Banks, W., Sharples, A., J . Appl. Chem. 16, 28 (1966). Kimura, S., Sourirajan, S., A.I.Ch.E. J . 13, 497 (1967). Kimura, S., Sourirajan, S., Ind. Eng. Chem. Process Design Develop. 7, 197 (1968). KopeEek, J., Sourirajan, S., J.A p p l . Polymer Sci., in press, 1969. Sourirajan, S., Ind. Erg. Chem. Fundamentals 3,206 (1964). Sourirajan, S., “Reverse Osmosis,” Logos Press, London, England, 1969 (in press). Sourirajan, S., Govindan, T. S., Proceedings of First International Symposium on Water Desalination, Washington, D. C., 1965; Office of Saline Water, U. S. Dept. Interior, Washington, D. C., Vol. 1, pp. 251-74. Sourirajan, S., Kimura, S., Ind. Eng. Chem. Process Design Develop. 6, 504 (1967).
curves obtained by surface layer-side and backside operations, especially with films giving greater than 60% solute separation for sodium chloride in low feed concentrations. Details of pretreatment and operating pressure affect the relative selectivity of a membrane for different solutes. I t may be possible to carry out backpressure treatment with brackish waters to give simultaneously potable water and an improved membrane for subsequent surface layerside reverse osmosis op’erations. Acknowledgment
The authors are grateful Pageau, and A. G. Baxter investigations. One of the National Research Council a postdoctoral fellowship.
to Nguyen N. Hung, Lucien for their assistance in these authors (J. K.) thanks the of Canada for the award of
literature Cited
RECEIVED for review November 29, 1968 ACCEPTED April 14, 1969
Agrawal, J. P., Sourirajan, S., Ind. Eng. Chem. Process Design Develop. 8 , No. 4, not published (1969).
Issued as N.R.C. No. 10903.
EFFECT OF TEMPERATURE AND CONCENTRATION ON THE CHLOROHYDRINATION OF ALLY LTRIMETHY LAMMONIUM CH LORlDE C .
1. M E H L T R E T T E R ,
T .
A .
M C G U I R E ,
C .
A .
G L A S S ,
A N D
C .
A .
W I L H A M
Northern Regional Research Laboratory, U . S . Department of Agriculture, Peoria, Ill. 61604
Reaction of allyltrimethylammonium chloride with chlorine in aqueous solution in the temperature range of 25’ to 32’C. gave a nearly quantitative yield of CI 3 to 2 mixture of N-( 2-chloro-3-hydroxypropyl)- and N-( 3-chloro-2hydroxypropyl)trimethylammonium chlorides. Chlorohydrination at 42’ to 46’ C. formed an appreciable quantity of the chlorination product, N( 2,3dichloropropyl)trimethylammonium chloride. A change in concentration of allyltrimethylammonium chloride from 15 to 35% did not affect the course of chlorohydrination at lower temperatures. Proton magnetic resonance spectra established the structure and composition of the products. The improved synthesis should allow more efficient production of cationic starches and ion exchange cellulose.
PARTHEIL (1892) made allyltrimethylammonium chloride
ogous synthesis in which allyltrimethylammonium chloride in aqueous solution was treated with chlorine a t 69°C. This chlorohydrination procedure, although more practical, introduced competitive chlorination, since the isolated product was a mixture of 68% MCC and 32% N-(2,3dichloropropyl)trimethylammonium chloride (DCC). The importance of MCC for producing cationic starches
(Weiss, 1892) react with hypochlorous acid in the cold and obtained a mixture of monochlorohydroxypropyltrimethylammonium chlorides (MCC), as shown below, which were separated and isolated as crystalline platinum chloride double salts. I n 1967, Shildneck and Hathaway described an anal-
+ [(CH3)3NCHzCH=CH*]Cl- + HOC1
-
+
[(CH3)3N-CH2CHOH-CH2Cl]Cl-
+
[(CH,),N-CHLCHCl-CH?OH]Cl-
(I) (11)
VOL. 8 NO. 3 SEPTEMBER 1969
279
(Paschall, 1959; Shildneck and Hathaway, 1967) and ion exchange cellulose (McKelvey and Benerito, 1967) prompted the authors to investigate the effect of reaction temperature and concentration of allyltrimethylammonium chloride on the course of the chlorohydrination. Proton magnetic resonance spectroscopy was used to characterize the products of the reaction. Experimental and Results
Preparation of Allyltrimethylammonium Chloride. Although similar to the patented process of Shildneck and Hathaway (1967), the authors' synthesis is described here. I n addition, however, allyltrimethylammonium chloride in crystalline form was quantitatively recovered. I n a 500-ml., three-necked, round-bottomed flask, equipped with a magnetic stirrer, reflux condenser, dropping funnel, and thermometer, was placed 216 ml. of 25% aqueous trimethylamine (50 grams, 0.85 mole). T o this solution was added dropwise during 2 hours, 65 grams (0.85 mole) of allyl chloride while stirring and keeping the temperature of the reaction a t 20" to 28" C. by outside cooling with tap water. The solution was then stirred a t 40" to 45°C. for 1 hour, cooled, and filtered from about 1 gram of gummy material. The combined clear yellow filtrate and washings of the residue (325 ml.) had a pH of 10.3 and contained 0.60 gram of free trimethylamine. Analysis for ionic chlorine indicated that reaction had occurred to the extent of 97.5% of theory. A 100-ml. portion of the reaction mixture was concentrated in vacuo to dryness to give 34.5 grams of product. Suspension of the crude salt in acetone, followed by filtration and washing with acetone, yielded 33.6 grams of a white crystalline product after drying (96% of theory). ANALYSIS.Calculated for C6H1&1N: N, 10.33%; ionic C1, 26.14%. Found: N , 10.57%; ionic C1, 26.3%. Chlorohydrination of Allyltrimethylammonium Chloride. REACTION AT 25" TO 30" C. The chlorohydrination reaction was carried out in a 600-ml. side-inlet gas-washing bottle equipped with a 60-mm.-diameter fritted disk. A 370ml. aqueous solution containing 55.6 grams (0.41 mole) of allyltrimethylammonium chloride (15%) was placed in the reaction bottle in a circulating cold-water bath to maintain the reaction temperature a t 25" to 30°C., and 32 grams of chlorine was passed through the solution during 2.5 hours. The theoretical amount of chlorine required is 29.2 grams. The solution was aerated to eliminate free chlorine and then removed from the bottle. After the bottle was washed thoroughly with water, the washings were combined with the colorless reaction solution (pH 0.4) and adjusted to a volume of 500 ml. with water. A 50-ml. portion of this solution was concentrated in vacuo to a sirup which was dissolved in 50 ml. of isobutyl alcohol and again evaporated to remove residual water as the azeotrope. The crystalline mixture of N-(2chloro-3-hydroxypropyl)- and N-(3-chloro-2-hydroxypropyl) trimethylammonium chlorides that formed was triturated with acetone. Supernatant acetone was then decanted, and the residue was dried in vacuo a t 100°C. for 2 hours to obtain 7.55 grams of the product (yield 98%). ANALYSIS.Calculated for CsH15ClzNO: N , 7.45%; ionic C1, 18.84%. Found: N, 7.54%; ionic C1, 19.0%. T o determine the effect of increasing the concentration of allyltrimethylammonium chloride on the course of the chlorohydrination, a second low-temperature run was made 280
I & E C P R O D U C T RESEARCH A N D D E V E L O P M E NT
(27" to 32°C.) with 70.6 grams (0.52 mole) of allyltrimethylammonium chloride in 200 ml. of solution (35%). The combined volume of the aerated reaction mixture and bottle washings was 250 ml. A 25-ml. portion of this solution, worked up like the first, gave 9.44 grams of crystalline MCC (yield 96.5%). ANALYSIS.Found: 7.36%; ionic C1, 18.8%. Analytically, the MCC products of both lowtemperature experiments contained no significant amount of DCC and were obtained in high yield. REACTION AT 42" TO 46°C. A similar experiment was performed a t 42" to 46°C. with 174 ml. of solution that contained 28.2 grams (0.208 mole) of allyltrimethylammonium chloride (16.8%). Over 1.5 hours, 17 grams of chlorine gas was led into the solution that was maintained a t 42" to 46°C. by an outside water bath. The reaction mixture was then aerated, and the combined solution and bottle washings had a volume of 265 ml. A 50-ml. portion of the solution was concentrated in vacuo to a sirup in a rotary evaporator and processed in the manner of the previous runs. The white crystalline product that was obtained weighed 6.80 grams (yield 92%). ANALYSIS.Found: N , 7.25%; ionic C1, 18.1%; total C1, 42.1%; calculated total C1 for MCC, 37.70%. Total chlorine and ionic chlorine analyses indicate the presence of 32 to 43%, respectively, of DCC in the final MCC product. To characterize the products in a more definitive manner, a nuclear magnetic resonance (NMR) study was made. Proton magnetic resonance spectroscopy allowed a comparison of the spectra of the several products in solution. For this purpose DCC was required as a reference compound and it was prepared by chlorination of allyltrimethylammonium chloride in glacial acetic acid (Shildneck and Hathaway, 1967). Nuclear Magnetic Resonance Study. The N M R spectra were recorded on a Varian Associates HA-100 spectrometer operating a t 100 MHz with 6% solutions of the products in deuterated dimethyl sulfoxide (DMSO-&, Merck, Sharp & Dohme of Canada, Ltd., Montreal, Canada) and a probe temperature of approximately 33" C. The instrument was operated in a proton lock mode with tetramethylsilane (TMS) as the internal standard. Chemical shifts were reported in parts per million on the 6 scale. Figure 1 illustrates the spectra for allyltrimethylammonium chloride (A), the chlorohydrination product from the second low-temperature run (B), the high-temperature chlorohydrination product (C), and DCC (D). The spectrum of A has the normal pattern for terminal olefinic protons a t 66.3 to 65.5, except that the protons have been deshielded about 0.8 p.p.m. by the proximity of the nitrogen nucleus. The spectrum for compound D has a multiplet a t 65.25, equivalent to one proton. This resonance has a low field position, which indicates that it is produced by the CHCl proton. Decoupling experiments showed this proton to be coupled to two groups of protons. A comparison of the spectrum of product B with those of the known compounds A and D reveals the absence of olefinic protons in B and only a trace of a signal at 65.25, corresponding to the presence of a slight amount of DCC. Signals a t 66.40 (doublet) and 65.95 (triplet) represent the splitting of the methine and methylene protons by the adjacent hydroxyl protons. The doublet is produced by the secondary hydroxyl in isomer I of the equation and the triplet, by the primary hydroxyl in
Conclusions
1
C
A
I
6.40
' I
5.95
I
d
5.25
b
Acknowledgment
I
4.70
The authors thank Clara McGrew for the elemental analyses.
Figure 1. NMR spectra of 6% solutions of four products in DMSO-& A.
B. C. D.
The chlorohydrination of allyltrimethylammonium chloride at 25" to 32OC. affords a mixture of MCC in nearly quantitative yield and of DCC by-product in only slight amounts. At about 45"C. and higher temperatures, from 32 to 43% of DCC is produced, which would be expected to decrease the efficiency of the over-all product in subsequent reactions. The structure and composition of the products were confirmed by N M R spectroscopy. I t was also shown that concentration of the unsaturated precursor was not critical for obtaining optimum results a t the lower temperatures of reaction. The procedure developed shows promise for use in the commercial production of cationic starches and ion exchange cellulose.
Allyltrimethylammonium chloride Lower temperature chlorohydrination product Higher temperature chlorohydrination product N-(2,3-Dichloropropyl)trimethylammonium chloride (DCC)
isomer 11. The ratio of primary to secondary hydroxyl in B is approximately 3 to 2. The spectrum of C has the same general pattern as that of product B, but the multiplet signal a t 65.25 is much stronger and clearly indicates the presence of a significant quantity of' DCC in product C. The absence of the olefinic line position in the spectra of both chlorohydrination product mixtures shows that complete reaction of allyltrimethylammo nium chloride had occurred.
literature Cited
McKelvey, J.B., Benerito, Ruth R., J . Appl. Polymer Sci. 11, 1693 (1967). Partheil, A., Ann. 268, 152 (1892). Paschall, E. F. (to Corn Products Co.), U. S. Patent 2,876,217 (March 3, 1959). Shildneck, P. R., Hathaway, R. J. (to A. E. Staley Manufacturing Co.), U. s. Patent 3,346,563 (Oct. 10, 1967). Weiss, J., Ann. 268, 143 (1892). RECEIVED for review December 19, 1968 ACCEPTED April 28, 1969 The Northern Laboratory is part of the Northern Utilization Research and Development Division, Agricultural Research Service, U. S. Department of Agriculture. Mention of firm names or commercial products does not constitute an endorsement by the U. S. Department of Agriculture.
PLASTICIZING EFFECT OF PERMEATES ON THE SELECTIVITY OF POLYMERIC MEMBRANES NORMAN
N .
L I
Corporate Research Laboratories, Esso Research and Engiwering Co., Linden, N . J . 07036
POLYMERIC membranes have long been used to separate hydrocarbon mixtures. The separation, however, cannot be predicted from the permeation data of single compounds, for the reason that the permeates plasticize the membrane (Li and Henley, 1964; Li and Long, 1969; Long, 1965; Reilley, 1965; Robeson, 1967; Vetter and Kim, 1967). Plasticizing of the membrane changes the solubility and diffusivity of the permeate. This paper presents the solubility results of both single gases and gaseous mixtures and shows how plasticizing effect can be quantitatively defined from the solubility data. Experimental
Materials. All gases were of the C.P. grade made by the Matheson Co. The plastic films contained no additive
or filler. The Teflon film was Du Pont's Teflon F E P with a density of 2.1056 grams per cc. The cellulose acetate film was the reverse osmosis membrane manufactured by Aerojet General, with a density of 0.2362 gram per cc. The polyethylene and polypropylene films were made by Enjay Polymer Laboratory. Their respective densities were 0.9288 and 0.8920 gram per cc. and crystalline contents were 55 and 64%. The polyolefin films (polyethylene and polypropylene films) were preannealed to ensure their structure stability by the procedure described by Li and Long (1969). Briefly stated, the film was submerged in a solvent and held a t 85°F. for 24 hours. The solvent used in annealing polyethylene was isopropyl alcohol. For polypropylene, three solvents (n-heptane, xylene, and isopropyl alcohol) VOL. 8 NO. 3 SEPTEMBER 1969
281
