Impact of Polymer Graft Characteristics and Evaporation Rate on the

---

pubs.acs.org/Langmuir © 2010 American Chemical Society

)

)

Impact of Polymer Graft Characteristics and Evaporation Rate on the Formation of 2-D Nanoparticle Assemblies Satyajeet Ojha,†,^ Benjamin Beppler,‡,§,^ Hongchen Dong, Krzysztof Matyjaszewski, Stephen Garoff,‡,§ and Michael R. Bockstaller*,†

)

† Department of Materials Science and Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, Pennsylvania 15213, ‡Department of Physics, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, Pennsylvania 15213, §Center of Complex Fluids Engineering, Carnegie Mellon University, 5000 Forbes Avenue Pittsburgh, Pennsylvania 15213, and Department of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, Pennsylvania 15213. ^ These authors contributed equally to the work.

Received May 14, 2010. Revised Manuscript Received June 28, 2010 The effect of polymer functionalization on the two-dimensional (2-D) assembly of uniform as well as highly asymmetric binary colloidal mixtures with both neutral and incompatible polymer grafts is presented. In ordered assemblies of uniform particle brush systems, the observed size-segregation is analogous to that of hard sphere colloidal systems, suggesting that lateral capillary interactions are responsible for the crystal nucleation in the early stages of assembly formation. Structure formation in binary blends of asymmetric particle brush systems is found to be strongly influenced by three major energetic contributions, that is, the interfacial energies associated with the particle brush/air boundaries, the interfacial energies between the distinct brush components, as well as the elastic energy associated with the stretching of the polymer-brush to fill the interstitial regions within locally ordered particle arrays. Our results demonstrate the relevance of capillary interactions in soft particle brush systems but also highlight distinctive differences in the order formation as compared to hard sphere colloidal systems. In particular, the compliant response of grafted polymer chains is shown to promote phase separation in binary blends of incompatible and asymmetric colloidal systems.

Introduction Nanoparticle assemblies provide an intriguing model system to study the organizing principles of colloidal systems in which the structure formation is determined by the subtle interplay of capillary as well as particle-particle and particle-substrate interactions. Particular interest has been focused on the reversible crystallization of monodisperse uniform and binary hard sphere systems for which entropic effects have been shown to be responsible for the formation of ordered lattice or (for a size ratio in binary colloidal blends close to dA/dB = 0.58, with di denoting the diameter of particle “i”) superlattice structures. As size-polydispersity in particle systems increases, crystallization is only expected in the presence of (weak) attractive forces. For example, Gelbart and co-workers observed the size-selective order formation in systems of polydisperse short-chain aliphatic (dodecyl) coated gold nanocrystals drop-cast on amorphous carbon films and interpreted the observed particle size segregation (with larger particles populating the center of hexagonally ordered particle arrays) as being driven by the maximization of van der Waals interactions.1 Nagayama and co-workers studied the two-dimensional order formation process of polydisperse micrometer-sized latex spheres using video microscopy and proposed a two-stage mechanism to rationalize the observed size-segregation of particles.2 In this mechanism, the initial formation of an assembly nucleus occurs through (attractive) lateral capillary interactions between large particle species; subsequent growth of the assembly *To whom correspondence should be addressed. E-mail: bockstaller@ cmu.edu. (1) Denkov, N. D.; Velev, O. D.; Kralchevsky, P. A.; Ivanov, I. B.; Yoshimura, H.; Nagayama, K. Langmuir 1992, 8, 3183–3190. (2) Ohara, P. C.; Leff, D. V.; Heath, J. R.; Gelbart, W. M. Phys. Rev. Lett. 1995, 75, 3466–3469.

13210 DOI: 10.1021/la1019372

occurs through directional motion of smaller particles to the boundary of the array.3-6 A common thread in these previous studies has been the focus on colloidal systems in which the stabilization of particle species is facilitated by charge interactions or short-chain aliphatic ligands. While these (approximately) hard sphere-type systems are representative for a wide range of colloidal materials, recent progress in grafting-to and grafting-from polymerization techniques has increased the relevance of polymer-stabilized colloidal systems for applications ranging from hierarchically structured nanocomposites to photonic crystal materials.7,9-12 Since polymer functionalization exerts a multifaceted effect on the interactions in colloidal systems, such as soft repulsive potentials or reduced van der Waals interactions, deviations in the formation of assemblies are expected. Since tailoring of the chemical constitution of the polymer grafts facilitates the selective variation of the dominant interaction mode (enthalpic or entropic), polymer-grafted nanoparticles (3) Wong, S.; Kitaev, V.; Ozin, G. A. J. Am. Chem. Soc. 2003, 125, 15589–15598. (4) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Nature 1997, 389, 827–829. (5) DeGennes, P. G. Rev. Mod. Phys. 1985, 57, 827–863. (6) Reiss, H. J. Chem. Phys. 1951, 19, 482–487. (7) Fink, Y.; Urbas, A. M.; Bawendi, M. G.; Joannopoulos, J. D.; Thomas, E. L. J. Lightwave Technol. 1999, 17, 1963–1969. Bockstaller, M. R.; Mickiewicz, R. A.; Thomas, E. L. Adv. Mater. 2005, 17, 1331–1349. Bockstaller, M. R.; Kolb, R.; Thomas, E. L. Adv. Mater. 2001, 13, 1783–1786. Bockstaller, M. R.; Thomas, E. L. Phys. Rev. Lett. 2004, 93, 166106. (8) Voudouris, P.; Choi, J.; Dong, H.; Bockstaller, M. R.; Matyjaszewski, K.; Fytas, G. Macromolecules 2009, 42, 2721–2728. (9) Foulger, S. H.; Kotha, S.; Swervda-Krawiec, B.; Baughmann, T. W.; Ballato, J. M.; Jiang, P.; Smith, D. W. Opt. Lett. 2000, 25, 1300–1302. (10) Ma, Q. G.; Remsen, E. E.; Kowalewski, T.; Wooley, K. L. J. Am. Chem. Soc. 2001, 123, 4627–4628. (11) Colvin, V. L. MRS Bull. 2001, 26, 637–641. (12) Zhang, J. H.; Sun, Z. Q.; Yang, B. Curr. Opin. Colloid Interface Sci. 2009, 14, 103–114.

Published on Web 07/16/2010

Langmuir 2010, 26(16), 13210–13215

Ojha et al.

Figure 1. Illustration of relevant length scales in the semidilute particle brush system SiO2-S770.

also hold opportunities to illustrate the respective relevance of material and interaction parameters. The objective of the present contribution is to elucidate the effect of polymer functionalization on size-sorting in two-dimensional deposits of both uniform as well as highly asymmetric binary colloidal mixtures with either neutral or incompatible polymer grafts. Our results confirm the relevance of capillary interactions in soft particle brush systems but also highlight distinctive differences in the order formation as compared to hard sphere colloidal systems. In particular, the compliant response of grafted polymer chains (in the semidilute brush regime, see below) is shown to promote phase separation in binary blends of incompatible, and asymmetric colloidal systems. In order to provide the context for the choice of materials as well as for the interpretation of distinctive features in polymergrafted colloidal systems, we briefly review the structure of polymer-grafted particles (or “particle brushes”). In general, particle brushes are categorized depending on the polymer grafting density (F) and degree of polymerization (N). In the limit of small grafting densities (the “dilute brush” limit), polymer chains assume an approximately random coil conformation (mushroom). Since interactions in these dilute brush particles are complicated due to only partial screening of surface interactions, these systems will not be considered in the present contribution. As F increases and chains start to overlap (i.e., σ0 < Na2, with σ0 = F-1 denoting the surface area per grafted chain and a the length of one repeat unit), a transition to the “semidilute brush” regime is observed in which segmental interactions become increasingly important and result in stretching of the polymer chains. Fukuda and co-workers demonstrated that the chain endto-end distance (Re) in the semidilute brush regime approximately follows approximately the relation Re ∼ N0.7a.13 In the limit of high grafting densities, the “concentrated brush regime” is observed which is characterized by segmental interactions and non-Gaussian chain characteristics with Re ∼ N0.83a. Since for spherical particle brushes the effective area per chain decreases with increasing distance r from the particle surface according to σeff = σ0(d/r)2, where d denotes the particle core diameter, a transition from the concentrated to the semidilute brush regime is expected if the brush height R - d/2 exceeds a critical distance rDC (see Figure 1). The latter was first proposed by Daoud-Cotton14 for (conceptually similar) star polymer systems as rDC = υad(πσ0)1/2, with υ being the excluded volume parameter that is approximately given by υ = b/a (b is the Kuhn segment) for athermal solvents. (13) Morinaga, T.; Ohno, K.; Tsujii, Y.; Fukuda, T. Macromolecules 2008, 41, 3620–3626. (14) Daoud, J. M.; Cotton, J. P. J. Phys. (Paris) 1982, 43, 531–538.

Langmuir 2010, 26(16), 13210–13215

Article

Blends of four distinct particle systems are the subject of the present study, that is, polystyrene-coated gold nanoparticles (Au-PS, d = 3.5 nm, N = 10), polyisoprene-coated gold nanoparticles (Au-PI, d = 3.5 nm, N = 15), and two populations of polystyrene-coated silica nanoparticles (SiO2-S10, SiO2-S770) with d = 20 nm and N = 10 and 770, respectively. Because of the respective short ligand length, three of the four particle systems (Au-PI, Au-PS, and SiO2-S10) will be considered as hardsphere-like while the fourth (SiO2-S770) is determined to be within the semidilute brush regime with a hydrodynamic radius of RH = 33 nm as confirmed by dynamic light scattering. In the following, we report the structure formation in drop-cast film deposits of binary asymmetric blends of these particle systems as a function of solvent evaporation rate. While the results for the organization of soft particle brushes supports the mechanism of a capillary-driven ordering process suggested by Nagayama and coworkers, the ability of the semidilute brush to stretch will be shown to induce major differences in the structure formation of incompatible asymmetric binary blends of semidilute particle brushes as compared to hard sphere colloidal systems.1

Materials and Methods Materials. All chemicals were purchased from Sigma-Aldrich Co., unless otherwise specified. Benzene was distilled and dried prior to application. All other chemicals were used as obtained. Thiol-terminated oligostyrene (PS-SH) ligands were synthesized by anionic polymerization at 25 C in anhydrous benzene. Adding ethylene sulfide to the reaction mixture terminated the reaction. The obtained thiol-terminated oligostyrene ligands were characterized by size exclusion chromatography (SEC, not shown here) using PS calibration yielding a number-average molecular weight of Mn = 1.04 kg/mol corresponding a degree of polymerization of N ∼ 10 and polydispersity index PDI = 1.08. The obtained product was purified before further reaction by precipitation out of benzene solution following the addition of methanol. Thiolterminated oligoisoprene (PI-SH) was synthesized under analogous conditions. Based on previous studies using nonpolar solvent conditions, predominantly 1,4-isomerism of PI is expected.15 The synthesis of polymer-functionalized gold nanoparticles was performed using the phase-transfer approach developed by Schiffrin and co-workers.16 The procedure for synthesizing Au-PS (Au-PI) nanoparticles with particle-core diameter d = 3.5 ( 0.2 nm was as follows: An aqueous solution of tetrachloroaureate (30 mL, 30 mmol/L) was mixed with a solution of tetraoctylammonium bromide in toluene (80 mL, 50 mmol/L). The mixture was stirred until all the tetrachloroaureate was transferred into the organic layer and 1 mmol of thiol-terminated oligostyrene (oligoisoprene) was added to the solution. A freshly prepared aqueous solution of sodium borohydride (25 mL, 0.4 mol/L) was added under vigorous stirring. After stirring for 3 h, the solution was evaporated to 10 mL and mixed with 400 mL of ethanol to remove excess ligand. The brown precipitate was filtered and washed with ethanol/tetrahydrofuran (THF). Silica nanoparticles (d = 20 ( 10 nm) were obtained from Nissan Chemicals (MIBK-ST) and used without further modification. Grafting of PS-ligands of degree of polymerization N = 10 and 770 (the corresponding sample abbreviations are SiO2-S10 and SiO2-S770, respectively) was performed using atom-transfer polymerization as described in the previous work of Bombalski et al.17 The molecular weight of the PS-ligands was determined by (15) Listak, J.; Hakem, I. F.; Ryu, H. J.; Rangou, S.; Politakos, N.; Misichronis, K.; Avgeropoulos, A.; Bockstaller, M. R. Macromolecules 2009, 42, 5766–5773. Avgeropoulos, A.; Paraskeva, S.; Hadjichristidis, N.; Thomas, E. L. Macromolecules 2002, 35, 4030. (16) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801–802. (17) Bombalski, L.; Dong, H.; Listak, J.; Matyjaszewski, K.; Bockstaller, M. R. Adv. Mater. 2007, 19, 4486–4490.

DOI: 10.1021/la1019372

13211

Article

Ojha et al.

Table 1. Structural and Compositional Characteristics of Particle Samples Used in the Present Study sample ID DP F (nm-2) d (nm) RH (nm)

Au-PS 10 ∼0.7 3.5 ( 0.2 ∼4 (TEM)

Au-PI 10 ∼0.7 3.5 ( 0.2 4 (TEM)

SiO2-S10 10 0.5 20 ( 10 nm 12 (DLS)

SiO2-S770 770 0.5 20 ( 10 33 (DLS)

SEC (not shown here), and the grafting density of ligands was determined by thermogravimetric analysis (TGA, not shown here). Table 1 summarizes the characteristics of all samples used in our study. The polydispersity index of the polymer ligands was determined to be PDI ≈ 1.1. Film Preparation. Particle solutions were prepared in toluene and mixed in appropriate ratios to result in the final particle concentrations (c = 5 mg/mL). Particle films were prepared by placing carbon coated TEM grids (Electron Microscopy Sciences) on a glass slide with a hydrophobic coating and subsequent deposition of a 10 μL drop of suspension onto the grid. The drop was found to fully wet the grid and to pin on the glass surface just beyond the grid edge. The solvent was allowed to evaporate under controlled conditions. In ambient conditions, solvent was found to completely evaporate within about 4-5 min, producing an evaporation rate of ∼0.025 μL/s. Slower evaporation rates (>30 min, ∼0.0014 μL/s) were achieved by enclosing the setup in (approximately) saturated toluene atmosphere. Faster evaporation rates (