COMMUNICATIONS TO THE EDITOR
July 5, 1961
2961
termination requires studies of the kinetics of polymerization in a non-stationary stage. In addition, there is a doubt whether the approach based on the conlposition of the copolymer is reliable in studies of ionic co-polymerization. We now have developed a technique which perI mits the direct determination of the co-polymerizawith dilute sulfuric acid, followed by ether extrac- tion rate constant kl,* in an anionic polymerization. tion, washing with cold 570 sodium bicarbonate The method is illustrated by an example of styrene solution, concentrating in vacuo and recrystalliz- addition to a-methylstyrene polymer. ing with 95y0 ethanol gave 72y0 of 1,4dinitroA solution of “living” a-methylstyrene tetramer butane, m.p. 33-34’, lit. value1 m.p. 33-34’ of a known concentration of “living” ends is (based on starting cyclopentanone) . Treating mixed rapidly in a three-way T-shaped stopcock Ia with base resulted in very little cleavage and ( 2 mm. bore) with a dilute solution of styrene converted it to dipotassium a ja’-dinitrocyclo- monomer in tetrahydrofuran. The mixture flows pentanone. On the other hand, potassium 2,6- through a capillary into a beaker containing wet dinitrocyclohexanone (Ib, n = 3) and potassium tetrahydrofuran which terminates the reaction 2,7-dinitrocycloheptanone(IC,n = 4) underwent instantly. If CHJ is used instead of water, titracleavage in basic as well as acidic media. Dis- tion of NaI determines the concentration of solving I b in hot water, basifying with 85,% potas- “living” ends a t the end of the capillary. Thus it sium hydroxide a t room temperature, acidifying has been shown that no “killing” takes place in with glacial acetic acid, extracting with ether and the reaction. The time of mixing is less than 5% distilling gave 78y0 1&dinitropentane, Its in- of time of flow. The time of reaction, z.e., the frared spectrum was superimposable with t h a t of total time of flow, is varied from less than 0.1 of a authentic 1,%dinitropentane. Similar treatment sec. to less than a second by changing the pressure of IC gave 7576 1,6-dinitrohexane, m.p. 37-39’, of pure nitrogen above the ingredients solutions. lit. value,l m.p. 38-39’. The final concentration of the unreacted monomer The basic cleavage of negatively substituted is determined by measuring the optical density a t ketones has been known for some time. Hauser, 290 mp. To correct for a very small absorption et u Z . , ~ have noted the effect of ring size on the site due to the polymer, a solution of the polymer of the of basic cleavage of negatively substituted ketones. same concentration has been inserted in the second The cleavage of the five-membered compound I a in beam of the spectrophotometer. acidic medium only is the first example of the effect In all these experiments the concentration of of ring size on the pH a t which cleavage occurs. monomeric styrene in the capillary was smaller Extension of this ring opening reaction to the than that of the “living” ends; thus, on the synthesis of substituted a,w-dinitroalkanes, which average less than one molecule of styrene was are a t present unknown or which are accessible added to the “living” a-methylstyrene. The log only with difficulty, is in progress. of the ratio initial styrene concentration over its Support of this work by the Office of Naval final concentration, plotted against time gives a Research is gratefully acknowledged. straight line, a t least for the initial stages of the ‘3) P J Hamrick Jr , C. F Hauser and C. R. Hauser, J . Oug reaction. The slopes of these lines divided by the Chem , 24, 583 (1959). concentration of “living” ends give the respective DEPARTMENT OF CHEMISTRY copolymerization rate constant kl,* (1-a-MS, 2PURDUE USIVERSITY HENRY FEUER Styrene). The data obtained by this method are LAFAYETTE, INDIANA ROYSCOTT ~NDERSON
0 II
RECEIVEDMAY13, 1961
THE DIRECT DETERMINATION OF T H E ABSOLUTE CO-POLYMERIZATION RATE CONSTANT IN ANIONIC POLYMERIZATION’
Sir: The conventional method for determining the absolute co-polymerization rate constants kl,z and k 2 , l is tedious and laborious.2 I t involves two steps, (1) analysis of the co-polymer composition from which one calculates the reactivities ratios, e.g., k ~ , ~ / k l , ~ refers to the reaction w*M1 111 ---All.Afl, whereas k l , z to the reaction .ww111 AI2 -+ ---ATl.l12’): aiid (2) determination of t h e absolute value of the rate constant k l , ~of the respective lioinopolymerization. The latter entity is not easily available and its de-
-+
+
(1) This work was supported by the National Science Foundation through Grant G-11393 and by Quartermaster Corp. Contract D.419129-Q.M:1297. ( 2 ) T. Alfrey. J. J. Rohrer a n d H. hlark, “Copolymerizatiun,” Interscience Publishers, Inc., S e w York, N. Y.,1952.
[“living”ends]X 103 N [styene] X l o 3 d!l kl.2 1. moles-1 sec. -l [“living”endsl X 103 [styrene] X lo3
M
1. moles-’ sec. -l
kl.2
1.3
1.3
1.3
3.0
3.0
3.5
1.1
1.0 1 . 1
2 2
2.6
3.0
1100
10‘70 990
780
780
i80
4.0
4 . 0 5 . 4 7.0
7 . 3 9 . 5 13.0
2.0 3 . 0 3 . 0 3.0 2 , 6 5.5
3.0
565 655
640
665 660
610
645
The increase i n the rate constant for very low coilcentration of “living“ ends was observed also iii the homo-polymerization of “liviiig” p ~ l y - s t y r e n e . ~ The reason for this phenomenon is not yet clear. It is interesting to compare these kl,? values to those obtained for styrene homo-polymerization4 (i.e., k2,2j. A t the same temperature and concentration of “living” ends k l , z is slightly larger than k?,z (c.g., 770 1. moles-’ sec.-l as compared (3) R . K. Graham, D. I.. Dunkelherger and W. E. Good, J. A m . Clzrin. Soc., 82, 400 (1960). (1)C. Geacintov, J. Smid and hl. Szwarc, ibid , 83, 1253 (1961)
2962
COMMUNICATIONS TO THE EDITOR
with 560 1. moles-' set.-'). This may indicate that in spite of some steric strain the electrondonating effect of the methyl group makes the poly-a-MS-anion more reactive toward styrene than the polystyrene anion. It was frequently claimed that for ionic polymerization the reactivity ratio's product r1.r2 is close to unity.5 This need not always be the case. For example, in anionic co-polymerization of a-methylstyrene and styrene a t 25' the k1,1 is about 2.5 1. moles-' set.-' (extrapolated value from data of Worsfold and Bywaters). I n the same units we find kl,z = 770, k2,2 = 5603 and the preliminary data for kz,l give kz,l -20. Thus, r1.r~is probably less than 0.1. Studies of other co-polymerization systems are in progress. The method may be applied also for determining the rate constant of homopropagation as a function of DP (for low-molecular weight polymers). Such a work is also in progress in our laboratory, and i t will be reported in due course.
PI
VOl. 83
H
That partial asymmetric synthesis was achieved in the above condensation is not surprising since this has been observed in mechanistically similar reactions.s However, as will be shown below, changing the solvent medium from toluene to that of N,N-dimethylformamide (DMF) or to nitrobenzene resulted in a complete reversal of sign in the resulting optically active trans-cyclopropane1,2-dicarboxylic acid.= In a typical experiment dry potassium t-butoxide (5) G. M . Burnett, "Kinetics and Mechanism of Polymerization," (6.7 g., 0.06 mole) was added slowly, in order to Interscience Publishers, Inc., New York, N. Y. (6) D. J. Worsfold and S. Bywater, Can. J . Chem., 86, 1141 (1958). maintain the temperature a t 30°, to a solution of DEPARTMENT OF CHEMISTRY C. L. LEE (-)-menthyl chloroacetate (11.6 g., 0.05 mole) STATE UNIVERSITYCOLLEGE OF FORESTRY and ethyl acrylate (5.0 g., 0.05 mole) in 15 ml. of AT SYRACUSE UNIVERSITY J. SMID toluene. After stirring for three hours the reaction SYRACUSE 10, N. Y . M. SZWARC mixture was washed with water and the solvent RECEIVED APRIL28, 1961 was removed in vacuo. The residue was completely saponifieds by potassium hydroxide in CYCLOPROPANES. XI. SOLVENT EFFECT IN boiling ethyleneglycol and yielded, upon acidiPARTIAL ASYMMETRIC SYNTHESIS' fication and extraction, 3.9 g. (60%) of transSir: cyclopropane-1,Pdicarboxylicacid, m.p. 173-174', CramI2Prelog3and their co-workers have studied [ a l Z s-1.5' ~ (c, 7.1, water). The infrared specthe course of asymmetric syntheses in a variety of trum was identical with that of an authentic systems and have found that the configuration sample.'O Treatment of the acid with diazomethof the newly created asymmetric center was ane produced the dimethyl ester, b.p. 94-98' a t dependent on the conformation of the original 15 mm., n% 1.4368 [ a I z 5-2.3' ~ (c, 8.0, ethanol). asymmetric center. The concepts derived from The identical procedure was followed for the exthis work have been applied, with some success, periments in DMF and nitrobenzene. Table I to the correlation of configuration. However, summarizes the results and shows the excellent certain anomalous results recently have been re- reproducibility of the experiments. The marked ported. For example, Collins5 has shown that in increase in optical yield in DMF should be noted. the addition of Grignard reagents to phenylacetoin The mechanistic implications of this solvent the use of phenylmagnesium chloride or phenyl- effect in asymmetric syntheses are currently under magnesium bromide resulted in a preponderance (7) This type of condensation to yield derivatives of cyclopropane of the threo isomer whereas with phenylmagnesium was first described by E. R. Buchman and D. Deutsch [Experie?zlin, iodide the erythro isomer predominated. There- 6 , 462 (195O)j. Currently, this method is being exploited a s a general fore a subtle change of reagents has resulted in a method, see R. Fraisse and R. Jacquier, Bull. SOC.Chim. (France), (1957); L. L. McCoy, J. A m . Chem. Soc., 80, 6568 (1958); D. T. reversal of the stereoselectivity. That tempera- 986 Warner, J. Org. Chem., 24, 1536 (1959). ture also can have a marked effect on changing the (8) F. J. Impastato, L. Barash and H. M. Walborsky. J. A m . Chevr. course of an asymmetric synthesis recently has SOC.,81, 1514 (1959). We also have observed that the Darzens been demonstrated by Pracejus6 We wish, a t this condensation of (-)-menthyl chloroacetate with acetophenone, by saponification, leads to the formation of sodium @-methyl, time, to report that a change in solvent can also followed 8-phenylglycidate, [ a I Z 4 D -0.61° (c, 6.4, H10). Lithium aluminum result in a reversal of stereoselectivity in a partial hydride reduction of the condensation product resulted in the formation asymmetric synthesis. The synthesis involves of a mixture of glycols consisting of 95% 3-phenylbutane-1,3-diol and 5% 3-phenylbutane-1,2-diol. Removal of the 1.2-glycol by periodate a base catalyzed Michael type addition of (-)oxidation yielded the pure 1.3 glycol [CZ]~'D - 9.1' (c 4.2 CHClr). menthyl chloroacetate to ethyl acrylate.' NOTEADDED IN pRoop.-Prof. K. Sisido has informed us t h a t he has (1) This work was supported in part by a grant from the National Science Foundation. (2) D. J. Cram and F. A. Abd Elhafez, J. A m . Chcm. Soc., 74, 5828 (1952). (3) V. Prelog, et al., Helw. Chim. Ada., 36, 308 (1953). (4) For a review of this work, see J. A. Mills a n d W. Klyne, Ch. 5 in "Progress in Stereochemistry," Vol. I. Academic Press, Inc., New Yvrk, N . Y., pp. 19&201. (5) J. H.Stocker, P. Sidisunthorn, B. h i . Benjamin and C.J. Collins, .I. A ~ ZChrm. . SOC., 82, 3913 (1960). ( 6 ) 11. Pracejus, Ann., 634, D (1960).
also observed asymmetric synthesis in the Darzens reaction. (9) L. L. McCoy, J . A m . Chcm. Soc., 82, 6416 (1960), has observed an exciting but much less subtle effect of solvent in this type of condensation. McCoy showed that changing the solvent medium from benzene to N,N-dimethylformamide changed the cis:lrans isomer ratio. In our work we could only isolate minor to trace amounts of the cis acid. (10) K . B. Wibert, R. K. Barnes and J. Albin, J. A m . C h r m . S n c . , 79,4944 (1957). report m.p. 176-177O for the racemic acid. E. Buchner and R. von der Heide, B e y . , 38, 3112 (1DO5), report m.p. 178' for the active acid, [ a I z r D 84.5'.
