Studies on cyclohexane derivatives

STUDIES ON

CYCLOHEXANE

DERIVATIVES

H. VAN BEKKUM

19 59

2 1

Studies on cyclohexane derivatives

Studies on

cyclohexane derivatives

Proefschrift

ter verkrijging van de graad van doctor in de technische wetenschappen aan de Technische Hogeschool Delft, op gezag van de rector magnificus ir. H.R. van Nauta Lemke, hoogleraar in de Afdeling der elektro-techniek, voor een commissie uit de senaat te verdedigen op woensdag 27 januari 1971 te 16.00 uur door

Herman van Bekkum

scheikundig ingenieur geboren te Rotterdam

/f^^?

sx / t'

1970

Universitaire Pers Rotterdam

Wolters-Noordhoff, Groningen

Dit proefschrift is goedgekeurd door de promotor PROF. DR IR B.M. WEPSTER

C O N T E N T S

INTRODUCTION 7 I. PREPARATIVE METHODS 9

Catalytic hydrogenation of benzene and cyclohexene

derivatives 9 Reduction and reductive alkylation of benzoic acid and

alkylbenzoic acids with lithium in liquid ammonia 14 E s t e r hydrolysis using concentrated sulfuric acid 17 Separation of i s o m e r i c cyclohexane d e r i v a t i v e s by

selective inclusion into thiourea 19

Miscellaneous 23 Epimerization techniques 23 Koch c^^rboxylation 24 Diels-Alder addition 25 Cyclohexane derivatives a s s t a r t i n g m a t e r i a l s 25 n . CONFIGURATIONAL ANALYSIS 27 Chemical proofs of s t r u c t u r e 27 Equilibration 28 Length determination in thiourea 28

NMR s p e c t r a 30 Mass s p e c t r a 30 GLC retention t i m e s 31 Density and refractive index 31

m . CONFORMATIONAL ANALYSIS. pK*-MEASUREMENTS 33 1,4Ditbutylcyclohexane and 1,3, 5 t r i t b u t y l c y c l o

-hexane 33 p K J - v a l u e s of 4 - and 3-alky 1-substituted

cyclohexane-carboxylic acids 34 2-Alkylcyclohexanecarboxylic acids and model compounds 35

Trimethylsilylcyclohexanecarboxylic acids 38 1-Alkylcyclohexanecarboxylic a c i d s . P r e d i c t i v e p K J - r u l e s 39 Alkyl-substituted cyclohexaneacetic a c i d s 42 2, 5-Cyclohexadiene-l-carboxylic acids 43 SAMENVATTING 47 REFERENCES 49

I N T R O D U C T I O N

This t h e s i s is based mainly on investigations on cyclohexane, c y c l o -hexene and cyclohexadiene derivatives the r e s u l t s of which have been published'''"* or a r e in the p r e s s 7 ' ^ Another seven p a p e r s ^ "•'•^ in which the p r e s e n t author participated is a l s o connected closely to the subjects to be d i s c u s s e d .

The p r e p a r a t i o n and p r o p e r t i e s of cyclohexane s y s t e m s with carboxyl a n d / o r alkyl substitution was a c e n t r a l theme in the investigations. Attention was drawn to, amongst o t h e r s ,

(i) the dependence of the ring conformation, i . e . the r e l a t i v e stability of both chair conforniations and, in some c a s e s , also of non-chair f o r m s , on the ring substitution applied;^'^'®

(ii) the influence of alkyl substituents (at the 1 , 2 , 3 and 4 -position) on the acidity of an equatorial or axial 1-carboxylgroup; » (iii) the assignment of configuration to the compounds studied; besides s o m e chemical proofs of configuration ' ^ ° ~ ^ s e v e r a l other r e l i a b l e configuration c r i t e r i a have been developed; ^'^'^ (iv) the efficiency, the selectivity and the mechanism of some of the p r e p a r a t i v e p r o c e d u r e s applied. ^~*

The cyclohexane and 1,4-cyclohexadiene d e r i v a t i v e s entered the investigations through the last-mentioned s t u d i e s . Cyclohexenes w e r e found t o play a r o l e a s i n t e r m e d i a t e s in s e v e r a l catalytic hydrogenations of aromatic compounds, ^ ' ^ w h e r e a s 2, 5 c y c l o -hexadiene-1-carboxylic acids ^ a r e intermediate products in a newly developed synthesis of 1-alkyl-substituted cyclohexanecarboxylic a c i d s .

In the p r e s e n t simimary the following o r d e r will be applied: I P r e p a r a t i v e Methods

n Assignment of Configuration

i n Conformational Analysis; p K j - m e a s u r e m e n t s .

I. P R E P A R A T I V E M E T H O D S

1. C a t a l y t i c h y d r o g e n a t i o n of b e n z e n e a n d c y c l o -h e x e n e d e r i v a t i v e s

A very useful method for p r e p a r i n g cyclohexane derivatives is the heterogeneously catalyzed hydrogenation of the corresponding aromatic compounds. ^^ .^'^ Starting out from d i - or trisubstituted benzenes, a mixture of s t e r e o i s o m e r s is generally obtained showing that the p r o c e s s involves m o r e than just a l l - c i s addition of six hydrogen atoms from the surface of the catalyst to the adsorbed side of the r i n g .

To account for the contribution of two-side addition of hydrogen the following m e c h a n i s m s have been suggested (i) ' r o l l o v e r ' of a d sorbed intermediate cyclohexenes^ '^® (ii) s e v e r a l ways of t o p -side addition of hydrogen to adsorbed unsaturated s y s t e m s , 2° ,21 and (iii) desorption of intermediate cyclohexenes followed by r e a d s o r p t i o n with the other side of the r i n g . ^^ > ^^

The reality of mechanism (iii) has been a s c e r t a i n e d experimentally by us in the liquid-phase hydrogenation of some 2-alkylbenzoic acids and methyl 2-alkylbenzoates over rhodium and ruthenium c a t a l y s t s . Scheme I gives (in a simplified formulation) r o u t e s towards c i s -and trans-2-alkylcyclohexanecarboxylic acid [ 5 ] on the b a s i s of cis-addition of hydrogen during each r e s i d e n c e of the s u b s t r a t e molecules on the catalyst s u r f a c e . T r a n s - i s o m e r is formed then through the cyclohexenes [ 3 ] and [ 4 ] . It m u s t be borne in mind, however, that t r a n s i t i o n s between the cyclohexenes [ 2 ] - [ 4 ] m a y occur by double bond migration on the catalytic surface.

The c o u r s e of the hydrogenation of [ l ] (Ri = t - B u , Rg = H) over rhodium at room t e m p e r a t u r e and a t m o s p h e r i c p r e s s u r e is r e l a -tively c l e a r as the a r o m a t i c acid [ l ] is found to be hydrogenated selectively with r e s p e c t to desorbed intermediate cyclohexenes. About 75% of [ 1 ] adopts six hydrogen atoms during one r e s i d e n c e o n t h e s u r f a c e y i e l d i n g c i s - [ 5 ] , 23% d e s o r b s a s cyclohexene [ 4 ]

yielding a 1 : 1 m i x t u r e of c i s - and t r a n s - [ 5 ] upon readsorption and hydrogenation. The cyclohexenes [ 2 ] and [ 3 ] w e r e detected but play minor r o l e s (< 1%) in the perceptible hydrogenation p r o c e s s . The graph of the reaction c o u r s e (Fig. 1) r e v e a l s at once that for this c a s e desorbing cyclohexene [ 4 ] is by far the

main s o u r c e of the t r a n s - [ 5 ] formed.

An analogous c o u r s e is observed for the hydrogenation of the c o r -responding methyl e s t e r ( [ l ] , R i = t-Bu, Rs = Me) over rhodium

a'

a:

Fig. 1. Hydrogenation of 2-t-butylbenzoic acid (1.8 g) over 0.3 g of 5% rhodium on carbon in ethyl alcohol (60 ml) at 22° and 1 atm. • , 2-t-butylbenzoic acid; • , 6-t-butyl-l-cyclohexene-l-carboxylicacid; X; cis-2-t-butylcyclohexanecarboxylic acid; A , trans-2-t-butyl-cyclohexanecarboxylic acid.

under the above conditions. The s t e r e o c h e m i s t r y of the hydrogena-tion of the intermediate cyclohexene [ 4 ] has been altered however: c i s - a n d t r a n s - [ 5 ] a r e f o r m e d f r o m [ 4 ] in a 4 . 5 : 1 r a t i o .

In rhodium-catalyzed hydrogenations of compounds [ l ] with s m a l l e r size of R l ( R i = Me, Et, i P r ; R s = H, Me) cyclohexene [ 2 ] a s well a s cyclohexene [ 4 ] have been observed a s i n t e r m e d i a t e s . ^* The maximum concentration of [ 4 ] in the r e a c t i n g m i x t u r e is s u b s t a n -tially lower than for R i = t-Bu but the total quantity of [ 4 ] desorbed from the surface is of the s a m e o r d e r of magnitude. Thus, with d e c r e a s i n g size of the alkyl substituent the e a s e of hydrogenation of [ 4 ] in the p r e s e n c e of [ l ] (governed by adsorption and surface reaction r a t e of [ 4 ] r e l a t i v e to [ l ] ) i n c r e a s e s . This is c o m p r e h e n -sible. Cyclohexene [ 2 ] attains a concentration of 5-6% in the above-mentioned hydrogenations. The maximum concentration is observed near the end of the reaction, consequently the total amount of [ 2 ] desorbed is only slightly m o r e . The low reactivity of [ 2 ] with r e s p e c t to [ 4 ] will be caused by the i n c r e a s e d substitution of the double bond. The very s m a l l r o l e of [ 2 ] in the hydrogenation of [ l ] (Rl = t-Bu) may be taken a s an indication that the addition of hydrogen chiefly s t a r t s at the carbon atom c a r r y i n g the t - B u g r o u p . ^ ^

When the a r o m a t i c carboxyl group is flanked by two methyl groups (2,6-dimethylbenzoic acid), the hydrogenation over rhodium is closely r e l a t e d to that of [ l ] (Ri = t B u ) . H e r e , 2 , 6 d i m e t h y l -1-cyclohexene-1-carboxylic acid d e s o r b s from the surface and a p p e a r s to be the main s o u r c e of the c 2 , t 6 d i m e t h y l r l c y c l o -hexanecarboxylic acid formed.

Data on the hydrogenation of [ l ] (Ri = t-Bu, Rs = H, Me) over rutheniimi, platinum and palladium a r e given in ref. 1. As to the manifestation of intermediate cyclohexenes, i . e . the maximvim concentration attained in the reacting m i x t u r e , the o r d e r of catalyst m e t a l s is Rh > Ru > P t , Pd. The total quantity of cyclohexene [ 4 ] desorbed is over ruthenium about the s a m e a s over rhodium but much l e s s over platinum. The s t e r e o c h e m i c a l r e s u l t s ( i . e . the c i s / t r a n s ratios) obtained f or the hydrogenations over palladium indicate that either desorption of alkenes takes place on a l a r g e scale (combined with relatively fast readsorption and reaction) or p r o c e s s (i) or (ii) plays an important r o l e . In view of the s e p a r a t e position of palladium among the other transition m e t a l s in exchange ^ ^ and hydrogenation p r o c e s s e s the last mentioned possibility cannot be excluded.

It may be mentioned that the hydrogenation of 1, 2-di-t-butylbenzene over rhodium, studied by van de Graaf, ^^ constitutes an excellent example of the appearance of two step next to one step ring s a t u r a -tion. As to this aspect this r e a c t i o n is closely connected to the hydrogenation of [ l ] ( R i = t - B u ) .

The two step p r o c e s s is predominating in the hydrogenation^ of 1 , 3 , 5 - t r i - t - b u t y l b e n z e n e [ 6 ] over rhodium, platinvun, and palladium. The g r e a t e r p a r t of the reacting molecules desorbs a s c i s 1 , 3 , 5 -tri-t-butylcyclohexene [ 7 ] after addition of four hydrogen atoms from the metal s u r f a c e . This follows from the c o u r s e of these hydrogenations (ref. 2, F i g . 1-3) a s well a s from the observation that s e p a r a t e hydrogenations of [ 6 ] and [ 7 ] yield the s a m e r a t i o of c i s , c i s - and c i s , t r a n s - l , 3 , 5 - t r i - t - b u t y l c y c l o h e x a n e (see Table

I)-Table I. Product composition in hydrogenations" of 1,3, 5 - t r i - t - b u t y l benzene and 1,3, 5-tri-t-butylcyclohexenes

Substrate Catalyst Products in %

c i s , cis c i s , t r a n s 1 , 3 , 5 - t r l - t - b u t y l b e n z e n e [ 6 ] Rh/C 84 16 1 , 3 , 5 - t r i - t - b u t y l b e n z e n e P t / C 93 7 1 , 3 , 5 - t r i - t - b u t y l b e n z e n e P d / C 98 2 c i s - l , 3 , 5 - t r i - t - b u t y l c y c l o h e x e n e [ 7 ] Rh/C 84 16 c i s - l , 3 , 5 - t r i - t - b u t y l c y c l o h e x e n e P t / C 93 7 c i s - l , 3 , 5 - t r i - t - b u t y l c y c l o h e x e n e P d / C 98 2 c i s - l , 3 , 5 - t r i - t - b u t y l c y c l o h e x e n e *> 98 2 t r a n s - l , 3 , 5 - t r i - t - b u t y l c y c l o h e x e n e [ 8 ] P t / C - 100 t r a n s - l , 3 , 5 - t r i - t - b u t y l c y c l o h e x e n e P d / C - 100 t r a n s - 1 , 3 , 5 - t r i - t - b u t y l c y c l o h e x e n e '' - 100 a. In n-heptane at 25° and atmospheric p r e s s u r e .

b . Reductions with diimide (formed in situ from p-toluenesulfonyl-hydrazine) for 24 hours at 1 0 0 ° .

Some insight in the appearance of [ 7 ] in the hydrogenation of [ 6 ] over platinum and palladium was gained^ by m e a s u r i n g the r a t e s of hydrogenation of [ 6 ] , [ 7 ] and some rtelated compounds and by c a r r y i n g out competition e x p e r i m e n t s . Deuterations of [ 6 ] and [ 7 ] otherwise showed that the addition p r o c e s s e s a r e accompanied by complex exchange phenomena in which in this c a s e also the t-Bu groups a r e involved. ^

It was interesting to note that in the hydrogenation of [ 7 ] over 12

platinum and palladium some dehydrogenation o c c u r s yielding [ 6 ] . This was not observed in the hydrogenation of t r a n s 1 , 3 , 5 t r i t -butylcyclohexene [ 8 ] . T h e r e f o r e , it may be concluded that with

[ 6 ] and [ 7 ] addition a s well a s elimination of hydrogen take place by c i s p r o c e s s e s .

The hydrogenation of [ 6 ] and [ 7 ] over ruthenixim has been i n v e s t i -gated by Hartog and Weterings. ^"^ Combination with the p r e s e n t work yields the following o r d e r of m e t a l s a s to the appearance of

[ 7 ] during the hydrogenation of [ 6 ] : Rh > Ru > Pt > Pd.

Other e x a m p l e s ^ * i n which the two step p r o c e s s plays an important r o l e in the catalyc hydrogenation of the a r o m a t i c nucleus include the rhodium-catalyzed hydrogenation of phthalic acid, dialkyl phthalates, and 1,4-di-t-butylbenzene [ 9 ] .

Compound [ 9 ] is distinghuished from s e v e r a l 1,4-dialkylbenzenes studied^*'®^ '^^ in yielding a relatively high percentage of c i s 1 , 4 -di-t-butylcyclohexane (cis-[lO]) when hydrogenated over platinum, rhodium and palladium at 25°. Hydrogenation of 1,4-di-t-butylcyclo-hexene [ l l ] yields c i s - and t r a n s - [ l O ] in s i m i l a r high r a t i o s . In p a r t i c u l a r the s t e r e o c h e m i s t r y of hydrogenation of [ 9 ] and [ l l ] over palladium is noteworthy a s 1,4-dialkylbenzenes and 1,4dialkylcyclohexenes generally yield the m o r e stable t r a n s 1 , 4 d i a l k y l -cyclohexane in e x c e s s when hydrogenated over palladium. °° > ^^ The s t e r e o c h e m i c a l r e s u l t s of the hydrogenation of disubstituted cyclohexenes ( i . e . the c i s / t r a n s r a t i o of the cyclohexanes obtained) cannot be r a t i o n a l i s e d in a simple way. F o r example the question whether adsorption/desorption equilibria a r e fast compared to the surface r e a c t i o n s and the problem which r e a c t i o n step is r a t e -controUing under a given s e t of conditions a r e still in discussion. When accepting the formation of the half-hydrogenated state a s the r a t e and productdetermining step for hydrogenations at a t m o s -pheric p r e s s u r e one has to consider two adsorbed s t a t e s each connected with two t r a n s i t i o n s t a t e s of addition of the f i r s t hydrogen atom. As to the geometry of the above s t a t e s , bond eclipsing around the c a r b o n a t o m s of the original double bond i s p r e f e r r e d ; a s a c o n -sequence the ring is in a boat conformation.^^'^^'^^

This model p r e d i c t s satisfactorily s e v e r a l s t e r e o c h e m i c a l t r e n d s observed in the hydrogenation of 1,4 and 2, 3substituted c y c l o hexenes, •""' ^® ' ^ ^ such a s the high p e r c e n t a g e of c i s i s o m e r o b -tained in the hydrogenation (Pt, Rh) of 1-t-butyl-4-X-cyclohexenes and, on the other hand, the predominant formation of t r a n s - i s o m e r in the hydrogenation (Rh, Ru, Pd, Pt) of 2 t b u t y l 3 X c y c l o -hexenes . Application of the above approach t o the hydrogenation of 1 , 3 and 2 , 4 s u b s t i t u t e d cyclohexenes, however, is l e s s s u c c e s s -ful. ==

Liquidphase hydrogenations at atmospheric p r e s s u r e in the t e m -13

p e r a t u r e range 0-90° show that with increasing t e m p e r a t u r e the amount of the m o r e stable cyclohexane i s o m e r generally i n c r e a s e s . Pronounced effects a r e observed for example in the hydrogenation of l - t - b u t y l - 4 - X - c y c l o h e x e n e s over platinum and rhodium. ^^ The effect of t e m p e r a t u r e is much l e s s when working at high p r e s -s u r e (200 atm H g ) . Thi-s i-s in accordance with the propo-sal^® that under such conditions the r a t e s of adsorption (for the two adsorbed states) a r e product-controlling.

Double bond migration phenomena on the catalyst surface have been studied by us by c o m p a r i i ^ the r e s u l t s of hydrogenation of 4 t b u t y l -1-hydroxymethylcyclohexene and l-t-butyl-4-hydroxymethylcyclo-hexene. In the f o r m e r s u b s t r a t e t h e C - O bond is at an allyl position and extensive hydrogenolysis o c c u r s ; in the other s u b s t r a t e double bond migration is r e q u i r e d to satisfy hydrogenolysis conditions. The o r d e r of migration activity obtained from these experiments is Pd > Rh > P t . F u r t h e r m o r e it is found that i n c r e a s e of t e m p e r -a t u r e i n c r e -a s e s the -amount of migr-ation.

Finally, it may be r e m a r k e d that 'trans-addition' of hydrogen (mechanism (ii)) could c r o s s s e v e r a l of the above considerations. In s e a r c h of the existence of this mechanism we have studied the heterogeneous hydrogenation of hexamethyl -Dewar-benzene

(hexamethylbicyclo [ 2 . 2 . 0]hexa-2, 5-diene).^* Unfortunately, the information obtained did not suffice for a decisive answer about mechanism (ii). At p r e s e n t we consider this mechanism l e s s p r o b -able and not intended for a major r o l e in heterogeneously catalyzed hydrogenations.

2. R e d u c t i o n a n d r e d u c t i v e a l k y l a t i o n of b e n z o i c a c i d a n d a l k y l b e n z o i c a c i d s w i t h l i t h i u m i n l i q u i d a m m o n i a

The p a r t i a l reduction of the benzene nucleus with solutions of alkali m e t a l s in liquid ammonia in the p r e s e n c e of a suitable proton donor (mostly ethanol) i s a wellknown p r o c e d u r e . ^ « ^ ^ Yet, r e l -atively few data a r e available for the m e t a l - a m m o n i a reductions of benzoic acid and alkylbenzoic acids, 3''-*° which generally a r e formulated a s

c o o " COO" c o o

-12 13

LIBRARY OF THE DELFT UNIVERSIW OF TECHNOLOGY

m

Please return this publication not later than the date printed overleaf to the above-mentioned address You are also allowed to send it back by post, properly wrapped up and sufficiently stamped.

You are urgently requested to return the publication not later than the date of expiry. If no reservations for the publication are outstanding, renewal of the loan period may be requested not later than the date of expiry.

If the publication Is not returned you will, much to our regret, be charged for the replacement value of the publication and an additional amount to cover administration and acquisition expenses

Bibliotheek Technische Hogeschool 0

Doelenstraat 101 2611 NS Delft, Nederland Postbus 98 2600 MG Delft, Nederland Afdeling Uitlening telefoon 015-785679

Gelieve dit werk uiterlijk op:

aan bovenstaand adres terug te bezorgen of goed verpakt en gefrankeerd terug te zenden.

U wordt dringend verzocht u zorgvuldig te houden aan de vervaldatum. Indien het werk inmiddels niet is gereser-veerd, kan uiterlijk op de dag van het verstrijken van de uitleentermijn verlenging worden verleend.

Indien u het werk niet terugbezorgt zijn wij helaas genoodzaakt u een rekening te sturen voor de vervangingswaarde van het werk vermeerderd met administratie- en venwervingskosten.

We have studied ^ this r e a c t i o n and the r e l a t e d reductive alkylation for p r e p a r a t i v e a s well a s for mechanistic r e a s o n s :

(i) It will be c l e a r that lithium-ammonia reduction followed by catalyc hydrogenation is a two step alternative for the d i r e c t hydrogenation of alkylbenzoic a c i d s . P a r t i a l hydrogenation of compounds [13] is a potential route to alkylsubstituted 2 c y c l o -hexene-1-carboxylic acids, w h e r e a s complete saturation yields alkylcyclohexanecarboxylic a c i d s .

(ii) One may wonder whether a proton donor i s a necessity for o b taining s p e c i e s [12] and [ 1 3 ] or whether ammonia itself can p r o t o -nate the negatively charged a r o m a t i c nucleus. Species [12] was considered p a r t i c u l a r l y important in view of the p r e p a r a t i o n (by alkylation of [12]) of 1alkylsubstituted 2, 5 c y c l o h e x a d i e n e l -carboxylic acids [ 1 4 ] . Hydrogenation of [14] would afford 1-alkyl-cyclohexanecarboxylic a c i d s , thus providing an alternative for the Koch synthesis (see 1.5.2).

(iii) Experiments with 4 and 3deuterobenzoic acid have been i n -cluded to provide information r e g a r d i n g the r e v e r s i b i l i t y of the protonation of the nucleus.

(iv) Ammonia, ammonia-ethanol, and a m m o n i a - w a t e r have been compared a s the media with special r e f e r e n c e to double bond migration phenomena in [ 1 3 ] .

The r e s u l t s of numerous lithium-ammonia reductions and reductive alkylations of benzoic acid and alkylbenzoic acids (Tables I and II in ref. 8) have led to the following picture for t h e s e r e a c t i o n s . When working in liquid ammonia without added proton donor (see Scheme II) the aromatic acid accepts two e l e c t r o n s and a proton yielding the dianion [ 1 2 ] . In [12] the e l e c t r o n accepting carboxylate group is situated at the most negative position*^ of the c y c l o

-hexadienyl anion. We a s s u m e that a trianion is r e q u i r e d to a b s t r a c t a proton from ammonia. *^ Experiments with 4-deuterobenzoic acid (Table IV in ref. 8) showed that this proton addition is i r r e -v e r s i b l e .

Scheme I

The dianion [12] can be protonated by e . g . ammonium chloride to yield the unconjugated system [13] in a kinetically controlled reaction, a s has been d i s c u s s e d ^ ^ ' ^ ^ f o r other b e n z e n o i d s u b s t r a t e s . When alkyl halides w e r e added to solutions of [12] in ammonia essentially quantitative alkylation at the 1-position was observed yielding [ 1 4 ] . A convenient method of preparing acids [14] was found to consist of adding lithium to a solution or suspension of the alkylbenzoic acid in ammonia until permanent blue colour and then adding excess of alkyl halide.

Under c l a s s i c a l conditions, i . e . in the p r e s e n c e of ethanol. Scheme m applies. Anion [12] is supposed to undergo almost complete protonation to yield [ 1 3 ] . With R = H and R = 3 - or 4-alkyl isomerization of [ l 3 ] i n t o [15] may occur, the extent of which was found to depend on the conditions applied. Longer r e s i d e n c e t i m e s in the ammonia-ethanol-ethoxide medium a s well a s higher t e m p e r a t u r e s promote isomerization of [13] into the conjugated s y s t e m [ 1 5 ] . The conservation of m o s t of the labeling in a sample of 1, 5cyclohexadiene1carboxylic acid obtained from 3 d e u t e r o -benzoic acid proved the slowness of the r e v e r s e reaction ( [ 1 5 ] -• [ 1 3 ] ) .

,000"

COO" 0 0 0 " COO"

Scheme I I

No isomerization [ 1 3 ] [ 1 5 ] was observed in the c a s e of R = 2 -alkyl. Taking the extent of isomerization - under given conditions a s a m e a s u r e of the basicity of the dianions involved (in other words of the equilibrium concentration of [13]) the following sequence is obtained.

This sequence has been rationalised on the basis of electronic *^ and s t e r i c effects.

When using water a s the proton donor, the isomerization [ 1 3 ] - [ 1 5 ] was not observed. Obviously, in the p r e s e n c e of water a much lower concentration of [13] is allowed than in the p r e s e n c e of the l e s s acidic ethanol. Reduction with the lithium-ammonia-water system was found to be the method of choice when 3 and 4 a l k y l -benzoic acids a r e to be reduced to the 1,4-dihydro p r o d u c t s . Experiments in which ammonium chloride was p r e s e n t d u r i n g the reduction revealed extensive reduction of the carboxyl group.®'^* Thus, 4-methylbenzoic acid yielded a mixture of 4-methylbenzalde-hyde, 4-methylbenzylalcohol, p-xylene, and 4 - m e t h y l - 2 , 5-eyelo-hexadiene-1-carboxylic acid. Under these c i r c u m s t a n c e s some undissociated carboxyl groups might be involved and attacked. Some thirty new alkyl-substituted 2, 5-cyclohexadiene-1-carboxylic acids w e r e p r e p a r e d in the c o u r s e of this study. Some of the acids together with t h e i r p K * ' s a r e listed in Table XII. F o r p r e p a r a t i v e and other details we r e f e r to ref. 8.

Reduction of a r o m a t i c s y s t e m s with lithium and e t h y l a m i n e * ' was applied for the synthesis^ of 1,4ditbutylcyclohexene and c i s -and t r a n s - 1 , 3 , 5-tri-t-butylcyclohexene ([7] -and [ S ] ) . In the lithium-ethylamine reduction of 1 , 3 , 5 - t r i - t - b u t y l b e n z e n e a 2.4:1 mixture of [ 7 ] and [ s ] was obtained which was separated by p r e p a r a t i v e GLC. As to the mechanism of this reduction a t h r e e step c o u r s e can be envisaged: reduction to a 1,4-cyclohexadiene derivative, isomerization to a conjugated s y s t e m , and again reduction.

Upon reduction of 2-methylbenzoic acid with lithivtm and ethylamine, the ' n o r m a l ' dihydro product, 2 m e t h y l 2 , 5 c y c l o h e x a d i e n e l -carboxylic acid, was obtained. Apparently, h e r e the isomerization r e q u i r e d for continued reduction does not occur. Reduction of 4 -t-butylbenzoic acid in this way also gave mainly dihydr o p r o d u c t s . In both experiments the conversion of the aromatic acid was far from complete, in spite of the u s e of^a large excess of lithium. 3 . E s t e r h y d r o l y s i s u s i n g c o n c e n t r a t e d s u l f u r i c a c i d °

F r o m m i x t u r e s of i s o m e r i c alkyl-substituted cyclohexanecarboxylic a c i d s , a s obtained by hydrogenation of unsaturated p r e c u r s o r s , g e n e r ally only one of the i s o m e r s can be obtained in a pure state ^ '^ by r e c r y s -tallization of the m i x t u r e of acids or of easily accessible d e r i v a t i v e s , e . g . benzylamine s a l t s . Additional i s o m e r s can be obtained by converting m i x t u r e s of acids (often mother liquors of the above r e -crystallizations) into mixtures of methyl e s t e r s and subjecting these to p r e p a r a t i v e GLC. Alternatively, m i x t u r e s of i s o m e r i c methyl

e s t e r s may be obtained by catalytic hydrogenation of methyl a l k y l -benzoates. •""

F o r the hydrolysis of samples of methyl e s t e r s a s isolated by p r e p a r a t i v e GLC we wished to have at our disposal a simple, rapid and non-epimerizing technique. It was found ^ that a short t r e a t m e n t with concentrated sulfuric acid at room t e m p e r a t u r e fulfils these requirements*® and provides a preparatively conven-ient method of hydrolysis.

Apparently the AAC 1 mechanism*^ is involved with the equilibria (see Scheme IV) in favour of the protonated acid due to the e x c e s s of w a t e r . / / H ffi ffi PC 9 : ; = = : RCO + MeOH2 OMe RCO® + H,0 : ; = = : RCOOH,® Scheme H

Some kinetic r e s u l t s for the hydrolysis of e s t e r s in concentrated sulfuric acid a r e shown in Table II (for additional data s e e Table I in ref. 3).

Table n . E s t e r Hydrolysis Rates in 95% Sulfuric Acid at 25°

F i r s t o r d e r r a t e ^^^^^ constant ( s ' ^ x l O * ) Methyl trans-4-t-butylcyclohexanecarboxylate 17.0 Methyl cis-4-t-butylcyclohexanecarboxylate 94 Methyl l , t - 4 - d i m e t h y l - r - l - c y c l o h e x a n e c a r b o x y l a t e 55 Methyl l , c - 4 - d i m e t h y l - r - l - c y c l o h e x a n e c a r b o x y l a t e 179 Methyl nonanoate 9. 2 Methyl 2-n-propylpentanoate 18.2 Methyl benzoate 0.65 Methyl 4-methylbenzoate 2 . 4 1 Methyl 3-methylbenzoate 1.51 Methyl 2-methylbenzoate 182 Methyl 3-bromobenzoate 0.032 Methyl 3-chlorobenzoate 0.035 Methyl 4-methoxybenzoate 10.2 Methyl mesitoate > 10 ^ 18

s t e r i c effects in the hydrolysis in sulfuric acid can be understood when it is accepted that the t r a n s i t i o n state of hydrolysis is l e s s s p a c e - d e m a n d i i ^ than the protonated e s t e r ; any van d e r Waals s t r a i n in the l a t t e r will then i n c r e a s e the r a t e of hydrolysis. Two p a i r s of e p i m e r i c methyl alkylcyclohexanecarboxylates provide examples of t h i s : the compounds with an axial methoxycarbonyl group a r e hydrolyzed distinctly f a s t e r than their i s o m e r s with an equatorial e s t e r group, obviously a s a consequence of r e l e a s e of s t e r i c s t r a i n during the reaction. The r e v e r s e i s known to be t r u e in hydrolysis in dilute acid^° and in alkaline hydrolysisf^»^^ F u r t h e r m o r e it may be noted that methyl t r a n s 4 t b u t y l c y c l o -hexanecarboxylate is hydrolyzed in sulfuric acid at the s a m e r a t e a s its acyclic analogue methyl 2-n-propylpentanoate.

A m o r e pronounced s t e r i c effect is observed for methyl 2 m e t h y l -benzoate which compound hydrolyzes 75 t i m e s f a s t e r than methyl 4-methylbenzoate.

As to electronic effects, a Hammett r h o value^^' ^^ of - 3.6 is calculated using r a t e constants of methyl benzoate and 3 s u b s t i -tuted methyl benzoates - showing that A^^l hydrolysis is m o r e sensitive towards substituent influences than the A AC2 hydrolysis in m o r e dilute acid. ^ The A^c 1 splitting may proceed in two ways,^ in either c a s e additional positive charge has been developed on the carboxyl carbon in the transition s t a t e . This explains the high rho value observed. F u r t h e r m o r e , a YukawaTsuno resonance p a r a -meter^^ of 0.35 is derived from the reaction r a t e constant of methyl 4-methoxybenzoate.

Finally, it may be noted that methyl m e s i t o a t e , possessing t h r e e activating methyl groups, is found to be hydrolyzed very fast in 95 a s well a s in 90% sulfuric acid, demonstrating that use of 100% sulfuric acid^® > ^® is not n e c e s s a r y .

4 . S e p a r a t i o n of i s o m e r i c c y c l o h e x a n e d e r i v a t i v e s by s e l e c t i v e i n c l u s i o n i n t o t h i o u r e a

We have found* that trans-4-alkylcyclohexanecarboxylic acids form adducts (inclusion compounds) with thiourea^ ° whereas the c o r r e sponding c i s i s o m e r s generally do not. This fact enabled s u c c e s s -ful separation of s e v e r a l p a i r s of epimeric 4-alkylcyclohexane-carboxylic a c i d s . In addition to the acids r e p o r t e d in ref. 4 also 4 - t - p e n t y l - ^ , 4-cyclohexyl-^ '^^, and 4-trimethylsilylcyclohexane-carboxylic acid^^ could be separated conveniently i n t h i s way into the c i s - and the t r a n s - i s o m e r .

According a s the dimensions of the alkyl group approach the inner dimensions of the thiourea channel m o r e stable adducts a r e o b -tained (probably a s a r e s u l t of overall i n c r e a s e of attractive van der

Waals forces between the guest molecules and the walls of the host). Thus, the solubilities of trans-4-R-cyclohexanecarboxylic acids in a s a t u r a t e d solution of thiourea in methanol at 20° depend on R in the o r d e r * Et > i P r > t-Bu < t - P e n t . The very stable adducts of t r a n s - and c i s - l , 4 - d i - t - b u t y l c y c l o h e x a n e provide other examples of the good accommodation of t-butyl groups in the thiourea channel. X - r a y rotation d i a g r a m s showed^ that the repeat periods of the trans-4-alkylcyclohexanecarboxylic acids in the thiourea adducts c o r r e s p o n d to twice the length of the molecule, apparently a s a consequence of dimerization by hydrogen bonding. Inspection of molecular models of the d i m e r s of e . g . c i s and t r a n s 4 t b u t y l -cyclohexanecar boxy lie acid [16] r e v e a l s (see F i g . 2) that the equatorial substituents i n t r a n s - [ l 6 ] e n s u r e a linear shape of the d i m e r (a) whereas the dimensions of d i m e r i c c i s - [ l 6 ] (b) - with the t-butyl groups equatorial and the carboxyl groups axial - do not allow accommodation in the thiourea channel. A non-linear shape

is a l s o exhibited by the mixed d i m e r (c).

V-H-0 ^^^^^s.

0-S '^"

Some other g e o m e t r i e s for a cis-4-alkylcyclohexanecarboxylic acid a r e to be considered which might allow at f i r s t sight inclusion into thiourea.

(i) Inclusion a s a monomer with the alkyl group equatorial and the carboxyl group axial. F r o m the view of space demand this should

be allowed, a s methyl cis-4-t-butylcyclohexanecarboxylate f o r m s a thiourea adduct.^ A r a t h e r high concentration of e s t e r (in methanol) is r e q u i r e d , however, to e n s u r e adduct formation. I n c r e a s e of c o n -centration of the corresponding acid, c i s - [ 1 6 ] , is m o r e limited due to lower solubility. In g e n e r a l , compounds with free hydroxyl groups s e e m not p a r t i c u l a r l y well suited to produce stable thiourea adducts; this may be exemplified by the low tendency of the 4 t -butylcyclohexanols to form adducts.

(ii) Adduction a s a d i m e r with the alkyl group R axial and the carboxyl group equatorial. When R is bulky (t-Bu, t-Pent) such a geometry would s e e m improbable in view of spatial demands of R and also because the ring conformation is expected to change. F o r R = Me, on the other hand, the d i m e r in the above conformation would s e e m to fit into the thiourea channel. Indeed, c i s 4 m e t h y l -cyclohexanecarboxylic acid was found to yield an adduct when applying a high concentration of acid and a low t e m p e r a t u r e (0°). Still, the adduct of the t r a n s - i s o m e r was found to be distinctly m o r e s t a b l e , enabling purification of the t r a n s - i s o m e r . In this p a r t i c u l a r c a s e d i r e c t crystallization, affording the t r a n s - i s o m e r , is p r e f e r r e d , however.

(iii) Inclusion a s a d i m e r with the rings in a non-chair conformation. Then, at the cost of about 3 k c a l / m o l e to the ring, a linear geometry would be obtained with a length comparable to that of the t r a n s -d i m e r s . So far, no evi-dence for this possibility has been obtaine-d for cis-4-alkylcyclohexanecarboxylic a c i d s .

Other examples of thiourea separation of i s o m e r i c alkyl-substituted cyclohexanecarboxylic acids a r e the separation of 4 t b u t y l l methylcyclohexanecarboxylic acid^* (the e a e i s o m e r with the c a r -boxyl group in equatorial position is selectively included), and the selective inclusion of the two i s o m e r s with equatorial carboxyl group (the e a e and the e e e i s o m e r ) of 4 t b u t y l 2 m e t h y l c y c l o -hexanecarboxylic a c i d . '

A useful separation in the cyclohexene s e r i e s is the selective i n -clusion^^ in thiourea of 4-t-butyl-3-cyclohexene-1-carboxylic acid from m i x t u r e s with 3 - t - b u t y l - 3 - c y c l o h e x e n e - l - c a r b o x y l i c acid (as obtained by Diels-Alder addition).

Several e s t e r s of c i s - and trans-4-t-butylcyclohexanecarboxylic acid have been tested^ for adduct-forming ability. It was invariably found that the c i s - e s t e r s display much l e s s tendency to form

thiourea adducts than the corresponding t r a n s - e s t e r s . This enabled s e p a r a t i o n of ethyl, t-butyl, and cyclohexyl 4-t-butylcyclohexane-carboxylate in the c i s - and t r a n s - i s o m e r s , by applying two-step r e c r y s t a l l i z a t i o n p r o c e d u r e s . In t h i s connection it may be mentioned that also ditbutyl 1,4cyclohexanedicarboxylate could be s e p a

r a t e d by selective inclusion of the t r a n s - i s o m e r into thiourea. The above selectivities would s e e m to originate from the fact that the c i s - i s o m e r s have to adopt a (less stable) non-chair conforma-tion to e n s u r e good accommodaconforma-tion in the thiourea host lattice.

Several p a i r s of l - t - b u t y l - 4 - R - c y c l o h e x a n e s have been examined for separation through thiourea adducts. It was found that when R = M e the c i s - i s o m e r is preferentially included, in accordance with studies of Russian w o r k e r s , ^ * whereas for R=iPr and R=t-Bu thiourea gives some preference to the t r a n s - i s o m e r . The latter a l s o holds for the 1,4-diisopropyl- and the 1,4-dicyclohexylcyclo-hexanes.

The above shift in preference of thiourea can be understood by examining molecular models of the hydrocarbons, with the d i m e n -sions of the thiourea channel in mind. When an axial methyl group is added to t-butylcyclohexane at the 4-position channel-fitting would seem to improve, when introducing other axial 4-alkyl groups in t-butylcyclohexane the c r o s s - s e c t i o n of the molecule is seen to be c r i t i c a l . On the other hand equatorial alkyl groups at the 4-position should be branched to provide a good contact with the thiourea w a l l s .

Length determination of guest molecules^ by X - r a y diffraction studies (cf. H. iii) of thiourea adducts confirmed that c i s 4 b u t y l -1-methylcyclohexane is in a chair conformation in thiourea. Un-fortunately, no decisive answer was obtained for the ring confor-mation of the other c i s - l , 4 - d i a l k y l c y c l o h e x a n e s .

The differences in adduct-forming ability between i s o m e r i c 1,4dialkylcyclohexanes a r e l e s s pronounced than for the a b o v e m e n -tioned acids and e s t e r s . Consequently repeated r e c r y s t a l l i z a t i o n of the adducts i s r e q u i r e d t o obtain the preferentially adducted i s o m e r in a pure s t a t e .

It may be mentioned that also selenourea - adducts of which compound w e r e p r e p a r e d for the first time in this laboratory ^ forms i n -clusion compounds with s e v e r a l 1,4-dialkylcyclohexanes. Though the difference in channel d i a m e t e r between thiourea and selenourea adducts s e e m s to be s m a l l , selenourea was found to be m o r e s e l e c -tive in the choice of its guest compounds. Some examples a r e given in ref. 15.

Some thiourea experiments with p a i r s of i s o m e r i c 1,3dialkylcyclohexanes d e s e r v e mention. In the c a s e of l t b u t y l 3 m e t h y l cyclohexane both i s o m e r s form an adduct with thiourea, the t r a n s -i s o m e r (w-ith the methyl group -in ax-ial pos-it-ion) be-ing preferent-ially adducted. On the other hand neither of the i s o m e r s of 1 , 3 d i i s o propylcyclohexane yields an adduct. In the c a s e of 1 , 3 d i t b u t y l -cyclohexane the t r a n s - i s o m e r is found to give an adduct, whereas the c i s - i s o m e r does not. Apparently, the non-chair ring

tion^^'^^of t r a n s - 1 , 3 - d i - t - b u t y l c y c l o h e x a n e allows inclusion into the thiourea channel.

Finally, attention may be drawn to the effect of some variation of the r i n g p a r t of guest compounds. F o r adducts of s i x - m e m b e r e d r i n g s y s t e m s the r u l e observed i s : the m o r e saturated the r i n g , the m o r e stable the thiourea adduct. F o r instance the adducts of c i s - and t r a n s - 1 , 4 - d i - t - b u t y l c y c l o h e x a n e a r e m o r e stable than the adduct of 1,4-di-t-butylcyclohexene which in its turn is m o r e stable than the adduct of 1,4-di-t-butylbenzene. These differences in

adductforming ability a r e sufficiently g r e a t to p e r m i t u s e in p r e p -a r -a t i o n s . ^

5. M i s c e l l a n e o u s

5.1 Epimerization techniques

Epimerization (configurational equilibration) at the a - c a r b o n either of the free acid or of the methyl e s t e r has been a p p l i e d ' for p r e -p a r a t i v e -p u r -p o s e s to many alkyl-substituted cyclohexanecarboxylic a c i d s . Acids can be e q u i l i b r a t e d ' by heating them with concentrated hydrochloric a c i d - a c e t i c acid (1 :3 by voliune) at about 150°. This technique h a s been applied to cis-2-alkylcyclohexanecarboxylic a c i d s , •'^'' c - 2 , c - 6 - d i a l k y l - r - l - c y c l o h e x a n e c a r b o x y l i c a c i d s , ' ' and c - 2 , c - 4 - d i a l k y l - r - l - c y c l o h e x a n e c a r b o x y l i c a c i d s , ' which compounds

(with axial carboxyl group) all a r e readily a c c e s s i b l e by catalytic hydrogenation of the corresponding benzoic a c i d s . The epimer with equatorial carboxyl group predominates in the equilibrium mixture and can be purified in general easily by r e c y r s t a l l i z a t i o n .

Configurational equilibration at C i of methyl e s t e r s is known to proceed by boiling with a solution of sodium methoxide in methanol. We used this method preferentially in c a s e s in which the equilibrium composition r e q u i r e d p r e p a r a t i v e GLC for separation of the e p i m e r s . Examples include the p r e p a r a t i o n ' of t 3 , t 5 d i e t h y l and t 3 , t 5 ditbutylr1cyclohexanecarboxylic acid starting from the c , c i s o m e r s , which a r e readily available by hydrogenation of the a r o -matic compounds.

Another example of the p r e p a r a t i o n of a cyclohexane s y s t e m with an axial methoxycarbonyl group starting from its epimer is the s y n -t h e s i s ' of s o m e me-thyl -t - 2 - R - -t - 5 - -t - b u -t y l c y c l o h e x a n e c a r b o x y l a -t e s

[ 1 8 ] . Again, the a l l - c i s acids, providing [ 1 7 ] , a r e a c c e s s i b l e by catalytic hydrogenation of the corresponding a r o m a t i c a c i d s .

17 18 COOMe

For R=Me and R=Et the eae-isomer [17] is preferred, for R=iPr and R=t-Bu, however, the equilibrium lies in favour of [18] showing substantial variation in the vicinal ea-COOMe-R interaction. For R=iPr, the equilibrium constant amounts to 5. 8 (-AG = 1.04 kcal/ mole); combination with the conformational energy of the methoxy-carbonyl group ' (1.27 kcal/mole) yields AG = 2.3 kcal/mole for the ea-COOMe-iPr interaction. For R=t-Bu it is doubtful whether [17] and [18] are in the chair conformation.

5. 2 Koch carboxylation

Tertiary carboxylic acids can be prepared by treating alcohols or olefins (or hydrocarbons in the presence of a lower tertiary or secondary alcohol) with carbon monoxide (often generated in situ from formic acid) in a strongly acidic medium. ^"''^ We used this method as an alternative to reductive alkylation (1.2) in the syn-thesis of some 1-alkyl-substituted cyclohexanecarboxylic acids, the acidities of which are discussed in section IE. Besides the known acids 1-ethyl- and 1,3-dimethylcyclohexanecarboxylic acid and the decalin-9-carboxylic acids, ''^ a number of new acids were prepared in this way.

Starting from butyl-l-methylcyclohexane as well as from 4-t-butyl-1-methylcyclohexanol we prepared c-4-t-butyl-l-methyl-r-1-cyclohexanecarboxylic acid.'^Similarly, t-3-t-butyl-l-methyl-r-1-cyclohexanecarboxylic acid was obtained from 2-methyl-4-t-butylcyclohexanol. Thus the Koch procedure provides selectively the aee-isomers with the carboxyl group in axial position. The corresponding eae-acids were obtained from mixtures of the two isomers prepared by reductive methylation of t-butylbenzoic acids followed by hydrogenation, and by carboxylation of the Grignard compounds of 4- and 3-t-butyl-l-bromo-l-methylcyclohexane. As an example of side-chain carboxylation we mention the prepa-ration of 2-methyl-2-(trans-4-t-butylcyclohexyl)propionic acid from 2-(trans-4-t-butylcyclohexyl)-2-propanol. The yield is low, however, due to the formation of many side products.

Complex mixtures of acids were also obtained when subjecting t -butylcyclohexanols to the Koch procedure. As found by Peters^' side reactions are suppressed by applying the Haaf modification'^ with slow stirring of the mixture. t-Butylcyclohexanols were con-verted in this way^ into 3 :1 mixtures of 1-t-butylcyclohexane-carboxylic acid and 2-methyl-2-(l-methylcyclohexyl)propionic acid. The synthetic importance of this technique is stressed by the fact that so far attempts at reductive t-alkylation of benzoic acid were unsuccessful.

5.3 Diels-Alder addition

The Diels-Alder addition'^ has been applied incidentally in the p r e s e n t work for the purpose of introducing a double bond at a given position in a s i x m e m b e r e d ring s y s t e m . F o r instance 4 t -butyl-3-cyclohexene-1-carboxylic acid can be b e s t p r e p a r e d ^^ by Diels-Alder r e a c t i o n of 2-t-butylbutadiene and acrylic acid. Another example is the preparation^ of 2 t b u t y l l c y c l o h e x e n e 1carboxylic acid by 1,4cycloaddition of butadiene and methyl t -butylacetylenecarboxylate, followed by p a r t i a l hydrogenation and h y d r o l y s i s .

Secondly, the retention of configuration of the r e a c t a n t s in the Diels-Alder r e a c t i o n can be applied to p r e p a r e a given i s o m e r by p r o p e r choice of the dienophile. Thus, l , t 2 d i m e t h y l r l c y c l o -hexanecarboxylic acid was conveniently p r e p a r e d by Diels-Alder r e a c t i o n of butadiene and tiglic acid ((E)-2-methyl-2-butenoic acid), followed by hydrogenation.

5.4 Cyclohexane derivatives a s starting m a t e r i a l s in some syntheses Alkylcyclohexanols (easily a c c e s s i b l e by hydrogenation of the c o r -responding phenols) would seem suitable starting m a t e r i a l s for the synthesis of alkylcyclohexanecarboxylic acids through replacement of hydroxyl by bromine followed by a Grignard r e a c t i o n . However, common p r o c e d u r e s for the f i r s t r e a c t i o n fail in that much

i s o m e r i z e d product " i s formed, for instance 4-t-butylcyclohexanol yields a. o. 3 - t - b u t y l - l - b r o m o c y c l o h e x a n e . An a l t e r n a t e p r o c e d u r e , which has been a p p l i e d ' in the p r e s e n t study, involves the p r e p a -r a t i o n of the tosylate which can be conve-rted into the b-romide by a solution of calcium bromide in DMF. ' ® Another method involves heating of the tosylate with sodium cyanide in N-methylpyrrolidone'^ yielding the cyanide which is hydrolyzed to give the carboxylic acid. F o r the synthesis of most cyclohexylacetic acids use h a s been made of the A r n d t E i s t e r t s y n t h e s i s . A low t e m p e r a t u r e (0°) is r e c o m -mended for the p r e p a r a t i o n of the starting acid chlorides in o r d e r to prevent epimerization. As might be expected the r e a r r a n g e m e n t of the diazoketones involved was found to occur stereospecifically.

n . C O N F I G U R A T I O N A L A N A L Y S I S

We have developed and applied s e v e r a l methods and c r i t e r i a in o r d e r to assign reliably the configuration to the various s e t s of epimeric acids and other cyclohexane d e r i v a t i v e s . Chemical a s well a s p h y s i -ical c o r r e l a t i o n s have been m a d e . A survey follows.

(i) Chemical proofs of s t r u c t u r e . A number of 4-alkylcyclohexane-carboxylic acids and 1,4-dialkylcyclohexanes have been c o r r e l a t e d

chemically'~'^°'^ with the 1,4-cyclohexanedicarboxylic acids, the configuration of which had been established with certainty. ®° As an example we give the chemical proof of s t r u c t u r e of the 4 e t h y l -and the 4-methylcyclohexanecarboxylic acids (see Scheme V). The r e a c t i o n s shown in the diagram w e r e c a r r i e d out for the c i s - a s well a s for the t r a n s - c o m p o u n d s . Moreover, some 1,2-disubstituted cyclohexanes have been c o r r e l a t e d chemically with the known

cyclohexanedicarboxylic a c i d s . e l 1.2-COOH COOH C H , CH2CH20H

i

The s t e r e o c h e m i c a l r e s u l t s (Table I) of the hydrogenation^ of c i s -and t r a n s - 1 , 3, 5-tri-t-butylcyclohexene ( [ 7 ] -and [S]) provide a direct proof of s t r u c t u r e for these compounds as well a s for both 1, 3, 5-tri-t-butylcyclohexanes. Compoimd [ 7 ] yields c i s , c i s - a s

well as cis, trans-1,3, 5-tri-t-butylcyclohexane upon hydrogenation or reduction with diimide whilst [8] gives exclusively the c i s , t r a n s -isomer.

In some cases the course of the hydrogenation of aromatic com-pounds (e.g. 2-t-butylbenzoic acid^) provided strong evidence for the configurations of the cyclohexane derivatives obtained.

(ii) Equilibration. Configurational equilibration (cf. 1.5.1) at the a-carbon has been applied' to many pairs of epimeric acids and methyl esters. The configuration with the carboxyl or methoxy-carbonyl group in equatorial position was assigned to the more stable epimer. This assignment was also found to hold for the 2-alkylcyclohexanecarboxylic acids, showing that reversal of stability, as has been observed for the 2-t-butylcyclohexanols, ®^ does not occur here. However, when vicinal interactions can be evaded in the epimer with axial carboxyl or methoxycarbonyl group exceptions to the above rule may occur; we mentioned already that methyl t 2 -isopropyl-t-5-t-butyl-r-l-cyclohexanecarboxylate was found to be more stable than its epimer with equatorial methoxycarbonyl group.

(iii) Length determination in thiourea.^'®^ It may be recalled that the inclusion or non-inclusion of a 4-alkylcyclohexanecarboxylic acid in thiourea may be looked upon as a strong indication of trans - or cis-configuration of the isomer in question. With many other pairs of 1,4-disubstituted cyclohexanes both isomers were found to give an adduct with thiourea (though generally of unequal stability). A nimiber of the adducts obtained were analyzed^ by performing C-and N-determinations C-and by taking X-ray rotation diagrams. From these diagrams the repeat distance of the guest molecules in the direction of the c axis was calculated. In general this repeat period was found to correspond to the linear dimension of one molecule, showing head-to-tail stacking of the guest molecules. Exceptions to this are observed for most carboxylic acids. Table HI gives some results.

When in the adducts both isomers are in the chair conformation, the trans-isomer (I) is expected to possess a longer period than the corresponding cis-isomer (Ha). About equal periods are expected, however, when the cis-isomer adopts a non-chair conformation such as nb in the thiourea channel.

I l a Qb 28

Table HI. Repeat periods of thiourea adducts No. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Guest compound t r a n s - 4 - i s o p r o p y l - l - m e t h y l c y c l o h e x a n e c i s - 4 - i s o p r o p y l - l - m e t h y l c y c l o h e x a n e t r a n s - 4 - t - b u t y l - l - m e t h y l c y c l o h e x a n e c i s - 4 - t - b u t y l - l - m e t h y l c y c l o h e x a n e t r a n s - 1 , 4 - d i - t - b u t y l c y c l o h e x a n e c i s - 1 , 4 - d i - t - b u t y l c y c l o h e x a n e trans-1,4-dicyclohexylcyclohexane cis-1,4-dicyclohexylcyclohexane methyl trans-4-t-butylcyclohexanecarboxylate methyl cis-4-t-butylcyclohexanecarboxylate t r a n s -1 - a c e t y l - i -t-butylcyclohexane c i s - 1 -acetyl-4-t-butylcyclohexane t r a n s - 1 - b r o m o - 4 - t - b u t y l c y c l o h e x a n e c i s - 1 -bromo-4-t-butylcyclohexane P e r i o d , X host guest 12.4 1 2 . 4 1 2 . 5 1 2 . 4 1 2 . 4 1 2 . 4 12.4 1 2 . 4 1 2 . 3 12.4 1 2 . 5 1 2 . 5 12.5 1 2 . 5 10.4 9.3 10.4 9.3 12.4 1 1 . 3 1 5 . 1 1 5 . 1 1 2 . 3 10.7 1 1 . 1 9.8 10.8 9.5

With adducts [ l 9 ] - [ 2 2 ] and [27]-[32] (see Table HI) the differences in repeat period are imderstood on the basis of chair conformations of the guest compounds: the cis-isomers occupy a distinctly shorter section of the thiourea channel than the trans-isomers. In fact, the cis-isomers in adducts [20], [22] and [32] occupy channel lengths which are only slightly greater than that of t-butylcyclohexane

(9.0 X). This is to be expected for a chair conformation with the methyl group ([20] and [22]) and the bromine atom [32] in axial position and thus not contributing significantly to the 'length' of the molecule.

These results demonstrate the potential use of length-determination in thiourea for configurational analysis of pairs of isomeric 1,4-disubstituted cyclohexanes.

With adducts [23]-[26] and some related adducts^ the question arises whether the ring in the cis-isomers is in a flexible or in a chair conformation in the adducted state. The identical period observed for the adducts [25] and [26] would strongly suggest a, non-chair conformation f or the central ring of the cis-isomer. The calculated length of cis-1,4-di-t-butylcyclohexane is 11.3 A in the case of a boat conformation and 10.4 A for a chair conformation (using 2.0 X for the van der Waals radius of the methyl group). In view of the length observed in its adduct [24] it is likely that a

chair conformation is involved in thiourea. It may be noted that in adduct [23] the guest has adopted the thiourea repetition. Clearly, the analysis of adducts [23]-[26] allows no conclusion regarding the configuration of the guest compounds.

(iv) NMR spectra. Non-equivalence of alkyl groups as revealed by NMR spectroscopy allowed assignment of configuration to a nimiber of cyclohexane derivatives.^"' For instance the NMR spectra of c-3,t-5-di-t-butyl-r-l-cyclohexanecarboxylic acid and of the cor-responding nitrile show two signals for the protons of the t-butyl groups, thus providing a proof of structure. Another example is r - 2 , c-4, t-6-tri-t-butylcyclohexanone ^ which compound is distin-guished from its two epimers by its NMR spectrum, which shows three signals for the protons of the t-butyl groups.

A general method of analysis ®* is based on the band width and the chemical shift of the signal due to the cf-proton in substituted cyclo-hexane derivatives. We have used this criterion in the configurational analysis of frozen alkyl-substituted cyclohexanecarboxylic acids. Equatorial a-protons (carboxyl group axial) were found to resonate at lower field (5 2.5-3.0 ppm) than the other ring protons and appeared as relatively narrow bands, whereas the signals due to axial Qf-protons usually overlapped with the signals of the other ring protons. In this way the position of the carboxyl group could be determined easily.

When the splitting patterns of the a-proton absorption is simplified as in 2, 6-dialkyl-substituted cyclohexane derivatives, the NMR spectrum generally allows assignment of structure at once. Examples include c 2 , c 4 , c 6 and t 2 , t 4 , t 6 t r i t b u t y l r l

-cyclohexanol. ^

(v) Mass spectra.®^ It has been found'"' that mass spectrometry is able to distinguish between the c i s - and trans-isomers of 3 - and 4tbutylcyclohexanecarboxylic acid. The fragmentation of the c i s -isomers appears to be largely governed by the ability of the t-butyl and the carboxyl group to interact. Transfer of a hydrogen atom to the (ionised) carboxyl group followed by loss of C4H7 leads to the base peak at m / e 129 in the spectra of c i s - 3 - and cis-4-t-butyl-cyclohexanecarboxylic acid. Interaction between the t-butyl and the carboxyl group is not possible (unless isomerization occurs) in the case of t r a n s - 3 - and trans-4-t-butylcyclohexanecarboxylic acid. Here, loss of the t-butyl group from the molecule ion, giving m/e 127, is an important fragmentation. '"'

Several sets of isomeric more highly substituted cyclohexane-carboxylic acids with t-butyl groups at the 3 - or 4-position have been found®® to show similar characteristic mass spectral

e n c e s , thus leaving no doubt about the mutual configuration of the t-butyl and the carboxyl group. In an analogous way m a s s s p e c t r a distinguished between the c i s and t r a n s i s o m e r s of 4 t p e n t y l -cyclohexanecarboxylic acids ^ and 4-t-butylcyclohexaneacetic acids.®®

(vi) GLC retention t i m e s . Relative GLC retention t i m e s of s e t s of i s o m e r i c cyclohexane derivatives have been found to be quite h e l p -ful for a tentative assignment of configuration. Two retention time

(and boiling point) r e g u l a r i t i e s to which no exceptions have been observed a r e : other things being equal, equatorial COOR > axial C O O R ^ ' ' ^ * ' ^ ' and axial Me > equatorial Me.^'^*'®® F o r instance the retention t i m e o r d e r of the methyl 3, 5-dimethylcyclohexane-carboxylates i s aee > eee > e a e .

F o r a mobile system such a s methyl cis-2-methylcyclohexane-carboxylate the r e l a t i v e retention time with r e s p e c t to its epimer cannot be given a p r i o r i : on the other hand the experimental o r d e r of these methyl e s t e r s cis > t r a n s indicates that the c i s - i s o m e r p r e f e r s the conformation with the methyl group in axial and the methoxycarbonyl group in equatorial position.

Several r u l e s have been postulated r e g a r d i n g r e l a t i o n s between physical p r o p e r t i e s , such a s density, refractive index and boiling point, and the configuration of epimeric cyclohexane derivatives.®^ An a c c u r a t e h i s t o r i c a l analysis of the origin and the scope of t h e s e r u l e s has been made by Verkade et al.'"'*

As to differences in boiling point between e p i m e r i c cyclohexane derivatives the data available allow the r u l e of thumb that for

polar substituents the e p i m e r with the polar substituent in euqatorial position (other things being equal) h a s the higher b o i l i i ^ p o i n t . ' ' ^ * ' ^ ° This r e s u l t is quite r e a s o n a b l e . Regarding alkyl groups other than methyl (see above) much data on p a i r s of epimeric dialkylcyclo-hexanes have been a s s e m b l e d by Russian workers®^ a s well a s in this l a b o r a t o r y . H e r e , the boiling point sequence is found to depend s o m e t i m e s on the way of ring substitution. Thus for the 1 , 2 - and 1,3-diisopropylcyclohexanes^'* the a e - i s o m e r is found to p o s s e s s a higher boiling point that the e e - i s o m e r w h e r e a s the r e v e r s e holds for the 1,4-diisopropylcyclohexanes.^'^^ A s i m i l a r r e v e r s a l is observed for the di-t-butylcyclohexanes^®'®^'^° in which c a s e the situation is of c o u r s e m o r e complicated due to the r o l e of non-chair ring conformations.

(vii) Density and refractive index. We have advanced the r u l e ^* > ^* that with cyclic s t e r e o i s o m e r s in which the substituents a r e bound to identical r i n g s the i s o m e r with the higher density and the higher refractive index is that which has the higher heat content. This

rule can be used with good confidence as has been shown a.o. for a series of alkylcyclohexylamines.®^ Generally the difference in density between epimeric cyclohexane derivatives is more pro-nounced than the difference in refractive index.

m . C O N F O R M A T I O N A L A N A L Y S I S . pK*-ME A S U R E M E N T S

1. 1 , 4 D i t b u t y l c y c l o h e x a n e a n d 1 , 3 , 5 t r i t b u t y l -c y -c l o h e x a n e

The i s o m e r i c 1,4-di-t-butylcyclohexanes ( c i s - and t r a n s - [ l O ] ) and c i s , c i s - and c i s , t r a n s - 1 , 3,5-tri-t-butylcyclohexane ([33] and [34]) w e r e p r e p a r e d ^ ' ^ by u s since we w e r e interested in the question whether in c i s - [ l O ] and in [ 3 4 ] the s t r a i n of an axial t-butyl group i s

so l a r g e that a non-chair conformation i s predominant.

C i s - [ l O ] has been examined® by various tools including infrared and NMR^® spectroscopy, X r a y and electron diffraction^' t e c h -niques, without obtaining a definite answer to the above question. The data p r e s e n t strong evidence, however, that c i s - [ l O ] p r e f e r s a nonchair conformation. This i s supported by some other p r o p e r ties® of c i s [ l O ] viz. melting point, heat of fusion, thiourea i n -clusion, and t h e difficult separation from t r a n s - [ l O ] . Therefore the enthalpy difference® between c i s - and t r a n s - [ l O ] is considered to be a good e s t i m a t e of the enthalpy difference between the c h a i r and a non-chair conformation of cyclohexane.

tBu

t B u ^ ^ tBuJ- tBuy-^^^^J-3A a 34 b

F o r compound [ 3 4 ] a non-chair conformation as [34a] a s well a s a chair conformation [34b] with one axial t-butyl group have to be considered. The NMR spectrum of [3-^] (60 MHz, -100 to +40°) p r e s e n t s s o m e evidence in favour of [ 3 4 a ] : the t-butyl groups a r e equivalent, and the r i n g protons appear a s one narrow band, whereas the NMR spectrum of [33] shows the ring protons over a much wider a r e a . This picture is acceptable when the flexible conformation [34a] is p r e f e r r e d . Other support of [ 3 4 a ] s t e m s from the relatively low melting point and heat of fusion of [ 3 4 ] and also from its s t r u c t u r a l relationship to t r a n s 1 , 3 d i t b u t y l -cyclohexane which compound is considered to prefer a non-chair form.®^'®®

2. p K * v a l u e s of 4 a n d 3 a l k y l s u b s t i t u t e d c y c l o -h e x a n e c a r b o x y l i c a c i d s

Methods of conformational analysis of cyclohexane derivatives have relied on the absence of any polar or steric effect of 3 - and 4-alkyl substituents on the functional group and its environment. ^® > ^ ^ This was accepted for frozen systems as t-butylcyclohexyl compounds as well as for the conformers of mobile cyclohexane systems. Recently, several departures from this postulate have been reported, deduced from kinetic studies•''°°"'''°^ and equilibrium measurements.•'•°°'''-°*

We have measured' the thermodynamic dissociation constants of forty 3 - and 4-alkyl-substituted cyclohexanecarboxylic acids in 50% ethanol-water (Tables I and II in ref. 7) in order to find out to what extent the acidity of an equatorial or axial carboxyl group is in-fluenced by an alkyl group at the 3 - or 4-position.

Effects of equatorial 4- and 3-alkyl groups on an equatorial 1-carboxyl group can be gathered from Table IV which table contains examples of such situations. 4-Alkyl groups are found to cause just a slight acid-weakening, whereas 3-alkyl substituents exert more substantial effects, which increase with growing bulk of the alkyl group. These ee-3-alkyl effects are foimd to be additive. Table IV. pK*-values of all-equatorial 4 - and 3-alkyl- and 3, 5-dialkylcyclohexanecarboxylic acids in 50% ethanol-water at 25° t-4-R H 4-Me 4-Et 4-iPr 4-t-Bu pK* 6.20 6.20 6.22 6.22 6.22 C-3-R 3-Me 3-Et 3-iPr 3-t-Bu pK* 6.23 6.27 6.29 6.30 c,c-3,5-di-R 3,5-di-Me 3,5-di-Et 3,5-di-iPr 3,5-di-t-Bu pK* 6.26 6.36 6.38 6.43

Examples in which axial 1-carboxyl groups are influenced by equa-torial 4 - or 3-alkyl groups are given in Table V. The value of the cis-4-t-alkylcyclohexanecarboxylic acids is believed' to be a good estimate of the pK* of the chair conformer of unsubstituted cyclo-hexanecarboxylic acid with the carboxyl group axial. The data reveal that equatorial 3-t-alkyl groups weaken the acidity of an axial 1-carboxyl group considerably.

Two factors can be envisaged' for the origin of the acid-weakening e e - and ae-3-alkyl effects: (i) the hydrophobic 3-alkyl group exerts some hindrance to solvation in the carboxylate anion, ^°® (ii) the 3 -alkyl substituent brings about a ring geometry which differs from that of the conformers of cyclohexanecarboxylic acid.

Table V. pK*-values of immobile c i s - 4 - a l k y l - , t r a n s - 3 - a l k y l - , and t r a n s , t r a n s - 3 , 5-dialkylcyclohexanecarboxylic acids c - 4 - R pK* t - 3 - R pKf t , t - 3 , 5 - d i - R pK* 3 , 5 - d i - E t 6.84 4 - t - B u 6.68 3-t-Bu 6.90 3 , 5 - d i - t - B u 7.06 4 - t - P e n t 6.69 3 - t - P e n t 6.93

As to ring deformation, valence-force calculations ^°® and NMR analysis ^°'' on methylcyclohexane show that the changes in the g e ometry of the ring p a r t around the 3position introduced by an e q u a -t o r i a l me-thyl group a r e negligible. In -t-bu-tylcyclohexane•'•°®' •'•°® such effects a r e m o r e substantial but still s m a l l . However, the axial proton a t the carbon c a r r y i n g the tbutyl group is bent s o m e -what toward the c e n t r e of the ring and might cause hindrance to solvation of a Y-syn carboxylate anion.

Ref. 7 contains a few examples in which an equatorial carboxyl group is influenced by an axial 3-alkyl group. The effect s e e m s to depend strongly on the c h a r a c t e r of the alkyl group but m o r e examples a r e r e q u i r e d to settle t h i s .

When the axial 3alkyl group is t e r t i a r y , as in c 3 , t 5 d i t b u t y l -r-1-cyclohexanecarboxylic a c i d , ' the question a r i s e s once m o r e

(cf. i n . 1) whether the s y s t e m p r e f e r s a nonchair or a chair c o n -formation.

3 . 2 - A l k y l c y c l o h e x a n e c a r b o x y l i c a c i d s a n d m o d e l c o m p o u n d s

In 2-alkylcyclohexanecarboxylic acids d i r e c t s t e r i c interaction between alkyl and carboxyl group may o c c u r . An e q u a t o r i a l e q u a t o r i a l interaction is likely to be involved in t r a n s 2 a l k y l c y c l o -hexanecarboxylic acids whereas the c i s - i s o m e r s will exist in general a s a conformational equilibrium ae Ï^ ea. The positions of such c h a i r / c h a i r equilibria have been estimated ' by comparing the acidity of 2-alkylcyclohexanecarboxylic acids with that of suitable frozen 2 , 4 and 2,5dialkylcyclohexanecarboxylic acids e . g . 2 alkyl4(5)tbutylcyclohexanecarboxylic a c i d s . Altogether p K * -values of s o m e fifty 2-alkyl-substituted cyclohexanecarboxylic acids w e r e determined (Table HI in ref. 7) in 50% ethanol-water at 2 5 ° . F i r s t of all the various model compounds enabled estimation of the four types of vicinal alkylcarboxyl interactions in t e r m s of ApK* -units (allowing for any 3 - and 4-alkyl effects). Table VI gives pK* - i n c r e m e n t s with r e s p e c t to the pK*-values of the equatorial and axial conformer of cyclohexanecarboxylic acid; values in p a

rentheses stem from compounds for which also non-chair confor-mations are to be considered. The increments are in part mean values; see ref. 7 for the separate values.

Table VI. pK* -increments resulting from vicinal alkyl-carboxyl interactions 2-R Me Et i P r t-Bu e-COOH, e 0.02 0.11 0.19 0.42 -R e-COOH, a 0.19 0.30 0.43 (0.66) ApK* -R a--COOH, e 0.39 0.40 0.66 0.69 -R a-COOH,a-R - 0 . 0 2 0.08 0.16 (0.16)

Some of the vertical differences in Table VI can be understood by examining the preferred conformations of the alkyl groups. Thus, Newman projections (see below) of the four ee-combinations show that the third y methyl group introduced must accept a syn posi-tion with respect to the carboxyl group as a result of which the acid strength is reduced appreciably.

„„„AvA „„„,AjA „ „ , i a ^ "ooc^X^

H-V-H H ' V ^ C H S H-'V^CHa HjC-V^CHj

The pK*-increments observed show that a 2-alkyl group weakens the acidity of a 1-carboxyl group in the order aa < ee < ea < ae

(position of carboxyl given first). This sequence can be understood when accepting®''.^°3.^^° I (9 = 0°) and H (cp = 120°) as the p r e -ferred conformations of equatorial and axial carboxyl and carboxylate groups, respectively. For the axial carboxylate group also confor-mation ni (cp = 90°) can be considered.® It is recognized then that quite different 2-alkyl-carboxyl interactions are involved which become more serious in the above order of ApK*-values.

^ A ^ A % ^

H H I I HI

The differences in ea- and ae-2-alkyl-carboxyl interactions have a bearing on the chair-chair equilibria of cis-2-R-cyclohexane-carboxylic acids and their anions (see Scheme VI), the positions of which have been estimated' for R=Me,Et, and iPr (see Table VII).

^ ^ R COOH COOH COOH ^ Z . - ' - O ^ A R K K eq ", R " COO coo- = ? = ^ /^^^^:r^R Scheme 3E \ ^ ^ '

Table VH. Conformational equilibrium constants of cis-2-R-cyclo-hexanecarboxylic acids No. 35 36 37 R Me Et i P r KcOOH 0.30 0.35 4 . 2 -AG, k c a l / m o l e - 0 . 7 - 0 . 6 +0.8 Kcoo-0.06 0.07 0.9 -AG, k c a l / m o l e - 1 . 7 - 1 . 6 - 0 . 0 6

The data show that acids [35] and [36] prefer the ea- to the ae-con-formation whereas the reverse is true for [37]. In compounds [35] and [36] the alkyl-carboxyl interaction will be more serious in the ae- than in the ea-conformer.'» ^°® The change in conformational preference from [36] to [37] is more abrupt than would be ex-pected on the basis of conformational energies®' and is due to a relative destabilization of the ea-conformer of [37], which is recognized easily.'

Non-chair ring conformations are supposed to contribute in some of the 2-alkyl-substituted cyclohexanecarboxylic acids studied. For instance for the c-2-t-butyl-c-5-R-r-l-cyclohexanecarboxylic acids (R=Me, iPr, t-Bu) two chair conformations (A, B) and a non-chair conformation (e.g. C) have to be envisaged.

" COOH '^" tBu R.^ Z - - - - ^ ACOOH