Emulsion Copolymerization of Small-Particle-Size, High-Molecular

mi-continuous emulsion copolymerization using persulfate ion initiator. ..... mine and clusters of up to thirteen molecules in higher solvents. This c...
0 downloads 0 Views 2MB Size
Chapter 14

Emulsion Copolymerization of Small-Particle-Size, High-Molecular-Weight Poly (alkylaminoalkyl methacrylate-co-alkyl methacrylate) Latexes 1,2

3

1

1,5

1

J. W. Vanderhoff , S. H. Hong , M. R. Hu , J. M. Park , I. Segall , S. Wang , and H. J. Yue

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

1

1

4

2

Emulsion Polymers Institute and Department of Chemistry, Lehigh University, Bethlehem, PA 18015 Research Directorate, U.S. Army Chemical Research, Development and Engineering Center, Aberdeen Proving Ground, MD 21010-5423 Air Products & Chemicals Company, Inc., 7201 Hamilton Boulevard, Allentown, PA 18195-1501

3

4

High-molecular-weight compositionally-homogeneous viscoelastic poly(alkylaminoalkyl methacry­ late-co-alkyl methacrylate)'s (e.g., 25:75 (mol­ ar ratio) poly(t-butylaminoethyl methacrylateco-isobutyl methacrylate)) were prepared by se­ mi-continuous emulsion copolymerization using persulfate ion i n i t i a t o r . The results were eval­ uated in terms of the expected proportionalities of the rate of polymerization to the number of polymer particles and the molecular weight to the number of p a r t i c l e s r e l a t i v e to the r a t e of radical generation. The rates of polymerization were determined gravimetrically or by gas chro­ matography. The average particle sizes were i n ­ ordinately small (18-86 nm), and the molecular weights were typically 10 daltons or greater. The small particle sizes were f i r s t attributed to the high emulsifier concentration and excellent s t a b i l i t y of the latexes, and later to the f a i l ­ ure of the latex particles containing only 8-11 polymer molecules to grow further. The high mole­ cular weights were attributed to the large number of p a r t i c l e s . The copolymer r a t i o , molecular weight, s o l u b i l i t y , and viscoelastic properties in solution were determined, and the effect of the type of i n i t i a t o r (water-soluble or o i l - s o l u ­ ble), polymerization temperature, emulsifier type and concentration, and monomer feed (emulsion or neat monomer) were investigated. The properties of laboratory copolymers were compared with those 6

5

Current address: Korea Chemical Company, Ltd.,C.R. I., San 1-9, Mabook-ri, Kuseongmyun, Yongin-kun, Kyunggi-do, South Korea 0097-6156/92/0492-0216$06.00/0 © 1992 American Chemical Society In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14. VANDERHOFF ET AL.

Emulsion Copolymerization of Latexes

217

prepared in larger reactors. Generally, the v i s ­ coelastic properties were satisfactory. Analysis of the homogeneity of the copolymers by C nu­ clear magnetic resonance spectrometry, FourierTransform infrared spectroscopy, d i f f e r e n t i a l scanning calorimetry, and transmission electron microscopy gave conflicting results.

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

1 3

The major cause of f a t a l i t i e s in a i r c r a f t crashes i s the f i r e which inevitably results when the get fuel tanks rup­ ture, and the fuel i s thrown into the a i r and ignites (1). Any organic l i q u i d can be ignited easily, provided i t i s dispersed in a i r to a sufficiently fine droplet size; how­ ever, viscoelastic polymers dissolved in jet fuel cause the breakup upon atomization to large rather than small droplets, which are much less l i k e l y to ignite. The polym­ ers to be used for this purpose must dissolve readily, yet have a hicjh enough molecular weight to produce viscoelas­ t i c solutions at low concentrations. Many different compo­ sitions can be used, but linear homogeneous copolymers of a p r o t i c electron-donating alkylaminoalkyl methacrylate with an aprotic electron-withdrawing dipolar a l k y l metha­ crylate have been shown to be particularly effective (2). Therefore, the objective of this work was to prepare linear compositionally-homogeneous viscoelastic p o l y ( a l ­ kylaminoalkyl methacrylate-co-alkyl methacrylate) with mo­ lecular weights of 4xl0 or greater and good s o l u b i l i t y in organic solvents. Emulsion polymerization is well-suited to the prepara­ tion of high-molecular-weight polymers at rapid polymeri­ zation rates. As a f i r s t approximation, the rate of polym­ erization Rp and the number-average degree of polymeriza­ tion X are given by ( 3 ) ; 6

n

Rp = k [M]RN

(1)

X

(2)

p

n

= 2k [M]RN/Ri p

where kp i s the rate constant for propagation, [M] the concentration of monomer in the particles, n the average number of free radicals per p a r t i c l e , N the t o t a l number of p a r t i c l e s , and Ri the rate of radical generation. Thus X can be increased by increasing N, increasing [M], or decreasing R i . High emulsifier concentrations give high values of N, but the value of [M] i s limited by the equi­ librium swelling volume of the latex p a r t i c l e s ; low i n i t i ­ ator concentrations and low polymerization temperatures give low values of R i . Therefore, the combination of a small latex particle size with a low i n i t i a t o r concentra­ tion and polymerization temperature would favor the prepa­ ration of high-molecular-weight polymer. Batch emulsion copolymerization gives a d i s t r i b u t i o n of copolymer compositions, according; to the disparity of the copolymerization reactivity ratios. Therefore, semicontinuous emulsion copolymerization, in which the rate of polymerization i s controlled by the rate of monomer addi­ t i o n , was used to give random compos i t ionally homogeneous copolymers. The rate of monomer addition was decreased un­ t i l i t controlled the rate of copolymerization. This meth­ od gives low values of [M], however, which favors the pre­ paration of low-molecular-weight polymer. This tendency n

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

218

POLYMER LATEXES

must be compensated for by increasing the value of N fur­ ther and thus decreasing the value of R i . Experimental Details Except where noted, the inhibitors were removed from the monomers before use; the methyl, ethyl, and isobutyl methacrylates (Rohm & Haas), ethylene glycol dimethacrylate (Rohm & Haas) , and t-butylaminoethyl and diethylaminoethyl methacrylates (Alcolac), were passed through an inhibitor removal column (Aldrich) before use. The Triton X-100, X405, and X-705 (octylphenol-polyoxyethylene adducts of i n ­ creasing chain length; Rohm & Haas), sodium persulfate i n ­ i t i a t o r (Fisher S c i e n t i f i c ) , and Wako V-044 (2,2'-azobis(N,N'-dimethyleneisobutyramidine) dihydrochloride; Wako V-50 (2,2 -azobis(2-amidinepropane) dihydrochloride), and Wako V-65 {2,2'-azobis(2,4-dimethylvaleronitrile) initia­ tors (Wako Pure Chemical Industries) were used as r e c e i ­ ved. Distilled-deionized water was used for a l l polymeri­ zations. The CM-120 polymerization recipe had been used e a r l i e r (4) for the semi-continuous polymerization of the 25:75 (molar ratio) poly(t-butylaminoethyl methacrylate-co-isobutyl methacrylate); i t comprised a 50:50 monomer/water r a t i o , 9.0% Triton X-405 emulsifier (based on water), and 0.15% sodium persulfate i n i t i a t o r ; 7% of the monomer was emulsified at room temperature in the water containing the dissolved emulsifier and potassium persulfate i n i t i a t o r in a four-necked 500-ml round-bottom flask equipped with a s t i r r e r , condenser, nitrogen i n l e t , and monomer syringe pump, heated to 60°C, and polymerized in batch to form the seed latex. The remaining monomer was then added neat at a constant rate over a four-hour period. After the monomer addition was completed, the latex was heated at 60°C to complete the polymerization. Polymerizations were carried out at 50 or 70°C, and batch polymerizations were also carried out in the same equipment. Other semi-continuous polymerizations were carried out adding an emulsion of the monomer rather than neat mono­ mer. In this case, the seed monomer was polymerized in batch using a proportionate amount of the water, emulsifi­ er, and i n i t i a t o r . Then, an emulsion of the monomer, pre­ pared using proportionate amounts of the water, emulsifi­ er, and i n i t i a t o r , was added continuously. The polymer molecular weights were determined from the i n t r i n s i c v i s c o s i t i e s (Limiting Viscosity Numbers; LVN) i n N-methyl pyrrolidone solution; the latex polymers were co­ agulated with isopropanol, f i l t e r e d , washed extensively to remove the emulsifier and other ingredients, and dried un­ der vacuum. The latex particle sizes were measured by pho­ ton correlation spectroscopy (Nicomp Submicron P a r t i c l e S i z e r ) , the copolymer ratio by C nuclear magnetic reso­ nance spectrometry (Bruker AM500) or Fourier-Transform i n ­ frared spectroscopy (Mattson Polaris), the p a r t i c l e mor­ phology by transmission electron microscopy (Philips 400), and the Tg by d i f f e r e n t i a l scanning calorimetry (Mettler). The rheological properties were measured using cone-plate geometry (Rheometrics Fluid Rheometer RFR-7800). The at­ tempts to measure the molecular weight by gel permeation chromatography were unsuccessful; the solutions f i l t e r e d slowly, and the f i l t r a t e s contained only low-molecularweight polymer. The f i n a l pH of the latexes was 9.40-9.50 except where noted.

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

,

1 3

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14. VANDERHOFF ET AL.

219

Emulsion Copolymerization of Latexes

Experimental Results and Discussion Small P a r t i c l e Size. Table I gives the molar compositions of alkylaminoalkyl methacrylate-alky1 methacrylate mix­ tures polymerized by semi-continuous polymerization (neat monomer addition) and the average p a r t i c l e sizes of the latexes produced. The number-average p a r t i c l e sizes were inordinately small (18-86 nm) . Typically, such semi-con­ tinuous polymerizations give number-average p a r t i c l e sizes of 100-150 nm. In the preparation of high-molecular-weight copolym­ ers, a volume-average particle size of 50 nm and a polymer density of 1.15 g/cm corresponds to 1 . 3 x l 0 particles/g polymer, a value about as high as i s p r a c t i c a l to achieve. Therefore, any increase in molecular weight must be made by decreasing the value of R£. Table II gives the variation with polymerization time of the number-average and volume-average p a r t i c l e size of a 25:75 poly(t-butylaminoethyl methacrylate-co-isobutyl methacrylate) latex prepared by batch copolymerization at 60°C. The p a r t i c l e size d i s t r i b u t i o n s were broad; the polydispersity indexes (PDI's) were 1.36-2.00 (average 1.59). The average particle size increased to 13 nm i n 15 minutes and to 32 nm in 20 minutes, and remained constant thereafter (average 25 nm over the range 20-420 minutes).

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

3

16

Table I. Average Particle Sizes of Copolymer Latexes Sample Composition

, \ (nm) Dl

116-27 116-17 116-15 116-26 116-14 116-21 116-23 116-24 116-19 116-28 116-20 116-22 116-25 MMA EMA iBMA tBAEMA DEAEMA

= = = = =

10/90 tBAEMA/EMA 25/75 tBAEMA/EMA 25/75 tBAEMA/EMA 50/50 tBAEMA/EMA 25/75 tBAEMA/iBMA 25/75 tBAEMA/iBMA 50/50 tBAEMA/iBMA 100/0 tBAEMA 25/75 DEAEMA/MMA 10/90 DEAEMA/EMA 25/75 DEAEMA/EMA 25/75 DEAEMA/iBMA 100/0 DEAEMA methyl methacrylate ethyl methacrylate isobutyl methacrylate t-butylaminoethyl methacrylate diethvlaminoethvl methacrylate

28 18 24 30 86 27 36 76 24 44 44 44 38

Table I I . Particle Size in Batch Polymerization Sample Polym. Time D PDI Dy n

1 2 3 4 5 6 7 8 final

(min) 15 20

30 40 60 75 90 150 420

(nm) 13 32 33 29 29 17 26 26 28

(nm) 26 45 45 43 41 34 40 41 44

2.00 1.41 1.36 1.48 1.41 2.00 1.54 1.58 1.57

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

Kl0"?6

3.1 0.88 1.5 2.0 2.2 3.9 2.5 2.4 1.9

220

POLYMER LATEXES

Similar results were observed for semi-continuous po­ lymerizations carried out at 60° and 70°C. Table III shows that the number-average particle size increased to con­ stant values of 32 nm at 60°C and 26 nm at 70°C after 20-30 minutes, which remained unchanged throughout the r e ­ mainder of the polymerization. Table IV gives the particle sizes corresponding to one or more poly(t-butylaminoethyl methacrylate-co-isobutyl methacrylate) molecules calculated assuming a molecular weight of 10 and a copolymer density of 1.15 g/cm . A single copolymer molecule i s equivalent to a sphere with a diameter of 14 nm, two molecules, a sphere with a diameter of 18 nm, etc. Thus, the 15-minute batch polymerization sample (Table II) had an average particle size correspon­ ding to a single copolymer molecule, and the succeeding samples had an average particle size of 32 nm correspon­ ding to thirteen molecules. Similarly, the 70°C semi-con­ tinuous polymerization had a seed particle size corres­ ponding to two copolymer molecules, and the succeeding samples had an average particle size of 26 nm correspon­ ding to seven molecules. Thus, the f i r s t particles i n i t i a t e d comprised single copolymer molecules. These particles grew u n t i l they com­ prised about 10 copolymer molecules and then, surprising-

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

6

3

Table III. Particle Size in Semi-Continuous Polymerization Sample Polym. Time D D PDI N/g. xlO-* (min) £nmj £nmj 60°C Polymerization Temperature 1 seed 19 27 1.42 0.58 2 20 28 39 1.39 0.20 3 40 26 36 1.38 0.31 4 60 28 41 1.46 0.29 5 90 29 41 1.41 0.78 6 120 31 44 1.42 1.4 7 140 41 52 1.27 1.1 8 175 37 50 1.35 1.2 final 420 38 51 1.34 1.2 70°C Polymerization Temperature 1 seed 18 27 1.50 0.58 2 30 28 37 1.32 1.0 3 60 21 36 1.71 2.9 4 90 24 40 1.67 2.1 5 105 25 39 1.56 2.5 6 130 19 35 1.84 3.7 7 160 34 45 1.32 1.8 final 420 34 46 1.35 1.7 n

v

Table IV. Spherical Particle Sizes of Copolymer Molecules Number of Copolymer Molecules Spherical Size turn) 1 14 2 18 3 20 5 24 10 30 20 38

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

6

14. VANDERHOFF ET AL.

Emulsion Copolymerization of Latexes

221

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

ly, stopped growing. This cessation of p a r t i c l e growth at an early conversion i s most unusual. Usually, the p a r t i ­ cles continue to grow u n t i l the supply of monomer or free radicals i s exhausted. If particle nucleation ceased early in the polymerization, the f i n a l latex would be of uniform size; i f the particle nucleation continued for a longer time, the f i n a l latex would have a broad p a r t i c l e size d i s t r i b u t i o n . The particle size distributions of these l a ­ tex samples were relatively broad; the average polydisper­ s i t y indexes were 1.54 for the batch polymerization (Table II) and 1.22 for the 70°C semi-continuous polymerization (Table I I I ) . Hypothesis f o r Cessation of Particle Growth. The reason for the cessation of particle growth at ten molecules i s not known. However, one possible explanation comprises a reversal of the mode of adsorption of the oligomeric s u l ­ fated radicals on the latex particle surface. The commonly accepted mechanism of i n i t i a t i o n using persulfate ion i n i t i a t o r (5.6) comprises the formation of sulfate ion-radicals in the aqueous phase, their slow pro­ pagation in the aqueous phase u n t i l they become surfaceactive and adsorb at the particle-water interface with the radical end oriented toward the monomer-polymer phase. The rate of propagation of these radicals i s slow in the aque­ ous phase because of the low concentration of monomer and rapid in the monomer-polymer phase because of the high concentration of monomer. In the present system, the i n i t i a l growth of the par­ t i c l e s would give an increasing number of positive amino surface groups from the t-butylaminoethyl methacrylate units, so that an acid-base interaction between the sur­ face amino groups and the sulfate groups of the oligomeric sulfated radicals would give a reversed mode of adsorp­ t i o n , with the sulfate end oriented toward the p a r t i c l e surface and the radical end oriented toward the aqueous phase. The radicals oriented toward the p a r t i c l e phase would grow into the particle rapidly because of the high concentration of monomer inside the p a r t i c l e s . In con­ t r a s t , the r a d i c a l s oriented toward the aqueous phase would grow only slowly because of the low concentration of monomer in the aqueous phase; moreover, they would be v u l ­ nerable to termination by oligomeric radicals or primary radicals from the aqueous phase. Termination of the adsor­ bed radicals by radicals from the aqueous phase would i n ­ h i b i t further polymerization inside the p a r t i c l e . Mean­ while, the nucleation of new particles would proceed con­ currently, so that particles would be nucleated, grow to a given size and then stop growing. This cessation of p a r t i ­ cle growth i s a new phenomenon not reported e a r l i e r to the authors' knowledge; therefore, this phenomenon i s termed limited growth emulsion polymerization. This mechanism would give a surface layer of different composition from the particle i n t e r i o r , which would be consistent with the indications of inhomogeneity i n the p a r t i c l e s observed by transmission electron microscopy, which w i l l be described later. Other Works. Other workers have reported the preparation of small-particle-size latexes, which grew slowly or not at a l l . Medvedev et a l . (7) prepared polysytrene latexes

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

222

POLYMER LATEXES

with diameters as small as 60 nm by batch emulsion polym­ erization of styrene using nonionic fatty alcohol-polyoxyethylene adducts as emulsifiers and oil-soluble benzoyl peroxide or 2,2 -azobisisobutyronitrile i n i t i a t o r s . These small p a r t i c l e size latexes were also reported to remain constant in diameter with increasing conversion. The for­ mation of these small particles was attributed to the f o r ­ mation of tiny emulsion droplets and their polymerization "as is" without transfer of monomer from one to another by diffusion through the aqueous phase. There was no indica­ tion that the particles stopped growing at any particular stage. Such a mechanism i s probably not operative in the present case because i t does not account for the cessation of p a r t i c l e growth. Also, Snuparek and Kleckova (8) found that the semicontinuous emulsion polymerization of acrylate ester/ethylene glycol dimethacrylate mixtures using monomer emul­ sion feed gave average particle sizes that o s c i l l a t e d be­ tween 200 and 600 nm with increasing conversion. This os­ c i l l a t i o n was attributed to competing processes of p a r t i ­ cle growth and particle flocculation. Similar o s c i l l a t i o n s have been observed in continuous emulsion polymerization, where they were attributed to the effect of emulsifier concentration on the number of particles nucleated r e l a ­ tive to the competing particle flocculation process (9. 10) . This o s c i l l a t i o n does not explain the present r e ­ sults, however, where the particles once nucleated grow to a given size and then stop growing, and where there i s l i t t l e or no tendency for the particles to flocculate.

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

/

Copolymer Molecular Weight. The molecular weights of the copolymers as measured by the i n t r i n s i c viscosity (Limi­ ting Viscosity Number; LVN) were a l l high, presumably be­ cause of the large numbers of particles and the r e l a t i v e l y low rate of radical generation. Generally, the molecular weights varied as expected with the number of particles and the rate of radical generation, increasing with i n ­ creasing number of particles and decreasing rate of r a d i ­ cal generation (low i n i t i a t o r concentration; low polymeri­ zation temperature). The molecular weight was s l i g h t l y higher when the monomer was added in the form of an emul­ sion as compared to when i t was added neat; this was at­ tributed to the addition of the i n i t i a t o r dissolved in the emulsion for the emulsion addition as compared to addition in the beginning of the reaction for the neat addition, the former giving a lower concentration of i n i t i a t o r at a given time and hence a lower rate of radical generation. Generally, however, the variation of molecular weight with these parameters, which had been well-established for a different system investigated e a r l i e r , was not as great as expected. Therefore, further experiments were carried out. Table V shows that there were no strong trends in the variation of particle size and molecular weight with sodi­ um persulfate i n i t i a t o r and Triton X-405 emulsifier con­ centrations for the batch emulsion copolymerization of 25: 75 poly(t-butylaminoethyl m e t h a c r y l a t e - c o - i s o b u t y l methacrylate); however, the averages of the parameters showed that the particle size increased with temperature from 28.9 to 33.7 to 41.6 nm for 50, 60, and 70°C, the width of the distribution decreased (the polydispersity index decreased from 1.65 to 1.48 to 1.29), and the Limi-

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14. VANDERHOFF ET AL.

Emulsion Copolymerization of Latexes

223

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

Table V. Effect of Emulsifier and Initiator Concentration Sample [SPS] [405] LVN (dl/a) m m (nm) 50°C Polymerization Temperature 50#1 50#2 50#3 50#4 50#5 50#2 50#6

0.09 9.0 25 0.15 9.0 28 0.27 9.0 27 0.15 4.5 36 0.15 5.8 31 0.15 9.0 28 0.15 13.5 27 60°C Polymerization Temperature

1.35 1.27 1.50 1.29 1.37 1.27 1.32

60#1 60#2 60#3 60#4 60#5 60#2 60#6

0.09 9.0 28 0.15 9.0 35 0.27 9.0 35 0.15 4.5 36 0.15 5.8 37 0.15 9.0 35 0.15 13.5 30 70°C Polymerization Temperature

1.04 1.39 1.32 1.25 1.20 1.39 1.30

70#1 0.09 9.0 70#2 0.15 9.0 70#3 0.27 9.0 70#4 0.15 4.5 70#5 0.15 5.8 70#2 0.15 9.0 70#6 0.15 13.5 [SPS] = sodium persulfate concentration T4051 = Triton X-405 concentration

43 39 33 63 44 39 30

1.44 1.64 1.61 1.58 1.40 1.64 1.43

ting Viscosity Number remained about the same (1.34 to 1.27 to 1.53). Thus, the average particle size and molecu­ lar weight were not sensitive to temperature i n the 5070°C range. The slight variation of p a r t i c l e size was i n the opposite direction to that expected, and the molecular weight did not decrease with increasing temperature as ex­ pected . The molecular weights of the copolymers were generally high, although the correlation between the Limiting V i s ­ cosity Number and the number-average or weight-average mo­ lecular weight has not yet been established. The best es­ timate i s that a LVN of ca. 2 d l / g i s equivalent to number average and weight-average molecular weights of l x l O and 4xl0 daltons, respectively. The copolymer solutions i n N-methyl pyrrollidone prepared for the LVN measurements (as well as those i n tetrahydrofuran prepared for gel per­ meation chromatography) were transparent, and without v i s ­ i b l e gels; however, the response of the solutions to shak­ ing suggested that they were not homogeneous, but compri­ sed swollen gels. Light scattering (11.12) showed that the copolymers associated in solution to give an average ag­ gregate or cluster size of four molecules i n isopropylamine and clusters of up to thirteen molecules i n higher solvents. This clustering suggested the p o s s i b i l i t y that the particles were not dissolved completely i n any of the solvents. Thus, further work i s required to demonstrate the homogeneity of the solutions. 6

6

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

224

POLYMER LATEXES

The Presence of Storage Inhibitor. Except where noted, a l l experiments of this work used monomer from which the stor­ age inhibitor had been removed; however, since the usual industrial practice i s to use monomer containing i n h i b i ­ tor for safety reasons, laboratory experiments using i n h i ­ bited monomer were carried out preparatory to the larger scale polymerizations. Earlier experience with inhibited monomer gave lower molecular weights and, in some cases, slower polymerization rates, according to the residual i n ­ h i b i t o r concentration. Table VI gives the results of experiments based on the CM-120 recipe polymerized at 60°C using monomer contain­ ing i n h i b i t o r . Generally, the molecular weight decreased with decreasing i n i t i a t o r concentration, decreased with slower monomer feed rates, and decreased with decreasing isobutyl methacrylate concentration in the monomer mix­ ture. Table VI. Effect of Polymerization Parameters [SPS] Feed Rate [tBAEMA] LVN Dn (dl/a) m (cc/min) (nm) 412 0.18 2.0 25 44 1.78 425A 0.18 1.1 21 41 1.90 425B 0.13 2.0 21 43 1.90 418 0.18 1.1 25 40 1.60 426A 0.13 1.1 21 46 1.73 430A 0.13 1.1 18 40 2.08 430B 0.13 0.5 21 45 1.35 426B 0.09 1.1 21 31 1.38 502B 0.09 0.5 18 40 1.32 418-S* 0.18 1.1 25 36 2.04 306-B4** 0.18 1.1 25 55 1.60 * = inhibitor-free monomer ** = batch polymerization; inhibitor-free monomer ftBAEMA1 = t-butvlaminoethyl methacrylate concentration Sample

Larger-Scale Polymerizations. Table VII gives the results of the larger-scale polymerizations. In the laboratory, the standard CM-120 recipe gave a particle size of 36 nm and a Limiting Viscosity Number of 2.04 d l / g . The corres­ ponding batch polymerization gave a larger p a r t i c l e size (54 nm) and a lower LVN (1.60 d l / g ) . Eight f i v e - l i t e r po­ lymerizations of the standard CM-120 recipe gave excellent reproducibility but a larger particle size (average 52; range 50-55 nm) and a lower LVN (average 1.68; range 1.471.90 d l / g ) , as expected. The use of monomer containing inhibitor gave s l i g h t l y lower Limiting Viscosity Numbers (1.90, 1.85 d l / g ) . The Morton laboratory polymerizations preparatory to the 100-gallon polymerization using i n h i b i ­ ted monomer gave lower Limiting Viscosity Numbers (0.701.53 d l / g ) , but the Morton 100-gallon polymerization gave a higher value (1.78 d l / g ) . Spray-drying the latex to r e ­ cover the polymer lowered the LVN; the decrease was great­ er for the higher temperature (300°F) of spray-drying. The t-butylaminoethyl methacrylate concentrations in the copo­ lymers measured by 1 C nuclear magnetic resonance spectro­ metry were smaller (average 20.1; range 17.5-21.5) than the 25% charged. 3

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14. VANDERHOFF ETAL.

Table VII. Larger-Scale Polymerizations LVN D (dl/a) (nm) EPI-315 (CM-120) 36 2.04 EPI-306 batch 54 1.60 EPI-423A* 8% seed 36 1.90 EPI-424A* 12% seed 39 1.90 Sample

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

S5-01 S5-02 S5-03 S5-04 S5-05 S5-06 S5-07 S5-08

n

5-liters 5-liters 5-liters 5-liters 5-liters 5-liters 5-liters 5-liters

Morton 278* Morton 287* Morton 288*

225

Emulsion Copolymerization of Latexes

[tBAEMA] m

50 50 53 51 52 55 53 53

1.47 1.58 1.96 1.75 1.78 1.50 1.72 1.73

20.9 21.5 19.9 21.3 17.8 20.6



1.00 1.53 0.70

- — -



Morton 100-gallons* 39 — spray-dried 250 — spray-dried 300 — room temo * = inhibited monomer

21.0 20.7 17.5

1.78 1.49 1.19 1.45

Miscellaneous Polymerizations. Table VIII shows that the water-soluble cationic Wako V-044 and Vazo V-50 i n i t i a ­ tors, and the oil-soluble Vazo V-65 i n i t i a t o r , gave lower Limiting Viscosity Numbers than the sodium persulfate i n i ­ t i a t o r . The oil-soluble i n i t i a t o r gave a s l i g h t l y larger average particle size. The substitution of Triton X-100, Triton X-705, and Triton X-405/sodium lauryl sulfate mix­ ture gave lower Limiting Viscosity Numbers than the Triton X-405 alone. Ethylene glycol dimethacrylate, which was added in 0.05% concentration to introduce branches into the copolymer and thus increase the Limiting Viscosity Number, actually gave a lower value. Homogeneity of the Copolymers. Homogeneous copolymers of random comonomer sequence distribution were desired for this application. The most widely used method to determine comonomer sequence distribution i s C nuclear magnetic resonance (NMR) spectrometry, which gives resonance s i g ­ nals corresponding to the diads and triads of the copolym­ er. The homogeneity of the copolymer can be calculated from the intensity of these resonance signals. The C NMR spectra of the 25:75 poly(t-butylaminoethyl methacrylateco-isobutyl methacrylate) showed no significant changes in the resonance signals compared with the spectra of the two homopolymers, which indicated that no information on the comonomer sequence distribution was obtained. This lack of signals was not unexpected because the backbone structures of the poly(t-butylaminoethyl methacrylate) and poly(isobutyl methacrylate) were the same; the side groups were different, but were located far enough from the carbon backbone skeleton that their electronic energy was insuf­ f i c i e n t to induce s p l i t t i n g for the corresponding diad or t r i a d resonances. Thus the C NMR spectrum for the pro1 3

1 3

1 3

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

226

POLYMER LATEXES

Table VIII, Miscellaneous Polymerizations Sample

LVN

Wako V-044 (pH 9.55) Wako V-50 (pH 9.30) Wako V-65 (pH 9.00) Triton Triton Triton Triton

142 39 51 50

0.99 1.24 0.90 0.91



1.12

X-100 X-405 X-705 X-405/sodium lauryl

E6DMA f0.050%)

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

(dl/a)

(nm) 46 42 87

sulfate

0.79 1.01 1.42

duct was essentially the same as the superposed spectra of the two homopolymers, and the product was a mixture of the two homopolymers or, more l i k e l y , a blocky copolymer. Figure 1 shows a subtle difference, however, between the C NMR spectrum of the 25:75 poly(t-butylaminoethyl methacrylate-co-isobutyl methacrylate) and those of the two homopolymers and a physical mixture of the two. The peak at 43.6 ppm in the poly (t-butylaminoethyl methacry­ late) spectrum corresponding to the -CH2-NH- structure was also observed in the spectrum of the copolymer, but was of lower height and shifted. This peak was attributed to the association of secondary amines with carbonyls; i t was not observed at a l l i n the p h y s i c a l mixture of the two homopolymers. This suggests that the morphology of the co­ polymer was different from that of the homopolymer mix­ ture, as expected. That the polymer comprised copolymers was supported by Fourier-Transform infrared spectroscopy. The secondary amines show a stretching mode at 3400-3500 wavenumbers and a bending mode at 1500-1600 wavenumbers. The association of the amines with the carbonyls shifts the peak for the amine bending mode upfield and that for the amine stretch­ ing mode downfield. Figure 2 compares the spectrum of the 25:75 poly(t-butylaminoethyl m e t h a c r y l a t e - c o - i s o b u t y l methacrylate) with that for a mixture of the two homopo­ lymers. The expected shifts for the amines are seen in the spectrum for the physical mixture but not in that for the copolymer; this result was attributed to the association of the amines with the carbonyls in the copolymer but not in the physical mixture. The transmission electron microscopy of these latexes gave marginal results because of the small size and poor contrast of these particles in the electron beam. Some mi­ crographs of stained particles, however, showed images of heterogeneous particles, consistent with the presence of two homopolymers; others showed homogeneous p a r t i c l e s . F i ­ gure 3 shows such an electron micrograph of heterogeneous p a r t i c l e s . These findings support the f i r s t conclusion of the 1 C NMR spectrometry that the particles comprised a blocky copolymer or a mixture of two homopolymers; how­ ever, the fact that each particle comprised only a few co­ polymer molecules suggests that these molecules can a r ­ range themselves so that some of the polar amine function­ a l groups are together, e.g., to form a polar s h e l l around a nonpolar core, and that this morphology gave the heter­ ogeneous appearance observed in a few samples. 1 3

3

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

Emulsion Copolymerization of Latexes

111

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

14. VANDERHOFF ET AL.

3

Figure 1. 1 C nuclear magnetic resonance spectra of the 25:75 poly(t-butylaminoethyl m e t h a c r y l a t e - c o - i s o b u t y l methacrylate), the two homopolymers s i c a l blend o f the two.

(A and B), and a phy-

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

POLYMER LATEXES

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

228

F i g u r e 2. F o u r i e r - T r a n s f o r m i n f r a r e d s p e c t r a of the 25:75 poly(t-butylaminoethyl methacrylate-co-isobutyl methacry­ l a t e ) and t h e p h y s i c a l b l e n d o f t h e two h o m o p o l y m e r s .

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14. VANDERHOFF ET AL.

229

Emulsion Copolymerization of Latexes

In contrast, d i f f e r e n t i a l scanning calorimetry of the purported copolymers suggested that they were copolymers. Table IX gives the values of the apparent second-order transition temperature Tg of these samples, and Figure 4 shows a t y p i c a l heat flow-temperature curve. A l l copolym­ ers gave single incisive T Q values of 60-63°C compared to the values of 3 8 ° and icfoc for poly (t-butylaminoethyl methacrylate) and poly(isobutyl methacrylate), respective­ l y , consistent with homogeneous copolymers. Table IX. Apparent Second-Order Transition Temperature Sample Composition T a

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

(°2) 116-44

poly(isobutyl methacrylate)

70

116-55 140-61 140-62 140-64 140-66 140-71 140-73

25:75 25:75 25:75 25:75 25:75 25:75 25:75

63 63 63 62 62 60 60

140-72

poly(t-butylaminoethyl methacrylate)

copolymer copolymer copolymer copolymer copolymer copolymer copolymer

P h y s i c a l blend of 116-44 and 140-72

38 38.70

F i n a l l y , semi-continuous polymerization should give homogeneous copolymers of t-butylaminoethyl methacrylate and isobutyl methacrylate because the copolymerization r e a c t i v i t y ratios r i = 1.2 and T2 = 0.8 (calculated from the Q and e values) were not too disparate. In contrast, i t was shown earlier (13.14) that, for the poly(vinyl ace­ tate-co-butyl acrylate) system, which has far more dispar­ ate copolymerization reactivity ratios of r i = 0.05 and ro = 5, batch copolymerization gave a mixture of a butyl aerylate-rich copolymer with polyvinyl acetate, indepen­ dent of the monomer r a t i o , whereas semi-continuous polym­ erization gave a single-peaked distribution of copolymer compositions, the Tg of which varied according to the mo­ nomer r a t i o . Thus, semi-continuous polymerization gave a homogenous copolymer of a monomer pair with far more d i s ­ parate copolymerization reactivity ratios than the 25:75 poly(t-butylaminoethyl methacrylate-co-isobutyl methacry­ late) . Therefore, i t i s expected that homogeneous copolym­ ers would be formed in the present system. In summary, the single Tg values of the polymers, the spectroscopic evidence for the association of the secon­ dary amine groups with the carbonyl groups of the C nu­ clear magnetic resonance and Fourier-Transform infrared spectra, and the use of semi-continuous polymerization suggested that the polymers were indeed homogeneous copo­ lymers. In contrast, absence of peaks for the diads and triads of the C nuclear magnetic resonance spectra and the heterogeneous appearance of the p a r t i c l e s i n some transmission electron micrographs suggested that the po­ lymers are mixture of homopolymers. Further work i s r e ­ quired to resolve this apparent discrepancy; however, on 1 3

1 3

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

230

POLYMER LATEXES

Figure 3. Transmission electron micrograph of Sample 11655 25:75 poly (t-butylaminoethyl methacrylate-co-isobutyl methacrylate). Temperature °C

H

e

a

t

F

l

o

w

Exothermal—>

Figure 4. D i f f e r e n t i a l scanning calorimetry heat flowtemperature scan of 25:75 poly(t-butylaminoethyl methacry­ late-co-isobutyl methacrylate).

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14. VANDERHOFF ET AL.

231

Emulsion Copolymerization of Latexes

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

balance, the formation of homogeneous copolymers i s consi­ dered to be the more l i k e l y result. Rheological Properties of Dilute Solutions. Table X shows, for dilute solutions of the larger-scale samples, the var­ iation of the apparent viscosity and f i r s t normal stress difference (FNSD)) with shear rate. The values of the ap­ parent viscosity of the f i v e - l i t e r samples were generally within the acceptable range, with the exception of sample S5-03, which showed an unexpectedly high apparent v i s c o s i ­ ty, but a FNSD value in the acceptable range; this sample also had the highest LVN value of the f i v e - l i t e r samples. The cause of this aberration is not known. The FNSD values of a l l f i v e - l i t e r samples indicated that these samples had a high enough viscoelasticity in the given solvents. Both the apparent viscosity and the FNSD value of the Morton 100-gallon sample were almost identical to those of the o r i g i n a l CM-120 sample, which indicated that the 100-gal­ lon polymerization was quite successful. Table X. Rheological Properties of Dilute Solutions Sample S o l Concn. FSND Apparent Viscosity vent (dyne/cm ) (poise) 2

(a/dl)

CM-120 Mix 1 Mix 2 S5-01 S5-02 S5-03 S5-04 S5-05 S5-06 S5-07 S5-08

Mix Mix Mix Mix Mix Mix Mix Mix Mix Mix Mix Mix Mix Mix Mix Mix

2.0* 3.0 3.0 3.1 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2

2.0 3.0 2.0 3.0 2.1 3.0 2.1 3.1 2.1 3.0 2.1 3.1 2.0 3.0 2.0 2.9

Shear Rate/sec 100 250 500 1000 155 530 1215 702 2296 204 1722 (68) 464 1232 2891 427 (74) 488 (39) 445 181 520 125 115 181 (40)

878 1220 350 229 476 252 731 971 1489 364 1009 1062 1857 282 857 853 1391 324 893 351 724 193 373 766 (87) 544 608 1115 435 1111

1329 497 570 1663 2043 2310 2760 2055 1767 2058 988 928 1148 1399 1399 2411

Shear Rate/sec 12.5 25 50 99 2.1 2.0 1.9 1.7 4.9 3.2 2.8 2.4 2.8 2.7 2.5 2.3 3.9 3.4 2.9 2.5 2.3 1.7 1.9 3.4 9.0 5.9 4.9 4.2 2.9 3.8 2.0 3.5 2.2 3.6 2.7 4.0

2.0 1.6 1.8 2.9 7.1 4.8 4.3 3.6 2.6 3.4 1.9 2.9 2.0 3.1 2.4 3.5

1.6 1.5 1.6 2.4 5.6 4.1 3.7 3.1 2.3 2.8 1.7 2.7 1.8 2.7 2.1 3.0

1.4 1.3 1.4 2.2 4.0 3.5 3.0 2.7 2.0 2.5 1.6 2.4 1.6 2.3 1.9 2.6

Morton Mix 1 2.0 150 511 1034 1555 2.5 2.3 2.1 1.8 100-gal Mix 2 3.0 280 824 1971 3.5 2.9 2.5 2.3 * estimated values from measurements of 2 g / d l solutions at 20, 30, and 40OC FNSD = f i r s t normal stress difference Mix 1. Mix 2 = proprietary organic solvent mixtures Conclusions High-molecular-weight 25:75 poly(t-butylaminoethyl meth­ acrylate-co-isobutyl methacrylate) was prepared by semi-

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

POLYMER LATEXES

232

continuous emulsion copolymerization. The number-average and weight-average molecular weights measured by Limiting Viscosity Number were ca. l x l O and 4xl0 daltons, respec­ t i v e l y , and the viscoelastic properties of the solutions were satisfactory; however, the polymer molecules appeared to associate in solution, forming clusters of up to t h i r ­ teen molecules in some solvents, and the molecular weights did not vary as sensitively with particle number and rate of r a d i c a l generation as expected. Moreover, the latex p a r t i c l e sizes were inordinately small (18-86 nm) instead of the expected 100-150 nm; the small sizes were a t t r i b u ­ ted to the rapid nucleation and growth of the particles to a size equivalent to about 10 copolymer molecules, follow­ ed by cessation of growth. The compositional homogeneity of the copolymers has not yet been demonstrated unequivo­ c a l l y : transmission electron gave evidence of heterogene­ ous p a r t i c l e s , and C nuclear magnetic resonance spectro­ metry showed no peaks for the expected diads and t r i a d s ; however, d i f f e r e n t i a l scanning calorimetry gave single i n ­ c i s i v e Tg's of the expected values, both C nuclear mag­ netic resonance spectrometry and Fourier-Transform i n f r a ­ red spectroscopy gave evidence for the association of the secondary amine groups with the carbonyl groups for the copolymers but not for the physical blend of the two homo­ polymers, and the semi-continuous polymerization method used has given homogeneous copolymers for other systems with more disparate copolymerization reactivity r a t i o s . 6

6

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

1 3

1 3

Literature Cited 1. Videocassette, Why Planes Burn. PBS, WGSA Educational Foundation, NOVA, 1988. 2. Chu, B.; Wang, J.; Pfeiffer, D. G.; Shuely, W. J . Ma­ cromolecules 1991, 24, 809. 3. Smith, W. V . ; Ewart, R. H. J . Chem Phys. 1948, 16, 592. 4. Hong, S. H. Review of Emulsion Polymerization: Optimi­ zation for Production of Ultrahigh Molecular Weight Poly(alkylaminoalkyl Methacrylate-co-Alkyl Methacry­ late), U.S. Government Report CRDEC-TR-88146, UNCLAS­ SIFIED Report, 1988. 5. van der Hoff, B. M. E. Polymerization and Polycondensation Processes. ACS Adv. Chem. Ser. 1962, 34, 6. 6. Vanderhoff, J . W.; van den Hul, H. J.; Tausk, R. J . M.; J . Th. G. Overbeek In Clean Surfaces: Their Prepa­ ration and Characterization for Interfacial Studies. Goldfinger, G . , Ed., Marcek Dekker: New York, NY, 1970, p. 15. 7. Medvedev, S. S.; Zuikov, A. V.; Gritskova, I. A . ; Dudukin, V. V. Vysokomol. Soyed (English transl.) 1971, A13, 1397. 8. Snuparek, J r . , J.; Kleckova, Z. J . Appl. Polym. Sci. 1984, 29, 1. 9. Gershberg, D. B.; Longfield, J . E. Symposium on Polym­ erization Kinetics and Catalyst Systems. 45th AIChE Meeting, New York, NY, 1961, preprint 10. 10. Greene, R. K.; Gonzales, R. A.; Poehlein, G. W. In Emulsion Polymerization. Piirma, I . , Ed., ACS Symp. Ser. 1976, paper 22. 11. Chu, B.; Wang, J.; Shuely, W. J . Macromolecules 1990, 23, 2252.

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14. VANDERHOFF ET AL.

Emulsion Copolymerization of Latexes

233

12. Chu, B.; Wang, J.; Shuely, W. J . Polymer 1990, 31, 805. 13. El-Aasser, M. S.; Misra, S. C.; Pichot, C . ; Vander­ hoff, J . W. J . Polym. Sci., Polym. Lett. Ed. 1979, 127, 567. 14. Misra, S. C.; Pichot, C.; El-Aasser, M. S.; Vander­ hoff, J . W. J . Polym. Sci., Poly. Chem. Ed. 1983, 21, 2382.

Downloaded by CORNELL UNIV on August 6, 2012 | http://pubs.acs.org Publication Date: June 3, 1992 | doi: 10.1021/bk-1992-0492.ch014

RECEIVED January 24, 1992

In Polymer Latexes; Daniels, E., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.