Controlling Morphology in Polycrystalline Films by Nucleation and

Aug 6, 2018 - Nanoscale Spin Manipulation with Pulsed Magnetic Gradient Fields from a Hard Disc Drive Writer. Nano Letters. Bodenstedt, Jakobi, Michl ...
0 downloads 0 Views 9MB Size
Download PDF
Letter Cite This: Nano Lett. XXXX, XXX, XXX−XXX

pubs.acs.org/NanoLett

Controlling Morphology in Polycrystalline Films by Nucleation and Growth from Metastable Nanocrystals Ajay Singh,†,‡,¶ Lukas Lutz,‡,# Gary K. Ong,†,§,# Karen Bustillo,‡ Simone Raoux,∥ Jean L. Jordan-Sweet,⊥ and Delia J. Milliron*,†,‡ †

McKetta Department of Chemical Engineering, The University of Texas at Austin, Austin, Texas 78712, United States The Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States § Department of Materials Science and Engineering, University of California−Berkeley, Berkeley, California 94720, United States ∥ Helmholtz-Zentrum Berlin für Materialien und Energie GmbH, Albert-Einstein-Strasse 15, 12489 Berlin, Germany ⊥ IBM Watson Research Center, 1101 Kitchawan Road, Yorktown Heights, New York 10598, United States

Downloaded via UNIV OF TOLEDO on August 7, 2018 at 01:51:15 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Solution processing of polycrystalline compound semiconductor thin film using nanocrystals as a precursor is considered one of the most promising and economically viable routes for future large-area manufacturing. However, in polycrystalline compound semiconductor films such as Cu2ZnSnS4 (CZTS), grain size, and the respective grain boundaries play a key role in dictating the optoelectronic properties. Various strategies have been employed previously in tailoring the grain size and boundaries (such as ligand exchange) but most require postdeposition thermal annealing at high temperature in the presence of grain growth directing agents (selenium or sulfur vapor with/without Na, K, etc.) to enlarge the grains through sintering. Here, we show a different strategy of controlling grain size by tuning the kinetics of nucleation and the subsequent grain growth in CZTS nanocrystal thin films during a crystalline phase transition. We demonstrate that the activation energy for the phase transition can be varied by utilizing different shapes (spherical and nanorod) of nanocrystals with similar size, composition, and surface chemistry leading to different densities of nucleation sites and, thereby, different grain sizes in the films. Additionally, exchanging the native organic ligands for inorganic surface ligands changes the activation energy for the phase change and substantially changes the grain growth dynamics, while also compositionally modifying the resulting film. This combined approach of using nucleation and growth dynamics and surface chemistry enables us to tune the grain size of polycrystalline CZTS films and customize their electronic properties by compositional engineering. KEYWORDS: Phase transformation, in situ X-ray diffraction, in situ transmission electron microscopy, Cu2ZnSnS4 (CZTS), chalcogenidometallates, Kissinger analysis Meanwhile, polycrystalline films from nanocrystals of complex copper chalcogenides have typically required aggressive annealing conditions, including the use of caustic selenium/ sulfur vapor, to facilitate grain growth. Even with these extreme annealing conditions, the resulting polycrystalline films contain irregular grains not optimal for high-performance solar cells.18−21 CZTS nanorods prepared in a metastable phase (i.e., wurtzite) have been shown to undergo a metastable-tostable phase transition during annealing in a thin film, with the tantalizing result being a polycrystalline CZTS thin film with uniform and sizable (∼0.3 μm) grains.2 Here, we show that this phase transformation proceeds by nucleation and growth of the stable kesterite phase, with grain growth propagating

C

olloidal nanocrystals, as solution-processable precursors, can be used to prepare semiconductor thin films following thermal annealing.1−6 This approach is particularly promising for materials with complex compositions, such as those based on CuInGaS2 (CIGS) or Cu2ZnSnS4 (CZTS), which often exhibit inhomogeneities when deposited by vapor phase methods.2,7−10 In nanocrystals, the composition is dictated by chemical control during nanocrystal synthesis,11−13 facilitating the deposition of compositionally uniform, singlephase polycrystalline films.2,14−16 Grain growth in thin films of the photovoltaic material CdTe is well-known to be facilitated by the addition of CdCl2 as a sintering agent, a process that has been extended to CdTe nanocrystals by employing (CdCl3)− as an inorganic ligand during film deposition.6 Inorganic ligands of like composition to the nanocrystal precursors have been shown to enable promising electronic properties, but not necessarily substantial grain growth, following annealing.17 © XXXX American Chemical Society

Received: May 10, 2018 Revised: July 4, 2018

A

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 1. Polycrystalline films of kesterite CZTS prepared from wurtzite nanocrystals. (a) ADF-STEM image of nanospheres, (b) ADF-STEM image of nanorods, (c) cross-sectional scanning electron microscopy (SEM) image of CZTS thin film obtained after annealing nanospheres at 400 °C for 15 min, (d) SEM of CZTS thin film obtained after annealing nanorods at 400 °C for 15 min, (e and f) XRD pattern of the thin film before and after annealing for nanospheres (e) and nanorods (f).

nanocrystal films but that carbonize during annealing, leaving significant residual carbon content that inhibits grain growth and is detrimental to electronic properties.30,31 Replacing native ligands with shorter chain ligands, by using an ex-situ ligand exchange protocol and subsequent sulfurization or selenization, can result in larger grains, but this strategy inevitably introduces a trade-off between forming high-quality nanocrystal films (demanding longer ligands) and successfully annealing those films to produce larger grains (demanding smaller ligands).32−34 For the CZTS materials system in particular, current strategies to drive crystallization and grain growth involve annealing in the presence of supplemental elements (Se, S, Na, etc.), which inevitably couples the grain growth process with compositional changes due to ion exchange or doping.6,7,18,19,33−42 Alternatively, metastable CZTS nanocrystals can be used to achieve significant grain growth at lower temperatures after short, few-second annealing times due to a thermodynamically favorable wurtzite to kesterite phase transformation.2 While previous reports have examined the influence of precursor or additional processing, such as annealing and selenization/sulfurization under Se/S vapor pressure,20,36−39 on final grain structure, here we sought to understand the underlying grain growth mechanisms for thin films generated from wurtzite CZTS nanocrystals.2,10,43 The apparent nucleation-and-growth mechanism suggests new strategies for tuning the morphology of polycrystalline films derived from nanocrystals. Our findings, while directly relevant for the CZTS system, may help to elucidate the underlying nucleation and growth mechanisms pertinent to the phase transformation of nanocrystal ensembles in general and facilitate future rational design of nanocrystal-derived semiconductor thin films.

from one nanocrystal to the next within a film. As a result, the activation energy for the transformation, which differs for nanorods and nanospheres, and the heating rate can be used to tune grain size and morphology in the resulting polycrystalline films. On the other hand, stabilizing the surface of the CZTS nanorods with inorganic chalcogenidometallate ligands (ChaMs) increases the activation energy for phase transformation, while facilitating classical grain growth by coarsening. Hence, by modifying the condition of surface sites the activation energy for nucleation and growth processes is shown to be a powerful strategy to realize favorable morphologies in polycrystalline films made from nanocrystal precursors. Solution-based thin film deposition routes have been used extensively for various metal chalcogenide and metal oxide chemistries as this low-temperature route offers significant advantages over traditional processes in high throughput rollto-roll processing, cost, and expansion to arbitrary device form factors.20−24 By using soluble precursors or nanocrystal inks, films can be cast on substrates of arbitrary geometry while nominally preserving the desired electronic and physical properties of their traditional thin film counterparts.21,25−27 However, current solution processing strategies often result in amorphous or fine-grained films that may exhibit compositional segregation and have poor electronic performance. Such films require a high-temperature processing step to improve phase purity and achieve crystallization and grain growth.27 Although presynthesizing nanocrystals of the target composition can avoid compositional inhomogeneity, the resulting films typically are fine-grained and coarsen only marginally with high-temperature thermal treatment.27−29 This lack of grain growth is largely due to the presence of long chain hydrocarbon ligands that are needed to deposit uniform B

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 2. Kinetics analysis of the wurtzite-to-kesterite phase transformation. (a) Kissinger’s plot for determination of the activation energy for CZTS nanospheres, (b) in situ XRD pattern shows the wurtzite (W) to kesterite (K) phase transformation for nanospheres when annealed at rate of 1 K/s (dashed white line indicates temperature), (c) Kissinger’s plot for determination of the activation energy for CZTS nanorods, and (d) in situ XRD pattern shows the wurtzite to kesterite phase transformation for nanorods when annealed at rate of 1 K/s.

CZTS nanocrystals were synthesized using previously published methods in the wurtzite phase, the metastable hexagonal modification of the tetragonal kesterite phase.2,36 Nanorods and nanospheres with similar diameters were synthesized with similar compositions, specifically with a Cupoor and Zn-rich composition that is commonly used for constructing high efficiency CZTS solar cells.22 Annular darkfield scanning transmission electron microscope (ADF-STEM) images (Figure 1a,b) revealed highly uniform nanocrystals with an average diameter of 8 ± 0.5 nm (Figure S1). Nanocrystal films were prepared on silicon substrates by drop casting from solvent dispersions, then these films were annealed at variable temperatures and ramp rates under argon flow to induce the wurtzite to kesterite phase transformation and associated grain growth (see further details in the Supporting Information). By holding all variables with the exception of nanocrystal shape constant, we sought to study the influence of nanocrystal shape on the morphology of the resulting thin films. During annealing, wurtzite nanocrystals of both shapes transformed to kesterite thin films with grain sizes that were significantly larger than the original nanocrystals. This observation is in stark contrast to annealed films of nanocrystals synthesized in the kesterite phase, which show no appreciable grain growth under identical annealing conditions (Figure S2). Wurtzite nanorods in particular exhibited large grains (360 ± 65 nm) after annealing that nearly span the thickness of the film (Figure 1d). Meanwhile, wurtzite nanospheres transformed to films with an average grain size of only 100 ± 20 nm (Figure 1c) although the film thickness was comparable (between 500 amd 600 nm). X-ray diffraction analysis of the annealed films from both shapes of nanocrystals demonstrate complete transformation from wurtzite to kesterite (Figure 1e,f) with the disappearance of all wurtzite reflections.

To uncover the mechanism underlying this difference in grain size, we used in situ synchrotron X-ray diffraction to monitor the onset of phase transformation. The phase transition temperature during a linear temperature ramp was determined by the maximum rate of change in the peak intensity of the (112) kesterite reflection (Figure S3). A Kissinger analysis of the dependence of the phase transformation temperature on the heating rate allows determination of the activation energy for nucleation of the kesterite phase.44−49 A striking difference in the barrier to transformation was found between nanospheres (1.85 ± 0.1 eV, Figure 2a) and nanorods (2.14 ± 0.12 eV, Figure 2b). This difference in activation energy can explain the difference in grain size between the nanorod and nanosphere films if one asserts that the kesterite grains are formed by nucleation and growth of the more stable phase. The higher barrier to nucleation found for nanorods reduces the nucleation rate, thereby suppressing the phase transformation until the temperature is high enough that growth becomes kinetically competitive. At a ramp rate of 1 K/s, for example, the phase transformation occurs at 402 ± 5 °C for nanorods (Figure 2c) and only 362 ± 3 °C for nanospheres (Figure 2d). In addition to measuring differences in activation energy, we also examined the exothermicity of the phase transformation, expecting it to vary with nanocrystal shape due to differences in specific surface area and surface energies. The heat released by the wurtzite-to-kesterite transformation was determined by differential scanning calorimetry (DSC, Figure S4). Both nanosphere and nanorod DSC results show distinctive double peak signatures suggestive of a nucleation and growth mechanism, with the lower temperature peak (assigned to nucleation) more dominant for nanospheres and the higher temperature peak (growth) dominant in the nanorod DSC C

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 3. Nucleation and growth controls grain size. (a−d) SEM image of a cross-section of CZTS thin film obtained (from nanorod film) when annealed with varying heating rate (from 0.2 K/s to 9 K/s). (e and f) TEM images shows the CZTS nanorod before and after initial nucleation of kesterite phase (at tips).

trace.50,51 Interestingly, nanospheres release more than twice as much energy (12 kJ/mol) overall as nanorods (5.25 kJ/mol). These differences can be ascribed to the larger specific surface area of the nanospheres and the larger representation of highenergy facets for low-aspect ratio nanocrystals. In wurtzite nanocrystals, the polar (001) and (00-1) facets that terminate the ends of nanorods have a higher specific surface energy than the nonpolar facets on their sides. The heat released is modest compared to chemical conversion processes, like those in sol− gel reactions, so local heating is unlikely to have a significant influence on the phase transformation. Rather, the kinetic analysis and DSC suggest that nucleation takes place more readily in nanospheres, while a larger fraction of the wurtziteto-kesterite transition occurs during the growth phase for nanorods. If the grain size is determined by nucleation and growth of kesterite from the wurtzite nanocrystals, classical nucleation and growth theory predicts that it should be possible to tune the grain size by changing the ramp rate. For instance, finegrained glass-ceramics are formed by nucleating many crystalline grains in glass at low temperature (ramped or held constant) before heating to high temperature to convert the remaining metastable glass to a polycrystalline material by growth of the nucleated grains.52,53 We tested whether this classical theory could be applied to our metastable nanocrystal films by varying the ramp rate used to anneal identical films of wurtzite nanorods. Varying the heating rate over nearly 2

orders of magnitude, from 0.2 to 9 K/s (Figure 3a−d), we observed a remarkable variation in grain size, consistent with the nucleation and growth mechanism, specifically as the heating rate increased, the average grain size increased. To further confirm that phase transformation in this system is driven by initial nucleation of the kesterite phase and subsequent growth of existing nuclei into the remaining metastable wurtzite matrix, we examined the initial nucleation events by ex-situ annealing and subsequent examination in TEM. For wurtzite nanorods, we observed initial nucleation only at the nanorod tips followed by propagation to neighboring nanocrystals (Figure 3e,f). The small number of high-energy tip sites constrains nucleation as a mode of transformation resulting only in a small exothermic change apparent in the DSC results. In addition, when nanorods lie close to each other in an ordered fashion (side by side or tip to tip), the transformation at the tips propagates into the neighboring nanorods specifically by initial coalescence of nanorod tips (blue arrow in the Figure 3f). This suggests that the close proximity of nanocrystals in thin films will be beneficial for grain growth, once the ligands have been thermally removed. Further, it has been previously shown that the sulfurization of compact NC thin film results in bigger grain size as compared to noncompact NC film, which highlights the importance of internanoparticle distance in the grain growth process.54 D

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 4. (a and b) SEM showing conversion from sheets of nanorods to sheet-like kesterite crystals and (c) in situ XRD pattern shows the complete phase conversion of (002) wurtzite peak of nanorod to (112) kesterite peak. The zoom in area (gray box) shows the shift in (112) reflection.

clusters (ChaMs), specifically [Ge2Se6]4−, through solution phase ligand exchange (Figure 5a), the surface atoms can be constrained. In particular, [Ge2Se6]4− is known to condense to form amorphous GeSe2 between 200 and 275 °C,57 so the nanorod surfaces were encased in and strongly bound to this glassy inorganic shell before reaching the wurtzite-to-kesterite transition temperature. In these inorganic-capped CZTS nanorod films, phase transformation temperatures were roughly 100 °C higher than their organic-capped counterparts (Figure 5c). Based on a Kissinger analysis, the activation energy for phase transformation was 2.48 ± 0.9 eV (Figure 5d), well above the 2.14 ± 0.12 eV found for organic-capped nanorods. Despite the suppression of nucleation until temperatures that were higher than those used with the ligand capped rods, we did not observe very large crystal grains immediately following the wurtzite-to-kesterite phase transformation or after the same annealing time (15 min) applied to pristine nanorod and nanosphere films (Figure 5e and Figure S6). The presence of GeSe2 coating between the nanorods appears to hamper grain growth during phase transformation by the propagation mechanism outlined above. That said, the absence of any carbon residue in these all-inorganic films allowed classical grain growth (i.e., coarsening) to proceed far more efficiently, likely involving the interdiffusion of Ge and Se with the components of the CZTS akin to a form of in situ selenization that leads to Cu2Zn(Sn1−xGex)(S1−ySey)4. The resulting films contained dense, micron-scale grains that spanned the film thickness (Figure 5f,g), and Raman analysis confirms the presence of Ge and Se as well as the absence of binary (ZnS, ZnSe, or GeSex) or ternary (Cu3SnS4, Cu2SnSe3) impurities commonly found in vacuum processed CZTS films

To further investigate the phase transformation propagation between closely packed nanorods, we sought to exploit this phenomenon to direct the crystalline morphology of the polycrystalline thin films produced after annealing by preassembling wurtzite nanorods into ordered superstructures. In this regard, we synthesized sheets of perpendicularly oriented nanorods55 that are laterally close packed (see detail in the Supporting Information), which were then deposited on substrates to produce nanocrystal films in which the wurtzite c axis is predominantly oriented perpendicular to the substrate (Figure 4a, XRD in Figure S5a). When annealed, the nanorod sheets convert to extended planar crystals of kesterite, as observed by the slight change in peak position from the wurtzite (002) to kesterite (112) characteristic reflection (Figure 4b and Figure S5b). The dominance of this peak indicates that the kesterite crystals are oriented with the (112) axis perpendicular to the substrate (Figure 4c). This preservation of directional crystalline texture by exploiting assembly of the nanocrystal precursor further motivates future studies to specifically engineer the directional propagation of phase transformations in solids made from nanocrystal building blocks. Considering the greatly reduced activation energy for phase transformation of nanospheres compared to nanorods and the observation by TEM of nucleation at nanorod tips, we hypothesized that nucleation may be surface-mediated and therefore sensitive to any constraints on the mobility of surface atoms. Based on thermogravimetric analysis, the organic ligands used in the synthesis were desorbed before the nanorods started the phase transformation at around 400 °C.56 By replacing the organic ligands with chalcogenidometallate E

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 5. Phase transformation of inorganic nanocomposites−grain growth follows transformation: (a) photographs and schematics of ChaMs illustrating the ligand exchange process of CZTS nanocrystals, (b) DSC of the composite, (c) Kissinger plots for the nanocrystal composites to calculate the activation energy, (d) in situ XRD pattern shows the complete phase transformation of the nanocrystal composite film from wurtzite to kesterite, and (e−g) cross-section SEM images shows typical grain growth for composite films with high [∼95%, (e and f)] and low NC [∼70%, (g)] loading, respectively, after annealing at 550 °C for 15 min (e) and 60 min (f and g).

(Figure S7).58,59 Considering that both Ge and Se doping are known to favorably enhance the optoelectronic properties of CZTS in thin film solar cells by increasing the band gap (Ge doping) and removing the trap states, which are generally found in sulfide materials (Se filling), these films may be suitable for application in high-efficiency photovoltaics.60−62 Transformation of nanocrystals from a metastable to stable phase has been shown to follow a nucleation and growth pathway when the nanocrystals are deposited in dense films or assemblies. In the case of the spherical NCs, there are more nucleation sites, which effectively lowers the surface energy and thus reduces the free energy barrier to facilitate the generation of a large number of nuclei compared to a nanorod film (where the lower number of nuclei leads ultimately to bigger grain size). The grain growth in nanorod films is further propagated more readily owing to the possibility to form compact films due to the close proximity of the lateral facets. In the case of inorganic ligand capped nanorod films, a classical grain growth by ripening is more dominant as the surface atoms are more constrained in these composite structures. This process may occur in other materials systems undergoing analogous transformations. For example, films of small amorphous

nanoparticles of the phase change memory material, GeTe, were found to yield a coarser crystalline morphology after annealing to induce crystallization.63 Furthermore, controlled nucleation and growth may be combined with pressureinduced densification of nanocrystal assemblies to maximize the grain size in resulting polycrystalline films.64 A remarkable aspect of this analysis is that the growth of crystals of the stable phase is apparently propagated through the film across boundaries between individual nanocrystals to form the resulting large grains. The microscopic mechanism underlying such phase propagation and whether such behavior might be observable in assemblies of nanocrystals undergoing reversible (e.g., crystal−crystal) phase transformations are intriguing open questions.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.8b01916. Full synthetic and experimental details, DSC data, XRD patterns, and Raman spectra (PDF) F

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

■

(14) Stolle, C. J.; Harvey, T. B.; Pernik, D. R.; Hibbert, J. I.; Du, J.; Rhee, D. J.; Akhavan, V. A.; Schaller, R. D.; Korgel, B. A. J. Phys. Chem. Lett. 2014, 5, 304−309. (15) Foos, E. E.; Yoon, W.; Lumb, M. P.; Tischler, J. G.; Townsend, T. K. ACS Appl. Mater. Interfaces 2013, 5, 8828−8832. (16) Jasieniak, J. J., MacDonald, B. I.; Mulvaney, P. Nanocrystals, Layer-by-Layer Assembly, and Photovoltaic Devices. In Nanomaterials, Polymers, and Devices: Materials Functionalization and Device Fabrication; Kong, E. S.W., Ed.; John Wiley & Sons, Inc.: Hoboken, NJ. 2015357−394, DOI: 10.1002/9781118867204.ch14. (17) Dolzhnikov, D. S.; Zhang, H.; Jang, J.; Son, J. S.; Panthani, M. G.; Shibata, T.; Chattopadhyay, S.; Talapin, D. V. Science 2015, 347, 425−428. (18) Kida, T.; Horita, K.; Suehiro, S.; Yuasa, M.; Quitain, A. T.; Tanaka, T.; Fujita, K.; Ishiwata, Y.; Shimanoe, K. ChemistrySelect 2016, 1, 86−93. (19) Harvey, T. B.; Mori, I.; Stolle, C. J.; Bogart, T. D.; Ostrowski, D. P.; Glaz, M. S.; Du, J.; Pernik, D. R.; Akhavan, V. A.; Kesrouani, H.; Vanden Bout, D. A.; Korgel, B. A. ACS Appl. Mater. Interfaces 2013, 5, 9134−9140. (20) Chernomordik, B. D.; Ketkar, P. M.; Hunter, A. K.; Beland, A. E.; Deng, D. D.; Aydil, E. S. Chem. Mater. 2016, 28, 1266−1276. (21) Rehan, S.; Kim, K. Y.; Han, J.; Eo, Y. J.; Gwak, J.; Ahn, S. K.; Yun, J. H.; Yoon, K.; Cho, A.; Ahn, S. ACS Appl. Mater. Interfaces 2016, 8, 5261−5272. (22) van Hest, M. F. A. M.; Ginley, D. S. Future Directions for Solution-Based Processing of Inorganic Materials. In Solution Processing of Inorganic Materials; Mitzi, D. B., Ed.; John Wiley & Sons, Inc.: Hoboken, NJ, 2008; DOI: 10.1002/9780470407790.ch14. (23) Lin, Z.; He, Q.; Yin, A.; Xu, Y.; Wang, C.; Ding, M.; Cheng, H. C.; Papandrea, B.; Huang, Y.; Duan, X. ACS Nano 2015, 9, 4398− 4405. (24) Yu, X.; Marks, T. J.; Facchetti, A. Nat. Mater. 2016, 15, 383− 396. (25) Kim, M. G.; Kanatzidis, M. G.; Facchetti, A.; Marks, T. J. Nat. Mater. 2011, 10, 382−388. (26) Zhang, R.; Szczepaniak, S. M.; Carter, N. J.; Handwerker, C. A.; Agrawal, R. Chem. Mater. 2015, 27, 2114−2120. (27) Fu, J.; Fu, J.; Tian, Q.; Wang, H.; Zhao, F.; Kong, J.; Zhao, X.; Wu, S. ACS Appl. Energy Mater. 2018, 1, 594−601. (28) Akhavan, V. A.; Goodfellow, B. W.; Panthani, M. G.; Reid, D. K.; Hellebusch, D. J.; Adachi, T.; Korgel, B. A. Energy Environ. Sci. 2010, 3, 1600−1606. (29) Korala, L.; Braun, M. B.; Kephart, J. M.; Tregillus, Z.; Prieto, A. L. Chem. Mater. 2017, 29, 6621−6629. (30) Akhavan, V. A.; Goodfellow, B. W.; Panthani, M. G.; Steinhagen, C.; Harvey, T. B.; Stolle, C. J.; Korgel, B. A. J. Solid State Chem. 2012, 189, 2−12. (31) Gu, E.; Yan, C.; Liu, F.; Liu, Y.; Su, Z.; Zhang, K.; Chen, Z.; Li, J.; Liu, Y. J. Mater. Sci.: Mater. Electron. 2015, 26, 1932−1939. (32) Kaelin, M.; Rudmann, D.; Kurdesau, F.; Zogg, H.; Meyer, T.; Tiwari, A. N. Thin Solid Films 2005, 480-481, 486−490. (33) Huang, T. J.; Yin, X.; Tang, C.; Qi, G.; Gong, H. ChemSusChem 2016, 9, 1032−1041. (34) Tiong, V. T.; Zhang, Y.; Bell, J.; Wang, H. X. RSC Adv. 2015, 5, 20178−20185. (35) Dun, C.; Huang, W.; Huang, H.; Xu, J.; Zhou, N.; Zheng, Y.; Tsai, H.; Nie, W.; Onken, D. R.; Li, Y.; Carroll, D. L. J. Phys. Chem. C 2014, 118, 30302−30308. (36) Williams, B. A.; Smeaton, M. A.; Trejo, N. D.; Francis, L. F.; Aydil, E. S. Chem. Mater. 2017, 29, 1676−1683. (37) Hages, C. J.; Koeper, M. J.; Miskin, C. K.; Brew, K. W.; Agrawal, R. Chem. Mater. 2016, 28, 7703−7714. (38) Williams, B. A.; Trejo, N. D.; Wu, A.; Holgate, C. S.; Francis, L. F.; Aydil, E. S. ACS Appl. Mater. Interfaces 2017, 9, 18865−18871. (39) Clark, J. A.; Uhl, A. R.; Martin, T. R.; Hillhouse, H. W. Chem. Mater. 2017, 29, 9328−9339.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Ajay Singh: 0000-0002-5168-7522 Lukas Lutz: 0000-0002-3466-4775 Delia J. Milliron: 0000-0002-8737-451X Present Address ¶

A.S.: Materials Physics & Applications Division: Center for Integrated Nanotechnologies, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, United States.

Author Contributions #

L.L. and G.K.O. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was performed at the University of Texas at Austin and, in part, at the Molecular Foundry, Lawrence Berkeley National Laboratory, a user facility supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy (DOE) under Contract No. DEAC02-05CH11231. A.S. was supported by the Bay Area Photovoltaics Consortium, sponsored by DOE EERE. G.K.O. was supported in part by a National Science Foundation Graduate Research Fellowship under Grant Number DGE1106400. Use of the SSRL is supported by DOE under Contract No. DE-AC02-76SF00515. In situ synchrotron X-ray diffraction was performed at the National Synchotron Light Source, a DOE Office of Science User facility operated for the DOE Office of Science by Brookhaven National Laboratory under Contract No. DE-AC02-98CH10886. Additional support from the Welch Foundation (Grant F-1848) and the National Science Foundation (Grant CHE-1609656) is gratefully acknowledged.

■

REFERENCES

(1) Jasieniak, J.; MacDonald, B. I.; Watkins, S. E.; Mulvaney, P. Nano Lett. 2011, 11, 2856−2864. (2) Mainz, R.; Singh, A.; Levcenko, S.; Klaus, M.; Genzel, C.; Ryan, K. M.; Unold, T. Nat. Commun. 2014, 5, 3133. (3) Panthani, M. G.; Kurley, J. M.; Crisp, R. W.; Dietz, T. C.; Ezzyat, T.; Luther, J. M.; Talapin, D. V. Nano Lett. 2014, 14, 670−675. (4) Panthani, M. G.; Korgel, B. A. Annu. Rev. Chem. Biomol. Eng. 2012, 3, 287−311. (5) Kim, B. H.; Staller, C. M.; Cho, S. H.; Heo, S.; Garrison, C. E.; Kim, J.; Milliron, D. J. ACS Nano 2018, 12, 3200−3208. (6) Zhang, H.; Kurley, J. M.; Russell, J. C.; Jang, J.; Talapin, D. V. J. Am. Chem. Soc. 2016, 138, 7464−7467. (7) Guo, Q.; Ford, G. M.; Hillhouse, H. W.; Agrawal, R. Nano Lett. 2009, 9, 3060−3065. (8) Miskin, C. K.; Yang, W. C.; Hages, C. J.; Carter, N. J.; Joglekar, C. S.; Stach, E. A.; Agrawal, R. Prog. Photovoltaics 2015, 23, 654−659. (9) Redinger, A.; Berg, D. M.; Dale, P. J.; Siebentritt, S. J. Am. Chem. Soc. 2011, 133, 3320−3323. (10) Schubert, B.-A.; Marsen, B.; Cinque, S.; Unold, T.; Klenk, R.; Schorr, S.; Schock, H.-W. Prog. Photovoltaics 2011, 19, 93−96. (11) Singh, A.; Singh, S.; Levcenko, S.; Unold, T.; Laffir, F.; Ryan, K. M. Angew. Chem., Int. Ed. 2013, 52, 9120−9124. (12) Ibáñez, M.; Cadavid, D.; Zamani, R.; García-Castelló, N.; Izquierdo-Roca, V.; Li, W.; Fairbrother, A.; Prades, J. D.; Shavel, A.; Arbiol, J. Chem. Mater. 2012, 24, 562−570. (13) Singh, A.; Coughlan, C.; Milliron, D. J.; Ryan, K. M. Chem. Mater. 2015, 27, 1517−1523. G

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters (40) Carrete, A.; Shavel, A.; Fontané, X.; Montserrat, J.; Fan, J.; Ibáñez, M.; Saucedo, E.; Pérez-Rodríguez, A.; Cabot, A. J. Am. Chem. Soc. 2013, 135, 15982−15985. (41) Zhou, H.; Song, T.-B.; Hsu, W.-C.; Luo, S.; Ye, S.; Duan, H.-S.; Hsu, C.-J.; Yang, W.; Yang, Y. J. Am. Chem. Soc. 2013, 135, 15998− 16001. (42) Altamura, G.; Wang, M.; Choy, K. Sci. Rep. 2016, 6, 22109. (43) Coughlan, C.; Ibáñez, M.; Dobrozhan, O.; Singh, A.; Cabot, A.; Ryan, K. M. Chem. Rev. 2017, 117, 5865−6109. (44) Kissinger, H. E. Anal. Chem. 1957, 29, 1702−1706. (45) Friedrich, I.; Weidenhof, V.; Njoroge, W.; Franz, P.; Wuttig, M. J. Appl. Phys. 2000, 87, 4130−4134. (46) Wuttig, M.; Lusebrink, D.; Wamwangi, D.; Welnic, W.; Gillessen, M.; Dronskowski, R. Nat. Mater. 2007, 6, 122−128. (47) Kao, K.-F.; Lee, C.-M.; Chen, M.-J.; Tsai, M.-J.; Chin, T.-S. Adv. Mater. 2009, 21, 1695−1699. (48) Raoux, S.; Welnic, W.; Ielmini, D. Chem. Rev. 2010, 110, 240− 267. (49) Bergerud, A.; Selbach, S. M.; Milliron, D. J. ACS Nano 2016, 10, 6147−6155. (50) Ma, E.; Thompson, C. V.; Clevenger, L. A. J. Appl. Phys. 1991, 69, 2211−2218. (51) Fisher, K.; Barron, S. C.; Bonds, M. A.; Knepper, R.; Livi, K. J. T.; Campbell, G. H.; Browning, N. D.; Weihs, T. P. J. Appl. Phys. 2013, 114, 243509. (52) Lewis, M. H.; Johansen, J. M.; Bell, P. S. J. Am. Ceram. Soc. 1979, 62, 278−288. (53) Marghussian, V. 1-Glass Crystallization. In Nano-Glass Ceramics; William Andrew Publishing: Oxford, U.K., 2015; pp 1− 62, DOI: 10.1016/B978-0-323-35386-1.00001-3. (54) Huang, T. J.; Yin, X.; Tang, C.; Qi, G.; Gong, H. J. Mater. Chem. A 2015, 3, 17788−17796. (55) Singh, A.; Singh, A.; Ong, G. K.; Jones, M. R.; Nordlund, D.; Bustillo, K.; Ciston, J.; Alivisatos, A. P.; Milliron, D. J. Nano Lett. 2017, 17, 3421−3428. (56) Jayabharathi, J.; Sundari, G. A.; Thanikachalam, V.; Jeeva, P.; Panimozhi, S. RSC Adv. 2017, 7, 38923−38934. (57) Mitzi, D. B. Inorg. Chem. 2005, 44, 3755−3761. (58) Yin, X.; Tang, C.; Sun, L.; Shen, Z.; Gong, G. Chem. Mater. 2014, 26, 2005−2014. (59) Altamura, G.; Vidal, J. Chem. Mater. 2016, 28, 3540−3563. (60) Neuschitzer, M.; Marquez, J.; Giraldo, S.; Dimitrievska, M.; Placidi, M.; Forbes, I.; Izquierdo-Roca, V.; Pérez-Rodriguez, A.; Saucedo, E. J. Phys. Chem. C 2016, 120, 9661−9670. (61) Hages, C. J.; Levcenco, S.; Miskin, C. K.; Alsmeier, J. H.; AbouRas, D.; Wilks, R. G.; Bär, M.; Unold, T.; Agrawal, R. Prog. Photovoltaics 2015, 23, 376−384. (62) Liu, X.; Feng, Y.; Cui, H.; Liu, F.; Hao, X.; Conibeer, G.; Mitzi, D. B.; Green, M. Prog. Photovoltaics 2016, 24, 879−898. (63) Jeyasingh, R. G. D.; Caldwell, M. A.; Milliron, D. J.; Wong, H. S. P. In Proceedings of the European Solid State Device Research Conference, September 2011; pp 99−102. (64) Wu, H.; Wang, Z. W.; Fan, H. Y. J. Am. Chem. Soc. 2014, 136, 7634−7636.

H

DOI: 10.1021/acs.nanolett.8b01916 Nano Lett. XXXX, XXX, XXX−XXX