I 1
CbHk CeHc
CgHs— C— CH— Cri3GH3CH— C— Cg H5 I I
CeH5 C6H5
C6H5-C=CH-CH=CH-CH=C-C6H5 C6H5 C6H5
The hexadiene may also be converted to the hexatriene by heating it with jp_-benzoquinone at 170-180°.
Fluorenone was also made to react with the Grignard solution of 1,4-dibrornobutane to form l,6-difluorenyl-l,6~hexanediol. This turned out to be so very insoluble that ordinary recrystallization was quite difficult. Hence the impure alcohol was heated with benzene sulfonic acid in acetic anhydride, which eliminated the water to give the hexadiene. This was converted to the hexatriene by means of Se03 in glacial acetic acid as above. The result was 1,6- dipiphenylene hexatriene.
.
iV,Xz
. ■
■
i '■
I ,
.
, ■■•-■yl\
«
t i ,
•'• a»l*-,^:
ft*-
" t. : ■■•*"' .■■■"■>'■■ ■"" "' k.
!>v
-. -,:v ...; .-;■ ■ ... .:t.- /<-:,
ff«t/i
I;.
\ x . •
■ y •/'•- . i!vri:o ,f; r :r ..... ;.> ; .
.-.-: . ' ,u/-
• ■
- 2
0;
45
CO + BrMgCH2CH3CH3CH3MgBr
*7%>.,
X ^
ca
/
NC-CKSCHSCHSCH3-C
V
<n
-<
^
C=CH-CH=CH-CH=C/'
O
V
Si
/
Acetophenone and benzaldehyde were also used in this reaction; good yields of the hexatriene were obtained in each case. It is interesting to note that the l,6-diphenyl*-l,6-dimethyl-l,6~ hexanediol from the acetophenone gives rise to two forms — one a racemic modification and- the other a meso form. This results from the two similar asymmetric carbon atoms which the molecule possesses.
OH ?H
C6H5-C*~CH2-CH3-CH2-CH2~C*-C6H5 l I
CH3 CH3
There are few classes of natural products that are harder to syn- thesize than the carotenoids.
This is mainly due to the long branched carbon chain and the ex- tensive system of conjugated unsaturations which are characteristic of this class of compounds.
One of the most important of the carotenoids is ^-carotene which
can be converted by the body into Vitamin A, The new synthesis of
hexatriene s gives us a good means of producing 0-carotene.j 0Ha CH3
R
OH
OH
\
■ CH:=CH-C=CH-CH=CH~
CH3 CHa
R— C— Cn3CH2CH2CH3— C—R
CH:
CH«
R-C=0 + BrMgCH2CH2CH2CH2MgBr CH3
R-OCHCH3CH3CH=C~R
CH,
l
CH,
R-C=CH~CH=CH~CHs=C-R . i i
CH3 CH3
|3-carotene
However it should be noted that the ketone with which this syn- thesis starts is not know but it has been w ell established that it could be produced by condensing £-ionylidene acetaldehyde with acetone;
» _ r.,
:. - ■ '
•
' j'l &
f-JV.f ' '..'
■ i
«,1 ',
;-*0- ■ ■
•. . '
-
. s-iv
» .r - > f
f
; - '"
gfc-. J W1'
' <'
,vi.'
,. i
-/,-.•:
■ .. i ■ . -
;;,; '• .. ;, '■ ■• ' . ; ':. ■■■:■ I: ■'J
46
CH3 CH3
CH=CH-C=0 \ CH9
CH3 CK3
,CH=CH-C=CH-CHO
CK3CHO
CHS
CH.
{5-ionylidine acetaldehyde
*
0
CH3 C--CH3
CH3 CH3
CH=CH~C=CH-CH=CK-C=0 CH3 CH3
^CH3
It would seem simpler to use l,4-dibromo-2-butene and thereby omit the dehydration, but this is not possible - only the pinacol is isolated.
Squalene is an unsaturated hydrocarbon which occurs in the liver oil of fish of the shark family and also in yeast. It is made up of six isoprene units and gives crystalline compounds with HC1 and HBr that are suitable for its characterization.
It is interesting to note that the Barbier reaction is employed instead of the G-rignard reaction in this synthesis of squalene. In the Barbier reaction the 1,4-dibromobutane is added to the geranyl acetone in an absolute ether solution of Mg with a crystal of iodine as catalyst, and the "nascent" G-rignard reagent reacts in a normal manner before complicating side reactions can set in. A vigorous reaction takes place and the diol is formed. This is converted into squalene by distillation under diminished pressure.
The geranyl acetone is first formed from geraniol by an aceto- acetic ester synthesis: q 0
CH3c£cH2-C^OC2H5 CH3C*CHCH3CHaC=CHCH2OH = R'CH R'CH PBra , R*Br — = —
CK3 CH3
CaHKOH
genanlol
,0
•c
I
R' 1
i^°
CH,c£cH + (£oC,Hs Ba(°H)3
co:
H30 Ketonic cleavage
CH3C=CHCH3CH2C=CHCH2CHsC=0 Cn3 CH3 CH3
geranyl acetone
Then the Barbier reaction is carried out, using 1,4- dibromobutane with the above product:
•
.1* k
-
. . . » V.
.
■'•• •
[ \
o
b. 4 -
OH OH
R'CH2C=0 + BrCH3CH3CH3CH3Br — ^-— ► R1 CH3C~CH2CH2CH3CH3-CCH3Rf
CH3 j CH3 CH3
■ — > CH3C— CHCH3CH3C— CHCH3CH3C— CRCH3GH2CH=C— CH2Cn3CH=CCH2CH3CH^. CH3 CK3 CK3 CH3 CH3 Qjfa
CH3 Squalene
The advantages of this squalene synthesis are:
(1) Cheaper and more easily obtainable starting materials
(2) Squalene is produced in better yields and in purer crystal- line form.
The old preparation started with farnesol — > farnesyl bromide
MgBr
► squalene. See reference to Karrer.
Bibliography
Josef Schmitt, Ann., 547, 103 (1941).
Jose;? Schmitt, Ann., 547. 115 (1941)
M. T. Bogert, G-ilman - Organic Chemistry, 1158.
Karrer, Organic Chemistry, 53,
All other references are given in Ann.
47
Reported by P. F-» Warfield October 22, 1941
' ■ n J ■-
■: ■ -. it S ;
■ .
- <-. ■ ' -
'*. .1
s .1 '
48
ATTEMPTS TO DETERMINE THE STRUCTURE OF PHTHIOIC ACID From the 1940 Presidential Address by Robert Robinson
Isolation
Phthoic acid was isolated by R. J. Anderson. He extracted moist Hue tubercule bacilli with an equal mixture of alcohol and water. The residue was further extracted with chloroform. The chloroform fraction yielded a wax. The alcohol-ether extract was treated with acetone which caused a precipitation of phosphatides and left in solution various glycerides.
The phosphatide precipitate was decomposed by hydrolysis, yielding 33% of water soluble constituents, consisting of carbo- hydrates and glycerophosphoric acid, and 67% of a mixture of fatty acids. This mixture of acids was converted into a corresponding mixture of lead soaps. This mixture was extracted with ether, leaving a residue, which yielded pulmitic acid on decomposition with acid. The ether extract was hydrogenated catalytically and the re- sulting mixture of acids was converted to the lead salts and extracted with ether. A residue was left which, when decomposed, yielded stearic acid. The stearic acid resulted from the hydrogenation of oleic acid present Juthe original mixture of fatty acids* The ether extract was decomposed yielding a yellow oil which solidified on cooling. The melting point of the resulting solid was 21°, the rotatory power La.]so£ was +11.96° and analysis showed an empirical formula C26H530a. Anderson named this product phthioic acid.
residue
CHC13
residue wax extract
tubercule bacilli
alcohol
ether
acetone extract > extract containing glycerides
a precipitate
hydrolyzed
33% water soluble gly- cerides and glycero- phosphoric acids
67% fatty acids
fatty acid mixture
Pb
Pb pulmitate Pb oleate Pb phthioite
ether
extract
H+
Pb pulmitate oleic acid phthioic acid
cat.
stearic acid phthioic acid
Pb
++
ether ►
extract
H
Pb stear- ate
phthioic acid
V-
J >■ v i ■._
....
-
•
n
'•
rJ
49
- 2 -
A mixture of fatty acids was also obtained from the wax frac- tion. On distillation at low pressure two fractions were isolated; a low "boiling one, which proved to be a single compound, was called tuberculostearic acid, the high boiling fraction was shown to be identical with phthioic acid*
Spielman oxidized tuberculostearic acid with chromic acid and obtained n-octyl .ketone, azelaic and j^-octoic acids. He, therefore, showed tuberculostearic acid to be —
H CH3(CH3)7C-(CHs)8COOH
CH3 Q
H v nv,n CK3 (GH2) 7C-CH3
CH3(CH3)7C-(CH3)8C00H K^r°7
I H3S04 CH3(CH3)6COOH
L3
CH
3
H03C(CH3)7COOH
Attempt to determine the structure
Phthioic acid, when oxidized with chromic acid, gave a compound with the formula C1:LH3303| which was claimed to be different from n-undecoic acid because two derivatives melted 20° too low. This "evidence is not thought to be conclusive in view of the small quanti- ties used and the probability of mixtures.
Charguff synthesized a number of C36 acids and, from the fact that the melting point of the n-hexacosanic acid is lowered 20-30° from 88° by the introduction of one side chain, concludes that phthioic acid must have at least three hydrocarbon chains.
Wagner-Jaureyg found by the Kuhn-Roth method of estimating side chain methyl that phthioic acid gave 2.4 moles of C3H403 per mole while 'tuberculostearic acid gave 1.4 moles of C3H4,0S per mole* This is suggested as evidence for three carbon chains in the molecule,
Anderson believes that there is a methyl group in the alpha position and another in the neighborhood of the eleventh carbon atom.
X-Ray reflections from multilayer films of the barium salt showed that the length of the molecule was that of a chain of twelve to fourteen carbon atoms.
Phthioic acid differs from known fatty acids in that it is in thin films. On water it forme a very compressed unimolecular layer collapsing at an area of ZQk0* per molecule. On the other hand, n- decyl-n-dodecylacetic acid forms a much more expanded film collapsing at about 6OA0 , The surface dipole moment- of this film was much smaller than that formed by phthioic acid. Stenhagen suggests the presence of a small alkyl group in the a-position to the carboxyl because this might account for the observed close packing of the chains. He suggested the formula —
&SftJ
0 i.
► i i : :■. .
I *
■
-. i ti .
■ : i. |
. g . .. ; • |
||||
! |
|||||
*. K |
■ |
V |
:.". ■' ' |
||
I ':'■ ' |
j |
||||
W( |
. |
\ ■■:< U '■ |
|||
• f.u |
.■: |
■.: v' ' |
•. - i - ;: |
'! |
;.;Ci ■:•
3 >., v
J. v
■ :) ;•
- 3 -
CH3 (CHe) x CH3(CH3)y __^CC03H
CH3(CH8)2^
in which x and y are about 10-12 and different and z = 0 or 1, the most probable formula thought to be the one in which z = 1, x = 9, y = 11.
Synthesis of Substances of the phthioic acid type
With the ultimate purpose of synthesizing phthioic acid, Birch synthesized several compounds of the phthioic acid type. Methyl di-n-octyl acetic acid was produced by application of the method devised by Reichstein.
CeHi
7
s
C8H17-CC1 +
KMnO
4
v! >-C03Me N0
Aici3 , S8^72c m '
CHa^ ^0^~C°3Me
CqEx 7 ^ CgHi? — ' — CCOgH CH3 ^
This acid formed the compressed films of the phthioic acid type.
a,a-Dimethyl n-decyl acetic acid was synthesized according to the directions of Haller.
NaNOs CH3 \ — 2* CH3~C-C03H
\\ /CH3 |
CH3N //0 |
C€H5C-C-CH3 + NH3Na - |
-> CHa-C-C |
C10H8i |
C10HS1/ N |
NHNa H3S04 C10H3"^ This acid also produced films of the type shown by phthioic acid.
••• i •
*- .-, ■ —
X V i
■
T.
I
'7 V
51
After several attempts £^-~^i-£^-octylbutyric acid was
synthesized.
0 CH3-COsH ethyl sodio
C8H17iS-CH3 GUareSCH CH3-9-.CH17 - ester - ester
chloride a-acetyi
CH2C0sH n-heptoate
Clemmensen
0 CH
hydrolysis , j 3 Clemmensen
reduction
CfiH
8n*7
CH3 (CH2) 7 \
CH3(CH2)7 CCH3C03H
This acid also gives films of the phthioic acid type.
CQn.cXu£..ion
., Tuberculosis and leprosy are caused by infection with bacilli characterized by the possession of a fatty or waxy envelope. In the case of leprosy, chaulmoogric acid and its derivatives prove to be bactericidal. It has been shown that the bactericidal action is not characteristic of any one group of atoms but rather due to the physical character of the molecule. The bacilli probably suffer an impairment of the fatty envelopes in the presence of chaulmoogric acid and its derivatives. It has also been shown that the best results were obtained with molecules containing 16 or 17 carbon atoms* Molecules of the phthioic acid type synthesized by Robinson and Birch, especially £, p-di-n-octyl butyric acid, are found to be bactericidal to tubercule bacilli. From the above analogy it is expected that the compounds of this type containing 16 or 17 carbon-- ( atoms may be even more active when synthesized.
Bibliography:
Anderson, J. Biol. Chem., 74, 525 (1927); ibid., 83, 169 (1929),
Anderson and Charguff, ibid,, 85, 77 (1929-30); ibid.. 106, 87 (1934).
Vl^Quff, Ber., 65, 745" (1932).
Wagner- Jaureyg, Z. Physiol. Chem., 247, 135 (1937).
Reichstein, Helv. Chim. Acta, 18, 271 (1938)*
Haller, Ann. Chim,, 1, 15 (19147.
Shriner and Adams, J. Am. Chem. Soc, 42, 2727 (1925).
Adams and Stanley, J. of Pharmacology and Exp. Therap., 45, 121-162
Robinson, J. Chem. Soc, 505-j(l940),
(1932)
Reported by R. G. Chase October 29 f 1941
I*
• -
.
. .
■
•£ >•;.:
^ « .......
t ■*' .
••• ! .
• ' -' .
" % ■- .'
. * ... ■ ' i. ■ . ,
:
52
THE OPTICAL ISOMERS OF CIST9-METHYL-1-DECAL0NE
Plentl, et al
In recent years the synthesis of steroids has been of special interest, but the question of their optical isomerism has not been so fully studied.
The ultimate aim of all of these syntheses was a comparison of the synthetic steroid with its corresponding natural compound. Since the latter, in all cases, are optically active the synthetic steroid must be a pure substance and not contain stereoisomers impurities.
cis-9-Methyl-l-decalone 8 first reported by Chuang, was selected as a raw material for study to obtain an isomer of definite composi- tion. This compound has the advantage of possessing an angular group at C9 which gives dissymmetry as well as being a general characteristic of all steroids. From this compound, methods are available for the preparation of both perhydrophenanthrenes and cyclopentenophenanthrenes.
Methylcyclohexenylbutyric Acid
The only methods reported for the preparation of 9-methyl-l- decalone are dependent on the synthesis of methylcyclohexenylbutyric acid; consequently the method previously worked out by Elliot and Linstead was tried. In this a Grignard reaction is run with 5- bromo-1-pentene and methylcyclohexanone with permanganate oxidation of the tertiary alcohol, followed by dehydration of the hydroxy acid.
CH3=CHCH2CH2CH2Br -> CH2=CHCH2CH2CH2MgBr -+
CH3 OH
CH2CH2CH2CH:=Cri2 — ►
However, appreciable amounts of methylcyclohexanol were found in the Grignard reaction, which is in agreement with the observation of Butenandt that aliphatic magnesium bromides with moderately long side chains have reducing properties. The dehydration of the hydroxy acid also forms a spirolactone. This* need not be discarded but is simply converted to ethyl methylcyclohexenylbutyrate in very good yields by boiling with thionyl chloride in benzene solution and pouring the mix- ture into absolute alcohol. ^"Lactones are usually split in this way and apparently the procedure is equally as good with C -lactones.
Simultaneously with the above study a new synthesis of this acid was attempted by Plentl and Bogert which proved superior in many ways. This is shown in Flow Sheet A. Essentially the method is the elonga- tion of the side chain of methylcyclohexenylacetic acid by two successive Arndt-Eistert rearrangements. The substituted acetic acid had previously been prepared by Chuang as follows:
(' I
1 -
. lii r
^ >
,-)
• ■ ..■ -.;/
OC 53
CKa 0 CH3 cHsC02Et ?Hs
/^
BrCH2C02Et Zn
cis-9-Methyl-l-decalone
- 2 -
St
Z_\ / ^SrCHBCOaEt
KHSO4
This preparation is also outlined on Flow Sheet A, and is Essentially a cyclization of the acid by the Darzens reaction, as modified by Cook and Lawerence, to the corresponding chloroketone which after removal of the HC1 yields the unsaturated ketones.
Resolution
Only the purest fraction of cis.-9-methyl-l-decalylamine was used for the resolution. Although Hueckel aid Kuehn were able to resolve a-decalylamine using camphorsulfonic acid, this reagent was unsatis- factory here because of the excessive solubility of its salts. However, bromocaraphorsulfonic acid formed salts which were sufficiently difficultly soluble in ethanol to be separated.
The amines were regenerated from these salts and allowed to re- act with nitrous acid to give a mixture of a hydrocarbon and a partially inverted alcohol. The hydrocarbon was probably Z^1'2 -9-methyloctalin and was not further studied. The conversion to the alcohols seemed to involve partial inversion at Cx, since the mixtures obtained from the d- and 1-amines could not be brought to equal and opposite rotation. However since oxidation of the mixture gave quite pure d- and 1-9- methyl decalone, this indicated that the Walden inversion was confirmed to Ox,
These facts are in agreement with Hueckel 's observations that in compounds of this type partial inversion occurs, although he was unsuccessful in isolating the alcohol. Further evidence is found here that the compound is eis and not trans .
since in trans compounds of this type Hueckel showed that no inversion occurs.
Bibliography
Plentl and Eogert, J. Org. Chem., 6, 669 (1941).
Elliot and Linstead, J. Chem. Soc.7 1958 f 660.
Butenandt, Cobler, and Schmidt, Ber. , 69, 448 (1936).
Cook and Lawerence, J. Chem. Soc, 1955 1637.
Hueckel and Kuehn, Ber., 70 2479 (1957).
Hueckel, Ann,, 53J3, 1 (1958),
Chuang, Tien, and Ma, Ber., 69, 1494 (1956).
Reported by J. D. G-arber October 29, 1941
Uli-J
* ♦
t-...-^....'.<jt. ( ' ' • ■ *;.! ' »,«*«■ « '-- r .
.- ', ■ t *r • J • ' ■- »
-'-■'-■■'■ *
CH
FLOW SHEET A CH3
CH3
0
CH.
COCHN2 I ■CH2
CONH:
CH3 0
/n/s
CI
>
. t '
•
'•'■■
,
. '
J.vj
FLOW SHEET B
55
HON (Han
4-
NOH
CH
NHS
W
CH3
■»
Co*
NH:
CH3J
Bromocamphor sulfonate [aD]HP0 + 75°
Bromocamphor sulfonate [aD3H 0 + 59*2°
\K
(~)cis-9-Methyl-l-decalyl-
amine hydrochloride
o [a ] - 6.9
/
Stereoisomeric decalols
(-)ci8-9-Methyl-l~decalone
Bromocamphor sulfonate [aD]H,0 + 68-8°
( +)jgis-9-Methyl-l-deca- lylamine hydrochloride
£aD^H20 + 7-°
Stereoisomeric decalols
V
( + ) cis-9-Methyl-l« decalone
CaD]Et0H " 3«9
^aD^Et0H + 4*2
:\ *.; . ;%>.:
r :
V' /
'"' ■'''■'>'
56
STRUCTURES OF PYRETKRINS I AND II
Pyrethrins I end II ere the active insecticidal constituents of Pyrethrum flowers. It is important to determine their structures so that attempts may be made to produce them synthetically.
Staudinger and Ruzicka did the first work on the structure of the pyrethrins and arrived at formulas I and II for the respective pyrethrins:
H ?Hs „ CH3
\i H H3 H H * 4 L
H2-C' V-C— C=C=C-CH3 Ha-C/f'V— C-C=C=C-CH.
II ■ II
H-C — C=0 K-C— C=0
0 0
c=o c=o
AH /GH3 (CH3)eC<C-Sg-C=C/ *"
(GH3)3C^0-C=CN _ H NC0CH:
H ^CH3 b
I II
Structure of the Acid Component s> — Staudinger and Ruzicka on saponification of pyrethrin extracts were able to isolate but one alcohol which they called pyrethrolone and two acids which they named chrysanthemum monocerboxylic acid and chrysanthemum dicar- boxylic acid corresponding to the acid portions of I and II. Some chrysanthemum dicarboxylic acid monomethyl ester was also separated, which indicates which -C00H group is attached to the pyrethrolone nucleus* Ozonolysis of chrysanthemum monocerboxylic acid yielded acetone and 1-trans-ceronic acid; ozonolysis of chrysanthemum di- . carboxylic acid yielded 1- trans.- car onic acid and pyruvic acid, thus establishing the structures of the acids as shown in I and II. LaForge and Kaller discovered a third acid constituent whose formula is C16H3002 but have not identified it*
Structure of the Pentenolone Nucleus 4 — The results of Staudinger and Ruzicka indicated the structure shown in I and II for the cyclo- pentanolone nucleus. LaForge and Heller showed that the analyses of pyrethrolone, tetrahydropyrethrolone, and their semicarbazones indicated two less H atoms than represented by their empirical for- mulas. They reduced pyrethrolone (III) to tetrahydropyrethrone (V):
CH3 CH3 CH3
H2-C "C *" C5H7 p rr Hg— C C— C5H11 Hg— C C— C5H11
HOC C=0 HOC — C=0 C1C C=0
1
1
8 H H
III IV
Zn
*
tr. i
i
' 'i •
, ■ , ' ■; ■
j > ... ; ; f.j
1 •
I
:f . ••;
■i ;
' R
■ -
"L
! •
I -
"
i ■ Si y ■ ) -
• i > ■' ..•-*
+i
57
Zn
CH3
/£%
Kg-^C C-CsHix
I / V
Hg-C — C=0
Mixed melting points showed the semicerbazone of tetrahydropyre- throne (V) to be identical to thgt of dihydrojasrnone of known struc- ture V.
structure of the SldecbeiEU — Studies of the structure of the sidechein have attracted considerable attention, for if it has the structure indicated by VI, it will be the first natural product known to contain a cumulated system of double bonds.
Staudinger and Ruzicka postulated structure VI in the side- chein from their studies. Lfter Ruzicka and Pfieffer abandoned VI in favor of VII .
CH3
fT-T
H2HH pHHHH
H3-C XC-C-C=C=C-CH3 H3-C '0-0=0-0=0-0^
[12346 j j 1 3 3 4 S
HOG 0=0 HOC 0=0
I i
H H
VI VII
LaForge and Haller's early work was evidence against a. con- jugated system but did not allow a definite choice,
Ozonolysis yielded acetaldehyde, establishing the position of the 3,4-double bond. Further oxidation with hydrogen peroxide yielded malonic acid which is evidence for the cumulated system. Only in one instance was an acid of probable structure VIII isolated in small yield from the ozonolysis.
CH3
A
H2-C C-CK2C00H
| I VIII
AcOC — 0=0
H
Treatment of pyrethrolone (VI ) with aluminum amalgam should have given 1,4-addition if a conjugated system were present. Sub- sequent reduction of the -OH group would have resulted in jasmone. Reduction with aluminum amalgam yielded pyre throne (IX) however. Pyrethrone on catalytic hydrogene tion furnished a tetrehydro deriv- ative identical with dihydro jasmone. (Nuclear double bond is very resistant to hydrogenation, )
'.
•
• ■ . -t
■ •
,)■• i. •
.-■•■•■• ■ . -
rt
•■:
; ,.
. . ..
..-";
■
'
■■
■>
'
^y 58
CH3 fH3
Hs-C C-C5H7 A1 tj H3-C C-C5H7
I I . A1 rtg > ! I
HOC C=0 H3~C — C=0
H
VI IX
Pyrethrolone and pyrethrone do not form characteristic products with raaleic anhydride or ^naphthoquinone,
Pyrethrone absorbed one mole bromine readily in an indifferent orgrnic solvent. If 1,4-addition had occurred, the product should have yield_ed jasmone when reduced with zinc in acetic acid, but the original pyrethrone resulted. Evidently bromine added on adjacent cprbon atoms and no choice could be made.
Addition of one mole of bromine in ethanol solution yielded p. monobromo compound plus nearly one mole of hydrogen bromide, Con- sequently the bromine reaction was considered as one of substitution. Addition of two moles of bromine gave a dibromo compound plus nearly two moles of hydrogen bromide. Both bromo derivatives gave pyrethrone when reduced with zinc. Only the sidechains are involved in that tetrahyclropyrethrolone and tetrahydropyrethrone do not decolorize bromine.
Pyrethrone was heated with sodium in a sealed tube and the product was treated with carbon dioxide resulting in the formation of an acid. This is typical of ^C=CxCHCH3.
No conclusion can be reached from the above results except that no such behavior has been noted with a conjugated system. The liter- eture furnishes no clues of the behavior of cumulated systems toward halogens except that a lien e is stated to add four atoms of bromine to give an unstable tetrabromide. Therefore, Acree and LaForge preppred l-phenyl-l,2-butadiene, l-cyclohexyl-2,3-pentadiene, and 2,3-pentadiene to compare their reactions with halogens with those of pyrethrone and pyrethrolone.
The above allenes did not react with maleic anhydride or a-naphthoquinone (cf, pyrethrone above).
All three of the allenes mentioned gave dibromo and dichloro pddition products with bromine and chlorine in indifferent organic solvents in the cold as did both pyrethrolone and pyrethrone.
Conant and Jackson and later Jackson and coworkers have re- ported that certain compounds with ethylenic linkage yield methoxy- bromo derivatives (a) as well as the normal dibromo derivatives (b) when treated with bromine in methyl alcohol*
(a) ^C=CC + ROH + Br s -» ^C— CC + HBr
6 Br R
(b) ^C-£^ -h Br, -* ^C-CC
Br 3r
u
« •
4,-
I
; '
-4- q 59
Both reactions proceed at a. slow rate*
Both the dibromo and the methoxybromo products were formed when each of the above mentioned alienee was treated with bromine in methrnol. Liberated hydrogen bromide amounted to 60-70$ equivalent, showing that reaction (a) predominates* In contrast to ethylenic compounds, the allenes reacted instantly in the cold. Pyrethrone reacted anrlogously in methanol to give a methoxybromo and a dibromo derivative. Titration of liberated hydrogen bromide (60-70^ equiv- alent) and methoxyl and bromine determine tions showed that reaction (a) predominated. Reduction of the mixture of methoxybromo and dibromo products with zinc regenerated pyrethrone in excess of the amount expected from the dibromo derivative present. Evidently zinc gives an analogous reaction with the alkoxybromo compound. Analogy for this reaction is found in an article by Dykstra, Lewis and Boord who reported the t a, p-alkoxybromo compounds are readily dehalogenated by zinc to form a double bond between the carbon atoms that carried the substituents*
All the above reactions of allenes and pyrethrone are com- patible and the cumulated system of double bonds seems the most likely arrangement in the sidechain.
LaForge and Acree have reported the reactions of allenes with lead tetraacetate but have not reported comparable reactions for pyrethrone yet.
Bibliography
Staudinger and Ruzicka, Helv. Chim, Acta, 7, 177 (i), 201 (II),
212 (III), 236 (IV), 245 (V), (1924). LaForge and Ha Her, J. Org. Chem . , 2, 56 (1937); J.Am«Chem»Soc. , 58,
1061 (1936). Ruzicka and Pfeiffer, Helv, Chim. Acta, 16, 1208 (1933). LaForge and Haller, J. Org, Chem,, 2, 546 (1938). Acree, F., Jr., and LaForge, J. Org. Chem., 5, 430 (1940); 6, 208(1941)
Reported, by D. W, Hein November 5, 1941
* ■> I .'•.. ■■: '
■
,.f.
'
'. v
. ■ .y '!
'•:':'.> '■:/
i
-
i
'Iv -
;
i ♦
■t". :i''
-(
■ ' ■'" <£ r '• ' ' •' i. ) ■'■
■ ... -'- '.'■
60
THE MICHAEL CONDENSATION: SOME RECENT INVESTIGATIONS Connor, et al. University of Pennsylvania
The Michael condensation may be generally represented by the equation
i i -C^C-Li
+ L3CHL3
Piperidine or NaCR
-C-C-Li i
H
Lft—C—Lo
in which Llt L3, and L3 are labilizing groups. Examples have been reported in which Lx is -COOR, -COR, -CN, -CONK3, ~N03, -S03R> and in which L3, L3, or both, are -COOR, -COR, -CN, -CONH3, -N03, -S03R, -CHO. The acceptor may be acetylenic rather than olefinic, or it may be a quinone , Either the acceptor or addendum may be vinylogs of these structures.
Influence of Experimental Conditions and the Structure of the Acceptor on the Condensation
Connor's work has shown that secondary amines, for instance, piperidine, are the safest catalysts; they seldom cause any reaction other than normal condensation. However, amines often fail to • bring about reactions that occur in the presence of NaOR, and the rate is so slow, even in favorable cases, that a long reflux is necessary. One-sixth to one-third of an equivalent of NaOR may bring about condensation when amines do not; the use of one equi- valent of NaOR is most likely to cause condensation, as well as side reactions. With sodium alkoxides as catalysts, the best re- sults are obtained by permitting the reaction mixture to stand at room temperature for twenty to one hundred and fifty hours. Higher temperatures may give lower yields, probably because side reactions and retrogression are favored.
Nature of Lx x
Arrangement of groups in order of activation of double bond is not generally possible but unsaturated ketone^ corresponding- ester > nitrile. Examples in Table I.
TABLE I
Acceptor
Addendum
% Yield Conditions
1. 0CH=CHCO0
2. 0CH=CHCOOEt
3. 0CH=CHCOOEt
4. 0CH=CHCN
5. p_-O3NC6H4CH=CHCO0
0CH3COOEt 0CH3COOEt 0CH3COOEt 0CHaCOOEt CH3(COOEt )s
6. p_-03NC6H4CH=CHCOOEt CH3(COOEt)
90 |
A |
0 |
A |
85 |
C |
0 |
c |
90 |
A |
0 |
A |
.,■ -
id
-v jz
■
; ■■
>'')
; ■■ ■'■ ft
■
•'...V
•
; ui
1 j , v
- 2 -
6i
Conditions
A. Piperidine catalyst, long reflux
B. Small amount of sodium alkoxide, stand at room temperature
C. Equivalent amount of sodium alkoxide, hot.
Substitution on a and g atoms
(1) Reactivity of acceptor decreases as hydrogens are replaced*
(2) Reactivity decreased if substituent is alkyl, aryl, carbethoxyl or acyl.
Typical results are shown in Table II.
TABLE II
Acceptor
Addendum
% Yield Conditions
1. 0CH=CHCO0
2. 0CH=C(COOEt)CO0
3. 0CH=CHCOOEt
4. 0CH=C(0)COOEt
5. CH3CH=CHC00Et
6. CH3CH=C(CH3)C00Et
7. (CH3)2C=CHC00Et
Remote Substitution
0CH2CCOEt
0CH2COOEt
0CH3COOEt
0CH2COOEt
0 CHgCOOEt
0CH3COOEt
0CH3COOEt
90 |
B |
0 |
C |
85 |
C |
0 |
C |
90 |
C |
40 |
c |
20 |
c |
Groups not directly attached to the double bond of acceptor may have greater effect than would be expected (Cf. Table III).
TABLE III
Acceptor
Addendum
% Yield Conditions
o-03NC6H4CH=CHC00Me m-02NC6 H4 CH=CHC00Me p_-02NC6H4CH=CHC00Me
0CH=CHCO Mes
CH2(C00Me)2 CHa(C00Me)a CH2(C00Me)3 CH2(C00Me)2
70 |
B |
95 |
B |
0 |
B |
70 |
B |
Instability of some Addition Products
In some cases, there is an unusual difference in reactivity between a substituted active methylene compound and the next higher homolog.
I : i !'
■
i
■
'
'. ,-i
■ . ■ : , ..i
".'J i
U • :
*K\ .. :■ .
"■. r
.1 ■ " s ' . ."si
.iV: '
- • '
■' : 1 ! » ".:
-3-
0CH=CHCO0
+ RCH(CO0Et)g
N fcOEt
0CHCH3CO0 R-C(COOSt)3
I R = CH3- ou/o II R = C3H5 0%
If the difference is due to spacial interference, II, once prepared, would be expected to be stable. An addition product of a Michael condensation was alkylated, as a possible synthesis of IL
0CHCH3CO0 CH(COOEt)3
III A
0CH=CHCO0
+ CK3(COOEt )a
B
0CRCH2CO0 C3H5C(COOEt};
/|\
ir i
V
0CH=CHCO0
+ C3K5CH(C0OEt)3
II
None of compound II was i sole ted even at -78 . Steric hin- drance is improbable when one considers the results of Connor and Andrews obtained by the reaction of the sodium derivative of ethyl ethylmalonste with benzalacetophenone to give ethyl cc-ethylcinnama te and ethyl benzoylacetate which must be the results of a Michael condensation.
Thus the data indicate ttu t the expected product is so readily cleaved by NaOR that isolation is impossible and steric hindrance does not prevent reaction.
Activation of the Methylene Group by Carbon -Carbon Unsaturation
The possibility of L2 and L3 both being aromatic or olefinic was investigated by Connor. The reactivities of fluorene, cyclo- pentadiene and 1, 4-pentadiene were studied.
Fluorene reacted with benzalacetophenone, benzal-p_-bromo- acetophenone, and benzalacetone in the presence of one equivalent of sodium ethoxide (yields 2-27%). No reaction occurred with a, p-unsaturated esters or m- or p_-nitrobenzalacetophenone.
0CH=CHCO0
+
CH3
NaOEt>
J
0CHCH3CO0 CH .
\S
%,
IV
Cyclopentadiene reacted with a, p-unsaturated ketones when piperidine was used as a catalyst. This indicates that it is a highly reactive compound. The reaction ves carried out under pressure to prevent the loss of the hydrocarbon (yields 25-30%).
■y
-4~ 63
0CH=CHCOC6H4Br(p_) 0CHCH3-COC6H4Br (p_)
. + I
^CH^. — ■ > /GHX
CH CH (CH3)5NH CH CH
I. II |! !l
CH CH CH— CH V
Pentediene-1,4 reacted with benzal. p.~bromoacetophenone in the presence of en equivalent amount of sodium ethoxide. The same compound was obtained by using the sodium derivative of the diene.
0CH=CHCOC6H4Br(p_) Ich^CH^CH^ > 0CHCH3COC6K4Br (pj
CH3=CH-CH-CH=CH3 (CH3=CH)3CHNa , vi VI
Bibliography
Andrews and Connor, J. Am, Chem . Soc, , 57, 895 (1935); 5_6, 2713 (1934)
Connor and McClellen, J. Org. Chem,, 3, 570 (1939).
deBenneville, Clagett rnd Connor, ibid"., 6, 690 (1941).
Ingold, Perren and Thorpe, J. Chem, Soc.,^119, 1976 (1921)*
Ingold and Powell, ibid,, 121, 1771 (1921).
Taylor and Connor, J. Org/ Chem., 6, 696 (1941).
Reported by R. E. Foster November 5, 1941
Mr
■?
64
REACTION OF ALKYL BENZOATES WITH SODIUM aLKOXIDES McElvain, et al, University of Wisconsin
A study of the products formed when alkyl benzoates are heated with the corresponding sodium alkoxide led McElvain at 'Wisconsin to suggest that a reverse Tischtschenko occurred. The "benzaldehyde and aliphatic aldehyde or ketone formed may then take part in one or several of the following reactions:
1. Forward Tischtschenko
2. Mixed aldol condensations followed by loss of CO to give alkyl phenyl carbinols which in turn give ketones.
3. Further condensation of the ketone, as acetophenone, with un- reacted ester to give 1,3-diketones,
4. Saponification of the ester by water formed in various conden- sations.
5. Condensation of the alcohol or its ester with sodium alkoxide through the Guerbet reaction.
6. Acyl exchanges between esters and 1,3-diketones.
Adickes had treated ethyl benzoate with sodium ethylate and ob- tained not only more sodium benzoate than expected but also a 20 per cent yield of dibenzoylmethane. McElvain found that heating 4 moles of the ester with one mole of the alkoxide at 175-180° for two hours gave a maximum yield, 40$, of dibenzoylmethane from ethyl benzoates. These conditions were then employed for treating the methyl, ethyl, ' propyl, icopropyl, butyl, isobutyl, and neopentyl esters. Table I is a summary of most of the results.
The formation of benzylbenzoate as a common product is cited as evidence that although the reactions are apparently quite different they may follow a common initial course, a reverse Tischtschenko, and the benzaldehyde formed could then give' benzylbenzoate ■ :
by a forward Tischtschenko. Reverse Tischtschenkos with neopentyl benzoate and isobutyl benzoate would give trimethylacetaldehyde and isobutyraldehyde. These aliphatic aldehydes have either none or only one alpha hydrogen and might give esters hy forward Tischtschenko reactions. Neopentyl trimethylacetate was found in good yield and enough isobutyl isobutyrate was formed to indicate that isobutyralde- hyde might have functioned in a forward Tischtschenko.
If ethyl and propyl benzoates took part in a reverse Tischt- schenko reaction the resulting aliphatic aldehydes would be capable of entering into aldol condensations with the benzaldehyde. The following mechanism is proposed to explain the origin of a 40 per cent yield of dibenzoylmethane from ethyl benzoates.
C6H5C00C2H5 -> C6HBCK0 + CH3CH0
C6HsCH0 + CH3CH0 -* C6HSCH0HCH2CH0 ~C0 > C6HsCH0HCH3
C6H5CH0HCH3 -^22 ► C6H5C0CH3
In a similar manner propiophenone would be formed from propyl benzoate.
;r .
■i ■
1
■
■ .
■■ 'J ,' ;; '«
''ri "<5i*5fli.4fi
5/Vi
65
* 2L-
Dibenzoylmethane, produced by the ethyl ester, could have been formed by a Claisen condensation,,
C6H5COCH3 + C6H5COOC3H5 Na0CsH5 > (C6H5CO)2CH2
It is recognized that the loss of carbon monoxide by an alpha hydroxyaldehyde is without precedent in the literature. Those esters giving high yields of the carbinol also gave high yields of carbon monoxide. Trimethylacetaldehyde formed in the neopentyl benzoate re- action could not give an aldol condensation to form an alpha hydroxy- aldehyde and this reaction did not liberate carbon monoxide.
The results from isopropyl benzoate are of particular interest be- cause the acyl exchange reaction suggested proved to be of considerable importance when studied further.
C6H5COOCH(CH3)3 NaOCH(CH3)2 ? ^^0 + (CK3)2C0
C6HBC00CK(CH3)2 Claisen ) C6h5cooCH3COCH3
C6H5C0CH8C0CH3 Acyl jlxchanSe , (C6H5C0)3CH2 + CH3C00CH(CH3 ) 2
The presence of 2-methyl-l-pentanol among the reaction products of the n-propyl ester and of 2-ethyl-l-hexanol from the n-butyl ester can be explained as being formed by the G-uerbet reaction.
Acyl Exchanges Between Esters with 1.5-Diketones and Esters with g- Keto Esters
The acyl exchange reaction mentioned above was investigated in more detail. If the reaction proposed is correct it should be possible to prepare 1, 3-diketones and {3-keto esters by such reactions as follows
(1) C6H5C0CH2C0CH3 + C6H5C00C2K5 -> C6H5C00CH2C0C6H5 + CH3C00C2H5
(2) CH3C0CH3C0CC3H5 + C6H5C0CC3H5 -* C6HeC0CH2C00C3H5 + CH3C00C3H5
o Attempts to obtain the exchange products by heating at 130-160 with sodium ethoxide caused the ester, e»g« ethyl benzoate, to take part in the reverse Tischtschenko and subsequent reactions to such an extent that the acyl exchange was obscured. When the less basic sodium enolates of the 1,3-diketone or j3~keto ester were employed good results were obtained. The sodium enolates were heated with a sufficient excess of the ethyl ester to form a homogeneous reaction mixture. The temperature employed was sufficiently high to permit the ethyl acetate formed to be removed by distillation. Results of the exchange are shown in Tables II and III.
In Table II it is to be noted that there are good yields and correlations between yields in runs 1,4, and 5 where the acetyl group is replaced by benzoyl and a-furyl. Since js-chlorobenzoyl did not replace benzoyl in run 7 it is believed that the d-j>-chlorobenzoyl- methane formed in run 2 (see footnote a) was formed by the reverse Tischtschenko and subsequent reactions.
The yield obtained in run 1 of Table III suggests a useful method of preparing acylacetic esters provided the esters employed have a
•,
, • . ' -.. -....> I,
T
r r
! . >•:-''!
~ 3 -
boiling point sufficiently high to permit the ethyl acetate formed to distil from the reaction mixture*
A mechanism is proposed for this exchange which accounts not only for the acyl exchange but also explains the origin of considerable alcohol which always distilled along with the volatile ester* A0 Na+ ^O-Na
CSH5CT + :CHCOC6H5 ► C6H5C-~
66
■OEt
COCK.
CH3COOEt + C6H5C(ONa)=CHCOC6H
OEt
•CHCOC6H5 COCH3
I
^ COCH3 C6H5C(ONa)=C-COC6H5 + EtOH
II
Intermediate I formed through a carbanion mechanism could decompose by path (a) to give the acyl exchange product II or by path (b) to give ethyl alcohol. Evidence is lacking for the formation of a triacyl methane, II,. but it may have been hydrolyzed into a diacyl compound and an acid. Acids roughly equivalent to the alcohol dis- tilled were separated from the mixture.
Under somewhat different conditions ethyl isobutyrate and ethyl benzoyldimethylacetate form ethyl benzoate and ethyl isobutyryliso- butyrate. The net result is an acyl exchange. This reaction pro- ceeds in the presence of sodium ethoxide and triphenylme thane at ordinary temperature and is explained by Hauser as a series of forward and reverse acetoacetic ester condensations.
TABLE I
Ratio of Moles Product to Moles Unrecovered Ester for the Reaction of Four Moles of C6H5C03R with One Mole of RONa at 175-180°
Iso- |
||||||
R~ |
Ethyl |
Prolyl |
Propyl |
n- Butyl |
Isobutyl |
Neopentyl |
Unrecovered |
||||||
Ester, Moles |
2,10 |
2.60 |
1.75 |
3.08 |
3.00 |
2,10 |
Benzoic Acid |
.26 |
.35 |
.34 |
.87 |
.34 |
.41 |
ROH |
.08 |
.60 |
1.00 |
,41 |
.47 |
.40 |
1,3-Diketone |
.19 |
. » • . |
»09a .05^ |
%■'■••• |
... |
• * » |
Esters from |
||||||
Tischtschenko |
,07° |
Trace |
.06° |
.013° |
.05° |
.07° |
Reaction |
.10d |
.06| .46* |
||||
RCH0HC6H5 |
• • • |
.08 |
• • • |
.11 |
.06 |
|
.12§ |
.138 |
..09S |
• • * |
|||
CO |
.27 |
.41 |
• • • |
.34 |
• • • |
• • • |
Guerbet |
||||||
Products |
• • • |
.015 .09S |
t f f |
.03 .08& |
• • • |
• • • |
Cl ■
'■• L -
3 r
-j.'.(
j - »
r-'ifl
■'>••' ;
' »
'••"•■■■
I '; • , '
.->
Table I (Conf) aC6H5COCH2COC6K5 ^CexHBCOCHsCOCHa cBenzyl benzoate
- 4 -
Isobutyl isobutyrate eBenzyl trimethylacetate ^Neopentyl trimethylacetate ^Benzoic ester of
Run
TABLE II
Acyl Exchange Between the Sodium Enolate of 1,3- Diketones and Esters
RCOCHNaCOR' + R^COgEt -
R'
R»
C6H5 p_-C6H4Cl
CsHgCHa
C4H30 [a- C4H30 CH3 jp.-C6H*Cl
(a) a 4:0% yield of ethyl benzoate was also obtained
(b) a mixture of CsH5COCHaCOC6H4Cl and C1-C6H4C0CH2C0C6H4-C1 was obtained. Yield not given.
1 |
C6H5 |
CH3 |
2 |
Q«HB |
CH3 |
3 |
C6HS |
CK3 |
4 |
C6H5 |
CH3 |
5 |
CH3 |
CH3 |
'6 |
C6H5 |
C6H5 |
7 |
C6H5 |
C6K5 |
RCOCHNaCOR" |
+ R |
fC03Et |
|
% yt< |
3 Id 1 |
of |
|
R'C02Et |
RC0CH2C0R" |
||
49* |
48 |
||
33 |
b |
||
43 |
b |
||
'uryl)51 |
47 |
||
72 |
32° |
||
0 |
0 |
||
0 |
0 |
TABLE III Acyl Exchange between the Sodium Enolate of p-Keto-Esters and Esters,
RCOCR'NaC03Et + R"C02Et
R"C0CR'NaC02Et + RC02Et % yield of
.un |
R |
R' |
R" R!C02Et |
R"C0CHK,C |
02Et |
1 |
CH3 |
H |
C6H5 56 |
49* |
|
2 3 4 5 6 |
CH3 (CH3^CH CH3 CH3 C2H50 |
C2H5 C2H5 H H H |
C6H5 10a C6H5 72 C4H3O (a-Furyl) 66 C8H4N (|5-Pyridyl)73 C6H5 10 |
61b 50 38 16 |
(a) yield of \0% ethyl acetate and 60* ethyl butyrate
(b) This product was ethyl benzoylacet&te .(c) Main product was a non-distillable tar.
Bibliography :
Magnani and McElvain, J, Am, Chem. Soc, £0, 813 (1938). McElvain and Weber, ibid, t 63, 2192 (1941)7 Adickes, Mullenheim, and Simeon, Ber., 66B, 1904 (1933). Hauser and Hudson, J. Am, Chem. Soc, 62, 62 (1940),
Reported by G, L* Schertz November 12. 1941
u
- 8
.
•••••' .•;
*£:
■' i.. •
■'■ Jb *-. >'- ■•-; - .,- ■•
• - .
63
THE STRUCTURE OF LIGNIN
Hibbert - McGill Freudenberg - Heidelberg
Brauns- Inst, of Paper
Chemistry Adkins - Wisconsin
Lignin is a constituent of the woody portion of plants. Besides b0-5b% of cellulose, wood contains 2£-3C$ of lignin, 15-20$ of sugars and other low molecular weight carbohydrates, and a few per cent of resins, fats, and proteins. ; Since wood contains so muqh lig- nin the structure and properties of lignin have been studied exten- sively by the paper industry, both with a view to finding more efficient ways of separating it from cellulose and to discover uses for it after it is removed.
Lignin is not readily being chemically combined w or glycoside linkages, and cellulose. Rather drastic heating a number of hours w chloride or various acids, doubt but that the amorphou differs from the lignin as the nature of lignin varies and its method of isolation nin is a highly complex sub and controversial one.
separated from the other wood constituents, ith the lower carbohydrates through ether the whole being intimately mixed with the chemical methods are necessary, such as ith sodium bisulfite, alcoholic hydrogen
Whatever method is used, there is little s brown solid which is finally obtained it originally occurs in wood. Moreover,
quite a bit depending both on its source For these reasons and the fact that lig~
tance, the lignin problem is a difficult
In spite of this, most authorities seem agreed as to the essentia}? structure of lignin, differ though they may in some details. Just as a protein may be regarded as a condensation product of amino acids and a polysaccharide as a condensation product of monoees, so lignin is regarded as a condensation product of a number of closely related aromatic compounds. The units are believed to be derivatives of phenylpropane. Typical examples are:
I RCOCHOHCH, II RC0C0CH3
R is guaiacyl (IV) or syringyl (V)
CH,0
HO
^
III RCH=CHCH2OH
OH CH— C-CH:
CHOH-CO-CH3
These units are thought to be joined together as in VI,. Most of the evidence indicates that this explanation is funda- mentally correct and that lignin is comprised of structures such as VI and polymers of VI. Unpolymerized compounds like I, II, and III may also be present.
The proof of the above hypothesis rests upon analytical data and the general reactions of lignin as well as upon the products obtained by the degradation of lignin by caustic fusion, oxidation,
o -f-
1 «
. t
- 2 -
69
hydrogenation, sulfonation, and alcoholysis. Studies of known sub- stances with structures similar to that assumed for lignin have also proved helpful.
Analytical Data and General Reactions
Lignin has a methoxyl content of around 16$* It has an hydroxyl content of about 10$, Color reactions and a certain acidity indicate the presence of phenol hydroxyl groups, but while dimethyl sulfate almost doubles the methoxyl content of the lignin, diazomethane in- creases it by only about 3$. This is taken to mean that most of the phenol hydroxyls of the guaiacyl and syringyl radicals are tied up in ether linkages as in VI.
Lignin can be chlorinated, bromlnated, nitrated, sulfonated, and mercurated in a way reminiscent of sensitive aromatic compounds. The absorption spectra of lignin sulfonic acid indicate an aromatic struc- ture. A positive haloform reaction shows the presence of CH3-C0- or CH3-CHOH-.
Brauns has isolated a lignin by merely soaking wood in alcohol for several days. On the basis of analytical data, he assigns it the formula C42H3306 (0CH3)4{0H)4 (CO) and says that it contains one phenol hydroxyl and an enolizable carbonyl group. This he believes is the fundamental building stone of lignin, an opinion not shared by Hibbert.
Degradation by Oxidation
Methylation of phenol groups with diazomethane, treatment with alkali to break ether linkages, another methylation, and finally oxi- dation with permanganate, gives veratric acid ,VII* / isohemipinic acid VIII, and trimethylgallic acid IX. The yields are very low, which is attributed to the fact that the acids themselves are not very resistant to permanganate. The isolation of isohemipinic acid is good evidence that the side chain of one phenylpropane unit has con- densed with the benzene ring of another.
COOH
COOH
COOH
0CH3 VII
HOOC
f
0CH3 VIII
0CH:
All three of these acids are obtained from hard wood lignin, but only veratric acid and isohemipinic acid are obtained from soft wood lignin. This is in line with other facts indicating that the guaiacyl radical IV is the only important radical in soft wood lignin, while both the guaiacyl radical and the syringyl radical V are important in hard wood lignin.
I '•'
.-vj •;- .1
•;
( i .
■
70
~ 3 ~
Degradation by Hydro aenat ion
Adkins recently reduced lignin catalytically. His lignin was ex- tracted by the relatively mild reagent methanol-dry hydrogen chloride. The lignin took up more hydrogen than a corresponding weight of ben- zene, and besides a large amount of methanol, a 40$ yield of X, XjC, and XI J, all cyciohexylpropane derivatives, was obtained. Unidentified compounds of higher molecular weight were also found*
CH3CH3CH3 CHSCH3CH3 ^CHsCHgCHsOH
OH X
OH XII
When he hydrogenated lignin obtained industrially by the rather severe soda process, he isolated small amounts of X and XII plus some cyclohexanol, 4-methylcyclohexanol, and 4-ethylcyclohexanol. However, most of the products were alcohols and glycols related to polycyclic hydrocarbons with 20 to 70 or more carbon atoms in the molecule. None was identified. Ke concludes that lignin isolated by mild means is made up of phenylpropane units joined in chains, and that the soda process causes considerable cyclization, giving mole- cules very stable to hydrogenolysis. This is good evidence that lig- nin has a skeleton of many carbon atoms.
Degradation by Bisulfite and Alkali
Lignin sulfonic acid from the sulfite liquor of paper manufacture when treated with alkali in the presence of m-nitrobenzenesulf onic acid, gives as high as a Ab% yield of vanillin, XIII, and syringic aldehyde, XIV. This process is actually being used commercially to prepare vanillin.
Small amounts of acetovanillone, XV, and acetosyringone, XVI, were also found.
CHO
x J-0CH3 CH30
OH XIII
CO-CH3
OCH3 CH3O
OCH,
Degradation by alcoholysis
Refluxing wood with et handle hydrogen chloride, a relatively mild method of isolating lignin, gives a quantity of water soluble products, some of which have been identified recently by Hibbert as vanilloyl methyl ketone, XVII, syringoyl methyl ketone, XVIII, a- ethoxypropiovanillone, XIX, and a-ethoxypropiosyringone, XX.
J
»/•;•
■
v.
i
■
i
I
. ■ •■. ■ • '• • ; - lOV:
; • -
»-* ' t • ■ -
-f ti ' .'
!
. . .<
.1 ,. .1
i
»
■
■ . ■■ ■■'■;%■
■* ' ' ...
#■,».' . . I, , ' * J,
'-";• •• : , , J.j. .
■ /.:..■
* « «.
71
i |
-C0-CH3 |
CO- l |
-CO-CH3 1 |
,CO-CHOEt- 1 I |
-CH |
3 CO- A |
-CHOEt CH3 |
|
1 T OH |
^OCHs |
CH30/^fNOCH3 OH |
kJ-OCH3. OH |
c*° |
W^OCHg OH |
|||
xvi : |
c |
XVIII |
XIX |
XX |
This is further proof that lignin is built up of phenylpropane units such as I, II, and III (this was originally postulated on some- what theoretical grounds) and is the first good indication of the structure of the side chain of the phenylpropane units.
The ethoxy group in XIX and XX probably came from the ethyl alcohol with which the wood was refluxed.'
Studies of Lignin Models
ri'he acid XXI has a structure very similar to that postulated for lignin. When heated with alkali, methylated, and then oxidized with permanganate, it gives the same products as does lignin under the same conditions (veratric acid, VII, and isohemipinic acid, VIII).
Coniferyl alcohol, XXII, which occurs in plants as the glucoside feritl, is readily condensed by a trace of acid to an amorphous iubstancs similar to lignin. Thus this substance gives the same oxida- tion products, reacts with bisulfite to give similar sulfonic acids, and has about the same composition. There seems to be a definite relationship between coniferin and lignin, but so little is known about the life processes of plants that it is hard to say just what it is. Likeivise, although there is no lack of theories regarding the forma- tion of lignin and its possible function in' the plant, not much can be said with any degree of certainty.
coni^fc
CH,0
\
m3o<f >-CH CH~CH3
0
-n
CH30
CH=CHCH3OH
S
■COOH
XXI
OH XXII
-OCH3
Bibliography: Phillips, Chem.
Revs., 14, 103 (1934).
Harris, D'J^nni, and Adkias, J. Am. Chem. Soc, 60, 146? (1938).
Freudenberg, {Inn. Rev. Biochem., 8, 88 (1939).
, 61, 2120 (1939).
J. . Am. Chem. Soc. ,
Hibbert, Paper Trade Journal, July 24, 1941. ■
Brauns Adkins
Am* Chem. Soc, Frank, and Bloom,
63, 549 (1941).
Reported by B. McKusick November 12, 1941
, •
*;• ■
••,■••
- i ■■*■".
♦'
V
i
«
- .ti?
: j
......
.. ■','>'■ ,l\
'
SYNTHESES IN THE TRIPHENYLENE SERIES
72
Until very phenylene series, pared. However, ties of derivativ prepare compounds Several useful me have been issued for the preparati dye intermediates the chemistry of
recently little attention had been paid the tri-
and few derivatives of triphenylene had been pre- following the discovery of the carcinogenic proper- es of 1,2-benzanthracene , an effort was made to
in the triphenylene series isomeric with these, thods of synthesis have resulted. Moreover, patents
within the past few years for several processes on of triphenylene derivatives suitable for use as
Thus, there is evidence of increasing interest in the triphenylene derivatives.
Triphenylene, the parent compound of the series, occurs to the extent of one to three per cent in the chrysene fraction of coal tar. The first useful synthesis of triphenylene was that developed by Mannich in 1907, although the formation of small amounts had previous- ly been reported in the pyrolysis of benzene, and in treatment of bromobenzene .with sodium. Mannich prepared it by heating cyclo- hexanone with a 30$ solution of sulfuric acid in methanol. The re- action is apparently exactly analogous to that in which mesitylene is formed from acetone:
H3SO4
CH3OH
^ f
HsS04 CHaOH
Cu
0
450 - 50(
^ .0 <
/»
*s
L
\
<
II
The product of the condensation, 1, 2, 3,4, 5, 6, 7, 8, 9, 10, II, 12-dodeca- hydrotriphenylene (I), is obtained in a yield of about Sf. By passing the vapors of this product in an atmosphere of carbon dioxide over a copper catalyst at 450-500C, Mannich dehydrogenated it almost quanti- tatively to triphenylene (II). Mannich assigned these compounds their correct structure on the basis of their oxidation to mellitic acid by fuming nitric acid in a sealed tube.
A modification of this method of preparation has recently been patented. Cyclohexanone or 1-cyclohexylidenecycylohexanone is con- densed to the same dodecahydrotriphenylene by heating under pressure in the presence of a dehydration catalyst and a rare earth oxide.
Most of the recent work on the triphenylene series concerns the synthesis of derivatives by ring closure methods which could leave no doubt about the position of substituent groups. Com-pounds to be tested for carcinogenic activity must, of course, be of known
r\
_ o _
structure. Moreover, it was considered desirable to obtain reference compounds which could be used to determine the structures of new com- pounds in this series.
A few general methods have been developed for syntheses of this type. The first of these is that introduced by Bergmann and Blum- Bergmann, and subsequently extended and improved by Fieser and Joshel. Triphenylene , the 1- and 2-methyltriphenylenes, and 1,2- climethyltriphenylene were prepared by the series of reactions shown in flow sheet A, starting with 9-bromophenanthrene. The percentages given under the arrows indicate the yields obtained.
Bachmann and Struve have prepared 1-methyltriphenylene by a similar series of reactions. Their synthesis, however, began with the preparation of £-[9~(l, 2,3,4-tetrahydrophenanthroyl )] propionic acid by a Friedel-Crafts reaction between l,2,354--tetrahydrophenan~ threne and succinic anhyaride. Phenanthrene itself could not be need, since, in the Friedel-Crafts it gives substitution mainly in the 3-position and to a lesser extent in the 2-position, not in the '^••position as desired here.
To prepare 1,4-dimethyltriphenylene, Fieser and Joshel were forced to adopt a different scheme after attempts to add the methyl Grignard reagent to the carbonyl group of |3-(9-phenanthroyl) propionic acid and its methyl ester had failed. The method finally employed is shown in flow sheet B. It involves an unusual aldehyde synthesis.
A novel method of forming the triphenylene skeleton is that investigated by Bergmann and Bergmann. When various 9-vinylphenan- threnes were treated with maleic anhydride, a Diels-Alder reaction occurred. The necessary 1,3-diene system was furnished by the vinyl group of the side chain and the 9,10-bond of the phenanthrene nucleus. Examples are shown in the following equations:
CH— C
X,
II ^0
boil
toluene \^ {U% yield)
\7
- 3 -
I
*s
„0
CH-— C ' + H ^
CH-— C'
boil
V
'A
V
toluene \ (58$ yield)
74
The most recent contribution to the field is the method worked cut by Rapscn. As in Mannich's original synthesis of triphenylene , the starting material was cyclohexanone . This was condensed to 1- c .■/,;:1.chexylidenecyclohexanone in almost quantitative yield when saturated with dry hydrogen chloride and allowed to stand. When this v :'->ciuct is treated with arylmagnesiumbromides, carbinols result Vuich can be cyclized to triphenylene derivatives. The equations in tlow sheet C show the syntheses accomplished by Rapson using this procedure. Yields in the cyclization step seem to have been low in general, although they were not reported in all cases. This scheme, however, involves fewer steps than the other general methods, and gives more convenient access to some derivatives.
Within the last six years a number of processes for making derivatives in this series have appeared in patents, 2-Triphenylene- eulfonic acid is made by the sulfonation of triphenylene in nitro- benzene solution. 2-Hydroxy-3-triphenylenecarboxylic acid is made by treating the alkali salt of 2-hydroxy triphenylene with carbon rdoxide at high temperature and pressure. Various anilides of this acid are made by condensing it with substituted anilines such as the toluidines, the anisidines, and the chloroanilines. Triphenylene- carboxylic amides can be made directly by a Friedel-Crafts reaction between triphenylene and two moles of carbamic chloride. This method can be applied to other compounds having three or more con- densed aromatic rings, or two benzene rings condensed with a hetero- cyclic ring. All of these products have been patented as dye inter- mediate s.
Bibliography:
W.nich, Ber., 40, 159 (1907).
Bergmann and Blum Bergmann, J. Am. Chem. Soc., 59, 1441 (1937).
Fieser and Joshel, ibid. . 61, 2958 (1939).
Bergmann and Bergmann, ibid., 59, 1443 (1937),
Rapson, <T. 'Chem. Soc, 1941. 15.
lei--. Farbenind. A.G,, French Pat. 790,565 (1935); Chem. Abs. 50.
2994 2 (1936). I.G, Farbenind, A.G., German Pat. 654,283 (1937); Chem. Abs., 32,
23722 (1938). 1,0-. Farbenind.A.G., German Pat. 654,715 (1937); Chem. Abs., 32,
36301 (1938). I.G. Farbenind. A.G. , French Pat. 799,598 (1956); Chem. Abs., 50,
7585* (1936). I.G. Farbenind, A.G,, French Pat. 797,072 (1956); Chem. Abs., 30,
6589 7 (1936).
Reported by P. L. Southwick, November 19, 1941.
■
_ 4 - FLOW SHEET A
75
o
/XCH9 zn;Hg \
1\
0)
KC1
0 |
•d |
■H |
•H |
c |
u |
- |
n |
0 |
K, |
o |
£ |
3 |
d |
ULI |
cti |
^CHs (79jg) C02H
/CH2
^CHS
dry HF 0°
VII
/CHs (87#) ^ C02H
Zn-Hg HC1 .
II
III
N'
t
s p,r?o°
«
la'
2200
CH3 MgCl
320
Se
-H20
(42$ yield in I last three steps) CH3
OH
./^
IV
-H'
V
V
76
- 5 - FLOW SHEET B
/X
V
•V
MgBr CH3OCH3CN (66#)
CH< I , CH
./
CHO
III
0 G
k G yy\/ XCH2-OCH;
/NaHSOa <\
KHSO* ^180° \
(58Jg)
HC
V
\
CHgMgCl
(95%) ^CH3
^CHsOCHg
/
II
I
CH3'C03H)3 Pip .ridine
(eof)
CH;
I
CH
CH
II
CH / C03H
IV
W*
H3
y <
CH3
'S
VII
Hs
Adams ' catalyst
/
S 250°
N
-Hi
Uslm hf _^
(B2%)
(. 2202
' -Hs0
(35$ over-all yield in last three steps)
VI
- 6 -
FLOW SHEET C
^~^s
/
<
OH
M/
II
CS2
AICI3
0° (19$
VIII
V\A
0°
/,-
\\R
/*
III
cs2 A1C13
( 29 fo when v R = CH3O)
Pd + C 300o
(quant. )
IX
\'
Pd + C 300° (quant. )
77
(R = CH3 or CH30)
XI
'
HIGH OCTAINE AVIATION FUELS
78
It has been known for many years that hydrocarbons of different structures have varying efficiencies as fuel for the internal combustion engine. Thus, gasoline obtained by direct distillation of petroleum, because of' its tendency to knock, can be used only in engines of low compression ratio. The power output and efficiency of the engine, however, improve as the compression ratio is increased. In :i s'cudy of the antiknock qualities of pure hydrocarbons of the gasoline range, it has been found that highly branded hydrocarbons tire superior fuels. Among the best is isooctane, while n-hepfcane is one of the poorest. In standardizing gasolines the fuel is compared 30 a mixture of isooctane and n-heptane.
The cracking process has brought a great improvement in the auality of gasoline. The unsaturated hydrocarbons produced have a I. rod antiknock rating, and the cracking is accompanied by isomeriza- ticn to more highly branched compounds. New processes for the production of airplane gasolines of very high quality involve the i".1i Liza lion of gases from the cracking units. The first of these \.ec: developed for the production of isooctane. When isobutylene is r.c-Gsed over sulfuric or phosphoric acid supported on inert material it .reacts with itself to give a mixture of isooctenes. These are hy- drogenated in the presence of a catalyst and isooctane is obtained.
."uw Temperature Sulfuric Acid Alkylatlon Process
A valuable modification of this process consists in a similar ureatment of the fraction containing all the four-carbon hydrocarbons. Isobutylene reacts preferentially with the butylenes. At the same time isobutane adds to the butylenes yielding saturated hydrocarbons.
H SO Isobutylene + butylene — ~ — $~* ootenes
Isobutane + butylene HasQ4 > octanes
The last reaction is accompanied bv the formation of higher and lower hydrocarbons (pentanes and decanes). Since n-butane does not enter into the desired reactions, it is converted to isobutane and butylenes by the following processes.
n-butane ^^ ► butylenes + H2
oxide catalysts
n-butane -►- isobutane
A1C13
The commercial development of the sulfuric acid process has gone forward very rapidly and it has been estimated that by the fall of 1940 plants had been installed capable of producing over 7,200,000 barrels per year of 92-94 octane fuel (unleaded) from C4 olefins and isobutane alone.
The entire process may be summarized as follows: With pure hydrocarbons it has been shown that the reaction in the presence of sulfuric acid may be generally applied to the lower olefins except ethylene. Thus propylene, isobutene, 1- and 2-butene , trimethyl- : ethylene, together with lower polymerides of isobutene, di- and
- 2 -
triisobutene, and the butene-isobut.ene copolymer have all been shown to react with isobutane to give good yields of saturated products possessing octane numbers of 90+. Of the paraffins investigated, isobutane, isopentane, and isohexane reacted smoothly although the octane number of the saturated product falls off rapidly with in- creasing length of the carbon chain- used.
The olefin is passed into a well-stirred mixture of the iso- paraffin and 96-97$ sulfuric acid kept at temperatures varying from -10 to 30°C. The overall yield of gasoline to the point where conversion was no longer considered economical in experiments with a ! 4 cut containing 56$ of unsaturated compounds, was 310$ by weight cf the acid used.
High Pressure Hydro ge nation Process
The production of hydrogenated fuels to augment the natural supply has been made possible by the development of new catalysts tor* the high pressure hydrogenation operations. These catalysts possess the property of converting petroleum oils boiling outside the naptha range into lower boiling cyclic and branched chain com- pounds imparting desirable high octane values to the product. The fuels obtained by this method generally possess antiknock qualities superior to those of straight run naphthas.
A typical run gives conversions cf 50-75$ per pass and yields 80-95$ of 75-78 octane gasoline as compared to 74 for natural aviation naphtha. The desirability of hydrogenated fuel is increased by its high octane number response on additions of lead tetraethyl.
In addition to the production of aviation naphthas high pressure hydrogenation is currently employed on a commercial scale for the preparation of blending agents. One of the principal blending agents thus made results from the hydrogenation of octenes available from copolymerization of iso- and n-butylenes. The hydrogenated codimer is not an aviation fuel in itself since it does not possess the boiling point range available for standard engine use. It is, therefore, principally employed as a blending agent to increase the octane number level of natural or hydrogenated aviation naphtha. In this case the increase in octane number is from 84 for the ccdimer to 100 for the hydrogenated polymers.
Mechanisms of Low Temperature Alkylations
McAllister and his coworkers have postulated the following reaction mechanisms for these alkylations:
1. Simple addition of the isoparaffin to the double bond of the olefin. Most of the products obtained can not be explained by this mechanism.
2. Polymerization and depolymerization. The fact that a large proportion of octanes are obtained by the alkylation of isobutane with butylene trimers is evidence for this reaction.
3. Rearrangement of the primary products. This reaction un- doubtedly occurs but not to a large extent under the given conditions.
_ 5 -
80
4. Carbon-carbon cleavage and addition of the fragments to the double bond. Pauling has calculated that it takes less energy to cleave a C-C bond than it does to cleave a C-H linkage, hence this reaction is more plausible than dehydrogenation. The reaction of iso- fcutylene and isobutane can be postulated as follows:
5 ?
C-C-C -+ C-Cr- + c-
c-c-c-C-C
C c
l i C-C-c-c-C
isooctane
Caesar and Francis propose a method whereby the formation of the observed compounds could be easily explained. They state that the olefin is able to wedge itself in between a methyl group and the rest of the isoparaffin so that the methyl group adds to one side of the double bond and the rest of the isoparaffin to the other. The methyl group farthest from the tertiary carbon is the one split off in the case of isopentane. Because it is necessary to use an isoparaffin they suggest as the first step that there is a bonding between the tertiary carbon, or its lone hydrogen atom, and the catalyst.
The following table gives the paraffins found and the manner in which they are explained by this method.
1. Isobutane + ethylene
C
i
C~C-c-c-C
2. Isobutane^* isobutene C c
■C-c-c-<
C-C-c-c
C c C-C-c-c-C
3.
Isopentane + propylene C
C-C- C-c-c-c
C-i-C-'c-c-C
Consideration of the thermodynamic equilibria of isomeric paraffinic hydrocarbons show the following interesting point: "In any group of isomeric paraffins formed by this process," the relative amounts of the isomers agree closely with those computed by thermo- dynamic equilibria when those isomers are excluded which are not permitted by this mechanism." Thus in the case of the hexanes:
81
n-hexane 2-methylpentane 3-methylpentane 2, 2- -dime thy lbutane 2,3-aimetnylbutane
AF (n = |
- 4 - o) |
Eqi P |
jiilib sr Ce |
rium nt |
Per |
Ce |
nt at |
in Alk-.l- ;e |
Calc |
a. |
Found |
||||||
0 |
4 |
0 |
0 |
|||||
-558 |
11 |
26 |
10-25 |
|||||
-558 |
11 |
0 |
0 |
|||||
-1341 |
42 |
0 |
0 |
|||||
-1165 |
32 |
74 |
75-90 |
bibliography
Miirphree, G-ohr, and Brown, Ind, Eng. Chem., 31, 1083 (1939). li-ch, et al., ibid. , 31, 884, 1081 (1939). SSbAllister, et al., J. Org. Chem., 6, 647 (1941). -oar and Francis, Ind. Eng. Chem,, 33, 1426 (1941).
Reported by Q,. Soper November 19, 1941
32
THE MECHANISE FOR THE COUPLING OF DIAZONIUM SALTS WITH AROMATIC AMINES AND PHENOLS
Bartlett and Wistar Hauser and Breslow
The diazo compounds have been a most interesting subject of investigation since their discovery by Griess in 1858. They have been proven to be a most important and useful group of com- pounds.
The structure of these compounds has been a very cont rovers if.:, matter. Griess thought that both nitrogen atoms were attached di- rectly to the benzene ring, but this idea was soon disproven by Kekule, who postulated the structures CeH5N=NX for the salt, and CsH5N=NOH for the free base. Shortly, afterward, Blomstrand' Erlanger, and Strecker independently suggested the identical struc- ture, C6H5N=X, but for different reasons.
X
For a number of years, little attention was devoted to the structure of these compounds, the major interest being in their uses However, the issue was raised again in 1892 by von Pechmann, and for many years a tumultous controversy raged between Hants oh, Bamberger, Angeli, and many others over various suggested struc- tures. The conflict between Hantsch and Bamberger was perhaps the most acrimonious in the history of Organic Chemistry. The matter has not been entirely clarified even at the present time.
The subject is complicated by the several isomeric compounds which are known, either directly or through their derivatives. Although many others have been suggested, supporting evidence exists for the following structures:
(1) Kekule' s dizao formula, ArN=NOH
(2) Blomstrand' s diazonium formula, ArN=N
' X
(3) Hantsch' s stereoisomers modification of Kekule1 s diazo formula
(4) von Pechmann' s nitrosamine formula, ArNHNO
Probably all these formulas are correct under certain conditions. Taylor and Baker summarize the relationship in the following manner:
Mineral Acid
S \ &
x Ar OM Ar Ar^
Diazonium , &~Diazotate Isodiazotate Isodiazo- Nitrosa- Salt hydrate mine
The limited scope of this report precludes a discussion of the evidence for and against the various structures which have been suggested from time to time. For further discussion of this point your attention is directed to the General References listed with ' this report.
'• •.
- .... .t^
■-.-• \ • (T '3
•. •
,, ,-; o. uwuao *>v~ 5*? --•^— *
••■:'',.
K* OKA 3311M* 0i:-MO«i :al»
I- •
.. I -.. d k-i ■» • v."-
',:■-:. 1 ' i
i •
. IS \ O li J. .-I
■
• &»•'•' •
y
'I
• . . r ; .
. . I . .. •■ -
JO I .:. .. It.i
- ■
... I? • '
f ' .V-
. , . .■ i
"■ I.
!-!,'/s
i ■
, J.*?;, j.
• ■■•'
■:., '{-Mr. i'- ;-y- ; '
* C u •*
i
'.'.:■. "■ J . ' ''
r - .
" ; :. ■ j , . ..
.jjil'.tfcj -; '"koaA
J If- J .
'1 'V.
■I :i:'
■in
m*t+ i&
. - 1; ■
83
- 2 -
The primary interest of this discussion is not the possible structure of the diazo compounds, but the mechanism by which one of their most useful reactions occurs. It was discovered in 1870 by Kekule and Hidegh that diazo compounds would unite with sub- stances such as phenols and amines to form stable azo compounds. Many hundreds of these • substances have been and are now used as dyestuff s.
This coupling reaction occurs in weakly acidic or weakly basic solutions. The mechanism by which it occurs has been a source. of uncertainty. It was generally supposed that the coupling took place between undissociated diazohydroxide and the phenol or amin^ The reaction has been usually represented as follows:
ArN=N0H + ArOH -* ArN=NArOH + HOH or ArN=N0H + ArNR3 -► ArN=NArNR3 + HOH
Since the reaction did not go well in strongly acid solutioni it was presumed that the addition of alkali to reduce the acidity simply converted the diazonium salt into the diazohydroxide* and thus promoted the coupling reaction
This point of view does not appear consistent with the modern electronic conceptions of aromatic substitution, which require that the substituting reagent be an electron acceptor attacking a region of high electron availability. It seems more logical that the di- azonium ion, which can readily serve as an electron acceptor is the 2!lw?«-rea?en/ lnJhe paction* Tt -1* known that electron attracting S^inUS?if (SuCh *! the Altro ST™?)* whi<* would increase the g electrophilic character* activate the diazonium component. It is also known that the same substituents, which would reduce the elec- tron donor tendency, deactivate the amine or phenolic component.
The diazonium ion may be represented in two resonance forms,
+
..+
ArjNj-N f ~> Ar:N::N
<*j (B)
tuS°ofh^eS?nanCLf0rm £ Probably contributes largely to the struc- le of the ion it may be assumed that at the approach of an elec- tron-donating molecule, resonance form B becomes the major struc- ture.. Similarly, the activation of the ion by the presence of electron attracting substituents in the aromatic ring can be explained by such a structure. The electron attracting substituents would
bvUILwl6nTah'e f°r? B t0 contribUte ffiore to the ionic structure by drawing the electrons closer to the aromatic nucleus.
It also appears that the free amine rather than the substituted Stv°^ or the phenoxide ion rather than the free phlnol
be mor^TfkPif ?at6r tendenc? for electron donation, and hence vould be more likely to serve as the other active componeAt in the
^onicnf r^ctio4i- Similarly, the presence of electron attracting groups in the aromatic nucleus would obviously reduce the tendency of this component to act as an electron donor*
.
'
84
- 3 -
The most obvious mechanism for the coupling reaction w ould appear to be
<CIJ>^r +H£2^N'(CH3)s "* <C>N=N^^3>=^+(CH3)3
^X
__^>N=N<^3>N(CH3)2 < "H+
Hauser and Br e slow have shown that the diazonium ion is capable of entering directly into the coupling reaction. They usea anhydrous media (pyridine), and observed that phenyldiazonium chloride couples readily with either p-naphthol or sodium ^j-naphtho^ ide to form l-benzeneazonaDhthol-2.
The argument might be made that a molecular oxy-azo compound (analogous to the diazohydroxide in aqueous solution) between the diazonium ion and the (3-naphthol might be formed first, followed hy a reaction between the oxy-azo compound and unchanged |5- naphthol. ' In answer to this possible objection, investigation of the reaction of the nitrogen-azo compound benzenediazopiperidide (analogous to the hypothetical oxy-azo compound postulated ab'ove) with p-naphthol and its anion was made. Coupling did not occur. The nitrogen-azo compound was used because of the difficulty in obtaining the oxy-azo compounds in a pure state.
These results become even more convincing when we consider that addition of pyridinium chloride to the solution of benzene- diazopiperidide and (3~naphthol (or sodium p-naphthoxide) brings about some coupling almost immediately. The pyridinium ion effects the decomposition of the benzenediazopiperidide to form piperidine and benzenediazonium ion, which is then free to couple with the p- naphthol or sodium 6-naphthoxide.
C6K5N=N-NC5H10 CsH5NH > C6H5N=N-NC5H10 -> C6H5NS+ + C5H10NH
The above results seem to show conclusively that the diazonium ion is capable of serving as the active reagent in the coupling re- action, at least in anhydrous media.
Further evidence in favor of this mechanism has been obtained by Bartlett and Wistar, making use of an entirely different method of approach, i.e., from a kinetic study of the reaction. The fact that diazonium coupling with phenols and amines takes place only when the acidity of the solution was kept below a certain value suggested to Conant and Peterson that the rate of coupling is a function of the H+ activity of the solution,, They investigated this point with different diazonium salts and phenols. Heavily buffered dilute solutions of constant ionic concentration were used, and the progress of the reactions was followed colorimetrically. It was found that the reaction is strictly bimolecular and free from complications. Over the pH range investigated, when log k was plotted against pH, curves were obtained which had a slope of 1.
'
-
.
• .
4 -
(1) Diazotized o-anisldine + disociium- 2-naphthol-3-6~di sulfonate
(2) Diazotized o_-anisidine + sodium- 1- naphthol-4-sulfonate
13) Diazotized sulfanilic acid + di sodium 2-naphthol-3-6- disulfonate
(4) Diazotized sulfanilic acid + sodium l-naphthol-4-sulfonate
A probable mechanism was suggested -
=* ArN=iMOH + X"
Figure I
(1) ArN3X + 0H~ *
(Fast and reversible)
(2) ArN=NOH + P qp#t ArN=NP + Hs0
(Relatively slow and irreversible)
The equilibrium in the first step would be a function of the (OH""") activity, which is a direct function of 1.0 pH. The rate controlling reaction (2) would depend directly on the concentration of ArN=N0H, and thus the rate would be a direct function of the pH value of the solution.
It is to be noted that both of the react ants are capable of acid-base equilibria.
p o
log k
3.0
ArNs
and
+ OH ArOH
ArHs0H
ArO~ +
H
However, the phenol equilibrium does not reach appreciable dissocia- tion in the pH region in which coupling occurs, and hence is of little importance in the actual kinetic study.
Recently Bartlett and Wistar have investigated the coupling reaction using amines instead of phenols. Unlike the phenols, the acid-base equilibria of amines reaches dOfo in the pH range where the coupling reaction occurs.
ArNH.
+ H"
a ArNH3+
Hence, this equilibrium must be of practical importance in the re- action, and the curve, log k vs pH, cannot be a straight line, as in the case of phenols.
There are four possibilities for actual reacting components in diazonium salt-amine coupling.
(1) Diazohydroxide couples with the substituted ammonium ion
(2) Diazonium ion couples with the free amine
86
- 5 -
(3) Piazonium ion couples with the substituted ammonium ion
(4) Diazohydr oxide couples with the free amine
In interpreting the results obtained, it is advantageous to consider the general shape of the curves (log K vs pH) resulting if the various possible combinations above are the active reactants. For simplicity, a model case with the following dissociation con- stants can be set up.
For the substituted ammonium ion K = 1 x 10~4
For the diazohydroxide » .& = 1 x 10"3
For the phenol K = 1 x 10~8
7/e can calculate the fractions of the total amine existing as free base and as. substituted ammonium ion in buffer solutions of different pH values* Similarly we can calculate the fraction of the diazonium compound existing at; the diazonium ion and as undissociated diazohydroxide in these same buffers.. Likewise the fractions of phenol and phenoxide ion can be calculated, "Vie re- sults of these model case calculations are tabulated below „
Effect of pH on fraction of each component in acid and basic form
U)
(B)
(C)
(D)
(E) (F)
ArNH2
ArNHa+ + ArNH.
ArNK
+
?!±~3L
ArNH3 '" ■*■ ArNH 3 ArO~
ArOH + ArO ArOH ArOH + ArO""
ArggOH
ArNaOH + ArN3+
ArN:
+
ArNoOH + ArN=+
Referring back to the possible combinations of diazonium com- pounds and amines, if Case (l) represents the facts, then the velocity constant for the rate of coupling v/ill change with pH in proportion to the change in product
(ArN3OH) (ArNH3+)
t;
(ArN3OH) + (ArNp )
(ArNH3+) + (ArNHp)
Corresponding expressions will be obtained for the other three cases. Plots of these products, therefore, indicate the general slope of the Rate-pH curve which should result from each of the eligible mechanisms. The theoretical curves are listed below.
•
- 6
C7
(1) ArN3OH + ArNH3
(2) ArN2+ + ArNHs
(3) ArNs + + ArNH,
(4) ArNsOH + ArNH;
Figure 3
log k values plotted are actually the fraction product illustrated above.
The general shapes of these curves would be the same even if the ionization constants were not the same as in the assumed case, only the points of inflection being shifted. Actually the assumed values are near the constants for the amines and salts actually usei(. To be perfectly accurate, it is necessary to note that these curves are not strictly correct, since the acidic character of the diazohydroxide is neglected. Only the alkaline portion of the curves would be affected, however. Experi- mentally this is of no importance as the pH regions investigated included only those regions where this effect is negligible, i.e., pH range 2.04-6.25. It is to be noted that the curves for mechanisms (l) and (2) are identical with each other in general shape, but are quite different from those of mechanisms (3) and (4).
The experimental work consisted in a determination of the velocity constants at various pH values for the coupling of
(1) 1-Faphthyl amine 4~sulfonic acid with diazotized sulfanilic acid.
(2) 1-Naphthyl amine 8-sulfonic acid with diazotized aniline
Determinations, were made in heavily buffered solutions of constant ionic concentration and temperature, duplicating the conditions used by Conant and Peterson exactly. Progress of the reaction was followed colorimetrically. Results were obtained as shown by the following curves.
M
3 |
Figure 4 |
|
2 |
_^*- — •£ |
|
1 - |
' ^ |
|
0 - |
pH 1 1 1 h- |
-4— |
Figure 5
12 3
. ;
■
- 7 -
88
Figure 4 (Cont)
1-Naphthyl Amine~4**Sulfonic Acid + Diazotized Sulfanilic Acid
Figure 5 (Cont)
1-Naphthyl Amine-8-sulf onic Acid + Diazotized Aniline
(Solid circles are experimental points. Open circles are theoretical superposed values)
Only a portion of the curve is shown in each case as it is obviously impossible to obtain experimental results over the entire pH range. However, this portion is sufficient to establish the general shape.
Of the theoretically eligible mechanisms, it is seen that the curves for mechanisms (l) and (2) are similar to the experimental curves. In fact, by making the slight necessary change in assumed K, the exact coincidence between the theoretical and experimental curves is obtained. This has been done in the graphs above. Mechanisms (o) and (4,) are definitely eliminated from consideration, although (4) has been usually considered as the most probable.
In deciding between mechanisms (l) (Diazohydroxide coupling with substituted , ammonium ion) and mechanism (2) (Diazonium ion coupling with free amine), there appears to be little doubt that mechanism (2) is the correct one. In those cases where aromatic substitutions have been carried out on anilinium ions, these ions have proven both unreactive and powerfully meta directing. It appears most unlikely that such a substance could act as an active intermediate in a reaction which fails to occur except with highly activated benzene derivatives, and which always results in para or ortho substitution. On the other hand, mechanism (2) is perfectly in accord with the modern electronic theories of substitution.
The phenol coupling reactions studied by Conant and Peterson are subject to the same analysis. Theoretical curves are obtained as follows:
Figure 6
-4
M -8
ho o
rH
-12 --
-16 --
(5) ArN3 + ArOH
(6) ArN2+ + ArO~
(7)ArN20H + ArOH
(8) ArNaOH + ArO~
log k is actually the fraction product of the reacting pairs
PH
89
Although for obvious reasons, complete experimental data is lacking, it is seen that only curves (6) and (7) have any por- tion that is linear with a slope of 1 as observed. The choice here is between the reactive pairs, diazohydroxide and phenol, or diazoni- um ion and phenoxide ion. On theoretical grounds, the second mech- anism is to be preferred here also. Certainly it is auite definitely shown that the results of Conant and Peterson are not incompatible with the postulated mechanism.
Since two entirely independent and different lines of approach to the problem have led to the same conclusion, and since this conclusion is thoroughly in accord with modern electronic theory, it appears to be definitely established that the dlazonium ion and the free amine or phenoxide ion are the active agents in the dlazonium coupling reaction.
Bibliography:
Hauser and Br e slow, J. Am. Chem. Soc, 63, 418 (1941 ), Wistar and Bartlett, ibid. , 63, 413 (i94l'jr Conant and Peterson, ibid. , 52, 1220 (193 o) . Hauser and Ereslow, ibid., 62, 2389 {l940), Hammett, "Physical Organic Chemistry", p. 314
General References:
Saunders, "The Aromatic Diazb Compounds"
Taylor and Baker, "Sidgwick's Organic Chemistry of Nitrogen", Chapter XIII
Reported by W. E. Blackburn November 26, 1941
A REVIEW OF THE ORGANIC CHEMISTRY OF ARSENIC
Since there are no naturally occurring organic arsenic com- pounds which might serve as starting material in a synthesis, one is immediately faced with the problem of preparing a compound with the desired 3~As linkage. With this end in view a large number of arsonation reactions have been attempted with varying success. Of these many reactions only a few with general applicability can be picked out,.
I. Methods of Arsonation.
A0 Aliphatic
Cahours and Richie in 1854 found that by heating an alkyl halide, usually the iodide, with sodium arsenide they obtained a mixture of arsinesc
RI + Na3As — > R2As-AsR3 + R3As 4 R4As I"
This reaction was important in pointing out the relationship between the various types of arsines then known, but is of little preparative value because of the mixture obtained. In 1859 Cahours announced an improvement of the synthesis.
RI + Zn3As3 -» R4As+IcZnIa •—y™^ R4As+I~ ^|^» R3As + RI The cadmium salts could also be used,
Michaelis and Reese in 1882 advanced another more versatile method for preparing these arsines „
3RX + bNa + AsX3 -» R3As + 6NaX
Using the following types of arsine halides, which are available from other reactions, RAsX3 and RR;AeX, one can obtain tertiary arsines of the type RAsR2 and R'R* rR: f "As respectively.
The most widely used synthetic method which is also capable of industrial application is the Meyer reaction.
As-ONa + RI -> Nal + RAS-ONa ^ONa ^ONa
Meyer carried the reaction out in an aqueous medium usually under pressure to avoid the loss of the alkyl iodide being used. Dehn extended this reaction and found that he could improve the yields by using dilute alcohol solutions and potassium rather than sodium arsenite. He thus obtained an homogenous reaction medium.
The procedure as modified by Dehn still left much to be de- sired, because the higher homologs were obtained in increasingly small yields and because a considerable amount of alkyl iodide was lost due to ether formation with the solvent alcohol. Valeur and Delaby found, in contrast to Dehn, that they obtained better re- sults with aqueous solutions which allowed better agitation and re- frigeration. Adams and Quick selected this method for preparing higher aliphatic arsonic acid and found that better yields could be obtained by the use of alkyl bromides or chlorides. The sodium salts of the arsonic acids produced are soluble in the reaction
.
.
.
- 2 - SI
mixture and difficult to isolate. When the alkyl chloride or brom- ide was used these sodium salts could be converted to tie free acids which are much less soluble. When the iodides were used the hydrogen iodide immediately reduced the ar sonic acids.
The reaction can be extended to further alkylation as follows:
RAs-ONa + R'l -* RR'As-ONa + Nal xONa
RR'As-ONa + R"I -> RR'RnAs=0 + Nal
Still another method of importance because of its application especially to the production of lethal compounds of military value is the addition of arsenic halides to acetylenes.
R-C=C-RT + AsCl3 -> R-C«
CI .
*C-R!
ASCI;
The reaction has also been carried out on ethylene
B.. Aromatic.
The oldest method of aronation in the aromatic series is that of Bechamp.
^>NH3+As04Hs- 19°-S°°° ■ NHs<^/>As03Hs
The reaction can be carried out on nuclear substituted amines with an open £ or o position. If both the c_ and the jo positions to the amine group are filled the reaction fails.
By far the most important method of ar.omatic arsonation is the Bart reaction or one of its various modifications.
<^ ^-N-NOH + K3As03 Ka° i <^ "\as03K3 + N2 f + KOH
According to Bart's original procedure a solution of the di-
azonium salt was prepared and then made alkaline and heated to a
rather high temperature, then with continued heating the arsenite
solution was poured into the diazo solution. He later suggested
the use of copper, nickel, silver, and cobalt and their salts as catalysts.
Since Bart's original procedure appeared it has undergone cany modifications in the hands of other workers. Schmidt suggests the use of a neutral or even faintly acid medium with no catalyst, Mouneyrat uses alkaline, neutral, or acid reaction medium with a dual catalyst consisting of a metallic salt and reducing agent chosen with reference to the pH of the solution.
No blanket statement can be made in regard to the best con- ditions for carrying out the Bart reaction. The choice in any given case must be made after experimentation with the particular
,
-
- 3 -
amine under consideration. Even then the successful completion of the reaction is dependent on the technique employed so that two operators have difficulty in reproducing yields.
As in the case of the Meyer reaction the higher arylated arsinic acids and arsine oxides can be prepared by proper choice of starting material*
92
^> N=NOK
J— *s
N=NOK
K2O3As0
KO-As-02
y0
02As-OK + N8
03As=O + N2
Chatt and Mann recently developed
iodides of the type As (abed) stereochemically.
AICI3 -*
These
1 + |
AsCl3 + |
V |
|
AsCl2 + |
I |
V |
|
RR'AsCl |
+ |
V |
KI
a method for making arsonium compounds are of interest
+ T~
04AsTI
KI + _ + AICI3 -» — * 03RAs I
+ AlClo ~*
KI
+T-
* RR'02As I
R,R,,RMAs +
A
Br
+ A1C1:
KI
RR"R"0As+I~
V
Many attempts at direct arsonation of aromatic compounds by the Meyer reaction have been made. Rosenmund was successful in isolating benzene arsonic acid from bromobenzene and aqueous po- tassium arsenite. Hamilton and Leudeman confirmed this report and attempted to extend the reaction. They found that c_-chlorobenzene- arsonic acid could be arsonated in aqueous medium to arsonophthalic acid in good yields but that other compounds with equally active halogen gave poor yields or none at all. It thus appears that direct arsonation is not a general reaction in the aromatic series.
II. Transformations in the Arsenic Series
The presence of arsenic in an aromatic nucleus does not materially alter the aromaticity of the ring. The arsonic acid group has about the same m orienting effect as the sulfonic acid
- 4 - 93
group and activates the ortho and to a lesser extent the para positions. p_-Chlorobenzenearsonic acid is sufficiently reactive to react with aromatic and aliphatic amines and with alcohols i The amino arsonic acids are of especial synthetic value since they undergo all of the characteristic aromatic amine reactions. The presence of an amine group or hydroxy group in the para position seems to weaken the C-As linkage and in this case the use of halogen containing reagents must be avoided since by some of them (i.e., PC13, SOsClg, S0C12, CI, Br) the arsonic acid group is cleaved from the ring being replaced by halogen.
There are numerous derivatives of arsenic known due to its existence in multivalent states. The arsonic acids are the most important derivatives since they are most readily prepared. Since interconversion in the series is quite easy, all of the other derivatives may be prepared from the arsonic acids. There are note- worthy differences between the aliphatic and aromatic arsenicals of the same class, but they are differences, as a rule, of stability and degree rather than of kind.
The types of arsenic compounds, their nomenclature, and their interconversion are summarized in Tables I and II.
III. A Comparison of Nitrogen and Arsenic Chemistry
As can be seen from Tables I and II, there are several types of compounds similar in structure to well known nitrogen compounds. There are also many arsenic compounds whose nitrogen analog is not known and vice versa.
Arsenic compounds react very readily v/ith halogens to give haloarsenic compounds; a reaction not observed in the nitrogen series. This is to be expected since inorganic arsenic halides are more stable and easier to prepare than nitrogen chlorides. The ability of arsenic compounds to become oxidized to a higher valence state is also more pronounced than in the nitrogen compounds. In- deed, with certain arsenic compounds their air oxidation occurs so frapidly that they are spontaneously combustible.
A study of the reactions and properties show that compounds of similar structural formula often show fundamental differences in their chemistry.
The arsonic acids most closely resemble the nitro compounds. The reactions due to the acidic hydrogens of arsonic acids, of course, have no parallel in the nitro compounds. The electronic arrangement of nitro compounds is suitable for the formation of a dibasic acid and the inability to do so has been attributed to the small atomic size of the nitrogen atom which will not permit it to coordinate the required number of groups. Arsonic acids can exis'-j in the anhydride form RAs02, but unlike the nitro compounds, these are hydrolyzed by water to the arsonic acid. The arsonic acid group has no chromophonic properties, all the arsonic acids being white solids. Like the nitro group, the arsono group exerts a m directing effect on orientation and like the nitro compounds the arsonic acids can be reduced to any desired lower valence state by the choice of suitable reagents.
c u
CD Eh
P C
CD rH Cd > •H
EH
p G (D
rH Cd >
cd P
d
CD PM
rH O CO <
I
•H
rH <&
>> <-
is! CD CD
•H vH CO
cd
J
o
co m
rH-H
cd co •H ^ fH cx3
p
X
o
o
II
co
< n
isi CD
H CO
cd >H Tj
•H CO -H
Sh rH X
P cd O
X
PS
>i-h w
-5-.P O
•H
«3
cd
CD P
cd a?
1
x
CQ
s
•H C
o
CO
CO
J-3
rH
co
S4
H
co H
cd o
•H
c
CD CO
u <
W PQ o
S
•H
P
cd Xi ft •H
P CJ CD
H
cd >
•H Sh Eh
73777* /y/./aT"
rH
cd
•H !h
Ph
+^ CJ 0)
rH
Co >
cd p C5 CD
w
cd
co
<«J
I
O
I
CO
<
CO
Pd
I
S-(CD
cdT3
•H rHX
rH < O
rHCD W Cd£ w-»-H «C>CQ
O
r~
7^777"
CD H TJ
>s CD CD >H
- o .yc-dc
CO
ro
rH «H »H cd
Cd tO rH >/
•H Si «i o
cv
rH
O
ft; 'O ajx!
H
pc ^ a?
CO rH £
<t; cd -h
I cd co
co U U
< p aj
0> CD -H
ft! Eh ^d
V CO
pc;
>3 CD rH -H
cd to
•H fs
•a cd
H O
co
CD I CD
C cdrd
•H Sh-H co+3rH £h CDcd
cd pjcJ o
CO
H
■m |
n |
|||||||||||||||||
c: cd |
, 1 |
U PQ |
rH o ^ 1 |
M |
~6- |
J |
||||||||||||
rH |
/*" |
\ ! |
||||||||||||||||
Cti > •H |
As iary sine |
- < |
/ |
/ |
,« |
|||||||||||||
\ |
||||||||||||||||||
SH |
\ |
|||||||||||||||||
Eh |
CO C ^ pc{ 4^> cd |
H |
||||||||||||||||
>> |
— X |
W |
>s \ \, |
|||||||||||||||
Sh |
7* |
/^ |
u \ <$ |
|||||||||||||||
03 |
^/ |
f |
\ |
§ \ , |
||||||||||||||
0) |
/y |
v\-r |
CD |
o V+« |
||||||||||||||
Eh |
/ |
•H |
cd co |
|||||||||||||||
+J |
i |
■ CO |
r$ <C |
|||||||||||||||
S |
rH |
*\ |
" ?-i |
C3? 'f |
||||||||||||||
CD |
cd |
o |
cd |
oc; |
||||||||||||||
rH |
t |
ra |
CD |
|||||||||||||||
Cd |
re r- |
m |
— ^ O |
rH |
t* |
|||||||||||||
> |
rH |
>s'd CD |
>> |
H |
||||||||||||||
cd |
o |
rH tf |
II |
C |
X |
|||||||||||||
•p |
CO |
cd CD-H |
rH |
CO |
cd |
o |
||||||||||||
CD |
CO |
•H CjrH Sh-h cd |
o |
-•4 10 |
•H Jh |
x |
||||||||||||
\ |
* |
|||||||||||||||||
Ph |
Ph |
4J crj£j <; |
oc; |
4^> |
||||||||||||||
CD |
rH |
|||||||||||||||||
CJ |
CO |
o |
||||||||||||||||
•H |
CD |
•^ |
||||||||||||||||
4-3 |
re |
CO SH CD |
i >> |
1 rH |
C c» |
V |
A |
|||||||||||
c rt |
CTJ<d |
a tn |
u o |
OJrH |
NJ Ph |
|||||||||||||
CD |
M |
•H |
o |
cd u: |
rH |
TJ |
CO |
*o |
||||||||||
rH <! |
rH X |
CO rH |
o hx; |
<C |
k |
"-*-' |
||||||||||||
a5 O |
>>o |
< >>>> |
CO |
>S |
1 |
CO |
||||||||||||
CO |
> |
TJ |
,H |
re rn |
X CD K |
<n |
^H CD |
aj |
<d |
.") |
||||||||
rH |
5 <c |
cd |
OS cd |
o c: o |
re |
cd a |
<rj |
re |
. J |
|||||||||
Cd |
u |
0J |
•H |
•H |
c-h ,w |
# |
■H ^H |
re |
• — ■■ |
CD |
||||||||
O r* CD |
En 0d t3 |
TJ "d CO \ HO* |
_/ |
rj CO |
7 |
13^ |
Cc |
|||||||||||
cd |
s/ |
r |
||||||||||||||||
Ti |
« |
o/ |
»"3 |
|||||||||||||||
o •H |
Secon ent |
0 |
CD rH CD |
cr. |
||||||||||||||
rH |
M«tj |
' 1 |
||||||||||||||||
o |
cd |
Ch-H |
1 |
|||||||||||||||
> crj 4-3 a |
CO |
cdrH |
°S |
o — — |
||||||||||||||
rH rH£i O >y,H |
^1 < - |
|||||||||||||||||
CO N |
||||||||||||||||||
CD |
CO |
rHrH |
«3 |
|||||||||||||||
Ph |
< re . Ph r |
cdn-3 H |
re Ph |
|||||||||||||||
X |
||||||||||||||||||
tto if n pi * q |
1/ c\. ^~ |
|||||||||||||||||
+i |
""*" |
CD |
/DM |
* |
J.' t/ |
|||||||||||||
E5 CD rH ^ |
13 1 -H U X |
l C, CO |
\ 1 1 o "* ?H -H rO |
<x |
CO CD |
0 |
||||||||||||
cd |
cd o |
W K |
cd ;3 |
re c |
T5 rd -H W |
a |
O CJ |
■ |
||||||||||
> c |
> |
O O |
o |
rH |
M z rH |
.<3 |
A |
CJ CD |
re p^ |
' ^) |
>' ! |
|||||||
•H 1 |
1 |
rH CD |
N 1 |
rH -Hnd |
iD |
1 , O rH CD O ^ |
1 |
CD Nl |
w |
-T |
^r- i |
|||||||
C c |
a |
t» c |
& |
>> CJ-H |
-> CO J |
>)drl |
a |
) |
CO CJ |
CO |
ry- |
l_ i |
||||||
Eh < P |
S |
in -H cd CO |
C, CD.O cd co cd |
/ r |
>/ pq cd co o .* |
-j PC |
i |
?H CD cd,o |
.-s ^ |
ri i |
||||||||
Sr)0 |
f |
|||||||||||||||||
£ |
"*"^ |
■— |
*-.-— . - |
i |
i? |
t s^ |
BoV>* |
,! |
||||||||||
c(7 J |
r?7 |
~j , |
||||||||||||||||
>i |
> |
* |
^-/ |
c |
X |
r |
-•T-i |
|||||||||||
cd 6 •H |
X. |
M V |
kj X |
^ |
o -r cx |
J ■0 ..r *^> — |
"I |
|||||||||||
Ph 4-> |
' / |
V |
^ |
o |
r 0 V / |
<v |
r -a r p •J |
41 |
i , i |
|||||||||
CD |
CD 1 |
^N |
•v |
/ |
s |
r> |
^7 |
|||||||||||
rH |
a 5UCD |
s |
0C |
>w |
t£-\. ' |
o i |
||||||||||||
cd |
•h or5 |
( |
^0 |
i |
||||||||||||||
> |
<* |
CO HH |
., |
3V |
1 |
> i |
||||||||||||
cd |
rH o |
cd o |
CO |
2 |
0 ' |
O \ |
||||||||||||
CD |
CO |
cd |
O - o |
^1 / |
rH Orrj |
w^ |
n |
^ ! |
||||||||||
~ X |
=<. |
* |
||||||||||||||||
Ph |
Ph |
>.-^> |
re |
g |
>5 C0.H |
<r |
>u |
|||||||||||
r-< cd cd 4J |
^ |
u u o . |
ry |
oz |
I |
|||||||||||||
cd cd cd |
S5
96
The arsines bear little chemical resemblance to the amines. They are nearly devoid of basic properties. Methylarsine and the tertiary arsines are basic enough to form salts with halogen acids, but these salts are completely decomposed by water. The tendency to oxidize increases as the number of organic groups is increased. Arsine itself is stable to air oxidation, the mono organic arsines are fuming liquids which must be protected from air to preserve them, and the di-organo arsines are spontaneously combustible. The tendency to air oxidation is less in the aromatic series.
On the other hand, the arsonium compounds bear a striking resemblance to the ammonium compounds. The salts are highly ionized, and the arsonium hydroxides are strong bases absorbing water and carbon dioxide from the air. The arsonium iodides are not decomposed by aqueous alkali, but must be distilled from potassium hydroxide to decompose them.
The areeno compounds are quite dissimilar to the azo compounds. The arseno group is not chromophoric although some arseno compounds are colored. Oxidation of the arseno compounds occurs under very mild conditions, silver oxide being reduced by an arseno compound,, The comparative instability of t^e arseno group is shown by the fact that halogens, sulfur and hydrogen cleave the As=As linkage and by the fact that heating a solution of two different arseno compounds causes an exchange reaction.
RAs=AsR + R'As=AsR' v=^ 2RAs=AsRf
Likewise, the arsine oxides bear only structural resemblances to the nitroso compounds. The arsine oxides are amphoteric, giving with halogen acids the dihalogen arsines, and with bases the salts of hydroxy arsines R2As-ONa or arsenous acids RAs(ONa)s. In the aliphatic series the acids corresponding to these salts cannot be isolated, but esters can be prepared with an alkyl halide under suitable conditions. In the aromatic series the arsenious acids. and hydroxy arsines are not stable unless nuclear nitro or carboxyl groups are present. In the latter event the arsine oxides may be hard to obtain.
Compounds having a formal resemblance to the dia 20, diazonium, hydrazo and azcxy groups and to the hydrazines are not known in the arsenic series.
In general it may be said that analogous reasoning from the nitrogen compound to its arsenic analog is not justified, although significant resemblances can be found.
IV. The Stereochemistry of Trivalent Arsenic
To date the best physical evidence indicates that trivalent arsenic has a rather flattened pyramidal configuration with the arsenic atom at the apex. The bond angles are calculated to be 100° + 3°. Recent chemical evidence seems to support this view.
Lesslie and Turner have resolved compounds of type I and found . them to be very stable to racemization.
97
- 8 -
w
CaHBI
H
\Aas\^C°*
1 CH3 also C3H5
The optical activity (a) The three rings to the asymmetry of behind the plane.)* the rings would have asymmetry of the mol activity. (c) The "asymmetric arsenic in the absence of an
0
r^V V%
II s N
CH3 C2H5
-COaH
of these compounds could be due to three causes: could be coplanar with the optical activity due the molecule (i.e., methyl group in front or
(b) If arsenic has a pyramidal configuration
to be folded along the 0 to As axis. The ecule would still be the cause of optical
optical activity could be due to an atom" which would impart optical activity even
asymmetric molecule.
Kamai attempted the resolution of compounds of the type R-A»-Rf-C03H but failed. This negative evidence, of course, does
R"
not rule out the third possibility. The evidence in favor of (b) is rather meager. Lesslie and Turner found that the optical isomers of I were very stable to racemization, being only slowly racemizel in boiling benzyl alcohol. However, in presence of an alkyl iodide
ethyl alcohol in the racemization was rapid. Treatment
of
with ethyl iodide would produce II in which the arsenic atom, being no longer trivalent, would probably not have the pyramidal configura- tion. If II has a tetrahedral arsenic atom, then it should be resolvable, but all attempts at resolution have met with failure c
Chatt and Mann have advanced interesting evidence in favor o f the pyramidal arsinic atom. They synthesized 9, 10 di-j>- tolylarsanthrene (III, IV, or V) and found it to consist of a mix- ture of two forms which they separated by fractional crystallization, Assuming the pyramidal construction the following three compounds are indicated.
f I *
k<
in
IV
The left hand ring is projecting from the plane of the paper toward the observer and a dotted line to the ]>-tolyl group indi- cates that it extends behind the plane of the paper and toward the center of the molecule. From an examination of the model for IV it is seen that the two p,-tolyl groups are almost coincident in space. They offer this as an explanation of their failure to obtav a third form.
.
-
S3
- 9 -
It thus appears that trivalent arsenic does have the pyramidal form. However, this pyramidal form might be stabilized by the fused ring systems present in I, III, and IV. The evidence on hand points to the fact that I is resolvable due to the asymmetry of the molecule but does not yet prove or disprove the existence of an "asymmetric arsenic atom."
Bibliography
For coverage of literature to 1923 see - Raiziss and Gavron, J. Am, Chem. Soc. Monograph. . Lesslie and Turner, J. Chem. Soc, 1954^ 1170; 1955. 1051;
1956. 750, Chatt and Mann, J. Chem. Soc, 1940. 1192. Hamilton and Leudeman, J. Am. Chem. Soc, 52, 5284 (1930). Kamai, Ber., 68B, 960 (1935); 68B, 1893 (1935). Tatum and Cooper, J. Pharmacol., 50, 198 (1934). Eagle, ibid.. 66, 423 (1939).
Reported by C. W. Theobald November 26, 1941
99
ADDITION PRODUCTS BETWEEN KETENES AND UNSATURATED HYDROCARBONS
Staudinger prepared several addition products between ketenesr and unsaturated hydrocarbons and suggested 'chat they were substi- tuted cyclobutanones. In the past few years the structure of the c,e addition products has been definitely established.
Dipheny.lketene and Cyclopentadiene. — Staudinger condensed
and cyclopentadiene and obtained a dicyclic unsat-
i&lble st.
dime thy Ike tene
urrted ketone, C9H130. He suggested I and II as two poo
tures, where R=CH3. After unsuccessfully attempting to character
ize the degradation products of Staudinger' s compound, Simonsen
studied the compound in which R=C6H5 jrnd proved the structure to
be I.
CH
' V.
RoC-CH CH
0=C-CH-CH2
permono-
phtna acid
lie 0 CH""/
CH2
/ \
R3C-CH CH
i ! I
0=C-CH-CH II
CHOH 02CH-CH CHOH
glycol
alkali
H > \ 02C-CH CH
III
0=C-CH-CH2
fusion
^ H03C-CH-CH2
a and p
III
CH3C00H>dlecetfte H3S04 03CH-CHCOOH
a and p forms CH2COOH
CHCOOH
:2«
KMnO,
J
lead
tetra
acetate
IV
The a and p forms of IV were synthesized by the hydrolysis of the condensation product of bromodiphenylmethane and methyl sodio-propane-a,a, P,X -tetracarboxylete, one of the carboxyl groups being eliminated.
CfiH
6rl5
^6^5
CH Br
COOCH.
Na I -C i
-CH-CH2COOCH.
COOCH3 COOCH3
HC1
■CO.
03CH-CHCOOH CHCOOH
CH2COOH
Dlpheny Ike tene and Styrene. — Structures A and B were suggested by Staudinger for the condensation product of diphenylketene and styrene. Hydrolytic fission (NaOH, H20) of the adduct gave an acid, C22H2002, which he suggested was a, a,^ -triphenylbutyric acid (I) or a, a, p-triphenylbutyric acid (II).
03C- C=0
I I
> 0CH-CH2
02C-CO2H I
H2C-CH3
i 11
02C-
-C=0
CH2-CH0
B
02C-CO2H
I
HC-CH20
.
-2-
r
100
Both of these acids were synthesized by Bergmann and found to differ from Staudinger's acid.
Addition of the ketene might involve reaction at the nuclear double bond (as with benzophenonephenyllmj.de end phenylnagne&iwp bromide, and ethyl azodicarboxylate and styrene derivative s) to give compounds like:
0CH
'c=o
CH;
/
CH0
Bergmann prepared a series of compounds which would result hydrolytic fission of the above rnd analagously formed addition products; none Was identical with Staudinger's acid.
b;
Simonsen in his work on the adduct with cyclopentadiene had found that hydrolytic fission occurred between the diphenyl carbon and the carbonyl group, A similar fission of A would give the acid III. This acid was synthesized by Bergmann and found to be identical with the acid derived from the addition product of di- phenylketene and styrene, thus establishing A as the structure o.p the addition product.
0SCHOCH3 + Na
08CHNa A 0CK=CHCO3CHS
02CH C02H 0-Ch— CH2
III
Structure B was also ruled out by the consideration that it would not account for the isomer! zation of the original adduct to C6H5CH=CHCCH(CeH5)3. 0
Using a different method of degradation Smith confirmed Simonsen' s structure for the adduct with cyclopentadiene. Smith also found that diphenylketene reacted with cyclohexadiene and with 2, 3-dimethylbutadiene. However, 1,4-dimethylbutadiene did not react and neither did cyclopentadiene react with ketene.
Farmer and Farooq found that diphenylketene reacted with cyclohexene, cyclopentadiene, cyclohexadiene, dimethylbutadiene and pioerylene. However tetramethylethylene would not react,
Diphenylketene and Phenylacetylene. — In an attempt to obtain further evidence thr t styrene and diphenylketene reacted to give cyclobutanone derivatives Agre and Smith tried the reaction be- tween diphenylketene and phenylacetylene. The compound obtained was soluble in methanolic potassium hydroxide and gave a positive phenol test (Folin). The G-rignsrd machine showed one active hydrogen and no addition of reagent. The principal reactions
-3-
101
are below: OH
N=NC6H4S03Na
0 KaCa<807/ \0
f ^Pb_(oAc)4 o J&
COOH
"+ 0COOK acetrte and /^ ^ >COC6H5 meth^ ether j l|
^
NrOAc (AcO)20
II
acetate
and methyl ether
OAc
OA: 0
0
All the reactions were adequately explained by structure I, but the reactions would be just as consistent with the l^-naphthol
Following the lead of earlier work by Franssen, Smith and Hoehn synthesized I by the addition of 1, 4-naphthoquinone to an excess of 5 moles of phenyl magnesium bromide.
/VS
0MgBr
oxime
COOK
+ 0COOH COC6Hs
Because of the unexpected n.-ture of the addition product an attempt was made to determine the mechanism of the reaction. The first step was to determine the location of the aryl group from the acetylene. The reaction between p_-tolyl acetylene and di;Dhenyl-
-4-
10;
ketene indicated that the aryl group from the acetylene goes to the 3-position in the naphthalene.
OK
alkaline
Jc6h5ch3(d)
-^.
KMnO,
COOK
ICrHi
+
x' 0
COOK
COOK
Diphenylketene reacts with many compounds with active hydrogen to give compounds containing the diphenylacetyl group. This might be the first step in the reaction. The compound which would be formed if this were the first step, 02CKCC=C0, was prepared by the
0
reaction between 0CsCMgBr and 02C=C=O, Attempts to cyclize the phenyldiphenylacetylacetylene with fused ZnCl2 in acetic acid failed. This eliminated this compound as a possible intermediate o
The fact that the dieubstituted acetylene reacted in tne same manner as the monosubstituted acetylene indicated that the active hydrogen of the phenyl acetylene was not necessary to the reaction.
OK
00^30 + 02C=C=O
vV0
| 02C=C=O
The addition between diphenylketene and phenylacetylene re- sembles the Diels-Alder reaction in that no catalyst is required gnd that elevated temperatures are not needed. The diphenylketene supplies a 1,4 conjugated system, if a double bond of a ring is included. But the diene synthesis could only lead to p-naphthols.
^\ *CH
-Vv^0
c-
C6HB
-5-
103
The work of SUauci.ing.er, Simons en, Farooq, end Bergmann had demonstrated that ketenes react with ethylenic compounds to give substituted cyclobutanone s. It is therefore quite probable that a cyclobutanone is an intermedir te of the reaction with acetylenes.
+ CH
0
\
'CH
CC6H5
0
^
'or
/
CH II
°*L^
H
H
0
//
.*/
x~
The cyclobutanone is the first and most important inter- mediate. It explains the formation of an cc-naphthol, why the rcetylenic aryl group goes to the 3 position, why both mono- and diary lacetylenes give analagous products. The bond which is bro] in subsequent steps is the same one which breads upon hydro lytic fission of the adducts of diphenylketene and ethylenes.
Attempts to obtain direct evidence for the existence of the cyclobutanone intermediate have been unsuccessful.
.en
Bibliography
Lewis, Raraage, Simonsen and Wainwright, J. Chem . Soc . , 1837 (1937).
Farmer and Farooq, Chem. pnv Ind. , 1079 (1937 ).
Bergmann anc"! Bergmann, J. Chem. Soc., 727 (19^8).
Smith, Agre, Leekley anc1 Pri chare1, J. Am. Chem. Soc, 61, 7 (1939).
Smith anc1 Hoehn, ibir1 . , 61, 2619, (1939); 63, 11*75 (1941).
Reported by W, H, Kaplan December 3, 1941
104
SOLVOLYTIC REACTION MECHANISM
Hammett-Columbiaj Ingold, Hughes-University College, London;
Bartlett-Harvard
Second-order replacement reactions by (a) anions or (b) electron donating neutral reagents are quite well established as occurring by a bimolecular inversion mechanism* This Lewis-London- Polanyi-Olson mechanism explains the Walden inversion.
yH3 CH3
(a) I* + C6H13-CHI ^zr±. C6H13CHI* + i"
(b) NH3 + RC1 -> RNH3+ + Cl~ (Menschutkln )
It is almost equally well established that not all anion re- placement reactions occur by this mechanism. For example: cc- phenylethyl chloride is hydrolyzed to the alcohol at a rate inde- pendent of the hydroxyl ion concentration, converted to the ethyl ether by alcoholysis at a rate independent of the ethoxyl ion con- centration, and to the acetate in glacial acetic acid at a rate independent of the acetate ion concentration. ■
Such first-order replacements occur in solvents such as water, alcohols, phenols, carboxylic acids, and sulfur dioxide and the reaction rate is influenced by the solvent concentration when a diluent is used.
To explain this first-order (in the presence of large excess solvent) replacement, Ward in 1937 proposed a solvolytic ioniza- tion mechanism, which Hammett and Ingold have accepted with modifications. It is unlike the bimolecul:-r mechanism where the energy to eliminate the departing anion is partly furnished from the energy of the bond forming with the simultaneously entering anion.
In the solvolytic mechanism the rate-determining and energy- providing step is the solvation with insipient ionization of the anion-formlng part of the molecule. The electron donating or
R R+
Step 1. R-CX ^± RC X" (solvated) slow
R R
R R
Step 2. Y" + RC — X -* YCR + x" rapid
R R
"nucleophllic" reagent Y" is not involved in the rate-determining step 1.
' !
it
■ . . f j| a
;•• ;•'.
-2-
105
The hypothesis that step 2 occurs with the anion being removed not over 1 A away explains why inversion of configuration occurs with large but incomplete racemization when optically active com- pounds are being substituted on the asymmetric carbon.
The ionization mechanism will be favored relatively tft the bimolecular mechanism by large electron-release from R, strong electron-affinity in X, low nucleophilic activity (basicity) and low concentration of Y, and high ionizing capacity of the solvent* Such a mechanism is not possible when the group is held very firmly as in aryl hs.lides and applies essentially to substitution at a saturated carbon atom and in solution.
Primary halides tend to react by the bimolecular mechanism; secondary and tertiary, by the ionization mechanism. The "mecha- nistic critical point" is illustrated in work of Hughes and Ingold on the decomposition of sulfonium hydroxides and salts in water at 100°.
.+,
OH" + RS R3» (CH3)3S+ -JL
ROH + RS'S CH30H second-order
( C2H5 )3S
+
ki
-^ C3H50H
CH3S(C3H,-i)8 _E3^C3H70H (CH3)2£c4H9-t -^±_>C4H90H
second-order
first-order
first-order
ki = 9 ks
k4 = 2600 k3
Trimethyl sulfonium salts were decomposed in alcohol
basicity
anions: .OH > 0C6H5
decreasing rate, second -order
first-order and equal rate
In addition to the three characteristics illustrated, (l) an ionization favoring solvent, (2) a first-order reaction rate independent of the concentration of the replacing ion, (3) en inductive effect in reverse to that of bimolecular replacement, the solvolytic mechanism can be diagnosed stereochemically.
d-sec-octyl bromide IN NaOH 80
60f EtOH acid
\ j±.
\!/
^-alcohol 65^ racemized
1-alcohol lesa than
racemized (probably due to secondary
reaction )
-3-
ICS
Some of the oth^r evidence for the mechanism is as follows:
1. In a mixed solvolysls experiment a linear rate dependency on water was observed in the alcoholysis of benzhydryl chloride; however; it was not a case of two superimposed bimolecular reactions because less than 1/4 the theoretical proportion (calculated from the relative rates) of benzhydrol to benz- hydrylethyl ether wps formed. The water facilitated ioniza- tion but entered step P. to a minor extent.
?.. In the solvolytic hydrolysis of p_,p_f-dimethylbenzhydryl chlo- ride in aqueous acetone the two steps do not have widely dif- ferent rrtee; therefore, a mass effect can be observed. Due to building up chloride concentration there is a progressive decrease in the overall reaction rate; furthermore a 0.05M addition of chloride ion depresses the reaction rate one-third.
3. Only those halides which cause Friedel-Craf t alkylation were effective in racemlzing a-phenylethyl chloride and the order of racemizing effectiveness coincided with alkylrting effec- tiveness (SbCl5> BC13 > SnCl4 > ZnCl3? HgCl). The solvolytic ionization is analogous to that caused by these strongly elec- trophilic reagents; it is presumably less in degree and due
to hydrogen bonding.
Bartlett has found that phenols exhibit hydrogen bonding with hydrogen chloride.
4. Because the rate of hydrolysis of benzhydryl chloride in ace- tone increases nearly thirtyfold for a tenfold increase in water concentration, and other similar results, Hammett refers to "polymolecular solvolysis." Dime.ric water must be more effective in solvolysis than monomeric.
An unusual case is the alcoholysis of d-bromopropionic acid to methoxypropionic acid. The reaction is mainly second-order at sodium methoxide concentrations from 0.5 to 1 M, but mainly first- order from 0,03 to 0.06 M. The second-order reaction inverts the configuration but the first-order reaction does not and is accom- panied by no racemization* The interpretation of the first-order reaction involves cyclization and double inversion; the reaction may or may not be solvolytic.
CH
3
Br ,p H 0 /P
d-CH3~C — C -» 1-CH3-C-— JC=0 -* d-CK3-C-C
n 0- ti 0-
The hydrolysis of ethyl benzene sulfonate shows the charac- teristics of solvolytic replacement.
-4-
10'
First-order elimination reaction? have been formulated by Hauser as occurring by tne solvolytic initial step.
As a result of this work, to quote Hammett, "It seems very probable that every displacement on an asymmetric carbon the rate of which is proportional to the concentration of the nucleophilic displacing ion or molecule involves an inversion of configuration."
Another step was taken toward the chemical explanation of homogeneous catalysis when it was shown that the solvent is kinetically important in first-order replacements.
Bibliography
Hammett, Physical Organic Chemistry , McGraw-Hill Company, New York
(1940).
Watson, Ann. Reports, |?, 236 (1940);. 35, 208 (1938).
Hughes, Trans. Faraday Soc, 34, 185, 202 (1938).
Hammett and McCleary, J. Am. Chem . Soc, 63, 2254 (1941 ).
Farinacci, ibid. . 63, 1799 (1941 ).
Bartlett and Dauben, ibid. , 62, 1339 (1940).
Reported by G. D. Jones December 3, 1941
103
AROMATIZATION OF ALIPHATIC HYDROCARBONS
In recent years steps toward more efficient utilization of natural hydrocarbon resources have led to development of such processes as cracking, nitration, chlorination, and automatiza- tion of gaseous hydrocarbons. The conversion of aliphatic hydro- carbons to aromatic hydrocarbons is by no means a recent discovery but the practical solution of the problem is the latest success of the petroleum industries.
A bulk production of aromatics from open chain compounds in industry obviously must make use of suitable naphtha cuts from petroleum sources, light oils arising from low temperature dis- tillation of coal, and the fractions of light spirit derived from either high pressure hydrogenation of coal or from synthetic processes for hydrocarbon production starting with water gas (Fischer-Tropsch synthesis).
Aromatization is best approached by consideration of thermal decomposition of paraffin hydrocarbons. This might proceed in the following manner in the case of n-hexane:
1. 2C6H14-> C13H36 + Ha, 6. C6H14~> C6H10(cyclic ) + 2 ESiJ
2. C6Hj.4 — ► C6H6 + 4H3, 7. C6Hj.4 — > C5H10 + CH4, ?. C6Hi4 — * C6H12 + H3, 8f 2C6H14 — + C5H13 + C7H1S,
4. C6H14 -> C6H12(cyclic) f H3, 9. C6H14-^ 6C + 7H3.
5. C6H14 -+C5H9CH3 (cyclic) + H3,
The maze of conceivable reactions may be simplified to a certain extent by a thermodynamic consideration of the possibility of each transformation at different temperature intervals. By means of linear equations expressing the free energies of formation of the hydrocarbons from graphite and hydrogen in the range 300° to 1000° A, the standard free energies accompanying the various transformations ca.n be determined. The conclusions to be drawn for a thermal decomposition at atmospheric pressure are: a re- forming reaction (l) cannot proceed since conditions favor the reverse reaction; cyclization and dehydrogena tion to benzene (2) begin about 500° A and the equilibrium is well on the aromatic- side at 600°A; dehydrogena tion to hexene (3) or cyclohexene (4' is not appreciable below 800 A; cyclization to methylcyclopen- tane (5) sets in about 350° A; cracking (?), disproportionation (Q), and decomposition to carbon and hydrogen (9) are appreci- able above 400° A.
,
-2-
IC9
Reactions 3,4,5 serve to produce substances which can form benzene by further dehydrogenation so that successful solution of the problem of aromatization of n-hexane hinges on the sup- pression of reactions 7,8,9 which involve fission of the G-C bond. The partial or complete elimination of these reactions will depend upon the choice of catalyst. The catalyst, moreover, must be specifically a dehydrogenation catalyst since the ener- gies of the C-C and the C-H bonds are, respectively, 58600 and 87300 cal./mole. Such selective action has been realized in mixtures of oxides of elements of groups VI, V, and IV (in order of their effectiveness) of the periodic table. The most common catalyst is amorphous Cr203 suspended on an inert carrier such as alumina, magnesia, or silica gel but mixtures containing molybdenum, vanadium, and sometijr.es titanium and cerium in ad- dition to the chromium give the best results. The method of preparation of these catalysts as well as the ratio of the dif- ferent oxides is very important to the activity of the catalyst.
Because investigators have employed different catalysts, reaction temperatures, and rates of flow, their results are not usually directly comparable. Nevertheless, the following table summarizes yields obtained in a few dehydrocyclizations for a single passage of the hydrocarbon over the catalyst.
Hydrocarbon
2-Methylpentane n-Hexane
2-Methylhexane 2,5-Dimethyl- hexene
n-Heptane
Composition of liquid product in weight per cent of starting material
Aromatic s
4.3 17.0 (benzene)
28.0 (toluene )
46.0
(80^ p_- xylene)
Naph- |
Ole- |
Parst- |
thenes |
fins |
fins |
2.2 |
14 |
53 |
2.2 |
14 |
54 |
0 |
18 |
44 |
0 |
16 |
26 |
1 |
9 |
43 |
0 |
11.5 |
75.5 |
30.0 (toluene) 26 *12.1
2,6-Dimethylheptsne 82.0 (m-methylisopropylbenzene )
n-Octane 38.0 (55^ p_-xylene, 35^ c_-xylene, 5% m-xy-
lene, b% ethylbenzene )
n-Nonane 52.0 (greater than 90^ methylethylbenzene )
Butylbenzene 12.0 (naphthalene)
What would seem to be low yields of aromatics in the table are in most cases satisfactory since the olefins and paraffins appearing in the product are converted to aromptics on recycli^e 0 Thus, the latest experimental data on the dehydrocyclization or n-heptane (starred drta in table), show that there was obtained 98.1 weight per cent of liquid products consisting of 12.1 weight per cent of toluene. The calculated recycle yield (yield ob- tained by repassing olefins and paraffins over the catalyst) of toluene is 89 weight per cent or 97 per cent of the theoretical..
-o-
b
no
All paraffin hydrocarbons with the exception of methane have "been converted to aromatice with the expected variations in yield. Those hydrocarbons whose structure permits formation of a six-membered ring are aromatized to a marked extent. Branched chain paraffins having less than six carbons in a straight chain give low yields presumably due to the diffi- culty of the necessary isomer! zation to a six-membered chain. Extent of aromrtization increases with the number of carbons and generally the more ways a cyclization can take place the greater the extent to which it occurs. Nevertheless, all possibilities of ring formation are not realized to an equal extent. It seems that the shortest sidechains (on aromatic rings) are formed preferentially*
The nature of the predominating aromatic products is governed by the reaction temperature. High temperatures fevoi? formation of naphthalene and anthracene. However, the pre sen-: value of the conversion of aliphatic s to pure aroma tics lies in the production of toluene and benzene.
The arometization of paraffins may be conceived as taking place in one of various ways. Thermodynamically, all are
Paraffin
\S i \
Naphthene <■ h Olefin
Aromatic
possible. Evidence indicates that the olefin is an inter- mediate in the reaction and not merely a by-product. The next step, cyclization, is considered to involve the double bond in a 5 or 6 position. This is in harmony with the observation that those olefins that have not undergone reaction and are isolated in the product contain the double bond in a 2 or 3 position. These remain only because a shift of the double bond must constitute a step preliminary to cyclization.
In no case have appreciable quantities of naphthenes been isolated from the reaction product. This does not eliminate' the naphthene structure as an intermediate in the proces*s since it would be dehydrogena ted more easily than the other reactants (probably before it could leave the surface of the catalyst). Under a given set of conditions the extent and ease of aroma tization from the following starting materials increases in the order: paraffin< corresponding olefin < six-membered naphthene < corresponding cycloolefin.
Aromatization of relatively pure hydrocarbons proceeds smoothly without formation of large amounts of by-products. On the other hand, the economic cracking of heavy petroleum residues to aroma tics depends on the disposal of considerable quantities of g°s, tar, "and carbon.
I I !
•4-
The latest variation of the dehydrocyclization reaction is the hydro forming: process, a dehydrogenation carried out in the presence of hydrogen. By this process, utilizing low octane heavy naphtha, there may be obtained either a. product consist- ing of 40 to 50 per cent aromatics (high grade gasoline) or one having upwards of 80 per cent aromatics (toluene production)
Bibliography
GoXdwaeser and Taylor, J. Am. Chem. Soc. , 61, 1766 (1939). Hoog, Verheus, and Zeiderweg, Trans. Far. Soc, 35, 993 (1939). Taylor and Turkevich, ibid. , 35, 921.(1939). Grosse, Morrel, and Mattox, Ind. Eng. Chem., 32, 528 (1940). Kazan sky, Losek, Zelinsky, Compt. rend. acad. sci. URSS., 27, 565 (1941); C.A., 35, 305 (1941). .
Reported by S. P. Rowland December 10, 1941
112
TKIAZOLES: SOKE SYNTHESES AND REACTIONS
M. T. Bogert - Columbia University E. Ochiai - University of Tokyo T. B. Johnson - Yale University
Thiazoles contain a f ive-membered ring with two hetero- atoms, nitrogen and sulfur, separated by a carbon atom. In num- bering the ring, one begins with the sulfur atom and proceeds in a manner such that nitrogen is the number three atom.
Recently there has been a great amount of interest in the pharmacological properties of thiazoles. Johnson has synthe- sized many thiazoles with a B, 4-dihyd poxy phenyl group in the 2-position. They possess definite analgesic and anesthetic ac- tion and are equally effective when administered intravenously; subcutaneously or orally. Three of the most recent chemothera- peutic agents are thiazoles — 2-sulfanilamidothiazole (sulfa- thia zole ), 2~sulfanilamido-4~methylthiazole ( sulfamethylthia- zole ) and ?.-sulfanilamido-4-phenylthiazole ( sulfaphenylthiazole ). They are superior to sulfanilamide and sulfapyridine in their bacteriostatic action on pneumococci types I, II and III, on p-hemolytic streptococcus A and on gonococcus. The 4-methyl derivative is even more effective then sulfathiazole itself, 4-Methyl-5~( p-hydroxy ethyl )-thiazole is the sulfur-containing portion of vitamin Bx.
The thiazole nucleus is important in the dye industry. A yellow dye, primuline, war the first thiazole dye. There is still a demand for Thioflavine T discovered by Green in 1889 because it is one of the few dyes giving yellow shades with a prized greenish tone,
In 192.1 it was reported that mercaptobenzothiezole was a very powerful vulcanization accelerator. At present the organic accelerator most widely used in industry is mercaptobenzo thia- zole, known to the trade as "Captax." Other popular accelerators are its zinc salt (Zenite) and its benzoyl derivative (Ureka C)0 These accelerators have nearly flat vulcanization curves making the vulcanization of thick rubber articles practical.
Syntheses of Mononuclear Thiazoles. — (l) The condensation of a-halogen aldehydes and ketones with thioacid amides.
This, one of the most important methods, was worked out by Hantzsch, The reaction is vigorous and is controlled by the use of a diluent such as alcohol, water or ice.
2-Methyl-4-phenylthiazole is obtained from cc-bromoaceto- phenone and thioacetamide.
-2-
I 13
HCBr
C6H5COK
HS,
HC— S
HN^
^C-CH.
C„HfiC— N^
,CCH3'HBr + H20
'6115'
Thiazoles in which the 2-position is unsubstituted ere obtrined when thiof or .tip mid e are used.
HCC1
HCOH
HS,
*CH
HN'
HC-S.
HC -N'
GH
The thiazole portion of vitemin Bx has been synthesized by three groups of workers, Buchman^ synthesis, reported to be the best, yields 8 - 12 grams of the thiazole from 100 grams of ethyl acetoacetate.,
CH3C0CH2C00C2K5
H0CH2CH2C-Sv CH3C-N/
0 NgOCaHs CH3COCHC^
/ >
CH2 CH2
S02C12
Cn2— CH-> 0
0 CH3C0CC1C^
1 >
CH3 Ch2
HC:
*NH.
0 CI CH3C~CCH2CH2OH H
Severpl recent patents appear to be based on this synthesiso
Johnson has made 4~(a-chloromethyl )- thiazoles by using symmetrical dichloroacetone and thiobenzarnide or its derivatives,.
HCCL HS
+ "CC6HS
HC — S.
\
ClCHoCOH
HN
f
C1CH2C— N
S
CCSH5
(2) Condensation of cc~helogen aldehydes and ketones with thioureas.
This synthesis wgs first employed by Traumann in the preparation of 2-aminothiazole s. It is illustrated by the synthesis of 2~amino~4-methylthiazole given in Organic Syntheses.
HCC1 CH3C0H
HS
HN
)PNH2
HC— Sv
CH„C— N^
CNH-
In a search for nevj sulfanilamide derivatives, Ziegler Iia s used it recently for the preparation of 2-amino~4~?lkylthia zolt r
-5-
I 14
CE3COCERCOOC2E5
Br:
CS.
4 BrCH2C0GHRC00C2H5
NHsGSNKj
->
EC=CCERCOOC2E5 I I
NEa
NaOE
EC = CCERCOONe
EC1
RCHaC— S>
S N \*
NH2
50-600
>
CNH-
EC— N^
(?) Treatment of a-acylated amino aldehydes and amino- ketones with phosphorus penta sulfide,
This reaction was first used by G-abriel in the synthesis o: 2-methyl*-5-phenylthia zole from acetaminoacetophenone.
C«H=COCEPNHCOCE.
P2S
3^5
C«E*C
6lA5
170°
EC— N^
CCE-
Similarly:
RC0NECH2C0CH3
P2S
CE,C— S
S'-'S
~>
X
CR
EC — N^
Syntheses of Benzothia zoles. — (l) Condensation of p_-amino- phenylmercaptans with acids, acid anhydrides, acid chlorides, aldehydes and ae tones.
Eofmann developed this reaction for the preparation of ben- zothiazoles. Condensing acetaldehyde with p_-aminophenylmercap~ tan results in the formation of 2-methylbenzothia.zoles.
+ CEoCEO
r V \
vV
CCE, + EP + EP0
However, these mercaptans are frequently unavailable since they ere usually prepared from the benzothiazoles and are often unstable. With some modification, this method has been especi- ally fruitful in the hands of Soger t and his coworkers.
N02C6E4SSC6E4N02
Zn
EOAc
■$»
L
(V
Zn
-■
I 15
-4-
RCHO
->
or RCOC1
CR
N
/
(2) Heating aroma tic amines and their derivatives with sulfur.
Bogert and Abrahamson report that the fusion of benzanilic.e with sulfur is the best method for the synthesis of 2-phenyl- benzothiazole .
c6h5c;
.0
NHCfiH
_S
6Xi5
V
%-
\
N
/
CCRH
6Xi5
Similarly, benzotniazole itself would be obtained from f ormanilide.
The principal product from the fusion of p_-toluidine with sulfur for the manufacture of thiazole dye intermediates is dehydrothio-p_~toluidine .
CH3
S
£— > CH3C6H^ lCC6H4NH2(p_)
I\
*w
NH-
A larger molecule is obtained upon further heating with sulfur » The sodium salt of the sulfonic acid obtained by the action of fuming sulfur id acid on this molecule is primuline0 An entire series of dyes is obtained by diazotizing and coupling primuline with amines and phenols.
(3) Rearrangement of compounds with the group CH-N-C-S-
1 A
Jacobson found that thioanilides and phenylthiourethans are oxidized by alkaline potassium ferricyanide at room temperature to benzothiazoles. Bogert and Mey?r used this method in tht synthesis of 2-(p_-tolyl )-benzo thiazole.
C-NHC6H4CH3
^\/S
K3Fe(CN)
■■A
\
/
CC6H4CH3
N
-5-
//
CH3C-NHC6H6
K3Fe(CN),
CCH3
Properties and Reactions of Thiazoles. — The benzothiazoles resemble the quinolines. They are stable to acids but yield an c_-aminophenylmercapte.n end an acid when boiled with alcoholic alkali or fusing with potassium hydroxide.
The thiazoles resemble the pyridines. Their aqueous solu- tions react neutral and they form stable salts with acids which have an acid reaction. They are oxidized by potassium perman- ganate. They are unaffected by most reducing agents but sodium amalgam and alcohol cleave the ring with the formation of an amine and a mercaptan. 2-Aminothiazole s can be diazotized and then treated as any primary aromatic amine. 2~Amino~4-elkyl~ thiazoles couple in the 5-position with diazotized amines to form azo dyes. Two molecules of the 2-amino -4-alkylthiazole will also condense with an aromatic aldehyde to form a compound of the type
ArCHf-C — S
RC— N^
"CNH:
J2
The chlorine atom in the 2-phenyl-4-(cc~chloromethyl )~thiazoles is comparable to that in benzyl chloride. Ammonia will replace it with an amino group at room temperature.
Most of the work on the substitution zoles has been done in the last three year coworkers. Bromine ti on s were conducted by zole in ice cold dilute sulfuric acid with it to stand overnight and by allowing the overnight with chloroform and bromine. Tii methyl thiazoles were unaffected while the aminothiazoles gave the 5-bromo derivative are obtained when thiazoles are bromine ted
at 250° and 400u ben zothie zole is
The 2-bromo brominated a t
derivative 450°.
reactions of thia- s by Ochiai end his
treating the thie-
bromine and allowing thiazole to stand e 4-methyl and 5-
2-hydroxy and 2~ 2-* Br omo thiazoles
in the vapor phase is also obtained when
4-Methylthie zole cannot be nitrated, 2-Hydroxy-4-methyl- thiazole gives the 2-hydroxy-4-methyl~5~nitrothie zole . The analogous 2-a.minothiazole gives the corresponding 4-methyl-2- nitremino-5-nitrothie zole. When a phenyl group occupies the 4- or the 5~position, a p_-nitrophenyl derivative is obtained.
4~Methylthiezole~5-sulfonic acid is obtained when 4- methylthiazole is sulfonated. The 2-amino derivative yields 2~sulfemino^4-methylthiezole when sulfonated in the cold.
I 17
-6-
Further treatment produces the corresponding 2-sulf amino -4- mrthylthiazole-5~sulfonic pcid.
A thiazole unsubstitut ed in the 2-position is obtained when 2~mercaptothiazoles are oxidized by acid or neutral hydro- gen peroxide. Oxidation with alkaline hydrogen peroxide pro- duces the 2- sulfonic acid,
2-Hydroxy-4-methylthiazole undergoes the Friedel-^Crafts reaction with acetylchloride to give 2-hydroxy~4-methyl-5-acetyl~ thiazole; with benzoyl chloride to give 2-hydroxy~4~methyl-5~ benzoylthiazole. 4-Methylthiazole would not undergo the Gattermenn reaction but 2 ~hydroxy-4-m ethyl thiazole did, forming the 5-eldehyde.
An amino group is introduced in the 2-position both of thiazoles and benzothia zoles when they are treated with sodamide at 140-150°.
Fox and Bogert have observed that 6~alkoxy-7-nitroben- zothiazoles, in the presence of alcohols and small amounts of dilute caustic alkali, exchange their alkyls for those of the alcohol used.
Bibliography
Bogert and Meyer, J. Am. Ghem. Soc, 44, 1568 (1922).
Bogert and Abrahamson, ibid., 44, 826 (1922),
Fox and Bogert, ibid . , 63, 2996 (1941).
Olin and Johnson, ibid., 63, 1475 (1931 ).
Ochiai, Kakude, Nekayama and Masuda, J. Pharm. £>oc . , Japan,
59, 462 (1939); Chem. Abs., 34, 101 (1940). Buchmrn, J. Am. Chem. Soc, 58. 1803 '1936), Ziegler, ibid. , 63, 2946 (1941).
Lawrence, Proc. Soc. Exptl. Biol. Med. . 43_ 92 (1940), Cadwell and Temple, A. C. S. Monograph, 74., Chaot, VIII. Meyer and Jacobson, "Lehrbuch der Organisohen Chemie," Waltei
de Gruyter and Company, Berlin and Leipzig (1920), Vol. Z.
Part 3, p. 535.
Reported by G, W. Cannon December 10, 1941
.
I 18
HIGH PRESSURE HYDROGENhTIONS OVER NICKEL AND COPPER CHROMITE
Adkins et. al. University of Wisconsin
The use of hydrogen at high pressures and temperatures in laboratory syntheses has been made practical by the work of Adkins and his students-. For the most part, this seminar is con- cerned with the results which they have obtained. In a previous seminar, the reactions of organic molecules over platinum, palladium, and other catalysts at low pressures were covered. These data will not be included in this report.
Hydrogen reacts with organic molecules either by addition to a multiple linkage (h.ydrogenat ion) or by cleavage of the molecule (hydro genolysis) . For convenience, the reaction of hydrogen with common functional groups will be considered before discussion of the selective hydro genat ion of polyfunctional molecules.
The most widely used catalysts for these hydro genat ions are copper chromite and nickel. Nickel catalyst is prepared either on a support such as kieselguhr or as "Raney" nickel. It is a highly active, versatile catalyst even at room temperatures and at low hydrogen pressures. In many partial hydrogenations it is difficult to stop the reaction when the calculated amount of hydrogen has been taken up. In cases like these the "limited (16) hydrogenation" procedure of Adkins and Durland can be used. This procedure is carried out by filling the bomb with nitrogen at 100 atmospheres and then adding the calculated amount of hydrogen a The nickel functions as a catalyst until the partial pressure of hydrogen approaches zero.
Copper chromite (Adkins5 catalyst) 3s a highly useful, stable catalyst prepared by the decomposition of copper ammonium chromate. It is active at temperatures above 100° and at hydro- gen pressures greater than 20 atmospheres, and is particularly useful for the hydrogenation of esters to alcohols and amides to amines. It is notably inactive for the hydrogenation of the benzenoid nucleus and, therefore, may be used for the hydrogena- tion of aryl compounds without much danger of saturating the ben- zene ring. As an example, naphthalene can be hydrogenated over copper chromite to tetralin, at which point the reaction stops. With nickel as catalyst tetralin can only be obtained by in- terrupting the reaction when the theoretical amount of hydrogen has been absorbed. A review of the preparation and properties of these catalysts is given in recent publications. (1,2)
"
-2-
! (9
Hydrogenation of the alkene linkage: In general the alkene linkage is easily and quantitatively reduced. A selected list of examples is given below. Of all the hydro carbons, the only carbon-carbon bond which is easily cleaved is the ethane bond in phenylated ethanes. For example, tetraphenylethane undergoes hydrogenolysis (cleavage) of the carbon-carbon bond. The corres- ponding cyclohexyl derivatives have a stable ethane bond*
Hvdrogenation of Alkenes
Products**
ethyl benzene ethyl benzene 1,1,2,2,-tetra- phenyl ethane
1,1,2,2-tetra- phenyl ethane cyclohexane
*Ni(k) - nickel on kieselguhr **The yields averaged 96-100$
Compound |
°c |
Time(Min. ) |
Catalyst* |
styrene |
20 |
25 |
NiU) |
styrene |
125 |
? |
CuCrO |
1, 1,2,2, -tetra- |
100 |
130 |
Ni(k) |
phenyl ethylene |
|||
1,1,2, 2-tetra- |
150 |
15 |
CuCrO |
phenyl ethylene |
|||
cyclohexene |
165 |
1 |
CuCrO |
(3) (3)
(3)
(3)
(4)
Hydrogenation of carbonyl compounds: The carbonyl group in aldehydes, ketones, esters, and lactones is of varying reactivity. Hydrogenation of the carbonyl group in amides will be discussed separately.
(a) Aldehydes and ketones generally react with hydrogen under conditions as mild as those required for the reduction of the alkene link. The yield of alcohol is practically quantita- tive even with aldo or keto esters and alcohols. Carbonyl com- pounds which are reduced to benzyl alcohols or to 1-3 or 1-4 glycols can undergo hydrogenolysis of the C-OH bond. This side reaction is a function of the conditions; the higher the tempera- ture and the longer the duration of the reaction, the greater will be the amount of hydrogenolysis. For example, benzflldehyde is reduced in 92$ yield to benzyl alcohol; however, reduction of ethyl benzoate requires high temperatures and gives a quantitative yield of toluene rather than benzyl alcohol.
o (b) Esters are hydrogenated over copper chromite at 200- 250 to give almost quantitative yields of the corresponding al- cohols or glycols. Esters which are reduced to unstable alcohols or glycols give lower yields. lb reduce diethyl succinate to tetramethylene glycol (74$ yield) the reaction is carried out as rapidly as possible in order to cut down the cleavage of the 1-4 glycol.
(c) 1-3 Diketones can be reduced either to the keto al- cohol, or to the 1-3 glycol as well as undergoing hydrogenolysis of the carbon-carbon bond. Keto alcohols have been isolated in the reduction of a series of 1-3 diketones in 35-68$ yield. The corresponding 1-3 glycols could be prepared in 50-90$ yield. If the methylene carbon of the dlketone is substituted, hydrogenolysis
'
1
.
120
- 3 -
takes place quite readily. The labile carbon-carbon bonds in 1-3 diketones are shown by dotted lines in the formula below —
|l H
R— C . . . C . H
jj
c7
50-90^
<JH H OH
* R-C-— £^C— R1 H H h
0 H OK
58~68^ -> R-C-C-C— R'
H H
(11) (12)
Compound acetone
cyclopentanone acetophenone
aaetophenone
H.vdrogenation of Carbonyl Compounds T°C Time Catalyst Yield
Product
125 100 150
13 m 6 hr 30 m
Ni(k) Ni(k) CuCrO
110 lOmn Ni(R)
100
100
95
91
propanol-2 (3)
cyclopentanol (6)
phenylmethylcarbi- nol (7)
phenylmethylcarbi- nol (7)
diphenylcarbinol (6
1,2-diphenylethyl- ene glycol (b)
benzyl alcohol C4) toluene (5) toluene (8) ethylbenzene diphenylme.thane (4 hexanol-1 (8) octadecanol-1 (9)
tetramethylene gly- col (10) butyrolactone
hexanediol-1-6 (10
2-isopropylbutane- diol,l-4 (10)
2-phenylbutanediol?
1-4 3-phenylbutanol-l (10)
The functional group in cyanides, imines, hydro xylamines, oximes and nitro compounds is readily reduced under mild condi- tions to give good yields of amines. Formation of secondary amines is a side reaction in the reduction due to the reaction of an imine with the primary amine. This secondary amine formation can be reduced to a large extent by carrying out the reduction
benzophenone |
160 |
1 hr |
Ni(k) |
87 |
benzil |
125 |
1 hr |
Ni(k) |
90 |
benzaldehyde |
180 |
1 m |
CuCrO |
92 |
benzylalcohol |
125 |
10 m |
Ni(k) |
88 |
ethyl benzoate |
250 |
— |
CuCrO |
100 |
acetophenone |
175 |
6 m |
Ni(k) |
88 |
benzophenone |
175 |
1 hr |
CuCrO |
97 |
ethyl caproate |
250 |
5 hr |
CuCrO |
95 |
ethyl stearate |
250 |
3 hr |
CuCrO |
95 |
diethyl succinate |
250 |
1-1/2 hr |
CuCrO |
74 18 |
diethyl adipate |
250 |
— |
CuCrO |
95 |
diethyl a- methyl succinate |
250 |
3 hr |
CuCrO |
30 |
diethyl a-phenyl succinate |
250 |
6 hr |
CuCrO |
12 67 |
\2\
_4_
rapidly, and by using methanol- ammonia as solvent. By this procedure alkyl cyanides can be reduced to primary amines in yield as high as 95$.
R-C=N
K R-C=NH
R-CHa-NH:
H H ^
R-C-N-CHa-R I
NH3
H R-CHa-N-CHs-R
n-butyl cyanide 125
n-butyl cyanide benzal aniline benzal aniline
methanol-ammonia
65 5m Ni(k) 175 25 m CuCrO
30 m Ni(R) 67$ n-pentyl amine (8)
16 di-n-pentyl amine
95 n-pentyl amine (13)
9? phenyl benzyl amine
97 phenyl benzyl amine .
(15)
Reduction of amides to amines: Although this hydrogenation requires drastic conditions (temperature 250-265°, pressure 200- 300 atmospheres) the yield of amines is very good. With substi- tuted amides, the yields range from 80-95$ while with simple amides the yield is lowered due to formation of secondary amines at the high temperature necessary for the reduction. In addition to reduction of the carbonyl, the linkages shown by dotted lines in the formula below are subject to hydrogenolysis. For the most part side reactions due to cleavage of these bonds are small* (l)
0=C,..N...R*
i H
Reduction of Amides (CuCrO catalyst)
Lauramide Heptamide Kept amide
250 250
250
Salicylamide 250 Lauroylpiperadine 250
N-diethylheptamide 250
42 m |
48 |
4 hr |
49 |
4 hr |
39 |
58 |
|
2 hr |
80 |
1 m |
2 |
92 |
|
1 hr |
64 |
4 |
|
25 |
n-duodecylamine (10) di-n-duodecylamine n-heptylamine (10) di-n-heptylamine q-cresol (10) n-duodecyl alcohol (10) N-n-duodecyl piperadine n-heptylethylamine (10) n-heptyldiethylamine dl-n.-heptylamine
Reaction of glycols with amines: Secondary and tertiary amines can be prepared in 30 to 70$ yields by reaction of gly- cols with amines in the presence of copper chromite. When 1-4, 1-5, or 1-6 glycols are used the corresponding pyrrolidine, piperidine, and hexahydroazepine is produced. The yield of the seven member ring compound (hexahydroazepine) is low (17$), but that of pyrrolidines and piperidines averages 50-75$.
■
••■
•
■ .
-
' -
I
.. ..
!
■':■
^B^fmm
122
CH3v.
R-NH3 + HO-CH3-(CHs)n-CH20H -> R-N ' /(°Hs)n
ur«3
Pentane 1-4 diol 250 CuCrO 60 l-n-amyl-2-methylpyrroli«-:. n-amylamine dine (17)
hexane 1-5 diol 250 CuCrO 75 l-n-amyl-2-methylpiperi- n-amylamine dine (17)
Hydrogenation of N-substituted amides of dicarboxylic acids : When the amides are prepared from 4,5,6-carbon chain dicarboxylic acids, hydrogenation gives excellent yields of pyrrolidines, piperidines, and hexahydroazepines. By this method £ and fr substituted piperidines can be prepared by starting with the appropriate substituted glutaramide.
0 0 '
H M i , , h H ^CH2-C
R-N-C-C-C-C-C-N-R -» R-H l-C"
I I I VCH3-S'
R amyl, benzyl, phenylethyl
Adipamides cyclize in this reaction to give hexahydroazepines in 35^ yield
In the case where the piperidine amide of these acids is reduced, ring closure is impossible and straight chained substi- tuted cadaverines are formed,
CH3-CH3 || jj CH3-CH3v CH3 >-C-(CH3) -C-Nsx CH3 -* py-CH3-(GH3)n-CH3-py ^CR3-CK3 CH3-CH3
N-n-amylsuccimide 200 30 m CuCrO 79 N-n-amylpyrroli-
dine (10 glutaramide 250 2 h CuCrO 70 piperadine (14
di-N-n-amyl- ^ , £ dimethyl 250 4-1/2 CuCrO 69 l-n-amyl-4 , 4 » - glutaramide h dimethyl piperi-
dine (14) £-methylglutarimide 250 1-1/2 CuCrO 50 4-methyl piperi- dine (14)
Hydrogenation of pyrrolldones and pioeridonas: (14)
0
CHa-C . CH3-CH3
f ^N-R -* I ) N-R
CH3-CH3 CH3-CH3
0
Cri3— C * CH3— CH3
CH3 N-R -* CKa SN-R
NCH3-CK3 NCHS-CHSX
It is not impossible that these cyclic imides would be in- termediates in the hydrogenation of amides of dicarboxylic acids to cyclic amines. However, the yield of cyclic amines by
.
.
-
■
-6- 123
reduction of these compounds is less than the overall yield from the straight chained amides. The piperidones and pyrrolidones are prepared in excellent yield by partial reduction of the appropriate cyclic imides. Nickel is used as a catalyst in this reduction,
0 .0
s CHa-C* Ni „ CH3-C* CHSn 7J-R r-* CH3^
CH2-C; 220° NCH2-CH3 0
Preparation of amines from aldehydes and ketones: The re- duction of aldehydes and ketones in the presence of ammonia or amines (Mignonac's method) is a very satisfactory preparative method for secondary and tertiary amines.
butyraldehyde 125 2 hr Nl(k) 70 dicyclohexylamine (15)
piperidine 93 N-n-butylpiperadine (15)
cyclohexanone 125 2 hr Ni(k) 70 dicyclohexylamine (15) cyclohexylamine
In hydrogenation reactions today the critical problem is not better, but more selective hydrogenation. By selective hydro- genation is meant the use of the proper catalyst and conditions so as to bring about the preferential hydrogenation of the important functional groups in the presence of each other.
Control of reaction: Success in selective hydrogenation de- pends primarily upon the selection of the catalyst, temperature, and duration of reaction, and secondarily upon the medium of reaction.
(a) Catalyst: Proper choice of a catalyst is perhaps the one most important factor to be recognized in choosing con- ditions for a selective hydrogenation. For an example, copper chromite is relatively inactive toward benzenoid nuclei, hence aldehydes, ketones, esters, and amides containing an aryl
group may be hydrogenated to the corresponding alcohols or amines containing the benzenoid ring. Nickel, on the other hand, is in- active toward oxygen-containing groups as the amido and carbalkoxyl groups, so compounds of these types containing aryl groups may be converted to the corresponding amides or esters containing cyclohexyl groups. Other selective catalysts will be encountered later.
(b) Temperature: In the discussion of hydrogenation re- actions it was noted in many instances that one functional group reacts at a temperature sufficiently below that required for another functional group to make selective hydrogenation easily attainable. An example of this is the selective hydrogenation
of the alkene linkage in preference to other unsaturated carbon- carbon linkages as in benzene, etc.
(c) Duration of reaction: In the preparation of p- phenylethyl alcohol from ethyl phenylacetate, the optimum yield of
124
~7-
phenylethyl alcohol ie obtained only by interrupting the hydro- genation before all the ester has undergone the first step in the following reaction: (29)
250°
(1) C6H5CH3C00C2H5 + 2H3(CuCrO) > C6H5CH3CH30H + C2H50H
(2) C6H5CH3CH30H + H3(CuCrO) 2-50° > C6H5CH3CH3 + HOH
(d) Modification of reaction medium: If diethyl furfural acetal is hydrogenated two principal products are obtained:
(1)
>-CH(0CaH5)3 + 2K:
1
I > CH(0C3HB)3
(2)
\,
CH(0C3H5)3+ 3Hj
CH30C3H5 + C3H50H
Over a nickel catalyst reaction (2) takes place to give the ether in 95$ yields. However, if a little sodium carbonate, sodium ethoxide, or any one of several amines is added, reaction
(l) takes place and the saturated yields. (5)
.cetal is obtained in 30-80
7"
1. |
Alkynes |
s; |
2, |
Alkenes |
9. |
3. |
Smides |
10. |
4, |
Oximino |
11. |
5. |
Nitro-nitroso |
12. |
6, |
Cyanides |
13. |
7. |
Aldehydes |
14. |
Structure and ease of hydrogenat ion : Attempts to summarize the relative ease of hydrogenat ion of a series of monofunctional compounds will embody contradictory facts which have no immediate explanation. However, on the basis of present data, the following qualitative list may be proposed:
Ketones
Furanoid
Pyridinoid (subst.)
Benzenoid
Pyrroioid (N-subst.)
Esters
Amides
The relative ease of reaction of two compounds with different functional groups taken separately is not a safe basis of pre- diction as to the relative rates of hydrogenation of the two com- pounds in a mixture or of the two groups when both are in the same molecule. d-Alpha pinene is hydrogenated much more rapidly than cinnamic acid, yet in a mixture of the two, cinnamic acid is completely hydrogenated to the exclusion of the pinene. Likewise in a single molecule the functional group which taken alone is most reactive is not always the most active when the molecule contains a second functional group.
Selective hydrogenations:
1. Hydrogenation of unsaturated esters: Despite the fact that the carbon to carbon double bond in alkenes is
hydrogenated under as mild conditions as is any other functi group, and despite the fact that the carbalkoxyl group requi a temperature of over 200° and pressures exceeding 100 atmospheres, certain unsaturated esters have been hydrogenat to the corresponding unsaturated alcohols using zinc chromit as a catalyst. (18)
o Butyl oleate 300 11 h ZnCrO 65 Butyl erucate 300o 11 h ZnCrO 68
125
onal res
ed e
octadecenol
ducosenol
2. Hydrogenat ion of benzenoid type corn-pounds : The benzenoid nucleus requires more drast j c conditions for hydro- genation than most other functional groups.
1 h Ni(R) 100 eye lone xane (30) 5 m Ni(R) 100 me thylcyclohexane (30) 4 h Ni(R) 92 1, 3 , 5-trime thylcyclo- hexane (30)
Benzene 150 Toluene 175° Mesitylene 200°
9-10 dihydro- phenanthrene 90$
*VScP"
sym-octahydro-
phenanthrene
85$ jJHiiRL.
t e trade cahydro- phenanthrene 85-90$
(a) Selective Hydrogenatioh of Phenanthrene (16)
' (19) dodecf hydro-
>i phenanthrene
a s ymm- oc t ahydr o-
s^s phenanthrene
y Ni(R) v 30$ r3TJo->
1,2,5, 4-tetra- hydrophenanthrene 40$
5. Hydrogenat ion of furanoid genation of the furanoid nucleus occurs conditions than those used for the benz differs in that the furan ring is a eye ject to hydrogenolysis.
o 2-Kethyl furane 150
o Furfuryl alcohol 125
o Furoin (in EtOH) 150
type compounds: Hydro- under somewhat milder 3noid ring and also lie ether and hence sub-
1 h
Diethyl furfuryl acetal (l g. amyl amine )
175
Ni(k)
2.5h Ni(k) 0„5h ^7i(k)
83 85 93
5 h Ni(k) 76
2-methyl tetrahydro-
furane (20) t e tr ahy dro furfuryl
alcohol (21) 1; 2-dihydroxy-l, 2- di t e tr ahydr o furfuryl ethane (21) diethyl tetrahydro- I'urfural acetal (21)
-9-
126
4. Hydrogenation of gyridinoid type compounds: Pyri- dine requires a somewhat higher temperature for conversion to piperidine than does the transformation of benzene to cyclo- hexane , However, derivatives of pyridine are generally more readily hydrogenated than are derivatives of benzene. This is evidenced by the preferential hydrogenation of the pyridine ring in compounds such as quinoline and phenyl pyridine.
Pyridine compounds with substituents benzyl, etc., in the 2 or 2,6-positions a at lower temperatures than is pyridine it
due to the effect of of the nitrogen to
the
"poison",
sub st it ue i.e..
nt a to
pyridine 3-acetyl
pyridine
2-methyl pyridine 2-benzyl pyridine 2, 6-dicarbethoxy pyridine
o 200 145°
200 100( 137 '
7 h 4.5h
0.6h 3 h O.Olh
Ni(R) Ni(R)
Ni(R) Ni(R) Ni(R)
in
com
83 28 61
90 85 66
, viz., carbethoxy, re hydrogenated rapidly self. This i s no doubt lowering the tendency bine with, the catalyst,
Refs. (22)(4) piperidine
3-ethyl piperidine 3-piperidyl methyl ke- tone 2-methyl piperidine 2-benzyl piperidine 2, 6-dicarbethoxy piperidine
Copper chromite is also active for hydrogenation of the pyridine ring in compounds such as quinoline.
5. Hydrogenation of i&yrroloid type compounds: Pyrroles are far less reactive toward hydrogen than are the derivatives of benzene, pyridine, or furane. The 2,3,4 and (or) 5-alkyl pyrroles react with hydrogen over nickel or copper chromite at 200-250°. However, the 2,3,4 and (or) 5-carbethoxy pyrroles re- quire such drastic conditions th:. t hydrogenation of the nucleus does not occur without simultaneous hydrogenation and hydro- genolysis of the carbethoxyl group. On the other hand, pyrroles bearing a carbethoxy or aryl group on the nitrogen atom readily react with hydrogen from room temperatures upward.
x-
Because of the resistance of the nucleus toward hydrogena- tion, acyl pyrroles have been converted to alkyl pyrryoles in e cellent yields over copper chromite. Similarly carbethoxy pyrroles have been converted to methyl pyrroles although careful control of the reaction is necessary so as to avoid excessive formation of pyrrolidines.
a.
b.
c. d.
e.
h h
Pyrrole 180 1
1-Phenyl 135° 1
pyrrole 0
1-Carbethoxy 70 0.3h
pyrrole 0
2,4-diacetyl- 160 0.3h 3,5-dimethyl
pyrrole 0
Ni(R) Ni(R)
Ni(R)
CuCrO
47 63
93
94
2,4-dicarbeth- oxy-3,5-di- methyl pyrrole
220 0.3h CuCrO 53
pyrrolidine (25) 1-phenylpyrrolidine (23)
1-carbethoxypyrrolidine
(24) 2,4-diethyl-3,5- dimethyl pyrrole (23)
3-carbethoxy-2,4,5 tri- methyl pyrrole (23)
2,3,4,5-tetramethyl pyrrolidine
-10-
127
f. 1, 2-dicarbeth- oxy pyrrole
200
1 h
CuCrO 42 2-methyl pyrrole (25)
. <_ a &« — Hydrogenation of substituted amides; As previously oon P«AohydJ0ge£ under 200-300 atmospheres reacts with amides at
°I °ner the i?flue2ce of copper chromite to give amines. A study of tne reaction of various substituted pyrrolidones and piperidones and open chain amido esters with hydrogen offers an excellent example of selective hydro genat ion. (25)
r
o=c.
CuCrO
N /~C00C3H5 210-220 30 m.
o=c
IT H
93^
r
o=c
H
H CONC5H
1 1
/ Np-C00C3H5
■CH,
C5Hn
CuCrO
210°
35 m.
r
o=c
H sN/*-CKaN-C6Hlx
H
m%
CuCrO
240-250" 20 m 0=C
H r—C0NCBHxl
0-C
CuCrO
KJ
N
235-240° 60 m
\
C5H1X
CHpOH
+
F 7 — VI
\E,
NG5H.L1
'— CH,0H
C 5 H ! x
0
C8H50C0CH3CH3C-^
V
x —
CuCrO 57/^ KOCH3CH3CH2CK2-N 230 35m 20^ H0CH3CH3CH3CH30H
^ti/fnmth«%ab0Vf^ata.il: iS observed ^at a carfcethoxy or group in the 5-position in a pyrrolidone-2 is more reactive J hydrogen than is the lactam group of the ring. In the"r>1-er: rtp£wVhl8 taT relationship holds with respect to the r^rhl
ttlol^ZLi* Tth the araid° ^ivatives, preferential h/£ ation of the lactam group in the ring occurs. In the oren >i
cm^?t!nJerS\bf h the 6Ste^ gT0Up a*d amid0 group undergo was fouSf fL h^^gen?1!i0S iand h^ogenolysit) since no evic was found for either the hydroxy amide or the amino ester.
am i do toward 1. c? i n 9 sthoay
■O.c p;>-
•c.ln lence
123
(26)
•11-
D'l^nni and Adkins/have shown that under the influence of either fUney nickel or copper chromite various N-pentamethylene amides having a hydroxyl group in the a, ft %X t oppositions are converted to the corresponding amino alcohols in yields of 51- 79$. In the case of the ^-hydroxy amide the hydroxyl group is eliminated and the chief product is the alkyl piperidine.
7. Hydro e; enat Ion of P.vrones (27)
0
.A/V
•CpH
3iA5
-^ O V
/y vcaH, w92,
2-ethyl chromone
\X/-CH3(CH2)3
CH;
'O
&
*$
J^
Flavone (2-phenylchromone ) can be hydrogenated to the corresponding series of compounds.
(28) 8. Hydrogenation of 6-ketoni trilea: Wiley and Adkins/ have found that the products obtained from the easily reduced ketonitriles depend upon the temperature used in the hydrogena- tion process, At 35-40° over nickel the ketoamines were iso- lated in 10-60$ yields, but hydrogenations carried out rapidly at 150-250° gave 30-60% yields of the aminoalcohols. Hydro- genolysis also occurred to some extent at 150-200°.
Bibliography;
Adkins "Reaction of Hydrogen" University of Wisconsin Prsss.
Madison, Wisconsin, 1937. (.I.-*
Adkins and Shriner, Gilman chapter (in press). (°.) Adkins and Zartman, J. Am. Chem. Soc, 54, 1668 (1932) (;>'
Adkins and Connor, ibid., 53, 1091 ( 19317. (•*)
Adkins, Covert, and Connor, Ibid, 54, 1651 (1932). ,(5j
Adkins and Cramer, ibid., 52, 4349 (1930). /6)
Covert MS Thesis, University of Wisconsin v7)
-12-
Burdick, MS Thesis, University of Wisconsin Adkins ana Folkers, J. Am. Chem. Soc, 54, 1145 Wojcik, ibid. , 55, ■ 4349 (1933). and Sprague, ibid., 56, 2669 (1934). and Stutsman, ibid. . 61, 3303 (1939). and Schwoegler, ibjdH 61, 3499 (1934). and Paden, ibid., 5J., 2487 (1936). and Winans, ibid. , 55 , 2051 (1934). Durland, ibid, 60, 1501 (1938)* Hill, ibid., 60, 1033 (1938). Sauer, ibid., 59, 1 (1937). Durland, ibid., 59, .135 (1938).
129
Adkins,
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins,
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins
Adkins,
Adkins,
and
and
and
and
and
and et
and
and
dnd
and
and
and
Wojcik and Covert, Zartman and Cramer,
Connor, ibid. . 54, 4678 (1932, Eurdick, ibid., 56. 438 (193' al., ibid., 56, 2425 (1934). Signaigo, ibid., 58, 709 (1936). Rainey, ibid., 61, 1104 (1939).
Sauer, ibid., 60, 402 (1938). D'lanni, ibid., 61, 1675 (1939).
Mozingo, ibid. , 60 Wiley, ibid. , 60
669 (1938). 914 (1938). ibid., 55, ibid., 53
1669 1425
(1933). (1931).-
(8) (1932). (9 (10
(11 (12 (13 (14 (15 (16 (17 (18 (19 (20 (21 (22 (23 (24 (25 (26 (27 (28 (29 (30
Reported by C. M. Himel
R. C. G-unther December 17, 1941
I »
|
130
THE CHEMISTRY OF ORGANOBORON COMPOUNDS
Organic compounds containing boron comprise a field which for many years has received relatively little attention, and is still generally unfamiliar. la recent years interest in these compounds has growr.'? for several reasons. For one thing boron is situated next to carbon in the periodic table and hence is very similar in effective nuclear charge ant atomic radius.; therefore, one may expect studies of organic boron compounds to be of value in interpreting the behavior of analogous carbon electronic sys- tems. Another emphasis is placed on boron compounds because of the ease of effecting nuclear disintegration of the boron aton, Kruger has recently shown that when neoplastic tissue which has been impregnated with boric acid is bombarded with slow neutrons the tumor cells are killed very effectively by the boron disin- tegration products. His experiments were carried out in vitro . bu he concludes that the same results could be obtained in the living body if the boron could be applied to the diseased tissue. This suggests the possibility of developing organic compounds contain- ing boron which may be specifically absorbed by the tissue.
Therefore, it is the purpose of this report to briefly outline the preparation and properties of the different classes of organoboron compounds., with reference also to certain mechanism explanations.
I. Trial^yl and Triaryl Borings (R.^B)
A. Preparation* The first organoboron compounds were pre- pared by Frankland in 1862: he obtained trimethyl- and triethyi-- borine by the interaction of zinc alkyls and ethyl borate. Ifrai.se and Nitsche have more recently obtained good yields of the berimes by the action of Grignard reagents on the etherate of boron trifluoride. Johnson, Snyder, and Van Carapen have also produced these compounds from methyl borate and Grignard reagent s; but find the BF3 method preferable,
3ZnR3 + (EtC)3B ~~^-> R33 ■-'- ?7.nPX'Et
3RMgX + Et20:EF3 .'-a— » R,!E •■? l<EzXJ -'- Et3C (80# y\-ld)
3RMgX + (MeO)3B -l&—> RSB + 3MgX0Me {50% yield)
(R = alkyl or aryl)
B. Properties: The aliphatic borine s have been stucied more thoroughly recently. Trimethyl borine is a gas - the only gaseous organometallic compound; the higher members of the series are colorless liquids.
(l) These compounds are so easily oxidized that they are spontaneously inflammable in air. (2) They are stable toward water, in sharp contrast to thy organic derivatives of the
13!
- 2 -
neighboring elements Be and Al. (3) With dry HBr a hydrocarbon is produced by the removal of one alkyl group:
R3B + HBr -> R3B-Br + R-H
In the presence of water the boron bromide is hydrolyzed to R2B-OH, but with anhydrous HBr the reaction is clear-cut, and the dialkyl boron bromide produced is a new type compound which may be important in the synthesis of other new organoboron com- pounds, (5) With dry Brs more complex reactions take place, with some substitution of bromine in the alkyl groups forming; HBr which may then react with the borine as shown above however, it is definitely shown that two alkyl groups aia.- s^iii off. This will be mentioned later. (6) As would be expected., NH3 forms an addition complex; a more unusual reaction is the addition cf the alkali metals - the latter is specific for t; aromatic borine s, it seems.
'OP
II. Boronic and Borinic Acids, Alkyl Boron Oxides [RB(OH)s, R5BOH, RBO]
1. Aromatic
A. Preparation: These were first prepared by Michaelis and Becker in 1880 by use of an organomercurial. Several bettei methods have been developed since then: Krause obtained the acids by careful oe-cdation of tria..'"vj. bonines ore-oared frcn BF3 and iUSgX-
R3B 4^~* R2E~0R --'- RE.(TK)8 •-— ** fiE'-">H;c
Khotinsky and Me lame ci used methyl borate and aryl magnesium bromide,
RMgBr -: 'LleC}33 — ; RD(CMe)P ~3a£u RB(0Il)a
Konig and Scharrnbeck modified :.".:. ;, procedure by the use of i-.o- butyl borate, and Bean and Johnson y Lie use of n--butyi borate. The last method is probably the b.^j • of the lot.
Diaryl borinic acids are f 01 ned as by-products in the above reactions.
B. Properties: The acids are crystalline white Solaris, soluble in organic solvents, sparingly soluble in water They are fairly stable substances, for benzene boronic acid has been nitrated and the nitro compound reduced to the amine without cleavage of the 0-3 bond by Johnson and his coworkers. This bond may be cleaved, however, upon warming with metallic salts such as HgClg, Zn012, and by aqueous H2Os or bromine water. They do not undergo atmospheric oxidation. Upon treatment with ammoniacal Ag20 the hydrocarbon R-H is formed.
•
- 3 -
2. Aliphatic
A. Preparation: Frankland first prepared an aliphatic boronic acid by the careful oxidation of triethyl borine, followed by hydrolysis. They are now prepared by the same , methods used for the aromatics. Snyder, Kuck, and Johnson have obtained the best results using the method of Khotinsky and Melamed, with slight modification, to prepare a series of ali- phatic boronic acids. The borinic acids may be prepared by the action of aqueous HBr on trlalkyl borines, but are very easily dehydrated, giving R3B-0-BR3. Their esters may be obtained by careful oxidation of the borine, as shown above.
B. Properties: The aliphatic boronic acids are much weaker acids than their aromatic analogs, as evidenced by the fact that the latter can be titrated with standard alkali upoi the addition of mannitol, whereas the former cannot. They are not cleaved by reagents which cleave the aromatics (HgCl3, Br3-H30), but do resemble them in being cleaved by H303. They undergo autooxidation in the air very rapidly.
RB(OH)3 H*°* , ROH * H3B03 Hs°s' * = ^ °r a"yl air air, R = alkyl only.
They react quite differently with ammoniacal Ag30 also, giving the R-R hydrocarbon instead of R-H:
2n-C4H9-B(0H)3 + Ag(NH3)3 + 2H30 -> n-CeHie + Ag + 2H3B03 + 2NIL
An interesting change takes place if the aliphatic borer?-;...; acids are dried over P30s or concentrated H3S04;
RB(OH)3 -HoO, RBO
The produce ere called alkyl boron oxides. Although they b^v? the comj /vsvticn RBO, molecular weight determinations show ii.am to be trimeri - This suggests a cyc.lrc ':^t >. .er ara."' o^ous to the paraldehyde .-. r?hi-.a it is seen that the aliphatic corcnic acids show a defini'-e similarity to the al-phatio aldehydes^ {x.) Th ;; combine with mvv.ecular oxygen, (2) the}' .-.-duce aAQ!oaf.ar-?l j*g?0. (3) they undergo cyclization to a pVienberea ring structure. This may be attributed to analogous 3?:.ectrt*nio configurations.
To shew ::he distinction between s.rooa.tic and allpr.nu: o boronic acids 3 Gill more clearly, Johnson Van demp-in, aid Grummitt pr^pavec' benzyl-, t-butyl •■ ?■ furyl-, and'" 2-tb.i^ny2 - boronic acids md tested them with Ag(KH, ;."'.' and aii! :>iiat^ci. The first two a ere shown to react definitely as aliphatic, tbi last pair as aromatic - confirming the wel- knowr a^.'macio character of these groups.
III. Mechanism Studies
Mention was made on page 2 of the reaction of dry bromine with trlalkyl borines, and it was noted that the significant point of the reaction is that two alkyl groin™ *rP m^vp^ fVnm a
• •
.
'
--•
' ■ I
.
!
133
- 4 -
This dibromide formed (RB-Br3) must come from the action of bromine on the monobromide (RaE— Br) since the latter has been shown to be inert toward KBr.
These reactions may be considered as proof of the theory that many organic reactions take place through a mechanism in which one group acts as an electron acceptor, the other as a donor. Here, the electron shell of the B atom may be completed by an unshared electron pair of a Br atom. This would cause a surge of electron density in the direction B — ■> R, and increase the mobility of the potential alkyl anion; an irreversible a - y shift within the complex would complete the reaction.
' R 1 R
:Br: -* B:R j -» :Br:3:R f R:Br:
• • • • > •
; Br : R
L " J
Failure of HBr to effect cleavage of more than one group may be attributed to the diminuished acceptor activity of B after the compound R3B-Br is formed, due to an internal resonance effect:
R
• • • • •©
;Er:Br: + B:R * • « • • • •
R
+
R:B:3r: e ^ R:3::8r: R* •' R*
Br 2 undergoes coordination even with this weaker acceptor effective enough to produce sufficient mobility of the alkyl group, but HBr does not.
Investigation was carried further in this field by Johnscrj and Van Campen, who studied the results of oxidation by aqueous H303 and autooxid^tion in air:
R3B -^-U R3BOR ■ (°< > RB(OR)3 I %i
They concluded that these reactions also go through a coordina- tion complex, because they found that the reaction would be stopped at stage I in the presence of water, and would not go at all with the ammonate of the borine (R3B £-> NH3 ) .
A survey of the behavior of various alkyl boron compounds toward HBr, Br3, H303, 0S, etc., indicates a definite gradation of reactivities; R3B <R3BOH >RB(OH)3 , and this is nicely ex- plained by reference to resonance possibilities within the molecule, leading to lessened reactivity.
IV. U se_ of. _D_lazonlum Borofluorides in Synthesis f
Dunker, Starkey, and Jenkins have s.aown that the diazonium borofluorides may be readily prepared as follows:
Ar_NH3 + NaN03 + 2HBF4 > Ar-N3BF4 + Na3F4 + 2H30 (90-97$
00 yields)
They were interested in these compounds because when they are
■ '
- 5 -
134
obtained. More recently, Starkey has used these diazonium com- pounds for synthesizing molecules in which the nitro group is ortho or oara to a nitro, carbonyl, or similar group. For example, his method is included in Organic Syntheses for the preparation of p_- and o-dinitro "benzenes in yields of 67-82$ and 33-38$ respectively. His reactions are carried out according to the following equation:
N0a
NO.
NaNOg -i- Cu pow d e r
25
efficient stxrring
v
N03
+ Na + NaBF4
Bibliography:
Frankland,
Krause and
J. Ohem. Nitsche .
Soc.
Ber.
15,
363 (1862). 2784 (1921);
55, 1261 (1922). Johnson, Snyder, and Van Campen, J. Am. Chem. Soc, 60, 115
(1938) Michaelis and Becker, Ber,, 13, 58 (1880); 15, 180 (1882).
ibid.f 42, 3090 (19095*.
J. prakt, Chem,, 128
Khotinsky and Kelamed,
Konig and Scharr'nbeck,
Bean and Johnson, J. Am, Chem. Soc., 54,
Snyder, Kuck, and Johnson, ibid.
153 (1930). 443.5 (1932).
60, 105 (1941). G-rummitt, ibid, ,
0 111 (1938).
Johnson, Van Campen, ana
Johnson and Van Campen, ibid.
Dunker, Starkey, and Jenkins,
Starkey, ibid., 59, 1479 (1937*57
Organic Synthesis, Vol. 19, p. 40.
Kruger, Proc. Natl. Acad. Sci,, U. S., 26, 181 (1940).
60, 121 (1938), ibid., 58, 2308 (1936).
Reported by Royston M. Roberts January 7, 1942,
CARDIAC AGLYCONES OF THE STROPHANTHIDIN GROUP
135
W. A. Jacobs - Rockefeller Institute for Medical
Research R.C. Elderfield - Columbia University
Plants of the digitalis-strophanthus group contain glycosides which have a characteristic and powerful action on cardiac muscle. The nucleus, which is common to all the agly- cones of this group, is given below (l):
.0
HSC-C
\
C=CH
/
Recent evidence indicates an a,, p unsaturated lactone.
The following table lists the best known aglycones.
Aglycone |
OH |
Grouos |
Ca-OH/R* |
Rings A/B* |
R |
Plant Source |
|
Digitoxigenin |
3, |
14 |
trans |
cis |
CH3 |
Digitalis |
|
Thevetigenin |
3, |
14 |
cis |
cis |
CH3 |
Thevetia |
|
Uzarigenin |
3, |
14 |
cis |
trans |
CH3 |
Uzara tree |
|
Digoxigenin |
3, |
11, 14 |
trans |
cis |
CH3 |
Digitalis |
|
Gitoxigen |
3, |
14, 16 |
trans |
cis |
CH3 |
Digitalis |
|
Periplogenin |
3, |
5, 14 |
trans |
cis |
CH3 |
Periploca |
|
Sarmentogenin |
3, |
11, 14 |
trans |
cis |
CH3 |
Strophanthus |
|
Strophanthidin |
3, |
5, 14 |
trans |
cis |
CHO |
Strophanthus |
|
* |
Probabl |
e structures |
In the glycosides the sugar residue is linked through the
C3-0H.
In 1934 Tschesche elucidated the cyclopentanoperhydro- phenanthrene ring system by selenium dehydrogenation of mono- anhydrouzarigenin to Diels' hydrocarbon (II) and also by degrada- tion of the same material to etioallocholanic acid (II!
COpH
ft
III
- 2 -
Correlation of the above results principally by W. A, Jacobs from infer the complete structures of
with the extensive work, 1922 on, made it possible to the more important aglycones.
136
The highly important and characteristic lactone group may be titrated with alkali at moderately elevated temperatures. It absorbs one mole of hydrogen to give a saturated lactone. The 3, Y unsaturation was inferred from the behavior of the agly- cones on treatment with sodium nitroprusside (Legal* s test) and ammoniacal silver solution (Tollens* reagent) which resembled
that of ^^,V angelica lactone (V) and not that of ^ a» ^ angelica lactone (IV).
CH=^CH
H3CCH C=0 ^ 0'
CH » H3CC
IV
so'
CK3 i
c=o
Therefore, the side chain appeared to be the lactone of an enolized aldehydo acid. Upon saponification and then acidifi- cation or merely by treatment with alcoholic alkali without saponification, a characteristic rearrangement of either the aglvcone or the unhydrolyzed glycoside to a saturated isomer (VI) takes place. Jacobs inferred that it was preceded by a stereochemical inversion on C17 by which the lactone and the C14-OH, originally trans, assumed a cis relationship.
rt, ^ 9 ?H2
c=o
KOH
X
A
CH-
.0=0
OK
CH
/wK
CH CH:
0-4-CH c=o
-o"
VI
The structure of the isomer (VI) is well established.
Evidence for the positions of the other functional groups of strophanthidin follows:
C14-0H. One molecule of water is split out on mild treat- ment (alcoholic HCl) of strophanthidin. The anhydrogenin cannot be isomerized. The C14 position alone allows the hydroxyl to be both tertiary and y or J with respect to the aldehyde group of the side chain,
C3-0H. The secondary hydroxyl has been located at C3 by degradations involving the opening of ring A,
. ...
■
137
- O -
C10-CHO, Wolff-Kishner reduction of a strophanthidin derivative gave a derivative of periplogenin. Conversion of this to a derivative of digitoxigenin was accomplished as follows:
.OH
HO
OH
CrO:
y^\ .,
0 OH
Hft0
^N^X/
H.
/*
CH
r
0
Digitoxigenin has been degraded to etiocholanic acid (a stereoisomer of III).
C5-OH. A second tertiary hydroxyl has been demonstrated. The extremely easy dehydration of the hydroxyketone in the previous transformation suggested a § hydroxyketone. To be both tertiary and p to C3 the OH must be attached to C5.
Recently Elderfield has considered the synthesis of cardiac
aglycones by transformations of other steroids. His attempts to
f*> v prepare model j=> substituted jx* unsaturated lactones have
thrown new light on the structure of the lactone ring in the
aglycones. His more successful syntheses made use of the fact
that half-ethers of primary-tertiary glycols are converted
smoothly into aldehydes when heated with acids;
CsHlxMgBr + CH3OCH3CN -* CeHj. iCOCHsOCHg
CsHnCOCHaOCHa + BrCH3CC2C2H5
KHSOj
SweT
^6^1 i^H
-* C6H11C(OH)CH2C03C3H£
CeHnCH CHO
VII
a^y-Saponff^- CH30CH3
tion
CH
N/
saponification then HBr
HOCH CO , .
^0s C6H5-C(OH)-CH C6HnC
VIII 1 t + I
.CH
C=0
CH.
CO
o-
IX \ HBr * X
It was not possible to dehydrate VII to an unsaturated lactone.
The following considerations have led Elderfield to
suggest that the lactone in the aglycones is really /£ unsaturated:
138
- 4
The comparison of the lactones of Keto acids (angelica lactones) with lactones of aldehydo acids is not justified. X and strophanthidin gave identical color reactions with a modified
6 Y
Legal test and with ferricyanide while £v ' angelica lactone
differed remarkably. £\a>fc angelica lactone gave only a slow response to Legal ' s reagent.
Representative aglycones, X, and ethyl crotonate have similar absorption spectra. Vinyl acetate (comparable to
Z\ lactones) differed greatly.
Hydroge nation of X and the aglycones yielded exclusively the saturated lactones. When the double bond is at the point of
lactonization (as in the A^' type), varying amounts of desoxy-acids are obtained.
The action of alkali on X parallels that noted in aglycone derivatives.
^6^1 i~C
-CH
CH-
0
CO
KOH in
CH3UH
ao. NaOH
Cs^i iGH
CHO
CeHn-C:
■CHa
COOH
=CH
CHoOH COCNa
The isomerism of the glycosides and aglycones by alkali may be formulated as follows:
=rCH
CH3 / CO
/CH
a:
c-
It
CH
' o'
CH3 I
c=o
CH.
OH
/
/
0
A
CH — CH;
I ! CH CO v 0'
Here the lactone is assumed cis to the C14 hydroxyl and no in- version on C17 is necessary. The observed failure of the gly- coside uzarin and of allostrophanthidin to isomerize is probably due to a trans configuration. Allostrophanthidin is produced by the action of an enzyme found in strophanthus seed upon strophanthidin. Allostrophanthidin has been shown to be a diastereoisomer of strophanthidin for which C17 or C14 is responsible.
Bibliography:
G-ilman, Organic Chemistry. Wiley, New York, (1958) pp. 1314-1337,
Elderfield et al., J. Org. Chem. , 6, 260 (1941).
Chen, Robbins, and Worth, J. Am. Pharm. Assoc, 27, 189 (1938),
Reported by R. F. Phillips
• .•
139
PREPARATION OF NITRILES BY THE USE OF CUPROUS CYANIDE J. F. Koelsch, University of Minnesota
Cuprous cyanide has been used as a reagent to produce nitriles
by the replacement of halogen atoms in aryl halogen compounds,
vinyl halides, allyl halides, and acyl halides. These reactions will be discussed in order.
The Rosenmund-Von Braun nitrile synthesis,
2 ArX + Cu2(CN)2 -> 2 ArCN + Cu2X3
is best known. Reactions reported in the literature vary widely in experimental conditions. Koelsch found that iodo- and bromo- diphenylindones mixed with cuprous cyanide and heated at 240-250°C« for three hours gave nearly quantitative yields of the nitriles. Fieser and Seligman found that when pyridine was added to a mix- ture of 4-chloro-7-methylhydrindene and cuprous cyanide the tem- perature necessary for reaction was lowered from 265° to 225°C, the time of heating could be cut from eighteen hours to two hours, and the yield of nitrile was higher. Newman has given the most detailed directions for the use of pyridine in the Rosenmund-Von Braun synthesis. Quinoline has also been used as a solvent to aid reaction.
Recently Koelsch and Whitney carried out a study to determine the mechanism of the transformation involved in this synthesis, and optimum conditions for the reaction. Though the reaction could be simply written, it was not free of complications. Reaction be- tween cuprous cyanide and p_-bromo toluene was only 15 per cent com- pleted after sixty minutes at 250°, but was 75 per cent completed in the next thirty minutes. This induction period indicated an auto-catalytic reaction.
The catalyst might be the aromatic nitrile, forming a complex with cuprous cyanide and bringing the solid into solution. To test this assumption p_-tolunitrile was added to the reaction mix- ture. Small amounts increased the rate of reaction; larger amounts were less advantageous. The induction period was not entirely eliminated.
The authors believed that peroxides might act as anticatalytic agents. These if present at the start of the reaction would be destroyed during its course. Accordingly, small amounts of hydro- quinone were added. The induction period was almost doubled.
. .
-2-
140
This suggested the use of an oxidizing agent. Cupric sulfate was found to have a pronounced catalytic effect. The actions of both cupric sulfate and p_-t©lunitrile on the reaction of p_- (C6H5 )3CHC6H4Br with two equivalents of cuprous cyanide are shown in the graphs below. The rate of reaction for this compound without catalysts is also shown, illustrating the induction period,
10Q
<h P O
cti
CD
U
p pn
c •
o>
o
Jn
CD Ph
0 I.
100 r
/
0 90 180 270 *60
Time, minutes
p_-(C6H5)2CHC6H4Br 2 equiv. CuGN
P |
||
O |
||
ctf |
||
CD |
||
u |
||
p |
50J |
/ |
/ |
||
O |
/ |
|
i I |
/ |
|
ft |
s |
, ,v |
0
0 0.2 0.4 0.6
C7H7CN or CuS04, grams
0.5 g. p_-(C6H5)3CHC6H4Br with I. p_-C7H7CN, 45 mins II. CuS04j 60 minutes
A mechanism was suggested to explain the catalytic effect of cupric sulfate in the Rosenmund-Von Braun synthesis:
++
ArX + Cu
(ArX ->Cu)++ + Cu+
(ArX ->Cu)
Ar+ + CN"
+
(ArX-^ Cu)++
(ArX ->Cu)+ + Cu++ Ar+ + CuX ArCN
This mechanism rests on assumptions that; (l) The aryl halide does not form a complex through the halogen atom with cuprous copper; (2) The cuprous complex can decompose to give an aryl ion
The study led to the formulation of a practical synthetic procedure. To a mixture of an aryl halide with excess cuprous cyanide is added a few drops of tolunitrile, and a little cupric sulfate* The whole is placed in a bath at 250°C. Completion of the reaction, as indicated by marked dimunition in volume of solid copper salts and formation of a dark liquid phase follows rapidly (ten to thirty minutes).
-3-
14 i
The nine compounds used in Koelsch' s experiments were rated as to relative reactivity with cuprous cyanide by measuring the time required for fifty per cent of a substance to react with cuprous cyanide. The order of reactivity was: p_-bromotriphenyl- methane < m-bromotoluene <p_-bromobenzophenone / p_-bromotoluene <^ bromobenzene <_ bromomesitylene / cc-bromonaphthalene / p_-bromobenzoic acid.
Many examples in the literature show that bromine or iodine is much more easily replaced by the nitrile group than is chlorine.
The reaction of vinyl halides with cuprous cyanide should be similar to the reaction of aromatic halogen compounds. Koelsch found that triphenylvinylbromide heated two hours with cuprous cyanide at 240° gave a quantitative yield of triphenylacry lonitrila. However, cc~p_-bromophenyl-|3, p-diphenylvinylbromide under similar conditions gave only a black resin and at lower temperatures did not react.
Allyl cyanide has been made by shaKing allyl chloride for eight days with a concentrated solution of potassium cyanide. The yield was ten per cent. Bruylants obtained a 95 per cent yield by refluxing allyl iodide or bromide one hour with a five per cent excess of cuprous cyanide,
ct-Ketonitrile s have been prepared from the lower members of the fatty acid series in sixty to eighty-five per cent yields by refluxing the acyl bromide and cuprous cyanide, without solvent, for one and a half to two hours. This is the method of Tschelinzeff and Schmidt. Previously silver cyanide had been used with acid chlorides. An autoclave was necessary, and there were many side reactions.
Bibliography:
Koelsch and Whitney, J. Org. Chem,, 6, 795 (1941). C. F. Koelsch, J. Am. Chem. S0c . , 58, 1528 (1956). Fieser and Seligman, ibid. , 58. 2482 (1936). Newman, ibid, . 59, 2472 (1957).
Bruylants, Bull. soc. Chim. Belg., 51, 175 (1932). Tschelinzeff and Schmidt, Ber., 62, 2210 (1929).
Reported by A. V. Mcintosh, Jr. January 14, 1942
•
142
the reduction of multiple carbon-carbon bonds
Because of the ease of replacement of the hydrogen atoms of acetylene by sodium or by the MgX group, substituted acetylenes m?.j be readily prepared in a pure state. These should afford a source of the corresponding olefins, provided that a satisfactory reagen' or catalyst for partial reduction can be found.
The ordinary catalysts, nickel, palladium, and platinum, in their various forms, have been used successfully for this process Zalkind has reduced a great many substituted acetylenic glycols using the noble metal catalysts and in all cases has found that there is a definite break in the curve at half-reduction. Campbeli and O'Connor have continued some of the earlier work of d'uPont us:;;;
Raney nickel on variously substituted acetylenes. Their results may be summarized as follows:
Type Half-reduction
R-C=C-H Very slight change in slope
R-C=C-R» Noticeable change, especially if R = R-
Ph-C=C-H and Ph-C=C-Me No change
Ph-C=C-Ph Stops at one mole
Analysis and Raman spectra indicate that there is better than 99.5$ olefin at the half-way point.
Less familiar methods include the use of chromous chloride, "Raney" iron, and sodium in liquid ammonia. The iron catalyst, prepared exactly as Raney nickel, was developed by Paul and Hilley and was found to be almost specific for triple bonds.. It failed in only a few isolated cases;, for example, tolane is completely re- duced to diphenyleth-.ne in the presence of this catalyst . Thompson and Wyatt have extended the use of the catalyst to the reduction of various alkynes, enynes, and dienynes.
Alkynes
OK OH
' _ / o
CH3-C-C=C-C~CH3 150 C 80$ olefin
CH3 CH3 14°0#
H i
o
H-C^C-C-OH 100 C 2nd mole of H enters
1000# one fourth as fast as
CH
, i . first
(CH3)2
EnyneS QH 100°C
CHa=C-C=CH 1000# 50$ isoprene
R 100°C
R-CH=C-C=C-R 1000// Stops at diolefin
- 2 -
Dienynei
r_C=C-C=C-C=C-R
H H K H (R is GHa or C6H1X)
100 C 1000#
'f ?
R-C=C-C=C=C-C~R
k
H
U
143
Sodium in liquid ammonia is a reducing agent developed by Lebeau and Picon and has been used to prepare olefins, especially trans olefins, by Campbell and Eby. Several trans octenes, hexenes, and decenes as well as butadiene have been prepared from the corres- ponding acetylenes by this method. They recommend it, incidentally, as the most satisfactory for obtaining terminal ethylenic compounds,
may be
As regards the stereochemical course of reduction reactions, it said that in general catalytic methods lead to cis forms whereas chemical methods produce trans modifications. Bourguel, G-redy, She rr ill and Zalkind have all used metal catalysts and have obtained cis compounds in 90-100$ yields. The melting points of known compounds and the Raman spectra data on the unknowns have been offered as evidence.
Interesting work in this field has been done by Schroter, Ott, and Farkas. Shroter found that by using the same batch of catalyst Over and over in the reduction of ethyl acetylenedicarboxylate, the ratio of ethyl furuarate to ethyl maleate gradually increased. Changing the amount of catalyst or the introduction of a catalyst poison such as carbon monoxide or hydrogen sulfide diminished the rate of hydrogen absorption but did not change the nature of the product. This led Ott to postulate in a later paper that the production of the more labile form is favored by an increase in the activity of the catalyst. He recognized that the velocity of the reaction on a catalytic surface is not measurable in terms of hydrogen absorbed. He assumed that the activity of catalysts is similar in nature to the reduction potential of the ordinary re- ducing agents. A metal of low potential should give a stable trans form, one of higher potential would give more of the cis isomer, and one high in the electromotive series which gives hydrogen of extremely high potential would carry the reduction all the way to the ethane. Propiolic acid was reduced chemically by the following combinations with the results indicated:
100% trans cinnamic acid
Mn-Alkali 50% cis. 4^ trans, and an oily mixture Mg, Na, etc. gave mixtures of the saturated hydro-
Cr-HCl
Zn-alkali 90^ trans 50% cis.
cinnamic acid and unreduced propiolic acid.
Thus between zinc and manganese lies the critical area, within which ordinary catalysts probably belong as far as reduction potential or activity is concerned.
Farkas and Farkas have also reviewed the literature and summarized the results rather briefly. Catalytic methods add two hydrogen atoms simultaneously, thus giving a cig form; nascent
- 3 -
M.
hydrogen on the other hand adds stepwise giving the opportunity for
able
modification to result form is the
and with most oiefinic
the fum^roid form the
the more
pairs the malenoid form is the more labile,
more stable. This may be carried further into the reduction of
olefins to ethanes and in general the me so form resembles the cis
and the racemic modification the trans. Most reductions conform
to the following scheme:
Modification reduced
C=C
C=C (cis)
C=C (trans)
If the temperature is catalytic reduction m- modification.
Product from Nascent hydrogen Catalytic hydrogen
trans dl-pa ir me so
CIS
me so dl-p^ir
too high or if enolization is possible, the: y lead partially or totally to the trans
So far, the work on these methods of partial reduction has been confined to the laboratory only. The industrial research in this field has been limited almost entirely to attempts to prepare butadiene from vinylacetylene. Most important patents are to du T ••"» Pont for the catalytic reduction of vinylacetylene, to Jasco for reduction by metallic zinc in alkaline solution, and to I. G-. Far- benindustrie for the reduction by sodium and zinc in the presence of butyl naphthalene sulfonate, Hurukawa has published results in- dicating a 60/6 yield of the diene from an electrolytic reduction at a palladium black cathode* There is also a process for the semi- reduction of most of the common acetylenes by means of palladium on clay and there is another patent for the reduction of several substituted dienynes to mono-olefins in the presence of nickel.
Bibliography
Zalkind, Vishnyakov and Morev, J. Gen. Chem. Campbell _.nd O'Connor, J. Am. Chem, Soc, 61, Campbell and Eby, ibid, . 63, 216 (1941). Thompson and Wy:-.tt, ibid. . 62, 2555 (19*0).
(USSR), 3, 91 (1933). (1939).
2897
Sherrill and Launspach, ibid.
2562 (1938). rend.. 157, 137 (1913).
Lebeau and Picon, Cou4pt. Bourguel, ibid., 180, 1753 (I925~T7 Paul and Hilley, ibid., 206, b08 (1938); Bull. Soc. Chim. [5] 6, 218 (1939).
[5] 2, 1029 (1935). er.. 60B, 624 (1927).
67B, 1669 Farkas and Farkas, Trans. Far. clay Soc, Hurukawa, J. Electrochem. Soc, Japan,
G-redy, ibid.
Schroter,
Ott, B^rth and Clems er, ibid. .
USP
FP
(1934). 33, 887 (1937). 7, 346 (1939). 2,207,070 cf. C.A. 34, 7932' (1940} 1,920,242 27, 4818 (1933).
2,167,067 33, 8625 (1939
834,111 33, 3393 (1939
837,196 33, 6872 (1939
Reported by John C. Robinson, Jr. January 14, 1942
. .
145
POLYENES AND CUMULENES Richar* Kuhn and Coworkers - Kaiser Wllhelm Institut
REVIEW OF POLYENES
Kuhn defines 'polyene' as referring to those substances th contain many ethylenic linkages in open chains. It is usually understood that the double bonds are conjugated.
Methods of Synthesis. —
Type A - C6H5£CH=CH3-nC6H5 (n = 1,2,3,4,5,6,7,8,11,15)
1. Treatment of hydrobenzoins with phosphorous iodide (P2I4):
2 C6H5CH=CH-CHO -^~* C6H5CH=CH-CH-CH-CH=CHC6K5 5 4 >
6h OH
C 6H 5CH=CH-CH=CH -CH=CHC 6H 5
This method is suitable for 1,4- and l,6-glyc«ls also. They are prepared by the action of BrMgC=CMgBr and BrMgC=C-C=CMgBr on polyene aldehydes.
2. Condensation of polyene aldehydes with h00C-CH2£CH=CH3ftCKa- COOH (n = 0,1,2) using lead oxide:
C6H5CH=CH-CHO + CH2— CH2 + OHC-CH=CE-C6H5 ->
COOK COOK
C6Hs-CH=CH-CH=CK-CH=CH-Ch=CH-C6HB
3. Action of benzylmagnesium chlorides on pnlyene aldehydes:
(a) C6HBCH=CH-CHO + CH3-CH=CK-CHO H|||r|dine )
C6HsfCH=CH3.5CH0 and C6H5f CK=CK9-7CH0
(b) C6H5-£CH=CH}nCHO + C6H5CH2MgCl -> C6K5fCH=CK}-nCHOHCK2C6H5 -> C6HB£ CH=CH}-n+1CeH5
4. Coupling of thio- and seleno-aldehydes:
C6H5£-CH=CH]-CHO HsS(H3Se)> C6K5£-CH^CH1 CHS °U> CaC°3 >
piperidine
CeHsfUH-CH^^j^CeKs
-2- 1 46
In this manner, C6HS£CH=CH3-15C6H5 was prepared. It is the highest known member of the polyene series. The solid is green- ish black and its solutions are violet-red.
Type B - CH3£CH=CH}nCH0 (n = 1,2, 3,4, 5, 7, 9) and
CH3-ECH=CH>nCOOH (n = 1,2,3,4,5,6,8)
Aldehydes, chiefly crotonaldehyde are condensed by means of piperidine acetate to give higher polyene aldehydes, CK3£CH=CH}nCHO. These aldehydes will condense with malonic acid:
—CO CH3fCH=CH}nCHO + CHa(COOH)a -> CH3-£CH=CHi-nCH=C(C00H )a ■ s >
CH3-ECR=CH}n+1C00H
By this method the total synthesis of stearic acid and cetyl alcohol was effected.
Type C - CH3£CH=CH}nCH3 (n = 1,2,3,4,6)
Polyene aldehydes react normally with alkylmagnesium ha- lides to form the carbinols. These are dehydrated by treating with a l-2# solution of p_-toluene sulfonic acid in ether:
-Ha0 CH3£CH=CH3nCH0 + CH3CHaMgBr -► CH3GCH=CH^CH0HCH3CH3 Z >
CH3fCH=CH9n+1CK3 (n = 2, m.p, 52°)
Type D - H03C£CH=ClQnCOOH (n = 1,2,3,4,5,7) and H03C-CH3-£CH=CH3nC00H (n = 1,2,3,4)
1. Claisen Condensation: EtOaC-C03Et + CH3tCH=CK^nCOaR
Et02C-C0CHa£CH=CH3nC0aR
pyridine AcaO
Et02C-C=CH£CH=CH4 COaR A1» H£ > OAc
Et03C-CHtCH=CH3 CHaC05R R'OH, NaOH >
OAc
R'OaCiCH=-CHan^C03R»
147
- 3 -
The free oxalo-polyene carboxylic acids are obtained by saponification. Treatment with hydrogen peroxide removes carbon monoxide :
H02C-COCK2-CH=CH-COCH K3°s » H02C-CH2-CH=CH-C02H
2. Eiological Oxidation:
Kuhn discovered that if polyene mono-carboxy acid amides were fed to rabbits, the mono-amide of the corresponding dicar- boxy acid would be obtained from -the rabbit's urine:
CH3£CH=CH33C0NHa -> H02CtCH=CH33C0NH2
Yields of 20-80 per cent have been obtained.
CUMULENES
Cumulenes are compounds containing an uninterrupted series of double bonds.
Methods of Synthesis: —
The first method used by Kuhn was the treatment of acetylene or diacetylene glycols with P2I4:
(c6h5)2c-c=c-c=c-c(c6h5)2 PgI*> (C6H5)2C=C=C=C=C=C(C6H5)2
(0.2-0.3^)
OH
/
OH
The yields were greatly improved by treating (I) in ether solu- tion with dry HC1 and VC12 or CrCl2. Yields of around 90 per cent were obtained. In this manner the following cumulenes were pre- pared:
Ryy
\
\R
_. C— C— C=?C— C— Cx ■ n
H R = CI OH,
=c=c=c=c=c
148
- 4 -
Attempts were made to prepare unsymmetrical cumulenes in order to test out the van't Hoff theory. Unsymmetrical benzo- phenones were treated with 3rMgC=CMgBr and the corresponding acetylene diols obtained. These, however, could not be converted into the butatrienes.
Kuhn attempted to prepare cumulenes from diphenylketene by the following reactions:
2(C6H5)2C=C=0 + BrMgC=CMgBr -> (C6H5 ) 3C=C-C=C-q=C (C6H5 )
II
OH
HG1
CrCl3
( C6H5 ) 3C=C=C=C=C=C ( C6H5 ) 3
Instead of (II), however, he obtained a diol of some other struc- ture, as yet undetermined*
Properties of Cumulenes ~.na lolyenes :
As compared to the corresponding saturated compounds, polyenes and cumulenes are more highly colored, have higher melting points and lower solubility, and have absorption bonds in the higher wave lengths.
Cumulenes are much more highly colored than polyenes and have higher melting points. They give a negative Baeyer test, do not react with maleic anhydride, and give no color reaction with C(N03)4. They are destroyed by K202 and are readily reduced.
Bibliography:
Richard Kuhn, Angew. Chem., 50, 703 (1937). A review. Richard Kuhn, J. Chem. Soc., 605 (1938). A review. Kuhn and Wallenfels, Ber., 71B, 783 and 1510 (1938). Kuhn and Platzer, Ber., 73B, 1410 (1940).
Reported by Stanley B. Speck January 21, 1942
149
REACTIONS OF PYRIDINE
Pyridine is a well-known organic base which occurs in small amounts in coal tar. It is a weaker base than ordinary tertiary amines but does form soluble, stable salts. It may be considered more aromatic than benzene, as shown by the oxidation of quinoline, and may be compared in many of its reactions and properties to nitrobenzene.
It is, however, more easily reduced than benzene. The selective reduction of phenyl and benzyl derivatives of pyridine has been accomplished by sodium and alcohol and by hydrogenation over Raney nickel. The salts may be reduced by Adams1 platinum catalyst. The ease of reduction is quite different for isomeric alkyl or aryl derivatives. Thus it has been reported that 4- phenylpyridine is not reduced with Adams' catalyst under conditions which effect reduction of 4-benzylpyridine. With Raney nickel the following is the order of decreasing ease of reduction of the pyridine ring: 2-benzyl, 2-methyl, 2-phenyl, pyridine, 4-phenyl. The reduction and alkylation of pyridine to methylpiperidinium formate by refluxing a mixture of methanol, formic acid, and pyridine has been reported.
Pyridine undergoes a number of substitution reactions, few of which may be carried out except under strenuous conditions. In general, the product formed by chlorination, bromination, iodination, sulfonation, or nitration is the three or the three- five derivative. However, conditions and catalysts have a pro- nounced effect upon the nature of the product. Under the strenuous conditions necessary to obtain any substitution at all it is diffi- cult to prevent the simultaneous formation of some di- and poly- substituted derivatives. The mechanism of the substitution of pyridine is not known but sufficient data are at hand to show that it is not a simple one.
The sulfonation of pyridine is best carried out in the presence of mercuric salts with fuming sulfuric acid at a tempera- ture of 270°. The main product is pyridine-3-sulfonic acid, al- though some of the 2-derivative is reported. No di sulfonic acids are produced. The compound (I) has been prepared by treatment of
S02 CgHsN' j (I) ^0
pyridine with (S03)3 and has been proposed as an intermediate in the sulfonation reaction. A few reactions of (I) which show the nature of the problem encountered are listed* It is noteworthy that one of the best methods for the iodination of pyridine in- volves the use of fuming sulfuric acid as an oxidizing agent. In this case the predominant product is 3-iodopyridine.
- 2 -
150
90%
NaOK
cold
SO,ONa
C^fUN/
'SllB
'oh
II
CH=CH-N-S02ONa CH=CH-CHOH
CH-CHO
H CH-CH=CHONa
V 79$
H20 ^ u-
50
CH-CH=NS02ONa CH-CH=CHONa
A
IV
I
c-
V
CH- VI
80$
-> C5H5N + Na2S04 60$
V
-CH=N
= CH
41$
If pyridine is chlorinated at 170-220 , 3- and 3,5-deriva- tives are formed while at 260-420°, 2- and 2, 6-derivatives are formed. When bromination is carried out in the absence of metallic salts, some of which tend to lower the temperature necessary for substitution and affect the course of the reaction, definite transition ranges are noted. Thus at 300° the 3- and 3,5-deriva- tives are formed; at 400° a mixture, the 2-, 3-, 2,6-, and 3,5- derivatives are obtained; while at 500° the product is mainly 2- and 2,6. It is apparent from these facts that some sort of a polar mechanism as opposed to a free radical type mechanism must apply and that possibly the character of the nitrogen atom is changed when the higher temperatures are reached. The formation of 2-chloropyridine by treatment of pyridine with sulfuryl chloride has been advanced as evidence that substitution in this position takes place on some sort of an N-oxide or peroxide derivative.
Nitration of pyridine produces practically negligible yields of 3-nitropyridine but if the pyridine ring is substituted it is possible to obtain the nitro compounds in yields as high as 90 per cent.
o When pyridine is heated with sodamide to 160-180 2-amino- or 2,6-diaminopyridine is obtained depending upon the ratio of the reactants. This reaction may involve an addition of the sodamide across the C=N linkage.
Treatment with mercuric acetate produces only the 3-mercuri- derivative under mild conditions and tars under more strenuous conditions. An intermediate, probably VII, is first formed.
0C0CH3 CSH5N / VII \HgOCOCH3
151
- 3 -
In general the orientation effects of groups already present are the same as the same groups in nitrobenzene and good yields of the substitution products may be obtained if the substituent is one which activates the ring.
Cases of 1-2 and 1-4 addition to the bond system of pyridine are encountered when pyridine is treated with phenylmagnesium bromide or phenyl lithium.
Since all three derivatives are isolated in most cases, free radical substitution probably does occur when pyridine is treated with diazotized aromatic amines in alkaline solution*
The comparison to nitrobenzene may be extended to the activity of the groups' already present, Thus halogen in the 2- or 4-position is activated much the same as in p_-chloronitrobenzene, Methyl or alkyl groups in these positions undergo active methylene condensations with aldehydes and esters and this method may be used to separate p-picoline from the mixture which is obtained from coal tar in the same manner that m-nitrotoluene is separated from the p_- and p_-isomers.
The treatment of amino pyridines with nitrous acid gives unstable diazonium salts which behave in replacement reactions much like the corresponding benzene derivatives. The conversion of 5-chloro-2-aminopyridine to 5-chloro-2-nitropyridine in 40^ yield by treatment with sulfuric acid and hydrogen peroxide has been reported.
The other amino pyridines may be obtained from the corres- ponding halogen compounds by treatment with ammonia in an auto- clave. 3-Aminopyridine may be obtained by the Hofmann degradation of nicotinamide and 4-aminopyridine is obtained by treatment of 4-pyridyl-N-pyridinium dichloride with ammonia (VIII).
S0C1
CI I
2C5H5N bU"15) y9 ^N — ^ \\N-HC1 + SO
VIII 94^ Some reactions of VIII are shown:
•4-aminopyridine + pyridine 41%
VIII — > 4-hydroxypyridine + pyridine
150° 84%
\
\ CfiH50K
4-phenoxypyridine + pyridine 56%
152
- 4
Pyridine-2-sulfonic acid has been produced by oxidation of the thiol obtained by treatment of 2-bromopyridine with sodium hydrosulfide. The sulfonic acids may be fused with alkali, metal hydroxides, or cyanides to produce the corresponding hydroxide or cyanide.
The hydroxy derivatives raay be produced by tiia&oti nation of the amine, by alkali fusion of the "sulfonic, acid 'or by hydrolysis of compounds of the type VIII.
The preparation of nicotinonitrile by the heating of 3- bromopyridine with cuprous cyanide has been reported. Nicotinic acid may also be produced by the treatment of 3-bromopyridine with butyl lithium followed by carbonation and hydrolysis, a process which is reported to produce yields as high as 70 per cent.
The treatment of 2-bromopyridine with magnesium apparently produces a polymer to which structure IX was assigned. It was
''/-Br
MgEr
possible to produce phenyl-2-pyridylcarbinol in 50 per cent yields by treatment of this polymer with benzaldehyde.
A table showing the yields and isomers obtained by some of the substitution reactions of pyridine is included.
TABLE
Reaction |
Product |
% yield |
Remarks |
Ref |
chlorination |
2-chloro 2,6-dichloro |
46 1 |
copper tube 270^ |
10 |
bromination |
3-bromo 3,5-dibromo polybromo |
39 12 4 |
mercury salts pumice 300° |
13 |
iodi nation |
3-iodo 3,5-diiodo |
18 |
fuming sulfuric |
9 |
sulfo nation |
3-sulfonic acid |
40 |
mercuric sulfate 230° |
7 |
nitration |
3-nitro |
3 |
18# fuming HN03 160-180° i i |
14 i |
- 5 -
Table (Cont) Reaction
Product
% yield
Remarks
153
Ref.
amination
phenylmagne sium "bromide
ethylmagnesium bromide
phenyl lithium
sodium benzene diazotate
benzyl chloride
mer curat ion
2-amino 2,6-diamino
2-phenyl
2-ethyl
2-phenyl
2-phenyl 3-phenyl 4-phehyl
2-benzyl 4-benzyl 2,6-dibenzyl 2,4-dibenzyl
3-mercuri
50 51
44
40-50
24 10
10
50
trace trace trace
49
autoclave
autoclave
110 -toluene
benzyl chloride heat and copper powder
160-180
15 16
18
18
19 20
17
Bibliography
Soc, 56, 2425
1. Adkins, Kuick, Farlow and *tfojick, J. Am. Chem (1934),
2. McElvain and Bailey, ibid. . 52, 1633 (1930).
3. Adams and Hamilton, ibid. , 50, 2260 (1928).
4. Overhoff and Webarit, Rec. trav. chim. , 50, 957 (1931).
5. Mayo, J. Org. Chem., 1, 496 (1936).
6. Baumgarten, Ber., 59B, 1166 (1926); 72B, 567 (1939).
7. Machek, Monatsh, 72, 77 (1938).
8. van Gastel and Wibaut, Rec. trav. chim., 53, 1031 (1934).
9. Rodewald and Plazek, Ber. , 70B, 1159 (19377.
10. Wibaut and Nicolai, Rec. trav. chim., 51. 709 (1930).
11. Colonna, Chem. Abstracts, 34, 7290 (1940).
12. McElvain and Englert, J. Am. Chem. Soc, 51, 863 (1929).
13. Wibaut and Kertog, Patent, Chem. Abstracts, 27, 4542 (1953).
14. Plazek, Ber., 72B, 577 (1959).
15. Wibaut and Dingernanse, Rec. trav. chim., 42, 240 (1923).
16. Shreve, Reichers, Rubenkoenig and Goodman, Ind. Eng. Chem., 32 173 (1940),
17. Shreve, Skeeters and Swaney, ibid., 32, 360 (1940).
18. Bergstrom and McAllister, J. Am. Chem. Soc, 52, 2845 (1950).
19. Organic Syntheses, 18, 70 (John Wiley and Sons, New York, 1958)
20. Haworth, Heilbron and Hey, J. Chem. Soc, 1940. 349.
21. J. Gen. Chem., 10, 1101 (1940) (USSR).
22. Koenigs and Greiner, Ber., 643, 1049 (1931).
23. McElvain and Goese, J. Am. Chem. Soc, 63, 2283 (l94l).
24. JSilman and Spatz, ibid., 62, 446 (19^0).
25. Overhoff and Proost, Rec. trav. chim., 57, 199 (1938).
26. von Braun and Pinkemelle, Ber., 64B, 187i(l93l).
Reported by F. J. Wolf, January 21, 1942
SEMINAR REPORTS
II Semester 1941-42
Pa pre Epinephrine and Related Compounds: The Influence of Structure on 1 Physiological Activity
K. E. Hamlin, Jr.
Attack at the Alpha Carbon Atom of Alpha, Beta-Unsaturated Carbonyl 9 Compounds
J. E. Mahan
Hydrogen Bonding by C-H 14
E. Ginsberg
Acidic and Basic Constituents of Petroleum 16
Billie Shive
Cleavage of the Alkyl -Oxygen Bond in the Hydrolysis of Esters 22
W. E. Blackburn
Some Reactions of Hydrocarbons in an Electrical Discharge 28
R. G. Chase
^ar Gases 34
C» F. Jelinek
Developments in Quantitative Organic Analysis 39
G. E« Inskeep
Recent ^thods for Making Acids 44
G. W. Cannon
The Prins Reaction 49
R. E. Foster
Chlorosulfonic Acid as an Organic Reagent 52
S. S. Drake
Some Derivatives of the Polyhydroxybenzenes 58
J. D. Garber
Relation between uptical Rotatory Power and Constitution of the Sterols 63
• H. Kaplan
Reduction of Carbonyl Groups to Methylene Groups 69
B. C. McKusick
The Replacement of Alkyl Groups during Nitration 72
s. I. Meltzer
Page Ketene Acetals 77
G. L. Schertz
Biphenylene 81
F. W. Spangler
The Constitution of Usnic Acid 85
C. W. Theobald
Alkyl Carbonates in Synthetic Chemistry 92
R. M, Roberts
Glycerol Derivatives 98
J. F. Shekleton
Determination of Branch Chain ^thyl, Citric Acid, and Organic "alogen 102
C. W. Smith
Recent Developments in the Identification of Organic Compounds 106
Q. F. S0per
Sulfadiazines 111
M. Chiddix
The Effect of Catalysts on the Grignard Reaction 116
A. B. Spradii.ng
The Chemistry of Explosives 121
R. S. voris and P. F. Warfield
Recent Developments in the Study of Vitamin E 133
F. W. wyman
The Acylation and Alkylation of the Sodium Enolates of Aliphatic Esters 138
John Whitson
EPINEPHRINE AND RELATED COMPOUNDS The Influence of Structure on Physiological Activity
Alt h o u gh r ap i d p r o g . ? ess ha s "been
m o rl
e, the relationship
ix'ucr.ure and it ill not
hiysiolo < leal activity of under stood. However,
between the chemical medi c ina 1 pro du 2 1 s i evidences that there is some such relationship are apparent in several well-known chemical series, such as the alcohols, amines, barbiturates, phenols, and "sulfas." The individual importance and purpose of each intimate group or radical within each series has yet to be determined- That this relationship is of value to discover, is essential 00 the chemist synthesizing products which combat disease. B;y such information, scientific approach can be made towards specific medicinal s where formerly success was often only by chance.
The earliest and best known study in attempting to correlate chemical constitution and physiological behavior was made with "pressors" or compounds which produce a rise in blood pressure. Interest in these compounds was first arcu sed when Oliver and Schaefer in 1894 found that extracts of the suprarenal glands produced a rise in blood pressure when injected into the blood vessels of animals. Immediately an intensive study of this gland in an effort to isolate its active principle was made by chemists, physiologists, and pharmacologists, credit finally going to Abel and his coworkers, who first isolated the hormone (as the poly- benzoyl derivative) in 1897. Once isolated, the formula for epinephrine was quickly established and this formulation confirmed by synthesis of the, hormone is as follows:
carried out by Stolz in 1904, and
OH
1 j/^N.0H ClCHaCOCl
/'
P0C1.
OH
/ns,oh
Q.h\ NH:
/<\Ch
"2
\
O
Al . Hg
C-CH3C1 0
y — C R 2 I'm liC II 3
0
CHOHCHs- NHCH3
Epinephrine
Since the levo compound is the only one used in medicine, the racemic mixture obtained is resolved through d-tartaric acid.
Biologically, epinephrine is an extremely active compound. It has been found to exert an effect on the isolated frog heart in dilution as low as one part in five billion. Its value as a drug is based on the effects that it produces. These are chiefly, (l) its action on smooth muscles to relieve asthma, hay fever and severe colds; (2) its hyperglycemic effect to convert muscle and liver glycogen into glucose; and (3) its pressor ac- tion to cause a high rise in blood pressure or locally to arrest hemorrhage.
2
While the interest in epinephrine was increasing, a substance isolated from putrified meat, acting as a pressor, was identified by Barger and Dale to be tyr amine . Their investigations on a group of compounds of putrefactive origin were reported in a classic which first shows this correlation of physiological character and structural similarity. Accordingly, when Chen in 1923 brought to light the medicinal virtues of ephedrine, a drug used by the Chinese for over five thousand years, workers in this field intensified their efforts to coordinate these physiological- ly related chemical compounds.
A comparison of the structure of the better known naturally occurring pressors (which follows) reveals that the p-phenyl- ethylamine skeleton is corrmon to all.
<^ "^> CHaCHaNHjs HOK^"^ CH3CKaNHa
p-Phenyle thylamine Tyr amine
/> CHOHCK3NHCE3 <f ^> CHOHCHCH3
NKCH3
Epinephrine Ephedrine
As i s noted, however, there are significant structural differences between each compound. How, then, do these differences affect their relative physiological behaviors? Such modifications of structure will next be considered.
It seems appropriate at this point to discuss very briefly the usual method by which such compounds are compared. Since pressor action is common to all of these substances, assay is made by administration to a test animal and recording the resulting rise or fall in blood pressure. As epinephrine is the animal hormone and has been most widely studied, it serves as the comparative standard during the assay. It must be realized that in this, as in other biological assays, variations occur as a result of many factors. Hence, in accumulating evidence from many sources, con- tradictions and inconsistencies are unavoidable. Thus, for comparison, the different results of a single experimenter are the most reliable. Also, in interpreting results, the relative activities ana toxicities must be compared before assigning a chemical compound a specific therapeutic value.
Considering now variation from the p-phenyle thylamine nucleus, Barger and Dale investigated a series of compounds in which the relative positions of the phenyl and amino groups were varied. Aniline was without effect, benEylamine was slightly active, a- phenyle thylamine more active, |5-phenyle thylamine had maximum activity, while Y~phenylpropyIamine was again much less active. They concluded, therefore, that, in such a series, optimum pressor
3
activity was shown by compounds having a phenyl group removed from an amino group "by two carbon atoms. Mere recently, these observations nave been substantially confirmed by other workers, thus establishing the ^-phenyl e thy lamine skeleton as essential for optimum activity.
Naturally occurring compounds of the type under consideration include tyr amine, epinephrine, and herdenine Qhordenine or anhaline is an alkaloid isolated from germinating barley having the formula:
HO<s /> CH3CH3N(CH3 ) 2 ]
primary, secondary, and tertiary amines, respectively then consider the variation in the amino group.
Let us
Barger and Dale in their early reports stated that for the non-phenolic compounds, me thylati on of the amino group makes no appreciable difference in pressor potency. However, a lengthy survey of reports uy more recent workers indicates that for both phenolic and non-phenolic substances of this type (with one exception), the secondary amines are much less active and more toxic than the corresponding primary amines. Introduction of the higher alkyl groups in place of methyl is a step in the wrong direction. Toxicity becomes higher and depressor action results. Two methyl groups also work to a disadvantage, as in the case of hordenine which is distinctly of less value than tyr amine. The effect of alkylating the amino group may best be summarized by Table I.
TA3LE I Effect of Substitution in Amino Group
Compound
Pressor Activity Relative Toxicity Epinephrine = 1
0CHSCH3NH2 (p-phenylethylamine ) 1/350 0CH3CH3NHCH3 1/350
Toxicity doubles on substitution
0CHCHCH3KH3 0CHOHCH3NHCH3
1/350 1/700
Toxicity about same
0CHOHCHCH3 (Propadrine)
•"'2
./80
§0% more toxic than primary amine
Table I (Conf )
£ (conf )
0CHOHCKCH3
N xl L* il 3 0OHOHCHCH3
N(CK3)3
0CHCHCHCH3 1
NHC3H5
0CKOHCKCH3 NHC3H7
1/95
1/600
1/150
Depressor
Toxicity progress- ively increases with alkylation
D HO
/-CH2CH2MHS (Tyr amine)
1/150
H0<^~ ~ />CH3CH3NHCH3 rlO^' ~^>CH3CHSN(CH3)
1/150
1/700
Toxicity increases with alkylation
H0<. ^> CHsCHgNHCsHs
1/200
.t of the investigations of it c^n be stated that the
A second principal difference in chemical structure in the naturally occurring pressors, is the presence of one or more phenolic hydroxy 1 groups. As a resu! Barger and Dale, Alles, and Taint er, intensity of action of phenolic pressor substances is much greater than that of the corresponding non-phenolic compounds. Although the ortho-hydroxyl has no apparent pressor influence, it does serve to increase the toxicity. The met a isomer increases the pressor activity as much as five times but also increases the tox- icity. Finally, the, para hydroxyl confers a pressor activity be- tween the ortho and me t a and, most important, it lowers the toxicity. Thus, it is not too surprising to find that the 3,4- dihydroxy compound indicates a maximum activity with a rather low toxicity. Further phenolic substitution only serves to decrease activity. With phenolic substitution, however, the action is more transitory, the duration of response being lowered. Table II summarizes such comparable compounds.
- 5
Compound
table ii
Effect of Phenolic Hydroxyls
Pressor Activity Epinephrine = 1
A
fCH8CHaNHa (I) <^~^X CE8GHaNH8
~7"
OK
1/550
1/350
Relative Toxicity
More toxic than
H0<^ ' ^- CE8CHaNH8 (Tyr. amine )
« ..!,— .-J.
^>— CH3CH8rJH8
HO
1/150
1/70
Less toxic than I
More toxic than I
L3^'-2-i'±2
H0<C ^>-0H»GHaNHs
HO
1/3!
Toxicity only slightly more than
H0<^ \^>-CH0H0H3^HCHa (II)
' — ** ( 3y n ep hr i n e ,
1/55
2?~ CEOKCHaNHOHa
( N e o t yn e p hr i n e )
HO
1/1
More toxic than II
H0<x x>-GHOHCHsNH0Ha
T~ (Epinephrine)
HO
More toxic than II
0CHOHCHCH3 (III)
"2
(it oo9 drine )
HG^ >*> - CEOKCHC!
1/80
1/50
1/5 £?.s t jyic as III.
- 6 - Table II (conf)
6
C (conf )
<^ ^>- CHOHCHGH3
/ NHa
HO
1/85
3 times as toxic as III
HO
<x x^~ CHOHCHCH3
NH3 (Cobefrine)
HO
1/4
l/lOO as toxic as epinephrine
In the use of epinephrine as a therapeutic agent, its chief handicap is its complete inactivity when administered by mouth. Since ephedrine, a propane derivative, is
orally active, is the length of the side chain responsible for this physiological difference? As studied by the Council on Pharmacy and Chemistry of the American Medical Association, phenylethanolamine is not active after oral administration, whereas phenylpropanolamine is. Chen, et al, in their investiga- tions on the ephedrine s attribute the oral efficacy of ephedrine to the presence of the third carbon of the side chain. Hartung and Munch from their results with the phenylpropylamines found that phenyl-l-amino-2-propane was orally active. Thus, such compounds as ephedrine, propadrine and benzedrine are all active when ad- ministered by mouth. However, further extension of the side chain gives results of a negative character. Thus, phenylbutanolamine is nearly inactive while phenylpentanolamine is a depressor with a marked toxicity.
Finally, in a discussion of these pressor substances, it i s noted that for some, a side chain hydroxyl group is in evidence. The effect of this group is not too well defined but all evidence points to an augmentation of the pressor activity and a lowered toxicity. An examination of Table III illustrates this rather clearly.
TABLE III Effect of -alcoholic Hydroxyl
Compound
Pressor activity
Epinephrine = 1 Relative Toxicity
0CHaCH3NHa 0CHOHCHaNKa
1/350 1/350
1/2 as toxic as simple amine
- 7 - Table III (Conf )
B
0CH2CHCH3 (Benzedrine)
NK:
0CKOHCKCK3 (Propadrine)
1/350 1/80
l/o as toxic as simple amine
C
HC
HO
<C~>c-
y
CHaNHCHj
3uil3
. IH
//
Ss
CHOHCHpNHCH.
1/150
1/35
Less toxic than simple amine
D
HO
1/35
HO <. /> CHOHCHgNHg
HO vArterenol)
dl-art. = 3 dl-epi. 2
dl-art. — 1
1-epi. 1.
Less toxic than s:
.e amine l/3 toxicity of epinephrine
The above discussions and the resulting conclusions may be summarized as follows:
1. The optimum pressor activity (production of a rise in blood pressure) is found in those compounds in which the aromatic nucleus and the amino group are attached to neighboring or adjacent carbon atoms, thus Ar-C-C-N.
2. The primary amines are more active and less toxic, generally, than corresponding methylated, secondary amines.
3. Substitution of hydroxyl groups in the 3, 4-positions of the phenyl nucleus confers optimum pressor activity.
4. Compounds with three carbon atoms in the side chain are much more active on the circulation after oral administration than are the homologs with only two carbon atoms in the side chain.
s
5. The alcoholic .hydroxy 1 group increases the activity and decreases the toxicity or both.
In applying this information, the following two compounds are ch&sen as representatives of the synthetic field to surplant perhaps in their individual usefulness, the natural products:
HO
CHOHCHpNH.
HO
HO<T >>• CHOHCHCH3
HO
NH:
Arterenol
Nearly as active as
epinephrine
l/o as toxic as epinephrine
Acts twice as long as
epinephrine
Cobefrine
Orally active
1/100 as toxic as epinephrine
Acts twice as long as
epinephrine
1/4 as active as epinephrine
Bibliography:
Earger and Dale, J. Physiol. 41, 19 (1910).
Chen, K.K., Wu, C. and Henri ks en, E., Journal of Pharmacology and
Experimental Therapeutics, 36. 363 (1929). Hartung, Iv.K., Chem. Rev., 9, 369 (1931) 251 ref. Hartung Munch, Miller, Crossley, J. Am. Chem. Soc, 53_, 4149
(1931). Tainter, Arch. Internat. Fharmac. Ther., 4JL, 365 (1931). Tainter, ibid., 46, 192 (1933). Alles, G-ordon a., Journal of Pharmacology and Experimental
Therapeutics, 47, 339 (1933). Alles, G-ordon a. and Knoefel, Peter K. , U. of Cal. Pub. in
Pharmacol., 1, 101 (1938). G-unn, J. a., Brit. Med. J., 2, 155, 214 (1939). Jenkins, G-.L. and Hartung, W.H., The Chemistry of Organic
Merlieinal Products, John S. Swift and Co., 274, (1941).
Reported by K. E. Hamlin, Jr. February 11, 1942
ATTACK AT THE ALPHh CARBON ATOM OF
alpha, beta-un saturated carbonyl compounds
a, ^-Unsaturated carbonyl compounds are well characterized by the addition reactions which they undergo with many different types of reagents. In a large number of cases addition occurs in such a manner that the elements H and A, of the reagent HA, finally become attached to carbon atoms 3 and 4, respectively.
4 3 3 1
c=c-c=o
+ HA
-C-CH-C=0
A
Kohler and his associates have definitely established that the initial reaction is a 1,4 addition in the case of G-rignard reagents, and hence, it is generally assumed that most other HA reagents react by the same mechanism. However, there are a few- reagents which apparently attack conjugated carbonyl compounds at the alpha carbon atom, and thus, are in contrast to the general scheme outlined above.
A. The first of these to be mentioned is the reaction of aromatic diazo compounds with substances containing a conjugated carbonyl system. The reaction was discovered in 1899 by Borsche who found that instead of obtaining an azo dye by the action of benzene diazonium chloride on p_~nitrosophenol, a carbon-carbon linkage was established and 2-phenyl-4-nitrosophenol resulted along with some 2,6-diphenyl-4-nitrosophenol.
0 NgCl
alkaline solution
0
IJ
/ \
0
+
s
NOH
NOH
The nature of this reaction has been further studied and extended to various benzo- and naphthoquinones by Kvalnes; while Meerwein and coworkers have established definitely that the attack is at the a- carbon atom. The reaction is also the subject of a few patents in connection with the preparation of substituted quinones to be used as dye intermediates. The work of Meerwein, mentioned above, was reviewed by Mr. Rabjohn as a seminar topic in the fall of 1939, so only a few of the more interesting reactions are listed below:
0 CH-fiH-CHO -(- e~C1.0N2C1
acetone
solvent
0-CH=C-CHO + Ns 0C1
+ KC1 (2)
*
0-CH=CH-C=N + p_-C10I<2Cl
- 2 -
76^
0-CH=C~G=N + N8 + HC1 (3)
0C1
LO
0-0 „.,w« -. Acidic HC
+ £-C10i\i3Cl ►
Solution
Umbellif erone
+
Ns + HC1 (4)
Methyl cinnamate and oinnamic acid are also attacked at the alpha carbon atom, however, in the case of cinnamic acid carbon dioxide is eliminated and a stilbene results. This reaction has been used in this Laboratory by Fuson and Cooke to prepare the methyl and ethyl esters of 4-carboxystilbene.
0-CH=CH-COOH + C1NS
X
CO OR ^ 0-CH=CH0COaR + HOI +
OO—u dr/o
+ CO,
(5)
It should be pointed out that although aromatic diazo com- pounds attack the conjugated system at the a-carbon atom, ali- phatic diazo compounds attack these substances in such a way that a new carbon link is established at the p-carbon atom. Thus benzalacetophenone and diazomethane give the two pyrazolines I and II.
0-CH=CH-C-0 It 0 |
+ CH2N3 - |
■* 0-CH- CH-C- CH2 N ° |
-0 |
+ |
0-CH— -C — C-0 ! jl 8 CE3 N \ / N TI H ^X |
i |
B. In 1927 Dufraisse and Moureu discovered that a, £- unsaturated ketones could be converted into a-diketones (40-80^ yields) by bromination, treatment with piperidine, and subsequent hydrolysis. After a careful examination of the reaction the authors formulated it as follows:
RCE=CH-C0R
Br
-» RCKBrCHBrCOR
-HBr
CBH1XN
RCH=CBr-COR III
C5HUN
RCH,CBr-COR — 5-
NC5H10 IV
HiiN
R-CH-
■CH-COR + RCH=C-C0R
some V
NC5H10
mainly VI
- 3 -
hydrolysis
11
[R-CH=C-COR] OH
■* R-CHo-C-C-R
0 0 VII
The French investigators wrote V as the a,a-dipiperidino compound but the recent work of Cromwell indicates rather definitely that it is the a, p-isomer. The transformation of III into IV is of particular interest since it involves an inverse addition to the conjugated carbonyl system. Compound VI (R=0) is a deep red crystalline compound and was first prepared by Watson, although he considered it to be fj-piperidinobenzalacttophenone .
Kohler and Addinall refused to accept either the mechanism of Dufraisse and Moureu or their structure VI (R=0) for the red piperidino compound. Instead Kohler and Addinall, who had been
studying the the reaction
RCK=CBr-COR III
action of alcoholates on as follows:
R RCH=C — C-NC5H10 Br OH IVa
formulated
C5H1XN
^s-dibro moke tones,
R
i-» R-CK=C— C-NC5H10
V
Via
hydrolysis
R [R-CH=C— C-NC5H10]
OH OH
[RCH-C-COR] 0 H
RCHaC-C-R 0 0
VII..
In order to prove which of these tivo mechanisms was correct a rigorous proof of structure of the compound represented by VI or Via was necessary and in so doing the questionable inverse addition would be clarified. This work was done by Kohler and Bruce and after an exhaustive study they reluctantly concluded that the structure represented by VI as assigned by Dufraisse and Moureu was correct. In brief, their most conclusive proof of structure was as follows:
0-CH=C-CO0
5MgBr
/
0
NC5H10
$,-CH-C=C-CMgX NC5H10
NK4C1
03-CH-CH
0-0
II
NCSH10 0
VI
■•X
OH
NC5H10 VIII
0Mg3r
OH
.0:
0-CH=C C-0 2 stePs^ 0-CH2-C-Cx
Villa
- 4 - 12
To decide which of the two formulas, VI II or Villa, was correct for the carbinol which they obtained was easy; oxidation with chromic acid gave two moles of benzophenone instead of one as would be expected from Villa.
More recently Cromwell has further investigated the reactions of a-bromo-a, ^-unsaturated ketones with amines. He has demon- strated that morpholine reacts, in cold ether, or petroleum ether, with a-bromobenzalacetone , IX, and o-bromobenzalacetophenone , X, to give the a-bromo-a-morpholino compounds XI and XII by inverse addition.
Br 0-CH=CBrCOR ^4K80 > 0-CHg-C-COR . Na0ii9 0-CH=C-COR
I Ale. ( NC4H80 NC4K80
(IX; R = CK3) (XI; R = CH3 ) (XIII; R = CH3 )
(X; R = 0) (XII; R = 0) (XIV; R = 0)
When XI and XII were treated with sodium acetate in alcohol solu- tion the cc-morpholino-a, ^-unsaturated ketones XIII and XIV re- sulted. The latter two compounds, however, could not be induced to add another molecule of morpholine, nor could this base be added tot £-morpholinobenz,-.lacetophenone which also contains the conjugated carbonyl system. Pyrrolidine has been found to add to a-bromobenzalacetophenone X in the same manner as morpholine, but the weaker* base, 1,2,5,4-tetrahydroquinoline will not react.
Now when benzalacetone and benzalacetophenone react with morpholine only saturated p-morpholino ketones result, thus indicating that amines attack the a-carbon atom of a, ^-unsaturated ketones only when there is a halogen atom substituted on the a-carbon.
In studying the reaction of msthoxylamine witn a, ^-unsaturated ketones, Blatt has shown that this reagent adds smoothly and reversibly in the absence of a catalyst to produce saturated fj- methoxy amino ketones.
R-CH=CH-CCR + CH30NHa -» R-CH-CH2-C-R thl00ri3 , R-CE=C-COR
I jl -CH3OH I
NH 0 NH3
0 CH3
^hydrpj^sis^ RCH2-Crg-R
0 0
However, when this reaction is carried out in the presence of a base or when the £-methoxyamino ketone is treated with sodium ethylate, methyl alcohol is removed, and an a-amino-a, j3~ unsaturated ketone is formed, which can be further hydrolyzed to an a-diketone. This shift of an amino group from the p to the a carbon is easily accomplished in yields which in meny cases approximate 90^ and thus constitutes a useful method for trans- forming a, ^-unsaturated ketones into a-dike tones.
13
- 5 -
C. It is reported in the paper of Meerwein, et al., that diphenyl ioclinium iodide reacts with a, ^-unsaturated carbonyl compounds to give substitution at the a-carbon atom just as in the case of diazo compounds. Thus
0-CK-CH-COOB + &0J3I -> 0GH=CH0 +01 + HI
There was no experimental detail, or reference to previous work given, merely the statement that the reaction was under in- vestigation. To date nothing further on this reaction has "been published.
Bibliography :
W. Bore che, Eer., 32, 2935 (1899); Ann., 312, 211 (1900).
U. S. Patent 1,735,432; C.A., 24, 732 (l9307. German Patent
508,395; C.A., 25, 712 (1931)". English Patent 480,617;
C.a. 32, 6262 (193S).
D. E. Kvalnes, J. Am. Chem. Soc, 56, 2478 (1934).
C. Dufraisse and K. Moureu, Bull. soc. chim. , (4) 41, 457, 850, 1370 (1927).
E. P. Kohler and C.R.Addinall, J. Am. Chem. Soc, 52, 3728 (1930) E. R. Watson, J. Chem. Soc, 85, 1322 (1904).
C. Dufraisse and R. Netter, Bull. soc. chim., 51, 550 (1932). S. P. Kohler and "i.F. Bruce, J. Am. Chem. Soc, 53, 1994 (1931). N. H. Cromwell, ibid., 62, 1672, 2897, 3470 (1940}"; 63, 837,
2984 (1941). V. E. Stewart and C. 3. Pollard, ibid., 59, 2702 (1937). A. H. Blatt, ibid., 61, 3494 (1939). H. Meerwein, E. Buchner and K. van Emster, J. prakt. Chem. 152.
237 (1939). H. G-. Cooke, Ph.D. Thesis, University of Illinois, 1940.
Reported by John E. Mahan February 11, 1942
HYDROGEN BONDING BY C-H
Marvel, Copley and Zellhoefer
The concept of association through hydrogen bonding proposed by Latimer and Rodebush in 1920 to explain the abnormal properties of certain liquids has been utilized with considerable success in understanding the behavior of liquids and solutions. In particular it has proved useful in explaining and predicting ab- normally high solubility of certain organic substances in various types of solvents, high heats of mixing, and other abnormal
a
olution behavior.
The types of hydrogen bonds that have been found are 0H<— 0 or N, XHfr-0 or N, NHe- 0 or N, SH -> 0 or N, and CH<-0 or N. The present work deals with the latter type.
There are a number of methods of determining the presence and strength of hydrogen bonds. One method involves shift in the C-H absorption frequency in the infra-red regions. The methods used at this school include measurement of the heat of mixing of liquid mixtures, vapor pressure - composition studies of a volatile solute in a relatively non-volatile solvent, and solubili- ty measurements of polymers in various solvents. All of these methods give relative values of the strength of bond. Qualitatively they are in excellent agreement. Absolute values for the strength of a CH(-0 bond, say for chloroform ether, are about 5,000 to 10,000 calories per mole of complex formed.
The CH hydrogen atom is capable of acting as an acceptor only when it is activated by some negative group such as halogen, nitro, nitrile, acetylenic linkage, and to a very slight extent the phenyl group. It has been found that as acceptors, CHC13 > CHBr3 > CHI3 and CH3C12 > CE3Br3 > CH3I3. This is quite in keeping with the order of electronegativity of the halogens: CI > Br>I. It is the presence of the strongly electron-attracting halogen atoms on the carbon atom which loosen the hydrogen atom and make it available for coordination to the donor atom.
One very unusual case of an active CH hydrogen atom is in benzotrichloride. This compound is a vinylog of chloroform and most probably owes its bonding activity to the vinylogous action of the halogen atoms on the para (or ortho) hydrogen atom.
In the case of donor solvent s the order of donor-ability appears to be, roughly and with a degree of overlapping: alkyl phosphates > 3C aliphatic amines ^ N,N-dialkyl amides > 1° ali- phatic amines > ethers, ketones and esters > aromatic amines> nitriles > alkyl sulfates and sulfonates > nitrates and nitro com- pounds.
Solvents which form strong intramolecular hydrogen bonds such as alcohols, unsubstituted amides, and glycols exhibit little tendency to bond intermolecularly with other acceptors and hence are poor solvents for such compounds as haloforms. However, they
- 2 - JLO
are good solvents for strong donors such as amines. Dinitriles, -~ _ aliphatic nitro compounds, aromatic amines, and other compounds which form weak intramolecular hydrogen bonds show less tendency to form int ermolecular bonds with either donor or acceptor mole- cules than do unassociated solvents. Similarly, chelation tends to decrease the solubility of diketones and keto esters in acceptor solvents.
A steric effect has been noted in the heat of mixing of chloroform and polyethylene glycol ethers. The curves indicate that every other oxygen atom bonds to a chloroform. This is similar to the result obtained with donor solvents and sym- tetrachloroe thane. In this case with two adjacent active hydrogen atoms only one appears to bond.
Hydrogen bonding has been of some help in obtaining solvents for polymeric materials. In general, if the polymer contains a donor group an acceptor compound will be a good solvent. However, hydrogen bonding is only a partial explanation for solubility of polymers.
Hydrogen bonding is a very useful aid to the organic chemist, but it must be remembered that it is only a part explana- tion of solubility and not an infallible rule.
Bibliography:
Latimer and Rodebush, J. Am. Chem . Soc, 42, 1419 (1920).
McLeod and Wilson, Trans. Faraday Soc, 51, 596 (1935).
Gordy, J. Am. Chem. Soc, 60, 605 (1938).
Buswell, Rodebush and Roy, ibid, , 60, 2448 (1938).
Zellhoefer, Copley and Marvel, ibid. , 60, 1357 (1938).
Zellhoefer and Copley, ibid., 60, 1343"Tl938).
Copley, Zellhoefer and Marvel, ibid. , 60 2666, 2714 (1938).
Copley and Hoi ley, ibid. , 61, 1599 (1939).
Copley, Marvel and Ginsberg, ibid. , 61, 3161 (1939).
Copley, Zellhoefer and Marvel, ibid., 62, 227 (1940).
Marvel, Dietz and Copley, ibid. , 62, 2273 (1940).
Marvel, Copley and Ginsberg, ibid. , 62, 3109, 3263 (1940).
Copley, Ginsberg, Zellhoefer and Marvel, ibid., 63, 254 (1941).
Marvel, Harkema and Copley, ibid., 63, 1609 (1941
Marvel and Harkema, ibid. , 65, 222lTl941 ) .
Reported by E. Ginsberg February 18, 1942
ACIDIC AND BASIC CONSTITUENTS OF PETROLEUM
16
At the University of Texas, under the direction of Dr* H. L, Lochte and the late Dr. J. R. Bailey, sixty-five indi- vidual acidic and "basic constituents of petroleum have been isolated and identified. Several more have been isolated but not yet identified. Only sixteen of these compounds had been previously reported, and thirteen of these were phenols and aliphatic acids. The compounds which have been isolated are listed in Table I.
TABLE I
Bases
Acids
Quinoline Quinaldine
8-Dimethylquinoline
3-Dimethylquinoline
•Methyl-8-ethylquinoline
3 , 8-Trimethylquinoline
3-Dimethyl~8-ethylquinoline
3-Dimethyl-8-n-propylquinoline
4-Dimethylquinoline
4, 8-Trimethylctuinoline
4-Dimethyl-8-ethyiauinoline
4-Dimethyl-8-n-propyiquinoline
4-Dimethyl-8-s.-butylquinoline
3,4,8-Tetramethylquinoline
3,4-Trimethyl-8~ethylquinoline
3,4-Trimethyl~8~n~propylquinoline
3,4-Trimethyl-8~l-propylquinoline
3,8-Trimethyl~4-ethylquinoline
3-Dimethyl-4,8-diethylquinoline
3-Dimethyl-4-ethyl-8-n-propylouinoline
Isoquinoline
2-Picoline
4-Picoline
2, 6-Dime thy lpyr idine
2, 5-Dimethylpyr idine
2,4-Dimethylpyridine
3, 5-Dimethylpyridine
2,4 , 5-Trimethylpyridine
2, 4 , 6-Trimethylpyridine
2-s_e_c.-3utyl-4, 5-dimethylpyridine
2, 4-Dimethyl-6- " trans T'-2, 2, 6- t rime thy lcyclahexvlpyr idine
2,3-Dimethylbenzo (h)ouinoline
2, 4~Dimcthylbcnzo (h)quinoline
Formic
Acetic
Propionic
n- Butyric
iso-Butyric
Valeric
.1 so- Valeric
n-Hexanoic
2-Methylpentanoic
3- Me thylp en tano i c
n-Heptanoic
SrMe t ay liitex&nol c
3-Methylhexanoic
4-Me t hy lhe xano i c
5-Methylhexanoic
n-Octanoic
n-Nonanoic
Dimethylmaleic anhydride
Phenol
o_~Cresol
m-Cresol
p_-Cresol
2,4-Xylenol
2,5-Xylenol
3,5-Xylenol
Cyclopentanecarboxylic
Cyclopentaneacetic
3-Methylcyclopentane-
acetic 2,3-Dimethylcyclopen-
taneacetic solid 4-methylcyclo-
hexanecarboxylic "trans "-2, 2, 6-Trimethyl-
cyclohexanecarboxylic cis-2, 2 , 6-Trime thyl-
cyclohexanecarboxylic
.
- 2 " 17
One of the interesting problems in connection with this research was the assignment of structure to a C16HS5N base, 2,4-dimethyl-6~trj^~2,2,6-trimethylcyclohexylpyridine, and the structural correlation of this base with a C10H1802 acid, trans- 2,2,6-trimethylcycl@hexanecarboxylic acid, also isolated from petroleum.
The positive contributions toward determining the structure are diagramed on the Flow Sheet and may be listed as follows:
1. Isolation of pyridine-2,4, 6-tricarboxylic acid from the oxidation products of the base with dilute nitric acid proved alkylation at positions 2,4, and 6.
2. Formation of a phthalone established the presence of a methyl group at position 2 or 6.
3. Condensation of the base with formaldehyde and nitric acid oxidation of the product yielded a dicarboxylic acid, G14H19N(COOH)3, which was decarboxylated first to C§ iHi9N(COOH)3, which was decarboxylated first to Cl4H20NCOOH and then to C14H21N, indicating the presence of two reactive methyl groups.
4. Formation of a dibenzal derivative also indicated the presence of two reactive methyl groups.
5. High pressure hydrogenation in the presence of Raney nickel as a catalyst resulted at 250° in the consumption of six atoms of hydrogen per molecule of the base. Dehydrogenation yielded the original base. The base with one mole of methyl iodide formed a quaternary ammonium salt which was not converted into a methylated free base by sodium hydroxide. The non-reactivity of the base with ammonium iodide and hydriodic acid in an atmosphere of carbon dioxide showed the absence of an N-:.lkyl group. All of these facts indicated that the base was a substituted pyridine.
One of the' confusing facts about the base was its stability towards oxidizing agents. Alkaline permanganate as well as chromic acid has no effect at room temperature. Acid permanganate attacks the base but no product except carbon dioxide could be isolated.
Data to this point indicated that the compound contained a pyridine nucleus with two methyl groups and a C9H17-alicyclic radical as substituents in positions 2,4, and 6, the exact location of the individual groups being indefinite.
6. Ring cleavage and removal of the nitrogen atom was achieved by the method of von Braun, by which benzoylpiperidine may be converted to pentamethylenedibromide. The dibromide ob- tained was converted to a diene by alcoholic KOH and ozonized to obtain a CloH1802 acid, confirming the presence of n C9H17- substituent on the pyridine nucleus. Direct ozonization of the base followed by hydrolysis in alkaline peroxide gave the amide of the acid, establishing the C0H17-group as being .in the °-> position
- 3 - AS
7. The C10H1803 acid obtained from degradation of the base was found identical with one obtained from petroleum.
8. Conversion to the C9H17NK3 amine was accomplished by the action of hypobromite on the amide or hydrazoic acid upon
the acid. Treatment with nitrous acid gavn an unsaturated hydro- carbon, C9H1S, which gave 2-methylcyclopentyl methyl ketone, leaving only five possible structures (listed in the flow sheet) for the acid. Since the acid was not esterif iable by methyl alcohol-hydrogen chloride, the ester prepared from the acid chloride was not saponified by alcoholic KOH, the e ster was not hydrogenated under 4QC0 pounds pressure at 250° in the presence of copper- chromium oxide, and the amide could not be prepared by ammonolysis of the ester, it appeared at first to be a tertiary acid. Whitmore's method of distinguishing between secondary and tertiary acids indicated that this acid was tertiary; so the structurally possible tertiary acids were synthesized.
After elimination of these tertiary acids, von Braun's imide chloride method showed the acid to be secondary, whereupon the two isomers of 2, 2,6-trimethylcyclohexar.ecarboxylic acid were synthesized. The acid from petroleum and from the degrada- tion of the base was identical with the higher melting form, tentatively assigned the trans configuration as the higher melting form.
Bibliography:
Shive, Horeczy, Wash, and Lochte, J. Am. Chem. Soc, 64, 385 (1942).
Shive, Roberts, Kahan and Bailey, ibid, (in press)
Shive, Crouch, and Lochte, ibid. , 63, 2979 (1941).
Wash, Shive, and Lochte, ibid. , 63, 2975 (1941).
Hancock and Lochte, ibid. , 61, 2448 (1939).
Schutze, Shive and Lochte, Ind. Eng. Chem., Anal. Ed., 12, 262
(1940). Lake and Bailey, J. Am. Chem. Soc, 55, 4143 (1933). Armendt and Bailey, ibid,, 55, 4145 TlS33). Thompson and Bailey, ibid, ,55, 1C02 (1931). Perrin and Bailey, ibid,, 55.;.' 4136 (1933' Bratton and Bailey, ibid., 59, 175 (1937 Lackey and Bailey, ibid. . 56, 2741 (1934 Axe and Bailey, ibid., 60, 3028 (1938). Axe, ibid-, 61, 1017 (1939). Axe and Bailey, ibid. , 61, 2609 (1939). G-lenn and Bailey, ibid., 61, 2612 (1939). Glenn and Eailey, ibid., 65, 637 (1941 J . Glenn and Bailey, ibid., 63, 639 (1941). Schenck and Bailey, ibid., 61, 2613 (1939). Schenck and Bailey, ibid. .. 52, 1967 (1940;. Schenck and Bailey, ibid.'. 63, 1364 (1941). Schenck and Bailey ibid., 63, 1365 (1941).
Reported by Billie Shive February 18, 1942
19
FLOW SHEET Degradation COON
!
CH3
tSSK^00
+ other oroducts
%/-C00H
^>
N^-G^3
phthalic
anhydride
Phthalone
JH^O
CilgCrigOH
KNO*
C9HX7-^ y-CffaCH8OK
CBH17^N^COOH CbHxtV*^
o ixi o
in W
CD "O
-^ C9H17\Ny-CH=CK-C6H5
soaa lime
C9K17-\N^
^16^35^ + 3H3 — » C16H31N C16H30NCBr3(C6H5) - distillation
CrHcCCCI prv.
-S-S ; C16H30N-C0-C6H5 -i^^^
in vacuum
* C16H30Br;
Me OH
0,
HoO
KOH
C9H17C00H
3^3
,Qa
1#Hj 7-A^ >/*- CNS
* ozonide j?a^H ■> C9H17-C0NH3 + 2CH3C00H +
H3°3
2C0o
^
' C9H17-CONH3^Jto£.
SB
C9H17-C00H
G9Hx 7NH;
„*^\x6*H&** C9H17i*(CH3)a
CH-
■CH:
0.
CH2 Cri-CH3
CH-C:
CH-
On ■
■CI
ri-
Cri2
/\
H2C CH
CH3HC C(CH3)2
vc6
T COOK
CHS
H2C/\CH(CH3)
I ! H2C G (CH3 ) 2
Y
CO OH
CH2
CHS — CH2
CH CH ( CH3 )
II
H3C-OH-CH3 COOH III
CH-CH3 CH-C^CH3
CHg CK2
i !
CH, CH(CH3)
^CH I CH3-C-COOHI
CH3 IV
CH2 CH2
1 I
CH2 CH(CH3)
CH »
CH3-CK-CK2~COOH V
SYNTHESIS OF I
(CH3)2C=CHCHaCH3~C=CH-C-H ife^L (QH3 )3C=CHCH3CH3-C~CH~COOH f orm^c >
C HC NCH2
CHa-C C(CHs)a
XCH
COOH
/
CH3
Pt
H
C
HoC -CHp
I I
(CH3)HC /C(CH3)2
CH COOH
ra.p. 74-75°
CH3 Hac, ac20
acid
trace HC1
H,
cx
H2C CH2
(CH.) HC C(CH3)2
COOH
m.p. 82-83'
PROPOSED SYNTHESIS OF BASE
21
CH3
(CH3)HC ,C(CH3)a
+
H
0
Ui-3
\^
•>- CH
Al.Hg
Uxlg
H2C v CH3
I i
(CH3)HC C(CH3)3
XC^
CH
CH3
«Li £»•— -CH3
22
CLEAVAGE OF THE ALKLY-OXYGEN BOND IN THE HYDROLYSIS
OF ESTERS
Cohen and Schneider (Harvard )
Although it appears that hydrolysis of sulfonic and other strong acid esters occurs in the following manner,
-SOa-0-3|+HJ5ff-* t«6020H+R0H
it is generally accepted that hydrolysis of carboxylic esters and the formation of these esters are reactions which do not involve rupture of the alkyl carbon-jjiygen bond hut involve an acyl carbon-oxygen mechanism,
R-C^O-R'+HOH-* RCOOH+R'OH
although the same products could be produced by the alkyl carbon- oxygen rupture.
The evidence for the acyl carbon-oxygen mechanism can be summarized as follows:
1. Reid's work on the esterif ication of thio-acids and mercap- tsns, and the hydrolysis of thio-esters.
(a) Esterif ication of thio-acids produced oxygen esters and hydrogen sulfide, rather than thio-esters and water,
*P-
RC-S-H+HpR' -^RCOOR»+H3S
(b) Carboxylic esterif ication of mercaptans produced thio- esters and water, rather than oxygen esters and hydro- gen sulfide.
RC-.0-H+HS-R1 -*RC-S-R»+H20
(c) Hydrolysis of thio-psters produced carboxylic acids and mercaptans, rather than thio-acids and alcohols.
s
RC-^-R'+HDH -^RCOOH+R'SH
2. Retention.of configuration when carboxylic esters of the
/O H
type R-C-0-£*_rt sre hydro lyzed. It is evident tha t if at sty time during the hydrolysis, the asymmetric carbon become free, complete retention of the original configuration is not possible. Thus, retention of configuration is an indica- tion that the alkyl-oxygen bond is not broken during the hydrolysis. There are many cases of such hydrolyses in the literature, and there is no simple instance of appreciable inversion of configuration of the alcohol.
-2-
23
3. Demonstration by Ingold 'and co-workers that single alcoholic oroducts are obtained when esters of the 0
R~C?-0-Ctt2-C=CR2 R type are hydrolyzed. If the resonating ion becomes free at any time during the hydrolysis, S+ + H
R:C::C:6« >R:C:C::C
R R H R R H a mixture of alcohols would be obtained.
4. Demonstration by Quayle and Norton that esterif ica tion of various acids with neopentyl alcohol produced no unsaturated products, and that hydrolysis of the resulting esters pro- duced neopentyl alcohol es the only alcohol. "Whitmore and his collaborators have demonstrated that a positive neo- pentyl group invariably rearranges to tertiary amyl, the rearrangement being accompanied by the formation of un- saturated derivatives.
9H3 + +
CH3C-CH2 > CH3Q-CH3CH3
CH3 CH3
Apparently the oxygen-neopentyl bond is not broken during esterif ica tion or hydrolysis, even though the a.cid concerned is as strong as trichloracetic acid.
5. Direct evidence employing isotopically distinguished dxygen.
(a) Alkaline hydrolysis ©f amyl acetate in an aqueous medium enriched in 018 resulted in the obtaining of amyl alcohol with the normal isotope ratio, thus indicating that the oxygen present in the alcohol was originally attached to the alkyl group in the ester.
(b) Similar results were obtained by Ingold in the acid- catalyzed hydrolysis of methyl hydrogen succinate.
(c) Urey and Roberts esterified benzoic acid with 018 enriched methyl alcohol, and found that the heavy isotope was not present in the water produced in more than the normal rati©, thus indicating that the oxygen of the water produced was obtained entirely from the benzoic acid.
There are a few exceptions to this mechanism in the case of some compounds, which because of special structural features, show unusual reactivity. These include ^-lactones, the hypo- thetical cc-lactones, and esters of a secondary allylic alcohol ( CGHs~CH-CH~gH0H~CH3 ) ,
The general conclusion for the mechanism has been based on a study of esters of primary and secondary alcohols, and apparently is perfectly valid for such cases. However, there is one instance reported in the literature in which a very limited racemization of a secondary optically active alkyl group occurred during esterif ication, i.e., the esterification ©f
.
-3-
p-n~octyl alcohol. An alternative mechanism was indicated. Since the tertiary alkyl-oxygen bond ie more easily broken than is the secondary , an investigation of tertiary rlkyl esters was thought to be worth-while.
Cohen and Schneider have made such investigations. Although the nature of the hydrolysis products would throw no light on the point of interest (unless optically active alkyl groups or iso topically distinguished oxygen were employed), the nature of the products obtained from a primary rlcoholysis of an ester would provide pertinent information.
The two possibilities are:
(a ) Ester Interchange
s$ *o
RC ~p-R M-HpR" -* RC-O-R" +R T OH a nd ,••••••
( b ) Ac I-* *n9 Ether .Formation P ,0
RC ~0-jR \ +t Rn0h -> RC-O-H+R * OR" The occurrence of ester interchange 'would* 'indicate the validity of the general conclusion as applied to tertiary esters, while the formation of acid and ether by alcoholysis would indicate a different mechanism, i.e. Alkyl carbon-oxygen cleavage.
Alcoholysis of i.-butyl benzoate with methyl alcohol in initially neutral solution produced a good yield of t.~ butyl methyl ether and appreciable quantities of benzoic acid. Al- though considerable methyl benzoate was produced, this could have been formed by esterif ication of the benzoic acid produced by the acid-ether mechanism. It was also shown that t_-butyl alcohol and1 methanol do not form J>-butyl methyl ether in the presence of benzoic acid. No t-butyl alcohol (which would result from ester interchange) could be detected. Thus it Bpp-peers that ester interchange does not play a very important part in this reaction, end that the alkyl carbon-oxygen cleavage Is probable. Similar results were obtained with t~butyl acetate.
On the other hand, similar alcoholysis in the presence of alkf li resulted in the normal ester interchange, and no evidence was found for the formation of t.-butyl methyl ether, thus indicating the "normpl" cleavage.
steric hindrance
■
2o
~4_
Result? of the hydrolysis of esters of 2,4,6-trimethyl benzoic acid were as follows:
(1) Acid hydrolysis of JCu-butyl 2, 4, 6-trirnethylbenzoe te produced a nearly quantitative yield of the acid.
(2) Similar acic1 hydrolysis of the methyl ester was un- successful.
(3) Comparable basic hydrolysis of the t-butyl ester was unsuccessful. These results demonstrate that there is a deep-seated difference between the alkaline and the acid hydrolyses of the t_-butyl ester, and that an equally important difference exists between the acid hydrolyses of the tertiary ester and of the. primary ester.
When t-butyl 2,4,6-trimethyl benzoate was subjected to alcoholysis, results pointing to the same conclusion were observed, namely
(1) Initially neutral alcoholysis gave products obtainable from the acid-ether mechanism, although the yields were not nearly fb great.
(2) Base catalyzed alcoholysis did not occur, indicating that the ester interchange mechanism is not active in the case of this st erica lly hindered compound.
The base-catalyzed ester interchange and hydrolysis re- actions of esters probably proceed by a nucleophilic attack an alkoxide or hydroxide ion on the carbonyl carbon atom of ester.
by
the
:0: R:C:p:R'+:0:Hz
„
• 0:.. R:C:0:H
5 0:R! ^
:0: tR:cV6:H+ :0:R1
A Stuart model of ^-butyl 2, 4, 6~trimethylbenzoate shows that the carbonyl carbon atom is strongly hindered, and the reason for the failure of attempts at alkaline hydrolysis is readily apparent. The probable mechanism for the "normal" acid- catalyzed 'hydro ly si s also involves the same difficulty, and the resistance of primary esters of this acid is understandable.
A possible course for the alcoholysis of tertirry esters to form the corresponding carboxylic ^cids and ethers ie as follows, First, a slow reaction occurs in neutral solution involving hydrogen bonding between en alcohol molecule and the carbonyl oxygen atom of the ester, followed by ejection of the £~butyl group and the formation of the products. This uncatalyzed or solvolytic mechanism accounts for only a very small portion of the products. However, the acid produced, catalyzes the faster and more important reaction.
~5- ,.^v^
:0: +:0:K
♦ • • « »
OeH8.;0:0:G(CHs)3+H;A: ^ — > C6H5:C:0:C(CH3 }3+:A:
K
0- ••■ ■+■
C6H8:C::0:C(CH3)3 -> < •
•0:
+ C6H5C:0:K+(CH3)Bt3
+ ## +
(CH3)3C +CH3:b:H -* CK3: 0:C(CH3 )3+H
The model of Jj,- butyl 2, 4, 6~trimethylbenzoa te shows that the carbonyl oxygen atom is not strongly hindered. Since the above mechanism involves this atom and not the carbonyl carbon, the above mechanism seems plausible, ?n0 would account for the acid hydrolysis of this ester ps well as the alcoholysis pro- ducts obtained from t^butyl benzoate in initially neutral solu- tion.
There are relevant kinetic data available for hydrolysis of esters of the type CK3C00R. As the group R is varied from CH3 to C3K5, i_-C3H7 and J>-C4Hg, the rate of the base-catalyzed hydrolysis falls off, while the rate of acid hydrolysis falls through a slight minimum and rises again, that of the. t_~butyl ester being about 15$ faster than that of the methyl ester.
These facts are consistent with the following interpreta- tion: in the base-catalyzed ractions, a single mechanism oper- ates, the attack of the hydroxy 1 ion on the carbonyl carbon atom followed by the ejection of the alkoxyl group. The rate of this reaction diminishes as the alkyl group is changed from primary to tertiary. In the acid-catalyzed reaction, there is one mechanism available for esters of primary end secondary alcohols involving a. rupture of the acyl-oxygen bond, and there is the alternative mechanism available for the tertiary alcohol ester, involving rupture of the alkyl-oxygen bond and ejection of the alkyl group.
Further investigation employing optically active tertiary esters or isotopically distinguished oxygen would provide further information on this Question.
BIBLIOGRAPHY
Cohen and Schneider, J. Am. Chem. Soc . 63, 3332 (1941) Annual Reports on the Progress of Chemistry for 1940, London
38, ?29 (1941) Reid, Am. Chem. J. 43, 489 (1910*
JS7
-6-
Polanyi pnc1 Szabo, Trans. Faraday Soc . 30, 508 (1934) Roberts and Urey, J. Am. Chem. Soc. 60, 2391 (1938) Olson and Miller, lipid. 60, 2687 (1938) Winstein, ibid. 61, 1635 (1939)
Bean, Kenyon and Phillips, J. Chem. Soc. 303 (1936) Kenyon, Partridge and Phillips, ibid. 85 (1936) Burton and Ingold, ibid. 904 (1928) Hughes, Ingold and Masterman, ibid. 840 (1939) Ingold and Ingold, ibid . 756 (1932)
Datta, Day and Ingold, ibid . 838 (1939) Quayle and Norton, J. Am. Chem. Soc. 62, 1170 (1940) Whitmore, ibid. 61, 1586 (1939)
Hammett, "Physical Organic Chemistry", McGraw-Hill (1940), pp. 211, 213, 356
Reported by ¥. E. Blackburn February 25, 1942
SOME REACTIONS OF HYDROCARBONS IN AN ELECTRICAL DISCHARGE
28
It is possible to excite molecules, causing them to undergo subsequent chemical reactions, by exposing them to the action of heat, light, sound waves, a particles, and electrical dis- charges. Considering only the last method, we find that there are two distinct types of electrical discharge, the silent or non- disruptive and the disruptive discharge. The silent discharge may take the form of the ozonizer, the semi-corona, the corona, the glow, or the electrodeiess discharge, while the disruptive discharge may occur as the arc or the spark. The chemical re- actions which take place in these two types of discharges are essentially different. In a given reaction, the important variables to be considered are: (l) the effective discharge in- tensity; (S) the effective residence time in the discharge.
Reactions of Hydrocarbons in the Silent Discharge- Ethyl ene is a fairly active hydrocarbon and requires only mild excitation to cause it to react. With ethylene and higher olefins reactions occur under mild conditions and are of a com- paratively simple nature. In a static system, that is a system in which the ethylene was allowed to stay in the reaction chamber during the entire course of the discharge, it was found that when using a high frequency course discharge, the ethylene had completely reacted after a period cf about ten hours. There appeared to be an induction period of about two hours which could be considerably shortened by the introduction of hydrogen into the reaction chamber. An analysis of the gaseous products showed 67% hydrogen and 20% saturated hydrocarbons. The liquid fraction consisted of a dark oil with a molecular weight of the order of five hundred.
The results of an experiment using a dynamic system are more informative. The ethylene was passed through the discharge in a stream and the products were condensed at -?0°C. In the particular experiment cited a high frequency ozonizer discharge was used. The following products were obtained:
Uncondensed gases (H2, C3H3, C2H6 ) 13%
Butane 45%
1-Butene 15%
a fraction boiling in the range
34-45 °C 4%
a C6 fraction 15%
Higher hydrocarbons 8%
In the C6 fraction 1-hexene and paraffin hydrocarbons were iso- lated. In other experiments in which somewhat different experi- mental conditions were used, it was found that the chief product was acetylene or butadiene. In all cases the energy consumption was about twenty kilowatt hours per kilogram of ethylene which reacted. The products identified are thought to be evidence for the theory that ethylene undergoes two primary reactions:
29
- 2 -
[1) Dehydrogenation of ethylene to produce acetylene and hydrogen; .2) Polymerization to give higher molecular, weight hydrocarbons. Egloff and others suggest that (l) may be accounted for by the assumption that the energy absorbed from the electrical discharge disrupts the C-H bond in ethylene leaving the vinyl radical.
II]
H H
X
H
K
H
Hx
C=(
u
H'
The vinyl radical then decomposes to give acetylene and more hydrogen.
H H
•C=CH + H-
It is quite possible, however, that the C-H bond is not actually broken but merely activated. Under these conditions it is possible that two molecules of ethylene might react as follows:
H H K
H
vc=c;
X
ri
.H
c=c-'
\
C=C H * ,*H
Nc=cr"
H' H
rigC— Crl— C113CH3
1-butene
from
The butene is isolated in a great many cases. Furthermore, in the presence of hydrogen, hydrogenation to butane is possible which would account for the formation of this substance. It is also possible for the butene to react with another molecule of excited ethylene forming a hexene and then a hexane. However, this theory is not sufficient to account for all the observed products. In many cases the liquid products absorb oxygen f] the air, a property common to acetylene polymers but not characteristic of ethylene polymers. This fact suggests the possibility that acetylene, after being formed, reacts to give polymers of the butadiene type.
With higher olefins, reactions similar to those of ethylene take place. In a static experiment using propene, it was found that 85-90$ of the olefin reacted to form a liquid possessing an average molecular weight of 232. Isobutene polymerizes readily to give, as the largest fraction, a mixture of di- and triiso- butenes.
Acetylene was found to react readily in an electrical dis- charge. When the reaction mixture was cooled to -60°C, a 70fo yield of a colorless liquid was obtained. This liquid had a molecular weight of a triaer and from the reactions of the liquid it was concluded that it contained 1, 5-hexadiyne, methyl pentadiyne and unidentified products.
When the re temperature, both of the products s to the benzene ri action of a silen If the reaction p and cooled, relat allowed to react compounds or arom A spectroscopic e is in the dischar carbon atom, and
- 3 -
action products were allowed liquid and solid products w eemed to have an unsaturated ng. Thus it seems probable t discharge on acetylene is roducts are quickly removed ively simple polymers may be further, long-chain, highly atiea with unsaturated side xaminati on of the light prod ge indicates the presence of the hydrogen atom.
iO
to stand at room ere formed. Some
side-chain attached that the primary to form polymers, from the system
isolated., . If unsaturated, aliphatic chains may be formed. uced when acetylene
the C+ ion, the
Benzene reacts in the silent electric discharge, commonly identified product is diphenyl but it is always accompanied by complex resinous material. In the gaseou_
considerable quantities.
The most In the gaseous products
hydrogen and acetylene are found in
the reaction products are quickly cooled, dihydrobiphenyl can be
obtained. .
vy
H H
>
dihydrobiphe nyl
products
net cooled
H:
Thus benzene aeems capable of undergoing two types of reactions.
> — > polymerization and dehydrogenation (biphenyl)
C6H6 <^ 4Q^
x _Zr — ,. depolymerization (acetylene)
A spectroscopic study of the light emitted during the reaction shows the presence of C ions, carbon and hydrogen atoms, C2 and CH molecules. This would indicate the presence of a complex mixture.
Paraffin hydrocarbons , being less reactive than olefins and acetylenes, require a more intense die charge to initiate a re- action. With the more intense discharge a high temperature is developed and thermal activation is at least partially responsible for the observed reactions. When methane is subjected to an ozonizing discharge at room temperature, the reaction is thought tc take the following general course.
_ 4 -
31
polymerizes > iiqUid and solid products
CH4 -» H3 + C3H4 +ris > C3H6
"0 h.4
CaH
3^8
G^Kp, + .7
Aiia_i_^3 c4h10
Very little ethylene was actually isolated, but it is not sur- prising for, if forced, it should immediately react in any dis- charge which is intense enough to activate methane. In a glow discharge methane can toe made to yield acetylene but the products must be quickly removed from the reaction zone and cooled to prevent the acetylene from undergoing further reaction.
Ethane can undergo both dehydrogenation due to activation of the C-H bond and demethanation cue to the activation of the C-C bond.
CsH6 — > Csn^ + H3
2C3H6 ~* OH* + C3K8
The latter reaction is characteristic of the saturated hydro- carbons in an electrical discharge. This reaction can possibly be accounted for by assuming the ion-cluster mechanism which was proposed to account for the reactions of hydrocarbons with a particles.
„ „ discharge _ ,T + ,„ *
C3n6 1 ^=-> C3H6 + e (1; ionization
0aHe+ + C3H6 -► [C3H:6.C3Hg] {2) clustering
[C3H6.C3K6] ~* LCr^.CgHe] (3) rearrangement of
the valence bonds
[CH4.C3H8] + e — + OH4 + C3HS (4) decomposition
It must be assumed that the ion-cluster is stable for a long enough period of time to allow the valence bonds to rearrange. Propane undergoes the same type of reactions.
Reactions of Hydrocarbons in Disruptive Discharges
The disruptive discharges consist of the arc and the spark. While there may be electrical effects on hydrocarbons in the spark discharge, there is reasonable doubt that there is any such effect in the arc. Both discharges cause the same type of re- action as does high teaperatur e, i.e. above 1500°C. Carbon and hydrogen are the main products in most cases and there is usually a considerable amount of acetylene formed. Traces of ethylene, dipropargyl (HC=CCH3CH2C=OH), benzene, and naphthalene have been found when using methane, indicating a thorough-going disruption and recombination. In arc and spark discharges, the electrode
- 5 -
material is of considerable importance and there is seme evidence to indicate that in the case of graphite electrodes, they may actually take part in the reaction. It has been noticed that benzene has a very distinct tendency to form carbon and hydrogen.
Activation of. the _Hc 1 eou.1 e s
Most of the reactions typical of rn electrical discharge take place in the silent discharge, Taking the gl cw discharge as an example o^ this type the process of activation will be considered in a little more detail, The activating electrons are produced ?t i;he cathode Moving under the force of electro- static repulsion away fsori. the negative polo. they migrate to the Grookes dark space Tills area i : the region of greatest potential drop and the electron experiences .an accelerating force toward the anode. During this acceleration, the electrons under- go three types of collisions with molecules,
(1) elastic collisions in which no energy is lost or gained
(2) ionizing collisions in which a secondary electron and a positive ion are formed
(3) activating collisions in which the quantum state of the molecule is raised
Each electron leaving the cathode is Known to produce from 50 to 100 secondary electrons by ionizing collisions and since the energy of ionization is much greater than that required for activation, it is assumed that many times that number of activat- ing collisions take place. At all pressures above .1 mm of Hg, electrons are thought to be the sole factors producing activa- tion and ionization. The positive ions present are so heavy that there is not a sufficiently high potential drop to give them a high velocity, and they are large and, therefore, must hit other molecules very often, Both of these conditions operate to prevent the positive ions from ever acquiring sufficient energy to cause an ionizing or activating collision Elastic collisions may be neglected as only the ionized and activated molecules have the energy necessary for chemical reaction. Very little is known about the reaction of hydrocarbon ions except what little can be deduced from the nature of the products. The course of the reaction of tne activated molecules can be followed in many cases by analogy with reactions which take place- under the influence of thermal activation.
Commercial Anplications
The action of the silent discharge is being used on a commercial scale to produce and improve lubricating oils. The process is known as "Siectricnization" in Belgium and as "Voltolizati on'-' in Germany. The oil is put in a large mechanical device which is arranged in such a way as to provide thorough mixing and a short exposure to the silent discharge. The reaction is run at about 80°C and the pressure is about 60- -65 mm. The electrical energy is furnished by a 500 cycle discharge at approxi- mately 4000 volts. The action of the discharge dehydrogenates the molecules which then polymerize. Fifteen per cent of the
- 6 - 33
product is blended with untreated oil. The mixture possesses the following desirable characteristics: (l) the viscosity is increased, (2) the viscosity temperature coefficient is decreased, (3) the pour point is lowered, (4) the sludge-forming tendency of the oil is decreased, (o) the tendency to form emulsions with water is increased.
The production of acetylene from methane on a commercial scale can be accomplished at the present time. In the carbide process we have the reaction:
o CaO + 5C -» CaC3 + CO ^F29e - 88,400 cal/mol
For the discharge process:
o 2CH4 -* C2H2 + 3Ha AF298 - 75,400 cal/mol
Translated into electrical units this means the carbide process requires theoretically 5.6 kilowatt hours as against 4.7 for the discharge process per cubic meter of acetylene. In actual practise the energy consumption is considerably larger, requir- ing from 9 to 18 kilowatt hours, depending on the type of furnace. Various experiments have been carried out with methane. When the methane is diluted with two volumes of hydrogen to cut down the carbon forming reaction, it was found that o&% of the methane which reacted, bl% of Che methane charged was converted to acetylene giving an energy consumption of 12 kilowatt hours per cubic meter of acetylene in an arc. It was possible with a glow discharge tube to obtain yields up to 90$ of methane charged. Using liquid hydrocarbons, a process has been developed which re- quires 15.5 kilowatt hours per cubic meter of acetylene but, at the same time, 1.4 cubic meters of hydrogen are produced as a by- product.
Several other processes, i.e. the preparation of diphenyl from benzene, butadiene from fuel oil, and various processes for the preparation of gasoline hydrocarbons by cracking in an electrical discharge, have been patented. As far as is known, " none of these are in actual operation at the present time.
It may be seen that an electrical discharge is a powerful tool in causing the activation and subsequent reaction of hydro- carbons. It is possible to expose the reactants to a wide variety of experimental conditions. The primary reactions seem to be de- ny drogenation and polymerization. In many cases these are followed by various secondary reactions which give rise to a large number of products. One commercial process, i.e. Voltolization, has been in operation for a number of years and other useful products may be awaiting only the proper experimental conditions for large-scale production.
Bibliography:
Harking and Gans, J. Am._Chem. Soc. , 5_2, 2578 (1930).
de Saint-Ceunay, Chi mi e Industrie 29, 1011 (1933).
Brewer, Chem. Rev., £1, 213 (1937).
Thomas, Egloff and Morrell, ibid , , 28, 1 (1941).
Reported by R. G> Chase February 25, 1942
WAR GASES 34
Perhaps the commonest classification of war gases is the physiopathological classification; that is, a classification based on the most characteristic action of the war gas on human beings. By this method, the gases usually are divided into the following classes:
(1) Lachrymators, or tear gases;
(2) Sternutators, or "sneeze" gases;
(3) Lung injurants;
(4) Toxic gases-those gases having a harmful systemic effect;
(5) Vesicants, or substances producing blister's on the skin.
This classification is not rigid, since the biological action of these substances is complex. For instance, chloropicrin is a lachrymator, a toxic gas, and a lung injurant. , Nevertheless, this classification is useful in helping to present a clear pic- ture of this subject.
(A) Lachrymators
The many substances which have been used as lachrymators include a- end (3-halogenated ethers, ethyl bromoacetate, acro- lein and its halogenated derivatives, oc-halogenated ketones, the benzyl and xylyl halides, o_-nitrochlorobenzene, bromobenzyl cyanide, and phenylcarbylamine chloride.
In general, it has been found that the aromatic compounds are more satisfactory than the aliphatic compounds. Of the substances mentioned above, bromsbenzyl cyanide and chloro- acetophenone are perhaps the most satisfactory lachrymators.
Bromobenzyl cyanide can be prepared in the following way:
C6H5 - CH.ei-gjS-* C6HS - CH2CN -. ,Br« vC^-CH^' yu ultra-violet 6 5 Vg
Bromobenzyl cyanide is a low-melting solid. One method in which it may be used is to dissolve it in chloropicrin, which in itself is a war gas. On heating to 160°, it decomposes to form dicyanostilbene:
,CN C6H5 - C - CN
2 CeH5 - CH > II + 2 KBr
NBr C6H5 - C - CN
For this reason, its use is limited because of its low sta- baility t© the explosion of a bursting shell. Another drawback is that it attacks metals, se that specially-lined containers-
% -2- ^>0
have to be used.
Chloroacetophenone can be prepared by the fallowing method,
C6HG + CICH3 - C0C1 A1C]:3 .,c6Hs - GO - CH2C1 + KC1.
This substance is also a sclid. One method of using it is to spray a benzene solution of it into the air. "When the benzene evaporates, the chloroacetophenone is dispersed in a state of fine subdivision.
(B) Sternutators
Some of the representative sternutators are phenyl dichloro- arsine, diphenyl chloroarsine, diphenyl cyanoarsine, pnd phena;:*- sazine chloride (Adamsite).
( 1 ) Phenyl Dichloroarsine and Diphenyl Chloroarsine
A mixture of the above two compounds was prepared by the Allies in the last war by the following method:
AsCl3 + 3 C6HBC1 + 6Na- >(C6H5)3 As + 01a CI
(C6H5)3 As + AsGl3 ~^^~* C6H5 ~ AsCl3 + vC6H5)2 -AsCl
60 - 65^ 35 - 40^
This mixture, which was used without further treatment, solid- ifies to form a low-melting mass. When it was dispersed in the form of fine particles, it passed through the gas masks then in use, causing sneezing and vomiting. Since then, layers of felt have been put into the gee mask to mechanically filter the sternutator .
( 2 ) Diphenyl Cyanoarsine
This substance can be prepared by the reaction of hydrogen cyanide with diphenyl arsenious oxide as shown by the following series of reactions:
2 (C6H5)2 - AsCl + H30 —+ HC1 + [(C6H5)2 As]2 0
[(CSH5)2 As]2 0 + 2 HON * 2 (C6H5)2 AsCN + H20
This also is a low-melting solid, and is utilized in the same way as diphenyl chloroarsine.
(3) Phenarsa yine Chloride (Adamsite )
This compound was studied in 1918, but wp s not used in the last war. It wee prepared at Edgewood by the reaction between diphenylamine and arsenic chloride:
36
T Yield, Q0%
Later, it was found the t diphenylanine hydrochloride dis- sociates into the amine and hydrogen chloride at about 100- , so the following method was developed: £6kU
/ \
2 (C6HS)S NH.HC1 + Ass 03 >3H30 + 2HN AsCl
^6H4X 95$
This substance melts at 190 , end is also employed in the form of a fine dust.
( C ) Lung Injurants
Substances which have been employed as lung irritants include chlorine, thiophoegene , chlcropicrin, phosgene, and tri- chloromethyl chlorof ormate.
Due to its low persistence, chlorine was not used to a great extent. In addition to being a lung injurant, phosgene is very injurious to th- heart."' Phosgene was used extensively, but it has the disadvantage of hydrolyzing easily.
On the other hand, trichloromethyl chlorof orma te is quite resistant to hydrolysis at ordinary temperatures. This compound can be prepared in the following way:
0
CH30H + C0C12 0
■» HC1 + CI - C
!1:
\)CHo Ultra-violet
CI - c
\
0-CC1.
This substance is also known as diphosgene, since on heating it decomposes into two moles of phosgene. It is completely re- tained by filters of active carbon.
Chlcropicrin was prepared by the action of calcium hypo- chlorite on picric acid, but now ca-n be prepared by the action of chlorine on nitrometnane :
CH3 - N03 + Cl3 C? (0C1),5.) Cl:
C
NO.
Chloropicrin is not as toxic as phosgene or diphosgene, and of the war gases is one of those most easily held back by active carbon.
(D) Toxic Gases
Among the toxic gases are cyanogen chloride, dimethyl sul- fate, carbon monoxide, hydrogen cyanide, and tetrachlor'9 dinitro- ethane.
3*«
~4~
Carbon monoxide and hydrogen cyanide °re too volatile. Cyanogen chloride had a limited use Gs a war gas due to its tendency to trimerize to form cyaruryl chloride.
CI
N *N
3C1CN -> « i
C1~C J3-C1
Dimethyl sulfate is unsuitable because of its low volatility*
Since the last war, it has been found that tetrachloro- dinitroetha.ne is a very effective toxic gas. It hr s been pre- pared as follows:
2L -- - „„, N02 C13„C-N03
CI3C-CCI3 > ci2c = cci, -T6^ ci;j_N0;
This substance is six times as toxic and eight times as lach- rymatory as chlorcpicrin.
(E) Vesiccnt s
The most important vesicant used in the last war was mus- tard gas, P, |3r~dichlorodiethyl sulphide* In addition to being a vesicant, mustard gas is five times as toxic as phosgene.
The method of preparation used by the Allies was that developed by Pope in England and Levinstein in America, Ethylene is bubbled into sulfur monochloride while the temperature is kept below 35°, According to Conant; the following reactions take place:
S3C13 : S + SC13
/C1
CH3 = CH2 + SC12 -* S
NCH* - CHa - CI
CH3 = CH3 + s' ^H3 - CH3 - CI
XCH3 ~ CK3 - CI -* ?"
vH3 — CH3 — CI
The Germans prepared it from thiodiglycol and concentrated hydrochloric acid by the following series of reactions:
HC1 20H - CH3 - CH3 -CI + Na3S->S(CH3 - CH3OH)3 — -**-»
50
S(CH3 - CH3C1)
3
The Allied method was cheaper, but the German method was more easily controlled. Recently, a cheap method of preparation of thiodiglycol has been developed:
~5- 2H3S + 1 N)->S(CH3 - CHsOH)s oS
ch/
Consequently, the thiodiglycol method of preparation of mustard gas may now be the preferable method.
Bleaching powder reacts with mustard gas both as an oxidizing agent anc* as a chlorinating agent, rendering it harm- less. For this reason, bleaching power is used to decontaminate areas infected by mustard gas.
Chloramine-T reacts with mustard gas to form a sulfilimine:
CK3 -C6H.i-30^f/ + S(CH3 - CH3Cl)3^CH3 - C6H4 - S03 - XC1
N<~S(CH3 - CH3C1)
Thus, chloramine-T can also be used as a decontaminant .
Lewisite is ^-chlorovinyl dichloroarsine, Cl3-As(CH=CHCl)3, which is prepared from acetylene and arsenic trichloride:
^- „TJ . „, A1C1, fCl3As - CH = CHC1 CH= CH + AsCl3 ±±2±L^ ClAe(CH = CHCl)3
(As(CK = CHC1)3
The chlorovinyl dichlorearsine can be separated from the other two products by fractional distillation. The secondary and tertiary arsines can be converted into Lewisite by heating with arsenic trichloride, so that the over-all yield is satisfactory.
On oxidation, Lewisite yields chlorovinyl arsenic acid, CI ~ CH = CH ~ As03H3,. which is innocuous.
Lewisite, besides being a vesicant and a toxic gas, is an irritant to the eyes and lungs, so that it is potentially a very effective war gas. This gas was developed too late to be used in the last war, so that it has not been tested in actual combat.
The vesicants proved to be the most effective gases in the last war. For instance, in the last year of the war, 77% of the British gas casualties were caused by mustard gas.
Bibliography
Sartori, _fhe War Gases, D. Van Nostrand Co., (1939).
Prentiss, Chemicals in War, McGraw-Hill Book 0o0, (1937).
B. M. Vanderbilt (to Commercial Solvents Corporation). U. S.
Patent, 2,181,411 (1939). Ch em, Aba., 34, 1993, (1940).
Nenltzescu and Scarla.tescu, Antigaz. 9, 12, (.1935).
Othmer and Kern, Ind . Sng. Chem. 32 ', 160, (1910).
Reported by C. F. Jelinek March 4, 1942
39
DEVELOPMENTS IN QUANTITATIVE ORGANIC ANALYSIS
This report is an attempt to note only some of the high spots in recent developments of quantitative analytical tech- nique, For clarity, brief descriptions of some methods in common use will be included.
Signer has introduced a process of molecular weight deter- mination which, although not new, has received little attention. Clark describes the method, together with simplified apparatus for its application. Solutions of the unknown and of a known compound in the same so]. vent are placed in a sealed, isc thermally insulated system so that their vapors are in contact. Following Racult's law. the solvent from one solution will distill into the other until equilibrium is reached. A measurement ; of the volume of each solution will then give the molecular weight of the unknown according to the equation
M
G-XMV x ="GV7
0
where M is molecular weight, G is weight, and V is solution volume of the standard, and the symbols with subscript x refer to these values for the unknown. The advantage of the meth©d is its high accuracy (reported error usually less than 1%, frequently less than 0.4$); its principal disadvantage lies in the time consumed in reaching equilibrium (several days).
Most of the recent cha.nges in the standard carbon-hydrogen combustion determination have been introduced to take care of the nitrogen oxides from N-containing compounds, because the lead dioxide fails to do so satisfactorily. Proposed substitutes include other metal oxides to absorb the nitrogen oxides or metals to reduce them, or an absorbing agent placed in a U-tube between the anhydrone and the ascarite absorption tubes. Tests by Elving and McElroy en a large number of these materials showed them to be unsatisfactory, either because of incomplete removal or because of a tendency to absorb C02 or water. They recommend a strong oxidizing agent (potassium permanganate or dichromate) in concentrated sulfuric acid solution placed in a U-tube between the two absorption tubes. Niederl and Whitman mix the sample with copper oxide and employ a stream of nitrogen; the oxide oxidizes the sample and the freshly reduced copper reduces the nitrogea oxides.
Several wet oxidation methods have been developed to simplify the apparatus required for combustion. Oxidizing agents used include permanganate, iodate, dichromate, or persulfate, in solution in concentrated sulfuric or phosphoric acid. The vilume of evolved C03 may be measured, or it may be determined gravi- metrlcally or volumetrlcally . Christensen has recently combined several previous methods in the following microprocedure: the sample is oxidized with potassium iodate and sulfuric acid; the carbon dioxide is absorbed in standard barium hydroxide solution and the excess base is titrated. Since such a procedure gives
~c -
40
only carbon content, Williams back-titrates the excess oxidizing agent (iodate or dichromate) and from this value of equivalent oxygen 0R he calculates hydrogen content..
foil - (#uR 4- loc .- 11/3 '^C)/3 93
This equation is applicable to compounds containing only C; H, and 0. He reports that the accuracy of the latter value is greater than that from the usual combustion procedure.. Williams also uses the oxidation equivalent E and the exao I; molecular weight M of an unknown compound OHO to obtain its formula*
12x + y + 15 z = M 2x + 0t5y - z = M/E
This oxidation equivalent E is the number of grams of the com- pound equivalent to one gram-atom of required oxygen. M must be even, and M/E must be a whole number. The equations are solved for x and y_ in terms of z and a few trial values of z readily show which is correct. Compounds containing S and N can be determined similarly provided the content of these ele-' ments is known and certain corrections are made. Wet combustions fail with certain compounds. Acetic acid, phyhalic acid, and pyridine types are unresponsive to one or more of the oxidants mentioned. Also, dichromate frequently produces some carbon monoxide, and the use of phosphoric acid medium leads to oxida- tion of part of the ammonia from nitrogen compounds.
Dumas nitrogen microanalysis often gives high results, especially when the sample contains more than 40% N. Fischer ascribes this ta oxygen from a carbon monoxide- cuprous oxide- oxygen equilibrium at the high temperature in the combustion tube and suggests a separate furnace kept at only 200°C for the farther end of the tube. He reports a result 0,28^ high on a sample containing 80#N.
In the standard Kjeldahl amino nitrogen procedure the sample is digested with sulfuric acid and a catalyst t*> pro- duce ammonium sulfate; the solution is made alkaline with strong caustic and the ammonia steam-distilled into a standard acid solution. This method is limited to amines, amides, and amino acids; nitro groups must be absent, and heterocyclic nitrogen is in general not attacked. Sometimes the method is used for total nitrogen by first reducing it, as in the Friedrich pro- cedure which uses HI and. red phosphorus. The excess HI is removed with sulfuric acid, and heat. Belcher and Godbert have applied this method to a wide range of compounds, including those with hetero nitrogen; they recommend as a catalyst for the Kjeldahl digestion a mixture of potassium sulfate, mercuric sulfate, and selenium. The reported results were rather con- sistently low by about 0.1$ N.
Huffman has perfected a way by which sulfur can be deter- mined during carbon-hydrogen combustion. The oxides of sulfur are absorbed on silver pellets to form silver sulfate, and this
41
is electrolyzed in dilute edueous isopropyl alcohol (so that an adherent plete may be obtained.) Halogens must be absent.
The standard Zerewitinof f procedure has been modified by Evans, Davenport, and Revuke s for improving the accuracy of micro determinations of active hydrogen. They burn the hydro- carbon evolved and determine C and H in the usual manner. n-Butyl Grignard is used to obtain more weighable product per active hydrogen than methyl G-rignard would give.
The classical method for determination of acetyl is by trans- eeterif ication with "evthenol, distillation of the ethyl acetate and its saponification in standard alkali. Ma.tchett and Levine have eliminated the unwieldy procedure of introducing ethanol vapors during trans-eeterificstion by employing an efficient fractionating column with total reflux; small fractions of the distillate at the top are removed at Intervals. This also per- mits the use of HC1 catalyst in place of the less available aromatic sulfonic acids.
Methoxyl or ethoxyl content is commonly found by a modifi- cation of the Z-isel method. The alkoxyl is converted, by HI to the alkyl iodide, which is distilled into bromine solution. The excess bromine is destroyed by formic acid, potassium iodide is added, and the liberated iodine is titrated with thio sulfate. The re actions are:
RI + Br2 ->RIBr2 -> RBr + IBr
IBr + 2Br2 + 3H20->HI03 + 5HBr
HI03 + 5HI -> ftla + 3K20
Lisle reports * rapid method for approximate determination of methoxyl. Evolved methyl iodide i's massed over a tegrfc paper saturated with a palladium chloride-pyridine solution, which turns brown. After drying it is compered with a set of standards. The question of whether the methio group will react to the Zeieel procedure has been investigated. Arndt shows the relationship of structure to ease of splitting: S-elkyl> S-aryl > S-cerbonyl . He concludes that the method is generally not applicable, but that at the same time methio groups will interfere in alkoxyl determinations .
An illustration of the applicability of absorption spectra to quantitative organic analysis is given by the work of Gore and Fatberg on the determination of toluene purity. Most im- purities are aliphatic hydrocarbons. The aliphatic C-H bond shows a marked absorption peak in the Infrared region at 3.4y*, and another smaller peak at 3.49X'. The aromatic C-H bond exhibits a much lower oeak at 3.3,*, Fig. 1 compares the absorp- tion curves for cyclohexpne (upper) and toluene. Small quantities of paraffin present in toluene will show themselves in the peak at 3.4^ in spite of the toluene methyl. By arbitrarily assuming a single compound es the impurity (e.g, cyclohexane ) it is
-4-
possible to estimate the degre? fif purity of the toluene from thp relative absorption at this w?ve length according to Beer*s lav:
log I0/I = kc
where I, and I reprfsent intensity of radiation through pure toluene in solution and that through the sample containing im- purity of concentration c. Results from pure toluene with added traces of cyclohexane gave fig. 2. The procedure is rapid and indicates relative purity of the toluene with suffi- cient accuracy for industrial purposes.
:;
exti'nct. coeff.
10-
Fi3
wave
Bibliography: Pregl, Roth, and
P. Blakiston's Son Niederl and Niederl, "Micromethods
Elementary Analysis," John Wiley
Daw, "Quantitative and Go . ,
Orgfnic Micro-analysis," Phila., 1937, 2nd Eng. Ed.
of Quantitative Organic and Sons, Inc., New York,
1938
Ed. 2
Shriner, "Quantitative Analysis of Organic Compounds," 1941, Williams, J. Am. Chem. Soc, 59, 288 (1957); Williams, Rohrman,
and Christensen, ibid., 59, 291 (1937); Christensen, Williams,
and King, ibid., 59, 293 (1937). Arndt, Loewe, and Ozansoy, Ber., 72B, 1860 (1939). Lisle, Analyst, 64/ 876 (1939) Fiech£>2-, Chem. Fsbrik, 1940. 154.
Belcher and Godbert, J.^Soc. Chem. Ind. (Trans.) 60, 196 •<3J^iBterrt?en. Wong, and Facer, Ind. Eng. Chem., Anal. Ed.
364 (1940). Huffman, ibid. . 12, 53 (1940). Evans, Davenport, and Revukae, ibid., 12, 301 (1940 )
(1941) 12,
43
-5-
Matchett and Levine, ibid. . 13, 98 (1941). Christeneea ?nd Wong, Ibid. , 13, 444 (1941) Elving pnd McElroy, Ibid. . 15, 660 (1941). Cxore ?nd Prtberg, ibid., 13. 768 (1941). Clerk, ibid., 13, 820 (1941).
Reported by G-. E. Inskeep Merch 4, 1942
RECENT METHODS FOR MAKING -ACIDS I. Condensations with Oxalyl Chloride
44
In the presence of light and at room temperature oxalyl chloride and typical saturated hydrocarbons react with the form- ation of an acid chloride and the liberation of carbon monoxide and hydrogen chloride. The reaction appears to be general for paraffinic end cycloparaff inic hydrocarbons and may be represented^ p s
H2 H3
H H
A
H2
H3
Hs
+
(C0C1);
H.
H C0C1
H.
+ CO + HC1
H
2 Ky
K2
H,
The yields are quantitative— that is, a mole of acid chloride is produced for each mole of oxalyl chloride used. The conversions, however*, vary from a few per cent for n-heptane to more than fifty per cent for cyclohexane* The chief cause of low conversions is the formation of colored matter in the reaction mixture which shields trie oxalyl chloride from effective radiation. •
Benzene inhibits the reaction, apparently by the absorp- tion and degradation of the effective radiation. Toluene and related compounds Co not react. This is presumed to be due to the effect of the aromatic ring on the transmission of the radiation reouired for activation of the oxalyl chloride.
From a consideration of the energy involved the following mechanism is proposed
Primary process:
(COCl)s -^2-COCl
(C0C1);
I0C0C1 + .CI
Secondary process;
.♦COd'-* CO + CI- •C0C0C1 -> 2 CO + CI- RH + CI* -**•»• + HC1 R- + (COCl)a -+RC0C1 •C0C1 -> CO + CI- CI- + RH -*R« + HC1
-'r
:oci
It was found that the addition of small quantities of benzoyl peroxide to a reaction mixture of cyclohexane and oxalyl chloride induced a reaction in which the conversion was practi- cally complete and the yield was sixty-five per cent.
Neither light nor peroxides h°ve any apparent effect upon the action of oxalyl chloride- with unsaturated compounds. Gentle refluxing, however, brings ;bout a reaction which may be
-2-
represented ps
RHC=CH3 + (COCl)s ->RKC=CHC0C1 + CO + HC1
Yields very from over fifty per cent for 1,1-diphenyl- ethylene to six per cent for 1-methylcyclohexene.
Phenyl acetylene reacts somewhat differently
C6H5C=CH + (COCl)8 ->C6HSC=CC0C1 + CO
CI NH
Most unsaturated compounds do not react under these mild conditions but the effect of typical catalysts on the conden- sation was not studied. A highly polar double or triple bond- is a prereauisite for reaction. The reaction is unaffected by the usual catalysts and inhibitors for reactions involving atoms and free radicals. The authors, therefore, believe that the reaction is strictly of the polar type.
II. Tertiary Cerboxylic Acid Esters
Aston ''n<3 his workers have shown the t sodium alcoholates in ether act on a-bromo-a, a-oialkyl ketones to yield the ester of a tertiary acid. The following table gives some esters with the yield obtained using this method
Isopropyl trimethylacetate 64^
Methyl time thy lac eta te 39^x
Methyl trimethylacetate 61. Z%
Methyl methyl~t-butylacetate 73^
Methyl ethylmethylpropylacetate 75^2
x-the low yield is due to an excess of alcohol used in
preparing the alcoholate. 3-a trace of methoxy ketone formed in the reaction is
included in this yield.
The steps postulated for the reaction are:
R'(R»)CBr-CO-R" ' + NaOR -> R-7 (rM )CBr~C(OR ) (ONe )~RT T T
I -» R1 (rU )c_C (OR)-R ■ « t _> riri i.R. . iCC00R \/ II III
Two other reactions are possible:
1. Normal metathesis; e.g. 3-alkoxy-2-butanone from
3~bromo-2-butenone 2.
R*R'»C — -C(OR)-R'»l-+ R,,^OH-^R'R"C(OH)~C(OR)(OR'»").
\) II
15
46
-3-
When no alcohol is present (sodium alcoholate suspended in ether), reaction (2) which requires alcohol is not possible. The possibility that a change of medium alone is the factor influencing the rearrangement to the ester is eliminated by the fact that if any alcohol is present a corresponding amount of hydroxy ecetal is obtained in ether medium.
The formation of addition product (I) rather than the product of metathesis is ascribed to steric hindrance. That the presence of hydrogen on the a-earbon is not the determining factor is shown by the seventy-three per cent yield of the pure methyl ester of methyl-t~butylacetic acid obtained from 4, 4-dimethyl-3-bromo-2-pentanone. The rearrangement is retarded by making the rearranging alkyl or aryl group larger. Thus cc-bromoisobutyrophenone gave only the a-methoxyisobutyrophenone.
III. Glycolic Acid
Glycolic acid is produced by heating formaldehyde, carbon monoxide and water at 160-170° for one hour at pressures between five and fifteen hundred atmospheres. An inorganic acidic material such as sulfuric acid, dissolved in an organic acid such as acetic acid, serves as a catalyst. Other suitable catalysts are boron trif luoridc: , hydrochloric, phosphoric, formic and glycolic acids.
IV. Substituted Acetic Acids
Substituted acetic acids are formed when carbon monoxide, formaldehyde or one of its polymers and an inorganic acid react in the presence of en acidic catalyst according to the equation
n (ECHO) + n (CO) + HnX -> X(CHsCOOH )
When sn organic acid is used, acyloxyacetic acids are obtained. Thus propionoxyacetic acid would be obtained from formaldehyde, carbon monoxide and propionic acid.
V. Acetic and Propionic Acids
Acetic acid is produced when methyl alcohol and carbon monoxide are heated at a temperature of 125-180° and pressures greater than twenty-five atmospheres, 700-900 atmospheres being most suitable. Boron trifluoride and water are used es the catalysts.
By the use of ethyl alcohol and carbon monoxide under similar conditione, propionic acid is produced. Propionic acid is also made from a hydra ting agent such as water, carbon monoxide, and ethylene using boron trifluoride and one to three moles of water as the catalyst.
In an analogous manner, polycerboxylic acids are obtained from polyhydroxy alcohols such as ethylene glycol, propylene glycol and. glycerol.
-4-
The boron trifluoride catalyst is prepared by treating anhydrous liouid hydrogen fluoride with boric acid, its anhy- dride or a borate at temperatures below 10°. The reactants are used in such a proportion that BF3*2H30 or BF3*3H20 will be formed.
Dreyfus has patented a catalyst consisting of traces of copoar end ammonium phosphate for the production of acetic acid by this procedure. The Eastman Kodak Company has a catalyst containing small amounts of the oxides of zinc and manganese and larger amounts of the oxides of copper with a binder such as sodium silicate or cellulose acetate. Other suitable catalysts are the fluorides of magnesium, calcium and titanium and the halides of numerous other metals.
VI. Maleic Acid
A process recently patented by the Standard Oil Develop- ment Company consists in passing a mixture of oxygen, an oxid- izable unsaturated hydrocarbon containing at least four carbon atoms and water vapor over a catalyst at a temperature of 250-400°. The oxides or salts of vanadium, nickel, tungsten, chromium, manganese, molybdenum or mixtures of these are suitable catalysts.
VII. Fumeric Acid
Fuma.ric acid is produced by the oxidation of furfural. This is accomplished, by moderate heating in a chlorate solution using vanadium trioxide and one or more of manganese trioxide, aluminum oxide or ferric oxide as the catalyst.
47
VIII. Soap-forming Carboxylic Acids
British and German patents describe a method for the preparation of soap-forming carboxylic acids or their salts by the oxidation of technical mixtures of hydrocarbons and fusion with alkali metal or alkaline earth metal basss. The hydro- carbon must contain one unsaturated link in en aliphatic chain and at least one saturated aliphatic chain not less than six carbon etome in length. The unsaturated tar oils boiling at 200~?50° obtained from the distillation of lignite, cracking-gas oil fractions boiling at 200~;7>250 (seventy to seventy-five per cent of which are olefins having twelve to eighteen carbon atoms) and olefinic products from the hydrogenation of carbon 'monoxide are suitable starting materials..
Permanganates, oersulf stes, hypochlorites, hydrogen peroxide, chlorine, oxygen and air are used as oxidizing agents and in some cases tne hydrocarbons are emulsified before oxid- ation.;; The use of solvents or diluents such ss mineral oils or saturated hydrocarbon? which may be present in the initial hydrocarbon mixture is beneficial.
48
-5-
In a typical example, a hydrocarbon having a boiling point above 250° is oxidized at 115° by means of air in the presence of heavy metal salts of unsaturated acids. The dried product may then be fused with alkali at 250-280° * The product consists of salts of stearic, palmitic, myristic", lauric and capric acids,
IX. a-Chloroacrylic Acid
Paraformaldehyde, trichloroethyiene; 98^ sulfuric acid and metallic copper (in the ratio 30' 132-84-1 parts) are heated at 25-30°; the temperature finally being raised to 45°. cc-Chloroacrylic acid is obtained from the reaction mixture
Bibliography
Aston, Clarke, Burgess
64, 300 (1942) Kharasch, Kane and Brown, i French Pat. 831,474 (1938); U.S. Pat. 2,153,064 (1939)
and Greenburg, J. Am. Chem. Soc
bid. ,
Chem. ibid 3
British Pat. 508,383 British Pat. 527,644 British Pat. 527,645 U. S. Pat. 2,135,453 U. S. Pat. 2,170,325 British Pat. 490,544
1939 1940 1940 1939 1939 1938 1940 1939 1939 1939 1940 1939 ( 1940 (1938 U. S. Pat. 2,211,624 (1940 French Pat. 845,230 (1939);
u. |
s. |
Pat. |
2> |
217, |
650 |
u. |
S. |
Pat. |
2 |
135, |
459' |
u. |
S. |
Pat. |
2, |
135, |
454 |
u. |
s. |
Pat. |
2, |
165, |
428 |
u. |
s. |
Pat. |
2, |
260, |
409 |
Br |
iti |
sh Pa |
t. |
498, |
398 |
Ja] |
Danese Pat. |
137,755 |
|||
Briti |
sh Pat. |
491, |
927 |
ibid ibid ibid ibid ibid ibid. ibid ibid ibid ibid, ibid, ibid, ) ibid. ; ibid, J ibid, ibid.
64, 333 (1942).
Abs. 33, 2539 (1939) 3, 5006 (1939).
34, 451 (1940).
35, 6981, (1941). 35, 6981 (1941).
33, 995 (1939).
34, 116 (1940). 33, 646 (1939).
35, 1070 (1941). 33, 996 (1939). 33, 1112 (1939). 33, 3214 (1939).
36, 784 (1942). 33, 4272 (1939). 35, 1806 (1941). 33, 1347 (1939). 35, 1068 (1941). 35, 1070 (1941).
Reported by G-, W. Cannon March 11, 1942.
.
:•
.
19
THE PR INS REACTION
This reaction, discovered by Prins, is the condensation of formaldehyde with an olefin.
RCH=CHR + CH30 -> RCH CHR H3° » RCH-CHROH
I 1
CHo— 0 CHoOH
SRCH CKR^ /
I \ 7
nu Viij
^0
In some cases the 4-membered ring may rearrange, or the glycol may dehydrate, to give an unsaturated primary alcohol. . The condensation is brought about by sulfuric acid in acetic acid, formic acid, or water. Prins carried out his initial experiments on styrene, anethole, isosafrol, a-pinene, camphene, and cedrene.
With unsymmetrical olefins, for example styrene, the addition might proceed in two ways, giving rise to the products:
A. QfCH—CK3 -* 0CH— CH30H
If I
CK3~0 CH20H
B. 0CH— CK2 -> 0CH— CK2CH30H 6.
-CH3 OH
Only one product was obtained; Prins believed it to be of the structure A, since it gave a diacetate with boiling acetic anhy- dride, instead of cinnamyl acetate.
Fourneau, Benoi't and Ferminich conclusively proved the structure to be the unsymmetrical glycol B. They based their conclusion on the following evidence:
1. Saponification of the diacetate yielded large quantities of cinnamyl alcohol.
?. The glycol, 1-phenyl-l, ^-propanediol prepared by
Rupe by reducing 0COCH=CHOK gave a dibenzoate of the same melting point (51°) ^s that from Prins' glycol.
?. The unsymmetrical glycol was prepared from a chloro- hydrin of known structure.
50
-2-
C1CH3CH2CH0 + 0MgBr -^ 0CH-CH3CH3C1 ^COgNa> 0CH-CH3GH3OCO0
OH OH I S0G13
0CHCH3CH3OCO0 0 ^3 * ' •0CH-CH3Cti3OCiO0 i(CHa), ci
Prins' glycole -^>Chlorohydrin (CH3)?iNH> JGOGJ,
3,
0CHCH3CH3OCO0
N(CH3)3
The two amino esters were identical.
Prins later offered an explanation as to the mode of addition of formaldehyde to an un symmetrical olefin derivative. He assumed that the negative oxygen atom would add to the more positive carbon atom of the ^C=G<I and he further assumed that a carbon atom would be the more positive when a hydrogen was substituted by an atom with a high electron affinity. Thus the reaction was studied with di-, tri~, and tetrachloroethylene The neutralization of the cha.rge on the carbon to which the oxygen became attached, activates the chlorine atom so that immediate hydrolysis occurs.
The reaction with dichlorcethylene gave only a resin. Trichloroethylene yielded CH30HCHC1C00H which lost water to form the ethereal acid 0(CH3CHC1C00H ). The oxygen attacks the GCl2=group almost exclusively. Tetrachloroethylene yielded the dichloroacid CHa0HCClPC00H.
Recent Developments and Applications .
Marvel, Harmon, and Riddle found that acetaldehyde con- denses with vinyl acetate in the presence of sodium to give a cyclic acetal reminiscent of the classical Prins reaction.
2CH3CK0 + CHa=CH0C0CH3 -> CH3CH-CHaCH-0C0CH3
Or J)
I CH3
Acid condensing agents had no effect, and no other aldehyde could be made to undergo the reaction.
Nenitzescu and Przemetzky, in searching for new routes to 8-methyl~l~hydrindanone and similar compounds, discovered that the best method involved the Prins reaction.
CH3
+ (CHaO).
M
-3-
/\
CH-
t >CH30H
V
PBr.
51
Ma Ionic ^NyCHa
ester
A patent has been issued for the f or ma. t ion o'f a synthetic plastic from formaldehyde and a cracked petroleum oil fraction. The reaction is brought about by an acid condensing agent in the presence of a promoter such as acetic acid.
Cyclic formals can easily be synthesized by reacting one mole of an aliphatic monoolefin with about two moles of formaldehyde in the presence of an acid condensing agent. Similarly, a source of 1, 3-butanediol is the reaction of formaldehyde hydrate with propylene at super-atmosphere pressures using HX and ZnCl2.as catalysts.
The important acid, ot-chloroacrylic acid may be manufactured by heating formaldehyde with trichloroethylene in concentrated sulfuric acid, and heating the subsequent mixture in the presence of water.
Bibliography:
Prins, Chem. Weekblad, 16, 1510 (1919)
Prins, Proc. Acad. Sci . Amsterdam, 22, 51 (1919); C. A. 14,
1662 (1920 J Fourneau, Benoit and Ferminich, Bull. Soc. Chim., 47, 860 (1930 ) Rupe and Miller, Helv. Chim. Acta, 1921, 4, p. 841 Prins, Rec. Trav. Chim., 51, 469 (1932) Marvel, Harmon and Riddle, J. Org. Chem. 4, 252 (1939) Nenitzescu and Przemetzky, Ber. 74B, 676 Tl941 ) U. S. Pat. 2,035,123 (1936); C.A. 36,3120 (1936) British Pat. 507,571 (1939); Ibid, 34, 449 (1940) U. S. Pat. 2,143,370 (1939); ibid. 33, 2914 (1939) French Pat. 845,230 (1939): ibid, 35, 1070 (1941) U. S. Pat. 2,233,835 (1941); ibid, 35, 3651 (1941)
Reported by R. E.Foster March 11, 1942
Ghloro sulfonic Acid As An ■
Organic Reagent ^>&
I. Preparation of chloro sulfonic acid.
II. Action as a sulfonating and chloro sulfonating agent on
Hydrocarbons and derivatives Hydroxy compounds Aldehydes and ketones Acid derivatives
III. Reactions involving oxidation and chlorination.
IV. Reactions as a qualitative organic reagent.
I. PREPARATION
Numerous examples have been cited in the literature for the preparation of chloro sulfonic acid, C1S03H, but all are essentially based upon the production of hydrochloric acid and its reaction with sulfur trioxide:
HC1 + S03 = CISO3H
For example, dry hydrogen chloride is passed into oleum until no further absorption occurs, and the chlorosulfonic acid separated by distillation in a stream of hydrochloric acid. Perhaps a more widely used method is the addition of solid sodium cnloride to oleum containing 60-80 mole per cent of sulfur trioxide and finally separation by distillation.
Chlorosulfonic acid itself is a colorless, fuming liquid which boils at 158°, melts at -88°, and has a sharp unpleasant odor. It is very corrosive and therefore must be handled with care.
II.. SULFONATION AND CHLORO SULFONATE ON
The action of chlorosulfonic acid as a sulfonating reagent, stressed by Gebauer-Fuelnegg and Haemmerle, is only one of its extensive and varied reactions.
Starting first with the aromatic hydrocarbons, the initial step in the reaction is .the elimination of hydrochloric acid and. the formation of the corresponding sulfonic acid which, in the presence of excess chlorosulfonic acid, reacts to form the sulfonyl chloride, thus:
C6H6 + CISO3H = C6H5S03H + HC1, (ben?enesulfonic acid) C6HBS03H + C1S03H = C6H5S03C1 + H3S04 (benzenesulfonyl
chloride )
53
The mechanism ie perhaps oversimplified, for the products depend intimately upon the experimental conditions: the time of reaction, the temperature, the quantity of reagents, and the nature and order of additions of the reacting substances. The type of reaction is determined by the temperature of the reaction and the amount of chlorosulfonic acid. The results may be summarized by the following; observations:
A. At low temperature, that is, room temperature although many reactions are run at 0° and below, and in the presence of equivalent quantities of the reacting sub- stances, the sulfonic ^cid is produced.
B. At low temperature and in the presence of an excess of chlorosulfonic acid, the acid chloride is produced.
C. At higher temperatures, 150°, and in the presence of equivalent amounts of reacting substances, the sulfone is produced which may be chlorosulfonated further.
D. At the higher temperatures, and in the presence of excess chlorosulfonic acid, the product obtained is the sulfonyl chloride. Under these forced conditions, more than one sulfonyl group may be introduced.
To illustrate these:
room
CeH6 + C1S03H (excess) — - * C6H5S03C1 (76#)
temp. (benzenesulfonyl chloride) 150°
C6H5 + C1S03H (excess) -» * m-C6H4(S02Cl ) + sulfone.
3 hrs. Tbenzene-m-disulfonyl chloride)
150° excess
C6H5(20 g. ) + C1S03H(15 g. ) — 5> n
10 min. 5 hrs.
(diphenylsulfone-m-di sulfonyl chloride )
Other organic compounds are indicated in the following table:
54
-3-
Hydrocarbon
Product s isolated
Toluene
o-xylene m-xylene p.- xylene
Naphthalene
Biphenyl
Chlorobenzene Chlorobenzene and Bromobenzene l-Chloro-4-nitro
benzene Ethyl phenyl sulfone Aniline m-phenylenediamine
o- sulfonic acid
p_-sulfonie acid
2, 4-disulfonyl chloride
di (p_~methylphenyl-m~sulfonyl chloride )
sulfone 3, 5-disulfonyl chloride 1, 3-dime thy lphenyl~4- gulf one 1 , 4-dimethylphenyl-2-sulf one 2-eulfonyl chloride 1-sulfonic acid 1-sulfonyl chloride 1, 5-disulfonyl chloride mono sulfonic acid 4, 4'-disulfonyl chloride ?., 4-disulfonyl chloride p-mono sulfonic acid diholosulfobenzide 2-eulfonyl chloride
m- sulfonyl chloride
2,4, 6~tri sulfonyl chloride
4,6 disulfonyl chloride
In its action on aliphatic compounds, chlorosulfonic acid is similar to sulfuric acid. Branched chain hydrocarbons are much more reactive, and these offer a possible means of purifi-
cation. Ethylene
adds
C1S03H to form CSH50S03C1
Hydroxy compounds of very extensively and seem following mechanism: A mo followed by the eliminatio fate which undergoes a Eri acid, whence the sulfonyl selected preferentially.
(1) C6HsOH + C1S03H
the aromatic series have been studied to bring about sulfone t Ion by the lecular addition compound is formed n of hydrochloric acid to form & sul- es rearrangement producing the sulfonic chloride, the para, position beimg For example: OS nr*CeHBOS03H +j3-K0C6H4S03H
-15
CH.
phenyl hydrogen sulfate
4-hydroxy benzene sulfonic acid
(2)
,0H
V*
+ C1S0 (exc
3H - ess;
AH3
, 2-hydroxytoluene- 2 ^ i d e
+ ClSOoH
110° i
1 Hr„
QH3 /\s, 0 SO 3 f^s .SO s c 1
s
0
JO, _
J
CK.
ra-cr e so lsulf onylide-di sulfonyl
-4-
The sulfate of equation (l) is formed if the reaction is carried, out in the presence of pyridine or some other tertiary base. Other examples are given in the following"
Reactant Prcduote
i \
p_-cresol j)-toluyl hydrogen sulfate
4,4' -dime thyldiphenyl-1 , 2 , 1 ' , 2 ' -
sulf onylide-6, 61 -disulfonyl chloride 2,3-dimethylphenol 4, 6-disulfonyl chloride
2, 3, 2', 3 '-tetrp.me thyldiphenyl-1, 6,1' , 5,-sulfonylide-4, 4'-disulfonyl
chloride 2,4-dimethylphenol 6-sulfonyi chloride
1,3,1' , 3'-tetramethyldiphenyl- 4, 5 , 4 • 5 » - sulf o ny lide
In the aliphatic series, hydroxy compounds tend to form the sulfuryl chloride, as ethyl alcohol produces C2H50S02C1. One interesting application is its use to produce sulfated sugars. These are converted to the acetates when treated with acetic anhydride.
Formaldehyde reacts with chloro sulfonic acid to produce C1CH30S0C1, CH2S04, (C1CH20)3S02, and (C1CH2)30 as if the hypo- thetical C1CH30H had been formed. Homologues are attached in the alpha position, producing sulfonic acids; however chloral, being stable in this position, produces (CCl3CKCl)30 and chloralide.
Acetophenone is of interest because of its unique reaction iY°-CH* ; 01S0.H „ ffV-CH^O.CX
\s k.^-so3ci
Reactions on acid derivations are indicated in the following table :
Reactant Products
Sodium acetate acetic anhydride
CHgOCOCl CHgOSOfcCl
ArS03K ArSCUCl
Salicylic acid 3, 5 -disulfonyl chloride
CH3(CONHR)2 di sulfonic acids
The most spectacular reaction of this group is the action of chloro sulfonic acid on acid chlorides.
nu nr\n% C1S03H 45O poo
CH3C0C1 *--> CH3C03S03C1 -2£-> CH2 (S03H )C0C1 -^—>
CH3S03C1 + C02
03H
-5-
56
Above 60°, s 3^ yield of 2-methyl-l, 4--pyrone-6~acetic acid was obtained. Prooionyl chloride reacts similarly to form CH3CH(S03K)C0C1 and a by-product of ot'-ethyl-p, B •■ -dimethyl- phorone; butuyryl chloride does not undergo these condensation reactions, but rather forms dipropyl ketone.
III. OXIDATION AND CKLORINATION
These reactions of oxidation or chlorination are more or less exceptions to the rule rather than general in application, however their presence is real and must be considered; besides they are interesting and unusual.
Res c tent
Products
Hydroquinone
Nitrobenzene
p-phenylenediamine
p-phenylenediamine (ClS03Na)
Pbloroglucinol
Resorcihol
Naphthalene
Cyclohexadiene
Thiophenetole
Thiophenol
Chloranil
Penta chlorophenol
Chloranil
Chloranil
Tetrachlorophenylendiamine
Pentachlorophenol
Hexachlorobenzene
Tetrachlorophthalic anhydride
Benzenesulfonyl chloride
Pe ntachl or 0 thiophenetole
Diphenyldi sulfide
Huntress and co-workers recently applied this versatile and reactive substance to the identification of organic compound: aryl halide, alkyl aryl ethers, and alkyl benzenes. The results have proven very satisfactory, and a procedure has been stan- dardized for the identification of such compounds. It consists of adding an excess of chlorosulfonic acid to a cold solution of the unknown in chloroform solution, separating the sulfonyl chloride and converting it to the amide by means of solid ammonium carbonate or concentrated ammonium hydroxide.
C1S03H » XArSOoCl
Ar X
(NH4)aC0;
4 XA.r S02NH.
He found that under these possible is invariably formed.
conditions the para isomer where
In general the halid.es of benzene, toluene, and naphthalene were studied. The iodo compounds all reacted abnormally, thus
X
(1)
S\
I
(2) C6H5I
chlorination, -. ^
(p-IC6H4)sS0;
57
-6-
Constituents in the ortho and mete positions of iodo com- pounds reacted according to (l); sulfonea as in (2) were also produced from f luorobenzene , and from o_-dichloro- and p_-dibromo benzene.
Some forty-two alkoxybenzenes were studied and found to react analogously, with only biphenyl compounds behaving pecuriarly.
r> a
In the case of the alkylbenzenes, the para orientation w. proven by permanganate oxidation to the corresponding sulfamido benzoic acid where possible. That only one sulf amide group had been introduced under tne conditions of the reaction was established by a nitrogen determination.
It was pointed out that even in the cases where abnormal reactions had taken place, the products served just as well for the purposes of identification.
Bibliography
Huntress and Autenrieth, J. Amer, Chem. Soc . , 63, 3446 (1941)
Huntress and Carten, ibid. , 62, 511, 603 ( 1940T~
Pollak and Gebauer-Fuelnegg. Monatsh.,47, 109, 537 (1927),
48, 187 (1928) Lustig and Katscher, ibid. , 48, 87 (1927)
Pollak, Krauss, Katscher, and Lustig, ibid. , 55, 358 (1933) Burkhardt, J. Chem. Soc, 1933, 337 Young, ibid. . 75, 172 ( 18997" Chem. Reviews, 25, 67 (1939)
Reported by S. S. Drake March 18. 1942
SOME DEPIV&TIVES 0r THE POLYHYDROXYBENZENES
Wilson Baker, et al Oxford University
This paper attempts to cover only a few of the many possible polyhydroxy benzene derivatives. It is chiefly con- cerned with certain tetra-, pent a- and hexamethoxybenzenes and some naturally occurring compounds that may be derived from them
Tetrahydroxybenzene Deri vat iv es
Compounds derived from tetrahydroxybenzene are not often found in nature, but of these substances the derivatives of 1 :?: 3: 5-tetrahydroxybenzene occur most frequently. Many of these have been determined, and found to be flavones, flavonols, and isofiavones.
Derivatives of 1:2: 4:5- tetrahydroxybenzene are contained in certain lichen coloring matter, and in embeiic acid (2:5- dihydroxy-;3-lauryl-p_-benzoc'Uinone ) . The latter- has been syn- thesized by Hasan and Stedman (9) as follows:
GO0lxH2; ^XOH ^
HO
V 0
C0CxlHS3
3n25
HO
fs/ ^v
0*\j?
</
o^MHMe
C 12^-2 s(n) G13H25
Derivatives of 1:2:3:4 -tetrahydroxybenzene are repres- ented in nature by parsley apiole (l :4-dimethoxy-2: 3-methylene dioxyallylbenzene ), fraxetin (7:B-dihydroxy-5-methoxycoumarin ), and dill apiole ( 3:4-dimethoxy-l :2-methylenedioxy-5-allyl-benzene ). T^o syntheses of the basic cc-mpound .1: 2: 3:4: tetramethoxybenzene are Bhwn below v' 1 , 2 ) :
OH CHO
Ar OMe .^
>0H
v/0H
OMe
Ac
V
OMe
OMe
-2-
I
OMe \OMe
OMe
i, }
OMe
f
I
OMe OMe
OMe
A OMe
OMe
!>OMe
The product, (l:2-dihydroxy-3:4~dimethoxybenzene ) , resulting from the first series of reactions, was used by B? ker and Jukes (?) in their synthesis of dill apiole, which occurs in dill plants, matico oil, and sea-fennel oil.
OH
/Nop
H
OH
r
OH
OMe
3Hp=0HCH
OMe
s^ /OMe V
OMe
CHa=CHCHg
OMe
Dill Apiole
Pentahydroxybenzene Derivatives
Many derivatives of pentahydroxybenzene have been isolated from plant sources. Some of these include nobiletin (5:6:718:3': 4,-hexamethoxyflavone ) , spinulosin (?:6-dihydroxy-4-methoxy- 2:5-toluquinone ) , erianthin (5:7-dihydroxy-?:6:8:3' :4T~ pentamethoxy flavone), pedicin (2:3~dihydroxy-4:5:6-trimethoxy- phenyl styryl ketone), and the closely related compound pedicellin (dimethyl ether of pedicin).
Four separate methods have been recorded in the literature for the preparation of pentahydroxybenzene derivatives, the yields varying and in general not good. Baker (5) has devised two relatively simple procedures which heve been outlined in Flow Sheet A. Flow Sheet B shows some simple reactions of these compounds, and in Flow Sheet C are found two syntheses of pedicellin (5) and also one of funagatin as worked out by Baker and Raistrick (6). This latter product, 3~hydroxy-4- methoxy-2: 5-toluquinone, can be isolated as maroon-colored needles from a mould metabolic product.
Hexahydroxy benzene Derivatives
The only derivative to be considered in this paper is hexamethoxybenzene . This compound has been prepared by Robinson f nd Va s ey ( 4 ) , a no by Ba ke r ( 5 ) .
60
MeO
OMe
8
-3-
MeO,/ Br
Y
OMe Br
MeO
OMe MeCO^S.OMe
MeoL JiOMe
OMe
MeO MeO
Ac 0
^\.0Me
U
OMe
/
/
0 Ac
OMe MeOf^\.OMe
MeO
OMe
OMe
Bibliography
1.
2.
3.
4.
5.
6.
7.
8.
9. 10. XI. 12.
Brker and Smith, J. Chem. Soc Baker, Kir by, end Montgomery, Baker and Jukes, ibid. \ 1934, 1681. Robinson and Vesey, ibid., 1941, 660. Baker, ibid. , 1941, 662. Baker and Raistrick, ibid. , 1941 , Organic Syntheses, 14, 40 (1934) Einhorn, Cobliner, and Ffeiffer, Hasan and Stedman, J. Chem. Soc,
, 1951, 2542. ibid. , 1932, 2876.
670.
Ber., 37, 119 (1904) 1951, 2113.
Chapman, Perkind, and Robinson, ibid. . 1927, 5028. Graebe and Kess, Ann. 340, 237 (1905). Dakin, Araer. Chem. J., 42, 477 (1909).
Reported by John D. Gerber March 18, 1942
Gi
Flow Sheet A
OH HO^\
OMe
V
OH
OMe
s
MeOi-
V
OMe -OMe
OMe
♦ f ij0He
MeO^J>OH COMe
OMe
OMe
>
MeO^x
OH
Me0\/;
OMe
>OMe
OMe
C03H
COMe
HO^v >OH OH
COMe
f>
<<v>OMe OMe
— >
COMe
rAoH
,
IO^v _/OMe OMe
■nOMe 4?
H ^V OH
U-OH OMe MeoLJJoMe
OMe OMe
Flow Sheet B
OMe
A
OMe
M©0^. 1LOH
OH
0
Moo.
OMe
'OH
II
0
OMe ?0\loAc
OH
A
MeO^^y OH
OMe OAc
MeO
OAc
rS V
OAc
OMe
OAc
02
OMe
MeO
OMe
V0H
COMe
Flaw Sheet C
HO HO
OMe ^OMe
-sy
OH
COMe
COMe
MeO MeO
OMe
*w*OMe OMe'
0 0 8cKaC0
MeOr^ 3 OMe Meotv >OMe
OMe
MeO-k AOVie OMe
OH MeCOr^ j|OMe
MeO-LjoMe OMe
CCH=CH0 MeOr^r\0Me
MeO ^/ OMe
OMe Fedicellin
Me
OMe
V/OMe
OMe
COMe Mer5^ >rOH
OMe
OMe
OH Me^\
OH
OMe
0
Me^ ^OH \ J^OMe
Y
Fumagatin
.
•-
.
'
63
RELATION BETWEEN OPTICAL ROTATORY POWER AND CONSTITUTION
OF THE STEROLS
Evidence for the correlation of optical rotatory power with constitution of the sterols was presented by Callow and Young in 1936 after a study of known pairs of compounds between which the only difference was an inversion of carbon atoms 5, 4, 5 or 17 or the introduction of a double bond. In 15 out of 18 cases there was an increase of dextro-rotatory power by a change of the 3 hydroxyl from cis to trans relative to the 10 methyl. Introduction of a double bond at the 4:5 position results in an increase in dextro-rotation, while introduction at the 5:6 position gives a decrease in dextro-rotation, A double bond at 8:14 brings a small decrease and at 14:15 a small increase in d-rotation. Changing the 17 ketone group to a secondary carbinol brings a decrease in dextro rotation.
However, the results of Callow and Young were merely quali- tative; the sign of the change but not the magnitude could be predicted. And exceptions were not rare, even in the limited data available, although many of these might be ascribed to impure materials or uncertainty as to structure.
Gorin, Kauzmann and Walter showed the t optical superposition depended on certain very special conditions to which the carbo- hydrate molecules readily conform. Steroid molecules in general cannot conform to these conditions and any method of optical superposition will not succeed.
Berstein, Kauzmann and Wallis have worked out a method which in effect divides the steroid molecule into two parts, the ring system and the side chain, each of which parts functions as a unit optically and is independent of the other part.
Consider the pairs of compounds, stigmastane and ergostane, stigma stanone and ergostanone (see next page). In each case the only difference is in the side chain, one having an ethyl the other a. methyl on C-24, If the molecule is considered as being made up of two independent regions of asymmetry, the atoms in the neighborhood of C-3 and the side chain, 'then the differ- ence in molecular rotation between the substances in each of the above pairs should be due only to the difference in effect of an ethyl and a methyl at C-24, and therefore both differences should have the same value.
stigmastane +10,470° stimastanone +17,000°
ergostane + 7,670° ergostanone +13,960°
+ 2,800° + 3,040°
Since an error of 1000° in molecular rotation corresponds to about 2.5° of specific rotation, the agreement of these values is very good.
-2-
G4
Now compare the pairs ergoetane snd ergostanone, stigmastene and stigmastanone. In this case the difference in rotation should be due to the effect of a methylene group and its environment and a carbonyl group and its environment at C-3.
ergostrnone ergost?ne
+13,960° + 7,670° + 5,290°
stigmastanone stigma stane
+17,000° +10.470° + 6,530°
31
CH.
ai ^ 23 » /
OHCH3CH2CHCH
CH.
r^^k^
I N
CH.
CH.
OM
H
CH.
X
CHCHpCHoCH
I \
4
CH.
CH, CH;
Ergostanone
Stigma stane
pH3 CH3
CHCH2CH3CHCH
NCHS
\
2H5
CH.
CH;
CHCHpCHpCHCH
AA
I CH3 C3H5
Stigmastanone
Whenever the environment of the C<*3 is the some as in ergostane — that is, with saturated rings and with the 0-5 hydro- gen trans to the C-10 methyl — the difference in rotation between the methylene and carbonyl compound should be +6,400°.
cholestanone cholestane
+15,840° + 9,160° + 6,680°
Tsitostanone ysi to stane
+15,760° + 8,090° + 7,670°
05
-3-
However, if the C-5 hydrogen is cie to the C-10 methyl, we would expect the difference between the rotations of the carbonyl and the methylene compound to have a different value.
coprostanone +14,010° coorostane ± 9.430°
+ 4,580°
In order to use these relationships more easily, we can set up a system of notation. Cholestane is taken as the reference compound and its molecular rotation is denoted by the letter C. In the change to cholestanone, we can symbolize the change in rotation as K3t, Thus the rotation of cholestanone is C + K3
The change in rotation in going from cholestane to ergostane can be called Erg. Thus the rotation of ergostane is C + Erg. Then knowing the numerical" values for the rotations of these molecules, we can easily find the value for the individual con- stants.
cholestane = +9,160° = C cholestanone = 15,840° = C + K3t
K3t = 15,840° - 9,160° ergostane = 7,670° = C + Erg Erg = 7,670° - 9,160°
Using this notation, we can indicate the rotation of ergo- stanone as C + K3 + Erg + e where e represents the difference in interaction of the ergostane and cholestane side chains with the carbonyl and methylene groups at C-3. But since two centers of asymmetry which are far apart do not influence one another's contribution to the optical rotation, e should be negligible. Then since we already know the values of the other three symbols, we should be able to calculate the molecular rotation of ergostanone.
C + K3t + Erg = +14,350° observed value = +13,960°
In this manner a table was set up (Table I) for the values of a large number of symbols which were assigned to specific changes on the sterol framework and then these values were used to calculate the rotation of a number of sterols (Table II). The calculated and observed values agreed fairly well in most cases. The authors ascribed large discrepancies to inaccuracies in obeerved values (difficulties of purification) or to errors in assigned structures.
In their most recent paper, Berstein, Wilson, and Wallis adapt a similar method to the calculation of the rotation of derivatives such as acetates, benzoates and m-dinitrobenzoates (Tables III and IV) and apply the method to recent experimental results.
66
o |
o |
O |
o |
O |
O |
o |
c |
O |
|
CD |
CD |
D- |
02 |
<st< |
O |
O |
CD |
0> |
o |
3 |
rH |
Ol |
<# |
W |
rH |
to |
tO |
C5> |
Oi |
H |
•* |
||||||||
01 |
OJ |
BO |
CM |
•* |
<j> |
LO |
|||
> |
rH |
Oi |
LO |
||||||
+3 |
+ |
H- |
+ |
1 |
V |
-I- |
+ |
1 |
1 |
■■d q |
|||||||||
C cd |
|||||||||
to -P |
GO |
||||||||
03 |
. . |
||||||||
rH C |
l^ |
||||||||
O O |
- |
||||||||
/3 o |
in |
tu |
U) |
||||||
F; |
• • |
• . |
9 . |
||||||
>i<H |
o |
43 |
o |
43 |
rf |
if) |
Ifl |
||
co o |
o |
PQ |
f — i |
3 |
M |
w |
n |
« |
P |
CD H1 lO lO Q ....
n a O) Q) -^ LO
G O rH lO WW
i_iO + I + +
to |
H |
CD |
IV |
|
Po |
» |
• |
9 |
• |
—I rH |
to |
LO |
^ |
O |
£ cd |
rH |
LO |
OJ |
CO |
_iO |
+ |
I |
+ |
+ |
OJ |
0 |
0 |
0 |
0 |
|||||||||||||||
G CO |
00 |
0 |
*M |
03 |
rH |
||||||||||||||
q 4J |
• • |
t> |
CO |
r-H |
CD |
||||||||||||||
<D £ |
^ |
i — 1 |
. |
•N |
•\ |
•» |
•> |
||||||||||||
EH cu |
p) CO |
LO |
to |
O |
0 |
||||||||||||||
-P |
in |
ID |
to |
t ^ |
r^ |
02 |
rH |
rH |
|||||||||||
a co |
.. |
1 |
0 |
+ |
[ |
+ |
+ |
||||||||||||
•H C |
■J 4-3 |
O 4-3 «" |
If) |
in |
|||||||||||||||
o |
PQ |
12 |
> s |
w |
W |
Q |
Q |
n |
|||||||||||
CD |
, |
||||||||||||||||||
i — » |
+ |
+ |
+ |
+ |
+ |
+ |
+ |
+ |
|||||||||||
i—J o |
o |
o |
C |
> o |
o |
o |
O |
O |
0 |
O O |
O rH |
O |
O P- |
||||||
i — o |
o |
o |
c |
o |
o |
o |
O |
O |
0 |
Q C |
** |
* |
•> |
•» |
|||||
ro |
CD |
to |
a |
1 w |
CD |
CD |
•^ |
to |
-sF |
r — 1 |
r-l |
H1 |
rH |
O |
CO |
||||
rH |
H |
SM |
LC |
C3 |
o? |
H^ |
r>- |
CD |
ts |
S |
cd |
oa |
H |
rH |
|||||
fio |
1 ) |
0 |
+ |
I |
+ |
+ |
|||||||||||||
H |
^ O |
a> |
en |
a |
I CD |
Cvi r-l |
rH rH |
to w |
O W |
CD |
H H |
• |
|||||||
1 CD |
i i — «• |
+ |
+ |
+ |
+ |
+ |
+ |
+^ |
! |
1 |
|||||||||
^ rH |
CD |
||||||||||||||||||
i & |
rH |
||||||||||||||||||
cd |
CU |
,Q |
|||||||||||||||||
E-t |
o q CD CD Cu <»-l XS 03 o p:; rH OJ |
cd EH |
O CO |
(0 10 0 |
•H |
•H |
|||||||||||||
cu 05 |
CO |
CO |
i-^ |
Q |
+ |
4-3 |
43 |
||||||||||||
4-' |
— . |
*— >* |
c |
q |
q |
43 |
CD |
CO |
|||||||||||
q cd |
* — » |
CO |
col |
Cu |
^"~N |
co |
CD |
co |
W |
4-3 |
|||||||||
CD d |
co |
■H |
CO |
•H |
q |
Cfi |
q |
CD |
EH |
+3 |
•H |
+ |
+ |
||||||
e cd |
•H |
O |
•H |
oj |
4-3 |
•H |
43 |
. . |
q |
■f |
M |
||||||||
C 4-3 |
O |
fc— o |
v-»» |
x_f O |
- |
t> |
q |
cti |
1 |
4-3 |
43 |
||||||||
O aj |
•H |
4-> |
PQ |
r** |
f^\ |
Px] |
|||||||||||||
q cu |
ffi |
as |
W |
us |
LO |
CD |
CD |
CO |
|||||||||||
•H r-l |
W |
OK |
o |
o UJ |
0 |
1 — 1 |
q |
+ |
+ |
+ |
+ |
||||||||
> o |
I |
1 1 |
i |
1 1 |
1 |
^ |
LO |
lO |
S |
0 |
|||||||||
q ,q |
in |
co in |
CO |
co in |
CO |
I |
1 |
1 |
1 — »o |
O |
O |
0 |
O |
||||||
W o |
o |
oo |
o |
o o |
0 |
Pr-, |
Er- |
a> q |
|||||||||||
4-3 |
CD |
r-i |
|||||||||||||||||
c a |
•H |
O |
|||||||||||||||||
M CC |
>d |
q |
|||||||||||||||||
+3 |
H |
rH |
evj |
- CO |
|||||||||||||||
rO W |
o |
O |
4^ |
rH |
+3 |
||||||||||||||
cd c |
q |
q |
0> |
CD |
CO |
CD |
O |
CO |
|||||||||||
cu o |
cd |
CD |
q |
q |
4) |
q |
q |
cd |
|||||||||||
P o |
i rH |
rH |
43 |
4-3 |
CD ■ |
JD |
rl |
CD |
03 |
ad |
S |
||||||||
CD |
CD |
O |
O |
CO |
CO |
+3 |
43 |
O |
O |
CD |
43 |
43 |
bfi |
||||||
CD '/J |
q |
q |
q |
q |
o |
CD |
aj |
rjj . |
..q |
q |
q |
0J |
CO |
•H |
|||||
o q |
CD |
cu |
c. |
oj |
q |
rH |
uD |
CD |
0 |
CO |
cu |
0 |
cu |
p |
|||||
q -h |
4-3 |
«P |
+3 |
43 |
P |
O |
rH |
H |
CO |
4^> |
q |
+3 |
e |
to |
|||||
cu q |
CO |
CO |
• CO |
CO |
b |
£ |
O |
O |
M |
CO |
hG |
•H |
bD |
1 |
|||||
■P .H |
CD |
o |
o |
CU |
o |
O |
XJ |
^ |
p |
CD |
CO |
•H |
•H |
||||||
CO Cu |
rH |
Sh |
Ih |
rH |
i |
t |
O |
O |
^ |
^s |
q |
l |
4J |
Ph |
|||||
.a p |
O |
p< |
P |
O |
.,-( |
•H |
I |
1 |
to |
co |
p-. |
CO |
Pt3 |
||||||
P rQ |
A |
O |
o |
H A-4 |
£> |
Ph |
t3 |
"1 |
• . |
||||||||||
en o |
o |
o |
P |
o |
M |
H |
"O |
67
to
"tf
W
lO |
<# |
m |
to |
o |
rH |
rH |
rH |
to |
OJ |
+ |
+ |
+ |
i |
+ |
tO CD
00 CD
CO "<^ CO to O) H CVi rH tO CM + + + 1 +
o |
O |
o |
o |
O |
to |
!> |
lO |
ID |
CD |
CVi |
to |
CO |
CD |
O |
tO |
m |
IN |
H |
co |
+ |
4- |
+ |
t |
+ |
+3 |
|||||
c |
|||||
oj |
|||||
p |
|||||
03 |
|||||
fl |
|||||
o |
|||||
o |
CD |
||||
u |
r^ |
||||
o |
•N |
||||
<+H |
p |
m |
to |
||
H |
+2* |
in |
in |
||
o |
P |
p |
10 |
Pi |
Pj |
ro |
S |
53 |
to |
s |
|
0 |
co |
co |
PQ |
CO |
CO |
1>S |
o |
CO |
*=—* |
CJ |
o |
CD |
<«; |
pq |
n |
<$ |
<ti |
o to
+
o o
cx> co
+
IN
+
O
to
rH I
o to
CO
+
M
rH
CD ■H
cd
EH
o
CD £
4-5 CO
CO p
O co
r— i £
S o
I
o
bD |
|||||
!h |
|||||
b0 |
W |
||||
•H |
|||||
P |
4^> |
+ |
|||
•H |
CD |
||||
bD |
bD |
CD |
in |
||
U |
^ |
1 |
+ |
H |
|
W |
W |
>*> |
o • |
||
CO |
-* |
||||
+ |
+ |
+ |
• • (0 |
||
P |
p |
P |
Q |
p |
|
13 |
w |
S |
53 |
||
+ |
+ |
+ |
+ |
+ |
|
o |
O |
o |
o |
o |
|
nd |
|||||
OJ |
|||||
3 |
|||||
C |
|||||
•H |
|||||
4^ |
rH |
||||
c |
O |
||||
O |
G |
H |
rH |
rH |
|
O |
OS |
O |
O |
O |
|
H |
4-3 |
£ |
U |
£ |
|
O |
CO |
CO |
CD |
0) |
|
M |
c |
o |
4^ |
p |
p |
H |
CO |
t£ |
Oj |
Oj |
Oj |
P |
fn |
O |
O |
Q |
|
CD |
CO |
OJ |
■P |
+j |
bfl |
rH |
o |
I |
•H |
■H |
Sh |
P |
fcu |
•H |
Oj |
CO |
oj |
CO |
S* |
p, |
I |
1 |
I |
E-t |
W |
w |
y. |
CO. |
GLi. |
> 03
•H 3
Sh rH
CD CO
O >
O
o
rH
O
o |
o |
o |
to |
o |
CM |
cm |
to |
co |
* |
■k |
|
cm |
to |
rH rH |
1 |
1 |
+ |
CD |
|||||
CD |
P |
CD |
|||
P |
CO |
P |
CD |
||
CO |
CO |
o |
CO |
P |
|
p |
p |
to |
p |
CO |
|
^ £ |
CD |
C |
0j |
p |
|
o a- |
O |
iD |
O |
CD |
|
<H P |
a? |
rO |
CD |
CO |
O |
CO |
^ P |
CO |
|||
nd a |
TJ |
■d |
£ CO |
'd |
|
CD o |
c |
C |
crj O |
C |
■d |
aj o |
co |
CO |
to |
CO |
a |
£> |
rH C |
c0 |
|||
bD |
rH |
rH |
O CD |
rH |
|
to £ |
O |
O |
£ fit |
O |
H |
0) -H |
£ |
£ |
d o |
?H |
o |
CJ P |
Co |
CD |
4--> Jh |
0j |
Sn |
£ co |
p |
P |
CO p |
P |
CD |
CS rH |
co |
CO |
CO' .H |
ro |
P |
P 3 |
CD |
CD |
B CS |
CD |
CO |
oj o |
<H |
1 — i |
CD-r-t |
rH |
O |
P rH |
O |
O |
•H rc |
o |
bD |
3 tf |
.3 |
^1 |
■^ !! |
41 |
^ |
SDO |
O |
CJ |
03 S |
o |
W |
CaIc.[a]D Calc [a]D using; obs. using Calc'd.
-6- Table IV
Obs [o].D CM]d of L*JD of
Compound (CHC13) Sterol Sterol
Stigmastanol acetate +15.4 +13.5 +13.4
tf-sitostanol acetate + 9.0 + 7.8 + 8.2
Ergostanol acetate + 6.8 (av.) + 4.8 + 7.5
aj.-sit06tan.ol acetate +39.4 +15.6
ergostanol
jjl-dinitrobenzoate +13.5 (av. ) + 6.6 + 8.6
Stigmasterol acetate -55.6 -51.3
p-sitosterol acetate -41.0 -40.5 -38.2
Brassicasterol acetate -65.0 -64.7 -60.3
(all tables abridged)
Bibliography
Callow and Younr:, Froc . Roy Soc. . (London]) A, 157,194 (1936) Bernstein and Wallis, J. Am. Chem. Soc, 61,2303 (1939) Gorin, Kauzmann, and Walter, J. Chem. Phys. , 7., 327 (1939) Kauzmann, Walter and Eyrinr, Chem. Pie v. , J^,339 (1940) Bernstein, Kauzmann, and Wallis, J. Orr. Queni. , £..319 (1941 ) Bernstein, Wilson, and Wallis, ibid. , 7,103 T~1942)
G8
Reported by W. H. Kaolan March 25, 1942
REDUCTION OF CARBONYL GROUPS TO METHYLENE GROUPS , — .
The three chief methods of converting carbonyl groups to methylene groups are the Wolff-Kishner reduction, the Clemmensen reduction, and high pressure hydrogenation over copper chromite. Other methods which have occasional value are reduction with sodium and alcohol, zinc and alkali, or hydriodic acid end red phosphorus.
The Wolff-Kishner Reduction (1,2,3,4,5')
The classical method of Wolff is to heat the hydra zone of an alddhyde or ketone with 5-10^ sodium ethoxide at 180-200° in a sealed, tube.
>C=0 + N2H4-H20_» >C=NNH3 N^0Et . )CH3 + N3
180-200a 2 3
The semicarbazone can also be used. In this case, the first step is decomposition to the hydrazone, which then reacts as above If there is insufficient heating, the reaction may stop at the hydrazone stage.
)C=NNHC0NH3 -* >C=NNH2 + C02 + NH3
Often it is not necessary to isolate the hydrazone. The
compound can be heated at 180-200° with sodium ethoxide and a.
slight excess of hydrazine hydrate in absolute alcohol, and the reduction product is obtained, in one step.
Because of the sealed tube feature of the reaction, only small batches of a compound can be reduced, usually about a gram at a time. Bigger batches can be reduced if a bomb is available. Ruzicka recommends the use of sodium benzylate instead of sodium ethylate. Benzyl alcohol has a high enough boiling point (205°) so that the reaction can be run in an open vessel. A commoner means of doing the reaction at ordinary pressure is due to Kish- ner. He found that heating a hydrazone with powdered potassium hydroxide accomplishes the desired result. One refluxes the carbonyl compound with hydrazine hydrate in alcohol, distils off alcohol, water, and excess hydrazine, and adds powdered potassium hydroxide to the residual oil. The mixture is then heated until nitrogen is evolved, usually at 160-180°.
The reaction gives 40-90^ yields with all ordinary types of aldehydes and ketones, including those which also contain hydroxyl groups, carboxyl groups, or double bonds. The product is readily purified. The chief side reaction is azine formation caused by water being present.
>C=NNH3 + HOH -» N2H4 + >C=0 'C=NNH2> >C=N-N=C<'
The Clemmensen Reduction (5,6,7)
In this reaction, the compound to be reduced is treated with zinc amelg&m and hydrochloric acid. The oldest method is merely to reflux the compound six or eight hours. It is often well to
-2-
70
dissolve the compound in toluene, particularly if it is a com- pound melting above the boiling point of the hydrochloric acid solution or if it is very insoluble. This has the advantage that since the compound is in the toluene layer, well removed from the zinc, little if any of the resinous by-product often accompanying the older method forms* Another modification employs alcohol, dioxan, pcetic acid, or other solvent miscible with the hydrochloric acid solution in order to form a homogeneous reaction mixture.
An advantage of the Clemmensen reduction is its extreme simplicity. The amalgam is preprred by shaking the zinc a few minutes with a 5-10# mercuric chloride solution containing a little hydrochloric acid. This polution is poured off, 20-40^ hydrochloric acid and the compound (with or without toluene) are added, and the whole is refluxed. No stirring is necessary because the hydrogen evolved provides good agitation. The amounts of zinc end acid are unimportant as long as there is an excess of each.
The reaction has not been used with much success on aldehydes except for a few phenolic ones. Low molecular weight aliphatic ketones go fairly well, higher ones with difficulty. Many ali- cyclic compounds, including various sterols, have been reduced successfully. Aromatic-aliphatic ketones, including the important 0-aroylpropionic acids, are usually reduced in good yield. Aro- matic ketones are undependable , some reacting smoothly and others not at all.
Most other functional groups are safe from attack. Excep- tions are: the hydroxy groups of a-hydroxy acids and benzyl alcohols; halogens alpha to a carbonyl or carboxyl group; and the double bonds in pyrroles, isoquinolines, and a~P unsaturated ketones or acids.
Catalytic Hydrogenation (8,9,10)
This is a relatively recent method which has not had much practical application as yet, but it seems destined to grow in importance as time goes on. Its use is limited to carbonyl groups directly attached to an aromatic ring. The compound is hydrogenated at 1 50-800° under a pressure of 2000-3000 pounds over a copper chromite catalyst. Yields are usually 80^ or better and en easily purified product results. Furan nuclei are completely reduced. Pyrrole and benjjene rings are not ordinarily affected.
Fieser found that keto acids were best reduced as an aqueous solution of the sodium salt. In other cases alcohol or no sol- vent at all was used.
Low pressure hydrogenation over a platinum catalyst sometimes reduces a ketone to a hydrocarbon in low yield, the main product being a carbinol.
71
-3-
Soci i urn end Alcohol (11,12)
This is an excellent method for reducing diary 1 ketones to the corresponding diary 1 methanes. Other ketones usually stop at the carbinol stage, as, indeed, do a few diaryl ketones. The sodium reduces even hindered ketones not susceptible to the Clem- mensen or Wolf f -Kishner methods. The yields ere usually good, but are sometimes low due to cleavage of the ketone.
One merely drops chunks of sodium into a boiling ethyl or e.myl alcohol solution of the ketone. The hot reaction mixture is poured into water and the product is filtered or extracted with ether.
Zinc and_ Alkali ( 6 )
This method often works veil on keto acids and enthrones. The zinc is activated by shaking it with a copper sulfate solu- tion. When the blue color disappears, this solution is poured off and a sodium hydroxide or ammonium hydroxide solution of the compound is then refluxed 12 to 48 hours. High melting com- pounds insoluble in alkali are dissolved in toluene.
Hydriodic Acid end Red. Phosphorus
Sometimes ketones which are hard to reduce to the methane are readily converted to the carbinol. The carbinol can then be reduced by means of hydriodic acid and red phosphorus. The carbinol may be changed to the halide first.
A method occasionally used as a last resort is to convert the carbonyl group to a 1,1 dichloride with phosphorus pentachloride and then reduce the product with the above reagent. An even more desperate measure involves the ketone, hydriodic acid, red phos- phorus, and a sealed tube at 200°.
Bibliography:
1. Suter and Weston, J. Am. Chem. Soc, 61, 232 (1939).
2. Dutcher and Wintersteiner, ibid. , 61, 1992 (1939).
3. Ruzicke and Goldberg, Helv. Chim. Acta, 18, 668 (1935).
4. Schmidt, Hopp, and Schoeller, Ber., 72B, 1893 (1939).
5. Fieser, "Experiments in Organic Chemistry," D. C. Heath and Co., New York City ( 1941), "Vol. II, p 420.
6. Martin, J. Am. Chem. Soc, 58, 1438 (1936).
7. Mike ska, Smith, end Lieber, J. Org. Chem., 2, 499 (1938).
8. Adkins, "Reactions of Hydrogen," University of Wisconsin Press, Madison, Wisconsin (1937), pp. 51, 69, 129.
9. Fieser and Heymann, J. Am. Chem. Soc, 63, 2333 (1941 ).
10. Packendorff, Ber., 67, 905 (1934).
11. Klages and Allendorff, ibid. . 31, 993 (1898).
12. Asahina, ibid.., 69B, 1643 (19367.
Reported by B. C. McKusick March 25, 1942
•••
THE REPLACEMENT OF ALKYL GROUPS DURING NITRATION
Y*it
Halogens, methoxyl, carboxyl, sulfonic acid, and alkyl groups are all known to migrate or to be replaced by the nitro group during nitration in certain cases. The removal of an
replacements have been observed.
Although the methyl groups of toluene and of the xylenes have been replaced by the nitro group in electrolytic nitrations, such replacement is not the rule when more usual methods are used.
The trimethylbenzenes have never been known to lose a methyl group. If 5- or 6-bromopseudocumene is added to fuming nitric acid and concentrated sulfuric acid and heated, a reaction proceeds as follows:
H.
Br
CH-
CH3
OsNy/NrCH.
CH3
Br
11no;
0SN<//A\CH3ON03
+
Br
CH.
NO;
The 3-bromopseudocumene does not give this result.
Continuing on to the tetramethylbenzenes, durene is not demethylated on nitration, but a nitric acid ester is obtained. If bromodurene is nitrated with fuming nitric acid or a mixture of nitric and sulfuric acids, not only cm the nitric acid ester be formed, but on standing at room temperature in the presence of the acid, ?~bromo~5, 6-dinitropsuedocumene is formed.
H,C
H,C
Br
U
CH.
CH.
HNO.
Br
h3cY
A
0°
H*C
NO.
Br
Uroom temp.v |
CH30N03 h3S04 H,C
v
CH3 N03
NO
3
In connection with the effect of halide substitution on the ease of replacement of alkyl groups, Qvist has found that the more chlorine substituted on the ring of p-cymene, the easier the replacement of the isopropyl group by tha nitro group .
Pentamethylbenzene and pentamethylbromobenzene act in an an analogous manner. The latter can have two methyl groups
■
. !
-2~
73
replaced by nitration with sulfuric and fuming nitric acids in chloroform at 0-5°.
In those cases where dinitro compounds were produced by the elimination of one or two methyl groups, the ortho compound is formed to the exclusion of the para . In the case of hexa- and pentaethylbenzene , the para dinitro compound is formed to the exclusion of the ortho . The best method of preparing the p_-dinitro, p_-diamino, or p_-quinone of tetraethylbenzene is through nitration of pentaethylbenzene rather than through nitration of 1,?,4, 5-~tetraethylbenzene.
The elimination of the isopropyl group was mentioned above. The latest reaction reported is the f ollowlng :
HN0,-H,S04v
CH,
NO.
82$ +
CH(CH3)
3 /2
CK(CH3)S
CH3-CH0H-CH; CH,-C0-CH,
Barbier has done work on the elimination of butyl groups. He found that only groups meta to the first entering nitro group, were replaced. This is well shown by two examples from his work.
HaC
\y
H9(iso ) C0CH3
£R3 92# HNO.
0°
C4H9(isp_)
H3cr Noch,
CH3C0»
u
C4Hg
OaN/^SCOCH.
H,C
N03
CH.
A°8
H3c/^fOCH.
\ywOg
CH^CO
Cihg
HaC
N02
N0S CH3
C,H9
H3CY >0CH3
\/Jno3
0aN
Amyl and heptyl substituent groups have also been observed to be affected by nitration of the benzene ring. In the latter compound, the heptyl group was eliminated but a nitro group did not take its place
(CoH5 )aG
HNO 3 HaS04
0aN
"
.
-3-
MECHANISMS
To date five mechanisms have been advanced to account for different instances of this type of replacement.
The first mechanism suggested was a Jacob-sen Rearrangement mechanism.
/4L
H3C H3C
CH
CH-
HMO 3 -Eg SO/
CK3 0° 30 mm CH CHC13
+
CH3
•*>
Ch CH3
CH.
3 ^8
— >
03N 03N
ch3
CH3 *CH3
The yield in this particular nitration is 70$ of the theoretical based on p en tame thy 1 benzene. Such a yield would require that more than the prehnitene itself be nitrated. Therefore, a mechanism would still be required to explain the conversion of hexamethyl benzene to dinitroprehnitene. Further- more, the low temperature and short time required for the reaction would permit only a negligible Jacobsen Rearrangement to take place.
Another mechanism suggested is the one by Alfthan proposing that the alkyl group is first oxidized to a carboxyl and that this carboxyl is then replaced by the nitro group.
A X3
CH(CH3)3
V
COCH.
CH.
COOK
•0
N0:
co;
Alfthan was unable, however, to nitrate methyl~p_~tolyl ketone to the nitrotoluene. It has been shown that in the case of polymethyl carboxylic acids, the carboxyl group can be almost quantitatively replaced. The objection to this mechanism lies in the fact that nitro replacement of a methyl group takes place in polymethyl compounds even under conditions which do not favor oxidation of the methyl group.
Barbier has advanced a mechanism which is in agreement with results of a large number of experiments. From his own work and from work on the elimination of halogens, carboxyl, alkoxy, and alkyl groups during nitration, Barbier suggested that elimination mpj take place when the alkyl group is met a to a nitro group which has entered the ring. This is in agreement with such results es the following.
CH3
I
OH
CH(CH3)
OpN
3 J3
CH(CH3)
75
3 /S
r
*k
OH(CHa)
3 / 3
OpM
£H3
NO 3 CH(CH3)3
+
CH3 rSNOa
NO.
0H3 OgNf V]N03
OH CH(CH3)3
0.
OaN
N03
CH*
CH3 N0S
In certain cases, however, this mechanism fails to account for the products of a reaction, e.g,, the nitration of p-cymene to give p_-nitrotoluene and the nitration of chloro-substituted p_-cymenes which have no free positions meta to the isopropyl group. In the case of polymethyl and of polyethylbenzenes, the mechanism does not explain the facts at all. Pentamethyl- benzene gives only tne ortho dinitro compound, and pentaethyl benzene gives only tne para dinitro compound.
From their work with polymethyl and L. I. Smith and coworkers have also adva propose that the first step is the nitra if a position is open. After this nitra if no position was open), another nitric to form a benzoyl alcohol ester of nitri then decomposes to give the nitrobenzene reaction are illustrated in the nltratio above., This particular nitration consti that the reaction c°n take place in tne Another similar nitration in which the i were isolated is the nitration of pentae
polyethylbenzenes, need a mechanism. They tion of the benzene ring tion (or the first step
acid molecule reacts c acid „ This ester B The step 3 of the n of bromodurene shown tutes one of the proofs steps postulated^ ntermediate products thylbromobenzene .
-5~
^,
76
Br H3Cp |CH
H3C\XCH3
CH3
3 fuming HN03 vCHa CKC1
alcohols obtained + by hydrolysis of these eaters
inseparable
nitro
compounds
Sn + KCi' GK
CH3 identified
This mechanism obviously does not explain such orientation of reaction as shown in connection with Barbier's work.
The latest mechanism proposed suggests that the replacement occurs as a hydrolysis of the a.lkyl group from the benzene ring. The proponents of this mechanism state as their only support that p_"Cymene on treatment with nitric and sulfuric acids at zero degrees gives about Q% of p_-nitrotoluene. Isopropyl alcohol can be recovered as well as acetone.
None of the above mechanisms by itself can account for all observed results. One mechanism, however, does not necessarily exclude all others and. it is possible that a more complete mechanism may be formulated by incorporating elements from the different mechanism so far proposed.
Bibliography
Alfthan, Ber . , 55, 78 (1920)
Barbier, Helv. Chim. Acta, 11, 152, 157 (1928)
Doumani and Kobe, Ind. Eng. Chem., 31, 257 (1939); J. Org. Chem.,
7, 1 (1942) Gottschalk, Ber., 20, 3286 (1887) Huender, Rec . trav. chim., 3_4, 9 (1915)
Ovist, Acta Acad Aboensis Math, et Phys. 6, sec 11 (1932) Rinkee, Rec. trav. chim., 57, 1405 (1939), 58, 218, 533 (1939) Smith, et al . , J. Am. Chem. Soc, 57, 1289 (1935); 59, 1082 (l937)j
57, 1293 (1935); 62, 1349 (19407; 62, 2635 (l9407.
Reported by R. April 1, 1942
I. Meltzer
■ -
■
KETENE ACETALS
McElvain et al, University of Wisconsin
The most satisfactory method for preparing ketene diethylacetal is:
, CH33rCH(0Et)3 + KOC(Ch3 )3 ..^^ L±9-°*)°l) CK3=C(0Et )3 + HBr
This ketene adds H30, EtOH, RCOOh, HX, CsHsOH, CH3COGH3COsEt , CH3(C03Et)3) NH3; C6H5NH3, CBH10NH, and H3 across the double bond. In the presence of Cct013 aimers, trimers, and polymers have been produced and a head- to- tail structure has been indi- cated. These properties were discussed in this seminar last year. This report is concerned with the addition of active halogen compounds, acids, and a, p-unsaturoted carbonyls.
Many of the additive reactions can be interpreted on the basis of the polarization that is characteristic of the hetero- enoid structure of ketene diethylacetal:
^ f°
CH2kCV,
X
xEt
1« Addition of active bromides. --The reaction with active bromides can be summarized as follows:
R = allyl or benzyl
CH3=C(OEt)3 + RBr -* [RCH3C(OEt )3Br] ->RCH3C03ET + EtBr
RCH=C(OEt)3 + HBr (not isolated)
\RBr
[CH3CBr(OEt)3] [R3CHC3r (OEt )3] CH3Cv53Et + HBr R3CHC03Et + EtBr
RCH=C(OEt)3 + EtOH -> RCH3C(OEt)3
Origin of the alcohol used in the last reaction is poly- merization of some of the ketene.
2. Addition of CHaCOCl. — The products isolated indicate the following course of the reaction:
CH3=C(OEt)3 * CH3C0C1 -^-4 CH3COCH3C03Et + EtCl
(b)//CH3COCl
CH3C(OCOCH3)=CHC03Et + HCl
-
78
-2-
The hydrochloric acid formed by reaction (b) appears to react in two ways, (c) end (d)n Notice- that reaction (d) involves s 1,4 addition ecronc two molecules of the acetal. This unusual reaction is substantiated by oxher additions.
(c )
CH3=C(OEt)3 + hCl -^-MGH3C03St + EtCl
CH3 (EtO)3C **KC1 (a) v ,
CH3 it
^^C(0Et)3
CH3C(OEt)=CHC03Et + EtCl + EtOH
The validity of reaction (b) is indicated by the fact that acetoacetic ester is acetylated by acetyl chloride with either pyridine or ketene diethylacstal as reagents for the removal of KC1.
3. Addition of Acids. — 1,4 addition of acids across two molecules is further indicated by similar addition to pure ketene diethylacetal by nine acids as, for example, hydrochloric, hydrobrornic , trichloroacetic, formic, and acetic in yields of 10-38^, A minimum of acidity seems to be required for this
type of addition since phenol and p_-bromophenol added only as in (c) above but tribromophenol gave a 26f yield of the 1,4 product.
4, Addition to a, ^-unsaturated carbonyl compounds. — The ad.dition of maleic anhydride can also be considered as a 1,4 intermolecular addition in which the establishment of the 2,3 double bond of the ordinary type is replaced by a single bond that joins the two molecules of the ketene involved. The work with maleic anhydride is summarized on the flow sheet at the end of the abstract. Because the addition product, II, is insoluble in ether, only one mole of the anhydride adds in tnis solvent. In benzene II will add a second mole of maleic anhy- dride to give III. Chief support for the structure of II is its dehydrogena tion to the known compound, 3, 5-diethoxyphthalic anhydride, IV, and hydrolysis of IV to the corresponding phthalic acid. Heating either the anhydride or the acid with aniline gave the corresponding N-phenylphthalimide . Since reaction (b) involves the loss of an a-hydrogen, possible addi- tion to dimethyl maleic anhydride was studied. No addition product was obtained.
Addition to quinone and proof of structure was as follows:
'**''•'
79
•3-
( + 2CHs=C(0Et); >
pOH3 + CK3C(OEt)3
WCOSt
Colorless
H20 fa
rf
Ester ^MhL. ||
GKpCOpH A
0 K
It is interesting to notice that I, a colorless compound,
can be considered as a vinylog of an ester while quinone,
from which it is derived, is vinylogous to the colored 1, 2-diketones.
The addition to dibenzalacetophenone and proof of struc- ture was as follows:
C6H5CH— €H~r-C0CK=CEC6H5
II \
CH-a-C(OEt)a \
\HC1 Br2 \Ha0
C6H5CH-CHBrCOCH=CHC6H5 C6H5CH-CH3COCH=CHC6H5 CH3COsEt + EtBr CHsC03K + 2EtOH
Benzalacetophenone added in a similar manner.
Bibliography:
McElvain and Kundiger, J. Am. Chem. Soc, 64, 254 (1942). McElvain and Cohen, J. Am. Chem. Soc . , 64, 260 (1942).
Reported by G-. L- Schertz April 1, 1942
oO
OH,
(EtO)aC
//
-4-
FLOW SHEET
CH— CO
|| p Ether Solution ,
CH-C6 (a)
Benzene
maleic ,,pnnydride Eenzene
H3
EtOC NCHCO=,H II i
III 60^
H\yCH2
35(
I
H3
A
EtOC7 XHC03H
II I
HC XJHCOgK
NT
il
0
(Et0)3C XCH~CO T) HaG ^CH-Ctf XCtOEt)2 I
(tfl
EtOC xCH~Q0
|i j ^0 + 2EtOH
C Jl—CV
xctm \
715? \
II
HpO
/CH2 EtOC' ^CKCOoH
II I
HC ^C-C03H
OEt
CH3=C(0Et)2
CH3C(0Et)3
/
Et0/^\\C0
0 /Et IV
HpO
36H5NH3 Imide
Acid
'
81
BIPHENYLENE
Investigation of the interesting hydrocarbon biphenylene (I) has been carried out because it might contribute to the theory of aromatic chemistry as a whole, supplying in its central ring a possible cyclobutadiene system and at the same time affording an extreme case of the Mills-Nixon effect.
JK.
V^
(I)
The history of its preparation is a series of repeated failures and of only one isolated e nd iireproducible success. Hosaeus, in. 189^, carried out a Wurtz reaction with o_-dibromo- benzene, obtaining biphenyl as the. only hydrocarbone. Niementowski , in 1901, on treating the diazonium salt of 2,2'- diaminobiphenyl with copper powder obtained carbazole. An attempted dehydration of 2~hydroxybiphenyl by Cullinane, Morgan, and Plummer also failed.
Dobbie, Fox, and Gauge in 1911 rep of the hydrocarbon in 1007 yield by the freshly cut sodium on 2,2' -dlbromobiphe
based their proof of
method of synthesis,
and on its analysis.
Indicated a. strained
the ring to give the
derivative of the hydrocarbon. The
the hydrocarbon gave dibenzofuran,
structure of the c on its oxidation i Reactions carried central ring. Tre starting material
ac di
derivative, Attempts by other investig of Dobbie, Fox, and Gauge failed oomple of a compound containing a cyclobutadie able.
orted the preparation
prolonged action of nyl in dry ether. They ompound C13HQ on the n part to phthalic acid,
out on the compound atrnent with bromine opened as well as a d.ibromo tion of nitric acid on nitro, and a tetranitro a. tors to repea.t the work tely so that the existence ne ring became question-
Lothrop has recently investigated biphenylene in connection with his work on the Mills-Nixon effect in aromatic compounds. His attempts to eliminate the bromine atoms from 2,2 '-dibromo- biphenyl by the action of hydrogen, lithium, sodium, and potassium were all successful, but the product in each case was biphenyl and a bromine containing oil. Magnesium reacted with only one of the bromine atoms and calcium, zinc, and pure copper had no effect. Cuprous oxide, however, g:eve a low yield of a new hydrocarbon which was volatile in steam and formed a scarlet picrrte. Pyrolysls of biphenylene iodonium iodide with cuprous oxide gave a larger yi<=ld of the hydrocarbon with small amounts of 2, 2'-diiodobiphenyl .
ex.
NH3 HaNV/
Picrate"
M,P. 121~122°C.
M.P. 109~110°C. (I)
Molecular weight determine tione in camphor and benzene agreed with the formula for biphenylene rather than for dimoleculrr coupling possibilities. Analyses of the hydro- carbon end its picrate agreed with the formula C13H8. Oxidation with chromic oxide ^pve phthalic acid, thus establishing the presence of ortho substitution. Reduction with hydrogen over red hot copper gave biphenyl.
CrO
3,--
(I)
^
COOH
OCH
2 . 7 -Dim ethyl biphenylene (II). — The 2 , 7 ~di me thy 1 derivative of biphenylene wee prepared in the same manner In order to see whether the dehalogena tion reaction could be considered general. This compound was prepared by two different routes, thus veri- fying the structural formula of biphenylene. This compound gave a deep crimson picrate.
. 1
-3-
83
H3C
H3C
Piers te M.P. 110°C.
f—
3C^\NH8 H3C NO*
<f-
\y
K>
N02 OaN
J\
\'/
H3C^V/>NH3 H2N^^CH
HgC^X
H*C
M.P. 112-113 C. (II)
/^SGK
NOo 0aN.
/y
— >
VfCH.
/^CH;
The formulas written for these hydrocarbons pre supported by the complete agreement found between prediction and experi- ment, and suggest that the cyclobutadiena ring may not be too strained to exist if fused with two benzene rings (A), or that the Mills-Nixon effect may operate to stabilize the molecule in a less strained cyclobutane form (B), or that the coplaner molecule has considerable resonance energy.
/V_A
Ss
%S — \f>
(A)
(B)
..
/•
-■'.
-4-
81
The fused system of three rings would seem to be confirmed by the formation of such stable and highly colored picrates.
2 , 3 ■ 2 ' . 3 ■ •■ Bin/mhthy len s ( III ) . ---The only other compound reported in the literature having a structure similar to biphenylene is 2, 3,2 ' , 3' -binaphthylene, prepared bj Rosenhauer, Bre.un, Pummerer, and Riegelbauer in connection with polymerize bi on studies. 1,4-Nephthoquinone was condensed to form 2, 3, 2* ; 3 ! --binaphthylene- 1, 4,1 ' , 4 ' -diquinone which, on distillation with zinc dust, gave 2*5,2' i 3 l -binaphthylene in 15.5$ yield. This compound formed a n un s t a bl e r o d p i c r a t e .
CflHsNO.
Py. + HOAc
Distil Zinc dust
v
Picrate M.P, > 260°C
Bibliography :
,/vv
W
M.P. 365°C (III)
Hosaeus, Monatsh,, 14, 323 (1B93).
Niementowski, Ber . , 34, 3325 (1901).
Dobbie, Fox, and Gauge, J. Chem. Soc . , 99, 693, 1615 (1911);
105, 36 (1913). Mills and Nixon, J. Chem. Soc, 2510 (1930). Schwechten, Ber., 65, 1605 (1932). Mascarelli and Getti, Gazz. chim. ital,, 63, 654, Cullinane, Morgan.., and Plummer, Rec . trav, chim., Rosenhauer, Braun, Pummerer, and Riegelbauer, Ber Lothrop, J. Am. Chem. Soc, 63, 1187 '( 1941). Hammett, "Physical Organic Chemistry", McGraw-Hill Book Co., Inc., New York, N. Y., 1940, p. 19
661 (1933). 56, 627 (1937). , 70, 2281 (1937).
Reported by F. W. Spangler April 8, 1942
85
THE CONSTITUTION OF USNIC ACID
Usnic acid is a very widely distributed naturally occurring compound which has been found in more than ?0 lichens. Usnic acid was isolated first by Rochleder and Heldt in 1843 and has been the subject of much investigation since that time. Widman in 1902 first advanced a formula for usnic acid (I), based on the observations of several earlier workers and on the re- sults of his own degradation experiments. In 1927 Shopf, Heuck and Kraus turned up several new experimental facts which made it obvious that Widman :s interpretation of his data was erron- eous.
CO — 0
C03H CH3COC=C-C=CCH--C8H11
0 C
■CO
C03H CH3COCH=C~C=CCK-CeH 1 %
0 — to
I II
Kraus demonstrated by a Zerewitinoff determination that usnic acid had three active hydrogens. Formula I would indicate only two if we assume complete enolization of the methyl Ketone group. Moreover, one of the characteristic degradation products, decarbousnic acid, represented by Widman as II, was demonstrated by Shopf and Heuck to be an acidic enol, not a carboxylic acid.
The structure of usnic acid is most readily elucidated from its degradation products which are quite numerous. The more important are set forth in Chart I for future reference.
The decomposition through usnetic acid to usnetol has been of greatest aid in elucidating the nuclear structure of usnic acid and its derivatives. The structure of usnetol follows from the following degradations and synthesis.
Ci3Hi404 O3 usnetol
I
~CH3C03H H30 (B)
A?
->
>yCnHiaOB + CH3C02H Alcoholic (Ident. as
^HCl silver salt)
%0 1 C13H1406
4 N'-NaOH
y\°H
CijH.,0, — >di^cetate Uen&dl
0^ JLsOs
^NaNOa ^
OpN
f\
CH3^H3S04 *(h)
N03
CHa
NO-
v I
■
[ > ;
■
■ ■
.
- ■
.
■ '■
86
Eh
< XX O
T3 |
|||||
•H |
O |
||||
O |
rH |
CD |
|||
CCS |
1 |
1 |
|||
CD |
O) |
<* |
CD |
||
O |
o |
o |
« |
O rH |
i> |
■4 |
•H |
OJ |
o |
> -* O |
rH |
H |
P |
o |
H +3 |
||
X |
CD |
t |
I |
JH CD |
• |
•4 |
c |
Ph |
m c: |
Ph |
|
H |
CO |
• |
H CO |
||
o |
2 |
JS |
O. 3 |
£5 |
/IS
o
O
CM
+
ffl |
ON |
||
o o |
\ |
||
to |
m oa |
ffi |
|
w |
H |
(9 |
|
O |
O Ov |
||
o |
o w |
||
o « |
ntrj |
||
to O |
w + |
||
w o |
o |
||
O 1 I |
1 |
-d
^ O
O co
CD m o
co a
o
^ -sh 0.? I I • tO W OJ Oi O CO rH CV2 <# + 1
Ph Ph II
• a
r<-i »2h I — I I 1
• • «
O -Pi— J CO o SU CO'
<d |
o |
|||
•H |
O |
|||
O |
O |
|||
CO O |
1 CD |
-V |
o |
|
lO-H |
a |
to |
CO |
|
O £ |
rH |
ro |
o |
l> |
W CO |
o |
CO rH |
rH |
|
H 2 |
o |
... •-• O |
||
ffl o |
• — |
_ j _ |
~*XX CD |
• |
ro *h |
Ph |
■-« a |
PL, |
|
* t>. |
« |
H CO |
• |
|
o p |
S |
O 3 |
is* |
w O
o
I
p Pi o
o en
o c o o
o m to
+ CD
<d |
|||
•H |
iO CO |
||
O |
• 1 o |
||
Co |
to to O |
||
^ |
r-t O ^ |
||
o |
O |
Cv> C\2 CV |
|
ID |
•H |
||
rH |
rH |
||
CO |
O |
f |
|
£ |
Ph |
||
rH |
C/J |
• |
|
o |
^ |
||
CD |
|||
1 |
c |
||
0 O |
<3 |
•H 3 rH |
|
o 1 |
O O |
||
N |
• |
•H |
|
3 |
|||
<y |
o |
|||
•H |
|||
G |
o |
||
CO |
iC |
||
CD 3 |
£N |
||
O O |
rH |
||
■n Q |
|||
h *h |
_ o |
||
iX co |
Tj |
• |
« |
^ o |
•H |
Ph |
ffl |
H CO |
O |
• |
t |
o -d |
CO |
s |
o |
P |
|
a |
S |
|
CO |
o |
o |
in 3 |
co |
o |
o O |
o |
CD |
C0rQ |
Cvi |
•d |
* fc |
||
K co |
• |
|
t^ o |
Ph |
|
»h a> |
« |
|
P ^ |
•d |
|
•H |
|
CO o |
|
O co |
o |
CD |
o |
■"• O |
CO. |
tX -H |
rH |
CO rj |
|
■"• O |
• |
o c |
Ph |
CO |
• |
2 |
*>' |
•
• ; :■
■1
-2-
87'
The isolation of the product C13H140€ after decomposition of the ozonide indicates a double bond in a ring system since no carbon atoms were lost. The ease with which this compound is hydrolyzed by alcoholic hydrochloric acid indicates an oxygen containing five or six memebered ring with one double bond. The alternative route of degradation (B) indicates one acetyl group, not present as an acetate. Structures III, IV or V would satisfy all of the above facts. The isolation of acetic acid and the compound G11¥{130B by acid hydrolysis of the ozonolysis product rules out an isomeric structure with an ethyl group and a hydro- gen at C-2 and C-3 instead of two methyls as shown*
CHCH
COCH.
V
The correct structure (ill) was established by the follow- ing synthesis of Curd and Robertson after they had established the structure of usneol as being III minus the acetyl grouo at C-7.
CH.
CH,0^
V
CH3
OH CHsO-^Vq NCCH3
OH
V
OH
COCH.
CI9HCH3 HO OCCH3
KpCO, CH3
COCH3
CH3I
CH.
.3CO3
CH3 *CH3
dimethyl ether of III (no depression on mixed
M.P.)
The establishment of the orientation of these substituted phloroglucinols is due to work by Robertson et si.
Usnetic acid must then have the same nuclear structure as usnetol with the addition of C03 to make the carboxyl group shown to be prepent. Because of the synthetic difficulties, usnetic c:ciu itself was not synthesized but the structure of pyrousruc «cid dimethyl ether (VI} arising from usnetic acid by loss of the nuclear acetyl group at C-7 has been established as follows.
8S
ch3o
CH3
r*\
CH3 CO
0
M^Ls^
V
0CH3
CH.O
CH.
V
-3-
OH c^30^^oCH3C03Et
COl3!!;"~>CH3iv
v
COCH.
OGB
OCH.
CHaO^y^NCHO CH3^ '
oc:
CH30
^v~^
^K/-
CH30f*
0
CH.
Gatterman
>
^
CH3
^CHO
CH*
OCH;
OCH,
hippuricy acid
• - 0
CH.
OCH.
10^ ^H oAA
NaOH h3°
■»
CH.
V
CH3COC03Et *CH3
CHaOr^ This structure for usnetic acid (VII) has been confirmed by Asahina by the CH3vv
following degradation.
CH2C02H
KOH y mzo<
CH3 Ha0
3^3
HOpC
CH3C03H *CH3
CH3N3 CBHxlONO NaOEt
-4-
89
\r
C03CH3
CHg
JkO^
acet
HOgC^
0
!i I'
anhyd, KOECj liCH.
.0
/ \C02H
3J
CH3
The structure of decarbousnic acid, has not been established by synthesis but there can be little doubt as to its structure . The following most pertinent facte must be considered in formu- la ting a structure for decarbousnic acid.
(1)
(2) (3)
(4)
(5)
Ketonic reagents indicate a 1,3 diketone structure.
a. Hydro xylamlne gives an ieoxazole derivative.
b. Phenylhydrazine gives a pyrazole derivative. c „ Semicarbazide gives ? pyrazole derivative. Mono--d.i-tri-tetraacetyl derivatives are known, Decarbousnic acid is optically inactive. Decarbousnic acid is a dibasic acid containing no GC2K group.
Ozonolysis of dlacetyl decarbousnic acid shows that decarbousnic 'acid has the characteristic substituted benzofurane nucleus.
^nd
The only formula which seems to fit all mass of minor experimental facts is VI
these criterea Experimental
evidence for- the side chain structure will be given with the ozonolysis of diacetyl usnic acid.
in connection
COCH
•Ha0
cone .
K3S04 60°
! I I
3 X/ CK3 SCX
OH
CK.
TV 1ft
The cyclization of decarbousnic acid VIII to decarbousnol IX (postulated) was without precedent until Foster and Robertson (and Kealy) showed that this type of reaction did occur and that the compounds so obtained were entirely analogous to decarbousnol and usnolic acid, of similrr structure..
The structure of usnic acid (X) itself is fairly well established without presupposing the structure of decarbousnic acid and the evidence thus obtained supports the proposed, formula for decarbousnic acid. Shopf and Ross performed the following degradation.
-5-
90
COCH
X
9\
CO
l , .CHGOCHg
Oa on
diacetpte
EtOH
->
CH3C02
CH.
COCHL
A/-
OCOEt \o 4. P°
„C.H3 CH3COCH3
+
co3
Y
H
XI
XII
The isolation of the lpctone XI pnd of acetoneoxalic eeter XII clearly indicates the presence of the double bond as shown and consequently of the benzodihydrof uren nucleus, end the 1,3 diketpnic nature of the side chain. The fact that all the degradation products of usnic acid have been demonstrated to be substituted ben^ofurans does not constitute a serious draw- back since the reversion to the more stable conjugated form is to be logically expected when possible.
v\A
*p°
h
XIII
The nature of the degradation products indicates the partial" structure XIII of usnic acid which must be considered as 3CHCOC_Ha derived from XIII by the replacement of
2H by the -CO- group. One of these H atoms must be from C6 in XIII in order to stabil- ize the dihydrocoumarin ring system. There are then four possibilities left; a) By enolization and removal of the H at C2 a sixmembered cyclic lactone results, process at C4 an 8 membered cyclic lactone removal of an H at C3 a 6 membered cyclic trike- tone {X) results, d) by the same process at C5 an 8 membered cyclic triketone results. Both lactone formulas are readily ruled out since the 1,3 diketonic system characteristic of usnic acid as well as decarbousnic acid would not be present. Of the remaining two the choice of the 6 membered triketone is pre- ferred on two grounds, first the more ready occurence of 6 membered rings, but this choice is even more strongly indicated since 1,3, diketones in a ring structure do not show the char- acteristic behaviour of straight chain 1,3 diketones with ketonic reagents (i.e. dihydroresorcinol ).
b) by the same results, c) by (X)
-6-
9i
The structure advanced has some de necessary to assume that the linkage be is rather weak since in all the degrada linkage is broken. This does not seem assumption since the linkage under ques hindered quaternary C atom which is und of several active ting groups-. The only out explanation is the easy recemizeabi again may be considered as a. property o Cs-Cilj. linkage but no adequate analogy
batable features. It is tween G3 and Cu in X tions of usnic acid this to be an unjustifiable tion is to a highly er the combined influence
serious objection with- lity of usnic acid.. This f the weakness of the is known.
Two
other compounds in Chart I which have had only "nuisance value" in the elucidation of the structure of acid are usnolic acid XIV and usnonic acid XV a or b postulated only ) .
a
usnic ( structures
COCH
COGH
C-COCH,
COCH3
CHCOCH.
0 C
CO
.CH~ CH^CT COCH
6
XIV
XVa
XVb
Bibliography
Recent literature only. For earlier references see Schopf and Heuck, Ann. 459, 233 (1927)
Schopf, Clemens Institut fur organische Chemie der Techn. Hochschule in Darmstadt
Schopf and Heuck (A. Kraus), Ann., 459, 233 (1927) Schopf and Ross, Natur., 26, 772 (1938) Schopf and Ross, Ann., 546, 1 (1940-41)
Robertson, U. of Liverpool
Curd and Robertson, J. Chem. Soc . , ,1953, 437, 714,1173
Birch, Flynn and Robertson, ibid. , 1936, 1834
Curd and Robertson, ibid;, ,1937, 894
Birch and Robertson, ibid. , 1958, 306
Foster and Robertson (Healy) ibid. . 1959, 1594
Asahina, Yasuhika; Pharmaceutical Institute, U. of fokio
Asahina, Miyasaka and Sekisawa, Ber., 693, 1643 (1936)
Asahina and Yanagita, ibid. , 70, 66, 1500 (lgs17)
Asahina, Yanagita, and Mayede, ibid. , 70, 2462 (1937)
Asahina and Yanagita, ibid, , 71. 2260 71933.-: 72, 1140 (1959)
Yanagita, ibid., 71, 2269 (1933)
Proc. Imp. Acad. TTokio ) 13, 270 (1937); 15, 311 (1939)
Perkin and Everest, Natural Organic Colouring Matters, (1938) p. 530 Reported by C. W. Theobald, April 8, 1942
•
92
ALKYL CARBONATES IN SYNTHETIC CHEMISTRY Wallingford, Homeyer, and Jones — Mallinckrodt Chemical Works
Until very recently the alkyl carbonates have found little application in synthetic organic chemistry. Within the last year a series of investigations have been carried out which have greatly extended the uses of this class of compounds and have made easily available certain types of compounds which before were difficult if not impossible to prepare,
!• Synthesis of Malonic Esters.
The first attempt to condense alkyl carbonates with esters was made by Wielicenus in 1887; he was unsuccessful, and con- cluded that alkyl carbonates were not suitable for Claisen type condensations. However, since then, several investigators (Lux, Nelson and Cretcher, Skinner) have succeeded in carrying out such condensations by use of sodium or potassium in benzene or ether solution, producing malonic esters in low yield. Wallingford and his associates point out that the difficulties encountered were due to the following factors: (l) The ester may condense with itself, (2) The metallic sodium and potassium used readily decompose alkyl carbonates, (?) A metal elcoholate in alcohol solution may not be used as a condensing agent because it ha s been shown by Cope and McElvein and by Connor that sub- stituted malonic esters undergo alcoholysis according to the equation
RCH(COOEt)2 + EtOH NR^Bt RCH2COO£t + (EtO)2CO
The Mallinckrodt workers have demonstrated that this reaction may be reversed. By applying the principles of mass action, they have developed a process which is quite general, and have pre- pared a large number of malonic esters by condensing alkyl car- bonates with a variety of esters. The general reaction is
R'CH2COOR + (RO)3CO + MOR -> [Rf C (COOR)3]M + 2R0H
The essentials of the new procedure are (l) the use of a large excess of alkyl carbonate as a reaction medium or solvent and (2) removal of alcohol from the reaction mixture by distillation. The validity of the applied principles of moss action is shown by the fact tha t an 86% yield of diethyl phenylaceta te was obtained from ethyl phenylacetate and diethyl carbonate whereas Nelson and Cretcher have shown that these two will not condense at all in an alcoholic solution of NaOEt.
It was noted in the carbalkoxyla tion of some of the lower aliphatic esters that an appreciable amount of alkylation occurred. For example, from ethyl n-butyrate and diethyl car- bonate there was obtained diethyl diethylmalonate along with the expected diethyl ethylmalonetp. This observation led to the work which will b° mentioned later in this report.
93
Scope and Limitations: It may be seen from Table I that the cerbalkoxyl group is introduced most readily into aryl substituted acetic esters; but the reaction also goes well with alipnatic esters up to ethyl steerate. The a-carbon must contain two hydrogen atoms. If the £--carbon of the ester is tertiary the yield is low. The esterified alcohol of the ester . nd of the carbonate should be the same.
I I ♦ Synthesis of £-Keto Esters .
It would be expected that if alkyl carbonates will condense with esters that they should condense with ketones, and such is the case. The general reaction may oe represented as follows:
R'C^CK3R" + (R0)3C0 + MOR -> [R'C^C-R' ' ]M + 2R0K
GOOR
Welling ford °nd his associates have used the same procedure used in the cerbelkoxyle tion of esters and have prepared a large number of p-keto esters from various ketones end alkyl carbonates; a few of these are shown in Table II. Such a general method of organic synthesis is particularly important since the products are one of the most reactive classes of organic compounds.
Limitations: Ketones which are so active that they condense with themselves or polymerize are not satisfactory. Unreactive ketones such as camphor give low yields. Forcing by heating ?bove lOOo j_s undesirable because of ether formation by the reaction between elcoholaae and alkyl carbonate at that temperature ^-Acylation takes place with certain ketones (cyclohexanone , for exemole ) giving the carbonic ester of the enol form of the ketone.
III. Synthesis of a-Cyeno Esters.
Alkyl carbonates may be condensed with nitriles to give the corresponding cc-cyano esters as follows:
R'CH3CN + (R0)2C0 + MOR
[R'C-CN]M + 2R0H COOR
Although several authors have reported the condensation of diethyl carbonate with phenylacetonitrile by various procedures, none indicated thet this reaction was applicable to other types of nitriles. The Mallinckrodt workers have employed the tech- nique -uaed with esters and. ketones for nitriles and have pre- pared a variety of cc-cyano esters, some of which are shown Table III.
in
Alcohol introduced with the metal alcoholete and that pro- duced in the condensation is removed by distillation. Cope and Hancock have shown that cyeno esters undergo an alcoholysis
-3-
91
analogous to that of malonic esters, producing an alkyl car- bonate and a nitrile. Therefore, here again the distillation of the alcohol favors completion of the reaction according to the mass law and prevents cleavage of the reaction product.
Scope end Limitations: Vinylecetonitrile and p_-nitrophenyl- acetonitrile appear to "be too active and result in tars. cc-Phenylbutyronitrile failed to react. A general statement may now be made in regard to the use of alkyl carbonates for carbalkoxylation:
(1) The method is general for esters, ketones, and nitriles having two alpha hydrogen atoms.
(2) Na and K alcoholates are equally effective as con- densing agents; Mg and Al alcoholates are not effective.
(.3) Primary alkyl carbonates react satisfactorily; the use of secondary carbonates is not practical.
(4) Lower members of each series give poor yields of cerbelkoxylated products either because the low boiling point prevents forcing the reaction, or because of polymerization.
IV. Alleviation of Malonic Esters by Alkyl Carbonates.
It was noted in the synthesis of ms Ionic esters with alkyl carbonates that as a side reaction the sodio derivative was alkylated; this was found to take place in the following manner:
[R'C(COOR)3]M + (RO)3CO -> R»C(COOR)3 + RMC03
R
This is a new type of alkylation; organic esters, including alkyl carbonates, are known to act as alkylating agents in the Friedel-Crafts reaction, but metal enolates have not been alkylated by esters of organic acids before.
The alkylation may be carried out on the metal derivative of a malonic ester, or a mono-substituted acetic ester may be used as starting material since carbalkoxyla tion gives the malonic derivative. The reaction is carried out by refluxing at temperatures of 125-175° with a large excess of the alkyl carbonate.
Scope ant? Limitations: Ethyl, butyl, isobutyl, ieoamyl, and benzyl groups have been introduced into a variety of esters (see Table IV). Secondary carbonates are poor alkylating agents. Malonic ester itself reacts to form tricarbalkoxymethane . If the first substituent is a s~aliphatic group, alkylation is poor. Best results are obtained if the alkyl carbonate, metal alcoholate, mr1 malonic ester all have the same alkoxy group. Preliminary experiments with substituted a-cyanoacetic esters and acetoacetic esters gave negative results.
-4- 95
Vi Alkyl Carbonates as Solvents for Metallation and Alkylation.
In the foregoing synthesis it will be noted that the product is in the form of the metal derivative. Not only is it easy to alkylate directly by means of the alkyl carbonate solution itself, but it has now been shown that alkyl carbonates are superior solvents for alkylation by means of alkyl halides. •
Previously, esters containing a branched substituent such as the .g_-butyl group have been very difficult to alkylate further by the usual procedure using sodium alcoholate in alcohol. This has been ascribed to the incomplete formation of the sodio derivative according to the equation
R'CH(COOR)2 + NaOR^=*[R'C(COOR)2]Na + ROH
The alkyl helide when added reacts to a large extent with the NaOR present. Furthermore, as previously noted, the alkylated malonic ester formed may undergo alcoholysis. Roth of these conditions may be avoided by using alkyl carbonates as reaction media. The above reaction may be forced to completion by removing the alcohol by distillation; the metal derivative may then be treated with the alkylating agent in alkyl carbonate solution with no danger of alcoholysis.
The alkylation of malonic, p-keto, and cc-cyano esters were carried out in alkyl carbonate solution with r.lkyl halides as the alkylating agent. Excellent yields were obtained as shown in Table V. It may be seen that ethyl and allyl groups are readily introduced into diethyl s-butylmalonate whereas this is practically impossible in alcoholic solution. The method has also enabled Wallingford to introduce two s.-alkyl groups as substituents, which has hitherto been impossible.
As would be expected, slight alkylation may take place by action of the alkyl carbonate, hence it is desirable to use as reaction medium the alkyl carbonate corresponding to the group to be introduced.
Bibliography
Wallingford, Homey er, and Jones, J. Am. Chem. Soc . , 63, 2056 (1941 ).
Wallingford, Homey er, and Jones, ibid. . 65, 2252 (1941).
WiFlicenus, Ber., 20, 2930 (1887).
Lux, ibid. , 62, 1827 (1929).
Nelson and Cretcher, J. Am. Chem. Soc, 50, 2758 (1928).
Skinner, ibid., 59, 322 (1937).
Cope and McElvain, ibid., 54, 4319 (1932).
Connor, ibid. , 55., 4597 (1933).
Wallingford, Jones, and Homey er, ibid. , 64, 576 (1942).
Wallingford and Jones, ibid. , 64, 578 (1942).
Wallingford, Thorpe, and Homeyer, ibid., 64, 580 (1942).
Cope and Hancock, ibid. , 61, 96,776 (193977
Reported by R. M. Roberts, April 15, 1942.
•5-
96
TABLE I.
Starting Material
Product
% Yield
Ethyl acetate
Ethyl butyrate
Ethyl stearate
Ethyl phenylacetate
Ethyl p_-methylphenylacetate
Diethyl raalonate 25
Tricarbethoxymethane 10
Diethyl ethylmalonate 45
Diethyl di ethylmalonate 10
Diethyl cetylmalonate 50
Diethyl phenylraalonate 86 Diethyl p_-methylphenylmalonate 65
TABLE II.
Starting Material
Pro due t
t Yield
Diethyl ketone
Methyl isobutyl ketone
Di-n -propyl ketone
Acetophenone
Dibenzyl ketone
Propiophenone
Cyclohexsnone
Ethyl p-keto-a-methylvalerate 20
Ethyl p-keto-^-methylcaproate 60
Ethyl (3-keto-a-ethylcaproate 45
Ethyl benzoylacetate 60
Ethyl a, Tf -diphenylacetoacetate 45
Ethyl a-benzoylpropionate 37 Ethyl 1-phenyl-l-propen-l-yl
carbonate 25
Ethyl 1-cyclohexen-l-yl carbonate 20
Starting Material
TABLE HI. Product
% Yield
Acetonitrile Butyroriitrile Capronitrile Stearonitrile Phenyl acetonitrile
Ethyl cyanoacetate Ethyl a-cyanobutyrate Ethyl a-cyanocaproate Ethyl a-cyanostearate Ethyl cc-cyanophenylacetate
10 40 54 75 78
TABLE IV.
Starting^ Material
Ethyl butyrate Diethyl ethylmalonate Ethyl ^ -methylceproa.te Diethyl .s-butylmalonate Dibutyl ethylmalonate Di butyl cetylmalonate Fluorene
Product
% Yield
Diethyl diethylmalonate 36 Diethyl diethylmalonate 54 Diethyl ethylisoamylmalonate 45 Diethyl .s-butylethylmalona te Poor Dibutyl butylethylrnalona te* 42 Dibutyl butylcetylmalona te* 83 Butyl 9-bu tyl-9-fluor en ecar boxy- late 45
9'
-6-
TABLE V,
Starting Materiel Alky! Kallde Product % Yield
Diethyl ethylmalonate EtI Diethyl di-:-thylmalonate 83
Diethyl ethylmalonate i-AmBr Diethyl ethylisoa.mylmalonate 75
Diethyl e.-butylmalonate EtBr Diethyl .s-butylethylmalonate 95
Diethyl ^.-butylmalone te AllylBr Diethyl e-butylallylmalonate* 86
Diethyl .s-butylmalona te s-3uBr Diethyl di~s.~butylmalonate* 25
* New compound
GLYCEROL DERIVATIVES
98
The specificity of triglycerides for various animal species has been shown. Structurally similar animals frequently eilatorati similar kinds and proportions of fatty acids combined as tri- glycerides. LiKewise it has been shown that the more complex fatty acid mixtures are found in fats of the simplest forms of plant and animal life, and a. gradual simplification is observed both in structure and number of component acids on ascending the evolutionary scale of development. With recent developments in molecular distillation and fractional crystallization it is reasonable to assume that more work will be done in the near future in isolation and purification of the triglycerides, and their structures will have to be checked by syntheses.
Accordingly, some workers have been engaged in fundamental research on triglycerides. A brief summary of their successes in synthetic problems constitutes this seminar report.
Synthesis of Simple Triglycerides
No special procedures need be resorted to in preparing simple triglycerides. Merely heating three moles of the acid with one mole of glycerol yields the tri-ester in nearly quanti- tative amounts. The slight oxidation which occurs at such tem- peratures may be avoided by use of an inert atmosphere such as
M
2 )
CO
2 )
or SO
2 •
CH3OH
CHOH +
I
CH3OH
3 RCOOH
2 CO'
6om
CH30C0R
\
CHOCOR
I
CH2OCOR
Syntheses of Symme t rical Mixed Triglycerides
The preparation of mixed triglycerides cannot be effected by the above method, i.e., by heating in either order a mole of glycerol with a mole of one acid and then with two moles of another acid. Such a reaction yields a variety of triglycerides, and the usual methods of fractional crystallization and dis- tillation can not be applied with any degree of success in attempting to purify any one product.
The following methods have been used:
CH3OH CHOH
CH3OH
Ha SO 4
CH30S03H
CHOH
!
CH3OS03H
2 R'COOH
CH3OCOR»
CHOH
t
CH3OCORf
RC0C1
.
-?.-
CK20H
CHOH
L
OH
C6HsCHO
CKaQ
CHOH piC6H5
CH30'
RCOC1
Pyridine
CH30
choco:
CHoO-
HCpH,
H3
/
CR3OH
CHOC OR i/R'^Cl
CHoOH
CH2OH
CHOH (C6HjJ3CCl
CH3OH
CH2OC(C6H5)3
HOH H2OC(C6H5)3
RCOC1
R'COCl
CH2OC(C6H5)3
i CHOCOR
CH3OC(C6H5)
CH3OH
CHOCOR
I
CH30H
The final product in each of the above three equations is the same symmetrical triglyceride. The latter method is of more recent date and appears to be successful for a wider variety of glycerides.
Syntheses of Unsymmetrical Mixed Triglycerides
The preparation of unsymmetrical mixed triglycerides involves first, of course, the formation and isolation in the pure state of either the a-monoglyceride or a, p-diglyceride.
Among the following syntheses, I is only of historical interest, II, III, and IV represent fairly satisfactory methods of syntheses, and V*s authenticity has been questioned because of its similarity to VI, which yields a product whose formation seems to involve a so-called acyl shift.
H3C1
H2S04
CH3OS03H
CHOSO3H
CH2C1
RCOOH
CH2OCOR
(JjROCOR
CH3C1
R'COOK
II.
5H3OH
3HOH (C6H5)3CC1
CH3OH
CH2OC(C6H5)3
CHOH 2 RCOC
T -*
CHpOH
1 I
H2OC ( C6H5 );
CHOCOR CH2OCOR
K;
Fd
~3~
CH2OH
CHOCOR
CH3OCOR
R'COCl
III. CH8ONa I CHOH
CH2OH
C6H5CH30C0C1
CH2OOCOCHsC6H5
CHOH
GHoOH
CH3OOCOCH3C6H5 CHOCOR " CH20C0R
2 RCOC1
H>3 Pd
CH2OH
CHOCOR
CHoOCOR
R'COCl
IV.
CH2OH
I
CHOH
acetone
CHpOH 1^ HC1
CHoQ
I >C(CH3)2
CHO^
I
CH2OH
R'COCl
cold quinoline
CH2OH
choh 2 R-aoci
I
CH2OCOR
v.
f|H2
CH I
CHpOCOR
HOI
CHgOH
CHI I
CH2OCOR
RCOOK
CH3OH
I
CHOCOR
CHpOCOR
R'COCl
VI.
CH2OH I
CHOH
CH3I
2RC0C1
CH3OCOR
I
CHOCOR
CH21
AgN03 EtOH
3H3OCOR
taoH
JHaOCOR
The a~g Acyl Shift In Dlglvcerldes
The expected product from VI (above) would be the a,§- diglyceride, end isolation of the a, a' -diglyceride to the exclusion of eny of the expected substance can be explained only by spying that the acyl group first went into the beta position and sub- sequently shifted to the al-pha position.
-4- 101
1
H30H
ch3o m CH3OCOR
|"vV -* CHO^ \R -* CHOH CH3OCOR CH3OCOR CH3OCOR
Fischer postulated the above mechanism of acyl shift several decades ago. Hibbert and Grieg, in 1931, substantiated Fischer's hypothesis by obtaining a stable ring structure in the acylation of ethylene glycol with trichloroacetic acid.
CH3OH CH30 /OK
I + ClgCCOOH -> I NC + H30
CH3OH CH3(/ XCC13
Apparently the polarity occasioned by the presence of the three chlorine atoms serves to stabilize the oxolone structure. Hibbert and Grieg presume that in all such acylations of a lpha - dihydroxy compounds there is an equilibrium between such an oxolone structure and the normal ester.
This a-p shift, seemingly, is applicable only to aliphatic acyl groups; several a, (3-aryl diglycerid.es have been found to be stable. Further, this acyl shift is not encountered during such catalytic hydrogenations as shown in equations II and III..
Bibliography
Grun, Ber., 40, 792 (1907)
Fischer, Ber., 53, 1621 (1920)
Helferich and Sieber, Z. physiol. Chem., 175, 311 (1928)
Garner, J. Soc . Chem. Ind . , 47, 278, 801 (1928)
Bergmrnn and Carter, Z. physiol. Chem., 191, 211 (1930)
Hibbert and Grieg, Can. J. Research, 4, 254 (1931 )
Jackson and King, J. Am. Chem. Soc, 55., 678 (1933$
Verted e et al, Rec. trav. chim., 54, 716 (1935)
Daubert and King, J. Am. Chem. Soc, 61, 3328 (1939)
Daubert, ibid, 62, 1713 (1940)
Daubert and King, Chem. Revs., 29, 269 (1941 )
Reported by Jos. F. Shekleton April 15, 1942
DETERMINATION OF BRANCH CHAIN METHYL, CITRIC ACID, AND ORGANIC HALOGEN
102
Determination of Branch Chain Methyl
Kuhn and L'Orsa developed a method for determining side- chain methyl groups in natural products (Vitamin A, Carotenoids, etc*) composed of isoprene units. The method consists in oxidiz- ing the substance with chromic acid in sulfuric acid, then dis- tilling and titrating the acetic acid formed. Pregl and Roth have adapted the process to micro-analytical work. King has applied the method to volatile materials by carrying out the oxidation in a micro bomb.
A variety of compounds has been analyzed in this way (Table I).
Compound C2H5OH
C2H50C2Hs C2H5OC- — R
CH3C^— OR
^0
CH3C~CH2R
CH3CH-CH-R
I I
OH 0!
OH CH3CH=CH-R
CH 3
=CH-C=CH
CH3C
CH
TABLE I.
Moles of Acetic Acid Moles of Acetic Acid
per mole of Compound Compound per mole of compound
^<2>
& N^CH.
1
1
.95-1
1
.85
.95
.85
.90
.10
,12
QK3/CH3
1
.40
CHo JDH9
c
lC^C-CH=CH-C
CH.
-CH.
CH^-C
!5h 2.70
CH I
CH2 CHpOH
C H o C H q
CHa NC~CK=CHC~
L3 I
CHS
CH.
H
i-CH.
2.00
on3
W N^vNH.
.70
103
It is evident from the figures given that the method is not as precise as might be desired. It has "been most useful in choosing one of several possible formulas by comparing the amount of acetic acid found with the t calculated from the figures given in Table I. For example, a-ionone would be expected to yield (0.4 + 0.90 + 0.85) - 2.15 molecules of acetic acid per molecule. There were found 2.0 molecules.
Determina tion of Citric Acid
Of the various methods used for the determination of citric acid, the pentabromoacetone method has produced the most satis- factory results. The details of the procedure are given by Deysher and Holm in a recent publication. The method consists of the oxidation of citric acid with potassium permanganate in the presence of bromine, under controlled conditions. The acid is converted quantitatively into pentabromoacetone, a white crystalline solid which can be determined gravimetrically„
CKS-C00H
H0-C-C00H
CHs-COOH
KMn04 KBr
H3S04
CHs-C00H
I
c=o
I
CH3-C00H
P CBr3(f— CBr3H
Kunz was the first to utilize the reaction to determine the citric acid content of milk, wines, and other food products* The citric acid content of urine also has been determined by this method .
Because of the slight solubility and volatility of the penta- bromoacetone, no absolute method can be prescribed for the com- plete recovery of citric acid as pentabromoacetone under all conditions. The method must be standardized with respect to the conditions and products employed. A complete discussion of various other methods and modifications is given by Lampitt and Rooke.
Kometiani determines the pentabromoacetone by treating it with hydrogen iodide in alcoholic solution to liberate six atoms of iodine per molecule. The iodine may then be determined volume trically.
It seemed probable that compounds similar to citric acid might be determined in a like manner but a study of the litera- ture disclosed no such examples.
Deter.ii3Jna_ti.on. of Organic Haloget-
The determination of organic halogen by the liquid ammonia- sodium process is both rapid and accurate. Chablay was the first to employ this reaction quantitatively. The method of Vaughn and Nieuwland, an adaptation of the Chablay method, does away
-"&M
104
with all special apparatus, and gives increased rapidity and accuracy of analysis of insoluble materials. A small amount, 0.1 to 0.4 g. of the halogen-containing organic -material is introduced into liquid ammonia contained in a beaker. If solu- tion does not take place upon stirring, some organic solvent (as ether) inert toward sodium in liquid ammonia is slowly added until the material is dissolved. One gram of sodium is then added in small pieces and the covered beaker allowed to stand until the reaction is complete (usually 30 seconds to two minutes). The excess sodium is removed by ammonium nitrate, and the solution allowed to evaporate to dryness. The solids are taken up in water acidified with nitric acid and the halogen content deter- mined with silver nitrate either gravimetrically or volumetri- cally by the method of Fajen or Volhard.
In some cases organic solids are produced during the decom- position. These must either be filtered off or put in solution by the addition of aldehyde-free acetone or other suitable solvent before precipitation of the halide.
Clifford noted that cyanides were formed with certain highly chlorinated compounds as carbon tetrachloride and hexachloroethane end caused high halogen values. He devised a method for removal of the cyanide which consists in boiling a. nitric acid solution. Dains and Brewster investigated the action of sodium in liquid ammonia on 12? organic halogen compounds and found cyanide pro- duced from only eight. In ell these cases where cyanide was produced two or more halogens were attached to a single carbon atom. Methyl cyanide, benzyl cyanide, cyanoacetic ester evidently splits off sodium cyanide directly.
Organic fluorides also may be determined by following this procedure except that the fluoride is precipitated from the water solution with calcium nitrate as calcium fluoride and determined gravimetrically.
Some of the results of Vaughn and Nieuwlend may give an idea of the reliability of the procedure.
Compound % Halogen Present % Halogen Found
Trichloracetic acid
Chloralhydrate
Hexachloroethane
Chlorobenzene
p_-Dichlorobenzene
Hexa chlorobenzene
Ethyl bromide
3, 5-Dibromobenzoic acid
Styrene dibromide •
n-Butyl iodide
Methylene iodide
Phenyl iodide
Fluorobenzene
1,3-Dime thy 1-5-bromofluoro benzene
o.-Chlorophenylf luorof orm
65.11 |
65.19 :' |
64.33 |
64.21 |
89.86 |
89.91 |
31 . 53 |
31 . 42 ' |
48.28 |
48.23 |
74.73 |
74.80 |
73.37 |
73.46 |
57.15 |
57.00 |
60.56 |
59.95 |
68.98 |
68.86 |
94.77 |
94.84 |
62.23 |
62.23 |
19.79 |
19.78 |
9.36 |
9.30 |
31.58 |
31.36 |
-4- j05
The average per cent difference for all determinations run was .14 although the large majority ran below 0.1^.
This method ha s been used in this labor? tory with good results.
Bibliography
Kuhn and L'Orsa, Z. Angew. Chem., 44, 847 (1931)
Pregl and Roth, Quantitative Organic Microanalysis, P. Blakiston's
Son and Co. Inc., Philadelphia, Ed. Ill, 1937, p. 201. Kuhn and Roth, Ber., 65, 1285 (1932)
Deysher and Holm, Ind. Eng. Chem. Analytical Ed., 14, 4 (1942) Kunz, Arch. Chem. Mikroskop., 7, 285 (1914) Lampitt and Rooke , Analyst, 61, 654 (1936) Kometiani, Z. anal. Chem., 86, 359 (1931 ) Chablay, Ann. Chim. (9) 1, 510 (1914) Clifford, J. Am. Chem. Soc, 42, 1573-9 (1920) Vaughn and Nieuwland, Ind. Eng. Chem. Analytical Ed., 3, 274 (1931)
Reported by Curtis W. Smith April 22, 1942
106
RECENT DEVELOPMENTS IN THE IDENTIFICATION OF ORGANIC COMPOUNDS
I. Acid s
Dewey bromid. e as reagent is When a hot acqueous so the derlvat The £~ehlor used "but bo of the salt
and Shasky suggest the use of p-bromopseudothiouronium a reagent for the identification of acids. The prepared from p_-bromobenzyl bromide and thiourea, alcoholic solution of this bromide is added to an lution of the sodium or potassium salt of the acid, ive precipitates out in the pure state on cooling, opseudothiouronium chloride reagent has also been th have the disadvantage that the melting points s formed cover only
a small range (about 140-170 ).
In addition tc It? us-"1 as a oerbonyl reagent, phenyl-- hydrazine has been used to form acid derivatives.. Steffipel and Schaffel tried this compound on aliphatic carboxy acids when they observed the success of its use on aliphatic sulfonic acids. To prepare the phenylhydra zide it is only necessary to boil a solution of the acid in phenylhydra zine for thirty minutes. The derivative precipitates on cooling or by the addition of benzene.
II. Alcohols
In order to identify the cellosolves, carbitols, and related glycols, Seikel and Huntress developed a procedure for the formation of the solid trityl ethers. The derivatives are formed by heating the alcohol and triphenylmethyl chloride in a mixture of pyridine and ether.
By means of a modified Schotten-Baumann reaction Lipscomb and Baker have discovered a good way to identify alcohols in an aqueous solution such ag would result from the saponification of an ester. A solution of 3, 5-dinitrobenzoyl chloride in benzene is made then d little ligroin is added to prevent freez-
ing of the mixture during the reaction. This with the aqueous solution of alcohol, sodium
is ace
shaken at ;ete, and ,
0 ■lkali
Alcohols may be identified by the use of saccharin chloride The reagent is easily prepared from insoluble saccharin and phosphorus pentachloride. It reacts with alcohols in the absence of water to give saccharin alkyl ethers in which the alkyl gr-oup is attached to the carbon through the oxygen bridge.
ROH -* ^\^%^R
+ HC1
-2-
The isomeric nitrogen ethers can be formed by the reaction of the alkyl helide on sodium seech? rin. Saccharin chloride was used to identify primary and secondary alcohols.
Ill- Aldehydes and Ketones
By the use of the methyl p.- to luenesulf onate addition product of nicotinic acid hydra zide it is possible to secure
107
CONKNH.
0 rfli <y bJ ■
solid derivatives of aliphatic aldehydes having a melting point
40 higher than the corresponding 2,4-dinitrophenylhyd.re zones.
It has the advantage in that the aldehyde is very easily recovered
or it can be converted directly into another derivative such
as the 2,4-dinitrophenylhydrazone. It gives oils with unsaturated
ketones and the range of melting points for the C6-C10 aldehyde
derivatives is narrow.
Sah added another carbonyl reagent to his long list when he described the preoaration and use of p_-iodobenzhydrazide. This compound is easily prepared from methyl p_-iodobenzoate and 40^ aqueous hydrazine hydrate. Refluxing a mixture of the hydra- zide, excess aldehyde, and two drops of acetic acid in pure alcohol gives the desired derivative. This reagent has two distinct advantages: (a) the yields are so high that it is easy to operate with 300 mg. of the aldehyde and still get enough material for a study of physical properties and analysis, and (b) the melting points of the derivatives of the elkanel series from C6-C10 are widespread. This is not true of the 2,4-dinitro- phenylhydrf zones or semicarbe zones for this series.
A new optically active carbonyl reagent of this type has been described by Woodward, Kohmen, and Harris. This is prepared from .1-menthol by conversion to l_-menthylhydrazide. It gives very good derivatives with nice melting points end has the added advantage of having a characteristic specific rotation. It is of interest in that it was used in the first successful resolution of dl-cemphor.
A new type of carbonyl derivative has been developed by Henze and Speer. In their work with keto ethers, they found that certain of their compounds did not give solid derivatives with the ordinary carbonyl reagents but did form solid hydentoins. The carbonyl compound was warmed in dilute alcoholic potassium cyanide and ammonium carbonate:
~3~ 108
R' R-CO-R' + KCN + (NH4)3C03 -> R-C CO
I
NH-CO-NH
(R1 can be K or R)
In most cases the hydantoin separates In a state of sufficient purity to eliminate further recrystellization. This is not useful for formaldehyde, certain unsaturated aldehydes, a few nitro- and hydroxy-aryl aldehydes and pyruvic acid. However, levulinic acid and ecetoecetic ester are converted to the corresponding 5, 5-di substituted hydantoine.
IV. Amines
By the dia zoti zation of ]>-±odobenzhydrazide, one can get p_~iodobenza zide. When this reagent is refluxed in an anhydrous medium, usually toluene, the corresponding substituted urea is formed. Again, as in the case of the hydra zide, only a email amount (200 mg. ) is necessary for a complete physical examination pnd analysis.
Billmen et al used 2, 4-dinitrobenzenesulfonyl chloride as a reagent for the identification of amines. The reagent is prepared from 2,2 ' ,4, 4'~tetrenitrodiphenyl sulfide and chlorine.
, ^s N03
°3N<C-3>~S~C1 + 2RNHs ~* °^<^l [^,-S-H-R + RNH2.HC1
H
It can be used even With a 30f aqueous solution of the cmine.
V- Hydrocarbons
Although it is not a new reagent and identification work with it is not new, trinitrobenzene has been found useful in identifying aromatic compounds. The derivatives, which are much like the corresponding picrates, have been listed for certain aromatic compounds: acids, aldehydes, amines, ethers, hydrocarbons, ketones, and nitriles; mostly for amines, hydro- carbons, phenols and phenol ethers. One interesting fact about this reagent is that the number of trinitrobenzene molecules used does not vary with the number of functional groups but with the number of aromatic nuclei!. Certain pyrrole derivatives, which do not give picrates, do form the trinitrobenzene compound.
The identification of pure paraffinic hydrocarbons has usually been made only through the use of physical constants of the compounds. Huntress has developed desoxycholic acid as a reagent for the preparation of derivatives of these hydrocarbons. He has applied it to three pentanes, five hexanes, seven heptanes, eight octanes, three nonanes, eight decanes, and three higher
-4-
paraffins, making a total of thirty-seven hydrocarbons. The derivative is produced when desoxycholic acid in methanol is allowed to react with the hydrocarbon.
VI . Phenols
Pnenols can be identified by the use of p_-iodobenzazide (see first reagent under amines). In this case the result is a substituted urea.
VII. Sulfonic Acids
Although characterization of carboxylic acids by the formation of their p_-nitrobenzyl esters has been a standard procedure, the analogous preparation of the corresponding sulfonates has hitherto proved impossible. By employing the silver salt of the sulfonic acid with p_-nitro benzyl chloride in pyridine, and taking advantage of the fact that the resultant silver chloride is less soluble in hot than cold pyridine, and that the solvent combines with the sulfonate to yield a readily crystallizable pyridium salt with sharp and significant melting point, Huntress has characterized twenty common aromatic sulfonic acids.
Chambers and Watt have extended the use of benzyliso- thiouronium hydrochloride to include 34 sulfonic acids. Impure sulfonic acids can be used but if there is an amino group present, the method usually fails.
If there is an amino group present and ordinary identification methods fail because of inner salt formation or sensitivity of the amino group, then Allen and Frame suggest the replacement of the amino group by a chlorine atom througn the Sandmeyer reaction. The sulfonyl chloride, sulfonamide or sulfonr nilide can then be made. The yields are good enough so th: t 1.5-2 gms. are all of the original acid that is necessary.
VIII. The Iodoform Reaction
Rothlin has developed a method whereby it is possible to distinguish between R-CO-CH3 and R-CHOH-CK3 by means of the iodoform reaction. Using a reagent mpde of one part potassium cyanide, four parts iodine, five parts ammonium hydroxide and fifty parts of water, a positive test can be obtained for the methyl ketone only. It may also be used quantitatively to determine the amounts of these two compounds in a mixture. By the ordinary iodoform reaction, the total amount of the two types can be determined; then by means of this second reagent the amount of methyl ketone can be found.
~fi-
110
Bibliography
Dewey and She sky , J. Am. Chem. Soc . , 63, 5526 (1941 ).
Stempel end Schaffel, ibid., 64, 470 (1942).
Seikel and Huntress, ibid. , 63, 593 (1941).
Lipscomb and Baker, ibid'. . 64, 179 (194?,).
Meadow and Reid, Memphis Meeting, American Chemical Society,
April, 1942. Allen and Gates, J. Org. Chem., 6, 596 (1941). San and Ksu, Rec. Trav. Chim., 59, 349 (1940). Henze and Speer, J. Am. Chem. Soc*, 64, 522 (1942). Woodward, Kohman, and Harris, ibid. , 65, 120 (1941). Sah, Rec Trav. Chim., 59, 364 (1940). Sudborough, J. Chem, Soc, 109, 1339 (1916). Billman et al, J. Am. Chem. Soc, 63, 1920 (1941). Huntress, Memphis Meeting, American Chemical Society, April, 1942 Sah, Rec. Trav. Chim., 59, 357 (1940). Chambers and Watt, J. Org. Chem., 6, 376 (1941). Allen and Frame, J. Org. Chem., 7, 15 (1942). Rothlin, see C. A., 35, 5091 (1941).
Reported by Q. F. Soper April 22, 1942
ill
SULFADIAZINES
The antistreptococcic activity of Prontosil, 2,4-diamino- azobenzene-4' -sulfonamide was discovered in the laboratories of I. G-. Farbenindustrie by Mietzsch, Klarer, and Domagk in 1935. That same year Tre'fouel, Trefouel, Nitti, and Bovet at the Pasteur Institute while synthesizing Prontosil derivatives found that sulfanilamide was just as active. During the next five years synthesis of about 1300 new derivatives was published. Among the most successful of these derivatives were sulfapyradine (2-sulfanilamidopyridine), sulfathiazole (2- sulfanilamide thiazole ), sulfaguanidine , and sulfadiazine (2-sulfanilamidopyrimidine ).
After its synthesis by Roblin and co-workers, sulfadiazine proved to be more effective, more soluble, and less toxic in general than the other active sulfa-drugs. As a result a great many pyrimidine and pyrazine derivatives of sulfanilamide were prepared.
(l) N-CH(6) N=£H HN-g=0 OH
N-CH 6 ) N=CI
hi,, ! '
IC CH 5 HC CI
(2) HC CH(5) HC CH H3N~C CH r=* H3N // N^
, , i L, , n ii it i n*=^
(3) N=CH(4) HC-N
pyrimidine pyrazine isocytosine
Synthesis of Isocytosine.
For the synthesis of isocytosine, an intermediate in the synthesis of several pyrimidine derivatives, Roblin and co-workers used a condensation of malic acid with guanidine sulfate in the presence of fuming sulfuric acid.
Isocytosine C00HCH3CH0HC00H + H3NC(=NH)NH3 -> [C00HCH3CH0 + HN=C(NH3)3] $
Among the earlier methods reviewed by Johnson and H^hn was the alkaline condensation of guanidine carbonate with the sodium salt of ethyl formylacetate.
C8H60C0-CH=CH0Na + HN=C(NH3)3 -> isocytosine
Davidson and Baudish developed a synthesis for uracil from urea and malic acid.
NHeCONH2 + H00CCHSCH0HC00H f""-HsS04 ) mcll (2,4-di ke topyr imidine )
Using this method Hilbert and Johnson synthesized cytosine and isocytosine from uracil according to the following reactions:
-2-
±±a
ale. Mixture of the
FU01a > 2,4~dichloropyrim, NH3 2-amino and 4-
amino chloropy- rimidines
NeOCHa ) separate the H30,HC1 > H0_ / \ and ieocytosine
" methyl ethers XN — 4/
NH
3
cytosine
Pyrimidlne synthesis
In the preparation of 2-aminopyrimidine Roblin vnd his co- workers treated isocytosine Kith P0C13 and then reduced off the chlorine in the (4) position with H2 and Pd(OH)2 on CaC03.
The same series of reactions starting with cytosine would lead to 4-aminopyrimidine . The reduction can also be carried out in either case with Zn and H20.
Using a procedure of Hale and Brill to obtain the sodium salt of 2-hydroxy-5-nitropyrimidine, Roblin worked out a synthesis of 5-aminopyrimidine.
NH2CONH2 + CHOCH(NOs)CHO ■ Efe0H.> NaO_^ ^-U02 • F0C1* y
2~chloro-5-nitro- Fe.HOAc ; 2~chloro-5~amino ^~> 5-amino- pyrimidine pyrimidlne ^ pyrimidlne
Nitromalondialdehyd e can be made by the action of sodium nitrite on mucobromic acid according to Hill who used pyromucic acid and bromine as a source of the acid. Simoniswas able to make muco- bromic acid in good yield by the action of bromine on furfural.
furfural + 5 Br3 + 3 H20 — > [pyromucj-c acXdl —> CEOCrzCCOOH
Br Br
Other condensations lead to alkyl and aryl pubstituted pyrimidines, Gabriel has investigated a number of reactions between amidines and various diketones.
±13
-3-
CH NH=CyNH2 + CH3COCH2COCH3 M&L> ^_^f "^ 3 y=GH3 , phenyl, etc.
=^CCH
3
NH=CyNHa + CH3COCH3COOC2H5 -*• y.
OH CH,
OH NH=CyNH3 + NaOCH=CHCOOC2H5 -> y_
^N — ^O
-N
Caldwell and co-workers made 4, 5-dialkylpyrimidines by refluxing the appropriate hydroxymethylene ketones with guanidine carbonate in alcohol. The ketones were prepared by a Claisen condensation of ethyl formate with the proper straight chain or cyclic ketone.
RCOCH3R» + C2H5OOCH Nf* RCOC(R! )=CHOH KN=C(NHS)^ HgN^^^X^-Rt
** ~^N £r~
Ballard and Johnson synthesized 2-a.mino-5~carboxypyrimidine as follows:
w ' ... OH NH=C(SR)NH2 + C3H50CH=C(C00C3H5)3 Na0G*K*> RS-^ \-COOC2H5
% y
2-alkylmercepto- N .
POCla ^ Zn.HaOy 5-carbethoxy- Cla , Cl^T "N^__COOC3H5
pyrimidine ^N— .X^
a 1 o . NH a \ ale .and' 10% KOH ^ 2-amino~5~carboxypyrimidine.
The preparation of ethoxy methylene me Ionic ester was worked out by Wheeler and Johns.
CH3(C00C3H5)2 + CH(C00C2H5)3 + (Ch3C0)20 + ZnCl2 -> C2H50CH=C(C00Et ),
Previously C2H50CH=C(CM)C00C2H5 and H0CH=C (CKO )C00C2H5 had been used in the condensation. In the pseucothiouree R may be hydrogen, ethyl, or benzyl.
1
-4-
114
Roblin, Winnek and English obtained 5-chloro-2-eminopyrimidine from the condensation of chloromalondialdehyde with guanidine carbonate in fuming sulfuric acid. The chloromalondialdehyde was prepared from tetrachloropropene .
CHCl3C(Cl)=CHCl + 95^ H2S04 -> CE0--CHC1-CH0
CH0CHC1CH0 + NH=C(NH3)3 -* 5-chloro-2-eminopyrimidine .
Pyrazine Synthesis*
Kail and Spoerri prepared aminopyrazine starting with quinoxa line.
KMnO* s
N
COOH
sublime v
/ %-COOE
\?
, (70^ yield)
pyrazoic ?cid CKaQH^ ester
NH.
HC1
CH,0H
^ amide , NaOCl
./N^-NHCOaNa
H
aminopyra zine
The authors say that this is the first time a stable sodium carbamate has been isolated from the Hofmann hypobromite reaction,
Sulfanilamide derivatives. N4-substituente:
(4) 5 6
NH
4NV //l
(It
S03NH3
3 2
The synthesis of N -pyrazinoyl sulfanilamide by Daniels and Iwamoto illustrates one method of preparing N4-deriva tivee.
pyrazinoyl chloride + sulfanilamide dry
pyridine
w
-5-
4 115
4-chloropyrimidine is used in the synthesis of 2-N -sulfon-
emidopyrimidine according to the method of de Suto-Nagy end Johnson.
4-chloro- + sulfanilamide ale, v N — . NH >^* "%y_S02NH2
pyrimictme >v '
N -substltuents: ■
There are three m^ in methods of preparing N -derivatives, (l) The use of ecetyleulfanylil chloride on an amine followed by hydrolysis,.
}l Hi
N'-<^3> + CH3C0NH_^~X_.S03C1 pyridine^ suifadi8zlRe
(2) The use of p_-nitrobenzene8ulf onyl chloride followed "by reduction
2-amino-5-chloro- , v
pyrimidine + 03N_.< 'X-S02C1 _py_v Fe.HCl , 5'-chloro-
n^ ^ * dil.alc/ sulfadiazine
(3) The action of an acyl chloride on N -acetylsulfaniiamide.
pyrazinoyl , ^
chloride + CH3C0NPL_X V_S03NH3 py . \0% NaOH
N S x ^
N -pyrazinoyl sulfanilamide Bibliography
Amundsen, J. Chem. Ed., 19, 167 (1942).
Ballard and. Johnson, J. An. Chem. Soc . , 64, 794 (194-2),
Caldwell, Kornfeld and Donnell, ibid. , 6.3, 2188 (1941 ).
Daniels and Iwamoto, ibid . , 63, 257 (1941).
Davidson and B?udisch, ibid.., 48, 2379 (1926).
de Suto-Nagy and Johnson, ibid. , 65, 3235 (1941).
Gabriel and Coiman, Ber. , 32, 1533 (1899).
Gabriel, ibid. . 37, 3641 (1904).
Hale and Brill, J. An. Chem. Soc, 34, 82 (1912)..
Hall and Spoerri , ibid. , 62, 664 (1940 ).
Hilbert and Johnson, ibid, , 52, 1153 (1930).
Hill, Am. Chem. J., 22, 95 (1899).
Johnson and Hphn, Chen. Rev.. 13, 193 (1933).
Morthey, ibid. , 27, 85 (1940).
Roblln, Willlamr, Winnik, and English, J. An. Chem. Soc, 62,
2002 (1940). Roblin, Wlr.nik and English, ibid . . 64, 567 (1942). Roblin and winnik, ibid. , 62, 1999 H940). Simonis, Ber., 32. 2085 (1899). Wheeler and Johns, An. Chen. J., 40, 238 (1908).
Reported by M. Chiddix April 29, 1942
THE EFFECT OF CATALYSTS ON THE GRIGNARD REACTION ±1_6
Kharasch et al, University of Chicago
It is known that several metals and metallic compounds have a significant effect upon the Grignard. reaction -- the nature of the products, the rate of reaction, and the yield. In order to study this effect more closely Kharasch and his co- workers investigated the catalytic effect of several metallic halides upon certain Grignard reactions.
The first reaction to be studied was that of isobutyl magnesium bromide with benzophenone. Ordinarily this reaction goes as follows:
CH3CHCH2Mg3r + 0CO0 -* 0-OHQH
92% yield
CH3
The reaction is thus a 2-electron Grignard reduction. However, when the reaction was run in the presence of manganous chloride, benzopinacol ae well' as benzohydrol was obtained, and the yield of benzopinacol varied directly with the amount of manganous chloride present up to 2 mol per cent of the latter. Chromic chloride and ferric chloride gave similar results but with smaller yields of benzopinacol. Cuprous chloride shewed no effect upon the reaction.
The second reaction studied, was that between methyl magnesium bromide and. benzophenone. The reaction ordinarily yields diphenylcarbinol, but the presence of small amounts of metallic halides altered the reaction to yield either benzopinacol or the carbinol in quantitative amounts.
0CO0 + CH3MgBr, : »# .OH
X
f XCH3
OH OH
2 mol %
catalyst upoq f. pinacol % carbinol
Mg 0 95
CuCl 0 93
MnCl3 0 93
FeCl3 65 21
CoCl3 93 2
11
The me thy Imp g repgents amounts o a yield o duct was various o diagram a
GH,
CH.
~2~
last reaction studied was that of i nesium bromide. Ordinarily the eddi to conjugated cyclic unspturated ket f the 1,2 and 1,4 addition products, f 71$ of the tertiary alcohol and it obtained. However, in the presence ther products were obtained as shown nd table .
0 + CH3MgBr
ophorone with tion of Grignard ones yields varying
a nd i n thi s case s dehydration pro- of metallic haiides
in the following
CH-
GH
3<x
Gil 3
II
, GH
Yields
Mol % |
1,2 ad |
dn |
• |
||||
Catalyst |
Diene |
C |
prbinol |
Cpd.V |
Pinacol |
1,4 addn. |
Total |
none |
48.2 |
42.6 |
90.8 |
||||
1.0 FeCl3 |
2.2 |
81.6 |
9.46 |
94.0 |
|||
1.0 CuCl |
6.96 |
82 . 5 |
89.5 |
||||
1.0 NiCl3 |
7.3 |
22.6 |
61.1 |
4.73 |
96.0 |
||
1.0 CoCl3 |
16.1 |
78.5 |
94,6 |
||||
20 xs Mg |
55.5 |
22.7 |
1.45 |
80.0 |
|||
1.0 AgCl |
57.7 |
35.0 |
92.7 |
||||
1.0 MnCls |
56.0 |
28.5 |
84.5 |
||||
ordinary |
•85 . 5 |
4 . 35 |
1.45 |
91.3 |
|||
Mg |
Determination of the structure of compound V proved to be difficult. On the basis of chemical evidence this compound, should be either the er.ol form of isophorone or the isomer in which the double bond is in the (3, /"position. If the latter structure is correct, then a remarkable rearrangement tak.es place when isophorone is treated with methyl magnesium bromide in the presence of 2 mol per cent of ferric chloride:
ilb
Ch^
Ch3 i 3
* < >°
t
jH3
Following is the evidence for the existence of compound V as distinct from i gopher one:
1. Carbon end hydrogen pnalysse show an empiric? 1 formula CgH140, the same ^ s that of isophorone.
2. The physical properties °re different from those of isophorone .
b,pa °C, b.o>(atm.) nS° d30
compound V 38.0-38*4 (4 mm.) 181-185 1,4620 0.9083
isophorone 69.0-69.3 (5mm.) 210-211 1.4775 0.9215
3. Compound V absorbs oxygen much faster than isophorone. This property caused much difficulty in securing analyses, and the compound he d to be distilled and Kept in high vacuum to get consistent results.
4. Compound V el owl j changes to isophorone at room temp- erature. This conversion is accelerated by a trace of acetic acid, and in -the presence of potassium bisulfate at 150°, the change is complete in one hour.
5. The possibility that compound V might be the alcohol /\3-3, 5, 5, -trimethyicyclohexanoi f orn ed by the reduction of iso- phorone by methyl magnesium bromide v° s disproved by comparison of the physical properties of the two.
B,,p. °C, n^° d30
Compound V 33-38,4 (4mm.) 174820 0.9033
Alcohol 69 (5 mm.) 1.4717 0.9144
6. A comparison of the reactivity of compound V end iso- phorone toward semi carbamide was carried, out. Compound V gave
a precipitate immediately, out one to three hours were required for isophorone to form any precipitate. Both s em icarba zones melted at 186-i87°C, and they did not depress each other's mel- ting poinrs.
7. Compound V is evidently an isomer of isophorone. The enol form should give a test with ferric chloride, end compound V did not, but this test is not conclusive.
3. Each compound reacted with hydro xylamine to give en oxime melting at 78-79°, cut an equimolar mixture of the two oximes melted at 00-52?-
-4-
Thus no definite conclusion could be reached as to the structure of compound V. The best method of deciding on one of the two proposed structures would seem to be their absorption spectra since tne infra-red should reveal the presence of a hydroxy 1 group and the ultra-violet should show the presence of conjugation.
In attempting to explain the effects of the metallic halides, certain considerations must be kept in mind. The coupling by metallic halides has been studied many times, but in every case at least one mol of the halide was used:
2RMgX + MX3 -> R-R + 2MgXs 2 mols 1 mol 1 mol
However, when one mol of bromobenzene is dropped into a mixture of phenyl magnesium bromide (one mol) plus 3 mol% of cobaltous chloride, the bromobenzene acts as an oxidizing agent in converting the phenyl magnesium bromide to bi phenyl. That the biphenyl is formed exclusively from the C-rign- rd reagent is shown by the fact that the bromobenzene may be replaced by p_-tolyl bromide, ethyl bromide, isopropyl bromide, and others with exactly the same results. The reaction is applicable to other biaryls, and n-Q1 bitolyl, SL-SL* bitolyl, 4-4' bianisyl, 2-2' biphenetyl, were obtained in yields better than those obtained using other coupling agents.
This reaction essentially involves the transfer of an electron to the halide from the organic radical of the Grignard reagent. The metallic halide acts as an oxidation-reduction catalyst since it is reduced by the G-rignard reagent and oxid- ized back by tne organic halide. The need of only small amounts of the metallic halide suggests a chain mechanism:
C6K5MgBr + CoCl2 -» C6K5CoCl + MgBrCl
2C6H5CoCl -> C6H5-C6H5 + 2 CoCl-
CoCl' + C6H5Br -> CoOlBr + CSH5«
XC6H5' — > C6E6, C6H5-C6H5, C6H5-C6H4-C6H5, etc.
This mechanism also explains the pinacol rather than the carbinol in the first reaction studied:
CHgMgBr + CoCl 2 -» CH3CoCl + MgBrCl
2CH3CoCl -* C2H6 + 2C0C1-
2CoCl. + 208CO + MgBrCl -* 2C0C12, + (03C-OKgBr)3
Not all Grignard reagents would react with bensophenone in the presence of cobaltous chloride to give the pinacol even
11
_5- 120
1
though the primary requirement, reaction of the Grignard reagent Kith the metallic halide, were fulfilled. The seconc" step, decomposition of the RC0C1 into R-R and CoCl, must be sufficiently rapid or the normal addition reaction will take place exclusively. Thus in attempting to alter the normal course of a Grignard reaction, the rate of normal condensation as well as the stability of the intermediate must be considered.
Bibliography
Kharascn, Kleiger, Martin, and Mayo, J. Am. Chem. Soc . , 63, 2305
(1941). Kharascn and Tawney , ibid. . 63, 2308 (1941). Kha.rasch and Lambert, ibid,, 63, 2315 (1941). Kharascn and Fields, ibid. , 63, £316 (1941). Linsteed, J. Chem. Soc, 1603 (1930).
Reported by A. B. Spradling April 29, 1942
J2i
THE CHEMISTRY OF EXPLOSIVES
Because of the contemporary importance of explosives, their chemistry, both organic end inorganic, is a topic of extreme interest. Inasmuch as exact information of modern developments is unavailable, this report will summarize the practices of the explosives industry during and following World War I.
History
Historically, tradition records that Berthold Schwarz, a German monk, discovered black powder around 1250, but various documents indicate that Roger Bacon in the thirteenth century was the first European scholar acquainted with the use of saltpeter in incendiary mixtures. Even earlier, around the eighth century, the Byzantines and Greeks described various flame-throwers and pyrotechnics. After guns began to be used in the fourteenth century, the best proportion of potassium nitrate, charcoal, and sulfur (6:1:1 ) was found and later developments were chiefly in the manufacturing methods. As an improvement, the potassium nitrate end sulfur were replaced by ammonium nitrate to produce Ammonpulver, a cheap, powerful, flashless, smokeless propellent.
The discovery of smokeless powder by Felouze (1838) and its development by Abel in the form of guncotton (nitrocellulose) marked the beginning of modern firearms; Nobel was the first to gelatinize guncotton end nitroglycerin with acetone.
Properties.
The important physical properties to be considered are:
1. Shattering Power (brisance)
a. Strength of Detonation- depends on volume of gas evolved and amount of heat liberated. Determined by measurement of the distension after explosion
in e lead block (Treuzl test), by recoil of a pendulum in a ballistics gun, or by means of a crusher gauge in a manometric bomb.
b. Velocity of Detonation- depends on constitution end packing density. Measured by a rota ting-drum chronograph or by comparison with standard samples (Deutriche test).
2. Sensitivity a . Impa c t
b. Temperature of Ignition
.?. Stability
a. Time end temperature for N02 evolution
b. Hygroscopic nature
\22
•2~
Classification
Explosives may be classified conveniently by reference to their responses to the stimuli causing the .explosions. Thus, there ere primary explosives, propellents, and high explosives (with an intermediate sub-class of boosters).
The behavior of the various classes of explosives is best illustrated by considering a typical ammunition shell containing both a. propelling charge and a bursting charge.
Igniter
Propellent
Primer
Driving
Band
Booster /
u :
Bursting charge
Detonator Fuze
Thus ,
flame
sets
burni
ga. s e s
collo
inal
acqui
the s
is ne
the p
the fir , igniti fire to ng of th
from th ided nit perfora.t re a rot oft cop-o cessary ro j^ctil
ing pin cau ng the blac the igniter e grains of e combust! o rocellulose ions, cause a ting mot le- er driving to insure a e.
ses the primer cap to produce a small k powder in the primer. This in turn , from which the large, hot flame causes
smokeless powder, the propellent. The n of the smokeless powder, which is
of cylindrical grains with 1-7 longitud- s the projectile to move forward and to n as the rifling of the gun bites into band. The proper burning of the powder
gradual and powerful acceleration of
If the shell is intended to explode a definite time after leaving the gun, a firing pin in the fuze undergoes set-back, striking a primer cap containing a primary explosive which ignites the slow-burning powder train adjusted by time rings. When this flame reaches the detonator and bursting charge, the shell explodes in flight. For detonation on striking the target, 8 percussion element is included. The centrifugal force causes a firing pin to rise up so that inertia, will cause it to strike the cap, produce fire in the powder charge, and result in an explosion. The usual high explosive must be insensitive to tolerate the shock of the set-back. In general TNT is used for this purpose along with a booster of tetryl which aids in the detonation and reduces the quantity of primary explosive necessary The bursting charge may also contain shrapnel or a war gas.
Naturally, there is some overlapping in the classification of explosives. For example, nitrocellulose may function as a propellant if ignited by fire, or as a high explosive if detonated by a primary explosive such as mercury fulminate. In adoption of a particular explosive depends to a large its stability, sensitivity, and reaction to climactic in addition to its strength.
all cases, extent upon conditions
-3- 123
I* The Prime- ry Explosives (initiators, detonators)
These compounds explode or detonate when heated or subjected to shock, but do not burn. Their uses vary with their brisance; mercury fulminate, lead azide, ana nitromannite are by far the most common initiators..
1. Mercury fulminate - Hg(ONC)3 ~ used with potassium chlorate and sometimes with powdered glass. Prepared
by dissolving mercury in 60% nitric acid, adding alcohol, and separating the grey oowder. Eleven times as sensitive as TNT".
2. Lead Azide - Pb(N3 )s - half as sensitive as fulminate. Prepared by 0
NftNH8 + N80 -+ NaNi Pb^N°*-H Pb(N3)s
3. Nitromannite - the hexe-nitrate of mannitol, which is made by electrolytic reduction of mannose. Nitrosorbitol is also used.
Others of lesser importance are lead picrate, trinitroresor- cinol, m-nitrophenyldia zonlum perchlorate, tetracene, nitrogen sulfide, copper acetylide, and nitrosoguanidine.
II. The Propellants (low explosives)
These are combustible materials containing within themselves all the oxygen for their combustion; they burn but do not explode and function by producing gas which explodes.
1. Black Powder - in the United States, it usually contains 75# sodium nitrate, lb% carbon, and 10$ sulfur. It is now used chiefly in coal mining, as time fuses, and in shrapnel shells. Its products of combustion are 56^ finely divided solids.
2. Smokeless Powder - Cellulose in the form of clean cotton or wood chips is nitrated by a nitric-sulf uric acid mixture to give a product containing 12.8^ N, which corresponds to the formula C24H30010(N03 ) 10. This "nitrocellulose", plasticized by alcohol-ether and stabilized by diphenylamine, is forced through perforated dies, cut, and dried to give the desired guncotton. An improved military rifle powder now being made contains 13.1?? N.
3. Cordite - a double-base powder developed by the British which contains nitrocellulose and about 30^ of nitro- glycerin, gelatinized by acetone. Burns more readily and faster and is well suited for trench mortars.
-4-
4. E. C. powder - semi-gelatinized nitrocellulose with small amounts of potassium or barium nitrate; used in hand grenades. NH
5. Nitroguanidine (NH3CNHNOs) made from guanidine nitrate, which comes from cyanamide by the addition of ammonia, then subsequent treatment with nitric acid. May be used with a chlorate or perchlorete.
III. The Hi gh Exp 1 o g 1 v e g ( bur sting cha rge s )
These substances detonate under the influence of a suitable primary explosive. They do not function by burning and sometimes are not even combustible. They ere not readily exploded by heat or shock but are extremely brisant once they are detonated. The chief members of this class are the dynamite and liquid oxygen types used for industrial purposes, and the nitro aromatic s and newly developed nitro aliphatic s employed in warfare.
1. NITROGLYCERIN AND DYNAMITE. -~C3H5 (N03 )3 - The trinitrate of glycerine is too sensitive; Nobel, however, discovered dynamite, a solid made by absorbing the liquid in kleeelguhr. The filler now used in the United States
is wood meal and sodium nitrate and acts as a cushion for the explosive during handling. A dynamite cartridge is fired by an electric spark in the sequence of primer mixture, fulminate, end dynamite.
To eliminate the hazard of frozen nitroglycerin, ethylene glycol dinltrate or glycerin dini tromono- chlorohydrin is usually added to lower the freezing point.
2. AMMONITES* — This term induces mixtures of ammonium nitrate with aromatic nitro compounds or combustibles such as sawdust and coal.
dynamon - ammonium nitrate with wood chips, widely
used in Russia . amatol - 20-B0i mixtures of ammonium nitrate and TNT. ammonal - 15# TNT, 17^ Al, 65% NH4N03, Z% C. alumatol - same as above except for 3% Al; used in hand grenades.
3. BLASTINE.— 60# NE4C104, U% TNT, 22# MaN03, 1% wax; used in mines.
4. LOX. — Liquid oxygen absorbed on carbonaceous material contained in a canvas wrapper. Retardents such as diammonium phosphate solution may be used to reduce handling and storage hazards. Used chiefly in open-air mining.
5. CARDOX.-- (liquid C03) in steel cylinders is ignited by a. special electric spark; sometimes used in mining.
~5~
1.25
6. NITROSTARCH. used as filler for hand grenades and mortar shells.
7. NIB - NITROGLYCERINE.— ( nitroisobutylglycerol trinitrate; (KOCHg)3CN03) obtained from CH3N03 and CH30 is nitrated.
8. TETRANITROMETHANE, — from acetic anhydride and nitric acid.
9. NITROGUAM1DINE, NITROCELLULOSE. —listed previously as propellents.
10. TETRACENE (guenyl-nitro saminoguanyl-l-tetracene.) —
This substance ha s a wide range of sensitivity depending upon its density. It varies from a slow burning com- pound to a detonator. However neither tetracene nor its mixtures with potassium chlorate are satisfactory substitutes for mercury fulminate. Commercial prepar- ation:
H3N~C-NHNH3
m
(HNOa)
3 /S
NaNCU v
H30^ HO AC 10°
H3N~C-NH~NH-N=N~C~NH-NH-NO
« ii
NH NH
tetracene
11. HEXOG-EN (trimethylenetrinitramine) (Syn. Cyclonite) Except for Penthrit this is one of the most brisant of all explosives. This high brisance makes it a useful underwater explosive for use in mines and. in depth bombs. For this purpose it is usually mixed with 30-40% of TNT. It can be made quite cheaply because the apparatus required is very similar to that used in the high pressure process for making methanol. Commercial preparation:
6CH30 + 4NK3 ■?► '^ti "?
'CH3 CH3"
W
CH3 CH3 J3HZ
3
Cone .
H exam ethylene tetramine
03N-|}— CHa-N-N08 CH3-N— CH3
tog
Hexogen
12. PENTHRIT fpentr ery thritol tetra nitrate ) This is the most powerful of the known military explosives. Like Hexogen, it has a high brisance which makes it a useful
126
underweter explosive, enC also makes it useful as a bursting charge in artillery shells. It is stable, easily stored, can be prepared from coal, water, air, limestone and sulfuric acid, and hence can still be made by any country even though that country's imports are completely cut off. Commericfl preparation:
CH3CH0 + CHaO Cg(0H.^) C(CHgOH)4 -^ * > C(CH3ONOs)4
Hg SO4
13. PENTRI NIT.— This is the name applied to mixtures of
Penthrit and Nitroglycerol in various ra'tios. Pentrinit 60/40 (60°/ Penthrit ?nd. 40% Nitroglycerol) has the highest rate of detonation of any of the known explosives.
The Nitro Aroma tics
Although several new and very good explosives have been developed in the aliphatic class, the aromatic nitro compounds still constitute the largest ahc! most important class of the high explosives*
1. TNT (2,4.6-trinitrotoluene) (syn. Trotyl, Tolite).-- This is probably the mort widely used of the nilitary high explosives. It is widely used in demolition shells," the TNT constituting as much as 60^ of the weight of the shell.
Commerical preparations A 3-stage direct nitration pro- cess is generally used in the commercial preparation of TNT froro toluene, although 1- and 2-stage processes have been used.
Purification of product: The crude product which separates from the reaction mixture usually contains several impurities, the chief of which are: (3~(2, 3, 4-) , Y~(3,4,6~), and £~(2,3?6~) TNT, TNB (l, 3, 5-trinitro- benzene) and DNT (s, 4-dinitro toluene ) . There may also be impurities from the toluene used. However, this latter class of impurities is usually removed by an initial purification of the toluene. In spite of this, some TNX ( trinitro-m-xylcne ) is usually present in the final product, but in such a small amount that it does no harm. , ■ Some of the DNT is removed by washing with sulfuric acid, and whet little is left causes no trouble. It is not all removed, because too large an amount of , TNT would be dissolved by the sulfuric acid. The ^-TNT is present in too small an amount to cause any trouble; however, the p- and y-THT must be removed. This is accomplished by treating the crude product with sodium "sulfite. The products formed from the p- and Y-TNT are dissolved in water, and then converted; to m-methyl tetryl:
-7-
±2
* t
3H3
P-TNT
3
Na8S03 warm
CHaNH^
»
CH3 1 /CH3 |
|
N°2 CH3 |
|
03N_/V_N03 1 TVH- |
02N
CH3
N0S Y-TNT
.N03 warm
0*N
)
l_S03Ha
CH^NH^
OpN
CHS
/
^0.
NO.
NO
.N
^CH3
\
H
these are H20 sol.
m-methyl tetryl
2. TNB (l, 3, 5-trinitro benzene ) This is the most powerful
explosive among the nitro aromatics, and is less sensitive to impact than TNT.
Commercial preparation: CK3
[
COOH
N03 02N HgSOA
Na2Crs07 40-50°
-CO.
OaN
-/\v-NO
NO,
Although benzene can be nitrated directly to TNB, the process is too expensive of acid and heat to be used commercially.
5, TNX ( trinitro-m-xylene ) This compound has a large excess of carbon in each molecule, so it is used with oxidizing agents such as NH4N03, It is also used to lower the m.p. of compounds such as TNT and PA, However, it attenuates the explosive power of these compounds to some extent. TNX is., prepared by a 3-stage nitration of m- xylene.
' .
■'.
,.
»;•
- •
A
!
'
-
-8-
4. HEXANITROBIPHENYL (2,2,J4,4*,6,6,~) This compound is said to have explosive propertieB superior to those of hexil. It is also non-toxic and very stable chemi- cally. It cannot be made by direct nitration of biphenyl since only the tetranitro compound is obtained. Commercial preparation:
Cu -powder
5. PA (Picric Acid) (Syn. Metinite, lyddite, shimose) This is a very powerful explosive, but has been largely supplanted by TNT, even though the latter is somewhat less explosive. This is because PA forms salts with the common metals (except Sn end Al ) , which are very sensitive to shock. (in fact lead picrate is an excellent detonator. ) Because of this reaction with metals, use of PA in shells necessitates an inside coat- ing of a varnish.
The straight nitration of phenol is not a satis- factory method for preparing PA for two reasons: 1. The -OH group weakens the ring and makes it more sensitive to oxidation, 2. Some p_-nitrophenol is always lost because of its volatility. Therefore PA is made commercially by an indirect method:
CI
A
HNO.
■»
H2S04 40-95°
NsOH
\^
4
autoclave
N0;
0J-/V
'NO,
NO;,
Picric Acid
' " ■ 1
H
.
-9-
129
There is elso a catalytic process which is commercially feasible. This process gives the same overall yields as the above method.
HNO.
Hg(N03):
■>
6. AMMONIUM PICRATE, — This explosive is less sensitive to shock than PA. In fact it is not even detonated by mercury fulminate, and a booster such as compressed PA or Tetryl has to be used to explode it. It is very suitable for use in armor piercing shells, especially for coast guard guns. A mixture of ammonium picrete and picric acid is known as "Explosive D" and is quite widely used in artillery shells in this country. Commercial preparation: PA is suspended in hot water and an excess of strong NH4OH is added. The ammonium picrate separates on cooling..
7. TNA (tetranitroaniline ) This compound ha s a very high velocity of detonation, but it is too reactive chemically to be of much use. It does find Borne use as a booster, however.
Commercial preparation:
N02
NaaS
S°4
NH.
^ HNO a
H3S04'
V 0aN^
^
N08
Y
N03 _N0S
8. TETRYL (l, 3, 5-trinitrophenylmethylnitramine ) probably the most important of the boosters, powerful and more brisant than TNT or PA. Commercial preparation:
This is It is more
CH,~N-H
CH,NH:
CHa-N-NO
HNO.
0aN
H2SO,
. . ■
:;.i.J'
1
t
i
■
■
.
i
•
■
;
■!
.
.
[
.
,.
..
■
-10-
180
9.
HEXIL (2,'2l,4r4,,6l6,-he«s.nitrodiphenylaffllne) This is another important booster like tetryl. It is slightly less sensitive to shock then tetryl, but it has the disadvantage of attacking the skin. It is made commer- cially from aniline and dinitrochlorobenzene:
Gl
+
NO,
V
NO.
Oaggg
60°
OpN
HNO
NO,
N03 NQa ^
NH_
EXPLOSIVE RIVETS
An important industrial use of explosives that has recently been patented is the use of explosive rivets. These are extremely valuable in the manufacture of air- planes, since they can easily be used when the riveter can reach only one side of the rivet.
The explosive rivet consists of a rivet with a hollow shank wherein is placed a very small charge of lead azide. The rivet is then placed in the hole and a riveting iron is applied to its head. This iron heats the rivet, causing the charge to explode within 1.5 to 2.5 seconds. The explosion causes the part of the shank that has passed through the plates to expand, thus lock- ing the rivet firmly in place. This expansion of the rivet can be controlled to within 0.02 inch.
Bibliography
Clark, Ind. Eng. Chem., £5_, 1385 (1933) Desvergnes, Chimie et induetrie, 2.8, 1038 DeWilde, ibid.,,, 30, 1034 (1933) Rinkenbecn and Burton, Array Ordnance, Storm, ibid. . gl, 20 (1941) Denues and Huff, Ind, Eng. Chem., News Ed. Lewis, ibid. . 19, 782 (1941 ) Klrkpatrick, Chem. and Met. Eng., 47, 744 Berl, ibid. , 46, 608 (1939) Woodward, ibid. , 4£, No. 2, 117 (1941 ) Rollend, ibid., 48, No. 6, 96 (l94l) Be.bie, ibid., 48, No. 10, 76 (1941 ) Hardy, ibid. . 49., No. 4, 76 (1942) Cullen, J. Soc, Chem. Ind,, 5£, 812 (1939)
(1932) 12, 120 (1931)
, 18, 1114 (1940) (1940)
i •. i
•
•
:.■
■
!
, . ■- -
■■•'...•' 3
: . j ■':"- ... i* fe.
'■':■. . . v -■
. . '■' " ■.'■■'■ • '■
:■'." ■ ■ ' ' '.
/" .■'■■ s
131
-n~
Marshall, Nitroglycerine and Nitroglycerine Explosives, Naoum,
Williams and Wilkine Co. (1928) Encyclopedia Britannica, Edition 14, Vol. 8 (1929) Riegel, Industrial Chemistry, Reinhold Publishing Corp., N. Y.
(1937) Davis, The Chemistry of Powder and Explosives, Vol. I, Wiley and
Sons, N. Y. (1943 )
Reported by R. S. Voris
P,, F. War field May 6, 1942
^ \ o — ffi — ca
P to O O
CO
«:
CD
Eh
o
o a
CU
o
M Eh <
Eh
W
s
PC
w
4-3 4-3
•H -H
CO O
S to
o
(U
CO
o
£} H
o 3
U rH
CO rH ■P CD CO O O O Ih Sh 4-3 +J •H «H C C
4j +j
c a
T t
rH rH O O 4^> 4-3 ■H -H
C rO
cd o
Tt T
C/J CD <D XJ Q «i oq O 3 O O
h a ^
CU CO o
CO 4-3 C£i o B Cti
CO
w
Eh
K O
>H
o
PQ
o
132
' ■■■
r
■
133
RECENT DEVELOPMENTS IN THE STUDY OF VITAMIN E
Although many p_~hydroxychromans show slight vitamin E activity, it is not possible to modify the structure of cc-tocopherol very much and without loss of part or all of the vitamin E activity . Karrer and coworkers have synthesized homologs of cc-tocopherol in which R (formula 1^ has been modified by addition or subtraction of isoprene units. Also they have synthesized derivatives in which R*,RS and R3 has been varied. A summary of the homologs they have prepared is given in Table I.
R3k\ A ,?-CH,R
In cc-tocopherol, R?',Ra;R3 = CH3
and R = C15R31 "3 isoprene units"
All of these compounds were obtained by condensing the proper hydroquinone derivative with the :-llyl bromide or alcohol.
The general methods used for the synthesis of the hydro- ouinones are as follows:
A
ilOH
The substituted hydroquinone s were all synthesized by one of these methods, with only one exception, that being 2-ethyl- 3,6-dlmethyl— ; D-hydroquinone . In this case, the tri-substituted phenol necessary for the above syntheses could not be obtained readily by synthesis. For this particular case the following procedure was employed:
•2-
lol
H.
HO
"V
CH
CH30
OCH.
OCH.
CH.oOr/
CH^CHOHSN ^OCH
CHoO/^
BrMg
^
CH3
loc»s
CH30
u
CH3
OCH3
HO
ZllB
CH3
0
CH3
OH
An application of Fischer's synthesis of pnytol wrs used for the synthesis of fllyl bromides. For example, l-bromo-3, 7, 11- trimethyldodecene-2 was preprred in the following manner:
OH
CH3-CH(CH3)3CCH3 + HC=CH 'V2 CH3CH(CH3 )3C-C=CH
CH3 0 CH3 0H3
II
III
)H
li3^t CHaCH(CH3)3A~CH=CH3 P-^3 CH3Ch(CH* 3C=CH-CK3Br
cr
JH3 CH3 IV
CH3 CH3 V
135
-3-
CpH
a«s
CH3COCHNaCOOEt P NaNH2
— > CH3CH(CH2)3CH(CK2)3CH(CH2)38cH3 J 1 ^
end reduction
CH3 CH3
VI
OH
H2(-Pt
CH3CH(CH2 )3C-C=CH "3*r \
CH3 CH3 VIII
VII
PBr3
, ) CH3CK(CH2)3CH(CH2)3C=CH-CH2Br
CH3 Cn3 CK3
IX
The l~brorno-3,7-dimethylpentene-2 which ws s needed for the synthesis of the homo log in which R(Formula I) is C5Hn appears in the above scheme. The higher homolog, i.e., the one in which R is C20H41, was obtained from phytol by the same method as above.
One compound containing an unsaturated hetero-cyclic ring has been prepared by Karrer. The method of synthesis here was the condensation of trimethyl hydroquinone with the bromide obtained from the action of PBr3 upon 3,7, 11, 15-tetramethyl- hexadecyn-l-ol-3. The latter was obtained by treating the ketone, resulting from the ozonolysis of phytol with acetylene and sodamide. It also occurs as one of the intermediates in the synthesis of phytol from compound IX above.
A New Method of Synthesis of Tocopherals
Smith and coworkers have devised another method of synthesis of 6-hydroxychromens which does not depend on the condensation of an allyl compound with hydroquinone s. This method is as follows:
..*'.-
'
-4-
136
CH30 CH3
v
CH3
CH3CH3OH
OCH.
-y
ch3o
CH3
CH3 CH3
CK3CH2MgBr CH30
H3 CH3
A
OCH-
"* CH.
Ch3Ch3C— R
V
CH3
OCH
OH
HO CH.
CH3 CH3
This method so far has been applied only to chromrns con- taining methyl groups in the 2,5,7, and 8 positions. By varying the nature of the ketone, five 2, 5,7, 8-tetramethyl-2-alkyl-6- hydroxychromans have been prepared, namely, R=CH3, C2H5, n_-C3H7, i_sp_-C4H9, end C16H33(4,8, 12-trimethyl decyl).
For the preparation of cc-tocopherol, the ketone was obtained from phytol by ozonolysis. The carbinol was prepared from trimethylhydroquinone dimethyl ether by bromination, conversion to the Grignard reagent and treating with ethylene oxide.
From a synthetic point of view, the proof of the structure of cc-tocopherol until now has been based upon the assumption that phytol and its derivatives would behave in the condensation reaction as the simpler analogs do, i.e. to form a chroman ring. Since the new method of synthesis leads unquestionably to a chroman structure, the preparation of cc-tocopherol by this method affords further proof, by synthesis, that the hetero ring in a- tocopherol is a chroman.
Bibliography
Smith and Miller, J. Am. Chem. Soc . , 64. 440 (1942).
Smith and Renfrow, ibid. . 64, 445 (1942).
Karrer and Yap, Helv. chim. acta., 23, 531 (1940); 24, 640 (1941 ).
Karrer and Hoffman, ibid. . 23, 1126 (1940 ).
Karrer and Schlapfer, ibid.., 2_4, 298 (1941).
Reported by F. May 13, 1942
W. Wymen
, i
TABLE I
13?
R
C 2 0^4 1 C 15^3 x
C 15H3 1 C 15H3 1
C 15^3 1 C15H3 1
Cl5^3 1 C15H3 1
C 15H3 1 C15H3 1 C 15H3 1
C 10^2 1
C5H1 1
on 3 |
CH3 |
CH3 |
CH3 |
H |
CH3 |
CH3 |
C2H5 |
CH3 |
n u ^2^-5 |
-,zns |
CH3 |
t~i T T un3 |
R* |
R3 |
Complete Activity |
No. Activity |
|
CH3 |
CH3 |
30 mg. |
3 |
mg-. |
CH3 |
CH3 |
3 mg. |
1 |
mg. |
H |
CH3 |
6 mg . |
||
CH3 |
H |
5-10 mg. |
||
CH3 |
CHa |
8-10 mg. |
< |
|
CHg |
C2H5 |
16 mg. |
||
C2H5 |
CH3 |
|||
CH3 |
riti on 3 |
|||
C2H5 |
CgH5 |
10 mg. |
||
CH3 |
P H |
10 mg. |
||
C2H5 |
H |
10 mg. |
4 |
mg. |
CH3 |
CH3 |
20 |
mg. |
|
CH3 |
CH3 |
40 |
rag. |
138
THE ACY1ATI0N AND ALKYLATION OF THE SODIUM ENOLATES
OF ALIPHATIC ESTERS
Hauser — Duke University
The scope of reactions of the sodium enolates of esters was limited until a base was found which would convert esters largely into their sodium enolates. The Claisen condensation of simple esters, for instance, required the presence of two a loha hydrogen atoms, if the condensing agent used was sodium ethoxide. Two esters qualifying in respect to having two aloha hydrogen atoms but failing to condense were ethyl isovalerate and ethyl t-butyl acetate. Also, esters belonging to the class which has only one alpha hydrogen atom did not condense.
In 1931 Schlenk, Hillemann and Rodlof'f showed that sodium triphenylmethyl converts methyl diphenylaceta te into its sodium enolate, consequently the use of sodium triphenylmethyl has been extended to the condensations, acylations and alkylations of those aliphatic esters requiring a strong base to form the enolate. The enolate is formed according to the following equation:
(C6HB)3CNfj + H-C-COOR -> Na[C~COOR] + (C6H5)3CH
Ethyl isobutyrate, which has only one a loha hydrogen a torn, has been converted to ethyl isobutyrylisobutyrate by sodium tri- phenylmethyl; and ethyl isovalerate, which was noted above to be unaffected by treatment with sodium ethoxide, similarly undergoes self -condensation to give ethyl isovalerylisovalerate :
(CH3)2CHCHsCOOEt +
Na
CHCOOEt
I
CK(CH3)S
_63^
(CH3 )3CHCH3COCHCOOEt CH(Ch3)3
+ NaOEt
Spielman and Schmidt have shown that this same self -condensation also takes place in the presence of mesitylmagnesium bromide, cl though the yield is lower.
The acylation of the sodium enolate of an ester by a different ester is satisfactory only when the latter has no active hydrogen, for the sodium enolate apparently attacks an available active hydrogen of the second ester more readily than the carbonyl group to give a mixture of two enolates and two free esters, from which four beta-keto esters might be formed.
~9_
139
Nitriles having hydrogens alpha to the -C=N group likewise exhibit this hydrogen interchange:
CH3(CH2)2CH2CN + Na[CHaCOOC(CHB)»] -> Na[CH3 (CH2 )2CHCN] +
n-valeronitrile podium enolate of CK3C0OC(CH3 )3
t-butyl acetate
Na[CH3(CH3)2GHCN] + CH3COOC(CH3 );
0 CH3COCHCN
CH2CH2CH3 + NaOC(CK3) cc-acetyl-n-valeronitrile
Ethyl oxalate, having no a lpha hydrogens, reacts easily to give ethoxalyl derivatives, but ethyl formate gives low yields of formyl esters:
COOEt
COOEt + Na[C(CH3)3COOEt] -}'''
sodium enolate of ethyl isobutyrate
COC(CH3)2COOEt
GOOEt
+ NaOEt
HCOOEt + Na[C(CH3)2COOEt]
1&%
HCOC(GH3 )2GOOEt + NaOEt
Hauser hap also carried out e series of acylations with acid chlorides and the sodium enolftes of esters having only one alpha hydrogen atom. This method, in contrast to that using esters as the acylating agent, gave satisfactory yields. The general reaction is as follows, and the results are summarized in Table I:
RC0C1 + Na[CR2C00Et] -> RCOCR2COOEt + NaCl
The a, a-disubstituted beta-keto esters so produced may be sub- jected to ketonic hydrolysis to give certain ketones of the type RCOCHR2. The complete procedure represents an extension of the acetoacetic ester method of synthesis of these ketones. Table II lists the experimental results.
Finally, certain other reactions of the sodium enoli tes of esters might be listed:
110
•3~
1. With phenyl isocyanate the enolate of ethyl isobutyra.te gives cc,a-dimethylmalonanilide ethyl ester:
C6H5N=C=0 + Na[C(CH3)3COOEt] ^ C6H5NHCOC (CH3 )3COOEt
2. With ethyl benzenesulf onate the enolate of ethyl isovalerate gives ethyl cx-e thy li so valerate:
C2H5OS02C6H5 + Na
CHCOOEt CH(CH3)a
33^
C2HBCHCOOEt C
H(CH3)2 + Na.OS02C6H5
This method represents a gain of 11$ in yield over that obtained when the same enolate was alkylated. .with ethyl iodide.
3. With ethyl cc-bromoisobutyrate the enolate of ethyl isobuty- rate gives diethyl tetra.methyleuccinate:
0 CH,
EtOC— C~Br + Na[C(CH3)2COOEt] CI
IH.
30^
0 CK3 CH3
II I I
EtO-C — -C C C-OEt
I I
CHn CHa
I
+ NsBr
4. With ethylene oxide the enolate of ethyl ieobutyrate gives a, a-dimethylbutyrolactone :
CH3
CH2~CH2 + Na[C(CH3)2C00Et] -* NaOCH2CH2C-COOEt
CH,
0
NaOEt +
CH3
;h3ch2c-c=
CH3
•0'
A previous method due to Blanc for production of this compound involved the reduction of cc,a-dimethylsuccinic anhydride, but this procedure also results in the formation of the isomeric P, P-dimethyllaotone.
-4- 14i
Bibliography
Blanc, Bull. soc. chim., 33, 893 (1905)
Hudson and Hruser, J. Am. Chem. Soc, 63. 3156 (1941)
Hudson and Hauser, ibid., 63, 3163 (1941)
Roberts and McElvain, ibid., 59, 2007 (1937)
Schlenk, Hillemann and Rotfloff, Ann., 497, 135 (1931)
Spielman and Schmidt, J. Am. Chem. Soc, 59, 2009 (1937)
Reported by John Whit son May 13, 1942
14^
TABLE I
Results of Experiments with Na[CR3COOC2K5] and RC0C1
Acid
-Ester Used, Ethyls (C6H5 )3CNa Chloride Used
Isobutyrate 23
Mole Mole
0.198 0.195 Acetyl
Yield .of Keto Ester. B.P,
.190 .185 n-Butyrl
.95 .93 Isobutyryl 102
.875 .85 Benzoyl 123
Ieobutyrate 22
Isobutyrate 110
Isobutyrate 102
Methylethyl- 26 .20
acetate
Methylethyl- 26 .20
acetate
Methylethyl- 25.6 .197 .197 Benzoyl
acetate
Diethyl- 28 .20
acetate
0 Mole G % °C M'n
32 0.408 15.7 51 75-6 15
183-4 760
21.5 .20 20.2 58 109-11 29
.95 128.0 74 92-94 15
373 122.0 65 146-8 15
.20 Propionyl 18.5 .20 19.3 52 97-102 15
.20 Isovaleryl 24.6 .20 21.7 51 116-19 15
27.7 .197 24.0 52 164 18
.195 Benzoyl 28 .20 28.8 59 175-77 20
TABLE II Ketonic Hydrolysis of Beta-Keto Esters
p-Keto Ester Used, Ethyl
Ethylbutyryl-
isobutyrate n-Butyryldi-
methylacetate
Propionyldi-
methylacetate
Isoveleryl-
methylethyl-
acetate
Benzoyldi-
methylaceta te
Benzoylmethyl-
ethylacetate
Benzoyldi-
ethylacetate
, — Hydrolysis Mixture, cc.^
CH3- 50f Time, G. H3S04 H20 COOH HP Hr. Ketone
Yield 0 B.P. G % C Mm
14.8 10
21.0 5
14 . 2 4
14.0 8
20.0 10
13.0
15.0
10 30
10 55
4 40
5 38
5 30
4.0 Di-isopropyl 7.0 78 121-25 760
3.5 n-Propyl .. i0.2 79 134-36 760
iso*-propyl 8.0 Ethyl iso- 6.7 78 134-36 760 propyl 14.0 Isobutyl 7.0 75 165-67 760 s- butyl
3.0 Phenyl iso- 13.4 81 102 15
propyl 218 760
75 75 48.0 Phenyl 6.0 69 109 10 s- butyl
75 75 48.0 Diethyl- 3.0 75 117-18 10
acetophenone 247-49 760