Qmrnmrnmrnmi STABLE crcwHExmiTHiw comkhihbs

AHBROVBDt

Gradu»t# Committeei

ofeaeor

Minor Piv®r«s»©r )wi^4

til Gosmiti©©u MeiaWr

Qommitt#© Member : Jy"

W /C C 'V ' t-r Xi' I L' <• • 'L

Director o£/fc^ Department of Ciiemistry

Deais of the Graduate 3c&ool

V/ ON FORMAT IONA.LLY STAPLE CYCLOTlEXYJiLJTH I UK COMTOOHDS

DISSERTATION

Presented to the Graduate Council of the

North Texas State University in Partial

Fulfillment, of tho Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

By

Charles M. Selruarx, B. 3.f M. S.

Denton, Texas

Januar.y , 1Q6&

TABLE OP CONTENTS

Page

LIST OP TABLES iv

LIST OP ILLUSTRATIONS v

Chapter

I. INTRODUCTION 1

II. EXPERIMENTAL PROCEDURES 28

Descriptions of Reagents and Synthesis Procedures

Experimental Methods Employed in Charac-terizing the Organolithiuiu Compounds and Their Derivatives

Experimental Methods Employed in Deriva-tive Porination of the Organolithium Compounds

III. RESULTS AND DISCUSSION 70

BIBLIOGRAPHY 118

1X1

LIST OF TABLES .

Table ' Page

I. Table of Letsinger's Data 3

II. Stereochemistry Investigated Alkenyl

Lithium Compounds . . . . . 11

III. p-Menthane and p-Menthene Retention Times . . . . 45

IV. Characteristic p-Menthene Infrared and Proton Magnetic Resonance Absorptions 46

V. Data Summary for Cyclohexyl Bromide Derivatives 62

VI. Characteristic Rotational Values for

Kenthyllithium Solutions . . . . 75

VII. Spin-Spin Coupling Constants 30

VIII. Calculated Inversion Energies 38

IX. Carbonation Products from Kenthyllithium . . . . 95

X. Bromination Products from Solutions of. Menthyllithium and 4-tert-Butyl-cyclohexyllithium . 103

XI. Product Distribution from Electrophilic Substitution Reactions of the Cyclo-hexyllithium Compounds 110

IV

LIST OP ILLUSTRATIONS

Pigure Page

1. Cyclohexyl Derivatives Studied . . . . . 73

2. c<~Methinyl Proton Resonance Spectra of Ken thy 1 and Neomenthyl Chloride Isomers (a), Bromide Isomers (b)t Iodide Isomers (c), and Alcohol Isomers (d) 81

3. Proton Kagnetic Resonance Spectra of Menthyl Chloride (a) and Ileoiuenthyl Chloride (b) . . . 32

4. Spin-Spin Coupling Diagram of Resonance Patterns 83

5. Proton Magnetic Resonance Spectra of Crystalline l.-entliyllithiuin Dissolved in Toluene (a) and Dimethyl Ether (c). Also Enlargements of o<-Methinyl Proton Spectra in Toluene (af)» Diethyl Ether (b), and Dimethyl Ether (e) 85

6. Inversion Pathways . . . . . . . . . . . 89

7. Infrared Spectra of 3-deutero-p-Menthane in Hydrocarbon Solvent (a) and Dimenthyl Ether Solvent (b) 97

8. <^-Methinyl Proton Resonance of 4-tert-Butylcyclohexyllithiuni Samples of an Original Reaction Solution in Benzene (a) and Crystalline Material Dissolved in Benzene (b), Diethyl Ether (c) Dimethvl Ether (d). . . . . . „ . 101

9. Infrared Spectrum of 4-tert-Butylcyclohexane-1-di in the Carbon-Deuterium Stretching Region . . . . . . « . . . 104

CHAPTER I

INTRODUCTION

Organolitnium compounds have been employed in synthetic

worK for many years. However only during the last decade

has mucn progress been made in establishing the mechanistic

pathways for the reactions of these compounds. To date

only a few detailed mechanism studies are known.

This lack of knowledge may be traced to three major

problems encountered in studying these reactions. The first

of the problems is the inherent sensitivity to moisture and

oxygen displayed by these compounds. A second pernicious

characteristic afforded by these systems with respect to

mechanism studies is their high specific rate constants.

Although both of these are very serious problems they can

be minimized by careful design of the techniques and

apparatus used in conducting such investigations.

The third problem involves elucidation of the conforma-

tional stability of organolithium compounds. The carbon-

lithium bond has been characterized for many systems as

being a fairly labile bond. The necessity of knowing the

conformational stability for such a compound prior to the

initiation of a mechanism investigation is obvious. ?or

this reason configurational stability studies of organolitnium

i

as well as all group IA-IIIA ;netal-orgaiiic compounds have

proven to be a subject of much current interest.

There are a number of important physical methods which

are used in conformational analysis studies. Of these methods

one of the oldest and most easily applied is the measurement

of optical rotation. This technique relates the experimentally

determined rotation to the molecular architecture usually

termed the conformation. Of course absolute configuration

assignments require the application of one or more of the

other techniques, as for example the currently very popular

nuclear magnetic resonance spectroscopy. However the over-

all stereochemical pathways from reactants to products for

many reactions can be followed very easily by rotation

measurements.

Studies ox the stereochemistry of substitution reactions

of organolithium compounds have long been plagued by the

fact that attempts to form optically active compounds, in

which the lithium atom is bound to an asymmetric tetra-

hedrally hybridized carbon resulted in racemic products.

Evidence for compounds possessing this property has been

observed only when the compounds are prepared and consumed

immediately. As a consequence, only the net stereochemistry

from alkyl halides or other starting materials to final

products could be studied; therefore the stereochemical course

of each individual reaction has been long delayed.

The research proposed for this investigation deals with

this problem of synthesizing ana isolating a conformationally

stable al^yllithium compound. Such compounds have not been

previously reported, perhaps because the carbon lithium

bond possesses an adequate decree of ionic character to

cause it to behave as a labile bond when solvated. However

if this is not a true characterization it way be that a

sufficient understanding of the conditions necessary to produce

such a system is still concealed in the unexplored chcmistry

of such compounds. Hints of these conditions can probably

be found in the published literature. Therefore a thorough

knowledge of the literature concerned with this subject is

imperative for the design of experiments which might yield

informative data. Having pointed out the importance assessed

to earlier investigations of this type it is believed that

these studies should be reviewed.

During the early 1930's the entire outlook on organo-

lithium chemistry was changed by the published work of Ziegler

and Colonius (43). Before this time the syntheses of organo-

lithium compounds were very laborious procedures and investi-

gations were directed primarily to understanding the nature

of these compounds. The work of Ziegler and Colonius (43)

outlines a direct synthesis procedure from lithium metal and

alkyl halide. The results of this work yielded an easy

synthetic method for these very reactive compounds. Shortly

' 4

after this method rras reported C-ilnan and co-workers (19, 20)

evaluated the relative reactivities for a number of these

compounds. They observed these reagents to he reactive as,

and in some cases more reactive than the already well in-

vestigated Grignard reagents.

Grignard reagents had already "been characterized as

very useful synthetic reagents. However, as early as 1911

a problem in understanding the stereochemistry of reactions

employing this reagent was realized by Pickard and Kenyon (31).

They generated the Grignard compound in the conversion of

an optically active alkyl iodide to the corresponding alcohol

and obtained only racemic product. Several years later

Porter (32) reported the results from a similar investigation.

In the hope of synthesizing an optically active compound

with one atom of hydrogen and one atom of deuterium attached

to the asymmetric carbon atom, the Grignard reaction was

applied to several optically active organic bromides. Complete

racemization of the bromides was observed in the formation of

the Grignard reagents.

Schwartz and Johnson (33) have also reported the formation

of a racemic anilide product in the reaction between phenyl

isocyanate and Grignard reagents prepared from optically

active 2-bromooctane. This investigation was conducted in

an attempt to gain a better understanding of reactions between

Grignard reagents and phenyl and -6-napthyl isocyanates. These

reactants were being employed at that time as analytical

reagents to detect and determine the concentration of an

organoiaagneaiuia lialide in solution.

In the period from 1900 to 1930 the chemistry of organo-

sodium compounds occupied the interest of several investigators.

With respect to the stereochemistry question Ott (30) reported

that sodium compounds appeared to react with more stereo-

specificity than evidenced for magnesium compounds. He showed

that the action of sodium on optically active «<-phenylethyl

chloride yielded optically active 2,3-diphenylbutane, along

with the meso form of the hydrocarbon, while the recovered

chloride was found to be extensively racemized. Treatment

of the active o(-phenylethyl chloride with magnesium in moist

ether gave only inactive diphylbutane. In this work he

postulated organosodium intermediates, of which Morton (26)

later demonstrated the existence.

V/allis and Adams (39) during this period employed organo-

sodium compounds in the study of a very fundamental stereo-

chemistry problem. This problem involved the question of the

spatial arrangement of groups in compounds containing a

tricovalent carbon atom. Much interest was attached by the

authors to the possibility of the existence of a non-planar

configuration in these compounds. They noted that if such

a spatial arrangement were true, asymmetry would be possible

when the three groups around the central carbon atom were

6

different. Therefore a study of the behavior of such com-

pounds toward polarized light would enable them to determine

the degree of stability of the antirneric configurations.

The authors realized that it was not possible experi-

mentally to resolve such compounds. Therefore they set out

to prepare the corresponding enantiomorphic quadricovalent

compounds, and then to determine whether the rotatory power

of such compounds is retained or lost upon transition into

the tricovalent state and return to the quadricovalent state.

These transitions seemed to be achieved through the formation

of organosodium compounds in the studies of earbanion and

free radical intermediates. The earbanion intermediates were

reported to be derived from a sodium triarylmethyl compound

prepared from an optically active reagent. Evidence for the

existence of a planar configuration was not observed. However,

in a later publication Wittig, Vidal, and Bohnert (41)

demonstrated this observation to be in error. At this time

it was believed that the V / u r t z reaction proceeded by a free

radical mechanism; therefore, this reaction was employed to

generate these intermediates. Only racemic product was produced

in this investigation thereby demonstrating the existence of

a planar figuration for this tricovalent species.

Tarbell and Weiss (36) reported the first investigation

of organolithium stereochemistry. A study of this type had

not been reported earlier because of the lack of a good

7

synthesis method for the lithium compounds. Also, since most

of the stereochemistry previously studied for similar systems

was of magnesium compounds, no encouragement for any success

had been offered. Then the works of Ziegler (4-3) and Ott (30)

were reported and thereby provided a synthetic method and

evidence for conformational stability in organoaliali chemistry.

It is interesting to note that even though the results re-

ported by Y/allis and Adams (39) contradicted Ott's (30)

observations, his wor c was cited as the basis for this study.

These authors found to their disappointment that when

optically active 2-chlorooctane was reacted with lithium,

the acid formed by carbonation was inactive, and the unre-

acted chloride which was recovered had a lower rotation than

the starting material. They pointed out that the reacemi-

zation might have occurred during the carbonation rather than

during the formation of the lithium compound. However, the

latter alternative seemed the more probable, especially in

view of the results reported in earlier magnesium work. The

formation of the active 2,3-diphenylbutane reported by Ott

(30) was then explained as the result of a reaction between

an optically inactive organosodium compound and an optically

active organic chloride,

Letsinger (25) can be credited with providing the

first insignt into the conditions required to demonstrate

conformational stability in organolithium compounds. He

a

was able to prepare optically active 1-methylheptyllithium

from active 2-iodooctane and sec-butyllithiaiii in a 6 per

cent diethyl ether-petroleum ether solvent mixture at -70

degreea centigrade. At these conditions a 20 per cent net

retention of configuration was observed. This work reported

in 1950 actually opened up the area of organolithium stereo-

chemistry.

Characterization of the lithium compound was accomplished

through the carbonation of derivative. The rotation measure-

ments reported were on the isolated acids and not on the

solutions of lithium compound. The carbonation reactions

were initiated two minutes after completion of the tempera-

ture or time variable studies conducted on the lithium

solutions. A summary of these data is listed in Table I.

TABL3 I

TABLE OF LETSIHGER1S DATA

Run Rotation of Temperature tfime at Listed Acia Number Alkyl Iodide (°C) Temperature Rotation

1 -44.36 -70 1 minute -1.13 2 -41.00 -70 60 minutes -0.96 3 -39.85 0 20 minutes -0.00

These data illustrate that time and temperature are certainly

two very real effects on the conformational stability of

organolithium compounds. However, this work does not provide

any evidence as to whether predominant retention or two

9

successive inversion pathways are followed in the conversion

of the organic iodide to the corresponding acid.

Interest in the stereochemistry of lithium alKenyls

became intense shortly after Letsinger (25) published his

work. Investigations by Braude and Coles (7) and Cur tin

and co-workers (11, 12) were reported in 1951. These works

all indicated that a definite increase over the lithium

alkyls in conformational stability was characteristic of

these compounds. This evidence opened up an interesting

and virtually uninvestigated area of research which dominated

the research in organolithium stereochemistry for the next

ten years. Due to the importance of a number of findings

reported during this time period a summary of these works

will be given.

Prior to the 1950's only the alkenyl lithium derivative

of y^-bromostyrene had been synthesized by Wright (42). Charac-

terization by carbonation of the products derived from the

reaction between lithium metal and this halide indicated only

the formation of the trans-alkenyl acid. Since the study of

this system involved the search for a way to obtain stereo-

isomers of l-phenyl-l,3-butadiene the-interest to continue was

subdued.

However, in 1950 Braude and co-workers (6, 8) reported

the synthesis of isobutenyllithium and cyclohexenyllithium

from the corresponding organic bromides and lithium metal.

10

The work was originated because of the desire to employ

©(-alkenyl organometallic derivatives for the synthesis of

various ethylenic systems. The well-known fact that alkenyl

halides containing a halogen atom adjacent to a ethylenic

"bond are very unreactive toward magnesium definitely limited

the selection of soluble metal derivatives to those of

lithium. The outlook for these compounds was somewhat bleax

since the only alicenyl-lithium system known was styryllithium,

which could be syntehsized using magnesium. Furthermore,

Gilman (18) had reported that in an attempt to synthesize

vinyllithium only dehydrohalogenation product was recovered.

Therefore these successful syntheses demonstrated that a limi-

tation to vinyl halides which also yield Grignard reagents did

not exist.

This success then stimulated Baude and Coles (7) to

determine whether the formation of these lithium derivatives

was limited toy?y£-disubstituted and cyclic alkenyl halides

which cannot easily undergo dehydrohalogenation. This led

to the fortunate choice of cis-propenyllithium, which was

found to be conformational^" stable over long periods of

time in refluxing ether. Curtin and co-workers (11, 12) were

also simultaneously investigating similar systems with an

interest in comparing their stereochemical stabilities. Over

the next ten years Curtin proved to be the principal in-

vestigator of the stereochemistry of alkenyl lithium compounds.

Table II includes a list of the systems studied.

11

TAB lull II

STEREOCHEMISTRY INVESTIGATED ALKSNYL LITHIUM COMPOUNDS

Entry Compound Empirical Formula References A trans-Cg^HC: CLiH (42) B cis- and trans-CH^IIC :CLiH (6, 10) C D 3

cis- and trans-ClC6H5C:CLiC6H5 cis- and trans-C6H5HC:CLiC6H5 cis- and trans-CHjHCiCLiCHj

(11, 13., 14) (12, 14, 28, 29) (5, 16)

These compounds were all prepared by the reaction of

the corresponding alkenyl bromides with either lithium

metal or n-butyllithiuin in ether solvent. A definite dif-

ference in the products formed by the two lithium reagents

was observed for the bromides with structure (&). When

lithium metal is employed as the reactant with these bromides,

where both the R»s may be any combination of phenyl, methyl,

and hydrogen, all the lithium compounds, with the exception

of the R = methyl and R' = phenyl case, are successfully

prepared. However when n-butyllithium is employed the

disubstituted acetylenes are obtained as the major products.

Propenyllithium has proven to be an exception. The case

where R = R* = methyl has not been studied.

Rf II /C = C"

R (&•) "Br

No explanation for the difference in these two lithium

reagents has been proposed by any of the investigators. Curtin

and Koehl (14) did postulate why it was possible to prepare

1 2

the two propenyllithium isomers "by the halogen metal exchange

reaction. Their explanation involves a metallation reaction

of the unreaeted bromides (&•) in which all the alkenyllithiums

containing ^-protons except the less reactive pr ope ny11ithi urn

reagents participate.

In reviewing these stereochemical data reported for

these systems three primary observations are projected. The

first factor which appears to affect their conformational

stabilities is the nature of the organic substituents. For

example an isomerization mechanism appears to be available

for the aryl substituted compounds. Curtin and Koehl (14)

have proposed that this mechanism involves the ionization

of the carbon-lithium bond followed by a shift of an aryl

substituent through a linear transition state (b), which is

diagrammed below. In the 'formation of this intermediate the

P-CIC6H4 li ;c=< •

C6H5 C6H5

p_.Ci-.C6H4

7/ C6H5 v- / 7

(b)

(-) P-CI-C6H4 ^C6H5

;c=c^ C6H5 li

negative charge can be distributed in part on the aryl sub-

stituent, thereby generating a favorable situation for

isomerization to proceed.

The second critical factor demonstrated from these

studies was the temperature. At low temperatures (-40 degrees

to -60 degrees centigrade) all the lithium aIkenyIs were found

to exist as conformationally stable cis— and trans—isomers in

13

diethyl ether. Above 0 degrees centigrade, however, inter-

conversion of the aryl substituted lithium a lice ny Is becomes

prominant. For these compounds the cis isomer showed the

greatest tendency to isomerize. In contrast to this the cis-

propenyllithium appeared to be stable in refluxing ether and

the trans-isomer demonstrated only a slight increase in its

tendency to isomerize.

The last and most significant factor was first reported

by Curtin and Koehl (14, 15). They demonstrated the existence

of a definite solvent effect on the stereochemical stability

not only of the lithium alkenyls but also of lithium alkyls.

It was reported that solutions of cis-o^stilbenyllithium and

of cis- and trans-l-p-chlorophenyl-l,2-diphenylvinyllithium

demonstrate greater stereochemical stability in hydrocarbon

solvents than in diethyl ether. Tetrahydrofuran is still more

effective than diethyl ether in promoting isamerization of

these compounds.

These results suggested that a saturated lithium alicyl

might possess more conformational stability than reported

by Letsinger (25) if it existed in an ether-free hydrocarbon

medium. Under these conditions it was observed that optically

active sec-butyllithium could be prepared and carbonated with

greater than 30 per cent retention of configuration. This

synthesis was achieved by employing the metal-metal exchange

reaction between optically active sec-butylmercury and racemic

2-octyllithium in n-pentane solvent.

14

The works of Letsitiger (25) and Cur tin and co-workers

(10, 14, 15) were important pioneering investigations and

illustrated the complexity of the stereochemistry of organo-

lithiui;! compounds. Port her more, the volume of stereochemical

research on both Groups IA and IIA organometallic systems

published since these works is certainly indicative that

these investigators revealed a very challenging area of

research.

One fact that must be emphasized concerning the works

of these two investigators is that they have provided in-

formation regarding only the net stereochemistry from alkyl

halide to final product. At no time were they able to make

any measurements on the organolithium solutions themselves.

Therefore these stability data are derived from determinations

by chemical means only. The fact that the stereochemical

course of each individual reaction could not be determined

by the employed analytical technique initiated investigations

of other methods. Hesmeyanov and Borisov (28) applied their

method of "odd and even cycles" and Allinger and Hermann (2)

employed infrared spectroscopy techniques. However the

results recorded in both of these studies were highly criti-

cized .

Seyferth and Vaughan (35) were among the first to apply

nuclear magnetic resonance spectroscopy to investigate this

stereochemistry. They were able to show unequivocally that

15

the preparation and carbonation procedures employed for the

propenyllithium system involved retention of the geometric

configurations. The oases for these stereochemical assign-

ments were provided "by the differences observed in the

coupling constant values for the two lithium isomers.

This technique has also been employed to study the

conformational stabilities of alkyllithiura compounds in

diethyl ether- An investigation of this type reports the

temperature at which the ^-methylene protons become magnet-

ically equivalent. Non-equivalency of these protons can

be derived primarily from either an unequal population of

rotamers about the carbon-carbon bond or by some inversion

mechanism whereby the protons can exchange places with one

another. For organolithium compounds such a mechanism in a

polar solvent can be conceived through the formation of a

carbanion and a solvated lithium cation. Positions of the

spectral lines as a function of temperature are used to

determine whether changes in population conformers occurs.

Therefore a distinction between these two causes may be

easily obtained.

Most of the published data have been reported on the

inversion rates of alkyl Grignard reagents and dialkyl

magnesium compounds. However Cram (9) has reviewed a private

communication from Roberts, -hitesides, and Witanowski which

reports such a study on 3,3-dimethylbutyllithium in diethyl

16

ether. The lithium compound was found to undergo rapid

inversion above zero degrees centigrade. Frankel, Adams,

and Williams (17) have also claimed such an investigation,

but a review of the data illustrate that magnesium compounds

were primarily the system being examined.

In the early 1950*s Walborsky became interested in the

stereochemical fate of a pair of nobonding electrons in an

orbital which is part of a three-membered ring (37). The

exo-orbitals of, such, a system have been characterized as

possessing more 11 s" character than is reported for tetra-

hedrally hydridized orbitals. Therefore this type of molecule

has a great deal of double bond character associated with it

which should lead to a higher barrier of inversion for such

a pair of electrons.

By I960 both Walborsky and co-workers (37, 38) and

Applequist and Peterson (4) had anticipated that cyclo-

propyllithium compounds might exhibit conformational stabil-

ities comparable to lithium alkenyls. Walborsky and co-

workers (37, 33) investigated the optical stability of a

l-methyl-2,2-diphenylcyclopropyllithium enantiomer. This

study was conducted by chemical means through protonation

with methanol, carbonation, bromination and iodination.

These reactions were carried out in a variety of solvents

such as diethyl ether, 1,2-dimethoxyethane, benzene, petroleum

ether, and various combinations of these solvents over a

17

•temperature range from 0 degrees to 30 degrees centigrade.

Polarimetry measurements were not made on the cyclopropyllithium

solutions. The results recorded from these investigations

demonstrated that the lithium compound could be prepared and

made to undergo electrophilic substitution with essentially

complete stereospecificity.

Applequist and Peterson (4) investigated the configura-

tional stability of the geometrical isomers of cis- and

trans-2-methylcyclopropyllithium. These compounds were

generated by exchange of the bromide isomers with isopro-

pyllithium under various conditions of ether-hydrocarbon

solvent mixtures and temperatures. The resulting solutions

were analyzed by brominolysis and carbonation. These

carbonation reactions were reported to proceed stereospecif- .

ically with retention. However a.lack of over-all stereo-/

specificity was reported for the brominolysis reaction.

The inescapable conclusion for these investigations is

that the cyclopropyllithium reagents do not interconvert at

a significant rate under the conditions employed, oeyferth

and Cohen (34) confirmed these findings by a nuclear magnetic

resonance investigation of the spectrum of cyclopropyllithium

in tetrahydrofuran.

A disagreement on the stereospecificity of the

brominolysis of these lithium reagents is obvious from

these investigations. Applequist speculated that the reason

,d>

for the non-stereospocificity of the brouinolysis was

derived from the existence of either a free radical or a

carboniuiij ion reaction intermediate. The radical arid

carbonium ion possibilities, it pas thought, could be

differentiated in the 2-norbornyl case. According to the

work of rCooyiuan and Vegfcer (23) this system should give

considerable endo-norbornyl bromide in any reaction of

norbornyl radica] with bromine; but exclusively exo-

product was predicted by "finsxein and co-workers (40) if

a bromine and norbornyl cation reaction was predominant.

In the 2-norbornyllithium vjorm Applequist and Chmurny

(3) investigated the stereochemistry of a number of electro-

philic substitution reactions of this lithium compound. The .

direct synthesis method was employed at 20 degrees and 32

degrees centigrade, the temperature of refluxing n-pentane

solvent, for preparing the lithium reagent, /t 20 degrees

different ratios of.endo- and exo-lithium isomers were reported

from the two different chloride isomers. An intermediate ratio,

which was independent of the chloride isomers, of endo- and

exo-lithium epimers was reported for the refluxing solvent

procedure. Average product yields of only 10 per ccnt required

19

that the lithium isomer distributions be determined by

chemical means. To make these isomer distribution claims

for the lithium compounds these authors made two assumptions.

The first assumption is that car Donation of alkyllithium

compounds proceeds by retention of configuration. The second

and more important of the two assumptions is that each isomer

reacts with the same degree of stereospecificity. Neither of

these assumptions was demonstrated experimentally to be valid.

Having solutions which were believed to contain varied

concentrations of the two lithium epimers these investigators

were then able to study the stereochemical pathways afforded

by these isomers. One observation made from these electro-

philic substitution data was that different reagents can give

different product mixtures. This indicates that not all the

reactions of this type proceed with simple retention of con-

figuration. It was also reported that the stereospecificity

of the brominolysis reaction appeared to proceed with prefer-

ential inversion. This unexpected result definitely eliminated

the cation intermediate postulated. However from the stereo-

chemical data alone it was not clear whether or not a radical

or carbonation intermediate could be postulated. It was con-

cluded that concerted aliphatic electrophilic substitutions,

m contrast with nucleophilic substitutions, have both inversion

and retention pathways available.

The nature of 2-norbornyllithium was also studied bv

Mulvaney and Garlund (27) prior to the work reported by

20

Applequist and Chmurny (3). They were interested in the

stereochemistry of alkyllithiuni addition to nonconjugated

olefins. It was their hope that employment of the highly

reactive norborene would facilitate determination of this

stereochemistry. They were able to execute this addition

reaction and characterize the 2-norbornyllithium product

"by chemical means. Evidence for the exo-derivatives only

was reported and the existence of a conformational^" stable

lithium compound was postulated from this observation.

Y ithin the last two'years Hill (21), Krieghoff and

Cowan (24), and Jensen and Nakamaye (22) have all made sig-

nificant contributions to the understanding of the stereo-

chemistry of norbornylmagnesium compounds. Nuclear magnetic

resonance spectroscopy was the primary tool employed in these .

studies. Application of this technique afforded these in-

vestigators a method of studying the organornetallic compounds

themselves instead of their derivatives. Por example Jensen

and Nakamaye (22) reported that each of the isomers in a

mixture of exo- and endo-norbornyl Grignard reagents can readily

be observed and identified from a proton magnetic resonance

spectrum. They were also able to demonstrate by the slow re-

appearance of the selectively destroyed exo-isomer that a

slow equilibrium between the two isomers exists. Armed with

these data they were then able to demonstrate easily the stereo-

chemistry of any individual electrophilic substitution reaction.

21

To date no investigation of this type has been reported

for alkyllithium compounds. Seyferth and co-workers (34, 35)

have determined the stereochemical stabilities of propenyl-

lithiuju and cyclopropyllithium by nuclear magnetic resonance

spectroscopy. There is no direct evidence that the results

obtained from these investigations can be extrapolated to

predict the stereochemistry of alkyl systems. Therefore it

is still unknown whether the observed stereochemistry for

alkyllithium compounds results from two reactions which

both occur by retention or by inversion of configuration.

It is thus obvious that an investigation of this type is

needed.

Such an investigation will require an alkyllithium com-

pound with structural stability comparable to the previously

reviewed systems. The stabilities of these systems appear to

be derived from their molecular geometries—for example, the

bridged norbornyl and trigonally hybridized alkenyl systems.

High energy barriers to most inversion mechanisms are evidently

associated with these geometries. The degree to which this

effect actually exists is definitely of interest, especially

in associated organolithium systems. "Thile it is desirable

to choose a system whose stability to inversion can be varied,

at the same time it must have approximately the bonding

character of simple alkyllithium systems. For such a system

not only an investigation of the stability effects but also

22

the stereochemistry of electrophilic substitutions of simple

alkyllithium compounds would be attainable.

It has been known for some time that the conformational

stabilities of cyclonexyl derivatives are dependent upon the

substituents attached to the ring. Also short of the bridged

derivatives, these systems approximate the bonding character

of secondary al^yl systems. Therefore an investigation of

substituted cyclohexyllithium compounds in a variety of

solvent and temperature conditions has been proposed for

this investigation. Although problems—such as those experi-

enced by Alexandrou (1)—are expected, it seems that definite

contributions to the stereochemistry of alkyllithium com-

pounds are attainable using this system. Furthermore fulfill-

ment of the quest to isolate a conformational^ stable alkyl- •

lithium compound might be observed from the synthesis of a

substituted 'cyclohexyllithium compound.

CHAPTER BIBLIOGRAPHY

1. Alexandrou, N. 13. , "Reaction of n-Butyllithium with cis- and trans-4-tert-Butylcyclohexyl Bromide and Mercuric Bromide," Journal of Organonetallic Chemistry, V (1966), 301. ,L

2. Allinger, Norman I. and Robert L. Hermann, "Infrared Spectra of cis- and trans-Proyenyllithium Stereo-chemistry of the Reaction of Vinyl Chlorides with Lithium Metal," Journal of Organic Chemistry. XXVI (1961), 1040. 6 >L

3. Applequist, Douglas E. and Gwendolyn Heal Chmurny, "Stereochemistry of Electrophilic Substitutions on 2-Norbornyllithium," Journal of the American Chemical Society, LXXXIX (1967) ,"575".

4. Applequist, Douglas E. and Alan H. Peterson, "The Configurational Stability of cis- and trans-2-methyl-cyclopropyllithium and Some Observations on the Stereochemistry of their Reactions with Bromine and Carbon Dioxide," Journal of the American Chemical Society, LXXXIII (1§6I)", Wo2~

5. Bordwell, P. G. and Phillip S. Landis, "Elimination Reactions. VII. A Comparison of Some Eliminations in Open-Chain and Cyclic Systems," Journal of the American Chemical Society, LXXIX (1957), 1593".

6. Braude, E. A. and J. A. Coles, "Alkenylation by Use of Lithium Alicenyls. Part IV. The Condensation of cycloliexeny 1-lithium with Aldehydes and Ketones, and Some Reactions of the Resulting Carbinols. (Studies of Molecular Rearrangement. Part V.)," Journal of the Chemical Society.(1950), 2014.

« "Alkenylation Employing Lithium Alicenyls. Part V. The Formation and Some Reactions of cis-Propenyllithium," Journal of the Chemical Society (1951), 2078.

23

24

3. Brauae, E. A. and C. J. Timmons, "Alkenylation "by Use of Lithium Alkenyls. Part I. The Condensation of isoButenyl-lithium with Benzaldehyde and Acetophenone, and the Oxotropic Rearrangement of the Resulting Carbinols. (Studies in the Molecular Rearrangement. Part III.)," Journal of the Chemical Society (1950), 2000,

9. Craw, Donald J., Fundamentals of Carbanion Chemistry, New York, Academic Press, Inc., 1965.

10. Curtin, David Y. and John W. Crump, "Configurational Stabilities of Stereoisomeric Vinyllithium Compounds." Journal of the American Chemical Society, LXXX (1953), U&T.

11. Curtin, David Y. and Elbert E. Harris, "The Preparation of Stereoisomeric Alx^enyl Lithium Compounds," Journal of the American Chemical Society, LXXIII (1951), £l±b.

12. , "Preparation of Stereoisomeric Alkenyl-lithium Compounds. II. cis-and trans-l,2-Diphenylvinyllithium and U- and /?-Styryllithium," Journal of the American Chemical Society, LXIII (1951), 45197

13. Curtin, David Y. , Harry Johnson, and Edwin G. Ste.iner, "Stereoisomeric Vinyllithium Compounds. III. Reactions with Aldehydes, Ketones, and Methyl Iodide," Journal of the American Chemical Society, LXXVII (19550, vser.

14. Curtin, David Y. and William J. Xoehl, "Effect of Solvent on the Steric Stability of Lithium Reagents," Journal of the American Chemical Society, LXXXIV (1962), 1967.

15. , "Optically Active sec-ButylIithium," Chemistry and Industry (i960), 262.

16. Dreiding, Andre S. and Richard J. Pratt, "The Carboxyl-ation of cis- and trans-2-Butenyl-2-lithium. A Stereospecific Synthesis of Angelic Acid," Journal of the American Chemical Society, LXXVI (1954), 1902.

17- Praenkel, Gideon, David G. Adams, and James Williams, "Structure of Grignard Reagents and Organolithium Compounds," Tetrahedron Letters (1963), 767*

25

18. Gilnan, Henry and A. U. Kaubein, "Ethynyllithium Com-pounds by Halogen-Ketal InterconversionsJournal of the American Chemical Society, LXVII (1945), 1420.

19. Gilman, Henry and R. H. Kirby, "The Relative Reactivities of Organolithium and Organomagnesiun Compounds," Journal of the American Chemical Society, LV (1933), i r o _

20. Gilman, Henry and Myrl Lichtenwalter, "Relative Reactivities of Organometallic Compounds. X. Lithium and Magnesium." Recueil des Travaux Chemiques des Pays-bas, LV (193b), 561.

21. Hill, Alexander B., "Nuclear Magnetic Spectra of the Bornyl and Norbornyl Grignard Reagents," Journal of Organic Chemistry, XXXI (1966), 20.

22. Jensen., Frederick R. and Kay L. Nakamaye, "Preparation of Geometrically Isomeric Grignard Reagents and the Stereochemical Courses of their Reactions," Journal of the American Chemical Society, LXXXVIII (1966),

23. Kooyman, E. C. and G. C. Vegter, "Bicyclanes-I. The Halogenation of 2:2:l-Bicycloheptane (Norbornane)," Tetrahedron, IV (1958), 382.

24. Krieghoff, Nilde G. and Dwaine 0. Cowan, "Stereochemistry of the Grignard Reagents from exo- and endo-Norbornyl Chlorides," Journal of the American Chemical Society, LXXXVIII (19bb),13227

25. Letsinger, Robert L., "Formation of Optically Active 1-Methylheptyllithium," Journal of the American Chemical Society, LXXII (1950), T$42.

26. Morton, Avery A. and Graham M. Richardson, "Condensation by Sodium. XVI. The Formation of Decane in the Wurtz Reaction," Journal of the American Chemical Society, LXII (1940), 1237"

27. Mulvaney, J. E. and Z. G. Gardlund, "Alkyllithium Compounds and Nonconjugated Olefins. The Nature of a Norbornyllithium Compounds," Journal of Organic Chemistry, XXX (1965), 4321.

26

28. Nesmeyanov, A. N. and A. 3. Borisov, "Stereochemistry ox Jilectr ophilic and Homolytic Substitution at the Olefinic Carbon," Tetrahedron, I (1957), 158.

29. Nesaeyanov, A. W., A. S. Borisov, and N. A. Volkenau, "Synthesis of Stereoisomeric Grganomercury Compounds from Organolithium Compounds," Izvestiva Akademii Hauk Soyuza Sovetskiieh Sutsialisticheskikh Respublik r r y w r w .

** • <

30. Ott, Srwin, "Uber die Additionreaktionen der Athylen-Doppelbindung unter Bildung von Athan-Verbindungen rait zwei, durch den Additionsvorgang entstehenden, asymmetrischen ivchlenstoffatomen und der ijinfluss der Reaktionsgeschwindigkeit auf den Reaktionsverlauf," Berichte der Deutschen Cheraischen G-esellschaft, LXI (1923), 21??.:

31. Pickard, Robert Howson and Joseph Kenyon, "VII -Investigations on the Dependence of Rotatory Power on Chemical Constitution. Part I. The Rotations of the Simplest Secondary Alcohols of the Patty Series," Journal of the Chemical Society, LXXXXIX (1911), 45.

32. Porter, C. W. , "Racemization in the Preparation of the Grignard Reagent," Journal of the American Chemical Society, LVII (1935),1436.

33. Schwartz, A. M. and John R. Johnson, "Characterization of Alkyl Halides and Organoinagnesium Halides," Journal of the American Chemical Society, LIII (1931), 1063.

34. Seyferth, Dietmar and Harvey M. Cohen, "The Stability of Cyclopropyllithiujii in Diethyl Ether and in TetrahydrofuranJournal of Organometallic Chemistry, I (1963), 15. ^

35. Seyferth, Dietmar and Lawrence G-. Vaughan, "The Isomeric 1-Propenyllithium Reagents: Their M R Spectra and the Stereochemistry of Their Formation by Direct Reaction and by the Transmetalation Reaction," Journal of Organometallic Chemistry, I (1963), 20T.

36. Tarbell, D. S. and Marvin Weiss, "The Action of Lithium on an Optically Active Aliphatic Chloride," Journal of the American Chemical Society, LXI (1939), 120"3.

2 7

3 7 . " ' a l b o r s ^ y , T i . M . a n d ? . - J . 7 . : : ; p a n t a i o , " ' " y c l o p r o p a n o n . V I . R e t e n t i o n o f O p t i c a l A c l i v i t y r . t t i C n n f i g u r a t i o n i n - c h o O y c l o ' o r o p y l C a r b a n i o n , " j o u r n a j o f t n e A m e r i c a n • d h o . - a c a l S o c i e t y , I X X X I ( 1 ^ 5 ) , 5 3 5 5 * " . '

3 8 . " r a ] l ) o r s . c y f i l . i f . , 1 ? . J . l u p a a t a t o , a n d A . Y o u n g , " C y c ? o p r c p a n e s . X V . T h e O p t i c a l S t a b i l i t y o f l - l l c t l u y 1 , . ? - a i p i i o n y i - o / c l o p r o p y 1 1 i t h i u m , " J o u r n a l o f t h e A m e r i c a n C h o r i o n 1 S o c i e t y , I i X X X V I ( 1 9 6 4 ; ,

3 9 . ' . ' ' a l l i s , ' v v e r e t I S . a n d F r e d e r i c I I . A d a ^ s , " T h e S p a t i a l C o n i i g u r a t i o n o f t h e V a l e n c e s i n T r i e o v a l e n t C a r b o n C o m p o u n d s J o u r n a l o f t h e A m e r i c a n C h e m i c a l S o c i e t y , I V ( 1 9 3 3 ) , 3 3 5 3 " : "

4 0 . W i n s t e i n , S . , 33. C l i p p i n g e r , R o b e r t H o w e , a n d E l l i o t V o g e I f a n g e r , " T h e H o n c l a s s i c a l h o r b o r n y l C a t i o n , " J o u r n a l o f t h e A m e r i c a n C h e m i c a l S o c i e t y , L X X X V I I

3 / 6 .

4 1 . F i t t i g , G e o r g e , P e d e r i c o V i d a l , a n d E r v / i n B o h n e r t , " f i b e r d i e k o n f i g u r a t i v e B e s t a n d i g ^ e i t m e t a i l o r g a n i s e h e r V e r b i n d u n g e n , " B e r i c l i t e d e r D e u t s c h e n C h e m i s c h e n G e s e l l . s c h a f t , i r m i T T l ^ ) 7 ~ 3 y j ;

4 2 . V / r i g h c , G e o r g e 3 ? . , " O r g a n o m e t a l l i c C o m p o u n d s o f S t y r e n e J o u r n a ] o f O r g a n i c C h e m i s t r y , I ( 1 9 3 6 ) , 4 3 7 .

4 3 . Z i e g l e r , K . a n d H . C o l o n i u s , " U n t e r s u c h u n g e n u b e r a l . c a l i -o r g a n i s c h e V e r b i n d u n g e n . V . ? 5 i n e " b e q u e m e S y n t h e o e e i n f a c h e r L i t h i u m a l ^ y l e J u s t u s L i e b i g s A n n a l e n d e r C h e m i e , C C C C L X X I X ( 1 9 3 0 ) , ± 5 5 1 ~~ "

CHAPTER II

.EXPERIMENTAL PROCEDURES

Descriptions of Reagents and Syntheses Procedures

Liaterials

The quality of materials used is a critical factor in

preparing and studying the chemical properties of oxygen

and moisture sensitive organometallic compounds. This is

exemplified by the fact that these compounds are character-

ized primarily through studies of their derivatives. There-

fore considerable attention has been given to determining

the quality of reagents used in this research investigation.

The lithium sand employed for the preparation of the

menthyl- and 4-tert-butylcyclohexyllithium compounds was

prepared from one-half inch diameter lithium rod containing

2 per cent sodium metal (16, 23). The lithium rod, purchased

from the Lithium Corporation of America, was cut into small

chips and added to a 500 milliliter flask containing 100

milliliters of heavy mineral oil. After heating the flask

to a temperature slightly above the melting point of the

lithium metal it was shaken vigorously and cooled very

rapidly, yielding a finely dispersed sand. The sand was

then transferred onto a coarse fritted funnel and washed

free of the mineral oil with dry n-pentane. The lithium

28

29

was kept suspended in the pentane solvent throughout this

washing period and during its addition to an argon-purged

reaction vessel.

Pure grade n-pentane obtained from the Phillips

Petroleum Company was dried by distilling it from lithium

aluminum hydride and stored under a nitrogen blanket. Other

hydrocarbon solvents used in various applications were also

purchased from Phillips Petroleum Company and treated in the

same manner before use.

Prepurified argon and nitrogen gases were used for all

processes which required atmospheres free from water and

oxygen.

Carbon dioxide, compressed air, dimethyl ether, carbon

tetrachloride, bromine, deuterium oxide, and anhydrous

magnesium sulfate were used as obtained from commercial

suppliers. Ethylene dibromide, methyl iodide, and diethyl

ether were always distilled prior to use.

(-)-Me nthy1 Chloride

The procedure reported by Smith and Wright (21) for

the preparation of highly levorotatory menthyl chloride was

utilized for synthesizing this diastereomer. These inves-

tigators found they could increase the levorotation of the

chloride isomer over other known synthetic methods by

treating (-)-menthol with a solution of zinc chloride in

hydrochloric acid, commonly known as the Lucas reagent. After

30

five hours at 35 degreeo centigrade the menthol is con-

verted in 90 per cent yield to a product with a specific

rotation equalling -45 degrees.

T.he procedure for a typical menthyl chloride preparation

consists of dissolving 226 grams of anhydrous zinc chloride

in 154 milliliters of concentrated hydrochloric acid. This

procedure is carried out in a 500 milliliter three—neck

round-bottom flask fitted with a Teflon stirring gland and

a Ilirschberg stirrer. The flask is immersed in an ice water

bath throughout the preparation of the lucas reagent. To r* 2.*

this solution 78 grams of (-)-menthol [<*1^ =-54.8°( n-pentane,

c. ,2.88)] , purchased from the Aldrich Chemical Company,

is added and the temperature is raised to 35 degrees centi-

grade by replacing the ice water bath with a heating mantle.

The reaction vessel is maintained at this temperature with

stirring for five hours.

The product from this reaction was isolated by distillation.

In order to facilitate this process, prior to distillation the

product was extracted with n-pentane solvent and treated with

concentrated sulfuric acid to remove any unsaturates. This

was accomplished by washing the n-pentane extract with 25

milliliters aliquots of sulfuric acid until no yellow

coloration appeared in the acid layer. The hydrocarbon la-'er

was then washed with water and dried over anhydrous magnesium

sulfate. The magnesium sulfate was removed by filtration and

31

the menthyl chloride was concentrated prior to distillation

oy evaporating the n—pentane solvent with a dry air purge.

Kenthyl chloride distillate was obtained at 98 degrees

centigrade and 10 torr, and the specific rotation of this

distillate, recovered in 89 per cent yield was a £ [o<]o =

-43.5°(n-pentane,c. ,5.02)]. A proton magnetic resonance

spectrum of this product demonstrated the characteristic

axial o<-methinyl resonance at 6.75 "3"-

This distillate was also analyzed on a 10 foot by one

quarter inch aluminum tubing gas liquid chromatography

column pa Cited with 20 per cent XF-1150 cyanosilicone substrate

on 60-80 mesh chromosorb P packing material. A temperature

of 110 degrees centigrade and a helium flow of 300 milliliters

per second were employed to obtain retention times of 18.75

minutes for the neomenthyl isomer and 20.5 minutes for the

menthyl isomer. Less than 5 per cent of the neomenthyl

isomer was detected.

( )-Neomenthyl Chloride

A selective synthesis of the neomenthyl chloride

diastereomer has been reported by Horner, Oediger, and

Hoffman (14)• The conversion of (-)-menthol to neomenthyl

chloride was effected by triphenylphosphine dichloride which

proves to be a very selective reagent. The authors reported

a 93 per cent theoretical yield of product with specific

rotation equalling +42.5 degrees. Although the procedure

*7 2

reported by these investigators was followed as closely

as possible typical yields averaging only 60 per cent were

obtained. A number of variations were tried to improve

the yield for this reaction with only limited success.

The procedure followed for preparing this isomer requires

the saturation of 230 milliliters of carbon tetrachloride

solvent with chlorine gas. The saturation point was determined

to be 0.08 grams of chlorine per milliliter. This solution

is added dropwise to a solution consisting of 65.5 grains of

triphenyl phosphine dissolved in 300 milliliters of carbon

tetrachloride solvent under a nitrogen atmosphere at 0 degrees

centigrade. Upon completion of this addition, thereby form-

ing the triphenyl phosphine dichloride reagent, 39 grams of

(-)-menthol are added and the reaction flask is stirred

vigorously and heated to 75 degrees centigrade for a period

of thirty minutes.. This reaction is carried out in a three-

neck round bottom flask equipped with the same stirring

assembly employed in the menthyl chloride preparation and

a Friedrichs reflux condenser equipped with a nitrogen

inlet adapter. A thermometer occupies the other neck of

the reaction flask. Upon cooling, the triphenylphosphine-

hydroxyl chloride by-product precipitates from solution as

a white solid.

Isolation of the neomenthyl chloride is achieved by

distillation following the removal of the- triphenylphsphine

33

hydroxylchloride by filtration. The carbon tetrachloride is

stripped from the solution under reduced pressure and the

diastereomer is distilled over at 91 degrees centigrade and

15 torr. The product recovered in 58 per cent yield was r

observed to possess a specific rotation of <*<1 =4-53.7 (n-

octane ,c. ,1.27) • The characteristic equatorial c<-methinyl

resonance at 5.5^ was observed from a proton magnetic

resonance spectrum of this distillate. Furthermore less

than 5 per cent of the menthyl isomer was detected using

the gas liquid chromatography analysis described in the

menthyl chloride analysis section.

cis-4~tert-Butylc.yclohex.yl Chloride

The method of Horner, Oediger, and Hoffman (14) was

also applied to the preparation of cis-4-tert-butylcyclohexanol

chloride. The only commercially available 4-tert-butyl-

cyclohexanol consists of an 80 per cent trans: 20 per cent cis

mixture of isomers. For this reason one of the selective

chlorinating reagents utilized in the menthyl chloride

synthesis was employed. The phosphine dichloride reagent

proved to be superior to the Lucas reagent for this particular

system.

In a typical synthesis triphenylphosphine dichloride

was prepared by the same procedure as described for the

neomenthyl chloride synthesis. While this reagent is Itept

at 0 degrees centigrade under a nitrogen atmosphere, 35.5

34

grams of commercial 4-tert-butylcyclohexanol dissolved in

250 milliliters of carbon tetrachloride are added, The

mixture is heated to 75 degrees centigrade and allowed to

reflux for two hours. The chloride product is isolated in

the same manner as the neomenthyl chloride. The reaction

supplies only a 35 per cent yield of product which distills

over at 31 degrees centigrade and 10 torr. Characterization

and purity of this distillate was determined by infrared,

nuclear magnetic resonance, and gas chromatographic techniques.

The infrared spectra of both the cis- and the trans-4-

tert-butylcyclohexyl chlorides have been reported by Greene,

Chu, and Walia (13). The infrared spectrum of the chloride

distillate corresponded well with the spectrum given for the

cis-chloride isomer. The strong carbon-chlorine stretching

absorption was observed at 639 wave numbers. According to

the published findings of Larnaudie (17), this corresponds

to the carbon-chlorine stretching region observed for axial

monochlorocyclohexane derivatives.

Separation of the chloride isomers on a gas liquid

phase chromatographic column was accomplished on a 10 foot

by one-quarter inch aluminum tubing column packed with

20 per cent XF-1150 cyanosilicone substrate on 60-80 mesh

Chromosorb P packing material. A temperature of 118 degrees

centigrade and a helium flow of 300 milliliters per minute

were employed to obtain retention times of 17 minutes for

35

the cis isomer and 19*5 minutes for the trans. In a typical

preparation the isolated distillate usually contained from

7 to 10 per cent of the trans isomer. Unreacted cis-4-tert-

batylc/clohexanol has been identified in the chloride

distillate and although its concentration may vary it never

exceeds 10 per cent.

Nuclear magnetic resonance was also employed as one of

the primary methods for characterizing the chloride isomers.

For these two chloride isomers values of 5.75 and 6.55

were recorded for the cis and trans c<-methinyl resonances

respectively. A detailed discussion of the application of

this analytical method to substituted eyclohexyl systems

will be included in a later section.

Preparation of (4-)-Menthyllithium

Menthyllithium was prepared in dry n-pentane under an

argon atmosphere from menthyl or neomenthyl chloride and

lithium sand according to the method originated by Gillian,

Langham, and Moore (12). This method was employed in

preference to other organolithium preparative methods in an

attempt to retard the competing elimination reaction reported

by Alexandrou (1) for a similar system. Yields ranging from

50 to 60 per cent were usually obtained for the menthyl system

by this method. It was observed tnat preparation of the

lithium sand and distillation of the menthyl chloride isomer

36

immediately prior to conducting the men thy llithium synthesis

facilitated the attainment of higher yields.

A one-quarter mole preparation of menthyllithium is

carried out using a 500~milliliter three-neck round bottom

flask equipped with standard taper 24/40 ground glass joints.

The flask is equipped with a Hirschberg stirrer in a Teflon

Asco stirring gland which seals with Neoprene rubber "o"-rings.

A 150-milliliter cylindrical separatory funnel with pressure

equalizing line is used in one neck for the addition of the

menthyl chloride epimer. An argon gas inlet connected by a

standard taper 24/40 joint to a Friedrichs reflux condenser

fills the third neck of the reaction flask. The reaction

flask is heated by a Glascol mantle controlled by a rheostat.

The first step in the preparation is to wash five grams

of lithium sand into the argon purged reaction flask with

dry n-penta-ne. A 43 gram sample of menthyl or neomenthyl

chloride is diluted to 125 milliliters in dry n-pentane and

this solution is loaded into the addition funnel. After

adapting the addition funnel to the reaction flask it is

purged with argon, and the entire reaction assembly is then

adjusted to remain under a positive argon pressure of ap-

proximately one centimeter of mercury. Stirring is started

and the contents of the flask are brought to a slow reflux.

After the reflux has begun, 10 per cent of the menthyl chloride

solution is added to the reaction flask to initiate the re-

action.

37

The initiation of this reaction is slower and not so

obvious as is demonstrated by many of the organolithium

compounds prepared by this method. A color change from

metallic gray to a purplish gray is observed after approx-

imate1^* thirty to forty-five minutes. Upon addition of the

halide the heat is usually reduced, thereby diminishing the

reflux rate. Upon evidence of the slight color change an

increase in the reflux rate can also be observed. If neither

of these two indications of initiation is observed after

forty-five minutes then evidence may- be seen by stopping the

stirring and dropping the heating mantle. If the reaction

has been initiated some metallic particles will sink in the

pentane solvent.

Upon evidence that the reaction has been initiated the

inenthyl chloride solution is added dropwise at a rate which

will maintain reflux of the solvent. It is believed that

reflux should be maintained at all times to eliminate the

production of hot spots in the flask which can cause decom-

position of the organolithium compound and elimination of

hydrogen chloride from the menthyl chloride reactant. Addition

is generally completed after three or four hours. To insure

that the reaction is complete the contents of the flask are

brought to reflux for one hour and then stirred without reflux

for two additional hours.

The reaction product is worked up in-the dry box. It

is first filtered through a medium fritted funnel to separate

38

the filtrate away from the lithium chloride by-product. A

clear yellowish-colored filtrate is usually obtained. If

the solution is- cloudy it should be refiltered to insure

that all of the lithium chloride has been removed. The

solution is then concentrated by removing the pentane

solvent by evaporation. This process, which cools as well

as concentrates, is carried out by applying vacuum to the

flask containing the solution and catching the volatile

solvent in a dry-ice and acetone trap. As the solution is

concentrated and the temperature drops the menthyllithium

is crystallized from the solution. Agitation of the flask

during the evaporation process appears to facilitate the

crystallization step. After the solution has been concen-

trated from a volume of approximately 250 milliliters to

approximately 60 to 70 milliliters, the solution will take

on the appearance of a milky orange liquid if a good yield

of menthyllithium is present. At this point the sample is

filtered through a coarse fritted funnel as rapidly as

possible.

The crystalline menthyllithium isolated on the funnel

frit usually contains p-menthene impurities. The greater

the impurity content the more muddy the consistency of the

product. Washing this material can be accomplished with

pentane solvent. The menthyllithium is very soluble in

n-pentane and therefore care must be taken not to wash the

3y

lithium compound too freely. The usual procedure is to

-.-/ash dorm the sides of the runnel and the solid material

very slo'.vly with a pipette wi.ii.Le vacuum is using applied

to the system. After washing, the vacuum is used to dry

the material for several minutes.

The procedure described above for isolating the

crystalline menthyllithium is repeated two or three times

depending upon yield of carbon bound lithium obtained.

Usually about one-half of the menthyllithium synthesized

can be isolated in this manner. After this quantity has

been removed the concentration of the menthenes becomes

large enough to keep the remaining product in solution.

Isolation of 10 grams of crystalline material is usually

considered a good yield of product.

Recrystallization of the isolated menthyllithium can

be achieved using n-pentane as the solvent. Due to the

high solubility of menthyllithium in this hydrocarbon the

process can be a difficult one to effect. However, it has

been found that if a high concentration of tnenthyllithium

is used in the solution of redissolved crystalline product

the attainment of recrystallized material is greatly

facilitated. The procedure used in obtaining recrystallized

material is the same as described for isolating crystalline

menthyllithium from the reaction mixture. Typical yields

of the recrystallized product averaged about 10 per cent of

the redissolved crystalline material.

40

tr a n s - 4 -1 e r t -B u t y 1 cy c 1 one xy 11 i t h i urn

Alexandrou (1) reported a 25 per cent yield of 4-tert-

butylcyclohexene from the halogen metal exchange reaction of

the trans-bromide with n-butyllithium. He also tried the

direct reaction with copper activated lithium metal as

described by Curtin and Xoehl (6), and a metal-metal exchange

reaction with isomers of 4-tert-butylcyclohexyl mercuric

bromides and n-butyllithium with no improvement. The lacA

of evidence for the formation of the organolithium compound

was indicative that 4-tert-butylcyclohexyllithium would be

more difficult to synthesize than menthyllithium. Since

elimination appeared to be the major reaction, a change from

bromide to chloride reagent was made in an attempt to retard

this process. Also a change to the same direct method re-

ported by Gilman, Langham and Moore (12) to be employed for

the synthesis of menthyllithium was used for this preparation.

One change in the direct method was made after two synthetic

trials yielded only 10 per cent product. This change was to

increase the volume of solvent from 500 to 1000 milliliters.

A reduction in elimination and coupling by-products was noted

by an increase in organolithium compound yield to 50 per cent

when synthesized in the more dilute system.

Because lower yields of organolithium compound are

obtained for this system and the product obtained is only

sparingly soluble in the n-pentane solvent, changes in the

41

isolation procedure for the ^-tert-butylcyclohexyllithiom

compound are required. Separation of the crystalline solid

from the 4-tert-butylcyclohexene by-product was successful

only when the yield of organolithium reached 30 per cent.

In a typical run 1.35 grams of the white crystalline material

was isolated by adding warm n-pentane to the cold cyclohexene-

pentane filtrate. The filtrate had been concentrated by

removal of almost all of the pentane solvent by evaporation.

At this point very little crystalline material had precipitated

from solution. The solution was filtered anyway to claim the

crystals present. As the warm n-pentane being used to wash

the frit filtered into the solution crystalline material im-

mediately appeared. For this same preparation additional

organolithium compound was obtained by repeated washing of

the lithium metal-lithium chloride mud remaining in the re-

action flask.

Experimental Methods Employed in Characterizing the Organolithium Compounds and

Their Derivatives

Double Titration Analysis

For several years there has been a continuing interest

in the analysis of organolithium solutions. This interest

has been stimulated by the fact that mechanistic and struc-

tural studies on these materials are becoming increasingly

more common in the published literature. Since these compounds

42

are very susceptible to oxygen and moisture it is a necessity

to be able to determine the concentration of carbon bound

lithium in the presence of the other basic impurities. The

most common of these impurities are lithium hydroxide, lithium

alkoxide, and lithium oxide. Brown, Ladd, and Newman (3)

have demonstrated that the impurities derived from oxidation

have considerable solubility in solutions of organolithium

compounds in hydrocarbon solvents.

Any one of several methods (4, 5, 9, 10, 11) can be used

to determine the carbon bound lithium concentration with vary-

ing degrees of accuracy and convenience. The "double titration"

method developed by Grilman and Cartledge (10) was the method

chosen for the analysis in this study. One modification was

made in the procedure by the substitution of dry n-pentane

for ether as the solvent. Comparable accuracy for the samples

diluted with the hydrocarbon solvent is achieved with this

modification and the problem of ether purity reported by

Kamienski and Esmay (15) is eliminated.

The procedure involves withdrawing a 3 milliliter

aliquot of the solution to be analyzed and adding it to 10

milliliters of dry n-pentane. A second 3 milliliter aliquot

is withdrawn and also diluted in 10 milliliters of n-pentane.

To this second solution is added dropwise 1 milliliter of

ethylene dibromide. The solution is swirled gently as the

halide is added and then allowed to stand for two minutes.

Both of these solutions are prepared in the dry box.

43

Upon removal of these samples from the dry box they are

hvdrolvzed with distilled water. The hydrolysis reaction (£t)

vieIds tf

R Li + HOH —» HH •+ LiOH (Ct)

hydrocarbon and lithium hydroxide. At this point all of

the lithium is converted to lithium hydroxide with the

exception of the carbon bound lithium previously reacted

with ethylene dibromide in sample number two. This reaction (b)

KLi 4- BrCH2CH2I3r—* RBr + LiBr + CH2CH2 (b)

converts all of the carbon bound lithium to organic halide

and lithium bromide. The concentration of hydroxide in

each hydrolyzed san^le is then determined by standard acid

and base titration methods using phenolphthalein indicator.

The agitation called for by Gilrnan and Cartledge (10) when

the second sample is near its end point is supplied by a

magnet stirrer and stirring bar. The difference between the

basic normalities of the two samples yields the molarity of

the carbon bound lithium and the normality of the ethylene

dibromide sample is a measure of the normality of oxygen

bound lithium.

Gas Liquid Chromatography

Some of the major by-products generated during the

menthyllithium or 4-tert-butylcyclohexyllithium syntheses

are the olefin derivatives of these compounds. Therefore

the second analytical method developed to determine the

44

parity of solutions derived from these compounds is designed

to determine the olefin concentration. It is also used as a

cliec,-; for the concentration of hydrocarbon bound to lithium

through a carbon lithium bond. This analysis employs gas

liquid chromatography for evaluating the olefin content

in these solutions.

Zubyk and Conner (24) reported the separation of cis-

and trans-p-menthanes, 3-p-menthene, and 1-p-menthene

using a packed Carbowax 4000 chromatograph column. An eight

foot by one-quarter inch aluminum tubing column packed with

10 per cent Carbowax 4000 substrate on 30-60 mesh size

Chromosorb W was prepared for this analysis. Comparable

separations to those reported by Zubyk and Conner (24) were

achieved at a temperature of 90 degrees centigrade and with

a helium carrier gas flow of 120 milliliters per minute.

A Model A-350B Aerograph chromatograph equipped with dual

thermal conductivity detectors was the instrument used for

this analysis.

The retention times of the components of a p-menthane

and p-menthene mixture for this column are listed in Table III.

It is evident from these data that the 2-p-menthene and

3-p-menthene are not separated on this column. Therefore

only the total amount of 2- and 3-p-menthenea can be deter-

mined by this analysis.

45

TABLE III

p-LIENTHANE AND p-MENTHSNB RETENTION TIMES

Compound Retention Time (Minutes)

cis-p-menthane . 3.00 trans-p-menthane . 4.05 trans-8(9)-p-ftienthene 5.25 2-p-menthen e 5.05 3-p-menthen e . . 5.85 1-p-menthene . . 8.70

The p~menthene compounds listed in Table I are charac-

terized by first concentrating each component by fractional

distillation on a Nester/Paust Annular Teflon Spinning Band

distillation column. The fractions, which do not represent

pure p-menthenes due to the inefficiency of the distillation,

are then analyzed on the Carbowax 4000 column, and infrared

and nuclear magnetic resonance spectra recorded for each

fraction. The data recorded from the infrared and proton

magnetic resonance analyses, along with the characteristic

boiling points reported for these compounds, are summarized

in Table 17.

Good agreements between these infrared wave length

values and those reported by Eschinazi and Pines (8) are

observed. Therefore from these data the menthene compositions

OA each distillate faction are established and assignments

of the olefinic proton resonances are made.

46

TABL^ IV

CHARACTERISTIC p-LIENTHENE INFRARED AND PROTON MAGNETIC RESONANCE ABSORPTIONS

Boiling Infrared Proton Compound Point Wave Lengths Resonances Compound

(°C) (0) (Liicrons) (4" Values)

1-p-uenthene 176 8.69 11.0 12.56 4-69 2-p-menthene 169 9.0 12.05 13.79 4.52 3-p-menthene 166 8.81 11.18 12.38 4.69 trans-8(9)-p-menthene 165.5 6.15 11.3 5.38

The p-inenthanes were characterized primarily from the

hydrogenation of a mixture of p-menthenes. The procedure

reported by Sauvage, Baker, and Husaey (20) for the hydro-

genation of cyclohexene over platinum oxide at one atmosphere

of hydrogen in acetic acid was used. A 10 gram sample of

the p-menthene mixture was shaken under these conditions

for approximately thirty hours. The sample was then

extracted in hydrocarbon solvent and analyzed by gas chroma-

tography and infrared spectroscopy. The retention times

reported in Table I for the two p-methanes were observed

and the infrared spectra were compared with those reported

in Sadtler (19) as spectra numbers 1846 and 20221 and by

Eschinazi and Pines (8).

When the hydrocarbon layer of a hydrolyzed sample from

a typica1 menthyllithium solution is analyzed by gas chroma-

tography the concentration of p-menthenes can vary from only-

trace quantities to 10 to 15 per cent. This concentration is

47

dependent upon the age and origin of the sample. The con-

centration of 4-tert—butylcyclohexene is also dependent

upon these two factors. Characterizations of the 4-tert-

butylcyclohexene and the saturated compound were accomplished

by employing the same techniques as those described for the

menthyl compounds.

Polarimetry

The possibility exists that the reaction between

optically active menthyl chlorides and lithium metal can

yield optically active enantiomers of the lithium compounds.

In order to determine whether this phenomenon occurs an

analytical procedure for measuring the rotations of the air

and moisture sensitive solutions was designed. For other

organic compounds standard procedures were employed for ob-

taining rotation data.

Optical rotatory dispersion curves were measured for

solutions of menthyllithium prepared from both menthyl and

neomenthyl chloride. These curves were measured over a

range of 600 to 320 millimicrons using a Gary Model 60

instrument.

Routine optical rotations were taken on a Model 80

Rudolph Visual High Precision Polarimeter equipped with

both sodium and mercury light sources. The instrument can

be read to 0.01 degrees directly. Polarimoter tubes adapted

with tapered ground glass porta and ntoppera and modified

48

with Teflon 'washers sealing the standard 15.5 millimeter

glass cndplates were used. Dry benzene, n-octane, and n-

pentane were the solvents used for these determinations.

All of the samples studied were prepared in the dry box.

The first step in the procedure employed to obtain

these readings is to dissolve the recrystallized rnenthyl-

lithiom in dry hydrocarbon solvent. The solution is then

filtered through a medium fritted funnel in an attempt to

eliminate any insoluble or suspended impurities. Aliquots

are tali.en for titration and gas chromatography analyses to

determine the concentration of carbon bound lithium and

olefin and alkoxide impurities. Another sample of the

solution is loaded into the polarimeter tube and taken out

of the dry box. Ten or more rotation readings are taken

and recorded as rapidly as possible. As soon as this is

completed the tube is rinsed several times and loaded with

solvent; rotation readings are then recorded for this

standard.

These optical rotational data are reported as specific

rotations which can be calculated from the observed rota-

tional readings and the concentration of carbon bound

lithium. The specific rotation can be determined from the

following expression:

t 100 - T T

C<

1'J

t — specific rotation of a substance* at a A specified wave length (A) and temperature (t)

— til® observed rotation measured in decrees at A a specified wave length (A) and temperature (5)

1 = length of liquid path measured in decimeters

c = concentration of solute expressed in grains per 100 milliliters of solution

Infrared Spectroscopy

This analytical technique was used to aid in the charac-

terizations not only of the organolithium compounds, but

also of the organic derivatives derived from these compounds.

The samples were prepared for these investigations using

well-known techniques, such as distillation, isolation from

gas-liquid phase chromatographic columns, and crystallisation.

Exceptions were the air and moisture sensitive compounds which,

were prepared in the dry box. These samples were run either

in tightly sealed cesium bromide liquid cells (0.1 millimeter

path length) or as postassium bromide pellets. These pellets

were prepared in the dry bos using a Mini-Press purchased

from the 7/ilks Scientific Corporation.

Two Perkins-31mer Infrared Spectrometers were used to

record these spectra. The Model 621 instrument was used

primarily with some spectra recorded on the Model 237. Care

was taken to insure that the instruments were calibrated

properly with the recommended standards.

50

Nuclear iJagnetic Resonance Spectroscopy 7*

The principal physical method used in the conformational

analyses of the organolithiuin compounds was nuclear magnetic

resonance spectroscopy . As in most of the reported conforma-

tional applications of nuclear magnetic resonance spectroscopy,

the hydrogen nucleus was utilized. The accessibility of in-

strumentation was the factor which dictated the choice of

a proton magnetic resonance study.

Most of the spectra recorded in this investigation were

obtained from a Yarian Associates A-60 Model High Resolution

Nuclear Magnetic Resonance Spectrometer. This instrument

was equipped with a variable temperature controller. However

the instrument's probe was limited to a temperature range of

-60 to 200 degrees centigrade. Spectra were also recorded on •

an A-60A Spectrometer which was equipped with a variable

temperature probe that could be cooled to -100 degrees centi-

grade. For both spectrometers the sweep was properly cali-

brated in the usual manner using tetramethylsilane and

chloroform absorptions. The variable temperature controller

was calibrated with methanol.

Because of the design of the detection system of the

A-60 and A60A Models a definite phase problem is encountered

at low temperatures. In order to eliminate this problem

spectra recorded over a temperature range from room tempera-

ture to -100 degrees centigrade were obtained from a Model

51

IIR-oO Varian Spectrometer. The Varian HA-100 Model High

Resolution Spectrometer was also employed in this in-

vestigation. This instrument was used primarily in studies

where changes in chemical shifts would aid in the charac-

terization of the compounds under investigation.

The procedure for the preparation of samples for this

analytical method are mostly routine. Only one sample

preparation procedure for the oxygen and moisture sensitive

compounds needs elaboration. This is the procedure employed

when dimethyl ether is used as the solvent.

The requirements of this procedure are to /.eep the

organolithium solution free from oxygen and moisture and at

the same time to Keep the sample below a temperature of

-25 degrees centigrade. This latter requirement is neces-

sitated by the low boiling point of dimethyl ether. For a

typical run these requirements are fulfilled by first sealing

a nuclear magnetic resonance sample tube onto an outer 14/55

standard tapered ground glass joint. A sample of crystalline

organolithium compound is then loaded into the sample tube.

An adapter consisting of a Kontes 4 millimeter hollow plug

vacuum stopcocii adapted with both male and female 14/55

standard tapered ground glass joints is fitted on xhe sample

tube. The stopcoc^ is closed in order to retain the dry box

atmosphere in the sample tube, and the entire assembly is

then removed from the dry box. It is then immediately

52

connected "by the remaining female glass joint to a liigh

vacuum manifold and the sample tube is evacuated.

After the system has been evacuated it is cooled to

the temperature of liquid nitrogen by raising a Dewar flask

filled with the liquid around the sample tube. Dimethyl

ether gas is condensed through a manifold port onto the

crystalline sample. The extremely cold temperature solid-

ifies the ether solvent so that the sample tube can be

evacuated once again. While continuing evacuation the

sample tube is sealed off with a torch directly below the

glass joint which had been adapted to the tube. The sample

is then warmed in a dry ice-acetone bath to -78 degrees

centigrade. ' ,7ith snaking the crystalline solute dissolves

at this temperature in the dimethyl ether solvent. The

temperature-of the solution can then be regulated to any-

desired temperature by adjusting the temperature of the

acetone bath with either dry ice or acetone additions.

Dry toluene and cyclopentane were used as solvents

for low temperature studies requiring hydrocarbon solvents.

n-Pentane can also be used but interference from side bands

in the high magnetic region limits its usefulness. The

samples for these studies are dissolved in hydrocarbon solvent

in the dry box at room temperature and special tight fitting

stoppers purchased from M R Specialties Incorporated are used -

to seal the sample tubes.

53

Experimental Llethods Employed in Derivative Formation of the Organolithium

Compounds

Car conation

Solutions of redissolved crystalline compounds, of the

remaining mother liquors, and original reaction solutions

for both of the organolithium systems studied are carbonated

in two ways. The first method consists of preparing a

mixture of crushed dry ice in n-pentane solvent and then

placing it inside a plastic glove "bag under a nitrogen

purge. The solution to be carbonated is then removed from

the dry box in a tightly stoppered flask and also placed

into the glove bag. The final step is to remove the stopper

and very quickly pour the contents of the flask over the

dry ice-n-pentane slurry. The reaction flask is then removed

from the glove bag, covered with a loosely fitting cover, and

allowed to sit for several hours while the excess carbon

dioxide evolves.

In order to determine whether the stereochemistry of

» this reaction is altered by changes in reaction procedures

a second carbonation procedure was employed for the menthol

system. This method differs from the low temperature pro-

cedure described above first, by the fact that it is conducted

at room temperature, and second, in the use of carbon dioxide

gas as the reactant. In a typical run the solution of organo-

lithium is loaded into a round bottom flask in the dry box.

54

" It is then capped, brought oat of the box, and secured to

a lattice rac;:. A gas dispersion tube purging with carbon

dioxide gas is then quickly inserted into the reaction flask

and the flow of gas is continued until the solution gives

a negative Gilman test.

Solvent effects on the reaction pathway were studied

as well. The method of carbonation using a dry ice-n-pentane

slurry is employed for these investigations. Bolh diethyl

and dimethyl ethers are used as solvents. The procedures

for preparing solutions in these two solvents differs from

that used for hydrocarbon solvents.

The preparation of a solution using dimethyl ether as

the solvent haa been essentially outlined in the procedure

for preparing nuclear magnetic resonances samples in this

solvent. However, due to a number of appartus changes made

to facilitate carrying out this process in larger quantities,

the following description is included.

In the dry box a sample of crystalline organolithium

compound is loaded into a 100 milliliter single-necic round

bottom flask equipped with a 24/40 standard tapered outer

joint. The flask is then sealed with an adapter that consists

of a Lontes 4- millimeter hollow plug vacuum stopcock sealed

to inner 24/40 and outer 14/35 standard tapered joints. Keep-

ing the stopcock closed, the flask and adapter are removed from

the dry box and connected to a high vacuum-manifold. The at-

mosphere above the lithium compound is then evacuated.

55

After the system has been evacuated it is cooled to the

temperature of liquid nitrogen by raising a Dewar flask filled

with the liquid around the f]ask. Dimethyl ether gas is con-

densed through a manifold port onto the crystalline sample.

The extremely cold temperature solidifies the ether solvent

so that the flask can be evacuated once again. At this

point the manifold and flask are pressurized to atmospheric

pressure with argon. Under this argon atmosphere the system

is warmed to the desired temperature. The flask is isolated

from the room atmosphere by dosing the adapter stopcock and

taken into the glove bag. The carbonation procedure is the

same as for other samples. Except for a few alterations this

same procedure is used in carrying out the carbonation reaction

on diethyl ether solutions. The only major change is neces-

sitated by the physical property differences of the two

solvents. A bulb to bulb distillation is required to transfer

the diethyl ether solvent into the reaction flask. In ad-

dition, when diethyl ether is the solvent, only a temperature

of below 0 degrees centigrade is required.

Characterization of the products produced from these

reactions is pursued by two different methods for the menthyl

system. The first method involves isolating the carboxylic

acids formed; the procedure used is the method outlined by

Smith and Wright (21).

The crude acid is obtained in yields as high as 90 per

cent. It .possesses a melting point between 62 and 65 degrees

56

centigrade and a specific rotation of -56.1°(CH3CH20H,

c. 5.2). Sruitn and Wright (21) reported the welting point

and specific rotation values for the 3-p-menthane car "boxy lie

acid isomer to "be equal to 60 to 63 degrees centigrade and

-4-2.6°(CH3CH20H, C. 816). They also reported the 3-p-

heomenthane acid isomer to have a higher melting ^oint and a

positive specific rotation value. A lacx of evidence for the

formation of the highest melting acid placed serious doubt

on the degree of sensitivity offered by this analysis method.

This is the reason the second and most effective method was

incorporated.

The procedure for this method employs the conversion

of the acid diastereoners into their respective esters. This

is accomplished by first carbonating a pentane solution of a

lithium compound over dry ice and then extracting the residual

hydrocarbon solution with dilute hydrochloric acid. The

hydrocarbon layer is then separated from the aqueous layer

and dried over anhydrous magnesium sulfate. This drying agent

is later removed by filtration and the carboxylic acid-n-

pentane solution is added dropwise to a cold ether solution

of diaamethane. This reagent is synthesized using either

the method reported by Moore and Reed (18) or the method of

Arndt (2). The former was found to be the favored method.

After the dropwise addition of the organic acid solution has

been completed the resulting solution is allowed to sit from

57

six to ei^ai hoars. Au thir> ti„.e ilic remaining excess

diazoi.i2 oiir,us is 'Invon oA.' by x'Si ] uxmg line solution v/ioxi

a steam bath. "Disappearance of the yellow coloration in-

dicates unao this atop Las been comp]eted and the reaction

pro duct 3 are concentrator; by 3olvont evaporation an .1 analyzed.

T-lie menthyl est or epimer mixture is easily isolated from

other menthyl derivatives by fractional distillation. This

is accomplished at 93. p to 94 degrees centigrade and J tori-

pressure. A typica"1 product isolated in such a fraction

possesses a 'specific rotation of -39• 7°( 11-C5H12,c. 5.0) and

characteristic ester infrared absorptions. However the

identity of the ester is confirmed by a .100 megacycle proton

magnetic resonance spectrum which shows an axial <-iuetliinyl

proton resonance at a lvalue of 2.33. This resonance is

obscured by other absorption in the 60 megacycle spectrum.

A detailed discussion on conformational analysis of cyclo-

ne xyl systems by nuclear magnetic resonance spectroscopy will

be included in the following chapter.

A gas liquid chromatographic analysis of the esters is

made at 145 degrees centigrade with a 240 milliliter per

minute helium flow on a ten foot by one-quarter inch aluminum

tubing column packed with a 15 per cent polypropylene glycol

substrate on 30-60 mesh Oiiromosorb V7 pacIcing material. The

use of the thermal conductivity detector instrument allows

the characterization of the two ester isomers by isolation

53

of the column effluent into micr o-infrared cells. This

technique was especially useful in characterizing the low

concentration isomer whose concentration ranged from only

5 per cent to 30 per cent, depending upon the solvent

employed in preparing the organolithiuij solution. Again,

both compounds demonstrated characteristic infrared ab-

sorptions of organic esters.

A second chromatographic analysis which gives better

separation of the ester compounds and the menthyl alcohol

impurities is carried oat at 105 degrees centigrade with

2 milliliters per minute carrier gas flow rate on a 150

foot by 0.01 inch Carbowax K-1540 G-olay column. A Perkin-

Slmer Model F-ll gas chromatograph equipped with a hydrogen

flame detector is used in this analysis. Retention times

of 11 and 13.5 minutes on the polypropylene glycol column

and 16 and 18.7 minutes on the Carbowax column are recorded

for the two isomers.

Gtolow and Boyce (22) have reported the physical

properties and the nuclear magnetic resonance spectra for

the cis and trans isomers of methyl 4-tert~butylcyclohexane-

carboxylate. Using these data, along with infrared and

carbon and hydrogen analyses, the compounds separated on

the Carbowax K-1540 and polypropylene glycol chromatography

columns were identified. Retention times recorded on these

columns using the conditions described above were 23.0 and

59

26.0 minutes on the Carbowax Ii-1540 column and 24.8 and

28.5 minutes on the polypropylene glycol column for the

cis and trans isomers.

Deuteration

The reactions between deuterium oxide and the menthyl

ana tcrt—buti;-lcyclohexyl lithium compounds undor investi-

gation were studied. The use of both n-pentane and dimethyl

ether as solvents for these reactions requires, as was

described for carbonation, two different procedures for

carrying out tnese investigations. The sample preparations

are conducted in the same manner as those described for

carbonation in the two different solvents. After these

steps are completed the deuterium oxide is introduced onto

the samples through a rubber septum cap with a syringe.

Therefore the reaction flask used for the dimethyl ether

solution must have a second opening which is sealed off with

the septum cap. In addition, magnetic stirring bars are

placed into the- flasks prior to sample introductions for both

systems. These are used for stirring the solution as the

deuterium oxide is being added.

The hydrocarbon layer containing the product is isolated

from the excess deuterium oxide with a separatory funnel and

dried over anhydrous magnesium sulfate. The drying agent

is later removed by filtration and the sample is concentrated

by evaporating the solvent with a stream of dry air. This

60

sample is then analyzed using the infrared spectrometer,

gas chromatography, and optical rotation instrumentation

previously described. The infrared stretching frequencies

of the caroon—deuterium bond are the most important data

recorded from these analyses. These values occur between

2200 and 2100 wave numbers for the deutero oyclohexane

derivatives studied. Carbon tetrachloride is the solvent

used in recording these spectra.

Bromination

Bromination procedures for both molecular bromine and

ethylene dibromide reagents are employed using both n-pentane

and dimethyl ether solvents and temperatures ranging from

room temperature to -70 degrees centigrade.

The preparation of the organolithium solutions in the

two different solvents is accomplished with the same methods

described for the carbonation reactions. One-hundred-milliliter

two-neck round bottom flasks equipped with 14/35 tapered ground

glass joints were used. One neck of the flask is modified

with a 4- millimeter stopcock inserted between the flask and

the joint. This entry port is used primarily to keep an

atmosphere of argon above the sample throughout the reaction.

Both of the brominating agents are introduced into the

reaction flask by the dropwise addition of n-pentane solutions

of these two reagents. The addition funnels employed for

this process are first loaded with the n-pentane-bromination

61

reagent solutions and then attached to the reaction vessel

through the second nock of tne flask. This is accomplished

while the entire s;/stem is being purged with argon gas.

The syste:;i is then equilibrated with an argon atmosphere

and the reaction is completed by the addition of the brominat-

ing reagent over a period of thirty minutes. The solution

in the reaction flask is stirred with a magnetic stirring

bar and the desired temperature is achieved by immersing the

flask in either a dry ice-acetone or ice water bath.

The reaction products are isolated by evaporating the

n-pentane solvent and, if large enough quantities of re-

actants are initially used, the products can be distilled.

The menthyl products are isolated at 32 and 83 degrees

centigrade and 7 torr; the 4-tert-butylcyclohexyl bromide

isomers are isolated at 99 to 102 degrees centigrade and

13 torr. The isolated products are then characterized using

infrared, nuclear magnetic resonance, and gas chromatographic

techniques. Infrared characterization data for substituted

cyclohexyl bromide systems has been reported by Eliel and

Haber (7). These data along with an analysis of the charac-

teristic nuclear magnetic resonance absorptions observed for

the different ©<-methinyl proton conformations support the

quantitative chromatographic determinations for the dis-

trioution of products. A summary of the findings from nuclear

magnetic resonance and infrared studies is included in

Table V.

62

TABLE V

DATA SIOIARY FOR CYCL0H3XYL BR Oil IDE DERIVATIVES

C ompound Characteristic

Infrared Wave dumber (cm"-'-)*

Characteristic^ Proton Magnetic Resonance Value

tie n thy 1 Bromide 686 6.30

Neomenthyl Bromide 665 5.62

trans-4-tert-Butyl-cyclohexyl Bromide 690 6.07

cis-4-tert-Butyl-cyclohexyl Bromide 663 5.38

*31iel and Haber (7) iiave reported characteristic infrared ranges of 690-710 cm~l and 670-685 cm~l for equa-torial and axial cyclohexyl bromide derivatives.

Separation of the bromide isomers of the menthyl and

4-tert-outylcyclohexyl systems is achieved on a 150 foot by

0.01 inch K-1540 Carbowax Golay chromatographic column. The

conditions employed for these analyses are a temperature of

88 degrees centigrade with a carrier gas flow rate of 2

milliliters per minute. Retention times of 16.2 and 17

minutes for the neomenthyl and menthyl bromide isomers, and

18.6 and 19-0 minutes for the cis- and trans-4—tert-butyl-

cyclohexyl bromide isomers are observed.

Ox idation

The procedure used to carry out the reaction between

molecular oxygen and menthyllithiuiu solutions is identical

to the one used for the room temperature carbonation of

63

mentliyllithiuw-n-po;itane solutions. Samples from tho orig-

inal reaction solution before crystallization and samples

from solutions of redissolved crystalline material in n-pentane

solvent are studied. Dry compressed air is bubbled through

these solutions at room temperature for one hour • Dui in^

-this time the solution is stirrod with a magnetic s lining

bar.

Upon completion of the oxidation the contents of the

flask are hydrolyzed and acidified with dilute hydrochloric

acid. The aqueous layer is then removed and the remaining

hydrocarbon solution is dried over magnesium sulfate. This

drying agent is easily removed by filtration and the filtrate

concentrated by solvent evaporation. The reaction products

are analyzed by gas liquid chromatography and nuclear mag-

netic resonance methods. An 8 foot by one-quarter inch

aluminum tubing column packed with 10 per cent Carbowax 4000

substrate on 30-60 mesh size Chromosorb V/ at 135 degrees

centigrade with a helium carrier gas flow of 105 milliliters

per minute is used for the chromatographic analysis. Re-

tention times of 16.7 and 21 minutes for the neomenthol and

menthol isomers are recorded for this analysis. Character-

istic ®<-methinyl proton resonances for the neomenthyl and

menthyl isomers occur at 5.97 and o.SQ't respectively.

u -

Iorti-LithiLU.I 'Bxchange

This reaction is carried out by adding a "benzene solution

of methyl iodide to a solution of mentliyllithium in n-pentane

solvent. The methyl iodide solution was added by pipette

and allowed to react for one minute. The product is then

quiclcly filtered through a fritted funnel. This entire

procedure is carried out in the dry box.

Upon removal of the filtrate from the dry box it is

hydro!yzed and the hydrocarbon layer is dried over magnesium

sulfate. The menthyl iodide solution is isolated from the

drying agent by filtration and then isolated from other

menthyl derivatives by fractional distillation. The menthyl

iodide fraction is collected at 115-116 degrees centigrade

and 10 torr pressure. This fraction is analyzed by proton

magnetic resonance and polarimetry methods. The character-

istic «=< -methinyl proton resonance patterns are observed from

the neomenthyl and menthyl iodide isomers at 5.24 and 5.89^.

The specific rotation value determined for this fraction

equals -31.5°(CD3C1, c. 13.0).

Decomposition

The decomposition is carried out on samples of crystal-

line methyllithium dissolved in dry n-decane. Both sealed

nuclear magnetic resonance sample tubes and. the same two-neck

flasks employed for bromination reactions are used as re-

action vessels. The organolithium solutions are prepared

65

and loaded into these reaction vessels in the dry box. At

the same ti^e samples for carbon "bound lithium concentration

and gas chromatographic analyses are taken.

The nuclear magnetic resonance sample tubes used for

reaction vessels are adapted with taped glass joints and

stopcock" adapter as described for the sample tubes used in

nuclear magnetic resonance studies of the organolithium

compounds in dimethyl ether solvent. As in these studies,

when the samples are removed from the dry box they are

frozen, evacuated, and then sealed with a glass torch. The

samples in the larger volume reaction flask are immediately

purged and then equilibrated, under an argon atmosphere when

brought out of the dry box.

The variable temperature attachment on the nuclear

magnetic resonance spectrometer is employed in carrying out

the decompositions of the samples sealed in the nuclear

magnetic resonance sample tubes. Spectra of the olefinic

proton absorptions are recorded repeatedly at the desired

elevated temperatures until the reaction is complete. For

the larger volume samples the elevated temperatures are

furnished by heated oil baths. These samples are stirred

with magnetic stirring bars.

The reactions are terminated by the addition of water

and then the hydrocarbon layers are separted and dried in

the same manner as the deuterated compounds described earlier.

66

The anal.-sis of Vut> p-.:ietheties and p-nienthane ooppositions

fron a topical reaction is ;.,ade using gas liquid ciiroi,.a-

tography. Tito inability of the packed Carbowax 4000 column

described earlier to separate the 2- and 3-p-i.ientiienes re-

quired the use of another column. This separation is achieved

on a 200 foot "by 0.01 inch 95 per cent Squalane and 5 per cent

Atpet Golay column at 83 degrees centigrade and with a carrier

gas flow of 2 milliliters per minute. The same infrared and

nuclear Magnetic resonance techniques described earlier are

again applied for identification of the elution order of

these compounds. Retention times of 13.5 and 18.8 minutes

are recorded for the p-3-menthene and p-2-nienthene.

CHAPTER BIBLIOGRAPHY

1. Alexandrou, h. E. , "Reaction of n-But,, llitnium with c is — and tr a ns -4~tsrt-Bat y lc,/clohex;-1 Bromide and Iviercoric Bromide," Journal of Or^anowctallic Chemistry, V (1966), 301.

2. Arndt, P., "DiazomethaneOrganic Syntheses, XV (1935), 3-

3. Brown, Theodore L., J. A. Ladd, and G. K. Newman, "Interaction of Alkyllitnium Compounds with Base. Complex Formation Between Ethyllithiuin and Lithium Ethoxide in Hydrocarbon Solvents," Journal of Organo-r.ietallic Chemistry, III (1965), 1.

4- Clifford, A. L. and R. R. Olsen, "The Analytical Chemistry of Or ganome tallies. Quantitative Determination of Organoalkalies," Analytica 1 Chernistry,, XXXII (19-60), 544.

5. Collins, Peter P., Conrad Xamiens.^i, Donald L. Esmay, and R. B. Ellestad, "Oxidimetric Determination of n-Butyllithi..,:. in Hydrocarbons Using Vanadium .Tcr.toxide," Analytical Chemistry, XXXIl'I (1961), 463.

6. . Cur tin, D. Y. and J. hoenl, "Effect of Solvent on the Steric Stability of Lithium Reagents," Journal of the American Chemical Society, LXXXIV (1962), I967.

7. Sliel, Ernest L. and Ralph G. Haber, "Conformational Analysis. V. The Reaction of cis- and trans-4-t-Butylcyclohexanol and trans-4-Methylcyclohexanol with Phosphorus Pentabromide. Syntheses of Alkylcyclohexyl Bromides," Journal of Organic Chemistry, XXIV (1959) 143.

8. Sschinazi, H. E. and Herman Pines, "Study in the Terpene Series. XXVI. Disproportionation of d-Limonene in the Presence of Palladium Hydroxide-Barium Sulfate Catalyst," Journal of the American Chemical Society, LXXVII (1956)","1176:

9.. Everson, W. L., "Determination of Butyllithium in Hydrocarbons by Thermometric Titration with Butanol," Analytical Chemistry, XXXVI (1964), 054.

67

68

10. Oilman, Henry and j'rank K. Cart ledge, "The Analysis of Organolithium Compounds," Journal of Organometallic Cher, is try, II (1964), 447. '

11. Gilman, Henry and A. U. Haubein, "The Quantitative Analysis of Alkyllithium Compounds," Journal of the American Chemical Society, LXVI (1944), 1515>.

12. Gilman, Henry, 7,rright Langham, and Ared W. Moore, "Some Interconversion Reactions of Organolithium Compounds." Journal of the American Chemical Society, LXII (1940), T5TT*

13* Greene, Frederick 3 . , Chin-Chium Chu, and Jasjit V/alia, "Decomposition of Alkyl Hypochlorites. The 4-t-Butylcyclohexyl Radical," Journal of Organic Chemistry, XXIX (1964), 1285.

14. Horner, von Leopold, Hermann Oediger, ana Hellmut Hoffman, "Reaktionen Mit Triphenyl-Phosphin-Dihalo-geniden," Justus Liebigs Annalen der Chemie, DCXXVI (1959), 26"

15. Kamienski, Conrad and Donald 1. Esmay, "Analysis and Stability of n-3utyllithium Solutions in n-Heptane," Journal of Organic Chemistry, XXV (i960), 115.

16. _ , "Effect of Sodium in the Preparation of Organolithium Compounds," Journal of Organic Chemistry, XXV (i960), 1807.

17. Larnaudie, M. Marcel, "Isome'rie des derives mono-substitues der cyclohexane," Comptes Rendus Hebdomadaires Des Se'ances De PAcad&nie Des Sciences. 0 o m V (1052) ,"1541

18. Moore, James A. and Donald "3. Reed, "Diazomethane," Organic Syntheses, XV (1935), 3.

15* Sadtler Standard Infrared Spectra, Philadelphia. The Sadtler Research Laboratories, Spectra Numbers 1846 and 20221.

20. Sauvage, James Frederick, Robert H. Baker, and Allen S. Hussey, "The Hydrogenation of Cyclohexene over Platinum Oxide," Journal of the American Chemical Society, LXXXII (1$66)',""6( 01

69

21. Smith, J. G. and George ?. //right, "The Diastereomeric Menthyl Chlorides Obtained j?roia (-)Menthol," Journal of Organic Chemistry, XVII (1952), 1116.

22. Stolow, Robert D. and Charles B. Boyce, "Preparation of cis- and tran3-4-t-Butyl-<<,«<-diirtethylcyclohexane-methanol," Journal of Organic Chemistry, XXVI (1961), 7426.

23. Walborsky, II. II. and M. S. Aronoff, "Cycl opropanes XIX. Reaction of Lithium Metal with Optically Active Halides," Journal of Organometallic Chemistry, IV (1965), 415T

24. Zubyk, W. J. and A. Z. Conner, "Analysis of Terpene Hydrocarbons and Related Compounds by Gas Chroma-tography," Analytical Chemistry, XXXII (I960), 912.

CHAPTER III

RESULTS AND DISCUSSION

Derivatives of cyclohexyllithium have been proposed as

the compounds to incorporate in this investigation of the

conformational stability and the stereochemistry of a number

of characteristic reactions of alkyllithium compounds. This

selection evolved from the requirement that the organic

moieties of these lithium compounds possess different degrees

of conformational stability. The barrier to inversion for

the substituted cyclohexyl products has been known for some

time to be dependent upon the nature of the organic sub-

stituent. Therefore a model system for this investigation

should be provided by a series of these compounds.

The first objective of this investigation is to determine

whether a configurationally stable alkyllithium compound

can be synthesized and isolated. From the viewpoint of the

lithium compound there appear to be three main factors which

can enhance this stability. The first factor is the experi-

mental conditions employed in synthesizing and isolating the

compound. It is obvious from the review of earlier investi-

gations of this type that the selections of both the tempera-

ture and the solvent are critical.

70

71

The second factor, which Curtin and Koehl (9) predicted

to be an integral ingredient of the inversion mechanism of

these systems, is the degree of association demonstrated by

these compounds. The extent of association will be primarily

predetermined by the choice of the solvent,. Even so this

phenomenon might still influence the populations of the

different conformers available to the organic constituents,

especially for systems as large as cyclohexyl derivatives.

The last factor concerns the effect of the conforma-

tional stability of the organic moiety on the overall

stability of the alkyllithium compound. This effect would

definitely be a primary contributor to the overall stability

of non-bridged cyclic systems. Therefore with the initial

objective of synthesizing a configurationally stable ,

cyclohexyllithium compound this effect is of primary im-

portance in- selecting the compound to study.

Conformational analysis has been applied most intensively

to cyclohexane and its substitution products. Because of

steric interactions and an increase in the number of gauche

conformers the axial position of a substituent is less

favored than an equatorial position. For the mono-substituted

products these effects become pronounced when an isopropyl

substituent is considered. If a still bulkier substituent

is examined, such as the tert-butyl group, then an axial

7 0

orientation is no longer possible. For di- and. tri-

substituted products these effects are magnified and

preference for the equatorial substituent may be observed.

The first substituted cyclohexyl product to be in-

vestigated as a lithium derivative is the menthyl system,

diagrammed in Figure 1 (a, b). This sytem is capable of

exhibiting not only conformational stability but also

optical activity. Incorporation of the optical property

into the system provides a convenient method for determining

whether one conformation of the compound is preferred over

another. It will also serve, along with proton magnetic

resonance and chemical derivative data, as evidence for

conformational stability.

Some concern might originate because this highly sub-

stituted molecule contains asymmetric centers other than the

carbon atom to which the lithium atom is bound. It is

reasonable to question whether neighboring group and asym-

metric induction effects potentially complicate the stereo-

chemistry studies being undertaken. Gram (8) pointed out in f

his review on the stereochemistry of organomercury compounds

that these same questions arose. In these cases simpler

systems were investigated and the same conclusion regarding

the stereochemistry for both groups was reached. For the

lithium system the same logic will be applied and in this

investigation both menthyl and 4-tert-butyl.cyclohexyl systems

will be studied.

73

Selection of the menthyl derivative also involved the

availability of the starting organic chloride. Smith and

Wright (25) and Horner and co-workers (12) have reported

stereospecific syntheses for menthyl chloride (a^) and

neomenthyl chloride (th) epimers from (-)-menthol (a2).

This alcohol is commercially available in a very pure form.

In contrast only an 80:20 isomer mixture of the trans- (c2)

1 2 3 4 5 6 CI OH D Br I C02H

(c) (d)

Fig. 1—Cyclohexyl derivatives studied

and cis-4-tert-butylcyclohexanols (d2) is available. Further-

more syntheses of the corresponding chlorides, (cj) and (di),

have not been studied so extensively and are therefore less

refined than those listed for the menthyl system.

Carbon bound lithium yields ranging from 50 to 60 per

cent were obtained from the direct reaction of lithium metal

with either menthyl- or neomenthyl chloride. This productivity

74

decreased to only 30 per cent for the cis-4-tert-butyl-

cyclohexyl chloride reaction. From these reaction solutions

crystalline, pyrophoric organolithium compounds were isolated.

The 4-tert-butylcyclohexyllithium product was observed to have

a relatively lower solubility in hydrocarbon solvent than the

inenthyllithium product. This property has a definite effect

on the amount of material which can be crystallized from a

solution. Only about half of the soluble menthyl derivative

can be isolated while almost all of the 4-tert-butylcyclohexyl

compound is obtainable.

Although the syntheses of these two cyclic organolithium

compounds have not been reported, this accomplishment comprises

only half of the first objective of this investigation. Com-

pletion of conformational analysis investigations of these

systems constitutes the second and most demanding half of

this study. This obviously requires installation of other

analysis methods. One of the oldest and most accessible

techniques employed in investigations of this type is a

measure of the rotation of polarized light. Since the menthyl-

lithium products could demonstrate optical activity, polarimetry

studies were pursued first.

Hydrocarbon solutions of the redissolved crystalline

lithium products derived from both the menthyl and neomenthyl

chloride isomers were observed to be optically active. Further-

more, both solutions were observed to be highly dextrorotatory

75

and the near identities of their specific rotations suggest

that the composition of the two solutions is identical. Only

small differences were recorded between the specific rotations

of an original reaction solution and a solution comprised of

dissolved crystalline material. Due to the high dextro-

rotatory specific rotational values reported "by McNiven and

Reed (19) for the p-2-menthene and p-3-menthene compounds

no measurements were recorded for the mother liquor solutions.

Table VI contains characteristic data recorded in these studies.

TABLE VI

CHARACTERISTIC ROTATIONAL VALUES FOR IviENTHYLLITHItM SOLUTIONS

Entry Chloride Isomer

Specific Rotation Menthyllithium Solution Specific Rotation

1 + 50.8° (n-C8H18,c. 5.24) +53.4° (C6H6, c. 3.40)^

2 -39.9 (C6H6,C. 5.00) +42.1 (C5H10, C. 6.28)-1

3 -40.0 (n-CgHgOgjC. 5.00) (**) (n-C5H12, c. 5.41)

-40.0 (n-C0pI]i g,c. 5.00) +56.0 (n-C H-j j* c. 3.65)^

^Solution of dissolved crystalline material.

**The rotational character for this original reaction solution was found to be dextrorotatory. Due to the presence of highly dextrorotatory p-menthenes in the solution a specific rotation value would be meaningless.

The accuracy of these values is only as good as the sampling

technique employed for obtaining the rotational values and

concentration values for the inenthyllithium. solutions.

76

Special attention was given to standardizing these techniques

from one ana3.ysis to another in order to minimize experi-

mental errors. This effort was then extended "by attempting

to determine whether the specific rotation values recorded

were subject to variation with time. Therefore a time

versus specific rotation study was conducted on a raenthyl-

lithiuu-cyclopentane solution. No change in rotation was

observed over a period of one hour.

Optical rotatory dispersion analyses were employed to

investigate the observation that the specific rotations were

effectively equivalent, irrespectibe of the chloride isomer

reagent. It was possible that one of the solutions could

possess a plane curve which starts on the positive side of

the zero rotation axis in the visible and crosses over this

axis in the ultraviolet region. However both solutions gave

almost identical dextrorotatory plane dispersion curves

from 600 to 300 millimicrons.

V/ith respect to the conformational analysis determination

the conclusions derived from these rotational studies were at

their best disappointing. The data indicates that the com-

position of the two solutions is probably identical. 7hether

this composition is comprised of a rapid or slowly equilibrating

equilibrium mixture or of a high concentration of one of the

menthyl- or neomenthyllithium isomers is not known. Further-

more, no predictions with respect to any of these possibilities

can be made from the size of the observed rotation values.

77

Head (21) has reported the rotational contributions of

tlie menthyl and isopropyl substituents for the neomenthyl

alcohol and amine isomers. Even with the assumption that

these contributions remain unchanged for the lithium compounds,

a lack of evidence of rotational contributions of lithium

substituents renders these values ineffective, furthermore

an investigation of the changes in rotational values with

different substituents for almost any system simply clouds

the situation even more. An example may be found in a

review of rotational values reported by V/alborsky and co-

workers (26) for the following conformational^ equivalent

compounds.

V

- CH~

^ / / W / /

X H Br I C02II

*1 -128° +116° +171° -36°

Infrared spectra of crystalline lithium compounds

suspended in a potassium bromide disc, a solution of dis-

solved crystalline material, and an original reaction

solution sample were recorded. In these spectra the charac-

teristic cyclohexyl ring and carbon-lithium absorptions were

observed in their respective frequency regions. Total analyses

of these spectra to establish complete structures were not

attempted. Furthermore, due to the broad shapes of the carbon-

lithium absorption, conformational assignments were not attempted.

78

These spectra were recorded primarily for purposes of com-

parison. No major changes were observed between any of

these three scans. This illustrates that the composition of

the crystalline material constitutes the primary composition

for both the original reaction solution and the solution

prepared from the crystalline material. It is also evident

that these data complement the findings reported in the

optical activity studies. However, the infrared data re-

quires simultaneous crystallization of the two epimeric

lithium compounds if a rapidly attainable equilibrium

exists in solution.

It is obvious at this point that another method of

investigation is needed. From a review of the published

literature a great deal of conformational analysis data for

cyclohexyl systems has been reported from nuclear magnetic

resonance investigations. Furthermore this tool has proven

to be very effective in a number of organometallic stereo-

chemistry studies. For these reasons incorporation of

proton magnetic resonance spectroscopy as a conformational

analysis instrument was one of the primary objectives of

this investigation.

Experimental observations by Lemieux and co-workers (17,

18) and Brownotein and Miller (5) demonstrated the usefulness

of proton magnetic resonance spectroscopy in studying con-

formations of substituted cyclohexyl ring compounds. Lemieux

79

and co-workers (17, IS) demonstrated the existence of a

chemical shift difference between the signals for protons

in axial and equatorial conformations. They observed that

the signal for protons in an axial orientation occurs at

higher fields than those for equatorial hydrogens.

Employment of this property as a conformational analysis

tool is often ineffective due to the existence of small

chemical shift differences and the complexity of the spin

coupling. For similar compounds Brownstein and Miller (5)

noted that conformational determinations could "be made "by

recording the widths of the two proton absorptions. They

reported that the width of the resonance peak of the ring

protons was always considerably greater in the disubstituted

isomers which had diequatorial conformations than in isomers

with axial-equatorial conformations. They further demon-

strated that spin-spin coupling is greater between two protons

that are axial to one another than between axial-equatorial

protons, which in turn is greater than for two equatorial

protons. Since the completion of these early investigations

by Lemieux and co-workers (17, 18) and Brownstein and filler

(5) a number of investigations of these types have been con-

ducted on a variety of similar compounds. Dryer (10) has

reported approximate coupling values for such systems; these

values are listed in Table VII.

80

TABLE VII

SPIN-SPIN COUPLING CONSTANTS

Type (XY) Coupling Constants (eps)

a,a 5-10 /—V a»e 2-4 < > e,e .2-4

K-C— C-H I I x y

These observations are applicable in characterizing

the menthyl conformers. Illustrations of the observed

o<-nenthinyl proton signals for a few menthyl epimers are

shown in Figure 2. Complete spectra of the menthyl and neo-

menthyl chlorides are shown in Figure 3. In agreement with

the previously described observations it can be seen in

Figure 2 that the broadest absorption occurs at a higher

field strength. This absorption therefore corresponds to

the o<_menthinyl proton in the axial position. Likewise the

absorption occurring at the lower field position is a much

sharper signal corresponding to the equatorial proton. The

axial and equatorial absorptions for various menthyl isomers

generally demonstrate signal widths of 30 and 10 cycles per

second respectively.

These width differences are readily explained in a

breakdown of the coupling constants generating each resonance

pattern. The high field absorption is derived from two axial-

axial and one axial-equatorial splitting. However the low

d.

c.

b.

Amu

81

5.0 6.0 7.0 C

Pig. 2 — o(-.Methinyl proton resonance spectra of menthyl and neomenthyl chloride isomers (a), bromide isomers (t>), iodide isomers (c), and alcohol isomers (d).

82

3--Proton magnetic resonance spectra of menthyl chloride (a) and neomenthyl chloride ("b).

83

field signal evolves from two axial-equatorial and one

equatorial-equatorial couplings. Using coupling values from

Table YII and the qualitative diagrams shown in Figure 4,

the two resonance patterns can be duplicated..

O A

axial D.e

0 to 20 L

30 0

quatorial e,e

10 20 JO

,?ig. 4—Spin-spin coupling diagram of resonance patterns.

Both of these characteristic patterns have been computer-

synthesized by Glaze (27). The IA0C00U II program reported

by Castellano and Botimer—By (6) was employed in these

calculations. Employing this procedure, while using vicinal

axial-axial and axial-equatorial coupling constants of 13 and

4 cycles per second and a geminal coupling of -15 cycles per

84

second, a high field pattern that could be almost super-

imposed upon the proton magnetic resonance spectrum was

generated.

A typical 60 megacycle proton magnetic resonance

spectrum of a solution of the crystalline lithium compound

dissolved in hydrocarbon solvent is illustrated in Figure 5a.

The K-methinyl proton signal appears as a multiplet signal

30 cycles per second wide at 10.5^• This high field ab-

sorption is characteristic of protons attached to a caroon

atom bound to lithium. Benzene, toluene, and cyclopentane

solvents were used in recording spectra of this type.

The broad triplet pattern of the o<-methinyl proton

closely resembles the high field patterns recorded for other

menthyl conformer compounds. However the absence of a second .

o<-methinyl resonance complicates such an interpretation. An

observation reported by Krieghoff and Cowan (16) is definitely

related to any speculations concerning where a second ab-

sorption might occur. They found the differences in chemical

shifts between the two norbornyl Grignard c<-proton absorptions

to be consistent with the corresponding differences for a large

number of exo-endo pairs of norbornyl compounds. Prom Figure 2

the chemical shift differences between menthyl epimers appear

to be approximately 45 to 47 cycles per second. Bo evidence

for a second absorption either up or down field from the

10.5 T absorption was observed.

85

MAW/****"1"'**

C.

7 8 10 11 Pig. 5—Proton magnetic resonance spectra of crystalline

menthyllithium dissolved in toluene (a) and dimethyl ether (c). Also enlargements of oi-methinyl proton spectra in toluene (a')» diethyl ether (b), and dimethyl ether (c).

oo

not

The trend reported by Krieghoff and Cowan (16) need

necessarily exist for the menthyl system. The possi-

bility exists that the broad nultiplet absorption might be

comprised of two absorptions overlapping one another. A

100 megacycle spectrum was obtained in benzene solvent to

investigate this possibility. The c<-proton resonance at

this frequency could be superimposed upon that obtained at

60 megacycles. These two spectra at different frequencies

were recorded over a time duration of several hours. During

this time period no visible change of the oO-methinyl resonance

was observed. This indicates that if an equilibrium mixture

of the two isomers exists, it does so in a rapidly equili-

brating manner. Otherwise some changes in the resonance

pattern may be expected as the slowly equilibrating equili-

brium mixture is attained.

An investigation was initiated at this point to determine

whether the observed resonance pattern was effected by changes

in the solution temperature. Such an investigation should be

very informative if the observed resonance is derived from a

coalescence of the two epimer proton resonances. An increase

in the temperature would definitely facilitate any slow in-

version process. Likewise a decrease in the rate of a rapid

inversion process should be obtained by a lowering of the

temperature. Either effect would be easily noted by a definite

broadening of the resonance pattern. Spectra were recorded

87

from 50 decrees centigrade to -35 decrees centigrade with

no observable changes in the resonance width. These spectra

were recorded over a time period of several hours. The fact

that the -methinyl resonance does not alter in width from

50 to -35 degrees centigrade over a period of several hours

is conclusive evidence that a slow interconversion between

the menthyl- and neomenthyllithium compounds does not occur.

However, this observation does not preclude a very rapid

interconversion with a barrier energy of less than 12

kilocalories per mole.

This energy barrier was determined by assuming that the

resonance observed is that of the menthyl epimer and that the

neomenthyl isomer resonance occurs 45 to 47 cycles per second

down field. This frequency difference is approximately the

chemical shift difference observed for a number of equatorial-

axial pairs of menthyl compounds. Recalling that the difference

in chemical shift is directly related to the rate constant for

a first order process allows the activation energy to be

computed using the Arrhenius equation. Table VIII lists the

energy values computed for the expected range of unimolecular

frequency factors.

An assessment of this energy barrier must be made before

determining whether the observed resonance is derived from a

single stable conformer or a rapidly equilibrating mixture

of both conformers. To assess this value.some idea of the

63

TABIiS VIII

CALCULATED INVERSION ENERGIES

Frequency Factor Energy* (sec-1) (kcal/mole)

11 o 1 10XJ" 3. 1012 . 9.0 101 3 9.7 101 4 10.6 101 5 11.6

*Values determined at -35 degrees centigrade and a rate constant equalling 45 sec-1.

process requring such energy must be envisioned. In accordance

with popular concepts of the nature of the carbon lithium bond

this inversion pathway can be envisioned through a carbanion

intermediate. However the fact that organolithium compounds

exist as associated species in hydrocarbon solvent would re-

quire a complex dissociation step prior to the formation of

the intermediate. Curtin and Koehl (9) have speculated that

a complex ion containing a number of alkyl lithium units is

involved in the racemization mechanism of optically active

sec-butyllithium in ether solvents. Both of the proposed

processes are diagrammed in Figure 6.

Miller and co-workers (15) have recently investigated

barriers to intramolecular motion which leads to inversion.

They reported barrier energies for XY3 acyclic carbanions

to be in the range of 10 kilocalories per mole. They also

reported that if carbanions are involved in the inversion of

a. Carbanion pathway

89

H I c;

.'Lk

H

H

Li

if-)

^ C - H

H C " 1 Li

Li— H ^ I

Li I C—H

"b. Complex ion pathway

R Li n n

R_ i Li rH n

w c~>

R n L L n n

Pig. 6—Inversion pathways

*The prime (R *) indicates an aggregate containing at least one organic moiety whose conformation is inverted from R.

90

cyclic specics, the "barriers would be higher than those

reported for acyclic analogs. It should be emphasized that

the 12 kilocalories per mole energy value reported earlier

is derived from a different process than the 10 kilocalories

per mole value reported here. This lower energy value in-

cludes only the energy required to invert the carbanion and

not the energy to form this radical.

It is well-known that alkyllithium compounds exist as

associated species in hydrocarbon solution. Therefore the

first step in forming a carbanion intermediate is to break

up these associated systems. Morton and Fetters (20) have

reported an enthalpy of dissociation of 37 kilocalories per

mole for dimeric polyisoprenyllithium. Brown (4) estimates

from nuclear magnetic resonance studies that the dis-

sociation of hexameric ethyllithium to a dimer and a tetramer

requires 16 to 18 kilocalories per mole. Furthermore ad-

ditional energy will be required in forming the carbanion

from an alkyllithium monomer or dimer. One therefore gets

the impression that the inversion barrier would be much

higher than the 12 kilocalories per mole value determined

if carbanion fo^.ation ie involved.

On the other hand the complex ion pathway in hydrocarbon

solvent would definitely represent an even higher energy

process than the carbanion pathway. Prom a mass spectrum

analysis of ethyllithium Brown and co-workers (3) have

91

reported the energy required to dissociate the hexameric

aggregate of this compound. This ionization process (a)

diagrammed below requires 111*5 kilocalories per mole of

Li^Rg + e' • LigR^" •+• R + 2e'

(a)

energy. This example illustrates the approximate energy

requirement necessary to consider the complex ion pathway.

It is evident from this discussion that the observed

resonance is not derived from a rapidly equilibrating mixture

of the two lithium epimers. The correct assignment therefore

appears to be the «<-methinyl proton absorption of a con-

figurationally stable menthyl- or neomenthyllithium compound.

Based solely on the shape and width of the signal one might

predict it to be the menthyl conformer.

Evidence of the absorption frequency and characteristic

shape of the other epimer resonance would certainly enhance

the configurational assignment for both signals. It seemed

that employment of & more basic solvent, such as an ether,

might facilitate some exchange pathway and thereby yield an

inverted product. Such an effect could generate a sufficient

concentration of the second epimer so that a resonance signal

would be observed.

The first solvent tried was diethyl ether. Above zero

degrees centigrade this solvent reacts in a very vigorous

manner with the lithium solute. At lower temperatures this

reaction is retarded and resonance signals were recorded.

92

No change in the characteristic shape of the resonance from

that observed in hydrocarbon solvent was recorded. However

a shift in absorption frequency was observed with the resonance

occurring at 11.00 ^ (5b) instead of 10.5 (5a*) as reported

earlier. This type of shift for these electron deficient

systems in more polar solvent is not uncommon. The facts

that no change in the absorption pattern is observed and

that no other signal appears are certainly in agreement with

a similar investigation reported by Cram (8). He cited from

a private communication of Roberts, Whitesides and Witanowski

that 3,3-dimathylbutyllithiuin appears to be configurationally

stable in diethyl ether below zero degrees centigrade.

Application of dimethyl ether, which io known for its

ability to solvate polar systems, as the solvent produced

the long awaited evidence for a second absorption. The new

absorption occurred at 10.13 "2" (5c) and had the characteristic

shape and width expected for a neomenthyl o<-methinyl proton

absorption. The characteristic broad multiplet, which oc-

curred at 11.00 ^ (5b) in diethyl ether, was also present at

10.95 jh (5c) in this solvent. Integration of these two

signals indicated a 75:25 ratio of the menthyl-neomenthyl

methinyl resonances. These results were recorded at -45

degrees centigrade.

The fact that these two absorptions were separated by

50 cycles per second, and that they demonstrated the expected

93

signal widths and shapes of a raenthyl-neonienthyl epimer

pair was very informative. In order to determine whether

the same results could "be observed for another ether solvent,

tetrahydrofuran, which can also be characterized as a polar

solvent, was employed. As with dimethyl ether both resonances

were observed and the two spectra could be superimposed. The

fact that identical spectra were obtained in two different

ether solvents provides good evidence that the second

resonance is not derived from a lithium derivative of the

solvent. To further demonstrate that these two resonances

are derived from the menthyl- and neomenthyllithium oC-protons

an investigation employing chemical means was undertaken.

Seyferth and co-workers (23, 24) have demonstrated

that carbonation of alkenyl- and cyclopropyllithium compounds

proceeds with retention of configuration. Although this

stereospecific pathway has not been demonstrated for cyclic

lithium compounds it has been demonstrated for norbornyl

Grignard reagents by Jensen and Nakamaye (14). These data

certainly predict a retention pathway for the menthyl com-

pounds under investigation. Therefore the first attempt to

demonstrate the origin of the observed resonances by chemical

means employs this reaction.

Solutions originating from a variety of sources, such as

the original reaction solution, solutions derived from dis-

solving crystalline material in different hydrocarbon and

94

ether solvents, and mother liquor solutions were carbonated.

Two procedures were employed in carrying out these reactions.

The first process involved pouring the solution of lithium

compound over a slurry of dry ice and n-pentane at -70 degrees

centigrade; the other procedure involved "bubbling carbon

dioxide gas through the organolithium solution. The resulting

acid products were either isolated through crystallization or

esterified with diazomethane. The results obtained from

these systems are recorded in Table IX.

It is easily observed from these data that 3-p-methane

carboxylic acid is the primary derivative produced from

carbonation of any of the three solutions containing the

lithium compound in hydrocarbon solvent. This evidence for

only one isomer is definitely complemented by the optical

rotation and infrared data reviewed earlier. Furthermore,

upon replacing this solvent with dimethyl ether, a substantial

increase in the other acid epimer is produced. This ob-

servation together with the fact that the same isomer dis-

tribution i,n dimethyl ether is observed by chemical means

and from proton magnetic resonance spectra is very informative.

It establishes that the resonance assignments made earlier

from proton resonance spectra analyses were correct. However,

the most important conclusion to be drawn from these data is

that the menthyllithiuin exists as a conformational^ stable

compound in hydrocarbon solvent.

9 5

CO

o

Ph

X EH

a

rH

c 0

o •H

n o5 O A1 CD

£h !H •* 0

r H £ >> O

,-q ffi - P H S3 0

*P *"0 O 0 3 Esl

* d t>5 O H ^ cd

PM C2 <cj

S3 O O w

•H - P cd

CD

O T5 rO 0) !h O Cd O

O Jh Ph

• P a 0 >

r H O

CO

a , a o w

*H •P 3

H o

CO

a • H ^ 5

*H u

o

03

0 i H T3 CO - H •H r H - p eg *H W S3

M

<11, T3 rC> nd t : ;

O r H i n KO r H i—I m !>- i—I r-H m r H H~ K \ J>5 * • • * W ** t>5 t>> • • • • • •

r a i n ^ i r \ k n r c j & i i \ & v o o

a d S3 J3 c 0) 0 0 0) 0 0)

f-^ f~HS »-h|

?h ?H %-i ^ £-r rC5 0 d ) r d n d C D <D <13 <d 0 <d 0 0 *H - P P *H *H P - P - H «H - P *H *P *P O C Q C D O O C Q C D O O G O O C O C O

< $ p q i 3 < i } < t | p q | i q < $ « 4 p £ | < $ w p q

M H I H M

c i a a a S3 S3 o o o o o o

P P p P P r Q ^ ^ ^ fn ?H ^ 03 O O Sh

<d * d * d »d f d f d * d ^ t>3 t>>

j x j i x j w w W f r j UJ

d s3 $3 o o o

fn Jh ^ cd cd cd o o o o o o u u u

nd «d * d

M M H

Sh Jh 0) CD ,£} <3 - P •P W W

r H H >5 l ^ r^H ^ - P

£ 5 8 > v H *H

W f l f i

a , o

rQ u 03 i O O ,

< i j < j ; f q o o o < a J < ^ < ^ f q o < i J < t J

c d o 5 c d a 3 a 3 c d r O r Q r j D r Q r Q o 3 a 3 M M M M M f H I — I M H M H H M

a) CO • 03

• H 0 H H i H 3 - P cd T * O

p 0 d CO o nd t>» o O

u o • P4 P^

S n h CD O 0 cd *H JH > t i ) a

CH # h 03 0

o n3 0 0 (D *H r*H J> £1( N - P

* •H CO O 0 ^ CD • H CO

* d 0 Js 0 *H $ rH 0 £3 r H o CD U O 05

1—1 a d > r*H O - P JH O • H X 03 S3

P -H ^ a

O o • H

P -H ^ a • H !H

r H • P 0 o a 03 ^

- P ro o H • P ^ £3 •H 1—J o o CD - p - p JH c H 6 S3 o •% £3 o a) o3 0 O G CD 0} 0 H •H O S3 3 t H

u T i • H H P4f~-l 0 O >> CD 03 CJ 0 ,Cj

rQ u a o p 4 - p M ' H CD S3

O ^ 0 p4 0 • H H3 S

CD r c ! JH 0 <3 * o S3 O S3 cd 0 *H d 03 A

*C5 - P O ^ CI S3 £3

H pq 03 0 •H J>> pq 03 Pi O r-\

O • * f P4«H Jh a

r H O i i*0 cd CD ^ 0 S3 S fn O"1 o *H - H

+D CD *H • H •P JH S3 •f--5 H H P4 CD •P t>3 0

CD SH pi CO r H 0 nd cd

r'3 O 4^ O

03 r«3 O JH 0 W 6H

««

^ - x r H

M tS3 • H JH

^ 0 03 rO 03 O • d - P

**—"*H Jh 0 p»

05 g

w o 03 JH 05

^ 3 o

96

In order "to obtain additional chemical evidence for

these observations the reaction between deuterium oxide and

(+) -menthyllithium was studied. This reaction was selected

because a large amount of conformational analyses data for

the carbon-deuterium stretching frequencies in cyclohexyl

compounds has been reported. Corey, Sneen} Danaher, Young,

and Rutledge (7) have published evidence that the stretching

frequencies for the monodeutero compounds exhibit two bands.

In the case of the axial deuterium these bands are separated

characteristically by 25 wave numbers and in the case of

equatorial deuterium they are separated by 15 wave numbers.

Furthermore, the absorptions due to axial deuterium for the

compounds studied in this investigation occur at nearly the

same place; those due to equatorial deuterium occur at higher

frequencies. The final distinction noted was that the higher

frequency band of the doublet is more intense for axial whereas

the low frequency band is more intense for equatorial deuterium.

Armed with these conformational analysis data and a very

similar system the following results were recorded.

3-deutero-p-Menthane is produced in the reaction between

deuterium oxide and menthyllithium. "The infrared spectra

illustrated in Figure 7 demonstrates that the axial and

equatorial deuterium product distributions are dependent

upon the solvent employed. Deuterolysis in hydrocarbon or

diethyl ether solvent produces a fairly broad single absorption

97

a.

L

2500 cm 2000 2500

b.

2138

2167

cm -i 2000

Pig. 7—Infrared spectra of 3-deutero~p-rienthane in hydrocarbon solvent (a) and dimethyl ether solvent ("b).

9$

at 2167 wave numbers. The width of this absorption evidently

maslcs out the low intensity band, thereby eliminating charac-

terization of the absorption with respect to the position or

the characteristic separation from the unobserved low in-

tensity band. However the observed absorption does occur

in the frequency range characteristic of equatorial deuterium

compounds.

Two absorptions are observed in the spectrum of the

deuterium products isolated from the same deuterolysis re-

action conducted in dimethyl ether solvent. A distinct

shoulder absorption on the 2167 wave number band was recorded

at 2138 wave numbers. The .fact that only two and not all

four absorption bands are present eliminates an inanediate

characterization. However the observation that these two

absorptions are separated by 29 wave numbers does exclude

the possibility that they both belong to a single deuterium

compound. Furthermore the fact that the 2138 wave number

absorption is found in the frequency range characteristic

of axial deuterium cyclohexyl derivatives is in agreement

with this observation. Therefore the assignments of these

two bands as the strong equatorial and axial deuterium

stretching frequencies can be confidently made from these data,

It is noteworthy that the carbonation and deuterolysis

chemical evidence definitely compliments the results reported

from the proton magnetic resonance investigation. Reviewing

99

these data along witli the optical activity and infrared

results, the following conclusions regarding the inenthyl system

are derived:

1. Isomerization of the two chloride reagents occurs

in their reaction with lithium metal in hydrocarbon solvent.

2. This reaction yields a single optically active

menthyllithium isomer as the primary product of the reaction.

IIo more than 5 per cent of the net carbon "bound lithium

resides in an axial conformation.

3. The (+)-menthyllithium isomer prepared demonstrates

configurational stability for indefinite time periods in

hydrocarbon and cold diethyl ether solvents.

4. In contrast to the last observation some isomeri-

zation process exists at temperatures below -20 degrees

centigrade in dimethyl ether and tetrahydrofuran solvents.

This process yields a mixture of the menthyl- and neomenthyl-

lithium epimers. Both of these compounds exhibit configurational

stability on the nuclear magnetic resonance time scale at these

temperature.

5. Both carbonation and deuterolysis reactions of these

lithium isomers proceeds by configurational retention pathways.

A conformational analysis investigation of the 4-tert-

butylcyclohexyllithiuia product was undertaken upon completion

of the menthyllithium characterization. Due to the success

experienced with the menthyl system the same techniques were

100

employed in tlie investigation of the 4-tert-butylcyclohexyl

compound. However a few procedure changes were required be-

cause of the inherent low solubility property demonstrated

by this compound in hydrocarbon solvents. ?or example,

carDonation and deuterolysis reactions were conducted on

the crystalline product suspended by agitation in such

solvent instead of on true solutions of the dissolved

crystalline material.

Proton magnetic resonance spectra of the o<-methinyl

resonances in various solution are recorded in Figure 8.

A spectrum of the original reaction solution is diagrammed

i*1 illustration 8a. Due to the problem of side bands occur —

2 ing in the field region of interest when n—pentane solvent

is employed this solution had been concentrated by removal

of a large portion of this solvent; subsequently benzene

standard was added. The spectrum shows a broad unresolved

multiplet resonance at 10.59^ and two other absorptions 43

and 55 cycles per second down field from this resonance.

Employment of the frequency difference of 48 to 50 cycles

per second recorded in the menthyllithium investigation

provided no evidence for the characterization of either of

these two lower field resonances. It was believed that these

two resonances were derived from by-products of the synthesis.

In an effort to investigate further the possibility that

both of the lithium isomers exist in hydrocarbon solution a

101

b.

iiti , I a_ I • • - • 1 i mJLm _ a _ J i i a.

10.0 11.0

-ig. -8— ©<-Methinyl proton resonance of 4-tert-butyl-cyclohexyllitlviuia samples of an original reaction solution in benzene (a) and crystalline material dissolved in benzene ("b), diethyl ether (c), dimethyl ether (d).

102

spectrum of dissolved crystalline product was required. At

this point the low solubility property of the crystalline

material was realized. Spectrum 8b was recorded on a typical

saturated benzene solution of this material. It was soon

observed however that addition of diethyl ether solvent to

a proton magnetic resonance sample tube containing undissolved

crystalline material in a saturated benzene solution yielded

a recordable spectrum as shown in spectrin 3c. No appreciable

concentration of the two resonances in question from spectrum

8a were observed. Also the broad multiplet is resolved into

the characteristic broad triplet resonance observed in the

menthyl spectra. It is noteworthy that no shift in the ab-

sorption frequency of this resonance occurred upon addition

of diethyl ether.

These data compiled from spectra 8a, 8b, and 8c suggest

that solutions prepared from the crystalline material in

either hydrocarbon or diethyl ether solvents are primarily

composed of the trans-lithium isomer. If either of the

other low field resonances are derived from the cis-isomer

then either its concentration is generated during the synthesis

or a slow equilibrium exists between these two isomers.

Dimethyl ether solvent was employed at this point to

determine whether either of these two low field absorptions

were produced. Spectrum 8d illustrates the resonances re-

corded in this solvent. The characteristic broad triplet

103

resonance occurs at 10.05/" and a second resonance at 10.03^.

Obviously a frequency shift of the resonances occurs in this

solvent. It ras also noted that the 10.03 jh resonance shifts

to 9.S5 Ih with changes in temperature over extended time

periods. The resonance position of this particular absorption

corresponds to the position of the 10.03^ resonance recorded

in spectrum 3a. These two resonances exhibited an approximate

30:20 high field to low field resonance ratio at -50 degrees

centigrade. An inversion of this ratio can be observed as

the temperature for such a solution is warmed from -70 degrees

to -30 degrees centigrade. The observation that this new

resonance could be concentrated yields an ideal system for

characterizing this new absorption by chemical means.

In contrast to the results recorded for the menthyl

system, the applications of carbonation and deuterolysis re-

actions produced only evidence for the existence of the trans-

lithium isomer in all of the previously described solutions.

For example, no observable difference in the carbon-deuterium

stretching frequency spectra of the 4-tert-butylcyclohexane-

1-d] product derived from crystalline lithium compound dissolved

in any of the solvents of interest could be detected. A typical

spectrum of the carbon-deuterium absorption region is illustrated

in Figure Q. The observed absorption doublet may be easily

characterized from the works of Corey and co-workers (7) as

the symmetric and asymmetric stretching modes for the equatorial

deuterium conformation.

104

v — v / V

2184

2167

2500 cm 2000

Fig. 9—Infrared spectrum of 4~tert-butylcyclohexane-l-d- in the carbon-deuterium stretching region.

105

In the case of 'the carbonation reaction the only

product detected after esterification with diazoraethane

was the trans-ester isomer. This result was observed to

he independent of the solvent employed. However a new

component in the product mixture of a dimethyl ether solution

was detected "by gas liquid chromatography. This solution h&d

"been characterized "by a proton magnetic resonance analysis

to be concentrated in the unassigned low field resonahce. A

complete characterization of this component was not obtained

through employment of preparative chromatography combined with

other analysis techniques, but sufficient evidence was recorded

to demonstrate that it was not the cis-ester isomer.

The conclusions to be drawn from these data are that the

4-tert-butylcyclohexyllithium compound prepared from the cis- .

chloride isomer and lithium metal is the trans compound.

Furthermore, in hydrocarbon and diethyl ether solvents this

isomer demonstrates configurational stability and indicates

a definite preference for the lithium atom to reside in the

equatorial position. However in dimethyl ether solvent the

behavior of the lithium compound is not so well understood.

The recorded data suggest the existence of some interaction

between dimethyl ether and the lithium compound. Reactions

of this type, such as metalation and methylene insertion

reactions, have been reported by Ziegler and Gellert (28).

However these reactions occurred only at temperatures above

106

-20 decrees centigrade. The fact that some similar process

could occur at slightly lover temperatures appears reasonable,

but whether the interaction involves the cis- or trans-lithium

isomer is the question of interest. It is unfortunate that

the evidence presented does not indicate whether isomerization

occurred in this solvent.

Combining the conformational analysis data reported from

the investigations of the menthyl and 4-tert-butylcyclohexyl

systems, one main conclusion is evident. This noteworthy ob-

servation is the preference demonstrated by the lithium atom

to reside in an equatorial conformation in substituted cyclo-

hexyllithium compounds. One might find it difficult to explain

this phenomenon from the size of the covalent radius of the

lithium atom. However, recalling the fact that these compounds

exist as associated species in hydrocarbon solvent, makes it

easier to illustrate that this functionality is large enough

to demand a particular conformation. One other observation

common to both systems is that carbonation and deuterolysis

reactions bpth proceed with configurational retention pathways.

Initially there was some concern over complication of the

stereochemistry of the menthyl system reactions by neighboring

group and asymmetric induction effects. These results obviously

indicate that if such effects are present they are definitely

insignificant for these two reactions.

107

It is evident that sufficient data has not been reported

to postulate mechanisms for these two reactions. However a

prediction that these two reactants form similar assisted

intermediates does not appear to "be an unrealistic explanation

for the recorded results. Obviously such a prediction could

not be made for a' reaction which yields configurationally

inverted products. Applequist and Chiuurny (1) have reported

evidence for such a reaction involving bromine and norbornyl-

liohium in hydrocarbon solvent. In order to determine whether

the stereochemistry of such a reaction is complicated by

neighboring group and asymmetric induction effects the

bromination reactions of menthyl- and 4-tert-butylcyclohexyl-

lithium compounds were investigated. The results recorded

from these reactions are listed in Table X. These results

demonstrate that a definite temperature effect on the stereo-

chemistry of this reaction exists. At low temperatures pre-

dominant inversion of configuration is observed for the bromide

derivatives of these two lithium reagents. However at higher

temperatures this product distribution is reversed for the

4—tert-butyIcyclohexyl bromides. Therefore from these data

and the findings reported by Applequist and Chmurny (1) it

is evident that solvent, temperature, and the nature of the

organic moiety of the lithium reagent can all effect the

stereochemistry pathway followed in this electrophilic sub-

stitution reaction.

103

TABLE X

BROMIHATIOF PRODUCTS PR OK SOLUTIONS OP IvIENTIiYLLITHIUM AND 4-TERT-BUTYLGYCLOHEXYLLITHIUM

Lithium Compound Reactants and Conditions

Product Ratio Equatorial Axial

4-tert-butyl-cyclohexyllithium

Original reaction solution, n-pentane solvent, -70°C. 23/77

4-tert-"butyl-cyclohexyllithium Crystalline material,

n-pentane solvent, 70°C. 21/79

Mother liquor solution, n-pentane solvent, -70°C. 21/79

Crystalline material, n-pentane solvent, 0°C. 37/63

Mother liquor solution, n-pentane solvent, 0°C. 58/42

Crystalline material, dimethyl ether solvent. -70°C. 59/41

Menthyllithium Crystalline material, n-pentane solvent, -70°C. 38/62

Menthyllithium Crystalline material, n-pentane solvent, 0°C. 65/35

109

Nov/ the question arises as to whether the nature of *A

the electrophile should be included in this list of variables

when electropirilic substitution reactions of alley H i thiuin

compounds are being considered in general. Data concerning

this question are included in the recorded product dis-

tributions of Table XI. These data were obtained from an

investigation of substitution reactions between the mentliyl-

and 4-tert-butylcyclohexyllithium reagents and a variety of

electrophiles.

One explanation for the observed variations in the

reaction products listed in Table XI has been published by

Applequist and Chmurny (l). These authors have reported

that retention and inversion pathways for such substitution

reactions appear to be inherently equal. However one is

favored over the other by changes in the conditions and re-

agents employed in carrying out the reaction. For example,

they speculated that carbon dioxide way have an affinity for

lithium great enough to proceed by a four-membered transition

state in spite of some inherent preference for inversion.

It is noteworthy that this explanation is based on the

assumption that the lithium reagent did not isomerize in the

presence of the bromide electrophile. The fact that this

assumption is not only questionable but unnecessary is

demonstrated by the conformational stability data and the

observed variations in product distributions data for different

110

TAB HE XI

PRODUCT DISTRIBUTION PROM ELECTROPHILIC SUBSTITUTION REACTIONS OP THE CYCLOHEXYLLITHIUE COMPOUNDS

Electrophile Conditions Products* Product Ratio Equatorial Axial

Carbon dioxide n-C5Hi2,25°C &6, b 6 95/5 Carbon dioxide (CH3CH2) 0,-40°C a6 »b6 96/4 Carbon dioxide (CH3)20,-40°C a6,b6 70/30 Carbon dioxide n-C5Hi2,25°C °6 » 6 95/5 Carbon dioxide (CIl3)20,-40OC C6,d6 95/5

Deuterium oxide 11-0511x2,25°C a3 ,b 3 * *

Deuterium oxide (CH3CH2)20,-40°C a3,b3 *-x Deuterium oxide (CH3)20,-40°C a3,b3 **

Deuterium oxide n-C5ll3 2,25°C C3,d3 * *

Deuterium oxide (CH3CH2)20,25°C °3>d3 Deuterium oxide (CH3)20,-70

oC c3,d3 **

Ethylene dibromide n-C5H12,25°C a4 >^4 97/3 Ethylene dibromide (CH3)20,-70°C a4 ,b4_ 88/12 Ethylene dibromide n-C5Hi2,25°C c4 »^4 95/5

Bromine n-C H-, 2 ,-70°C a4 j^4 38/62 Bromine n-C5Hx2,0°C a 4 >0 4 65/35 Bromine n-C5Hl2,-70°C c 4»d4 21/79 Bromine n-C5Hi2,25

0C C4»d4 37/63 Bromine (CH3)20,-70°C c4»d4 60/40

Oxygen n-C5H12>25°C a2>'b2 80/20

Kethyl iodide n-C5Hi2j25°C a5,b5 80/20

*A description of the products can be found in Figure 1.

**Since these samples were analyzed by infrared spectro-copy only a qualitative estimate of the ratio can be reported, Except for the spectrum of the dimethyl ether solution of menthyllithiura the carbon-deuterium absorption was primarily derived from an eqiaatorial conformation.

Ill

elcctrophiles all included in this investigation. These

data suggest that the character of the electrophilic agent

determines the pathway through an isomerization effect on

the lithium reagent, furthermore this effect is facilitated

to different degrees "by the conditions employed in conducting

the reaction. The proposal of this effect requires that the

stereochemistry pathway followed in such reactions "be one of

retention. Such a process can he easily envisioned for alkyl-

lithium compounds as proceeding by an assisted mechanism.

Although these conclusions are only speculations they seem

more consistent with the prevailing concepts of organometallic

mechanisms.

Pyrolytic elimination of lithium hydride from lithium

alkyls has been a characteristic reaction of interest since

the work of Schorigin (22) was reported in 1910. An in-

vestigation of this reaction has been initiated for the con-

figuration lly rigid (4-)-menthyllithiuLi compound. ]?rom this

study two main goals are sought. The first of these is to

report the stereochemistry of a characteristic elimination

reaction along with all the previously listed substitution

reaction data. The second and most important goal is to

compare the results recorded from the decomposition of this

conformationally rigid system with results reported by Glaze

and co-workers (11) for sec-butyllithiurn. These authors

proposed that sec-butyllithium decomposes by a cis-elimination

112

mechanism. They further speculated that the association

character of this alkyllithium compound had a definite effect

011 the pathway exhibited for this reaction. Whether a similar

pathway will he exhibited for the menthyllithium compound can

be demonstrated from an analysis of the decomposition products,

The following illustration demonstrates how this can be deter-

mined.

H

Li II

pseudo-trans

il

H

2-p-nienthene

CH3

H H

Mixture of 3-p-Menthene 2-p-menthene

Evidence for a cis-elimination mechanism has also been

demonstrated for the thermal decomposition of (-)—menthyl

chloride by Barton and co-workers (2). This mechanism

assignment was based on the 3:1 ratio of p-3-rnenthene and

113

These p-2-monthene products .Corned in the decomposition.

auoiiors concluded that the (-)-menthyl chloride pyrolyzed

by a Juonogeneous unimolecular mechanism. They proposed

the following four centered transition state (a) as being

the product yielding intermediate. They explained the

C^=C i i i « II--CI (a)

preference for a cis- over a trans-elimination process for

this system as being derived from the nature of the cyclo-

hexy.;. ring. This system was believed to be capable of

minimizing the activation energy for the proposed transition

state by allowing this intermediate to lie in a single

plane. On the other hand Huckel and co-workers (13) have

demonstrated that under the conditions of an E2 reaction

with sodium ethoxide a trans-elimination mode of reaction is

preferred. This assignment was also based on the observed

pioduco distribution, which in this case yields only

p-2-menthene.

A 45:55 ratio of p-2-menthene to p-3-menthene products

was recorded in the thermal decomposition of (+)-menthyllithium(

Although this distribution is somewhat different from that

recorded for the chloride derivative it definitely indicates

that the lithium compound decomposes by a cis-elimination

mechanism. Whether the proposed pathway is derived by the

114

effects noted by Barton and co-workers (2) or by Glaze and

co-tvor Iters (11) is not known at this time. .Furthermore

the effect of alkoxide concentration on such a process is

unknown. This concentration of alkoxide might produce the

variation in product distributions between the chloride i

reaction and the lithium decomposition by facilitating

some E2 reaction products. Both of these questions are

interesting but additional data will be required before they

can be answered.

CHAPTER BIBLIOGRAPHY

1. App.Tequist, Douglas E. and Gwendolyn Weal Ghuiurny, "Stereochemistry of Electrophilic Substitution on 2-Norbornyllithiuin," Journal of the American Chemical Society, LXXXIX (1967}, 075. '

2. Barton, D. R.H., A. J. Head, and R. J. Williams, "Stereospeeificity in Thermal Elimination Reactions, Part II. The Pyrolysis of (-)-Menthyl Chloride," Journal of the Chemical Society (1952), 453.

3. Berkowitz, Joseph, D. A. Bafus, and Theodore L. Brown, "The Mass Spectrum of Ethyllithium Vapor," Journal of Physical Chemistry, LXV (I96I), 1380.

4. Brown, Theodore L., "On the Mechanism of Organolithium Reactions in Hydrocarbon Solution," Journal of Organometallic Chemistry, V (1966), 19I.

5. Brownstein, S. and R. Miller, "Conformational Analysis of Disubstituted Cyclohexanes by Proton Resonance Spectroscopy," Journal of Organic Chemistry, XXIV (1959), 1886. b ^

6. Castellano, S. and A. A. Bothner-By, :Analysis of HMR Spectra by Least Squares," Journal of Chemical Physics, XXXXI (1964), 3863.

7. Corey, E. J., R. A. Sneen, M. G. Danaher, R. L. Young, and R. L. Rutledge, "A Method for the Location of Deuterium Atoms Attached to Oyclohexane Rings," Chemistry and Industry (1954), 1294.

8. Cram, Donald J., fundamentals of Carbanion Chemistry, Hew York, Academic Press, Inc., Ifj6^.

9. Curtin, David Y. and 7'illiam J. Koehl, "Effect of Solvent on the Steric Stability of Lithium Reagents," Journal of the American Chemical Society. LXXXIV (1962), T$6T.

10. Dryer, John R., Applications of Absorption Spectroscopy of Organic Compounds, Englewood Cliffs, Prentice-Hall, Inc., 1965.

115

11. Glaze, 'Yilliam H., Jacob Lin, and 3. G. .Pelton. "The yi olysis of Unsolvated Alkyllithium Compounds," Journal of Organic Chemistry, XXXI (I966), 2643.

12. Horner, von Leopold, Hermann Oediger, and Hellmut Hoffman, "Reaktionen Mit Triphenyl-Phosphin-

DCXXVIg( 19^9 j1'"2^ustus L i e b i^ s Annalen der Chemie ,

15* Walte]f> Werner lappe, and Gunter Legutke, Abspalbungreairtionen und Ihr Sterischer Verlauf "

Eiebi^s Annalen der Chemie, DXXXXIII (1940),

14. Jensen, Frederick R. and Kay L. Nakamaye, "Preparation Isomeric Grignard Reagents and the

Stereochemical courses of their Reactions," Journal ' of^t^e American Chemical Society, LXXXVIII (I966)t

15. Koeppl G. i. D. 6. Sagatys, G. S. Kirshnamurthy, and filler, "Inversion Barriers of Pyramidal

(XY3 and Related Planar ( XY) Species," Jo^nafof the American Chemical Society. LXXXIX (I967), 33967

16* ^ild® a M Dwaine 0. Cowan, "Stereochemistry Reage^ts from exo- and endo-Norbornyl

ixx^in (xgroTr&ssf — 4-merlc?-n- Sississi society,

17# Le .ieux, R. U., R. K. Eullnig, H. J. Bernstein, and I' +* "Configurational Effects in the ?ro*°P Magnetic Resonance Spectra of Acetylated carbohydrates," Journal of the Amerinan Society. LXXIX (IgbV), 1UU5Z ^nemical

18.

|?o^nTE|netio'Resoiiantrsp«tra of Sii MeSfcered S o c f e ^ ; P ™ 3 ( i 9 f g f ^ 9 § £ African Cheaical

IS5(ai9l2TrS MineS'" ^£ilft"e 3o£e,ioal

117

20. Morton, M. and L. J. Fetters, "Homogeneous Anionic Polymerization V. Association Phenomena in Organo-lithium Polymerization," Journal of Polymer Science, A2 (1964), 3311. ;

21. Read, John, "Recent Progress in the Menthone Chemistry,,!

Chemical Reviews, ¥11 (1930), 1.

22. Schorigin, P., "Uber "bis Natiumalkyle Hire Reaktion hi it der Atheru," Berichte "Der Deutschen Chemischen Gesellschaft, XXXXIII (l'jl&TT W 5 T .

23. Seyferth, Dietmar and Harvey M. Cohen, "The Stability of Cyclopropyllithium in Diethyl Ether and in Tetra-hydrofuran," Journal of Organometallic Chemistry, I (1963), 15.

24. Seyferth, Dietmar and Lawrence G. Vaughan, "The Isomeric 1-Propenyllithium Reagents: Their iMR Spectra and the Stereochemistry of Their Formation by Direct Reaction and by Their Transmetalation Reaction," Journal of Organometa11ic Chemistry, I (1963), 201.

25. Smith, J. G. and George F. '/right, "The Diastereomeric Menthyl Chlorides Obtained from (-)Menthol," Journal of Organic Chemistry, XVII (1952), 1116.

26. Walborsky, H. L I . , F. J. Impastato, and A. E . Young, "Cyclopropanes. XV. The Optical Stability of 1-Methy1-2,2-diphenyl-cyclopropyllithium," Journal of the American Chemical Society, LXXXVI (1§64), 3283.

27. William H. Glaze, unpublished notes, Department of Chemistry, North Texas State University, Denton, Texas, 1967.

28. Zeigler, von Karl and II. G. Gellert, "XVII. Reaktionen Zwischen Lithium-alky leu und Athern," Justus Liebigs Annalen der Chemie, DLXVII (1950), 1851

BIBLIOGRAPHY

Books

Cram, Donald J., Fundamentals of Carbanion Chemistry, New York, Academic Press, Inc., 1965.

Dryer, John R., Applications of Absorption Spectroscopy of Organic Compounds, Englewood Cliffs, Prentice-Hall, Inc., vjoTI

Sadtler Standard Infrared Spectra, Philadelphia, The Sadtler Research laboratories, Spectra Numbers 184-6 and 20221.

Articles

Alexandrou, N. E., "Reaction of n-Butyllithium with cis-and trans-4-tert-Butylcyclohexyl Bromide and Mercuric Bromide," Journal of Organonieta11ic Chemistry, V (1966), 30TI

Allinger, Norman L. and Robert L. Hermann, "Infrared Spectra of cis- and trans-Propenyllithium Stereochemistry of the Reaction of Vinyl Chlorides with Lithium Metal," Journal of Organic Chemistry, XXVI (1961), 1040.

Applequist, Douglas E. and Gwendolyn Heal Chmurny, "Stereo-chemistry of Electrophilic Substitutions on 2-Norbornyl-lithium," Journal of the American Chemical Society, LXXXIX (1967), 8?5:

Applequist, Douglas E. and Alan H. Peterson, "The Configura-tional Stability of cis- and trans-2-methylcyclopropyl-lithium and Some Observations on the Stereochemistry of their Reactions with Bromine and Carbon Dioxide," Journal of the American Chemical Society, LXXXIII (I96I), 862.

Arndt, J*., "Diazomethane," Organic Syntheses, XV (1935), 3.

Barton, D. R. H., A. J. Head, and R. J. Williams, "Stereo-specificity in Thermal Elimination Reactions. Part II. The Pyrolysis of (-)-Menthyl Chloride," Journal of the Chemical Society (1952), 453.

118

119

Berkov/itz, Joseph, D. A. Bafus, and Theodore L. Br own, "The Ilass Spectrum of Ethyllithium _Vapor," Journal of Physical Chemistry, LXV (I96I), 1380.

Bordwell, P. G. and phillip S. Landis, "Elimination Reactions. VII. A Comparison of Some Eliminations in Open-Chain and Cyclic Systems," Journal of the American Chemical Society, LXXIX (1957), 1593.

Braude, 3. A. and J. A. Coles, "Alkenylation by Use of Lithium Alkenyls. Part IV. The Condensation of cycloHexenyl-lithium with Aldehydes and. Ketones, and Some Reactions of the Resulting Carbinols. (Studies in Molecular Rearrangement. Part V.)," Journal of the Chemical Society (1950), 2078.

, "Alkenylation Employing lithium Alkenyls. Part V. The Formation and Some Reactions of cis-Propenyllithium," Journal of the Chemical Society (1951), 2078.

Braude, E. A. and C. J. Timmons, "Alkenylation by Use of Lithium Alkenyls. Part I. The Condensation of isoButenyl-lithium with Benzaldehyde and Acetophenone, and the Oxotropic Rearrangement of the Resulting Carbinols. (Studies in the Molecular Rearrangement. Part III.)," Journal of the Chemical Society (1950), 2000.

Brown, Theodore L., "On the Mechanism of Organolithium Reactions in Hydrocarbon Solution," Journal of Organometallic Chemistry, V (1966), 1^1.

Brown, Theodore L., J. A. Ladd, and G. II. Newman, "Inter-action of Alkyllithium Compounds with Base. Complex Formation Between Ethyllithium and Lithium Ethoxide in Hydrocarbon Solvents," Journal of Organometallic Chemistry, III (1965), 1. '

Brownstein, S. and R. Miller, "Conformational Analysis of Disubstituted Cyclohexanes by Proton Resonance Spectro-scopy," Journal of Organic Chemistry, XXIV (1959), 1886.

Castellano, S. and A. A. Bothner-by, "Analysis of M R Spectra by Least Squares," Journal of Chemical Physics, XXXXI (1964), 3863.

Clifford, A. L. and R. R. Olsen, "The Analytical Chemistry or Organometallics. Quantitative Determination of Organoalkalies," Analytical Chemistry, XXXII (i960), 544.

120

Collins, Peter ?., onrad Eamienski, Donald L. Esmay, and R. B. Ellestad, "Oxidimetric Determination of n-Butyllithium in Hydrocarbons Using Vanadium PeQt-oxide," Analytical Chemistry, XXXIII (I96I), 468.

Corey, S. J., R. A. Sneen, M. G. Danaher, R. L. Young, and R. ,L. Rutledge, "A tlethod for the Location of Deuterium ..toms Attached to "yclohexane Rings," Chemistry and Industry (1954), 1294.

Curtin, David Y. and John 7. Crump, "Configurational Stabil-ities of Stereoisomeric Vinyllithium Compounds," Journal of the American Chemical Society, LXXX (1953), 1922.

Curtin, David Y. and Elbert E. Harris, "The Preparation of Stereoisomeric Alkenyl Lithium Compounds," Journal of the American Chemical Society, LXXIII (1951), 2716.

, "Preparation of Stereoisomeric Alkenyl-lithium Compounds. II. cis-and trans-1,2-Diphenylvinyllithium and o<- and y^-Styryllithiuin," Journal of the American Chemical Society, LXIII (1951), 45197"

Curtin, David Y., Harry Johnson, and Edwin G. Steiner, "Stereoisomeric Vinyllithium Compounds. III. Re-" actions with Aldehydes, Ketones, and Methyl Iodide," Journal of the American Chemical Society, LXXVII (1955),

Curtin, David Y. and William J. Koehl, "Effect of Solvent on the Steric Stability of Lithium Reagents," Journal of the American Chemical Jociety, LXXXIV (19627" 1967.

.» "Optically Active sec-Butyllithiura," Chemistry and Industry (i960), 262.

Dreiding, Andre S. and Richard J. Pratt, "The Carboxylation of cis- and trans-2-Butenyl-2-lithiun. A Stereospecific Synthesis of Angelic Acid," Journal of the American Chemical Society, LXXVI (1954)," 19027"'

Eliel, Ernest L. and Ralph G. Haber, "Conformational Analysis. V. The Reaction of cis- and trans-4-t-Butylcyclohexanol and trans-4-Methylcyclohexanol with Phosphorus Penta-bromide. Syntheses of Alkylcyclohexyl Bromides," Journal of Organic Chemistry, XXIV (1959)» 143.

121

Eschinazi, E. E. and Herman Pines, "Study in the Terpene Series. XXVI. Lisproportionation of d-Limonene in the Presence of Palladium Hydroxide-Barium Sulfate Catalvst," Journal of the American Chemical Society, IlXXVIII (lg^b;, 11757

Everson, W. L., "Determination of Butyllithium in Hydrocarbons "by Thermometric Titration with Butanol," Analytical Chemistry, XXXVI (1964), 854.

Fraenkel, Gideon, David G. Adams and James Williams, "Structure of Grignard Reagents and Organolithium Compounds," Tetrahedron Letters (1963), 767.

Gilman, Henry and Frank K. Cartledge, "The Analysis of Organolithium Compounds," Journal of Organometallic Chemistry, II (1964), 447. •

Gilman, Henry and A. U. Ilaubein, "Ethynyllithium compounds "by Halogen-Lietal Interconversions." Journal of the American Chemical Society, EXVII (19?5), 142(57

"The Quantitative Analysis of Alkyllithium Compounds," Journal of the American Chemical Society, LXVI (1944), 1515.

Gilman, Henry and R. H. Kirby, "The Relative Reactivities of Organolithium and Organomagnesium Compounds," Journal . of the American Chemical Society, LV (1933), 1265.

Gilman, Henry, Wright Langham, and Fred W. Moore, "Some Interconversion Reactions of Organolithium Compounds." Journal of the American Chemical Society, M i l (1940), 2327.

Gilman, Henry and Ivlyrl Lichtenwalter, "Relative Reactivities of Organometallic Compounds. X. Lithium and Magnesium," Recueil des Travaux Chemiques des Pays-bas, LV (1936), 561.

Glaze, William H., Jacob Lin, and E. G. Helton, "The Pyrol-ysis of Unsolvated Alkyllithium Compounds," Journal of Organic Chemistry, XXXI (1966), 2643.

Greene, Frederick D. s Ohin-Chium Chu, and Jasjit Walia, "Decomposition of Alkyl Hypochlorites. The 4-t-Butylcyclohexyl Radical," Journal of Organic Chemistryt XXIX (I964), 1285.

122

Hill, Alexander E., "Nuclear Magnetic Spectra of the Bornyl and Norbornyl Grignard Reagents," Journal of Organic Chemistry, XXXI (i960), 20.

Horner, von Leopold, Hermann Oediger, and Hellmut Hoffman, "Reactionen Mit Triphenyl-Phosphin-Dihalogeniden," Justus Liebigs Annalen der Chernie, DCXXVI (1959), 26.

Huckel, von Walter, Werner Tappe, and Gunter Legutke, "Abspaltungreaktionen und liar Sterischer Verlauf," Justus Liebigs Annalen der Chemie, DXXXXIII (1940), 191*

Jensen, Frederick R. and Kay L. Nakamaye, "Preparation of Geometrically Isomeric Grignard Reagents and the Stereo-chemical Courses of their Reactions," Journal of the American Chemical Society, LXXXVIII (1966), "345F.

Kamienski, Conrad and Donald L. Esmay, "Analysis and Stability of n-Butyllithiurn Solutions in n-Heptane," Journal of Organic Chemistry, XXV (I960), 115.

, "Effect of Sodium in the Preparation of Organolithium Compounds," Journal of Organic Chemistry, XXV (i960), 1807.

Koeppl, G. W., D. S. Sagatys, G. S. Kirshnamurthy, and Sidney I. Miller, "Inversion Barriers of Pyramidal (XY3) and Related Planar ( XY) Species," Journal of the . American Chemical Society, LXXXIX (I967), 5396.

Krieghoff, Nilde G. and Dwaine 0. Cowan, "Stereochemistry of the Grignard Reagents from exo- arid endo-Norbornyl Chlorides," Journal of the American Chemical Society, IiXXXVIII (196b) ,13227

Larnaudie, M. Marcel, "Isomerie des derive's monosubstitues des cyclohexane,"fComptes Rendus Hebdomadaires Des Seances De Ti'AcademieDes Sciences, CCXXXV (1952) 154-.

Lemieux, R. U., R. K. Kullnig, H. J. Bernstein, and W. G. Schneider, "Configurational Effects in the Proton Magnetic Resonance Spectra of Acetylated Carbohydrates," Journal of the American Chemical Society, LXXIX (1957)« 1005.

, "Configurational Effects on the Proton Magnetic Resonance Spectra of Six-Membered Ring Compounds, "Journal of the American Chemical Society, LXXX (1958), 6098..

123

Let singer, Ro"bert L., "Formation of Optically Active ?-Methylheptyllithium," Journal of the American Chemical Society, LXXII (1950), 45771 '

McNiven, Heal t,. and John Reed, "Researches in the Menthane Series. Part XVII. Configurations of Menthols and l.Ienthylamines," Journal of the Chemical Society (1952), ] 53.

Moore, James A. and Donald E. Reed, "Diazomethane," Organic Syntheses, XV (1953), 3.

Morton, Avery A. and Graham M. Richardson, "Condensation by Sodium. XVI. The Formation of Decane in the Wurtz Reaction," Journal of the American Chemical Society, LXII (1940), 123':

Morton, M. and 1. J. Fetters, "Homogeneous Anionic Polymer-ization V. Association Phenomena in Organolithium Polymerization," Journal of Polymer Science, A2 (1964), 33U.

Mulvaney, J. E. and Z. G. Gardlund, "Alkyllithium Compounds and Uonconjugated Olefins. The Nature of a Norbornyl-lithium Compound," Journal of Organic Chemistry, XXX (1965), 4321.

ITesmeyanov, A. H. and A. E. Borisov, "Stereochemistry of Electrophilic and Homolytic Substitution at the Olefinic Carbon," Tetrahedron, I (1957), 153.

ITesmeyanov, A« N.» A. E. Borisov, and II. A. Volkenau, "Synthesis of Stereoisomeric Organomercury Compounds from Organolithium Compounds," Izvestiya Akademii Nauk Soyuza Sovetskikh Sotsialisticheskikh' Respublik (1954)

Ott, Erv/in, "liber die Additionreaktionen der Athylen-Doppelbindung unter Bildung von Xthen-Verbingdungen mit zv/ei, durch den Additionsvorgong entstehenden, asyiruiietrischen Kohlenstoffatomen und der Einfluss der Reaktionsgeschwindigkeit auf den Reaktionsverlauf," Berichte der Deutschen Chemischen Gesellschaft, TjXI TO2S), 2T2T.

Pickard, Robert Howson and Joseph Kenyon, "VII - Investigations on the Dependence of Rotatory Power on Chemical Consti-tution. Part I. The Rotations of the Simplest Secondary Alcohols of the Fatty Series," Journal of the Chemical Society, LXXXXIX (1911), 45.

124

Porter, 0. W., "Raccmisation in the Preparation of the Grignard Reagent," Journal of the American Chemical Society, LVII (1935), H36.

Read, John, "Recent Progress in the Menthons Chemistry," Chemical Reviews, VII (1930), 1.

Sauvage, James Frederick, Robert H. Baker, and Allan S. Kussey, "The Kydrogenation of Cyclohexane over Platinum Oxide," Journal of the American Chemical Society, ixxxii (19 0), smo~.

Schorigin, P., "Uber bis Hatiumalkyle Ihre Reaktion mit der Atheru," Berichte Der Deutschen Chemischen GeselLschaft» XXXXIII (I9IO), 19517 *

Schwartz, A. II. and John R. Johnson, "Characterization of Alkyl Halides and Organomagnesium Halides," Journal of the American Chemical Society, 1III (1931), 1065.

Seyferth, Dietmar and Harvey M. Cohen, "The Stability of Cyclopropyllithium in Diethyl Ether and in Tetrahydro-furan," Journal of Organometallic Chemistry, I (1963), 15.

Seyferth, Dietmar and Lawrence G. Vaughan, "The Isomeric 1-Propenyllithium Reagents: Their KMR Spectra and the Stereochemistry of Their Formation by Direct Reaction and by the Transmetalation Reaction," Journal of Organometallic Chemistry, I (1963), 20TT

Smith, J. G. and George P. Wright, "The Diastereomeric Menthyl Chlorides Obtained Prom (-)Menthol, Journal of Organic Chemistry, XVII (1952), 1116.

Stolow, Robert D. and Charles B. Boyce, "Preparation of cis- and trans-4~t-Butyl-o(,«<-dimethylcyclohexanementhanol,n

Journal of Organic Chemistry, XXVI (19 61), 7426.

Tarbell, D. 3. and Marvin Weiss, "The Action of Lithium on an Optically Active Aliphatic Chloride," Journal of the American Chemical Society, LXI (1939), 1203.

Walborsky, H. M. and M. S. Aronoff, "Cyclopropanes XIX. Reaction of Lithium Metal with Optically Active Halides," Journal of Organometallic Chemistry. IV (1965), 418.

Walborsky, H. M. and P. J. Impastato, "Cyclopropanes. VI. Retention of Optical Activity and Configuration in the Cyclopropyl Oarbanion," Journal of the American Chemical Society. LXXXI (1959), 5&55~.

125

\7a I b o r S I' , h.. • • , ? • J « I».iptHS"Li3 SO , 3 il(. 1 . E • /OUQjj j^ 11 Gj d o — "oropa i ies . X V . T h e O p t i c a l S t a b i l i t y ox 1— r!s ohy J.— 2 , 2 - d I p i i e n y l - c y c " * o p r o p y i l i t n i c u , " J o u r n a 7 o f t h e A m e r i c a n gi' iei-i ica]. S o c i e t y , LXXXV1 ( I ' joA- ) t~~y2'o2 •

V-fa"11 i s . E v e r e t t 3 . a n d ? r e d e r i c I I . Ada:. iS, " i ' l io S p a t i a l ;on f i g u r a t i o n o f the V a l e n c e s i n S r i o o v a l e n t C a r b o n

OoiJipoundsi" J o u r n a l o f t h e A m e r i c a n G h e r . i c a l S o c i e t y , LV ( 1 9 3 3 ) , 3 6 3 6 .

E i n s t e i n , S . , 75. G l i p p i n g e r , R o b e r t i l o u e , a n d E l l i o t V o g e l - ^ f a n g e r . " T h e H o n c l a s s i o a l I T o r b o r n y l C a t i o n , " J o u r n a l o f t h e A n e r i c a n C h e m i c a l S o c i e t . y , L X X X V I I ( 1 9 6 5 ) 1 3 7 b .

W i t t i g , G e o r g e , F e d e r i c o V i d a l , a n d E r v / i n B o h n e r t , " i f b e r d i e k o n f i g u r a t i v e B c s t a nd i g k e i t rue t a i l o r ga n i s c h e r V e r b i n d u n g e n , " B e r i c h t e d e r D e u t s c h e n G h e m i s c h e n G e s e l l s c n a i t , I i X X X I I I (~1 "J > 0 ) "j 3 5 9 *

W r i g h t , G e o r g e 3?., " O r g a n o u e t a l l i c Compounds o f S t y r e n e , " J o u r n a l o f O r g a n i c C h e m i s t r y , I ( 1 9 3 6 ) , 4 5 7 .

Z i e g l e r , k . a n d I I . C o l o n i u s , " U n t e r s u c h u n g e n u b e r a l ^ a l i -o r g a n i s c h e V e r b i n d u n g e n . V . E i n e bequeme S y n t h e s e e i n f a c h e r L i t h i u m a l k y l e J u s t u s L i e b i g s A n n a l e n d e r G h e m i e , GCGGLXXIX ( 1 9 3 0 ) , T W .

Z i e g l e r , v o n k a r l a n d I I . G . G e l l e r t , " X V I I . R e a k t i o n e n

Z w i s c h e n L i t h i u m - a l k y l e n und A t h e r n , " J u s t u s I d e b i g s A n n a l e n d e r C h e u i i e , D I t X V I I ( 1 9 5 0 ) , 1 8 5 .

U n p u b l i s h e d M a t e r i a l s

V f i l l i a m H . G l a z e , u n p u b l i s h e d n o t e s , D e p a r t m e n t o f C h e m i s t r y , K o r t h T e x a s S t a t e U n i v e r s i t y , B e n t o n , T e x a s , 1 9 6 7 .