New insights into the nucleation and growth of PS

CREATED USING THE RSC ARTICLE TEMPLATE (VER. 2.1) - SEE WWW.RSC.ORG/ELECTRONICFILES FOR DETAILS

ARTICLE

www.rsc.org/softmatter | Soft Matter

New insights into the nucleation and growth of PS nodules on silica nanoparticles by 3D cryo-electron tomography Jean-Christophe Taveau,a David Nguyen,b,c Adeline Perro,b,c Serge Ravaine,c Etienne Duguetb and Olivier Lambert*a 5

10

15

Receipt/Acceptance Data [DO NOT ALTER/DELETE THIS TEXT] Publication data [DO NOT ALTER/DELETE THIS TEXT] DOI: 10.1039/b000000x [DO NOT ALTER/DELETE THIS TEXT] The nucleation and growth of polystyrene (PS) nodules on 170-nm silica seeds in emulsion polymerization conditions have been investigated for the first time by cryo-electron tomography. 3D arrangements were reconstructed from samples collected at several polymerization times (from 5 to 120 min). Early samples display the presence of small PS nodules bounded to silica particles in random distribution. For longer polymerization times, the number of PS nodules per silica seed decreases to lead to octopod-like morphologies. The tomographic method allowed to measure the contact angle which the growing PS nodules form with the silica bead surface. The average value of 142.4° remains constant all over the observed period of the polymerization reaction. This contact angle appeared to be one of the key parameters for controlling the morphology of PS/silica biphasic particles.

Introduction 20

25

30

35

40

In the last few decades, the interest for hybrid organicinorganic particles that combine organic and inorganic parts has increased considerably due to the potential benefits of these nanoobjects in multiple areas of material science. 1 Biphasic nanoparticles made of a metal oxide core and organic polymer shell are of particular interest. 2 In most cases, such core-shell particles are prepared according an emulsion process route allowing confining the polymerization reaction onto the surface of the inorganic cores, which act as seeds. In order to improve its affinity with regard to the surfactant and monomer molecules, the inorganic surface needs to be previously modified by the adsorption or the grafting of promoting molecules.2 In the case of particles made of polystyrene (PS) and silica cores, we have reported that, when the surface grafting density of these promoting molecules is low, PS nodules are able to nucleate onto the surface, but not to encapsulate the inorganic core by a homogeneous coating of constant thickness. In fact, PS nodules grow from the surface leading to original raspberrylike morphologies. 3 More interestingly, we have demonstrated that the number of PS nodules per silica seed may be controlled through experimental parameters, such as the silica seed concentration. 4 Those multipod-like morphologies were observed after several tens of minutes after the polymerization initiation, i.e. at relatively high monomer conversion. a Molecular Imaging and Nano-Bio-Technology, IECB, UMR-CNRS 5248, Université Bordeaux-1, 2 rue Robert Escarpit, F-33607 Pessac (France) Fax:+33 540 003 484; Tel: +33 540 003 490; E-mail: [email protected] b ICMCB, CNRS, Université Bordeaux-1, 87 avenue du Dr Albert Schweitzer, F-33608 Pessac Cedex (France). Fax: +33 540 002 761; Tel: +33 540 002 651; E-mail: [email protected] c Centre de Recherche Paul Pascal, Avenue du Dr Albert Schweitzer, F33600 Pessac (France). Fax: +33 556 845 600; Tel: +33 556 845 667; E-mail: [email protected]

This journal is © The Royal Society of Chemistry [year]

45

50

55

60

65

70

75

Conventional transmission electron microscopy (TEM) appeared to be a suitable technique for their morphology characterization even if sometimes (i) the single 2D projection collected for each object is difficult to correlate to its real 3D arrangement and (ii) the drying process of the aqueous dispersion disturbs the 3D arrangement of the PS nodules in particular when they are too soft. This last drawback was particularly critical at low monomer conversions, when the PS nodules are highly swollen of non-polymerized monomer. These nascent nodules were not easily imaged by conventional TEM. Thus a more sophisticated electron microscopy technique can make a breakthrough in the understanding of their nucleation and growth processes. Cryo-transmission electron microscopy (cryo-TEM) is a suitable method for the preservation of soft or fragile structures by maintaining the specimen in hydrated state with its embedding within a fine ice layer. 5 This approach developed for investigating the structure of biological complexes emerges as a powerful technique for characterizing functionalized nanomaterials. 6 Three-dimensional analysis of nanometer-scale structures is possible with electron microscope and tomography methods, a leading method for 3D imaging single objects. 7 In materials science, this approach was promoted by the studies of mesoporous materials, 8 nanocomposite particles, 9 carbon nanotubes, 10 dispersions of silica particles in natural rubber. 11 3D morphology and chemical mapping were retrieved from the 3D reconstruction. With the recent development of automated data acquisition, the tomographic methods are applied to specimen in frozen hydrated state. Cryoelectron tomography (cryo-ET) combining cryo-TEM with tomographic methods has been developed genuinely for biological specimen 12 and is thus a suitable method for investigating the structure of sensitive nanomaterials.

Soft Matter, [year], [vol], 00–00 | 1


80

85

90

In order to get insight into the growth of PS nodules on silica submicronic particles, images of early morphologies were recorded by freezing samples at short polymerization times. The present paper describes an overview of the nucleation and growth of the PS nodules on 170 nm silica seeds in experimental conditions in which octopod-like biphasic particles are mainly obtained (i.e. 8 PS nodules bounded to each silica seed). The 3D arrangements were determined by cryo-TEM combined with a tomographic approach. Their evolution with time was correlated to known growth phenomena in conventional emulsion polymerization processes, i.e. non-seeded processes. Original data, such as the contact angle which the growing PS nodules form in contact with the silica surface, were extracted from the tomographic reconstructions.

140

145

150

Experimental 95

100

105

110

115

120

125

130

135

Materials The reagents for synthesis of the Stöber silica sol: absolute ethanol (VWR), ammonium hydroxide (28-30%, JT Baker) and tetraethyl orthosilicate (TEOS, Fluka) were of analytical grade. Styrene (99% purity, inhibited with 4-tertbutylcatechol, Aldrich) was distilled under vacuum at 55°C. Methacryloxymethyltrimethoxysilane (MMS, A.B.C.R) and sodium peroxodisulfate (Fluka) were used without further purification. The nonylphenol poly(oxyethylenic) non-ionic surfactant (Synperonic NP30, M W= 1540 g.mol-1, Fluka) was of analytical grade and used as received. Ultrapure water (18.2 MOhm.cm at 25°C) was obtained with a Direct-Q3 system (Millipore). Synthetic procedures First, 170 nm silica particles were synthesized according to the well-known Stöber procedure. 13 250 mL of absolute ethanol and 17.5 mL of ammonia were introduced in a threeneck round flask of 500 mL. The mixture was stirred at 400 rpm to homogenize. Then 8.5 mL of TEOS were added into the solution and the reaction occured at room temperature under continuous stirring. The functionalisation of the silica beads was carried out by adding MMS directly to the particles dispersion. The MMS amount was previously calculated such as their surface density on silica particles was assumed to be nominally 0.1 molecule.nm-2 (i.e. 0.17 µ mol.m-2). After the mixture was stirred for 5 h at room temperature, the reaction was heated at 90°C for 1 h in order to promote covalent bonding. 14 When the reaction was completed, the main part of ethanol and ammonia was removed through evaporation under reduced pressure. The silica suspension was then dialyzed against water until neutral pH in order to remove the remaining reactants and replace ethanol by water. Next, emulsion polymerization of styrene (100 g.L-1) was performed in a three-neck round flask of 250 mL equipped with a refrigerating system in the presence of MMSfunctionalized silica beads (10 g.L-1) and NP30 (3 g.L-1). This surfactant concentration was fixed in order to promote the growth of 250 nm latex particles when polymerization was completed. The mixture was purged with nitrogen and heated to 70°C before addition of sodium peroxodisulfate (0.5% wt 2 | Soft Matter, [year], [vol], 00–00

155

160

165

170

175

relative to monomer) to initiate the reaction. This iniator addition is considered as time zero. 5-, 10-, 20-, 30-, 40-, 50and 120-minutes samples were collected with a syringe under nitrogen current. These samples (10 mL) were immediately mixed with 1 mL of a solution of hydroquinone (1 g.L-1) for quenching the polymerization reaction and stored at 4°C. The monomer-to-polymer conversions were determined gravimetrically after drying a known volume of the sample in an aluminium dish to constant weight at 80°C. The average diameter (Dn)TEM of the PS nodules was measured directly from the transmission electron micrograph. TEM experiments were performed with a JEOL 2000 FX microscope (accelerating voltage of 200 kV). Samples were diluted 100 times in ethanol and one drop was deposited on a copper grid coated with a carbon membrane and liquid was let to evaporate. The number of PS nodules per unit volume of water Np was calculated by the following equation : Np = [M - M silica] * 1021 / [(π/6) (Dn)TEM3 ρ] where M (g) is the total mass of solid, M silica (g) is the mass of silica particles used as seeds, ρ is the PS density (1.01 g.cm-3) and (Dn)TEM is expressed in nm. Cryoelectron microscopy and 3D tomography A 5 µ L sample of the aqueous dispersion of the biphasic particles was deposited onto a holey carbon coated copper grid. The liquid excess was blotted with a filter paper and the grid was plunged into a liquid ethane bath cooled with liquid nitrogen (Leica EM CPC). Specimens were maintained at a temperature of approximately –170 °C, using a cryo holder (Gatan), and were observed with a FEI Tecnai F20 electron microscope operating at 200 kV and at a nominal magnification of 19000X under low-dose conditions. Images were recorded with a 2Kx2K low scan CCD camera (Gatan). For cryo-ET, tilt-series were collected automatically from -60º to +60º at 2º intervals along the tilt axis using the FEI tomography software. The defocus was ~10 µ m and the magnification was set such that each CCD pixel corresponded to 1.08 nm at the specimen level. For image processing, using colloidal gold particles as fiducial markers, the 2D projection images, binned 2-fold, were aligned with the IMOD software package images15 and then tomographic reconstructions were calculated by weighted back-projection using Priism package 16 .

Results and discussion 180

185

190

The growth of PS nodules on silica bead was studied by electron microscopy at very short polymerization times. However, conventional TEM experiments were not possible because the monomer-free PS nodules are too small and often unstable under the electron beam. Hence, we used the cryoTEM technique since the embedding method in a thin ice layer allows the sample preservation in a state similar to that in the polymerization reactor: the aqueous continuous phase is converted into ice and styrene is not devolatilized, i.e. it remains partitioned as during the polymerization reaction. Therefore, each cryo-TEM image is the frozen representation of a small volume unit of the reactor at the moment the sample was collected. The overall morphology of these This journal is © The Royal Society of Chemistry [year]


Fig. 2 Biphasic particles composed of 8 PS nodules per silica core (octopod) as observed by cryo-TEM after 120 min of polymerization. A) Gallery of octopods. B) Three different views of 3D reconstruction of an octopod showing the geometrical arrangement in a square antiprism form. Scale bar 100 nm.

215

220

Fig. 1 Cryo-TEM and cryo-ET images of PS nodules growing onto silica seeds. Row A) a sample collected prior to the initiator addition in the reactor, i.e. for a polymerization time of 0 min. Rows B-G correspond to 5, 10, 20, 30, 40, and 50 min polymerization times, respectively. First column, images of typical views of biphasic particles. The projections appear as composed of a dense cores surrounded by less dense nodules which correspond to the silica and PS beads, respectively. Stereoscopic views of 3D architectures of hybrid structures in cross-eye (second and third columns) and parallel-eye (third and fourth columns). Black dots visible on cryo-TEM images correspond to gold particles used for image alignment. Scale bars 100 nm.

195

200

205

210

biphasic nanoparticles is expected to be preserved. Furthermore, the tomographic approach, i.e. 3D reconstruction from tilt series of images collected by cryo-TEM, provides an accurate description of their topology and geometry including a complete inventory of PS nodules even those hidden by the silica particle. Figure 1A displays a cryo-TEM image of a collected sample prior to the initiator addition in the reactor, i.e. for a polymerization time of 0 min. Only silica particles may be observed; neither styrene-swollen micelles nor styrene droplets are visible. Hence, when this picture is compared to those obtained from samples collected later at different polymerization times, i.e. 5, 10, 20, 30, 40, and 50 min (Figure 1B-G), small spheres observed on silica seeds likely result from the polymerization reaction and therefore represent the PS nodules. At low polymerization times, 10 to 15 PS nodules appear to be randomly distributed on the silica seed (Figures 1B, 1C). Such a distribution suggests that (i) they are small enough to ignore the presence of their neighbors and that (ii) they have nucleated on privileged sites on silica surface where the surface density of MMS grafted This journal is © The Royal Society of Chemistry [year]

225

230

235

240

245

250

molecules would be higher. Nevertheless, the presence of these MMS “islands”, although they are conceivable in such situations, remains unknown. At 30-min polymerization time (Figure 1E), biphasic particles composed of 7 to 8 PS nodules were frequently observed. Then the maximum number of nodules per silica seed (i.e. 8) stayed constant till the end of the observed period of the polymerization reaction, leading to more regular morphologies (Figures 1F, 1G). At 120-min polymerization time, the regular octopod-like morphology4 can be obtained as shown in Figure 2. As fast as the PS nodules are growing, their number per silica seed is decreasing. This phenomenon can be explained on the one hand by the styrene polymerization process and on the other hand by steric constraints. Indeed, emulsion polymerization of styrene in water necessitates the use of surfactant molecules at a concentration above their Critical Micellar Concentration (CMC) in order to form micelles capable of absorbing large quantities of styrene. So, three phases are established: (i) the aqueous continuous phase in which small quantities of surfactant and styrene are molecularly dissolved, (ii) large styrene droplets maintained in suspension by adsorbed surfactant molecules and agitation and (iii) small styrene-swollen micelles. In the aqueous phase, the initiator molecules dissociate by thermolysis and react with styrene to produce oligomeric radical species which then diffuse into monomer-swollen micelles to initiate polymerization. This process is known as heterogeneous nucleation of PS nodules. The evolution of Np calculated from the monomer conversion determined gravimetrically and from the PS nodules diameter nodules (Dn)TEM (Table 1) with polymerization time is drawn in Figure 3. It may be observed that – despite the value uncertainty – Np is maximal after a few minutes of polymerization and then decreases as fast as the monomer conversion increases. The most probable interpretation is that a coagulative growth occurs by coagulation of PS nuclei initially obtained by heterogeneous nucleation. 17 It may explain slight increases in size polydispersity, because the diameter of the resulting nodule is 25% higher than that of both precursor nodules. In our case, it Soft Matter, [year], [vol], 00–00 | 3


280

285

290

Fig. 3 Evolution with polymerization time of the number of PS nodules in the reactor, when the polymerization is performed in the presence of silica seeds. These data were calculated according to the formula given in the experimental section from 20 min up to 1200 min. For very short polymerization times (0-20 min) the Np number is likely higher than 13*1015 according to cryo-ET data and is schematically illustrated by the hashed line.

255

260

265

270

275

is important to underline that this coagulation phenomenon occurs between PS nodules interacting with silica surface and very probably, for statistical reasons, between PS nodules bounded to the same silica seed. On the other hand, the decrease observed by cryo-ET of the nodule number per silica seed could be linked to their increasing size: some nodules that are in close contact would undergo steric constraints. The consequence of these constraints would most likely induce a release of PS nodules from the seed particle, as confirmed by the observation of a few free PS nodules on images. Therefore, the decrease of the PS nodules number could result from a combination of both mechanisms – coagulative growth and steric release – and unfortunately we could not explain the relative importance between these two processes during the polymerization reaction. According to this phenomenon, we can even consider that the PS nodules number could be higher than the 10-15 observed in cryo-ET for very short polymerization times (i.e. 0-5 min). Consequently, the Np number in the range of 0 to 20 min polymerization times which is not accessible experimentally is likely higher than 13*10 15 as suggested by the hashed line in Figure 3. The average diameter of the PS nodules (Dn)cryo-TEM was determined as a function of polymerization time (Table 1). At equivalent polymerization times, (Dn)cryo-TEM is systematically

295

300

305

310

315

320

higher than (Dn)TEM. This result is consistent with the fact that the PS nodules observed by cryo-TEM are swollen with styrene. Indeed, polymerization propagation within the nodules is supported by absorption of styrene from the aqueous phase, there being concurrent diffusion of styrene droplets into the aqueous phase to maintain equilibrium. This mass transfer rate is substantially higher than the polymerization rate, and hence a styrene partition between the different phases of the system according to thermodynamic equilibrium represents a volume fraction of 0.6 in the PS nodule.17 This maximal value is constant and begins to decrease when the styrene droplets have disappeared, i.e. when the monomer-to-polymer conversion reaches 40 %. Therefore, in our time course, the monomer conversion ratio is less than this 40%-threshold value (Table 1) meaning that PS nodules collected from the reactor are particularly soft. Unlike in cryo-TEM, a drying process occurs during sample preparation in conventional TEM inducing monomer devolatilization. A rough calculation indicates that a volume fraction of styrene of 0.6 corresponds to a diameter increase of 35 %, which is quite consistent with the difference between (Dn)cryo-TEM and (Dn)TEM. From 3D data calculated by the tomographic method, we are able to provide a synthetic view of the PS nodule growth on silica bead in order to investigate the interaction of PS nodules with silica surface. Central cross-sections passing through the centers of each PS nodule and the silica particle were extracted and averaged over 180°. A gallery of nodule sections sorted by their size represents snapshots illustrating various steps of the growth of PS nodule on silica bead (Figure 4). The contact area increases with the PS nodule radius, but remains small even for large radii. That observation becomes obvious by superimposing the all 100 sections (Figure 4B). Moreover it is clearly visible that the nodule centre is shifted along the vertical axis as a function of its radius. For a more accurate analysis, we determined the contact angle of the supported PS nodules with respect to their radius. Assuming that silica seed and PS nodules can be comparable to perfect spheres, we measured their radii and their center coordinates from the 3D tomograms. The contact angles were calculated for more than 80 nodules whose diameters were comprised between 40 to 200 nm (Figure 5). The contact angle determination was not performed for the smallest nodules because of their non-spherical shape. The mean value is 142.4° and mean square root deviation is 7.88. This data

Table 1 Evolution with polymerization time of the average number and average dimensions of PS nodules. Polymerization time (min)

5

10

20

30

40

50

120

Conversion (%)

-

1

3.2

4.7

6.9

8.1

17.8

(Dn)TEM (nm)

-

-

80

95

105

110

150

(Dn)cryo-TEM (nm)

30

60

85

110

130

140

185

PS nodule 10108≤8 number per 15 15 10 silica particle 4 | Soft Matter, [year], [vol], 00–00

≤8

≤8

≤8



350

X

with the silica surface. Further experiments are currently in progress for checking the contact angle value is directly dependent on the surface grafting density of the MMS promoting molecules on silica.

Acknowledgements

Fig./Scheme XX Caption.

Fig. 4 Snapshots of PS nodule growth. A) Gallery of averaged crosssections of supported PS nodules. For each PS nodule, the axis passing through the centres of the PS nodule and the silica bead was vertically oriented. The XY median sections of the 3D volume were extracted every one degree about Y axis and averaged. B) Superimposed cross-sections. Scale bars 50 nm.

355

The authors are very grateful to Elisabeth Sellier and Michel Martineau for conventional TEM experiments (CREMEM, Talence, France). David Nguyen and Adeline Perro were supported by grants from the French Ministry of Education, Research and Technology and the Conseil Régional d’Aquitaine, respectively.

References 1

θ contact angle (degree)

360

2

3 365

4

370

(Dn)cryo-TEM (nm)

Fig. 5 Plot of contact angle θ as a function of the PS nodule radius. PS nodules and silica nanoparticles were assimilated to perfect spheres whose radius and centre coordinates were determined from the 3D tomograms. For each PS nodule, the contact angle measured between the tangents at the contact point was calculated using trigonometric relationships knowing silica and PS nodule radii R1, R2 and the centre-tocentre distance d (according to the sketch).

5 6 375

7 8

380

325

330

335

variability is in the range of the precision of the radius measurement corresponding to ±1 pixel. The distribution of contact angles in a first approximation could be fitted to a linear regression with a slope of 0.01. These results indicate that the contact angle is constant for all PS nodule size meaning that PS nodules grow at a fixed contact angle. It suggests that (i) the growing PS nodules may be considered at thermodynamic equilibrium all along the polymerization reaction and (ii) the contact angle of the final particles could be tuned thanks to the control of the experimental parameter(s) which fix this contact angle in the polymerization reactor. In particular, the relation between the contact angle and the MMS surface grafting density needs to be investigated, as already suggested. 2

9 385

10 11 12

390

13 14 395

15 16 17

Conclusions 400

340

345

Original cryo-TEM images and 3D reconstructions were obtained from samples collected from a polymerization reactor where styrene was polymerizing in the presence of silica seeds. Early samples display the presence of small PS nodules bounded to silica particles in random arrangements. For higher polymerization times, the number of PS nodules per silica seed decreases to lead to octopod-like morphologies. The tomographic method allowed to measure for the very first time the contact angle which the PS nodules form in contact This journal is © The Royal Society of Chemistry [year]

A. Perro, S. Reculusa, S. Ravaine, E. Bourgeat-Lami and E. Duguet, J. Mater. Chem. 2005, 15, 3745-3760. E. Bourgeat-Lami and E. Duguet, in Functional coatings by polymer microencapsulation, ed. S.K. Ghosh, Wiley-VCH, Weinheim, 2006, ch. 4, pp. 85-152. S. Reculusa, C. Poncet-Legrand, S. Ravaine, C. Mingotaud, E. Duguet and E. Bourgeat-Lami, Chem. Mater., 2002, 14, 2354-2359. S. Reculusa, C. Mingotaud, E. Bourgeat-Lami, E. Duguet and S. Ravaine, Nano Lett., 2004, 4, 1677-1682; S. Reculusa, C. PoncetLegrand, A. Perro, E. Duguet, E. Bourgeat-Lami, C. Mingotaud and S. Ravaine, Chem. Mater., 2005, 17, 3338-3344; E. Bourgeat-Lami, M. Insulaire, S. Reculusa, A. Perro, S. Ravaine and E. Duguet, J. Nanosci. Nanotech., 2006, 6, 432-444. J. Dubochet, M. Adrian, J. J. Chang, J. C. Homo, J. Lepault, A. W. McDowall and P. Schultz, Quat. Rev. Biophys., 1988, 21, 129-228. S. Mornet, O. Lambert, E. Duguet and A. Brisson, Nano Lett., 2005, 5, 281-285. G. Thornton, Science, 2003, 300, 1378-1379. K. P. de Jong and A. J. Koster, Chem. Phys. Chem., 2002, 3, 776780; M. Weyland, T. J. V. Yates, R. E. Dunin-Borkowski, L. Laffont and P. A. Midgley, Scri. Mater., 2006, 55, 29-33; M. Weyland and P. A. Midgley, Ultramicroscopy, 2003, 96, 413-431; P. A. Midgley, M. Weyland, T. J. Yates, I. Arslan, R. E. Dunin-Borkowski and J. M. Thomas, J. Microsc., 2006, 223, 185-190. K. Kaneko, W. J. Moon, K. Inoke, Z. Horita, S. Ohara, T. Adschiri, H. Abed and M. Naito, Mater. Sci. Eng., 2005, A 403, 32-36. K. Kaneko, R. Nagayama, K. Inoke, W. J. Moon, Z. Horita, Y. Hayashi and T. Tokunaga, Scri. Mater., 2005, 52, 1205-1209. S. Kohjiya, A. Katoh, J. Shimanuki, T. Hasegawa and Y. Ikeda, Polymer, 2005, 46, 4440-4446. W. Baumeister and A. Steven, Trends Biochem. Sci., 2000, 25, 624631. W. Stöber, A. Fink and E. J. J. Bohn, Colloid Interface Sci., 1968, 26, 62-69. S. L. Westcott, S. J. Oldenburg, T. Randall Lee and N. J. Halas, Langmuir, 1998, 14, 5396-5401. D. N. Mastronarde, J. Struct. Biol., 1997, 120, 343-352. H. Chen, D. D. Hughes, T. A. Chan, J. W. Sedat and D. A. Agard, J. Struct. Biol., 1996, 116, 56-60. J. C. De la Cal, J. R. Leiza, J. M. Asua, A. Butte, G. Storti and M. Morbidelli, in Handbook of Polymer Reaction Engineering, ed. T. Meyer and J. Keurentjes, Wiley-VCH, Weinheim, 2005, ch. 6, pp. 249-317.

Soft Matter, [year], [vol], 00–00 | 5


Single column figure/scheme (below) X Fig./Scheme XX Caption.

Double column figure/scheme (below) X Fig./Scheme XX Caption.

405

Single column image (no caption) (below) X

Double column image (no caption) (below) X Single column numbered equation/reaction (below) (X)

X

Single column table (below) Table XX Caption X a

Footnote text.

Double column table (below) Table XX Caption X a

Footnote text.

6 | Soft Matter, [year], [vol], 00–00
