Encapsulation of Single Nanoparticle in Fast-Evaporating Micro

Subscriber access provided by AUSTRALIAN NATIONAL UNIV

Article

Encapsulation of single nanoparticle in fast-evaporating microdroplets prevents particle agglomeration in nanocomposites Ming Pan, Xinjian Shi, Fengjiao Lyu, Ben Louis Levy-Wendt, Xiaolin Zheng, and Sindy K. Y. Tang ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b07773 • Publication Date (Web): 13 Jul 2017 Downloaded from http://pubs.acs.org on July 15, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Encapsulation of single nanoparticle in fastevaporating micro-droplets prevents particle agglomeration in nanocomposites Ming Pan1‡, Xinjian Shi2‡, Fengjiao Lyu2, Ben Louis Levy-Wendt2, Xiaolin Zheng2* and Sindy K. Y. Tang2* 1

Department of Materials Science and Engineering, Stanford University, Stanford, CA, 94305,

USA 2

Department of Mechanical Engineering, Stanford University, Stanford, CA, 94305, USA

KEYWORDS: nanocomposite, micro-droplets, single encapsulation, dispersability, photocatalysis

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 37

ABSTRACT

This work describes the use of fast-evaporating micro-droplets to finely disperse nanoparticles (NPs) in a polymer matrix for the fabrication of nanocomposites. Agglomeration of particles is a key obstacle for broad applications of nanocomposites. The classical approach to ensure the dispersability of NPs is to modify the surface chemistry of NPs with ligands. The surface properties of NPs are inevitably altered, however. To overcome the trade-off between dispersability and surface-functionality of NPs, we develop a new approach by dispersing NPs in a volatile solvent followed by mixing with uncured polymer precursors to form micro-droplet emulsions. Most of these micro-droplets contain no more than one NP per drop, and they evaporate rapidly to prevent the agglomeration of NPs during the polymer curing process. As a proof of concept, we demonstrate the design and fabrication of TiO2 NP@PDMS nanocomposites for solar water-splitting reactions with high photocatalytic efficiency and recyclability arising from the fine dispersion of TiO2. Our simple method eliminates the need for surface functionalization of NPs. Our approach is applicable to prepare nanocomposites comprising a wide range of polymers embedded with NPs of different composition, sizes, and shapes. It has the potential for creating nanocomposites with novel functions.


2

Page 3 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. Introduction Nanocomposites, which comprise inorganic nanomaterials (e.g., nanoparticles) embedded in an organic polymer matrix, enable the design and customization of desired properties, such as the integration of conductivity and flexibility by embedding silver nanoparticles in an elastomeric polymer.1-2 Nanoparticles (NPs) provide the function that arises from their unique properties at the nanometer scale, such as catalytic reactivity and surface plasmon resonance (SPR).1, 3 The organic component serves as a medium to disperse the NPs, and acts primarily as the structural support. Nanocomposites, as a result, have shown promising applications in a wide range of areas, including sensors, optoelectronics, catalysis, self-healing materials, biomedical devices and materials.1-9 The biggest challenge in the design and fabrication of these nanocomposites is the dispersal of NPs in polymer without agglomeration. In contrast to polymer blends consisting of mixtures of two or more polymers, the high surface energy of NPs in organic polymer medium tends to lead to agglomeration, which is a thermodynamically favorable process driven by the minimization of NPs surface energy.1 Agglomeration is more prominent in nanocomposites than in composites comprising microparticles due to the high surface-to-volume ratio of NPs. Agglomeration compromises the performance of nanocomposites, as it reduces the surface area of NPs and leads to the loss of their functions. As such, there is a critical need for establishing simple and facile fabrication techniques to enable the fine-dispersion of various NPs in different types of polymer matrices. So far, several approaches have been developed to improve the dispersability of NPs. The first approach is direct physical mixing, where as-synthesized solid NPs are incorporated into the


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 37

polymer host with the aid of a high energy input, such as heat and ultrasound.6, 10-12 The high energy input breaks bulk solid into small particulates. The further disassembly of small particulates into individual NPs in the polymer host was not achieved, however.1-2 To tackle the surface incompatibility between NPs and polymer, a second approach is to functionalize NPs with appropriate ligands, which are either small organic molecules or polymeric capping agents, to render the NPs surface compatible with the polymer matrix.1, 3, 8, 10, 13 Nonetheless, surface functionalization of NPs has multiple limitations: (1) The functionalization can be tedious, especially when the transfer of NPs between different phases is required. (2) The choice of ligands relies on the specific surface chemistry of the NPs. As such, there is no universal ligand that works for all NPs. (3) Importantly, the improved dispersability of NPs in the polymer matrix is achieved at the cost of their surfaces being occupied by ligands, which compromise their surface functionality. This limitation is particularly problematic in applications that require target molecules to be adsorbed onto the NP surface, such as catalysis or sensing. A third approach is to synthesize NPs inside the polymer matrix by introducing NP precursors and other reagents in the polymer matrix. This approach is widely adopted to incorporate metallic NPs in a polymer. Polymer cross-linker is often used as the reducing agent. Upon nucleation and growth inside the polymer matrix, metallic NPs are formed and dispersed.10, 14-16 While this approach eliminates the need for additional ligands in some cases, the precise control over the size and morphology of the resulting particles can be difficult. For example, the presence of cetyltrimethylammonium bromide (CTAB) micelles in aqueous solution is critical in guiding the growth of gold nanorod (AuNRs) crystals and in controlling the AuNRs’ aspect ratio.17 Due to the limited water solubility which constrains CTAB micelle formation in polymer matrices that are immiscible


4

Page 5 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with water, this approach cannot be adopted easily for the synthesis of nanocomposites containing AuNRs. The ideal solution to tackle the above challenges is to develop a strategy that allows: (1) the incorporation and fine dispersal of pre-synthesized NPs, without surface modifications, into the polymer matrix so to preserve NP functionality. (2) The incorporation of NPs with a wide range of sizes, shapes, and surface chemistries in different types of polymers. In this work, we show that the above criteria can be met by dispersing NPs in a volatile solvent, which forms an emulsion upon mixing with the polymer precursor. The rapid evaporation of the volatile solvent (referred to as “co-solvent” thereafter) leaves the NPs in the polymer matrix in a finely-dispersed state (Figure 1a). Complete evaporation of the co-solvent is possible as the polymer matrix is permeable to the co-solvent molecules. The use of emulsion eliminates the need for surface modifications of NPs. Our strategy is applicable to a wide range of particle geometries and surface chemistries. As a proof of concept, we demonstrate the use of the strategy in making nanocomposites with enhanced photocatalytic performance and recyclability for solar watersplitting reactions, where the dispersability of photocatalyst is critical to achieve high catalytic performance. While the use of co-solvents is common in the fabrication of composite materials, to our knowledge, no work has reported the use of co-solvents to form micro-droplets to disperse NPs during nanocomposite formation. The strategy presented here is simple, preserves NPs functions, and can broaden the choice of NP materials that can be incorporated into polymers. It can further facilitate the characterization of structure-property relation critical for the rational design of nanocomposites.


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 37

2. Results and discussions 2.1. Encapsulation of single NP in micro-droplets and dispersability of NPs in polymer We fabricated the nanocomposites by mixing the uncured polymer precursor, curing agent and a volatile co-solvent containing NPs of interest in their pristine form without surface functionalization (Figure 1a). We chose poly(dimethylsiloxane) (PDMS) as the model polymer as its fabrication was well-established.18 The model co-solvent was ethanol (EtOH). First, we characterize the size distribution of co-solvent drops formed. Green fluorescent molecules (fluorescein-linked polyethylene glycol, fluorescein-PEG) were added to EtOH to facilitate measurements of the size of EtOH drops. Upon mixing with PDMS precursor, both gas bubbles and EtOH droplets were formed (Figure 1a, step 1). The size of EtOH droplets was controlled, in part, by the energy applied during mixing. An increased stirring rate led to a decrease in the diameter of co-solvent drops (Figure S2). Gentle mixing (~ 50 rpm) led to the formation of large drops with size on the order of 10s – 100s µm (Figure S1) and vigorous mixing (~ 500 rpm) led to the formation of micro-droplets with a mean diameter of 680 nm (Figure 1b, see experimental section for details on the determination of droplet size distribution). It is likely that drops smaller than 500 nm were present but could not be detected due to the limited optical resolution of our imaging setup. Nevertheless, the distribution here represents the upper limit of droplet sizes formed using our method. We observe that the drops were distributed uniformly in three dimensions and remained relatively stagnant in the polymer matrix. Despite the lower density (~0.789 g/cm3) of EtOH compared with uncured PDMS (~1.03 g/cm3), the drops did not move due to the high viscosity of the uncured PDMS (~3500 cP). The drops also did not coalesce despite the close proximity between adjacent drops. The green


6

Page 7 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fluorescence diminished completely within 15 min after mixing, indicating the complete evaporation of EtOH drops, since uncured polymers are permeable to co-solvent molecules for complete evaporation (Figure S3). Second, we demonstrate the formation of a nanocomposite comprising a fine dispersion of NPs by repeating the process with a dilute NPs dispersion in co-solvent. 210 nm polystyrene (PS) NPs conjugated with red fluorescent molecules were dispersed in EtOH containing fluorescein-PEG. Red fluorescent NPs were chosen to distinguish themselves from green fluorescent EtOH drops. Figure 1c and 1d show that most of the drops did not contain any NP as they emitted green fluorescence only. For the drops that contained NPs, most of them contained a single NP (Figure 1d, indicated by white arrows). There were a few drops that contained multiple NPs, but most of them were large drops with diameters larger than 2 µm (indicated by yellow arrows). We then cured the mixture using the same method as that for curing undoped PDMS. Figure S4b shows that the cured nanocomposite (red fluorescent PS@PDMS) was similar to a “frozen dispersion” where NPs did not move inside the polymer matrix. The red fluorescent NPs maintained its original size and separation distance before curing, indicating good dispersability of NPs in cured PDMS (Figure S4). To confirm that our method can be applied for dispersing multiple types of NPs simultaneously, we used two types of commercial fluorescent PS NPs premixed at a molar ratio of 1:1 and dispersed in undoped EtOH. The NPs were 200 nm and 500 nm in size, and had red fluorescence and green fluorescence, respectively. Two types of fluorescent NPs were chosen to facilitate fluorescence imaging to determine if there was severe agglomeration after their incorporation into PDMS. The nanocomposite (red-green fluorescent PS@PDMS) was fabricated using the same process as that for red fluorescent PS@PDMS. Similar to red


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 37

fluorescent PS@PDMS, the cured nanocomposite was similar to a “frozen dispersion” where NPs did not move inside the polymer matrix (Figure 2b). Most of the NPs emitted either green or red fluorescence, indicating good dispersability of NPs in cured PDMS. There were a few spots that emitted both red and green fluorescence and appeared yellow, which may have resulted from red and green fluorescent NPs in close proximity to each other, or from a drop containing a few NPs. The absence of large particle clumps that emitted both red and green fluorescence indicates the absence of significant agglomeration. The concentration of NPs was similar in different focal planes within the polymer, indicating the uniform distribution of the NPs in three-dimensional space (Figure S5). In contrast, when the two types of NPs were dispersed into PDMS precursor directly without prior dispersion in EtOH, significant agglomeration was observed (Figure 2c). 2.2. Requirements for the micro-droplet-based strategy Previous work has shown that the drying of droplets containing multiple PS microparticles (MPs) led to their agglomeration.19 The fine dispersion of NPs in cured PDMS in our approach resulted from the encapsulation of single NP in most of the micro-droplets, and the subsequent immobilization of the NPs in the polymer matrix upon drying of the co-solvent. In order for our method to work, the following requirements should be met: 1. NPs should be dispersable in the co-solvent without agglomeration. The upper limit of particle loading is set by the colloidal dispersibility of NP in the solvent, typically 1 - 10% by weight for typical solvents for dispersing NPs. 2. The incorporation of NPs and co-solvent should not affect the curing process of the polymer. Co-solvent and uncured polymer should be able to mix at a sufficiently high


8

Page 9 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


volume ratio without causing the swelling of the polymer nor inhibiting its polymerization (Table S1 and S2). 3. Once dispersed, the movement of co-solvent drops in the uncured polymer host should be minimized before complete evaporation of the drops. Collison among NP-containing drops could lead to their coalescence, forming drops with multiple NPs. Also, creaming or sedimentation of co-solvent drops leads to uneven distribution of NPs. In general, droplet motions can be prevented by choosing a polymer precursor having a high viscosity and a co-solvent that is volatile, such that the drops are evaporated completely before two drops come into contact. We can estimate timescale of droplet movement as follows. In our system, one driving force that causes the drops to move and come into contact is buoyancy force arising from the difference in densities between the co-solvent drops and the polymer. The terminal velocity u of a drop with diameter d ~ 1 µm rising or sinking in a medium of viscosity µ ~ 103 cP and density difference from the drop ∆ρ ~ 0.2 g/cm3 is estimated to be u ~ g d2 (∆ρ)/µ ~ 10-9 m/s, where g is gravitational acceleration.20 At this speed, it would take 104 seconds for the drop to cream/sediment through a 10-µm height in the polymer. This time scale is much longer than the time for the drops to evaporate, which was typically < 102 seconds for the combination of co-solvent and polymers used in our studies.21-22 Indeed, we found that the viscosity of uncured PDMS (~3500 cP) was sufficiently high to immobilize the EtOH drops in our system. 4. The co-solvent drops should be small enough so that the most of the drops contains one or zero NP. This criterion can be met by choosing co-solvents with low interfacial tension and/or increasing the energy supplied during mixing to create a fine emulsion. In our case,


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 37

moderate agitation by manual stirring was sufficient to break bulk EtOH into small drops where the majority of the drops had a size less than ~1 µm. While the precise size distribution of EtOH micro-droplets was difficult to survey due to both the limited optical resolution of our imaging system and the fast evaporation rate of these drops, we can use the mean diameter of EtOH drops determined from fluorescence microscopy (680 nm, Figure 1b) to estimate the average number of NPs per drop (λ) for the sample shown in Figure 1d. For the concentration of NPs used, λ was about 0.017. This result was consistent with the observation that most of the drops were free of NPs (Figure 1d). Assuming a uniform droplet size distribution and a Poisson encapsulation process for the NPs, we can also make rough estimations of the maximum NP loading in EtOH (before incorporation into PDMS) so that at least 99% of the drops contain no more than one NP per drop (see detailed estimation in Note S1). We found the maximum NPs concentration to be about 1.5 nM (Table S3). For the combination of co-solvents and polymers tested, the volume fraction of co-solvent in the final nanocomposite mixture was approximately 10% (see Table S1 and S2 for details). The concentration of NPs in the nanocomposite achievable using our approach was, therefore, in the range of 100 pM to 1 nM. We note that these concentrations were independent of the size, density and surface properties of the NPs. 2.3. Application to a range of particles and polymers Figure 3 shows the application of our approach to incorporate various types of NPs possessing different compositions and sizes into different types of polymers. First, we used PDMS as the polymer host to disperse NPs of different sizes and surface chemistries. To maintain high


10

Page 11 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dispersability of NPs in the co-solvent, in addition to EtOH to disperse hydrophilic NPs, we also used hexadecane as the model organic co-solvent to disperse highly oleophilic NPs and HFE7500 as the model fluorinated co-solvent to disperse fluorophilic NPs. In each case, the cosolvent was volatile to ensure its rapid evaporation. We then used the same approach to disperse 1 µm PTFE particles into various polymers with EtOH as the co-solvent (Figure S6). In addition to PDMS, we selected Ecoflex to represent thermally-curable polymer and Norland Optical Adhesive (NOA-68) and SU-8 to represent UV-curable polymers. We also chose different types of SU-8 with viscosities ranging from 102 to 105 cP to verify that the dispersion of NPs was successful in a large range of viscosities. For all these cases, the colors of the nanocomposites were uniform, and were similar to the original colors of the NPs when dispersed in co-solvents. To show quantitatively that our method prevented agglomeration, we compared the absorption spectra of AuNPs in colloidal suspension and in PDMS nanocomposite. The agglomeration of AuNPs was known to cause a red shift of the absorption peak.23 After incorporating 20 nm AuNPs in PDMS, the resulting AuNPs@PDMS nanocomposite exhibited an absorption peak at ~530 nm, similar to that of 20 nm colloidal AuNPs in water without agglomeration (Figure 3b). This result suggests that the AuNPs were unlikely to be agglomerated in our nanocomposite. We expect similar dispersibility of NPs in other nanocomposites tested, though it was challenging to perform direct measurement of their dispersibility. While electron microscopy techniques such as SEM and TEM are often employed to characterize composites, they are primarily used for the confirmation of the presence or absence of particles in the polymer matrix, rather than for the quantification of the spatial distribution and/or agglomeration of NPs in nanocomposites. The latter is challenging due to the small size of the NPs and significant sample


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 37

charging due to the non-conductive nature of the polymers. Nevertheless, in our system, since the NPs were confined within the micro-droplets which were immobilized during their rapid evaporation, the spatial locations of individual NPs in the nanocomposite was determined by the locations of individual co-solvent drops. In addition, the high viscosity of the polymer prevented subsequent motion of NPs during the curing process after co-solvent drops were evaporated. As such, the spatial location of NPs in the cured nanocomposite should not depend on the composition of the NPs for the same type of co-solvent and polymer used. The spatial distribution of NPs in the nanocomposites shown in Figure 3a should, therefore, be similar to that for red fluorescent PS@PDMS shown in Figure 1c and Figure S4b, since they were all PDMS-based nanocomposites by using EtOH as the co-solvent. 2.4. Application of porous nanocomposite for photocatalysis The benefits of nanocomposites with well-dispersed NPs are examined in the application of photocatalytic solar water-splitting. For the proof of principle, we chose TiO2 NPs as TiO2 is one of the commonly used catalyst to split water to H2 and O2 under solar illumination. Most previous studies using TiO2 as photocatalysts either directly dispersed NPs in a colloidal dispersion or supported them on the surface of solid substrates.24-27 Here, we used our emulsion method to disperse TiO2 NPs in a PDMS host, which is optically transparent and is permeable to small molecules. We fabricated three control TiO2 NP/PDMS nanocomposites (Figure 4 A, B and C) to compare their catalytic activity and stability. The co-solvent dispersion was TiO2 NPs in EtOH (average diameter: 102.8 nm measured by Dynamic Light Scattering). Both nanocomposites A and B were prepared with our micro-droplet approach to incorporate TiO2 NPs in PDMS sponge, but with different TiO2 loading concentrations. Nanocomposite A had a low TiO2 loading concentration (TiO2% = 450 ppm in weight; average number of NPs per drop λ


12

Page 13 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


≈ 0.247 assuming drop diameter ≈ 680 nm, with 97.4% of drops contained no more than a single NP) and nanocomposite B had a high TiO2 loading concentration (TiO2% = 1.2 × 104 ppm, λ ≈ 6.8, with 99.12% of drops contained more than one NP). After fabricating the TiO2@PDMS sponge, the PDMS sponge was slightly etched in a TBAF/NMP solution to expose the embedded TiO2 NPs (refer to Figure S7 for more details). Nanocomposite C was fabricated by dipcoating28-30 a pre-fabricated PDMS sponge in a colloidal dispersion of TiO2 NPs followed by air dry (referred to as “TiO2 ads-PDMS sponge” thereafter). We compared the catalytic performances of nanocomposites A, B and C by measuring the amount of H2 produced per gram of TiO2 NPs for four catalytic cycles. Figure 5a compares the H2 generation rate per unit mass of TiO2 initially loaded between nanocomposite A and B. Clearly, the mass normalized H2 generation rate for nanocomposite A (~1.6×10-3 mol /(g TiO2 • hr)) was about 2.7 times higher than that for nanocomposite B in which TiO2 NPs may have formed larger agglomerates due to the high loading concentration. The mass normalized H2 generation rate for nanocomposite A was similar to those from TiO2 NPs colloidal dispersion (1.5-2.5×10-3 mol/(g TiO2 • hr)) in previous work,31-33 indicating good dispersability of TiO2 NPs in PDMS sponges. We further tested if using PDMS host was capable of improving the recyclability of TiO2 NPs, since a known challenge for TiO2 NPs colloidal dispersion is the difficulty in collecting the TiO2 NPs and re-using them in other solutions. We define one cycle of testing as removing the nanocomposites from the used glycerol/distilled water mixed solution, rinsing them with distilled water, and immersing them in a fresh glycerol/distilled water mixed solution. After four cycles, nanocomposite A showed less than 10% reduction in its H2 generation rate, but nanocomposite C showed as much as a 80% reduction (Figure 5b). This result is consistent with our expectation


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 37

that embedding TiO2 NPs with PDMS, rather than adsorbing them on PDMS surface, greatly improved the physical bonding between TiO2 and PDMS, leading to better recyclability. 3. Conclusions We have developed a simple strategy to incorporate NPs into polymer hosts by the rapid evaporation of co-solvent micro-droplets. The encapsulation of single NPs in most of the cosolvent micro-droplets and the rapid evaporation of the co-solvent prevented NPs from agglomeration. We note two potential strategies to further increase NP concentration in the nanocomposite using our method while maintaining their fine dispersibility: 1) Decrease the size of droplets in the co-solvent emulsion. As an example, one can increase the energy input during the mixing process by using a homogenizer. 2) Use a co-solvent that allows the mixing with uncured polymer at high volume fractions (> 10%). Table 1 summarizes the advantages of our work over traditional fabrication techniques. By adjusting the composition of co-solvent, our approach is compatible with a wide range of NPs. It eliminates the need for surface functionalization of NPs and, therefore, preserves the desired surface properties of NPs. We believe our method will enable the fabrication of new nanocomposites with a diverse set of NPs and polymers for a wide range of applications from sensing to catalysis.

4. Experimental Section Materials. All materials were used as purchased without purification. Sylgard 184 silicone elastomer kit was purchased from Dow Corning Corporation. Ecoflex-0030 kit was purchased from Smooth-on Inc. Norland optical adhesive 68 (NOA-68) was purchased from Norland


14

Page 15 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Products Inc. SU-8 photoresist with different viscosities was purchased from MicroChem Corp. Absolute ethanol (99.5%) and hexadecane (99%) were purchased from Sigma-Aldrich. NovecTM 7500 Engineered Fluid (HFE-7500) was purchased from 3M. The NPs used in this work were obtained either commercially or synthesized in-house. For commercial ones, platinum nanoparticles (PtNPs, 3nm, 1000 ppm in water) and polytetrafluoroethylene particles (PTFE powder, average diameter: 1µm) were purchased from Sigma-Aldrich. Fluorescent polystyrene beads (PS, Fluoresbrite® Carboxylate Microspheres, 2.5% solids in water) was purchased from Polysciences Inc. Fe3O4 NPs (ferrofluid EMG 304) was purchased from FerroTec Inc. Titanium oxide (TiO2, anatase, 99.9% trace metals basis) was obtained from Alfa Aesar; Cobalt Oxide (Co3O4, 99.8% trace metals basis,

Encapsulation of Single Nanoparticle in Fast-Evaporating Micro

Recommend Documents