Synthesis of High Molecular Weight Poly(glycerol monomethacrylate) via RAFT Emulsion Polymerization of Isopropylideneglycerol Methacrylate.

High molecular weight water-soluble polymers are widely used as flocculants or thickeners. However, synthesis of such polymers via solution polymerization invariably results in highly viscous fluids, which makes subsequent processing somewhat problematic. Alternatively, such polymers can be prepared as colloidal dispersions; in principle, this is advantageous because the particulate nature of the polymer chains ensures a much lower fluid viscosity. Herein we exemplify the latter approach by reporting the convenient one-pot synthesis of high molecular weight poly(glycerol monomethacrylate) (PGMA) via the reversible addition-fragmentation chain transfer (RAFT) aqueous emulsion polymerization of a water-immiscible protected monomer precursor, isopropylideneglycerol methacrylate (IPGMA) at 70 °C, using a water-soluble poly(glycerol monomethacrylate) (PGMA) chain transfer agent as a steric stabilizer. This formulation produces a low-viscosity aqueous dispersion of PGMA-PIPGMA diblock copolymer nanoparticles at 20% solids. Subsequent acid deprotection of the hydrophobic core-forming PIPGMA block leads to particle dissolution and affords a viscous aqueous solution comprising high molecular weight PGMA homopolymer chains with a relatively narrow molecular weight distribution. Moreover, it is shown that this latex precursor route offers an important advantage compared to the RAFT aqueous solution polymerization of glycerol monomethacrylate since it provides a significantly faster rate of polymerization (and hence higher monomer conversion) under comparable conditions.

Introduction

Water-soluble polymers can be used for a wide range of commercial applications, including as flocculants in brewing,[1] for dewatering in paper manufacture[2−4] or for municipal water purification.[5−7] High molecular weight (>105 g mol–1) polymers are particularly efficient and include nonionic, anionic, or cationic polyacrylamides,[7−9] poly(ethylene oxide) (PEO),[10] and poly(diallyldimethylammonium) chloride (PDADMAC).[11] Such polymers induce aggregation via a bridging flocculation mechanism.[12−14] Water-soluble polymers are also widely employed as viscosity modifiers.[15−17] For example, PEO[18] and poly(acrylic acid) (PAA)[19] are commonly used as thickening agents in cosmetics, while polyurethanes (PU)[20] and poly(vinyl alcohol) (PVA) are utilized in paint formulations.[21] In such applications polymers often confer the additional benefit of acting as steric stabilizers for other components of the formulation, e.g., oil droplets or pigments.[22]

Poly(glycerol monomethacrylate) (PGMA) is a water-soluble polymer that is highly biocompatible and nonfouling and has been utilized for the manufacture of soft contact lenses.[23−25] Glycerol monomethacrylate (GMA) is a relatively expensive specialty monomer. In principle, it can be obtained via hydrolysis of a cheap commodity monomer, glycidyl methacrylate, in aqueous solution,[26] but in practice it is actually prepared via a protected precursor, isopropylideneglycerol methacrylate.[27] In the field of biomaterials, GMA-based copolymers have been used to prepare hydrogels that act as corneal substitutes,[28] for the design of amphiphilic networks that serve as suitable substrates for dermal fibroblasts,[29−31] and grown in the form of a hydrophilic brush layer from tissue culture polystyrene in order to improve cell adhesion.[32] Canton et al. demonstrated that human stem cell colonies enter stasis within 16 h of their immersion within PGMA-based block copolymer worm gels.[33] In addition, the cis-diol moiety of PGMA has been utilized for metal binding to magnetite[34] and other iron-based materials.[35] Recently, Deng and co-workers reported that 4-aminophenylboronic acid can bind to PGMA-based block copolymer vesicles in alkaline aqueous solution, hence inducing various morphological order–order transitions.[36]

Polymerization-induced self-assembly (PISA) is a well-recognized and versatile platform technology for the efficient synthesis of a wide range of block copolymer nano-objects.[37−44] PISA formulations based on RAFT aqueous emulsion polymerization involve chain-extending a water-soluble precursor polymer with a water-immiscible monomer to produce an amphiphilic diblock copolymer in situ.[45−56] This drives self-assembly to produce sterically stabilized nanoparticles. In principle, the copolymer morphology should simply depend on the relative volume fractions of each block, with more asymmetric blocks forming either worms or vesicles.[38,47,57−60] However, in many cases only kinetically-trapped spheres are accessible.[45,61−67] For example, Cunningham and co-workers[61] explored the scope of RAFT aqueous emulsion polymerization by chain-extending a PGMA macro-CTA with benzyl methacrylate (BzMA) at up to 50% solids. Only spherical nanoparticles were obtained, but the particle size was readily tunable by varying the DP of the core-forming PBzMA block. A maximum Mn of 117 000 g mol–1 (measured by DMF GPC vs near-monodisperse PMMA standards) could be achieved when targeting a PBzMA DP of 1000. More recently, Davis and co-workers[62] have used a similar PISA formulation to produce “ultrahigh” molecular weight polystyrene in the form of a low-viscosity dispersion of sterically stabilized nanoparticles. More specifically, a statistical copolymer of ethylene glycol methyl ether acrylate (EGA) and N-hydroxyethyl acrylamide (HEAA) was chain-extended via RAFT aqueous emulsion polymerization of styrene. Block copolymers with Mn values exceeding 106 g mol–1 were obtained with dispersities of less than 1.40.

Destarac and co-workers recently reported the synthesis of high molecular weight water-soluble polymers via the RAFT solution polymerization of acrylamide.[68]Mn values of more than 106 g mol–1 with relatively narrow molecular weight distributions (typically Mw/Mn < 1.30) were achieved reproducibly at 10 °C by utilizing high monomer concentrations and a relatively low initiator concentration. However, the final reaction solutions were relatively viscous. Cunningham and co-workers[69] offered a potential solution to this problem by utilizing RAFT aqueous dispersion polymerization to prepare a high molecular weight water-soluble polymer above its lower critical solution temperature (LCST). Thus, a PGMA macromolecular chain transfer agent (macro-CTA) was chain-extended with N-(2-(methacryloyloxy)ethyl)pyrrolidone (NMEP) at 70 °C to yield a low-viscosity dispersion of partially hydrated spherical PGMA–PNMEP nanoparticles. PNMEP exhibits an LCST of around 55 °C. Thus, cooling such aqueous dispersions induced particle dissolution to produce molecularly-dissolved copolymer chains at 20 °C, with a concomitant significant increase in solution viscosity. Although not a true homopolymer, the mean degree of polymerization (DP) of the PNMEP block could be systematically varied from 100 up to 4500, which substantially exceeded that of the PGMA stabilizer block. Moreover, DMF GPC analysis indicated relatively low dispersities (Mw/Mn < 1.50), and high NMEP conversions (>98%) could be achieved for such PISA formulations. However, literature examples of the preparation of high molecular weight water-soluble homopolymers in low-viscosity form using wholly aqueous formulations are rather rare.[70,71]

Herein we examine such a strategy for the synthesis of high molecular weight PGMA of relatively narrow molecular weight distribution. More specifically, a water-soluble PGMA stabilizer is chain-extended with isopropylideneglycerol methacrylate (IPGMA) using RAFT aqueous emulsion polymerization at pH 4 to produce amphiphilic PGMA–PIPGMA diblock copolymers in the form of sterically stabilized nanoparticles (see Scheme ). Optimization of this PISA formulation enabled the mean DP to be maximized while achieving at relatively high monomer conversions. Subsequently, the hydrophobic PIPGMA block can be deprotected to afford a water-soluble PGMA homopolymer via selective hydrolysis at low pH. This approach bears some similarity to that employed by Zentel and co-workers, who copolymerized IPGMA to form pH-responsive nanoparticles that undergo dissociation on addition of acid.[72] In this context, it is also worth noting a recent report by Rimmer and co-workers, who prepared polystyrene–poly(isopropylideneglycerol methacrylate) core–shell latexes via conventional aqueous emulsion polymerization.[73] Subsequent deprotection of the methacrylic residues in the shell at low pH led to PGMA-stabilized PS latexes that proved to be highly resistant to protein fouling.

Scheme 1

Synthesis of PGMA–PIPGMA (G39-I) Diblock Copolymer Nanoparticles via RAFT Aqueous Emulsion Polymerization of Isopropylideneglycerol Methacrylate (IPGMA) at 70 °C Using a PGMA Chain Transfer Agent as a Steric Stabilizer

Experimental Section

Materials

Glycerol monomethacrylate (GMA, 99.8%), and isopropylideneglycerol methacrylate (IPGMA, 97.8%) were donated by GEO Specialty Chemicals (Hythe, UK) and used without further purification. 4,4′-Azobis(4-cyanopentanoic acid) (ACVA, 99%) and dichloromethane were purchased from Sigma-Aldrich (UK) and were used as received. 2-Cyano-2-propyldithiobenzoate (CPDB) was purchased from Strem Chemicals Ltd. (Cambridge, UK) and was used as received. Deuterated DMF and methanol were purchased from Goss Scientific Instruments Ltd. (Crewe, UK). All other solvents were purchased from Fisher Scientific (Loughborough, UK) and used as received. Deionized water was used for all experiments.

Protocol for the Synthesis of a PGMA Macro-CTA

A PGMA39 (or G39) macromolecular chain transfer agent (macro-CTA) was synthesized as follows: CPDB RAFT agent (0.829 g, 3.70 mmol) and GMA monomer (30.0 g, 187.3 mmol) were weighed into a 100 mL round-bottomed flask and purged under N2 for 30 min. ACVA initiator (210 mg, 0.75 mmol; CTA/ACVA molar ratio = 5.0) and anhydrous ethanol (46.6 mL; previously purged with N2 for 30 min) were then added, and the resulting red solution was degassed for a further 10 min. The flask was subsequently sealed and immersed into an oil bath set at 70 °C. After 100 min, the GMA polymerization was quenched by exposing the flask to air, immersing it in liquid nitrogen for 30 s, and dilution of the reaction solution with methanol (100 mL). A final GMA conversion of 69% was determined by 1H NMR analysis by comparing the integrated monomer vinyl signals at 6.1–6.2 ppm to oxymethylene signals adjacent to the methacrylic ester groups of polymerized GMA residues at 3.8–4.3 ppm (see Figure S1 in the Supporting Information). The methanolic solution was precipitated into a ten-fold excess of dichloromethane. After filtering and washing with dichloromethane, the crude polymer was dissolved in water and the residual dichloromethane was evaporated under vacuum. The resulting aqueous solution was freeze-dried overnight to yield a pink powder. 1H NMR analysis indicated a mean degree of polymerization of 39 ± 1 for this purified PGMA macro-CTA, by comparing the integrated aromatic protons assigned to the RAFT CTA end-group at 7.3–8.0 ppm to that of the polymerized GMA repeat units at 3.8–4.3 ppm (see Figure S2). DMF GPC analysis confirmed that this GMA homopolymerization was well-controlled (Mn = 11 100 g mol–1, Mw/Mn = 1.13).

Preparation of PGMA39–PIPGMA Nanoparticles via RAFT Aqueous Emulsion Polymerization

PGMA39–PIPGMA1000 (G39-I1000) was synthesized as follows: PGMA39 macro-CTA (0.026 g, 4.00 μmol), IPGMA monomer (0.80 g, 3.99 mmol), and ACVA initiator (0.28 mg, 1.00 μmol) were weighed into a 10 mL round-bottomed flask and dissolved in deionized water (3.30 mL). The resulting solution was purged under N2 for 30 min before being sealed and immersed in an oil bath at 70 °C for 5 h. The polymerization was quenched by exposure to air and cooling to 20 °C. A final IPGMA conversion of more than 97% was determined by 1H NMR analysis by comparing the integrated monomer vinyl signals at 6.2–6.3 ppm to that of the six methyl protons assigned to the acetal group of the polymerized IPGMA residues at 1.5–1.7 ppm (see Figure S3). These PGMA39–PIPGMA1000 spherical nanoparticles were used without further purification.

Deprotection of PGMA39–PIPGMA1000 Nanoparticles To Afford Water-Soluble PGMA1039

A 20% w/w aqueous dispersion of PGMA39–PIPGMA1000 diblock copolymer spheres (4.0 mL; initial pH 3) was transferred into a 10 mL round-bottomed flask and adjusted to pH 1 by addition of concentrated HCl. The resulting acidic solution was immersed in an oil bath at 70 °C for 3 h. 1H NMR analysis indicated that 99% of the IPGMA residues were converted into GMA residues, yielding a 16% w/w aqueous acidic solution of water-soluble PGMA1039 homopolymer.

One-Pot Protocol To Afford Water-Soluble PGMA1039 via RAFT Aqueous Emulsion Polymerization of IPGMA Followed by Acid Hydrolysis

PGMA39 macro-CTA (0.026 g, 4.00 μmol), IPGMA monomer (0.80 g, 3.99 mmol), and ACVA initiator (0.28 mg, 1.00 μmol) were weighed into a 10 mL round-bottomed flask and dissolved in deionized water (3.30 mL). The resulting solution was purged under N2 for 30 min before being sealed and immersed in an oil bath at 70 °C for 6 h. A final IPGMA conversion of more than 99% was determined by 1H NMR analysis. The polymerization was quenched by exposure to air. The solution was adjusted to pH 1 by addition of concentrated HCl. The resulting acidic solution was maintained at 70 °C for 3 h. 1H NMR analysis indicated that 99% of the IPGMA residues were converted into GMA residues, yielding a 16% w/w aqueous acidic solution of water-soluble PGMA1039 homopolymer.

RAFT Aqueous Solution Polymerization of GMA

CPDB RAFT agent (11 mg, 5.00 μmol) and GMA monomer (0.80 g, 5.00 mmol; target DP = 1000) were weighed into a 10 mL round-bottomed flask and purged under N2 for 30 min. ACVA initiator (0.35 mg, 1.25 μmol; CTA/ACVA molar ratio = 4.0) was dissolved in deionized water (3.21 mL) and added to the monomer solution. The resulting solution was purged under N2 for 30 min before sealing the flask and immersing it in an oil bath at 70 °C for 5 h. The polymerization was quenched by exposure to air and cooling to 20 °C. A final GMA conversion of more than 97% was determined by 1H NMR analysis.

NMR Spectroscopy

All 1H NMR spectra were recorded in either deuterated methanol (for the PGMA macro-CTAs) or deuterated DMF (for the series of PGMA–PIPGMA diblock copolymers and for monitoring the acid-catalyzed deprotection of the PGMA–PIPGMA diblock precursor to afford PGMA homopolymer) using a 400 MHz Bruker Avance-400 spectrometer (64 scans averaged per spectrum).

Gel Permeation Chromatography (GPC)

Copolymer molecular weights and dispersities were determined using an Agilent 1260 Infinity GPC system equipped with both refractive index and UV–vis detectors. Two Agilent PL gel 5 μm Mixed-C columns and a guard column were connected in series and maintained at 60 °C. HPLC-grade DMF containing 10 mM LiBr was used as eluent and the flow rate was set at 1.0 mL min–1, with DMSO used as a flow-rate marker. The refractive index detector was used for calculation of molecular weights and dispersities by calibration using a series of ten near-monodisperse poly(methyl methacrylate) standards (with Mn values ranging from 625 to 618 000 g mol–1; see Figure S4). UV GPC chromatograms were obtained simultaneously by detection at a fixed wavelength of 309 nm, which corresponds to the absorption maximum for the dithiobenzoate RAFT end-groups.

Transmission Electron Microscopy (TEM)

Copolymer dispersions were diluted 50-fold at 20 °C to generate 0.20% w/w dispersions. Copper/palladium TEM grids (Agar Scientific, UK) were coated in-house to produce a thin film of amorphous carbon. These grids were then treated with a plasma glow discharge for 30 s to create a hydrophilic surface. One droplet of each aqueous diblock copolymer dispersion (12 μL; 0.20% w/w) was placed on a freshly treated grid for 1 min and then blotted with filter paper to remove excess solution. To stain the deposited nanoparticles, an aqueous solution of uranyl formate (9 μL; 0.75% w/w) was placed on the sample-loaded grid via micropipet for 20 s and then carefully blotted to remove excess stain. Each grid was then carefully dried using a vacuum hose. Imaging was performed using a FEI Tecnai Spirit TEM instrument equipped with a Gatan 1kMS600CW CCD camera operating at 120 kV.

Oscillatory Rheology Experiments

An AR-G2 rheometer equipped with a variable temperature Peltier plate, a 40 mL 2° aluminum cone, and a solvent trap was used for all experiments. Temperature sweeps were conducted at an angular frequency of 1.0 rad s–1 and a constant strain of 1.0%. The temperature was increased by 1.0 °C between each measurement, allowing an equilibration time of 2 min in each case. A solvent trap was required to prevent evaporation of water on the time scale of these experiments.

Dynamic Light Scattering (DLS)

Measurements were conducted at 25 °C using a Malvern Instruments Zetasizer Nano series instrument equipped with a 4 mW He–Ne laser (λ = 633 nm) and an avalanche photodiode detector. Scattered light was detected at 173°. Copolymer dispersions were diluted to 0.10% w/w. Intensity-average hydrodynamic diameters were averaged over three runs and calculated via the Stokes–Einstein equation.

Results and Discussion

The goal of this research was to synthesize high molecular weight PGMA homopolymer in aqueous solution via deprotection of PGMA–PIPGMA diblock copolymer nanoparticles, thus circumventing the problem of high solution viscosity usually associated with an aqueous solution polymerization route.[68] Moreover, given that an emulsion polymerization protocol was employed to prepare the intermediate sterically stabilized nanoparticles, a significantly faster rate of polymerization was anticipated compared to that obtained via aqueous solution polymerization owing to the well-known effect of compartmentalization, which leads to a significant reduction in the rate of termination and hence allows access to high molecular weight polymer chains.[74,75]

Optimization of PGMA–PIPGMA Diblock Copolymer Synthesis

First, a well-defined PGMA macro-CTA (Mn = 11 100; Mw/Mn = 1.13) was prepared at 70 °C in ethanol using CPDB as the RAFT CTA. In principle, a trithiocarbonate-based RAFT agent should also be suitable for the RAFT emulsion polymerization of IPGMA. However, a dithiobenzoate-based CTA was chosen for this study in view of the well-controlled RAFT emulsion polymerizations obtained for other water-immiscible monomers such as benzyl methacrylate or 2,2,2-trifluoroethyl methacrylate.[61,76] The mean DP of this water-soluble homopolymer was determined to be 39 by 1H NMR spectroscopy. Subsequently, this PGMA39 precursor was chain-extended via RAFT emulsion polymerization of IPGMA at 20% w/w solids. Like the majority of RAFT aqueous emulsion polymerization formulations, only spherical nanoparticles were obtained using this protocol.[40,61,62] In the context of the present study, this kinetically-trapped morphology is actually an advantage because it ensures that a relatively low dispersion viscosity is maintained during such syntheses. In each case, high monomer conversions (>97%) were determined by 1H NMR spectroscopy, as judged by the disappearance of the vinyl proton signals at 5.9 and 6.2 ppm. In addition, DMF GPC analysis indicated low dispersities (typically Mw/Mn < 1.29), while DLS studies confirmed the formation of near-monodisperse spheres (polydispersities typically below 0.10). Thus, good control was achieved over both the molecular weight distribution and the particle size distribution during such heterogeneous polymerizations.

The kinetics for the RAFT emulsion polymerization of IPGMA were monitored when targeting a final diblock composition of PGMA39–PIPGMA1000 by extracting aliquots from the reaction solution at regular time intervals. After quenching the polymerization via dilution and cooling, these samples were analyzed in turn by 1H NMR, DLS, and DMF GPC (see Figures a, 1b, and 1c, respectively. 1H NMR spectra recorded at various reaction times (and hence monomer conversions) are shown in Figure S5.

Figure 1

Analysis of aliquots extracted during the PISA synthesis of PGMA39–PIPGMA1000 nanoparticles via RAFT emulsion polymerization of IPGMA at 70 °C showing (a) conversion vs time curve and the corresponding semilogarithmic plot against time as determined by 1H NMR spectroscopy, (b) evolution of intensity-average DLS diameter against time, and (c) evolution of Mn and Mw/Mn against conversion determined by DMF GPC using a series of near-monodisperse poly(methyl methacrylate) calibration standards. The theoretical Mn is shown by a dashed line. Conditions: 20% w/w solids; ACVA initiator; macro-CTA/ACVA molar ratio = 4.0.

1H NMR studies confirmed that more than 95% conversion was achieved within 2 h at 70 °C. DMF GPC analysis indicated the linear evolution of molecular weight with conversion expected for a well-controlled RAFT polymerization, with relatively low dispersities (Mw/Mn < 1.40) being maintained throughout the reaction. Somewhat broader molecular weight distributions were observed above 90% conversion, as judged by the significant increase in dispersity (from Mw/Mn ∼ 1.23 up to Mw/Mn ∼ 1.38). This is attributed to chain transfer to polymer, which becomes more likely under monomer-starved conditions. Close inspection of the semilogarithmic plot revealed a significant rate acceleration between 90 and 120 min. In the case of RAFT dispersion polymerization formulations, such data have been interpreted in terms of the onset of micellar nucleation.[77−81] However, the concomitant DLS studies indicate the presence of (presumably) monomer-swollen nanoparticles of around 120 nm in the reaction solution after just 20 min (which corresponds to the time at which the first aliquot was extracted). Such early nucleation is not atypical for RAFT emulsion polymerization syntheses.[61,82] For the present formulation it is also physically realistic because the monomer conversion observed after 20 min is approximately 22%, which corresponds to a mean DP of 220 for the hydrophobic PIPGMA block. Between 80 and 100 min there is a discernible increase in the rate of IPGMA polymerization. There are only a few literature examples of PISA formulations exhibiting faster polymerization kinetics after the onset of micellar nucleation.[80,83,84] This unusual behavior is not fully understood, but it is worth emphasizing that we have observed such behavior for both aqueous and nonaqueous PISA systems.

In a second set of experiments, a series of PGMA39–PIPGMA diblock copolymers were prepared by targeting PIPGMA DPs ranging between 100 and 2000 while maintaining an overall solids concentration of 20% w/w. Given that the PGMA macro-CTA/initiator molar ratio was fixed at 4, this means that lower initiator concentrations are utilized when targeting higher DPs. This leads to progressively slower RAFT polymerizations, and at some point the radical flux becomes so low that the final monomer conversion becomes rather irreproducible for such formulations.[69] Indeed, high IPGMA conversions (at least 97%) could be achieved when targeting DPs up to 1000, with narrow molecular weight distributions being maintained (see Table ). However, a substantially lower conversion (42%) was obtained when targeting a DP of 2000. For an intermediate target DP of 1500, a final IPGMA conversion of 95% was achieved in one particular synthesis, but several attempts to repeat this result were unsuccessful (Table contains details of the best results achieved for this PISA formulation, which is on the cusp of irreproducibility owing to the relatively low initiator concentration). Thus, high conversions could only be reproducibly achieved when targeting DPs of up to 1000. For this upper limit DP the Mw/Mn was 1.20, which indicates relatively good RAFT control.[85−87]

Table 1

Summary of Monomer Conversion, Molecular Weight, and Intensity-Average Particle Diameter Data Obtained Using 1H NMR Spectroscopy, DMF GPC (Refractive Index Detector; Poly(methyl methacrylate) Standards), and Dynamic Light Scattering (DLS), Respectively, for a Series of PGMA39–PIPGMAX (G39-I) Diblock Copolymer Nanoparticles Prepared at 20% w/w Solids via RAFT Aqueous Emulsion Polymerization of IPGMA at 70 °C

sample no.	target composition	conv (%)	GPC Mn (g mol–1)	Mw/Mn	diameter/nm
1	G39	69	11100	1.13	N/A
2	G39-Il00	99	21500	1.24	42
3	G39-I200	99	30500	1.26	66
4	G39-I300	99	42100	1.29	98
5	G39-I400	99	51800	1.23	132
6	G39-I500	99	61500	1.21	163
7	G39-I600	99	71900	1.22	207
8	G39-I700	98	86400	1.28	247
9	G39-I800	97	90000	1.28	297
10	G39-I900	97	99200	1.22	315
11	G39-Il000	97	125000	1.20	363
12a	G39-Il500	95	159500	1.32	717
13	G39-I2000	42	96200	1.16	1172

Attempts to reproduce this formulation led to significantly lower monomer conversions.

DMF GPC analysis of the first eleven samples shown in Table revealed a linear evolution in Mn with increasing PIPGMA block DP (see Figure a), which is similar to that previously observed for the RAFT aqueous dispersion polymerization of NMEP.[69] Moreover, DLS studies indicated a linear correlation between the intensity-average diameter and PIPGMA DP for this series of spherical nanoparticles (see Figure b). Cunningham and co-workers also reported a monotonic increase in particle size with core-forming block DP for the synthesis of PGMA–PBzMA diblock copolymer nanoparticles prepared via RAFT emulsion polymerization. However, the mean hydrodynamic sphere diameters obtained in this earlier work were much smaller than those observed in the current study for similar core-forming block DPs.[61] Unlike the PNMEP-core particles reported by Cunningham et al.,[69] it seems unlikely that the PIPGMA-core particles are appreciably hydrated. However, we cannot rule out the possibility that some degree of deprotection of the IPGMA residues occurs in situ during the RAFT aqueous emulsion polymerization. If this were the case, it would introduce hydrophilic GMA units within the core-forming block, which could lead to some degree of particle swelling. However, such GMA units in the core-forming block would be spectroscopically indistinguishable from those in the stabilizer block.

Figure 2

(a) Evolution of Mn and Mw/Mn with PIPGMA DP, where the theoretical Mn is shown by a dashed line. (b) Correlation between intensity-average DLS diameter against PIPGMA DP for a series of PGMA39–PIPGMA spherical nanoparticles prepared via RAFT aqueous emulsion polymerization of IPGMA at 70 °C (see Table ).

Transmission electron microscopy images obtained for the PGMA39–PIPGMA1000 diblock copolymer nanoparticles (see entry 11, Table ) are shown in Figure . This confirms the well-defined spherical morphology for such nanoparticles.

Figure 3

Representative TEM images obtained for the PGMA39–PIPGMA1000 diblock copolymer nanoparticles.

Systematic Variation of the Copolymer Concentration

PIPGMA DPs of 1000, 1500, and 2000 were targeted in turn at 30% w/w solids using PGMA39 as the steric stabilizer block. However, such formulations only led to the formation of thick pastes, rather than free-flowing colloidal dispersions. Similar results were obtained at 25% w/w solids. Empirically, it was found that free-flowing dispersions could only be obtained at 20% w/w copolymer concentration when targeting PIPGMA DPs of 1000. Attempts to confer greater steric stabilization by utilizing a PGMA63 macro-CTA at 20% w/w solids also proved to be unsuccessful when targeting DPs of 1500 or 2000: free-flowing dispersions were obtained under such conditions, but conversions proved to be substantially incomplete. Using a low-temperature initiator (2,2′-azobis[2-(2-imidazolin-2-yl)propane]dihydrochloride; VA-044) at 50 °C combined with this longer stabilizer block enabled a final conversion of 84% to be achieved when targeting a DP of 2000. DMF GPC analysis of this PGMA39–PIPGMA1680 copolymer indicated an Mn of 203 000 but a relatively broad molecular weight distribution (Mw/Mn = 1.65), which suggests significant loss of RAFT control under such conditions. In summary, the optimal conditions for the RAFT aqueous emulsion polymerization of IPGMA at 70 °C involves using the PGMA39 macro-CTA at 20% w/w solids. This formulation reproducibly affords a final conversion of at least 97% within 2 h when targeting a DP of 1000, which produces an apparent Mn of around 125 000 g mol–1 and an Mw/Mn of 1.20–1.37 (e.g., see entry 11 in Table ).

Deprotection of PGMA–PIPGMA Spheres

It is well-known that acetal protecting groups are readily removed on addition of aqueous acid.[88] Indeed, the industrial manufacture of GMA monomer is achieved via acid-catalyzed deprotection of IPGMA,[27] and Hoogeveen et al. reported the preparation of PGMA-based diblock copolymers from PIPGMA-based precursors via acid hydrolysis at ambient temperature for 72 h.[89] Very recently, Russell and co-workers reported the deprotection of IPGMA residues in a series of polystyrene–PIPGMA (PS–PIPGMA) diblock copolymers using HCl in 1,4-dioxane.[90] Of particular relevance to the present study, a similar strategy was recently utilized by Rimmer and co-workers for the synthesis of sterically stabilized PS–PGMA latexes from precursor core–shell PS–PIPGMA particles.[73] In this case, acid hydrolysis was conducted in aqueous solution at approximately pH 1 for 4–8 h at 60 °C, but no kinetic studies of the extent of deprotection were reported.

Initial deprotection experiments involved adjusting the solution pH of a 20% w/w aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles to pH 1 via addition of HCl. This acidified turbid dispersion was then stirred for several days at 20 °C, but there was no discernible change in its appearance. In principle, successful deprotection of the acetal groups on the hydrophobic PIPGMA block should result in nanoparticle dissolution to form a transparent solution because the resulting PGMA homopolymer is water-soluble. This transformation was subsequently achieved for the same copolymer formulation by heating to 70 °C at pH 1. It is perhaps worth noting that the volatile acetone byproduct (bp 56 °C) is removed from the reaction solution at this temperature, which helps to drive the reaction toward completion. The extent of acetal deprotection under such conditions was monitored by extracting aliquots from the reaction dispersion/solution at predetermined time intervals for analysis by 1H NMR spectroscopy (in d7-DMF), DMF GPC, and DLS (see Figures and 5).

Figure 4

(a) Reaction scheme for the acid-catalyzed deprotection of PGMA39–PIPGMA1000 nanoparticles at 70 °C to afford water-soluble PGMA1039 chains after 3 h at pH 1. (b) 1H NMR spectra in d7-DMF recorded for the initial PGMA39–PIPGMA1000 nanoparticles and the final water-soluble PGMA1039 homopolymer obtained as a result of this acid-catalyzed deprotection.

Figure 5

(a) Gradual reduction in particle size and derived count rate observed during the deprotection of PGMA39–PIPGMA1000 nanoparticles under the conditions described in Figure . (b) DMF GPC curves indicating the apparent increase in Mn and reduction in Mw/Mn during the acid-catalyzed deprotection of PGMA39–PIPGMA1000.

The disappearance of the pendent methyl proton signals assigned to the IPGMA residues at 1.55 and 1.62 ppm relative to the methacrylic copolymer backbone proton signals at 0.93–1.43 ppm in the 1H NMR spectra allowed the extent of hydrolysis to be determined during the course of the acetal deprotection reaction. This analysis confirmed that more than 98% of the acetal groups were removed within 2 h at 70 °C. As expected, the initially turbid dispersion gradually became less opaque and eventually became transparent as water-soluble GMA-rich copolymer chains (and ultimately PGMA homopolymer) were formed toward the end of the reaction. Surprisingly, DMF GPC analysis of the initial PGMA39–PIPGMA1000 diblock copolymer, intermediate copolymers, and final PGMA1039 homopolymer indicated an apparent increase in Mn during acid deprotection. This is clearly an experimental artifact because the GMA repeat unit (160 g mol–1) is less massive than the IPGMA repeat unit (200 g mol–1). Presumably, DMF is a significantly better solvent for the PGMA chains (which hence occupy a larger hydrodynamic volume) than for the PIPGMA chains. It is perhaps also noteworthy that the molecular weight distribution remains essentially unchanged after deprotection, which confirms that no chain scission or cross-linking of the (co)polymer chains occurred under the hydrolysis conditions. Finally, DLS enabled the nanoparticle dissolution process to be conveniently monitored. The initial intensity-average diameter of 270 nm was reduced to just 30 nm within 150 min at 70 °C, while the scattered light intensity (or derived count rate) was reduced by more than two orders of magnitude over this time period. Moreover, the DLS polydispersities exceeded 0.50 after 120 min, which approximately corresponds to the time at which a significant reduction in solution turbidity is observed. Clearly, the size data shown here are rather noisy compared to the scattered light intensity (derived count rate), which most likely indicates the formation of transient, weakly scattering hydrogen-bonded complexes in aqueous solution. Overall, these observations are consistent with complete dissolution of the PGMA39–PIPGMA1000 diblock copolymer nanoparticles to afford molecularly-dissolved PGMA1039 homopolymer chains.

Deprotection of the IPGMA residues was also examined under milder conditions. A 20% w/w aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles was adjusted to pH 2 using HCl and heated to 70 °C. As expected, the rate of acid hydrolysis was significantly slower, but nevertheless 93% deprotection was achieved within 10 h (see Figure ). This presents a facile deprotection method for such polymers where the milder reaction conditions can be utilized at the expense of longer reaction times.

Figure 6

Conversion against time curves determined for the acid hydrolysis of a 20% w/w aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles at 70 °C by 1H NMR spectroscopy: (a) at pH 1 (red squares) and pH 2 (blue diamonds).

One-Pot Polymerization and Deprotection Protocol

Given that the RAFT aqueous emulsion polymerization and subsequent acid hydrolysis are both performed in aqueous solution at 70 °C, the feasibility of developing a convenient one-pot polymerization and deprotection route to high molecular weight PGMA homopolymers was examined, as outlined in Figure . Thus, IPGMA was polymerized using the same PGMA39 macro-CTA targeting a DP of 1000 for the PIPGMA. After 6 h, an aliquot of the resulting turbid dispersion was extracted for analysis by 1H NMR and DLS. The former technique indicated more than 99% conversion, and the latter suggested the presence of relatively uniform nanoparticles (338 nm, polydispersity = 0.053). DMF GPC analysis indicated an Mn of 128 000 and an Mw/Mn of 1.37, which indicates a somewhat broader molecular weight distribution than that reported in Table (see entry 11) but still suggests reasonable RAFT control. This hot 20% w/w aqueous dispersion was then exposed to air and immediately adjusted to pH 1 using HCl. The reaction temperature was maintained at 70 °C for a further 3 h before taking an aliquot from the resulting transparent solution for analysis. 1H NMR spectroscopy indicated more than 99% acetal deprotection, while both DLS studies and visual inspection confirmed loss of the original nanoparticles (see Figures a and 7, respectively). Finally, DMF GPC analysis of the final water-soluble PGMA1039 homopolymer obtained after an overall reaction time of 9 h at 70 °C had an apparent Mn of 154 000 and an Mw/Mn of 1.42.

Figure 7

A one-pot wholly aqueous synthetic protocol for the preparation of high molecular weight PGMA starting from a PGMA39 macro-CTA at 20% w/w solids. First, RAFT aqueous emulsion polymerization of IPGMA is conducted at 70 °C to produce PGMA39–PIPGMA1000 nanoparticles at pH 4, and then acid hydrolysis of the IPGMA residues is performed at the same temperature at pH 1. The latter deprotection reaction leads to nanoparticle dissolution and the formation of a transparent aqueous solution comprising water-soluble PGMA1039 homopolymer chains.

Advantages over Conventional Solution Polymerization

As described above, a wholly aqueous two-step one-pot synthetic route to high molecular weight water-soluble PGMA has been developed. At this point, it is pertinent to ask whether this strategy offers any useful advantage(s) over the RAFT aqueous solution polymerization of GMA. Thus, the RAFT aqueous solution polymerization of GMA was conducted using the same GMA concentration (16% w/w solids) as that achieved after acid hydrolysis of the PGMA39–PIPGMA1000 nanoparticles. To circumvent its limited water solubility, the CPDB RAFT agent was first dissolved in GMA monomer prior to addition of water and ACVA to make up the initial reaction solution. Aliquots were periodically taken for 1H NMR and DMF GPC analysis to determine the kinetics of GMA polymerization and hence enable a direct comparison to be made with the overall time scale required for the two-step one-pot protocol utilizing the precursor PGMA39–PIPGMA1000 nanoparticles (see Figure ).

Figure 8

Conversion vs time plots obtained for (a) PGMA39–PIPGMA1000 nanoparticles prepared by RAFT aqueous emulsion polymerization of IPGMA (red squares) and (b) PGMA1000 prepared via RAFT aqueous solution polymerization of GMA (black diamonds). Both syntheses were conducted at 70 °C at identical molar concentrations of monomer (either IPGMA or GMA); this corresponds to 16% w/w solids for the PGMA1000 chains and 20% w/w solids for the PGMA39–PIPGMA1000. The loss of the acetone protecting group during acid hydrolysis of the IPGMA residues in the latter synthesis accounts for the difference in solids content.

The RAFT solution polymerization of GMA (targeting PGMA1000) proceeded to 81% conversion within 5 h at 70 °C, whereas the RAFT emulsion polymerization of IPGMA (targeting PGMA39–PIPGMA1000) attained 97% conversion within 2 h at the same temperature. It is well-known that emulsion polymerizations typically proceed significantly faster than the equivalent solution polymerization.[80,91] This rate acceleration is attributed to compartmentalization, which reduces the instantaneous number of propagating polymer radicals per growing nanoparticle and hence lowers the rate of termination relative to that of propagation.[74,75] This homopolymer has an Mw/Mn of 1.27 at 81% conversion (see Figure S6), which is somewhat higher than that achieved for the final PGMA1039 homopolymer obtained via the RAFT aqueous emulsion polymerization of IPGMA (Mw/Mn = 1.20 at 97% conversion). Importantly, the overall time scale required for the synthesis of PGMA1039 chains using the two-step one-pot synthesis protocol is significantly shorter than that required for the RAFT aqueous solution polymerization of GMA. Assuming that first-order rate kinetics holds for this solution polymerization (which is the best case scenario), a further 5.5 h at 70 °C would be required to achieve 97% conversion. Thus, it is clear that significantly higher final monomer conversions can be achieved using the former route within shorter overall reaction times.

Finally, rheological studies were performed to compare the viscosity of the aqueous solution of PGMA1000 to that of the aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles. In addition, the final water-soluble PGMA1039 homopolymer obtained after acid hydrolysis of the PGMA39–PIPGMA1000 nanoparticles was also examined (see Figure ).

Figure 9

Viscosity vs. shear rate curves obtained for: (a) a 16% w/w aqueous solution of PGMA1000 prepared via RAFT solution polymerization of GMA (blue diamonds); (b) a 20% w/w aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles prepared via RAFT emulsion polymerization of IPGMA (red squares); and (c) a 16% w/w aqueous solution of PGMA1039 obtained after acid hydrolysis of a 20% w/w aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles (black triangles). Rheological measurements were performed at 20 °C, and the differing solids concentrations correspond to approximately equimolar polymer concentrations.

The 20% w/w aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles exhibits a significantly lower viscosity than either of the two PGMA homopolymer solutions across the entire range of shear rates investigated (101–103 s–1). Interestingly, an approximately Newtonian response is displayed by both these water-soluble homopolymers under these conditions, whereas shear-thinning behavior might have been expected.[92,93] This could be simply because the range of shear rates examined in the present study is too narrow. Alternatively, it may indicate extensive interchain interactions (e.g., hydrogen bonding). In addition, the upturn observed at low shear rates for the PGMA39–PIPGMA1000 dispersion suggests weakly interacting nanoparticles. In summary, the one-pot synthesis of high molecular weight water-soluble PGMA via precursor PGMA–PIPGMA nanoparticles can be conducted with faster kinetics, higher final monomer conversions, and lower viscosities than those achieved during the RAFT aqueous solution polymerization of GMA. This study provides a further example of the advantages offered by PISA formulations compared to conventional polymer syntheses conducted in homogeneous solution.

Conclusions

RAFT emulsion polymerization of IPGMA at 70 °C affords well-defined PGMA39–PIPGMA diblock copolymer spherical nanoparticles at 20% w/w solids. High final monomer conversions (at least 97%) could be reproducibly obtained when targeting PIPGMA block DPs up to 1000. A monotonic increase in both Mn and mean particle diameter was observed up to this critical value, with relatively low dispersities (Mw/Mn < 1.29) being maintained. However, irreproducible results were obtained when targeting DPs of 1500 or 2000, so a target DP of 1000 appears to represent an upper limit, at least for this specific PISA formulation. Acid hydrolysis of the aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles at 70 °C converts almost all (>98%) of the hydrophobic IPGMA residues into hydrophilic GMA residues within 2 h at pH 1. This leads to nanoparticle dissolution and the formation of an aqueous solution of PGMA1039 homopolymer.

Furthermore, a one-pot protocol was optimized whereby a highly viscous aqueous solution of PGMA1039 can be prepared at 20% w/w solids within 9 h from PGMA39–PIPGMA1000 nanoparticles, which act as a low-viscosity precursor. Importantly, the relatively fast kinetics achieved during the RAFT emulsion polymerization of IPGMA means that the overall time scale for this one-pot synthesis is significantly shorter than that required for the synthesis of PGMA1039 via RAFT aqueous solution polymerization, despite the requirement for post-polymerization deprotection of the IPGMA residues. Moreover, the viscosity of an aqueous dispersion of PGMA39–PIPGMA1000 nanoparticles at 20% w/w solids is significantly lower than that of PGMA1000 prepared via RAFT aqueous solution polymerization. In summary, we report a new wholly aqueous synthetic route to relatively high molecular weight PGMA via RAFT aqueous emulsion polymerization that offers significant advantages in terms of both overall kinetics and lower viscosity compared to the RAFT aqueous solution polymerization of GMA.