5472
BERNARD D. COLEMAN AND RAYMOND M. Fuoss [ COSTRIBUTION NO, 1313 FROM
Quaternization Kinetics. I.
THE STERLING CHEMISTRY
LABORATORY O F YALE
VOl. 77 UNIVERSITY]
Some Pyridine Derivatives in Tetramethylene Sulfone *
BY BERNARD D. COLEXAN AND K.%YMOND 11. Fuoss RECEIVED J U N E 15, 1955 The quaternization at 25,50 and 75" of pyridine, 4-picoline and 4-isopropylpyridine by n-butyl broniidc in tetraniethylcnc sulfone (dielectric constant 42 a t 50') quantitatively follows second-order kinetics; a t 25", k z = 4.7 X 8.4 X l./mole min. for these three bases, respectively. The corresponding values of AE are 16.0, 15.95, 15.6 kcal./ and 9.0 X mole; of AS$,-30.0, -29.3 and -30.3 e.u. The quaternization of poly-4-vinylpyridine starts with about the same ratc as the other 4-alkylpyridines, b u t the rate decreases as the reaction proceeds. The concentration range was higher than that at which diffusion of alkyl halide t o individual polymer coils would control the rate. Ultraviolet absorption spectra for 4-picoline, 4-ethylpyridine, 4-isopropylpyridine, 4-vinylpyridine, poly-4-vinylpyridine, 4-methyl-N-methylpyridinium bromide and 4-methyl-N-n-butylpyridinium bromide are reported.
Introduction X considerable part of our research program on polyelectrolytes' has been based on quaternized poly-4-vinylpyridine; during the preparation of these materials, a number of puzzling incidental observations were made which suggested a systematic study of the quaternization of pyridine derivatives. Hinshelwood2-6 and co-workers have studied the quaternization of pyridine itself by methyl iodide in a variety of solvents, and found secondorder kinetics in most cases. We were, however, unable to locate any information on quaternization by n-butyl bromide, which offers certain advantages in the preparation of polyelectrolytes. The purpose of this paper is to present results of a study of the quaternization of pyridine, 4-picoline and 4-isopropylpyridine in tetramethylene sulfone a t 25, 50 and 75" by n-butyl bromide. This solvent has a dielectric constant of 42 a t 50' (sufficiently high to give useful rates) and does not involve side reactions such as appear with nitrobenzene and dimethylformamide. Second-order kinetics were found. Preliminary experiments on the quaternization of poly-4-vinylpyridine under the same conditions indicate that the reaction begins a t about the same rate as that of the simple 4-alkylpyridines; as the reaction proceeds, however, the rate begins to decrease. At extremely low concentrations, this effect might be explained on the assumption that the alkyl halide molecules were uniformly distributed, while the basic nitrogen atoms were concentrated locally in the individual polymer coils; hence diffusion of halide from the bulk of the solvent to the coils might control. But our experiments were made a t concentrations sufficiently high to permit individual coils to overlap; therefore it must be the influence of charged nitrogens on neighboring as yet unyuaternized atoms which decelerates the reaction. I t is not likely that an inductive effect (*) Office of h-aval Research Project NR 051-002, Paper No. 48. Reproduction in whole or in p a r t is permitted for any purpose of t h e United States Government. Results presented in this paper are aiistracted f r o m a dissertation presented b y Bernard D. Coleman t o the Graduate School of Yale University in partial fulfillment of t h e requirements for t h e Degree of Doctor of Philosophy, June, 19.54. (1) R.hi. Fuoss. Discs. F o r a d n y S O C . 11, , 125 (1951). (2) C. A . IVinkler and C. PT. Hinshelwood, J . Chem. SOC., 1147 (1935). ( 3 ) X, J . T. Pickles and C . N. Hinshelwood, ibid., 1353 (1936). (4) R . A . Fairdough and C. N. Hinshelwood. rbid , 538, 1573 (1937). ( 3 ) K . J , Laid er and C.N. Hinshelwood, ?6cd., 858 (1938); K .I. Laidler, ibid., l78fi (1938) (6) H . C. Raine and C. A'. I l i i i s h e l w o u d , ibtd., 137s (1939!.
could be strongly transmitted through a methylene group; possibly the electrostatic field of a qu'a t ernized nitrogen can directly affect the transition state of a nearby nitrogen-butyl bromide complex. \Vork with simpler model substances (a,w-dipyridyl alkanes) is in progress. Experimental Preliminary.-Both for quantitative work on kinetics a n d for the preparation of compounds, i t is desirable to use a solvent in which the reaction is reasonably rapid, one in which reagents and product are soluble and above all, one which does not react with reagents or products. Menschutkin7 and all subsequent workers8 have found t h a t the quaternization reaction goes faster the higher the dielectric constant of the solvent. Methanol has a dielectric constant9 of 32.63 a t 25" and is a good solvent; it proved t o be unsuitable, however, because alcoholysis of the alkyl halide contaminated the product with hydrobromide. Kitrobenzeneg ( e ' = 34.82) aiid dimethyIformamide10 ( e ' = 109.5) likewise proved to be unsatisfactory due to side rcactions. We were unable to locate a n y previous comment on side reactions with nitrohenzene, although several previous studies of quaternization have been made in this solvent; some of our observations are therefore summarized helow. Nitrobenzene (7.3 ml.), n-butyl bromide (1.8 nil.) and 4-picoline (0.8 ml. j gave a practically colorless solution on mixing a t room temperature. On standing, color developed as follows: 30 min., yellowish-green; 60 min., dark green; 3 hr., very dark green and fluorescent (blue-green t o reflected light, red-brown to transmitted light). Potentiometric titration showed about 10:; conversion to quaternary salt a t this stage. Color did not develop on mixing a n y pairs of the above three reagents; Eurthermorc, substitution of pyridine or $-isopropylpyridinc for 4-picoline eliminated the color; the 4-methyl gronp thus appears t o be necessary for colored products. Different lots of C.P. nitrobenzene gave the same sequence of colors but a t various rates; carefully purified nitrobenzene," however, gave only a reddish-brown on standing with &picoline and butyl bromide. The blue-green product appears to be due to iinpurities normally present in C.P. nitrobenzene, because mere redistillation gives a solvent which does not show thc above colors. IVith nitromethane as the solvent for n-butyl bromide and &picoline, fhc mixture bccamc yellow in a few hours and deep orange in a vveek. Diniethylforniamide, used immediately after distillation, give\ no color with a mixture of n-butyl bromide and 4-picoline or Trith S-butyl-4-picoliiiiuin bromide. But a sample ivhich had iiceii distilled 9 tno. prcviously gave a pink color within 3 min. a f t u adding 0.167 g. of the above salt to 2.0 rill. of solvent; within a n hour, the solution was deep orange (peak a t 470 mh). The picolinium concentration was unchanged, accordiiig to the ultraviolet spectrum; hence only trace quantities irere involved in the colored productb. . ~.
~
( 7 ) S .Llenschutkin, %. p i i y s i k . C h e m . , 6 , 41 (18!10). (8) S Glasstone, K . J. Laidler aud H. Eyring. " T h e Theory of Kate Processes," lfcGraw-Hill Book Co., S e w I'ork, S . Y . . 1941, 11p. 41.4423, 449-450, 4.i7. I Il aLi i d 14, A . A . l l a r y o t t a n d E. K . Smith, . V o l l . G , L I C C S
(
iv
G . R . Ixader, THLS J O U R N A L , 73,8.56 (19,>1J. I ) C. I
