In Situ Observation of Directed Nanoparticle Aggregation During the

Jan 17, 2014 - In Situ Observation of Directed Nanoparticle Aggregation During the Synthesis of Ordered Nanoporous Metal in Soft Templates. Lucas R. P...
1 downloads 13 Views 3MB Size
Download PDF
Article pubs.acs.org/cm

In Situ Observation of Directed Nanoparticle Aggregation During the Synthesis of Ordered Nanoporous Metal in Soft Templates Lucas R. Parent,*,† David B. Robinson,‡ Patrick J. Cappillino,‡ Ryan J. Hartnett,‡ Patricia Abellan,§ James E. Evans,∥ Nigel D. Browning,†,§ and Ilke Arslan§ †

Department of Chemical Engineering and Materials Science, University of California−Davis, One Shields Avenue, Davis, California 95616, United States ‡ Sandia National Laboratories, P.O. Box 969, Mail Stop 9291, Livermore, California 94551, United States § Fundamental and Computational Science Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States ∥ Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland, Washington 99352, United States S Supporting Information *

ABSTRACT: The prevalent approach to developing new nanomaterials is a trial-and-error process of iteratively altering synthesis procedures and then characterizing the resulting nanostructures. This is fundamentally limited in that the growth processes that occur during synthesis can be inferred only from the final synthetic structure. Directly observing real-time nanomaterial growth provides unprecedented insight into the relationship between synthesis conditions and product evolution and facilitates a mechanistic approach to nanomaterial development. Here, we use in situ liquid-stage scanning transmission electron microscopy to observe the growth of mesoporous palladium in a solvated block copolymer (BCP) template under various synthesis conditions, and we ultimately determined a refined synthesis procedure that yields extended structures with ordered pores. We found that after sufficient drying time of the casting solvent (tetrahydrofuran, THF), the BCP assembles into a rigid, cylindrical micelle array with a high degree of short-range order but poor long-range order. Upon slowing the THF evaporation rate using a solvent-vapor anneal step, the long-range order was greatly improved. The electron beam induces nucleation of small particles in the aqueous phase around the micelles. The small particles then flocculate and grow into denser structures that surround, but do not overgrow, the micelles, forming an ordered mesoporous structure. The microscope observations revealed that pore disorder can be addressed prior to metal reduction and is not invariably induced by the Pd growth process itself, allowing for more rapid optimization of the synthetic method.

■

INTRODUCTION Materials with tailored nanoscale structure show enhanced properties in applications including catalysis,1−3 hydrogen storage,3−6 electrochemical energy storage,7 biomedicine,8 and water purification.9 For example, nanoporous palladium can absorb and release hydrogen much more rapidly than the bulk metal, and it exhibits lower hysteresis.4−6 Various synthesis methods have been established for the growth of high-surfacearea, porous, inorganic nanostructures for such applications, including assembly of modular building blocks,1,2,10 nanoparticle self-assembly and consolidation,11,12 and growth within hard or soft templates.3−6,13−16 Slight alterations of the synthesis conditions and procedures can cause significant changes to the products, and the range of conditions leading to optimal properties is often narrow. The conventional approach to achieve this optimum consists of iterations of synthesis followed by characterization of the final product by methods including electron microscopy. Subtle but important growth stages and mechanisms that may occur © 2014 American Chemical Society

during synthesis are imperceptible with such techniques and can be inferred only from the final structure. In situ nanoscale characterization techniques that enable real-time, direct observation during structural growth in liquid (and solid− liquid) synthesis environments could provide the means to debug the process rapidly and to optimize the properties of the products. Recent advancements in atomic force microscopy (AFM),17−20 scanning tunneling microscopy (STM),19 X-ray diffraction (XRD),21,22 X-ray photoelectron spectroscopy (XPS), 23 small-angle X-ray scattering (SAXS), 22 X-ray absorption spectroscopy (XAS),21,24 and (scanning) transmission electron microscopy ((S)TEM)25−33 have made in situ characterization of certain liquid−solid systems possible. In particular, (S)TEM liquid-sample holders have enabled Received: October 24, 2013 Revised: January 13, 2014 Published: January 17, 2014 1426

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433

Chemistry of Materials

Article

anneal conditions (experiment 3), immediately after applying the droplet of solution to the Si chip, the chip was transferred using Teflon tipped tweezers onto a pedestal of glass slides inside a Petri dish containing a pool of THF/H2O (∼3:2 v/v). The cover of the Petri dish was placed on top, and the Si chip with solution was left inside for 60 min (21 °C). The Si chip was removed from the Petri dish and placed back on the lab bench for 5 min of ambient air exposure at 21 °C. For all three conditions, a second Si chip (10 nm thick and 50 × 50 μm2 SiNx membrane window, 60 s Ar/O2 plasma clean) with SiNx membrane facing down was carefully placed on top of the first Si chip (once the respective preparation time had elapsed), sealing the sample (now dried to a thin film) between the two SiNx membranes. This sandwich of Si chips and sample was transferred into the tip of the Hummingbird Scientific in situ liquid stage. Using an optical microscope and Teflon tweezers, the alignment of two transparent windows was checked and corrected as needed. The tip of the in situ liquid stage was sealed, and the holder was pumped to vacuum in the prepumping chamber. After achieving a value below 4.0 × 10−6 mbar, the stage was loaded into the FEI Titan/Cs STEM. Microscope. All in situ growth and imaging was performed using a Cs-corrected FEI Titan running FEI TIA software in STEM mode. The corrector was aligned using a Au standard sample for resolution below 1.1 Å. The microscope was operated at 300 kV using the highangle annular dark-field (HAADF) detector and 50 μm condenser (probe forming) aperture. The STEM probe was adjusted to have an average beam current of 5.7 pA. All images (and frames in Supporting Information Movies S1−S3) were taken under one of three conditions: 512 × 512 pixel image and 3 μs dwell time (Figures 1A,B, 2A,B, 3A−C, and S8A,B), 1024 × 1024 pixel image and 5 μs dwell time (Figures 1C−F, 2C,E, and 3D−F), or 1024 × 1024 pixel image and 9 μs dwell time (Figures 2D,F, 4A,B, and S6A,B). All images and movies were acquired at a corner of the SiNx windows, where the least membrane bulging occurs and sample thickness is minimized. The screen-capture software CamStudio was used to record video of the TIA scanning images using a capture rate of 2 frames per second. All STEM images presented (with the exception of Figures 2D,F, 4A,B, and S6A,B, which are TIA record images) are frames from the CamStudio movies. The full movie files are available as Supporting Information Movies S1−S3, which have been speedadjusted using Camtasia Studio 8 video-editing software. The Fourier transformations and radial integrals in Figures S2 and S5 were generated using FEI TIA software. The Fourier transformation in Figure 4C was generated using Gatan DigitalMicrograph software for the indicated blue region of the respective TIA record image (Figure 4B).

numerous in situ studies of the electron-beam-induced growth of nanostructures or nanoparticles (NPs) from aqueous precursor solutions28,30−33 and of the motion, aggregation, and local ordering of particles in liquid.25,29,32 In a previous study,32 we used an in situ STEM platform to investigate the growth processes that occur during synthesis of mesoporous Pd by reduction of a Pd salt in a concentrated surfactant (Brij 56) template that is known to form densely packed cylindrical micelles.4,5 When metals such as platinum or rhodium are grown using a Brij 56 template, the resulting pore arrays are also densely packed, but for palladium, the pores are disordered, and this pore disorder contributes to their thermal instability.4,5,34 In the STEM, Pd seed particles were observed to nucleate and then grow rapidly to sizes exceeding the expected dimension of the micelle array. This was thought to be a major contributing factor to the disorder of the final pore network, and we suggested that augmenting template rigidity or limiting the transport of metallic Pd through the template could help maintain long-range micelle order during synthesis. Greater rigidity can be expected from a template that forms larger micelles. For example, it is known that polymersomes (water-containing spherical shells of block copolymers) are more mechanically robust than liposomes (analogous structures formed from lipid bilayers).35 Several reported approaches to nanoporous noble-metal synthesis use a polystyrene-blockpoly(ethylene oxide) (PS-b-PEO) block copolymer (BCP) soft template.6,36 By adjusting the molecular weight of the constituent polymeric blocks, the BCP micelle diameter can be tuned, with reported values including 7 and 13 nm.6 Use of this template led to a pronounced increase in the long-range pore order compared to Brij 56-templated palladium, with many particles exhibiting hexagonally packed pore arrays.4−6 The increased pore size and order contributed to significant improvements in thermal stability.5,6 Forming the larger micelles requires use and removal of a cosolvent such as tetrahydrofuran (THF). If the removal is not performed under a narrow range of conditions, then variations in pore order can result. This complicates attempts to change batch size and other conditions that may aid practical use of the reaction. In the work presented here, we use in situ liquid-stage STEM to observe the reduction of aqueous palladium salts to metal in a PS-b-PEO micelle template, to determine the mechanism of growth of the metal nanostructure, and to use these results to identify improvements to pore order. We found that the BCP template is sufficiently large and rigid to prevent overgrowth of Pd metal, but the prevention of overgrowth and the selfassembly of long-range micelle order occur only after optimizing the rate and extent of THF removal.

■

■

RESULTS AND DISCUSSION We present three experiments of electron-beam-induced growth of metallic Pd from a solution containing THF, water, PS-b-PEO (BCP), hydrochloric acid, and a Pd salt using an in situ liquid stage and STEM. Each of the three experiments involved different drying conditions prior to in situ growth and observation. In the first two experiments, a 0.5 μL droplet of solution on a silicon nitride membrane was allowed to evaporate in air for 5 (experiment 1) or 15 min (experiment 2). These evaporation times are comparable to the rotary evaporation times used in ref 6. In the third experiment, the 0.5 μL droplet of initial solution was subject to a 60 min solventvapor anneal (SVA) followed by 5 min of evaporation in air. Further details are described in the Experimental Section. Each sealed sample was then loaded into the STEM for direct observation and video recording of the electron-beam-induced Pd growth within each of the three different BCP templates. The 300 kV electron probe used for nanoscale STEM imaging actively interacts with any of these samples by breaking bonds and freeing valence electrons, generating reactive radical and ionic species. In liquid samples, any radical species created

EXPERIMENTAL SECTION

Samples. A reactant solution was prepared by mixing 16.5 mg of polystyrene-block-poly(ethylene oxide) in 0.924 mL of tetrahydrofuran (THF) with 4.5 mg of (NH4)2PdCl4 in 0.076 mL of 0.02 M aqueous HCl. The polymer was obtained from Polymer Source, with a polystyrene block size of 2.3 kDa and poly(ethylene oxide) block size of 3.1 kDa. To prepare each of the three different experiments, one Si chip with a 10 nm thick (50 × 50 μm2 lateral dimensions) SiNx membrane window (Ar/O2 plasma cleaned for 60 s) was placed on a clean surface with the SiNx membrane facing up. Using a micropipet, a 0.5 μL droplet of initial reactant solution was applied to surface of the SiNx membrane, being careful not to rupture the window. For the ambient air-exposure conditions (experiments 1 and 2), the Si chip, with sample solution on the surface, was left on the lab bench at 21 °C for 5 (experiment 1) or 15 min (experiment 2). For the solvent-vapor 1427

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433

Chemistry of Materials

Article

Figure 1. DF STEM images of the growth of Pd nanostructure (appearing as high intensity) by electron-beam-induced reduction of Pd salts within the PS-b-PEO BCP template (appearing as low intensity) after 5 min of ambient air exposure. A seventy second total growth period is displayed with no electron-beam illumination prior to capturing image A. The cumulative incident electron dose for each image is displayed in the lower right corner, and the time from initially opening the electron beam for each image is displayed in the upper right corner.

nuclei grow into dendritic particles. Plausible growth mechanisms include autocatalysis, where reducing species contribute electrons to the surface and then the metal salt accepts those electrons to form metal on the surface (growth from solution), and an aggregation mechanism, where small particles attach together or to a larger particle. A network of darker areas is manifest in the scan area, presumably occupied by the BCP/THF, which have relatively low Z number compared to Pd. The dark areas apparently restrict the volume available for Pd growth, limiting aggregation and flocculation of the particles to structures of dimensions smaller than the space between BCP regions and restricting the potential locations of initial nucleation. The disordered but interconnected BCP/ THF phase is deformed and displaced by the growing Pd particles (Figure 1D−F). The BCP clearly has not selfassembled into an ordered or aligned array of cylindrical micelles. As electron-beam-induced growth slows (presumably because of local Pd-precursor depletion), these large BCP/THF veins (∼10−20 nm in diameter) remain as a highly irregular and interconnected pore network in the synthesized Pd film. The STEM electron-beam-induced growth results for the 15 min air evaporation sample are illustrated in Figure 2 over an 82 s interval (Movie S2), starting with the first frame after opening the beam (Figure 2A). In much the same manner as observed with the 5 min sample (Figure 1), growth arises via homogeneous nucleation of Pd nanoparticles over the entire scan area within the first few seconds of electron-beam

are mobile within the solution and inevitably undergo secondary reactions that can oxidize or reduce other species within the sample. Electron-beam interactions with the aqueous Pd-salt solution can produce a variety of oxidizing species including Cl2−•, HO•, H2O2, and O2−• as well as reducing species such as H•, H2, and solvated free electrons.37−40 Incident electrons that experience sufficient scattering events in the solution can become trapped as highly reducing aqueous electrons. Although the electron beam creates both reducing and oxidizing radicals, the net direction of secondary reactions between these radical species and the liquid sample is toward the reduction of tetrachloropalladate ([PdCl4]2−) precursor into metallic Pd0, the same product obtained on gram-scale ex situ synthesis using H2 gas reduction.31,32 Figure 1 shows a series of dark-field (DF) STEM images from an electron-beam-induced reduction and growth movie for the 5 min air exposure sample over a 70 s interval (Movie S1), starting with the first scan after opening the beam. This area of the sample had no prior exposure to the electron beam, and the field of view contains two pre-existing ∼10 nm particles (Figure 1A). Although these particles could have served as seed locations for heterogeneous growth, Pd growth proceeds instead through apparent homogeneous nucleation over the entire scan area (Figure 1B), similar to previous results of electron-beam-induced Pd growth from aqueous Pd-salt solutions.32 After further reduction (Figure 1C) and using longer beam dwell time, Pd nucleation subsides as existing 1428

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433

Chemistry of Materials

Article

Figure 2. DF STEM images of the growth of Pd nanostructure (appearing as high intensity) by electron-beam-induced reduction of Pd salts within the PS-b-PEO BCP template (appearing as low intensity) after 15 min of ambient air exposure. An eighty-two second total growth period is displayed with no electron-beam illumination prior to capturing image A. The magnification was increased just before capturing image E to zoom in on the region marked by the red box in image D. The cumulative incident electron dose for each image is displayed in the lower right corner, and the time from initially opening the electron beam for each image is displayed in the upper right corner.

evaporation) are more structurally robust. Once the micelle template becomes visible, the morphology of the array is not substantially disrupted by the growing Pd particles (Figure S4). The longer drying time has likely allowed for greater association between neighboring hydrophobic blocks, with less interference from residual solvent, which may decrease between 5 and 15 min. The additional time alone may have permitted hydrophobic blocks to assemble into more welldefined and extended micelles. In both cases, the micelle template prevents particle migration, limits aggregation, and maintains metal-free space in the growing Pd structure. Despite this, the micelle template does not significantly inhibit transport of the Pd ions or Pd-salt molecules within the liquid phase (the aqueous Pd-salt solution). Pd growth is predominantly uniform across the scan area, with only a slight decrease in Pd intensity (thinner) in the center, where Pd ion diffusion distances from the periphery of the scan area are greatest (following local precursor depletion). We observed that allowing more time for solvent evaporation and polymer self-assembly produces a more homogeneous Pd nanostructure with consistent pore sizes and short-range order. However, the absence of long-range order in the micelle template limits the achievable pore density of the final synthetic structure and likely results in more junctions or other defects than would be expected in a pore array that has long-range order. The reduction of such defects is expected to contribute to thermal and mechanical stability of the product.

illumination (Figure 2B). With continued exposure to the STEM probe, nucleation gives way to growth as the prevalent process. Dark regions become contrasted against the bright intensity caused by scattering from metallic Pd (Figure 2C−F). The contrast reveals that the BCP has assembled into a relatively rigid array of 6.0 ± 1.6 nm diameter micelles (Figure S1), which limit the initial Pd nucleation sites and confine the available volume for further growth. This array exhibits shortrange order but lacks long-range order, evident by the diffuse ring in the Fourier transformation of Figure 2D (Figure S2) and the wide distribution of spacing distances between adjacent micelles (Figure S3, micelle spacing of 4.56 ± 1.2 nm). If the geometry of an individual micelle can be described as cylindrical, which is expected from prior studies of this polymer,6,36 then the cylinders are apparently oriented perpendicular to the substrate (parallel to the optical axis). We do not have a direct measure of the height of the cylinders, but the observed lack of Pd in these regions suggest that the micelles extend at least above and below the ∼13 nm depth of field of the STEM probe.41 Plasma cleaning the SiNx substrate was necessary to obtain micelles of this morphology. Fifteenminute THF evaporation under ambient air allows for shortrange organization but does not provide sufficient time for long-range ordering. Compared to the BCP/THF regions in Figure 1 (5 min air evaporation), the BCP micelles in Figure 2 (15 min air 1429

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433

Chemistry of Materials

Article

Figure 3. DF STEM images of the growth of Pd nanostructure (appearing as high intensity) by electron-beam-induced reduction of Pd salts within the PS-b-PEO BCP template (appearing as low intensity) after 60 min of solvent-vapor anneal and 5 min of ambient air exposure. A sixty-one second total growth period is displayed with no electron-beam illumination prior to capturing image A. The magnification was increased just before capturing image F to zoom in on the region marked by the red box in image E, part of which is out of the view area in E. The cumulative incident electron dose for each image is displayed in the lower right corner, and the time from initially opening the electron beam for each image is displayed in the upper right corner.

Using a vapor environment of THF/H2O (∼3:2), we exposed the initially deposited solution to a 60 min SVA followed by a 5 min additional evaporation time under ambient air to provide an amount of drying time comparable to the previous samples. A mixed-vapor environment containing both water and THF was intended to suppress the evaporation of water from the aqueous Pd-salt phase and to help maintain the original Pd reactant concentration. The in situ STEM growth results for this SVA sample are shown in Figure 3 over 61 s of electron-beam exposure (Movie S3). Growth proceeds in a nearly identical manner to that observed for the 15 min air evaporation experiment (Figure 2), with initial homogeneous nucleation of Pd seed particles across the scan area and the appearance of rigid 6.63 ± 0.7 nm diameter micelles, presumably the 2D projection of cylindrical micelles oriented parallel with the optical axis (Figures 3A−C and S1). As the intensity of the Pd increases (from increased thickness) and the contrast relative to the micelles improves, regions of close-packed long-range micelle order are evident (Figures 3D−F and 4A) in which micelles two or more micelle diameters away from each other are aligned. The long-range order is most prominent in patches about 100 nm wide, as indicated by the spots in the Fourier transformation of the region enclosed by the yellow box in Figure 4A (Figure S5). Compared to the 15 min air exposure conditions, the 60 min

The longer solvent-evaporation time described earlier is essential for the assembly of robust cylindrical micelles, but with any additional time, we expect that the polymer chains are insufficiently mobile to anneal into a more ordered state. Thermal annealing, optimized BCP film thickness, and modifications in substrate surface chemistry are known to improve BCP assembly kinetics and to foster micelle ordering.42−45 Most notably, solvent-vapor annealing has been shown to induce the formation of highly ordered PS-bPEO micelle arrays.46−49 In this process, the BCP/THF film is enclosed in a solvent-vapor environment (rather than ambient air), slowing the rate of THF evaporation by keeping the vapor phase near saturation and modifying the interfacial energy of the PS and PEO blocks at the BCP/vapor interface.46 Additionally, the solvent vapor accelerates BCP assembly kinetics by swelling the BCP film, increasing the available free volume, and lowering the glass transition temperatures of the blocks to improve polymer mobility, altering interactions between the blocks and tempering detrimental substrate surface−BCP film interactions.43,47,50 The intent of solventvapor annealing is to maintain an optimal range of residual solvent that is sufficient to accelerate assembly but not disrupt it through dissolution, providing enough time to complete the process of self-assembly. 1430

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433

Chemistry of Materials

Article

Figure 4. (A) DF STEM image of the Pd nanostructure grown by ∼30 s of electron-beam illumination in the confines of the ordered BCP micelle template after 60 min of SVA and 5 min of air exposure (same sample as Figure 3). Yellow and red boxes indicate two regions of distinct long-range order. (B) High-magnification DF STEM image of 2.77 ± 0.4 nm Pd nanoparticles (Figure S6) grown by ∼15 s of electron-beam illumination at lower magnification and in a different region of the same sample as image A (60 min SVA). The blue box indicates one crystalline Pd NP aligned near the zone axis showing (111) lattice fringes. (C) Fourier transformation of the region enclosed in the blue box in image B. Red circles indicate Pd (111) spots.

SVA micelles are slightly larger and more homogeneous in diameter (Figure S1). The SVA micelles are also significantly more closely packed overall and have a smaller distribution of individual spacing distances between adjacent micelles (Figure S3, SVA micelle spacing of 3.51 ± 0.9 vs 4.56 ± 1.2 nm for 15 min air evaporation). At higher magnification (Figures 3F, 4B, and S6), it is apparent that small Pd seed particles (2.73 ± 0.4 nm diameter) have formed and then loosely aggregated with adjacent particles. Growth has occurred only in the free space between the BCP micelles, without significantly disturbing or altering the micelle array (Figure S7). The short- and long-range order of the template is maintained even as the small Pd particles begin to impinge upon each other and flocculate, forming multigrained Pd networks around the BCP micelles. These Pd networks are composed of adjoining crystalline metallic Pd particles, revealed by the (111) Pd lattice fringes of particles oriented near the zone axis, highlighted by the blue box in Figure 4B and corresponding Fourier transformation in Figure 4C. A similar fine structure of adjoining Pd grains is also seen in ex situ TEM studies of mesoporous Pd particles grown by H2 gas reduction in the PS-b-PEO micelle template (Figure 7 in ref 6), indicating that the electron-beam-induced growth processes at the dose rates used here, as detailed in the Experimental Section and Supporting Information, closely simulate the growth processes during gram-scale synthesis of this material. For all three of the in situ STEM experiments here, the PS-bPEO micelle template is not visible until sufficient growth of metallic Pd has occurred (Figures 1B, 2B, and 3B), and even at that point, the low Z-number BCP is not directly observed; the BCP micelles are presumed to occupy the regions of lower intensity that lack Pd particles. Our evidence does not unequivocally show that the micelle template remains fixed in

the same form before and after growth is initiated or rule out that the growing Pd induces order in the film that did not exist previously. The sensitivity of the results to the preparation conditions prior to introduction to the microscope provides indirect but strong evidence that the structure is essentially fixed, but the question can be addressed further. Ex situ AFM,43,46−49,51 scanning electron microscopy (SEM),48 and energy-filtered TEM51,52 have been used for directly imaging the morphology and topology of spin-coated PS-b-PEO thin films on single substrates, revealing hexagonally packed micelle arrays for certain solvent-vapor annealed films. These experiments also present indirect evidence that order may be present in our system prior to growth, although our samples are expected to contain more residual water and may be perturbed by the sandwiching of the samples between two membranes. Sealing the BCP film between the two SiNx membranes can apply compressive and shear forces on the film, potentially distorting the original template morphology. Within a given sample, there will be local variations in the forces experienced by the BCP film based on the local film thickness, the local film density, and the presence of dust or other spacer particles on the SiNx membranes. We do observe distinct differences in the final Pd pore structures for the three different preparation conditions used (Figures 1−3), indicating that at least some regions of the BCP films generally retain their initial (prior to being sandwiched) micelle morphologies when sealed in the liquid stage and that further template assembly for all three samples is largely arrested at that point. It is possible that the electron beam plays a role in the template assembly process or in some form of template reformation after being enclosed in the liquid stage. However, in a separate 60 min SVA sample (Figure S8, different sample than Figure 3 but same preparation method), an ordered 1431

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433

Chemistry of Materials

Article

micelle template appears within the first STEM scan after being exposed to a low electron dose (7.88 e−/Å2 cumulative dose for first scan, with 0.94 s scan time). This suggests that the micelle array morphology was present prior to electron-beam illumination and was not a product of extended electronbeam irradiation. Compared to the previous in situ growth study of mesoporous Pd structures using the Brij 56 template,32 the results here indicate that PS-b-PEO BCP self-assembles into a larger, more rigid micelle template that more effectively confines Pd growth and controls the final pore structure and Pd morphology. Even with short drying times (Figure 1), the BCP veins are considerably more structurally robust in the presence of the growing Pd structure than the organic solventfree micelles of Brij 56, where Pd clusters were seen to migrate at an average ∼0.6 nm/s and cover distances on the order of 100 nm.32 At the longer template assembly times in this study (15 min air-dry, Figure 2, and SVA, Figure 3), the array of formed micelles is only marginally altered during Pd growth and successfully directs the developing Pd nanostructure into a more uniform mesoporous morphology with few defects or pore junctions. Larger micelle size allows for greater entanglement and reduced aqueous solubility of hydrophobic blocks. Larger spacing between micelles allows growing Pd particles to deplete reactants locally and to reduce their surface energy before encountering micelles, thereafter favoring aggregation that conforms to the template versus continued growth that may disrupt the template. Either or both of these factors (perhaps among others) may contribute to a correlation between micelle size and template stability.

annealing steps or similar efforts to control the extent and rate of solvent removal carefully to aid polymer self-assembly and to improve ordering in the resulting pores. The increased order is expected to improve the thermal and mechanical properties of the products. We have shown that in situ liquid-stage STEM is a powerful characterization tool for directly observing the nanoscale processes that occur during template-directed synthesis and for understanding the relationship between synthesis conditions, such as drying, annealing, and reaction times, and the products of the reaction, enabling a more rapid optimization of those conditions.

■

ASSOCIATED CONTENT

S Supporting Information *

Micelle diameter histograms, micelle spacing histograms, Pd nanoparticle diameter histograms, and image-to-image comparisons for Figures 2 and 3; Fourier transformations for Figures 2D and 4A; image pixel size, electron dose rate, and cumulative dose for Figures 1−3 and S8 (PDF); movie files and captions of in situ DF STEM growth of Pd nanostructure (MOV). This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

■

CONCLUSIONS Using electron-beam-induced reduction via in situ STEM imaging, we have observed the nanoscale PS-b-PEO BCP template morphology and the real-time Pd growth processes for three different synthesis conditions. The 5 min air exposure condition did not lead to the assembly of rigid or ordered BCP micelles and instead produced nonuniform, disordered, and interconnected BCP/THF veins that were deformed and to some degree displaced by the growing Pd particles. A significant improvement in the BCP template was seen after 15 min air exposure, which allowed for the assembly of a comparatively rigid, cylindrical micelle array that guided the growth of mesoporous Pd structures. The micelles were of relatively uniform size, shape, and orientation (oriented perpendicular to the SiNx substrate) but lacked long-range order. Adding a THF/H2O solvent-vapor anneal step produced a generally analogous micelle array in terms of micelle size, morphology, rigidity, and short-range order but with regions exhibiting closepacking and long-range order. The growth process under electron-beam illumination for both fully formed micelle templates (15 min air exposure and SVA with 5 min air exposure) was very similar. Growth began with homogeneous nucleation of Pd seed particles followed by continued growth by tetrachloropalladate reduction and local particle aggregation within the confined space of the template. As particles became ∼2.7 nm, they aggregate with adjacent particles, forming an interconnected network surrounding the micelles. The SVA step is of potential synthetic value when forming nanoporous metal films, which may have chemical sensing, optical, energy storage, magnetic, or other applications. For the synthesis of bulk (3D) nanoporous materials that rely on polymer templates, this work suggests the possible utility of

ACKNOWLEDGMENTS This work was conducted in the William R. Wiley Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by DOE’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. The Pacific Northwest National Laboratory is operated by Battelle for the U.S. Department of Energy under contract DE-AC05-76RL01830. This research was funded in part by the Presidential Early Career Award for Scientist and Engineers for I.A., the University of California Academic Senate and the University of California Laboratory fee research grant, the Laboratory-Directed Research and Development program at Sandia National Laboratories, and the Chemical Imaging Initiative at Pacific Northwest National Laboratory. Sandia National Laboratories is a multiprogram laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

■

REFERENCES

(1) Seo, J. S.; Whang, D.; Lee, H.; Jun, S. I.; Oh, J.; Jeon, Y. J.; Kim, K. Nature 2000, 404, 982. (2) Chui, S. S. Y.; Lo, S. M. F.; Charmant, J. P. H.; Orpen, A. G.; Williams, I. D. Science 1999, 283, 1148. (3) Joo, S. H.; Choi, S. J.; Oh, I.; Kwak, J.; Liu, Z.; Terasaki, O.; Ryoo, R. Nature 2001, 412, 169. (4) Robinson, D. B.; Fares, S. J.; Ong, M. D.; Arslan, I.; Langham, M. E.; Tran, K. L.; Clift, W. M. Int. J. Hydrogen Energy 2009, 34, 5585. (5) Robinson, D. B.; Langham, M. E.; Fares, S. J.; Ong, M. D.; Jacobs, B. W.; Clift, W. M.; Murton, J. K.; Hjelm, R. P.; Kent, M. S. Int. J. Hydrogen Energy 2010, 35, 5423.

1432

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433

Chemistry of Materials

Article

(40) Khan, H. M.; Waltz, W. L.; Woods, R. J. Radiat. Phys. Chem. 1986, 27, 41. (41) Lupini, A. R.; Borisevich, A. Y.; Idrobo, J. C.; Christen, H. M.; Biegalski, M.; Pennycook, S. J. Microsc. Microanal. 2009, 15, 441. (42) Seppala, J. E.; Lewis, R. L.; Epps, T. H. ACS Nano 2012, 6, 9855. (43) Albert, J. N. L.; Epps, T. H. Mater. Today 2010, 13, 24. (44) Sivaniah, E.; Hayashi, Y.; Iino, M.; Hashimoto, T.; Fukunaga, K. Macromolecules 2003, 36, 5894. (45) Kelly, J. Y.; Albert, J. N. L.; Howarter, J. A.; Kang, S. H.; Stafford, C. M.; Epps, T. H.; Fasolka, M. J. ACS Appl. Mater. Interfaces 2010, 2, 3241. (46) Peng, J.; Han, Y. C.; Knoll, W.; Kim, D. H. Macromol. Rapid Commun. 2007, 28, 1422. (47) Kim, B.; Hong, S. W.; Park, S.; Xu, J.; Hong, S. K.; Russell, T. P. Soft Matter 2011, 7, 443. (48) Ghoshal, T.; Shaw, M. T.; Bolger, C. T.; Holmes, J. D.; Morris, M. A. J. Mater. Chem. 2012, 22, 12083. (49) Guo, Z. Y.; Lin, M. S.; Chang, Y. T.; Sun, Y. S.; Chiang, Y. W.; Chou, C. Y.; Woon, W. Y.; Chuang, M. C. Polymer 2012, 53, 4827. (50) Gotrik, K. W.; Hannon, A. F.; Son, J. G.; Keller, B.; AlexanderKatz, A.; Ross, C. A. ACS Nano 2012, 6, 8052. (51) Allen, F. I.; Ercius, P.; Modestino, M. A.; Segalman, R. A.; Balsara, N. P.; Minor, A. M. Micron 2013, 44, 442. (52) Allen, F. I.; Watanabe, M.; Lee, Z.; Balsara, N. P.; Minor, A. M. Ultramicroscopy 2011, 111, 239.

(6) Cappillino, P. J.; Hattar, K. M.; Clark, B. G.; Hartnett, R. J.; Stavila, V.; Hekmaty, M. A.; Jacobs, B. W.; Robinson, D. B. J. Mater. Chem. A 2013, 1, 602. (7) Chmiola, J.; Yushin, G.; Gogotsi, Y.; Portet, C.; Simon, P.; Taberna, P. L. Science 2006, 313, 1760. (8) Gupta, A. K.; Gupta, M. Biomaterials 2005, 26, 3995. (9) Asanuma, H.; Hishiya, T.; Komiyama, M. Adv. Mater. 2000, 12, 1019. (10) Peng, X. G.; Manna, L.; Yang, W. D.; Wickham, J.; Scher, E.; Kadavanich, A.; Alivisatos, A. P. Nature 2000, 404, 59. (11) Klajn, R.; Bishop, K. J. M.; Fialkowski, M.; Paszewski, M.; Campbell, C. J.; Gray, T. P.; Grzybowski, B. A. Science 2007, 316, 261. (12) Krishna, K. S.; Sandeep, C. S. S.; Philip, R.; Eswaramoorthy, M. ACS Nano 2010, 4, 2681. (13) Lee, J.; Orilall, M. C.; Warren, S. C.; Kamperman, M.; Disalvo, F. J.; Wiesner, U. Nat. Mater. 2008, 7, 222. (14) Attard, G. S.; Bartlett, P. N.; Coleman, N. R. B.; Elliott, J. M.; Owen, J. R.; Wang, J. H. Science 1997, 278, 838. (15) Attard, G. S.; Glyde, J. C.; Goltner, C. G. Nature 1995, 378, 366. (16) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Nature 1992, 359, 710. (17) Jaafar, M.; Martinez-Martin, D.; Cuenca, M.; Melcher, J.; Raman, A.; Gomez-Herrero, J. Beilstein J. Nanotechnol. 2012, 3, 336. (18) Arruda, T. M.; Heon, M.; Presser, V.; Hillesheim, P. C.; Dai, S.; Gogotsi, Y.; Kalinin, S. V.; Balke, N. Energy Environ. Sci. 2013, 6, 225. (19) Atkin, R.; Borisenko, N.; Druschler, M.; El Abedin, S. Z.; Endres, F.; Hayes, R.; Huber, B.; Roling, B. Phys. Chem. Chem. Phys. 2011, 13, 6849. (20) Hayes, R.; Borisenko, N.; Corr, B.; Webber, G. B.; Endres, F.; Atkin, R. Chem. Commun. 2012, 48, 10246. (21) Shannon, I. J.; Maschmeyer, T.; Sankar, G.; Thomas, J. M.; Oldroyd, R. D.; Sheehy, M.; Madill, D.; Waller, A. M.; Townsend, R. P. A. Catal. Lett. 1997, 44, 23. (22) Liang, Y. C.; Juan, Y. W.; Lu, K. T.; Jeng, U. S.; Chen, S. A.; Chuang, W. T.; Su, C. J.; Liu, C. L.; Pao, C. W.; Lee, J. F.; Sheu, H. S.; Chen, J. M. J. Phys. Chem. C 2012, 116, 26649. (23) Brown, M. A.; Jordan, I.; Redondo, A. B.; Kleibert, A.; Woerner, H. J.; van Bokhoven, J. A. Surf. Sci. 2013, 610, 1. (24) Mukerjee, S.; Srinivasan, S.; Soriaga, M. P.; McBreen, J. J. Electrochem. Soc. 1995, 142, 1409. (25) Park, J.; Zheng, H. M.; Lee, W. C.; Geissler, P. L.; Rabani, E.; Alivisatos, A. P. ACS Nano 2012, 6, 2078. (26) Radisic, A.; Vereecken, P. M.; Hannon, J. B.; Searson, P. C.; Ross, F. M. Nano Lett. 2006, 6, 238. (27) Williamson, M. J.; Tromp, R. M.; Vereecken, P. M.; Hull, R.; Ross, F. M. Nat. Mater. 2003, 2, 532. (28) Zheng, H. M.; Smith, R. K.; Jun, Y. W.; Kisielowski, C.; Dahmen, U.; Alivisatos, A. P. Science 2009, 324, 1309. (29) Zheng, H. M.; Claridge, S. A.; Minor, A. M.; Alivisatos, A. P.; Dahmen, U. Nano Lett. 2009, 9, 2460. (30) Evans, J. E.; Jungjohann, K. L.; Browning, N. D.; Arslan, I. Nano Lett. 2011, 11, 2809. (31) Jungjohann, K. L.; Bliznakov, S.; Sutter, P. W.; Stach, E. A.; Sutter, E. A. Nano Lett. 2013, 13, 2964. (32) Parent, L. R.; Robinson, D. B.; Woehl, T. J.; Ristenpart, W. D.; Evans, J. E.; Browning, N. D.; Arslan, I. ACS Nano 2012, 6, 3589. (33) Woehl, T. J.; Evans, J. E.; Arslan, L.; Ristenpart, W. D.; Browning, N. D. ACS Nano 2012, 6, 8599. (34) Klein, M. P.; Jacobs, B. W.; Ong, M. D.; Fares, S. J.; Robinson, D. B.; Stavila, V.; Wagner, G. J.; Arslan, I. J. Am. Chem. Soc. 2011, 133, 9144. (35) Discher, D. E.; Ahmed, F. Annu. Rev. Biomed. Eng. 2006, 8, 323. (36) Takai, A.; Yamauchi, Y.; Kuroda, K. J. Am. Chem. Soc. 2010, 132, 208. (37) LaVerne, J. A. Radiat. Res. 2000, 153, 196. (38) Delcourt, M. O.; Belloni, J.; Saito, E. J. Phys. Chem. 1976, 80, 1101. (39) Watanabe, R.; Saito, K. Radiat. Phys. Chem. 2001, 62, 217. 1433

dx.doi.org/10.1021/cm4035209 | Chem. Mater. 2014, 26, 1426−1433