Monomer Complexes in Aqueous Polymerizations

Polymerization was conducted under ultrapure N2 atmosphere (5−8 psi) and allowed to ... A 10 mL sample of either CMA, 2EHMA, or tBuMA was added to a...
0 downloads 0 Views 388KB Size
Biomacromolecules 2000, 1, 615-621

615

Carbohydrate/Monomer Complexes in Aqueous Polymerizations: Methylated-β-cyclodextrin Mediated Aqueous Polymerization of Hydrophobic Methacrylic Monomers Phillip H. Madison and Timothy E. Long* Department of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061-0212 Received May 26, 2000; Revised Manuscript Received September 19, 2000

Hydrophobic methacrylic monomers were polymerized in aqueous media using methylated (1.8)-βcyclodextrin (MeCD) additives. Hydrophobic monomers tert-butyl methacrylate (tBuMA), cyclohexyl methacrylate (CMA), and 2-ethylhexyl methacrylate (2EHMA) were each dissolved in chloroform with MeCD. Chloroform was then evaporated to yield solid monomer/cyclodextrin complexes. Complexes were shown by 1H NMR and thermogravimetric analysis (TGA) to have molar ratios of monomer to MeCD as high as 0.72/1.00. The water-soluble complexes were readily polymerized in aqueous media using free radical initiation. During polymerization, hydrophobic methacrylic polymers precipitated and the majority of MeCD remained in solution. Poly(alkyl methacrylates) synthesized via this method exhibited numberaverage molecular weights ranging from 50 000 to 150 000 with polydispersities from 3.2 to 5.5 depending on monomer structure, and isolated yields were as high as 86%. Additionally, corresponding methacrylic/ carbohydrate films were prepared and examined. High molecular weight poly(tBuMA), poly(CMA), and poly(2EHMA) were blended with MeCD to produce optically clear films with as high as 20 wt % MeCD. Differential scanning calorimetry (DSC) characterization indicated that the glass transition temperatures of these novel carbohydrate blends were controllable over a 20 °C range depending on the relative concentration of each component. Introduction The demand for environmentally benign processes is growing as society becomes increasingly aware of environmental issues involving conventional organic solvents. The chemical industry is encouraged to look for new means to the same end for many of its traditional processes that either produce environmentally unfriendly industrial products or result in toxic byproducts.1,2 In an effort to overcome such potential obstacles with minimal expense, research is directed toward the replacement of traditional organic solvents with environmentally benign compounds such as carbon dioxide,3 biomolecules,4 and water. Carbohydrate mediation offers a viable alternative to organic solvents. Recent efforts have focused on the incorporation of carbohydrates in polymer backbones5 or in polymer side chains.6-11 Cyclodextrins are a family of cyclic carbohydrates that have received a great deal of industrial attention12 and are well-known to solublize hydrophobic compounds in aqueous media.13,14 This attribute has been utilized in several areas including stereoselective synthesis,15 dual phase catalysis,16 water purification,17-19 and rotaxane chemistry.20-22 Ritter et al. have reported the CD mediated aqueous polymerizations of a limited number of monomers including phenyl and cyclohexyl methacrylates,23 N-methacryloyl-11aminoundecanoic acid and N-methacryloyl-1-aminononane,24,25 and isobornyl and n-butyl acrylates.26 They have also calculated reactivity ratios (r) for the copolymerization of

two complexes26 and proposed a polymerization mechanism that involves the dethreading of CD as the radical chain propagates.24 However, preliminary polymerization studies involving cyclohexyl methacrylate in the presence of MeCD did not include a discussion of attainable molecular weights or molecular weight distributions.23 Also, Rimmer and coworkers have recently examined the emulsion polymerization of n-butyl methacrylate in the presence of β-CD, without prior cyclodextrin-monomer complexation.27 However, the reaction temperature dependency and a systematic study of the effect of the ester alkyl group in methacylic/MeCD complexes on polydispersity and attainable molecular weights have not been addressed earlier. This report describes the aqueous polymerization of tertbutyl methacrylate, cyclohexyl methacrylate, and 2-ethylhexyl methacrylate mediated with methylated (1.8)-β-CD in order to determine the effects of monomer structure, polymerization temperature, and initiator concentration on the polymerization process, molecular weight, and molecular weight distribution of the final products. These monomers are of particular interest because both tBuMA and 2EHMA are commercially important acrylic monomers utilized widely in microlithographic processes, and adhesive applications, respectively. The blending of methylated (1.8)-β-CD with various poly(alkyl methacrylate) homopolymers to form novel, optically clear carbohydrate/acrylic films will also be described, and the preliminary thermal properties of these films will be presented.

10.1021/bm0055706 CCC: $19.00 © 2000 American Chemical Society Published on Web 11/01/2000

616

Biomacromolecules, Vol. 1, No. 4, 2000

Experimental Section Purification. All monomers and initiators were obtained from Aldrich. MeCD was generously donated by WackerChemie and was used as received. All monomers were vacuum distilled (0.5 mmHg) from CaH2 using the freezethaw method to degas the monomer and stored at -25 °C under nitrogen. Potassium persulfate was used as received, and stored at -25 °C. Complexation. Complexations of tBuMA, 2-EHMA, and CMA were preformed in 200 mL of chloroform with concentrations of approximately 0.04 M of both MeCD and monomer. Approximately 10.0 g (7.6 mmol) of MeCD and 1 molar equiv of monomer were added to 200 mL of chloroform in a 500 mL Erlenmeyer flask. The solution was stirred for 1-6 days at room temperature. Chloroform was then removed using a rotoevaporator and the complex was dried in vacuo at 70 °C for 24 h. Isolated yields ranged from 90 to 96% and ratios of monomer to MeCD ranged from 0.50/1.00 to 0.75/1.00. Aqueous Homopolymerizations. Homopolymerizations of the complexes were conducted in a 50 mL 1-necked, round-bottomed, flask with 20.0 mL of deionized water, 1.00-9.00 wt % K2S2O8 compared to monomer, and 2.00 g of complex. The polymerization reactor was then septum capped and sparged with ultrapure nitrogen for 5-10 min prior to initiator addition. Polymerization was conducted under ultrapure N2 atmosphere (5-8 psi) and allowed to proceed for approximately 24 h. The precipitated polymer was filtered and dried in vacuo for 18 h at 60 °C. MeCD was recovered by removing the water with a rotoevaporator and drying in vacuo for 18 h at 80 °C. Bulk Homopolymerizations. A 10 mL sample of either CMA, 2EHMA, or tBuMA was added to a 50 mL 1-necked, round-bottomed flask with 3.00 mg of BPO dissolved in 1.0 mL of THF. The polymerization reactor was sealed with a septum and sparged with ultrapure nitrogen to eliminate oxygen. The reactor was then heated under nitrogen to 70 °C and stirred for 24 h. The product was cooled and THF was stirred over the viscous polymer. Poly(tBuMA) was subsequently precipitated from THF into 1:1 methanol:water (10× compared to polymer solution). Poly(CMA) and poly(2-EHMA) were precipitated from THF into methanol due to their more nonpolar nature. Methylated (1.8)-β-Cyclodextrin/Polymer Blends. Poly(tBuMA)/MeCD, poly(2-EHMA)/MeCD, and poly(CMA)/ MeCD blends were produced by dissolving 1.000 g of polymer and 0.050-0.200 g of MeCD in 5.0 mL of chloroform. The clear solutions were then poured into Petri dishes and covered. The solvent was allowed to evaporate slowly and the films were dried in vacuo for 24 h at 70 °C. Characterization. Molecular weights and molecular weight distributions were determined in NMP/P2O5 (0.02 M) at 60 °C with a Waters Gel Permeation Chromatograph (GPC) equipped with an external 410 RI detector and Viscotek 150 R viscometer using a flow rate of 1.0 mL/min. DSC analysis was preformed using a Perkin-Elmer Pyris 1 under N2 atmosphere, from 25 to 200 °C at a heating rate of 10 °C/ min. All reported glass transition temperatures (Tg) were

Madison and Long

based on the second heat profile. All NMR spectra were obtained in CDCl3 using a Varian 400 MHz NMR. Thermogravimetric analysis (TGA) was done under N2 atmosphere using a Perkin-Elmer TGA7, and heating cycles varied. Results and Discussion Methacrylic/MeCD complexes were prepared in chloroform solvent (Scheme 1). The cyclodextrin utilized in this study was a methylated derivative of β-cyclodextrin (1.8 out of 3.0 hydroxyl groups present on each glucose ring were converted to methoxy groups). Complexation times of 6 days were used initially as previously described by Ritter et al.;23 however, this length of time was not required due to the observation that complexations of CDs generally tended to reach rapid equilibrium according to 1H NMR data. It is also well-known that complexation of a hydrophobic guest within the CD cavity is relatively unfavorable in chloroform.13 Therefore, it is proposed that the resulting solid MeCD/ methacrylic complex was formed as the result of the increasing concentration of the two species as cholorform was removed. Consequently, complexation times were reduced to 1 day for convenience. Complexation may require only a fraction of this time, but it has been observed that whether complexation time is 1 or 6 days the resulting solid complexes contain nearly identical ratios of monomer to MeCD (Table 1). However, the exact structure of the complex requires further characterization and will be reported later. TGA was utilized to determine the ratio of monomer to MeCD in the resulting complex from Scheme 1. First, TGA of pure, dry MeCD was performed as a reference. The cyclodextrin/monomer complex was carefully dried to remove any residual solvent and water and was analyzed under identical conditions. Figure 1 illustrates the thermal stabilities of the CMA/MeCD complex (0.72/1.0) and pure MeCD. The CMA/MeCD complex is represented and indicates an approximate 8% weight loss compared to MeCD at time equal to 55 min (T ) 250 °C). It is proposed that this weight loss resulted from dissociation of monomer from the complex. This 8% weight loss interestingly occurred approximately 35 °C above the boiling point of pure CMA, indicating a favorable association between monomer and MeCD. The TGA analysis is further supported using 1H NMR spectroscopy (Figure 2). Spectra 1 and 2 represent CMA and MeCD respectively prior to complexation. Spectrum 3 was obtained on the product produced from the complexation process (Scheme 1). Integration of the unsaturated protons (HA and HB) (Figure 2) of the monomer and MeCD (HD) minus contribution of proton (HC) of the CMA monomer was calculated and a ratio of these were determined. From these data the molar ratio of CMA to MeCD was calculated and the weight percent of CMA was easily determined. In the case of the CMA/MeCD complex that was shown to undergo an 8% weight loss at 250 °C by TGA, the weight percent of CMA in the complex was determined to be 8.2% by 1H NMR spectroscopy. This correlation demonstrated that

Biomacromolecules, Vol. 1, No. 4, 2000 617

Carbohydrate/Monomer Complexes

Scheme 1. Preparation of the Monomer/MeCD Complex (Step 1) and Subsequent Aqueous Polymerization (Step 2)

Table 1. Comparison of Results from Complexation Procedures That Involved Stirring for Either 1 or 6 days in Chloroform monomer

number of days

ratio monomer:MeCD

tBuMA tBuMA CMA CMA

6 1 6 1

0.60:1.00 0.65:1.00 0.72:1.00 0.70:1.00

TGA was a valuable complementary analytical method for facile characterization of alkyl methacrylate/MeCD complexes. The alkyl methacrylate/MeCD complexes were polymerized in aqueous media using potassium persulfate as the free radical initiator (Scheme 1). It was observed that methacrylic/ carbohydrate complexes were completely soluble in aqueous media. However, when monomer and water were mixed together in the absence of the complexing cyclodextrin, an immiscible two-phase system was produced. Reaction temperatures, monomers used, and initiator concentrations were varied and are summarized in Tables 2-4. Initially, a relatively large amount of initiator was used (9 wt % to monomer) as reported earlier by Ritter and co-workers,23 and these high levels are often considered dangerous and not recommended. Subsequent polymerizations utilized 1 wt % initiator to monomer and, as expected, resulted in higher molecular weights. A variation in glass transition temperature (Tg) was initially observed, and glass transitions varied as much as 10 °C compared to the control polymer produced in bulk. Each monomer was homopolymerized in bulk to obtain high molecular weight products that were utilized as controls and for subsequent film and blend production. The glass transition temperatures for reference polymers poly(tBuMA), poly(2EHMA), and poly(CMA) were +118, -5, and +112 °C, respectively. Also, 1H NMR showed that

varying amounts of MeCD remained complexed to the polymer (Figure 2). Therefore, it was concluded that residual MeCD in the isolated polymer had a significant effect on the Tg. This conclusion was supported by the preparation and analysis of approximately 1 mm thick methacrylic/MeCD films with up to 20 wt % MeCD. Again, the thermal analysis of these films indicated there was a significant dependence of the MeCD present on the Tg. It is proposed that a low weight percent of MeCD acts as a plasticizer. However, at higher concentrations, approximately 5 wt %, the presence of MeCD caused an increase in Tg (Tables 5 and 6). It was observed that the glass transition temperatures of polymers prepared by MeCD mediated aqueous polymerizations varied as much as 21 °C depending on the amount of residual MeCD. Gel permeation chromatography indicated that the residual cyclodextrin was not covalently bonded to the polymer backbone via a chain transfer mechanism. In addition, redissolution of the polymer products containing 3-8 mol % in tetrahydrofuran and subsequent precipitation into water quantitatively removed the residual cyclodextrin. Also, 1H NMR analysis indicated the absence of resonances associated with residual cyclodextrin. Moreover, a direct correlation between the monomer utilized for polymerization and the amount of residual MeCD in the precipitated polymer was not observed. The results of the aqueous polymerizations of the monomer/ MeCD complexes are summarized in Tables 2-4. Conditions A and B are examples of aqueous radical polymerizations of the specified complex. A was initiated with 1.0 wt % K2S2O8 to monomer while B was initiated with 9.0 wt % K2S2O8 to monomer. As expected, a comparison of A to B at the same temperature shows that molecular weights can be controlled by variations in initiator concentrations.

618

Biomacromolecules, Vol. 1, No. 4, 2000

Madison and Long

Figure 1. Thermogravametric analysis comparing a CMA/MeCD complex (0.70 mol/1.00 mol) and uncomplexed MeCD.

Figure 2.

1H

NMR spectra of CMA, MeCD, CMA/MeCD complex, and poly(CMA) from aqueous polymerization of CMA/MeCD complex.

Table 2. Data for the Polymerizations of tert-Butyl Methacrylate under Specified Conditions

Table 3. Data for the Polymerizations of Cyclohexyl Methacrylate under Specified Conditions

polymer

conditions

temp (°C)

Mn

Mw/Mn

Tg (°C)

% yield

polymer

conditions

temp (°C)

Mn

Mw/Mn

Tg (°C)

% yield

poly(tBuMA)1 poly(tBuMA)2 poly(tBuMA)3

A B A

50 50 60

137 000 90 000 48 000

3.45 3.69 3.62

107 120 125

77 85 65

poly(CMA)1 poly(CMA)2 poly(CMA)3

A B A

50 50 60

142 000 93 000 69 000

5.91 4.39 6.24

107 110 105

50 75 72

Comparison of two As at different temperatures indicated that an increase in reaction temperature also resulted in lower molecular weights. This is attributed simply to the fact that the half-life of K2S2O8 is estimated to be approximately 1 order of magnitude less at 60 °C than it is at 50 °C;28 hence, there is a higher concentration of radicals during polymerization. It is also encouraging that all polymerizations tend

to exhibit similar molecular weight trends depending on initiator concentration and polymerization temperature, and that these trends are typical for traditional free radical polymerizations. For example, under condition A at 50 °C both poly(CMA) and poly(tBuMA) exhibit number-average molecular weights of 142 000 and 137 000, respectively, and increases in either temperature or initiator concentration

Biomacromolecules, Vol. 1, No. 4, 2000 619

Carbohydrate/Monomer Complexes Table 4. Data for the Polymerizations of 2-Ethylhexyl Methacrylate under Specified Conditionsa polymer

conditions

temp (°C)

Mn

Mw/Mn

Tg (°C)

% yield

poly(2EHMA)1 poly(2EHMA)2 poly(2EHMA)3

A B A

50 50 60

140 000 56 000 53 000

3.56 3.20 2.75

2 5 -8

74 80 86

a Conditions: (A) polymerization from the complex in water initiated with 1.0 wt % K2S2O8; (B) polymerization from the complex in water initiated with 9.0 wt % K2S2O8.

Table 5. Dependence of Tg on Weight Percent of MeCD Present in Filma wt % MeCD

Tg (°C)

0 5 10 20

118 105 117 124

a Approximately 1 mm Thick films were optically clear as cast from CHCl3 with up to 20 wt % MeCD.

Table 6. Dependence of Tg on Weight Percent of MeCD Present in Filma wt % MeCD

Tg (°C)

0 5 10 20

112 100 105 110

a Approximately 1 mm thick films were optically clear as cast from CHCl 3 with up to 20 wt % MeCD.

resulted in the expected decrease in number-average molecular weight. 1H NMR (0.8-1.2 ppm, -CH ) analyses of polymer 3 products indicated that the cyclodextrin mediated aqueous polymerizations resulted in poly(alkyl methacrylate)s with stereochemistry similar to that obtained in conventional bulk polymerizations. For example, poly(tBuMA) produced from conventional bulk polymerization gave a 53% syndiotactic, 38% heterotactic, and 9% isotactic product, and poly(tBuMA) produced from MeCD mediated aqueous polymerization resulted in a product that was 55% syndiotactic, 40% heterotactic, and 5% isotactic. Isolated yields of these polymerizations were typically as high as 85% in water with 80-95% recovery of MeCD in all cases. 1H NMR analysis of recovered MeCD indicated that there was no change in the chemical composition, and therefore could be used repeatedly without purification. Experiments using recycled MeCD resulted in identical results. Transmission electron microscopy (TEM) was also employed to investigate the precipitant formed during aqueous polymerizations. A representative sample of reaction mixture was removed from the reactor and a single drop was placed on a carbon grid. The water was then allowed to evaporate leaving the solid polymer precipitate on the carbon grid. Figure 3 shows typical TEM photographs obtained at different magnifications of the precipitate obtained from the aqueous polymerization of the tBuMA/MeCD complex. Another interesting trend in the data is found in the polydispersity indices (PDI) of the different polymers. The trend in molecular weights under the different conditions was similar for each polymer, but when comparing one polymer

to another the polydispersities show a significant trend. The mean PDI increased 13% going from poly(2EHMA) (3.17 ( 0.41) to poly(tBuMA) (3.59 ( 0.12) and increased even more (53%) from poly(tBuMA) to poly(CMA) (5.51 ( 0.98). It is anticipated that PDI variations are due to the nature of the interactions between the polymer and MeCD as chain propagation occurs. Ritter et al. proposed a mechanism involving the dethreading of CD during propagation of the radical chain.19 It is believed that the affinity of the monomers to MeCD is driven largely by the hydrophobic-hydrophobic interaction between the monomer and the MeCD cavity.29 Therefore, a compound that fits snugly into the cavity will tend to have a stronger force of attraction. Corey-Pauling-Koltun (CPK) space filling atomic models were employed to investigate the relationship between monomer structure and PDI results obtained. CPK models are well-known atomic models that are utilized to examine the spatial relationships of molecules.13,30 CPK models of the tert-butyl group (Figure 4) suggested a good fit in the CD cavity due to the circumference of the tert-butyl ester alkyl. It is suspected to have a large affinity for the MeCD cavity. CPK models of the 2-ethylhexyl ester alkyl also indicated a suitable circumference containing peripheral hydrophobic groups and likewise suggested a large affinity toward the cavity of MeCD. The 2-ethylhexyl group however also has substantial length, which is believed to enhance its affinity toward MeCD. In addition, the CPK model of the cyclohexyl ester alkyl was shown to occupy the MeCD cavity less efficiently than either of the previous two, and has less affinity toward the MeCD cavity. The space filling models of the alkyl groups combined with the resulting PDI data suggest that there was a direct correlation between the size of the alkyl ester and the resulting PDI. It is proposed that the MeCD dethreading during chain propagation is significantly affected by the size of the methacrylate alkyl ester, and that complex strength between the alkyl ester and the cyclodextrin will affect the process of dethreading and thus the solubility of the propagating chain. Therefore, since the polymer precipitates during polymerization, it is reasonable to conclude that a monomer with a stronger complexing group will result in a more well-defined polymer than a monomer with a weak affinity toward complexation. On the basis of polymerization results and CPK spacefilling models, it is proposed that the tert-butyl and 2-ethylhexyl alkyl substituients are preferred for β-cyclodextrin mediation in aqueous media. An examination of a fourth monomer, n-butyl methacrylate, further supported these findings. The space-filling model indicated that it would have the least affinity toward the MeCD cavity compared to all other monomers. Therefore, the resulting polymer would be expected to exhibit an increase in PDI when compared to the previous three polymers. The mean PDI of poly(n-butyl methacrylate) was found to be 10.40 ( 0.74, which is greater than that for all polymers previously discussed. Future efforts involve the examination of emulsion type polymerizations of the monomers discussed here using carbohydrates as emulsifiers. Previous work has been done using cyclodextrin and high molecular weight dextran as emulsifiers.27,31 Our efforts will focus on using MeCD and a linear dextrin as

620

Biomacromolecules, Vol. 1, No. 4, 2000

Madison and Long

Figure 3. TEM analysis of precipitant formed from the aqueous polymerization of a tBuMA/MeCD complex.

as high as 0.72/1.00 were observed in the resulting solids. Data obtained from the novel characterization of these complexes using TGA correlated well with results obtained using 1H NMR spectroscopy. Polymers exhibited numberaverage molecular weights as high as 140 000 with PDI’s as low as 3.2. Hydrophobic, high molecular weight acrylic polymers were prepared in water with acceptable polydispersities and isolated yields as high as 86%. Additionally, 80-95% of the MeCD used in these reactions was recovered and 1H NMR data and subsequent polymerizations indicated that the carbohydrate could be recycled. Novel miscible methacrylic/carbohydrate films were also produced and shown to possess interesting thermal properties. Acknowledgment. The authors would like to thank Wacker-Chemie for their generous donation of the substituted cyclodextrins used in this project. Financial support provided by Jeffress Memorial Trust, 3M Co., National Science Foundation (CRIF CHE-9974632), and Virginia Tech Department of Chemistry is also gratefully acknowledged. References and Notes Figure 4. CPK molecular model of a possible tBuMA/β-CD complex.

surfactants in emulsion polymerizations in order to elucidate the effect of prior complexation and a homogeneous starting material on polymerization mechanism and final products. Conclusions It has been demonstrated that with complexation times of 1 day, solid complexes of tBuMA, CMA, or 2-EHMA with MeCD were obtained. Molar ratios of monomer to MeCD

(1) Chesnutt, T. Market Response to the GoVernment Regulation of Toxic Substances, The Rand Corp.; Santa Monica, CA, 1998. (2) Wagner, I. The Complete Guide to the Hazardous Waste Regulations, 3rd ed., John Wiley & Sons: New York, 1999. (3) Shaffer, K.; DeSimone, J. Trends Polym. Sci. 1995, 3, 146-153. (4) Mecoy, M. Chem. Eng. News 1999, 77, 25-27. (5) Aoi, K.; Takasu, A.; Okada, M. Macromolecules 1997, 30, 61346138. (6) Havard, J.; Vladimorov, N.; Frechet, J. Macromolecules 1999, 32, 86-94. (7) Aoi, K.; Tsutsumichi, K.; Aoki, E.; Okada, M. Macromolecules 1996, 29, 4456-4458. (8) Ohno, K.; Tsujii, Y.; Miyamoto, T.; Fukuda, T. Macromolecules 1998, 31, 1064-1069.

Carbohydrate/Monomer Complexes (9) Loykalnaunt, S.; Hayashi, M.; Hirao, A. Macromolecules 1998, 31, 9121-9126. (10) Renard, E.; Deratan, A.; Volet, G.; Sebille, B. Eur. Polym. J. 1997, 33, 49-57. (11) Bachman, F.; Hopken, H.; Kohli, R.; Lohmann, D.; Schneider, J. J. Carbohydr. Chem. 1998, 17, 1359-1375. (12) Hedges, A. Chem. ReV. 1998, 98, 2035-2044. (13) Bender, M.; Komiyama, M. Cyclodextrin Chemistry; Springer; Berlin, 1978. (14) Saenger, W. Angew. Chem., Int. Ed. 1980, 19, 344. (15) Shiraishi, Y.; Toshima, N.; Kawarmura, T.; Mihori, H.; Shirai, H.; Hira, H. J. Mol. Catal. A: Chem. 1999, 139, 149-158. (16) Tilloy, S.; Bertoux, F.; Mortreux, A.; Monflier, E. Catal. Today 1999, 48, 245-253. (17) Crini, G.; Bertini, S.; Torri, G.; Naggi, A.; Sforzini, D.; Vechi, C.; Janus, L.; Lekchiri, Y.; Morcellet, M. J. Appl. Polym. Sci. 1998, 68, 1973-1978. (18) Hattori, K.; Takahaxhi, K.; Inamaru, H.; Imajo, S.; Murai, S. J. Colloid Interface Sci. 1997, 190, 488-490. (19) Spencer, J.; He, Q.; Ke, X.; Wu, Z.; Fetter, E. J. Solution Chem. 1998, 27, 1009-1019.

Biomacromolecules, Vol. 1, No. 4, 2000 621 (20) Born, M.; Ritter, H. Macromol. Rapid Commun. 1996, 17, 197202. (21) Harada, A.; Kawaguchi, Y.; Toshiyuki, N.; Kamachi, M. Macromol. Rapid Commun. 1997, 18, 535-539. (22) Okada, M.; Kamuchi, M.; Harada, A. Macromolecules 1999, 32, 7202-7207. (23) Jeromin, J.; Ritter, H. Macromol. Rapid Commun. 1998, 19, 377379. (24) Jeromin, J.; Ritter, H. Macromol. Chem. Phys. 1998, 199, 26412645. (25) Jeromin, J.; Ritter, H. Macromolecules 1999, 32, 5236-5239. (26) Glockner, P.; Ritter, H. Macromol. Rapid Commun. 1999, 20, 602605. (27) Tattersall, P.; Rimmer, S. Polymer 1999, 40, 6673-6677. (28) Kolthoft, I.; Miller, I. J. Am. Chem. Soc. 1951, 73, 3055. (29) Amiel, C.; Sebille, B. AdV. Colloid Interface Sci. 1999, 79, 105122. (30) Cram, J.; Cram, D. Science 1974, 185, 4127. (31) Chern, C. S.; Lee, C. K.; Tsai, Y. J. Colloid Polym. Sci. 1997, 275, 841.

BM0055706