In situ small-angle X-ray scattering studies during the formation of polymer/silica nanocomposite particles in aqueous solution

This study is focused on the formation of polymer/silica nanocomposite particles prepared by the surfactant-free aqueous emulsion polymerization of 2,2,2-trifluoroethyl methacrylate (TFEMA) in the presence of 19 nm glycerol-functionalized aqueous silica nanoparticles using a cationic azo initiator at 60 °C. The TFEMA polymerization kinetics are monitored using 1H NMR spectroscopy, while postmortem TEM analysis confirms that the final nanocomposite particles possess a well-defined core–shell morphology. Time-resolved small-angle X-ray scattering (SAXS) is used in conjunction with a stirrable reaction cell to monitor the evolution of the nanocomposite particle diameter, mean silica shell thickness, mean number of silica nanoparticles within the shell, silica aggregation efficiency and packing density during the TFEMA polymerization. Nucleation occurs after 10–15 min and the nascent particles quickly become swollen with TFEMA monomer, which leads to a relatively fast rate of polymerization. Additional surface area is created as these initial particles grow and anionic silica nanoparticles adsorb at the particle surface to maintain a relatively high surface coverage and hence ensure colloidal stability. At high TFEMA conversion, a contiguous silica shell is formed and essentially no further adsorption of silica nanoparticles occurs. A population balance model is introduced into the SAXS model to account for the gradual incorporation of the silica nanoparticles within the nanocomposite particles. The final PTFEMA/silica nanocomposite particles are obtained at 96% TFEMA conversion after 140 min, have a volume-average diameter of 216 ± 9 nm and contain approximately 274 silica nanoparticles within their outer shells; a silica aggregation efficiency of 75% can be achieved for such formulations.

Laboratory-scale synthesis of PTFEMA/silica nanocomposite particles AIBA initiator (74.8 mg; 1.0 mol % based on TFEMA), Bindzil CC401 silica sol (4.71 g of a 40% w/w dispersion, or 1.89 g dry silica) and deionized water (55.3 mL) were weighed into a 100 mL round-bottom flask containing a magnetic stirrer bar. The reaction mixture was adjusted to pH 8.9 by addition of 200 μL of a 0.1 M NaOH solution and then degassed with N2 gas for approximately 30 min. After removing its MEHQ inhibitor, cold TFEMA was degassed separately using N2 gas for 30 min with the aid of an ice bath. Degassed TFEMA (4.53 g) was then added to the reaction mixture and a 0.1 mL aliquot was immediately extracted for 1 H NMR spectroscopy analysis. The reaction solution was degassed for a further 5 min prior to immersion in a 60 o C oil bath and stirred magnetically at 800 rpm. The 'zero time' (t = 0 min) for this polymerization was arbitrarily taken to be the point when the degassed reaction solution was first immersed in the oil bath, rather than the time at which the reaction solution had reached this temperature. Aliquots were subsequently removed under N2 via syringe at various time intervals for 1 H NMR, DLS and TEM analysis. Each 1.0 mL aliquot was quenched by cooling using an ice bath with concomitant exposure to air. For 1 H NMR analysis, 40 μL of each aliquot was diluted using CDCl3 (650 μL). A small quantity of anhydrous MgSO4 was added to remove water and each deuterated solution was then passed through a cotton woolplugged pipette to remove any solids. For both TEM and DLS analysis, each aliquot was diluted fifty-fold using deionized water at 20 o C to produce 0.20% w/w dispersions.
In situ SAXS studies of nanocomposite particle formation using the stirrable reaction cell AIBA initiator (3.0 mg; 1.0 mol % based on TFEMA), Bindzil CC401 silica sol (0.187 g of a 40% w/w dispersion, or 74.7 mg dry silica) and deionized water (2.21 mL) were weighed into a 14 mL sample vial. The reaction mixture was adjusted to pH 8.9 by addition of 10 μL of a 0.1 M NaOH solution and then degassed with N2 gas for approximately 30 min. After removing its MEHQ inhibitor, cold TFEMA was degassed separately using N2 gas for 30 min with the aid of an ice bath. Degassed TFEMA (0.18 g) was added to the reaction mixture, which was then transferred via degassed syringe to the stirrable reaction cell (already containing a magnetic flea and equipped with a magnetic stirrer unit) which had been separately purged with N2 gas for 20 min. This cell was then attached to the sample stage in I22 and aligned relative to the X-ray beam and TFEMA polymerization was initiated by using a water-circulating jacket to heat the cell up to 60 o C as the X-ray beam shutter was opened. The TFEMA polymerization was monitored until no further evolution in the 1D SAXS pattern was observed, at which point it was assumed that the reaction was complete.
In situ conductivity studies during the aqueous emulsion polymerization of TFEMA in the presence glycerol-functionalized silica nanoparticles The experimental protocol employed for determining the solution conductivity during the synthesis of silica-stabilized PTFEMA latex particles was as follows. A 100 mL two-neck roundbottom flask was fitted with a Primo 5 conductivity probe. Bindzil CC401 (6.41 g of a 40% w/w dispersion, or 2.56 g dry silica) and deionized water (70 mL) were weighed into the 100 mL round-bottom flask. The reaction mixture was adjusted to pH 8.9 by adding 0.1 M NaOH solution (100 μL) via micropipette and then degassing with N2 gas at 60 °C for approximately 30 min. AIBA initiator (0.102 g) and deionized water (5 mL) were weighed into a 14 mL sample vial and degassed with N2 gas for approximately 30 min. After removing its MEHQ inhibitor, cold TFEMA was degassed separately using N2 gas for 30 min with the aid of an ice bath. Degassed TFEMA (6.14 g) was added to the reaction mixture and allowed to stir at 500 rpm for 3 min. The degassed aqueous AIBA solution was then added to the reaction mixture. The 'zero time' (t = 0 min) for this polymerization was arbitrarily taken to be the point when the degassed initiator solution was added to the reaction mixture. Conductivity data from the Primo 5 were recorded using the camera facility on a mobile phone. 1 H NMR analysis indicated that a final TFEMA monomer conversion of 98% was achieved during this synthesis. S4

H NMR Spectroscopy
All NMR spectra were recorded at 298 K using a 400 MHz Bruker Avance-400 spectrometer (64 scans averaged per spectrum).

Dynamic Light Scattering (DLS)
DLS studies were conducted on 0.20% w/w aqueous dispersions at 25 °C in disposable plastic cuvettes using a Malvern Zetasizer NanoZS instrument that detects back-scattered light at an angle of 173°. Intensity-average hydrodynamic diameters were calculated via the Stokes−Einstein equa on using a non-negative least squares (NNLS) algorithm. All data were averaged over three consecutive runs. Intensity-average hydrodynamic diameters were converted into volume-average hydrodynamic diameters using the Malvern Zetasizer Software (Version 7.01).

Transmission Electron Microscopy (TEM)
Removed aliquots were diluted fifty-fold at 20 °C to generate 0.20% w/w dispersions. Copper/palladium TEM grids (Agar Scientific, UK) were surface-coated in-house to yield a thin film of amorphous carbon. The grids were then plasma glow-discharged for 30 s to create a hydrophilic surface. Individual samples (0.20% w/w, 5 μL) were adsorbed onto the freshly glow-discharged grids for 1 min and then blotted with filter paper to remove excess solution. To stain the aggregates, uranyl formate solution (0.75% w/v, 5 μL) was soaked on the sampleloaded grid for 20 s and then carefully blotted to remove excess stain. The grids were then dried using a vacuum hose. Imaging was performed on a Technai T12 Spirit instrument at 120 kV equipped with a Gatan 1 k CCD camera.

Thermogravimetric Analysis (TGA)
Thermogravimetric analyses were conducted using a Perkin-Elmer Pyris 1 TGA instrument. Excess non-adsorbed silica nanoparticles remaining after the TFEMA polymerization was removed via five centrifugation-redispersion cycles (5,000 rpm for 20 min), with each successive supernatant being carefully decanted and replaced with deionized water. TEM studies confirmed that excess silica nanoparticles were removed using this purification protocol. Higher centrifugation rates (> 8,000 rpm) and longer times (> 1 h) were avoided since these resulted in partial sedimentation of the non-adsorbed silica nanoparticles and also hindered redispersion of the sedimented nanocomposite particles. Purified nanocomposite dispersions were dried in an oven at 75 °C for 24 h and freeze-dried overnight remove unreacted TFEMA and water. Using the Perkin-Elmer TGA instrument, dried samples were heated in air up to 800 °C at a heating rate of 10 °C min -1 . The observed mass loss was attributed to the quantitative degradation of PTFEMA, with the remaining white incombustible residues being assumed to be that of pure silica (SiO2). The silica aggregation efficiency can be calculated from the residual mass, see Section 2.3.

Optical Microscopy
Aliquots were extracted from the reaction mixture at 60 °C and the TFEMA polymerization was quenched by cooling to 20 °C with concomitant exposure to air. Optical microscopy images were recorded immediately at 20 °C using a Motic DMBA300 digital biological microscope equipped with a built-in camera and Motic Images Plus 2.0 ML software.

Small-Angle X-ray Scattering (SAXS)
SAXS patterns were recorded at a synchrotron facility (station I22 at Diamond Light Source, Didcot, Oxfordshire, UK). A monochromatic X-ray beam (λ = 0.124 nm) and a 2D Pilatus 2M pixel detector (Dectris, Baden, Switzerland), were used for these experiments. A q range of 0.02−2.00 nm −1 was used for measurements, where q = (4π sin θ)/λ, corresponds to the modulus of the scattering vector and θ is half of the scattering angle. For these time-resolved measurements, a custom-designed SAXS cell was used as the sample holder, see Figure 1 in the main text. SAXS patterns were recorded every 60 seconds for 60 min and every 150 seconds thereafter until no further change in the SAXS patterns could be observed. X-ray scattering data were reduced (integrated, normalized, and background-subtracted) using Dawn software supplied by Diamond Light Source. The scattering intensity of water was used for absolute scale calibration of the X-ray scattering patterns. Irena SAS macros for Igor Pro were utilized for modelling and further SAXS analysis.

Calculating Silica Aggregation Efficiency from TGA
The silica nanoparticles used in this work lose 3.8% mass on heating to 800 °C in air, see Figure  S1a. This is attributed to a (i) surface moisture and (ii) pyrolysis of surface glycerol groups, which are present for this particular commercial grade. This mass loss must be taken into account when calculating the silica content of the nanocomposite particles. In order to account for this, the residual mass as determined by TGA for the purified nanocomposite particles must be divided by the silica weight remaining (96.2) divided by 100: where and are the mass fractions of the purified nanoparticles and silica particles determined by TGA (see Figure S1). Figure S1b shows the TGA curve recorded for the purified nanocomposite particles retrieved from the stirrable reaction cell after the in situ SAXS experiment (i.e., after the excess non-adsorbed silica nanoparticles were removed by five centrifugation-redispersion cycles). A residual mass of 21.9% was determined by TGA for the purified nanocomposite particles , thus s = 22.8%.
With the modified silica mass fraction of the nanocomposite particles calculated, the silica aggregation efficiency can be determined using: where and are the initial masses of monomer and silica respectively, is the TFEMA conversion, and is the modified silica mass fraction of the nanocomposite particles. At the end of the TFEMA polymerization, the silica aggregation efficiency can be calculated as follows: volume fraction of monomer within the growing nanocomposite particle cores (black squares) obtained from SAXS data fits using the population balance model.

Scattering Model for SAXS analysis
As outlined in the main text, a three-population scattering model is used to fit the in situ SAXS patterns recorded for the PTFEMA/silica nanocomposite particles, which acquire a welldefined core-shell morphology after a certain time point.
The scattered intensity I(q) is measured as a function of the scattering vector length = (4 ⁄ ) sin , where 2 is the scattering angle. After subtraction of the solvent background, the scattered intensity (or differential scattering cross-section per unit sample volume) for a colloidal dispersion can be approximately expressed as: where n is the number of types of particles present in the dispersion, ( ) is the structure factor arising from interparticle interactions, is the number density of scattering particles of the th population per unit sample volume, ( , ) is the scattering amplitude that describes the particle morphology, and Ψ ( ) is the size distribution function of scattering particles corresponding to the i th population.
The silica shell within the nanocomposite particles is particulate in nature and excess (nonadsorbed) silica nanoparticles are present in the aqueous continuous phase throughout polymerization. Thus the scattering intensity [eq (S2)] is best analyzed using a threepopulation model (n = 3). The first population (i = 1) describes the core/shell morphology of the PTFEMA/silica nanocomposite particles. However, this merely assumes a homogenous distribution of electron density within the shell layer. Thus, a second population (i = 2) describing the particulate nature of this shell is also required. 1 The third population (i = 3) is used to account for the gradual reduction in concentration of the non-adsorbed silica S21 nanoparticles that remain within the aqueous continuous phase as the TFEMA polymerization progresses.
The following functions and parameters were used for the three-population scattering model.

Population 1 -Core/shell nanocomposite particles
The scattering amplitude of the core/shell nanocomposite particles can be described as follows: where ( ) = 4 3 , ( ) = 3(sin + cos ) ⁄ is the normalized scattering amplitude of a homogeneous sphere and is the shell thickness. , and and are the scattering length densities of the medium, the PTFEMA core and the silica shell, respectively. Since the shell is composed of water molecules and silica nanoparticles, the (averaged) scattering length density of the shell can be expressed as: where is the scattering length density of the silica nanoparticles, is the scattering length density of water, and silica is the packing efficiency of the silica nanoparticles within the shell, which is defined as: where vsilica is the relative volume fraction of silica in the system, i.e. both free silica nanoparticles [(1-Ae)vsilica] and that which forms the silica shell (Aevsilica), Ae is the fraction of S22 adsorbed silica nanoparticles (or the silica aggregation efficiency), and vsol-shell is the relative volume fraction of water within the silica shell.
Initially, the TFEMA monomer mostly resides in relatively large (ca. 50 μm diameter) monomer droplets, with only a small fraction dissolved within the aqueous continuous phase.
However, after nucleation, monomer diffuses into the growing particles to solvate the PTFEMA chains and enable the polymerization to continue. Therefore, the scattering length density of the latex core (equation S6) is defined by the scattering length densities of the monomer and polymer and is described as: where is the volume fraction of TFEMA in the polymer cores. It was also assumed that the water content within such hydrophobic cores is negligible. Because the aqueous solubility of TFEMA monomer is relatively low, it makes no contribution to the scattering length density of the aqueous continuous phase ( med ). Thus To a good approximation, the polydispersity of the core/shell nanocomposite particles can be described by the polydispersity of the PTFEMA latex core radius. This is expressed in terms of a Gaussian distribution: where is the mean latex core radius and is its corresponding standard deviation.
Clearly, the mass of silica nanoparticles and the total mass of TFEMA/PTFEMA both remain constant during polymerization. Thus several constraints can be incorporated in the model.

S23
The ratio between the total volume of the shell, ∫ [ ( + ) − ( )]Ψ ( ) d ∞ , and the total volume of the core, ∫ ( )Ψ ( ) d ∞ , can be expressed by the ratio of volume fractions of the components comprising the particle shell and core: where vmon is the initial volume fraction of the monomer, conv is the monomer conversion, and and are the densities of TFEMA monomer and PTFEMA at the synthesis temperature, respectively. The left-hand term in equation S9 can be approximated by the ratio of the average shell volume, Vsh-avg, and the average core volume, Vco-avg, and hence rewritten as: Thus the shell thickness can be expressed using equation S10 via the fraction of aggregated silica nanoparticles: The number density of the core/shell nanocomposite particles, N1, for equation S2 can be expressed as: Owing to the relatively high targeted solids concentration (10% w/w), it is necessary to account for interactions between the core/shell particles. Thus a hard-sphere structure factor (solved using the Percus-Yevick closure relation), S1(q, DcsPY, csPY), was incorporated in the scattering equation (equation S2), where DcsPY is the distance between the centres of the core/shell nanocomposite particles and csPY is the effective volume fraction of these particles.

Population 2 -Silica particles located within the nanocomposite shell
The form factor for the adsorbed silica nanoparticles located within the shell, , can be described by the following equation: where ( ) corresponds to the silica nanoparticle volume. The Gaussian size distribution of the silica nanoparticles within the shell in equation S2 is described by the following expression: where is the mean silica nanoparticle radius and is the standard deviation. Both parameters are determined from the first frame of the time-resolved SAXS patterns recorded during synthesis ( Figure S4).
The number density of the silica nanoparticles, N2, for equation S2 can be expressed as: A structure factor, S2(q, DsPY, sPY), in equation S2 accounts for the relatively dense packing of the silica nanoparticles within the shell. This S2 term corresponds to a hard-sphere structure factor that is solved using the Percus-Yevick closure relation, where DsPY is the distance between the centres of the silica nanoparticles and sPY is the effective volume fraction of packed silica nanoparticles.
Population 3 -Non-adsorbed excess silica nanoparticles located within the aqueous phase The scattering amplitude of the non-adsorbed (free) silica nanoparticles located throughout the aqueous phase can be described as follows: where ( ) corresponds to the silica nanoparticle volume. The Gaussian size distribution of the silica nanoparticles is described in equation S2 by the same parameters used for population 2 (see equation S14): The number density of silica nanoparticles located throughout the continuous phase is described in equation S2 as: Because the concentration of excess silica nanoparticles is relatively low, it is assumed that there are no interparticle interactions, hence S3(q) = 1 in equation S2.

S26
Number of silica nanoparticles located within the shell of the core-shell nanocomposite particles The mean number of silica nanoparticles (Ps) can be calculated from the number density of adsorbed silica nanoparticles (equation S15) and the number density of core/shell nanocomposite particles (equation S12): If the ratio of the total PTFEMA core volume to the total volume of the aggregated silica nanoparticles can be approximated by the cube of the ratio of the mean PTFEMA core radius to the mean silica nanoparticle radius, then the mean number of silica nanoparticles within the shell can be calculated using:

SAXS analysis
During SAXS analysis, it is assumed that the silica volume fraction (vsilica), the silica nanoparticle radius and polydispersity (Rsilica, σsilica), the total mass of TFEMA and PTFEMA, and the mass densities and scattering length densities of the various components [water (ρ H2O and ), monomer (ρmon and ), polymer (ρpol and ) and silica (ρsilica and )] all remain constant during the polymerization. For the first SAXS pattern recorded prior to any polymerization, the scattering is dominated by the silica nanoparticles because the TFEMA monomer droplets are so large they lie outside of the experimentally accessible q range ( Figure S4). Thus this initial scattering pattern was used to determine Rsilica, σsilica and vsilica. Here only population 3 was included in the fitting process and Ae = 0 (i.e. all silica S27 nanoparticles are located within the aqueous continuous phase). The last frame recorded during the time-resolved SAXS experiment corresponds to the end of the TFEMA polymerization: this was used to define the initial relative volume fraction of TFEMA monomer (vmon) assuming a final conversion (conv) of 96%, as determined by postmortem 1 H NMR studies. All three populations in equation S2 were included during the fitting process, see Figure S5. Thus, all the intermediate SAXS patterns were analyzed using ten independent fitting parameters (Rc, σc, DcsPY, csPY, , conv, Ae, silica, DsPY, sPY) and the twelve fixed parameters summarized in Table S1. Our approach was to analyze the penultimate frame and continue backwards in terms of the reaction time until a SAXS pattern was identified for which the three-population model was no longer appropriate. This three-population model is not , , mon , silica , , silica , , mon , pol (from equation S11) , silica , , mon , silica , , , mon , pol (from equation S20) S28   Table S1. Summary of the twelve variables that are assumed to remain constant during the time-resolved SAXS experiment described in this study.   Table S2. Summary of the sixteen variables used for data fits to the time-resolved SAXS patterns recorded in this study.

Constant variable
Fitted variable Symbol PTFEMA core radius Rc PTFEMA core radius standard deviation σc Distance between the centers of the silica nanoparticles DcsPY Volume fraction of core/shell nanoparticles csPY Volume fraction of TFEMA monomer within the latex core TFEMA monomer conversion conv Aggregation efficiency of silica nanoparticles Ae Packing density of silica nanoparticles within the silica shell