. y - 9d u.‘...1*“'¢ a... m.*‘-‘ v J‘J‘ - THE MOLECULAR WEIGHT AND MOLECULAR WEIGHT DISTRIBUTiON OF SOME $TYRENEv-MALEIC ANHYDRlDE COPQLYMERS Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSETY Chades E. McCoy Jr. 196:6 'J B r 3" g V mmmmmmmnrmm L 31 ' ' f 293 01033 9805 UmVCISUT j --_— Metm 00p01ymer: 30 35,000 coPolymer: catalyst, 1 Mole ChI‘OmatOg measur‘eme fr“ gel °°rPelate by va por Mole the gel I An 6 in estima copolYme] ABSTRACT THE MOLECULAR WEIGHT AND MOLECULAR WEIGHT DISTRIBUTION OF SOME STYRENE-MALEIC ANHYDRIDE COPOLYMERS by Charles E. McCoy Jr. Methods for preparing styrene-maleic anhydride copolymers with average molecular weights from 5,000 to 35,000 have been developed. The effect on the copolymerization of varying solvent, temperature, and catalyst was studied. Molecular weights were determined via gel permeation chromatography, vapor pressure osmometry, and viscosity measurements. Number average molecular weights calculated from gel permeation chromatography were contrasted and correlated with number average molecular weights obtained by vapor pressure osmometry. Molecular weight distributions were derived from the gel permeation chromatograms. An approximate "K" and "a" were determined for use in estimating molecular weights of styrene-maleic anhydride capolymers by measuring viscosity of the copolymers in acetone solution. THE MOLECULAR WEIGHT AND MOLECULAR WEIGHT DISTRIBUTION OF SOME STYRENE-MALEIC ANHYDRIDE COPOLYMERS By Charles E. McCoy Jr. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department Of Chemistry 1966 T0 SHIRLEY, CAROL, AND LINDA ii ACKNOWLEDGEMENT The author expresses his sincere appreciation to Professor Ralph L. Guile for his guidance, understanding, and inspiration. The author is grateful to the Dow Chemical Company for financial support. A special thanks is extended to Dow's Polymer Analysis Laboratory for use of the gel permeation chromatography unit. Finally, thanks to Moustafa Sharabash for his valuable suggestions and close cooperation. iii TABLE OF CONTENTS Polymerization Technique Purification and Isolation Technique Preparation of Samples for INTRODUCTION. HISTORICAL. . REAGENTS. . . EXPERIMENTAL. I. II. III. A. B. C. D. E. F. IV. Copolymerization Copolymerization Copolymerization Copolymerization Copolymerization Testing . in Dimethoxymethane in Benzene. . . in Tetrahydrofuran. in Bromochloromethane in Isopropyl Benzene. Samples From Other Sources . . . Evaluation of Samples. . A.‘ B. C. Vapor Pressure Osomometry. . . . Gel Permeation Chromatography. . Viscosity Measurement. iv Page 10 15 16 l7 l8 l8 19 20 23 2h 25 28 28 35 #5 TABLE OF CONTENTS - Continued Page CORRELATION OF RESULTS AND DISCUSSION . . . . . . . . 53 I. General. . . . . . . . . . . . . . . . . . Sh II. Gel Permeation Chromatography. . . . . . . 57 III. Viscosity Measurements . . . . . . . . . . 59 IV. Chain Transfer . . . . . . . . . . . . . . 60 63 VI. Benzene Extraction . . . . . . . . . . . . 6h V. Isopropyl Benzene Stability. VII. Structure of the Polymer . . . . . . . . . 65 SUNWRY O O O C C O O O C O O C O C O O O O C C C C C 68 LIST OF TABLES TABLES l. 2. 3. A. 5. 6l 9. 10. ll. 12. Samples Prepared for Testing . . . . . . . . . Determination of K with Benzoic Acid for the Vapor Pressure Osmometer . . . . . . . . . . . Vapor Pressure Osmometry Results for Sample 11 Mn Values From Vapor Pressure Osmometry Data . Molecular Weight Calculations for Sample 12. . Molecular Weights From Gel Permeation Chromatography Data. . . . . . . . . . . . . . Results From Viscosity Measurements on Sample 12 O O O O O O O O O O O O .0. O O O O O O O O Intrinsic Viscosity Results and Calculated7Mn values 0 O C C C C O C O O . C C C O O O O O 0 Mn Estimated From ”JP/c at C 0.5g/lOO milliliters. 0 O O O O O O O O O O O O O O O 0 Results From Copolymerization in Mixed SCI-vents O O O O O O O O O O O O O O O O O O O Compagison of Mn by Vapor Pressure Osmometry with Mn by Gel Permeation Chromatography . . A Comparison of Mn Values Obtained From Intrinsic Viscosity Data with Mn Values for Vapor Pressure Osmometry Data. . . . . . . . . vi Page 27 31 32 3h A2 A3 A7 A9 52 67 69 72 qvn-v' filed} to #— C1). FIGURES l. 10. ll. 12. 13. 10. LIST OF FIGURES Schematic drawing of the apparatus used for the polymerization experiments. . . . . . . Calibration curve utilizing benzoic acid, (K - #19) for the vapor pressure osmometer. . . Vapor pressure osmometry plots. . . . . . . . . Waters liquid chromatography assembly . . . . . Gel permeation chromatogram of sample 12, calibration curve number 2. . . . . . . . . . . Calibration curves for gel permeation chromatography. . . . . . . . . . . . . . . . . Cumulative distribution plot for sample 12. . . Intrinsic viscosity extrapolations. . . . . . . Log D7] versus log M for samples 10, ll, l2, 15, l7, and 2h. 0 O O O O O O O O O O O O O O O ESP/c versus Mn at concentrations of 0.5 grams per 100 milliliters . . . . . . . . . . . . . . ma versus polymerization temperature in isopropyl benzene, no added initiator . . . . . Mn by vapor pressure osmometry versus Mn by gel permeation chromatography . . . . . . . . . Intrinsic viscosity versus Mn . . . . . . . . . Detection of benzene extractable materials via permeation chromatography . . . . . . . . . vii Page 26 30 33 39 #0 Al #4 #8 50 51 8? 7O 71' 73 INTRODUCTION Styrene-maleic anhydride copolymers have been of interest to this laboratory for several years. The chemistry of this copolymer system has been the subject of several publications (1). The purpose of this study was twofold: l. The synthesis of a broad range of low molecular weight copolymers with number average molecular weights less than 20,000. 2. The characterization of these copolymers by molecular weight and molecular weight distribution. Little has been published on the molecular weight axni molecular weight distribution of equal molar styrene- maleic anhydride copolymers. A better understanding of these properties and methods for controlling them will be important to future work in this laboratory and will fill a void in the existing literature. . ., .J. Misery... v2»... M,4... \ v , .. Masha?“ HISTORICAL l. Cepolymerization was first described by Klatte (2) in 19lh. Later studies revealed that copolymerization was different from homopolymerization in that some monomers, such as maleic anhydride, homopolymerize with difficulty or not at all but readily copolymerize with monomers like styrene and vinyl chloride (3). The first studies on the mechanism of copolymerization were conducted by Dostal (1.). Three independent publica- tions by Alfrey and Goldfinger (5), Mayo and Lewis (6), and Wall (7) in 19Al+ described a "copolymerization equation" to account for the observed copolymer composition from any two comonomers. 'Copolymerization has since been described ‘in many publications. Several explanations (8) have been advanced for the observed difficulty in the attempted homepolymerization of maleic anhydride and the observed ease of copolymerization ' With other monomers. Copolymerization with many comonomers favors 1:1 copolymers. Stilbene and maleic anhydride give 1:1 c0polymers regardless of the initial monomer composition (9) . Styrene and maleic anhydride give essentially 1:1 coPolymers except when the monomer mixture contains a high “1°18 percentage of styrene (> 80%) (10). Alfrey and Lanvin (11) have shown that four distinct processes are involved in Copolymerization and that these processes are governed by fOur prepagation rate constants. With styrene-maleic anhYdI‘ide systems, the four pr0pagation rates become three, Since maleic anhydride has practically no tendency to add to its own radicals--thus favoring the 1:1 copolymer composition. This conclusion has been supported by Ham, gt 31. (12) and Bartlett and Nozaki (13). Three sets of rl and r2 values are given for cepoly- merization of styrene and maleic anhydride: Styrene is monomer 2 r1 r2 reference 0 0.01 (11.) O 0.0h2 (11) 0 0.02 (15) If rlr2=0, each radical prefers to react exclusively with the other monomer and the initial copolymer from any concentration of comonomers will be alternating (16). The relative reactivity of the maleic anhydride radical for styrene is some twenty (17) times greater than the styrene radical for styrene. Maleic anhydride forms colored molecular complexes with styrene and other aromatic compounds (18). A suggested structure of these materials is one consisting of pairs of radical ions in which styrene has donated an electron to maleic anhydride. Mayo (lb) and co-workers have suggested that the larger the difference in polarity or donor acceptor properties between two comonomers, the greater will be the alternating tendency. They have devised a donor acceptor series in which the substituent groups are ranked as follows: Donor Acceptor Series R-O- >~. :3 H20=CH- .21 f3 06%" g R-CHz- E '0 H- 4 E 01- '5 g R-CO- g R-O-CO- "O A NEC- When two comonomers are well separated in this series, they will have a marked tendency to alternate in copolymerization. Styrene and maleic anhydride are well separated. The structure of a 1:1 copolymer of styrene and maleic anhydride can be illustrated by the following repeating unit Alfrey and Price (19) developed the cepolymerization parameters Q and e which fill the need for more general and constant factors for characterizing monomers. Q is a measure of general monomer reactivity and e depends on the polar pr0perties. These two parameters should be constant and unique for a given monomer regardless of the copoly- merization system. They are calculated as follows (20): e2 e e1 I (-1n rlrz)i 02 - Ql/rl exp [-el(e1-e2)] . Styrene is taken as the standard for the Q-e scheme and values of Q=1.00 and e'-0.80 are assumed. The r1 value for maleic anhydride is 0. In such cases, a small finite value must be assumed in order to obtain a reasonable value for e. Young (20) reports the Q and e values for styrene and maleic anhydride as follows: e Q Styrene -0.80 1.00 Maleic Anhydride 2.25 0.23 The positive 0 value for maleic anhydride indicates that it is an "electron poor" monomer. The large difference between the two e values (-0.80 to 2.25) suggests that 1:1 copolymers will be obtained when styrene and maleic anhydride are polymerized. This large difference in polarity supports the "radical complex" theory (21) which suggests that polymerization is preceeded by the formation of a "radical complex" which produces 1:1 copolymers virtually independent of any styrene excess available. While the polymerization parameters for styrene- maleic anhydride systems have been determined by a number of workers, the resultant polymers have not been characterized. The techniques for characterization are known, but apparently have not been applied specifically to styrene-maleic anhydride copolymers. The recent literature is dominated by process research and techniques (22) which give industrial copolymers for specific uses. The average molecular weights are sometimes reported (23), but not the copolymer distribution. A polymer or copolymer sample consists of a homologous mixture of molecules having similar or like repeating units. Determination of molecular weight by colligative methods (i.e., osmotic pressure, vapor pressure lowering, freezing point depression, etc..) provides an actual count of the number of solute molecules. The result is a value usually referred to as the number average molecular weight (Mn). When molecular weight is determined by light scattering, the nature of the process is such that the larger particles contribute more to the scattering than the smaller ones. The molecular weight obtained by this method is usually referred to as the weight average molecular weight, MW. nw is always greater than Mn except for monodisperse systems in which they are equal. The ratio Mw/Mn has been referred to as a measure of polydispersity. 9 Neither Mn or aw satisfactorily characterize a polymer system. Values for Mn.and MN on a given polymer sample give an indication of the molecular weight distribution, .but cannot be used to construct a complete molecular distri- bution curve. A new technique, gel permeation chromatography, allows rapid determination of molecular weight and molecular weight distribution. This is a technique for fractiona- tion of samples according to molecular size on a polymer gel column using the principles of liquid phase chromatog- raphy. Gel permeation chromatography was introduced by Moore in 196A (20). The growth and acceptance of this method has been rapid and commercial units are available (25). The importance that gel permeation chromatography has assumed in polymer chemistry has been paralleled by studies to ascertain the details of the mode of separation. Moore and co-workers have continued their work (26) and have been joined by several others (27). Publications relating to the application and utilization of gel permeation chroma- tography units have appeared (28). This method is rapidly supplanting the technique of polymer fractionation from solvent by fractional addition Of non-solvent and removal of the precipitated fractions. Average molecular weight values can also be calculated from gel permeation chromatography data. REAGENTS 10 (a) (ta) (c) 11 Styrene (The Dow Chemical Company) was washed with three successive portions of 10% sodium hydroxide to remove the inhibitor. This was followed by repeated water washings until the washings were neutral to litmus. The resultant styrene was stored over anhydrous sodium sulfate in a refrigerator for one week. The dry styrene was distilled under reduced pressure, and the portion distilling at 36-37°C at 30 mm. of mercury was collected and stored in the refrigerator over anhydrous sodium sulfate until used. Prior to use, the styrene was tested for the presence of polystyrene by addition of methanol to a small sample. Since polystyrene is insoluble in methanol, the absence of a precipitate indicated that the styrene was free of polystyrene. Maleic Anhydride Maleic anhydride (Fisher Scientific Company, reagent grade) was further purified by vacuum distillation. The sample used was collected under reduced pressure (15 mm. of mercury) at 56-580C. The melting point was 52.5:100. The sample was ground to a coarse powder, bottled, and stored in'a desiccator over calcium chloride. Benzoyl Peroxide Benzoyl Peroxide (Eastman Kodak, reagent grade) was used as received. The bottled material was stored in a refrigerator. (d) (e) (f) (s) 12 2,2'-azobis (2-methy1 propionitrile) The 2,2'-azobis (2-methyl propionitrile) (Eastman Chemical, reagent grade) was used without further purification. Benzene Thiophene free benzene was washed three times with concentnated sulfuric acid. This was followed by a wash with a 10% solution of sodium bicarbonate and then water washings until the washings were neutral to litmus. The resultant benzene was stored over calcium chloride and then over metallic sodium. It was distilled from metallic sodium immediately prior to use. The fraction used distilled at 80°c:0.5°c at atmOSpheric pressure. Tetrahydrofuran Tetrahydrofuran (Eastman Chemical Company) was stored over solid potassium hydroxide for at least one week. It was then refluxed with lithium aluminum hydride for a minimum of 12 hours and was distilled immediately prior to use. The fraction used distilled at 6A-65°C at atmospheric pressure. Carbon Tetrachloride Carbon tetrachloride (The Dow Chemical Company) was washed with a 10% solution of potassium hydroxide in a 3:1 mixture of water and ethanol. The mixture was stirred vigorously for 30 minutes at 50-5500. The aqueous layer was separated and the alcohol removed (h) (i.) (J) (1:) 13 from the carbon tetrachloride by several water washings followed by washing with small portions of concentrated sulfuric acid. Finally the carbon tetrachloride was washed with water to remove the last traces of acid. The resultant carbon tetrachloride was dried over calcium chloride. The product was distilled immediately prior to use. The fraction used distilled at 76-77°C at atmospheric pressure. Dimethoxymethane ' Dimethoxymethane (Eastman Organic Chemicals, reagent grade) was used as received. The boiling point was uhi0.5°C at atmOSpheric pressure. Isopropyl Benzene Isopropyl benzene (The Dow Chemical Company) was dried over calcium chloride, refluxed with lithium aluminum hydride for a minimum of four hours and finally distilled from lithium aluminum hydride under nitrogen. The material used distilled at 15210.500 at atmospheric pressure. Bromochloromethane Bromochloromethane (The Dow Chemical Company) was dried over calcium chloride and distilled immediately prior to use. The fraction used distilled at 69i0.5°0 at atmOSpheric pressure. I Petroleum Ether Petroleum ether (Eastman Chemical Company) was dried over sodium and was filtered prior to use. “The boiling range was 60-9000 at atmospheric pressure. (1) 1h Acetone Acetone (Eastman Chemical Company, spectro grade) was used as received. The boiling point was 5620.500 at atmo3pheric pressure. EXPERIMENTAL 15 a! 592‘ t-‘la UAJ LIQ 16 I. Polymerization Technique The following technique was used for all the copoly— merizations conducted during the course of this investiga- tion. Figure l is a drawing of the polymerization apparatus. The reaction vessel was a one liter, round bottomed, three necked flask. The three necks were fitted with a nitrogen bubbler, stirrer (ground glass joints), and reflux condenser (with calcium chloride drying tube). The temp- erature was raised to the desired level by means of an oil bath heated with a nichrome heating coil controlled by a variac. A 50:50 mole ratio of comonomers was used in each copolymerization. This was 21.6 grams of styrene and 20.h grams of maleic anhydride in 500 milliliters of solvent. Anhydrous conditions were maintained at all times. The type and amount of initiator, the temperature, and the reaction times were varied. The reaction flask was heated to the desired temp- erature, and nitrogen was passed through the empty vessel via a nitrogen bubbler. Next #00 milliliters of solvent were added to the reaction vessel and stirring was started. The temperature was equilibrated at the desired level; and the maleic anhydride, initiator, and styrene, all weighed to 0.001 grams, were added quantitatively in the order given. Solvent was used to effect the quantitative addition of the comonomers and the initiator to the reaction flask and 17 to bring the total volume of solvent in the reaction vessel to 500 milliliters. After the pre-determined time period had elapsed, the heat was removed and the copolymerizations were allowed to cool to room temperature, with stirring, for approximately 30 minutes. In some reactions, the copolymer precipitated during the course of the polymerization and in others remained in solution. Two procedures were deve10ped for obtaining the copolymers from the reactions as fine white powdered material. These procedures are described in the next section. II. Purification and Isolation Technique The copolymers that precipitated from the reaction solvent were vacuum filtered (Buchner). The filter cake was thoroughly washed with the polymerization solvent. The samples were stored in a vacuum desiccator over calcium chloride at room temperature and reduced pressure (0.2mm. of mercury) where most of the polymerization solvent was removed. Then the samples were extracted with benzene for 48 hours by Soxhlet extraction. Finally the copolymers were dried under reduced pressure (0.2mm. of mercury); first at room temperature and then at 56°C. They were stored in a desiccator over calcium chloride at atmospheric pressure while awaiting characterization. The reactions in which the copolymer precipitated were termed "heterogeneous". The copolymers that remained in solution were precipi- tated by a seven-fold volume of petroleum ether in a large 18 beaker. The polymer was filtered and treated as is described above. The reactions in which the copolymer remained in solution were termed "homogeneous". III. Preparation of Samples For Testing Copolymers were prepared in dimethoxymethane, benzene, tetrahydrofuran, bromochloromethane, and isopropylbenzene. The yields obtained and the copolymer designations are shown in Table 1. All copolymerizations were carried out in 500 milliliters of solvent. A. Copolymerization in Dimethoxymethane Two initiators were used, benzoyl peroxide at concen- trations of 0.21 and 0.A2 grams and 2,2'-azobis (2-methyl- propionitrile) also at concentrations of 0.21 and 0.42 grams. All four copolymerizations were conducted at AhiloC for 24 hours. All were "heterogeneous". The resultant cepolymers were labeled 1 through A respectively. The monomers and the initiators were very soluble in dimethoxymethane. All four reactions began as clear solu- tions but became turbid. Copolymerizations initiated with 2,2'-azobis (2-methylpropionitrile) were definitely turbid within a few minutes, while the benzoyl peroxide initiated reactions appeared hazy only after approximately two hours. After 1h hours, the viscosity of the copoly- merization mixtures had increased significantly and the stirrer speed had to be increased to insure proper mixing. 19 In general, reactions with 2,2'-azobis (2-methyl- propionitrile) were much faster and the yields were better. B. Copolymerization in Benzene Five copolymers were prepared utilizing three different procedures. The monomers and the initiators were very soluble in benzene. All five reactions were "heterogeneous". Procedure 1 Two copolymerizations were conducted at 50:100 for 24 hours. Benzoyl peroxide was the initiator in one at a concentration of 0.21 grams. The other contained 2,2'-azobis (2-methyl- propionitrile) also at a concentration of 0.21 grams. The resultant polymers were labeled 5 and 6 reapectively. Both copolymerizations were initially clear. The one initiated with 2,2'-azobis (2-methy1- prOpionitrile) was turbid within 15 minutes, while the one initiated with benzoyl peroxide was turbid within 25 minutes. Formation of copolymer in both was rapid during the course of the reaction, and the stirrer speed had to be increased after two (the 2,2'-azobis 2-methyl- propionitrile initiated reaction) and eight, (the benzoyl peroxide initiated reaction) hours to maintain pr0per mixing. 20 Procedure 2 Two cepolymerizations were carried out at 80i1°C for 1.5 hours. One was initiated with 0.84 grams of benzoyl peroxide and the other was initiated with 0.84 grams of 2,2'-azobis (2-methylpropionitrile). These two copolymers were labeled 7 and 8 respectively Although initially clear, both reactions became turbid within a matter of seconds. Within 30 minutes the copolymers had precipitated to the point where stirring was extremely difficult. The stirrer speeds were increased to the maximum extent, and 50 milliliters of solvent were added to each reaction before a satisfactory stirring rate could be achieved. One copolymerization was conducted at 800:100 for four hours with no added initiator. Turbidity was observed after 58 minutes. The formation of copolymer was very slow. This copolymer was sample 9. Copolymerization of styrene and maleic anhydride in benzene with initiator was rapid and nearly quantitative. The reaction without initiator was very slow but did proceed at 80°C. C. (Zopolymerization in Tetrahydrofuran Six cepolymerizations were carried out using three procfEdures, two in tetrahydrofuran, two in mixtures of 21 'tetrahydrofuran and isopropyl benzene and two in mixtures of tetrahydrofuran and carbon tetrachloride. In each case the temperature was 65:100 and the reaction time was 5.5 hours. Procedure 1 Two 00polymerizations, both "homogeneous", were conducted in tetrahydrofuran. One was initiated with 0.84 grams of benzoyl peroxide and one was initiated with 0.84 grams of 2,2'-azobis (2-methylpropionitrile). These copolymers were labeled 10 and 11 respectively. Procedure 2 In this case, combinations of tetrahydrofuran and isopropyl benzene were used as solvents for cepolymerization. Two cepolymerizations were ' carried out; one in a 3:2 ratio by volume of tetrahydrofuran to iSOprOpyl benzene (300 mifliliters of tetrahydrofuran: 200 milliliters of isopropyl benzene), and one in a 4.5:0.5 ratio by volume of tetrahydrofuran to iSOprOpyl benzene. The initiator in both was 0.84 grams of benzoyl peroxide. The copolymerization with the 3:2 ratio of solvents was initially clear but became turbid after about one hour. As more polymer precipitated, it began to agglomerate into large particles and was difficult to stir. When the reaction was finished, the c0polymer was in a single solid mass, 22 which could not be removed from the flask. The tetrahydrofuran-iSOpropyl benzene mixture was removed, filtered, and saved. The solids collected were returned to the flask and, along with the single solid mass, were dissolved in tetrahydrofuran. The copolymer was precipitated with petroleum ether and was labeled 12. The clear tetrahydrofuran-isopropyl benzene solution which was filtered and saved was diluted with petroleum ether and a second copolymer sample was obtained. This was Sample 13. Thus, two polymers were Obtained when a 3:2 ratio of tetrahydrofuran to isopropyl benzene was used as the solvent; the polymer that precip— itated and the polymer that remained in solution. A The copolymerization utilizing the 4.5:0.5 ratio by volume of tetrahydrofuran to isopropyl benzene remained clear throughout the 5.5 hour reaction. The cepolymer was precipitated with petroleum ether and was labeled 14. Procedure 3 In this case, combinations of tetrahydro- furan and carbon tetrachloride were used as copolymerization solvents. Two copolymerizations were conducted; one with a 3:2 ratio by volume of tetrahydrofuran to carbon tetrachloride and one with a 4:1 ratio by volume of tetrahydrofuran 23 to carbon tetrachloride. The added initiator in both reactions was 0.84 grams of benzoyl peroxide. The copolymerization with the 3:2 ratio of solvents was initially clear but became turbid within the first 1.5 hours. The copolymer agglomerated into large particles and stirring was difficult. When finished, the precipitated polymer had formed a single solid mass. The reaction solvent was filtered and saved. The solids on the filter paper were returned to the flask and along with the single solid mass were dissolved in tetrahydrofuran. This copolymer was precipitated with petroleum ether and was labeled 15. The clear tetrahydrofuran-carbon tetrachloride that passed through the filter was diluted with petroleum ether and a second cepolymer was isolated. This was sample 16. The copolymerization with a 4:1 volume ratio of tetrahydrofuran to carbon tetrachloride as the solvent remained clear throughout the polymerization. The copolymer was precipitated with petroleum ether and was labeled 1?. D, Copolymerization in Bromochloromethane One copolymerization was carried out in bromochloro- meChane at 691100 for 40 minutes with 0.84 grams of benzoyl (I. w u PupiialAflUPF P Vb r. ..J......»... ».' 3!th 2A peroxide as the initiator. The comonomers and the initiator were very soluble. The reaction was initially clear but became turbid within 20 minutes. The copolymer from this "heterogeneous" copolymerization was labeled 18. E. Copolymerization in Isopropyl Benzene Six copolymers were prepared using four procedures. The comonomers and initiator were very soluble in iSOpropyl benzene. All six copolymerizations were "heterogeneous". Procedure 1 Two copolymers were prepared at 800:100 for 1.5 hours. One was initiated with 0.84 grams of benzoyl peroxide and the other contained no added initiator. Both reactions were initially clear but became turbid, in-both instances, within 11 minutes. The resultant copolymers were labeled 19 and 20 respec- tively. Procedure 2 One copolymerization was carried out at 105°il°0 for 1.5 hours with no added initiator. Turbidity was observed within 15 minutes. The resultant copolymer was labeled 21. Procedure 3 A One cepolymerization was carried out at 135° 11°C for 1.5 hours with no added initiator. Turbidity was observed within one minute. The resultant copolymer was labeled 22. 25 Procedure 4 Two copolymerizations were conducted at 15201100 for 1.5 hours. One was initiated with 0.84 grams of benzoyl peroxide and the other polymerized without added initiator. Turbidity was instantaneous in both cases. With these two reactions, the copolymers adhered to the wall of the reaction vessel and were scraped off. The adherence to the wall of the reaction flask does not occur at 135°C or below. The resultant copolymers were labeled 23 and 24 respectively. Styrene and maleic anhydride copolymerize very readily in iSOprOpyl benzene. The ease of copolymerization is illustrated by the rapid reaction at 80°C without added initiator. F3 Samples From Other Sources A commercial sample, lOOOA from Sinclair (29), was labeled 25 and a sample previously prepared in tetra- hydrofuran in this laboratory, 4E (30), was labeled 26. ,0 2 \“vvxw----H-z-Hw1,.w.mswmmmm...MMu\\_/n,huu_ i .n_._ falls! ; 1 \\:..H A)... f -1.-..5ivl - nllll._ '3‘ "l’I‘I'OII 1' I all” I-‘|. “I mgr-m7 "2' 1‘? ..-~._~~.- , ‘ ‘1'. 1“- VI. lol‘i‘l050" -ii’ugi: L Schematic drawing of the apparatus used for the polymerization eXperiments. Figure 1. 27 Table 1. Samples Prepared for Testing Initiator g/500m1 Type Reaction Temp. % Sample Solvent of sol, TimeL Hours degrees,C Yield 1 DMM 0.21 32202 24.0 44 11.8 3 DMM 0.21 AZ 24.0 44 71.5 4 DMM 0.42 AZO 24.0 44 41.2 5 benzene 0.21 BZ 02 24.0 50 93.4 6 benzene 0.21 AZ 24.0 50 95.7 7 benzene 0.84 B2 02 1.5 80 91.0 8 benzene 0.84 AZ 1.5 80 89.4 9 benzene none 4.0 80 3.0 10 THF 0.84 BZ 02 5.5 65 47.4 11 THF 0.84 AZ 5.5 65 67.9 12 3/2 THF- 0.84 32202 5.5 65 54.8 and IPB and 13 7.6 14 4.5/0.5 0.84 32202 5.5 65 54.3 THF-IPB 15 3/2 THF- 0.84 132202 5.5 65 56.0 and CClh and 16 7.4 17 4/1 THF- 0.84 32202 5.5 65 51.0 CClh 18 BCM. 0.84 82202 0.7 69 27.5 19 IPB 0.84 82202 1.5 80 73.0 20 IPB none 1.5 80 39.5 21 IPB none 1.5 105 56.5 22 IPB none 1.5 135 88.5 23 IPB 0.84 32202 1.5 152 88.4 24 IPB none 1.5 152 88.4 25 & 26 were obtained from other sources (see page 25) B2202 - Benzoyl Peroxide AZO = 2,2' azobis—(2-methylpropionitrile) DMM - Dimethoxymethane THF - Tetrahydrofuran IPB - Isopropyl Benzene 0014 - Carbon Tetrachloride RCM I BrnmochloramcthAnm 28 IV. Evaluation of Samples The samples prepared for testing were evaluated using vapor pressure osmometry, gel permeation chromatography, and viscosity measurements. These techniques and the results obtained are reviewed individually. A. Vapor Pressure Osmometry Number average molecular weights were determined using a Mechrolab, High Temperature, Vapor Pressure Osmometer, Model 302. The Mechrolab brochure (31) and other references (32) describe the theory and method of Operation in detail. The vapor pressure osmometer Operates on the principal Of vapor pressure lowering. Solutions always have a lower vapor pressure than the pure solvent. In this unit a drop of copolymer solution in acetone and a drop of acetone were suspended, side by side, in a closed chamber saturated with acetone vapors. The two drops had different vapor pressures and a differential mass transfer occurred between the two drops and the acetone vapor phase. This resulted in lower evaporation from the copolymer solution drop than from the acetone drop, creating a temperature differential between the two drops. This temperature differential was preportional to the vapor pressure lowering and to the copolymer concentration. This is a colligative effect, dependent only on the number of dissolved molecules. Acetone was the solvent in all cases. The unit was <3a1ibrated with benzoic acid in accordance with the 29 Mcchrolab brochure (31) and a K was determined. Figure 2 is a calibration curve. The data used to construct the calibration curve are in Table 2. The K of calibration from this determination was 419. Recalibrations were made after several (8-10) determinations. The molecular weights were calculated using the equation: __- l = K ““3 M» (9.5). Where 0 is the concentration of the sample used and AR is the dekastat reading Obtained for a given-concen- tration, C. The value ($90 was obtained by extrapolation to zero concentration. The number average molecular weights were obtained "for samples with molecular weights within the range of the equipment (Mn < 20,000). 0f the samples prepared,- only those made in tetrahydrofuran and isopropyl benzene were in this category. A typical calculation is shown for sample 11, in Table 3. The correSponding plot is shown in Figure 3 along with all the other samples tested. The resultant Mn values for all the samples tested are shown in Table 4. 1!- 30 i r \0 see 50. db .80 a: ‘— GE) .memsosmo enammmmo noom> my pwom owoucon mcwuwawps m>pso cowpmgnwamo .N osswwm ocopmow cw nmaoz No. HO. -L F- own -- com 00: Dad 0 I owe Om: -- 0:: -- cm: 006 . one 31 Table 2. Determination of K with Benzoic Acid for the Vapor Pressure Osmometer Molar Concentration .04 .06 .08 A.R reading # l 16.77 25.14 33.55 A R reading # 2 16.78 25.14 33.56 Average A R ' 16.775 25 .14 33 .555 JAE 419.4 419.0 419.4 /V\ 32 Table 3. Vapor Pressure Osmometry Results for Sample 11 Concentration in grams per liter 20 .__&Q__ .__éQ_. 80 A R reading # 1 1.19 2.59 4.08 5.86 A R reading # 2 1.18 2.59 4.02 5.78 Average AR 1.185 2.59 4.05 5.82 .453 0.0592 0.0647 0.0675 0.0727 (6230 (from .figure 3) - 0.04 e. . 4,115,], .1. cg.) -1017. 32 1 33 ,/$-W ®-,-//Q -/l 030 '7 ’10 /// / .28 _/ I l 1. I 1 60 80 100 grams per liter in acetone 1 l I l 20 40 Figure 3. Vapor pressure osmometry plots. 34 Table 4. Mn Values From Vapor Pressure Osmometry Data Copolymer Designation Mn 10 13,500 11 10,470 12 17,010 13 5,000 14 _ 15,290 15 17,210 16 3,420 17 12,500 23 5,740 24 8,150 25 1,500 26 8,700 35 '8. Gel Permeation Chromatography A commercial instrument manufactured by Waters Associates was used in this study. The theory and operational procedure for this instrument have been described in the Waters instruction manual (25) and elsewhere (28). A schematic diagram of the unit used is shown in Figure 4. A four column circuit made up of four foot columns was used. The columns were arranged in series by pore size as follows: 106 A 104 .92 104 R 103 £3 The sample always passed through the larger pore size first. The columns were high quality with a rating of about 800 plates per foot. Tetrahydrofuran was used as the solvent for each determination. The sample was prepared by dissolving one weight per cent of the copolymer in tetrahydrofuran. The pumping rate was one milliliter per minute. The sample was injected into the sample valve, and at the proper time, the valve was opened for one minute, introducing one milliliter of the sample solution into the column. The time required for a sample to pass completely through the system was about two and one half hours. A typical chromatogram is shown in Figure 5. Solvent flow is plotted in increments of five milliliters versus 36 the change in refractive index. The flow rate is maintained constant by a pump. A differential refractometer con- tinuously compared the refractive index of the fraction to the reference solvent and gave a signal proportional to the amount of polymer in the solvent. The circuit was calibrated with polystyrene of known molecular weight, and the calibration was checked period- ically. Two runs, separated by several weeks, were made. The two calibration curves are shown in Figure 6. The polystyrene used for calibration was polymerized by an anionic mechanism (33) and was made available by The Dow Chemical Company. The calibration samples and their molecular weights are shown below: ngplg Mw Mn Mrms S-O 10,500 6,400 8,200 1.64 S-102 82,000 78,000 80,000 1.05 s-105 153,000 147,000 150,000 1.04 'Mrms = —V(Mn) (MN) The elution count correSponding to the peak of the cali- bration sample was assigned the corresponding Mrms value. Accordingly, Mn, Mn, and the corresponding—Mw/Mn values determined by gel permeation chromatography are based on calibration using polystyrene. This should give good estimated of Mw and Mn for the molecular weight range desired; however, the relationship between elution count and molecular weight for materials other than pure poly- styrene in tetrahydrofuran does not necessarily hold. 37 The number and weight average molecular weights were determined using a procedure that amounted to tabulating the data on the chromatogram, supplementing it with data from the standard curve, and then following the directions in the following derivation (25) (34). The height of an increment (i.e., the "height" above the base line),(u , is prOportional to the amount or mass,ln;, of material in the increment. h; a ‘m; h; a k ‘mL 0) The mass is equal to the number of molecules in the increment,7u,, times the molecular weight of the molecules,fW;(Assume that all molecules in an increment are of the same molecular weight). ‘m: e NM. (9.) Thus . be == m/M; (3) Number average molecular weight is defined by .4 _ 2: (h: M:)_ M h ' Z (m) (‘0 From (1) and (3) it is known that ‘n: = K “An; (3) By substituting (5) into (4): R __2_:(K kZ12: (’75) ‘g c :4; (hi) __ (6) V Z. (K hi/mz) Z (hi/Mi) Weight average molecular weight, Mw, is defined as *- s E (M: Mi)¢__ (7) W Z (w) 38 Substituting - _ 21mm“) M” .. 20121511) Next, substituting (5) -- _ 3: (mm:)__ M“ ' Z 02:) With the low molecular weight samples, the data were tabulated on each one-half elution count. With some of the higher molecular weight materials, data were tabulated on the full count. Typical calculations are shown in Table 5. These data were taken from the curve in Figure 5. The results from the gel permeation chromatography work are shown in Table 6. Data taken from the gel permeation chromatograms were used for plotting distribution curves (28) (35). The refractive indices (see Figure 5) were added to make a table of cumulative heights (see Table 5). The sum of the heights 2):; equalsEMJhisince the height of each interval equals the product of the average molecular weight of this interval (Mg) times the number of molecules (71") . The cumulative heights were normalized and plotted versus the molecular weight data Obtained from the calibration curve. Such a plot is shown in Figure 7. All of the chromatograms and some selected distribution plots can be found in appendices 1 and 2 respectively. 7/ _ L\/Ad A. 39 -_--— I-III-w- —-—I———--———~a-~-—————-1 .zanfiommm mzamnmopmsonco vfiswwa mncpma .e epomHm . l moeomjoo zoiodmudr ¥r mama; n n 11111111111111111 isull -ulnnuuunn . .\\1\uViHHu Amwu \ .jmo oeozd . . Ta mm>._<> momDOm :5... a a J. noEzoo 202.15 T. -1..- $5: 1. E dznd Lk fl 1. $358 (1058..-, l... H \Iml>.n.<>. (IL +. .JH w mnemeaoeommmmm a w 4538 _ N “MAN .m3<> mnazdm A sea. [.1 e are _ mode 850 . _ _ _, Zm>O embed: w .T. H . H32. “1.52% CMFp§o coprprumflw HmpwmpCH mOH x E S 1 1 _ _ all. .0— 1— db- .N wpzwwm _% 31191911 3 QB 45 C. Viscosity Measurement Viscosity measurements were made on dilute solutions in acetone at concentrations of 0.5, 1.0, 1.5, and 2.0 grams per 100 milliliters. A Cannon-Fenske viscometer with 0.50 millimeter capillary diameter was utilized at 25°C. Previously published procedures (36) were followed with regard to cleaning, filling, alignment, and measurement. The solution viscosities were measured by comparing the "efflux time t" required for a specified volume of polymer solution to flow through the capillary tube with the corresponding "efflux time to" for the solvent. The specific viscosity (95,.) was determined at several con- ‘95? = k/to'l Values for ”SP/c were plotted versus the concentration in centrations where: grams per 100 milliliters and extrapolated to infinite dilution to give the intrinsic viscosity,[0] . The results from the viscosity measurements on sample 12 are shown in Table 7. The plot of sample 12 is shown in Figure 8 along with the others. Evaluation of molecular weight for a polymer from viscosity measurements is usually made from the equation: mm This requires an evaluation of "K" and "a" for a Specific polymer solvent system using polymer samples of known molecular weight by an absolute method. 46 The log of the intrinsic viscosity is a linear function of the logarithm of the molecular weight: «an9{?i]“~9‘09’)< 'F'CLHQ07}/V\ Mn values for samples 10, ll, 12, 15, 17, and 24 had been previously determined by vapor pressure osmometry. The intrinsic viscosity was determined for these same samples and a plot of log M versus Log [’7] yields a straight line. The 510pe (0.79) is "a" and the intercept, when extrapolated to zero molecular weight, is K. The plot is shown in Figure 9. The extrapolation is not shown. The values of log M and log [9] were taken from a point on the straight line and K = 0.64 x 10'“ was calculated using equation 1 (see Figure 9). Extrapolation to the intercept gives the same value of K. The intrinsic viscosities and the.Mn values calculated from them are shown in Table 8. A typical calculation is also shown. Billmeyer (37) points out the possibility of estimating molecular weight from the specific viscosity at one con- centration. A plot of 1)st at 0.5g/100 milliliters versus molecular weight is shown in Figure 10. This appears to be a quick, easy method to get a relative molecular weight value on styrene-maleic anhydride copolymers for comparative purposes. This was used to estimate the Mh of three copolymers with the results shown in Table 9. 47 Table 7. Results From Viscosity Measurements on Sample 12 to ' 93.6 seconds y) SP 95 Pk Concentration Efflux Times Average t 2g/d1 1 . 53/01 lg/dl 0.53/01 126.1 126.2 116.8 108.4 100.5 0.348 0.248 0.158 0.074 0.174 0.165 00158 0.148 macaumaodmppxo mowmoom4> owmcfipucH .m opswwm mcopoom no mpopaaaaaas OOH mom madam N m.H ©.H :.H NM H 0.1H w1.0 QHO 4H0 NMO " w 1111 n -1. 1-1.. 1- .- 1:1-.1111- .1. .11 - -11111...1:1111.1.1. . -- .. . .- 1 :1- - . -- 3-1 - 1-1..-) - m4 “W111I1Mwwwmwauw) .0 .111. -111111118Mwo 44.43188 0 011-111.111.- . 116. 111 .1111))1.1.1Ma.\414.& .13 @1- 140!) 1. N e. 1.... 1.. .1- .2 u.& 48.411111111111191 1.11.1. 0111.11 Ed. 1.4434441W11111. a 01-111- -- 14 0 1|- )Illllnll!‘ 0 av: . :1. .4 . . 1. 01111111113 0.4..4.“-.-,.4W111111111111®11111111si1111 o 31)... 1:1- .. mam-549% . )\)1\))\\.... 411111111111 . 11.111 11.111111...- 8... 1. .011 421-244.4434.» WNW-11.11110111111- 9 61,--.. ,. . 4.4 302% ...m .. 4 1.. 0. 1.... 111.1111. -4“ \ .29 \\ __ \.\.\.\ a \ 1.1mm #9 Table 8. Intrinsic Viscosity Results and Calculated 17m Va lue s Copolymer Ln] Mn Cgigglated 10 0.1110 13,290 11 0.0955 10,620 12 0.1100 17,210 15 0°1398 17,210 17 0.1178 ~ 13,850 19 0.171 22,710 20 0'2525 36,410 21 0.1725 22,160 22 0-0975 10,650 24 0.0800 8,480 Typical Calculation, Sample 2b 0.08 - 0.6h x 10-h M-79 M'79 - 1257.86 .79 log M - log 1257.86 10% M '5 3.92855 M a 8,483 50 , ‘I|-TE- INN cam {A .3 .ma .3 .OH mmamsmm (Sm v~ woq 26%: T.”— moq .0 CE moq mm.¢ mm.4 gm.q Om.¢ 0H.: NH.¢ m@.4 #0.: own: ©@.m mm.m mm.m 4m.m .-...-,---+i- H a; .r mpsmwm 5-0:” $4.0 4.3 2 «£6 + 2 $4 ... r594 :6 «6‘ O : ON.HI A I 0H.HI . l S4- ,‘ 8.7 i 3.7 ‘. 84- .. 8.0- -1 No.0- ; 3.0.. ‘- $5- :. mpg meg . .m:ouoom CH mpopwawaawe 00H pan msmpm m.o mo mcowumpucmocoo um um mampw> .uxwmp .OH mpsmfim A H? 56pm vmpmHSonov .mOH x a: +3 m: 0: mm mm dm mm.om m.m.om.fim.m.mro_m ma 3 3 NH 3 m o : .aua:»:7-¢4-:x4qs;*, " . . »-:.¢-4|&1:z+‘vu¢1u.¢au;721x111;1a1;on.+1|c.:&a+91!1i;4lx;»!li NH. mafia cowpmaoawpuxm .L zpwmoomfl> owmcfippcw msp mmon B 5 mafia cowpmaoamppxo muwmoowfl> owwcwpucw on» conAu : NN. 52 Table 9. Fin Estimated From USP/c at c - 0.5g/100 milliliters Gogolgmer YEP/C Estimated Fin 5 ' 2.20 200,000 18 - 1.400 150,000 19 0.070 6,300 CORRELATION OF RESULTS AND DISCUSSION 53 5h I. General The choice of solvent in which the 00polymerization occurs is a critical factor in determining the molecular weight of alternating styrene-maleic anhydride copolymers. 0f the solvents used, only tetrahydrofuran and isopropyl benzene resulted in low molecular weight polymers within the desired range (Mn<20,000). The preparation of low molecular weight styrene-maleic anhydride samples in dimethoxymethane has been previously reported (37). However, all of the copolymers prepared in dimethoxymethane during this study had Mn values in excess of 37,000. In dimethoxymethane, the reactions initiated with 2,2'-azobis (2-methylpropionitrile) were much faster and the yields were better than when benzoyl peroxide was used as the initiator. The chromatograms in Appendix 1 and the MW/Mn ratios in Table 6 show a broad molecular weight distribution for all four copolymers (samples 1 through 4). The two samples initiated by 2,2'-azobis (2-methylpropionitrile) (samples 3 and h) have a much narrower distribution than the corresponding 00polymers initiated with benzoyl peroxide (samples 1 and 2). The capolymers prepared in benzene were also of high molecular weight. Sample 7 with an Mn value of 36,300 was the lowest molecular weight sample of the five prepared. Gel permeation chromatograms are shown for samples 7 and 8 in Appendix 1. The benzoyl peroxide initiated sample (sample 8) has the higher mOIecular weight (Mn - h2,000), but its 55 distribution is highly unsymmetrical as is evidenced by the shape of the chromatogram and by the Mw/Mn value of 5.1. The rate of reaction in benzene, (at comparable temp- eratures) was much faster than in dimethoxymethane. This is evidenced by higher yields, shorter reaction times, and reaction without added initiator. The one reaction in bromochloromethane resulted in an Mn greater than 150,000 (estimated from one Dsgk value, see Table 9). High molecular weights (Mn > 100,000) have been reported for copolymerization of styrene and maleic anhydride in methylene chloride (39). In the polymerization of styrene, halogenated hydro- carbons are reported to be active chain transfer agents (hO); much more active than simple hydrocarbons. This is not the case in styrene-maleic anhydride 00polymeriza- tions. The comparison of the molecular weight of copolymers prepared in bromochloromethane and isoprOpyl benzene is illustrative(samples 5 and 19, Table 9). This aspect will be discussed further under chain transfer. The use of tetrahydrofuran as a solvent resulted in low molecular weight (Mn< 15,000) copolymers. The copolymers initiated with benzoyl peroxide and 2,2'-azobis (2-methyl- propionitrile) had similar distributions (see the chroma- tograms and distribution curves for samples 10 and 11 in Appendices 1 and 2) but significantly different molecular weights. A 30% difference was observed in the.fin values of samples 10 and 11, with the benzoyl peroxide initiated polymerization (sample 10) giving the higher molecular weight. 56 Isopropyl benzene and carbon tetrachloride were evaluated as portions of the solvent in combination with tetrahydrofuran. The technique by which two polymers, one soluble and the other insoluble, are produced in mixed solvents has been termed "insitu fractionation". Table 10 reports the results from polymerization in both of the above mentioned mixtures to include data on homogeneous copolymerizations in which the ratio of the mixture has been varied enough to keep the 00polymers in solution. The molecular weight difference between samples 13 and 16 indicates that styrene-maleic anhydride copolymers are more soluble in isopropyl benzene than they are in carbon tetra- chloride. Obviously, the combination of various non-solvents with solvents can result in the isolation of polymers with varying molecular weights. This same technique is also useful for preparing polymers with narrow distribution. The narrow distributions of samples 12 and 15 are shown on the gel permeation chromatograms and distribution plots in the appendices. The chromatogram of sample 14 is essentially a composite of samples 12 and 13, assuming equal concentrations. The lack of symmetry in sample 13 is due to the loss of the higher molecular weight portion by fractionation. The polymerization of styrene and maleic anhydride proceeds rapidly in isoprOpyl benzene with and without added initiator. Low molecular weight polymers with En values varying from 6,000 to 36,000 are readily obtained. The results with no added initiator are shown graphically 57 in Figurell. It may be noted that the molecular weight decreases substantially as the temperature increases. This is an excellent method for preparing polymers with a predetermined molecular weight. The results with initiator are similar near reflux temperatures but differ at lower temperatures. The following is illustrative: Mn 0 6 Sample ' 152°C 80 Q_ 23 Initiated with 0.8h grams - benzoyl peroxide 6500 22,750 24 No added initiator 7500 36,000 II. Gel Permeation Chromatography In gel permeation chromatography a dilute solution of polymer is injected into a solvent stream flowing through a column packed with porous beads of inert, cross-linked, polymer gel of controlled porosities. Under a constant total flow rate, the permeation rate of the molecular species of different sizes in the polymer sample differ. The smaller molecules have greater accessible volumes in the gel packed column and permeate more slowly than large molecules. The larger species permeate the gel least and are eluted first. Styrene-maleic anhydride copolymers were not expected to permeate the gel at the same rate as polystyrene; however, this did occur. TablelJ.compares the Mn values from vapor pressure osmometry data with those calculated from gel permea- tion chromatography data. These same data are shown graph- ically in Figure 12. The correlation between the two methods 58 is obvious. In tetrahydrofuran at En values less than 20,000 styrene-maleic anhydride copolymers appear to have the same "effective" size as polystyrene on the column utilized. As was mentioned earlier, such a correlation was not expected. Just recently, Meyerhoff (Al) evaluated poly- styrene, cellulose nitrate, and nolymethyl methacrylate and in a paper on molecular parameters and gel permeation noted that neither molecular weight, intrinsic viscosity, radius of gyration, or diffusion coefficient permits any general correlation with the evolution volume independent of the chemical nature of the polymer. It is reasonable that different polymer structures will pass through the columns at different rates. The fact that styrene-maleic anhydride copolymers, with fin values less than 20,000, acted like polystyrene samples is at present uneXplained. For the column circuit used, the optimum resolution of polymers differing in molecular size occurred between elution counts 21 and 28. Only reasonable resolution was possible between counts 28 and 36. It was fortunate that the low molecular weight samples tested eluted largely in the area of reasonable resolution. The COpolymers with Mn values less that 20,000 have narrower distributions and lower Mw/Mn values that the higher molecular weight 00polymers. This trend is apparent from the data in Table 6. It may also be reflecting the loss of resolution due to the transition into the reasonable 59 resolution area; however, even though this is a factor it would not be expected to account for the entire trend. The samples with Mn below 20,000 have similar Ew/fin values (see Table 6) , samples 14 and 16 being exceptions. Sample 1L (MW/Mn = 2.10) resulted when a u.0/o.5 ratio of tetrahydrofuran to isopropyl benzene was utilized as the solvent. The in value was similar to that obtained for a c0polymer prepared in pure tetrahydrofuran (sample 10); however, Kw was larger by 5000. Thus, isopropyl benzene used in this fashion appears to broaden the distribution. This is readily observed from the chromatograms in Appendix 1 (samples 12 and 14). III. Viscosity Measurements Viscosity measurements cannot be used as an "absolute" method of molecular weight determination. The most general equation eXpressing the dependence of the intrinsic vis- cosity on molecular weight (A2) is: in = K M“ "K" and "a" are usually determined from data based on solution viscosity of narrow polymer fractions of known molecular weight. In this case, the copolymers used for determination of "K" and "a" were not fractionated. The molecular weights of the samples were determined by vapor pressure osmometry. Gel permeation chromatography data indicated that these copolymers had Mw/fin values varying from 1.75 to 2.01. The use of [913KMQ' and the experimentally determined values of "K" and "a" for styrene-maleic anhydride copolymers of similar polydispersity should result in 6O reasonable Mn values. Figure 13 is a plot of intrinsic viscosities versus Mn values calculated, using "K" = 0.6L x 10“ and "a" = .79. The values calculated from intrinsic viscosity data correlate well with the Mn values from vapor pressure osmometry in the range investigated (En c2 .HH madman mOH X CE mm sm 0N 0H NH w 4 All -.:: a-a,sz.; Tsaazs. -1. w l a l. a w a 1 e 0 a 0 : om ///// //// II. OOH //AV/ oo ///// i .QEmB [/4 /r.!. : omH 4 0.: r 00H 69 Table 11. Compagison of fin by Vapor Pressure Osmometry with Mn by Gel Permeation Chromatography Conolymer fin(VP0) fih (GPC) 10 13,500 13,500 11 10,500 9,100 12 17,010 16,560 13 5,000 5,720 11 15,290 13,750 16 3,u20 ‘ a,100 23 5,7uo u,u00 25 , 1,500 3,080 26 8,700 8,800 70 moa x.Aoeov :5 on mm 6H ea _ n a q NH OH n d .asdmmmoumaomco scammeEpoa Haw up cE m5mtm> >npmeosmo mpswmmna Loam> up CE : .ma eeeeaa m .7i:::+a!z:4 ..~H {OI X (OdA) ”fl 71 mm as I‘ O ’1”... Till!“ 0: 1? om ' 4 a: mamme> zpwmoomfi> owmcflLSCH .mH mnsmwm moH x :E mm mm 4N ON OH NH w A- )- P .1 a P ‘ 4) : NH. : 0H. 5 ..ON. .rmm. .rem. ..©N. 72 Table 12. A Comparison of Mn Values Obtained From Intrinsic Viscosity Data with Mn Values for Vapor Pressure Osmometry Data '_ -- Sample Mn (VPO) Mn (Vis) 10 13,500 13,290 11 10,470 10,620 12 17,010 17,240 15 17,210 17,210 17 12,500 13,850 24 8,150 8,480 73 9-1.1.6... 1,... in- Wuhan. u... .zgampmouwsoaso cowpmmemma Hem mfi> newsmaoa cw mamflnmpms oflL69>Hooncoc mo cowpoepmo .JH onswfim Ameeeaawwaae mmv ReneeaHaHMas away 0 30.7% %/ litanfirhni 1.; J». .. a..- \WWJJM//1/l/// .. m Heaeeenz mapmuomnpxm _, s I o T w .1, .. - S .1, u N H i7 xaput antaoeajel SUMMARY 71+ .1 HI ' 75. Styrene-maleic anhydride c0polymers were prepared with number average molecular weights varying from 3,000 to 36,000. The molecular weights within this range were varied as desired by varying catalyst concentration, temperature, and solvent. Number average and weight average molecular weights were calculated from data obtained from a polystyrene calibrated gel permeation chromatography unit with a general purpose column. Number average molecular weights were determined via vapor pressure osmometry. The number average molecular weights determined by gel permeation chromatography correlated well with those determined by vapor pressure osmometry. Molecular weight distributions were determined from the gel permeation chromatography data. Low molecular weight styrene-maleic anhydride copolymers are readily prepared in isopropyl benzene without added initiator. Mn can be varied from 8,000 to 36,000 by varying the polymerization temperature. The polymerization of styrene-maleic anhydride in tetrahydrofuran results in low molecular weight polymers (Mn = 10 to 13,000). ‘Mn can be varied by the choice of initiator. llm 76 Carbon tetrachloride, a very effective chain transfer agent in homopolymerization of styrene, did not act as chain transfer agents in the preparation of styrene- maleic anhydride copolymers. Viscosity measurements were made on a series of styrene- maleic copolymers (MW/Mn values ranging from 1.75 to 2.01). The constants in the equation Enl= KM“ ' were evaluated using samples with Nb values determined by vapor pressure osmometry and were K: 0.441(10“f ICC. :- 0:77. LITERATURE CITED 77 l) 3) 4) 5) 6) 7) 8) 9) 10) ll) 12) 13) 14) 15) 16) 17) 78 Meyer, E. 0., "Studies on Poly—(Maleic Anhydride Co Styrene)", M. S. Thesis, Michigan State University, (1965); Garrett, E. 8., "Studies on the Heteropolymer: Maleic Anhydride-Styrene", M. S. Thesis, Michigan State College, (1948). Austrian Patent, 70,348 (1914). Staudinger, H., and J. Schneiders, Ann., 5&1, 151 (1939). Dostal, H., Monatsh. pg, 424 (1936). Alfrey, T. Jr.,.and 0. Goldfinger, J. Chem. Phys. lg, 205 (1944). Mayo, F. R., and F. M. Lewis, J. Am. Chem. Soc. 66, 159A (1944). Wall, F. T., J. Am. Chem. Soc. éé, 2050 (l9hh). D'Alelio, G. F., "Fundamental Principles of Polymerization" John Wiley (1952), pp. 44-46. Wagner-Jauregg, T., Ber., 63, 3213 (1930). Moore, E. R., gtflgl., Official Digest, 37, Number 483, 410 (1965). Alfrey, T. Jr., and E. Lanvin, J. Am. Chem. Soc., 62, 2044 (1945). Ham, G. 8., Am. Chem. Soc., Div. Polymer Chem., Preprints one, Number 2, 219-31 (1960). 5.41.:- Bartlett, P., and K. Nozaki, J. Am. Chem, Soc. 68, 1495 (1946). Mayo, F. R., 33 31., J. Am. Chem. Soc., 19, 1529 (191+8) o .. ' "' Banford, C. H., and w. G. Barb, Disc. Faraday 800., Ag, 208 (1953). Billmeyer, F. W. Jr., "Textbook of Polymer Science", John Wiley (1962), p. 313. h 1.1.; L‘ mks“? '- Alfrey, T. Jr., et al., "Copolymerization", High Polymer Series, Vol. VIII? Taterscience (1952), p. 45. 18) 19) 20) 21) 22) 23) 24) 25) 26) 27) 28) 29) 30) 31) 32) 33) 34) 79 Walling, C., gpgal., J. Am. Chem. Soc., 29, 1544 (1948). Alfrey, T. Jr., and J. Price, J. Polym. Sci., g, 101 (1947). Young, L. J., J. Polym. Sci., g, 411-455 (1961). Walling C. £2 31. J. Am. Chem. Soc. 19 1537 (1948). ' ’ ’ ’ ' French Patent 1,322, 807, April 5, 1963; U. 8. Patent 2,971,939; U. 3. Patent 3, 085, 994; U. 3. Patent 3,046, 246. Neth. Appl. 6,506,790, Dec. 1, 1965. Moore J. C., J. Polym. Sci., Part A, 3, 835-843, (1964). Instruction Manual for the Waters Gel Permeation Chromatography Unit, Waters Assoc. Moore, J. C., and J. G. Hendrickson, J. Polym Sci. Part C, Number 8, 233- 241 (1965); Hendrickson, J. C., gndégS C. Moore, J. Polym. Sci., Part A- 1, 4, 167- 188 19 . Berger, H. L. and A. R. Schultz, J. Polym. Sci., Part A, 2, 364’ 3-3648 (1965); Smith W. B. and A. Kollmansberger, J. Phys. Chem., _9, Number 12, 4157- 4161 (1955) Maley L. E. , J. Polym. Sci., Part C, Number 8, 253- 268 (1965); Edwards, 0. D., J. App. Polym. ’Sei., 2. 3845- 3850 (1965). Product Data Bulletin, Number 1, Sinclair Petrochemicals, Inc., Jan. 1, 1964. Guile, R. L., Unpublished Results. Instruction Manual for the Mechrolab Vapor Pressure Osmometer, Model 302, Mechrolab, Inc. Bonnar, R. V. et a1. "Number Avera e Molecular Weights", Interscience, N. Y. (1958?. McCormick, H. W., J. Polym. Sci., 36, 341 (1959). Fleming, J., Private communication. Harmon, D. J., J. Polym. Sci., C, Number 8, pp. 243- 251 (1965). 43) 44) 80 Rafikou, S. R., et al., "Determination of Molecular Vieights and PolydiSpersity of High Polymers", Davey and Co. (1964) pp. 302- 327. Billmeyer, P. w. Jr., 22. cit. pp. 79-85. Meyer, E. G., 9p. cit. U. S. Patent 2,872,436. Billmeyer, F. W., o . cit. p. 279. Meyerhoff, G., Makromol. Chem., 82, pp. 282-284 (1965). Jen-Yoan, C., "Molecular Weights of High Polymers", S. Monson (1962) pp. 29- 33. Flory, P. J., "Principles of Polymer Chemistry", Cornell Press (1953) pp. 143- 145. Van Norman, G. B. and R. L. Guile, Unpublished Data. d APPENDICES 81 APPENDIX 1 Gel Permeation Chromatograms 82 A REFRACT IVE INDEX 83 ’_.I N +- l--' 00 \o o H l k -w---... 1--.”--- ..l .1... -.._,.J. 1.....- i i '31 22 32 34 33 Elution Count 21 20 19 Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 1 (calibration curve #1) AREFRACTIVE INDEX H O ._A.,_ -._..._-. ;.____.._...A .. . - . 8h 28 29 i 33 BL 31 20 Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM.OF SAMPLE 2 (calibration curve #1) 19 18 17 A REFRACTIVE INDEX 85 143' 13* 121 ll , 2 101 25 4‘ J 7 3‘ 22 l i 10 2‘9 32 31 Elution Count Solvent FIOW3 Increments of Five Milliliters ( 26 GEL PERMEATION CHROMATOGRAM OF SAMPLE 3 (calibration curve #1) i 19 8 A REFRACT IVE INDEX 15¢ 14 -- 13 12 113. 10- 23 22 w..-€:.‘» no“- i E i f 312 3‘1 Elution Count 20 19 1 Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 4 (calibration curve #1) A REFRACTIVE INDEX 15 14 5‘ ‘u—‘ .r' ‘DW‘W‘ n-C'JI 13 124' 11 -_.. “J 87 24 22 Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM 0F SAMPLE 7 (calibration curve #1) A REFRACTIVE INDEX 88 12-- ll-- 10- 26 25 9» 28 29 Elution Count 19 18 Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 8 (calibration curve #1) A REFRACTIVE INDEX 89 26 l 25 24 35 34 Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 10 (calibration curve #1) A REFRACT IVE INDEX 90 14* 13.. 12+ 114 10' 32 i 27 4 35 . 26 25 Elution Count Solvent Flow, Increments of Five Milliliters GEL.PERMEATION CHROMATOGRAM OF SAMPLE 11 (calibration curve #1) A REFRACT IVE INDEX h* h‘ H .\ a...“ . g I--. I--. I“ l—‘l—l Nb.) Lung-m... ,‘._ - —In)‘—¢- ‘- 4...... 1a.... 14. 91 27 31 32 26 l I 33 i 34 36 35 25 2 Elution Count Solvent Flow, Increments of Five Milliliters l 4 23 GEL PERMEATION CHROMATOGRAM 0F SAMPLE 12 (calibration curve #2) A REFRACTIVE INDEX 12: 11~ 10 92 och!” ,. l | 27 26 Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 13 (calibration curve #2) 11.. ' “P "' -. ' L1:W3flfinfi, . Lu A REFRACT IVE INDEX 15.; 13{ 12“- 11.? 10‘; OOKD J.........—.»- I _ 93 29 H ~-”- 3 32 g vk 5 l . ‘; Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 14 (calibration curve #2) A REFRACT IVE INDEX 9h Fr" '- I' . .d"'-"‘ I M M.» ‘ t. . r? I - 29 I l 28 27 Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE l6 (calibration curve #2) A REFRACT IVE INDEX 14 13 12: 114 10' 95 ” ._-\ .I,‘...-«~...o.—.. -.—.— ~‘—.-.-—--n..vs-.. X. .. 3. 35 Th“.--.‘, ._.,v.._._‘ I 36 27 26 Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 23 (calibration curve #2) if?" {SI-2+: 1231953723: _-. '— 4 I .. ! A REFRACTIVE INDEX 16~ 15 14 UI (JO-D - h-~<-'_'.‘l_'l" *‘~lI--‘II.b-.¢“*I-.~'~‘ ' U N H ‘ I) J 32 35 30 ‘2‘... .-_.‘ 1. w 36 28 27 26 ' Elution Count Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 25 (calibration curve #2) . . 3.1.913 A REFRACT IVE INDEX 134 12 4. 11.. 10“ 97 29 3O 31 28 32 33 37 3% 35 3‘4 2% H Elution Count 25 24 23 Solvent Flow, Increments of Five Milliliters GEL PERMEATION CHROMATOGRAM OF SAMPLE 26 (calibration curve #1) APPENDIX 2 Integral Distribution Curves l 98 99 / n :.;,