Quasi-Two-Dimensional Assembly of Bottlebrush Block Copolymers

Dec 19, 2018 - Block copolymer guided assembly of nanoparticles leads to the formation of nanocomposites with periodic arrangement of nanoparticles, w...
5 downloads 0 Views 3MB Size
Download PDF
Article Cite This: Macromolecules XXXX, XXX, XXX−XXX

pubs.acs.org/Macromolecules

Quasi-Two-Dimensional Assembly of Bottlebrush Block Copolymers with Nanoparticles in Ultrathin Films: Combined Effect of Graft Asymmetry and Nanoparticle Size Yaron Aviv,† Esra Altay,‡ Lea Fink,† Uri Raviv,† Javid Rzayev,*,‡ and Roy Shenhar*,† †

Macromolecules Downloaded from pubs.acs.org by UNIV OF SOUTH DAKOTA on 12/21/18. For personal use only.

Institute of Chemistry and the Center for Nanoscience and Nanotechnology, The Hebrew University of Jerusalem, Jerusalem 9190401, Israel ‡ Department of Chemistry, University at Buffalo, The State University of New York, Buffalo, New York 14260-3000, United States S Supporting Information *

ABSTRACT: Block copolymer guided assembly of nanoparticles leads to the formation of nanocomposites with periodic arrangement of nanoparticles, which are important for applications such as photonic devices and sensors. However, linear block copolymers offer limited control over the internal arrangement of nanoparticles inside their hosting domains. In contrast, bottlebrush block copolymers possess unique architectural attributes that enable additional ways to control the local organization of nanoparticles. In this work, we studied the coassembly of 8 and 13 nm gold nanoparticles with three bottlebrush block copolymers differing in the asymmetry of their graft lengths. Assembly was performed in ultraconfined films, where it occurs quasi-two-dimensionally. Our results indicate that graft asymmetry could be used as an additional tool to enhance nanoparticle ordering by forcing them to localize at the center of the domain regardless of their size. This behavior is analyzed in terms of the influence of the graft asymmetry on the average conformations of the blocks.

■

INTRODUCTION Metal and semiconductor nanoparticles (NPs) possess unique physical properties, which derive from their chemical composition, size, and shape.1,2 The collective properties of an ensemble of NPs are influenced by coupling interactions between adjacent NPs3−6 and are also sensitive to the geometry of NP organization within the superstructure. For example, NPs organized in parallel layers exhibit enhanced light absorption,7 behave as photonic crystals,8,9 and display nonlinear optical properties,10 whereas assembling NPs in a chainlike fashion enables tuning the wavelength11,12 and creating plasmon waveguides.13−17 One of the most promising approaches for organizing NPs into ordered superstructures relies on the utilization of block copolymers (BCPs) as organizing matrixes. BCPs are composed of chemically distinct sequences (blocks) and form periodic, nanoscale structures because of microphase separation. The bulk periodicity of the block copolymer (typically in the range 10−100 nm) depends on the polymer molecular weight, and the bulk morphology is dictated by the volume fractions of the blocks. Mixing BCPs with NPs that have been modified with surface ligands that make them compatible with one of the BCP blocks leads to inherently ordered nanocomposites, in which the NPs are organized in alternating domains.18,19 Various parameters influence the organization of NPs in the BCP matrixes. Simulations performed by Balasz et al. have © XXXX American Chemical Society

shown that large NPs segregate to the center of the domains whereas small NPs are more evenly distributed throughout the domain.20 This result was explained by the large entropic penalty associated with the need of the block to circumvent a large NP compared to circumventing a small NP. Experimental findings corroborate these conclusions.21−24 Kramer et al. have demonstrated that the distribution of the NPs across the domains becomes narrower at high filling fractions; this distribution was explained based on similar arguments.25 Tuning the chemistry of the ligands protecting the nanoparticles provides a more controllable way to influence NP location.2,26−29 For example, covering the NPs with a polymeric ligand that is identical in chemical composition to one of the blocks makes these NPs compatible with that block and thus directing them to the corresponding domains; functionalizing NPs with a mixed monolayer can be used to direct the NPs to the interface between the domains.26,27 The inclusion of NPs influences the nanocomposite periodicity by swelling the hosting domains. Ausserré et al. have shown that the lamellar periodicity increases linearly with NP filling fraction.30 At high filling fractions, it may also alter the nanocomposite morphology.29,31−33 In thin films, the inclusion of NPs often leads to perpendicular orientation of the Received: September 14, 2018 Revised: November 15, 2018

A

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Macromolecules

■

domains because of the tendency of NPs to sequester to the film surface to lower the nanocomposite surface energy.29,34 Although linear block copolymers provide various handles to tune the resulting structure in terms of morphology, periodicity, and NP location and distribution, there are two notable limitations. First, obtaining large periodicities requires BCPs with large molecular weights. However, such polymers with the desired chemical composition are not always accessible, and even when they are their phase separation is usually hindered by a high degree of entanglements.35 Additionally, the dependence of NP location on their size is sometimes restrictive (e.g., when trying to localize small NPs at the center of the domain). Manipulation of polymer architecture provides new avenues for tailoring material structure and properties. Bottlebrush copolymers (BBCPs) are an emerging class of branched macromolecules characterized by densely grafted polymeric side chains along a polymeric backbone.36−38 Steric repulsion between the side chains results in elongated backbone conformations, decreased entanglements in melt, and a strong tendency for forming alternating lamellae even with compositions that significantly deviate from symmetric.39−48 Owing to their large molecular size and fast assembly dynamics,49 BBCPs have facilitated access to periodic nanomaterials with large domain spacing for photonic applications.46,50−53 The nature of the bottlebrush architecture also allows for manipulating molecular shape and polymer composition in two ways: using backbone asymmetry or graft asymmetry.46,54−56 The latter has been shown to play a critical role in controlling the interfacial curvature during the self-assembly process and dictating the resulting polymer morphology both in solution and in melt.54,56,57 Thus, the strong propensity toward the lamellar morphology in BBCPs can be circumvented by using graft length asymmetry to favor the formation of cylindrical microstructures. Because of their size, rigidity, and easily controllable shape and composition, BBCPs can offer new pathways for affecting the arrangement of metallic nanoparticles in composite systems. Watkins and co-workers have conducted the main body of work on BBCP/NP composites.8,10,31,55,58 Taking advantage of the high periodicities displayed by BBCPs and employing hydrogen bonding to increase NP loading, photonic films with tunable properties were prepared.8−10 It was also demonstrated that the strong interaction between the NPs and the hosting polymer blocks could be used to either compensate for graft asymmetry and thus stabilize the lamellar morphology55 or induce effective graft asymmetry with symmetric BBCP to obtain a cylindrical phase.31 Furthermore, employing BBCPs consisting of low-Tg grafts on both blocks has been shown to lead to long-range ordering of the nanocomposite even at high filling fractions, which is seldom obtained with linear block copolymers.58 In this work we focus on the details of NP localization within BBCP-based nanocomposites in ultraconfined films, where the assembly is essentially quasi-two-dimensional. In particular, we wanted to explore the possibility to increase the level of ordering of the NP superstructures in the composite film using the graft asymmetry. To enable free distribution of NPs within their hosting domains, they were functionalized with polymeric ligands having the same chemistry as the hosting block. This modification enabled us to probe the NP size effect and compare it to what is known for coassembly with linear BCPs.

Article

RESULTS AND DISCUSSION

Three BBCPs with polymethacrylate backbones, polystyrene (PS) grafts on one block, and poly(lactic acid) (PLA) bifurcated grafts on the other were synthesized as reported previously (see the Experimental Section and Supporting Information Figure S1 and Table S1 for additional details).46 All three polymers featured nearly symmetric backbone lengths (average of 150 repeat units for the PLA-grafted block and 143−156 repeat units for the PS-grafted block) and constant PS graft length (40−42 repeat units on average). The PLA graft lengths vary from shorter to longer than the PS grafts (Figure 1). The polymers were termed according to the calculated volume fractions of the PLA-grafted blocks as BBCPPLA maj (f PLA = 68%), BBCPsym (f PLA = 46%), and BBCPPLA min (f PLA = 38%).

Figure 1. Chemical structures of PS-PLA BBCPs:46 (a) graftsymmetric BBCPsym, (b) graft-asymmetric BBCPPLA maj, and (c) graftasymmetric BBCPPLA min.

In addition to probing the graft asymmetry effect, we focused on studying the assembly in ultraconfined films, i.e., films of thicknesses that are on the order of the size of the NP. Confining the films to such thicknesses leads to quasi-twodimensional assembly for the polymers, which facilitates the extraction of structural dimensions from scanning electron and scanning force microscopy (SEM and SFM) images. For each polymer, we studied the assembly with polystyrene-functionalized gold nanoparticles (Au-PS NPs) of two core sizes (8 and 13 nm; see Figure S2) and variable filling fractions (4, 8, 12, and 16%). The average grafting density of PS-SH ligands (Mn 12.5 kDa) on both types of NPs was 1.27 ligands nm−2 (see the Supporting Information for further details), which makes them miscible in PS matrixes.59,60 Morphology of the Neat BBCP in Ultraconfined Films. To set a reference for the structural characterization of the nanocomposite films, we first studied ultraconfined films of the B

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 2. SFM height images (a−c) and SEM images (d−f) of ultrathin films of neat PS-PLA BBCPs after solvent annealing in THF vapor: BBCPsym (top row), BBCPPLA maj (middle row), and BBCPPLA min (bottom row). Values in the top right corner in (d−f) denote f eff PS. (g−i) Background-subtracted SAXS curves of the corresponding bulk samples after thermal annealing. Black curves represent the experimental data; red curves represent the best fitted lamellar model (see the Experimental Section for additional details); values on the principal peak denote the calculated lamellar periods in nm.

neat BBCPs on silicon wafers with native oxide. Films were cast at about 15 nm thickness, which is slightly above the estimated cross-sectional diameter of the BBCPs. For bottlebrushes with polystyrene side chains of similar lengths, Sheiko et al. determined by small-angle X-ray scattering (SAXS) that the cross-sectional radius of gyration is ∼5 nm.61 These bottlebrushes were found to spread on mica substrates, giving rise to a monolayer thickness of 6.5 nm with intermolecular distance of 8.2 nm. Russell et al. reported that a single layer of PS-PLA BBCP with 4 kDa side chains on a silica substrate has a thickness of 13 nm.62 Within this framework, we estimate that the ultrathin film thicknesses used in our work represent between one and two layers of BBCP chains lying flat on a surface. The films were annealed under THF vapor, which is slightly PS-selective (χTHF‑PS = 0.15; χTHF‑PLA = 0.62) but was found to be the closest to a neutral solvent.52,62,63 No dewetting was observed, and ultrathin films with uniform thickness were obtained by this method. Figure 2 shows SEM and SFM height images of the neat BBCP films. Both BBCPsym and BBCPPLA min produced well-defined lamellar morphologies, despite a slight asymmetry in the latter. This behavior is consistent with such bottlebrush block copolymers favoring the formation of flat interfaces, as reported previously.46 On the other hand, a large side chain asymmetry in the BBCPPLA maj copolymer forces the formation of a cylindrical morphology (Figure 2b,e). It is important to note that this BBCP arranges in a nearly hexagonal phase only because of the asymmetry between the PS and PLA graft lengths and despite the similar length of the respective backbones.

To understand the structural arrangement of the BBCP blocks in the film, we estimated the dimensions of the blocks in each polymer. SAXS analysis was performed on thermally annealed bulk samples (Figure 2g−i). The curves of the BBCPsym and BBCPPLA min were fitted to a multilamellar model, using 337 and 397 e−/nm3 as the electron densities of the PS and PLA domains, respectively (see the Experimental Section for additional details).64 The SAXS analysis enabled us to directly estimate the average length of each block. The ratios between the block lengths were then used with the lamellar periods calculated by Fourier transform analyses of the SFM images to estimate the dimensions of the polymer blocks in the ultraconfined films (Table 1). In the case of BBCPPLA maj, the SAXS curve could not be fitted to any classical phase (i.e., lamellar or cylindrical). We attribute this mismatch to the shape of PS domains, which are neither cylindrical nor lamellar (Figure 2e). For this BBCP we set the effective length of the PS-grafted block to be equal to that of the BBCPPLA min, as they share the exact same molecular attributes, and calculated the effective length of the PLA-grafted block using the periodicity determined by SFM analysis. Lastly, the effective heights of the domains were measured by SFM scratch profiling (Figure 3; see the Experimental Section for additional details). Modeling the general shape of the copolymer chain as a prism in the case of lamellae-forming BBCPs (BBCPsym and BBCPPLA min) and as a wedge in the case of BBCPPLA maj (Figure 3a−c) enabled us to calculate the effective volume fractions for each polymer. The resulting values are in good agreement with the volume fractions calculated using the Mn values and the respective block densities (1.04 and 1.25 g mL−1 for the PS and PLA domains, respectively; neglecting the volume occupied by the C

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

conformation compensates for the mismatch between the interfacial cross sections of both blocks, which is caused by the asymmetry in the PS and PLA graft lengths.55 The BBCPPLA maj packs into a nonlamellar morphology, consistent with the fact that it features the largest graft length asymmetry.54 The assumption that the PS-grafted blocks have the same effective lengths in both BBCPPLA min and BBCPPLA maj is supported by the volume fraction calculations (Figure 3d); assuming a more coiled conformation would lead to larger deviation from the volume fraction determined by SEC. Yet, this assumption seems to contradict the expectation that in asymmetric BBCPs the shorter-grafted block must coil to match the interfacial cross section. Hence, we conclude that the length of the PS grafts in these polymers does not allow the corresponding block in the BBCPPLA maj to coil further; therefore, adopting a nonlamellar conformation is favored in the case of BBCPPLA maj. Coassembly of NPs and BBCPs. The most straightforward polymer to analyze is the BBCPsym, which not only is symmetric in terms of the length of the respective backbones but also features PS and PLA grafts of similar size. In a sense, one could envision BBCPsym as the bottlebrush analogue of a linear block copolymer, with a thicker cross section. Figure 4 shows SEM and SFM height images of BBCPsym with varying amounts of Au-PS NPs of two sizes (8 and 13 nm). The lamellar morphology persists over the entire range of NP filling fractions that was probed, as expected from the calculated effective volume fractions of the PS-grafted blocks (0.56 < f eff PS < 0.62). At low filling fractions, the small NPs segregate to the center of the domains (Figure 4a,b). This could be explained by the difficulty of the hosting PS brushes to twist and circumvent the NPs.20,21 It should be emphasized that with linear BCPs this behavior is observed only for large NPs and is explained by the substantial entropic penalty associated with limiting the conformational space of the hosting blocks when they need to circumvent a large NPs. With BBCPs, the tendency to segregate the hosted NPs to the center of the domain is amplified by the stiffness of their backbones, which adds an

Table 1. Lamellar Periods and Effective Block Lengths Determined by Analysis of SAXS and SFM Measurements SAXS

SFM

polymer

Dlam [nm]a

leff PS [nm]b

leff PLA [nm]b

Dlam [nm]c

leff PS [nm]

leff PLA [nm]

BBCPsym BBCPPLA maj BBCPPLA min

84.0 72.6 73.9

23.6

18.5

20.7

16.3

81 69 67

23d 19e 19d

18d 16e 15d

a

Lamellar periods calculated as 2π/q from the location of the principal scattering peak. bEffective block lengths calculated from fitting the experimental curve to a lamellar model. cLamellar periods calculated from Fourier transform of the SFM images (see Figure S3 and Experimental Section for further details). dEffective block lengths calculated by using the Dlam from SFM and applying the same block length ratios obtained from the SAXS measurements. eThe effective length of the PS-grafted block was taken to be the same as the corresponding block in the BBCPPLA min as they share the exact same attributes; the effective length of the PLA-grafted block was calculated according to the corresponding Dlam and leff PS.

backbone), which substantiates the validity of the dimensional analysis. Confining the assembly process into films much thinner than the block copolymer periodicity and on the order of expected bottlebrush cross-sectional diameters enabled us to use SFM block lengths to evaluate the average block conformations of the BBCP macromolecules on the solid surface. The effective length of the PS-grafted block in the BBCPPLA min is shorter than the corresponding block in the BBCPsym by ca. 3 nm. As the PS grafts are similar in the number of repeat units (40 and 42, respectively), the difference in PS-grafted block length is attributed to the shorter backbone in the BBCPPLA min (143 vs 156 repeat units in the BBCPsym). Indeed, the calculated difference in contour length is 3.3 nm; this difference indicates that the backbone of this block is considerably stretched. The PLA-grafted block in the BBCPPLA min is also shorter than the PLA-grafted block in the BBCPsym. Here, however, both backbones consist of 150 repeat units. Hence, the difference in the effective block lengths indicates enhanced coiling of the PLA-grafted blocks in the asymmetric BBCP. The coiled

Figure 3. Simplified illustration of the self-assembled structures in ultraconfined films of neat BBCPs using simple building blocks: (a) BBCPsym, (b) BBCPPLA maj, and (c) BBCPPLA min. (d) Volume fractions of the PS-grafted blocks calculated from the compositional data (f PS SEC) and the microscopic characterization (f PS SFM). D

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 4. SEM and SFM height images of 15 nm thick films of the BBCPsym assembled with Au-PS NPs of two sizes: 8 nm (a−d, top: SEM, bottom: SFM) and 13 nm (e−h, top: SEM, bottom: SFM). Values in the top right corners denote the effective volume fraction of the PS domains (f eff PS).

enthalpic penalty (i.e., flexing a rigid chain, leading to local steric crowdedness between adjacent grafts) to the entropic penalty associated with the stretching needed to circumvent the NPs. Hence, with BBCPs the tendency to segregate to the center of the domain is noted also for relatively small NPs that would have been distributed more homogeneously in the hosting domain of a linear BCP. At higher filling fractions, the small NPs are more evenly dispersed in the PS domains65 and also located at the PS/PLA interface (Figure 4c,d; especially visible in the SEM images). Additionally, many NPs appear dim in the SEM images, which may infer their localization within the film. We note here that although the film thicknesses correspond to at most a bilayer of BBCPs, they still exceed the size of the small NPs, thus allowing them to be embedded inside the film at different locations along the direction normal to the substrate (see Figure S4 for additional details). Different behavior is observed in the case of the large NPs (Figure 4e−h). First, all NPs appear bright in the SEM contrast (i.e., located at the film surface) regardless of filling fraction. This is consistent with their diameter (∼20 nm including the ligand layer), which is larger than the film thickness (see Figure S4). Hence, the assembly of the large NPs is truly two-dimensional. Additionally, all the NPs are located at the center of the domains, forming chains of beads. When the filling fraction exceeds the capacity of the PS domains to host a single chain of NPs (Figure 4g,h), closepacked aggregates form at defect points in the structure but are still located far from the PS/PLA interfaces (in comparison,

the small NPs distributed randomly across the domains at the same filling fractions). In fact, the nearest-neighbor interparticle distance measured in these aggregates is ∼21 nm, which is similar to the interparticle distances measured in close-packed arrays of NPs assembled without a polymer (see the Experimental Section). This measurement suggests that these NPs are actually microphase separated from their hosting PS domains, within the confines of these domains. This behavior is in accord with the entropic and enthalpic penalty that is associated with the need of stiff brush blocks to circumvent large NPs. Figure 5 shows the coassembly results of films consisting of BBCPPLA maj and Au-PS NPs of the same sizes and filling fractions as described above. In this asymmetric BBCP, the PLA grafts are longer than the PS grafts. As shown above (Figure 2b,e), this architecture resembles the shape of a wedge, which results in the formation of distorted PS cylinders. For the small NPs at low filling fractions (Figure 5a,b), the effective volume fraction of the PS domains maintains the nearcylindrical phase. Unlike the assembly with the BBCPsym, here not all the NPs are segregated at the center of the domains, and many are located near the PS/PLA interfaces. At higher filling fractions, the film morphology becomes rather illdefined, with NPs distributed randomly throughout the PS domains (Figure 5c,d). SFM images suggest that the free surface of the film is enriched with the PS phase. This conclusion is corroborated by the higher selectivity of the SiOx substrate to the PLA and the lower surface tension of the PS (γPS = 40.7 mN/m, γPLA = 43.5 mN/m).66 E

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 5. SEM and SFM height images of 15 nm thick films of the BBCPPLA maj assembled with Au-PS NPs of two sizes: 8 nm (a−d, top: SEM, bottom: SFM) and 13 nm (e−h, top: SEM, bottom: SFM). Values in the top right corners denote the effective volume fraction of the PS domains (f eff PS).

are segregated at the center of the domains over the entire range of filling fractions. Compared with the other BBCPs, aggregation is largely avoided and occurs only at a few defect points in the nanocomposite films. Large NPs tend to form single chain-of-beads at low filling fractions, and hexagonally packed double and triple chains-of-beads as the filling fraction increases. The enhanced tendency of the NPs of both types to segregate to the center of the PS domains in the BBCPPLA min is unique, especially considering that the PS-grafted blocks have similar attributes for all three BBCPs probed. This property suggests that the reason for this behavior might be related to the PLA-grafted blocks. The PLA grafts in the BBCPPLA min are shorter than the PS grafts. The combination of the graft length asymmetry and the specific length of the PLA grafts in this blockwhich are the shortest in all BBCPs described in this workcauses the PLA-grafted blocks to be more flexible than the PS-grafted blocks in this polymer.55 As mentioned above, in the neat BBCPPLA min, this means that the backbones of the PLA-grafted blocks coil to match the difference in interfacial cross section caused by the differences in graft lengths. In principle, added NPs could be accommodated either at the center of the domain or between adjacent PS brushes. However, accommodating NPs between adjacent brushes would further increase the interfacial cross section between the PS and PLA-grafted blocks. This increase would impose on the PLA-grafted blocks to coil much more than in the neat BBCP, which is both entropically and enthalpically disfavored considering the relative stiffness of bottlebrush blocks. Hence,

The large NPs localize at the center of the PS domains in the films featuring low filling fractions (Figure 5e,f). A large fraction of the PS domains contain only one nanoparticle localized at the center. This is a unique arrangement that may be used for precise positioning of nanoparticles guided by bottlebrush block copolymer assembly. As the filling fraction increases, a transition to a wormlike morphology is observed, where NPs form extended aggregates throughout the PS domains (Figure 5g,h). In contrast to the corresponding assemblies of the BBCPsym (Figure 4g,h) no single chain-ofbeads are observed. This behavior and the ill-defined structures observed with the small NPs suggest that the asymmetric architecture of the BBCP, and particularly the fact that the domains hosting the NPs consist of the blocks with the shorter grafts, imposes fewer restrictions on the locations of the NPs. Increasing the effective volume fraction of the minority PS domains by compartmentalizing larger amounts of NPs (higher NP filling fraction) decreases the interfacial curvature and leads to a transition from a distorted cylinder structure to a networklike morphology. Because the PS-grafted block has a limited conformational freedom and a smaller cross-sectional diameter than its PLA counterpart, the formation of flat (or less-curved) interfaces is stabilized by intercalation of NPs between the PS bottlebrush chains throughout the PS domain. The nanocomposite films consisting of the BBCPPLA min display higher level of ordering than the previous two systems (Figure 6). First, the lamellar morphology persists even at relatively high filling fractions (corresponding to f eff PS as high as 0.68). Second, close inspection shows that both types of NPs F

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

Figure 6. SEM and SFM height images of 15 nm thick films of the BBCPPLA min assembled with Au-PS NPs of two sizes: 8 nm (a−d, top: SEM, bottom: SFM) and 13 nm (e−h, top: SEM, bottom: SFM). Values in the top right corners denote the effective volume fraction of the PS domains (f eff PS).

Figure 7. Effect of NP filling fraction on the normalized periodicity changes for (a) 8 nm NPs and (b) 13 nm NPs.

in the BBCPPLA min films are larger than those in the BBCPsym films. The tendencies mentioned above can be explained by considering that the volume added by NP inclusion in the nanocomposite film is not distributed equally along the three axes. First, the low film thickness, which is on the order of the NP diameter, severely limits the amount of volume that could be added along the vertical axis. Additionally, volume is not necessarily distributed equally along the lateral dimensions (i.e., along the domain director and along the axis that defines the nanocomposite periodicity). Hence, for a certain added volume, the increase in the periodicity is influenced by the degree of swelling in the other directions. NPs that segregate to

segregating the NPs to the center of the PS domain remains the only viable option. Additional insights into the coassembly process are obtained from analyzing the lamellar periodicity changes with NP filling fraction in the lamellae-forming BBCPs (BBCPsym and BBCPPLA min). Figure 7 shows the increases of the periodicity from that of the neat BBCP (L0) with increasing NP filling fractions, which were calculated from Fourier transform of SFM images and normalized to the effective diameter of the NPs. Interestingly, despite the normalization to the NP size, the periodicity changes strongly depend on the NP size, showing steeper increase with filling fractions for the large NPs than for the small NPs. Additionally, the periodicity increases G

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

In contrast, for the corresponding filling fractions, small NPs are both more uniformly dispersed in the PS domains.65 Additionally, their size allows a certain degree of freedom in their location along the thickness of the film, which distributes some of the added volume in this direction. As the contribution to the increase in periodicity is affected by the amount of volume added in the other dimensions, the overall effect is of a more gradual increase in periodicity (Figure 7a) compared to the steep increase noted forlarge NPs (Figure 7b). The higher increase in periodicity noted for the BBCPPLA min-based nanocomposite film compared to the BBCPsym-based film for both types of NPs provides insights into the average conformation of the PLA-grafted blocks in each system. In the large NP case (Figure 7b), the NPs are segregated at the center of the domains (Figures 4e−h and 6e−h). As the NP size and location and the PS-grafted block characteristics are the same for both the BBCPsym and BBCPPLA min, the only difference between these cases is the length of the PLA grafts. Hence, the lower slope in the BBCPsym indicates that the PLA blocks must have contracted. In the case of the small NPs (Figure 7a), the location inside the PS domains should also be included in the comparison. Small NPs segregate to the center of the PS domains of the BBCPPLA min (Figure 6a−d) but are more evenly distributed within the PS domains in the BBCPsym (Figure 4a−d). This situation further lowers the change in periodicity in the case of BBCPsym/small NPs, making the difference in trends between the two polymer systems (Figure 7a) more pronounced than with the large NP case (Figure 7b).

the center of the domains (Figure 8, bottom) directly contribute to an increase in the lamellar period (x-axis in

■

CONCLUSIONS The quasi-two-dimensional assembly of gold nanoparticles in ultrathin films consisting of backbone-symmetric bottlebrush block copolymers was investigated. The combined effect of nanoparticle size and copolymer architecture was probed, where the lengths of the PS grafts in the blocks hosting the nanoparticles were kept constant and the lengths of the PLA grafts varied. Insights gained from our analyses show that the unique architecture of the bottlebrush provides an additional factor that enables to control two aspects of nanoparticle organization: the distribution of their location across the hosting domain and at high filling fractionsalso the level of nanoparticle ordering. The important factor here is the relative length of the grafts on both blocksa factor that does not exist in linear block copolymer systems. In graft-asymmetric bottlebrush copolymers, the block consisting of the shorter grafts is less stiff and hence is forced to coil to match the crosssectional area of the other block at the interface between them. If the NPs are hosted in the domain consisting of the shorter grafted blocks (i.e., A-compatible NPs in BBCPA‑min), a more homogeneous distribution of the NPs inside these domains would be favorable because it would help to increase the effective cross-sectional area of these blocks and thus relieve the imposed coiling to a certain extent. In contrast, if the NPs are hosted in the domains consisting of the longer grafted blocks (i.e., A-compatible NPs in BBCPA‑maj), they have to segregate to the center of the hosting domains to avoid increasing the existing asymmetry between the effective crosssectional areas, which would impose on the shorter-grafted block to coil beyond its ability.

Figure 8. Cartoon illustrating the dependence of periodicity increase on the location of the nanoparticle in a symmetric BBCP: (center) neat BBCP as reference; (top) moderate periodicity increase when NPs are accommodated between adjacent brushes; (bottom) strong periodicity increase when NPs are forced to localize at the center of the domain. The cross-sectional area between the PLA-grafted and PS-grafted blocks is drawn as green rectangles.

Figure 8). Conversely, NPs that localize between adjacent PS brushes (i.e., off the domain centerline; Figure 8, top) swell the domain more along the direction parallel to the PS/PLA interface (y-axis in Figure 8) than along the periodicity axis, hence contributing less to an increase in periodicity. Moreover, spacing between adjacent brushes increases the cross-sectional area between the two blocks (green frame in Figure 8, top), causing the PLA-grafted blocks to contract (Figure 8, top). The overall effect is that the periodicity increases more gradually when NPs are allowed to disperse in-between the PSgrafted blocks. The NP size effect noted in Figure 7 can be explained as follows. Large NPs segregate to the center of the PS domains, thus directly contributing to an increase in the lamellar period. H

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

silicon wafers containing 4 nm thick native oxide (precleaned in NoChromix solution in H2SO4 for 24 h and then rinsed for 20 min in distilled water). Samples coated with the composite films were cut into five pieces, and each series of samples was annealed in a closed Petri dish under saturated THF atmosphere for various intervals of time (0, 20, 40, 60, and 120 min), after which they were rapidly quenched by fast removal from the solvent vapor chamber. Owing to the differences in the relaxation dynamics among the different BBCPs, the annealing times for each BBCP and its composite films were chosen to yield the most ordered samples for the corresponding neat BBCP (20, 60, and 120 min for the BBCPsym, BBCPPLA maj, and BBCPPLA min, respectively). All the experiments were performed on the same day to maintain fixed ambient conditions. We note that all films remained smooth under these conditions (i.e., did not develop islands and holes). Basic Characterization. 1H NMR spectra were recorded on a Varian INOVA-500 (500 MHz) spectrometer using CDCl3 as a solvent. Size exclusion chromatography (SEC) analyses were performed on Viscotek’s GPC Max and TDA 302 Tetradetector Array system equipped with two PLgel PolyPore columns (Polymer Laboratories, Varian Inc.) using THF as the mobile phase at 30 °C. The system was calibrated with 10 linear polystyrene standards with molecular weights ranging from 1.2 × 106 to 500 g/mol. Highresolution scanning electron microscopy (HR-SEM) images of the composite BBCP/NP films were acquired with a Sirion microscope (FEI Company) at 5 kV acceleration voltage. Film thicknesses were determined by performing SFM scans across scratches made with tweezers in the annealed composite films. Scanning force microscopy (SFM) images were acquired by a Dimension 3100 scanning probe microscope with a Nanoscope V controller (Veeco, Santa Barbara, CA) and were corrected by first-order flattening using the Nanoscope Analysis Program (V1.40, Bruker). Height histograms were obtained using the built-in Depth Analysis tool. Periodicity Determination. The periodicity of the neat BBCPs was determined by SAXS using an in-house setup described elsewhere.73 Radial integration of the 2D scattering image was performed using the FIT2D software.74 The sample-to-detector distance (∼1.85 m) was calibrated using silver behenate as a standard. For every measured sample, background measurement on an exposed area of the supporting Kapton sheet was subtracted from the corresponding curve. The experimental data of all curves were fitted to a model of a lamellar phase using the “X+” software;75 the data of the asymmetric BBCPs were also fitted to a cylindrical model using the “D+” software.76 The computed scattering intensity curve of each model was convoluted with a Gaussian function representing the instrument resolution function, with a full width at half-maximum (fwhm) of 0.02 nm−1. The lamellar model consisted of a stack of N infinite flat slabs in the xy-plane with uniform thickness and electron density (ED).77 The ED values used for the PS and PLA domains were 337 and 397 e−/nm3, respectively, calculated using their bulk densities (1.04 and 1.25 g mL−1, respectively). The fitting variables were the widths of the PS and PLA layers; the ED of the volume surrounding the stack of slabs was set to be the weighted average of the ED of the two layers. It was found that four PS/PLA bilayers (i.e., N = 8) was the minimum number of slabs needed to reproduce the width of the experimental peaks. This value represents the lower bound of the correlation length of the lamellae. The cylindrical model employed 100 cylinders, 50 nm long, packed in a hexagonal arrangement in a rhombus. The ED of the cylinders was set to either that of PS or PLA, and the ED of the surroundings was set to that of the other block. The fitting variables were the center-to-center distances between the cylinders and their diameter. The number of cylinders and their lengths were found to represent faithfully the infinite limit (i.e., marginal change in the scattering intensity was observed when larger values were used). As described above, the scattering curve of the BBCPPLA min eventually showed a better fit to the lamellar model (see Figure 2i),64 and the curve of the

The NP size effect is also affected by the bottlebrush architecture. Although the general trend remains the same as with linear block copolymers (i.e., large nanoparticles display a stronger tendency to segregate to the center of their hosting domains compared with small nanoparticles),20,21 when Acompatible NPs coassemble with BBCPA‑maj, even small NPs are forced to localize at the center of the A-domains. Hence, using BBCPs enables obtaining higher ordering in superstructures even with small nanoparticles. The insights gained from this study extend the knowledge gained from linear block copolymer-based nanocomposites and highlight the opportunities for structural control that become accessible with the bottlebrush architecture.

■

EXPERIMENTAL SECTION

Materials and Synthesis. Polystyrene-functionalized gold NPs with two different core diameters were used in this study (8 and 13 nm). The 8 nm NPs were synthesized with dodecanethiol ligands by the Brust−Schiffrin method67 followed by thermal growth.68 NPs were dissolved in tetrahydrofuran (THF) and purified from excess ligand three times by centrifugation in methanol at 8500 rpm for 5 min. The 13 nm, citrate-stabilized gold NPs69 were modified with hexadecyltrimethylammonium bromide (CTAB) ligands by an overnight stirring at room temperature in deionized water. A 2-fold excess of CTAB relative to the calculated number of the surface Au atoms was used; excess ligand was removed by centrifugation at 13000 rpm for 35 min in water at 35 °C. Thiol-terminated polystyrene (PS-SH), Mn = 12.5 kDa, was synthesized by anionic polymerization under a nitrogen atmosphere as described previously.70 The ligand exchange reaction was performed by stirring each type of nanoparticle in THF at room temperature overnight with a 2-fold excess of the PS-SH ligand relative to the calculated number of the surface Au atoms. Modified NPs were purified from excess ligand by performing two cycles of centrifugation at 9500 rpm at 15 °C for 10 min in THF and then thee more cycles using methanol. The solutions were then filtered by a 0.22 μm PTFE filter, and their concentration was determined by UV− vis (using εNP(527 nm) = 5.13 × 107 M−1 cm−1 and εNP(532 nm) = 2.80 × 108 M−1 cm−1 for the 8 and 13 nm NPs, respectively).71 The NP core diameters were determined by analyzing scanning electron microscopy (SEM) images of NPs deposited on silicon substrates (Figure S2) using ImageJ software.72 Average values were determined by histogram analysis to be 7.5 ± 0.9 and 13 ± 2 nm for the small and large NPs, respectively. The effective ligand thickness was estimated as ca. 3.5 nm from the center-to-center distance between close-packed NPs using Fourier transformation (Figure S2). The average number of PS ligands per particle and the PS ligand density on the surface of the particles were calculated from ligand weight fractions determined by TGA (performed on a Mettler-Toledo TGA2 analyzer, measuring weight loss under nitrogen flow between 120 and 600 °C at a heating rate of 20 °C min−1). PS-PLA BBCPs were synthesized by a combination of controlled radical and ring-opening polymerizations as reported previously.46 BBCPsym: Mn = 1439 kDa, Đ = 1.11; BBCPPLA maj: Mn = 2130 kDa, Đ = 1.09; BBCPPLA min: Mn = 1093 kDa, Đ = 1.17 (see Table S1 for additional details and Figure S1 for GPC and 1H NMR data). Bulk periodicities of the neat polymers were determined by small-angle Xray scattering (SAXS) on samples prepared on pieces of Kapton sheets and annealed under vacuum at 150 °C for 12 h. Preparation of Nanocomposite Films. Stock solutions of the BBCP (5 wt %) and Au-PS NPs (0.5−1.5 μM for the small NPs and 0.01−0.5 μM for the large NPs) were prepared in toluene. Films containing different NP filling fractions (ϕBBCPNPs = 0, 4, 8, 12, and 16%) were cast from 200 μL solutions containing aliquots of varying volumes taken from the BBCP and the Au-PS NP stock solutions (0.4−0.5 wt % BBCP concentration, resulting in film thicknesses in the range 15−20 nm). The solutions were allowed to equilibrate for 15 min and then spin-coated at 3000 rpm for 30 s onto hydrophilic I

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules BBCPPLA maj (Figure 2h) showed poor fitting to both the lamellar and cylindrical models attempted. The periodicity of each nanocomposite film was determined by analyzing SFM images, where the contrast is dominated by the BCP domains and less influenced by the presence of the NPs (compared to the analogous SEM images). Fourier transformation on 2 μm × 2 μm images (performed using ImageJ software)68 followed by radial integration using Fit2D program74 furnished one-dimensional plots (Figure S3), and the periodicity was calculated from the location of the principal peak. All the measurements were referenced to the periodicity of the BBCPsym (L0 = 81 nm), which was determined by Fourier transform performed on the SFM image using the Nanoscope Analysis program.

■

Silver Nanostructures for Plasmonic Applications. Chem. Rev. 2011, 111, 3669−3712. (6) Liu, K.; Zhao, N.; Kumacheva, E. Self-assembly of inorganic nanorods. Chem. Soc. Rev. 2011, 40, 656−671. (7) Bockstaller, M. R.; Thomas, E. L. Proximity effects in selforganized binary particle-block copolymer blends. Phys. Rev. Lett. 2004, 93, 166106. (8) Song, D.-P.; Li, C.; Colella, N. S.; Lu, X.; Lee, J.-H.; Watkins, J. J. Thermally Tunable Metallodielectric Photonic Crystals from the SelfAssembly of Brush Block Copolymers and Gold Nanoparticles. Adv. Opt. Mater. 2015, 3, 1169−1175. (9) Song, D. P.; Li, C.; Li, W. H.; Watkins, J. J. Block Copolymer Nanocomposites with High Refractive Index Contrast for One-Step Photonics. ACS Nano 2016, 10, 1216−1223. (10) Song, D. P.; Shahin, S.; Xie, W. T.; Mehravar, S.; Liu, X. H.; Li, C.; Norwood, R. A.; Lee, J. H.; Watkins, J. J. Directed Assembly of Quantum Dots Using Brush Block Copolymers for Well-Ordered Nonlinear Optical Nanocomposites. Macromolecules 2016, 49, 5068− 5075. (11) Nie, Z. H.; Fava, D.; Kumacheva, E.; Zou, S.; Walker, G. C.; Rubinstein, M. Self-assembly of metal-polymer analogues of amphiphilic triblock copolymers. Nat. Mater. 2007, 6, 609−614. (12) Thomas, K. G.; Barazzouk, S.; Ipe, B. I.; Joseph, S. T. S.; Kamat, P. V. Uniaxial plasmon coupling through longitudinal self-assembly of gold nanorods. J. Phys. Chem. B 2004, 108, 13066−13068. (13) Quinten, M.; Leitner, A.; Krenn, J. R.; Aussenegg, F. R. Electromagnetic energy transport via linear chains of silver nanoparticles. Opt. Lett. 1998, 23, 1331−1333. (14) Maier, S. A.; Brongersma, M. L.; Kik, P. G.; Meltzer, S.; Requicha, A. A. G.; Atwater, H. A. Plasmonics - A route to nanoscale optical devices. Adv. Mater. 2001, 13, 1501−1505. (15) Solis, D., Jr.; Paul, A.; Olson, J.; Slaughter, L. S.; Swanglap, P.; Chang, W.-S.; Link, S. Turning the Corner: Efficient Energy Transfer in Bent Plasmonic Nanoparticle Chain Waveguides. Nano Lett. 2013, 13, 4779−4784. (16) Solis, D., Jr.; Willingham, B.; Nauert, S. L.; Slaughter, L. S.; Olson, J.; Swanglap, P.; Paul, A.; Chang, W.-S.; Link, S. Electromagnetic Energy Transport in Nanoparticle Chains via Dark Plasmon Modes. Nano Lett. 2012, 12, 1349−1353. (17) Halas, N. J.; Lal, S.; Chang, W.-S.; Link, S.; Nordlander, P. Plasmons in Strongly Coupled Metallic Nanostructures. Chem. Rev. 2011, 111, 3913−3961. (18) Lauter-Pasyuk, V.; Lauter, H. J.; Ausserre, D.; Gallot, Y.; Cabuil, V.; Kornilov, E. I.; Hamdoun, B. Effect of nanoparticle size on the internal structure of copolymer-nanoparticles composite thin films studied by neutron reflection. Phys. B 1997, 241-243, 1092−1094. (19) Huh, J.; Ginzburg, V. V.; Balazs, A. C. Thermodynamic behavior of particle/diblock copolymer mixtures: Simulation and theory. Macromolecules 2000, 33, 8085−8096. (20) Thompson, R. B.; Ginzburg, V. V.; Matsen, M. W.; Balazs, A. C. Predicting the mesophases of copolymer-nanoparticle composites. Science 2001, 292, 2469−2472. (21) Bockstaller, M. R.; Lapetnikov, Y.; Margel, S.; Thomas, E. L. Size-selective organization of enthalpic compatibilized nanocrystals in ternary block copolymer/particle mixtures. J. Am. Chem. Soc. 2003, 125, 5276−5277. (22) Hamdoun, B.; Ausserré, D.; Joly, S.; Gallot, Y.; Cabuil, V.; Clinard, C. New nanocomposite materials. J. Phys. II 1996, 6, 493− 501. (23) Hamdoun, B.; Ausserré, D.; Joly, S. Nanoparticle-copolymer composites: Curvature. J. Phys. II 1996, 6, 1207−1217. (24) Bockstaller, M. R.; Ryu, H. J.; Ojha, S.; Choi, J. 1D periodic bimetallic superstructures by co-assembly of ternary block copolymer/ nanoparticle blends. J. Mater. Chem. 2010, 20, 9339−9341. (25) Chiu, J. J.; Kim, B. J.; Yi, G. R.; Bang, J.; Kramer, E. J.; Pine, D. J. Distribution of nanoparticles in lamellar domains of block copolymers. Macromolecules 2007, 40, 3361−3365.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.8b01988. Structural characteristics of the BBCPs (Table S1); GPC curves and NMR spectra of the BBCPs (Figure S1); size analysis of the NPs (Figure S2); characterization of ligand composition and density (Table S2); curves corresponding to the radial integration of the Fourier transformation of the SEM and SFM images of the neat BBCPs (Figure S3); quantitative height analysis on the assembly of BBCPsym with both types of NPs (Figure S4) (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: jrzayev@buffalo.edu. ORCID

Uri Raviv: 0000-0001-5992-9437 Javid Rzayev: 0000-0002-9280-1811 Roy Shenhar: 0000-0002-0631-1542 Funding

Financial support for this work was provided by the National Science Foundation (DMR-1409467). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Meirav Oded for a sample of the PS-SH ligands and Carmen Tamburu for the SAXS measurements. Y.A. thanks the Hebrew University for a doctoral fellowship; L.F. acknowledges the Levtzion Foundation for a doctoral fellowship.

■

REFERENCES

(1) Haryono, A.; Binder, W. H. Controlled arrangement of nanoparticle arrays in block-copolymer domains. Small 2006, 2, 600−611. (2) Kao, J.; Thorkelsson, K.; Bai, P.; Rancatore, B. J.; Xu, T. Toward functional nanocomposites: taking the best of nanoparticles, polymers, and small molecules. Chem. Soc. Rev. 2013, 42, 2654−2678. (3) Tang, Z. Y.; Kotov, N. A. One-dimensional assemblies of nanoparticles: Preparation, properties, and promise. Adv. Mater. 2005, 17, 951−962. (4) Bockstaller, M. R.; Mickiewicz, R. A.; Thomas, E. L. Block copolymer nanocomposites: Perspectives for tailored functional materials. Adv. Mater. 2005, 17, 1331−1349. (5) Rycenga, M.; Cobley, C. M.; Zeng, J.; Li, W.; Moran, C. H.; Zhang, Q.; Qin, D.; Xia, Y. Controlling the Synthesis and Assembly of J

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules (26) Chiu, J. J.; Kim, B. J.; Kramer, E. J.; Pine, D. J. Control of nanoparticle location in block copolymers. J. Am. Chem. Soc. 2005, 127, 5036−5037. (27) Li, Q. F.; He, J. B.; Glogowski, E.; Li, X. F.; Wang, J.; Emrick, T.; Russell, T. P. Responsive assemblies: Gold nanoparticles with mixed ligands in microphase separated block copolymers. Adv. Mater. 2008, 20, 1462−1466. (28) Ploshnik, E.; Langner, K. M.; Halevi, A.; Ben-Lulu, M.; Müller, A. H. E.; Fraaije, J. G. E. M.; Agur Sevink, G. J.; Shenhar, R. Hierarchical Structuring in Block Copolymer Nanocomposites through Two Phase-Separation Processes Operating on Different Time Scales. Adv. Funct. Mater. 2013, 23, 4215−4226. (29) Halevi, A.; Halivni, S.; Oded, M.; Müller, A. H. E.; Banin, U.; Shenhar, R. Co-Assembly of A-B Diblock Copolymers with B′-type Nanoparticles in Thin Films: Effect of Copolymer Composition and Nanoparticle Shape. Macromolecules 2014, 47, 3022−3032. (30) Hamdoun, B.; Ausserré, D.; Cabuil, V.; Joly, S. Copolymer nanoparticle composites: Lamellar period. J. Phys. II 1996, 6, 503− 510. (31) Song, D. P.; Lin, Y.; Gai, Y.; Colella, N. S.; Li, C.; Liu, X. H.; Gido, S.; Watkins, J. J. Controlled Supramolecular Self-Assembly ofLarge Nanoparticles in Amphiphilic Brush Block Copolymers. J. Am. Chem. Soc. 2015, 137, 3771−3774. (32) Diaz, J.; Pinna, M.; Zvelindovsky, A. V.; Asta, A.; Pagonabarraga, I. Cell Dynamic Simulations of Diblock Copolymer/ Colloid Systems. Macromol. Theory Simul. 2017, 26, 1600050. (33) Kim, B. J.; Chiu, J. J.; Yi, G. R.; Pine, D. J.; Kramer, E. J. Nanoparticle-induced phase transitions in diblock-copolymer films. Adv. Mater. 2005, 17, 2618−2622. (34) Lin, Y.; Böker, A.; He, J. B.; Sill, K.; Xiang, H. Q.; Abetz, C.; Li, X. F.; Wang, J.; Emrick, T.; Long, S.; Wang, Q.; Balazs, A.; Russell, T. P. Self-directed self-assembly of nanoparticle/copolymer mixtures. Nature 2005, 434, 55−59. (35) Edrington, A. C.; Urbas, A. M.; DeRege, P.; Chen, C. X.; Swager, T. M.; Hadjichristidis, N.; Xenidou, M.; Fetters, L. J.; Joannopoulos, J. D.; Fink, Y.; Thomas, E. L. Polymer-based photonic crystals. Adv. Mater. 2001, 13, 421−425. (36) Sheiko, S. S.; Sumerlin, B. S.; Matyjaszewski, K. Cylindrical molecular brushes: Synthesis, characterization, and properties. Prog. Polym. Sci. 2008, 33, 759−785. (37) Verduzco, R.; Li, X. Y.; Pesek, S. L.; Stein, G. E. Structure, function, self-assembly, and applications of bottlebrush copolymers. Chem. Soc. Rev. 2015, 44, 2405−2420. (38) Rzayev, J. Molecular Bottlebrushes: New Opportunities in Nanomaterials Fabrication. ACS Macro Lett. 2012, 1, 1146−1149. (39) Wintermantel, M.; Schmidt, M.; Tsukahara, Y.; Kajiwara, K.; Kohjiya, S. Rodlike combs. Macromol. Rapid Commun. 1994, 15, 279−284. (40) Wintermantel, M.; Gerle, M.; Fischer, K.; Schmidt, M.; Wataoka, I.; Urakawa, H.; Kajiwara, K.; Tsukahara, Y. Molecular bottlebrushes. Macromolecules 1996, 29, 978−983. (41) Runge, M. B.; Dutta, S.; Bowden, N. B. Synthesis of comb block copolymers by ROMP, ATRP, and ROP and their assembly in the solid state. Macromolecules 2006, 39, 498−508. (42) Rathgeber, S.; Pakula, T.; Wilk, A.; Matyjaszewski, K.; Beers, K. L. On the shape of bottle-brush macromolecules: Systematic variation of architectural parameters. J. Chem. Phys. 2005, 122, 124904. (43) Dalsin, S. J.; Hillmyer, M. A.; Bates, F. S. Linear Rheology of Polyolefin-Based Bottlebrush Polymers. Macromolecules 2015, 48, 4680−4691. (44) Hu, M.; Xia, Y.; McKenna, G. B.; Kornfield, J. A.; Grubbs, R. H. Linear Rheological Response of a Series of Densely Branched Brush Polymers. Macromolecules 2011, 44, 6935−6943. (45) Vlassopoulos, D.; Fytas, G.; Loppinet, B.; Isel, F.; Lutz, P.; Benoit, H. Polymacromonomers: Structure and dynamics in nondilute solutions, melts, and mixtures. Macromolecules 2000, 33, 5960−5969. (46) Rzayev, J. Synthesis of Polystyrene-Polylactide Bottlebrush Block Copolymers and Their Melt Self-Assembly into Large Domain Nanostructures. Macromolecules 2009, 42, 2135−2141.

(47) Dalsin, S. J.; Rions-Maehren, T. G.; Beam, M. D.; Bates, F. S.; Hillmyer, M. A.; Matsen, M. W. Bottlebrush Block Polymers: Quantitative Theory and Experiments. ACS Nano 2015, 9, 12233− 12245. (48) Gai, Y.; Song, D. P.; Yavitt, B. M.; Watkins, J. J. Polystyreneblock-poly(ethylene oxide) Bottlebrush Block Copolymer Morphology Transitions: Influence of Side Chain Length and Volume Fraction. Macromolecules 2017, 50, 1503−1511. (49) Yavitt, B. M.; Gai, Y.; Song, D. P.; Winter, H. H.; Watkins, J. J. High Molecular Mobility and Viscoelasticity of Microphase-Separated Bottlebrush Diblock Copolymer Melts. Macromolecules 2017, 50, 396−405. (50) Sveinbjornsson, B. R.; Weitekamp, R. A.; Miyake, G. M.; Xia, Y.; Atwater, H. A.; Grubbs, R. H. Rapid self-assembly of brush block copolymers to photonic crystals. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 14332−14336. (51) Runge, M. B.; Bowden, N. B. Synthesis of high molecular weight comb block copolymers and their assembly into ordered morphologies in the solid state. J. Am. Chem. Soc. 2007, 129, 10551− 10560. (52) Gu, W. Y.; Huh, J.; Hong, S. W.; Sveinbjornsson, B. R.; Park, C.; Grubbs, R. H.; Russell, T. P. Self-Assembly of Symmetric Brush Diblock Copolymers. ACS Nano 2013, 7, 2551−2558. (53) Xia, Y.; Olsen, B. D.; Kornfield, J. A.; Grubbs, R. H. Efficient Synthesis of Narrowly Dispersed Brush Copolymers and Study of Their Assemblies: The Importance of Side-Chain Arrangement. J. Am. Chem. Soc. 2009, 131, 18525−18532. (54) Bolton, J.; Bailey, T. S.; Rzayev, J. Large Pore Size Nanoporous Materials from the Self-Assembly of Asymmetric Bottlebrush Block Copolymers. Nano Lett. 2011, 11, 998−1001. (55) Song, D. P.; Gai, Y.; Yavitt, B. M.; Ribbe, A.; Gido, S.; Watkins, J. J. Structural Diversity and Phase Behavior of Brush Block Copolymer Nanocomposites. Macromolecules 2016, 49, 6480−6488. (56) Bolton, J.; Rzayev, J. Tandem RAFT-ATRP Synthesis of Polystyrene-Poly(Methyl Methacrylate) Bottlebrush Block Copolymers and Their Self-Assembly into Cylindrical Nanostructures. ACS Macro Lett. 2012, 1, 15−18. (57) Fenyves, R.; Schmutz, M.; Horner, I. J.; Bright, F. V.; Rzayev, J. Aqueous Self-Assembly of Giant Bottlebrush Block Copolymer Surfactants as Shape-Tunable Building Blocks. J. Am. Chem. Soc. 2014, 136, 7762−7770. (58) Song, D.-P.; Li, C.; Colella, N. S.; Xie, W.; Li, S.; Lu, X.; Gido, S.; Lee, J.-H.; Watkins, J. J. Large-Volume Self-Organization of Polymer/Nanoparticle Hybrids with Millimeter-Scale Grain Sizes Using Brush Block Copolymers. J. Am. Chem. Soc. 2015, 137, 12510− 12513. (59) Kim, B. J.; Bang, J.; Hawker, C. J.; Kramer, E. J. Effect of areal chain density on the location of polymer-modified gold nanoparticles in a block copolymer template. Macromolecules 2006, 39, 4108−4114. (60) Kim, B. J.; Fredrickson, G. H.; Kramer, E. J. Effect of polymer ligand molecular weight on polymer-coated nanoparticle location in block copolymers. Macromolecules 2008, 41, 436−447. (61) Sheiko, S. S.; Gerle, M.; Fischer, K.; Schmidt, M.; Möller, M. Wormlike polystyrene brushes in thin films. Langmuir 1997, 13, 5368−5372. (62) Hong, S. W.; Gu, W.; Huh, J.; Sveinbjornsson, B. R.; Jeong, G.; Grubbs, R. H.; Russell, T. P. On the Self-Assembly of Brush Block Copolymers in Thin Films. ACS Nano 2013, 7, 9684−9692. (63) Vayer, M.; Hillmyer, M. A.; Dirany, M.; Thevenin, G.; Erre, R.; Sinturel, C. Perpendicular orientation of cylindrical domains upon solvent annealing thin films of polystyrene-b-polylactide. Thin Solid Films 2010, 518, 3710−3715. (64) The reason the curve fitting in Figure 2 deviates from the experimental data in the mid-q range is that any kind of structural change could smear the model features. The fitting model is likely to be the dominant structure, owing to the fact that it reproduces both the locations and the intensities of the main peaks. However, coexisting structures arising from varying graft, block, or total K

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules copolymer lengths might have contributed slightly different scattering curves, which are reflected in the measured signal. (65) As the assembly is quasi-two-dimensional, the phrase “dispersed in the PS domains” refers only to the lateral (in-plane) directions, not the direction normal to the substrate. (66) Biresaw, G.; Carriere, C. J. Interfacial tension of poly(lactic acid)/polystyrene blends. J. Polym. Sci., Part B: Polym. Phys. 2002, 40, 2248−2258. (67) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. Synthesis of Thiol-Derivatized Gold Nanoparticles in a 2-Phase Liquid-Liquid System. J. Chem. Soc., Chem. Commun. 1994, 0, 801− 802. (68) Teranishi, T.; Hasegawa, S.; Shimizu, T.; Miyake, M. Heatinduced size evolution of gold nanoparticles in the solid state. Adv. Mater. 2001, 13, 1699−1701. (69) Frens, G. Controlled Nucleation for Regulation of Particle-Size in Monodisperse Gold Suspensions. Nature (London), Phys. Sci. 1973, 241, 20−22. (70) Pavan, M. J.; Ploshnik, E.; Shenhar, R. Nanoparticle Assembly on Topographical Polymer Templates: Effects of Spin Rate, Nanoparticle Size, Ligand, and Concentration. J. Phys. Chem. B 2012, 116, 13922−13931. (71) Liu, X. O.; Atwater, M.; Wang, J. H.; Huo, Q. Extinction coefficient of gold nanoparticles with different sizes and different capping ligands. Colloids Surf., B 2007, 58, 3−7. (72) Rasband, W. S. ImageJ, version 1.39u; U.S. National Institutes of Health: Bethesda, MD, 1997−2014; http://imagej.nih.gov/ij/. (73) Nadler, M.; Steiner, A.; Dvir, T.; Szekely, O.; Szekely, P.; Ginsburg, A.; Asor, R.; Resh, R.; Tamburu, C.; Peres, M.; Raviv, U. Following the structural changes during zinc-induced crystallization of charged membranes using time-resolved solution X-ray scattering. Soft Matter 2011, 7, 1512−1523. (74) Hammersley, A. P.; Svensson, S. O.; Hanfland, M.; Fitch, A. N.; Hausermann, D. Two-dimensional detector software: From real detector to idealised image or two-theta scan. High Pressure Res. 1996, 14, 235−248. (75) Ben-Nun, T.; Ginsburg, A.; Szekely, P.; Raviv, U. X plus: a comprehensive computationally accelerated structure analysis tool for solution X-ray scattering from supramolecular self-assemblies. J. Appl. Crystallogr. 2010, 43, 1522−1531. (76) Ginsburg, A.; Ben-Nun, T.; Asor, R.; Shemesh, A.; Ringel, I.; Raviv, U. Reciprocal Grids: A Hierarchical Algorithm for Computing Solution X-ray Scattering Curves from Supramolecular Complexes at High Resolution. J. Chem. Inf. Model. 2016, 56, 1518−1527. (77) Szekely, P.; Ginsburg, A.; Ben-Nun, T.; Raviv, U. Solution Xray Scattering Form Factors of Supramolecular Self-Assembled Structures. Langmuir 2010, 26, 13110−13129.

L

DOI: 10.1021/acs.macromol.8b01988 Macromolecules XXXX, XXX, XXX−XXX