Prog. Polym. ScL, Vol. 8, pp. 373-468, 1982 0079-6700/82/040373-96548.0/0 Printed in Great Britain. All rights reserved Copyright © Pergamon Press Ltd. PREPARATION, CHARACTERIZATION, SOLUTION PROPERTIES AND RHEOLOGICAL BEHAVIOUR OF POLYACRYLAMIDE W.-M. KULICKE, R. KNIEWSKE and J. KLEIN Institut fdr Chemische Technologie, Technische Universitiit, 3300 Braunschweig, F.R.G. Dedicated to Prof. Dr. Dr. h.c.F.X. Patat on the occasion of his 75th birthday ABSTRACT In this report a survey is given on structure and properties of polyacrylamide homopolymers (PAAm) in solution. However the review is restricted to all those papers, where a molecular characterization of the polymers has been achieved as a basis to correlate this fundamental information with applicational properties. Different polymerization methods are summarised in brief, the preparation and solution structure of long chain branched polyacrylamides as well as chemical modification reactions of linear PAAm are also mentioned. A number of experimental characterization methods (GPC, ultracentrffngation, intrinsic viscosity, and light scattering measurements) are described with special emphasis on the difficulties of the different procedures including some proposals for properly designed experimental techniques. The state of solution is discussed in view of experi- mental data obtained with different solvents. Moreover viscosity constant ~ is calculated for aqueous solution and the unperturbed dimensions are estimated. All available data on cross correlations (e.g. [~] --M, S O --M, (h2) t/2 --M) are collected with the intention to give a survey of established relations and, comparing the given relationships, to suggest the reliable ones of them. The phenomenon of long-term viscosity decrease of aqueous PAAm solutions has been investigated and discussed with regard to its molecular origin. The viscoelastic properties are discussed in dependence on molecular weight, concentration, solvent quality, and shear rate (<10 6. s -t ). Based on these data a simple equation was devel- oped for the ~o-c-M relationship, which can be applied to other polymer systems as well. It is further described that the elastic nature (first normal stress difference) may overwhelm the viscous nature (shear stress) at relatively low shear rates. This high elasticity can cause deviation from laminar flow conditions. Moreover, it can be demonstrated - based on instationary measurements as well as the comparison of steady shear flow with dynamic rheology - that energetic interactions (H-bonds) strongly influence the theological behaviour. CONTENTS Abbreviations 374 1. Introduction 375 2. Polymerization 376 2.1. General 376 2.2. Polymerization Methods 376 2.3. Long Chain Branching 379 2.4. Polymer Modification 380 373
`
`SNF Exhibit 1013, Page 1 of 96
`
`

`
`374 W.-M. KULICKE, R. KNIEWSKE and J. KLEIN 3. Characterization and Properties in Dilute Solution 3.1. Experimental Methods and Results 3.1.1. General 3.1.2 Gel Permeation Chromatography 3.1.3. Diffusion and Sedimentation 3.1.4. Intrinsic Viscosity 3.1.5. Light Scattering 3.2. State of Solution 3.3. Cross Correlations 3.3.1 [~]-M Relations of PAAm in Water 3.3.2. [~]-M Relations of PAAm in other Solvents 3.3.3 Other Cross Correlations 3.4. Instability Effects 3A.1. Time Dependence of Solution Viscosity 3A.2. Degradation 4. Solution Rheology 4.1. Steady Shear Flow 4.1.1. Viscosity Functions 4.1.1.1. Dependence on molecular weight, concentration Shear Rate 4.1.1.2. q0-M--c-Relationship 4.1.2. Elasticity Functions 4.1.2.1 Inertia 4.1.2.2. Comparison of Shear Stress and Normal Stress Data 4.1.2.3. Deviations from Laminar Shear Flow 4.1.2.3.1. Detection for onset conditions 4,2 Instationary measurements 4.3 Relation between Steady Shear Flow and Dynamic Rheology 5. Concluding Remarks References ABBREVIATIONS AAm A2 c. G It K /Co M Mn t, tSD Mw Mz Me N NL PAAm acrylamide second virial coefficient (bar. cm 6 • g-2 ) characteristic ratio diffusion coefficient (cm 2 • s -I ) storage modulus (Pa) loss modulus (Pa) optical constant for unpolarized incident beam 4~r =. n~.f. (dn/dc)= 1 + cos=O tmol .cm= / K= g, / unperturbed dimension constant (ern 3. g-i ) molecular weight viscosity average molecular weight number average molecular weight sedimentation and diffusion average molecular weight weight average molecular weight z-average molecular weight critical molecular weight degree of polymerization Loschmidt number (mol-') polyacrylamide PAAm/AAcNa poly(acrylamide--co-sodium acrylate) (R ~)w weight-average mean square radius of gyration and 382 382 382 382 385 388 390 397 404 404 409 410 412 412 424 424 426 426 426 434 437 439 442 444 447 451 457 461 462
`
`SNF Exhibit 1013, Page 2 of 96
`
`

`
`SURVEY OF POLYACRYLAMIDE 375 (R=)z RO sR T Ve V_F V~ a C C*, Cerit dequ[~] dn/dc d f (h 2) (h2)o (h2)of kH n $o t q 7 Te.adt ~0 virel I~*1 ~o p P Pp Ps o"!1---o22 (O 11 "--0'22 )i 0'12 0 CO z-average mean square radius of gyration absolute scattering intensity (cm -1 ) recoverable strain temperature (°C) elution volume (cm s) (GPC) floccuiation speed in Kaolin suspension (cm • rain -1 ) partial specific volume of the dissolved polymer (cm 2. g-i ) Mark-Houwink exponent concentration (g. cm -2), in sec. 4 % = g/100 cra 2 critical concentration (g" cm -~) equivalent sphere diameter ~um) refractive index increment (cm 2 • g-i ) diameter (cm) Cabannes factor mean square end-to-end distance (rim) (indices: see <R2)) unperturbed mean square end-to-end distance (nm) mean square end-to-end distance in the freely rotating state Huggins coefficient refractive index of solvent (n I ) and polymer solution (n2) sedimentation coefficient (Svedberg) time (s) polymolecuiarity correction factor shear rate (s -1 ) critical shear rate (s -1 ) zero shear viscosity (Pa. s)(r/* f(~)) ViScosity (Pa- s)07 = 01~1'~) upper limiting viscosity (2. Newtonian region) (Pa • s) relative viscosity specific viscosity intrinsic viscosity (cm 2/g) complex dynamic viscosity (Pa. s) viscosity constant for non-geussian coils (mo1-1 ) viscosity constant for gaussian coils (O-state) (tool -! ) wave-length in vacuo (nm) exponent in the (h2)112_M relationship density density of polymer solution (g • cm -2 ) density of solvent (g • cm -2 ) first normal stress difference (Pa) apparent first normal stress difference due to inertia (Pa) shear stress (Pa) conformation factor frequency 0 -1 ) cone angle (tad) ImPa.s = IcP 1 Pa = 10 dyn/cm 2 1. INTRODUCTION Polyacrylamide, as obtained by free radical polymerization of acrylamide, is one of the most widely used and technically important water soluble polymers. A large number of papers therefore has been published, covering scientific and technical or applicational topics. In preparing a progress report like this it is obviously impossible
`
`SNF Exhibit 1013, Page 3 of 96
`
`

`
`376 W.-M. KULICKE, R. KNIEWSKE and J. KLEIN to cover practically all the literature even for a limited time period of the last ten years. It is therefore important to confer to the reader those ideas which have been used by us as guidelines to narrow the topic in one or the other way. The first selection criterion is related to the definition of "polyacrylamide" itself. Based on the fact that practically all polymers of technical importance are produced as copolymers of acrylamide with anionic or cationic comonomers, the term "polyacrylamide" has developed to become really a family name including these "acrylamide derivatives" too. We felt, however, that it is the more important to emphasize a precise nomenclature and to contribute to a clear correlation of a polymer's name, its chemical structure and the respective properties by covering only those papers where polyacrylamide as the true homopolymer has been studied. The second criterion for paper selection has been our intention to discuss the properties of polyacrylamide in solution as a function of basic molecular para- meters (like molecular weight or molecular weight distribtion) in a quantitative fashion. So a large number of papers, where only poor or just no characterization data have been included, and which therefore only can be considered as qualitative contributions to the field, had to be omitted. Our own laboratory has been actively involved in the research on polyacrylamide solutions for several years. While the first contributions have been qualitative in nature as well I we soon felt the necessity to include the preparation and character- ization of the polymer samples in our studies, 2 thus adopting the above mentioned two criteria for our work. We therefore feel justified to conceive this progress report as a challenge to summarize all our earlier contributions - which have been appeared in different publications - in a comprehensive fashion and to critically review our results in the light of experimental and theoretical findings established otherwise in the literature. Though starting from a more personal point of view we hopefully have been able to present a complete picture on to-day's knowledge on solution structure and prop- erties of polyacrylamide. 2. POLYMERIZATION 2.1. General The monomer acrylamide is a colourless crystalline solid which is obtainable by partial saponification of acrylonitrile. It is easy soluble in water and alcohols and can be recrystallized from benzene. By exclusion of day light it is stable over long periods. The important physical properties of acrylamide are summarized in Table 1. 2.2 Polymerization methods Acrylamide readily undergoes vinyl polymerization. A short survey of different polymerization methods was recently given in ref. 2 (1978) in view to obtain
`
`SNF Exhibit 1013, Page 4 of 96
`
`

`
`SURVEY OF POLYACRYLAMIDE 377 TABLE 1. Physical Properties of Aeryiamide s Molecular weight 71,08 Melting Point 84~5 ± 0,3°C Boiling Point 87°C (2 Torr) 103°C (5 Tort) 116,5°C (10 Torr) 136°C (25 Torr) App. Density 1,122 g • cm "3 (30°C) Solubility Acetone 63,1 (grams]100g at 30°C) Benzene 0.346 Methanol 155 Water 215,5 Formula CH3=CH I C=0 I NH2 TABLE 2. Polymerization methods of acryiamide 1. Per- and Azo-compound initiated polymerizations Initiators: H2 022,4,7 , Persulfate4,6,7,s,9, Di_t_butyloxide~O, 4,4-Azo-bis--4---cyano- valenan acid I1 , AIBN 4 , Ce-IV l= 2. Redox initiated polymerizations Initiators: Permanganate/thioureat3 , Permanganate/Tataxic acidt4 , Permanganate/Thio - glyeolic acid I s, Permanganate/citric acid 16 , Permanganate/Oxafic acid a ~, Permanganate/Ascorbin acid (emulsion) 18 , Glycerol/Ce(IV) 19 , Potassium persulfate/2-mercaptoethanoP 8 , Potassium persulfate/2-mercaptoethylamine hydroehloride 21 , Potassium persulfate/Thioglycolic acid 23 , Cer(IV)/thiourea ~3 , KBr0s/Thioglycolie acid u , fNI-I 4 )2 $20e/Thiolactie acid 2s , Vanadium(V)/ Cyclohexanone 26 , Chlorate/Sulfite 3~ , Pinacol/Ce(IV) =8 H303/Fe(II) 4 . 3. Photopolymerization UV without sensibilisators or visible light with sensibilisators 29- s3. 4. Radiation induced polymerization X-ray or -r-ray in aqueous solution or in substance 4 ,e, ~,34,3s. 5. Electroinitiated polymerization ~--~9 . 6. Ultrasonic polymerization 4,4°,4t . 7. Other methods Polymerization in presence of Cer-salts 43 , Initiation by a cobalt complex 43 by nitrogen dioxide ~,4s , effect of Ag(I) and Cu(II) on the polymerization initiated by peroxodisulfate ions 46 , polymerization catalyzed by bisulfite 47 .
`
`SNF Exhibit 1013, Page 5 of 96
`
`

`
`378 W.-M. KULICKE, R. KNIEWSKE and J. KLEIN readily soluble, high molecular weight, linear polyacrylamide (PAAm) samples. In Table 2 a summary of polymerization procedures is given. Summaries of different polymerization methods are also given in ref. 3 (1969), ref. 4 (1954), ref. 5 (1962), and ref. 6 (1973). The polymerization with H2022'4'7 has the obvious advantage to produce linear, completely soluble polymers, ff the conversion is kept below 20%. Furthermore a wide molecular weight range is obtainable by means of various water-alcohol mixtures as polymerization medium. No residual initiator or ions can be present in the polymer. However, saponification as a side reaction takes place if the polymerization is carried out at temperatures exceeding 70°C. 4'48'49 The polymerization with persulfate, 4'6-9 AIBN, 4 and other initiators 1°-12 can lead to residual ions or radicals in the polymers. AIBN as initiator can be used for bulk polymerization but low molecular weight polymers are produced only. Using redox initiators 4'13-2s the polymerization can be performed at low temper- atures, but again the disadvantage exists that residual ions may be present in the polymer sample. Choosing the photo polymerization 29-33 method good results were obtained only by means of sensibilisators. This includes the danger of product contamination too. The radiation induced polymerization 4'6'7'34'3s is generally applicable for bulk polymerization. The advantages are low reaction temperatures and the formation of high molecular weight products. A recently applied initiation reaction is based on electro--chemical methods, ~-39 which represents a redox reaction. The initiation by ultrasonics 4'4°'41 can be accompanied by depolymerization effects, if the solutions are not strictly degased. Therefore, low molecular weight samples may be obtained. Various initiation systems 42~7 are collected in point 7, which have in general the lack to produce no high molecular weight samples. Additionally a contami- nation with residual initiator systems cannot be excluded. As far as the work from our laboratory is concerned, the PAAm samples were TABLE 3. Polymerization conditions of PAAm (150 g AAm was dissolved in 1500 cm s solvent. For the water/methanol mixtures the ratio of the components was 9:6 by volume. The polymers were precipitated two times in methanol and freeze dried. The conversion was kept below 20%) Sample Solvent H2 02/(cm 3 ) T/(°C) [7]/(cm 3" g-I ) 1 H 20/MeOH 6 67 232 2 H~ 0/MeOH 6 50 315 3 H 20/MeOH 6 40 345 4 H20 6 40 740 5 H20 5 40 950 6 H=0 5 40 1360 7 H20 3 40 1400 8 H~0 3 40 1500 9 H20 5 34 1580 10 H20 5 34 1676
`
`SNF Exhibit 1013, Page 6 of 96
`
`

`
`SURVEY OF POLYACRYLAMIDE 379 prepared in a free radical solution polymerization in water using H202 as initiator. 2 From the fact that under more drastical polymerization conditions - namely high amounts of a persulfate-bisulfite initiator at 70°C - branched PAAm with only a few long chain branches are produced, it is justified to conclude that linear PAAm will be formed by the mild polymerization mentioned above, see section 2.3. Table 3 presents examples of polymerization conditions, which have been used to prepare high molecular weight polymers. The samples prepared by this procedure are white, amorphous, and cotton-wool like solids. They are soluble in water, formarnide, and ethylene glycol. Furthermore they have a great affinity to humidity and should be stored in vacuo over a drying agent. 2.3. Long chain branching The statement given above, that linear polymers are obtained in the course of a regular free radical polymerization reaction at low conversion, is of critical import- anee for all the following discussions on solution structure of PAAm. It is there- fore important to present some experimental findings on PAAm samples, where branches have been deliberately incorporated into the polymer. Such investigations on PAAm with controlled chain branching have not been done excepting two papers. 48'49 The earlier experiments 49 were carried out with radiotagged acrylamide which was polymerized in a solution containing untagged PAAm. It could be shown that, by polymerizing acrylamide at 78°C with a persulfate-bisulfite initiator, significant amounts of branched polymer are formed by grafting. However, poly- merization at 50°C gave linear polymers only. In ref. 48 a similar graft polymerization in water is described preparing long chain branched PAAm. The best graft polymerization conditions were found to be 70°C using relatively high amounts of initiator (persulfate-bisulfite). The graft CH2=CH/c--o + xrCH2-Cm H']'x/ C=0 INITIATOR70°C" ~ ~(~ H2N H2N FIG. 1. Graft polymerization of preparing long chain branched polyacrylamide Polymerization condition 4s Backbone ~ = 794000g • tool -1 ) : 2,34g Acrylamide : 7,5 g Initiator (Redox system) : 0,124 g (NH4)~ $20e 0,063 g Na 2 $205 Solvent : 150 ml H~0 Temperature : 70°C Polymerization time : 60 min
`
`SNF Exhibit 1013, Page 7 of 96
`
`

`
`380 W.-M. KULICKE, R. KNIEWSKE and J. KLEIN formation occurred by polymerizing acrylamide in solution in the presence of a well defined PAAm backbone (Fig. 1). The separation of non-grafted branches has been achieved with "Hollow Fiber" tubes. The molecular weights of backbone and branches were 794 000 and 48 000 respectively. From light scattering measurements a weight average molecular weight of 1.8 • 106 was obtained and the average number of branches along the backbone chain was estimated to be around 20. This deliberately much branched PAAm exhibits a smaller hydrodynamic volume as compared with a linear one of the same total molecular weight. 4s The ratio of [r/] b/[~] 1 was estimated to 0,60 which serves as measure of branching density. Moreover, the flocculation activity of linear PAAm was shown to be better than of branched ones. s° The results give rise to the supposition that, using the free radical polymerization at low temperatures and small amounts of H202-initiator, linear samples will be produced only. 2.4. Polymer modification Pure PAAm can be modified in a def'mite manner by Hofmann degradation st , Mannich reaction s2, polymer analogous reaction (e.g. hydrolysis), and other methods. 6,s3 Applying the Hofmann degradation to PAAm by use of a very slight excess of sodium hypochlorite and a large excess of sodium hydroxide at 0 ° to -15°C for about 15h, polyvinylamine (95,6mo1--% amine units) is obtainable, sl Poly- vinylamine is readily soluble in water, acetic acid, and lower alcohols. NaOC1, NaOH -ltCH2--~ H-]-x 0 ° to--15°C, 15h ~ -I-ell2-({ H-]-x C=0 NH2 I NH2 The Mannich reaction of formaldehyde and dimethylamine with PAAm was studied with 13C NMR spectroscopy by s2 -['CH,-'~H=]-x ,, CH,(CH3)2NHX(oI-I), =[-CH2"-~H-]-x + H20 C=O C=O I I NH2 CH2--N(CH3)2 Acrylic acid groups can be introduced into PAAm by polymer analogous reactions, e.g. hydrolysis of amide or ester side groups, s4-ss The acidic hydrolysis was studied by. s4 Strong acidic conditions, however, would lead to a considerable amount of
`
`SNF Exhibit 1013, Page 8 of 96
`
`

`
`SURVEY OF POLYACRYLAMIDE 381 cyclic imide structures, s9 On the other hand an alkaline hydrolysis can be used to introduce acrylate groups into the PAAm. This saponification with aqueous NaOH at room temperature (23°C) leads to the sodium salts (poly(acrylamide-co-sodium acrylate)s) which have the same degree of polymerization as the basic PAAm. sT' 58,60,61 -[-CH2-CH-]-x Temp.,TimeNaOH -[-CH2--~H~H,--~H-]-z C=O C=O C=O I I I NI-I2 NH2 O- Na* The saponification products have a statistical distribution of acrylate side groups and practically no block structure will be formed, s7 The degree of hydrolysis depends on temperature and reaction time. An example for different polymer con- centrations and molarity of saponification agent is presented in Fig. 2. The results were obtained using high purity PAAm from our laboratory. o~ 100 0 E ,, 3-10"3g.tm "3 PAAm in 11'4 Na0H ¢~Jlt~ • 3.10"3g.cm "3 PAAm in 211 NaOH in 2N NQOH 100-t .~ .c 1 ~ 80- It 601~' ~ 40- I \\ ' ' ' ' I ' ' ' ' I ' ' ' ' ~.o-I \~_ o 5 lo . !s ~ __ ____~ ~---- ~-~.I~ I I : I- : . . :I I~- -:I - : .............. 10 2"f / m,n 20I ' ' ' ' I ' ' ' ' I ' ' ' ' I .... I ' ' ' ' I ' 0 SO 100 ISO 200 lO'Z'f I rain FIG. 2. Dependence of the degree of hydrolysis on the reaction time for polyacrylamide, (PAAm), at 23°C. It is of interest that the content of amide side groups doesn't decrease below about 30mol % acrylamide, even at saponification times of 30 days, see Fig. 2. The reason for the high conversion limit is may be caused in the "decreasing reactivity of the amide groups depending on the structure of the neighbouring groups", s7 Similar behaviour was also reported, see ref. 55 and ref. 5, Vol. 14/2, p. 711.
`
`SNF Exhibit 1013, Page 9 of 96
`
`

`
`382 W.-M. KULICKE, R. KNIEWSKE and J. KLEIN In a recently presented paper ss it was mentioned: "that in the beginning of the hydrolysis process a flower attack of hydroxyl ions may be present" and was compared with fictitious similar observations of another hydrolysis reaction. 62 This presumption was due to the kind of plotting half-logarithmic: mol % AAm = f(log t) and has to be reconsidered. A more meaningful plotting procedure may be mol % AAm = f(t), as it has been done in Fig. 2. From this one can conclude that the reactivity is faster in the beginning and slowing down with saponification time. The partial hydrolysis gives rise to a coil expansion in 0,5 M NaBr s7 and 0,5 M NaC16° aqueous solution with a maximum around 50mol % amide units. Corre- spondingly a minimum is observed in the sedimentation coefficients from ultra- centrifugations experiments. 6° Furthermore the flocculation speed of PAAm/AAcNa in Kaolin suspension depends on the saponification degree, s°'6a A set of [7]-M and so-M relations over a wide molecular weight range from Mw = 5"10 s to 5,5"104 and a compositional range from 100 to 30mo1% AAm is given in ref. 60. Several publications have been appeared on the effects of salts and temperature on conformational properties and chemical stability of PAAm/AAc- Na, 61'64 but should not be covered here in detail. 3. CHARACTERIZATION AND PROPERTIES IN DILUTE SOLUTION 3.1. Experimental methods and results 3.1.1. General - As a consequence of the practical importance of FAAm the demand of a precise characterization is obvious. The polymer analysis includes a number of methods to determine the molecular weight, mol.wt.distributions, coil size, thermodynamic interaction with the solvent, etc. The methods most frequently used are the viscometry and light scattering. The methods applied to the charac- terization of PAAm in dilute solution shall be shortly summarized in the following sections. It must be empba~zed that the comparison of results of different labora- tories cannot include in each case the discussion details regarding polymerization conditions, conversion, and so on. It must then be referred to the original paper. Additional to the published results from the literature some as yet unpublished data obtained in our laboratory, concerning measurements of viscosity, light scat- tering, refractive index increment, and partial specific volume, will be mentioned. In Table 4 several properties of PAAm are summarized. The IR-spectra of PAAm films ~? (deuterated and undeuterated) are shown in Fig. 3. The absorption band assignment is listed in Table 5. NMR-studies of aqueous PAAm solutions have been made by Bovey and Tiers. ss 3.1.2. Gel Permeation Chromatography - There are some difficulties in GPC analysis of water soluble polymers: in comparison to organic systems calibration
`
`SNF Exhibit 1013, Page 10 of 96
`
`

`
`SURVEY OF POLYACRYLAMIDE TABLE 4. Properties of PAAm 383 Solid properties Density Glass transition temperature Solvents Formula Freeze dried: White, amorphous, cotton-wool li~e. Great affinity to humidity predpitated in methanol and dried: glass hazd, partly transparent 1,302g • cm -3 at 23°C 3 153°C ~5 165°C 66 188 ° C sT Water Formamide Ethylene glycol -[-CH,-~H-]-x C=O loo_ 80 "~ /,0 ~ 20 0 I /.000 3000 2000 I I I 1500 1000 500 wavenumber / (cm "I) 100 80- 60- 4.O- t/) r 20- 0 /,000 I I 3000 2000 C=0 I NH 2 I I I 1500 1000 500 wovenumber / (cm-1) FIG. 3. IR-spectra of undeuterated (a) and deuterated (b) PAAm film
`
`SNF Exhibit 1013, Page 11 of 96
`
`

`
`384 W.-M. KULICKE, R. KNIEWSKE and J. KLEIN TABLE 5. Tentative absorption band assignment of the IR-spectra of PAAm and deuterated PAAm wave number intensity assignment [cm-' ] Fig. 3. PAAm H Fig. 3 PAAm D 3450 strong v(OH)H20 3330 strong Vas(NH 2 ) 3190 strong us(NH 2 ) 2930 weak ~(CH) 1660 strong ~(C=O) 1615 strong 6 (NH~) 1615 (weak) v(C--N) 1450 medium 6 (CH 2 ) 1410 medium v(C-N) 3450 medium v(OH)H~ O 2930 weak v(CH) 2550 strong ua s(ND~ ) 2500 weak v(OD)HD O, D~ O 2390 strong vs(ND 2 ) 1640 strong v(C=O) 1615 weak shoulder v(C--N) 1520 medium ~ (ND 2 ) 1450 medium 8 (CH 2 ) 1420 medium v(C--N) of aqueous GPC is somewhat difficult because no narrow distribution, extremely high molecular weight water soluble standards are commercially available (with the exception of poly(sodium styrene sulfonate)s, but up to M ~ 1 • l06 only). Cali- bration therefore requires time consuming fractionation or the application of mathematically tedious methods for calibration with polydisperse standards. The main factors that complicate aqueous GPC are adsorptive and thus enthalpic inter- actions between polymers and hydrophylic stationary phases. Consequently, GPC measurements are possible only when careful elaborated conditions are observed. Most of the literature in the field of aqueous GPC is concerned with the develop- ment of specific polymer separation systems by modification of stationary or mobile phases. In addition, the enormous variety of commercial available stationary phases reflects the special problems in this field. High speed GPC of water soluble polymers has been investigated by using TSK- Gel, type PW columns packed with small porous particles. 69 Besides other water soluble polymers like dextran, polyethylene glycol, polyvinylalcohol, and poly- vinylpyrrolidone, PAAm was separated according to molecular size with no evi- dence of adsorption. Semirigid and rigid column packings, such as ion-exchange resins, 7° poly- ethylene glycol dimethylacrylate gels, 71''n poly(2-hydroxyethyimethacrylate) gels, ~,~ polyacryloylmorpholine gels, 7s polyvinylalcohol gels, 76 porous silica beads, ~-79 and porous glasses, 79-as have also been evaluated for use in aqueous GPC, but suffer from drawbacks of pore size, adsorptivity, and resolution. 6s
`
`SNF Exhibit 1013, Page 12 of 96
`
`

`
`SURVEY OF POLYACRYLAMIDE 385 Various commercially available stationary phases for GPC were tested by Klein and Westerkamp s6 to determine their effectiveness in aqueous exclusion chroma- tography. It was found that controlled pore glass (CPG) is the most suitable material for the separation of PAAm, PAAm/AAcNa, dextran, and poly(sodinm styrene sulfonate) in 0,1 M aqueous Na2 SO4 solution. The following demands had to be satisfied by the suitable stationary phase: (1) Stability to high pressures, to allow short time analyses (2) Large pore diameter for the separation of polymers up to several millions in molecular weight (3) No tendency toward adsorption of polymers Using broad distributed PAArn standards a calibration curve was established by means of a trial and error procedure. Mw-range = 140000-5 500000. ~ To avoid artificial oscillations on the evaluated distribution curves a cubic B--spline represen- tation of the calibration curve was used. A 0,1 M aqueous Na2SO4 solution as mobile phase was found to be the best compromise to avoid exclusion phenomena on one side and adsorption tendencies on the other. Because GPC material is com- mercially available only with pore diameters up to 300 nm, separation is limited to molecular weights of several millions. Applying this system the solution instability of PAAm was followed by GPC. The effectiveness of aqueous GPC was demonstrated in an evaluation of thermal degradation measurements of PAAm. The results of these both investigations are presented in Section 3.4.1. and 3.4.2. Furthermore the validity of an universal calibration in the separation system described was tested with narrow distribution poly(sodium styrene sulfonate) standards on the basis of the standard calibration curve established for PAAm. The applicability of GPC to PAAm was also examined on the same stationary phase of controlled pore glass (GPC) by Onda eta/. sT'Ss Formamide s7 as well as aqueous solutions ss were used as mobile phase (eluent). The GPC investigations in formarnide solution 87 were made using fractionated samples of three commercial products. The aqueous GPC studies ss were made with fractions of samples, which were polymerized at pH 10 and 40°C for 8h, using sodium periodide as initiator. The measurements were performed with molecular weights from 0,14.106 to 1,59- 10 e at 30°C. The application of the separation system to the aqueous solution has been examined from both the viewpoint of the effect of salt addition and the GPC mechanism. An adequate addition of neutral salt, 5 • 10-3M KC1 to the eluent gave rise to the elution behaviours being in accord with the hydrodynamic volume concept of the GPC separation. 3.1.3. Diffusion and sedimentation - Another method to determine the molecular weight of a polymer sample is the measurement of the diffusion coefficient Do,
`
`SNF Exhibit 1013, Page 13 of 96
`
`

`
`386 W.-M. KULICKE, R. KNIEWSKE and J. KLEIN provided that there is a known relationship between Do and M. Such cross corre- lations are given below, but the molecular weight determination by this way is a less common method. Furthermore the molar masses can be obtained from sedi- mentation measurements by ultracentrifugation, however, one distinguishes different types of ultracentrifugation methods. (a) The method of sedimentation velocity given from Svedberg. Choosing that way, one has to determine the sedimentation coefficient so and to know the dif- fusion coefficient Do and can calculate the molecular weight by the Svedberg equation soRT M,v - Do(l -¢2p,) It is also possible to obtain the molecular weights from the sedimentation coefficient s o using a known so-M relation (see below). Co) The method of sedimentation equilibrium. An important parameter for the molecular weight determination using the Svedberg equation and the method of sedimentation equilibrium is the partial specific volume ~72. Therefore, the treatment of this quantity has been made first, followed by the presentation of diffusion - and sedimentation - molecular weight relationships. Partial specific volume: An as yet unpublished determination of the partial specific volume performed in our institute should be presented here. Density measurements were made at 20°C using DMA 02 Digital Precision Density Meter manufactured by A. Paar KG (Graz, Austria). The partial specific volume "q2 was calculated from the slope of a plot of pp versus c by the following equation pp = p, + (I --V/2p,)c (3-1) where p, and pp are the density of the solvent and of the solution respectively. The measurements were performed on six aqueous solutions of a PAAm with M w = 5 • 106. The polymer concentations were chosen between 0,06 and 1.2 g/l. No concentration dependence could be observed in this range. Repeating the deter- ruination three times we always calculated V2 = 0,693 cm 3- g-1 (Table 6). TABLE 6. Partial specific volume V2 of PAAm solvont V2/( cms" g-' ) T/(°C) references water 0,693 20 this work water 0,697 25 89 water 0.696 20 90 water 0,769 20 91 aqueous 0,2 M NaCl 0,687 20 90 aqueous 0,1 M NaCI 0,702 9. 92a
`
`SNF Exhibit 1013, Page 14 of 96
`
`

`
`SURVEY OF POLYACRYLAMIDE 387 This value is dose to that obtained by M~chtle s9 of V2 = 0,697 cm 3 g-i using the same PAAm with Mw = 5 • 106 . A determination published by Munk et al. 9° is supported by both values. They found '¢z = 0,696 cm3" g-i using molecular weights of 225000 and 616000. From these results one can conclude that no molecular weight dependence from 2- 10 s to 5 • 106 seems to be present. On the basis of these experiments the considerably higher value of ~¢z = 0,769 cm 3" g-~ reported by Scholtan 91 is not explainable. In salt solution values slightly differing from those determined in water were found s°'9~ (Table 6). Diffusion coefficient: A concentration dependence of diffusion coefficients was measured by Scholtan 91 and a relation between diffusion coefficient D o and molecular weight MSD derived. Recently a new relationship of diffusion coefficient as a function of molecular weight Mw was presented by Francois, Schwartz, and Weill. 93 A remarkable differ- ence in the constant and the exponent is obvious (see Table 14), Do = 8,46" 10 -4 MS°/~ 69 (cm 2- s -1) Water, 20°C 91 (3-2) Do = 1,24" 10-4M'w°w '$3 (cm 2" s -1) 0,1MNaC19a (3-3) but a comparison cannot be made, because no temperature was given by the authors 9a and saline solutions were used. Sedimentation measurements: The determination of sedimentation coefficients of PAAm in 0,5M aqueous NaC1 at 20°C by ultracentrifugation measurements were made by. 94 From these investigations a relation of the sedimentatio