THE STEREOCHEMISTRY OF THE
DECOMPOSITION OF 1-PYRAZOLINES
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
The writer wishes to express his deep appreciation for the valuable
assistance and inspiring direction of Dr. W. M. Jones who conceived this
research project. His sugestions and encouragement were generously
contributed in the completion of the research and in the writing of the
lie wishes to thank, also, the other members of his supervisory
committee for their constructive suggestions and criticism during the
writing of this dissertation and in particular Dr. W. S. Brey for his
assistance in obtaining and interpreting the N. M. R. data.
The writer is grateful to Mr. D. G. Baarda for his suggestions in
the preparation of the rough draft and to Mrs. R. L. Bielling for the
typing of the final copy.
- ii -
TABLE OF CONTENTS
ACKIOLEDGE)MENT . .. .* ... .. .
LIST OFIGURES . . *
I. IfTRODUCTION . *
II. RESULTS AND DISCUSSIONS .. . .
Preparation of 3-methyl-cis-3, 4.dicarbcmzethoa~ -5, 5-
diphenyl-1-pyramoline (VT. .. .*
Decomposition of 3-methyl-ci-3, 4-dicarbomethoxy-5, 5-
diphenyl-1-pyrazoline (V). . .
Preparation of 3-methyl-trans3, -dicarbomethoxy-5, 5-
diphCnyl-l-pyrazoline (VI). . .
Decomposition of 3-methyl-trans.3, 4-dicarbomethox-5, 5-
diphenyl-1-pyrazoline (VIIT . *
Preparation of benzaldehyde-hydrazone .
Preparation of phenyldiazomethane. ....
Preparation of 3- -thyl-cli-3, 4-dicarbomethoxy-5-
phenyl-1-pyrazoline (X). ..... .
Preparation of 3-methyl-trans-3,4-dicarbomethoxy-5-
phenyl-1-pyrazoline (XX I . .
Preparation of 1-methyl-cis-l, 3-dicarbomethoxy-3-
phenylcyclopropane (XI). ............... .
Isolation of l-methyl-trans-l, 2-dicarbomethoxy- 3-
phenylcyclopropane (XIII . .
Hydrolysis of trans-cyclopropane (XIII).. *
- iii -
Preparation of 3, 5-dicarbomethoxy-4-phenyl-5-methyl-
2-pyrazoline with phenyl and methyl groups cis (XX). 21
Preparation of 3, 5-dicarbomethoxy-4-phenyl-5-methyl-
2-pyrazoline with phenyl and methyl groups trans
(XXI). . . ...... 22
General procedure for gas chromatographic analysis
of the decomposition product ..... .. ..... 22
Decomposition of 3-methyl-cli-3, 4-dicarbomethoxy-5-
phenyl-l-pyrazoline (X),... .. ..... 27
Decomposition of 3-methyl-trans-3, i-dicarbomethoxy-5-
phenyl-1-pyrasoline (XI). . 28
Decomposition of the 2-pyraoline (XX) . 29
Decomposition of crude 2-pyrazoline (XXI). 30
LIV SSTO ALRYr ... ..................... g3
IV*OR EUPIARYL * *. 33
LIST OF REEHECNCESC .. .. #. ..# ... 35
BIOGRAPHICAL SWMH* . # ** #36
- iv -
LIST OF FIGURES
i ,ure Page
1. The standardization curve for the cis-cyclopropane (XVI). 24
2. The standarization curve for the trans-cyclopropane (XVII) 25
3. The standarization curve for 4-t-butylcyclohexanone. 26
4. An illustration for the estimation of the inarim peak area*. 27
It has been generally assumed that the decomposition of 1-pyrasolines
occurs stereospecifically.1,2 This assumption has been based only on the
examination of 1-pyrazolines with methylene groups adjacent to the nitrogen, 3
CHOOC COOCH3 CooH3
CH3ooC C 3 COOCH3
It is rather difficult to make this observation consistent with
either an ionic or a diradical mechanism for the decomposition of 1-
pyrazolines unless it is assumed that the ring-closure to cyclopropane
occurs more rapidly than rotation around the single bonds.
Van Alphen's work4 on the decomposition of a 1-pyrazoline with two
phenyl groups on the C-5 atom confirmed this assumption. According to
his results, the decomposition of 3-methyl-cie-3,4-dicarboumthoxy-5,5-
diphenyl-l-pyrazoline (V) yielded, rather than the anticipated cis-isomer,
exclusively l-methyl-trans-1, 2-dicarbomethoxy-3, 3-diphenylcyclopropne (VI),
indicating a lack of stereospecificity in this reaction.
CH300C Cooca3 30COC
C6"' H C3H 3CH3
c695:' C6H5 3H3
From a consideration of the kind of the substituents on the C-5 atom
of the afore-mentioned 1-pyrazolines, it appeared that the competition
between ring-closure and rotation is dependent upon the stability of the
In an attempt to rationalize the stereochemistry of the decomposition
of certain 5-phenyl-2-pyrazolines, Jones has assumed that 5-phenyl-l-pyra-
zolines decompose stereoselectively.5,6 Inherent in this assumption wa
the assumption that a single phenyl on the C-5 atom of the 1-pyrazoline
would not have enough stabilizing effect on the decomposition intermediate
to allow free rotation around the single bonds. However, in view of the
observations of van Alphen, it became apparent that the validity of these
assumptions might well be questioned. Thus, the purpose of this project
was to test the validity of these assumptions. The approach employed was
an examination of the stereochemistry of the decomposition of the isomeric
II. RESULTS AND DISCUSSIONS
At the time when this work was begun, van Alphen's result4 was the
only known example of a non-stereospecific decomposition of a 1-pyrazoline.
His work was repeated and compared with the decomposition of the isomeric
3-methyl-trans-3, -dicarbomethoxy-5, 5-diphenyl-l-pyraeoline (VII), an
addition product of diphenyldiazomethane and dimethyl mesaconate. The
purpose of the comparison was two-fold. One was to ensure that the addition
of diphenyldiazomethane to dimethyl citraconate was in a cia-fashion so
that the stereochemistry of the 1-pyrazoline (V) prepared by van Alphen
was as claimed. The other was to provide further proof that the decompo-
sition of van Alphen's 1-pyrazoline (V) was non-stereospecific. If the
decomposition of these pyrazolines (V) and (VII) were non-stereospecific,
both pyrazolines should yield the same product, namely, 1-methyl-trans-l-
2-icarbomethoxy-3, 3-diphenylcyclopropane (VI).
This has been found to be the case. When cis-pyrazoline (V, m.p.
1240C., dec.) and trans-pyrazoline (VII, m.p. 103C., dec.) were decom-
posed at their respective melting temperatures, both pyrazolines yielded
nearly one hundred percent yields of trans-cyclopropane. The identity of
both decomposition products was confirmed by infrared spectra and mixed
The non-stereospecificity of the decomposition of pyrazolines (V)
and (VII) my be explained as due to the stabilizing effect of the two
phenyl groups on the C-3 atom of the decomposition intermediate (VIII,
either ionic or free radical). The life-time of the intermediate could
thus become sufficiently lonG to permit free rotation around the single
bonds before the ring-closure could occur.
CH3000C COOCH3 CHC COOCH3
CH *32 CH3 / 2 cH3
y*3 CTe 3
On the other hand, the decomposition of 3-methyl-cis-3,4-dicarbo-
methoxy-l-pyrasoline (I) and its isomeric trans-pyrazoline (II) have been
found to be stereospecific.3 In this case, the ring-closure of the inter-
mediate (IX) to cyclopropane apparently occurs more rapidly than rotation
around the single bonds. This is reasonable in view of the system inves-
tigated since either a primary carbonium ion or primary free radical would
be expected to be extremely reactive.
The afore-mentioned two extreme results aroused interest in an inves-
tigation of the analogous 1.-pyrazoline with one phenyl group on the C-5
atom. Thus, although it has been suggested that a 5-phenyl-l-pyrazoline
probably decomposes stereoselectively,5,6 this has not been previously
For this purpose the stereochemistry of the decomposition of 3-methyl-
cis-3, l-dicarbomethox-5-phenyl-l-pyrazoline (X) and the isomeric 3-methyl-
trans-3,4-dicarbomethoxy-5-phenyl-l-pyrazoline (XII) were investigated.
The two pyraaolines X and XII were prepared from the addition reactions
of phenyldiazomethane with dimethyl citraconate and dimethyl mesaconate
respectively. (At this stage, the configuration of the phenyl group can
not be assigned. It will be determined later on.)
CH3000 1H3 CH300C CH3
C6115 Hc -- H3
The decomposition of the trsLn-1-pyraaoline (XII) was carried out at
180-185C. he analysis of the decomposition product by gas chromatography
using an 18 ft. column of 30-60 mesh Tide in a Perkin Elmer Model 154-B
Fracometer at 1900C. at an internal pressure of 25 p.s.i. revealed the
presence of a single cyclopropane peak with a retention time of 12.7 min.
The cyclopropane (XIII) was isolated from the distillate of the crude
reaction mixture of the trans-l-pyrasoline (XII) by liquid phase chroma-
tography and was confirmed by an elementary analysis and KMnO4 test. In
order to gain soe knowledge of the configurations of the two carbomethoxy
groups, the cyclopropane was hydrolyzed under alkaline conditions and an
attempt was made to close the resultant di-acid to an anhydride by heating
the di-acid at 200-2300C. and 30 mm.Ee. Only unchanged acid sublimed,
indicating that the two carbomethoxy groups in the cyclopropane were trans.
The possibility of the isomerization of the di-acid during saponification
was eliminated by converting the di-acid to the original cyclopropane by
methylation with diazomethane.
C6 "5 OCH3
When the cis-l-pyrazoline (X) was decomposed and analyzed under
conditions identical with those employed for the trans-l-pyrazoline (XII)
no cyclopropane was detected. In order to clarify the question as to
whether, in this case, no cyclopropane is formed or whether the cyclopro-
pane formed has an extremely long retention time, one of the isomeric cis-
cyclopropanes (XI) was synthesized and analyzed by gas chromatography.
The synthesis of the cis-cyclopropane (XI) was accomplished by the methy-
lation of a half-ester obtained by refluxing the cyclopropee anhydride (iV)
in methyl alcohol. The cyclopropane anhydride (XV), on the other hand,
was prepared from the decomposition of pyrazoline anhydride (XIV) result-
in& from the addition of phenyldiazomethane to citraconic anhydride. When
the cyclopropane thus obtained was analyzed by gas chromatography under
the condition described previously, it showed a peak with a retention time
of 17.7 min
This observation suggested that probably no cyclopropane was formed
under the conditions employed for the decomposition of the cis-l-pyrazoline
(X). Attempts were made, therefore, to find conditions under which the
cis-1-pyrazoline (X) would yield cyclopropane. The pyrazoline was decom-
posed at several different temperatures ranging from 1400C. to 260C. The
decompositions were also tried by refluxing the pyrazoline in xylene,
mesitylene and decalin. In no case could a cyclopropane be detected in
the v. p. c. chart. However, when the cis-pyrazoline (X) was introduced
into a test tube preheated at 280-2900C., the decomposition gave
he analysis of the above decomposition product by gas chromatography
revealed two peaks, the larger peak with a retention time of 17.7 min. and
the smaller peak with a retention time of 12.7 min. The retention time of
the large peak was identical to that of the cis-cyclopropane (XI) and the
retention time of the small peak to that of the trans-cyclopropane (XIII).
*When the crude cis-pyrazoline (x) with m.p. 104C. was melted, the
intensity of the -NH absorption in the infrared spectrum increased. It is,
therefore, thought that the failure of the formation of the cyclopropane
from the cis-pyrazoline (X) with d.p. 137C. decomposed at the temperature
below 260oc is due to its tautomerization to a 2-pyrazoline, thus prevent-
ing the formation of a cyclopropane.
Addition of previously prepared cis- and trans-cyclopropene (XI) and (XII)
to the reaction mixture enhanced the peaks already present, thus confirming
the assignment of these peaks to ci*- and trans-cyclopropane. Quantitative
analysis by gas chromatography revealed that the decomposition product of
the cis-pyrazoline (X) contained approximately 6 percent of cis-cyclopro-
pane (XI) and approximately 2 percent of trans-cyclopropane (XIII),
indicating the decomposition was stereoselective regarding the cis- and
trans-relationship of the two carbomethoxy groups.
It now becomes necessary to consider the stereochemistry about the
third and remaining carbon atom in the cyclopropane ring. There are four
possible isomeric cyclopropanes (XVI), (XVII), (XVIII) and (XIX) which
can be derived from the 1-pyrazolines (X) and (XII). Of these four cyclo-
propanes, the ones with structures (XVIII) and (XIX) are sterically less
HX33 3/ 3
CGs j OOCH3C COI
CH3 1 3
stable than the ones with structures (XVI) and (XVII), due to the fact
that there are three bulky groups on the same side of the cyclopropane
ring. Actually only two cyclopropanes were detected from the decomposition
products of the 1-pyrazolines (X) and (XII). It was, therefore, reasoned
that the cyclopropanes resulted from the decomposition of cis- and trans-
1-pyrazolines were most likely the sterically more favorable cyclopropanes
(XVI) and (XVII).
In order to verify the above stereocheeical assignments, the two 2-
pyrazolines (XX) and (XXI) were prepared by reactions of methyl diazoa-
cetate with methyl 0(-methyl cinnamte and methyl allo- (-methyl cinnamate
respectively. The decomposition of these 2-pyrazolines was effected at
C6H5 CA3 C6 COOCH3
/ T^cooc3 i 7^3
CH300C N 'H C CH3OOC
According to Jones, the stereochemical course of the decomposition
of 2-pyrazolines can be predicted by presuming that the reaction proceeds
by an initial tautomerization to the sterically favored 1-pyrazoline
followed by expulsion of nitrogen.56 This rule has been confirmed for
certain 3-carbomethoxy-2-pyrazolines. Since both 2-pyrazolines (XX) and
(XXI) are 3-carbomethoxy-2-pyrazolines, upon decomposition they could be
expected to yield predominately cis- and trans-cyclopropane (XVI) and
The results of the analyses of the decomposition products of 2-
pyrazolines (XX) and (XXI) by gas chromtography were in line with the
expectations. The analysis of the decomposition product of 2-pyrazoline
- 10 -
XX I / COOCH -- /
3CH3 -- 3 3
1, CH J= N CH 300C 3
(XX) indicated that the decomposition product contained a miaim of 79
percent by eight of the cis-cyclopropane (XI) and a maximum of 8 percent
by eight of the trans-cyclopropane (XIII) (or other material with similar
retention time), The results of the analysis of the decomposition products
of the 2-pyrazoline (XXI), on the other hand, indicated that the decompo-
sition product contained a mxtmYiu of 23 percent by eight of the rae-
cyclopropane (XIII) and a naxiumm of 9 percent by eight of the cis-
cyclopropane (XI) (or other material vith similar retention time). The
major cyclopropanes in the decomposition products of the 2-pyrazolines
(xx) and (XXI) were both isolated by liquid phase chromatography and
confirmed by their infrared spectra to be identical with cis- and trans-
cyclopropanes (XI) and (XIII) respectively.
The fact that the two 2-pyrazolines which were decomposed gave
different major cyclopropane products is a good indication that at no time
during the decomposition did equilibrating rotation around the single bonds
occur. Thus, the stereochemical relationship of the phenyl and the ring-
methyl must be the same in the product as in the reactant. If this is
true, then the fact that the cyclopopanes resulting from the decomposition
of the 2-pyrazolines are identical with those resulting from the decompo-
sition of the 1-pyrazolines leads to the conclusion that cyclopropanes
resulting from the decomposition of cis- at trans-1-pyrazolines actually
have the structure pictured in XVI and XVII, respectively.
The n..r. analyses also confirmed the above stereochemical assign-
ments for the cyclopropane (XI) and (XIII). The results of analysis
showed that the cis-cyclopropane (XI) had a coupling constant of 6.9 e.p.s.
and the trans-cyclopropane (XI) 7.2 c.p.s. for the two neighboring
cyclopropane hydrogens, demonstrating that in both cyclopropanes the two
hydrogen atoms were in the trans configuration.7
The consideration of the stereochemistry of the cyclopropanes assigned
above suggested that the phenyl groups on the C-5 atom of the original 1-
pyrazolines (X) and (XII) were trans to the carbomethoxy groups on the 604
atom of the pyrasolines. As the hydrogen atoms in the cyclopropanes
found in the decomposition product were trans with respect to one another,
it seems likely that the two hydrogen atoms in the parent pyrasolines also
assume the same stereochemistry. Otherwise, the decomposition of the
pyrasolines (X) and (XII) would require that there Vera only equilibration
between the 0-4 and the C-5 atom and no equilibration between the C-3 and
the C-4 atom of the pyrasolines taking place.
For this reason the structures (XXII) and (XXIII) are assigned for
the pyrazolines (X) and (XII) respectively. However, the assignments are
not very conclusive in view of the possibility that the rotation around
- 12 -
CH C OH3 C300C CH3
H 3 COOCH3
c6 N / 3 c6; 3
the bonds between the C-4 and the C-5 atom could conceivably be easier
than those between the C-3 and the C-4 atom of the pyrazolines due to the
presence of the bulky methyl group (or the carbomethoxy group for the
pyrazoline (XII)) on the C-3 atom as compared to tohe all hydrogen atom
on the C-5 atom of the pyrazoline.
Preparation of 3-methyl-cis- 3, 4-dicarbomethoxy-5, 5-diphenyl-l-pyrazoline
(V). To 10 grams (0.0556 mole) of dimethyl citraconate8 was added 10.8
Grams (0.0556 mole) of crystalline diphenyldiazomethane obtained from the
oxidation of benzophenone-hydrazone.9 The resultant solution was set
aside with occasional shaking. After two days, the whole solution had
solidified. The solidified mass was pink colored. After one week, the
slightly pink mss was filtered and the resulting solid was washed twice
with 50 ml. portions of ether and then recrystallized three times from
methyl alcohol to give 8.4 grams of coarse colorless crystals; m.p. 1240C.
with decomposition. Yield: 43.0 percent.
Analysis: Calculated as Cp20Q004N2: C, 68.18; H, 5.68; N, 7.95.
Found: C, 67.81; H, 5.95; N, 8.08.
The infrared spectrum indicated the absence of the -NH group.
Decomposition of 3-methyl-lia-3, 4-dicarbomethoxy-5, 5-diphenyl-l--yrasoline
UV. A sample (0.237 gram; 0.673 moles) of V was decomposed at 1250C.
until nitrogen evolution had ceased. The decomposition product, white
solid, weighed 0.217 gram equivalentt to 0.670 mole of the corresponding
cyclopropane (VI)). Recrystallization of the product fro methyl alcohol
yielded colorless needles, m.p. 114-1150C., which did not decolorize 1X0an0
solution.' Admixture with the decomposition product from the trans-pyra-
zoline (VII) shoved no depression in melting point. Infrared spectra of
the two decomposition products were also identical.
- 13 -
- 14 -
Preparation of 3-methyl-Jn- 3, ,-dicarbomethoxy-5, 5-diphenyl-l-pyrazoline
(VII). To 10 grams (0.0556 mole ) of dimethyl mesaconatel0 was added 10.8
grams (0.0556 mole) of diphenyldiazomethane.9 The resultant solution we
shaken well and set aside overnight. By the next morning, the whole solu-
tion had solidified. The solidified mass was slightly pink colored.
After one week all of the pink color had disappeared. 'he greenish solid
left behind was then washed with ether and recrystallized from methanol
to yield 9.1 grams of coarse crystals; m.p. 10o4C. with decomposition.
Yield: 46.4 percent.
Analysis: Calculated as C2T200 N2: C, 68.18; H, 5.68; N, 7.95.
Found: C, 68.28; H, 5.71; N, 8.22.
The infrared spectrum indicated the absence of the -NH group.
Decomposition of 3-methyl-trans- 3, 4-dicarbomethoxy-5, 5-diphenyl-l-pyra-
zoline (VII). A sample of 0.192 gram (0.546 mole) of VI was decom-
posed at 1050C. until nitrogen evolution had ceased. The decomposition
product, white solid, weighed 0.177 gram (corresponding to 0545 mole
of l-methyl-trans-l, 2-dicarbomethoxy-3, 3-diphenyl cyclopropane (VI)).
Recrystallization of the product from methyl alcohol yielded colorless
crystalline; needles, m.p. 1150C., which did not decolorize KMnO4 solution.
Admixture with the decomposition product from 3-methyl-ci-3, 4-dicarbome-
thoxy-5, 5-diphenyl-i.-pyrazoline (V) showed no depression in melting point.
The infrared spectra of the two decomposition products were also identical.
Preparation of benzaldehyde-hydrazone. To an ice-cooled solution of
anhydrous hydrazine (64.0 grams, 2.00 moles) was added dropwise with
stirring a solution of 106 grams (1.00 mole) of benzaodehyde in 300 ml.
- 15 -
of ether. After the addition of the benzaldehyde solution, the resultant
solution was dried over solid sodium hydroxide for three or four days
until the solution became reddish brown. The solution was then filtered
and the solvent and excess hydrazine were removed on a water bath under
the vacuum of a water aspirator. The residue was distilled at a pressure
of 12 am. Seventy-two grams of the distillate with b.p. 134-1360C. were
obtained. Yield: 60.0 percent. (Different preparations resulted in
different yields which ranged from 58.3 to 61.3 percent.)
Preparation of phenyldiazomethane. Fifty grams (0.458 mole) of benzalde-
hyde-hydrazone was suspended in 500 ml. of pentane. With stirring 100
grams (0.463 mole) of red mercuric oxide were introduced to the suspension
in a time interval of two and one-half hours. After the addition of
mercuric oxide the reaction mixture was further stirred for one-half of
an hour. The mixture was then filtered to remove insoluble materials.
The residue was washed with 50 ml. of pentane. The washing was combined
with the filtrate. The final volume of the pentane solution was 530 ml.
From titration with maleic anhydride, the concentration of phenyldia-
zomethane in pentane was estimated to be 0.313 X 10"4 anle/ml. Yield:
Preparation of 3-methyl-ci- 3, -dicarbomethoxy-5-phenyl-l-pyrazoline (X). -
To 14.8 grams (0.0939 mole) of dimethyl citraconate was slowly added 300
al. of ice-cold solution of pentane containing 11.1 grams (0.0939 mole)
of phenyldiazomethane. The resultant solution was shaken well and was
placed in the refrigerator. After three days at this temperature, a mass
of white solid contaminated with orange and yellow precipitates had formed.
- 16 -
The solids were filtered and washed several times with pentane until the
washing was only slightly yellow. The solid was then washed with ether to
remove the remaining yellow color in the 1hite solid. The white solid
after washing and drying in a vacuum desicator weighed 18.40 grams, mp.
104-1090C. The white solid was unstable in the air. Upon standing in
the air for three or four hours the solid changed to a brownish viscous
mess. It was also unstable in warm methanol, chloroform and carbon
tetrachloride solution. When the white solid, mp. 104-109oC., was dis-
solved in any of these solvents the solution changed from colorless to
The white solid with mop. 104-1090C. (15.65 grams) was dissolved in
approximately 800 ml. of ether. After most of the solvent was removed by
the slow evaporation of ether in the hood, the colorless crystals formed
were separated from the remaining yellow oil by filtration. The colorless
crystals after being washed with ether weighed 3.05 grams, m.p. 130C. vith
decomposition. The white crystals (1.47 grams), m.p. 1300C. with decooqo-
sition, were recrystallized from cold methanol solution to yield 1.30 gras
of the white crystals, m.p. 1370C. with decomposition. Yield: 12.1
Analysis: Calculated as C9416N204: C, 60.90; H, 5.80.
Found: C, 60.93; H, 5.83.
The infrared spectrum indicated the absence of the -NH group.
Preparation of 3-methyl-*|r L- 3,4-dicarbomethoy- 5-phenyl--yzoline
). -- To 24.50 grams (0.155 mole) of dimethyl mesaconate was slowly
added 218 ml. of ice-cooled solution of pentane containing 18.3 grams
- 17 *
(0.155 nole) of phenyldiazomethane. The resultant solution was shaken
well and placed in the refrigerator.
After three days, a mass of white solid contaminated with orange and
yellow precipitates deposited on the bottom of the flask. The solids
were filtered, washed several times with pentane until the washing solution
was only slightly yellow. The solids were further washed with ether to
give 38.5 grams of white solid which melted at 63C. (started to bubble
Ten grams of the white solid with m*p. 63C. were dissolved in approa-
imately 50 ml. of ether and the solution was placed in the refrigerator.
After about a week, the lhite solid which had separated was filtered and
recrystallized from acetone. M.p. 1750C. with decomposition. Yield; 1.I
grams (9.9 percent).
Analysis Calculated as C14,Hu6N4: C, 60.90; H, 5.80; N, 10.14.
Pound: C, 61.02; H, 5.69; N, 10.08. C, 61.15; H, 5.79; N, 10.22.
Infrared absorptions indicated the absence of the -NH group.
In some cases, even after weeks in the refrigerator, the reaction
product of dimethyl mesaconate and phenyldiazomethane failed to solidify.
In these cases, the two layers formed in the flask vere separated. An
approximately equal volume of ether was added to the reddish viscous oil
that had been the bottom layer, and the ether was slowly evaporated at
ice temperature for several days. The white solid separated was recrystal-
lized from acetone to give a white solid which decomposed at 175C.
In a typical run, 300 ml. of pentane containing 0.111 mole of phenyl-
diazonethane were added to 18.7 grams (0.111 mole) of dimethyl mesaconate
and 3.2 gram (yield: 10.4 percent) of the white solid with decomposition
point 1750C. were isolated by the afore-mentioned method.
Preparation of l-methyl-cs-1, 2-dicarbomethoxy- 3-penylcyclopropane (XI). -
To thirty grams (0.268 mole) of citraconic anhydride was added an equi-
molar amount of phenyldiazomethane in 500 ml. of pentane. The reaction
mixture was set in the ice room. The next day, 33.3 grams of white solid
contaminated with a emal amount of yellow crystals had formed. The
solid was isolated, dried in the vacuum desicator and decomposed at 90-
1000C, The decomposed product weighed 28.8 grams. Twelve gram of the
decomposition product was refluxed in 150 ml. of 95 percent methanol
solution for 6 hours. After refluxing the solvent was removed on the
water bath under the vacuum of a water aspirator. To the residue an
etheral solution of diazomethane was added until bubbles were no longer
evolved. The ether was removed and the residue was dissolved in acetone
and oxidized with 10 percent IKMO4 solution. The brown and black precip-
itates formed during the course of the oxidation were filtered. Water
was added to the filtrate and the resultant solution was extracted three
times with 20 ml. portions of ether. The ether extracts were combined
and dried over anhydrous calcium sulfate. After the removal of ether the
residue was distilled. The fraction which distilled at 160-1620C./6 am.
was collected. Yield: 3.2 grams (11.6 percent).
Analysis: Calculated as C1H1604:c C, 67.78; H, 6.45.
Found: C, 67.62; H, 6.72.
H1 Analysis: Coupling constant for neighboring hydrogen atoms; 6.9
- 19 -
Isolation of l-methyl-t g-1, 2-dicarbomethoxy- 3-phenylcyclopropane
(XI The oil (43.2 grams) resulting from the preparation of trans-
pyrazoline (XII) was fractionally distilled to give the following fractions.
Fraction B. P. C me. Wt. gr.
1 104-120 15 3.0
2 120-120 15 1.1
3 160-160 13 0.5
4 160-175 13 1.2
5 175-190 13 14.0
One gram of the distillate from the fifth fraction was chromatographed
by using a column (diameter, 1.5 cam; length, 30 cm.) packed with 40 grans
of acid-washed alumina. The sample was transferred into the column by
dissolving in a very small amount of 35 percent ether in hexane. Tenty
percent ether in hexane was used as eluent. The following fractions were
Fraction Volume, ml. Wt. of the residue, ag.
1 70 35
2 50 20
3 60 123
4 30 134
5 30 98
6 35 100
7 50 21
- 20 -
The residue from fractions 2-7 were dissolved in hexane and cooled
at 00C. over night to give 320 ag. of white precipitate. Recrystallization
of the white precipitate several times from Me01-H20 yielded 250 mg. of
colorless crystals, m.p. 41.5-43C., which did not decolorize KMn04
solution. Yield: 28.4 percent.
Analysis: Calculated as C14H1604: C, 67,78; H, 6.45.
Found: C, 68.16; H, 6.49.
NMR Analysis: Coupling constant for neighboring hydrogen atoms; 7.2
Hydrolysis of tjang-cyclopropane (XIII). A sample of 0.65 grams of the
trans-cyclopropane was dissolved in 5 ml. of 5 percent KOH-MeOH solution.
After 4 days, the above solution was poured into 20 ml* of distilled water
and acidified with 10 percent C01 solution until the solution was strongly
acidic. The solution which became turbid was then extracted three times
with ether. The ether extracts were combined, washed with distilled
water until the washings were neutral to litmus paper. The ether solution
was dried over anhydrous calcium sulfate over night. The residue obtained
after the removal of ether weighed 0.54 grams m.p. 104-1100C. Recrystalli-
zation from MeOH-H20 yielded colorless crystals with map. 113-1150C.
Yield: 93 percent.
Analysis: Calculated as C1 1204: C, 65.5; H, 5.46.
Found: C, 65.75; H, 5.64.
A small amount of the di-acid was converted to its dimethyl ester
with diazomethane.ll Its infrared spectrum, melting point, and mixed
melting point with pure trans-cyclopropane proved that no isomerization
had occurred during the hydrolysis. An attempt to close the di-acid to
- 21 -
the anhydride failed. Upon heating the acid at 200-2300C. and 30 m in
a sublimation apparatus, only unchanged acid sublimed, m.p. 113-1150C
Mied with pure acid; no depression in melting point.
Preparation of 3,5-dicarbomethoxy-4-phenyl-5-methyl-2-pyrazoline with
phenyl and methyl groups cia (xx). To 24.1 grams (0.137 mole) or methyl
oL-methylcinnamate1 was added 13.70 grams (0.137 mole) of methyl diazoac-
etatel3 prepared from the diazotization of methyl Elycinate hydrochloride.1Y
The resultant yellow solution was heated at 90-950C. for 3 days. The
completion of the reaction was tested by addition of a few drops of dilute
hydrochloric acid solution to a small portion of the reaction mixture. No
nitrogen evolution was observed. The reaction mixture was then eluted
with ether through a column (diameter, 3 cm.; length, 50 am.) packed with
200 grams of acid-washed alumina. The residue obtained after the evap-
oration of ether weighed 33.0 grams. A 17.4 grams portion of the residue
was distilled under nitrogen at a pressure of 4.9 m. to remove unreacted
methyl o -methylcinnamate. The fractions which distilled up to 1600C.
were collected. The residue remaining in the distillation pot weighed
8.1 grams. The residue was recrystallized from CC14 to give 2.35 grams
of colorless crystals of m.p. 113-116C. Repeated recrystallization from
CC14 yielded white crystals, m.p. 114.5-1150C. Yield: 11.8 percent.
Analysis: Calculated as C4i16042: C, 60.90; H, 5.80; N, 10.14.
Fund: C, 60.91; H, 5.97; N, 10.20.
Significant infrared absorptions: 2.95, 571, 5.81, 6.41 microns.
- 22 *
Preparation of 3,5-dicarbometho ---phenyl-5-methyl-2-pyrazoline vith
phenyl and methyl groups trans (XIC). To 2.93 grams (0.0176 mole) of
methyl allo- ( -methylcinnamte15 was added 1.76 grams (0.0176 iole) of
methyl diazoacetate. The resultant solution was heated at 80-850C. for 4
days. At the end of this time a small portion of the reaction mixture
gave no nitrogen evolution upon the addition of dilute hydrochloric acid
solution. The solution was then distilled at a pressure of 5 ma. to
remove unreacted methyl allo- t-methylcinamate. All material which
boiled up to 1060C (2.24 grams) was collected. The residue in the
distillation pot, weighed 1.75 grams. Significant infrared absorptions:
2.99, 5.70, 5.80, 6.15, 14.25 microns. Without further purification the
residue was decomposed and the decomposition product analyzed by gas
General procedure for gas chromatographic analysis of the decomposition
product. A Perkin Elmer Model 154-B vapor fractometer, operating at
188-192C. and 25 p.s.i. and utilizing an 18-foot coiled 1/4 inch copper
tube packed with 30-60 mesh Tide detergent (F & M Scientific Corp.) was
employed for analyses of the decomposition products.
Before the column was used for analysis, it was "baked" in the
heating chamber for about 18 hours at a temperature of 180-190C. and a
helium pressure of 10 p.s.i. to remove excess liquid phase from the deter-
gent packing. The liquid phase comes off initially in a discontinuous
fashion, finally leveling off to a constant flow which continues through-
out the operational life of the column. If not removed prior to the
column's use, this initial effluent causes an erratic pattern of thermistor
response which interferes with the analysis.
- 23 -
The column was standardized by preparing separately sales of the
two isomeric cyclopropanes and 4-t-butylcyclohexanone (used as an internal
standard) of known concentrations and calculating the areas of the peak
on the v.p.c. chart for from four to ten injections for each sample.
Plots of each area vs. number of micromole of sale were then constructed
from these results. Reasonably good straight line relationships of the
peak area vs. number of micromole of sales were obtained (Figures 1, 2,
In analyzing the reaction products a weighed amount of 4-t-butylcyclo-
hexanone was added to a given amount of the decomposition product. The
mixture was dissolved in chloroform solution and was analyzed by gas
chromatography. The area ratio of 4-t-butylcyclohexanone to cyclopropanes
was measured and their mole ratio found from the standard curves. From
the weights of 4-t-butylcyclohexanone and the decomposition product, and
their mole ratio, the weight percentage of the cyclopropane in the decom-
position product was calculated.
In cases of analyses for the decomposition products of the 2-pyra-
zolines (XX) and (XX), it was found that the cis- and trans-cyclopropane
peaks were not cleanly separated. In order to solve this difficulty the
following approximation was used: the line was extended along the curve
on the v.pce chart and the area under the peak was measured to estimate
the marimm percentage of the cyclopropane as illustrated in Figure 4*
Under the experimental conditions (temperature 188-192 C., 25 p.s.i.
and utilizing an 18-foot column, etc.) the retention times of trans- and
- 24 -
I I II I
4 8 12 16 20
Figure 1 The standarization curve for the cis-cyclopropane (XVI).
I I I I I
4 8 12 16 20
Figure 2. The standarization curve for the trans-cyclopropane (XVII).
o 8 -8
I I I I i
4 8 12 16 20
Figure 3. -- The standarization curve for 4-_tbutylcyclohexanone.
- 27 -
Figure 4. An illustration for the estimation of the max. peak
cis-cyclopropanes ranged from 13 to 16 min. and 15 to 18 min. respectively.
For this reason when the decomposition products were analyzed for qualita-
tive purposes, pure trans- and cis-cyclopropane were added to a portion
of the product to ensure the correctness of the results of the analyses.
Decomposition of 3-methyl1-i -3 -dicarbomethoxay-5-phenyl-l-pyraoline (X) -
No cyclopropane was formed when the cis-pyrazoline (X) was decomposed at
several different temperatures ranging from 140 to 2600C Refluxing the
cis-pyrazoline (X) in xylene, nesitylene and decalin also did not give
cyclopropane. However, when the pyrazoline was decomposed at 280-29GC.,
it was decomposed to yield cyclopropane.
A sample of 0.240 grams of the cis-pyrazoline (X) with d.p. 137oC.
was introduced in a test tube preheated at 280-2900C. The residue weighed
0.180 grams. From the analysis by gas chromatography, the residue va
found to contain approximately 6 percent by weight of the cis-cyclopropane
(XI) and approximately 2 percent by weight of the trans-cyclopropane (XIII).
Gas chromatographic analysis,
Wt. of the decomposition product 180.0 ag.
Wt. of 4-t-butylcyclohexanone 14.0 mg.
ci-eyelopropane to 4-tbutylcyclohexanone 0.439
trans-cyclopropane to 4-t-butylcyclohexanone 0.014
cis-cyclopropane to 4-t-butylcyclohexanone 0.469
trans-cyclopropane to 4-t-butylcyclohexanone 0.116
It was further observed that under the above experimental condition
the mole ratio of cis- to trans-eyclopropane products was not independent
of the purity, as indicated by the melting point of the cis-l-pyrazoline.
The mole ratio of cia : trans cyclopropane was observed to be 9.4 : 0.2
when the decomposition point of the cis-l-pyraoline was 1330C.; 8 : 1
when the melting point was 1040C.; and as previously described, 6 : 2
when the decomposition point was 1370C.
Decomposition of 3-methyl-trn-3, 4-dicarbomethoxy-5-phenyl-1-pyrazoline
(XI). A sample of 0.0823 grams of the trans-pyrazoline (XII) was
decomposed at 180-1850C. The residue weighed 0.0784 grams. From the
analysis by gas chromatography, the residue was found to contain approxi-
mately 12 percent of the trans-cyclopropane (XIII). No detectable amount
of the cis-cyclopropane (XI) could be found. The cis-cyclopropane (XI)
was also found to be absent in the decomposition product of the crude
reaction mixture of the trans-pyrazoline (XII).
Gas chromatographic analysis.
Wt. of the decomposition product 137.1 ag.
Wt. of 4-t-butylcyclohexanone 25.1 ag.
trans-cyclopropane to 4-t-butylcyclohexanone 0*513
trans-cyclopropane to 4-t-butylcyclohexanone 0*513
Wt. percentage of the cyclopropane 12.2
Decomposition of the 2-pyrazoline (XX). A sample of 0.617 grams of the
2-pyrazoline (XX) was decomposed at 240-2600C. until 53.30 ml. (at 753 a*
and 280C.) of nitrogen had been evolved. The analysis of the decomposi-
tion product by gas chromatography showed it to contain a maxjzm of
78.5 percent by weight of the cis-cyclopropane (XI) and a maximim of
8.4 percent by weight of the trans-cyclopropane (XIII). Due to the
presence of an uncharacterized impurity in the decomposition product the
peaks of the cis- and the trans-cyclopropane were not cleanly separated.
Therefore, the approximation described in "General Procedure for Gas
Chromatographic Analysis of the Decomposition Product" was employed to
estimate the maxinam percentage by weight of the two cyclopropanes in the
The major cyclopropane product, the cis-cyclopropane, was isolated
from the decomposition product by liquid phase chromatography by using
acid-washed alumina as an adsorbent and 20 percent ether in pentane as an
eluent. The infrared spectrum of the isolated cyclopropane was identical
to that of the cis-cyclopropane (XI).
- 30 -
Gas chromatographic analysis.
Wt. of the decomposition product 22.8 mag.
Wt. of 4-t-butylcyclohexanone 5.1 ag.
cis-cyclopropane to 4-t-butylcyelohexanone 2.10
trans-cyclopropane to 4- -butylcyelohexanone 0.28
cis-cyclopropane to 4-t-butylcyclohexanone 2.22
trans-cyclopropane to 4-t-butylcyclohexanone 0.24
Decomposition of crude 2-pyrazoline (XXI). A sample of 1.19 grams of
crude XXI was decomposed at 200-250C. until 89.80 ml. (at 759.7 nm. and
25OC.) of nitrogen had been evolved. The decomposition product weighed
0.94 grams. The decomposition product was analyzed by gas chromatography
and the presence of both trans- and cis-cyclopropane were confirmed.
However, quantitative evaluation of the cis- and trans-cyclopropane
presented two difficulties. First, under the experimental conditions
employed, the cyclopropane peaks were not cleanly separated to alow
accurate determinations. Second, there was a peak with a retention time
very close to that of the internal standard, 4-t-butylcyclohexanone. The
last difficulty was overcome by the following method: the area ratio of
the trans-cyclpoproae to the interfering peak was determined. From this
ratio, the area of the trans-cyclopropane peak and the sum of the areas of
- 31 -
the internal standard and interferring peaks, the area of the interfering
peak was calculated and eliminated. The approximation method described in
"General Procedure to Gas Chromatographic Analysis of the Decomposition
Product" was used to overcome the first difficulty. In this way the
maxtinm weight percent of trans- and cis-cyclopropanes wre estimated to
be 22.6 percent and 8.8 percent respectively.
The major cyclopropane product, the trans-cyclopropane, was isolated
from the decomposition product (0.80 gram) by liquid phase chromatography
by using acid-washed alumina (17 grams) as an adsorbent and 20 percent
ether in hexane as an eluent. The residue (ca. 0.10 gram) obtained after
the evaporation of the solvent from the first four fractions (total volume,
99 al.) was confirmed to be trans-cyclopropane by comparison of its infrared
spectrum with that of trans-cyclopropane (XII).
Gas chromatographic analysis.
Vt. of the decomposition product 49.0 iag.
Wt. of 4-t-butylcyclohexanone 76.9 ar.
cis-cyclopropane to 4-t-butylcyclohexanone 0.196
plus interfering peak
trans-cyclopropene to 4-t-butylcyclohexanone 0.578
plus interfering peak
trans-eyclopropane to interfering peak 1.38
cis-cyclopropane to 4-t-butylcyclohexanone 0.354
trans-cyclopropane to 4-t-butylcyclohexanone 0.812
The non-stereospecificity of the decomposition of 3-methyl-cis-3,4-
dicarbomethoxy-5, 5-diphenyl-l-pyrazoline (V) reported by van Alphen has
been confirmed by the comparison of the decomposition of his pyrazoline
with that of isomeric 3-methyl-trans-3, 4-dicarbomethoxy-5, 5-diphenyl-l-
pyrazoline (VII). Both pyrazolines decomposed to give the same product.
3-Methyl-cis-3,4-dicarbomethox0y-5-phenyl-l-pyrazoline (XXII) and 3-
methyl-tranes-3,-dicarbomethowry-5-phenyl-l-pyrazoline (XXIII) have been
synthesized and the stereochemistry of the pheryl groups in these pyra-
zolines has been tentatively assigned as trans to the carbomethoxy group
on the C-4 atom of the pyrazolines.
The decomposition of the cis-1-pyrazoline (XXII) yielded 6 percent
by veiGht of the cis-cyclopropane (XVI) and 2 percent by weight of the
trans-cyclopropane (XVII), vwhle the decomposition of the trans-1-pyra-
zoline (XXIII) gave 12 percent by weight of the trans-cyclopropane (XVII)
and no detectable amount of the cis-i smer, indicating that the decompo-
sitions in these cases are stereoselective.
The geometrical configurations of the cyclopropanes resulting from
the decomposition of the 1-pyrazolines have been assigned as a result of
synthesis of the cyclopropanc with cis-carboalkoxy groups from the
anhydride, stereospecific syntheses of each cyclopropane from appropriate
2-pyrazolines and attempted anhydride formation from the di-acid resulting
- 33 -
from hydrolysis of XIII. The assigned configurations have been further
confirmed by n*m.r. analyses.
LIST OF REFERENCES
1. K. v. Amwrs and F. Kxnig, .Ann, 27 (1932).
2. T. L. Jacobs in R. C. Elderfield, "Heterocyclic Copouands." John
Wiley and Sons, Inc., New York, N. Y., Vol. 5, p. 80, 1957.
3. K. v. Auvers and F. Knig, Ann., 4, 252 (1932).
4. J. van Alphen, Ree. tray. da., 62, 334 (1942).
5. Jones, J. m. A. em. soc., 8, 6687 (1958).
6. w. M. Jowes, ibid., 81 5153 (1959).
7. G. L. Closs and L. E. Closs, ibid., 82, 5723 (1960).
8. W. H. Perkin, Br., 14, 2541 (1881).
9. L. I. Smith and K. L. Howard, "Organic Syntheses." John Wiley and
Sons, Inc., New York, N. Y., Vol. 24, p. 53, 1944.
10. W. H. Perkin, Ber., 14, 2542 (1881).
11. T. J. Deboer and H. J. Backer, "Organic Synthese." John Wiley and
Sons, Inc., New York, N. Y., Vol. 36, p. 16, 1956.
12. L. Edeleano, Ber., 20, 619 (1887).
13. N. B. Searle, "Organic Synthese." John Wiley and Sons, Inc., Vol.
36, p. 26, 1956.
14. T. Curtius and F. Goebel, J. prakt. hem. (2), 150 (1888);
Beilstein, 4, 340.
15. R. Stoermer and 0. voht, Ann., i49, 54 (1915).
- 35 -
Wun-Ten Tai was born January 24, 1932, in Taipei, Taiwan. He
entered the National Taiwan University, Taipei, Taiwan, in September,
1950, and graduated in June, 1954, with the degree of Bachelor of
Science. After serving one year as a full time teaching assistant at
the National Taiwan University, he entered the Graduate School of the
University of South Carolina in September, 1956, and graduated in August,
1958, with the degree of Master of Science.
In September, 1956, Mr. Tai entered the Graduate School of the
University of Florida. He held a graduate assistantship until June,
1960, and the remainder of his work was supported by a National Science
Mr. Tai is married and the father of one child. He is a member of
Gamma Sigma Epilon.
This dissertation was prepared under the direction of the chairman
of the candidate's supervisory committee and has been approved by all
members of that committee. It was submitted to the Dean of the College
of Arts end Sciences and to the Graduate Council, and vas approved as
partial fulfillment of the requirements for the degree of Doctor of
June 5, 1961
Dean, College of Arts and efence
Dean, Graduate School
UNIVERSITY OF FLORIDA
Illl I IIIIIII li 1 1 1 lill 11 111ii
3 1262 08553 7818