Oxidative Cleavage of Cyclopropanes Witpi Mercuric Acetate
This dissertation has been microfilmed exactly as received 69 -11,697
ROBINS, Richard Dean, 1942- OXIDATIVE CLEAVAGE OF CYCLOPROPANES WITPI MERCURIC ACETATE.
Tlie Ohio State University, Ph.D., 1968 Chemistry, organic
University Microfilms, Inc., Ann Arbor, Michigan OXIDATIVE CLEAVAGE OF CYCLOPROPAKES
WITH MERCURIC ACETATE
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of the Ohio State University
By
Richard Dean Rotins, B.A., M.S.
The Ohio State University 1968
Approved ty
Adviser Department of Chemistry DEDICATION
To life.
ii ACKNOWLEDGMENT
To Professor Robert J. Ouellette I extend my gratitude for the origination of this problem, his invaluable assistance during the course of this investigation, and above all for his sincere effort in aiding me to become a competent chemist.
I would also like to extend a special note of thanks to Dr. Aubrey
South.
Ill VITA
November 19; 1942 Born - N. Manchester, Indiana
1964...... B.A.; Manchester College, N. Manchester, Indiana
1966 ...... M.S., The Ohio State University, Columbus, Ohio
1964-1965 ...... Teaching Assistant, Department of Chemistry, The Ohio State University, Columbus, Ohio
1965-1968 ...... Research Associate, Department of Chemistry, The Ohio State University, Columbus, Ohio
PUBLICATIONS
''Oxidative Cleavage of Cyclopropanes. IV. Kinetics of the Cleavage of Arylcyclopropanes by Merciuric Acetate,'' J. Am. Chem. Soc., 90, 1619 (1968).
''1 ,3-Acetoxyl Participation in the Solvolysis of Organomercury Compounds (l),'' Tetrahedron Letters, 397 (1968).
FIEKDS OF STUDY
Major: Organic Chemistry
Studies in Organic Chemistry. Professors Melvin Newman, Harold Shechter, Leo Paquette, Gideon Fraenkel and Robert J. Ouellette
Studies in Physical Chemistry. Professors Frank Verhoek and George MacWood
Studies in Inorganic Chemistry. Professors Daryl Busch and Andrew Wojcicki
iv CONTENTS Page ACKNOWLEDGMENT...... iii
VITA ...... iv
TABLES...... vi
ILLUSTRATIONS...... vii
INTRODUCTION ...... 1
KINETICS OF THE REACTION OF MERCURIC ACETATE WITH PHENYLCYCLOPROPANES...... 6
COMPARISONS OF SELECTED HEAVY METAL ACETATES...... 24
1,5-ACETOXYL PARTICIPATION IN TÎIE SOLVOLYSIS OF ORGANOMERCURY C O M P O U N D S...... 59
THE CLEAVAGE OF OTHER CYCLOPROPANES BY MERCURIC ACETATE...... 43
EXPERIMENTAL...... 49
APPENDIX...... 68
REFERENCES CIT E D ...... 69
V TABLES
Table Page
1. Decomposition of Mercuric Acetate ...... 10
2. Rate of Cleavage of Substituted Arylcyclopropanes . . . 15
5 . Activation Parameters for the Cleavage of Arylcyclopropanes ...... 15
k. Effect of Addends on Rate at 5 0 ° ...... I8
5 . Rate Constants for Cleavage by Ion Pair at 25° .... 35
6 . Lead Tetraacetate Decomposition at 2 5 ° ...... 55
7. Solvolysis of Organomercury Compounds ...... 4l
8. Rate of Bicyclic Cleavage by Mercuric Acetate ...... 45
9* Bicyclic Type Cleavage from Thallium Triacetate .... 44
10. Relative Rates of Cleavage by Cleavage T y p e ...... 45
11. Relative Rates of Tosylate Solvolysis ...... 46
12. Activation Parameters for the Cleavage of Selected Bicyclics ...... 47
1 5 . Visible Spectroscopy D a t a ...... 67
VI ILLUSTRATIONS
Figui'e Page
1. Typical Second Order Plot at 50*1° for the Reaction of p-methylphenylcyclopropane with Mercuric Acetate ...... 9
2. Kinetic Plot for Mercuric Acetate Decomposition at 75° Assuming 0.33 Order Reaction...... 11
3. Hammett a"*" Plot at 50.1°...... 17
h. Ultraviolet Spectrum of Phenylcyclopropane and Mercuric Acetate in Methanol ...... 23
5- Free Perchloric Acid Dependence on ZPb^^ at O.O6 I F [HClO^lo and 0.02 M[Pb+4]o...... 30
6 . Graphical Determination of k for p-Methylphenylcyclo- propane at 24 . 9 ° ...... 32
7. Hammett Plot of log k'*' versus at 2 4 . 9 ° ...... 35
VI1 INTRODUCTION
The chemistry of cyclopropanes has been extensively investigated in recent years. This study deals with the reaction of electrophiles with highly strained substituted cyclopropanes. Most electrophilic reagents that react with olefins also attack cyclopropane bonds to give products resulting from ring cleavage. Many such reactions are summerized in
Lukina's^ review on the structure and reactivity of cyclopropanes.
Physical and theoretical chemists became interested in cyclopro- panes and suggested models to explain their reactivity.^ Coulson and
Moffitt^ revised previous theories by considering that the hybridization of the carbon-carbon bonds is different than that of the carbon-hydrogen bonds, and that neither is sp^ nor spS. Using this model Ingraham® calculated that the bent endo-ring orbitals are sp^'i® and the exo-ring orbitals, sp®‘®®. Ingraham suggested that these p-weighted internal cyclopropyl bonds are well placed for conjugation with unsaturated groups. Conjugation between cyclopropane rings and unsaturated substi tuents was first observed by Kizhner^ in I915 a,nd later by Robinson^ in
1916. The review by Lukina^ presents several examples of cyclopropyl
conjugation effects. Cyclopropane rings have also been shown to act as
proton-acceptor groups in hydrogen bonding.®
Other physical evidence has been presented demonstrating that
cyclopropanes are more similar to olefins than alkanes. Linnett^ found
1 the force constant for carbon-hydrogen stretching in cyclopropane is
5 .0 X 10^ dynes/cm as compared to those of methane and ethylene, being
H .79 X 10^ and 5 .I x 10^ dynes/cm, respectively.® The C^®-H coupling constants are a measure, of hybridization of that bond.® Muller^® and
coworkers found a coupling constant of 16 I cps for cyclopropane, com pared to 125 and 156 cps found for methane and ethylene, respectively.
Roberts^^ and coworkers reported that based on carbon-carbon coupling
constants the hybridization of the orbitals used in forming the internal bonds is sp® and that for the external bonds is sp^. An elementary
molecular orbital treatment of cyclopropane by Handler^® also suggests
that the external bonds are close to sp^ hybridization. A recent sum
mary of the description of bonding in cyclopropanes was given by
Bernett.^® The strain of the small ring is the ultimate reason for all
the above anomalies relative to most cycloalkanes. Based on strain free
cyclohexane, Seubold^'^' found the strain energy to be 2J.k kcal/mole.
The direction of cyclopropane ring opening in substituted cyclo-
propanes by electrophilic reagents is of interest. The ring cleavage
can be considered to involve initial electrophilic attack to give an
intermediate carbonium ion. Therefore, the stability of the generated
charge should determine which bond is attacked and the direction of 1 opening.
The cleavage of a carbon-carbon single bond as the result of a
direct bimolecud.ar reaction with an electrophile is fundamentally a
simple reaction. Such an attack is designated, 8^2 in the Hughes-Ingold 3 terminology. In this type of reaction the electrophilic reagent may be thought to displace a carbonium ion.
I I + I I — Ç— C— + E -> — Ç+ + — C— E
Of the many possible combinations of electrophiles and leaving groups
that can be envisaged as participants in Sg2 processes, only electro philic attack on carbon-metal bonds has been examined in any detail.
The only case where carbon-carbon single bonds have been shown to be
cleaved by electrophilic reagents is in compounds containing a cyclopro
pane ring.
Cyclopropane ring cleavage by reagents, now classified as electro
philes, to yield adducts has been known since the 19th century.The
cleavage process can be interpreted in terms of initial electrophilic
attack to produce an intermediate of carbonium ion-like character fol
lowed by addition of a nucleophile. The degree of synchronization of“
attack by the electrophile and the nucleophile is a subject of some
interest. In general the direction of cyclopropane ring cleavage is
thought to reflect the stability of the incipient carbonium ion.® Most
early investigations dealt with the problem of position of ring cleavage
as a function of substitution. It has been generalized that
Markovnikov's rule can be applied to the reactions of cyclopropanes as
well as of olefins. However, the generalization clearly has to be modi
fied to include the effect of other factors, such as ring strain and
steric accessibility to the reagent. Of these two factors only ring strain has heen examined in detail. In a thorough study of the acid- catalyzed addition of acetic acid to hicyclo[n. 1.0]alkanes LaLonde^”^ and covorkers observed that the extent of internal "bond cleavage increases with decreasing values of n. In a similar manner the cleavage of the same class of compounds hy thallium triacetate and lead tetraacetate has heen observed to exhibit the same trends.Levina^® has cleaved the hicyclo[n.1 .0 ]alkanes with mercuric acetate, hut the details of this reaction have not heen examined as extensively as with acid, thallium triacetate and lead tetraacetate.
The stereochemistry of the ring cleavage of cyclopropanes has heen examined only recently. LaLonde and coworkers have shown that addition of the nucleophilic solvent occurs with inversion at the cyclopropane ring carhon. Addition of solvent in the reaction with thallium tri acetate and lead tetraacetate also has heen shown to occur with inver sion. The general question of the stereochemistry of addition of the electrophile is largely an open one. While it appears likely that elec trophilic attack will occur with net retention of configuration at the
carhon to which the electrophile becomes attached, the postulate has not heen experimentally demonstrated in a conclusive manner.
Depuy^° and coworkers have recently examined the stereochemistry of
acid and base-catalyzed ring opening of cis-2-phenyl-1-methyl-cyclopro-
panol. In this very complete investigation an optically active sub
strate was shown to undergo an 8g2 reaction with protons and configura
tion was retained. With base an Sgl reaction occurs with inversion of
configuration. The cyclopropanols were chosen over cyclopropanes themselves because they have some apparent advantages. First, the reac tion occurs rapidly in 50:50 dioxane-water at 50°, a temperature which is less drastic than those conditions reported for the acid cleavage of alkylcyclopropanes in acetic acid. However, in light of both the differences in solvent and the presence of an aromatic ring, this advantage may be artificial. The second apparent advantage is that nor mal cyclopropane ring cleavages lead to several paths for further reac tion. In the case of cyclopropanols the stable conjugate acid of an aldehyde or ketone is produced. Cyclopropanes may be designed to afford almost the same advantages. Indeed, the presence of an aryl substituent is enough to provide specific attack at one of the cyclopropane bonds and lead to a resonance stabilized benzyl type carbonium ion. Since the concern is largely with the stereochemistry of the electrophilic addi tion, the ultimate fate of the center to undergo nucleophilic attack is not germane. While cyclopropanols promise to yield bountiful informa tion, the presence of the functional and potentially basic hydroxyl group may alter the fundamental process of simple carbon-carbon bond
cleavage by an electrophile. Thus it is felt that the more fundamental
cyclopropane derivatives should continue to be examined. KINETICS OF THE REACTION OF MERCURIC ACETATE
WITH PHENYLCYCLOPROPANES
In this study an important adjunct to the stereochemical experi ments presently available is reported. The nature of the Sg2 process can be elucidated by kinetic examination of the effect of substituents on the rate of reaction. Attachment of aryl groups to the cyclopropane ring simplifies the problem of direction of ring cleavage. Only one of the two different types of bonds in phenylcyclopropane and related com
pounds is cleaved by thallium triacetate and lead tetraacetate.The
reaction potentially involves displacement of a benzyl type carbonium
ion by the electrophile metal salt. The degree of deposition of posi
tive charge at the benzyl position can be detected by the response of
rates of cleavage as a function of substituents on the aromatic ring.
0— + M(OAc )^ M(OAc )
or
0 V y M(0Ac)%_2 + OAc
Metal salts have been chosen for this study instead of protons.
The advantage of metal salts is that a variety of structural modifica
tions are possible to alter the geometry and electrophilicity of these
6 reagents. The reactions proceed conveniently in the 25 to 75° temper ature range in acetic acid. Since the cleavage by thallium triacetate and lead tetraacetate is followed by a second step which could cloud the information desired, mercuric acetate was chosen as an experimen tally simpler reagent to use. Studies with mercuric acetate were
initiated after it became apparent that the results with the other salts would be more meaningful if the simpler reaction was examined also.
Kinetic Methods
The kinetics of the cleavage reaction were determined by measuring
the thiocyanate equivalence titer of reaction solution aliquots. Excess
standardized sodium thiocyanate was added to each aliquot and back
titrated with standardized silver nitrate. Mercuric acetate reacts with
two equivalents of thiocyanate per mole, and the adduct reacts -v/ith one
equivalent of thiocyanate per mole. Thus the thiocyanate consumed as
determined by back titration with silver ion is equal to
2(Ao -~X) + X = 2Ao - X, where Ao represents the initial molar
concentration of mercuric acetate and X is equal to the molar concen
tration of adduct. Therefore, the order of the reaction can be deter
mined by substitution into the appropriate integrated rate expressions,
as the concentration of the cyclopropane is available from the knovm
initial concentration and the stoichiometry of the reaction.
X-0-< + Hg(0 Ac}2 > X-0— k ^ HgOAc 8
For mathematical convenience egnimolar quantities of the cyclopro pane and mercuric acetate were utilized in most runs. However, runs involving excess arylcyclopropane also were carried out.
The reactions were run in sealed ampoules at 75° and, for the longer runs, at ^0°. In those cases where evaporation of solvent was not a problem, a single flask containing the reaction solution was used, and samples were 'vn.thdra'vm periodically by means of a pipette.
The reaction was quenched by plunging the sealed tubes into an acetone-
Dry Ice mixture. Samples -vn-thdra™ by pipette were delivered into
flasks cooled to 0°C.
A representative kinetic run is shovm in Figure 1 for equimolar
concentrations (0.053 M) of mercuric acetate and p-raethylphenylcyclo-
propane at 50 .1°.
For the reaction of p-methox}'-phenylcyclopropane and for some of
the runs with p-methylphenylcyclopropane it was necessary to use a
partitioned flask. Solutions of mercuric acetate and the substituted
cyclopropane in acetic acid were placed in separate compartments and
allowed to equilibrate prior to mixing the solutions. The flask was
shaken to mix the two solutions. In order to quench such reactions the
flask was plunged into an acetone-Dry Ice mixrture.
Stability of Mercuric Acetate
Mercuric acetate reacts in acetic acid to reduce its effective
thiocyanate titer by a factor of one half. While this side reaction
does not interfere with the cleavage of activated phenylcyclopropanes,
it is a major competing reaction with the deactivated compounds. The 110
100
% oI
20 10 20 Time(min.) Fig. 1.--Typical second order plot at $0.1° for the reaction of p-methylphenylcyclopropane with mercuric acetate 10 nature of the reaction of mercuric acetate in acetic acid has not heen examined in detail as it is not our first interest. Using the effec tive titer change of one half for infinite reaction the function
[SCN“] = X + 2(Ao - X) was evaluated where X represents the molar concentration of the mercury species which requires one equivalent of thiocyanate in the titration and Ao - X is the concentration of mercuric acetate. The apparent order of the reaction is quite low. Adequate straight line fits are obtained for either 0 .3 5 oi' 0 .2 5 order reactions for two to three half lives in the concentration range studied. An example of a plot obtained for 0.33 order reaction is given by Figure 2. The apparent rate constants at 50 o and 75° are given in Table 1 for an assumed 0.33 order reaction. These limited values provide sufficient information to allow a determination of the experimental limitations in a study of the cleavage reaction.
Also the mercuric acetate v/as converted to mercurous acetate during the reaction.
TABLE 1
Decomposition of Mercuric Acetate
[Hg(0 Ac)2 lo T k(mole^/^/l^/^ sec)
o -7 0 .0199 50.0 ° 2.0 X 10 -6 0.0215 75.0° 4.0 X 10
0.0417 75.0° 3.1 X 10"° 11.0
10.0
o I— ! 9.0 X a'ico <
I I I !______!______I I______!______I______* - 0 — Î 20 60 100 120 140 160 180 200 220 240 260 280 $00 Time(min.) Fig. 2.--Kinetic plot for mefcnric acetate decomposition at 75° assuming 0.$$ order reaction
H 12
The above results are in discord with those reported by hitching and Wells. They report the reaction of mercuric acetate in acetic acid to be first order in mercuric acetate and possibly first order in acetic acid with a rate constant of 1. 5 x 10 ^ sec~^ at 95°. They also suggest that the product is a polymer of mercurated acetic acid.
Stability of Phenylcyclopropane
In order to ascertain whether the reaction being studied was in fact a mercuric acetate induced cleavage of cyclopropanes and not a more
complex sequence of reactions, the stability of phenylcyclopropane was examined. Although an acetic acid catalyzed isomerization to olefinic compounds followed by oxymercuration was regarded as unlikely, this possibility was checked. A solution of phenylcyclopropane in acetic acid was sealed in an nrar tube and maintained at 75° for eight days.
During this time period the sample showed no evidence of appreciable deterioration. The regions of the nmr spectrum where olefinic protons
resonate remained blank. In addition there was no evidence of any
acetic acid catalyzed cleavage products such as 1 -phenylpropyl acetate. OAc 0— + HOAc — 0--
Stability of Phenylcyclopropane-Mercmic Acetate Adduct
The stability of the phenylcyclopropane-mercuric acetate adduct is
a necessity for the kinetic analytical scheme used to be valid. At
100 ° the adduct in acetic acid undergoes no noticeable change in the
thiocyanate titer after five days. Therefore the decomposition of the 13 adduct cannot complicate the kinetic analysis. Decomposition to yield mercury and cinnamyl acetate does occur slowly at 135° with k = 8 .6 X 10 sec ^ .
■HgOAc > 0— xs /— OAc + Hg
For comparative purposes, the stability of $-phenylpropylmercuric
acetate in acetic acid was examined under the same conditions as those
used for the adduct. The rate of reaction is approximately 20 times
slower. Therefore it is likely that the ^-acetoxyl group in the adduct
assists in the decomposition. Such 1,3-acetoxyl participation serves as
a rationale for the decomposition product.
Ichikawa^^ and coworkers reported mercuric acetate increases the
decomposition rate of benzylmercurie acetate in acetic acid-water-
perchloric acid solvent systems. Therefore it was considered a possi
bility that mercuric acetate might react with the adduct formed in the
cleavage step. The adduct of mercuric acetate and phenylcyclopropane
was heated in the presence of mercuric acetate at 75° in acetic acid.
The thiocyanate titer decreases with time from three to two equivalents.
However the rate of this change corresponds exactly to the decomposition
reaction of mercuric acetate in acetic acid. By subtracting the titer
equivalent of the adduct from the total titer, only a titer change
corresponding to the mercuric acetate decomposition is indicated.
Therefore a second order decomposition of the adduct by mercuric acetate
in a manner analogous to the reaction observed by Ichikawa with benzyl Ik mercuric acetate does not occur under the experimental conditions.
OAc 0------HgOAc + Hg(OAc)s-----> HggOAcg + organic products
The analytical scheme used in this study is not complicated by the destruction of the adduct.
Results
The second order rate constants for the cleavage of the six aryl cyclopropanes studied are listed in Table 2. The temperatures employed were 25°, 50° and 75°• Each rate constant reported is the average of two or more rate constants obtained from two or more runs. Individual rate constants were reproducible to the extent of 2 to ^ . Equimolar quantities of mercuric acetate anc cyclopropane were used for the p-Meo, p-Me, m-Me and H compounds. However, it was necessary to use excess
cyclopropane for the p-Cl and m-Cl compounds to allow the rate of
cleavage to exceed the rate of decomposition of mercuric acetate.
The activation parameters, AH^ and A8^, for the compounds studied
are listed in Table 5 . The enthalpies of activation were calculated
from a plot of In k/T versus l/T by the method of least squares. While
the precision of the rates is approximately 2/0 , their accuracy is un
known. However, for an accuracy of 5^ the reported enthalpies of acti
vation are accurate to ±0 .^ kcal/mole over the fifty degree range
studied. The free energy of activation was calculated from the 50° data
and used with the enthalpy of activation to obtain the entropy of acti
vation for each compound. 15
TABLE 2
Rate of Cleavage of Substituted Arylcyclopropanes
Substituent 25.0 50.1 75.6
P-CH3O 5.7 X 10"^ 5.0 X 10“^ ■I - ' 1
P-CHg 2.2 X 10"® 2.2 X 10 “® 1.7 X 10“i
ITI-CH3 5.5 X 10"4 4.0 X 10 “® 3.7 X 10 “®
H 1.6 X 10"^ 2 .2 X 10 “® 2 .1 X 10 “®
p-Cl 4.7 X 10"5 7.0 X 10“4 7.5 X 10“®
m- Cl 6.7 X 10“® 1 .2 X 10“4 1.6 X 10 “®
TABLE 3
Activation Parameters for the Cleavage of Arylcyclopropanes
Substituent (kcal/mole) • A8^ (cal/deg mole)
2 -CH3O 16.0 -10.6
P-CH3 17.0 -13.6
m-CH3 18.5 -12.4
H 19.1 -11.7
2-Cl 1 9 . 9 -11.4
m-Cl 21.6 -9 .8 16
A Hammett type plot for the rate constants at 50° versus is shown in Figure J. The calculated response is p = -5-2 with a correla tion coefficient of r = O.9 8. Although the correlation is not exceed ingly good, the correlation with the alternate parameter o is poorer.
Clearly the magnitude of the response of rate to the substituent die- tates the use of c , The rate for p-methoxyphenylcyclopropane deviates to the largest extent from the correlation line. It may be that the p-methoxyl group is not as effective an electron donating group in the reaction medium. One possibility that was considered to explain the relative slowing of the p-MeO compound was a complex formed between mercuric acetate and the methoxyl group. Such complex formation should hinder the cleavage reaction. Introduction of anisole into the reaction of p-methylphenylcyclopropane and mercuric acetate at 50 ° does not
alter the rate of cleavage. It appearss that any decrease in the elec-
tron donating ability of the methoxyl group is the result of the
solvent itself. Acetic acid may retarcd partially the methoxyl group
from donating electrons by a specific solvent-solute interaction at the
methoxyl group.
The effect of added, lithium acetate on the reaction is unimportant.
The observed rate constant for p-methylphenylcyclopropane and mercuric
acetate is given in Table k and compared with the rate constants ob
served. in the presence of lithium acetate and anisole. 0.0
- 1.0
- 2.0
bO O
-4.0
- 0.8 - 0.6 -0.4 - 0.2 0.0 +0.2 en- + Fig. 3.--Hammett O' plot at 30.1° 18
TABLE U'
Effect of Addends on Rate at 50°
Mercuric p-Methylphenyl- Lithium Acetate Anisole k(l/m sec) Acetate(m/l) cyclopropane (m/l) (m/1 ) (m/l)
0.0334 0.0334 0.0 0.0 2.20 X 10“^
0.0398 0.0398 0.0 0.0398 2.26 X 10"^
0.0352 0.0352 0 .l4l 0.0 2.20 X 10 “^
Conclusions
The observed kinetic order for the cleavage of cyclopropanes by mercuric acetate and the known electrophilic nature of mercuric acetate
suggest that an 8^2 reaction is involved. In addition the magnitude of
p indicates the deposition of a large positive charge at the benzylic
carbon atom in the transition state. The p for the cleavage reaction
approaches the p for the solvolysis of aryldimethylcarbinyl chlorides + which Brown^^ used to establish the a scale.
The electrophilic reagent must be mercuric acetate itself or some + related ion pair. If HgOAc were the attacking reagent then a mass law
effect should have been observed with added acetate ion. Another in
teresting possibility which seems to be precluded on the basis of the
same experiments is the formation of Hg(OAc)^ . Mercury (ll) forms
tetrahedral complexes readily in aqueous solution. Such complex forma
tion in acetic acid would lead to a dianion which is isoelectronic with
the well known oxidizing agent lead tetraacetate. On the basis of 19 orbital availability of the covalent mercury (ll) salts as compared to
Pb(lV) compounds it would be expected that mercury (ll) acetate would be a more effective electrophile than lead (IV) acetate. Formation of
Hg(OAc)^ ^ should decrease the rate of cleavage of the cyclopropane
ring as the electrophilicity of the mercury (ll) has been decreased.
The formation of Hg(0Ac)3~ in acetic acid would be anticipated to be
more favorable than that of Hg^OAc)^"^ on the basis of the poor ion
solvating character of the medium. However, the formation of Hg(0Ac)3"'
appears unlikely as it is unreasonable to expect Hg(0Ac)2 and Hg(0Ac)3
to be of identical electrophilicity. Since the rate of cleavage is not
affected by added acetate ion the species Hg(OAc)^ ^ and Hg(0Ac)3 ^
probably are not formed in acetic acid at the concentrations investi
gated. The above arguments are largely speculative but must be
considered when comparisons are made with salts of T1 (ill) and Pb (IV).
Tv/o solvent molecules per molecule of mercuric acetate could be
attached to mercury forming a four coordinate mercury (ll) electrophile.
If this is the case, the mercury species could serve as an electrophile
only if it releases one ligand in the transition state. Displacement
of solvent would be expected to be more facile than displacement of
acetate ion.
The transition state for cleavage of phenylcyclopropane by mer
curic acetate is postulated to resemble structure 1. 20
HOAc I 6;%( 0Ac)g(H0Ac)
The magnitude of p dictates the deposition of considerable positive charge at the benzylic position. This in turn indicates that the carbon-mercury bond is significantly formed in the transition state and that the mercury atom is not symmetrically situated with respect to the edge of the cyclopropane ring. Replacement of the ligand acetic acid in the transition state is more palatable than acetate ion although this choice is speculative and the magnitude of the selectivity of such
a competitive process cannot be shovm. Bimolecular processes often are
associated with a large negative entropy of activation. In the case of
cleavage of arylcyclopropanes, the entropy of activation is not very
high and it is possible to rationalize this fact by postulating that a
ligand becomes effectively dissociated from mercury in the cleavage
step. This postulate is in agreement with the degree of positive
charge generated at the benzylic carbon and the extent of carbon-
mercury bond formation.
There are similarities between the cleavage of cyclopropanes by
mercuric acetate, which may be termed an oxymercaration, and the deoxy-
mercuration reaction investigated by Kreevoy.^® Oxymercurated adducts
of olefins have been deoxymercurated in water. The alkylsubstituted 21 adducts exhibit a p* = -2 .8. Although the solvent systems and struc tures of the compounds in Kreevoy's and this work are different there
are great similarities in the electronic requirements for the two reac
tions. The transition state suggested by Kreevoy is represented as
structure 2 and our postulated transition state is a homologated ver
sion of Kreevoy's.
OHg
R - Ch— CHa"' \ l 'HgX
levina^^ has determined that complexes form between mercuric ace
tate and cyclopropanes which do not involve bond cleavage. The conclu
sion is based on conductivity measurement and the calculated equilib
rium constant is very low. Under the kinetic conditions employed in
this study such a complex would constitute less than 1'^ of the total
species present. However, such complex formation could be accomodated
in the kinetic scheme by placing the complex in equilibrium with the
starting materials. The complex could be transformed into the acti
vated complex, or free starting material could proceed to the
transition state via an associative process which is different from
that involved in complex formation.
Cyclopropane complex formation has been reported in the case of
chloroplatinic acid.^® However, the apparent complex has been shown
to have resulted frati cleavage of the cyclopropane ring.^® 22
It was considered of interest to determine if complex formation with mercuric acetate might be detected. The ultraviolet spectrum of phenylcyclopropane and mercuric acetate in methanol is identical to that calculated from the summation of the spectrum of each compound.
These observations are illustrated in Figure . While this data does
not eliminate in a rigorous manner the presence of a complex without
ring cleavage, it is suggestive that such a possibility is not an
important consideration in the cleavage reaction studied. 0.8
0.6
rH CO s A < 0.4 o M A A O
0.2
0.0 200 250 WAVELENGTH, mU î^g. 4 .-- ___ = Ultraviolet spectrum of phenylcyclopropane and mercuric acetate in methanol; o = Sum of mercuric acetate and phenylcyclopropane spectra in methanol
ro 'OJ COMPARISONS OF SELECTED HEAVY METAL ACETATES
One of the reasons for the initiation of this study was to compare kinetically the heavy metal acetates; Fb(OAc)^; TlXOAc)^, and Hg(0Ac)2
in their reactions with cyclopropanes. Since attack by an electrophile
could he regarded as an oxidation, the order of electrophilicity might he predicted to he Fb(0Ac)4 > Tl(OAc)s > Hg(0Ac)2. However this order
neglects the structures of the salts which may he important in kinetic
comparisons. Quantitative studies of the electrophilicity of the ace
tates of Hg(ll), Tl(lll), and Pb(lV) in a common system would provide a
much needed comparison of the salts.
The second order rate constants for the cleavage of phenylcyclopro
pane hy mercuric acetate, thallium triacetate and lead tetraacetate at
approximately 25° are 1.6 x 10"'^, 2.0 x 10”^ and J.l x 10 ^ l/m/sec,
respectively, corresponding to relative rates of 1, 12 and 0.019,
respectively. Therefore, mercuric acetate and thallium triacetate
appear to he somewhat similar in terms of their relative electrophilic-
ities, whereas lead tetraacetate reacts at a significantly slower rate.
Regardless of the substrate or temperature thallium triacetate is a
better electrophile than mercuric acetate.
Mercuric acetate and thallium triacetate are similar in that they
both can act as Lewis acids, for they are coordinatetively unsaturated
if acetate is considered as a monodentate ligand. In acetic acid both
2 k 25 compounds may be four coordinate with the solvent occupying the remain ing coordination sites. When these two salts serve as electrophiles it
is likely that acetic acid would he replaced in preference to acetate
ion. Therefore, it is convenient to consider them coordinatively un
saturated and able to act as effective electrophiles with respect to
the cyclopropane bonds as base. The order of reactivities of mercuric
acetate and thallium triacetate probably is related to the higher oxi
dation state or nuclear charge of the latter compound.
On the basis of either oxidation state or nuclear charge the
electrophilicity of lead tetraacetate would be expected to be greater
than that of thallium triacetate. liovrever, there are important struc
tural considerations. Lead tetraacetate is coordinatively saturated
and would have to lose an acetate ion in order to incorporate the
cyclopropane in a coordination position. Thus the required loss of the
acetate ion should decrease the electrophilicity of lead tetraacetate.
In the above discussion only the covalent compounds have been
considered as active reagents. For each compound one or more related
free ions derived from ionization of acetate ions could be considered.
However, in acetic acid their concentration should be quite low.
Furthermore the absence of a common ion effect with added lithium ace
tate for both mercuric acetate and lead tetraacetate cleavages indicate
that free ions such as HgOAc’*' and Fb(OAc)s'^ are not kinetically im
portant species. The absence of a common ion effect in the case of
mercuric acetate also eliminates the formation of such species as
Hg(0Ac)3~ or Hg(0Ac)4~^ the latter being isoelectronic with lead 26 tetraacetate. Howeverj the rate of cleavage hy thallium triacetate is notahly depressed upon the addition of added excess lithium acetate.3°
It is likely that the species Tl(OAc)^ is formed which derives stabil ity from a completed octet.
It could be argued that ion paired metal cations are important contributors to the cleavage reactions. Neglecting coordinated solvent molecules for the sake of clarity, a general kinetic scheme such as that depicted below is consistent with the experimental facts obtained initially.
Ki M ( O A c _> [M(OAc)x_ 1 OAc]
A r - < ^ + r.M(OAc)^_^ OAc] products
Ar— + M(OAc )^ products
The apparent rate constant would be k^K^ if the ion pair is the sole reacting species (k'^'Kj^ » k°). Any interpretation of the rates of
cleavage of cyclopropanes would be complicated by changes in both
and k’*'. In the case of mercuric acetate and thallium triacetate there
is no compelling reason for utilizing ion paired electrophiles, as the
compounds are coordinatively unsaturated. While the ion pairs would be
expected to be more reactive than covalent material the magnitude of + should more than counterbalance the high value for k with respect
to k°. Kresge®^ and coworkers have calculated that the rate of mercura
tion of benzene by HgOAc^OAc" is 10 ’^'^ faster than by Hg(0Ac)2. However 27
is estimated to te 2 x 10"^, and it can be concluded that the mercu- rating agent responsible for aromatic mercuration is un-ionized mercu ric acetate. In light of this information it appears unlikely that ion pairs are responsible for the cleavage of cyclopropane in the case of mercuric acetate. This conclusion depends on the assumption that the ratio of relative electrophilicities of Hg(0Ac)2 and [HgOAc"^] in the
cleavage of cyclopropanes is comparable to the ratio for aromatic
mercuration. Fortunately the estimated low value for provides a
comfortable margin for any uncertainty in the possible variance of
relative electrophilicity ratio for covalent and ion paired mercuric
acetate.
Thallium triacetate in pure acetic acid most likely reacts as a
covalent entity. The formation constants for thallium (ill) complex
Ions are larger than for mercur-y (ll) in those cases where identical
ligands are compared.Therefore Kj_ for thallium triacetate must be
less than 1 0 "®, In order for [TlXOAc)^^ OAc"] to compete kinetically
with Tl(OAc)a the electrophilicity ratio for k^/k° would have to be in
the order of 10® and be dramatically different from the corresponding
electrophilicity ratio of 10^ for meicuric acetate. Such a change is
extremely unlikely. It might be argued that the electrophilicity ratio
in the case of thallium would be expected to be smaller than that in
the mercury case due to the fact that the thallium ion pair has two
coordinatively bonded acetates which may stabilize the formal positive
charge whereas the mercury ion pair has only one such coordinatively
bound acetate. 28
ïiead tetraacetate is coordinatively saturated and therefore might he expected to react hy an ion pair intermediate. The cyclopropane moiety could then he accepted in a coordination position. The low rate of cleavage hy lead tetraacetate relative to thallium and mercury could he attributed to the contribution of as = k"*'Kj_.
Experiments were designed to obtain k , the rate of cleavage of phenylcyclopropane hy Fb(0Ac)30Ac". Brown and Kresge^^ report the generation of HgOAc^ClO^ from mercuric acetate and perchloric acid in glacial acetic acid. The conversion to ion pair is essentially com plete under the conditions employed. The same approach was applied to / \ + lead tetraacetate. By generating PbCOAcjsClO^ , the equilibrium
constant for ion pair formation for lead tetraacetate in acetic acid,
could he calculated from the observed rate of lead tetraacetate
cleavage if it proceeds through the ion pair. The assumption is made
that Fb(0Ac)3C10% reacts in the same manner as Fb(OAc)3OAC and that
they have the same electrophilicities.
The reaction may he follovred hy noting the change in Pb"^^ titer.
The rate expression is:
■ S " k'^K^Ca - x)(x) 1 +^Kf[HC104]
where) x = EPb’*'"^, a = and Kf is the equilibrium constant for
ion pair formation for Pb(0Ac)4 + HCIO4 in glacial acetic acid.
The expression as it stands cannot he integrated because the
perchloric acid concentration is not constant throughout the course of 29 the reaction at the perchloric acid concentrations employed. However^ the free perchloric acid concentration may he written in terms of the total lead (iV) present at any given time. This is done hy solving the five following equations containing five unknowns simultaneously for an initial lead tetraacetate concentration of 0.02 M at a particular formal perchloric acid concentration.
^ ~([Ph(0Ac)4.] - x )([HC104] - x) where: x = [Ph(0 Ac)^^10 %]
Ph(OAc)srokUAc/s ______y y______^ ([Ph(OAc)s] - y)([HC104] - y)
where; y = Ph(OAc) CIO4 and...4 IT Pb(0Ac)2 ^ is, the equilibrium constant for
ion pair formation for Pb(OAc)g + HCIO4 in glacial acetic acid.
= [HCIO^] + X + y (4)
gPb"^^ = [Ph(0Ac)4] + X (5 )
ZPh^2 = rPb(OAc)s] + y (6)
The free perchloric acid concentration was calculated at 0 , 25, 50, 75
and lOC^ reaction. The free perchloric acid concentration was plotted
against the concentration of EPh’^^. The hest straight line for these
points was calculated. (Figure 5)* The equation of this line relates
the free acid concentration as a function of EPh^^ which is the experi
mentally observed quantity. CM O
4 .6
4 .4
4 .2 0.02 00.010A. 0.005 0.000 ^ Pb+%1 Fig. 5 -— Free perchloric acid dependence on Spb''^-' at O.O6I F [HClO^j and 0.02 M [pb J
Vj4 o 31
For the general ease then;
where b = [HC104 ]initial ~ [HC104 lfinal (EFb initial and c = [HC104 ]final.
Equation (l) may now be written as
The integrated form of this expression is
ab + c * bx + c gSb + ac
db / d tk K.p abc + c2 (bx to)-— log (x) = + constant (9) where d = 1 + cK^. The equilibrium constants associated with ion pair
formation in the presence of perchloric acid for Fb(0 Ac)4 and Fb(OAc)2
were obtained from visible spectroscopy measurements and were found to be
approximately 25 and 25O, respectively (see experimental).
Values for k ’*' were obtained from the initial slopes of plots of the
appropriate function of eq. (9) versus time (Figure 6). The values
obtained for k ’*' as a function of cyclopropane type, cyclopropane concen
tration and perchloric acid concentration are listed in Table 5 * 280 ON
270
250
210 10 12 Time(min.) Fig. 6.--Graphical determination of k"^ for p-methylphenylcyclopropane at 2 4 .9= 55
TABLE 5
Rate Constants for Cleavage "by Ion Pair at 25 °
Substituent [A]o Fb(0Ac)4 HC104 formal X 103 m/l m/l m/l l/m/sec
H 0.2211 0.0195 0.0610 7.84
H 0.1276 0.0196 0.0610 8.16
H 0.2461 0.0186 0.0970 9.65
H 0.1068 0.0208 0.0970 10.07
2-CHs 0.1757 0.0195 0.0610 19.17
m-Cl 0.5200 0.0186 0.0610 2.58
Stability of Lead Tetraacetate
The stability of lead tetraacetate was examined under the reaction conditions employed for the kinetic studies. The results are summarized
in Table 6.
TABLE 6
Lead Tetraacetate Decomposition at 25 °
Fb(0Ac)4 HC104 formal Min io Change in Fb'^'^ m/l m/l
0.0240 0.15 50 10
0.0265 0.57 50 9
0.0209 0.10 15 0 3h
O 4" The substituent response at 25 is given hy p = -1-3 when log k is plotted versus a ’*' for H and m-Cl (Figure j). The value of k"*" obtained for ^-methoxylphenylcyclopropane is much larger than expected and may not be correlated with the other rates. It is likely that another process is taking place in the g-OMe case. One such possibility is that the cyclopropane may be cleaved by acid initially to generate olefin which would undergo immediate attack by lead tetraacetate.
The substituent response found for the cleavage of substituted phenylcyclopropanes by lead tetraacetate in the absence of perchloric acid at 25 ° is given by P = -1 .9.^^ The differences in the response of rates to substituents in the presence and absence of perchloric acid may be accounted for in two ways. It may be that an ion-paired intermediate is involved in both cases and that Fb(0Ac)3ClG4 and Fb(OAc)gOAc react in different manners and that they have different electrophilicities. An alternative explanation is that different mechanisms are involved in the presence and absence of perchloric acid.
The electrophilicity of Pb(0Ac)3C104 may be greater than that of
Pb(0Ac)30Ac" for the perchlorate ion would not be expected to be as ti^tly held as the acetate ion. Therefore, reaction involving
Fb(OAc)3010% would likely exhibit a lower substituent response to cleav
age due to its less descriminating attack. If an ion-paired intermediate
is involved, then K%. in the absence of perchloric acid may be calculated 4" 4~ from the observed rate, k^^g and k since k^-yg = k Although the
value of k"*" must be obtained for the reaction of Pb(0Ac)3C104 and not
for Pb(0Ac)30Ac“, they ought to be of the same order of magnitude. The 35
+0.2
0.0
- 0.2
-0 .4 % bO O 1-1
- 0.6
- 0.8
- 1.0
-0.40 - 0.20 - 0.00 + 0.20
Fig. Y.--Hammett plot of log k"*" versus d at 24 .9' 5 6 value of in the absence of perchloric acid is 4.2 x 10"^ for k"*" = 8.0 X 10"® l/m/sec and kg-j^g = 3 *5^ x 10"® l/m/sec if ion pair inter mediate is involved. The formation constants for lead (iv) complexes
ought to be larger than for thallium (ill) and mercury (ll) complexes in those cases where identical ligands are compared.®® Therefore, Kj_ in the
absence of perchloric acid must be less than 10”® reported by Brown and
Kresge®^ for mercuric acetate. Furthermore, the equilibrium constants
for the reactions of Fb(OAc)^ and Hg(OAc)g with perchloric acid are about
the same (eq. (lO) and (ll).
Pb(0Ac)4. + HC1 0 4 ^ = = ^ Pb(0Ac)3 C104 + HOAc K^. = 23.? (lO)
Hg(0Ac)2 + HCIO4 Hg0Ac"^C10% + HOAc = 25.5 (u)
The ease with which lead tetraacetate loses an acetate ion and accepts a
perchlorate ion is about the same as for mercuric acetate to lose an
acetate and accept a perchlorate ion (eq. (l2 )).
Fb(0A c )4 + HgOAc’^ClO; Fb(0Ac)Jci0; + Hg(OAc)s K = 1.0 (l2)
In the absence of perchloric acid one may -v.uite:
Pb(0A c )4 + HgOAc'^OAc" Pb(0Ac)j0Ac" + Hg(0A c )2 K = 1.0 (1 3 )
Therefore, the equilibrium constant for ion pair formation of Pb(0Ac)4
in competition with Hg(0Ac)2 in the absence of perchloric acid is nearly
unity. If the value obtained by Brown and Kresge®-’- for the equilibrium
constant for the ion pair formation from mercuric acetate in the absence 57 of perchloric acid is correct then the equilibrium constant for ion pair ~ -s formation from lead tetraacetate should be 10 . Therefore, ionization
of FbLOAc)^ does not occur prior to cleavage of the cyclopropane. While
ion pair is not rigorously excluded as the reactive intermediate for
Fb(0Ac)4 , both the difference in substituent response for Fb(OAc)4 and
Pb(0Ac)3 C104 and the apparently unreasonable value for ion pair forma
tion from lead tetraacetate in the absence of perchloric acid indicates
that another reaction pathway is probably involved.
One might envisage a process as shown by structure 5 in which an
acetate ligand is displaced in concert with the introduction of the
cyclopropane moity into the coordination sphere of lead (iv).
Fb(OAc)s OAc
i
While lead tetraacetate has distinguished itself as a versatile and
povrerful reagent in the field of organic chemistry for nearly fifty
years, the mechanisms of its reaction have remained largely an enigma
except for the cleavage of diols.®^ The mode of reaction is strongly
dependent on the substrate and reaction conditions. There have been
numerous product studies of oxidative processes involving lead tetra
acetate which unfortunately have provided little insight into'the
nature of the reacting species. Many of the product studies did not
utilize modern techniques. In spite of the possibility of undetected
reaction products, the mechanisms of reaction and the exact nature of the 58 attacking species have heen -widely discussed.®^ Among the many active intermediates postulated, Fb(OAc)3 has received the most attention.
This species has heen ruled out for the reaction discussed above and its possible role in other reactions is no\T made more tenuous. 1,3-ACEïOXYL PARTICIPATION IN THE
SOLVOLYSIS OF ORGANOMERCURY COMPOUNDS
In a report on the stereochemistry and direction of cleavage of hicyclo[n.l.O]alkanes by lead tetraacetate and thallium triacetate^®
it was suggested that 1,3-acetoxyl participation might be important in
the decomposition of the intermediate organometallic derivatives of the
general structure k.
ÇH3 'C. O'' ' ^ 0
-M(OAc )X - 1
Such a 1 ,3 -acetoxonium ion intermediate derived from the solvolysis of
the carbon-metal bond seemed to be the most likely model with which to
account for the observed stereochemistry in the two step oxidative cleav
age mechanism proposed. The analogy between 1,2-acetoxyl and 1,3-ace-
toxyl participation has been demonstrated by isolation of 1,2-acetoxonium
salts.A six center-transition state would be formed in the case of
1 ,3 -acetoxyl participation. This transition state should be easily
achieved. A 1,3-acetoxonium intermediate has been postulated to account
for the rearrangement of 3,4,6-tri-0-acetyl-«-D-glucopyranose-l,2-0-
acetoxonium-hexachloroantimonate which when treated with alcohol in
5 9 ko pyridine gave 1 ,2 ,$-tri-0 -acety1-^,6 -0 -(l-ethoxy-ethylidene)-cc-D- idopyranose.®^
The work of Dolby^^ and coworkers in which acetolysis of trans-2- acetoxymethylcyclohexyl hrosylate-carbonyl-0 ^® yields cis-diacetate con taining nearly all of the 0 ^® may be attributed to 1 ,3 -acetoxyl partici pation.
In order to elucidate the decomposition pathway of the Y -acetoxyl organometallic intermediates obtained in the cleavage of cyclopropanes by metal salts, it was determined to seek kinetic evidence for 1 ,3 -acetox- onium ions. The net stereochemical consequences of the two step cleav age- oxidation pathway suggested could be interpreted by combination of steps other than those proposed. To date we have been unsuccessful in
our attempts to isolate the postulated organolead and organothallium
derivatives of the cleavage reaction by lead tetraacetate and thallium
triacetate. Therefore we turned to the corresponding mercuric acetate
adducts.
The adduct II of p-methoxyphenylcyclopropane and mercuric acetate
was isolated in $4^ yield from the reaction in acetic acid.
pAc CK3 -O— + Eg (OAc) 2 --> H3 C-O-/ O L ^ HgOAc
II
The structure of the compound was confirmed by elemental analysis, nmr
and reduction with sodium borohydride to 1-(p~anisyl)-propyl acetate. Ill
Similarly the adduct of phenylcyclopropane and mercuric acetate was pre pared and its structure confirmed hy elemental analysis and nmr spectro scopy. For comparative purposes, hutylmercuric acetate and 3-phenyl- propylraercuric acetate were prepared hy conventional means.
The rates of solvolysis of the three acetates at 135° and two perchlorates at 100° in acetic acid are listed in Tahle J.
TABLE T
Solvolysis of Organomercury Compounds
Compound k(sec"^) T(°C)
CHsfCHsïsHgClOa 4.8 X 10-5 102°
0CH2(CH2)2HgClO4 2.4 X 10"^ 101°
0CH2(CH2)2HgOAc 4.7 X 10"® 135° 0CH(OAc)(CH2)2HgOAc 8.6 X 10-7 135°
jp-MeO0CH( OAc ) ( CII2 ) 2HgOAc 4.4 X 10-4 135°
That hutylmercuric perchlorate solvolyzes at a faster rate than 3"
phenylpropylmercurie perchlorate indicates that aryl participation is not
important. This conclusion was arrived at hy Winstein®^ when he observed
that Ari-4 ''participations compete very poorly in solvolysis of cu-aryl-
1 -alkyl arylsulfonates in acetic or formic acid solvents.''
The 3"Phenyl-3-acetoxypropylmercuric acetate solvolyzes l8 times
faster than does 3-pbenylpropyl mercuric acetate. This rate acceleration
suggests that 1 ,3 -acetoxyl participation is important in the solvolysis
of the primary mercuric acetate derivatives. Furthermore, an additional
510 fold rate increase for 3 -(p-anisyl)-3 -acetoxypropyl mercuric acetate 1|2 over the analogous phenyl derivative substantiates the existence of a
1 ,3 -acetoxyl intermediate. Such a rate enhancement corresponds to p =-3.6 at 135° if a correlation is applicable. Clearly the develop ment of a substantial positive charge at the benzylic carbon atom in the transition state necessitates the formation of a structure in which the acetoxyl group is bridged from the benzyl to primary position. The sole organic product observed in the solvolysis of 3"Phenyl-3-acetoxypropyl- mercuric acetate is cinnamyl acetate and may be readily accounted for as shown in structure 5 by the proposed participation.
ÇH3
' ' 6” ô J X J r ---- Hg(OAc)
H THE CLEAVAGE OF OTI-EER CYCLOPROPANES
BY MERCURIC ACETATE
Bicyclo[ n. 1.0] alkane s
The bicyclo[n.l.O]alkanes where n = 4,$,2 were examined. As
expected, bicyclo[2 .1 .0]pentane reacts very rapidly with mercuric ace
tate. Its rate at. 25° was not measurable by employing any of the above
mentioned techniques. The reaction is over in seconds. A sample of the
compound was provided by Dr. Kevin Mansfield and Dr. Paul Gassman. A
priori one might expect that on the basis of strain bicyclo[3.l.O]hexane
would cleave more rapidly than bicyclo[^l-.1.0]heptane. The data obtained
at 50.1° and 24.$° are listed in Table 8 .
TABLE 8
Rate of Bicyclic Cleavage by Mercuric Acetate
Compound k l/mol/sec C°
Bicyclo[4 .1 .0 ]heptane 2.92 X 10"° 50.1
BicycloC3.1.0]hexane 3.42 X 10"4 50.1
BicycloC 4.1.0]heptane 2.25 X 10"4 24.9
BicycloC 3'1.0]hexane 3.77 X lO'S 24.9
Bicyclo[4.1.0]heptane reacts about 8.5 times as fast as
43 kh bicyclo[3 -l-0]hexaine at 50°.
Products arising from internaJL and external bond cleavage were observed when the bicyclics were treated with thallium triacetate.^®
The percentage distribution of the internal, and external bond cleavage resulting from reaction with thallium triacetate is outlined in Table 9*
TABTjE 9
Bicyclic Type Cleavage from Thallium Triacetate
Compound Cleavage Type 1o
4.1.0 internal 9
4.1.0 external 91
3 .1.0 internal 53.4
3 .1.0 external 46.5
The product distribution in the case of mercuric acetate has not been
obtained. Since the mercury data are not available, an approximation is
made that the product distribution for mercuric acetate would be about
the same as that found for thallium triacetate. The rate of cleavage of
the internal bond of bicyclo[3.l.O]hexane is given a value of one. The
approximate relative rates for the cleavage of the internal and external
bond for both bicyclics are given in Table 10. The external bond rela
tive rates have been corrected for the statistical, factor. 45
TABLE 10
Relative Rates of Cleavage by Cleavage Type
Compound Cleavage Type Relative Rates
4.1.0 internal 1.7
4.1.0 external 7.6
5 .1.0 internal 1.0
5 .1.0 external 0.5
The substituent response observed in the cleavage of substituted
phenylcyclopropanes with mercuric acetate is given by p = -5.2 at 50°.“^®
A larger response was observed for the reaction with thallium triace
tate. The sign and magnitude of the substituent response indicates
that a great deal of positive charge is developed at the benzylic-carbon.
Therefore, much positive charge should be developed in the transition
state of the cleavage of the cyclopropane moiety by mercuric acetate in
the case of the bicyclics. The stability of the carbonium ion formed
upon cleavage of the external or internal bond should greatly Influence
the direction of cleavage. A cyclohexyl carbonium ion is formed upon
the cleavage of the internal bond of bicyclo[5 .1 .0 ]hexane and the
external bond of bicyclo[4.1.0]heptane. A cyclopentyl and a cycloheptyl
carbonium ion are formed upon cleavage of the external bond of
bicyclo[3 .1 .0]hexane and the internal bond of bicyclo[4.1.0]heptane,
respectively. A measure of the stabilities of the respective carbonium k6 ions is given by the relative rates of solvolysis of cyclopentyl, cyclo hexyl and cycloheptyl tosylates. The relative rates at 70° are given in
Table 11.®°
TABLE 11
Relative Rates of Tosylate Solvolysis
Compound Relative Rates
CsHb OTs 14
CsHiiOTs 1
C7H13OTS . 25
On the basis of carbonium ion stability the external bond cleavage
should be favored over internal bond cleavage for the bicyclo[$.1 .0 ]-
hexane case. Hoirever, in the bicyclo[4.1.0]heptane case internal bond
cleavage should be favored over external bond cleavage on the basis of
carbonium ion stability. Table 10 reveals that the opposite is observed
for both compounds. An examination of models of the bicyclics indicates
that the strain in both systems is about the same with the smaller
bicyclic being slightly more strained. Therefore, on the basis of strain
one would expect the bonds of bicyclo[5 .1 .0]hexane to cleave more readily
than those of bicyclo[4.1.0]heptane. Carbonium ion stability predicts
that the internal bond of bicyclo[4.1.0]heptane should cleave faster than
the internal bond of bicyclo[5.l.Ojhexane. While Table 10 shows that
the order is as predicted, the order of magnitude is much less than that hj observed in the tosylate solvolysis to give the same carbonium ions.
Also external bond cleavage of bicyclo[$.1.0]hexane should be favored over exbernal bond cleavage of bicyclo[4.1.0]heptane when one considers carbonium ion stability. Table 10 shows that the opposite is in fact observed. Therefore, while carbonium ion stability is expected to influence the direction of cleavage, the results obtained indicate that other features are more important.
In all of the above discussion ster&z features have been ignored.
It is possible that the steric requirements of the transition state is the dominate feature governing the direction of bond cleavage. Examina tion of models does not readily reveal that steric features should differ greatly between the two bicyclics studied. Any further conclu sions await more detailed study.
The activation parameters were calculated and are given in Table 12.
The entropy of activation was calculated at $0°.
TABLE 12
Activation Parameters for the Cleavage of Selected Bicyclics
Compound kcal/raol AS^ cal/deg/mol
Bicyclo[3.1.0]hexane 16.2 “24.6
Bicyclo[U.1.0]heptane 18.9 -10.9 48
SpiroFx.glalkanes
It vould be interesting to examine the kinetics of the cleavage and the products obtained from the cleavage of the spiro[x.2]alkanes, where
X = 5^4,5^2 when they are reacted with the acetates of mercury (ll), thallium (ill) and lead (iv). Strain factors as well as steric effects are expected to play important roles in the cleavage of the spiranes by heavy metal acetates. The preparations of two of the spiranes of inter est are described in the experimental.
Stereochemistry of cleavage
It was thought that by isolating the organometallic intermediate resulting from the cleavage of certain tricyclics that the stereo chemistry of the cleavage product might be obtained. It is thought to possess trans stereochemistry but this has not been rigorously demon
strated.^® The compounds examined were nortricyclane and tricyclo-
[$.2 .1 .0^^^]octane. In each case mercury containing compound was
obtained by employing the technique described in the experimental.
However, examination of the isolated material in all cases reveals only
one acetate group resonance at about t 8.0. Also the elemental analyses
obtained gave results totally inconsistent with the expected structures.
The results are most spurious. EXPERIMENTAL
Preparation of zinc-copper couple
The procedure outlined by LeGoff^® was employed. To a hot, nearly refluxing solution of 1.0 g cupric acetate hydrate in $0 ml of glacial acetic acid was added 65-TO g (l.O mole) of Bakers $0 mesh granular zinc.
Best results are obtained by first cleaning the granular zinc with dilute hydrochloric acid. The mixture was shaken 1-3 min keeping it hot to prevent the precipitation of zinc acetate. - After decanting the acetic acid, the couple was washed with 100 ml of glacial acetic acid followed by four 3O ml portions of ether. The couple was dried by passing a stream of nitrogen over it. The couple was freshly prepared before use.
Purification of acetic acid
The anhydrous acetic acid employed as the solvent in the kinetic studies was prepared by refluxing 1.5 liter of glacial acetic acid con taining 75 ml of acetic anhydride and 3 g of para-toluenesulfonic acid for Zk hr. The acetic acid was subsequently distilled through a 60-cm glass-helice packed column. The fraction boiling at 117.5-118.0° was collected and stored.
Kinetic analysis of the cleavage of
phenylcyclopropanes by mercuric acetate
Solutions of the desired concentrations of arylcyclopropane and
mercuric acetate in acetic acid were prepared gravimetrically. The
49 50 concentration range employed was 0.004-0.08 M. Aliquots of 2 rnl were sealed in test tubes for runs at the higher temperatures. When vola tility was not a problem, 2 ml aliquots were taken directly from a
stoppered flask containing the reaction mixture. The reaction was
stopped by thrusting the test tubes into a acetone-Dry Ice mixture. The
samples were transfered into a 50 ml flask, containing an aliquot of
standardized sodium thiocyanate, by adequately washing out the test tubes
containing the sample with water. Titration of the excess thiocyanate
was carried out at 0° using a standardized silver nitrate solution. The
indicator was ferric ammonium sulfate in dilute nitric acid. Similar
procedures as described above were used to determine the stability of
mercuric acetate and the alkylmercurie acetates. The kinetics of the
cleavage of the other cyclopropanes studied were also followed in the
manner described above.
Preparation of bicyclor^.l.Olhexane
To a flask charged with 40 g (O.587 mole) of cyclopentene, 46 g
(0.700 mole) of zinc-copper couple and 100 ml of anhydrous ether was
added approximately 2 ml of methylene iodide. Gentle heating initiated
the reaction and more methylene iodide was addeddropwise over a 1 hr
period such that a total weight of 172 g (0.640 mole) had been added.
The mechanically stirred reaction was maintained at reflux for 72 hr.
After periods of 24 and 48 hr an additional 23 g (0.35 mole) of zinc-
copper couple, 26 ml (0.32 mole) of methylene iodide and 25 ml of
anhydrous ether was added to the reaction mixture. 51
The reaction mixture was suction-filtered with much difficulty due to gummy zinc salts formed during the course of the reaction. The fil trate was twice washed with saturated ammonium chloride, saturated sodium bicarbonate, many times with water and dried over anhydrous mag nesium sulfate. The majority of the ether was removed by distillation through a small Vigreaux column after which the resulting concentrate was distilled through a spinning band column. Approximately 11.4 g of material (bp IIO-I30) was collected and shown to contain considerable ether as indicated by nmr. The desired product was obtained kinetically pure by collecting it by vpc (lO' x 3/8 '', 2Qffo APIEZON J on 60/8O
Chromosorb ¥ column at 65°). The nmr shows a complex multiplet in the
t9-10 region and no absorption in the vinyl region. The preparation of this compound has been previously described by others4-o_
Preparation of spirofS.3]octane
A flask was charged with I3.6 g (0.21 mole) of freshly prepared
zinc-copper couple, 2 ml of methylene iodide and 80 ml of anhydrous
ether. The mixture was heated to gentle reflux after which 10 g
(0.10 mole) of methylene cyclohexane and 9-8 ml (O.15 mole) of methylene
iodide solution v/as added dropwise. After 20 hr an additional 3-4 g
(0.052 mole) of zinc-copper couple and 3 ml (O.O37 mole) of methylene
iodide was added. The stirred reaction mixture was maintained at
suetion-filtered and the filtered solids washed with ether. The combined
filtrates were washed twice with saturated ammonium chloride, twice with
saturated sodium bicarbonate, once with distilled water and once with
saturated sodium chloride solution. The ether layer was subsequently 52 dried over anhydrous magnesium sulfate. The majority of the ether was removed by slow distillation through a 15-cm Vigreaux column. The pro duct was collected hy vpc (lO' x 5/8*' APIEZOW J on 60/8O Chromosorh W at 75°)' The collected material weighed 1.5 g and the nmr spectrum shows a sharp singlet at T9.8, a multiplet at t 8-9 and no vinyl absorptions.
The preparation of this compound has been previously described by others.
Preparation of spirof^l-.2]hepta-l,3 diene^^
To a flask charged with 5OO ml of dry tetrahydrofuran and 78.1 g
(2.0 mole) of finely divided sodium amide; 66.1 g (l.O mole) of freshly distilled cyclopentadiene was added dropwise over a O .5 hr period. The rate of addition of diene was adjusted so as to maintain gentle reflux.
Care needs to be exercised at this point as occasionally there is an
induction period in the anion formation. The dropwise addition of I89.O
g (1.0 mole) of 1,2-dibromoethane to the stirred mixture was carried out
over a 5 hr period and was accompanied by considerable heat evolution.
The reaction mixture was allowed to stand and cool overnight during
which time the mixture divided itself into two layers. The upper organic
layer was decanted and the residual salt layer was dissolved in 5OO ml
of water. The aqueous layer was extracted with three portions of 20-60°
petroleum ether and the extracts combined with the organic layer. The
combined organic layers were then washed successively with distilled
water, 10% hydrochloric acid and dried over anhydrous magnesium sulfate.
The solvent was removed by distillation. The concentrate was distilled
through a 35 X 1.5-cm glass bead packed coluimi to produce 39.0 g, 42% 55 yield of splro[4.2]hepta-l,5 diene, Ip 82-90°. The nmr spectrum shows a sharp singlet at t 9-0 and a pair of well split doublets in the vinyl region.
Preparation of potassium azodicarboxylate
To a 2 liter erlenmeyer flask charged with 600 ml of hojo aqueous potassium hydroxide cooled to 0°, 100 g (0.86 mole) of azodicarbonamide was added slowly over a 1.5 hr period with stirring. The stirring was
continued for about 1 hr after complete addition. Ammonia ivas contin uously given off during the course of the reaction. The yellow solid was collected in a fritted glass funnel and washed thirteen times with
generous portions of ice-cold methanol. The salt was dried overnight in
a vacuum oven. The dry salt, I50 g, SQPjo yield was stored in a bottle.
Preparation of spiroT^!-.2lheptane
To a flask charged with 9-TT3 g (O.IO5 mole) of spiro[4.2]hepta-
1,5 diene, $20 g (I.65 mole) of potassium azodicarboxylate and 5OO ml of
methanol, 5OO g of glacial acetic acid was added dropwise. The total
addition time was 2 hr and the reaction was stirred for 1 hr after addi
tion was complete. Distilled water, 250 ml was added to the mixture.
The solution was extracted with three portions of ether. The ether
extracts were combined, washed twice with 10^ aqueous potassium hydrox
ide, once with distilled water, once with saturated sodium chloride
solution and dried over anhydrous magnesium sulfate. The material was
concentrated by distilling off the majority of the solvent. The nmr
spectrum of the concentrate indicates that the desired product was
obtained. A sharp singlet at t9*6 was observed and there were no 54 absorptions in the vinyl region. The preparation of this material has been described by others
Preparation of exo-tricyclo[3.2.1.0^'^loctane^^
A flask was charged with 66 g (l.O mole) of zinc-copper couple, 4-7 g
(0.5 mole) of bicyclo[2.2.l]heptene, 200 ml of anhydrous ether and 2 ml
of methylene iodide. The reaction mixture was heated until reaction was
initiated as indicated by the evolution of bubbles. To the rapidly
stirred mixture, I87.5 g (0.7 mole) of methylene iodide in anhydrous
ether was added dropwise. After 12 and 22 hr more zinc-copper couple,
53 g (0.5 mole) and 28 ml (O.55 mole) of methylene iodide was added. The
stirred reaction mixture was maintained at reflux for about 3O hr. Most
of the zinc salts were removed by suction-filtration. The filtrate was
successively washed with saturated ammonium chloride, saturated sodium
bicarbonate, saturated sodium chloride and dried over anhydrous magnesium
sulfate. The filtrate was concentrated by distilling off the majority of
the ether. The concentrate was distilled through a 15-cm Vigreaux
column. The best fraction bp 97-104° which was 20 ml in volume was shown
by nmr, t9-10 (complex multiplet), t3-5 (little absorption), to be at
least 95^ pure with a trace of olefin and ether present.
Preparation of endo-tricyclo|~5-2.1.0^-’'^''1oct-6-ene
The general method for the preparation described by Gloss and
Krantz^^ was employed. To a stirred suspension of 24.0 g (0.6 mole) of
sodium amide in 40 ml of mineral oil maintained at 80°, 46 g (O.6 mole)
of allyl chloride diluted by 5O ml of mineral oil was added dropwise
over a 10 hr period. A 35-cm glass-bead packed condenser cooled by a 55 circulating ice-water pump vras employed to keep the allyl chloride from flash distilling. A tygon gas escape tube was attached to the condenser and the open end of the tube immersed in a stirred solution of 13.2 g
(0.2 mole) of freshly prepared cyclopentadiene dissolved in 80 ml of n-pentane maintained at 0°. After complete addition of the allyl chlor ide the n-pentane was removed by distillation. The concentrate was dis tilled through a micro-distillation apparatus. The nmr spectra of the two higher boiling fractions indicates the presence of cyclopropyl pro tons based on the complex multiplet observed at t9-10 . The product was purified by vpc ( 3 ' x 5/8'' 20/o Garbowax 20 M on 6O/8O Chromos orb W column at 55°)* The collected material was a white to clear solid weigh ing about 1 g. The nmr spectrum agrees with that reported by Wiberg^® and is distinctly different from that reported^^ for the exo isomer.
Preparation of endo-tricyGlo[3.2.1.0^^'^'joctane
A flask was charged with I.05 g (O.OO95 mole) of endo-tricyclo-
[3 .2 .1 .0^^^]oct-6-ene prepared as described above y xnl of absolute methanol and I8.5 g (0.095 mole) of potassium azodicarboxylate. To the stirred reaction mixture, 22 g of glacial acetic acid was added dropwise over a 2 hr period. The reaction was stirred for 30 min after complete acid addition during which time the color of the reaction mixture changed
from yellow to white. Distilled water, 100 ml was added and the solution
transferred to a separatory funnel. At this time an additional 15O ml
of distilled water was added. The water layer was extracted three times
with ether. The combined ether extracts were washed twice with a lOfo
aqueous potassium hydroxide solution, once by distilled water, once by 56 saturated sodium chloride solution and dried over anhydrous magnesium sulfate. Most of the ether was removed hy distillation through a 15-cm
Vigreaux column. The product.was collected hy ypc (lO' x 5/8'' 20%
APIEZON J on 60/8O Chromosorh W at 120 °). The white crystalline product, weighing 0.26 g shows complex splitting in the T$-10 region of its nmr spectrum. Wo absorptions were observed in the vinyl region. The endo nmr spectrum differs greatly from that obtained for the exo isomer indi
cating that isomerization during vpc collection did not occur.
Preparation of 1,5-dihromo-l-phenylpropane^^
A solution of 59.6 g (O.5O mole) of l-hromo-5-phenylpropane, 53-2 g
(0.50 mole) N-hroraosuccinimide and a trace of henzoyl peroxide in 5OO ml
of chloroform was stirred at reflux overnight with continuous irradiation hy a 250 watt red sun lamp. The reaction was complete as indicated hy
the formation of succinimide which floats to the top of the solvent. The
mixture was cooled in an ice-hath and the succinimide removed hy suction-
filtration. The filtrate was concentrated under reduced pressure. The
concentrate was dissolved in ether and the ether layer washed twice with
distilled water to insure removal of all succinimide. The ether layer
was dried over anhydrous magnesium sulfate and concentrated under reduced
pressure. The concentrate was distilled, hp 100-101° (0.35 imm) under
reduced pressure through a 15-cm Vigreaux column. A 80% yield of 67 g
(0.24 mole) of the 1,3-dihromo-l-phenylpropane was obtained.
Preparation of phenylcyclopropane^^
A flask charged with 33*6 g (0.12 mole) of 1,3-dihromo-l-pnenyl-
propane, ik g (0.21 mole) of zinc-copper couple and TOO ml of anhydrous 57 ether was mechanically stirred at reflux for 20 hr. After cooling the solution was washed with saturated ammonium chloride, saturated sodium bicarbonate and dried over anhydrous magnesium sulfate. The majority of the ether was removed under reduced pressure. The concentrate was dis tilled, bp 108-5° (ill mm) under reduced pressure. The nmr spectrum of the product obtained shows a congplex multiplet in the t9-10 region indi
cative of cyclopropane. There is no absorption observed in the ?4-5 region where the benzyl-proton of 1,3-dibromo-l-phenylpropane is found.
The nmr spectrum also revealed that the product is at least 95^ pure with
ether as the impurity. Approximately 4.69 g (33% yield) of phenylcyclo-
propane was obtained.
Preparation of nortricyclanol^^
A flask charged with 104 g (I.132 mole) of bicyclo[2.2.l]hepta-2,5-
diene, 97 g of glacial acetic acid and 0.8 ml of boron trifluoride
etherate was heated on a steam bath for 6 hr. After cooling the reaction
mixture was diluted with 200 ml of ether and washed successively by two
portions of 3 N ammonium hydroxide, two portions of saturated sodium
chloride and dried over anhydrous magnesium sulfate. The ether was
removed under reduced pressure. The black concentrate was distilled
bp 50-61° (4.0 mm) under reduced pressure through a 15-cm Vigreaux
column to give about I30 g of a mixture of nortricyclyl acetate and
b icyclo[2.2.l]hepta-5-ene-2yl acetate.
The acetate mixture was dissolved in 350 ml of chloroform which
was then cooled to -10° by an ice-salt bath. Nitrocyl chloride was
bubbled through the solution with swirling until the color of the 5B solution changed from bright green to brownish green indicating an excess of nitrocyl chloride. A white precipitate was observed. Petroleum ether, 350 ml was added and the mixture cooled at -5° for an additional
15 min. The nitrocyl adduct of bicyclo[2.2.l]hepta-5-ene-2yl acetate was collected by suction-filtration. The filtrate was washed with two por tions of saturated sodium carbonate solution, one portion of saturated sodium chloride solution and dried over anhydrous magnesium sulfate. The solution was concentrated under reduced pressure and the concentrate was distilled under reduced pressure to give 90 g of nortricyclyl acetate as a faintly green liquid.
The nortricyclyl acetate obtained was added to a solution of O.5 g
of sodium metal in 3OO ml of anhydrous methanol. The reaction mixture was heated on a steam bath and the methanol was slowly distilled off
with the aid of a Dean-Stark trap. The residue was diluted with 30-60°
petroleum ether followed by two washings with distilled water and drying
over anhydrous magnesium sulfate. The solvent was removed at 25° under
reduced pressure to yield a solid. The absence of the sharp singlet
exhibited by acetates in the nmr at about t8 indicated that hydrolysis
was complete. The crude product was used as such in the oxidation to
produce nortricyclanone.
Preparation of nortricyclanone*^
The reagent employed in the oxidation of nortricyclanol to nortri
cyclanone was prepared by dissolving I8 g (0 .I8 mole) of chromium
trioxide in 25 ml of distilled water in a 250 ml beaker. The beaker was
immersed in an ice-bath and 30 g (O.3O mole) of concentrated sulfuric 59 acid followed h y $0 ml of distilled water was cautiously added with stirring. The solution was cooled to 0-5°.
A flask charged with $0 g (O.JO mole) of nortricyclanol, prepared above and I50 ml of anhydrous acetone was cooled to 0-5°. The cold oxi dizing agent >ras added dropwise to the cold vigorously stirred nortri cyclanol solution at a rate which maintained a temperature at or below
20°. Stirring was continued for 5 hr after complete addition. Sodium bisulfite was added in small portions until the brown color of chromic acid disappeared from the upper layer of the two-phase system. The top layer was decanted. The green, dense lower layer was extracted with three petroleum ether washings. The combined ether extracts were washed twice with saturated sodium bicarbonate solution, once with saturated
sodium chloride solution and dried over anhydrous magnesium sulfate. The majority of the solvent was removed under reduced pressure. The concen trate was distilled bp 50° (5.5 mm) under reduced pressure through a 15-
cm Vigreaux column to give I5 g of nortricyclanone. The infra-red
spectrum is identical with that of an authentic sample prepared by
Dr. Frank Zalar.
Preparation of nortricyclane
A flask was charged with 6.6 g (O.O61 mole) of nortricyclanone, 1 g
of potassium hydroxide, 8 ml of anhydrous hydrazine and HO ml of ethylene
glycol. The reaction mixture was stirred and slowly heated for 1 hr to
a temperature of 190° while a slow stream of dry nitrogen flushed the
product through a Tygon tube attached to the top of the condenser into a
trap imraersed in a Dry-Ice acetone bath. At various intervals over a 60
6 hr period the white; solid trap contents were removed hy rinsing with minimal amounts of n-pentane. The washes were stored in a stoppered flask maintained at Dry-Ice temperature. The product is a white solid which sublimes very rapidly. The difficulty in handling this material made obtaining spectral data impractical. This material has been pre pared previously by others^^.
Preparation of 3~phenylpropylmercuric acetate
To a flask charged with 0.80 g (O.O33 mole) of dry magnesium in 50 ml of anhydrous ether; 6.0 g (O.O3I mole) of l-bromo-3-phenylpropane in
ether was added dropwise. The Grignard was stirred at reflux overnight
and was then added dropvrise to an ether slurry of 10.85 g (O.O3O mole) of
mercuric bromide and stirred for O .5 hr at room temperature. The reac
tion was carried out under a positive pressure of nitrogen. The reac
tion mixture "VTas hydrolyzed by the addition of I50 ml of 3% acetic acid.
The ether and water layers were separated. The ether layer was washed
with six portions of water and dried over magnesium sulfate. The ether
was removed under reduced pressure and a white solid remained. The pro
duct was obtained in yield mp 4%° upon recrystallization from abso
lute methanol. Anal. Calcd for CgHnHgiBri: C, 27*0^; H, 2.77.
Found: C, 2 7 . H, 2.89.
A flask charged with 2.82 g (O.OO71 mole) 3 -phenylpropylmercuric
bromide and I.18 g (O.OO7I mole) of silver acetate in 50 of absolute
methanol was stirred at room temperature overnight in the absence of
light. The precipitated silver bromide was removed by suction-filtra
tion. The solvent was stripped off under reduced pressure revealing a 6l
colorless oil. The oil was dissolved in anhydrous ether, the ether was washed, four times with distilled water and dried over magnesium sulfate.
The ether was removed under reduced pressure and again a colorless oil
was revealed. The oil is found upon titration of the mercury II to he
greater than 95% pure.
Preparation of n-hutylmercurie acetate
n-Butylmercurie hromide was treated in a manner analogous to that
described above for the preparation of $-phenylmercuric acetate to pro
duce n-butylmercurie acetate. The n-butylrnercuric acetate was recrystal
lized from n-pentane rap 53-0-53-5j lit.^° mp 33.8-54.3°.
Preparation of 3~Phenyl-3-acetoxypropylmercuric acetate
A 100 ml, round-bottomed flask charged with 0.59 g (O.OO5O mole) of
phenylcyclopropane, 1.40 g (0.0044 mole) of mercuric acetate and 50 ml of
glacial acetic acid was put into a 50° bath. The reaction was allowed to
stand in the bath for 51*5 hr after which time a small amount of free
mercury was noted. The acetic acid was removed by vacuum transfer and a
white solid remained. The product, mp 87-88° was obtained in 30% yield
after recrystallization from n-heptane. The product was identified by
its nmr spectrum, t4.2 (t,benzyl proton), 8.1 (s,acetate protons), and
8.2 (s,acetate protons), and chemical analysis. Anal. Calcd for
OisHisHgiO^: 0, $ 5.65; H; 3.68; Hg, 45.90. Found: C, 55.6$; H, 5.57;
Hg, 45.65.
Preparation of 5-(p-anisyl)-3-acetoxypropylmercuric acetate
A 100 ml, round-bottomed flask charged with 2.95 g (O.OI96 mole)
of p-methoxyphenylcyclopropane, 5*64 g (O.OI77 mole) of mercuric acetate. 62 and 100 ml of glacial acetic acid was placed in a bath at 50° for 2 hr.
A trace of free mercury was noted and the solvent was removed at room temperature hy vacuum transfer revealing a white solid. The solid was purified hy stirring with 5^ ml of hexane for l6 hr. The solid was filtered and T.T5 g, 9^^ yield of material mp T0.0-T1.5°, was obtained analytically pure. Anal. Calcd for Ci^HisHgiOs: C, $6 .01; 5.89.
Found: C, 35*90; 11 ^ 3*98. The nmr spectrum of the adduct in pyridine
exhibited resonances at t^.3 (t^benzyl proton), 6.5 (s,methoxyl protons),
8.0 (s,acetate protons), and 8.1 (s,acetate protons). The remaining
protons exhibited complex multiplets in the t8-9 region. The
satellites were not determined.
The adduct was demercurated by the method described by Brown^^ to
produce l-(p-anisyl-propyl acetate which was identified by its nmr spec
trum and comparison to an authentic sample prepared by an independent
route. The nmr reveals absorptions at t9“10 (complex multiplet), 8.0
(s,acetates), 6.3 (s,methoxyl), 4.5 (t,benzyl), 2.7 (doublet of doublets
due to aromatic protons).
Preparation of 1-(p-anisyl)-l-acetoxypropane
To a flask charged with I.16 g (0.048 mole) of freshly pulverized
magnesium in 50 ml of anhydrous ether, 5*24 g (0.048 mole) of ethyl
bromide diluted by 50 ml of anhydrous ether was added dropwise. The
reaction was stirred at reflux overnight and the resulting Grignard
solution was filtered through a glass wool plug into a dropping funnel
and was added dropwise to a flask containing 6.5 g (0.048 mole) of ani
saldéhyde. After stirring for 2 hr, 50 md. of distilled water was added 6) to hydrolyze the mixture. The water layer was separated from the ether layer and discarded. The ether layer was washed twice with saturated ammonium chloride solution, twice with saturated sodium chloride solution and dried over anhydrous magnesium sulfate. The ether solution was placed in a flask along with 6 ml of pyridine to which was added $.8 ml
(0.048 mole) of acetyl chloride drop\fise with stirring. The reaction was stirred at room temperature overnight. Distilled water, ^0 ml was added and the water layer discarded. The ether layer was washed twice with saturated ammonium chloride solution and dried over anhydrous magnesium sulfate. The majority of the ether was removed by distillation. The product was distilled bp 82-85° (0.40 mm) under reduced pressure. The nmr spectrum of the mterial obtained after distillation reveals that
some impurity is present. This impurity could not be removed by vpc.
The principal resonances coincide with those obtained from the nmr spec
trum of the product arising from the demercuration of J-(p-anisyl)-^-
acetoxypropylmercurie acetate.
Preparation of 5-(p-methyl)-5-acetoxypropylmercuric bromide
A 250 ml, round-bottomed flask charged with 2.67 g (0,020 mole) of
p-methylphenylcyclopropane, I.96 g (O.OO6I mole) of mercvu-ic acetate and
150 ml of glacial acetic acid was placed in a bath maintained at 50°.
After 20 min the reaction vessel was removed and the solution frozen by
thrusting the vessel into a Dry-Ice bath. The solvent was removed by
vacuum transfer. A white, gummjr residue remained which was taken up in
ether. The ether insoluble material was filtered off and identified as
mercurous acetate. The material turned black upon exposure to 6k concentrated aramonlura. hydroxide. The ether was stripped off revealing an oil which defied all attempts to prompt it to crystallize. A vessel containing the oil was charged with 0.72 g (0.00)8 mole) of potassium hromide and UO ml of absolute methanol. The reaction mixture was stirred at room temperature in the absence of light for )6 hr. After this time the reaction vessel was immersed in a Dry-Ice bath. Initially a solid was noted which disappeared as the vessel began to warm up but upon further warming new crystals formed in abundance. A total of 1.25 g of white crystals were obtained from three crops. The best material was obtained by stripping off the solvent and recrystallizing from )0-
60° petroleum ether. The product, mp 61-62° was identified by elemental analysis. Anal. Calcd for CigEisHgiOgBri: 0, )0.55; H, ).21. Found:
C, 31.45; H, 3.39.
Product from the decomposition of 3~phenyl-
3 - a c et oxyp)r opy Imer cur i c acetate
A test tube containing 0.24 g of 3"PEenyl-3-acetoxypropylmercuric
acetate and 4 ml of glacial acetic acid was sealed and placed in a bath
maintained at 100° for 9 days. The reaction mixture was taken up in 3O
ml of ether. The ether layer was washed twice with saturated sodium
bicarbonate and dried over anhydrous magnesium sulfate. The ether was
removed under reduced pressure. The remaining material was shown to
have the same retention time as an authentic sample of cinnamyl acetate
prepared by Dr. Aubry South, Jr. A Garbowax column at 185° was employed
for the above comparison. A sample of 1-phenyl-1,3-diacetoxypropane, a
possible product, had a much different retention time than that observed 65 for the single product obtained above.
Kinetic analysis of the cleavage of
phenylcyclopropanes hy mercuric acetate
Preparation of substituted phenylcyclopropanes employed in this
study were prepared by Dr. Aubrey South, Jr.^^
Kinetic analysis of the cleavage of
phenylcyclopropanes by lead tetraacetate
A stock solution of perchloric acid was made up gravimetrically in
purified acetic acid. The perchloric acid employed was TC^ by weight.
Therefore, enough acetic anhydride was added to ensure that all of the
water was removed. A stock solution of lead tetraacetate was prepared
gravimetrically. Tlie desired amount of arylcyclopropane was weighed
out into a vessel. The cyclopropane was transferred into the vessel
containing the lead tetraacetate solution. Aliquots of the resulting
solution and of the perchloric acid solution were placed in respective
sides of a partition flask. After equilibration the contents were mix
ed and the lead (IV) titer change was determined after a specified
period of time had elapsed. The lead (IV) concentration at t = 0 was
in aJ-1 cases approximately 0.02 M. The cyclopropane concentration
varied from a 0.5 to a 15 fold excess depending upon the concentration
of perchloric acid and the structure of the substrate employed.
The lead (iV) titer change was determined by adding excess potas
sium iodide solution, 20 ml of 5^ KI, and titrating the iodine that
was liberated with a standardized sodium thiosulfate solution. Vitex
was employed as indicator and the color changed from green to a smooth
yellow. 66
Determinations of Eq.uili'brium Constants
The molar extinction coefficient of 4-chloro-2 nitroaniline at 410 mjj, is 4470 l/m cm. The optical density of a solution containing
$.7 X 10 ^ M indicator and 2.44 x 10 ^ M perchloric acid is O .166 corres ponding to a calculated Kg, for the indicator of 2.$6 x 10"^ l/m. For formal concentration of 5*7 x 10"^, 5>0 x 10”^ and 2.44 x 10 ^ m/l for the indicator, mercuric acetate and perchloric acid, respectively, the Hg(0Ac)+C10l , observed O.D. is 0.J2 corresponding to a Kf = 2).7 l/m. For formal concentrations of 5-7 x 10"^, 1.25 x 10 and 2.44 x 10 ^ rn/l for
the indicator lead diacetate and perchloric acid, respectively, the O.D.
is 0.20 corresponding to ^ ^ CIO4 _ l/m. For formal concentra
tions of 5.7 X 10"^, 2.72 X 10 and 2.44 x 10”^ m/l for the indicator,
lead tatraacetate and perchloric acid, respectively, theO.D. is 0.25
corresponding to _ 25.7 1 u.
The procedure employed was that previously described by Lemaire and
Lucas.The data obtained for Hg(OAc)g, Fb(0Ac)4 and Pb(OAc)g is given
in Table 13. 6t
TABLE 15
Visible Spectroscopy Data
Hg(OAc)g Fb(0Ac)4 Pb(0Ac)2
M M M
[l]o 5.T X 10" 3.7 X 10-4 2 . 7 X 10-4
[Hg(0Ac)2]o 5.0 X 10-2 [pb(QAc)^]o 2.72 X 10-2 [pb(0Ac)2]o 1.23 X 10-3
[HC104]o 2.44 X 10-3 2.44 X 10-3 2.44 X 10-3
n-5 [I] T.I5 X 10- 5.6 X 10-3 4.47 X 10-5 -4 [IH*] 3.0 X 10-4 3.1 X 10 3.3 X 10-4 n-4 -3 [H+] 9.9 X 10" 1.31 X 10 1.74 X 10-3 -4 [HgOA&Clo;] 1.13 X 10-3 [pb(0Ac)2fl0;] 8.2 X 10-4 [pbOAcClo;] 3.4 X 10
O.D. 0.32 0.25 0.20
K?®0A^C104 23.7 l/m 25.7 l/m 246 l/m 68
APPENDIX
Results of a typical kinetic run at ^0.1°
Ao = [Hg(0Ac)2]o = 0.04909 M
Bo = [0 " - 3o - 0.04909 M
Tvro milliliter aliquots contained in sealed tubes were periodically removed and thrust into an acetone Dry-lce bath. After addition of 5 ™1 of 0.02872 W sodium thiocyanate solution, the excess thiocyanate was titrated by a0,02560 N silver nitrate solution. A saturated solution of ferric ammonium sulfate in 50fonitric acid \tq.s employed asindicator.
The red solution was titrated to a white color.
quot ml AgNOs X [Ao - x] [Ao - x]-i Time m/l m/l m/l min
1 4.470 0.0085 0.0408 24.5 10.9
2 4.960 0.0146 0.0545 29.0 30.9
5 5.436 0.0207 0.0284 55.2 58.9
4 5.844 0.0259 0.0252 43.1 90.0
5 6.095 0.0291 0.0200 50.0 118.6
6 6.541 0.0522 0.0169 .59.2 165.5
7 6.700 0.0568 0.0125 81.5 254.0 REFERENCES CITED
1. M. Yu. Lukina; Russian Chemical Reviews; Wo, 8; 419 (1962).
2. C. A. Coulson and ¥. Moffitt; Phil, Mag,; W ; 1 (1949),
5 , L, L, Ingraham; ’’Steric Effects in Organic Chemistry;''M, S, We-VTman; Ed.; John Wiley and SonS; Inc.; New York; W. Y. ; 195&; p, 518,
4, N, M, Kizhner; Zhur. Russ. Fiz, Khim. Obshch,; 45; 957 (1915); Chem, Abstr.; J_, 3965 (1913)-
5, R, Robinson; J. Chem, Soc,; 109; 1042 (1916),
6 , L, JoriS; P, von R. Schleyer and R, Gleiter; J, Am, Chem.Soc,; 90, 327 (1968).
7 ,J, W, Linnett; Nature; I6O; 162 (1947).
8 , A, D, Walsh; Discussions Faraday Soc.; 2, I8 (1947),
9 , P, C, Lauterbur; J. Chem. Phys.; 26; 217 (1937).
10, N, Muller and D, E. Pritchord; ibid.; 31; 768 (1939).
11, F, J, Weigert and J, D, Roberts; J, Am, Chem, Soc,; 89;5962 (1967).
12, G, S, Handler and J, H. Anderson; Tetra,; 2, 345 (1958).
15, W. A, Bernett; J, Chem. Educ,; 17 (1967).
14, F, H, Seubold; Jr.; J. Chem, Phys.; 21; 1616 (1933).
15, A, Baeyer; Chem, Ber,; I8; 2277 (1855); R. Fitlig andF, Roder; Ann,; 227; 13 (I885); G. Gustauson; J, Prakt. Chem,; [2] 36; 300 (1887),
69 TO
16. R. A. Raphael; ''Chemistry of Carhon Compounds;'' Vol. II; E. H. Rodd; Ed.; Elsevier Publishing Co.; New York, N. Y., 1953 P* 26; E. E. Royals, ''Advanced Organic Chemistry,'' Prentice-Hall, Inc.; Englewood Cliffs, New Jersey, 195^j Chapter 5-
IT. R. T. LaLonde and L. S. Ferney, J. Am. Chem. Soc., 3T6T (1963); R. T. LaLonde and M. A. Tobias, ibid., 85, 3TT1 (1963)) R. T. LaLonde and M. A. Tobias, ibid., 86, 4 0 ^ ( 1964).
18. R. J. Ouellette, A. South, Jr., and D. L. Shaw, ibid., 8T,2602 (1965).
19. R. Ya. Levina, V. N. Kostin, and T. K. Ustynyuk, Zh. Obshchei Khim., 359 (i960); Chem. Abstr., 2443Tb (I960).
20. C. H. DePuy, F. W. Breitbeil and K. R. DeBruirn, J. Am. Chem. Soc., 33bT (1966).
21. R. J. Ouellette, D. S. Miller, A. South, Jr., and R. D. Robins, unpublished work.
22. P. R. Wells and W. Kitching, Aust. J. Chem., I8 , 320 (1965); P. R. Wells and W. Kitching, ibid., 20, 2029 ( l 9 ^ -
23. R. J. Ouellette and R. D. Robins, Tetrahedron Letters, 397 (1968).
24. K. Ichikawa and H. Ouchi, J. Am. Chem. Soc., 82, 38?6 (i960).
25. H. C. Brown and Y. Okamoto, ibid., T9, 1913 (1957).
26. L. L. Schaeleger, M. A. Turner, T. C. Chamberlin, and M. M. Kreevoy, J. Org. Chem., 3421 (1962).
27. R. Ya. Levina, Zh. Obshchei Khim., 2512 (1964); Chem. Abstr. 14701c (1964).
28. C. F. H. Tipper, J. Chem. Soc., 204-5 (1955)-
29. D. M. Adams, J. Chatt, R. G. Guy, and N. Sheppard, ibid., 738 (1961).
30. A. South, Jr., nd R. J. Ouellette, J. Am. Chem. Soc., (in Press).
31. A. J. Kresge, M. Dubeck, and H. C. Brown, J. Org. Chem., 32, 745 (1967).
32. L. G. Gillen and A. E. Martell, ''Stability Constants,'' The Chemical Society, London 1964. 71
35. R. Criegee, ’’Oxidation in Organic Chemistry,’’ Vol. 3-A, K. B. ¥iberg, Academic Press, New York, N. Y., I963, Chapter 5*
3^1-. C. B. Anderson, E. C. Friedrich, and S. Winstein, Tetrahedron Letters, 2037 (1963)} H. Meervrein, Angew. Chem. 67, 37^ (1955); H. Meerwein, V. Hederich and K. Wunderlich, Arch. Pharm., 291, 5^1 (1958); H. Meerwein, K. Bodenhenner, P. Borner, F. Kunert, and K. Wunderlich, Ann., 632, 38 (196O); H. Meerwein, V. Hederich, H. Morschel, and K. Wunderlich, ihid., 635; 1 (l96o)j H. Paulson, W. P. Trautwein, F. Garrido Espinosa, and K. Heyns, Tetrahedron Letters, 4131, (1966).
35. H. Paulson, W. P. Trautwein, F. Garrido Espinosa, and K. Heyns, ibid., 4137 (1966).
36. L. J. Dolby and M. J. Schwarz, J. Org. Chem,, 3581 (1965).
37. R. Heck and S. Winstein, J. Am. Chem. Soc., 79, 3105 (1957).
38. P. D. Bartlett, ’’Nonclassical Ions,’’ W. A. Benjamin, Inc., New York, N. Y., 1965 p. I8O.
39. E. LeGoff, J. Org. Chem., 29, 2048 (1964).
40. H. W. Pouncey, Jr., Dissertation Abstracts, 20, 3O8 (1959).
41. R. W. Shortridge, R. A. Craig, K. W. Greenlee, J. M. Derfer, and C. E. Boord, J. Am. Chem. Soc., 70, 946 (1948).
42. C. F. Wilcox, Jr., and R. R. Craig, ibid., 83, 3866 (1961).
43. Ya. M. Slobodin and M. V. Blinova, Zh. Obshchei Khim., 24, 621 (1954); Chem. Abstr. 53171 (1955).
44. H. E. Simmons and R. D. Smith, J. Am. Chem. Soc., 8I., 4256 (1959).
45. G. L. Gloss and K. D. Krantz, J. Org. Chem., 638 (1966).
46. K. B. Wiberg and W. J. Bartley, J. Am. Chem. Soc., 6375 (i960).
47. R. J. Ouellette, R. D. Robins, and A. South, Jr., ibid., 90, I619 (1968).
48. J. Meinwald, J. Crandall, and W. E. Hymans, Org. Syn., 77 (1965). 72 h9. P. von R. Schleyer, J. Am. Chem. Soc., 80, 17OO (1958).
50. R. J. Ouellette, Ph.D. Thesis, The State University, Berkeley, California, 1962.
51. H. C. Brp^m and ¥. J. Hammar, J. Am. Chem. Soc., §9^ 1522 (1967).
52. H. Lemaire and H. J. Lucas, ihid., 75, 5198 (l95l)j H. Lemaire and H. J. Lucas, ibid., 77; 959 (195577