Microstructure Evolution During Selenization of Cu2ZnSnS4 Colloidal

Subscriber access provided by GAZI UNIV

Article

Microstructure Evolution During Selenization of CuZnSnS Colloidal Nanocrystal Coatings 2

4

Boris D Chernomordik, Priyanka M Ketkar, Anne K. Hunter, Amélie E. Beland, Donna D. Deng, and Eray S. Aydil Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/acs.chemmater.5b02462 • Publication Date (Web): 03 Feb 2016 Downloaded from http://pubs.acs.org on February 8, 2016

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.

Chemistry of Materials 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 15

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

Chemistry of Materials

Microstructure Evolution During Selenization of Cu2ZnSnS4 Colloidal Nanocrystal Coatings Boris D. Chernomordik, Priyanka M. Ketkar, Anne K. Hunter, Amélie E. Béland, Donna D. Deng, and Eray S. Aydil* Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, Minnesota 55455, United States. ABSTRACT: Annealing colloidal nanocrystal coatings in a selenium-containing environment to form polycrystalline thin films of the earth-abundant solar absorber copper zinc tin sulfoselenide (CZTSSe) is an attractive approach for making solar cells. We used a closed selenization system to investigate how coatings comprising copper zinc tin sulfide (CZTS) nanocrystals evolve into polycrystalline CZTSSe thin films and studied the effects of selenium vapor pressure, annealing temperature, and heating rate. These studies revealed two different types of microstructures and two different grain growth mechanisms depending on whether the CZTS nanocrystals are exposed to selenium vapor only, or to both selenium vapor and liquid selenium. Coatings annealed in the presence of selenium vapor form a microstructure comprised of micron-size CZTSSe grains on top of a nanocrystalline, carbon-rich, CZTSSe layer. The film microstructure is controlled by concurrent normal and abnormal grain growth and the grain size distribution is bimodal, similar to that observed when CZTS nanocrystal coatings are annealed in sulfur vapor. The size of the abnormal crystals increases with selenium pressure and temperature to as large as 4 µm after annealing at 700 oC in 450 Torr of selenium. Carbon, initially present on nanocrystals as dispersion stabilizing ligands, segregates to the region between the CZTSSe grains and the substrate instead of desorbing from the coating as volatile reaction products such as CSe2. Experiments suggest that carbon segregation occurs due to the tendency for CSe2 to polymerize and form (CSe2-x)n. Coatings annealed in the presence of liquid selenium exhibit neither the bimodal grain size distribution nor the carbon-rich layer between CZTSSe grains and the substrate. In the presence of liquid selenium, the CZTS nanocrystals selenize, grow and coarsen to ~1 µm in size, forming compact CZTSSe films through liquid phase sintering, a mechanism wherein both grain size coarsening and film densification are mediated by the presence of a liquid phase.

INTRODUCTION Power conversion efficiencies of solar cells based on the earth-abundant light absorber copper zinc tin sulfoselenide [Cu2ZnSn(SxSe1-x)4, or CZTSSe] have increased rapidly from 2.67% in 20011 to 12.6% in 2013.2 While there are many approaches to forming thin CZTSSe films for solar cells,3–11 low-cost solution-based methods are emerging as compelling alternatives to vacuum thin film deposition. These solution-based methods include sol-gel deposition, electrodeposition, and coatings cast from colloidal nanocrystal dispersions, slurries, and molecular inks.10,11 All these approaches require post-deposition annealing at 400-600 oC in sulfur or selenium vapor to obtain largegrained (~1 µm) polycrystalline films. Indeed, the highest efficiency solar cells have been achieved with CZTSSe films synthesized from metalorganic slurries in hydrazine.2 In the approach where films are formed by annealing coatings cast from colloidal dispersions of CZTS nanocrystals, the material synthesis and film formation are separated into two steps. First, CZTS nanocrystals are

synthesized and dispersed in a solvent. Following, CZTS coatings are cast onto substrates from these dispersions and annealed in sulfur and/or selenium vapor to form large-grained thin films. The dispersions are stabilized by organic ligands on the surfaces of the nanocrystals and these ligands may incorporate into the nanocrystal coatings. The carbon and hydrogen in these ligands react with sulfur to form volatile H2S and CS2 and are removed from the film during annealing in sulfur.12 Curiously however, when CZTS nanocrystal coatings are annealed in selenium, a carbon-rich layer forms at the interface between the CZTSSe film and the substrate.13 The segregation of carbon to the region between the CZTSSe grains and the substrate, instead of its desorbing from the film as volatile species such as CSe2, contributes to the formation of the bilayer microstructure. This carbon segregation and bilayer microstructure formation have been observed in many studies and appears to be a universal feature of all film synthesis approaches employing annealing under selenium vapor where carbon containing ligands or precursors are present.14,15

ACS Paragon Plus Environment


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

Amidst the understandable flurry to improve power conversion efficiencies of CZTSSe solar cells, systematic exploration of the key process variables that control microstructure development in CZTSSe thin films during selenization of CZTS nanocrystals has been neglected. Herein, we use isothermal annealing of CZTS nanocrystal coatings in a closed selenization system to investigate the roles of selenium vapor pressure, annealing temperature, and heating rate in the formation of polycrystalline CZTSSe thin films. The closed system annealing approach provides a well-defined and reproducible annealing environment with known selenium vapor pressure and temperature. While difficult to implement on large production scale, the material-intrinsic lessons learned via systematic investigations using the closed-system annealing approach can inform other annealing methods. Our experiments suggest that the carbon-rich layer formation between the CZTSSe film and the substrate is due to the formation of nonvolatile compounds, such as polymerized CSe2 and is not due to physical trapping of carbon by the growing grains as has been suggested previously.13,16,17 Additionally, we show that under selenization conditions that favor the presence of liquid selenium within the film, CZTS nanocrystals selenize, grow, sinter, and densify into a compact film containing CZTSSe grains up to ~1 µm in size without the formation of a carbon-rich layer between the film and the substrate.

EXPERIMENTAL Materials Fused quartz polished plates and fused quartz tubing were purchased from GM Associates, Inc. and the soda lime float glass (SLG) substrates were purchased from Valley Design Corp. The solid selenium used for annealing was purchased from Sigma-Aldrich (99.999%). Colloidal CZTS Nanocrystal Synthesis The colloidal CZTS nanocrystal dispersions were prepared using the synthesis methods described in detail in our earlier work.18 In summary, stoichiometric CZTS nanocrystals were synthesized by dissolving copper, zinc, and tin-diethyldithiocarbamate precursor powders in oleic acid at 140 oC. After dissolution, the precursor mixture was injected into a separate flask containing oleylamine heated to 280-340 oC. After growth for 10 minutes, the reaction solution was cooled and the nanocrystals were crashed by adding ethanol followed by centrifugation. After two washing steps, the particles were dispersed in toluene to form a 30 mg/ml dispersion. The nanocrystals were 25-40 nm in diameter as calculated from the width of the (112) X-ray diffraction peak using the Scherrer equation. Film Preparation and Annealing The nanocrystals were drop cast onto quartz substrates. (A detailed discussion of the drop casting procedure can be found in a previous publication.18) We use quartz substrates to investigate a wider range of temperatures than Mo-coated soda lime glass (SLG) allows. Using quartz also simplifies the system by eliminating the role of alkali-

Page 2 of 15

metal impurities in grain growth and reactions of CZTSSe with molybdenum. However, select experiments were also conducted using SLG as the substrate to demonstrate that the microstructures observed on quartz could also be reproduced on SLG and is relevant to processing used in making solar cells. We described the annealing procedure in detail in an earlier publication where we studied annealing of CZTS nanocrystal coatings in sulfur vapor.12 This procedure is followed here but sulfur is replaced by selenium. Briefly, the CZTS nanocrystal-coated quartz substrates were placed in quartz tubes with a predetermined mass of selenium. The calculation of selenium vapor pressure in the ampule is discussed below. The tube was evacuated to ~10-6 Torr and flame-sealed to produce a sealed ampule. The film is then annealed in one of two ways hereafter referred to as the slow-ramping method and the hotloading method. In the slow-ramping method, the ampule is inserted into a cavity in an alumina block placed inside the furnace at room temperature and heated at 6 o C/min to the desired annealing temperature set point (e.g., 600 or 700 oC in the experiments described herein). After one hour at the annealing temperature, the furnace is turned off and the alumina block and the ampule are allowed to cool naturally with a time constant of approximately 90 minutes. The temperature decreases from 700 o C to ~350 oC in 1.5 hours; a schematic of the temporal variation of the furnace temperature is shown in Supporting Information Figure S1. In the hot-loading method, the furnace and the alumina block are preheated to the desired annealing temperature (e.g., 600 or 700 oC) for at least six hours. After preheating, the furnace door is opened, the ampule is quickly inserted into the alumina block, and the furnace door is closed. After 1 hour, the furnace is turned off and the alumina block and the ampule are allowed to cool naturally. When the ampule is inserted into the alumina block, ~1 cm of the ampule protrudes from the alumina block. During cooling, this end cools faster than the rest of the ampule because the ampule is in contact with the alumina block. Consequently, selenium condenses on this cold end and not on the film during the cooling period. Chalcogen Pressure in the Ampule During Annealing Figures 1a and 1b show the temperature dependence of the sulfur and selenium pressures, respectively, in a sealed quartz ampule. Within a closed isothermal system, the chalcogen pressure above the film can be determined from the equilibrium vapor pressure curve (the black lines in Figures 1a and 1b). Above its melting point (115 oC for sulfur and 221 oC for selenium), the chalcogen pressure in the ampule will be the equilibrium vapor pressure as long as there is enough chalcogen mass charged into the ampule to maintain liquid-vapor coexistence, i.e., as long as there is enough S or Se mass in the ampule to achieve the vapor pressure at that temperature. As the temperature is increased, liquid and vapor will coexist until the mass of the chalcogen needed to sustain the saturation vapor pressure exceeds the mass charged into the ampule. Above this temperature, all the liquid is converted into

2 ACS Paragon Plus Environment

Page 3 of 15

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


vapor and the pressure can be calculated from the ideal gas law using the mass of the chalcogen charge, temperature and the ampule volume. The temperature at which the chalcogen goes from two phases into a single vapor phase increases with increasing chalcogen mass charged into the ampule. In Figures 1a and 1b, the lines with different colors correspond to the chalcogen pressure in the ampule as function of temperature. For a given S or Se mass, the pressure in the ampule will follow the black saturation vapor pressure until all the liquid is converted into vapor. This happens at ~260 °C for 1 mg sulfur and at ~370 °C for 10 mg sulfur. Above these temperatures the sulfur pressure (Figure 1a) follows the colored lines (e.g., red or blue), which each represent a different mass of the chalcogen charged into the ampule. This is also the case for selenium (Figure 1b). Temperature-dependent speciation from Rau et al19,20 (see Supporting Information Figures S2 and S3; Tables S1 and S2) was used in these calculations. Entering the single-phase regime is an advantage of the closed system annealing because the annealing temperature and chalcogen pressure can be varied independently from each other by changing the amount of chalcogen charged into the ampule. For example, at 700 o C, 2 mg of selenium will produce ~50 Torr while 20 mg will produce ~450 Torr of selenium vapor. Characterization A Bruker D8 Discover system with a Hi-Star 2D area detector was used to collect X-ray diffraction (XRD) from nanocrystal coatings and annealed films. Raman spectra were collected using the Witec Alpha300R confocal Raman microscope equipped with an Omnichrome Ar ion laser (514.5 nm, ~300 nm beam spot size), a UHTS300 spectrometer and a DV401 CCD detector. An 1800 lines/mm grating was used to collect and disperse the scattered light, resulting in 0.02 cm-1 spectral resolution. The microstructure and elemental composition of the films were examined using a JEOL 6500 field-emission scanning electron microscope (SEM) equipped with a Thermo-Noran Vantage energy dispersive X-ray spectrometer (EDS). For both(f) imaging and EDS measurements, the electron energy was 15 keV. For films with small grains (e.g., 100 nm, we determined the average grain sizes from the SEM images.

EXPERIMENTAL RESULTS CZTS nanocrystal coatings can be transformed into polycrystalline CZTS films by annealing in sulfur.12 During this transformation, the nanocrystals grow, sinter and densify to form a polycrystalline CZTS film with grain sizes ranging from several hundreds of nanometers to microns depending on the annealing conditions.12 When the same CZTS nanocrystal coatings are annealed with Se vapor, selenium exchanges with the sulfur in the CZTS film during growth and sintering of the nanocrystals.

Thus, CZTSSe thin films can be made by annealing CZTS nanocrystals in selenium vapor.21

Figure 1. Vapor pressures of (a) sulfur and (b) selenium in the sealed quartz ampule as a function of temperature for different values of the chalcogen mass charged into the ampule. In both (a) and (b), the black curves are the equilibrium vapor pressure of sulfur or selenium, respectively. The pressure in the ampule follows the different colored curves each corresponding to a different value of the chalcogen mass charged into the ampule. The temperature dependence of the colored curves is calculated from the known ampule volume and the mass charged into the ampule using the ideal gas law, where temperature-dependent speciation of the 19,20 chalcogen vapor is taken into account in the calculations. For a given mass of sulfur or selenium, the pressure in the ampule first follows the equilibrium vapor pressure curve (black) and then continues on the colored curve corresponding to the mass charged into the ampule. Below the temperature where the colored lines intersect the black curve, the chalcogen liquid and vapor coexist. Above that temperature, the chalcogen is all gas.

There is a significant difference between the sulfur and selenium vapor pressures and their temperature dependence. This difference has important consequences for annealing, particularly in microstructure development. Specifically, the equilibrium vapor pressure of sulfur is greater than that of selenium at any given temperature. One consequence of this difference is that the transition temperature from vapor-liquid coexistence to single vapor phase occurs at a higher temperature for selenium than for sulfur (the intersection between the black vapor pressure curve and the different colored curves). For example,



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 15

Figure 2. Digital photographs, plan-view and cross-sectional SEM images comparing films annealed for 1 hour at 700 °C at different Se pressures using the slow-ramping method. The Se pressures in the ampule were (a-c) 450 Torr, (d-f) 250 Torr, (g-i) 50 Torr, and (j-l) 10 Torr, i.e., rows 1-4 top to bottom, respectively. All films were on fused quartz substrates. The scale bars in the plan view images (b, e, h, k) and in the cross-sectional images (c, f, i, l) are 20 µm and 1 µm, respectively.

with 1 mg of sulfur, the transition from liquid-vapor coexistence to a single vapor phase occurs at ~260 °C. In contrast, this transition is at ~480 °C with 2 mg of selenium. At 600 oC, both 1 mg of sulfur and 2 mg of selenium results in ~40 Torr in the ampule. Thus, when grain growth and sintering begins at ~350 °C, selenium liquid and vapor coexist in the ampule. The consequences of this will be discussed later in the article. In contrast, sulfur is all gas after 350 oC for 1-9 mg of sulfur charged into the ampule.

Annealing via the Slow-Ramping Method Figure 2 shows digital photographs and SEM images of nanocrystal coatings after they were annealed for 1 hour at 700 oC using the slow-ramping method with 450 Torr (Figures 2a-2c), 250 Torr (Figures 2d-2f), 50 Torr (Figures 2g-2i), and 10 Torr (Figures 2j-2l) of selenium. (See Supporting Information Figure S1a for temporal variation of the furnace temperature during annealing.) Each row in Figure 2 shows images for a different selenium pressure.


Page 5 of 15

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


For clarity and brevity these films will be referred to as 450s, 250s, 50s, and 10s, where “s” denotes the slow-ramp annealing method and the number preceding the “s” is the selenium pressure during annealing. The SEM images of the coatings annealed at selenium pressures greater than 50 Torr show micron-sized CZTSSe crystals (1-10 µm) on top of a layer comprising much smaller CZTSSe nanocrystals (1 µm) and the substrate during annealing in selenium but not during annealing in sulfur. For example, the CZTS nanocrystal coating annealed at 700 oC for 1 hour in 450 Torr of selenium (Figures 2c and 4e) shows a carbon-rich nanocrystalline CZTSSe layer between the large abnormal CZTSSe crystals and the substrate. In contrast, during sulfidation under identical conditions, carbon is almost entirely removed from the film and there is no sign of a carbon-rich layer at the filmsubstrate interface.12 While sulfidized films do not show a carbon-rich layer, they may still exhibit a bilayered morphology with abnormal large grains on top of a layer with smaller grains. This bilayered morphology may look similar to that obtained upon selenization but the important difference is the lack of significant amounts of carbon in the layer with small CZTS grains. While one would expect the reaction chemistry between carbon and selenium to be similar to that between carbon and sulfur, there are significant differences. These differences offer insight into the reasons for carbon segregation to the interface between the CZTSSe film and the substrate during selenization but not during sulfidation.12,13,24,46 Carbon segregation during selenization is not unique to CZTS nanocrystals and has been observed during selenization of other chalcogenide nanocrystal coatings including Cu(In,Ga)S247 and CuInS2.48 The segregation is also independent of the substrate: the carbon layer forms during selenization when the nanocrystals are on SLG, quartz or Mo. It has been suggested that the abnormal crystals grow quickly during selenization and form a “capping” layer that traps the carbon between the substrate and the CZTSSe crystals.13,16,17 However, this hypothesis is not sup-

ported by our observations. In our films, the abnormal CZTSSe crystal layer is not always continuous: occasionally we encounter gaps in this layer large enough to conduct elemental analysis on the underlying layer by EDS. We observe high concentrations of carbon even in places where the nanocrystalline layer is not covered by large abnormal grains. Apparently, carbon rich layer forms even when there are no large CZTSSe grains to trap the carbon or to block Se from reaching the nanocrystals. Absence of a carbon-rich layer in CZTS nanocrystal coatings annealed in sulfur vapor suggests that sulfur is more effective in removing carbon from these coatings than selenium. This is surprising because CSe2, CS2, OCS and OCSe, some of the expected products of the reactions between carbon and the chalcogen vapor, are all very volatile. Moreover, both sulfur and selenium vapor penetrate well into the nanocrystal coating. This is evident from the efficient exchange of sulfur atoms in CZTS with selenium to form CZTSSe as well as from the formation of MoSe2 and MoS2 at similar rates, during selenization and sulfidation, respectively, when the films are on molybdenumcoated glass. One significant difference between CSe2 and CS2, however, is that CSe2 polymerizes readily.49 Thus, the CSe2 molecules that form via the reactions between the selenium vapor and carbon in the ligands can polymerize to form a (CSe2-x)n matrix, trapping carbon and selenium in the film. Polymerized CSe2 has a weak Raman scattering peak at ~1475 cm-1.50 The Raman spectrum of the floor layer (Supporting Information Figure S9) exhibits a broad feature spanning this region. This broad feature peaks at ~1350 cm-1 and at ~1580 cm-1, characteristic of graphite with nanocrystalline domains.51,52 These strong broad graphite peaks overlap with Raman scattering from (CSe2x)n. It is also likely that some of this (CSe2-x)n polymer loses Se with time and some converts to graphite. Indeed when we anneal a 25 nm thick amorphous carbon film on quartz substrate at 700 oC in 450 Torr selenium for 1 hour using the hot-loading method, the Raman spectrum of this film between 1300-1600 cm-1 resembles the Raman spectrum, of the carbon-rich layer formed between the CZTSSe grains and the substrate.52,53 (See Supporting Information Figure S10 and compare to Figure S9.) While carbon segregation appears to contribute to bilayer microstructure formation by encasing the nanocrystals and imparting a significant kinetic barrier to sintering and grain growth, there may be other factors involved besides those discussed and supported by experiments discussed in this article. Indeed, there are some studies on CuInGa(SSe)2 and CuInSe2 that report small-grained layers adjacent to the Mo layer even in the presumed absence of carbon in the films.54,55 Selenium Condensation and Liquid Phase Sintering An important remaining issue is the reason for two distinct microstructures (e.g., Figures 4b and 4c vs. Figures 4d and 4e) on the same film during annealing using the hot-loading method. Here we consider various possibilities and discuss experiments that implicate selenium condensation within the nanocrystal pores and a mecha-



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

nism resembling liquid phase sintering. After eliminating various possibilities, we argue that selenium condensation leads to the microstructure observed in the smooth regions of the film annealed using the hot-loading method. One possibility is that the films annealed using the hotloading method and those annealed using the slowramping method spend different amounts of time at elevated temperatures (i.e. >350 oC). The temperatures of the ampule and its contents rise to the set point (700 oC) more rapidly during annealing by the hot-loading method than during annealing by the slow-ramping method. Thus, the film spends less time at elevated temperatures while annealing with the hot-loading method. To explore whether this time difference is related to the formation of the smooth regions, we conducted an experiment where the film was annealed using the hot-loading method but the time the film spends at elevated temperatures was increased to three hours. Figures 6a, 6b, and 6c show digital photograph and plan-view SEM images of a CZTS film annealed at 700 oC for 3 hours at 250 Torr of selenium using the hot-loading method. This film also exhibited both smooth and matte regions, thus ruling out the difference in annealing time as the reason for the two distinct microstructures in films annealed using the hotloading method. The elemental compositions of the various regions are listed in Supporting Information Table S6.

Figure 6. (a) Digital photograph of a CZTS nanocrystal coato ing, cast on fused quartz, and annealed at 700 C for 3 hours at 250 Torr selenium using the hot-loading method. Plan view SEM images of the (b) smooth and (c) matte regions of this film.

To test whether the development of the two distinct microstructures is unique to annealing of CZTS nanocrystal coatings, we annealed a 430 nm thick Cu-Zn-Sn alloy film that was deposited on a quartz substrate using cosputtering.56 This film was annealed at 700 oC in 450 Torr selenium for one hour using the hot-loading method. The resulting film morphology was uniform and exhibited micron size grains similar to the matte regions of the nanocrystal coatings annealed under the same conditions but, naturally, without the carbon-rich layer. Supporting Information Figures S11a and S11b, show digital photographs before and after annealing, while Figures S11c, S11d and S11e show a plan view SEM image, XRD pattern, and Raman spectrum from this film. The XRD pattern and Raman spectrum are consistent with CZTSe. We conclude that, the microstructure observed in the smooth regions of the films annealed using the hot-loading method is obtained only when annealing CZTS nanocrystal coatings. One important characteristic of the CZTS nanocrystal coatings is the presence of nanopores. One possibility is

Page 10 of 15

that the selenium vapor condenses in the small pores of the nanocrystal film due to the pressure difference across the high-curvature liquid-vapor interface (i.e., capillary condensation).57,58 To test whether capillary condensation may be responsible for the microstructure observed in the smooth regions of the film, we annealed a nanocrystalline coating using the slow-ramping method such that the selenium vapor pressure was always (during the entire heating and cooling cycle) on the saturated liquid-vapor coexistence equilibrium line (black curve in Figure 1b). Since capillary condensation occurs below that saturation vapor pressure, this would have guaranteed the condensation of selenium within the pores if they were small enough to cause condensation. Slow heating eliminates the possibility that the selenium is condensing due to temperature gradients. If selenium is condensing in the nanopores of the coating and is responsible for the microstructure shown in Figures 4b and 4c we should see this smooth microstructure at least in some parts of the film. Figure 7 shows a digital photograph and SEM images of a film annealed using the slow-ramping method at 600 oC in 240 Torr of selenium. (The mass of selenium charged into the ampule, 26 mg, was in excess of that necessary to establish the saturated vapor pressure, 13 mg.) The resulting film is entirely uniform and has the microstructure corresponding to the regions with matte appearance (Figures 4d and 4e), suggesting that capillary condensation is not the cause for the formation of the morphology with smooth appearance.

Figure 7. (a) Digital photograph of a CZTS nanocrystal coato ing cast onto fused quartz and then annealed at 600 C for 1 hour in 240 Torr of selenium using the slow-ramping method. (b) Plan view and (c) cross sectional SEM images of this film which uniformly had the microstructure with the matte appearance.

Selenium can still condense on the film during annealing while using the hot-loading method. Rapid temperature rise can create transient temperature gradients across the ampule and the substrate. We hypothesized that such gradients can lead to vaporization of selenium from the hot regions of the ampule and subsequent condensation on the colder regions including the substrate. During rapid heating, the temperature of the substrate can lag behind that of the region where the selenium is placed. Such a lag is plausible because the temperature of the thin-wall quartz ampule, in contact with the hot alumina block, increases faster than that of the thicker quartz substrate. As the temperature of the ampule increases rapidly, the selenium evaporates but then condenses when it encounters the colder quartz substrate. The visual clue implicating spatial gradients is the parabolic shape of the boundary between the smooth and


Page 11 of 15

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


matte regions: see for example, the digital photograph of the film annealed at 450 Torr in Figure 4a. Moreover, we found that the direction in which vertex of this parabolic boundary points is determined by the position of the solid selenium charge relative to the substrate: the vertex is always positioned on the side away from the solid selenium charge, i.e., the smooth region always appears closer to the solid selenium charge while the matte region is always on the farthest side. Figure 8 illustrates this observation regarding the shape of the smooth region and its relation to the position of the selenium charge with respect to the substrate. These observations suggest that selenium may indeed be condensing on the film during the temperature transient because the substrate temperature lags behind that of the region with the selenium charge.

Figure 8. An illustration depicting the shape of the region that has the microstructure with smooth appearance as a function of the location of the selenium shot with respect to the substrate. The vertex of the parabola-shaped boundary between the smooth and matte regions is always away from the solid selenium as shown in (a) and in (b).

Based on these experiments we propose that the two regions with different microstructures shown in Figure 4 develop through two different mechanisms: the morphology with the smooth appearance develops through liquidphase (selenium) sintering and the morphology with the matte appearance develops through selenium-vapor assisted grain growth. The latter is similar to the mechanism that was described recently for annealing CZTS nanocrystal coatings in sulfur. The former does not occur during annealing in sulfur because annealing conditions are typically too far from the liquid-vapor coexistence line (see Figure 1). These two fundamentally different mechanisms are revealed, serendipitously, in annealing experiments where the ampules are heated rapidly. Transient temperature gradients, caused by the rapid heating, condense selenium on the side of the substrate closest to the selenium source while condensation does not take place on the farther side. Liquid phase sintering is a mechanism wherein both grain size coarsening and film densification are mediated by the presence of a liquid phase.59 Coarsening refers to an increase in crystal size while densification necessarily implies the removal of pores and voids. The microscopic details of the mediation mechanism are material specific but the common feature is the presence of a liquid phase

that mediates atomic processes that lead to grain growth, densification and sintering. Liquid phase sintering typically involves 5-30 vol.% liquid60 and generally occurs in three stages: (1) rearrangement of the nanocrystals within the liquid to increase packing; (2) solution-precipitation, where atoms at particle contact points dissolve due to high capillary pressure and then re-precipitate in lowpressure regions of the solid particles leading to contact flattening and densification; and (3) a combination of solid phase sintering of the solid particle skeletal network, pore migration, and continuation of the solutionprecipitation process depending on liquid content, pore size, and solid-liquid solubility.60,61 For densification, it is desirable for the liquid phase to wet the solid particles. Solubility of the solid phase components in the liquid phase is also desired because liquid diffusivities are higher than solid diffusivities.59,61 Chemical gradients contribute an additional driving force for boundary motion and coalescence during sintering. Such gradients are present in our films as CZTS transforms into CZTSSe in the presence of selenium liquid and/or selenium vapor. During annealing using the hot-loading method, liquid phase sintering occurs during the time that the substrate temperature lags behind that of the ampule and the selenium vapor. Selenium condenses on the film and is taken up by the coating via capillary action. As the temperature lag decreases, the liquid selenium evaporates from the coating. Following, grain growth occurs predominantly by solid state diffusion and selenium-vapor phase assisted growth. To investigate the possibility of liquid phase sintering leading to the microstructure observed in the smooth regions of the films annealed by the hot-loading method (Figures 4b and 4c), a CZTS nanocrystal coating was annealed with several pieces of selenium shots placed directly on top of the coating in selected locations. The annealing conditions were chosen such that, in the absence of these selenium shots on the coating, a uniform film with the matte appearance would be obtained. Specifically, the film was annealed at 600 oC at the saturated vapor pressure of selenium (240 Torr) using the slow-ramping method. These are the same conditions used to anneal the coating in Figure 7. The total mass of selenium charged into the ampule was double the mass necessary to establish the saturated vapor pressure and to ensure liquid-vapor coexistence during annealing. As the temperature increased, the selenium shot melted on the CZTS nanocrystal film, forming a liquid selenium pool that partially covered the coating. Figures 9a and 9b, show the digital photographs of the film before and after annealing, respectively. The left half of the film, where the selenium shots were placed, exhibits the microstructure that appears smooth (Figure 9e) whereas the right half of the film exhibits the microstructure with matte-like appearance (Figure 9d). The emergence of the microstructure with smooth appearance where selenium has melted, even though the coating was annealed using the slowramping approach, conclusively associates this microstructure with the presence of liquid selenium on the coatings. The size, shape, and location of the region



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

where the smooth microstructure appears are consistent with liquid selenium spreading out towards the left side of the film after the shots have melted. The smoothmicrostructure is observed in circular patches where the selenium shots were placed. The optical and SEM contrast is due to the grain size differences between the CZTS film in the circular patches and the regions surrounding these patches. Within the circular patches, the average crystal size (~450 nm) is double that in the surrounding regions (~200 nm). We surmise that the size difference is due to the difference in the amount of time that these regions were exposed to liquid selenium: the regions within the circular patches, where the selenium shots were placed, were exposed to liquid selenium the longest and therefore exhibit larger grains. As the selenium shots melt, spread and evaporate, the liquid thickness would always be highest where the selenium shots were placed. There is no evidence of elemental selenium remaining in the film: all of the excess selenium eventually condensed on the opposite end of the ampule during the cooling period.

Page 12 of 15

CONCLUSIONS The effects of annealing temperature, selenium pressure and heating rate on the microstructure of CZTS films formed by annealing coatings cast from CZTS nanocrystal dispersions were studied in a closed system. Two different types of microstructures were discovered depending on whether or not liquid selenium is present in the nanopores of the CZTS nanocrystal coatings during annealing. Films, or portions of the films, annealed without liquid selenium condensation on the nanocrystal coating develop into a microstructure comprised of micron-size abnormal CZTSSe crystals on top of a nanocrystalline, carbon-rich, CZTSSe floor layer. Significant fraction of carbon, originally present as ligands on the nanocrystals polymerize to form (CSe2-x)n and nanocrystalline graphite and remain in the floor layer. The size of the abnormal crystals increases with selenium pressure and temperature until a continuous layer of large grains form provided that there is enough nanocrystals coated on the substrate. Continuous CZTSSe films with largest CZTSSe crystals (110 µm, 4 µm average) were obtained when films were annealed at 700 oC in 450 Torr of selenium. This type of microstructure had a matte appearance. In contrast, when liquid selenium is present on the nanocrystal coating during annealing, the nanocrystals in the floor layer grow rapidly and this rapid growth suppresses the abnormal crystal formation. Through liquid-phase sintering, the floor layer transforms into a dense film with up to ~1 µm CZTSSe grains. This microstructure has a smooth and shiny appearance. Either mechanism could be used to get large-grained films with microstructure suitable for solar cells though other factors such as the ability to remove carbon, to obtain low-defect density films and to maintain desirable cation stoichiometry also contribute to solar cell performance.

ASSOCIATED CONTENT Supporting Information. Additional calculations and characterization data, such as elemental compositions, XRD, Raman, and SEM images. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author Figure 9. Several pieces of the selenium shot (out of 26 mg total; circled in red) were placed on top of the CZTS nanocrystal coating prior to annealing, as shown in the digital photograph in (a). (b) The digital photograph of the film after annealing shows that the left half of the film is different in appearance than the right half, and there are circular patches in the locations where selenium pellets were placed. Plan-view SEM images showing (c) a circular patch on the left-half of the film, (d) the right-half of the film, which has the microstructure with matte appearance, (e) the left-half of the film, which has the microstructure with the smooth appearance and (f) inside the circular patch, which also has the the microstructure with smooth appearance but larger grains.

*E-mail: [email protected] (ESA).

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported primarily by the National Science Foundation through the University of Minnesota MRSEC under Award Number DMR-1420013 and partially by the Initiative for Renewable Energy & the Environment, IREE (RL0004-11). B.D.C. acknowledges financial support from the NSF Graduate Research Fellowship Program. Parts of this work were completed at the Characterization Facility at the University of Minnesota, which receives partial support from NSF through the MRSEC program. The authors would like to


Page 13 of 15

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


thank M. Johnson, C. Leighton, D. J. Norris and M. Manno for helpful discussions.

REFERENCES (1) Katagiri, H.; Saitoh, K.; Washio, T.; Shinohara, H.; Kurumadani, T.; Miyajima, S. Development of Thin Film Solar Cell Based on Cu2ZnSnS4 Thin Films. Sol. Energy Mater. Sol. Cells 2001, 65, 141–148. (2) Wang, W.; Winkler, M. T.; Gunawan, O.; Gokmen, T.; Todorov, T. K.; Zhu, Y.; Mitzi, D. B. Device Characteristics of CZTSSe Thin-Film Solar Cells with 12.6% Efficiency. Adv. Energy Mater. 2014, 4, 1301465. (3) Repins, I.; Beall, C.; Vora, N.; DeHart, C.; Kuciauskas, D.; Dippo, P.; To, B.; Mann, J.; Hsu, W.-C.; Goodrich, A.; et al. Co-Evaporated Cu2ZnSnSe4 Films and Devices. Sol. Energy Mater. Sol. Cells 2012, 101, 154–159. (4) Li, J. B.; Chawla, V.; Clemens, B. M. Investigating the Role of Grain Boundaries in CZTS and CZTSSe Thin Film Solar Cells with Scanning Probe Microscopy. Adv. Mater. 2012, 24, 720–723. (5) Su, Z.; Sun, K.; Han, Z.; Cui, H.; Liu, F.; Lai, Y.; Li, J.; Hao, X.; Liu, Y.; Green, M. A. Fabrication of Cu2ZnSnS4 Solar Cells with 5.1% Efficiency via Thermal Decomposition and Reaction Using a Non-Toxic Sol–gel Route. J. Mater. Chem. A 2014, 2, 500–509. (6) Jeon, J.-O.; Lee, K. D.; Seul Oh, L.; Seo, S.-W.; Lee, D.K.; Kim, H.; Jeong, J.-H.; Ko, M. J.; Kim, B.; Son, H. J.; et al. Highly Efficient Copper-Zinc-Tin-Selenide (CZTSe) Solar Cells by Electrodeposition. ChemSusChem 2014, 7, 1073–1077. (7) Miskin, C. K.; Yang, W. C.; Hages, C. J.; Carter, N. J.; Joglekar, C. S.; Stach, E. A.; Agrawal, R. 9.0% Efficient Cu2ZnSn(S,Se)4 Solar Cells from Selenized Nanoparticle Inks. Prog. Photovoltaics Res. Appl. 2015, 23, 654–659. (8) Cao, Y.; Denny, M. S.; Caspar, J. V; Farneth, W. E.; Guo, Q.; Ionkin, A. S.; Johnson, L. K.; Lu, M.; Malajovich, I.; Radu, D.; et al. High-Efficiency Solution-Processed Cu2ZnSn(S,Se)4 ThinFilm Solar Cells Prepared from Binary and Ternary Nanoparticles. J. Am. Chem. Soc. 2012, 134, 15644–15647. (9) Woo, K.; Kim, Y.; Yang, W.; Kim, K.; Kim, I.; Oh, Y.; Kim, J. Y.; Moon, J. Band-Gap-Graded Cu2ZnSn(S1-x,Sex)4 Solar Cells Fabricated by an Ethanol-Based, Particulate Precursor Ink Route. Sci. Rep. 2013, 3, 3069. (10) Xin, H.; Katahara, J. K.; Braly, I. L.; Hillhouse, H. W. 8% Efficient Cu 2 ZnSn(S,Se) 4 Solar Cells from Redox Equilibrated Simple Precursors in DMSO. Adv. Energy Mater. 2014, 4, 1301823. (11) Schnabel, T.; Löw, M.; Ahlswede, E. Vacuum-Free Preparation of 7.5% Efficient Cu2ZnSn(S,Se)4 Solar Cells Based on Metal Salt Precursors. Sol. Energy Mater. Sol. Cells 2013, 117, 324–328. (12) Chernomordik, B. D.; Béland, A. E.; Deng, D. D.; Francis, L. F.; Aydil, E. S. Microstructure Evolution and Crystal Growth In Cu2ZnSnS4 Thin Films Formed By Annealing Colloidal Nanocrystal Coatings. Chem. Mater. 2014, 26, 3191– 3201. (13) Mainz, R.; Walker, B. C.; Schmidt, S. S.; Zander, O.; Weber, A.; Rodriguez-Alvarez, H.; Just, J.; Klaus, M.; Agrawal, R.; Unold, T. Real-Time Observation of Cu2ZnSn(S,Se)4 Solar Cell Absorber Layer Formation from Nanoparticle Precursors. Phys. Chem. Chem. Phys. 2013, 15, 18281. (14) Wang, G.; Zhao, W.; Cui, Y.; Tian, Q.; Gao, S.; Huang, L.; Pan, D. Fabrication of a Cu2ZnSn(S,Se)4 Photovoltaic Device by a Low-Toxicity Ethanol Solution Process. ACS Appl. Mater. Interfaces 2013, 5, 10042–10047. (15) Ilari, G. M.; Fella, C. M.; Ziegler, C.; Uhl, A. R.; Romanyuk, Y. E.; Tiwari, A. N. Solar Cell Absorbers Spin-

Coated From Amine-Containing Ether Solutions. Sol. Energy Mater. Sol. Cells 2012, 104, 125–130. (16) Wang, W.; Han, S.-Y.; Sung, S.-J.; Kim, D.-H.; Chang, C.-H. 8.01% CuInGaSe2 Solar Cells Fabricated by Air-Stable Low-Cost Inks. Phys. Chem. Chem. Phys. 2012, 14, 11154–11159. (17) Ahn, S. J.; Kim, C. W.; Yun, J. H.; Gwak, J.; Jeong, S.; Ryu, B.-H.; Yoon, K. H. CuInSe2 (CIS) Thin Film Solar Cells by Direct Coating and Selenization of Solution Precursors. J. Phys. Chem. C 2010, 114, 8108–8113. (18) Chernomordik, B. D.; Béland, A. E.; Trejo, N. D.; Gunawan, A. A.; Deng, D. D.; Mkhoyan, K. A.; Aydil, E. S. Rapid Facile Synthesis of Cu2ZnSnS4. J. Mater. Chem. A 2014, 2, 10389–10395. (19) Rau, H.; Kutty, T. R. N.; Guedes De Carvalho, J. R. F. Thermodynamics of Sulphur Vapour. J. Chem. Thermodyn. 1973, 5, 833–844. (20) Rau, H. Vapour Composition and Critical Constants of Selenium. J. Chem. Thermodyn. 1974, 6, 525–535. (21) Guo, Q.; Ford, G. M.; Yang, W.-C.; Walker, B. C.; Stach, E. A.; Hillhouse, H. W.; Agrawal, R. Fabrication of 7.2% Efficient CZTSSe Solar Cells Using CZTS Nanocrystals. J. Am. Chem. Soc. 2010, 132, 17384–17386. (22) Thompson, C. V. Secondary Grain Growth in Thin Films of Semiconductors: Theoretical Aspects. J. Appl. Phys. 1985, 58, 763–772. (23) Humphreys, F.; Hatherly, M. Abnormal Grain Growth. In Recrystallization and Related Annealing Phenomena; Sleeman, D., Ed.; Elsevier Ltd: Oxford, Kidlington, UK, 2004; pp 368–378. (24) Van Embden, J.; Chesman, A. S. R.; Gaspera, E. Della; Duffy, N. W.; Watkins, S. E.; Jasieniak, J. J. Cu2ZnSnS4xSe4(1−x) Solar Cells from Polar Nanocrystal Inks. J. Am. Chem. Soc. 2014, 136, 5237–5240. (25) Khare, A.; Himmetoglu, B.; Johnson, M.; Norris, D. J.; Cococcioni, M.; Aydil, E. S. Calculation of the Lattice Dynamics and Raman Spectra of Copper Zinc Tin Chalcogenides and Comparison to Experiments. J. Appl. Phys. 2012, 111, 083707. (26) Wang, K.; Gunawan, O.; Todorov, T.; Shin, B.; Chey, S. J.; Bojarczuk, N. A.; Mitzi, D.; Guha, S. Thermally Evaporated Cu2ZnSnS4 Solar Cells. Appl. Phys. Lett. 2010, 97, 143508. (27) Grossberg, M.; Krustok, J.; Raudoja, J.; Raadik, T. The Role of Structural Properties on Deep Defect States in Cu2ZnSnS4 Studied by Photoluminescence Spectroscopy. Appl. Phys. Lett. 2012, 101, 102102. (28) Grossberg, M.; Krustok, J.; Raudoja, J.; Timmo, K.; Altosaar, M.; Raadik, T. Photoluminescence and Raman Study of Cu2ZnSn(SexS1−x)4 Monograins for Photovoltaic Applications. Thin Solid Films 2010, 519, 7403–7406. (29) Fairbrother, A.; Fontané, X.; Izquierdo-Roca, V.; Placidi, M.; Sylla, D.; Espindola-Rodriguez, M.; Marino-Lopez, S.; Pulgarin, F. A.; Vigil-Galán, O.; Perez-Rodriguez, A.; et al. Secondary Phase Formation in Zn-rich Cu2ZnSnSe4-based Solar Cells Annealed in Low Pressure and Temperature Conditions. Prog. Photovoltaics Res. Appl. 2014, 22, 479–487. (30) He, J.; Sun, L.; Chen, S.; Chen, Y.; Yang, P.; Chu, J. Composition Dependence of Structure and Optical Properties of Cu2ZnSn(S,Se)4 Solid Solutions: An Experimental Study. J. Alloys Compd. 2012, 511, 129–132. (31) Khare, A.; Himmetoglu, B.; Cococcioni, M.; Aydil, E. S. First Principles Calculation of the Electronic Properties and Lattice Dynamics of Cu2ZnSn(S1−xSex)4. J. Appl. Phys. 2012, 111, 123704. (32) Cheng, A.-J.; Manno, M.; Khare, A.; Leighton, C.; Campbell, S. A.; Aydil, E. S. Imaging and Phase Identification of Cu2ZnSnS4 Thin Films Using Confocal Raman Spectroscopy. J. Vac. Sci. Technol. A 2011, 29, 051203.



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

(33) Ganchev, M.; Iljina, J.; Kaupmees, L.; Raadik, T.; Volobujeva, O.; Mere, A.; Altosaar, M.; Raudoja, J.; Mellikov, E. Phase Composition of Selenized Cu2ZnSnSe4 Thin Films Determined by X-Ray Diffraction and Raman Spectroscopy. Thin Solid Films 2011, 519, 7394–7398. (34) Cheng, Y. C.; Jin, C. Q.; Gao, F.; Wu, X. L.; Zhong, W.; Li, S. H.; Chu, P. K. Raman Scattering Study of Zinc Blende and Wurtzite ZnS. J. Appl. Phys. 2009, 106, 123505. (35) Levi, Z.; Bineva, I.; Nesheva, D. Raman Scattering from ZnSe Nanolayers. Acta Phys. Pol. A 2009, 116, 75–77. (36) Fernandes, P. A.; Salomé, P. M. P.; da Cunha, A. F. Study of Polycrystalline Cu2ZnSnS4 Films by Raman Scattering. J. Alloys Compd. 2011, 509, 7600–7606. (37) Wang, C. Raman Scattering , Far Infrared Spectrum and Photoluminescence of SnS 2 Nanocrystallites. Chem. Phys. Lett. 2002, 357, 371–375. (38) Chandrasekhar, H. R.; Humphreys, R. G. Infrared and Raman Spectra of the IV-VI Compounds SnS and SnSe. Phys. Rev. B 1977, 15, 2177–2183. (39) Smith, A. J.; Meek, P. E.; Y, L. W. Raman Scattering Studies of SnS2 and SnSe2. J. Phys. C Solid State Phys. 1977, 10, 1321–1333. (40) Izquierdo-Roca, V.; Pérez-Rodríguez, A.; RomanoRodríguez, A.; Morante, J. R.; Álvarez-García, J.; Calvo-Barrio, L.; Bermudez, V.; Grand, P. P.; Ramdani, O.; Parissi, L.; et al. Raman Microprobe Characterization of Electrodeposited S-Rich CuIn(S,Se)2 for Photovoltaic Applications: Microstructural Analysis. J. Appl. Phys. 2007, 101, 103517. (41) Xue, C.; Papadimitriou, D.; Raptis, Y. S.; Richter, W.; Esser, N.; Siebentritt, S.; Lux-Steiner, M. C. Micro-Raman Study of Orientation Effects of CuxSe-Crystallites on Cu-Rich CuGaSe2 Thin Films. J. Appl. Phys. 2004, 96, 1963. (42) Fernandes, P. A.; Salomé, P. M. P.; da Cunha, A. F. A Study of Ternary Cu2SnS3 and Cu3SnS4 Thin Films Prepared by Sulfurizing Stacked Metal Precursors. J. Phys. D. Appl. Phys. 2010, 43, 215403. (43) Munce, C. G.; Parker, G. K.; Holt, S. a.; Hope, G. a. A Raman Spectroelectrochemical Investigation of Chemical Bath Deposited CuxS Thin Films and Their Modification. Colloids Surfaces A 2007, 295, 152–158. (44) Mernagh, T. P.; Trudu, A. G. A Laser Raman Microprobe Study of Some Geologically Important Sulphide Minerals. Chem. Geol. 1993, 103, 113–127. (45) Fernandes, P. a.; Sousa, M. G.; Salomé, P. M. P.; Leitão, J. P.; da Cunha, A. F. Thermodynamic Pathway for the Formation of SnSe and SnSe2 Polycrystalline Thin Films by Selenization of Metal Precursors. CrystEngComm 2013, 15, 10278. (46) Wang, C.; Manthiram, A. Low-Cost CZTSSe Solar Cells Fabricated with Low Band Gap CZTSe Nanocrystals, Environmentally Friendly Binder, and Nonvacuum Processes. ACS Sustain. Chem. Eng. 2014, 2, 561–568.

Page 14 of 15

(47) Klugius, I.; Miller, R.; Quintilla, A.; Friedlmeier, T. M.; Blázquez-Sánchez, D.; Ahlswede, E.; Powalla, M. Growth Mechanism of Thermally Processed Cu(In,Ga)S2 Precursors for Printed Cu(In,Ga)(S,Se)2 Solar Cells. Phys. Status Solidi - Rapid Res. Lett. 2012, 6, 297–299. (48) Niemegeers, A.; Burgelman, M.; De Vos, A. On the CdS/CuInSe2 Conduction Band Discontinuity. Appl. Phys. Lett. 1995, 67, 843–845. (49) Ives, D. J. G.; Pittman, R. W.; Wardlaw, W. The Preparation, Properties, and Chlorination Products of Carbon Diselenide. J. Chem. Soc. 1947, 1080–1083. (50) Iqbal, Z.; Correale, S. T.; Reidinger, F.; Baughman, R. H.; Okamoto, Y. Synthesis and Structure–property Aspects of Poly(carbon Diselenide). J. Chem. Phys. 1988, 88, 4492–4497. (51) Wang, Y.; Alsmeyer, D. C.; McCreery, R. L. Raman Spectroscopy of Carbon Materials: Structural Basis of Observed Spectra. Chem. Mater. 1990, 2, 557–563. (52) Chu, P. K.; Li, L. Characterization of Amorphous and Nanocrystalline Carbon Films. Mater. Chem. Phys. 2006, 96, 253–277. (53) Ferrari, A. C.; Robertson, J. Interpretation of Raman Spectra of Disordered and Amorphous Carbon. Phys. Rev. B 2000, 61, 14095–14107. (54) Lim Y. S.; Kwon, H.-S., Jeong, J.; Kim, J. Y., Kim, H.; Ko, M. J.; Jeong, U.;Lee, D.-K. Colloidal Solution-Processed CuInSe2Solar Cells with Significantly Improved Efficiency up to 9% by Morphological Improvement. ACS Appl. Mater. Interfaces 2013, 6, 259-267. (55) Goushi, Y.; Hakuma, H.; Tabuchi, K.; Kijima, S.; Kushiya, K. Fabrication of Pentary Cu(InGa)(SSe)2 Absorbers by Selenization and Sulfurization. Sol. Energ. Mater. Sol. C. 2009, 93, 1318-1320. (56) Johnson, M.; Baryshev, S. V; Thimsen, E.; Manno, M.; Zhang, X.; Veryovkin, I. V; Leighton, C.; Aydil, E. S. AlkaliMetal-Enhanced Grain Growth in Cu2ZnSnS4 Thin Films. Energy Environ. Sci. 2014, 7, 1931–1938. (57) Hiemenz, P. C.; Rajagopalan, R. Some Models of Capillary Condensation. In Principles of Colloid and Surface Chemistry; Hiemenz, P. C., Rajagopalan, R., Eds.; Taylor & Francis: Boca Raton, Florida, 1997; pp 438–439. (58) Gemici, Z.; Schwachulla, P. I.; Williamson, E. H.; Rubner, M. F.; Cohen, R. E. Targeted Functionalization of Nanoparticle Thin Films via Capillary Condensation. Nano Lett. 2009, 9, 1064–1070. (59) German, R. M. Liquid Phase Sintering; Plenum Press, 1985. (60) German, R. M.; Suri, P.; Park, S. J. Review: Liquid Phase Sintering. J. Mater. Sci. 2009, 44, 1–39. (61) Kingery, W. D. Densification During Sintering in the Presence of a Liquid Phase. I. Theory. J. Appl. Phys. 1959, 30, 301–306.


Page 15 of 15

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

Chemistry of Materials TABLE OF CONTENTS GRAPHIC

ACS Paragon Plus Environment

15

Microstructure Evolution During Selenization of Cu2ZnSnS4 Colloidal

Recommend Documents