Reactions of 2,2-dibromocyclopropyl carboxylic acids with methyllithium LEIVK. SYDNES' A N D SBLVISKARE Department qf' Chemistry, Urliversity ($Tromsm, P.O. Box 953, N-9001 TromsB, Norwny Received Janaury 3. 1984 This pnper is cledicc~tedto Projc.s.sor Pnul de M{lyo on the occrrsion of hi.r 60th birthdq

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

LEIV K. SYDNES and S 0 ~ v SKARE. l Can. J. Chem. 62, 2073 (1984). For reactions of 2,2-dibromocyclopropyl carboxylic acids with methyllithium, the course of reaction depends mainly on the position of the carboxyl group. When the COOH group is directly attached to the gern-dibromocyclopropane ring MeLi generally attacks the gem-dibromo moiety and gives the corresponding monobromocyclopropane as the principal product. When the reaction is performed above -80°C the monobromides are formed stereospecifically in the trnns configuration. 'The highest yields, as high as 80-90%. are obtained at O°C. When the carboxyl group is not directly attached to the cyclopropane ring most of the MeLi is consumed in an acid-base reaction with the COOH group.

LEIVK. SYDNES et S ~ L V SKARE. I Can. J. Chem. 62, 2073 (1984). L'orientation de la rtaction dtpend principalement de la position du groupe carboxylique. Lorsque le groupe COOH est attach6 directement au cycle gem-dibromocyclopropane, le MeLi attaque gkntralement la portion gem-dibromo et conduit principalement au produit monobromocyclopropane correspondant. Lorsqu'on effectue la reaction a des temptratures suptrieures a -80°C, les monobromures se forment stCr6ospCcifiquement en configuration trnrls. C'est i 0°C que I'on obtient les rendements les plus ilevCs, qui peuvent aller jusqu'i 80-90%. Lorsque le groupe carboxyle n'est pas attach6 directement au cyclopropane, la majorit6 du MeLi est utilist dans une riaction acide-base avec le groupe COOH. [Traduit par le journal]

Introduction Reactions of gem-dibromocyclopropanes with methyllithium generally occur with initial lithium-bromine exchange, resulting in formation of the corresponding bromolithio derivatives (1, 2). The subsequent reaction of such derivatives is significantly influenced by the nature of the substituents attached to the ring. When the substituents are nonpolar, like alkyl and aryl groups, the organometallic intermediates usually collapse by elimination of lithium bromide to yield allene and (or) bicyclobutane derivatives ( I - 10). On the other hand, a variety of polar groups, e.g. the carbonyl, alkoxy, and silyloxy moieties, may stabilize the lithio-bromocyclopropane by intramolecular coordination and thus prevent LiBr elimination (9- 13). In such cases the major product is generally the corresponding monobromocyclopropane, which is formed by hydrolysis of the lithium derivative during aqueous work-up. An exception to this general pattern is 2,2-dibromo-Imethylcyclopropanecarboxylic acid ( l b ) . When this compound was treated with methyllithium at O°C, a stereospecific formation of trans-2-bromo- I -methylcyclopropanecarboxylic acid (2b) was observed (14). According to Stein and Morton

I

2

3

trR=R1=H.n=O b R = H, R = CH2, rl = O c R = R' = CH2, rl = 0 d R = H, R' = Ph. rl = 0 e R = H, R' = CH3, 11 = I

this product resulted from an intramolecular quenching of a cyclopropyl anion prior to hydrolysis (14). It is thus conceivable that a COOH group attached to a gem-dibromocyclopropane can function as a general directing group and act 'Author to whom correspondence should be addressed.

Ph

_/

CHCI3

TEBA 50% NaOH

dph

NaOH/EtOb

CI2

Ph

/

-7CH(OEt), 4

as a vehicle for stereoselective preparation of a variety of monobromocyclopropanes. It was therefore of interest to study the reactions between methyllithium and various gem-dibromocyclopropanecarboxylic acids; the results of this investigation are reported here.

Results and discussion Most of the acids employed were prepared in good yields by literature procedures. 2,2-Dibromo- l -phenylcyclopropanecarboxylic acid ( I d ) , however, was obtained from styrene as outlined in Scheme I. The low overall yield (23%) results from inefficient addition of dibromocarbene to the carbon-carbon double bond of atropaldehyde diethylacetal (4); this is not surprising since dichlorocarbene addition to similar acetals is fairly ineffective ( 15). 2-(2,2-Dibromo- I-methylcyclopropyl)ethanoic acid ( l e ) was prepared by Jones oxidation of the corresponding primary alcohol, which was obtained by dibromocarbene addition to 3-methyl-3-buten- 1-01 under phasetransfer conditions.The cyclopropyl alcohol formed in the latter reaction was contaminated with a minor product tentatively identified as 2,2-dibromo- 1 -methyl- 1 -(2-dibromomethoxyethy1)cyclopropane (5) on the basis of analogy with earlier work ( 16); 5 is probably produced by dibromocarbene insertion into the 0-H bond (Scheme 2) (16). The gem-dibromocyclopropane acids were reacted with

CAN. J . CHEM. VOL. 62. 1984

TABLE1. Product distribution in reactions of 2,2-dibromocyclopropyl acids (1) with 1.2 equivalents of methyllithium

CHBr, TEBA

50% NaOH

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

&

O

H

&OH

+

A O C H B r - .

Products (% yield)" Acid Reaction (% unreacted)" temperature ("C) cis-2 trans-2 3 Unknown

SCHEME 3 methyllithium at - 1 15 to 25°C using, in most cases, 1.2 molar equivalents of the lithium reagent. The reaction mixtures were analyzed by gas-liquid chromatography (glc) before the products were isolated by distillation or column chromatography and identified by spectroscopic means. Most of the results are summarized in Table 1 , which reveals several interesting features. With the exception of l e , the acids generally react to give the corresponding monobromocyclopropanes 2 as the main products. When the reactions were carried out below -80°C the monobromides were formed as isomeric mixtures. However, above -80°C the monobromo derivatives were formed as single isomers with the bromine and carboxyl groups trans to each other. The stereochemistry is assigned by 'H nmr spectroscopy on the basis of two assumptions. Firstly, it is assumed that in a cyclopropane ring the cis coupling constant is larger than the trans, which is in accordance with calculations as well as experimental data (17). Secondly, the assumption is that for a pair of isomers the hydrogen atom geminal to the bromine atom gives rise to signals at a lower field for the trans than for the cis isomer, because of the anisotropy of the carbonyl group (12, 13, 17). Treatment of acid l e with methyllithium, on the other hand, gave no monobromocyclopropane at all; the predominant product was 3-methyl-3,4-pentadienoic acid (3e) which was isolated in low yield when the reaction was carried out above -5°C. An allene, 2-methyl-2,3-pentadienoic acid (3c), was also formed when l c was treated with MeLi at and above WC, but in this case the monobromide was the major product. The products isolated can be accounted for as outlined in Scheme 3. The first step involves an exchange of bromine with lithium, which has been reported to be a faster reaction than proton abstraction from the carboxyl and hydroxyl groups of various bromosubstituted acids and alcohols (14, 18-20). The organolithium intermediate is not formed stereospecifically since reactions performed at - 115°C result in formation of mixtures of isomeric monobromocyclopropane acids. Consequently, at - 115°C the stereoisomeric organolithium derivatives 6 and 7 equilibriate very slowly. However, at -78°C and above, the only isomer present is 7 , which is the more

"% of reaction mixture as determined by glc and 'H nmr analyses; for isolated yields, see Experimental. "Identified by spectra only.

TABLE2. pK, values of various cyclopropanecarboxylicacids in water at 25°C Acid

pKa

Cyclopropanecarboxylic acid 2-Bromocyclopropanecarboxylic acid 2,2-Dibromocyclopropanecarboxylic acid ( l a ) 2.2-Dibromo-I-methylcyclopropanecarboxylic acid (I b) trans- 1,2-Dimethylcyclopropanecarboxylicacid trans-2,2-Dibromo-l,3-dimethylcyclopropanecarboxylic

acid ( l c ) 2,2-Dibromo-l -phenylcyclopropanecarboxylicacid (Id) 2-(2,2-Dibromo-1-methylcyclopropyl)ethanoic acid (le) "From ref. 23. "From res. 22.

stable due to lithium-oxygen coordination. Stabilization of organolithium intermediates by adjacent oxygen functions has been extensively reported in the literature (9, 1 1 - 13). The final products result from two competing reactions following the formation of the lithio-bromocyclopropanes, viz. quenching of the cyclopropyl anion and allene formation via a cyclopropylidene intermediate. When the carboxyl group is directly attached to the ring, the former alternative is generally the predominant reaction; a-elimination of lithium bromide and subsequent allene production are only observed when steric interactions increase the internal energy of 7 and render opening of the ring a more favourable process. The amount of recovered starting material depends, in addition to the relative amount of methyllithium employed, mainly on two factors. More important is obviously the position of the carboxyl group relative to the gem-dibromocyclopropane ring. When the COOH group is directly attached to the ring, as in la-lci, the dibromo moiety and the carboxyl group will mutually interact; this is convincingly borne out by comparing the pK, values of the acids with those of similar cyclopropane-

SYDNES A N D S K A R E

TABLE3. The ratio between the acids before and after treatment of the l b / l e and 1b/ I , I-dibromo-2,2,3-trimethylcyclopropane (DTC) mixtures with MeLi Yield of

2. DCI

rr.ctt1.s-2 b

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

Mixtures

Before MeLi

After MeLi

H

9

(%I

carboxylic acids (Table 2) (21 -23). As a result of this interaction the bromine atoms become more positive and compounds l a - l d are therefore more prone to undergo lithium-bromine exchange than proton abstraction from the carboxyl group when treated with MeLi. On the other hand, in 1 e such interactions are impossible, and proton abstraction is therefore the predominant reaction when this compound is treated with methyllithiurn; consequently, most of the acid is recovered unchanged during the work-up. The reaction temperature also influences considerably the amount of recovered starting material. At and below -70°C the irreversible consumption of acid is generally very low, primarily because most of the methyllithium is consumed in reaction with the carboxyl group. When the temperature increases, however, the relative importance of the lithium-bromine exchange reaction increases and reaches, for no obvious reason, its maximum at 0°C. The attack on the getn-dibromo moiety is therefore a more temperature sensitive process than the abstraction of a proton from the carboxyl group. If the rationale above correctly describes the course of reaction between methyllithiurn and acids l a - l e , then 2,2dibromo- l -methylcyclopropanecarboxylic acid (10 ) should be more reactive toward MeLi than either acid l e or alkylsubstituted gern-dibroniocyclopropanes. Experimentally, this was found to be the case; when acid 1 0 was ~uixedwith l e and 1,l -dibromo-2,2,3-trimethylcyclopropaneand treated with MeLi, 1 0 was selectively attacked by the lithium reagent. This is borne out by a decrease in the relative amount of 1 0 in the mixtures and by a simultaneous formation of tra17s-20 as the main reaction product (Table 3). From acids 1a - 1d the monobrornocyclopropanes are formed by an acid-base reaction between the lithiocyclopropyl moiety and a proton source. From experiments performed with deuterated reactants at 0°C (Scheme 4) Stein and Morton concluded that the proton was supplied by the carboxyl group in an intramolecular process. This conclusion would have been correct if the lithium intermediates corresponding to 6 and 7 both had been present at 0°C. This is, however, not the case; the only cyclopropyl lithium derivative present at this temperature is 8, which will yield the trnr7s monobromides 9

and 10 as the only products whether 8 is quenched intramolecularly or intermolecularly. In order to shed more light on this problem, acid 1 0 was treated with MeLi at - 115°C and subsequently hydrolyzed, first with DCl/D?O and then with HzO. By comparing glc and 'H nmr analyses of the reaction

+

PC;_. q +

7%;

COOH

mixture the results summarized in Scheme 5 emerge. Most striking is the fact that trans-20 and a second product which from the similarity of its 'H nmr signals is almost certainIy the cis isomer, were both formed as a mixture of 2-bromo-lmethylcyclopropanecarboxylic acid and 2-bromo- l -methylcyclopropanecarboxylic-2-d acid. This clearly shows that the cyclopropyl anions abstract protons both prior to andduring the hydrolysis and, consequently, that the organolithium intermediates live long enough to permit intermolecular quenching. In fact, studies of models of the atlti cyclopropyl anion 11 clearly indicate that only intermolecular quenching is conceivable for this and other intermediates with the same configuration.

From the results outlined above it is evident that one can influence the course of reaction by employing a salt of the acids instead of the acids themselves. Such salts, prepared quantitatively by treating the acids with I equivalent of sodium hydride, were therefore treated with methyllithium in the usual way. When the reaction mixtures were worked up it turned out that the reduced loss of methyllithium did not increase the yield of trans-2-broniocyclopropanecarboxylicacids but rather the yield of other products. Typical examples are shown in Scheme 6: 2-acetyl- I , 1-dibromo-2-methylcyclopropane was formed as a minor product from the sodium salt of 1 0 at 0°C (24), whereas the yield of allene 3c tripled when sodium 2,2-dibromo-1,3dimethylcyclopropanecarboxylate was reacted under the same conditions. Consequently, the sodium salts react less selec-

2076

CAN. J. CHEM. VOL. 62. 1984

R

R=H

COONa

COOH

\

I. MeLi

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

R:CH3

2'

tively with methyllithium than the corresponding acids, but it is interesting to note that the monobro~nocyclopropanecarboxylic acids are formed exclusively in the trans configuration, i.e from cyclopropyllithium intermediates in which lithium coordinates with the sodium carboxylate moiety.

Experimental Gas chromatographic analyses were carried out on a Varian 3700 gas chromatograph equipped with thermal conductivity and flame ionization detectors. The columns were 2 m long and were packed with 3% OV17 on Chromosorb W-HP, 80/100. The ir spectra were recorded on a Shimadzu IR 420 spectrophotomcter. The nmr spectra were obtained on JEOL PMX 60 SI and JEOL FX 90Q spectrometers, using tetramethylsilane as internal standard. Mass spectra were run on a Micromass 7070 H spectrometer, operated in the El mode with an ionization potential of 70 eV. Boiling points and melting points are uncorrected. Melting points were determined on a Mettler FP61 apparatus. Elemental analyses were performed by llse Betz Microanalytical Laboratory, Kronach, West Germany. A PHM 62 pH-meter with a Radiometer SK2401 C glass electrode was used to determine the pK, values of carboxylic acids. Preparation of src~rtingttznterials

2.2-Dibromocyclopropanecarboxylic acid (1b), 2.2-dibromo- l methylcyclopropanecarboxylic acid (1b), and 2.2-dibromo- 1.3-dimethylcyclopropanecarboxylic acid ( l c ) were prepared as described in the literature (13, 25). 2,2-Dibromo-I-phenylcyclopropanecarboxylicacid ( I d ) was made in the following way. Dibromocarbene, from bromoform (43.0 g, 0.17 mol) and 50% aqueous NaOH (10.0 g, 0.225 mol), was added to atropaldehyde diethylacetal (4) (15.0 g, 0.07 mol) under phasetransfer conditions using TEBA (0.50 g, 0.002 mol) as catalyst (26, 27) to give 2.2-dibromo-I-phenylcyclopropanecarbaldehyde diethylacetal (10.6 g, 40%). bp 122- 125"C/0.2 Torr ( I Torr = 133.3 Pa); ir (film): 1602, 1500, 1450, 1370, 1300, 1240, 1200-1000, 695 cm-'; 'H nmr (90 MHz, CDCI,) 6: 1.13 (3H, t, J 7 Hz), 1.22 (3H, t, J 7 Hz), 2.10(2H, s), 3.57 (2H, q, J 7 Hz), 3.65 (2H, q , J 7 Hz), 4.49 (I H. s), 7.34 (5H. br s) ppm; the compound isolated was not pure enough for elemental analysis. A solution of the cyclopropanecarbaldehyde diethylacetal (4.6 g, 0.012 mol) in 80% formic acid (50 mL) was then refluxed for 3 h; after extraction (diethyl ether), washing (aqueous NaHCO,), and drying (MgSOJ, work-up in the usual way left a residue which was essentially pure 2.2-dibromoI-phenylcyclopropanecarbaldehyde (3.4 g, 97%) (25). Jones oxidation (17) of this aldehyde (3.4 g, 0.012 mol) gave 2,2-dibromoI-phenylcyclopropanecarboxylic acid ( I d ) (3.5 g, 97%), mp 12 1 - 122°C (from water) (lit. (25) mp 1 18- 120°C (sublimed)). 2-(2.2-Dibromo-I-methylcyclopropyl)ethanoic acid ( l e ) was prepared as follows. Dibromocarbene, from bromoform ( 1 16.0 g, 0.46 mol) and 50% aqueous NaOH (56.0 g, 0.70 mol), was added to 3-methyl-3-buten-1-01 (20.0 g, 0.23 mol) in the presence of TEBA

(0.70 g, 0.003 mol) (27) to give 2-(2.2-dibromo-I-methylcyclopropy1)ethanol (34.3 g, 58%) by distillation, bp 95-98"C/0.2 Torr; ir (film): 3300, 1730, 1460, 1390, 1 180, 1050,700 cm-I; 'H nmr (90 MHz, CDCI,) 6: 1.41 (3H, s), 1.44 (IH, d, J 8 Hz), 1.52 ( I H, d, J 8 Hz), 1.90 ( 1 H, br s), 1.95 (2H, 2t, J 7 Hz), 3.87 (2H, 2t, J 7 Hz), 4.38 (weak signal, t. J 7.5 Hz), 8.07 (weak signal, s) ppm; the compound gave no molecular ion in ms and was not obtained pure enough for elemental analysis; the ir absorption at 1730 cm-' and the nmr signals at 4.38 and 8.07 ppm are tentatively ascribed to cyclopropane 5. Jones oxidation ( 17) of the dibromocyclopropylethanol (34.0 g, 0.13 mol) afforded acid l e (18.0 g, 51%) by recrystallization from water, mp 94-96°C; ir (CC1,): 3500-2300, 17 10. 1410, 1260, 1220, 1 115, 1020, 695 cm-'; 'H nmr (90 MHz, CDCI,) 6: 1.52 (3H, s), 1 . 5 4 ( 1 H , d , J 8Hz), 1 . 5 9 ( I H , d , J 8Hz), 2.78(2H, s), 9.66(1H, S) ppm; "C nmr (22.5 MHz, CDCI,) 6: 23.1 (CCH,), 26.9(CCH3), 34.4 (CH2), 36.8 (BrCBr), 43.0 (CH2COOH), 176.9(CHzCOOH) ppm; ms m/e: 215(8), 213(14), 21 1(7), 193(19), 191(19), 151(27), 149(29), 133(12), 131(1I ) , 1 1 1(100), 82(73), 81(28). 80(77),79(29). Anal. calcd. for C6HxBrzOz: C 26.50. H 2.94; found: C 26.09, H 3.25. Treartnent of gem-dibrottzoc~cloprop~lacids with tnethgllithiutn; general procedure The reaction was carried out under pure nitrogen. An etheral solution (1.4 M) of MeLi, containing lithium bromide, was added dropwise to a stirred solution of acid in dry ether (15 mL ether/ I0 mmol acid) at and above -78°C and in a 1 1 : 3: 6 mixture of tetrahydrofuran, ether. and petroleum ether (5 mL/mmol acid) at - 115°C. The mixture was stirred for another 1.5 h at bath temperature and was then hydrolyzed with 1.0 M HCI and extracted with ether. The combined organic fractions were dried (MgSOJ. Evaporation of ether left a residue which was analyzed by glc prior to purification by distillation or recrystallization. Reactions of 1 a The acid was treated with MeLi (1.2 equivalents) at temperatures from - 1 15 to 25°C. The glc (column temperature 60- 180"C, 6"/min) and 'H nmr analyses gave the results included in Table I. The products formed at - 1 15OC were, according to 'H nrnr spectra of the reaction mixture, the cis and trans isomers of 2-bromocyclopropanecarboxylic acid (20) (23, 29). At and above -78°C the only product formed was trotzs-20, bp 105"C/ 1.5 Tom: mp 64.5 -65°C (sublimed); ir (film): 3600-2200, 1700, 1440, 1260, 1215, 1020, 945 cm-I; 'H nmr (60 MHz, CDCI,) 6: 1.15-2.30 (3H, very complex m), 3.20-3.55 (IH, m), 9.65 (IH, br s). When the reaction was carried out at -78, 0, and 25°C with 0.98 g (4.0 mmol) of l a , trans-2a was isolated by sublimation in 20, 90, and 30% yield, respectively. Reactions of 1b The acid was reacted with MeLi (1.2 equivalents) at temperatures from - 115 to 25°C; The glc analyses (column temperature 80- 180"C, IOO/min)gave the results compiled in Table I. When the acid (2.58 g, 0.010 mol) was treated with MeLi at - 115°C a l : l mixture (0.35 g, 20%) of the trtrtzs and (evidently) cis isomers of 2-bromo-I-methylcyclopropanecnrboxylicacid (2b) was formed, bp 64-65"C/lO Torr; ir (film): 3500-2500. 1690, 1460, 1405, 1310, 1185, 1040, 690 cm-'; 'H nmr (90 MHz, CDCI,) 6: 0.92- 1.99 (2H. m, cis and trclns), 1.40 and 1.47 (3H. 2s. due to cis and trans, respectively), 3.05 and 3.57 ( I H, 2dd, due to cis and trn~zs, respectively. J 6 and 7.5 Hz), 10.75 (IH. s). When reacted at -78 and 0°C. trrr~ls-2bwas the only product formed. The boiling point and spectroscopic data were in accordance with literature data (14). At -78°C 2.58 g (10 mmol) of l b gave 0.54 g (30%) of trans-2b by distillation. At 0°C 2.28 g (8.8 mmol) of l b afforded 1.21 g (81%) of the same product by distillation. When the acid (1.29 g, 5 mmol) was reacted at 25'C. 0.31 g (35%) of tratzs-2b was isolated by Kugelrohr distillation. In addition, glc analyses (column temperature 80- 180°C. IOO/min) showed that 5 additional products had been formed in approximately 5% yield; these

SYDNES PrND S K A R E

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

products were not identified Reactions (d 1c This acid was reacted with MeLi (1.2 equivalents) at temperatures ranging from -78 to 25°C. The glc analyses (column temperature 80- 180°C. IOo/min) of the reaction mixture prior to work-up gave the results presented in Table I . At -78"C, treatment of 1.36 g (5.0 mmol) of l c gave a reaction mixture from which 0.14 g (15%) of the trcrns isomcr of 2-bromo-1,3-dimethylcyclopropanecarboxylicacid (2c) was isolated by Kugelrohr distillation at 90°C/0.5 Torr. The product crystallized during the distillation, mp 98-99°C; ir (CCI,): 3400-2400, 1690, 1420. 1290, 1180. 600 cm-I; 'H nmr (90 MHz, CDCL) 6: 1.12 (3H. d . J 6 . 5 H z ) , 1.27(3H.s), 1.55-1.94(lH,m),3.73(lH,d,J8Hz), 12.28 (1H. s) ppm; "C nmr (22.5 MHz, CDCI,) 6: 10.00 (CHCH,), 10.54 (CCH,), 24.78 (CH,CH), 25.76 (CCH,), 36.51 (HCBr), 180.33 (COOH) ppm; ms m/e: 194(2, M'). 192(2, M'), 179(2), 177(2), 149(2), 1 13(100). 9 3 3 I ) , 82(6), 80(6), 67(66). Anal. calcd. for Cc,HoBrOz:C 37.34, H 4.66; found: C 37.27, H 5.13. When 1c (2.72 g, 10 mmol) was treated with MeLi at O°C distillation afforded trans-2c (0.56 g, 29%), bp 94"C/0.4 Torr, and 2-methyl-2.3-pentadienoic acid (3c) (0.15 g, 13%). bp 85"C/0.4 Torr (lit. (28) bp 74"C/0.03 Torr). When l c (2.72 g, 10 mmol) reacted with MeLi at 25'C, the yield of isolated trans-2c and 3 c dropped to 25% and 12%. respectively. Reactions of' 1d Reactions between this acid and MeLi (1.2 equivalents) were performed at temperatures ranging from -78°C to 25'C. The glc analyses (column temperature 100-200°C. 20°/min) of the reaction mixture, after quenching with water but prior to work-up, gave the results given in Table 1. Reactions on a preparative scale were only performed at 0 and 25°C. At 0°C acid l d (3.20 g. 10 mrnol) gave trtr1~.~-2-bromol-phenylcyclopropanecarboxylic acid (rrnns-2d). which was isolated in 70% yield (1.69 g) when recrystallized from water. mp 136.5- 137.5"C; ir (CCI,): 3350-2200, 1690, 1420, 1300. 1 180. 1055. 695. 650, 600 c m - ~, .I H nmr (90 MHz, CDCI,) 6: I .75 (2H, t , J 6 Hz), 2.24 (2H, dd, J 6 and 8 Hz), 3.80 (2H, dd, J 6 and 8 Hz), 7.33 (5H. br s), 11.60 (IH, br s) ppm; "C nmr (22.5 MHz. CDCI,) 6: 24.87 (HCH), 27.76 (HCBr), 34.16 (CPh), 128.10 (CH), 128.20 ( C ) . 131.45 (CH), 134.24 (C), 178.37 (COOH) ppm; ms ~ n / e :242(9. M'). 240(9, M'), 224(0.5), 222(0.5), 197(0.5), 195(0.6), 161(6). 160(7), 133(20), 117(13), 116(20). 115(100), l03(21), 91(8). Anal. calcd. for CloHsBrOr: C 49.83%. H 3.73; found: C 50.197'0, H 3.45%. When the same reaction was carried out at 25'C the isolated yield of trans-2d dropped to 58%. Reactiotls of 1e The acid was treated with I .2 equivalents of MeLi at temperatures between - 1 15 and 25°C. The glc analyses (column temperature 80- 180°C. IOO/rnin) gave the results compiled in Table I. Three products. formed in 6 and 4% yield at 0 and 25°C. respectively, could not be identified although there are indications that one of the compounds is 3-(2.2-dibromo-I-rnethylcyclopropyl)-2-propanone. The reactions performed at - 115 and -78°C gave no product except recovered starting material. When the acid (2.70 g, 10 rnmol) was reacted with MeLi at O°C, a product believed to be 3-methyl-3.4-pentadienoic acid (3e) (0.09 g, 8%) was obtained by distillation. bp 1000C/0.5 Torr: ir (film): 3500-2500, 1970. 1740,850 cni - I ; 'H nlnr (90 MHz. CDCI,) 6: 1.79 (3H, t, J 3.2 Hz). 3.03 (2H. t , J 2.3 Hz). 4.70 (2H. m), 10.05 (IH. s); the compound was not obtained pure enough for elemental analysis. An identical reaction carried out at 25'C gave 3e in 6'70 yield. p K,, tnecisuretnents The measurements were carried out by titration of aqueous solutions (25.0 mL) of the acids with 0. I00 M aqueous NaOH. The pK., values were determined graphically from plots of pH as a function of addcd base. The results are presented in Tablc 2.

2077

Renctio11.s of tnet/~yllit/~iutn rvit/l I b / l e and 1b l l . I -dibrotno-2,2,3tritnet/~ylc~~clopropnr~e rni.rrtlrre.s Both reactions were performed in the following way. An etheral solution (5 mL), containing approximately 1.00 mmol of each acid, was stirred at O°C and treated with 1.0 mL of a I .O M etheral solution of MeLi. The reaction was quenched and worked up in the usual way after 40 min. The composition of the reaction mixture was determined by glc (column temperature 80- 180°C, IOO/min). The results are summarized in Table 3. Treczttnent of I b with MeLi and DCI/DZO at - 115°C Acid l b (0.5 1 g, 2.0 mmol), dissolved in a dry mixture ( I 0 mL) of THF, ether, and petroleum ether, was stirred at - 115'C and treated with 2.0 mL of a 1.2 M solution of MeLi. The mixture was kept at bath temperature for 30 min and was then hydrolyzed at the same temperature with DCI/D20 (5 mL, 2.0 M). The hydrolysate was allowed to reach room temperature before 20 mL HZOwas added. The reaction mixture was then worked up in the usual way and analyzed. The gle analysis (column temperature 60- 180°C, 8"/min) indicated that a I .O: 1.0 mixture of the cis and trans isomers of deuterated and nondeuterated 2 0 had been formed in 26% yield. The ' H nmr (90 MHz, CDCI,) analysis showed, by comparing the integral of the double doublet at 3.57 ppm (HCBr due to trt111~-2b),the double doublet at 3.05 ppm (HCBr due to cis-20). and the doublet at 2.44 ppm (HCH due to l b ) , that the reaction mixture contained 16% of nondeuterated 2b in a translcis ratio of 5.0:3.0. Preparation of ether01 sol~ctionsof'sodium salts oj'2,2-dibrotnocyclopropanecarboxylic acids The sodium salts were prepared by treating a 0.20 M etheral solution of the acids at O°C with an equivalent amount of sodium hydride, which was obtained from a NaH dispersion by removing the mineral oil with pentane. The mixture was stirred for 45 min at ambient temperature to complete the reaction. Hydrolysis and adjustment of pH to 1 followed by standard work-up gave recovered acids in quantitative yields (30). Reactions qf t / ~ esodircm snlts of I b nrld I c ivirh methyllithium The reactions were carried out according to the general procedure for treatment of the corresponding acids with methyllithiurn. A 20% excess of MeLi was used. The glc analyses of the reaction mixtures (column temperature 80- 18O0C, IO0/min) gave the results presented in Scheme 6. When the sodium salt of 1 0 was reacted with 1.2 equivalents of Meli, 2-acetyl- I , I-dibromo-2-methylcyclopropanewas isolated in 15% yield, bp 33"C/0.2 Torr (lit. (3 I ) bp 38"C/0.6). I. 2. 3. 4. 5. 6. 7. 8. 9. 10.

I I. 12. 13. 14. 15.

W. R. MOOREand H. R. WARD.J. Org. Chem. 2 7 , 4 179 ( 1962). L. S K A T T E B ~Acta L . Chem. Scand. 17, 1683 (1963). W. R. MOOREand J. B. HILL.Tetrahedron Lett. 4553 (1970). M. S. BAIRD.J. Chem. Soc. Chem. Commun. 1145 (1971). M. E. HENDRICK, and M. JONES,JR.Tetrahedron D. W. BROWN, Lett. 3951 (1973). Tetrahedron Lett. 1371 D. P. G. HAMONand V . C. TRENERRY. (1974). J. Chem. Soc. Chem. Comun. M. S. BAIRDand A. C. KAURA. 356 (1976). M. S. BAIRD.J. Chem. Soc. Chem. Commun. 776 (1979). and L. K. SYDNES.Acta Chem. N. 0 . NILSEN,L. SKATTEBDL, Scand. B, 36. 587 (1982). Acta Chem. Scand. B, 36, 593 J. ARCTand L. SKATTEBDL. (1982). K. G. TAYLOR, W. E. HOBBS,and M. SAQUET.J. Org. Chem. 36, 369 (1971). L. K. SYDNES and L. SKATTEBDL. Acta Chem. Scand. B, 32,547 (1978). M. S. BAIRD and A. G. W. BAXTER. J. Chem. Soc. Perkin Trans. I , 2317 (1979). C. A. STEINand T. H. MORTON.Tetrahedron Lett. 4933 ( 1973). A. K. KHUSID,G. V . KRYSHTAL, V. A. DOMBROVSKY, V . F.

2078

16. 17. 18. 19. 20.

Can. J. Chem. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 01/28/17 For personal use only.

21. 22.

C A N . J. CHEM. VOL. 62. 1984

KUCHEROV, L. A. YANOVSKAYA, V. 1. KADENTSEV, and 0. S. CHIZHOV. Tctrahedron, 33, 77 ( 1 977). K. KLEVELAND. L. SKATTEBOL. and L. K. SYDNESActa Chem. Scand. B. 31. 463 ( 1977). L. K. SYDNES.Acta Chem. Scand. B, 32, 47 ( 1 978). A. R. ALLANand M. S. BAIRD.J . Chcm. Soc. Chem. Comrnun. 172 (1975). R. J . BOATMAN,B. J. WHITLOCK, and H. W . WHITLOCK, JR. J . Am. Chem. Soc. 99, 4822 (1977). D. SEEBACH.H. NEUMANN. and R. DAMMANN. Helv. Chim. Acta, 62, 1 162 (1979). Y. KUSUYAMA and Y. IKEDA.Bull. Chem. Soc. Jpn. 49, 724 (1976). R. N. MCDONAL-D and R. R. REITZ.J . Am. Chem. Soc. 98, 8144 (1976).

23. Y. KUSUYAMA. Bull. Chem. Soc. Jpn. 52, 1944 (1979). 24. C . TEGNER.Acta Chem. Scand. 6, 782 (1952). L . Chem. Scand. 25. K. H. HOLM,D. G. LEE. and L. S K A T T E B ~Acta B, 32. 693 (1978). 26. 1. CROSSLAND. Org. Synth. 60. 6 (1981). 27. M . MAKOSZAand M . W A ~ R Z Y N I E ~Tetrahedron ~C~. Lctt. 4659 (1969). 28. W. RUNGE,G. KRESZE,and E. RUCH. Justus Liebigs Ann. Chem. 1975, 1361 (1975). 29. K. B. WIBERG, R. K. BARNES,and J . ALBIN.J. Am. Chem. Soc. 79. 4994 ( 1 957). 30. C . MARCAILLOU, G. FONTAINE. and P. MAITTE.C . R. Acad. Sci. Ser. C , 267, 846 (1968). 31. R. BARLET.C. R. Acad. Sci. Ser. C , 278, 621 (1974).