’- ", THE FRAC'E‘IQNATION AND MGLECULAR WEIGHT DETERMINATION OF EMULSION P01.YMERIZED POLYSTYRENE Thesis hat the Degm 0* M. S. MEG-HGAN STATE COLLEGE Max WaiBace Km]! 1949 This is to certify that the thesis entitled The Fractionetion and Molecular Weight Determinetion of Emulsion Polymerized Polystyrene presented by Max Wallace Krell has been accepted towards fulfillment of the requirements for M . E .‘ degree in_0*rg_e 3:1 __q C h emi 5 try ——_ 422% 95/ f MaTjor professor Date - THE FRACTIONATION AND MOLECUIAR WEIGHT DETERFINATION OF EVUISION POLYMERIZED POLYSTYREFE By XAX'WALLACE KRELL 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 1949 tawny om ~“ . r--' 0‘? Ti} .f if; . 17’} '3 5'" x f (‘L‘ ACKNOWLEDGMENT The author wishes to express his appreciation to Doctor Ralph L. Guile for his counsel and guidance during the course of this investiga- tiono *##**#** ****** ssss ** nu mm‘- __,,,__ -»____.._.r..— wax aim TABLE OF CONTENTS Introduction.............................................. Historical................................................ Reagents.................................................. Preparation.of Polymer Samples..............;............. Determination of Precipitability of Samples............... Precipitability Curves............................... Precipitability Tangent Curves....................... Fractionation............................................. Determination of Molecular weights........................ Graphical Representation of Molecular weight Distribution Data...................................................... Molecular weight Distribution Curves................. Moleeular weight Distribution Tangent Curves......... Graphical Determination of Intrinsic Viscosities.......... Determination of Precipitability of Fractionated Samples.. Discussion................................................ Summary................................................... RefermOBSOOOOOI...0.0.000...O...0.0.0....OOOOOOOOOOOOOOOC Page 11 27 30 33 4O 5O 61 62 i i. s :1 x o ~sc ...- - s .. . . a tl . ... ————vw- ’ -_ INTRODUCTION During the past few'years the emulsion polymerization of styrene has been investigated extensively in this laboratory. 1’2’3’4’5 These investigations, for the most part, have been concerned'with the various phases of the reaction. Determina- tions of the average molecular weight have been made in con- junction with this work, and some preliminary efforts have been.made on the fractional precipitability of the polystyrene from a solvent system by a non-solvent. This paper reports the results of the first work in this laboratory to establish some relationship between the molecular weight distribution of a given sample of polystyrene and the 'precipitability" of that polymer from.solution. The fraction- ation of the polymer samples, and the determination of their molecular weights are necessarily an integral part of this investigation. The term."precipitability" as used in this work is defined as the extinction of light caused by the turbidity of a polymer solution, which is due to the precipitation of the polymer from solution by a non-solvent. HISTORICAL Styrene was first produced in 1831 by Bonastre as a distilla- tion product of storax, which was obtained from Liquidambar Orien- talis, a tree native to Asia Minor. Eight years later, Eduard Simon obtained (upon heating the monomer) what was thought to be an oxide of styrene, but which in reality was polystyrene. It was not until 1845 that the polymeric nature of the mater- ial was recognized by two Englishmen, Hofmann and Blyth. In 1869 Berthelot reported the preparation of monomeric styrene from ethyl benzene. The first patents on polystyrene were granted to Dr. F. E. matthews of London, England in 1911. Two French chemists, Dufraisse and Moreu were responsible for a great deal of develop- ment work on the polymer, and in work published in 1953 they dis- cussed agents which retarded styrene polymerisation. Commercial production of polystyrene was attempted in Germany about 1930, and by the Naugatuck Chemical Company in the United States in 1933. However, it was not until 1937, when the Dow Chemical Company perfected the ethylbenzene method for the pro- duction of the monomer that large scale production of the polymer could begin. Since 1937, many other companies have taken up poly- styrene production, many of them obtaining their monomer from Dow Chemical Company. The Dow method for the production of monomeric styrene consists of the pyrolytic dehydrogenation of ethylbenzene which is produced by a liquid phase reaction of benzene with ethylene under atmospheric pressure at 88°C. Any polyethylbenzenes produced are disproportion- ated by recycling so that the yield of ethylbenzene is high. The ethylbenzene is then catalytically dehydrogenated to styrene at tanperatures of 600 to 800°C. Yields of so to so percent are re- ported. Styrene will polymerize slowly on standing at room.tempera- ture but its polymerization may be inhibited by numerous antioxi- dants, hydroquinone and tertiary butyl catechol being the most im- portant commercially. They may be removed by distillation of the styrene over solid caustic. Styrene polymerization is accelerated both by heat and catalysts, average chain length of the polymer decreasing with increasing tem- perature and increasing concentration of the catalyst used. Air must be excluded in the polymerization, otherwise yellowing of the product will occur. The polymerization of styrene is thought to be a chain reaction which is initiated by the activation of the double bonds of a small number of molecules. These activated molecules provide nuclei for polymerization, adding monomer molecules with which they come in contact, and transmitting their energy of acti- vation to the larger molecules as they grow; As is often the case with commercial products, scientific re- search has lagged far behind the commercial development. Determina- tion of average molecular weight began in 1930 when Staudinger and Honor6 reported finding some relationship between viscosity of poly- meric solutions and molecular. weight of the polymer. In subsequent papers Staudinger7'8 published data on the relationship between vis- cosity of polymeric solutions and chain length of the polymers. The Staudinger method, made use of extensively in this thesis for the determination of molecular weight, will be discussed in a later por- tion of this work. In 1926, Svedbergg reported the derivation of the formula for obtaining molecular weights by sedimentation data using a high speed or "ultra" centrifuge. This first report was followed by several later ones along the same line, in which molecular weights were determined, and size distribution, sedimentation, and disper- sion studies were made. later developments brought applications of these methods of determination of molecular weight into the field of high polymeric materials. In 1935 Signerlo reported his work with the ultracentri- fuge, and the following year published results on the direct determination of the molecular weight of polystyrene with this instrument.11 In 1936 Schulz's worklz appeared, in which he had developed an equation for the molecular weight determination of highly poly- merized compounds from osmotic pressure measurements. The following -4- year Dobry13 applied the osmotic pressure method for the determina- tion of the mmlecular weight of polystyrene. The latest method to be devised for the determination of the molecular weight of polystyrene is the method of light scattering. The work of Debye14 is probably the most outstanding in this field. Because of the simplicity of the apparatus, and the ease of manipulation, the viscosity method is still the most widely used of the methods mentioned. All of the methods discussed give average molecular weights rather than absolute values. Quite often molecular weight values determined with the ultracentrifuge vary markedly from those ob- tained by viscosity methods. Probably the ultracentrifuge gives more nearly an absolute value, but the viscosity molecular weight of a typical polymer would not be seriously in error unless the polymer is extensively branched. In the case of polystyrene this error would not be large. Much work has been done on the molecular weight distribution of polystyrene, but little has been published on the actual frac- tionation of the polymer into components of different molecular weight. In an article appearing in 1936 Schulz and Hueemenn15 pub- lished their work in which they fractionally precipitated a butanone solution of polystyrene by adding methanol as a non-solvent. This was followed by a later article by Schulz and Dinglingerls in which essentially the same method was used. -5- Since these articles were published, little has appeared on the subject of polystyrene fractionation. The Dow Chemical Company is working on the problem from a commercial standpoint, but has not published any work regarding it. An article by morey and Tamblyn17 confirms, for the most part, the work of Schulz. Only a few studies on the precipitability of a polymer solution by a non-solvent have been made. Schulz18 published an article in 1937 in which he tried to relate solubility and molecular weight of high molecular compounds. Adams and Powers19 studied the polymer distribution of varnish resins. Morgan1 , of this laboratory, was the first to apply the method of Adams and Powers to polystyrene samples. REAG ENTS Styrene The styrene used in this work was obtained from the Dow Chemical Company. Before use, the monomer was distilled under re- duced pressure (20 man.) and that portion having a refractive index of 1.544 collected for use in the ermlsion polymerization. Water The water used for the aqueous phase of the emlsion was distilled from alkaline potassium permanganate (300 grams potassium hydroxide, 8 grams potassium permanganate per liter of solution) under an atmosphere of air or nitrogen, depending on which atmos- phere was to be used for the polymerization. Two liters of water were added to 50 mls. of the alkaline permanganate, the mixture re- fluxed for thirty minutes, and then allowed to distil. The' first 200 mls. of distillate were discarded, the receiver flushed with steam from the distilling system, and the distillate again collected. If a nitrogen atmosphere was used, the water was kept under nitro- gen pressure until used. Potassium Persulfate Merck's reagent grade was recrystallized from purified water (see above), filtered off, and dried at room temperature for 48 hours. Dupanol E This material, manufactured by DuPont, was used as the emulsifying agent. It is reported as being lauryl amine sulfate. The Dupanol G was stored as a liquid at 50°C. to provide for an easy method of handling the reagent. Aluminum.Chloride Baker's C. P. (A1C13-6H20) Toluene Baker's C. P. This reagent was redistilled under atmos- pheric pressure and the fraction having a refractive index of 1.498 collected for use in viscosity measurements. Ethanol U. s. P. 95% ( B. P. 76 - 77°C.) Methanol Merck's C. P. Butanone Eastman's C. P. Nitrogen water pumped. The nitrogen, before use, was passed through alkaline pyrogallol solution. This solution consisted of fifty grams of potassium.hydrcxide in 100 mls. of water, to which was added five grams of pyrogallic acid. PREPARATION OF POLYMER SAMPLES The method of enulsion polymerization employed for the prepara- tion of polystyrene‘used in this work was adapted from the methods 1:2’3'4'5 The polymerization was of coaworkers in this laboratory. carried out in a three necked, round bottom, one liter flask with standard taper ground glass joints. The flask was ifitted with a mercury sealed swivel stirrer, thermometer, nitrogen addition tube, (when.using a nitrogen atmosphere) and a reflux condenser. The con- denser was attached to a small double water trap to permit nitrogen to escape and prevent air from.entering the reaction vessel. The reaction flask was immersed in a constant temperature bath at 60°C;£0.2°C. throughout the polymerization. Polymerizations were carried out under both nitrogen and air atmospheres, with and with- out stirring. The emulsion was composed of eight parts of water to one part of styrene, with one percent Dupanol G emulsifier, based on the aqueous phase. The catalyst, potassium.persu1fate, was used at a concentration of 0.0017 M. based on the aqueous phase. In carrying out a polymerization, the Dupanol G and water were added to the reaction vessel which had previously been flushed out with nitrogen if a nitrogen atmosphere was to be used. The styrene was then added, and the reaction.mixture stirred until it reached operating temperature. At this time the calculated amount -9- of catalyst was added and timing begun. In the case where the re- action was not stirred, the stirrer was turned off two minutes after addition of the catalyst. Where partial stirring was employed, the stirrer was turned off at the same time, and then turned on for two minute periods at one-half hour intervals. Reaction times varied from 55 minutes to 14% hours, depending on the atmosphere used and whether or not the emlsion was stirred after the addition of the catalyst. At a recorded time, the polymer was precipitated from.the emul- sion by pouring the emulsion into approximately twice its volume of 95% ethanol to which had been added a trace of aluminum.chloride. After complete precipitation had occurred, the polymer was filtered off by the use of suction, washed twice with 95% ethanol, and then six times with distilled water. After final washing, the polymer was filtered off and allowed to dry at 50°C. for 48 hours. TABLE I Sample Length of Atmosphere Stirring Actual Percent Run (Hrs.) Yield Yield 1 3:20 Nitrogen None 59.0 g. 78.6 2 6:30 Air Complete 89.5 g. 89.5 3 14:30 Air Complete 66.3 g. 88.4 4 0:55 Nitrogen Complete 78.8 g. 87.6 5 1:40 Nitrogen Partial 81.6 g. 90.7 6 3:00 Air Complete 80.9 g. 89.9 -10.. DETERMINATION OF "PRECIPITABIIJTY" "Precipitability" curves were obtained by utilization of the method of Adams and Powerslawith modifications. Solutions of the polymer samples in butanone, 0.02 molar, were prepared. The molar- ity referred to is that of the "Grundmol", which is a molarity based on the weight of the recurring group in the polymer. In the case of polystyrene, a molar solution would be one containing 104 grams of polymer per liter of solution. To 125 mls. of this solution, methanol was added with stirring at 20°C. and the extinction of light passing through the solution ‘measured by means of a photoelectric cell and galvanometer. (See diagram.on following page.) Methanol was added until further addi- tion brought about no further extinction of light. Graphs were then plotted with percent methanol (based on total weight of solution) as the abscissa and extinction Log incident light (In) as the transmitted‘light(1) ordinates. In all cases incident light was equal to 100. The tangent or differential curves from the extinction curves were also plotted. The tangents were calculated between successive points on the extinction curves, and the values obtained plotted opposite the average percent methanol between these successive pairs of points. PRECIPITRBILITY APPARATUS To 6 volt E Transformer<$———— G ‘VR PC WIRING DIAGRAM G Galvanometer VR Variable Resistance (30,000 - 35,000 Ohms) PC Photoelectric Cell 512- M1. CH 03 0 0010000000 ravdrdoa mthNOCJCDPN O OCH 0 7 3 H 1.55 3.05 4.51 5.92 7.30 8.63 9.93 10.87 11.18 11.50 11.80 12.11 12.41 12.71 13.01 13.63 14.18 14.73 15.33 16.44 17.53 19.10 21.60 23.94 26.16 28.24 30.21 32.08 33.84 35.52 TABLE II I 100.0 100.0 100.0 99.5 99.0 98.5 98.0 87.0 78.0 71.0 65.0 59.5 55.0 53.0 50.0 45.5 41.5 38.5 36.5 33.5 31.0 29.5 27.5 26.5 26.0 25.5 25.0 24.5 24.0 24.0 POLXHER-# 1 Io/I 1.000 1.000 1.000 1.005 1.010 1.015 1.020 1.150 1.282 1.410 1.540 1.680 1.820 1.888 2.000 2.198 2.410 2.600 2.742 2.987 3.206 3.390 3.640 3.775 3.847 3.925 4.000 4.085 4.165 4.165 -13... logIo/I 0.0000 0.0000 0.0000 0.0022 0.0043 0.0065 0.0086 0.1079 0.1492 0.1875 0.2253 0.2601 0.2760 0.3010 0.3420 0.3820 0.4148 0.4381 0.4752 0.5060 0.5302 0.5611 0.5769 0.5851 0.5938 0.6196 0.6196 Av.%CH30H 0.78 2.30 3.78 5.21 6.61 7.96 9.28 10.40 11.02 11.34 11.65 11.96 12.26 12.56 12.86 13.32 13.90 14.46 15.03 15.88 16.98 18.32 20.35 22.77 25.05 27.70 29.22 31.14 32.96 34.68 Tangent 0.0000 0.0000 0.0000 0.0015 0.0016 0.0016 0.0016 0.0555 0.1522 0.1290 0.1270 0.1220 0.1159 0.0531 0.0834 0.0662 0.0727 0.0595 0.0388 0.0335 0.0282 0.0154 0.0124 0.0068 0.0037 0.0042 0.0040 0.0049 0.0048 0.0000 M1. %CH OH 3 3.05 5.92 8.63 9.93 11.50 11.80 12.11 12.41 12.71 13.01 13.31 13.63 13.89 14.18 14.73 15.33 15.89 16.44 17.53 18.59 20.12 21.60 23.94 26.16 28.24 30.21 32.08 33.84 35.52 37.12 - POLYMER # 2 TABLE III PRECIPITABILITY I 10/1 100.0 1.000 100.0 1.000 100.0 1.000 100.0 1.000 96.0 1.021 96.0 1.042 92.5 1.061 69.0 1.124 64.0 1.191 76.0 1.262 72.0 1.590 67.0 1.494 65.5 1.575 60.5 1.654 56.0 1.725 55.5 1.670 50.0 2.000 47.5 2.106 45.0 2.224 42.0 2.562 40.0 2.500 57.5 2.670 56.0 2.760 54.5 2.900 55.0 5.050 52.5 5.076 51.75 5.150 51.50 5.176 51.25 5.200 51.0 5.225 51.0 5.225 -14.. logIo/I Av. %06306 Tangent 0.0000 0.0000 0.0000 0.0000 0.0090 0.0179 0.0338 0.0508 0.0759 0.1079 0.1430 0.1744 0.1973 0.2185 0.2368 0.2718 0.3010 0.3235 0.3471 0.3769 0.3979 0.4265 0.4440 0.4624 0.4814 0.4883 0.4983 0.5022 0.5052 0.5085 0.5085 1.52 4.49 7.28 9.28 10.56 11.34 11.65 11.96 12.26 12.56 12.86 13.16 13.47 13.76 14.04 14.46 15.03 15.61 16.16 16.98 18.06 19.36 20.86 22.77 25.05 27.20 29.22 31.14 32.96 34.68 36.32 0.0000 0.0000 0.0000 0.0000 0.0072 0.0276 0.0532 0.0546 0.0838 0.1066 0.1171 0.1045 0.0716 0.0819 0.0630 0.0637 0.0486 0.0401 0.0431 0.0274 0.0198 0.0187 0.0118 0.0078 0. 0086 0.0033 0.0052 0.0020 0.0017 0.0020 0.0000 TABLE Iv - POLYMER # 5 PRECIPITABILITY Ml. 06306 %CH30H I 10/1 logIO/I Av.%CH30H Tangent 4.0 5.05 100.0 1.000 0.0000 1.52 0.0000 6.0 5.92 100.0 1.000 0.0000 4.49 0.0000 12.0 8.65 100.0 1.000 0.0000 7.26 0.0000 14.0 9.95 100.0 1.000 0.0000 9.26 0.0000 16.0 11.16 100.0 1.000 0.0000 10.56 . 0.0000 16.5 11.50 96.5 1.015 0.0065 11.54 0.0202 17.0 11.80 96.5 1.057 0.0158 11.65 0.0500 17.5 12.11 94.5 1.058 0.0245 11.96 0.0281 18.0 12.41 92.5 1.081 0.0556 12.26 0.0511 16.5 17.71 90.5 1.105 0.0454 12.56 0.0518 19.0 15.01 67.5 1.142 0.0577 12.86 0.0477 19.5 15.51 82.5 1.211 0.0851 15.16 0.0649 20.0 15.65 77.5 1.290 0.1106 15.47 0.0858 20.5 15.89 72.5 1.580 0.1599 15.76 0.1128 21.0 14.18 69.0 1.449 0.1611 14.04 0.0728 21.5 14.47 66.0 1.515 0.1804 14.52 0.0667 22.0 14.75 65.5 1.575 0.1975 14.60 0.0649 22.5 15.04 61.5 1.627 0.2114 14.86 0.0455 25.0 15.55 60.0 1.667 0.2219 15.18 0.0564 24.0 15.69 57.0 1.755 0.2445 15.61 0.0598 25.0 16.44 54.5 1.855 0.2656 16.16 0.0552 26.0 16.99 52.5 1.905 0.2799 16.72 0.0295 28.0 18.06 49.5 2.020 0.5054 17.52 0.0256 50.0 19.10 47.0 2.126 0.5276 16.56 0.0214 52.0 20.12 45.0 2.221 0.5466 19.61 0.0166 55.0 21.60 44.5 2.296 0.5614 20.66 0.0100 40.0 25.94 44.25 2.581 0.5766 22.77 0.0066 45.0 26.16 44.0 2.470 0.5927 25.05 0.0072 50.0 26.24 44.0 2.470 0.5927 27.20 0.0000 -15- -16- TABLE v - POLYMER # 4 PRECIPITABILITY M1. CH30H %06306 I 10/1 logIo/I Av.%CH30H Tangent 6.0 4.51 100.0 1.000 0.0000 2.26 0.0000 8.0 5.92 100.0 1.000 0.0000 5.22 0.0000 10.0 7.50 99.5 1.005 0.0022 6.61 0.0016 12.0 6.65 99.0 1.010 0.0045 7.96 0.0016 14.0 9.95 96.5 1.015 0.0065 9.26 0.0017 16.0 11.18 66.0 1.157 0.0556 10.56 0.0592 16.5 11.50 74.5 1.542 0.1278 11.54 0.2250 17.0 11.60 65.0 1.589 0.2011 11.65 0.2446 17.5 12.11 55.5 1.802 0.2558 11.96 0.1760 18.0 12.41 50.0 2.000 0.5010 12.26 0.1845 18.5 12.71 46.0 2.175 0.5575 12.56 0.1214 19.0 15.01 45.0 2.526 0.5666 12.86 0.0972 19.5 15.51 41.0 2.441 0.5876 15.16 0.0697 20.0 15.65 59.5 2.554 0.4056 15.47 0.0508 21.0 14.16 56.5 2.740 0.4578 15.90 0.0617 22.0 14.75 54.5 2.900 0.4624 14.46 0.0449 25.0 15.55 55.5 2.965 0.4749 15.05 0.0209 25.0 16.44 51.5 5.176 0.5019 15.88 0.0242 27.0 17.55 50.0 5.555 0.5228 16.96 0.0192 50.0 19.10 28.5 5.510 0.5455 18.52 0.0145 55.0 21.60 26.5 5.775 0.5769 20.55 0.0126 40.0 25.94 25.0 4.000 0.6021 22.77 0.0107 45.0 26.16 24.5 4.080 0.6107 25.05 0.0059 50.0 26.24 24.25 4.120 0.6149 27.20 0.0020 55.0 50.21 24.0 4.166 0.6197 29.22 0.0024 60.0 52.06 25.75 4.210 0.6245 51.14 0.0050 65.0 55.84 25.75 4.210 0.6245 55.46 0.0000 M . 0 1 CH3 H 4.0 8.0 10.0 12.0 14.0 15.5 15.0 16.5 17.5 18.0 18.5 19.0 19.5 20.0 21.0 22.0 24.0 26.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0 70.0 %CH CH 3 3.05 5.92 7.30 8.63 9.93 10.87 11.18 11.50 12.11 12.41 12.71 13.01 13.31 13.63 14.18 14.73 15.89 16.99 19.10 21.60 23.94 26.16 28.24 30.21 32.08 33.84 35.52 POLYMER # 5 TABLE VI PRECIPITABILITY I 0/1 100.0 1.000 100.0 1.000 100.0 1.000 100.0 1.000 99.5 1.005 95.0 1.052 90.0 1.111 84.0 1.190 77.5 1.290 69.5 1.459 65.5 1.575 58.0 1.725 54.5 1.854 52.0 1.925 48.0 2.084 44.5 2.225 40.0 2.500 56.0 2.652 34.5 2.900 52.5 5.080 51.0 5.225 50.0 5.555 29.5 5.590 29.0 5.450 28.75 5.480 26.5 5.510 28.5 5.510 -17- logIO/I 0.0000 0.0000 0.0000 0.0000 0.0022 0.0220 0.0457 0.0756 0.1106 0.1581 0.1973 0.2368 0.2634 0.2840 0.3189 0.3473 0.3979 0.4203 0.4624 0.4886 0.5085 0.5228 0.5302 0.5378 0.5416 0.5453 0.5453 Av.%CH30H 1.52 4.49 6.61 7.92 9.28 10.40 11.02 11.34 11.82 12.26 12.56 12.86 13.16 13.47 13.90 14.46 15.31 16.44 17.04 20.35 22.77 25.05 27.20 29.22 31.14 33.46 34.68 Tangent 0.0000 0.0000 0.0000 0.0000 0.0017 0.0211 0.0765 0.0935 0.0740 0.1582 0.1307 0.1322 0.0882 0.0643 0.0635 0.0517 0.0436 0.0203 0.0200 0.0105 0.0085 0.0064 0.0036 0.0039 0.0020 0.0021 0.0000 TABLE VII - POLYMER # 6 PRECIPITABILITY M1.06306 20H306 I 10/1 logIo/I Av.%CH30H Tangent. 2.0 1.55 100.0 1.000 0.0000 0.72 0.0000 4.0 5.05 100.0 1.000 0.0000 2.50 0.0000 6.0 4.51 99.75 1.002 0.0009 5.78 0.0006 8.0 5.92 99.5 1.005 0.0022 5.22 0.0009 10.0 7.50 99.25 1.006 0.0055 6.61 0.0010 12.0 8.65 99.0 1.011 0.0048 7.96 0.0010 14.0 9.95 98.75 1.015 0.0056 9.26 0.0006 16.0 11.18 96.5 1.016 0.0069 10.56 0.0010 16.5 11.50 98.0 1.021 0.0090 11.54 0.0067 17.0 11.60 69.0 1.124 0.0508 11.65 0.1591 17.5 12.11 78.5 1.275 0.1055 11.96 0.1765 16.0 12.41 70.5 1.420 0.1525 12.26 0.1559 18.5 12.71 65.0 1.589 0.2011 12.56 0.1626 19.0 15.01 58.5 1.710 0.2550 12.86 0.1065 19.5 15.51 55.5 1.870 0.2716 15.16 0.1295 20.0 15.65 50.5 1.980 0.2967 15.47 0.0776 21.0 14.18 46.0 2.075 0.5170 15.90 0.0570 22.0 14.75 42.5 2.555 0.5716 14.44 0.0994 25.0 15.55 40.0 2.500 0.5979 15.05 0.0459 24.0 15.89 56.5 2.600 0.4150 15.61 0.0504 26.0 16.99 55.5 2.820 0.4502 16.44 0.0521 26.0 16.06 54.0 2.942 0.4686 17.52 0.0169 50.0 19.10 52.5 5.078 0.4885 18.58 0.0189 55.0 21.60 50.0 5.555 0.5228 20.55 0.0156 40.0 25.94 26.5 5.510 0.5455 22.77 0.0096 45.0 26.16 27.5 5.658 0.5609 25.05 0.0070 50.0 26.24 26.5 5.775 0.5769 27.20 0.0077 55.0 50.21 26.0 5.850 0.5855 29.22 0.0045 60.0 52.06 25.75 5.865 0.5894 51.14 0.0021 65.0 55.84 25.5 5.922 0.5955 52.96 0.0025 70.0 55.52 25.25 5.960 0.5977 54.68 0.0025 75.0 57.12 25.0 4.000 0.6021 56.52 0.0027 -18.. -19.. 3‘ .3 . 52549307232 Hzmhwmmm 0.25% N). 3A 285 N) 5 mmnom QH n 5014090... .5852 H 4 54:38 '1‘ om ON 1: ..n- -_----14|1-J, 4 _ n n _ \r \ . a \ \ 4 a $253.04 .622 n 4 5538 I \ \ nfimnmoéq 5.... m 4 5.5.84 IIIII \ BBWU PH HM H93. Hm momma noxionIIxfl fl ECHH 0.0 H.“ mg 0% -20- —-0 mmnqm m mmzom m\~ H unaszflz mm BE>AOW1ZDA Huang Dd ON _ d _ _ fl _ q — nmflnnofi< m: o n 5.54.84 £24834 585.2 m x 5.238 lllll. mmmmnmoneq zmosmgHz n 4 mmzwnon WEE :0 Chg Hmdu. E Hummm N0 ILONLLXH C) -21- UN *4 (A C‘) .‘r-i oszDthm oz mason. QH m $6148.54 51.12.32 H 4 mafia man Wino MHHAHmmbo HWEDRSH. IL m ESE 3.0 3.0 154301171 -22- 0: oszmHhm m 4 amnion 28: SH 5 manumdgh4 mH4 m k mafimqom mom mbmpu memqumHmmm Mbmuo Hzm0249 0a.o In C) O C; lfiflDNVl 0H.o -23- 00 4. .I \4.I.\J.l.fl\...;.« 4....“ ..... 1 . ELL: hf b :9-.. I, 5.50. r. 3.270. oannEm ”1,5560 meg «\H 5H mfimnwoa< m: m 4 5.584 .404 EEB weHnHmHmmn m>m00 Hzmozmpo HBHHHm44HmHommn some mmeHmma mbmno eznmnm 04H00m44040unnn menu nmaH..:m mameo enmmz4e .54 N mmmch -26- 5. E—* M o-J -—4 b 1 [x] (Q :1: 4 LLI Ed ‘ . } '4 (L4 (1. (I) H C) U - < (:3 {-14 y m <1: 11* C1: (2‘. 1 :! r~4 2 , .-) <1: H l.’ (I. ——. 4 (1‘. _ r—q C I \O E4 Lt} U I 17' 1t: #4 fix] .. m‘ b. 53 E11 L13 I») 6". pg} rLi 'r‘ 0 CL] 5 J 2" :‘b " ) (L) Er“ ii. (3 I 7., p ”—4 94 [-1. (n 2’: Er: co CL) (1.] .0) f :3 L ) ‘f " i=1 2: - ' :1: 03 [if E?) ' L) M 8 J l l r4 0 IA 0 o-—-4 . C3 1 O. LNEflNYl 0.00 . n, H". 30 'U\ 15 V'Y '1'!“ \T FRAC TI OIIAT I OH Fractionation was accomplished by stepwise precipitation of the polymer from solution. Ten grams of polystyrene were dissolved in one liter of butanone, and methanol added, with stirring at 20° C.. to a predetermined percentage of non- solvcnt. The mixture was allowed to stand in a 20°C. con- stant temperature bath until complete sedimentation occurred. The supernatant liquid was then decanted off, the polymer fraction washed with methanol, and dried at 50° c. for 48 hours. Methanol was again added to the liquid mixture, and the process repeated. In this manner, each original poly- mer sample was divided into from four to seven fractions. After drying, the fractions were weighed, and the molecular weight determined by the viscosity method. The process of fractionation consumed approximately two weeks. Recovery of the original polystyrene sample ran from 92.74 to 99.33 perc ent. -2 7.. TABLE VIII - POLYMER ,, 1 FRACTIONATION IPraction Percentage Range ' Weight of of Nonsolvent Fraction 1 00.0 - 11.0 2.3887 g. 2 1100 ' 1200 5.1883 g. 3 12.0 "’ 1300 0.9154 g. 4 13.0 - 14.0 0.4532 g. 5 14.0 - 16.0 0.2341 g. 6 16.0 - 20.0 0.1308 g. 7 20.0 "' 35.0 0.0287 g. Percent Recovery - 93.39 TABLE IX - POLYI~£ER # 2 FRACTIONATION Fraction Percentage Range Weight of of Nonsolvant Fraction 1 00.0 - 11.5 4.0813 g. 2 11.5 "' 1200 1.5293 g. 3 12.0 - 13.0 2.2351 g. 4 13.0 - 14.0 0.7259 g. 5 14.0 - 15.0 0.3313 g. 6 15.0 - 20.0 0.3844 g. 7 20.0 - 36.0 0.0249 g. Percent Recovery - 93.12 TABLE X - POLIMER-# 3 FRACTIONATION 1EFraction Percentage Range weight of of Nonsolvent Fraction 1 00.0 - 11.5 2.5040 g. 2 11.5 " 12.0 . 2.5523 g. 3 12.0 " 13.0 2.2915 g. 4 13.0 - 14.0 0.8037 g. 5 14.0 - 15.0 0.4974 g. 6 15.0 "' 16.5 0.5248 g. 7 1605 " 28.0 0.7597 30 Percent Recovery - 99.33 -28- Fraction muoaeara Fraction 050100915?!“ Fraction OQU'IIbCIINl-J TABLE x1 - POLYMER # 4 FRACTI 0111111011 Percentage Range Of Nonsolvent 00.0 "' 1 1008 - 1 1105 ' 12 1200 - 340 Percent Recovery - 95.09 TABLE XII - POLYMER # 5 FRACTIONATION Percentage Range of Nonsolvent 00.0 - 11.0 11.0 " 11.5 11.5 12.0 12.0 13.0 13.0 - 14.0 14.0 - 35.0 ' Percent Recovery - 96.05 TABLE XIII - POLYMER # 6 FRACTIONATION Percentage Range of Nonsolvent 00.0 - 12.5 13.0 14.0 15.0 16.0 40.0 ldtdldidid ozoam.oaea O O O O O O1@500h3h4 Fraction O>OIQDOIO3FJ TABLE.XVIII - POLYMER # 4 MOLECULAR WEIGHT DETERMINATION Concentration 0.0101 M.- 0.0103 0.0101 0.0100 Efflux Time 11103 8°C. 93.8 78.7 69.5 TABLE XIX - POLYMER-# 5 MOLECULAR WEIGHT DETERMINATION Concentration 0.0101 M. 0.0101 0.0103 0.0101 0.0102 0.0103 Eff lux Time 121.1 sec. 102.5 88.3 78.7 70.3 64.9 TABLE xx - Pomms # 6 MOLECULAR WEIGHT DETERMINATION Concentration 0.0102 M. 0.0102 0.0103 0.0100 0.0101 0.0103 -32_ Eff In: Time 82.1 sec. 73.9 65.9 62.0 59.6 57.6 Molecular'Weight 593,100 406,300 259,100 164,100 Molecular‘Weight 693,000 504,800 349,500 259,900 170,700 114,900 Molecular‘Weight 290,000 207,300 125,500 88,200 62,500 41,400 GRAPHICAL REPRESENTATION OF MOLECUIAR WEIGHT DISTRIBUTION DATA The molecular weight distribution curves for the six samples used in this work were constructed by plotting the average mole- cular weight of the fractions obtained versus the weight frac- tions (based on the original sample weight of 10 grams) up to and including the fraction of the molecular weight being plotted. For the last weight fraction only, one-half of the weight of the last fraction was added to the cumulative weight, since this was the last point to be plotted on the curve. The tangent or differential curves from the molecular weight distribution curves were also plotted. The tangents were calcu- lated between successive points on the molecular weight distribu- tion curves, and the values obtained plotted opposite the average molecular weight corresponding to the midpoint between these suc- cessive pairs of points. -33.. Fraction 40:01.;smmH Fraction QOU‘IPMNH Fraction 40301»:an TABLE XXI - POLYMER # 1 Weight of Average Molecular Fraction 2.3887 g. 5.1883 0.9154 0.4532 0.2341 0.1308 0.0287 TABLB XXII - POLYMER-# 2 'Weight of Average Molecular Fraction 4.0813 g. 1.5293 2.2351 0.7259 0.3313 0.3844 0.0249 weight 559,000 404,000 228,500 156,800 98,800 68,300 63,000 ‘Weight 381,900 284,100 203,200 125,900 100,100 75,900 31,800 weight Fraction x 10 8.1449 6.9505 1.7622 0.8468 0.3936 0.1595 0.0287 Weight Fraction x 10 7.2715 5.2309 3.7016 1.4665 0.7406 0.4093 0.0249 TABLE.XXIII - POL¥EER~# 3 weight of Average Molecular Fraction 2.5040 g. 2.5523 2.2915 0.8037 0.4974 0.5248 0.7597 weight 300,700 219,900 144,200 90,700 71,100 49,500 22,700 -34- weight Fraction x 10 8.6814 7.4294 4.8771 2.5856 1.7819 1.2845 0.7597 Tangent 0.0771 x 10'4 0.2940 0.1275 0.0781 0.0768 0.2465 Tangent 0.2090 x 10 0.1890 0.2890 0.2815 0.1370 0.0872 Tangent -4 0.1553 x 10'4 0.3370 0.4285 0.4100 0.2300 0.1955 Fraction weight of Average Molecular Fraction POINH Fraction OO‘IfiCIINH Fracti on CDCJ'IprflNH 4.6539 g. 3.5409 0.7781 0.5357 weight of Average Molecular Fraction 3.2298 g. 3.4147 1.2267 1.0298 0.3823 0.3221 'Weight of Average Molecular Fraction 2.7341 g. 3.0238 1.9893 0.7555 0.4096 0.3620 Wei ght 593,100 406,300 259,100 164,100 Weight 593,000 504,800 349,500 259,900 170,700 114,900 TABLE XXVI - POLY? 'Weight 290,000 207,300 125,500 88,200 62,500 41,400 TABLE XXIV - POLYMER 5% 4 weight Fraction x 10 7.1816 4.8547 1.3138 0.5357 TABLdexv - POLYMER # 5 Wei ght Fraction x 10 7.9905 6.3756 2.9609 1.7342 0.7044 0.3221 # 6 'Weight Fraction x 10 7.9072 6.5402 3.5164 1.5271 0.7716 0.3620 Tangent 0.1241 x 10’4 0.2400 0.0819 Tangent 0.0858 0.2195 0.1368 0.1153 0.0686 Tangent x 10'4 0.1651 x 10'4 0.3660 0.2942 0.1895 FIGURE 9 (17 O I_ 5.0 -— I / WEIGHT / FRACTION / x10 4.0 4 / “ETC LFICU LA}? W? I CHT D It? T- RIBUTION CURVES OF ORIRINAL POLYMER SAMPLES. [U C) I J J l l 100. 200 500 1+00 500 £00 MOLECULAR "HEIGHT X 1'3 -36- {(11.12} HT I‘IUKCTI‘ON x 10 (3 C) 0.0 O I.» O to O 1.0 C) 0 FIGURE 10 TRI BUT IO N (”URI-"ES OR I ":1 NAL PO LY‘IER S AM T”LES . 1 l J l 200 300 1+00 500 MOLECULAR WEIGHT x 103 -37- MO LECULAR WE I GHT D 15 - OF‘ 700 -1 TANGENT x 10 “ FTGURE,11 O.hS C.MO 0.30 0.25 . 0.15 0.10 0.05 0.00 \ - -—-P0LY2ER # 1 ------ POLYMER # 2 |\ ————————Pc-mzm # 3 \ \ \ \ TAX yr coavas DERIVED FROM I \ kULECULAF WEIGHT EISTRIBUTICN | \\ CURVES a: rag ORIGINAL FOLYLER I \ SAgfLES I \ \ , \ | \ I \ t \ / ’1 \aK/ I / I l I l l 1 mm 2a) 3a) mm 5a) mOLhCULAH WEIGHT x 103 -38- LNT x 10"h qv-w U Y ! TAP 0.35 0.50 0.50 O o \A) O 0.10 0.00 IFIGURE 12 POLYMER # h —————— mum-LR # S - POLYMER # 6 TANGEINT CURVES DERIVED FROM MOLECULAR WEIGHT DISTRIBUTION CURVES OF THE ORIGINAL POLYMER SAEFLES l l l l 100 200 300 we 500 MOLECULAR WEIGHT x 103 -39- 600 GRAPHICAL DETERHIKATION 0F INTRINSIC VISCOSITY Intrinsic viscosities for the first three fractions of polymer=# 2 were obtained by determining the specific viscosi- ties of these samples at three different concentrations, divid- ing the specific viscosities obtained by the concentration (in grams of polymer per 100 ml. of solvent) and plotting this value of \Qsp/b versus the concentration. The points plotted approximated straight lines. The y-axis intercepts of these lines were the intrinsic viscosities of the polymer samples usado TABLE XXVII Determination of Intrinsic Viscosities of Three Fractions of Polymer # 2 Fraction Concentration Time of Efflux Rep Rep/C (go/100 m1.) (8°C.) 1 0.1058 90.9 0.700 6.52 0.0529 69.4 0.297 5.62 0.0264 60.2 0.125 4.84 2 0.1056 81.3 0.520 4.93 0.0528 65.3 0.221 4.18 0.0264 58.4 0.094 3.54 3 0.1052 73.3 0.370 3.51 0.0526 62.2 0.166 3.15 0.0263 57.1 0.0673 2.56 The intrinsic viscosities as determined by extrapolation.were 4.40 for fraction #1, 3.14 for fraction=# 2, and 2.30 for fraction # 3. -40- 3+0 IN'T‘WIN'S I" h) C) O O 7‘ DETER‘IINATION OF INTRINSIC VISCOSITY FOR VLDAITIOIE 1, 3 .9: .3 OF POLYMER f“ 7 l l l l l 1 l l J l .05 '- .10 “OKIO‘UFIRATION 1RA’.’.3/'1-JO ‘JL. DETERIJINATI ON OF "PRECIPITABILITY" OF FRACTI ONATED SMWLES "Precipitability" was determined on certain of the poly- styrene fractions in the manner described on page 11. The differential (tangent) curves were plotted. These results are shown in the following tables and graphs. -42- TABLE XXVIII - POLYESR FRACTION 3-1 PRECIPITABILITY 21.09303 %CH30H I 10/1 logIo/I Av.%03308 Tangent 00.0 00.00 100.0 1.000 0.0000 0.00 0.0000 15.0 10.56 100.0 1.000 0.0000 5.28 0.0000 16.0 11.18 100.0 1.000 0.0000 10.87 0.0000 16.5 11.50 87.0 1.149 0.0603 11.34 0.1883 17.0 11.80 72.0 1.389 0.1427 11.65 0.2747 17.5 12.11 68.0 1.471 0.1676 11.96 0.0804 18.0 12.41 63.5 1.575 0.1973 12.26 0.0990 18.5 12.71 60.0 1.667 0.2219 12.56 0.0820 19.0 13.01 57.5 1.739 0.2403 12.86 0.0613 20.0 13.63 52.0 1.923 0.2840 13.32 0.0706 21.0 14.18 48.0 2.082 0.3185 13.90 0.0628 23.0 15.33 43.0 2.325 0.3664 14.76 0.0416 25.0 16.44 40.0 2.500 0.3979 15.88 0.0284 30.0 19.10 36.0 2.775 0.4433 17.77‘ 0.0170 35.0 21.60 33.5 2.985 0.4749 20.35 0.0126 40.0 23.94 32.5 3.075 0.4878 22.77 0.0059 45.0 26.16 31.5 3.174 0.5016 25.05 0.0062 50.0 28.24 31.0 3.225 0.5085 27.20 0.0033 55.0 30.21 30.5 3.278 0.5156 29.22 0.0036 60.0 32.08 30.0 3.333 0.5228 31.14 0.0038 65.0 33.84 30.0 3.333 0.5228 32.96 0.0000 TABLE XXIX . POLYMER FRACTION 3-3 PRECIPITABILITY M1.CH30H %03303 I 10/1 logIo/I Av.%08303 Tangent 00.0 00.00 100.0 1.000 0.0000 0.00 0.0000 15.0 10.56 100.0 1.000 0.0000 5.28 0.0000 17.0 11.80 100.0 1.000 0.0000 11.18 0.0000 18.0 12.41 74.0 1.351 0.1307 12.10 0.2142 18.5 12.71 58.0 1.725 0.2368 12.56 0.3537 19.0 13.01 51.0 1.961 0.2925 12.86 0.1890 19.5 13.31 47.0 2.126 0.3276 13.16 0.1170 20.0 13.63 44.0 2.272 0.3564 13.47 0.0900 21.0 14.18 40.0 2.500 0.3979 13.90 0.0755 22.0 14.73 37.0 2.702 0.4317 14.46 0.0615 23.0 15.33 35.5 2.818 0.4499 15.03 0.0303 25.0 16.44 33.0 3.030 0.4814 15.88 0.0284 30.0 19.10 29.5 3.390 0.5302 17.77 0.0190 35.0 21.60 27.5 3.635 0.5605 20.35 0.0121 40.0 23.94 26.5 3.776 0.5770 22.77 0.0070 45.0 26.16 25.5 3.920 0.5933 25.05 0.0073 50.0 28.24 24.5 4.085 0.6112 27.20 0.0086 55.0 30.21 24.5 4.085 0.6112 29.22 0.0000 TABLE XXX - POLYMER FRACTION 3-5 PRECIPITABIIITY M1.CH 03 %CH OH I I /I logI /1 Av.%CH 0H Tangent 3 3 o o 3 00.0 00.00 100.0 1.000 0.0000 0.00 0.0000 19.0 13.01 100.0 1.000 0.0000 6.00 0.0000 20.0 13.63 100.0 1.000 0.0000 13.32 0.0000 20.5 13.89 98.0 1.020 0.0086 13.76 0.0330 21.0 14.18 91.0 1.099 0.0410 14.04 0.1117 21.5 14.47 82.0 1.220 0.0864 14.32 0.1562 22.0 14.73 72.0 1.389 0.1427 14.60 0.2162 22.5 15.04 62.0 1.612 0.2074 14.88 0.2035 23.0 15.33 55.5 1.801 0.2555 15.18 0.1660 24.0 15.89 48.0 2.082 0.3185 15.61 0.1125 25.0 16.44 44.0 2.272 0.3564 15.16 0.0690 27.0 17.53 40.0 2.500 0.3979 16.98 0.0381 30.0 19.10 36.5 2.740 0.4378 18.32 0.0254 35.0 21.60 34.0 2.940 0.4684 20.35 0.0122 40.0 23.94 32.0 3.124 0.4947 22.77 0.0112 45.0 26.16 31.5 3.175 0.5017 25.05 0.0032 50.0 28.24 31.0 3.224 0.5084 27.20 0.0032 55.0 30.21 31.0 3.224 0.5084 29.22 0.0000 TABLE XXXI - POLYMER FRACTION 4-1 PIECIPITABILITY M1.03303 309308 I Io/T logIo/I Av.%CH30H Tangent 0.0 0.00 100.0 1.000 0.0000 0.00 0.0000 5.0 3.81 100.0 1.000 0.0000 1.90 0.0000 10.0 7.30 100.0 1.000 0.0000 5.56 0.0000 15.0 10.56 98.0 1.020 0.0086 8.93 0.0026 15.5 10.87 97.0 1.031 0.0133 10.72 0.0152 16.0 11.18 71.0 1.408 0.1486 11.02 0.4364 16.5 11.50 63.0 1.588 0.2008 11.34 0.1630 17.0 11.80 57.5 1.739 0.2403 11.65 0.1317 17.5 12.11 53.0 1.886 0.2755 11.96 0.1135 18.0 12.41 50.0 2.000 0.3010 12.26 0.0850 19.0 13.01 45.5 2.198 0.3420 12.70 0.0683 20.0 13.63 42.5 2.355 0.3720 13.32 0.0484 21.0 14.18 40.0 2.500 0.3979 13.90 0.0471 23.0 15.33 37.5 2.665 0.4257 14.76 0.0242 25.0 16.44 36.0 2.778 0.4437 15.88 0.0162 30.0 19.10 33.0 3.030 0.4814 17.77 0.0142 35.0 21.60 31.5 3.175 0.5017 20.35 0.0081 40.0 23.94 30.5 3.280 0.5159 22.77 0.0061 45.0 26.16 30.0 3.333 0.5228 25.05 0.0031 50.0 28.24 30.0 3.333 0.5228 27.20 0.0000 -44- H1.CH 0H TAELE XXXII - PCLYHER FRACTION 4-3 flCHSOH 0.00 7.30 10.56 11.18 11.50 11.80 12.11 12.41 12.71 13.01 13.63 14.73 16.44 19.10 21.60 23.94 26.16 28.24 30.21 32.08 PRECIPITABILITY I 0/1 100.0 1.000 100.0 1.000 100.0 1.000 99.0 1.010 98.0 1.020 54.0 1.851 42.0 2.380 37.0 2.701 34.5 2.900 32.5 3.075 30.0 3.333 26.5 3.774 23.5 4.252 21.0 4.755 19.0 5.258 18.0 5.555 17.0 5.885 16.5 6.055 16.0 6.250 16.0 6.250 IogIo/T 0.0000 0.0000 0.0000 0.0043 0.0086 0.2674 0.3766 0.4315 0.4624 0.4878 0.5228 0.5768 0.6286 0.6772 0.7208 0.7447 0.7698 0.7821 0.7959 0.7959 Av.%CHson 0.00 3.65 8.93 10.87 11.34 11.65 11.96 12 .26 12.56 12.86 13.32 14.18 15.58 17.77 20.35 22.77 25.05 27.20 29.22 31.14 TABLE XXXIII - POLYMER FRACTION 4-4 308 03 / 3 0.00 10.56 11.18 11.50 11.80 12.11 12.41 12.71 13.01 13.31 13.63 14.18 14.73 15.33 16.44 19.10 21.60 23.94 26.16 28.24 30.21 PRECIPITABILITY I o/I IogIo/I 100.0 1.000 0.0000 100.0 1.000 0.0000 100.0 1.000 0.0000 99.5 1.005 0.0022 99.0 1.010 0.0043 98.5 1.015 0.0065 54.5 1.835 0.2636 42.5 2.352 0.3714 37.0 2.702 0.4317 34.5 2.898 0.4621 32.5 3.075 0.4878 30.0 3.333 0.5228 28.5 3.508 0.5451 27.0 3.705 0.5688 25.0 4.000 0.6021 22.5 4.445 0.6479 20.5 4.880 0.6884 19.5 5.130 0.7101 18.5 5.405 0.7328 18.0 5.555 0.7447 18.0 5.555 0.7447 -45- ‘ Av.% 3 0.00 5.28 10.87 11.34 11.65 11.96 12.26 12.56 12.86 13. 16 13.47 13.90 14.46 15.03 15.88 17.77 20.35 22.77 25.05 27.20 29.22 CH CH Tangent 0.0000 0.0000 0.0000 0.0069 0.0134 0.8627 0.3850 0.1830 0.1030 0.0847 0.0565 0.0491 0.0303 0.0183 0.0214 0.0102 0.0113 0.0059 0.0070 0.0000 Tangent 0.0000 0.0000 0.0000 0.0069 0.0070 0.0071 0.8570 0.3593 0.2010 0.1013 0.0803 0.0636 0.0406 0.0395 0.0300 0.0172 0.0162 0.0092 0.0102 0.0057 0.0000 t 416- mm. )Zw ”WW Orux WM 7 Fl...” M. r.“ 01f. _ q _ q — _ _ 1 ‘ J O.) 1 I. Ii. - o: . ..1 l A fit. o...HmmH..m neuduéo mm“? m) fl “Emma 35.4 m: m .4. 520,36. ,8 mun/waflwm15wmm mmxmh mom mambmuo .L. r-a Hmm.“ c- f1 680% a a 1 a ._ m 4850332 IIII \ 1 m z.-.HHU§w lquIII. \ . H Zagrwmdxw: III \ 7 .1 Jon. 1: TEE flirt} .' irrv 91:“ Y . 4:7- mm 63,603 8,343 0523 mm ”2.05854 5.5.855 n a Essen no mzomyox30 MHHAHmdeHmHommm ._ 228$; .l..I| m 220.03% 17!! H ZOHHQm Do 4m ma "mm—meg C ‘) 0H5, \1... la. ‘5? Once mm.o 111309171. Hzmnzomlzoz Bzmw. 5m 3 m l OH.O _ 1 3.0 _ _ _ mimmmwhm _ .l mh.o . mvmqmwfifi 08.5.2.2.“ mm mvawmmdn..uw< annexe/mph” 3 m mmuwuwi 03 ngmbufimfi mmumé , L mu>xou - . ,\rb >4 .Hqum:HHu.Hon.xm 311317....“ QM>-UH againd Smmzm. 0) —_ Al \(xfiu C 4.. .)4<) . I'llll ~v «it rLfo ,E .- “0 ’4‘]4.‘.\/. J‘WH " 1 :7 {.143 f nll _. a zaflbumflm O\ O Q MISSIWJ. DISCUSSION Polymer samples used in this work were made by emulsion polymerizing styrene under carefully controlled conditions. A considerable quantity of polymer prepared by a known and reproducible method was necessary for each sample. Conditions of polymerization were chosen so that the samples would vary in characteristics, yet be obtainable by procedures studied by other workersl'2'3’4’5 within this laboratory. The measurement of "precipitability" by light extinction required that the total amount of polymer in solution remain suspended as the non-solvent addition occurred. Sedimentation of polymer would result in a discontinous “precipitability" curve. Trials with solutions of different concentration indi- cated that the use of a two-hundredths molar solution of poly- styrene in butanone was most satisfactory. The use of a more concentrated polymer solution resulted in sedimentation of the solid polymer during the determination, and solutions of lesser concentration did not give a suitable extinction of light. Fractionation of the polymer samples was based on the work of Schulz and Dinglinger,16 who gave no specifications for actual procedure. Morey and Tamblyn17 stated that concentration of the polymer solution had little effect on the quality of the frac- tionation. It was necessary to develop a suitable procedure, and in its development it was discovered that the fractionation -50- was simpler if carried out in dilute solutions containing approxi- mately ten grams of polystyrene in one liter of butanone. The precipitation by methanol of concentrated solutions of polystyrene resulted in instantaneous precipitation in the. area where the methanol entered the solution. 'With.vigorous agitation, the precipitated polymer returned to solution, but further addition of methanol gave the same result, leading to a slow, laborious procedure. With the dilute solution.mentioned above, precipitation occurred gradually. Addition of methanol turned the poLymer solution milky white as precipitation occurred, and sedimenta- tion of the polymer took place in twenty-four hours or less. The poLymer separated as a gelatinous mass, which upon.washing with methanol quickly solidified. The quality of the fractionation varied with the original polystyrene samples used. The molecular weight range of a given polymer determdned how easily it could be fractionally precipi- tated, and into how many fractions it could be separated. As an example, the fractionation of pohwmer=# 4 is compared to the fractionation of polymer:# 3. Polymer=# 4 was separated into four fractions with distinctly different average molecular 'weights, 1.0. the "precipitability" tangent curves for three of the four fractions Obtained each showed distinctly different maximum. (Figure 17) The areas under the "precipitability" -51- tangent curves for the three fractions were very small laterally, which indicated that these polymer fractions had a narrow'range of molecular weight. The lateral width of the area under the "precipitability" tangent curve for polymer # 4 (Figure 6) was quite small, which indicated that the original sample had a narrow range of molecular weight. The conditions of polymeriza- tion for this sample were such that a polymer with a limited range of molecular weight was expected. The lateral width of the area under the "precipitability" tangent curve for polymer-# 3 (Figure 5) was much greater than that for polymer # 4, which indicated a greater range in.mole- cular weight. The “precipitability" tangent curves for three fractions of polymer # 3 (Figure 16) each showed a distinct maxi- mum, and lateral width of area under the curves indicated a much narrower range of molecular weight than in the original sample. Comparison of the "precipitability" tangent curves for fractions of polymer=# 4 (Figure 17) and polymer-# 3 (Figure 16) indicated a greater homogeneity of molecular weight in the fractions of polymer-# 4. The samples obtained by fractionation.might be fractionated a second time. A second fractionation into constituents of dif- ferent molecular weight would be possible on samples where the lateral width of the area under the "precipitability“ tangent curve for the sample in question was large. Some overlapping -52- of molecular weight between the last and first fractions from two successive samples would occur in the second fractionation. Where the lateral width of the area under the "precipitability" tangent curve was very small, little could be gained by a second fractionation, as almost complete precipitation would occur suddenly, giving little, if any, separation. In recent years controversy has arisen concerning the proper approach to methods of molecular weight determination of polymers. The values of average molecular weights determined by use of the ultracentrifuge and osmometer are more hearty absolute values than those determined by viscosity, but the length of time required for these determinations, and the in- accessibility of the necessary equipment often forces workers in this field to use other methods. The determination of average molecular weight by viscosity measurements has been the subject of more controversy than the determination of molecular weight by other methods. It is quite certain that the original Staudinger equation [54 = 161 should be modified, but investigators in this field cannot agree upon what the modification should be. Two other versions of the Staudinger equation have been proposed: (1) PL] = KIT/constant (2) [4] = m” At the present time equation (2) holds the greatest promise. It is quickly apparent that the original Staudinger equation is a special case of equation (2) with beta equal to one. As a comparison of the usefulness of these equations vis- cosity measurements were made on solutions of certain polymer samples at different concentrations, and their intrinsic vis- cosities obtained (Figure 13). These values of [\J were used in equation (2) along with values of K and beta obtained by Goldberg, Hohenstein and Mark.21 In certain cases molecular weights obtained from the Staudinger equation and equation (2) were in fair agreement, in other cases they varied widely. The determination of K and beta for equation (2) is still in the experimental stage and values available are not too reliable. As an example, the values of K and beta for polystyrene polymerized at 60° 0. are 1.28 x 10" 4 and 0.7, respectively. These constants, according to Goldberg, Hohenstein, and Mark,21 cover a molecular weight range of 550,000 to 2,000,000. 011 the first fraction from polymer # 7 the intrinsic viscosity was 4.4. Using equation (2) 4.4 = 1.28 x 10'4 x M9'7 a value of 3,022,000 was obtained, which was approximately eight times the value obtained by the use of the Staudinger equation. Since this pair of values for K and beta do not cover the molecular weight range of the polymer in question, good corre- lation was not expected. According to Goldberg, Hohenstein and -54- Mark,21 the values of K and beta covering this range of molecular 'weight are for polystyrene polymerized at 120°C. and are 5.5 x 10'3 and 0.8, respectively. Using these values and the same'value of 4.4 for the intrinsic viscosity in equation (2) 4.4 = 5.5 x 10'3 x Mo's a value of 4255 was obtained, which was just slightly over one percent of the value obtained by use of the Staudinger equation. Many factors affect the determination of the values of K and beta, and separate constants must be determined experimentally for polystyrene polymerized under various conditions. Type of system used in the polymerization, catalyst, catalyst concentra- tion, temperature, and time allowed for polymerization could all exert an influence on these constants. Another method for the determination of molecular weight of polystyrene from'viscosity measurements has been proposed by Kemp and Peters.23 It is based on the Arrhenius relation: logrLr = C where “Y is the relative viscosity. K Kemp and Peters proposed equation is M = logrLr x K C They determined the value of K to be .45 x 104 for low molecular ‘weight polystyrene in benzene or chloroform. Price and Adams24 modified K for toluene solutions of high molecular weight -55- polystyrene, and determined the value of the constant to be .74 x'104. Using this value of K for the determination of the average molecular weight of polymer~# l v = 1 84.1 74 104 1.. 0g 5—33 x e x 00101 a value of 144,000 is obtained, as compared to a value of 312,200 obtained from the Staudinger equation. The values obtained for the average molecular weights of the six polymers used in this work are listed below. Molecular'Weights by Polymer Sample Kemp-Peters Staudinger Equation Equation 1 144,000 312,200 2 98,500 198,700 3 105,100 215,400 4 202,000 486,500 5 159,000 354,700 6 85,500 167,900 The values for the average molecular weight as determined by the Kemp-Peters equation.vary from 41 to 51 percent of the value obtained for the value of the average molecular weight as determined by the Staudinger equation. This correlation is a great improvement over that obtained in the case of the modified Staudinger equation. In this work the primary objective was to obtain molecular weight comparisons rather than absolute molecular weight determina- tions. Therefore, the use of the Staudinger equation is justified -56- because it is as suitable as other pr0posed equations for compara- tive purposes. The same method for the determination of molecular weight is used in all instances, therefore, comparisons may be made with reasonable accuracy. The constant used in the Staudinger equation is one which was determined experimentally for polystyrene, and which has been used extensively. The molecular weight distribution curves obtained from the fractionation data were the typical S-shaped curves (Figures 9 and 10) comparable to those published in work by other investigators. Before comparison with ”precipitability" curves it should be noted that the distribution curves should be reversed. On the distribu- tion.curves the molecular weight increases from left to right on the "precipitability" curves the equivalent of molecular weight, i. e., solubility of polymer, decreases from left to right as the high molecular weight polymer was the first to be precipitated. This same characteristic was carried through to the tangent curves derived from.the "precipitability" and molecular weight distribu- tion curves. Actual point by point comparison of these tangent curves was not possible, as it was impossible to use the same units and scales as ordinates and abscissas. The comparison of the maxima of the tangent curves was made possible by a method based on average molecular weight in the following fashion: The maximum on the mole- cular weight distribution tangent curve for polymer # 3 (Figure 11) occurred at a molecular weight of 84,000. The average molecular weights of the fractions obtained from polymer-# 3 showed that this value was included between fractions four and five. (Table XVII). These two fractions were precipitated from solutions be- tween thirteen and fifteen percent non-solvent concentration (Table I). Interpolation for the value of 84,000 between the values of 71,100 (fraction 5) and 93,700 (fraction 4) gave a percent non-solvent concentration of 13.80%. This value is one- tenth of one percent from the value obtained for the maximum of the tangent curve as calculated from the "precipitability” curve (Table IV). The same use of data from the other five curves gave corres- ponding results. The agreement of percentage values was not as good in all cases as it was with polymer # 3, but all of the comp parisons were close to the values calculated from.the "precipi- tability“ curve. Better agreement could be obtained by more extensive fractionation, which.would result in more exact curves. Polymer Maximum Percent maximum Percent Difference Sample Non-solvent Non-solvent "Precipitability" Calculated from Tangent Curves Distribution Tangent Curves 1 11.02 11.99 - 0.97 2 12.86 13.54 0.68 3 13.76 13.86 0.10 4 11.65 11.42 0.23 5 12.26 11.43 0.83 6 11.96 12.99 1.03 -58- As the maxima of the two types of tangent curves are comparable, it is reasonable to assume ”precipitability" tangent curves could serve as an approximation to the distribution tangent curve with proper mathematical treatment. The tangent curves as derived in this work are graphical differentiations of the "precipitability" and molecular weight distribution curves. If the equations of the "precipitability" and molecular weight distribution curves could be determined, actual point by point comparison could be made and these equations might then be differentiated mathematically. If this were possible, a useful tool for fast approximations of mole- cular weight distribution of a given polymer sample would be pro- vided, eliminating the slow laborious procedure of fractionation. This method would necessarily be limited to polystyrene, but fur- ther work might extend it to other types of pomeers. Several extensions of this work are possible. There are many possibilities in the field of molecular weight, as present incon- sistencies are extensive. As considerable equipment is necessary for work with the ultracentrifuge, the osmometer would be the logi- cal choice for continuation in this work. Data and results obtained in this manner could be compared with data and results obtained from viscosity measurements. There is still a great deal to be done in the calculations of K values, as well as values of beta, providing it is proven that the equation using beta is the proper one. The fractionation procedure could be improved by developing some type of container which would facilitate separation of the precipitated polymer sample and the supernatant solution. It is also possible that fractionation could be carried out in conjunc- tion with the "precipitability" apparatus, resulting in a more uni- form separation. As has been.mentioned previously, a mathematical treatment of this work could lead to a valuable shortcut in determining molecular weight distribution of polymers. It is difficult to say how exten- sive this work would have to be to obtain the proper correlation between these two types of curves. -50- 1. 2. 3. 4. 5. SUVMARY A new and simpler method based on "precipitability" is proposed for determination of integral average molecular weight distribu- tion curves of emulsion polymerized polystyrene. The above method has been applied to differential distribution curve maxima and the differential "precipitability" curve maxima with suitable agreement. In no case was the variation equal to more than the equivalent value of 1.03%. Fractional precipitation of polystyrene is possible using butanone as a solvent and a lower member of the alcohol series as a non-solvent. The choice of alcohol has little effect upon the fractionation. The extent and quality of fractionation of a given polymer is dependent upon the individual sample, and the conditions under which it was prepared. It is advantageous to use relatively dilute solutions in the fractional precipitation, and still more dilute solutions in "precipitability" measurements. -51- 14) 15) 16) 17) 18) 19) 20) 21) 22) 23) 24) REF EREI‘TC ES liorgan, D. 11., 1.1.8. Thesis, Michigan State College (1946) Yang, P. T., M. S. Thesis, Michigan State College (1947) Hallenbeck, v. 0., M. s. Thesis, Michigan State College (1948) Mihina, J. 8., M. S. Thesis, Michigan State College (1948) Loring, T. M... 1.5. S. Thesis, Michigan State College (1948) Staudinger, J: Bauer, Ber. .6522, 222 - 34, (1930) Staudinger, Helv. Chim. Acta, 15, 213 - 21 (1932) Staudinger, Ber., £52, 267 - 79 (1932) Svedberg, z. Phipik Chem. 123, 65 — 77 (1926) Signer, Kolloid Z., 19, 24 - 6 (1935) Signer, Trans. Faraday Soc., §_2_, 296 - 307 (1936) Schulz, Angew. Chem., :9, 863 - 5 (1936) Dobry, Kolloid Z., §_1_ 190 - 5 (1937) Debye, J. Phys. 6: Colloid Chem., 21, 18 (1947) Schulz 8c Husanann, Z. Physik. Chem., @514, 187 - 213 (1936) Schulz d: Dinglinger, Z. Physik. Chem., 343, 47- 57 (1939) Morey a; Tamblyn. J. Phys. a Colloid Chem., 51, 721 - 46 (1947) Schulz, Z. Physik. Chem., 5319, 321 (1937) Adams 8: Powers, Ind. Eng. Chem. (Anal. Ed.) 15., 711 - 14 (1943) Alfrey, Bartovics Mark. J. Am. Chem. Soc. 65, 2319 - 23 (1943) Goldberg, Hohenstein and Mark. J. Poly. Sci. _2_, 503 (1947) Kemp and Peters. Ind. Eng. Chem. 33, 1263 - 69 (1941) Kemp and Peters, Ind. Eng. Chem. 33, 1097 - 1102 (1942) Price and Adams, J. Am. Chem. Soc. 61, 1675 - 80 (1945) -62 .- Md 2 6 ’51: Ml? LIBRARY IN , as '61. t v,” . meme gr .1111 7 9’64 hum KUG 2"1'1992' mm a ' b: i _ gang‘s ‘3'"93 ' . - T547-2 217598 1292 Krell 11W 11H]1111111111111111II 3 1293 02446 7411 ll!