". "mem‘M-.. ', l l‘ r l “‘an WWI WI» | l r } 118 231 THS EMULSEGN @‘fli‘s’MEREZAHCN OF STYRENE: ALCOHOLS AS DEAC'YE‘JAYEON AGENTS Thesis far {he Degree sf M. S- MlCHiGAN SYATE CC‘LLEGE icsaph St'egihen Mihina 1948 I - B—r‘wi 7113;»:th It-iichiggm Stag-e University This is to certify that the thesis entitled "Emulsion Polymerization of Styrene: Alcohols as Deactivation Agents” presented by Joseph S. Minina has been accepted towards fulfillment of the requirements for M. 5. degree iLChemi Btry ii i. 5’ ajor PIOfe—dsgl' 7 Date “£27 9 19,48 31-796 I 0 ll ‘. .‘I' ,O l , .- I: {'3}. .,'¥ 0 {a r.‘- V J . . ' -1 ' . .7. I \ ‘ f, i .\ \ \- .. ‘ s ‘ \- .4 ~ I O . l EMULSION POLYMERIZATION 0F STYRENE: ALCOHOLS AS DElCTIVLTION AGENTS By JOSEPH STEPHEN MII-IINA A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department. of Chemistry 191:8 7/ ”if?" C?!‘ ACKNOWLEDGMENT The writer wishes to express his appreciation to Doctor Ralph L. Guile, for his counsel and guidance. W W W 7 H at rt ., , 3*; 3 133-1 Am INTRODUCTION Polyneriaation proceeding in a hanogenous phase, such as in bulk or solution, is not as canplLicated as that in hetero- geneous phase as exemplified by suspension or emulsion polymeri- nation. There are many factors which can canplicate an mllsion polymerization, such as the type of manner, character of the aqueous phase, emulsifier, type and purity of initiator, atmos- phere, presence or absence of inhibitors, rate of stirring, tearperature of reaction and many others which are dependent on the particular system studied. However, it is possible to make a study of the effect of varying a single factor in a systen by keeping all other factors constant or reasonably so during the course of the reaction. The systems studied in this paper are those of the nulsion polymerization of styrene where the atmosphere is varied using nitrOgen, carbon dioxide, and a mixture of nitrogen and oxygen. The other variable factor us the composition of the aqueous phase Ihich was varied by the addition of uater soluble alcohols, such as methanol, ethanol, isopropanol and t-butanol. -1- HISTORICAL Styrene, orginally obtained frcn natural sources, was observed to polylerise in the presence of light and heat by snonl. This was the first observation of what later be- came known as "bulk" polymerisation. Berthelot2 synthesised styrene from acetylene and ben- sene and found that it polynerised on standing. Until recently, the major work in polymerizing styrene was carried out in bulk or in solution of styrene in carbon tetrachloridelo and various hydrocarbonsn. In these hano- geneous phase reactions the polymerization was catalysed by heat3, lighth, salts, such as atomic chlorides or zinc chlorideé, alkali metals? , canine8 , and organic peroxides9 . Early work on polymerisation theory led Standinger12 to suggest that the process occurred in three steps; chain ini- tiation, chain propagation, and chain termination. Shortly thereafter Whitbyj suggested that polymerisation was the step- wise addition of monomer to double bond, however, further studies of the mechanism and kinetics of polymerisation led to . the concept of the formation of active centers fraa which long polymer chains were tamed”. A further modification of the latter theory is that of chain transfer in which growing chains transfer activation energy on collision, causing branching, and thereby termination of their own growth. The molecular weight of the polymer at different stages of the reaction is significant for mechanism and rate studies. Many methods have been used to determine these molecular weights, the most important being emotion", viscoscimetricls, ultrac entrii'ugal16 , and light scattering” all of which give values that are average molecular weights rather than absolute values. The viscoscinetric method for reasons of simplicity of equipnent and operation is generally used for such molecular weight determinations. Standingerlg‘ developed an expression for molecular weight based on viscoscity which has generally been used in reporting average molecular weights. The average molecular weight of the polymer at any stage of the reaction is fairly constant, a fact that has been ex- plained by Schuls2L8 as being due to a constant ratio of chain propagations to chain terminations throughout the course of polymerisation. Heterogeneous polymerisation with particular reference to emulsion polymerisation, was first studied in connection with synthetic rubber. It was observed that a nonmer could be poly- merised in an emulsion system to give a latex fras which the polymer could be coagulated. Styrene later prepared by the caul- sion technique on coagulation gave a powder and not a rubber-like 20 product” . Mark and co-workers further developed the technique of mulsion polymerisation and a theory of mechanism. The emulsion systu consists of aqueous phase, moncmer, mulsifying agent, and initiator. It is preferred to use sane sort of initiator due to the fact that polymerisation takes place in a much shorter length of time, but in the cases of polymerisation by 11 ght or heat it is mitted. Hohenstein20b showed in an experiment that when styrene was placed in one container and water with initiator placed in another container, all under a single bell Jar, polymeri- sation occurred in the water. The explanation was advanced that styrene vapor diffused through the atmosphere and dissolv- ed into the water whereupon polymerisation took place. Further evidence that uulsion polymerisation proceeds in the aqueous phase was given by mung” who showed that the manor was solubilised by soap to give micelles in the aqueous phase prior to any polymer formation. This micellular theory was also used by Hawkins23 and Montrolla‘, who further postulated that poly- merisation began in the aqueous phase and that, beyond approxi- mately 20% , polymerisation occurred in the polymer-monomer (latex) particles. The kinetics and mechanim have been studied by man “rkw325,2h,15b,13ts° and in general the theories state that the monomer is first solubilised and activated by reactim with salts, light, or heat. Thereafter the activated monomer reacts in any of a number of ways among which are; reactim with unacti— vated monaner leading to the formation of long chains, reaction 4,- rith an inpurity or solvent causing deactivation, and dispropor- tionation resulting in deactivation. _ Emulsion polymerisaticns generally have an induction period; this is the interval during which the activated monmers are re- acting with inhibiting canponents until the latter are substan- tially neutralized, whereupon chain propagation becomes the pre- dominant reaction. The inhibiting agent most «manly encount— ered is oxygen26 but it has been shown that with proper care as to oxygen contamination and purity of the reagents used in poly- nerisation, the induction period should approach are”. Breiten- hoch21 polymerized styrene using a nitrogen atmosphere to prevent oxygen inhibition. The mechanisms proposed for chain termination reactions in unulsim polymerisation include local exhaustion of mom-er, dis- proportionation, reaction with solvent, and dissipatim of acti- vation energzsc’d. It has been found that activation energies for emulsion poly- merizations are generally less than those for bulk polymerisa— t1m28,20a. Although it is possible to use heat or light to initiate an anulsion polymerization, it is advantageous to use a water solu- ble per-salt or peroxide since the polymerisation starts in the aqueous phase. Among the best salts are the salts of per-acids, such as potassium persulfate and sodium perborate29, the former being more commonly used. Although the mechanism of persulfate -5- activation has not been solved, it has been demonstrated that the concentration of persulfate decreases very slightly during the course of polymerisation”. Price2513 and Kolthoff27’30‘ have shown through a series of experiments that the rate of polymerization of styrene is dependent on the square root of the persulfate concentration. Yang-31 carried out polymerisations on a system similar to that used by the above authors, and his results showed the same relationship between persulfate concen- tration and rate of polymerisation. The inhibiting effect of oxygen, which has been mentioned previously, has received a great deal of attention. It has been reported that the inhibiting effect results from the reaction of oxygen with activated monomer to fans peroxides26' 31 and poly- peroocides in which the oxygen is actually a part of the polymer32 EXPERIMENTAL 3.282932: smene - The styrene obtained frm Dow Chemical Canpamr was vacuum distilled, and that fraction distilling at 1.143% 0 lh-léms, or 55-58° C a 30-33 nun, n20: 1.5mm, was used. The styrene was used immediately after distillation, or if not used the same day, it was stored in a refrigerator under nitrogen. It was not considered good practice to use styrene that was more than one day old even when kept in a refrigerator since low mole- cular weight polymers form and contaminate the monaner. A test made on styrene which had been stored in a refrigerator for approx- imetely three months showed the presence of alcohol insoluble polymer. Potassium Persulfate - This salt was Merck's reagent grade which was recrystallised before being used in any reaction. It was noted that if the salt was taken directly fran the bottle and dissolved, the pH reading of the solution was 3 while the re- crystallised material when dissolved gave a pH reading of 7.0 a 0.5. The lowered pH was believed to be due to the decanposi- tion of the persulfate while in storage. in experinent conducted to determine the decomposition of potassium persulfate in solu- tion showed that the pH decreased fran 7 to 3 indicating that the persulfate as obtained free the stoclcroon was not pure enough to be used in polymerisation reactions. The procedure for purification of the salt was to make a saturated solution of the salt in water at 25° C and then cooling it in an ice bath for one hour with intermittent stirring to prevent calcing. The crystals were filtered, washed with cold water, and dried at room tmperature. Duponol-G -- This was an emulsifying agent manufactured by duPont and described as "an organic alcohol wine sulfate", the particular alcohol being lauryl. The reagent was used without purification. m -- The distilled water used was refluxed at least four hours with nitrogen bubbling through it to remove the dissolved oxygen. It was then distilled under nitrogen and kept under nitrogen until ready for use. Methanol -- Anhydrous Merck reagent grade was used (BP 614- 65°C), after treating it in the above manner in order to elimi- nate or minimise the amount of dissolved oxygen. Ethanol, USP 95% (BP76-77°C)3 Isopropanol. Bum Kodax (BP 81%—82%°c)3 t-Butanol, Eastman Kodar (BP 81-82°c). These alcohols were also treated as described above. Lluminmn Chloride - Baker and Adamson reagent grade A1013.3H20 Mercuric Acetat. "" Je Te Baker, Ce Po Carbon Tetrachloride - J. T. Baker, C. P. Sodium Chloride - Baker and idsmson reagent grade. The apparatus used for the polymerisation reactions was adapted from that used by Pricezsf and previous workers in .8... this laboratory31’33. It consisted of a four-necked flask with standard—taper ground glass Joints and was fitted with a mercury-sealed stirrer, reflux condenser, thermometer, a tube for introducing gas into the system, and a tube that ex- tended below the surface of the reaction for the purpose of sampling. . All reactions were conducted under a nitrogen atmosphere, except as designated, and at a constant temperature maintained by a water bath controlled to 60°C 1: 0.20. The nitrogen was passed through two gas absorption bottles containing alkaline pyrogallol to renove traces of oxygen. The general formula for the emulsion system based on the aqueous phase before addition of monomer was: 1% mulsifier one part monomer to nine parts aqueous phase potassitmi persulfate to make 0.001714. When the aqueous phase consisted of water and alcohol, both the weight of the aqueous phase and the ratio of monomer to aqueous phase were maintained constant. The particular con- centration of potassium persulfate used was chosen from Yang's31 work since the resulting rate of reaction was convenient for sampling. The emulsion was prepared by first weighing the emulsify- ing agent in the four-necked flask. The equipment was then assembled and flushed with nitrogen. A weighed amount of oxygen-free water was added to the flask followed by stirring -9- in the constant temperature bath until the temperature reached 60°C. The potassium persulfate was added and, as soon as it dissolved, a weighed amount of styrene preheated to 60°C was introduced. The time of addition of the styrene was noted and recorded as sero tine. The stirring in all experiments was at the sac speed using the same stirrer. The canposition of a typical mulsion was: aqueous phase, 678.0 gm; Duponol—G, 6.78 gm.) potassium persulfate, 0.3116 gum; and styrene, 75.3 gm. The technique of sampling was to aspirate duplicate sam- ples from the emulsion at various intervals of time. One sample was used in the determination of the pH of the system by a Beckmann pH meter, and the other caught in a tared flask containing a precooled (-10%), weighed amount of ethanol and 0.1 gm. aluminum chloride. The purpose of the alcohol and the aluninm chloride was to quench the reaction and coagulate the polymer. The chilling of the ethanol to a very low temperature assists in quenching by the rapid reduction of the taperature. The weight of the sample rmoved was detersined by weight difference. The liquid was separated as soon as possible no. polyner by centrifuga- tion and analysed by a nethod indicated below. The polymer was transferred to a flask and washed with ethanol several times by decantation generally allowing ten minutes to an hour to elapse batman decantations depending men the anount and rate -10- of settling of the polymer. It was then washed with water and finally ethanol in the above manner to complete the renoval of emulsifier, water soluble material, monomer, and other alcohol soluble material. The polymer was filtered, partially air dried, oven-dried for 21; hours at 75°C and weighed. The per- cent theoretical yield of solid polymer is determined by: Initial Wt. of Wt. of Polymer mulsion 7: Theoretical Yield of = in e 1e 1 Wei fit of sample 1 100 Solid Polymer Weight 6? §tyrene used In addition, the percent of styrene which polymerised dur- ing the course of the reaction was determined by a titration method which depended on the reaction of mercuric acetate with a double bond to liberate one molecule of acetic acid per double bond. The procedure followed in this determination was to re- move a 15 ml. sample fran the supernatent liquid obtained from centrifugation, to which was added 1; gm. of mercuric acetate followed by 20 ml. of absolute methanol. At this point the solution was agitated in order to dissolve the salts as rapidly as possible, and 20 ml. of a saturated solution of sodium chlor- ide were added. The solution was allowed to stand at least five minutes before adding 20 m1. of when tetrachloride. The acetic acid liberated in the reaction was titrated with standard sodium hydroxide using phenolphthalein indicator. The amount of double bond, calculated as styrene, was obtained from the umber of milliequivalents of standard sodium hydroxide used. .11.. In order to evaluate the accuracy of the titration method an mulsion systan was prepared from styrene, water, and «nul- sifying agent, omitting potassium persulfate. A sample was re- moved and the amount of styrene present in the sample was ob- tained by titration and canpared with the weight calculated to be present in the sample. Samples were also taken at various time intervals to see whether the results obtained were repro- ducible. A further evaluation of the titration schue was made by preparing a system in which freshly distilled styrene was weighed directly into a flask and an emulsion made to correspond to a 1-9 ratio of styrene to water and a 1% concentration of emulsi- fying agent. The weight of styrene obtained by the titration procedure was compared to that weighed into the flask. Another check was obtained by weighing monaneric styrene directly into a flask and then comparing this weight with that obtained frm the titration method. A blank was run in all cases. In order to determine the stability of the potassium persul- fate in solution a system was prepared consisting of 600.0 pl. of water, 6.0 gs. of Duponol—G, and 0.261; gm. (0.001711) potassium persulfate. The solution was maintained at 60°C and samples re- moved for pH determinations at various intervals of time. For the determination of the average molecular weight of the polymers, a 0.01M solution of polystyrene in toluene was prepared. A sample of polystyrene weighing 0.01m gm. was -12.. placed in a 100 ml. volumetric flask and covered with approxi- mately 50 m1. of toluene. The flask was then allowed to stand overnight at 10°C in order to dissolve the high molecular weight polystyrene, after which toluene was added to make 100 ml. at 20°C. A Cannon-Fenske-Ostwald viscoscity pipette, K-lOO, was used to determine the time of efflux of pure styrene and the polystyrene solutions. The. time was recorded by a timer to 0.1 seconds while keeping the pipette in a constant temperature bath maintained at 20°C a 0.1. The specific viscoscity was de- termined fran the following expression: _ Time of efflux of solution at 20°C g 1 ”l 8P " Tfie of efflux of sofvent it 20°0- Using Standinger' 5153 equation the average molecular weight was determined: M M : 01 3P ”lap 00 (05 average molecular weight specific viscoscity C - mole per liter of polystyrene in toluene -13- EXPERIMENTS & DATA Temperature 60° CI 0.2° (Weight of Water 678.0 gms (Styrene 75.3 gms (Duponol G 6.78 gms (Potassium persulfate 0.3116 gms (0.0017 M) based on the aqueous phase 41,. Epth #1 Nitrogen Atmosphere Tine - % Solid % Roasted Average Mole- (Mine Polymer Styrene pH cular Wt. like Do? ""- 6eh 25.2 0.9 15.0 6.h 314.8 21o]. 26e5 —- h17,000 h2.6 -—- 1.9.6 7.0 ----- h9.6 53.9: 69.1 7.0 h9h,000 57.6 67.1.; 714.2 7.0 h66,000 66.0 7h.7 85.2 7.5 h61,000 78.9 82.7 91.8 -- 1439.000 lmoo 91kg """' 803 h553m0 Ehcperiment #2 Carhon Dioxide Atmosghere Time 7% Solid % Reacted Average Mole- (Mins! Polner Styrene pH cular Wt. 3.0 0 19.7 h.9 20.0 5.h 22.2 h.6 25.0 10.7 26.1 14.9 268,000 30.3 18.5 35.1; has 310,000 35.1 26.2 1413.3 11.9 393,000 140.1; 3109 ""'" heé h03,000 ’4700 5101! 67.9 he3 5173000 5206 57e2 77e9 1‘02 “48,000 5900 66s? 81.)} 11.2 1530,” 73.2 76.9 91.7 14.2 h17,000 80.8 79.1 96.6 11.1 397,000 Experiment #3 95% Nitrogen & 5% 0137311 Atmosphere Time a: Solid % Reacted Average Mole- (Mine! Polymer Styrene 2H cular Wt. 21.6 0.2 6.6 3h.0 1.7 7.1 50.11. heh 609 66.3 -- 6.h 8h.9 3.0 7.1 101.h 5.6 6.7 136.8 7.9 6.h 33,000 181e8 15.6 -- h6,000 2h0.0 3h.2 6.2 150,000 299.6 62.8 5.2 111,000 317.1 71“? 11.6 122,000 h30.0 77.7 3.9 122,000 In the following experiments the percentage of alcohol in the aqueous phase is by weight of anmdrous alcohol. Went #11 5% Methanol Tine % Solid % Reaeted Average Mole- (Mine! Poyger Styrene pH cular Wt. 303 o laes 7.2 10.8 0.1; 12.9 7.5 25.0 2.6 15.0 7.3 110.2 12.0 28.1; 6.8 26h,000 118.8 21.6 118.7 7.2 321,000 56.2 38 .9 51.2 7.2 327,000 62.6 115.8 58.1 7.1; 326,000 69.8 56.3 68.1; 7.1; 328,000 714.0 -- 75.2 7.6 332,000 90.0 77.0 86.5 7.8 321,000 .15.. Time (Mine) 3.2 13.h 2h.5 39.9 h9.3 58.6 66.7 73.h 79.8 86.3 92.3 Ayerage Mele- cular'Wt. -16.. Experiment #5 10%'Methanol % Solid. 2 Reacted Polzgpr Stzrgne 2H 0 18.5 7.2 106 19013. 705 509 2101‘ 705 22.h 39.5 7.6 3607 52 05 706 h9.7 6h.2 7.8 6006 7305 705 6803 7906 707 73.6 85.0 7.7 III... 9001 708 8h.3 92.6 8.0 Experiment #6 15% Methanol % Solid. % Reacted ‘Eglzggg Stzrgne pH 0 -... 703 2.7 13.2 7.5 1205 """" 701% 15.6 35.0 7.1 26.0 h3.5 7.2 m 5200 702 h2.5 59.3 7.3 50.5 67.6 7.h 60.9 76.5 7.3 71.6 8h.8 7.h Experiment #7 20% Methanol % Sdlid % Reacted Polymer Styrene pg 0 10.9 7.h 2.3 16.7 7.1 1309 2906 701‘ 18.0 37.h 7.2 2807 181801 7.3 32.9 50.9 7.3 h2.9 59.2 7.h 52.0 66.0 7.3 5602 6907 705 6h.5 77.8 7.5 -.—- 8301 705 192,000 316,000 360,000 h05.000 386,000 383.000 380,000 h01,000 391,000 AyerageiMole- cular'Wte 25h.000 237.000 273,000 326,000 338.000 317,000 3hh,ooo 313.000 Average Mele- enlar‘Wte 2hl,000 282,000 297,000 2979000 303,000 300,000 31h3000 306,000 307,000 10.6 20.1 33.7 89.8 63.6 78.1: 85.8 95.1 117.3 Time (Mine) 8.1 28.2 h2.7 50.8 59.1 68.1 77.8 86.5 101.1 Average Mole- cular W13 0 Erperiment #8 30% Methanol % Solid % Reacted Polyrer Styrene EH 0 23.0 7.6 1.9 26.2 7.8 h.h 25.1 7.7 7.3 27.7 . 7.h 9.5 31.9 7.3 1207 3501‘ 7.3 17.7 h0.2 7.1 22.3 85.7 7.2 23.7 89.8 7.1 29.5 58.9 7.3 36.6 61.3 7.3 h6.2 67.3 7.3 Experiment #9 5% Ethanol % Solid % Reacted Polyrer Styrene pH 0 .0.- 701 0.8 12.7 7.0 1.0 13.9 7.2 '1'.“ 1805 701 29.8 h5.2 7.2 h7.6 63.3 7.3 62.5 78.6 7.h 72.2 86.1 7.h 80.8 92.1 7.5 8507 9606 .- Experinent #10 10% Ethanol % Solid % Reacted ngyrer sryrene pH 0 6.6 5.0 6.5 18.h 6.7 30.9 6.6 39.3 6.5 89.7 6.8 61.0 6.5 68.1 6.6 80.1 6.7 .17- 138.000 118,000 107,000 lh8,000 122,000 125,000 126,000 121,000 Average Mole- cular'Wtri 233,000 352,000 393,000 387,000 388,000 382,000 371,000 AverageiMole- 0111” Wte 188,000 288,000 350,000 3330000 350,000 328,000 361,000 3hh,000 Thee M 0.0 36.0 89. 7 62. 7 78. 3 95.5 111.2 157.0 Time (Mina) 3.1 30.5 no.2 50.6 61.1 71.7 81.9 92.5 106.5 123.0 Time (Mina! 1.9 10.0 30.3 85.6 60.1 75.2 95.9 115.6 136.3 155.0 175.1 203.3 266.6 martini: #11 10% Ethanol % Solid % Reacted Average Mole- Pglyrer SEyrene BE cular‘Wt. 0 6.7 0 6.5 13.0 6.7 205,m 25.3 6.6 320, 000 82.3 6.7 806, 000 60.7 6.7 817, 000 79.6 6.9 37S,W 87-3 6.9 293,W Experiment #12 15% Ethanol % Solid % Reacted Ayerag01Mole- Polymer Styrene pH cular‘Wt. 0 12.8 6.8 h.5 18.3 6.9 155,000 11.0 21.h 6.7 217,000 21.3 29.9 6.6 2hh,000 28.5 h1.9 7.0 262,000 37.8 51.2 7.1 266,000 88.5 59.9 7.0 275.000 58.6 70.2 -— 263,000 65.h 7h.0 7.7 288,000 76.9 88.1 7.7 275,000 Egperheent #13 20% Ethanol % Solid. % Reacted AyerageIMole- Polymer Styrene EH cular It. 0 13.6 7.1 0.9 12.5 7.1 1.7 16.8 7.1 703 2008 700 803 2700 702 711,000 11.7 30.h 7.0 81,000 lh.7 30.0 7.0 105,000 1907 ' 3908 609 100,000 --— h0.8 7.1 92,000 32.h hh.3 7.1 101,000 37.6 h7.8 7.2 103,000 h2.8 58.6 7.2 107,000 60.h 69.8 7.1 113,000 .18.. Tine (Minsk 5.2 9.3 31.3 83.8 57.6 71.8 85. 100.6 117.0 133.5 158.2 170.7 190.1. 211.1 253.6 (Mine) 8.1 20.7 28.5 no.0 53.3 67.2 75.8 83.3 99.7 112.7 127.3 Experiment #18 20% Ethanol % Solid % Beacted Average Mole— PolEer Styrene EH cular Wt . 0 11405 705 0 16.3 7.7 1.7 20.1 7.11 5.2 26.3 7.8 6.9 27.0 7.5 9.3 28.3 7.11 11.6 32.5 7.8 70,000 18.0 33.0 7.h 70,000 15.6 3h.h 7.5 76,000 18.8 314.2 7.11 75,000 25.1 37.2 7.14 93,000 31.0 113.6 7.6 35,000 30.0 ‘ 116.5 7.3 85,000 31.1 h6.7 7.2 85,000 39.3 56.6 7.0 79,000 Experiment #15 5% Isoprepanol % Solid % Reacted Average Mole- PolEer Styrene EH cular 1It. 0 7.3 007 7011 0.9 7.3 1.7 7.3 h.8 7.h 6.3 7oh 11.3 7.3 103,000 19.0 7.2 1h6,000 29.9 7.3 20h,000 5003 703 280,000 59.8 7.8 33h,000 Time (Mine) 14.17 15.0 30.0 36.0 178.3 53.3 63.3 73.3 86.6 93.3 103.8 Experiment #16 10% Isoprqpanol % Solid % Reacted Average Mole- Polyrer Styrene 2H cular Wt. 0 7.6 109 705 102 707 2.5 7.7 3.1 7.5 2.5 7.5 hol 70h h.6 7.6 hl,000 7.1 7.3 h0,000 9.3 7.5 h3,000 28.2 6.8 h7,000 Experiment #17 5% Tertiary Butanol % Solid % Reacted Average Mole- Polyrer Styrene 2H cular Wt. 0 m- 703 102 1908 7.3 3.6 20.5 7.h 8.7 23.0 7.h 180,000 13.5 30.8 7.h 233,000 2501 14209 705 3133000 311.9 57.1 7.5 3117.000 52.6 70.h 7.5 362,000 70.5 81.7 7.6 376,000 """"' 8502 707 355,” 75.0 90.8 7.7 353,000 .20- Experhnent #18 10% Tertiary Butanol Tine % Solid (Mine) Polyrer 3.2 0 1500 0.8 26.6 2.1 80.0 3.2 65.0 5.1 83.0 8.1 110.0 8.7 159.0 9.7 196.0 9.6 235.0 16.2 280.0 20.8 380.0 23.6 Experiment #19 Sample # Minutes 1a 3.2 lb 2a 15.0 2b 3a 9000 30 8 121.5 5 231.3 Egperiment #20 Sample # Minutes 1 9.0 2 26.5 3 80.6 % Reacted Average Mele- Styrene EH cular Wt. 19.9 7.8 18.6 7.5 ...... 705 22.6 7.8 26.8 7.0 82,000 28.1 7.0 87,000 30.2 7.0 80,000 36.0 7.1 83,000 37.8 7.2 50,000 83.2 7.3 80,000 87.0 7.2 51,000 50.0 7.2 58,000 Comparison in an emulsion system with no potas- sium.persu1fate of the calculated weight of styrene vs the observed weight obtained by titra- tion of the double bond. moulated mserved % Styrene ‘Weigrt 15231W°15ht15231 ‘p__ Present 1.80 1.20 6.9 85.7 1.80 1.21 86.8 1.10 0.90 7.0 81.8 1.10 0.89 80.9 1.90 1.67 7.1 87.9 1.90 1.66 87.8 1062 10148 609 91014 1.78 1.85 6.8 81.5 This is a duplication of the above experiment using a new emulsion sample. Calculated Observed % Styrene Weight (25) Weight( 32 pg Present 2.07 1.92 7.0 92.8 2.85 2.25 7.0 91.8 1.32 1.09 6.9 82.6 .21- Experiment #21 Il'his experiment is a variation of the above. Instead of determining the weight of styrene in an aliquot portion of a large sample, two samples were made up by weighing out styrene, water and Duponol-G in the proper ratio and then using the whole sample to determine the weight of styrene by titration. Sample # Wt. Styrene Wt. Calculated % Styrene Added (es) Fran Titration 1 1.11 1.08 97.1.; 2 1.11 1.08 97.1.1 Herman #22 To determine the effect of water and mulsi— fying agent on the determination of weight of styrene by the titration method, two samples were made «sitting the water and usulsifying agent. Pure monomeric styrene was weighed out and then applying the procedure for determin- ing the amount of styrene by titration, the weight of styrene was found and canpared. Sample # Wt. Styrene Wt. Calculated % Styrene Added 1mm Titration 1 1.08 1.02 98.8 2 0.83 0.78 98.0 Experiment #23 Decomposition of potassium persulfate vs time followed by pH. System was standard, i.e., except for styrene. me (M11180 ) EH 0 8 .52 25 8.32 57 8.00 85 7.60 161 5.31 178 8.72 362 3.28 -2 2.. Time (Mine 3.0 11.3 22.2 35.1 82.2 89. 56.5 63.8 71.0 91.0 Time (Mint) 3.0 12.6 22.2 28.9 35.2 82.6 89.8 56.5 68.8 72.8 91.0 Experiment #Zkg Nitrogen Atmosphere % Solid % Styrene Average Mole- Polmer Reacted EH cular Wt. 0 --- 7.2 007 """" 705 2.9 16.7 7.5 16.6 35.0 7.6 339,000 - 11700 706 1107,000 14308 6006 707 397’w0 Shah 7108 ’ 708 397,000 68.3 79.3 7.8 828.000 76.6 86.0 7.8 386,000 8000 91408 709 37730“) Experiment #25 Nitrogen Atmosphere % Solid % Styrene Average Mole- 2252255. ‘Reacted EH cular It. 0 m- 609 006 507 702 2.5 8.1 7.8 9.8 16.8 7.2 189,000 16.9 23.2 7.2 257,000 29.7 38.5 7.1; 311,000 81.8 52.6 7.7 353,000 -"'""" 6h02 707 1116,000 55.7 77.1 7.8 802,000 76.7 83.8 7.9 803,000 88.1 96. 7.9 362,000 Data from P. T. Yang's polymerisation of st 0 in a water, Duponol-G, and potassium persulfate system 1. The concentration of potassium persulfate was 0.001711. Tine i!1222 0 60 90 100 110 125 150 225 . % Solid PolEer 0 0.62 25.1 39.8 58.1 70.0 87.6 91.1 % Styrene Reacted -23. .181.. 6.9 6.8 6.3 7.1 7.8 7.6 6.8 7.5 Average Mole- cum Ute 336.000 378.000 376,000 378,000 353,000 310,000 5:0 352.» 3.2L. as one o: 00. a. 0. Or 00 on or 0' o o 9 O. Q 7.: 001:... on .z.$<$0.m 02E om modem {6.00.0 0.00 mink 36:33.5; 51002.2: 0: q t 1.320053. 0.“ 00 000 .00} 0e oooeor 000:0»... co 1 .0 .u no. °(. ‘6 O ygwivod 07409 '9 0313033 Jusyils 00‘ 7.2.2.312... 06 0r 0 T. 0- EEioenm; Marx .O moon is 710.00 0 or if: 0 z13 wnod one; 4 ' nevus ye cal—way a V AHZZLV milk M....1.\& or Gen a... 9 a on a; o... 0% e O 0 so «.0. 2666.0 000.? o o 2 0.0.0 .mEuL. A v 00010? omen tuenxo New. IIQO{L_Z ON Seminooxt< nfluuuflauuxu teCnNQmZEUefilo :15 on .0m. 0 3. E3 3. . OM 0005. 00 on on W. or on. 8138441007 OI‘IOS‘Oé o: n: 10 exam in: O“M.r sneer; 05C. ooh one 004 3.. re; a: o! 0: or H: he s 38¢. do ZoCcuatuxeeLied no MLQM on or 00. yak/(1007 (”'10 c; 04, 6000:") HOCOro 50f F713 A VEKAGE (‘1 out Cu LAR W’lvour AOLV '19 [00001.3 411137171121“ f1.)8.1§01-\L( “$5.1le OF POL75F,'KEN[ L42 618105711556 4 0 N2, \o 70 L0 30 H0 "0 60 7o 90 70 Z Poevmtmz 477071 F12. 5 «02.53 .38; or on one oi. on. 08 2.878818%. d1 MWCIm mDOmD o, E soarini 8m m.o .m i: X 6.00.0 0.60 .193; 1003:. c Lennione/x In J, ) Hx rflu.l,iw_.io,< 11 ,0 1- (M MK OOOJdm anew . 000 .mdm 06C 0k 90 OOONNm mmru om Ofi O) on Oh 0D or Do. UJWAt‘Vod 01109 00, 0.413ij ENQLMJ C “143 V d 2.1 Cried/1r... Q.) 0m J... Oh )0 be 02 3. on o: _ 3- Ohx O. tie/Conn; 120.: silk 24003.0( N. am: on“ 0‘ .h I. note: mt moon): {6.00.0 uooq mine 5. tnoozn noiai Nwamll who. m an. «(idol (with town 00080: oODch 00031.“ q me.“ n m 0 an 0001.20. flesh n n oi om 0m. or 0% Om am or 00. I ‘10“ E Nate/410d 0| 0 tray FUJUJJS ‘80 0:115 17 32.2» ucéh. ma; CT on. a: 93 o» o» ox. 3 0% 2. Ga 0d 2 O 4‘ O 0 v 7 o .EZwfimm; THIS; 95me 5:. 94% _ ,naom3a< ODONmm. 0 fl \ 3: S 40:5.qu emf: om no d. av. {2.30.0 0000 £qu 0m mmuzznori< 5903.: w 1:32 Emu: 2. zol3m m0 MES. Oh ooo‘rrm Us 000 .m‘m Hr SN Ob RS 8 ow 00. ABM/nod (wt-10904 (13.1%?“ 9N3UAJ.S% O V 3:33 NSC. r om Ofix OT‘ 3: 00» Q. Z__>ZN..AG.. ”mutt mmw sum com 03 :thmmy; “3.35 “.02: “uh: Eifiud . «.3 ”93m. bo amax 5.586 Vooo mZmF Damiano-{#1 zmoow‘tz thqNEmifom We mtg q :8 met (\H .L {SE u>¢bu 0009;” F G. 5 0: or. 0w. be. 0m 0» or 0% C O fiméu Chi 0006.: 000 @mm 3% 089nm 2.0m 08 in $3 89:; Ono .@ ‘ oooJ¢r 000 N; NW mo 6N 000 3r . My 0 ooo$nr A: QM QM 0m. Gm. 3 Oh om or no. (val/v4 70;! m 705' 04 THEORI AND DISCUSSION The starting point for the growth of polyleric chains has frequently been postulated to be activated manners from which polymeric chains are built up by the addition of single monomeric units. This particular growth of chains is inhibited by the presence of air, or more exactly, by oxygen. It was postulated, and later shown to be true, that the oxygen reacts preferentinfly with the activated nonmu' to fern peroxides26’ 27’30. Breitenbocth performed the emulsion reactions under nitrogen and obtained more consistent results. In no case was there any nention of the fact that the carbon dimcide of the air had any effect on the polymerizing system. In order to detemine whether or not carbon dioxide had any effect on an uulsion polymerization system, aperi- nents were prepared conducting the reaction under an atmosphere of carbon dioxide and comparing it with these reactions conducted under an atmosphere of nitrogen, and a mixture of 95% nitrogen and 5% oxygen respectively. The rate of polymerization and the induction period for Experiment #1 (Figure l, nitrogen atmosphere) and Experinent #2 (Figure 2, carbon dioxide atmosphere) are alnoet identical, in contrast to the effect of Experiment #3 (Figures 3 and 1;, oxygen atmosphere) which is to definitely decrease the rate of polymer- isation. -2h- The most pronounced effect is on the average molecular weight. The polymer from Experiment #3 (oxygen atmosphere) was over 300,000 units lower than those of Experiments #1 (nitrogen atmosphere) and #2 (carbon diouide atmosphere). This lowering must be attributed to the fact that the oxygen acts as a chain terminator and deactivator. The average molecular weight of the polyner from Experi- ment #2 (carbon dioxide atmosphere) averaged 30,000 units less than that free: mperinent #1 (nitrogen atmosphere). This lower- ing is probably due to the decrease in the pH of the former re- action resulting.fro| the formation of carbonic acid. Since pre- .tone are known to act as chain.terninators, a lowering of the molecular weight would.be expected. However, since carbonic acid is a'weak acid, it liberates only a few protons, thus accounting for the rather small lowering of the average mole— cular'weight. The pH lowering toward the end of the polymerization in Experiment #3 (oxygen atmosphere) may be attributed to the de- composition of the potassiun.persu1fate to potassium bisulfate which is acidic. Ordinarily it was found that in a reaction under a nitrogen atmosphere, not exceeding two hours in duration, the potassium persulfate could be considered substantially stable, but on standing at 60°C the decomposition'becones apparent and appreciable. -25- Since the styrene in an emulsion polymerization ferns peroxides in the presence of oxygen, a sample of the polymer obtained from Experiment #3 was tested'by a procedure des- cribed'by'Nosacki3h. The method consists of dissolving the polymer in acetic anhydride containing some chlorofonn. Sodium iodide is added and the solutions allowed to stand fro S to 15 minutes. The development of a yellow color due to iodine indicates the presence of peroxides. The poly-er 'which was tested fouled a yellow colored solution showing the presence of peroxide in the polymer. Since emulsion.polymerization is initiated in the aqueous phase and the polymeric chains grow by addition of mononer, also in the aqueous phase, an increase in the solubility of moncmer in the aqueous phase might tend to produce polymers of longer chains by making more monomer available at the site of the growing chains. In order to increase the solubility of monomer in the aqueous phase some solvent which dissolves styrene should be added to the aqueous phase. In the case of solution.polymerization the solvents must be such that both the monomer and polymer are dissolved, and therefore hydrocarbons, such as toluene and'benzene, were used for this purpose. The use of alcohols was reported only in one casegb . They proved ineffective since, although the monaner ‘was soluble in the alcohol, the polymer was insoluble and would precipitate in the course of polymerization. —26- In emulsion polymerization, hydrocarbons cannot be used due to their insolubility in the aqueous phase. Since a compound dissolving styrene and also soluble in water was necessary, the saturated alcohols with low molecular weights appeared satis- factory. The high molecular weight polymers may tend to pre- cipitate in the presence of these alcohols, but in an anulsion polymerization the mlsifying agent keeps the polymer in suspen- sion. Use of an alcohol-water solution as the aqueous phase should increase the solubility of styrene, but as a consequence intro- duces another variable into the system. The alcohol may, in addition to increasing the solubility of monomer, act as a chain terminator or deactivator causing a decrease in the molecular weight. Since the alcohols are soluble in both monomer and water, they would distribute themselves between the two phases, but would be concentrated principally in the aqueous phase because of the larger volume of the latter. Since dilute solutions were used, the. amount dissolved in the nonmer would be mall and therefore its effect as a solvent or contaminant of the mm was considered negligible. It has been shown (Experiments 5-18) that the effect of using various water soluble alcohols in the aqueous phase of an emulsion polymerization was to lower the molecular weight and reduce the rate of overall polymerisation. The rate of overall -27.. polymerisation, as considered in this paper, is the percent theoretical yield of solid polymer formed per unit time. This rate was calculated for the "straight line" portion of the curves of percent theoretical yield versus time. This percent theoretical yield of solid polymer is termed "% solid polymer" on these graphs. Florle" was the first to suggest the possibility of chain transfer in polymerisation whereby growing chains may be termi- nated by collision with solvent molecules. The activated solvent molecule may then initiate another chain by colliding with a nononer molecule and activate it or cause branching by colliding with a long chain. It has been shown by many workers using non-catalysed solu- tion polymerisation35 that while the rate of formation of polymer was slower than that of bulk polymerisation, the rate of chain growth was not affected by the solvent. The degree of polymerisa— tion, however, varied markedly with the solvent used. Mayo36 interpreted the work of Suesa‘u‘ and Schulz25b on solution polymerisation on the basis of chain transfer. Although Mayo was dealing with non-catalyzed polymerization of styrene, his arguments may be applied to emulsion polymerization if the effect of the potassium persulfate is considered negligible due to its relatively small concentration. He states that the growth of polymer chains proceeds by addition of styrene units as follows: .28.. (1) RI 4 C6HSCH -.-. CH2 -—-) (R-CHZ -CH-C6H5)* (a) designates an activated.molecu1e The reaction can then be terminated by one of the three followh ing mechanisms: (2) 2(R-CH2-CHC6H5)* ——> R-CH as CHC6H5 4 R-caz-CHz-cgns ( 3) R-CHz-ga-C6H5 + c112 :- CH-06HS ——> R—CHZ-CHz-Céfls - CH2=§H (h) (R-CHz-CH-06H5)* + S-H --9 5* + R-CHz-CHz-Céfls Reaction (2) has been shown to occur but only to a mall extent 15°. The importance of reaction (3) has not been determined. Reaction (1:) is that with solvent and is considered in detail in this paper. . If the activated solvent molecule reacts readily with styrene monomer, another chain would be started. ‘Were this the only effect, the result would.be a decrease in average molecular weight with no decrease in the overall rate of polymerization. If upon formation the activated solvent molecules do not react readily with.monomer, they will accumulate in the solution and, by-a reaction analogous to (2), destroy the reaction chain. The effect would.be to reduce the overall rate of polymerization. and the average molecular weight. The effect of reducing the overall. rate of polymerization may be approached from a different point of view. If an acti- vated monomer collides with a solvent molecule and this molecule absorbs part or all of the energ of activation, that monomer 'would.be effectively terminated. -29- The solvent molecule in absorbing the energ of activa- tion need not be activated to such a degree that it must react by a mechanism analogous to (2) in order to be deactivated, nor need it transfer the activation energy by reaction with mononer. It may instead, absorb the energy and retain it, re- maining inert insofar as the polymerization reaction is con- cerned. The net effect of such a process could be called de- activation. If deactivation occurs, the overall rate of polymerization would be decreased and the average molecular weight would be re- duced; the latter resulting from the fact that yowing chains would be terminated by solvent before a high molecular weight was attained. The lowering of the molecular weight would depend on the concentration of solvent since the probability of temina- tion increases with increasing concentration of solvent. Norrish and Smith-37 in the study of velocity constants of certain bimolecular reactions in solution observed that some sol- vents decrease the rate of reaction as carpored to the same re- action in the gaseous state. The decrease in rate was explained by the collision of reactants with solvent molecules. A line of reasoning similar to that of the above authors was applied to reaction rate calculations in this paper. Essentially the theory of reaction in the gaseous phase states that in order to have a reaction betwun two molecules it is necessary that they collide with a combined energy equal to -30. or greater than, E, the energy of activation. The proportion of molecules for which the kinetic energy exceeds, E, is”: {umber of effective collisions g 9' 1%. Total number of collisions Norrish and Smith following the method of Luis”, calculated the average rate of collisions occurring betwoen like stuns or molecules in the gaseous state by the equation: (5) 2 = 5 77 0’35. 112 where: Z - number of molecules entering into collision per cc per second a": diameter of the molecule u : root mean square velocity u (BET/Mfi n = number of molecules per cc Designating the rate of effective collisions as ”k", on expres- sion is obtained which can be used to calculate the rate of reaction betwoen like molecules in the gaseous state: (6) k :- 71'!an (QT/M); e. gr where k 3 number of effective collisions per sec per cc Substituting (5): (7) k : Z 0. ET This equation when applied to reactions in the gaseous state gave good agreement betwoen the observed and calculated rates. Norrish and Smith applied this equation to reactions in solutions assming that the solvent acted only as free space. They found that the rate of reaction in sale solvents was not affected as compared to that in the gaseous state, while other -31- solvents lowered the rate. A prdbability factor, P,'sas in- troduced to account for the discrepancies that occur between the observed and calculated rates due to factors which are not taken into account in calculation of the term, Z. The ex- pression (7) then becomes: E (8) kn P 2 e’ RT In the cases which Norrish and Smith studied the proba- bility factor was used to compensate for the deactivating effect of the solvent. They considered this effect to be con- nected with the removal of energy of activation at the moment of collision, the amount absorbed being stated to be dependent on the molecular structure. For collisions between unlike molecules the following ex- pression was obtainedbyHinshelwood39 for gaseous phase re- actions: E (9) k . NANB ag‘faan/MA . 1%)] § ." m where: k = number of effective collisions per cc per second number of molecules of A per cc NA NB : number of'molecules of B per cc 0TB “A mean molecular diameter, cm. molecular‘weight of‘A MB 3 molecular weight of B E 3 energy of activation, cals/mol -32- In the cases studied in this paper the polymerization of styrene in a.non-alcoholic water phase was used as the "standard" of comparison. Any reduction in the rate of polymerization and average molecular weight would be attributed to the addition of alcohol to the aqueous phase. It is assumed that any effect 'which the presence of water.may have on the reaction is constant in all the experiments due to its relatively great concentration as canpared to am one of the other components in the system. The results of this investigation show that the overall rate of polymerization is reduced by the addition of alcohols to the aqueous phase (Figures 6-11, 13-18, 30, 22-2h), indicat- ing that active centers are being removed from.the reaction. Since, in order to effect a deactivation, it is necessary for an active center and an alcohol molecule to collide, the rate of effective collisions may be calculated from equation (9). The implication is that two reactions are proceeding annul- taneously in the alcoholic emulsion polymerization system: 1) the reaction of growing chains with.monomer to lengthen the chain, and 2) the reaction of activated monomer with alcohol to remove activated centers. The rate of styrene polymerization was determined for both water and alcoholawater solutions. The observed decrease in the rate of polymerization of the alcohol- water systems is ascribed to the difference in rates of the two competing reactions. Calculation of the theoretical rate of deactivation by alcohol using equation (9) gives results inconsistent with experimentally determined deactivation "reaction" rates. If it is assumed that rate of polymerization and rate of deactiva- tion are independent and that deactivation occurs by reaction with activated monomer, it is possible to reconcile the cal- culated and experimental rates by the use of a specific prob- ability constant, P. This prdbability factor has been found to have a reasonably constant value for each of the alcohols (Tables I, II, III, IV) investigated. The deactivation effect of the different alcohols based on the calculated values of P was found to decrease in the order isopropanol)t-butanoDethanol) methanol. A similar study by Norrish and.Smith using the reaction of trimethylamine and p-nitrobenzyl chloride in methanol and ethanol solutions, gave calculated values for P which are re- markably close to those reported in this paper. It is rather difficult to postulate a.mechanism for deacti- vation. It may, however, be looked at in a qualitative way. At equi-molal concentrations the calculated rates for t—butanol and methanol are substantially identical. Such being the case, the larger deactivating effect Observed for tabutanol may be attributed to the greater ability of the larger structure to absorb activation energy on collision. TABLE I Methanol Solution 28 MOlS h h 5 Z x 10' Alcohol FKZ x 10 Kc K x 10 P x 10 20,h89 l.h 2.h 2.2h 0.7 3.2 h0,969 2.8 2.3 h.08 0.8 1.96 6l,h10 h.2 1.9 6.61 1.2 1.81 82,178 5.6 1.8h 9.11 1.26 1.38 121,859 8.h 0.5 lh.h6 2.6 1.79 Average 2.03 TABLE II Ethanol Solution Mole Z x 10"28 Alcohol K2 x 10h Kc K x 10h P x 105 16,708 0.97 2.15 1.90 0.95 5.0 33,u1u 1.9h 1.9 3.8 1.2 3.2 50,121 2.91 1.53 5.72 1.57 2.8 66,828 3.88 0.35 7.6h 2.75 3.6 . Average 3.6 TABLE III Isopropanol Solution Mols Z x 10""28 Alcohol K? x 10h Kc K x 10h P x 105 12,610 0.7h 1.5 1.hh 1.6 11.1 25,220 ' lens 0005 2086 3.05 1007 Average 10.9 TABLE IV t-Butanol Solution Mole Z x 10"28 Alcohol K2 x 10h Kc K x 10h P x 105 lo,h79 0.60 2.19 1.19 0.91 7.6 20,958 1.20 0.1h 2.39 2.96 12.0 Average 9.8 (See Note - Next page) -35- Note: "Standard" ratio of polymerization 3.1 x lO'h'mols K1 : per liter per sec. Z 3 Number of collisions per cc per second K2 3 Observed Polymerization rate, mols per liter per sec. Kc : Calculated rate, mole per liter per sec. : Difference between "Standard" - Observed rates, mols per liter per sec. Specific probability factor = AK/K C The molecular, diemeterbo, cf; of: methanol . 1.. 57 x 10" 8 ethanol : 5.17 x 10"8 isopropanol : 5.6h x 10'8' tebutanol : 6.18 x 10 .8 styrene g 6.h6 x 10"8 Sample Calculation, 10% Ethanol Data: Concentration of styrene is 0.95 mole/lit 20 molecules of styrene/cc Concentration of ethanol is l.9h.mols/1it 2 NB : 11.69 x 10 O molecules of ethanol/cc EA 10h.1 Molecular'Weight of Styrene MB : h6.07 Molecular‘Weight of Ethanol OI : 6.1;6 x 10"8 Molecular Diameter of Styrene (TB 2 5.17 x 10"8 Molecular Diameter of Ethanol E : 17,000 cele.2Sh T : 3330K -36- N H <8 7r mi (sl'.‘~""'§/2)2 um, + l/MBfi (We) 2 - (83.37 x 10“) (5.32 x 10'8) 2(.02c32)%(5.72) (11.69) x who 28 Z number of collisions per cc per sec. 33,181.14 X 10 E k = Z e" RT where k is the velocity constant 28 _ 17000 21 Kc :: 2.291 x 10 collisions per second of molecules with an energy greater than 17000 cals in one cc To convert to mole per lit per second multiple "k" by 1000 and divide by Avogadro's number 6.025 x 1023. Then: 2 Kc : 2.291 x 102116.025 x 10 O z 3.80 mole per lit per sec. Rate of polymerization of styrene (K1): 3.1 x 10‘“h mols lit/sec Observed Rate with 10% ethanol: (K2) 1.9 x 10"}4 9 n N Difference (AK): 1.2 x 10:3 Deactivation Constant, P, 3 AK or specific probability 'K'c— factor ; 1.2 1: 10473.80 3 3.3 a": 10'5 The values obtained for P by Norrish and Smith37 for a bi- molecular reaction in methanol and ethanol are: Methanol: O.h3 x 10'"5 Ethanol: 3.8 x 10"5 201 x 10. It has been shown that an increase of alcohol concentra- tion causes the rate of polymerization to approach zero, at which point it might be assumed that any active centers formed are immediately deactivated. The decrease of the average molecular weight with an in- crease of alcohol concentration in the aqueous phase (Figures, 12, 19, 21, 25) is consistent with the theory that the prehe- bility of chain termination is greater in the higher concentra- tions of alcohol. Factors that were variable and which might effect the poly- merization are the pH and decomposition of potassium persulfate. Each reaction.was allowed to reach its own pH, no attempt being made to add any material to adjust the pH to a specific value. Addition of buffering agents was avoided as yet another variable ‘would have been added to the system. ‘With the exception of the experiment using carbon dioxide as the atmosphere the reactions remained between a pH of 6.7 and 7.7. Price‘o'sb has shown that the variation of the pH between these values has no effect upon the system. It has been assumed that the potassium persulfate does not decompose to any appreciable extent over the period of polymerization, and the data (Experiment #23) shows this to be a reasonable assumption. In no experiment, except those which ran for more than three hours, was there any appreciable decrease in the pH which might be attributed to the decomposition of the persulfate. -38- In the reactions using alcohols as part of the aqueous phase the decrease in unsaturation of the polymerizing system was determined by a titration method. These results calcue lated as reacted styrene monomer paralleled those of the per— cents of solid polymer formed. It should be noted that all the curves obtained for percent of reacted styrene lie 5-15% above' those of the percent of solid polymer formed. The reason that they are not superimposed, as would be expected, may lie in the method of‘purifying the polystyrene. In this method 95% ethanol was used to wash the polymer and consequent leaching out of the alcohol soluble polymers. The data (Experiments 19, 20, 21, 22) shows that the accur- acy of the method of determining the’amount of styrene in the system by reaction with mercuric acetate is 1 5%. In order to obtain reproducible results the order of addition of chemicals must be taken into account in addition to the time allowed.be- tween additions of the chemicals. It is not possible to inter- pret the values obtained from the titration scheme on any abso- lute basis, but since the curves for the percent of reacted styrene parallel those for the percent solid polymer formed, they are a check on the rate of overall polymerization. Any discussion of the effect of deactivation on the rate of polya merization as shown by the solid polymer curves applys equally well to the curves for the percent of reacted styrene. It is possible to duplicate the polymerization reactions if the same chemicals are used under the same conditions. If it is necessary to use a new chemical, it is not possible to reproduce previous work exactly. The difference in rate of polymerization between the curves of Figure 27 and Figure 1, lies in the potassium persulfate used. The potassium persul- fate used in.both cases was from the same bottle, recrystallized on different occasions using identical procedures, but it was found that the recrystallized persulfates differed in catalytic properties. Reproducibility of results from a single crystal crop was good. Experiments 10 and 11 (Figure 1h) were run with.the same chemicals approximately three months apart and show an excellent duplication of the rates of polymerization. The difference in rates of polymerization in Experiment 13 (Figure 16) and Experiment 11; (Figure 17) is attributed to the fact that in the former experiment freshly distilled styrene was used, while in the latter the same styrene, which had been stored for one week in the refrigerator, was used. -110. 1. 2. 3. 7. CONCLUSIONS A carbon dioxide atmosphere has no effect on the rate of overall polymerization. Its effect is limited to the lowering of the pH of the system accompanied by a slight decrease in the average molecular weight of the polymer. The water soluble alcohols studied are chain terminators. The average molecular weight of the polymer decreases with increasing concentration of alcohol in the aqueous phase. The water soluble alcohols studied are inhibitors or de- activators. The rate of polymerization decreases with in- creasing concentrations of alcohol. The relative deactivating effect as calculated is isopropanol > t—butanol > ethanol > methanol. A ”specific probability constant", P, is determined for each alcohol and is characteristic of that alcohol. The amount of double bond.which has reacted, calculated as the percent of styrene reacted per unit time, is the same as the rate of solid polymer formation. It is possible to duplicate a particular polymerization reaction by using the same chemicals in the same manner. -hl- 1. 2. 8a. b. 9a. b. C. 10. 11a. b. 12. LITERATURE CITED Simon, E., Ann., 21, 267 (1839). Berthelot, H., Ann., 1h1, 1818 (1867). CentraIEIatt, 65, I, 1026 (1867). §§, I, 1081 (1867). Whitby, G. 8., Trans. Faraday Soc., 22, 315 (1936). Schulz, G. V. & Husemann, E., Z. Physik Chem., 1323, 187 (1936). Cohen, S. G., J. Am. Chem. Soc., 67, 17 (19155). Taylor, H. 5. 8c Vernon, A. A., J. Am. Chan. Soc., 53, 2527 (1931) Bruson, H0 & Standinger, Ac, Ind. 8: Eng. Chane, 18, 38 (1929). Sontag, D., Anal. Chem., 1359 (193h), 0. 1., 28, L716 (193k). Ziegler & Bohr, Ber., élg’ 253 (1928). Houtz, R. C. & Adkins, H., J. Am. Chem. Soc., §_S_, 1609 (1933). Kawamus, H., C. A., 21, 11372 (1910)- USP 1683h0h, Sept. h, 1928. Abere, J., Goldfinger, G., Naidus, U., & Mark, H., Jour. COth, 80 Ge, Jo Ania Chains 8000, 92, 1057 (19147). Breitenboch, J. 77., Springer, A. , Abrahamczik, E. , C. A., 6232 (1938). Suess, H., Pilch, K., Rudorfer, H., Z. Physik Chem., A179, 361 (1937). Breitenboch, J. W. & Rudorfer, H., Monatsch., 19, 37 (1937). Staudinger 8; Frost, Ber., _6__8_, 2351 (1935). .442... 13a. b. c. d. lb. 153. A b. C. d. 16. 17. 18. 19. 20a. b. 21. 22. 23. 2b. 253. b. Flory, P. J., J. Am. Chem. Soc., 65, 372 (19h3). Chalmers, H., J. Am. Chem. Soc., 56, 912 (193k). Standinger, H., Trans. Faraday Soc., 22, 97 (1936). Taylor, H. S. & Jones,'W., J. Am. Chem. Soc., 53, 1111 (1930). Schulz, G. V., Z. Physik Chem., 5116, 317 (1936). Standinger, H. & Hauer,'W., "Die Hochmolekularen Organis- chen Verbindungen'. Springer, Berlin 1932 p. 1179. Standinger & Schulz, Ber., ggg, 2320 (1935). Schulz,G. V. & Husemann, E., Z. Physik Chem., Egg, 19h (1936). Kemp, A. R. & Peters, H., Ind. & Eng. Chem., 2Q, 1097 (19h2). Signer, H. & Gross, H., Helv. Chim. Acta., 17, 335 (1931). 0.1., _2_8, 3059 (1930). Debye, Ewart, Roe & McCarinen, J. Chem. Phys., lg, 687 (l9h6). Schulz, G. V., z. Physik Chem., 229, 379 (1935). Staudinger, H. & Husemann, E., Ber., 68, 1691 (1935). Mark:& Raff "High Polymeric Reactions" Interscience, N. Y. 19hl. Hohenstein, w. 9., Siggia, s. & Mark H., India Rubber'WOrld,‘lll, 173-177 (19kb). Breitenboch, J. W., Kolloid Z., 109, 119 (l9hh) 0.1., 31, 5338“(19u7). Fryling, C. F. & Harrington, E. W., Ind. & Eng. Chem., 26, 118 (19hh). Harms, Jo, J. Chem. PhySo, 13, 381 (19115) IE, 17, 215 (19h6). Montroll, E. W., J. Chem. Phys., 12, 337 (19h5). Fikentacher, H. , Anger. Chemie, £1, [:33 (1938). Schulz, Dinglinger'& Husemann, Z. Physik Chem., E82, 385 (1939) -u3- c. d. e. f. g. h. 26. 27. 28. 29. 30a. b. 31. 32. 33. 3h. 35. .36. 37. 38. 39. 110. Price, C. C. 1 Durham, D. A., J. Am. Chem., Soc., fig, 2508 (19h2). Abere, J., Goldfinger, G., Naidus, U. & Mark H., J. Chem. Phys., 11, 379 (19u3). Pram, Ho F0, Salley, D. J. 6C Mark ,Ho, Jo M. Chen. 5°C., 66, 983 (l9hh). Price & Adams, J. Am. Chem. Soc., 61, 167h-80 (19h5). Corrin, M. L., J. Poly. Sci., 2, 257 (19h7). Frillette, W. J. at HOhenStein, We Po, Jo POIYO Seio’ 31 22 (19h8). Barnes, J. Am. Chem. Soc., 61, 217 (19h5). Kolthorr,, I. M. & Dale, W. J., J. Am. Chem. Soc., 62, hul (l9h7). Harman, R. A. & Eyring, H., J. Chem. Phys.,‘19, 557 (19h2). Kline, G. H., Brit. Plastics, 18, 101 (l9h6). Kolthoff & Dale, J. Am. Chem. Soc., 61, 1672 (1985). Kolthoff, Guss, May & Medulia, J. Poly. Sci., 1? 3h0 (19h6). Yang, P. T., M. S. Thesis, Michigan State College (19h7). Kolthoff & Bovey, J. Am. Chem. Soc., 62, 21h3 (19h7). Morgan, D. H.,Ifi. S. Thesis, Michigan State College (l9h6). Nozacki, K., Ind. 2 Eng. Chem., Analyt. Edit., 18, 583 (19h6). Schulz, Dinglinger & Husemann, Z. Physik Chem., Bh3, 385 (1939) Mayo, F0 R0, J. m. Chan. 13°C., fig, 232).; (19113). Norrish & Smith, Trans. Chem. Society, 128, 129 (1928). Lewis, w. C. M., Trans. Chem. Society, 112, h71 (1918). Hinshelwood, C. N., "The Kinetics of Chemical Change" Oxford Press pgs. 1h,17,h6-h7, 50-52 (19h0). Moelwyn-Hughes, E. A., Chem. Reviews, 19, 2h3 (1932). -hh-