Radical polymerization of miniemulsions induced by compressed gases

Abstract

Generation of miniemulsions (nanoemulsions) comprising hydrophobic vinyl monomers and water by use of the compressed gases carbon dioxide and ethylene, respectively, has been investigated in the absence of high energy mixing. It is proposed that transparent/translucent miniemulsions are formed at the transparency pressure (P_T) as a result of the refractive index of the dispersed phase being reduced due to expansion with carbon dioxide/ethylene, resulting in refractive index matched miniemulsions with enhanced stability. Radical polymerization of methyl acrylate at 40 °C in miniemulsions induced by carbon dioxide at P_T proceeds as a hybrid miniemulsion/emulsion polymerization system generating particles with diameters less than 100 nm. The experimental data are consistent with particle formation occurring mainly via monomer droplet nucleation (miniemulsion polymerization) but also via secondary nucleation in the aqueous phase (emulsion polymerization). Polymerizations at a range of pressures above and below P_T revealed limited pressure effects on the polymerization rate and particle size for both carbon dioxide and ethylene.

---

Introduction

Miniemulsions are kinetically stable but thermodynamically unstable dispersions of one liquid in another liquid, most commonly an oil dispersed in water, with typical droplet diameters in the range 50–500 nm.^1–3 Ugelstad and co-workers⁴ initially reported the concept of miniemulsion polymerization in 1973. They found that the polymerization could be initiated in monomer droplets (monomer droplet nucleation) during emulsion polymerization if an appropriate recipe is chosen and applied with stirring. Ideally, each monomer droplet is transformed into a polymer particle. Thus, monomer diffusion from monomer droplets to polymer particles via the aqueous continuous phase is no longer required, as in conventional emulsion polymerization. This process facilitates the synthesis of hybrid nanoparticles,^5–7 hollow polymer particles^8,9 as well as implementation of controlled/living radical polymerization.^10–20 High-energy input, however, for example high-pressure homogenization or ultrasonication, is usually required to generate a kinetically stable miniemulsion, which remains an impediment to industrial implementation of miniemulsion polymerization.³ In addition, precise control of particle size in miniemulsions still remains a challenge. Therefore, the synthesis of polymeric nano-objects by the use of miniemulsions based on alternative low energy emulsification processes has been attracting attention.

Various methods have been developed for generation of miniemulsions without ultrasonication. Most of them are based on catastrophic phase transitions, which can be induced during the emulsification process owing to the change in the spontaneous curvature of the surfactant because of the variation in the physicochemical properties of the system. Examples include change of temperature (phase inversion temperature (PIT) method^21–23), change of composition (emulsion inversion point (EIP) method^24–26) or change of pH or ionic strength²⁷ with additives. Another low energy approach is based on in situ formation of surfactant at the oil/water interface.^17,28–35 An organic acid dissolved in the organic phase undergoes reaction at the interface with a base located in the aqueous phase, generating the surfactant at the interface. Although great progress has been made, some of these methods are not easily implemented (EIP method) or incompatible with thermally initiated radical polymerization (PIT method).

Han and co-workers have reported^36,37 the first example of the use of compressed gases, such as CO_2, to induce the formation of miniemulsions. The method employs gentle stirring at moderate pressures (<10 MPa) and is applicable to a wide range of oil-to-water volume ratios in an emulsion system. The presence of CO₂ increases the stability of the miniemulsion by retarding phase separation. Moreover, addition of CO₂ leads to a reduction in the turbidity of the initial emulsion until it becomes completely transparent at the transparency pressure (P_T). There is a gradual transition to more turbid emulsions with further increases in pressure above P_T. It is generally assumed that the smallest droplets are obtained in the transparent region, although this effect has yet to be conclusively demonstrated. Similar results have been found using gases other than CO₂ (ethylene, ethane, and propane),³⁶ and the effect of CO₂ is thus not related to an induced reduction in pH of the aqueous phase.

An emulsion can be transparent for three reasons: (i) the dispersed phase comprises sufficiently small droplets/particles (as typically observed in a microemulsion³⁸); (ii) the concentration of dispersed entities is sufficiently low, and (iii) the refractive indices of the two phases are similar.^39,40 It has been proposed that the transparency induced by compressed gases is related to the insertion of gas molecules into the interfacial film, resulting in a change in the curvature and creating smaller droplets.³⁶ The transparent emulsion formed via CO₂ pressurization of TX-100 (poly(oxyethylene)iso-octylphenyl ether)/water/isooctane at P_T = 3.90 MPa was characterized by small-angle X-ray scattering (SAXS), yielding spherical droplets of diameter 92 nm.⁴¹ However, a typical miniemulsion with a diameter of 92 nm is not transparent but milky white.¹

Compressed gas-induced miniemulsion polymerization entails pressurization of a coarse macroemulsion to moderate pressure followed by radical polymerization. The method circumvents high pressure mixing approaches for miniemulsion generation and may also offer the advantage of droplet/particle size-tunability via the gas pressure. Our group has reported⁴² RAFT polymerization of styrene in aqueous miniemulsions induced by compressed CO₂, obtaining size-controllable polymer nanoparticles with good control/livingness. However, despite these promising results, questions remain regarding the mechanism underpinning the effect of CO₂ on miniemulsion formation, and it remains unclear to what extent this approach can be exploited as a general method.

In the present work, we report the use of compressed CO₂ and ethylene to induce the formation of miniemulsions involving vinyl monomers dispersed in water. Radical polymerization of methyl acrylate in compressed gas-induced miniemulsions is explored for nanoparticle synthesis at pressures above and below P_T. The results are discussed in terms of the mechanisms for miniemulsion formation and polymerization. In particular, we examine whether polymerization proceeds via monomer droplet nucleation (miniemulsion polymerization) or emulsion polymerization (particle generation in the continuous phase).

Experimental

Materials

Methyl acrylate (MA; Aldrich, 99.9%), methyl methacrylate (MMA; Aldrich, 99.9%) and lauryl methacrylate (LMA; Aldrich, 99.9%) were purified by passing through a column of basic aluminum oxide (Ajax). Dowfax 8390 (disulfonated alkyl diphenyloxide sodium salt; Dow Chemicals), hexadecane (HD, Sigma-Aldrich, 99%), 2,2′-azobis[2-(2-imidazolin-2-yl)propane]dihydrochloride (VA-044, Wako, 99%), CO₂ (Coregas, 99.5%) and ethylene (Coregas, 99.8%) were used as received. Distilled and deionized water were used in the experiments.

Synthesis of 4-stearoyl-TEMPO

Synthesis was performed as reported previously,⁴³ whereby diethyl ether (20 ml), OH-TEMPO (2.0 g), and pyridine (2.0 g) were mixed at 38 °C. Stearoyl chloride (5.3 g) dissolved in ethyl ether (20 ml) was added via a dripping funnel over 10 min, the temperature was maintained at 38 °C and mixing was continued overnight under refluxing. The resulting suspension was filtered and washed three times with 3 wt% hydrochloric acid aqueous solution and then three times with deionized water prior to drying under vacuum.

Phase behaviour

Observations of the phase behaviour of miniemulsions induced by compressed CO₂/ethylene were conducted using an in-house designed and modified Jerguson sight gauge reactor with an internal volume of 60 ml. A detailed description of the reactor setup is given elsewhere.⁴⁴ In a typical experiment, an emulsion consisting of monomer/HD/Dowfax 8390/water (Table 1) was mixed in a glass bottle using gentle magnetic stirring for 30 min to obtain a turbid emulsion. The mixture was then transferred into the reactor and stirring was initiated. The reactor was sealed and purged with low pressure CO₂ or ethylene (0.5 MPa) to remove oxygen. Heating was then applied and the temperature of the emulsion was adjusted to the desired set point. CO₂ or ethylene was slowly added into the reactor once the temperature was within a few degrees of the desired value.

Table 1 Recipe for compressed gas-induced miniemulsions

Dispersed phase	Quantity (g)
a VA-044 initiator omitted for the phase behavior experiments.
MA/MMA/LMA	1.0
HD	0.08 (8 wt% to monomer)

---

Continuous phase	Quantity (g)
Dowfax 8390	0.1 (10 wt% to monomer)
Water	8.62
VA-044^a	0.004 (0.4 wt% to monomer)

---

The transparency pressure for a given system was determined using a trial and error approach. In general, the emulsion was pressurized using small increments of CO₂/ethylene. The system was allowed to stabilize for 30 min for each increment of compressed gas. Continuous stirring (1000 rpm) was maintained to ensure equilibrium between the liquid and vapour phases. Once the P_T had been identified, the pressure was further increased in stages until it became evident that the turbidity of the emulsion was increasing. The pressure was then reduced to a value slightly below the P_T to check the reliability of the transparency measurement. The addition of small increments of CO₂/ethylene was repeated until transparency was re-established. The uncertainty in the value of P_T was found to be less than 0.2 MPa using this procedure.

Radical polymerization of MA in compressed gas-induced miniemulsions

The apparatus used for phase behavior observations was also used for polymerizations. In a typical experiment, MA/HD/Dowfax 8390/water/VA-044 (Table 1) were mixed in a glass bottle using gentle magnetic stirring for 30 min. The mixture was then transferred into the reactor and stirring was initiated. Prior to the addition of the emulsion, the reactor was first cooled to 15 °C. The reactor was sealed and purged with nitrogen for 30 min to remove oxygen. The sequence of heating, followed by slow pressurization with CO₂/ethylene to a desired pressure, was identical to that used for the phase behaviour experiments. In general, the time required to establish the required temperature and pressure for a given polymerization was around 15 min. The polymerizations were stopped at prescribed times by depressurization and subsequent cooling of the reactor to room temperature. Polymerization of MA was conducted at 40 °C and the corresponding P_T, as well as pressures above and below this condition. Control experiments were also performed at 1 atm under N₂ (no pressurization) using the same reactor.

Measurements and characterization

Monomer conversions were obtained by gravimetry. A sample after polymerization was transferred to an aluminum dish and covered with a piece of perforated aluminum foil, and subsequently dried to constant weight in a vacuum oven at 40 °C. Molecular weights and molecular weight distributions were determined by gel permeation chromatography (GPC) with a Shimadzu modular system with THF as eluent at 40 °C at a flow rate of 1.0 ml min⁻¹ with injection volume of 40 μl. The GPC comprised a DGU-12A solvent degasser, a LC-10AT pump, a CTO-10A column oven and an ECR 7515 A refractive index detector, and a Polymer Laboratories 5.0 μm bead-size guard column (50 × 7.8 mm²), followed by four 300 × 7.8 mm² linear Phenogel columns (10⁵, 10⁴, 10³, and 500 Å). The system was calibrated with commercial polystyrene standards ranging from 500 to 10⁶ g mol⁻¹. Offline dynamic light scattering (DLS) measurements to measure particle size distributions and number-average (d_n), volume-average (d_v) and intensity-average (d_i) particle diameters were performed at 25 °C using a Malvern Zetasizer Nano Series with a 4 mW He–Ne laser operating at a wavelength of 633 nm and an avalanche photodiode (APD) detector. The scattered light detecting angle was 173 °C. The samples were diluted with distilled water (0.1 ml emulsion in 1 ml water).

Results and discussion

Effect of compressed gas on P_T

The influence of the type of compressed gas on the transparency pressure was examined for the MA/HD/Dowfax 8390/water system at 40 °C (Table 1). A turbid emulsion was initially obtained from combination of the dispersed and continuous phases for this system. Upon pressurization with CO₂, the turbidity of the emulsion decreases until a completely transparent emulsion is obtained at 3.83 MPa (Fig. 1). Beyond this pressure it can be seen that turbidity again increases. The same general trend is observed using ethylene on the same emulsion (Fig. 2). The P_T obtained using ethylene is substantially higher at 4.91 MPa.

Fig. 1 Phase behavior of the MA/HD/Dowfax 8390/water system at 40 °C with compressed CO₂: (A) 3.40 MPa, (B) 3.53 MPa, (C) 3.76 MPa, (D) 3.83 MPa (P_T), (E) 3.95 MPa, (F) 4.13 MPa and (G) 4.25 MPa.

Fig. 2 Phase behavior of the MA/HD/Dowfax 8390/water system at 40 °C with compressed ethylene: (A) 4.75 MPa, (B) 4.80 MPa, (C) 4.86 MPa, (D) 4.91 MPa (P_T), (E) 5.02 MPa, (F) 5.10 MPa and (G) 5.21 MPa.

As explained more fully below, it is proposed here that the transparency effect is caused by the reduction of the refractive index (RI) of the dispersed phase, due to dissolution of the compressed gas in the monomer. It is well known that compressed gases such as CO₂ and ethylene are highly miscible with most organic solvents.^45,46 The solubility of such gases in solvents at elevated pressure is governed primarily by the density of the gas. The density of CO₂ at 40 °C and 3.83 MPa is 0.078 g ml⁻¹ while the corresponding density for ethylene is 0.051 g ml^−1−47,48 Since the density of ethylene is much lower than that for CO₂ at the same conditions, a higher pressure is required to achieve transparency of the emulsion. Consistent with this line of reasoning, the P_T obtained using ethylene is comparable to the pressure required (5.3 MPa) to achieve the same gas density as CO₂ at 3.83 MPa.

Effect of temperature and monomer type on P_T

P_T was measured as a function of temperature using different monomers. Fig. 3 shows P_T at different temperatures for three kinds of vinyl monomers with different alkyl chain lengths: MA, MMA and LMA. Turbid emulsions were also initially formed upon the dispersion of MMA and LMA in water using the given recipes (Table 1). In all cases, P_T increases with increasing temperature, consistent with previous reports.⁴² This trend can be attributed to the decreased solubility (and density) of CO₂ in solvents at elevated temperature and constant pressure. Consequently, the greater the temperature, the higher is the pressure required to cause a reduction in the RI of the monomer phase to match that of water (RI = 1.33 for water; RI ∼1.0 for CO₂ and ethylene). The RIs of the monomers in increasing order are:⁴⁹ MA = 1.402, MMA = 1.414 and LMA = 1.445. These values are also consistent with the trends observed in P_T. At constant temperature, the higher the RI of the monomer, the higher is the pressure required to reduce the RI of the monomer phase to match that of water. In the case of LMA, although the emulsion became translucent with the addition of CO₂, a completely transparent point could not be found. An attempt was made to measure P_T for the LMA system at 60 °C, as reported for MA and MMA (Fig. 3). However, the emulsion remained somewhat turbid at a pressure up to 11.5 MPa.

Fig. 3 Dependence of P_T on temperature for different monomers (methyl acrylate (MA), methyl methacrylate (MMA), lauryl methacrylate (LMA)); monomer content: 10 wt% rel. to total; Dowfax: 10 wt% rel. to monomer.

Mechanism of miniemulsion transparency

The relationship between the RI and density (molar volume) of a pure substance is often expressed by the Lorentz–Lorenz relation:⁵⁰where v_i is the molar volume, n_i is the RI and R_i is the molar refractivity. For liquid mixtures in which there is negligible change of volume on mixing, it is assumed that the refractivities are additive. This leads to the Lorentz–Lorenz mixing rule for estimating the RI of a liquid mixture (n_mix) from its pure component properties:^51,52where Φ_i is the volume fraction of component i. This mixing rule has also been applied in the calculation of phase densities from phase composition and RI data in vapour–liquid systems involving dense CO₂.^53,54 CO₂ is poorly soluble in water, even at elevated pressure (<5 wt% at reaction conditions), and therefore has a small effect on the RI of the continuous phase based on eqn (2). However, dissolution of CO₂ in the dispersed monomer phase will significantly affect the RI. For example, it has been reported that the refractive indices of CO₂-expanded ethanol and n-pentane at 25 °C decrease by ∼10% when pressurized to 5 MPa.^53,55 Thus, RI values of the continuous and dispersed phases approach each other as the miniemulsion is pressurized, with matching values at P_T. At pressures beyond P_T, the emulsion once again becomes turbid as the RI of the dispersed phase decreases below that of water. This is consistent with experimental observations.

Experimental data on the solubility of CO₂ in water and the RI of dense CO₂, over a wide range of temperature and pressure, are available in the literature.^50,56 Using this information, in combination with the density of CO₂,⁴⁷ eqn (2) can be used to estimate the RI of the continuous phase at P_T. The results for the three vinyl monomers discussed above are summarised in Table 2. It can be seen that the presence of CO₂ in the continuous phase at P_T reduces the RI by around 3–8% compared with pure water (1.33). Note that the effect of dissolved monomer/Dowfax on RI in the continuous phase is neglected here.

Table 2 Estimated RI of miniemulsions at P_T and corresponding phase compositions

Miniemulsion	PT^a (MPa)	RIPT^b	xCO2,CP^c	xCO2,DP^d
a PT of miniemulsion obtained at 40 °C using recipe given in Table 1.b RI of miniemulsion at PT, calculated from the Lorentz–Lorenz mixing rule (eqn (2)).c Mole fraction solubility of CO2 in the continuous phase at PT.^56d Mole fraction solubility of CO2 in the dispersed phase at PT, calculated from the Lorentz–Lorenz mixing rule (eqn (2)).
MA	3.8	1.23	0.014	0.11
MMA	4.4	1.23	0.016	0.15
LMA	8.1	1.28	0.024	0.57

---

Invoking the matching RI values at P_T, eqn (2) can also be used to estimate the volume fraction of CO₂ in the dispersed phase, and hence the solubility of CO₂ in that phase. In this calculation, the dispersed phase is assumed to consist only of monomer and CO₂ (HD neglected). The results are also presented in Table 2 in terms of the mole fraction solubility of CO₂. Overall, the solubility of CO₂ in the dispersed phase is around an order of magnitude higher than that in the continuous phase.

The preceding analysis also explains the enhanced stability of the miniemulsion observed in the presence of CO₂. The attractive van der Waals forces between droplets are dictated by the Hamaker constant (A) which is given by:⁵⁷

where ε_i is the dielectric constant of phase i, n_i is RI of phase i, a = 3kT/4, b = (3hv_e)/(16(2)^0.5), k is the Boltzmann's constant, T is the absolute temperature, h is Planck's constant, and v_e is the main electronic absorption frequency in the UV. In systems with matching refractive indices, the Hamaker constant approaches zero and as such the attractive van der Waals forces between droplets are minimized.^58–62 This concept has previously been employed by deliberately creating emulsions where the dispersed/continuous liquid phases have similar RI values.^58,59

With regards to miniemulsion formation under CO₂ pressure, an additional factor may be that the viscosity of the monomer droplets decreases due to the dissolution of CO₂. This may in turn expedite break up of micron-size monomer droplets into smaller droplets by gentle stirring, thus facilitating miniemulsion formation.^63,64

Radical polymerization of MA in compressed gas-induced miniemulsions at P_T

Among the monomers considered, the lowest P_T values were obtained for MA. This monomer was therefore selected for radical polymerizations in compressed gas-induced miniemulsions. Polymerization was initiated using the water-soluble initiator VA-044 and all polymerizations were conducted at 40 °C and the corresponding P_T for each gas. For a miniemulsion system, a low molecular weight hydrophobe (HD in this study) is typically employed to minimize Ostwald ripening.^1,65 In the absence of pressurization, the polymerization would be expected to proceed based on an emulsion polymerization mechanism, i.e. particles would be generated in the continuous phase via homogeneous or micellar nucleation.⁶⁶ If pressurization results in formation of sufficiently small droplets, one would instead anticipate a miniemulsion-type mechanism, i.e. predominant monomer droplet nucleation.^1–3

In all three cases (CO₂, ethylene, and no pressurization), the polymerizations proceeded rapidly without coagulation, reaching conversions >70% in less than 1 h (Fig. 4). The droplet/particle diameters as measured by DLS (offline, i.e. depressurized samples) were very similar for all three cases, remaining well below 100 nm (Fig. 5). The molecular weights were very high for both CO₂/no pressurization (Fig. 6), consistent with a compartmentalization effect (segregation of propagating radicals) on radical termination.^67,68 The higher M_n for the system in the presence of CO₂ may indicate a stronger compartmentalization effect for that system. Due to effects of compartmentalization, a decrease in particle size leads to higher polymerization rate and higher molecular weights in an emulsion polymerization (relative to the corresponding bulk system).^66,67 Such effects are typically not observed in miniemulsion systems, unless the droplets/particles are sufficiently small.⁶⁹ In the present systems, it is apparent that the particle sizes are similar in all three cases, and as such so are the polymerization rates and molecular weights.

Fig. 4 Conversion–time plots for VA-044 initiated emulsion/miniemulsion polymerization of MA at 40 °C (no pressurization, with CO₂ at P_T, and with ethylene at P_T) based on recipe in Table 1.

Fig. 5 Number-, volume-, and intensity-average diameters (d_n, d_v and d_i) vs. conversion for VA-044 initiated emulsion/miniemulsion polymerization of MA at 40 °C (no pressurization, with CO₂ at P_T, and with ethylene at P_T) based on recipe in Table 1.

Fig. 6 Number-average molecular weight (M_n) and dispersity (Đ) vs. conversion for VA-044 initiated emulsion/miniemulsion polymerization of MA at 40 °C (no pressurization, with CO₂ at P_T) based on recipe in Table 1.

4-Stearoyl-TEMPO (0.8 wt% rel. to monomer) was added as an oil phase inhibitor to investigate the polymerization mechanism in the presence of CO₂. This inhibitor cannot diffuse from monomer droplets to micelles due to its low water solubility. Thus, an emulsion type polymerization can still proceed via homogeneous or micellar nucleation, but a miniemulsion type polymerization based on droplet nucleation would be extremely slow or completely inhibited due to the presence of inhibitor in the monomer droplets.^32,70 The conversion for polymerization in the presence of CO₂ with inhibitor at P_T after 6 h was 24%, to be compared with ∼100% conversion without inhibitor. These results are consistent with predominant monomer droplet nucleation. In the absence of CO₂, however, the conversion reached 74% in 6 h in the presence of 4-stearoyl-TEMPO, indicating a large portion of micellar/homogeneous nucleation consistent with an emulsion polymerization mechanism. These results suggest that the polymerization carried out at P_T proceeded via significant monomer droplet nucleation, i.e. predominantly, but not exclusively, via a miniemulsion polymerization mechanism. That is, the droplet size of the initial emulsion becomes smaller due to the presence of CO₂, enabling monomer droplet nucleation to be a dominant mechanism consistent with previous reports.^36,42 However, the droplet size appears not to be sufficiently small for exclusive droplet nucleation to occur. Therefore, droplets and micellar/homogeneous nucleation co-exist in the system at P_T, i.e. it is effectively a hybrid miniemulsion/emulsion polymerization system.

Radical polymerization of MA at pressures above and below the P_T

It has been proposed in previous reports that the gas pressure can be used to tune the droplet/particle size in compressed gas-induced miniemulsions.^36,37 The minimum droplet/particle size is thought to occur at the transparency pressure. In our earlier study on RAFT polymerization of styrene in CO₂-induced miniemulsions,⁴² we observed that the smallest particles were obtained at P_T, although pressures beyond this value were not considered. Since the present MA system is likely to be a hybrid miniemulsion/emulsion polymerization, the role of conversion on particle size should also be considered when examining the effect of pressure on particle size.

Conversion data for polymerization of MA in miniemulsions induced by CO₂/ethylene at various pressures are presented in Fig. 7. These data represent the conversions attained after polymerization over a fixed period of 2.5 h at 40 °C. In the case of CO₂, there is a slight rate reduction at pressures below P_T but overall the conversion data are largely unaffected by the operating pressure. Furthermore, the conversions achieved are comparable to that obtained in the absence of CO₂ (∼80%). The conversion data for ethylene are also largely independent of pressure and similar to the control experiment. The corresponding particle size measurements for both systems are presented in Fig. 8 and 9. In CO₂ (Fig. 8), average particle sizes are similar to or perhaps slightly larger than those obtained in the absence of CO₂. However, there is some indication that the particle size approaches a minimum value at P_T. This result is also evident in the data for ethylene (Fig. 9).

Fig. 7 Polymerization of MA in miniemulsions induced by CO₂ and ethylene, respectively, at 40 °C and pressures above and below P_T (recipe in Table 1). Conversion data are for a reaction time of 2.5 h. The dotted line represents the conversion obtained at 1 atm under N₂.

Fig. 8 Particle size data for polymerization of MA in CO₂-induced miniemulsions at 40 °C at pressures above and below P_T (recipe in Table 1). The dotted line represents the particle size obtained at 1 atm under N₂.

Fig. 9 Particle size data for polymerization of MA in ethylene-induced miniemulsions at 40 °C at pressures above and below P_T (recipe in Table 1). The dotted line represents the particle size obtained at 1 atm under N₂.

Conclusions

Formation of aqueous miniemulsions of various vinyl monomers by use of the compressed gases CO₂ and ethylene (less than 6 MPa), respectively, has been investigated in the absence of the traditionally employed high energy mixing. Contrary to previous work,³⁶ it is here proposed that transparent/translucent miniemulsions are formed at the transparency pressure (P_T) due to the refractive index of the dispersed phase being reduced to become approximately equal to that of the continuous phase (water) as a result of expansion with CO₂/ethylene, resulting in refractive index matched miniemulsions with enhanced stability. In the case of CO₂ (ethylene not investigated), the value of P_T increases with increasing temperature due to the decrease in solubility (and density) of CO₂ at constant pressure. Moreover, P_T can be correlated with the refractive index of the vinyl monomer – the higher the refractive index of the monomer, the higher is the CO₂ pressure required to achieve dissolution of a sufficient amount of CO₂ to generate the required reduction in refractive index of the dispersed phase. The enhanced emulsion stability observed for both gases is a consequence of minimization of the attractive forces between droplets in such refractive index matched emulsions as governed by the Hamaker constant. The value of P_T is higher for ethylene than CO₂, which is thought to be a consequence of the density of ethylene being lower than for CO₂. It follows that a higher pressure is required for ethylene, given that the solubility of such gases in solvents at elevated pressure is mainly dictated by the gas density.

Radical polymerization of methyl acrylate at 40 °C in miniemulsion induced by CO₂ at P_T has been demonstrated to proceed as a hybrid miniemulsion/emulsion polymerization system generating particles with diameters less than 100 nm. The experimental data are consistent with particle formation occurring predominantly via monomer droplet nucleation (miniemulsion polymerization) but also via secondary nucleation in the aqueous phase (emulsion polymerization). The presence of compressed gas results in a significant decrease in droplet size (compared to a macroemulsion), but the droplets do not appear to reach sufficiently small sizes in these particular cases for monomer droplet nucleation to be the sole particle formation mechanism. Polymerizations at various pressures above and below P_T revealed limited effects of pressure on both the polymerization rate and particle size for both CO₂ and ethylene.

Acknowledgements

PBZ is grateful for a Future Fellowship from the Australian Research Council. We wish to thank Prof. Jerome Claverie (Université du Québec à Montréal) for stimulating discussions and enlightening comments.

References

1. J. M. Asua, Prog. Polym. Sci., 2002, 27, 1283–1346.
2. K. Landfester, Macromol. Rapid Commun., 2001, 22, 896–936.
3. J. M. Asua, Prog. Polym. Sci., 2014, 39, 1797–1826.
4. J. Ugelstad, M. El-Aasser and J. Vanderhoff, J. Polym. Sci., Polym. Lett. Ed., 1973, 11, 503–513.
5. K. Landfester, Angew. Chem., Int. Ed., 2009, 48, 4488–4507.
6. K. Y. van Berkel and C. J. Hawker, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 1594–1606.
7. R. Udagama, E. Degrandi-Contraires, C. Creton, C. Graillat, T. F. L. McKenna and E. Bourgeat-Lami, Macromolecules, 2011, 44, 2632–2642.
8. F. Tiarks, K. Landfester and M. Antonietti, Langmuir, 2001, 17, 908–918.
9. R. H. Utama, Y. Guo, P. B. Zetterlund and M. H. Stenzel, Chem. Commun., 2012, 48, 11103–11105.
10. H. de Brouwer, J. G. Tsavalas and F. J. Schork, Macromolecules, 2000, 33, 9239–9246.
11. M. F. Cunningham, Prog. Polym. Sci., 2002, 27, 1039–1067.
12. T. Nakamura and M. Okubo, Macromolecules, 2007, 40, 8663–8672.
13. R. W. Simms and M. F. Cunningham, Macromolecules, 2007, 40, 860–866.
14. P. B. Zetterlund, Y. Kagawa and M. Okubo, Chem. Rev., 2008, 108, 3747–3794.
15. P. B. Zetterlund, Y. Saka and M. Okubo, Macromol. Chem. Phys., 2009, 210, 140–149.
16. J. K. Oh, F. Perineau, B. Charleux and K. Matyjaszewski, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 1771–1781.
17. Y. Guo, J. Liu and P. B. Zetterlund, Macromolecules, 2010, 43, 5914–5916.
18. W. Li, J. A. Yoon and K. Matyjaszewski, J. Am. Chem. Soc., 2010, 132, 7823–7825.
19. M. A. M. Oliveira, C. Boyer, M. Nele, J. C. Pinto, P. B. Zetterlund and T. P. Davis, Macromolecules, 2011, 44, 7167–7175.
20. P. B. Zetterlund, S. C. Thickett, S. Perrier, E. Bourgeat-Lami and M. Lansalot, Chem. Rev., 2015, 115, 9745–9800.
21. P. Fernandez, V. André, J. Rieger and A. Kühnle, Colloids Surf., A, 2004, 251, 53–58.
22. L. Spernath and S. Magdassi, Polym. Adv. Technol., 2007, 18, 705–711.
23. L. Yu, C. Li, J. Xu, J. Hao and D. Sun, Langmuir, 2012, 28, 14547–14552.
24. I. Solè, A. Maestro, C. González, C. Solans and J. M. Gutiérrez, Langmuir, 2006, 22, 8326–8332.
25. S. Sajjadi, Langmuir, 2006, 22, 5597–5603.
26. S. Cheng, Y. Guo and P. B. Zetterlund, Macromolecules, 2010, 43, 7905–7907.
27. T. Nishimi and C. A. Miller, Langmuir, 2000, 16, 9233–9241.
28. N. I. Prokopov and I. A. Gritskova, Russ. Chem. Rev., 2001, 70, 791–800.
29. O. Saygi-Arslan, E. D. Sudol, E. S. Daniels, M. S. El-Aasser and A. Klein, J. Appl. Polym. Sci., 2009, 111, 735–745.
30. Y. Guo, V. L. Teo, S. R. S. Ting and P. B. Zetterlund, Polym. J., 2012, 44, 375–381.
31. Y. Guo and P. B. Zetterlund, Macromol. Rapid Commun., 2011, 32, 1669–1675.
32. Y. Guo and P. B. Zetterlund, Polymer, 2011, 52, 4199–4207.
33. S. R. S. Ting, E. H. Min and P. B. Zetterlund, Aust. J. Chem., 2011, 64, 1033–1040.
34. U. El-Jaby, M. Cunningham and T. F. L. McKenna, Macromol. Chem. Phys., 2010, 211, 1377–1386.
35. U. El-Jaby, M. Cunningham and T. F. L. McKenna, Macromol. Rapid Commun., 2010, 31, 558–562.
36. J. Zhang, B. Han, C. Zhang, W. Li and X. Feng, Angew. Chem., Int. Ed., 2008, 47, 3012–3015.
37. Y. Zhao, J. Zhang, W. Li, C. Zhang and B. Han, Chem. Commun., 2009, 2365–2367.
38. P. Y. Chow and L. M. Gan, Microemulsion polymerizations and reactions, in Polymer Particles, ed. M. Okubo, 2005, vol. 175, pp. 257–298.
39. K. L. Thompson, J. A. Lane, M. J. Derry and S. P. Armes, Langmuir, 2015, 31, 4373–4376.
40. J. Zou, C. C. Hew, E. Themistou, Y. K. Li, C. K. Chen, P. Alexandridis and C. Cheng, Adv. Mater., 2011, 23, 4274–4277.
41. J. Zhang, Y. Zhao, J. Li, G. Yang, B. Han, Z. Wu and Z. Li, Soft Matter, 2010, 6, 6200–6205.
42. S. Cheng, S. R. S. Ting, F. P. Lucien and P. B. Zetterlund, Macromolecules, 2012, 45, 1803–1810.
43. Y. Guo, M. E. Tysoe and P. B. Zetterlund, Polym. Chem., 2013, 4, 3256–3264.
44. D. Pu, F. P. Lucien and P. B. Zetterlund, J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 4307–4311.
45. A. Kordikowski, A. P. Schenk, R. M. V. Nielen and C. J. Peters, J. Supercrit. Fluids, 1995, 8, 205–216.
46. P. G. Jessop and B. Subramaniam, Chem. Rev., 2007, 107, 2666–2694.
47. R. Span and W. Wagner, J. Phys. Chem. Ref. Data, 1996, 25, 1509–1596.
48. M. Jahangiri, R. T. Jacobsen and R. B. Stewart, J. Phys. Chem. Ref. Data, 1986, 15, 593–734.
49. C. Wohlfarth, Refractive Indices of Inorganic, Organometallic, and Organononmetallic Liquids, and Binary Liquid Mixtures, Springer, 1996.
50. G. J. Besserer and D. B. Robinson, J. Chem. Eng. Data, 1973, 18, 137–140.
51. R. Mehra, J. Chem. Sci., 2003, 115, 147–154.
52. A. Z. Tasic, B. D. Djordjevic, D. K. Grozdanic and N. Radojkovic, J. Chem. Eng. Data, 1992, 37, 310–313.
53. G. J. Besserer and D. B. Robinson, J. Chem. Eng. Data, 1973, 18, 416–419.
54. G. J. Besserer and D. B. Robinson, Can. J. Chem. Eng., 1971, 49, 651–656.
55. M. T. Spuller, Resist and Residue Removal Using Gas-Expanded Liquids, Georgia Institute of Technology, 2003.
56. J. Kiepe, S. Horstmann, K. Fischer and J. Gmehling, Ind. Eng. Chem. Res., 2002, 41, 4393–4398.
57. P. C. Hiemenz, Principles of Colloids and Surface Chemistry, Marcel Dekker Ltd, New York, 2nd edn, 1985.
58. W. C. K. Poon, J. Phys.: Condens. Matter, 2002, 14, R859–R880.
59. D. J. Hibberd, A. R. Mackie, G. K. Moates, R. Penfold, A. D. Watson and G. C. Barker, Colloids Surf., A, 2007, 301, 453–461.
60. Y. Yan, H. Hoffmann, A. Makarsky, W. Richter and Y. Talmon, J. Phys. Chem. B, 2007, 111, 6374–6382.
61. A. Song, K. Reizlein and H. Hoffmann, Swelling of Aqueous L(alpha)-Phases by Matching the Refractive Index of the Bilayers with that of the Mixed Solvent, in Surface and Interfacial Forces - from Fundamentals to Applications, ed. G. K. Auernhammer, H. J. Butt and D. Vollmer, Springer-Verlag, 2008, vol. 134, pp. 111–119.
62. J. N. Israelachvili, Intermolecular and surface forces, Academic press, USA, 3rd edn, 2011.
63. R. Sih, M. Armenti, R. Mammucari, F. Dehghani and N. R. Foster, J. Supercrit. Fluids, 2008, 43, 460–468.
64. R. Sih, F. Dehghani and N. R. Foster, J. Supercrit. Fluids, 2007, 41, 148–157.
65. P. Taylor, Adv. Colloid Interface Sci., 1998, 75, 107–163.
66. S. C. Thickett and R. G. Gilbert, Polymer, 2007, 48, 6965–6991.
67. R. G. Gilbert, Emulsion polymerization: a mechanistic approach, Academic Pr, London, 1995.
68. P. B. Zetterlund and M. Okubo, Macromolecules, 2006, 39, 8959–8967.
69. N. Bechthold and K. Landfester, Macromolecules, 2000, 33, 4682–4689.
70. H. Maehata, X. Liu, M. Cunningham and B. Keoshkerian, Macromol. Rapid Commun., 2008, 29, 479–484.

---