1972 R Pettit Vol.

82

The 52-min. component was ethyl cis-(3-cyclobutylacryl- To a solution of 4.5 g. (0.031 mole) of N-cyclobutyl-N-nitro-
ate (XV), m28d 1.4578. The infrared spectrum (chloro- sourea in 100 ml. of ether at —45° was added a solution of
form) had absorption characteristic of an ,/3-unsaturated lithium ethoxide (from 0.5 g. (0.07 g. atom) of lithium) in
ester at 1700 and 1635 cm.-1.16 The n.m.r. spectrum (135 25 ml. of ethanol and 25 ml. of ether. The diazo compound
mg./ml. in carbon tetrachloride, G in Fig. 3) showed cyclo- formed immediately to give a cherry-red solution. The
butyl hydrogen signals from 239 to 293 p.p. 10s ethoxy hy- solution was turbid but homogeneous. Seventy-five ml.
drogen signals at 68 (quartet) and 357 (triplet) p.p. 10s and of the solution was transferred to another flask at —45°.
olefin hydrogen signals at —174, —152, —145, and —125 A solution of 1.5 g. (0.015 mole) of benzaldehyde in 15 ml.
(quartet) and —88 and —58 (doublet) p.p. 10s. of ether was added to this solution, and the color disap-
Anal. Caled, for C9H1402: C, 70.10; , 9.15. Found: peared in 3 min. The remaining solution of diazo compound
C, 70.49; H, 8.86. was allowed to warm to room temperature, by which time
Isolation of a Mixture of Cyclobutyl Ethyl Ether and the color of the diazo compound had disappeared. The
color was still present but faint at 0°.
Cyclopropylcarbinyl Ethyl Ether from a Photochemical Re- Both reactions were washed with 4 portions of 25 ml. of
action of Diazocyclobutane.—In the workup of reaction
mixtures from the irradiation of solutions of diazocyclobu- water, dried over anhydrous calcium chloride and filtered.
tane in ira»s-2-butene-ethanol,17 a number of volatile frac- Twenty ml. of freshly distilled decalin was added to each
tions were obtained. One of these, b.p. 80° (130 mm.) to of the solutions and they were distilled through a semi-
micro spinning band column to give, in addition to an ether
b.p. 110° (25 mm.), when it was further purified by vapor fraction (b.p. 34°), a fraction b.p. 42 to 160°. The weight
chromatography through a didecyl phthalate column (re- of the latter fraction was 0.833 g. for the reaction which in-
tention time 16.5 min. at 70° with a helium flow rate of 60 cluded benzaldehyde and 0.727 g. for the reaction which
cc./min.) was found to have the properties of a mixture of did not.
cyclobutyl ethyl ether and cyclopropylcarbinyl ethyl ether,
n23d 1.4042 (lit.30 »20d 1.4050, b.p. 99.5-100.8° for cyclo- Vapor chromatography of the foreruns through two di-
See [Link] for options on how to legitimately share published articles.

propylcarbinyl ethyl ether). The infrared spectrum (chlo- decyl phthalate columns at 80° showed only ether, at 6.6
minutes. Chromatography of the other fractions at 90°
roform) showed no absorption between 1500 and 2700 cm.-1
or above 3150 cm.-1. There was absorption at 3080, gave the following results. The yields are based on areas
Downloaded via UNIV FED DE SAO CARLOS on July 31, 2025 at [Link] (UTC).

of the peaks.
2970, 2860 and 1115 cm.-1.
Benzaldehyde Thermal
Anal. Caled, for CeHl20: C, 71.95; H, 12.08. Found: Component reacn., min. reacn., min. %
C, 71.74; H, 12.05.
Ether 6.1 17 6.0 23
The n.m.r. spectrum (215 mg./ml. in carbon tetrachlo-
ride, I in Fig. 3) showed a triplet at 353, 371 and 387 p.p. 10s Ethanol 7.4 47 7.3 36
and groups of signals from 114 to 181 p.p. 10s, 273 to 314 p.p. Mixed ethers 28.4,29.2 11 28.0,28.9 13
10s and 431 to 461 p.p. 10s. Decalin 120 25 120 28
A sample of this ether mixture when fractionated through
two didecyl phthalate columns at 70° with a flow rate of 50 The retention times of cyclobutyl and cyclopropylcarbinyl
cc./min. showed two peaks of about equal heights at 45.8 ethyl ethers from the photochemical reaction above at 90°
and 49 min. were 27.8 and 28.8 min. The estimated yields of these
Formation of Cyclobutyl Ethyl Ether and Cyclopropylcar- ethers for the benzaldehyde and thermal reactions were 92
binyl Ethyl Ether in the Preparation of Diazocyclobutane.— mg. (6%) and 94 mg. (6%), respectively.
(30) C. G. Bergstrom and S. Siegel, This Journal, 74, 145 (1952). Urbana, III.

[Contribution from the Departments of Chemistry at the University of Texas and the University of London,
Queen Mary College]

The Synthesis and Properties of the Perinaphthenylium Cation

By R. Pettit
Received September 16, 1959

The synthesis of perinaphthenylium perchlorate is described. This salt is thermodynamically stable; although very re-
active, it can be stored under nitrogen. The salt is insoluble in non-polar solvents though it readily dissolves in nitro-
methane and 60% perchloric acid. The perinaphthenylium cation reacts with water to produce perinaphthene and peri-
naphthenone; it reacts with zinc dust to form peropyrene.

Simple L.C.A.O. molecular orbital calcula- that the gain in ir-electronic energy in deriving
tions111predict that the perinaphthenyl cation each of the above species following loss of a hydride
(la), anion (lb) and radical (Ic) should all possess ion, hydrogen atom or proton from perinaphthene
the same value of the ir-electron delocalization (II) should he 1.7 ß or 34 [Link] These two facts
energy, namely, 5.83/3 (/3 20 keal.). This follows
=
together then predict that the perinaphthenyl
from the fact that like the allyl and triphenyl- cation, radical and anion might display unusual
methyl systems, the perinaphthenyl system is an stability and, in a suitable environment, each may
odd alternant hydrocarbon and the extra one and be capable of more than a transient existence.
two electrons possessed by the radical and anion, The successful preparations of a salt of the
respectively, occupy a molecular orbital of energy perinaphthenylium cation and of the peri-
zero (with reference to the energy of an electron in naphthenyl radical have recently been reported2-3
the pz-orbital of an isolated sp2-hybridized carbon and solutions containing the corresponding anion4
atom). Furthermore, similar calculations show have been made. Even prior to these reports
(1) (a) M. J, S. Dewar and R. Pettit, unpublished results. Other there was considerable evidence supporting the
independent calculations have been reported by V. Gold and F. L.
Tye, J. Chem. Soc., 2184 (1956); (b) based upon a value of 4.13 ß for (2) R. Pettit, Chemistry & Industry, 1306 (1950).
the -electron delocalization energy of l-vinylnaphthalene as given (3) D. H. Reid, ibid., 1504 (1956).
by C. A. Coulson and R. Daudel, “Dictionary of Values of Molecular (4) V. Boekelheide and C. E. Larrabee, This Journal, 72, 1245
Constants.” Vol. Ill, p. 68, (1950).
April 20, 19G0 The Perinaphthenylium Cation 1973

la, A =
-r II
b. A = -

c, A = ·

Hydrolysis of this ester followed by a Curtius

III IV rearrangement of the corresponding acid azide
molecular orbital Thus peri- furnished the isocyanate VII. Unlike cyclohepta-
predictions.
naphthenone (III), was found to be soluble in trienyl isocyanate, which can readily yield tro-
strong aqueous acids,5 presumably due to the pylium salts,11 the isocyanate VII showed no
formation of the hydroxy derivative of the peri- potentially ionic properties at all, and attempts to
remove an isocyanate ion from this molecule failed.
naphthenylium cation, a reaction entirely anal-
ogous to that of tropone with acids to form hy- Hydrolysis of the isocyanate afforded the hydro-
chloride salt of 3'-amino-1:2-cyclopropanoace-
droxytropylium salts.6 Ciar had also accumulated
evidence showing that several derivatives of the naphthene (VIII). It was hoped that treatment
of this amine with nitrous acid would furnish the
perinaphthyl radical were unusually stable.7 Boek-
elheide and Larrabee4 also showed that peri- alcohol IV (R =
OH) and that this could yield
the perinaphthenyl cation upon treatment with
naphthene (II) reacted with phenyllithium and
with triphenylmethyl to give presumably the strong acids. However reaction of the amine
anion lb and the radical Ic, respectively. hydrochloride with sodium nitrite in dilute acid
The present paper is concerned with the prepara- yielded only high molecular weight material.
tion and properties of the perchlorate salt of the Nevertheless, when the reaction was carried out in
concentrated hydrochloric acid, 3'-chloro-1:2-cy-
perinaphthenylium cation; a preliminary report
of some of this work has appeared earlier.2 A clopropanoacenaphthene (IX) was formed in fair
stable tribenz-derivative8 and solutions containing yield. This very fortunate result was a little
an indeno-derivative9 of the parent cation la unexpected; the most reasonable explanation
seems to be that an intermediate diazonium chlo-
have also been reported.
At the outset it seemed that the most suitable ride ion pair is formed in the concentrated hydro-
material which could be used for ascertaining the chloric acid and that this undergoes an Sní type
reaction. There can be no free carbonium ion
potential stability of the perinaphthenylium cation
would be a halogen derivative of perinaphthene type intermediate involved, for, as will be seen,
where the halogen atom was attached to the this would surely undergo immediate electronic re-
saturated carbon. However Boekelheide and Gold- arrangement to the more stable perinaphthenylium
man had previously tried to synthesize the bromo cation. The formation of the chloride IX was
derivative of II without apparent success.10 The fortunate in that it allowed the perinaphthenylium
cation to be formed under anhydrous conditions
synthesis we therefore chose to follow was one
similar to that employed by Dewar and Pettit11 and this greatly facilitated the isolation of a salt of
for their synthesis oí the tropylium cation and the cation.
aimed at preparing a molecule such as IV. It was The chloro compound IX showed no ionic proper-
hoped that if the group R in a molecule such as IV ties, it was readily soluble in non-polar solvents
could be removed as an anion, a cationic system and gave no immediate reaction with silver nitrate.
would remain which could be reasonably expected It also showed no unsaturated behavior and the
to give the perinaphthenylium cation upon elec- molecule undoubtedly possessed the covalent
tronic rearrangement. structure IX. The ultraviolet spectrum was very
The final successful synthesis is outlined in the similar to that of the carboxylic acid IV (R =

accompanying scheme. Diazoacetic ester was COOH) and to that of acenaphthene itself, show-
treated with acenaphthylene in decalin to give 3'- ing that no rearrangement had occurred. At-
carbethoxy-1:2-cyclopropanoacenaphthene (VI). tempts to prepare the corresponding bromo and
iodo derivatives directly, utilizing a Hunsdiecker
(5) J. W. Cook and C. L. Hewett, J. Chem. Soc., 365 (1934).
(6) W. von E. Doering and F. L. Detert, This Journal, 73, 876
reaction on the acid IV (R =
COOH), failed.
(1951). However, treatment of the chloro compound
(7) E. Ciar, “Aromatische Kohlenwasserstoffe,” 2nd Ed., Springer- with silver perchlorate in dry nitromethane at 70°
Verlag, Berlin, 1952, p. 431. for two hours did result in almost quantitative
(8) E. Ciar and D. G. Stewart, J. Chem. Soc., 23 (1958).
(9) D. H. Reid, Angew. Chem., 67, 761 (1955). precipitation of silver chloride. Addition of ether
(10) V. Boekelheide and M. Goldman, This Journal, 76, 604 to the remaining deep green solution caused pre-
(1954). These workers isolated a yellow solid which underwent rapid
decomposition. In view of the very facile decomposition of perinaph-
cipitation in good yield of a yellow microcrystalline
solid which, from the following properties, was
thenylium perchlorate it seems possible that the yellow solid above may
have contained the perinaphthenylium cation. shown to be the required perinaphthenylium per-
(11) M. J. S. Dewar and R. Pettit, J. Chem. Soc., 2021 (1956). chlorate.
1974 R. Pettit Vo], 82

The yellow solid contained the perchlorate anion, ture was cooled, stirred with acetone (400 ml.) and the large
it readily decomposed in moist air but could be amount of polymeric material which precipitated was fil-
tered off. The acetone was evaporated from the filtrate and
stored at room temperature under nitrogen with- the crude ester VI was stirred vigorously under reflux with
out apparent change. It was insoluble in such 10% XaOH solution (1000 ml.). The mixture was cooled
non-polar solvents as benzene, chloroform and to 5° and the slimy crystals of the sodium salt of the acid
ether though it readily dissolved in nitromethane. were filtered off. The salt was dissolved in hot water (500
ml.) and treated several times with charcoal until the fil-
Perinaphthenylium perchlorate readily dissolved in trate appeared dear. The free acid (62 g., 20%) was ob-
60% aqueous perchloric acid to give stable yellow tained by treating the filtrate with excess HC1 solution. The
solutions; but on dilution the salt decomposed acid crystallized from ethanol in white plates, m.p. 199-
and perinaphthene (II) and perinaphthenone (III) 201°. The material did not readily decolorize permangan-
ate solution; ultraviolet spectrum: max. µ (log e) 2SS
were isolated from the decomposition products. (3.95), 300 (4.10), 314 (3.95), 320 (3.63); min. µ (log e)
These compounds probably arose following dis- 258 (3.22), 292 (3.90), 210 (3.83), 319 (3.58).
proportionation of diperiuaphthenyl ether or of Anal. Caled, for ChHio02: C, 79.98; 11,4.79. Found:
perinaphthenol. Whatever their mechanism of C, 79.59; H.4.89.
formation, this decomposition has prevented the The picrate derivative formed orange plates from alcohol;
estimation of the pKa value of the perinaph- m.p. 158°.
Anal. Caled, for C20HlsN3O9: C, 54.67; H, 2.98; N,
thenylium cation. The lowest value of the atom 9.5/. Found: C, 55.21; , 3.06; N, 9.76.
localization energy of the perinaphthenylium cation Curtius Rearrangement of Acid IV (R =
COOH).—(a)
is 1.7 ß, which would thus predict that the pKa The above acid (30 g.) was refluxed for 1.5 hours with thi-
value of the cation should be somewhat less than onyl chloride (20 g.) in dry benzene (1.80 ml.) containing six
that for the tropylium and benztropylium cations drops of pyridine. The solvent was removed and the re-
but greater than that for 1:2, 4:5-dibenztropylium sidual acid chloride crystallized from petroleum ether.
Treatment with ammonia afforded the amide derivative as
cation.13 white needles from alcohol, m.p. 248°.
The ultraviolet spectrum of the salt taken in 60% Anal. Caled, for C14HnNO: C, 80.36; , 5.30; N,
H2SO4 was closely similar to that of perinaph- 6.69. Found: C, 79.99; H, 5.29; N, 6.80.
thenone (III) in the same solvent, confirming the (b) To the acid chloride (26 g.) dissolved in acetone (150
idea that the latter solutions contain the hydroxy ml.) a solution of sodium azide (8.5 g.) in water (15 ml.) was
derivative of the perinaphthyl cation. added dropwise with vigorous stirring. The temperature
was kept below 0° during the addition and the mixture was
In an attempt to prepare methylperinaphthene, then stirred for a further 20 minutes. Water (300 ml.) was
perinaphthenylium perchlorate was treated with then added and the acid azide which separated was extracted
both methyllithium and methylmagnesium bro- with benzene and the solution dried over sodium sulfate.
Most of the benzene was removed under reduced pressure,
mide, but although a reaction appeared to take dry benzene (300 ml.) added and the solution then refluxed
place, in neither case could an identifiable product for 2.5 hours. The solvent was concentrated to a volume
be isolated. In view of the ready rearrangements of about 50 ml., concentrated hydrochloric acid (150 ml.,
which seem to occur in the series of methylperi- 35%) was added to the residue and the mixture then was
stirred vigorously under reflux for 45 minutes. On cooling,
naphthenes13 it would have been interesting to find the amine hydrochloride VIII (20.5 g.) separated as light
which isomer was formed in such a reaction of the brown plates. A sample was recrystallized from dilute hy-
cation. drochloric acid.
Peropyrene was isolated following reaction of Anal. Caled, for Ci3Hl2NC¡: C, 71.72; H, 5.53; N,
perinaphthylium perchlorate with zinc dust. This 6.44. Found: C, 71.44; H, 5.59; N, 6.20.
hydrocarbon is also formed when the perinaphthenyl 3'-Chloro-1:2-cyclopropanoacenaphthene (IX).—The
above amine hydrochloride (10.0 g.) was dissolved in a mix-
radical3 is similarly treated and in the former reac- ture of concentrated hydrochloric acid (80 ml.) and glacial
tion the same radical is presumably also formed as acetic acid (40 ml.). Ether (200 ml.) was added, the mix-
an intermediate following a one electron transfer ture cooled to —5° and a solution of NaNO- (4.2 g.) in water
from the metal to the cation. (20 ml.) then added dropwise with vigorous stirring. After
15 minutes the mixture was stirred at room temperature for
Attempts have been made to isolate other salts of a further 30 minutes. Water was added and the mixture ex-
the perinaphthenylium cation, but due largely to tracted several times with ether. The combined ether lay-
the instability of the cation in water these have not ers were washed first with water then with dilute ammonia
been successful. Nitrobenzene solutions of the solution. The crude product obtained after removal of the
chloro compound IX, when treated with aluminum solvent was adsorbed on alumina and eluted with light pe-
troleum ether. The chloro compound (5.0 g.) was thus ob-
chloride, became deep green in color and also be- tained as colorless needles, m.p. 122-123°.
came conducting, suggesting that a salt was formed, Anal. Caled, for CI3HSC1: C, 77.8; H, 4.5; Cl, 17.7.
but no solid could be isolated. Also when treated Found: C, 77.7; H, 4.8; Cl, 17.5.
with boron trifluoride in ether, solutions of the The chloro derivative is very soluble in benzene, hexane
chloride acquired the green color typical of the and chloroform; it gives no immediate reaction with etha-
cation la, but again no solid salt could be isolated. nolic silver nitrate; ultraviolet spectrum: max. (log e)
286 (3.92), 298 (4.00), 311 (3.88); min. (log e) 256 (3.34);
290(3.88), 308(3.72).
Experimental Perinaphthenylium Perchlorate.—3'-Chloro-1:2-cyclopro-
1:2-CycIopropanoacenaphthene-3'-carboxylic Acid (IV, panoacenaphthene (1.1 g.) was dissolved in dry nitrometh-
R =
COOH).—To a mixture of acenaphthylene (250 g.) ane (30 ml.) and silver perchlorate (3.0 g.) added. The
and decalin (200 ml.) stirred at 140°, ethyl diazoacetate (130 solution was kept at 70° for two hours, with occasional
g.) was added dropwise over a period of 3.25 hours. Heat- shaking. Silver chloride slowly precipitated during this
time and the solution gradually became dark green. The
ing and stirring were continued until evolution of gas ceased.
(In a typical run about 20 1. of gas was collected.) The mix- silver chloride (0.79 g. 90%) was then filtered off. Dry
ether (250 ml.) was added to the filtrate. The yellow mi-
(12) D. Metiche, H. Strauss and E. Heilbronner, Helv. Chim. Acta, cro-crystalline precipitate of perinaphthenylium perchlorate
41, 57 (1958). which appeared was collected, washed several times with
(13) V. Boekelheide and C. E. Larrabee, This Journal, 72, 1240 dry ether and, while still wet, transferred under nitrogen to
(1950). a weighed flask. The excess ether was removed under vac-
April 20, 1960 0-PlIENYLENEDIAMlNES WITH DlBENZYL KETONES 1975

uum. The product weighed 1.2 g. (84%). The material The yellow benzene eluate gave yellow needles of peri-
can be successfully crystallized from warm nitromethane if naphthenone (0.17 g.) on evaporation, m.p. 156°. This
the operation is carried out rapidly and in an inert atmos- material dissolved in concentrated hydrochloric acid and
phere; ultraviolet spectrum; max. (log e) 226 (4.51), 400 could be recovered unchanged on dilution.
(4.67); min. 308 (2.67). There also occurs a shoulder at Anal. Caled, for Ci3H80: C, 86.7; H, 4.4. Found:
378 µ (4.21) and the spectrum “tails off” very slowly; C, 86.7; H, 4.2.
the log e value is still 2.56 at 540 µ (in which area the curve
also appears to have a broad shoulder). (b) Zinc Dust.—A freshly prepared sample of the salt
(approx. 0.50 g.) was stirred in ether (50 ml.) with zinc dust
Anal. Caled, for C13H9CIO4: CIO4, 37.6. Found: (5.0 g.) for 20 hours. The ether was evaporated and the
C104, 37.3,38.0. residue extracted with boiling benzene (3 X 50 ml.). Evap-
Deep green solutions were also formed when IX was oration of the benzene afforded crude peropyrene (0.14 g.).
treated with aluminum trichloride in nitrobenzene and with The compound was purified by sublimation followed by re-
boron trifluoride in ether. Although the development of crystallization from toluene. The substance dissolved in
color probably indicated presence of the perinaphthenylium coned, sulfuric acid to give blue-green solutions. The ul-
cation, only intractable materials could be isolated from traviolet spectrum in alcohol showed peaks at 230, 264, 277,
these reactions. 310, 324, 366, 388, 410 and 437 µ, in agreement with the
The perchlorate salt was insoluble in all non-polar sol- reported values.18
vents; it dissolved in 60% aqueous perchloric acid, and in Attempted Hunsdiecker Reaction on the Acid IV, R =

35% hydrochloric acid, though the latter solutions became COOH.—Bromine (3.1 g.) in dry benzene (30 ml.) was added
turbid on standing and both did so on dilution. dropwise to a stirred suspension of the silver salt of the acid
Reaction of Perinaphthenylium Perchlorate with: (a) (6.3 g.) in dry benzene (200 ml.). The mixture was stirred
Water.—The salt (0.47 g.) was added with stirring to a mix- and refluxed for 3 hours. The silver bromide was collected.
ture of water (20 ml.) and ethanol (20 ml.). After 10 min- The residue after evaporation of the benzene was almost com-
utes the mixture was extracted with ether, the ethereal solu- pletely soluble in sodium hydroxide. It crystallized from
tion then washed and dried. After evaporation of the ethanol in white prisms, m.p. 220°.
ether the residue was chromatographed over alumina, elut- Anal. Caled, for C14H9Br02: C, 58.1; , 3.1; Br, 27.7.
ing first with petroleum ether followed by benzene. The Found: C, 57.9; H.3.3; Br,27.2.
colorless first fraction on evaporation yielded perinaphthene
which crystallized from petroleum ether in white needles, The same compound was formed when the acid was
treated with bromine in acetic acid. The material would
m.p. 84°. From alcohol, the compound formed an orange- thus appear to be the 5-bromo derivative of the acid IV (R =
red picrate, m.p. 120° or 205° depending on the rate of
heating.14 The ultraviolet spectrum of the hydrocarbon COOH) though no further work was done on this com-
was identical to that previously reported.15 pound.
Only polymeric material was formed when iodine was
(14) G. Lock and G. Gergely {Ber., 77, 461 (1944)) report a value substituted for bromine in the above reaction.
of approximately 110° for the m.p. of this picrate, while V. Boekel- (16) See ref. 7, p. 362.
heide and C. E. Larrabee (see ref. 4) report a value of 205-207°.
(15) See ref. 7, p. 429. Austin, Texas

[Contribution from tiie Department of Chemistry of the University of Michigan]

The Reaction of o-Phenylenediamines with Ketones. V. Further Studies with
Dibenzyl Ketones1,2
By Robert C. Elderfield and Kenneth L. Burgess3
Received July 6, 1959

The thermal decomposition of a series of 2-benzyl-2-substituted benzylbenzimidazolines has been investigated. Analysis
of the products by vapor phase chromatography or mass spectroscopy with the aid of C13 has shown conclusively that the
decomposition is not subject to the Hammett equation as previously indicated on the basis of preliminary evidence. A
definitive comparison of the accuracy of vapor phase chromatography versus use of C13 as a tracer in the analysis of mixtures
of toluenes has been made. A critical study of optimum methods for the synthesis of a series of substituted phenylacetic
acids and of a series of l,3-diphenyl-2-propanones has been made.

In the preceding paper of this series4 a study of

the thermal cleavage of certain 2,2-dibenzylben- nh2 ,CH2CeHs
zimidazolines was reported. In particular, atten- + o=c -11,0 /CH2C6H5
tion was devoted to the decomposition of products \\h2
CH2CJ14X N \;.h2c6H4X
resulting from condensation of o-phenylenediamine II
with unsymmetrical dibenzyl ketones as typified H hi
by I-IV or I-V.5 H
(1) The material presented in this paper is taken from a dissertation i\
submitted by Kenneth L. Burgess in partial fulfillment of the require- /
\
ments for the degree of Doctor of Philosophy in the University of
x ¿*CCH2C„H.5 + CH3C6H4X
Michigan, June, 1957. N IV
(2) This investigation was supported in part by a contribution from
Parke, Davis and Co., Inc., to the Michigan Memorial Pheonix Project. or
(3) Procter and Gamble Fellow, 1955-1956.
(4) R. C. Elderfield and V. B. Meyer, This Journal, 76, 1887 II
(1954).
(5) The structure of the condensation products of o-phenylenedi-
amine and the dibenzyl ketones is arbitrarily represented by III for - ch3c6h5
convenience. This is probably an oversimplification. For a detailed [^JJ^ ^CCH2C6H4X
discussion of the tautomerism involved see ref. 4. V