3168 Wyman R. Vaughan and Randolph Perry, Jr. Vol.

75

at 10 mm. and afforded 2.7 g. (35%) of racemic camphene. b.p. 156°, for 5.5 hours.The reaction mixture was treated
The other data are given in Table I. asabove, and afforded 4.1 g. (56%) of sublimed camphene.
In another experiment, 7.3 g. of camphene-8-C14 was The other data obtained in this experiment are given in
heated with 1.5 ml. of pyruvic acid and 30 ml. of acetoni- Table I.
trile in a sealed tube immersed in refluxing bromobenzene, Cambridge 39, Mass.

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

Racemization in the Camphene Hydrochloride-Isobomyl Chloride Rearrangement1 2

By Wyman R. Vaughan and Randolph Perry, Jr.

Received January 3, 1953

A new route to optically active camphenilone is described. By means of C14-labeled methyl iodide, the optically active
camphenilone is converted into methylcamphenilol and thence by the xanthate dehydration to optically active camphene
labeled in the 8-position. Racemization by standard rearrangement procedures affords camphene which is shown by ozonoly-
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

sis and analysis of the resulting dimethylnorcampholide to have the isotopic carbon statistically distributed among the 8-, 9-
and 10-positions. It is concluded that the racemization occurs in part by migration of one of the gem-methyl groups in accord
with the Nametkin hypothesis and in part by an alternative process not involving methyl migration.
Downloaded via UNIV OF DELHI on September 23, 2019 at 13:33:47 (UTC).

Until comparatively recently “optically pure” using ( + )-tartaric acid and (+)-camphoric acid.
camphene has not been available for study and, of Since both resolving agents were subject to the
the published routes to it,3·4 neither admits of same objection (the fully resolved salts are all but
preparation of the optically pure substance labeled optically inactive), the resolution involved recrys-
in the 8-position. Inasmuch as it seemed desirable tallization until the specific rotation of the salt
to minimize the operations subsequent to intro- fell below an arbitrary minimum. Conversion to
duction of C14, it was decided to develop an entirely camphenilone of several such samples established
new route to optically active camphenilone which the acceptability of this procedure inasmuch as
then could be readily converted in one or two steps samples of this nature afforded camphenilone com-
to the desired optically active labeled camphene. parable with that of the highest reported optical
The preparation of a suitable starting material, purity and subsequently yielded camphene of
2,2-dimethyl-3,6-ewdo-methanocyclohexyl methyl correspondingly high optical purity.
ketone (I), has already been described.1·5 Con- In passing from camphenilone, [o:]26d 76.1° (ben-
version of I into N-eamphenilylacetamide was zene), to camphene, [ ]26 109.0° (benzene), use was
made of the Grignard reaction, using ordinary
methyl iodide. The methylcamphenilol thus pro-
duced was dehydrated to camphene by means of the
xanthate ester procedure which obviates racemiza-
tion at this critical stage. Slightly less “active”
camphenilone, [a]26o 62.4° (benzene) was used with
C14 Grignard to give slightly less active camphene,
achieved by means of the Schmidt reaction using [a]26o 103.5° (benzene).
sodium azide in concentrated hydrochloric acid. It is of interest to record that it may not be
Owing to the hygroscopic character of camphenilyl- necessary to isolate the alcohol from the Grignard
amine and its tendency to absorb carbon dioxide reaction when it is intended for subsequent xan-
rapidly, it was never handled as the free base but thate formation. If a higher boiling ether is
was identified by comparison of the infrared spec-
employed, the Grignard reaction mixture may be
trum of its N-benzoyl derivative and that of an treated with carbon disulfide and excess methyl
authentic sample. iodide to give the methyl xanthate directly without
In exploratory work the racemic amine was con- isolation of the alcohol and subsequent time-
verted directly to camphenilone in excellent yield consuming conversion to the potassium alkoxide.
by means of a modification of the Sommelet re- Unfortunately time did not permit full exploration
action in which hexamine and formaldehyde are of this aspect of the problem, but work is continuing
used to convert a primary amine to a carbonyl on this modification of the original Chugaev re-
function.6 action.
Subsequently camphenilylamine was resolved Once the optically active 8-labeled camphene was
(1) Presented in part before the Division of Organic Chemistry, obtained, it was racemized by a procedure worked
121st Meeting, American Chemical Society, Buffalo, N. Y., March 26, out by Meerwein and Montfort.7 To provide
1952.
(2) Abstracted from a portion of the Ph.D. dissertation of Randolph bulk in subsequent operations, the labeled cam-
Perry, Jr., University of Michigan, 1953. phene was “diluted” with a known quantity of pure
(3) XV. Hückel, W. Dull, S. Eskola and H. Weidner, Amh., 649, 199 racemic camphene and then was dissolved in ether
(1941), and converted to camphene hydrochloride with
(4) J. P. Bain, A. H. Best, B. L. Hampton, G. A. Hawkins and L. J.
Kitchen, This Journal, 72, 3124 (1950). dry hydrogen chloride at 0°. The crude hydro-
(5) W. R. Vaughan and R. Perry, Jr., ibid., 74, 5355 (1952). chloride was dissolved in cresol, and a portion was
(6) J. Graymore and D. R. Davies, J. Chent. Soc., 293 (1945);
S, J. Angyal, P. J. Morris, J. R. Tetaz and J. G. Wilson, ibid., 2141
placed in a 1-dm. polarimeter tube where the loss
(1950), (7) H. Meerwein and F. Montfort, Ann., 436, 207 (1923).
July 5,1953 Racemization in Camphene Hydrochloride-Isobornyl Chloride Rearrangement 3169

of optical activity was followed until the solution lactone, dimethylnorcampholide (II), are produced
became too dark to permit further observations. in about equal quantities.11
After 163 min. there was 16.6% residual rotation;
thus to assure complete racemization the solution
was allowed to stand for 10.75 hr. and then was
worked up to give racemic isobornyl chloride which
was reconverted to camphene by treatment with
sodium ethoxide.7 The resulting camphene was
purified and proved to be fully racemic. The notoriously high cryoscopic constant of
Two reasonable paths for racemization would
seem to be available: the first was proposed as a camphenilone coupled with its relatively low melt-
consequence of Nametkin’s study of the behavior ing point and high volatility render it a difficult
of 1-substituted camphenes.8 It is pictured (with- substance for small-scale study. Consequently
out regard for the concerted nature of rearrange- it was decided to isolate and purify the more tract-
able II as the major fragment from the ozonolysis,
ments) most simply as a shift of the endo methyl since it, like camphenilone, contains the two methyl
of the “camphenehydro” ion
groups of camphene into which the C14 may have
become distributed. The operations and results
duplicate the literature11 findings exactly, and thus
it was possible to obtain a suitably pure sample of
II for combustion.
Making suitable correction for the “dilution”
of the labeled sample as described above the
The second path embodies a hydride shift in the following data are obtained: Total radioactivity of
rearrangement stage or subsequent to it (again original sample (“standard” run on methyl-
without regard for the concerted nature of re- camphenilol) 2.34 X 108 c./min./mole; radioactiv-
arrangements). ity of dimethylnorcampholide (prepared from
racemized camphene, itself obtained from active
camphene prepared from the standard sample of
methylcamphenilol) 1.55 X 108 c./min./mole;
% radioactivity retained in dimethylnorcampholide
1.55/2.34 X 100 66.3%.
=

Thus 2/3 of the original radioactivity is retained in

Evidence in favor of the latter is adducible from II after the racemization. Accordingly the C14
the work of Ritter and Vlases,9 who report the in- originally at position 8 must have become statis-
activation of 8-substituted camphenes. Namet- tically distributed among positions 8, 9 and 10.
kin’s findings taken with those of Ritter and
Vlases suggest the possibility of concurrent modes of
rearrangement, since the two inactivations are
individually incompatible.
By labeling carbon-8 it becomes possible to con-
sider the two paths without recourse to camphene
derivatives, which it might be argued can behave
differently from camphene itself. The Nametkin Distribution of C14 among these three positions
hypothesis predicts formation of a camphene which abundantly indicates the necessity of involving
will retain a portion of its initial radioactivity in both racemization processes, since the trans migra-
the isopropyhdene bridge: 50% if one methyl tion principle in the Wagner-Meerwein rearrange-
is involved, as demanded by the stereospecificity10 ment admits only of migration of the endo methyl
of the general Wagner-Meerwein rearrangement. of the isopropyhdene bridge in camphene hydro-
The alternative hydride shift will not affect the chloride.
status of the original methylene group, which will The question of configuration in camphene hydro-
retain 100% of the radioactivity at the 8-position, chloride is not of serious moment inasmuch as the
if the Nametkin racemization is not involved. rearrangement to isobornyl chloride requires the
Thus ozonolysis will afford a radioactive dimethyl- chlorine to be exo; and even were there appreciable
norcampholide (and camphenilone) only if the quantities of the endo isomer present, it would be
methyl shift mechanism is involved, and retention converted to the more stable exo form, since both
of more than 50% of the initial radioactivity would methylcamphenilol (exo) and camphene hydrate
imply concurrent involvement of the alternative (endo) afford the same camphene hydrochloride
hydride shift mechanism. with hydrogen chloride.12
In the ozonolysis reaction camphenilone and the Without examination of partially racemized
(8) For a brief discussion and leading references see Sir John camphenes it is not possible to determine to what
Simonsen, “The Terpenes,” Vol. II, 2nd ed., Cambridge University extent methyl and hydrogen migration contribute;
Press, Cambridge, 1949, pp. 290-293. however, the near-statistical distribution suggests
(9) J. J. Ritter and G. .Vlases, Jr., This Journal, 64, 583 (1942); both are appreciably involved, which is in keeping
Cf. L. A. Subluskey and L. C. King, ibid., 78, 2647 (1951).
(10) For extensive discussion of the mechanism of 1,2-shifts, cf.
with Nametkin’s results on the one hand and
G. W. Wheland, “Advanced Organic Chemistry,” 2nd ed., John Wiley (11) C. Harries and Baron J. Palmén, Ber., 43, 1432 (1910).
and Sons, Inc., New York. N. Y., 1949, Ch. 12, esp. pp. 529-534. (12) O. Aschan, Ann., 383, 65 (1911); 410, 272 (1915).
3170 Wyman R. Vaughan and Randolph Perry, Jr. Vol. 75

Ritter’s and Vlases’ on the other, both investiga- ous reaction occurred. The volatile product was then
tions affording ample quantities of product re- slowly distilled into 10% hydrochloric acid until the tem-
perature of the reaction mixture reached 198°. Steam dis-
sulting from the particular process being established. tillation into fresh hydrochloric acid then removed all
The various isotopic antipodes and the relation- traces of the amine. The hydrochloric acid solutions were
ships between them are illustrated in Chart I combined and extracted with ether, made alkaline with so-
where (+)-A is arbitrarily taken to be the initial dium hydroxide and the liberated amine was extracted into
ether (87% yield by assay of an aliquot). The amine is
camphene-8-C14. strongly basic, as previously reported,15 and it absorbs car-
Chart I bon dioxide from the atmosphere rapidly. Thus it is best
stored as the hydrochloride which can be precipitated from
the dried ethereal solution as indicated in the preceding ex-
periment. If resolution by means of (+ )-tartaric acid is
desired, the ethereal solution of the racemate may be con-
verted into the diastereomeric tartrates by careful addition
of the ether solution to an alcoholic solution of ( + )-tartaric
acid.
Several derivatives were prepared. The benzamide,
m.p. 153.0-154.0° cor., melted somewhat higher than pre-
viously reported15·17 and a mixed melting point with a sample
(m.p. 148.5-152.0°)17 very kindly furnished by Dr. Noyce
melted at 150.0-151.0°. However, infrared absorption
curves for the present product and for Dr. Noyce’s sample
are identical. The picrate was prepared: m.p. 215.5-
216.0° dec. Previously the value was reported as 205-
207° for a picrate prepared from a mixture of the present
It is apparent that until at least one methyl migra- endo with the exo form of the amine.18
tion (e.g., A -* F or B E) has occurred, no Anal. Caled, for C15H20N4O7: C, 48.91; H, 5.47; N,
change in the distribution of C14 is possible although 15.12. Found16: C, 48.87; H, 5.28; N, 15.03.
racemization can occur (e.g., A -* B). Resolution of Camphenilylamine.—An ether solution of
36.1 g. (0.259 mole) of camphenilylamine was added slowly
Acknowledgment.—The senior author wishes to and with good stirring to a dilute solution of 45 g. (0.30
express his thanks to the Board of Governors of mole) of (+ )-tartaric acid in 90% ethanol. The white
the Horace H. Rackham School of Graduate solid (dec. 200-202°) which separated immediately was re-
Studies of the University of Michigan for the moved by filtration and dissolved in the minimum amount
of water at 80°. After standing for 2 to 3 hours at room
award of a Summer Faculty Research Fellowship temperature, the needle-like crystals were filtered off and
which enabled him to initiate this study and to the recrystallized from fresh water as before. The so-called
Michigan Memorial Phoenix Project for a grant to triangular scheme19 was carried out, and whenever the spe-
cific rotation of a given crop was found to be á +3°, it was
purchase counting equipment. removed from the pattern and consolidated with all other
material meeting this specification.
Experimental13 A sample of partially resolved tartrate, 13.94 g. (0.0482
N-Camphenilylacetamide.—The Schmidt reaction14 was mole), was converted into the (-f)-hydrogen camphorate
carried out on camphenilyl methyl ketone (I), 7.48 g. and the same method was then followed for further resolv-
(0.045 mole) of which was dissolved in 150 ml. of absolute ing the mixture, dioxane being used as the solvent. The
methanol previously saturated at room temperature with specific rotation was found to approach a minimum close to
dry hydrogen chloride. To this solution was added 4.0 g. 0°, and consequently the camphorate offered no advantage
(0.06 mole) of sodium azide in small portions over a period over the tartrate. Figures are given in Table I for compara-
of 6 hr. at room temperature. After standing overnight tive purposes.
there was added excess 10% aqueous alkali and the mixture Camphenilone.—(The following procedure was used for
was steam distilled. The distillate contained some un- preparation of optically active samples of camphenilone from
changed ketone and some camphenilylamine (from 6 to the salts used in resolution. Where the racemic compound
was used for study of conditions, the hydrochloride was
12%) which was recovered by ether extraction. The amine
was isolated from the ethereal solution, after drying over neutralized directly.) Camphenilylamine was liberated
magnesium sulfate, by addition of dry hydrogen chloride from 13.0 g. (0.045 mole) of the tartrate by the addition of
which caused the amine hydrochloride to precipitate. The excess alkali. It was taken up in ether from which it was
major portion of the rearrangement, however, afforded N- extracted by a threefold excess of 10% hydrochloric acid in
camphenilylacetamide which was recovered from the basic three equal portions. The acidic extracts were combined
distillant by cooling and filtering. Recrystallization from and neutralized to phenolphthalein with coned, alkali.
aqueous methanol yielded 5.22 g. (64%), m.p. 140.2- After the addition of 5-10 drops of glacial acetic acid and
141.2°. An alternative procedure in which 55 ml. of or- 5.15 g. (0.060 mole) of 37% aqueous formaldehyde8 the mix-
dinary concentrated hydrochloric acid is used as the reac- ture was chilled well in ice and shaken vigorously for 10-15
tion medium for 0.05 mole of ketone afforded a somewhat min. during which time a white solid separated. To the
lower yield of N-camphenilylacetamide, m.p. 143.0-143.5°. resulting slurry there was added 40 g. (0.029 mole) of hex-
Both melting points are rather higher than the previously amine and glacial acetic acid sufficient to bring the acetic
reported value of 135-136°.16 The microanalysis was car- acid concentration to about 50%. The mixture was then
ried out on the higher melting sample inasmuch as recrys- refluxed for ca. 1 hr., after which the flask was chilled and
tallization failed to raise the melting point. the contents made alkaline with coned, sodium hydroxide.
Anal. Caled, for CuH19NO: C, 72.88; H, 10.81; N, After three extractions with ether the ethereal extracts were
7.73. Found16: C, 72.70; H, 10.46; N, 7.72. combined and washed successively with 5% hydrochloric
acid, water and 5% sodium bicarbonate solution. After
Hydrolysis of N-Camphenilylacetamide .—The amide is drying over magnesium sulfate the ethereal solution was dis-
particularly resistant to all forms of acid-induced hydroly- tilled through an efficient column yielding 4.65 g. (75%) of
sis. A solution of 29.78 g. of the amide (0.164 mole) and camphenilone, b.p. 188-191° (atm.) uncor. The infrared
120 g. (1.83 moles) of potassium hydroxide in 220 ml. of spectrum is identical with that of camphenilone prepared
ethylene glycol was heated to 150-160° whereupon a vigor- in the conventional manner.8 The semicarbazone melted
at 221-223° and gave no depression with an authentic
(13) Melting points are uncorrected unless otherwise noted.
(14) We are indebted to Dr. P. A. S. Smith for several helpful sug- (17) D. S. Noyce, This Journal, 78, 20 (1951).
gestions regarding conditions for carrying out this reaction. (18) G. Komppa, Ann., 366, 71 (1909).
(15) W. Hückel, Ber., 69B, 2769 (1936). (19) Cf. A Weissberger, “Technique of Organic Chemistry," Vol.
(16) Clark Microanalytical Laboratory, Urbana, Illinois. HI, Interscience Publishers, Inc., New York, N. Y., 1950, p. 424.
July 5,1953 Racemization in Camphene Hydrochloride-Isobornyl Chloride Rearrangement 3171

Table I
Source Camphenilone 6·6 Methylcamphenilol M Camphene 6

Salt [a]23D0 M.p., °C. [ej26o M.p., °C. [a]*«.«D M.p., °C. [aJ26D

Hydrochloride 39.2-40.8 118-119 50.0-52.5

.....

Camphorate + 1.36° 41.0-42.8 +76.1° 110-115 +29.1° 47-51 +109.0'

Camphorate -
2.13 37-40 +62.4 48.0-52.8 + 103.5
Camphorate + 10.16 38-40 -30.5
Tartrate + 19.9 38-40 -30.5
Residues from tartrate resolutions 35-40 -20.6 49.2-51.5 27.1 -

0
Rotations taken in absolute ethanol, 1.5-2.0% solutions, 1 dm. tube. b
Rotations taken in benzene, 1.5-2.5% solutions,
d
1 dm. tube. "Reported3 values: m.p. 38-39°, [ ]20 66.7° (benzene). Reported3 values: m.p. 112.0-113.5°, [a]20d
32.1°. "
Reported values: m.p. 51-52°; [a]MD 107.7° (benzene)8; m.p. 47°, 12 —113.5° (ether), 117.5° (toluene.4)
·

Table II
Counting Data
Total Plate
wt., BaCOa, BaCOa, Sp. act. BaCOa, Weighted Molar sp. act. pdct.,
Compound No. mg. mg. mg. ct./min./mg. averages, % ct./min./mol.
Methylcamphenilol («active 1 67.8 922.0 a 13.7 111.2 =b 0.6 111.7 ± 1.5 2.34 X 10s ± 1.5%
camphene) b 19.1 112.1 ± 1.2
2 59.9 809.6 a 8.8 112.4 ± 1.2 112.3 ± 2.0 2.34 X 10s ± 2.0%
b 17.6 112,2 ± 1.2
Dimethylnorcampholide 72.8 885.8 a 4.85 21.2 ± 0.3 21.5 ± 4.5 1.55 X 10s ± 4.5%
(from racemized camphene) b 5.7 22.7 ± 0.5 Corrected for dilution
c 13.85 21.2 ± 0.3 by factor 4.338/
d 10.0 21.6 ± 0.5 1.126

sample. Rotational and melting point data are given in and the mixture was heated on the steam-bath overnight (11
Table I. hr.). At the end of this time the mixture was black with
Methylcamphenilol.—This was prepared according to the a dark greenish cast to it. The flask was cooled and
procedure given by Htickel, et al.,3 by means ot large excess aqueous methanol was added very slowly and with good
of methyl Grignard reagent acting upon camphenilone: stirring, the reaction being quite vigorous. Much solid
yield (based upon 2.00 g. (0.0145 mole) of camphenilone) separated at this stage. The mixture was then steam dis-
1.94 g. (86.6%), m.p. 118-119°, as previously reported,3 tilled and the distillate separated, the organic layer being
for racemic sample after repeated sublimation. Rotational dried, filtered and distilled through an efficient column.
and melting point data are recorded in Table I. The meth- At 138-148° a fraction which contained appreciable un-
ylcamphenilol-8-C14 used as counting standard and for con- saturated material distilled, while at 148-152° there was
version to labeled camphene was prepared in the same obtained 1.38 g. (33%) of camphene, m.p. 49-52°, whose in-
manner using methyl iodide-C14: specific activity 111.7 frared spectrum was identical with that of authentic cam-
ct./min./mg. BaCOg; molar specific activity 2.34 X 108 phene.
ct./min./mole methylcamphenilol. Optically Active Camphene-8-C14.—This was prepared
Camphene. (A) By Xanthate Ester Dehydration of from optically active camphenilone (2.757 g. (0.020 mole),
Methylcamphenilol.—A solution of 2.0 g. (0.013 mole) of m.p. 37-40°, [<*]26d +62.4° (benzene)). The product
methylcamphenilol in 5 to 10 ml. of dry benzene was re- was purified by distillation and repeated sublimation:
fluxed with excess clean potassium for at least 6 but prefer- m.p. 49.0-52.8°, [q:]26d +103.5° (benzene, molar specific
ably 12 hr., after which interval gas bubbles were no longer activity 2.34 X 10sct./min./mole(seemethylcamphenilol)).
formed on the surface of the molten potassium. The mix- This material was used in the racemization studies. The
ture was cooled and the hardened lump of potassium was methylcamphenilol taken as standard was removed from the
removed from the yellow-orange solution in which large reaction series and suitably purified, whereas the methyl-
clear crystals were sometimes observed. After warming camphenilol for xanthate preparation was used directly
slightly 5 ml. (a large excess) of carbon disulfide was added without purification.
causing immediate separation of a cream to orange colored Racemization of Optically Active Camphene-8-C14.—The
solid. Lumps were broken up by vigorous shaking and the racemization was carried out as described by Meerwein and
mixture was then refluxed for 2 hr. There was then added Montfort.7 A solution of 1.126 g. (0.00827 mole) of op-
5 ml. (a large excess) of methyl iodide, and the refluxing tically active camphene-8-C14 and 3.2119 g. (0.0236 mole)
was continued for 10 to 12 hr. longer. The mixture was of ordinary purified racemic camphene20 in 35 ml. of dry
then filtered and the solvent was distilled from the filtrate. ether was saturated with dry hydrogen chloride at 0°.
The dark, unpleasant smelling oil remaining after removal After 1 hr. the solvent and excess hydrogen chloride were
of the solvent was refluxed for 1 hr. with a solution of 7 removed by vacuum distillation leaving a residue of slightly
pellets of potassium hydroxide in 5 ml. of 95% ethanol, straw colored camphene hydrochloride, m.p. 124-128°, re-
after which it was poured into cold water. The resulting ported 125-127°.21 This was dissolved in freshly distilled
mixture was extracted with ether, the extract treated with o-cresol, and a portion was placed in a polarimeter tube (1
Norit, dried over magnesium sulfate, filtered and distilled dm.). As time passed the initial pinkish tinge deepened,
yielding 1.26 g. (71.5%) of camphene, b.p. 150-154° and the solution became so dark by the time the initial ro-
(atm.). The infrared spectrum was identical with that of tation of —1.20° had fallen to —0.20° (16.6%, 163 min.)
authentic camphene. that further observation was impossible. Since Meerwein
(B) By Conversion of the Grignard Product.—With 25 had shown that isobomyl chloride may be completely race-
ml. of «-butyl ether as solvent the Grignard reagent was mized under similar conditions in 190 min., and that cam-
prepared in the usual manner from 0.0618 mole of methyl phene hydrochloride rearranges to isobomyl chloride very
iodide. The mixture was heated on the steam-bath for 5 rapidly under these conditions, it appeared safe to assume
hr. to ensure complete reaction; then 4.25 g. (0.0308 mole) complete racemization, based on the relative rate of de-
of camphenilone in 10 ml. of «-butyl ether was added slowly crease in optical activity for the present sample as com-
and the heating was continued for 2 hr. During this pe- pared with Meerwein’s sample, after 10.75 hr. At the end
riod a white solid separated. To this slurry was added of this time the reaction mixture was poured into 200 ml. of
5 g. (0.0658 mole) of carbon disulfide, and heating was 15% potassium hydroxide, and the resulting mixture was
continued for an additional 4 to 5 hr. The mixture was at
first yellowish in color but gradually became almost black. (20) Kindly supplied by the Hercules Powder Co.
Next there was added 9.0 g. (a large excess) of methyl iodide, (21) H. Meerwein and K. van Emster, Ber., 52B, 1815 (1920).
3172 William A. Mosher and Clifton L. Kehr Vol. 75

extracted three times with ether. The combined ethereal melting point, 94.0-96.5°, previously 96.0-96.5°u and 95-
extracts were washed twice with 100 ml. of 10% sodium 96°22; specific activity 21.5 ct./min./mg. BaC03; sp.
hydroxide solution, and then the organic layer was dried molar activity 1.55 X 10' ct./min./mole dimethylnorcam-
and the ether was distilled using an efficient column leaving pholide, corrected for “dilution” of camphene.
crude isobornyl chloride which was reconverted to cam- Analytical Procedure.—Peroxide fusion in a Parr bomb
phene by refluxing for 2 hr. with excess alcoholic sodium was used. In each case the conversion to barium carbon-
ethoxide. Distillation of the alcoholic solution carried over ate and preparation of plates were carried out as described
most of the camphene. To the residue was added 20 ml. of by Calvin, et al.23 The plates were counted using an '-
water and the distillation was continued until an oil began Wood Model K flow counter and a Nuclear Instrument
to steam distil at about 83°. The alcoholic distillate was Chemical Co. model 161 scaler with attached Veeder Root
drowned with a large quantity of water and then extracted register and Precision Scientific Co. timer. The counter
thrice with ether. After drying over magnesium sulfate was operated at 1400 v., 125 v. above the threshold, under
the ethereal solution was distilled yielding 2.08 g. (0.0153 a constant stream of 3-5 bubbles of “Q” gas per sec. The
mole, 48%) of fully racemic camphene, b.p. 146-156° over-all calculated error for molar specific activity is
(atm.), m.p. 40-51°. The product was carefully sublimed ±4.5%.23
to give pure camphene, m.p. 49-51°, [<*]25d 0.00° (benzene).
Ozonolysis of Racemized Camphene.—The procedure of (22) F. W. Semmler, Ber., 42, 246 (1909).
Harries and Palmén11 was followed exactly with similar (23) M. Calvin, C. Heidelberger, J. C. Reid, B. M. Tolbert and P. F.
results. The dimethylnorcampholide was recrystallized Yankwich, "Isotopic Carbon," John Wiley and Sons, Inc., New York,
thrice from ether, five times from 1:1 ether-petroleum ether N. Y., 1949, pp. 90-92, 118, 283-291.
(90-100°) and then was sublimed without change in the Ann Arbor, Michigan

[Contribution from the Department of Chemistry, University of Delaware]

The Decomposition of Organic Acids in the Presence of Lead Tetraacetate*
By William A. Mosher and Clifton L. Kehr
Received December 8, 1952

Previous workers have shown that lead tetraacetate decomposes in acetic acid to give carbon dioxide, methane, acetoxy-
acetic acid and methylene diacetate. We have extended this reaction to include a variety of organic acids (formic acid, tri-
phenylacetic acid, trimethylacetic acid, isobutyric acid and isovaleric acid) in a study of the mechanism involved. The prod-
ucts obtained may be readily explained by the following ionic scheme: interchange first occurs between lead tetraacetate
and the organic acid present; decomposition of the Pb ++++ salt yields the corresponding Pb++ compound, the negative
ion of the acid and positive ion such as R-CO-O +. This electronically deficient oxygen-containing ion may then decompose
to carbon dioxide and a carbonium ion which may stabilize itself by losing a proton to form an olefin, by reacting with solvent
anion to form esters, or by abstracting a hydride ion from a molecule of solvent to give rise to a more stable carbonium ion
and hydrocarbon gas. No evidence of free radical decomposition was observed.

The decomposition of lead tetraacetate in acetic Because it is a well established fact that aldehydes
acid has been reported by Kharasch1 and co- under the conditions employed are stable to lead
workers, and a mechanism involving trivalent lead tetraacetate, it was suspected that the hydroxyl
radicals proposed. Previous work2 in this Labora- group must be the center of attack by the oxidizing
tory, however, has suggested that benzpinacolyl agent. The following generalized mechanism is
alcohol was oxidized by lead tetraacetate in good proposed
yield to benzaldehyde and triphenylcarbinol via Z° /°
an ionic cleavage" reaction involving an electroni-
Pb(OAc)4 + 4R—C—OH Pb(0—C—R)4 + 4HOAc
cally deficient oxygen intermediate. In an attempt
to determine the generality of ionic reactions of lead decomposes
tetraacetate, a systematic study has been made of i-"-1
the decomposition of tetravalent lead salts of or-
ganic acids of varying structure. Table I presents
a concise summary of the conditions and products
/° /?.
Pb(0—C—R)2 + RC—0:~ + R—C—0 +
/°.
of the reactions investigated. I II
A careful study of the products obtained in the
above reactions indicated a pronounced mechanis- ¡+H*
tic trend which led us to interpret the reaction
from the ionic viewpoint rather than the free radi- />
R—C—OH R+ + C02
cal concept advanced by Kharasch.1 III
The oxidation of formic acid8 appeared to be a
convenient starting point in the present mecha- There is other evidence4 for the rejection of the
nistic study because this molecule offers to the oxi- classical aldehyde theory of oxidation of formic
dizing agent only two possible points of attack— acid.
the aldehyde function and the hydroxylic function. In support of this mechanism, the tetravalent
* Presented in lead salt of trimethylacetic acid has been prepared
part before the Division of Organic Chemistry,
American Chemical Society, Atlantic City, September, 1952. and decomposed in trimethylacetic acid solvent.
(1) M. S. Kharasch, . N. Friedlander and W. H. Urry, J. Org. The products obtained from this reaction were
Chem., 16, 533 (1951).
(2) W. A. Mosher and H. A, Neidig, This Journal, 72, 4452
nearly identical to those found when lead tetra-
acetate itself was decomposed under similar con-
(1950).
(3) J. M. Grosheintz, ibid., 61, 3379 (1939). (4) Cf. Paul Haas, Nature, 167, 325 (1951).