Surface sulfate groups on poly(methyl methacrylate) and poly(vinyl acetate) particles from soap-free emulsion polymerization

Size: px

Start display at page:

Download "Surface sulfate groups on poly(methyl methacrylate) and poly(vinyl acetate) particles from soap-free emulsion polymerization"

Elizabeth Tyler
5 years ago
Views:

1 e-polymers 2005, no ISSN Surface sulfate groups on poly(methyl methacrylate) and poly(vinyl acetate) particles from soap-free emulsion polymerization Tadanobu Saiga, Atsushi Suzuki *, Kenji Kikuchi, Takuji Okaya The University of Shiga Prefecture, Department of Materials Science, School of Engineering, 2500 Hassaka, Hikone , Japan; Fax ; (Received: July 29, 2005; published: November 8, 2005) This work has been presented at the 2 nd International Symposium on Polymeric Microspheres, March 29-31, 2005, in Fukui, Japan Abstract: The amounts of sulfate groups on the particle surfaces of poly(methyl methacrylate) (PMMA) and poly(vinyl acetate) (PVAc) obtained in soap-free emulsion polymerization were determined. Polymerization was carried out at low monomer concentration using ammonium persulfate as an initiator. After the ionexchange procedure, conductometric titration was carried out. Amounts of surface sulfate groups were lower and particle diameter was larger in the PMMA system compared with PVAc. A large and remarkable difference appeared in the ratio of surface sulfate groups to decomposed sulfate radicals, viz. 90% for the PVAc system and 50% for PMMA. The ratios of sulfate groups per polymer molecule on PMMA and PVAc particles (functionality) were calculated. The functionality of the PMMA system was 70% of the theoretical value, while it was close to 100% in the PVAc system. This large difference might arise from the weak reactivity of the sulfate radical with MMA compared with that of VAc, due to the strong electrophilic nature of the radical. A side reaction of the sulfate radical with MMA is assumed. On the basis of these experimental results, the instability of the emulsion polymerization of acrylic monomers using poly(vinyl alcohol) (PVA) as a protective colloid is discussed in terms of the instability of the soap-free particles formed after the consumption of free PVA in water by the grafting reaction. Introduction In polymer emulsions, an advantage of using PVA as a protective colloid, in contrast to the common surfactants, is the strengthening effect of PVA on the emulsion films [1-2]. PVA has been widely used industrially in the emulsion polymerizations of vinyl acetate (VAc) and VAc/ethylene as a protective colloid. However, it has not been employed in the emulsion polymerizations of conjugated monomers such as acrylic derivatives and styrene/butadiene due to the lack of stability during the polymerization. This lack of stability has been believed to be caused by the weak ability of grafting of these conjugated monomers onto PVA, since the growing chain radicals from the monomers are not as reactive as that from VAc. In other words, the growing chain radical from VAc is reactive because of no conjugation in the polymer radical. 1

2 The authors had doubts about this, since there should be two grafting mechanisms: one is hydrogen abstraction from PVA with the growing chain radicals, and the other with the primary radicals. In order to demonstrate the ability of grafting acrylics onto PVA, the authors employed a model experiment utilizing MMA as a representative of acrylics, and a very low monomer concentration (MMA/PVA/water = 1/1/100) that can be regarded as the initial stage of emulsion polymerization, since emulsion polymerization starts in the water phase. As the authors have reported in previous papers [3-7], grafting took place during the model polymerization: about 95% of polymerized MMA and approximately 65% of PVA were grafted. Surprisingly there was almost no difference between the results in MMA and VAc systems. When the amount of monomers was increased from 1 to 5 (or 10) (water and PVA amounts being the same as before), a difference appeared: In the case of MMA, the particles became unstable and coagulated. On the contrary, in the case of VAc, emulsion polymerization proceeded to the last stage. This may have arisen from the decreasing amounts of free PVA in water that was consumed by grafting in the initial stage [8,9]. After the consumption of free PVA in water, the primary radicals seem either to enter into the polymer particles or to form soap-free particles in water. Accordingly the stability of the soap-free particles was thought to play an important role. The aim of this study is to measure the sulfate groups on the surfaces of PMMA and PVAc particles and to clarify the stability of soap-free emulsion obtained in the socalled model emulsion polymerization (in the absence of PVA). Finally the results will be discussed with respect to the lack of stability of the emulsion polymerization of MMA using PVA as a protective colloid. Experimental part Materials Methyl methacrylate (MMA) was distilled under reduced pressure before polymerization. Vinyl acetate (VAc) of polymerization grade supplied by Kuraray was used. Ammonium persulfate (APS) (Wako Pure Chemical) of GR grade was used as received. Soap-free emulsion polymerization of MMA and VAc Soap-free emulsion polymerization was performed under argon atmosphere using APS at 60 C for 2 h on the basis of the basic recipe: monomer / continuous phase = 1/100 (v/v). Amounts of APS were varied. Conductometric titration The purification of the soap-free particles of PMMA and PVAc was carried out by the ion-exchange method to remove several ions, NH 4 +, S 2 O 8 2- and SO 4 -, in the aqueous phase. The purified soap-free particles were titrated with sodium hydroxide solution. ICP measurements (inductively coupled plasma spectroscopy) To determine the amount of sulfate groups in the soap-free particles, the standard ICP method was used. The purified soap-free emulsions (25 ml) were mixed with sodium carbonate (Na 2 CO g) and sodium sulfate (Na 2 SO 4 ) (1-12 ppm) in a 2

3 melting pot. These samples were burned with an electric furnace for 12 h at 380 C. The sulfates in the ashes were soluble in ion-exchange water. The amounts of sulfate were determined with ICP (SEIKO Instruments SPS 4000, Plasma Spectrometer). Measurements Conversion was measured gravimetrically. Particle diameters were measured by dynamic light scattering (DLS, PAR-III, Otsuka Electronics). The molecular weights of the resulting polymers (M n ) were measured with gel permeation chromatography (GPC) (Tosoh, TSK gel GMH HR -M) using polystyrene as a standard for calibration in tetrahydrofuran at 40 C. The viscosity-average molecular weight of the polymers (M v ) was determined by viscometry. In the case of PMMA, M v was converted into M n, assuming that the ratio of M v to M n was 1.5. We confirmed a slight difference in the M v by GPC ( ) and in the M v by viscometry ( ). In the case of PVAc, M v can t be determined exactly by viscometry. M n was measured by GPC. There were two peaks, where the high value ( ) indicates the bridge-building polymers. The low value ( ) was used for functionality, considering a slight difference in M v by GPC and viscometry. Decomposition rate of APS There are three steps in the decomposition reaction [11], which can be written as follows: where k d is the rate constant of the decomposition reaction of APS. k d was measured by the conductometric titration with 0.01 M NaOH [10]. Fig. 1 shows the decomposition vs. time. The decomposition rate constant, k d, was measured to be s -1, which is slightly lower than that of potassium persulfate (KPS) [11]. + 2,5 [H + ] (10-5 mol L -1 ) 2 1,5 1 0,5 Fig. 1. Decomposition of APS. Recipe: APS g, NaSO g, H 2 O 400 g, 60 C Time (min) 3

4 Results and discussion For the particles obtained in the soap-free emulsion polymerization of MMA and VAc, the amounts of surface sulfate groups were measured. Fig. 2 shows examples of the conductometric titration curves for both systems. There are V-shaped lines that show strong-acid behaviour in the conductive titration. We confirmed the reproducibility of the V-shaped line under sufficient stirring during neutralization. From the titration experiments, the concentration of the strong acid, i.e., of the sulfate groups, on the particle surfaces was determined. In Tab. 1 are listed the concentrations of surface sulfate groups in both soap-free systems together with particle diameters. Furthermore, the ratios of sulfate groups on the particle surfaces to sulfate radicals formed by the decomposition of APS (I s /I d ) are given. 6 (a) 4 (b) Conductivity (µs/cm) 4 2 Conductivity (µs/cm) ,5 1 1, N NaOH (ml) 0 0 0,2 0,4 0,6 0, N NaOH (ml) Fig. 2. Conductometric titration curves of PMMA (a) and PVAc (b) polymer particles obtained via soap-free emulsion polymerization. Recipe: MMA 0.5 ml, VAc 1 ml, APS 0.05 g, water 100 ml, 60 C Tab. 1. Concentration of sulfate groups, their ratio to formed sulfate radicals (I s /I d ), and occupied area per sulfate group (S) on the particle surfaces of PMMA and PVAc. Recipe: MMA 0.5 ml, VAc 1 ml, water 100 ml, APS, 60 C APS in g D n in nm PMMA SO 4 - in µeq per g polymer I s /I d in % S in Å 2 D n in nm PVAc SO - 4 in µeq per g polymer I s /I d in % S in Å The concentrations of surface sulfate groups on PMMA particles were lower than those on PVAc particles. Particle diameters of PMMA were larger compared with those of PVAc to the extent of several ten percents. There exists a large and remarkable difference in I s /I d in both systems: the ratio is about 50% for PMMA and c. 90% for PVAc. This indicates that in the PVAc system the decomposed sulfate 4

5 radicals are efficiently utilized as surface sulfate groups, while in the case of PMMA only c. 50% of them are bound to the surface. There is not such a large difference in the occupied area per sulfate group in both cases. From these results, the following conclusion might be drawn: in the case of PMMA, the amount of sulfate groups on the surface is so low that moderate flocculation takes place to decrease the surface area, to result in an increase of the particle diameter and almost the same occupied area as in the case of PVAc. In fact it was observed that there was unstable soap-free emulsion polymerization in 1 ml of MMA system differing from the 1 ml VAc system. Thus, we used 0.5 ml of the MMA system instead of 1 ml. We think that during titration the sodium ions cannot easily penetrate into the soapfree particles because the degree of swelling of PMMA with water is less than 3% after long time. For PVAc particles, we reached the same conclusion. There may be a chance that charged groups are buried inside the particles. In order to check this possibility, we determined the total amounts of sulfate groups in the soap-free particles by ICP. Tab. 2 lists the ratio of surface sulfate groups to total sulfate groups in a particle. In the case of PMMA initiated by APS (0.05 g), 54% of sulfate groups are located on the soap-free particles. In the case of PVAc, 80% of sulfate groups are found on the surface. Thus, there is a high fraction of sulfate groups on soap-free PVAc particles in comparison with PMMA. This result may be related to the affinity of the monomers with sulfate groups. In both cases, from geometrical and entropic reasons, several thousands of polymer molecules without sulfate end group were coiled inside the polymer particle. Tab. 2. The ratio of surface sulfate groups to total sulfate groups in a particle. Recipe: MMA 0.5 ml, VAc 1 ml, APS 0.05 g, water 100 g, 60 C, 2 h D n in nm ICP (I total ) in µeq per g polymer Conductive titration (I s ) in µeq per g polymer I s /I total in % PMMA PVAc The ratio of sulfate end groups to polymer molecules, functionality, can be calculated, considering the total number of sulfate groups in a particle converted from the amount of surface sulfate groups on the particle with I total /I s (1.9 for PMMA, 1.3 for PVAc) in the low initiator concentration case (APS g), while the determination of the amount of sulfate groups inside the particle was impossible due to the limited sensitivity of ICP. From the estimated sulfate content inside the particle, particle diameter, and the number-average degree of polymerization of PMMA (DP n = ), the functionality of the PMMA system (APS g) was estimated to be 0.93, while that of PVAc system was 0.88 (DP n = ). Although the functionality of PMMA was smaller to some extent compared with that of PVAc, the difference looks not so large. However, we think that the actual difference is rather large considering the following issue: In the polymerization of MMA, the theoretical functionality is determined by the termination (combination and disproportionation) reaction, since chain transfer reactions can be neglected. Accordingly, the theoretical functionality reaches 1.35 [12]. The value 0.93 obtained in this study is 69% of the theoretical value. In the case of VAc, chain transfer reactions to monomer and PVAc take place 5

6 to the large extent. The theoretical functionality must be less than 1.0, although it is difficult to obtain it accurately. Accordingly the value 0.93 obtained in this study looks very close to the theoretical value. Let us consider the reason why sulfate radicals are not bound effectively to the end of PMMA molecules in contrast to PVAc molecules. In the reactions of the highly electrophilic sulfate radical with MMA and VAc as shown in Fig. 3, addition of the sulfate radical to VAc must be easier compared with that to MMA, since the double bond in VAc is of electron-rich nature (e = [13]) while that in MMA is of electrondeficient nature (e = 0.40 [13]). (a) MMA (Q = 0.78, e = 0.40) (b) VAc (Q = 0.026, e = -0.89) Fig. 3. Reactivity of sulfate groups with MMA and VAc McGinniss and Kah reported that the reaction rate constant of the sulfate radical with MMA ( L mol -1 s -1 ) was slightly lower than with VAc ( L mol -1 s -1 ) [14,15]. However, it should be noted that these are not the addition rate constants of the sulfate radical to the monomers but the constants of decay of the sulfate radical in presence of the monomers. In other words, their values may be over-estimated because they contain the rate constants of other side reactions. Accordingly the addition rate constant of the sulfate radical to MMA may be considerably smaller than the above-mentioned value because of the presence of the following side reaction: The highly electrophilic sulfate radical may attack an electron-rich portion in MMA instead of the electron-deficient double bond. Since the most electron-rich portion in MMA is the oxygen atom in the carbonyl group, following reactions may take place. As shown in Eq. (4), the first reaction may be the electron abstraction from oxygen in the carbonyl group of MMA with the electrophilic sulfate radical. The carbo-cation formed by rearrangement may react with water to form a new radical as shown in Eq. (5). From the newly formed radical, a methoxyl radical may be formed as shown in Eq. (6). The methoxyl radical may react with the double bond in MMA forming a PMMA molecule with no sulfate end group. Although we have no evidence that these side reactions actually take place, Eqs. (4) - (7) offer a possible explanation for two important experimental results: in the PMMA system, low ratios of surface sulfate groups to decomposed sulfate radicals (50%, compared to c. 90% in the PVAc system) and low sulfate end group functionality (70%, compared to c. 100% in the PVAc system) are observed. Palit et al. reported the presence of sulfate groups in polystyrene (PS) and PMMA prepared by persulfate initiator using the dye partition method [16-20]. The persulfate-initiated polymers contain very low amounts of sulfate end groups, about per polymer molecule for PS and for PMMA [18]. Several papers 6

7 reported the determination of the number of sulfate end groups on the particle surface of cleaned latex after ion-exchange [21-25]. Vanderhoff et al. [22] pointed out that the dye partition method in PS exhibits quite a lower estimate of sulfate end groups on PS than by other standard methods, such as X-ray fluorescence analysis and conductometric titration. Based on these facts, they argued that the dye-partition method of end-group analysis is not applicable to non-polar polymers. They concluded that the polarity of the polymer certainly affects the quantitative applicability of the dye partition method. Palit demonstrated that in the dye-partition method the most important factor is the polarity of the diluent and not the polarity of the polymer [26,27]. In fact they clarified that other systems with two initiators instead of persulfate yielded polymers with about 2 salt end groups per polymer molecule [26]. Reexamination of the end-groups of the emulsion polymers of PMMA initiated with potassium persulfate indicated sulfate radical end groups per molecule [26]. (1) (2) (3) (4) Sangster et al. [28] studied the reactions of styrene with hydroxyl radical that has also an electrophilic nature similar to the sulfate radical, and concluded that the main reaction is electron abstraction from the phenyl group in styrene to result in hydroxyl 7

8 ion and monomer radical with no bound hydroxyl group, and the addition to the double bond takes place less frequently. For the reason of the low content of sulfate end groups in PS, Palit et al. estimated that the sulfate radical participates predominantly in the electron-transfer reaction with styrene instead of adding directly to the double bond as proposed by Ledwith et al. [29]. To our knowledge, there are no papers reporting on the side reaction of sulfate radical with MMA in acidic solution. An electron abstraction from the monomer with the sulfate radical as described above may be possible. Tauer et al. [30] reported a change of conductivity during the entire experiments for styrene, MMA and VAc. The observed increase in the conductivity indicated hydrolysis that leads to the formation of methacrylic acid and acetic acid during the thermal equilibration period in pure distilled water. However, in our experimental results, the hydrolysis of monomers during the polymerization took place in acidic solution, because of the soap-free emulsion polymerization with persulfate initiator. To clarify the reason of the unsuitableness of utilizing PVA in the emulsion polymerization of acrylic monomers, the authors have studied the initial stage of the polymerization using a very low MMA concentration in the presence of PVA in terms of grafting the monomer onto PVA, since this grafting has been widely believed not to occur in the case of the emulsion polymerization of conjugated monomers such as acrylics and styrene/butadiene due to the low reactivity of the corresponding growing chain radicals. Unexpectedly, grafting of MMA onto PVA took place markedly in the initial stage, and no difference in grafting behaviour was observed in comparison with the VAc system [8,9]. During the initial stage, large amounts of PVA were consumed by grafting to result in particle formation. Accordingly, the amount of free PVA in the water phase decreases with increasing conversion. The sulfate radicals decomposed in the water phase either enter into emulsion particles or form soap-free new particles by adding monomer in the water phase followed by propagation until coiling up. In this regard, the stability of the soap-free particles may be the important key factor. Conclusion The result obtained in this paper may shed light upon the following: in the soap-free emulsions of PMMA and PVAc, there exists a large difference in the amounts of surface sulfate groups. In the case of VAc, sulfate radicals are utilized efficiently as surface sulfate groups bound to PVAc molecules and the particles are stable, while in the case of MMA, utilization of the sulfate radicals as surface sulfate groups bound to PMMA molecules are limited because of the side reaction as described in this paper. In the case of PMMA, the soap-free particles are not stable because of the low amounts of surface sulfate groups, resulting in partial flocculation thus decreasing the surface area. This flocculation seems to lead the emulsion system of acrylics using PVA as a protective colloid to coagulation in the course of emulsion polymerization. [1] Yuki, K.; Saito, T.; Maruyama, H.; Yamauchi, J.; Okaya, T.; Polym. Int. 1992, 30, 513. [2] Yuki, K.; Nakamae, M.; Sato, T.; Maruyama, H.; Okaya, T.; Polym. Int. 2000, 49, [3] Okaya, T.; Suzuki, A.; Kikuchi, K.; Macromol. Symp. 2000, 150, 143. [4] Okaya, T.; Suzuki, A.; Kikuchi, K.; Colloid Polym. Sci. 2002, 280,

9 [5] Okaya, T.; Suzuki, A.; Kikuchi, K.; Colloid Polym. Sci. 2002, 188, 288. [6] Suzuki, A.; Yano, M.; Saiga, T.; Kikuchi, K.; Okaya, T.; Progr. Colloid Polym Sci. 2003, 124, 27. [7] Suzuki, A.; Yano, M.; Saiga, T.; Kikuchi, K.; Okaya, T.; Colloid Polym. Sci. 2003, 281, 337. [8] Suzuki, A.; Saiga, T.; Kikuchi, K.; Okaya, T.; 39 th IUPAC Congress and the 86 th Conference of the Canadian Society for Chemistry 2003, p.163. [9] Suzuki, A.; Saiga, T.; Kikuchi, K.; Okaya, T.; IUPAC Polymer Conference on the Mission and Challenges of Polymer Science and Technology (IUPAC-PC2002), p [10] Kikuchi, K.; Ueno, Y.; Okaya, T.; Kobonshi Ronbunshu 2003, 60, 448. [11] Kolthoff, I. M.; Miller, I. K.; J. Am. Chem. Soc. 1951, 73, [12] (a) Bevington, J.; Melville, H.; Taylor, R.; J. Polym. Sci. 1954, 12, 449. (b) Bevington, J.; Melville, H.; Taylor, R.; J. Polym. Sci. 1954, 14, 463. [13] Polymer Handbook, 4 th edition, John Wiley & Sons, New York 1999, 316. [14] McGinniss, V. D.; Kah, A. F.; J. Coat. Technol. 1977, 49, 61. [15] McGinniss, V. D.; Kah, A. F.; J. Coat. Technol. 1979, 51, 81. [16] Ghosh, P.; Chadha, S. C.; Mukerjee, A. R.; Palit, S. R.; J. Polym. Sci., Part A2 1964, [17] Mandal, B. M.; Nandi, U. S.; Palit, S. R.; J. Polym. Sci. B 1967, 5, 677. [18] Palit, S. R.; Mandel, B. M.; J. Macromol. Sci., Revs. Macromol. Chem. 1968, C2, 225. [19] Mandal, B. M.; Nandi, U. S.; Palit, S. R.; J. Polym. Sci. A , 8, 67. [20] Roy, G.; Mandal, B. M.; Palit, S. R.; in Polymer Colloids, Fitch, R. M., editor; Plenum Press, New York 1971, p. 49. [21] Van den Hul, H. J.; Vanderhoff, J. W.; J. Colloid Interface Sci. 1968, 28, 336. [22] Van den Hul, H. J.; Vanderhoff, J. W.; Brit. Polym. J. 1970, 2, 121. [23] Van den Hul, H. J.; Vanderhoff, J. W.; in Polymer Colloids, Fitch, R. M., editor; Plenum Press, New York 1971, p. 1. [24] McCann, G. D.; Bradford, E. B.; Van den Hul, H. J.; Vanderhoff, J. W.; in Polymer Colloids, Fitch, R. M., editor; Plenum Press, New York 1971, p. 29. [25] El-Aasser, M. S.; Methods of Latex Cleaning, in Science & Technology of Polymer Colloids, NATO ASI Ser. E, Appl. Sci. 1983, 68, 422. [26] Banthia, A. K.; Mandal, B. M.; Palit, S. R.; J. Polym. Sci. 1977, 15, 945. [27] Chaudhur, A.; Palit, S.; J. Polym. Sci. 1980, 18, [28] Sangster, D. F.; Davision, A.; J. Polym. Sci., Symp. Ser. 1975, 49, 191. [29] Ledwith, A.; Russel, P. J.; J. Polym. Sci., Lett. Ed. 1976, 13, 109. [30] Tauer, K.; Padtberg, K.; Dessy, C.; ACS Symp. Ser. 801, Polymer Colloids Science and Technology of Latex Systems, Daniels, E. S.; Sudl, E. D.; El-Aasser, M. S.; editors; American Chemical Society, Washington, DC 2002, p

Effect of Polyvinyl Alcohol of Different Molecular Weights as Protective Colloids on the Kinetics of the Emulsion Polymerization of Vinyl Acetate

Effect of Polyvinyl Alcohol of Different Molecular Weights as Protective Colloids on the Kinetics of the Emulsion Polymerization of Vinyl Acetate K.A. Shaffie *1,A.B.Moustafa 2, N.H. Saleh 2, H. E. Nasr