Poly(methyl methacrylate) Latexes

May 13, 2005 - Cationic core/shell polystyrene/poly(methyl methacrylate) latexes (PS/PMMA) have been produced with a core/shell diameter ratio of 1:7...
14 downloads 34 Views 271KB Size
8538

Ind. Eng. Chem. Res. 2005, 44, 8538-8541

Small Core/Thick Shell Polystyrene/Poly(methyl methacrylate) Latexes Robert L. Sherman, Jr. and Warren T. Ford* Department of Chemistry, Oklahoma State University, Stillwater, Oklahoma 74078

Cationic core/shell polystyrene/poly(methyl methacrylate) latexes (PS/PMMA) have been produced with a core/shell diameter ratio of 1:7.5 by starved semicontinuous emulsion polymerization. A cross-linked 80/20 PS/PMMA seed 70 nm in diameter gave 530-nm particles in three steps of PMMA growth. Dynamic light scattering and scanning electron microscopy were used to measure particle sizes. Locations of the cores in the particles were imaged by transmission electron microscopy. Growth using larger core latexes increased particle diameter to 800 nm, but second generation latex particles prevailed in attempts at further growth. Introduction The goal of this research is to form monodisperse micrometer-sized polymer particles having a small core with a refractive index (RI) much different from that of a thick shell. The particles are designed for measurement of diffusion coefficients in concentrated dispersions by dynamic light scattering (DLS). Normally, diffusion coefficients of polymer colloids in concentrated dispersions cannot be measured because of multiple scattering, as shown in Figure 1A. However, from small core/ large shell particles with the RI of the shell equal to that of the medium, light is scattered only by the core, as shown in Figure 1B. The resulting decrease in multiple scattering will make it possible to measure diffusion coefficients in concentrated dispersions. Particles with a 1:10 ratio of core/shell diameters under these conditions scatter light as if the dispersion were diluted by 1/103 ) 1/1000. For this purpose, the core/shell latex must meet several conditions. First, there must be a large difference in RI between the core and the shell. Second, the core and shell must be immiscible in one another and be glassy, not crystalline. Third, the particles must be dispersible in water for synthesis by emulsion polymerization and, after surface modification, must be dispersible in a solvent with a RI that matches that of the shell and that swells neither the shell nor the core. To meet these conditions, we chose polystyrene/poly(methyl methacrylate) (PS/PMMA) core/shell latexes. The RI of polystyrene, 1.59, contrasts sharply to that of PMMA with a RI of 1.49.1 For RI-matching, cis-decahydronaphthalene, RI ) 1.48,2 with small amounts of a higher RI solvent has been used for PMMA particles.3 PS and PMMA are immiscible, and their respective monomers are soluble in both polymers.4 Composites of PS and PMMA have varied morphologies that are determined by how the polymers are made or mixed in materials such as latex particles from emulsion5-7 and dispersion8,9 polymerization, block and random copolymers,10 and homopolymer blends.11 Core/shell PS/ PMMA latex particles have been made via seed growth by monomer swelling,4,6,12,13 batch and semi-batch polymerization,4,6,13 and starved growth polymeriza* To whom correspondence should be addressed. E-mail: [email protected].

Figure 1. (A) Multiple light scattering by a high concentration of homogeneous particles. (B) Light scattering by a high concentration of core/shell particles with matched refractive indices of the particle shell and the solvent.

tion.5,6,13,14 Dispersion polymerization has been used to control the morphology of large particles,7,8 whereas other PS/PMMA composites have been formed by heterocoagulation10,15 and thermal annealing of preformed PS/PMMA composite particles.16 The many variables in the synthetic conditions allow for either kinetic or thermodynamic control of morphology. Under thermodynamic control, core/shell and inverted core/shell latex structures are formed.4,8 Kinetically controlled morphologies result from particle formation under conditions where polymer diffusion is too slow to attain the equilibrium structure. The kinetic structure of a latex often can be converted to the more stable structure. For example, multiple core latexes can be thermally annealed to give one core. The morphologies of composite latexes are controlled by many factors, including differences in the solubilities of the monomers and the polymers, the viscosity of the monomer-swollen polymer, the rate of radical transport into the polymer, crosslinking, the differences in polarity of the monomers and polymers, and the charge of the initiator. In this paper, we report an uncommon latex morphology of a small

10.1021/ie048867j CCC: $30.25 © 2005 American Chemical Society Published on Web 05/13/2005

Ind. Eng. Chem. Res., Vol. 44, No. 23, 2005 8539 Table 1. Compositions and Sizes of Seed Latexes sample

446

491

MMA (mL) styrene (mL) water (mL) VA-044 (mg) T (°C) diameter, DLS (nm) diameter, SEMb (nm)

1.0 3.0 40 20 65 327 320

2.0 8.0a 100 150 80 50 70

a 0.200 g of vinylbenzyl(trimethyl)ammonium chloride and 0.150 g of divinylbenzene comonomers. b Error limits not reported because of possible sample damage by SEM.

PS core and a thick PMMA shell. Large cores and thin shells are common. In the past, large-scale growth of polystyrene latex spheres has been accomplished by seeded emulsion polymerization in which a small latex seed is swollen to many times its original diameter with a monomer, a nonpolar solvent such as hexadecane, and an oil-soluble initiator and then polymerized.12,17-19 We were unable to swell polystyrene cores with large amounts of MMA and produce core/shell particles by this method. We succeeded, instead, by a starved semicontinuous method. Toernell et al. showed that a thin, uniform PS shell could be formed on PMMA latexes using starved semicontinuous emulsion polymerization in which monomer is added slowly to prevent extensive swelling of the particles. Concurrent with the slow addition of monomer, a rapid decomposition of water-soluble initiator causes rapid polymerization in solution and the capture of growing polymer chains by the particles. New polymer deposits as a shell on the seed polymer and continues to grow but cannot diffuse far because the particle is too viscous to allow diffusion of macromolecules from the shell to the core.5,6 Results and Discussion To promote the sticking of growing PMMA radicals to the surface of a seed particle, copolymers of 80/20 or 75/25 PS/PMMA were used as the seeds, which become the cores of the composite latex. The seed particles are reported in Table 1. Seed 491 contained vinylbenzyl(trimethyl)ammonium chloride for additional charge stabilization and divinylbenzene as a cross-linker to limit swelling. The use of pure PS latexes as seeds resulted in aggregation and no PMMA shell growth. We attribute the failure of a PMMA shell to grow on a PS core to slow addition of the MMA monomer, which did not allow MMA to penetrate the core and form a growth locus for a shell before the MMA polymerized on its own. A new second crop of PMMA particles could cause coagulation of the core particles. This hypothesis suggests that faster addition of MMA would have produced core/shell particles from PS cores, but we did not try the experiment after succeeding with the poly(styreneco-MMA) cores.

To generate the high concentration of radicals in water needed for starved polymerization of the PMMA shell, we used the initiator VA-044 {2,2′-azobis[2(2-imidazolin-2-yl)propane] dihydrochloride} (1), which has a 40-min half-life at 65 °C.20 To enhance the colloidal stability of the latexes, a cationic monomer, 2-(N,N-dimethylamino)ethyl methacrylate (2) was added. The results of the starved semicontinuous emulsion polymerizations are in Table 2. The first sample, 455, is a control experiment with no seed. Sample 458 used a 75/25 PS/PMMA seed, whereas the last three samples, 492-494, used an 80/20 PS/PMMA cross-linked copolymer core. A cross-linking monomer, ethylene glycol dimethacrylate (3), in the shell improved the stability of the core/shell particles. Cross-linking at the start of growth might produce PMMA “buds” on the surface of the core. Growth of the buds would give bumpy particle surfaces at intermediate stages of growth, followed by filling of the spaces between buds to produce spherical particles with acentric cores, as reported later in this paper. Growth of the shell worked best with slow addition of MMA. The initiator was added in one batch 10 min prior to monomer addition. The addition of a second batch of initiator during the monomer addition process resulted in second generation particles. The amount of initiator left at the end of the monomer addition over a 2.25-h period was adequate to carry the polymerization to completion. Growth of the 320-nm latex 446 (Figure 2a) gave the monodisperse 730-nm latex 458 (Figure 2b). The attempted growth of latex 458 by several different swelling methods and starved semicontinuous growth methods resulted in the formation of second generation particles. We attribute the secondary particles to insufficient surface area of the seed to capture all of the newly formed oligomer radicals from solution. To attain a core/shell diameter ratio of 1/10 and a shell diameter of less than 800 nm, we started with a smaller cross-linked seed 491. Growth of seed 491 by an approximate doubling of the diameter in three steps resulted in an increase from 70 nm to 160, 300, and 530 nm. A scanning electron microscope (SEM) image of the 530-nm particles is in Figure 3. To determine if a true core/shell morphology was achieved, the latex particles were dispersed into epoxy resin, the resin was cut into 70-nm-thick slices, and the PS was stained with ruthenium tetroxide21 for transmission electron microscopy (TEM). Negative images of sectioned sample 494 in Figure 4 show PS cores in much larger PMMA particles, which confirms the core/shell morphology. The particles in Figure 4 appear in different sizes as a result of the microtoming process. Only a particle cut at its equator appears full size. In a statistically valid sample, only 1/7.5 of the particles should show the core. Possible reasons for the appearance of cores in a large fraction of the particles in Figure

Table 2. Shell Growth by Starved Semicontinuous Emulsion Polymerization sample

seed (g)

volumea (mL)

VA-044 (mg)

MMA (mL)

addition rate (mL/h)

diameter DLS (nm)

diameter SEM (nm)

455 458 492 493 494

0 446 (1.0) 491 (1.0) 492 (1.0) 493 (1.0)

10 110 115 115 115

10 150 150 150 150

1.0b 9.0c 9.0c 9.0c 9.0c

0.5 4.0 4.0 4.0 4.0

330 780 101 255 507

730 160 300 530

a

Total volume of water, seed particles, and monomer. b 10 mg of 2 and 10 mg of 3. c 60 mg of 2 and 60 mg of 3.

8540

Ind. Eng. Chem. Res., Vol. 44, No. 23, 2005

Figure 4. TEM negatives at different magnifications of particles 494 after three growth steps.

Figure 2. SEM images of (top) 300-nm 80/20 PS/PMMA core latex 446 and (bottom) sample 458 from growth of a PMMA shell.

having a diameter of 530 nm and a core/shell diameter ratio of 1:7.5. The high RI contrast between the core and shell should enable the measurement of particle diffusion coefficients at high particle concentrations by DLS in a solvent matched to the RI of PMMA without the complication of multiple scattering. However, transfer of the core/shell particles from water to the RI-matched solvent cis-decalin will require coating with a hydrophobic steric stabilizer such as poly(12-hydroxystearic acid).3,22 Experimental Section

Figure 3. SEM image of particles 494 after three growth steps. The nonspherical particle shapes are due to damage of the sample by insufficient drying before Au coating.

4 are partial damage of the PMMA shells by the electron beam, distortion of the particles during microtoming, experimenter bias in looking for cores, or some combination of these factors. The acentric locations of cores seen in Figure 4 are common,4,13 resulting from immiscibility of PMMA and PS and faster growth of PMMA-rich regions than of PS-rich regions on the particle surface. Conclusions Small core/large shell PS/PMMA latexes have been synthesized by starved semicontinuous emulsion polymerization using a PS-rich copolymer seed for the core. The attempted growth of 300-nm cores to more than 800 nm resulted in second generation particles. Growth of a 70-nm core in three stages produced a PMMA shell

General Methods. Monomers were from Aldrich Chemical Co. and Fisher Scientific Co., VA-044 was from Wako Specialty Chemicals, and ruthenium tetroxide was from Polysciences. Divinylbenzene was 55% pure. Water was purified to a conductance of < 4 µohm-1 cm-1 using a three-column Barnstead e-pure system. Monomers were purified by passing them through a basic alumina column to remove the inhibitor. DLS sizes were measured using a Malvern HPPS 5001 instrument with 1 cm quartz cuvettes at 20 °C. TEM images were measured using a JEOL 100 keV microscope with Formvar coated nickel grids. SEM images were measured using a JEOL JXM 6400 microscope with samples cast onto aluminum stubs and coated with Au/Pd. Ultrafiltration was carried out using 100, 220, and 450-nm cellulose acetate/nitrate millipore filters. Non-Cross-Linked 300-nm Seed Latex 446. In a 100 mL round-bottomed flask submerged in a 65 °C oil bath and equipped with a nitrogen-purged condenser, 1.0 mL of methyl methacrylate, 3.0 mL of styrene, and 40 mL of water were stirred with a 1 in. magnetic stirring bar. After 20 min of heating, 20 mg of VA-044 initiator was added. The polymerization was carried out for 8 h. The latex was filtered through cotton and ultrafiltered for 24 h using 100-nm filters with multiple water changes. Shell Growth Latex 458. In a 250 mL two-neck round-bottomed flask equipped with a nitrogen-purged condenser and a syringe pump, 1.0 g of seed latex 446 (10 mL of dispersion) and 100 mL of water were heated and stirred with a 1.5 in. magnetic stirring bar for 10 min in a 70 °C oil bath. To the heated mixture was

Ind. Eng. Chem. Res., Vol. 44, No. 23, 2005 8541

added 150 mg of VA-044. After 10 min, the syringe pump was started and 9.0 mL of MMA, 60 mg of 3, and 60 mg of 2 were added at a rate of 4.0 mL h-1. After 3 h, the reaction was stopped and filtered through a cotton plug. Cross-Linked 70-nm Seed Latex 491. In a 250 mL round-bottomed flask submerged in an 80 °C oil bath and equipped with a nitrogen-purged condenser, 100 mL of water, 8.0 mL of styrene, 2.0 mL of methyl methacrylate, and 0.200 g of vinylbenzyl(trimethyl)ammonium chloride23 were stirred with a 1.5 in. magnetic stirring bar. After 30 min, 0.150 g of VA-044 was added. Once the mixture became turbid, 0.150 g of divinylbenzene was added. After 2 h, the mixture was cooled and filtered through cotton. Shell Growth Latex 492. Using the method of 458, 1.0 g of 491 (14 mL of dispersion), 100 mL of water, 150 mg of VA-044, 9.0 mL of methyl methacrylate, 60 mg of ethylene glycol dimethacrylate, and 60 mg of 2-(N,N-dimethylamino)ethyl methacrylate produced latex 492. Preparation of Latexes 491-494 for TEM.21 Epoxy embedded core/shell materials were prepared by drying latex samples 491-494 and dispersing them in ethanol. A small portion of the ethanol dispersion was dispersed into PolyBed 812 epoxy resin. The samples were cured overnight at 60 °C. Samples were ultramicrotomed to a thickness of 70 nm and placed on Formvar coated nickel TEM grids. The grids were placed in a Petri dish, and one drop of aqueous 0.5% RuO4 solution was placed 0.5 cm from each grid. The dish was covered and allowed to stand for 30 min in a fume hood. The excess RuO4 was removed by pipet, and the stained samples were allowed to stand overnight before TEM analysis. Acknowledgment We thank Phoebe Doss and Terry Colburg for TEM and SEM assistance and Penger Tong for valuable discussions. This research was supported by National Science Foundation Grants DMR-0102759 and EPS0132534. Literature Cited (1) Polymer Data Handbook; Mark, J. E., Ed.; Oxford University Press: New York, 1999. (2) Handbook of Chemistry and Physics, 73rd ed.; Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 1993. (3) Dullens, R. P. A.; Claesson, M.; Derks, D.; van Blaaderen, A.; Kegel, W. K. Monodisperse Core-Shell Poly(methyl methacrylate) Latex Colloids. Langmuir 2003, 19, 5963-5966. (4) Sundberg, D. C.; Durant, Y. G. Latex Particle Morphologys Fundamental Aspects: A Review. Polym. React. Eng. 2003, 11, 379-432. (5) Stubbs, J.; Karlsson, O.; Jonsson, J.-E.; Sundberg, E.; Durant, Y.; Sundberg, D. Nonequilibrium particle morphology development in seeded emulsion polymerization. 1: Penetration of monomer and radicals as a function of monomer feed rate during second stage polymerization. Colloids Surf., A 1999, 153, 255270.

(6) Joensson, J. E. L.; Hassander, H.; Jansson, L. H.; Toernell, B. Morphology of two-phase polystyrene/poly(methyl methacrylate) latex particles prepared under different polymerization conditions. Macromolecules 1991, 24, 126-131. (7) Joensson, J.-E.; Hassander, H.; Toernell, B. Polymerization Conditions and the Development of a Core-Shell Morphology in PMMA/PS Latex Particles. 1. Influence of Initiator Properties and Mode of Monomer Addition. Macromolecules 1994, 27, 1932-1937. (8) Okubo, M.; Takekoh, R.; Izumi, J. Preparation of micronsized, monodispersed, “onion-like” multilayered poly(methyl methacrylate)/polystyrene composite particles by reconstruction of morphology with the solvent-absorbing/releasing method. Colloid Polym. Sci. 2001, 279, 513-518. (9) Okubo, M.; Izumi, J. Synthesis of micron-sized monodispersed, core-shell composite polymer particles by seeded dispersion polymerization. Collloids Surf., A 1999, 153, 297-304. (10) Ito, F.; Ma, G.; Nagai, M.; Omi, S. Study of particle growth by seeded emulsion polymerization accompanied by electrostatic coagulation. Collloids Surf., A 2002, 201, 131-142. (11) Tsai, I. Y.; Kimura, M.; Russell, T. P. Fabrication of a Gradient Heterogeneous Surface Using Homopolymers and Diblock Copolymers. Langmuir 2004, 20, 5952-5957. (12) Durant, Y. G.; Sundberg, E. J.; Sundberg, D. C. Effects of Cross-linking on the Morphology of Structured Latex Particles. 2. Experimental Evidence for Lightly Cross-Linked Systems. Macromolecules 1997, 30, 1028-1032. (13) Karlsson, L. E.; Karlsson, O. J.; Sundberg, D. C. Nonequilibrium particle morphology development in seeded emulsion polymerization. II. Influence of seed polymer Tg. J. Appl. Polym. Sci. 2003, 90, 905-915. (14) Joensson, J.-E.; Karlsson, O. J.; Hassander, H.; Toernell, B. Shell-Layer Stability in Core-Shell Particles Prepared with Different Initiators. Macromolecules 2001, 34, 1512-1514. (15) Omi, S.; Fujiwara, K.; Nagai, M.; Ma, G.-H.; Nakano, A. Study of particle growth by seed emulsion polymerization with counter-charged monomer and initiator system. Collloids Surf., A 1999, 153, 165-172. (16) Lee, C.-F. The characteristic properties of poly(methyl methacrylate)/polystyrene core-shell composite polymer latex. J. Appl. Polym. Sci. 2003, 88, 312-321. (17) Okubo, M.; Shiozaki, M.; Tsujihiro, M.; Tsukuda, Y. Preparation of micron-size monodisperse polymer particles by seeded polymerization utilizing the dynamic monomer swelling method. Colloid Polym. Sci. 1991, 269, 222-226. (18) Ugelstad, J. Swelling capacity of aqueous dispersions of oligomer and polymer substances and mixtures thereof. Makromol. Chem. 1978, 179, 815-817. (19) Ugelstad, J.; Kaggerud, K. H.; Hansen, F. K.; Berge, A. Absorption of low molecular weight compounds in aqueous dispersions of polymer-oligomer particles, 2. A two step swelling process of polymer particles giving an enormous increase in absorption capacity. Makromol. Chem. 1979, 180, 737-744. (20) Wako Specialty Chemical Homepage. http://www.wakousa.com/specialty/index.html (2004). (21) Lee, S.; Rudin, A. Control of core-shell latex morphology. ACS Symp. Ser. 1992, 492, 234-254. (22) Markovic, I.; Ottewill, R. H.; Underwood, S. M.; Tadros, T. F. Interactions in concentrated nonaqueous polymer latices. Langmuir 1986, 2, 625-630. (23) Ford, W. T.; Yu, H.; Lee, J. J.; El-Hamshary, H. Synthesis of monodisperse cross-linked polystyrene latexes containing (vinylbenzyl)trimethylammonium chloride units. Langmuir 1993, 9, 1698-1703.

Received for review November 24, 2004 Revised manuscript received March 31, 2005 Accepted April 16, 2005 IE048867J