Research Article www.acsami.org
Enhanced Conductivity in CZTS/Cu2−xSe Nanocrystal Thin Films: Growth of a Conductive Shell Lasantha Korala,† J. Tyler McGoffin,‡ and Amy L. Prieto*,† †
Department of Chemistry and ‡Department of Physics, Colorado State University, Fort Collins, Colorado 80523, United States S Supporting Information *
ABSTRACT: Poor charge transport in Cu2ZnSnS4 (CZTS) nanocrystal (NC) thin films presents a great challenge in the fabrication of solar cells without postannealing treatments. We introduce a novel approach to facilitate the charge carrier hopping between CZTS NCs by growing a stoichiometric Cu2Se shell that can be oxidized to form a conductive Cu2−xSe phase when exposed to air. The CZTS/Cu2Se core/shell NCs with varying numbers of shell monolayers were synthesized by the successive ionic layer adsorption and reaction (SILAR) method, and the variation in structural and optical properties of the CZTS NCs with varying shell thicknesses was investigated. Solid-phase sulfide ligand exchange was employed to fabricate NC thin films by layer-by-layer dip coating and a 2 orders of magnitude rise in dark conductivity (∼10−3 S cm−1 at 0 monolayer and ∼10−1 S cm−1 at 1.5 monolayers) was observed with an increase in the number of shell monolayers. The approach described herein is the first key step in achieving a significant increase in the photoconductivity of as-deposited CZTS NC thin films. KEYWORDS: CZTS, nanocrystals, core/shell, thin films, conductivity, solar cells
■
absorber layer.4,5 However, high-temperature annealing of the as-fabricated CZTS NC films in Se vapor (selenization) is required to enhance the grain growth in order to reduce the grain boundaries and surface defects, thereby providing facile transportation for photogenerated charge carriers.6 Although selenization yields Cu2ZnSn(SxSe1−x)4 solar cells with appreciable efficiencies (>8%),7−9 high-temperature annealing significantly increases the manufacturing cost. Furthermore, elimination of the selenization step would enable the use of flexible substrates and consequently reduce the cost of solar modules. Steinhagen et al. have reported CZTS/CdS heterojunction NC solar cells with comparatively lower efficiencies (0.23%) that were fabricated without annealing the absorber layer.5 The highly insulating nature of the NC surface ligand coverage is likely a key contributor to the low efficiency of these devices, in addition to surface and bulk defects. In binary lead chalcogenide NC-based devices, the exchange of native longchain ligands with short-chain organic molecules10−12 or inorganic anions13−15is widely employed to improve the conductivity in NC thin films and to passivate the NC surface states. However, the higher intrinsic defect density of quaternary CZTS NCs relative to binary lead chalcogenide NCs16 makes it difficult to sufficiently improve the charge transport properties of NC thin films solely by ligand exchange.17 In this report, we introduce a novel method to improve the conductivity in CZTS NC films without the selenization step
INTRODUCTION An ever-increasing demand for energy has brought tremendous attention to the exploitation of renewable energy sources due to the environmental impact and limited supplies of fossil fuels. Among renewable energy sources, solar energy is the cleanest and most widely available energy source, yet efficient and costeffective solar energy harvesting is a great challenge. Solar photovoltaic electricity generation has gained considerable attention all over the world as a viable approach to sustainable energy production. However, novel materials and processing strategies, and innovative technologies, are required to reduce the cost and increase the widespread commercialization of solar cells. Over the past few years, Cu2ZnSnS4 (CZTS) has emerged as a promising material for thin film solar cells due to its nearoptimum direct band gap (1.5 eV) and large absorption coefficient (∼104 cm−1).1 Moreover, CZTS is composed of earth abundant and nontoxic elements, which is ideal for the scalable manufacturing of solar cells. Shin et al. have reported the fabrication of pure CZTS solar cells with 8.4% efficiency by a vacuum process.2 However, solution-phase thin film deposition processes are more desirable for the manufacturing of inexpensive solar cells. Recently, more than 12% efficiency has been achieved in thin film devices using a Cu2ZnSn(SxSe1−x)4 absorber layer deposited by a hydrazine solutionbased process.3 It would be ideal to avoid the use of hydrazine as a solvent because of concerns about its toxicity and explosiveness. Inks composed of high-quality, phase-pure CZTS nanocrystals (NCs) produced via a milder solution-phase synthesis provide a low-cost and less toxic alternative to fabricate the © 2016 American Chemical Society
Received: December 2, 2015 Accepted: January 8, 2016 Published: January 8, 2016 4911
DOI: 10.1021/acsami.5b11037 ACS Appl. Mater. Interfaces 2016, 8, 4911−4917
Research Article
ACS Applied Materials & Interfaces
the nitrogen glovebox by using anhydrous solvents. The NCs were first precipitated with ethanol followed by centrifugation. The resultant solid was then re-dispersed in toluene and centrifuged, and finally the supernatant was precipitated with ethanol. The NCs were vacuumdried and stored in a nitrogen glovebox for further use. Characterization. X-ray diffraction (XRD) was performed on a Scintag X-2 advanced diffraction system equipped with Cu Kα radiation (λ = 1.54 Å). The NC dispersions were dropcast onto zero background SiO2 sample holders for XRD analysis. Low-resolution transmission electron microscope (TEM) analysis was carried out on a JEOL JEM 1400 instrument operated at an accelerating voltage of 100 kV. Carbon-coated copper TEM grids were dipped into dilute NC dispersions for a few seconds and air-dried for TEM imaging. Highresolution TEM images were obtained with a JEOL JEM 2100F TEM operated at an accelerating voltage of 200 kV. Semiquantitative elemental analysis of NC samples was obtained by a JEOL JSM 6500F field emission scanning electron microscope (FE-SEM) equipped with an EDAX Genesis energy dispersive spectroscopy detector. Thermogravimetric analysis (TGA) measurements were performed on a TA Instruments TGA 2950 thermogravimetric analyzer under nitrogen flow. The temperature of the NC samples (5−10 mg) was increased to 600 °C at a rate of 10 °C/min. Absorption spectra were obtained with an Agilent 8453 UV−vis spectrophotometer and a dual beam Cary 500 UV−vis−near-IR spectrophotometer on the NC dispersions. The core/shell NC dispersions were prepared in the glovebox and exposed to air just before spectra were recorded. FT-IR spectra were obtained using a Thermo Scientific Nicolet 380 FT-IR spectrometer on OLA- and sulfide-capped CZTS NC films, and spectra were normalized according to UV−vis absorption of the films. XPS spectra were obtained using a Physical Electronics ESCA 5800 system with monochromatic Al Kα (E = 1486.6 eV) X-ray source. Dry NC powders were placed on double-sided carbon tape adhered to the sample holder. The core/shell NC samples were prepared in a nitrogen glovebox and transferred to the sample chamber of the XPS instrument exposed to the air only for several seconds. High-resolution scans were performed with a pass energy of 23.5 eV and a step size of 0.10 eV/step. Data were processed using Multipak software, version 9.3.03. All spectra were shifted using inorganic carbon as a reference at 284.80 eV, and data fitting was performed assuming Gauss−Lorentz profiles with Shirley backgrounds. Thin Film Fabrication and Hall Measurements. The glass substrates were cleaned by sonication in soapy water, isopropanol, and DI water for 15 min, respectively. Substrates were stored in fresh DI water after final sonication and blow-dried with nitrogen before film deposition. Films were fabricated by a layer-by-layer approach using a mechanical dip coater. The substrates were first dipped in NC solution (2 mg/mL in hexane) and then in ammonium sulfide solution (1 mM in methanol), and finally in fresh acetonitrile solution to complete one cycle. This process was repeated to deposit films with ∼100 nm thickness measured by profilometry. The electrical properties of NC films were measured with an Ecopia 3000 Hall measurement system. Device Fabrication and Measurements. FTO glass substrates coated with n-type CdS layer (∼130 nm) grown by close space sublimation28 were used to deposit p-type NC films. The substrates were dipped in a NC solution (6.5 mg/mL), ammonium sulfide solution (1 mM in methanol), and acetonitrile as described before. After 50 cycles, gold electrodes (∼250 nm) were deposited on to the NC layer (∼250 nm) by thermal evaporation at a base pressure of 1 × 10−6 Torr through a shadow mask. The device area was 0.5 cm2. The current density−voltage (J−V) characteristics were recorded using a Keithley 2401 source meter under a solar simulator with standard test conditions (A.M. 1.5 Global spectrum with 100 mW cm−2 intensity) at 25 °C.
by overcoating the CZTS nanocrystals with a Cu2Se shell that can dynamically change conductivity upon exposure to air. It has been reported that the air oxidation of Cu+ cations in stoichiometric Cu2Se NCs generates copper vacancies in the valence band,18 resulting in an increase in the free carrier density, leading to a significant enhancement in electrical conductivity in Cu2Se NC films.19 The CZTS NCs described herein were coated with different numbers of monolayers of Cu2Se shells via the successive ionic layer adsorption and reaction (SILAR) method, and the effects of the Cu2Se shell on the structural and optical properties of the CZTS NCs were investigated. The variation in both dark current and photoconductivity as a function of Cu2Se shell thickness was determined by Hall measurements on NC films and current density−voltage (J−V) measurements on NC-based solar cells. Here we show that the conversion of stoichiometric Cu2Se shells on CZTS NCs to conductive Cu2−xSe shells due to air oxidation significantly improves the charge carrier hopping between CZTS NCs.
■
EXPERIMENTAL SECTION
Materials. Copper(II) acetylacetonate (99.99+%), Copper(I) chloride (≥99.995%), zinc acetate (99.99%), tin(II) chloride dihydrate (99.99%), sulfur powder, 1-octadecene (ODE) (90%), oleylamine (OLA) (70%), and ammonium sulfide (20 wt % in water) were purchased from Sigma-Aldrich. Se (99.99%) was purchased from Strem Chemicals. ACS grade chloroform, methanol, acetonitrile, ethanol, toluene, and hexane were obtained from Fisher Scientific. All chemicals were used as purchased. Synthesis of Cu2ZnSnS4 (CZTS) Nanocrystals. CZTS NCs were synthesized following the method published by Steinhagen et al. In a typical reaction, 0.52 g (2 mmol) of copper(II) acetylacetonate, 0.29 g (1.6 mmol) of zinc acetate, 0.18 g (0.8 mmol) of tin(II) chloride dihydrate, 0.13 g (4.0 mmol) of sulfur, and 40 mL of OLA were added to a 100 mL three-neck flask and degassed under vacuum for 2 h. The reaction mixture was then purged with N2 for 30 min at 110 °C and heated to 280 °C. The NCs were grown for 1 h and then cooled to room temperature. The product was isolated by precipitation with ethanol, followed by centrifugation. The precipitate was re-dispersed in chloroform and centrifuged again to remove unreacted precursors and large agglomerates. This procedure was repeated one more time, the final dispersion was precipitated with ethanol, and purified NCs were dried and stored in a nitrogen glovebox for further use. Synthesis of Cu2ZnSnS4/Cu2Se (CZTS/Cu2Se) Core/Shell NCs. A Cu2Se shell was grown on to the CZTS NCs by the successive ionic layer adsorption and reaction (SILAR) method. Amounts of copper and sulfur precursors required for the growth of each monolayer of shell (thickness of 0.3 nm) were calculated assuming spherical NCs. The average radius of CZTS NCs was determined from Scherrer analysis, and the weight percentage of organic ligands was estimated by TGA analysis (Supporting Information (SI) Figure S5). In a typical Cu2Se shell synthesis, 0.25 mmol of CZTS was mixed with 5 mL of OLA and 5 mL of ODE in a 100 mL three-neck flask, and the mixture was degassed under vacuum for 1 h and heated to 280 °C under N2 flow for shell growth. In order to prepare Cu and Se precursor solutions, CuCl (2 mmol) and Se (1 mmol) powder were added to 10 mL of OLA in separate three-neck flasks. The mixture of CuCl and OLA was heated to 120 °C under vacuum for 1 h, and the resultant clear CuCl solution was kept under N2 and heated to 150 °C prior to injection. The Se powder was dissolved in OLA by first heating the mixture to 150 °C under vacuum for 1 h and then increasing the temperature to 250 °C under N2 flow. After Se was fully dissolved, Cu and Se precursor solutions were injected dropwise and alternatively (e.g., 0.7 mL for growth of 1 monolayer) into the flask containing CZTS cores at 280 °C for the synthesis of the desired number of shell monolayers. After final injection, the reaction mixture was cooled to room temperature. The core/shell NCs were purified in
■
RESULTS AND DISCUSSION
CZTS NCs were synthesized following a literature procedure reported by Steinhagen et al.5 (see Experimental Section for more details). As-synthesized OLA-capped NCs were purified and thoroughly characterized in order to investigate the 4912
DOI: 10.1021/acsami.5b11037 ACS Appl. Mater. Interfaces 2016, 8, 4911−4917
Research Article
ACS Applied Materials & Interfaces
Figure 1. Low-resolution TEM images of (a) CZTS NCs, and CZTS/Cu2Se core/shell NCs with (b) 0.5, (c) 1, (d) 1.5, and (e) 2 monolayers. (f) High-resolution TEM image of a CZTS/Cu2Se core/shell NC with 2 monolayers.
conditions). Purified and dried CZTS NCs were dispersed in an OLA and octadecene (ODE) mixture, and the solution was heated to 280 °C. Simultaneously, Cu and Se precursor solutions were prepared by dissolving CuCl and Se in OLA at 150 and 250 °C, respectively. Each monolayer of the Cu2Se shell was grown by alternating injections of cation and anion precursor solutions. Panels b−e of Figure 1 show the TEM images of CZTS NCs overcoated with 0.5, 1, 1.5, and 2 monolayers of Cu2Se shell. XRD patterns of core/shell NCs (Figure 2) show peaks corresponding only to the kesterite CZTS structure. However, diffraction peaks showed an ∼0.1° shift toward lower angle between 1 and 1.5 monolayers indicative of replacement of some of the surface S atoms with Se20 (SI Figure S1). Additionally, according to the elemental analysis of core/shell NCs (SI Table S1), the Zn composition decreased notably between 1 and 1.5 monolayers, and this reduction could be due to the leaching of Zn to the solution as a consequence of cation exchange with Cu. Nevertheless, the Cu/Sn ratio increased with the addition of shell precursors, indicative of iteratively increasing the number of shell monolayers. Based on these observations, we speculate that core/shell NCs have a Cu-rich and Zn-deficient Cu2ZnSn(SxSe1−x)4 (CZTSSe) intermediate layer between the CZTS core and the Cu2Se shell at 1.5 and 2 monolayers. UV−vis−near-IR spectroscopy was used to explore the effect of structural modifications of CZTS NCs caused by the addition of Cu2Se shell monolayers on the optical properties of the particles. Figure 3 and SI Figure S2 show the UV−vis− near-IR absorption spectra of dispersions of core and core/shell NCs in toluene before and after exposure to air. The band gap of CZTS NCs is ∼1.35 eV as determined by extrapolating the linear portion of the band edge region of the plot of αhν2 versus hν (Figure 3b). This value is consistent with reported values of 1.45−1.6 eV.4 Note that the unambiguous selection of the linear section of the band edge region is not trivial due to lack
structure, morphology, and optical properties. Figure 1a shows a low-resolution TEM image of CZTS NCs, which have comparatively wide size distribution in the range of 5−20 nm. The kesterite structure was confirmed by the XRD pattern of CZTS NCs as shown in Figure 2. Scherrer analysis of the (2 2
Figure 2. PXRD patterns of CZTS NCs and CZTS/Cu2Se core/shell NCs with different shell thickness (0.5−2 monolayers). Black and red reference lines correspond to the kesterite CZTS structure (PDF no. 00-026-0575) and the cubic Cu1.8Se structure (PDF no. 01-088-2045). All of the peaks can be indexed to the kesterite CZTS structure or the cubic Cu1.8Se phase.
0) peak indicates that the crystallite size of the particles is ∼10.7 nm. Compositional analysis using SEM-EDS revealed that the as-synthesized CZTS NCs are Cu deficient and Zn rich with Cu/(Zn + Sn) ≈ 0.87 and Zn/Sn ≈ 1.15 (SI Table S1). The Cu2Se shell was grown on the CZTS NCs by the SILAR method where copper and sulfur precursor solutions were alternatively injected into the CZTS NC solution (see Experimental Section for a full description of the reaction 4913
DOI: 10.1021/acsami.5b11037 ACS Appl. Mater. Interfaces 2016, 8, 4911−4917
Research Article
ACS Applied Materials & Interfaces
Figure 3. (a) UV−vis−near-IR absorption spectra of CZTS NCs and CZTS/Cu2Se core/shell NCs (2 monolayers) dispersed in toluene. A LSPR peak evolved under ambient conditions due to the gradual oxidation of the Cu2Se shell, and peak intensity and maxima continuously increased and blue-shifted with time. Plots of αhν2 as a function of hν for (b) CZTS NCs and (c) CZTS/Cu2Se core/shell NCs (2 monolayers) show a decrease in the band gap of the NCs upon coating and oxidation of the shell.
Figure 4. (a) High-resolution XPS spectrum containing both S 2p and Se 3p core-level regions of CZTS/Cu2Se core/shell NCs (2 monolayers). Evolution of XPS spectra of (b) Cu 2p and (c) Se 3d core levels as a function of air exposure time. The gradual oxidation is evident by the broadening and shifting of Cu (+I) and Se (−II) peaks and emergence of new peaks due to oxidized species.
resonance (LSPR) that appears in the near-IR region of the absorption spectra.18 Similarly, when exposed to air, oxidation of the Cu2Se shell creates Cu vacancies that consequently increase the mobile carrier concentrations of the core/shell NCs. We did not observe a detectable LSPR peak in the UV− vis−near-IR absorption spectra of core/shell NCs up to 1 monolayer (SI Figure S2). However, upon addition of 1.5 and 2 shell monolayers, a distinct LSPR peak appears as a result of a significant increase of the free carrier density. Furthermore, the LSPR peak intensity slightly increased and the peak maxima slightly blue-shifted with time at 2 monolayers as a
of a distinct excitonic feature. The absorption onset red-shifted only after the addition of 1.5 shell monolayers, which is consistent with the peak shifts observed in the XRD patterns confirming the substitution of some of the core-surface S atoms with Se at that point as shown in SI Figure S2. The band gap of CZTS/Cu2Se NCs (2 monolayers) is determined to be ∼1.06 eV (Figure 3c). Oxidation of copper chalcogenide NCs under ambient conditions causes an increase of free carrier density (holes) due to formation of Cu vacancies.21 This self-doping process in copper chalcogenide NCs leads to a localized surface plasmon 4914
DOI: 10.1021/acsami.5b11037 ACS Appl. Mater. Interfaces 2016, 8, 4911−4917
Research Article
ACS Applied Materials & Interfaces
Carrier concentration (n) and conductivity (σ) of NC films (thickness ∼ 100 nm) were obtained by Hall-effect measurements at room temperature, and the results are summarized in Table 1. To the best of our knowledge, this is the first time that
consequence of continuous oxidation of the Cu2Se shell (Figure 3a).22 The progressive oxidation of the Cu2Se shell was further monitored by X-ray photoelectron spectroscopy (XPS). SI Figure S3 shows HR-XPS spectra of Cu, Zn Sn, and S for CZTS NCs and the peak location and the peak splitting values [for Cu 2p (2p3/2, 932.3 eV; 2p1/2, 952.05 eV), Zn 2p (2p3/2, 1022.1 eV; 2p1/2, 1045.1 eV), Sn 3d (3d5/2, 486.15 eV; 3d3/2, 494.6 eV), and S 2p (2p3/2, 161.6 eV; 2p1/2, 162.7 eV)] are consistent with the existence of Cu (+I), Zn (+II), Sn (+IV), and S (−II) oxidation states.4 As shown in Figure 4, after addition of 2 monolayers of a Cu2Se shell, Se 3p3/2 and 3p1/2 peaks appeared at 160.2 and 165.8 eV, respectively, and Se 3d5/2 peak was observed at 54.2 eV, suggesting the presence of a Se (−II) oxidation state.23 Air oxidation of the Cu2Se shell caused a broadening and shifting of the Cu 2p3/2 and 2p1/2 peaks along with appearance of satellite peaks due to formation of Cu2+ species as previously reported.19 Similarly, the oxidation of Se2− resulted in a broadening and shifting of the Se 3d5/2 peak accompanied by the emergence of a new peak between 58 and 59 eV, which can be assigned to a Se (+IV) oxidation state.24 We evaluated the intrinsic electrical properties of thin films fabricated from both CZTS NCs and CZTS/Cu2−xSe core/ shell NCs by Hall-effect measurements. The NC thin films were deposited on glass substrates by a layer-by-layer dip coating technique using a mechanical dip coater under ambient conditions (see Experimental Section for more details). Briefly, clean glass substrates were dipped into NC solutions to form a near-monolayer NC film and then dipped in to an ammonium sulfide solution to exchange insulating native ligands with S2− anions followed by rinsing in acetonitrile. This cycle was repeated until a desired film thickness was achieved. The ligand exchange step is not efficient, as shown in the CH stretching region in the FT-IR spectra of both OLA- and sulfide-capped CZTS NC films (SI Figure S4). Figure 5 shows the UV−vis− near-IR absorption spectra of NC films that were taken before Hall measurements. As in the case of the NC dispersions, a prominent LSPR was observed in the near-IR region for core/ shell NC films with 1.5 and 2 shell monolayers.
Table 1. Electrical Properties of NC Films electrical properties sample CZTS CZTS/Cu2−xSe CZTS/Cu2−xSe CZTS/Cu2−xSe CZTS/Cu2−xSe
(0.5 monolayer) (1 monolayer) (1.5 monolayers) (2 monolayers)
carrier concentration (cm−3) 6.3 1.7 8.1 3.8 5.7
× × × × ×
1015 1017 1017 1018 1019
conductivity (S cm−1) 1.6 7.7 1.4 2.0 1.6
× × × ×
10−3 10−3 10−2 10−1
the charge transport properties of thin films based on ligandexchanged CZTS nanocrystals have been studied without any type of annealing. The carrier concentration of sulfide-capped CZTS nanocrystal films (6.3 × 1015 cm3) is comparable with annealed CZTS NC films,25 and the conductivity is 1.6 × 10−3 S cm−1. The dark conductivity increased with the addition of shell monolayers along with the carrier concentration, and a significant, 2 orders of magnitude enhancement in conductivity (2.0 × 10−1 S cm−1) was observed after the addition of 1.5 shell monolayers. In fact, the conductivity of CZTS/Cu2‑xSe (1.5 monolayers) films is an order of magnitude higher than that of annealed CZTS NC films.26 Finally, we fabricated CZTS and CZTS/Cu2−xSe NC solar cells with a simple device architecture of FTO/CZTS/CdS/Au. The NC layers (∼250 nm) were deposited on CdS-coated FTO glass substrates following the layer-by-layer dip coating technique as described before (see Experimental Section for more details). Figure 6 shows current density−voltage curves of solar cells fabricated using CZTS and CZTS/Cu2−xSe (0.5−1.5 monolayers) NC films. The sulfide-capped CZTS NC solar cells showed an impressive open-circuit voltage (VOC, 589 mV) and a relatively good fill factor (FF, 41.8%) although a low short-circuit current density (JSC, 1 mA/cm2) resulted in an efficiency of only 0.25%. The photocurrent slightly increased with the addition of shell monolayers, but the VOC decreased as the number of shell monolayers increased, contributing to the decrease in cell efficiency after 0.5 monolayers. The core/shell NC solar cell with 0.5 monolayer of Cu2−xSe shell displayed the highest efficiency, 0.32%, which is a record for CZTS/CdS heterojunction NC solar cells fabricated without postannealing treatments. The positive effect of Cu2−xSe shells on the dark conductivity of NC thin films did not result in a corresponding increase in photoconductivity as expected. This is primarily due to two reasons: (1) inefficient ligand exchange and (2) cation exchange of core Zn2+ with Cu+ in the solution. As discussed previously, a significant portion of the highly insulating native ligand coverage is present on the NC surface after sulfide exchange hindering the hopping of photogenerated charge carriers between NCs even with a conductive shell. Currently, we are trying to understand the complicated surface chemistry of quaternary CZTS NCs and are in the process of developing efficient ligand exchange strategies. This could also enable achieving an appreciable efficiency with a minimum number of shell monolayers. However, the presence of the Cu-rich intermediate layer that is formed as a consequence of cation exchange between Zn2+ and Cu+ in the shell synthesis has
Figure 5. UV−vis−near-IR absorption spectra of thin films fabricated by CZTS NCs and CZTS/Cu2−xSe core/shell NCs with different shell monolayers. 4915
DOI: 10.1021/acsami.5b11037 ACS Appl. Mater. Interfaces 2016, 8, 4911−4917
Research Article
ACS Applied Materials & Interfaces
Figure 6. Current−voltage characteristics of (a) CZTS and (b−d) CZTS/Cu2−xSe NC solar cells under AM 1.5 illumination.
remained a major obstacle. The decrease in Zn content in the core while the Cu content increases with the addition of shell monolayers has an adverse effect on both VOC and JSC due to the formation of detrimental [2CuZn + SnZn] defect clusters that act as recombination centers.16,27 Presently, we are investigating methods including room temperature synthesis of a conductive shell (e.g., Cu2S) to eliminate or minimize cation exchange.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected].
■
Notes
CONCLUSIONS In summary, we present a novel method to enhance the charge transport in CZTS NC absorber layers by growing a stoichiometric Cu2Se shell on individual CZTS NCs, and, upon exposure to air, the shell becomes conductive via a selfdoping mechanism due to air oxidation. This conductive shell facilitates the charge carrier hopping between CZTS NCs as evident by the increase in dark conductivity as a function of the number of shell monolayers. The progressive increase in photoconductivity with the addition of shell monolayers was hampered by the significant changes to the CZTS core structure due to the cation exchange between Zn2+ and Cu+. However, addressing this issue could potentially increase the carrier collection efficiency in CZTS NC solar cells without costly postannealing treatments.
■
persions, high-resolution XPS spectra of core NCs, FTIR spectra of core NC films, and TGA curve of core NCs (PDF)
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was supported by the Research Corp. for Science Advancement through a Scialog Award. We sincerely acknowledge Kevan C. Cameron at CSU Next Generation PV Center (NGPV) for providing substrates for solar cell fabrication and Professor James Sites (Department of Physics, Colorado State University) for helpful discussions about solar cell measurements. We also thank Dr. Roy Geiss (Central Instrument Facility, Chemistry Department, Colorado State University) for help with the high-resolution TEM imaging.
■
REFERENCES
(1) Walsh, A.; Chen, S.; Wei, S.-H.; Gong, X.-G. Kesterite Thin-Film Solar Cells: Advances in Materials Modelling of Cu2ZnSnS4. Adv. Energy Mater. 2012, 2, 400−409. (2) Shin, B.; Gunawan, O.; Zhu, Y.; Bojarczuk, N. A.; Chey, S. J.; Guha, S. Thin Film Solar Cell with 8.4% Power Conversion Efficiency Using an Earth-Abundant Cu2ZnSnS4 Absorber. Prog. Photovoltaics 2013, 21, 72−76. (3) Wang, W.; Winkler, M. T.; Gunawan, O.; Gokmen, T.; Todorov, T. K.; Zhu, Y.; Mitzi, D. B. Device Characteristics of CZTSSe ThinFilm Solar Cells with 12.6% Efficiency. Adv. Energy Mater. 2014, 4, 1301465.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.5b11037. Table containing EDS elemental analysis data of core and core/shell NCs and figures showing enlarged XRD patterns of core and core/shell NCs, UV−vis−near-IR absorption spectra of core and core/shell NC dis4916
DOI: 10.1021/acsami.5b11037 ACS Appl. Mater. Interfaces 2016, 8, 4911−4917
Research Article
ACS Applied Materials & Interfaces (4) Riha, S. C.; Parkinson, B. A.; Prieto, A. L. Solution-Based Synthesis and Characterization of Cu2ZnSnS4 Nanocrystals. J. Am. Chem. Soc. 2009, 131, 12054−12055. (5) Steinhagen, C.; Panthani, M. G.; Akhavan, V.; Goodfellow, B.; Koo, B.; Korgel, B. A. Synthesis of Cu2ZnSnS4 Nanocrystals for Use in Low-Cost Photovoltaics. J. Am. Chem. Soc. 2009, 131, 12554−12555. (6) Zhou, H.; Hsu, W.-C.; Duan, H.-S.; Bob, B.; Yang, W.; Song, T.B.; Hsu, C.-J.; Yang, Y. CZTS Nanocrystals: a Promising Approach for Next Generation Thin Film Photovoltaics. Energy Environ. Sci. 2013, 6, 2822−2838. (7) van Embden, J.; Chesman, A. S. R.; Della Gaspera, E.; 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. (8) Hsu, W.-C.; Zhou, H.; Luo, S.; Song, T.-B.; Hsieh, Y.-T.; Duan, H.-S.; Ye, S.; Yang, W.; Hsu, C.-J.; Jiang, C.; Bob, B.; Yang, Y. Spatial Element Distribution Control in a Fully Solution-Processed Nanocrystals-Based 8.6% Cu2ZnSn(S,Se)4 Device. ACS Nano 2014, 8, 9164−9172. (9) 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 2015, 23, 654− 659. (10) Zhang, J.; Gao, J.; Church, C. P.; Miller, E. M.; Luther, J. M.; Klimov, V. I.; Beard, M. C. PbSe Quantum Dot Solar Cells with More than 6% Efficiency Fabricated in Ambient Atmosphere. Nano Lett. 2014, 14, 6010−6015. (11) Fischer, A.; Rollny, L.; Pan, J.; Carey, G. H.; Thon, S. M.; Hoogland, S.; Voznyy, O.; Zhitomirsky, D.; Kim, J. Y.; Bakr, O. M.; Sargent, E. H. Directly Deposited Quantum Dot Solids Using a Colloidally Stable Nanoparticle Ink. Adv. Mater. 2013, 25, 5742−5749. (12) Zarghami, M. H.; Liu, Y.; Gibbs, M.; Gebremichael, E.; Webster, C.; Law, M. p-Type PbSe and PbS Quantum Dot Solids Prepared with Short-Chain Acids and Diacids. ACS Nano 2010, 4, 2475−2485. (13) Oh, S. J.; Berry, N. E.; Choi, J.-H.; Gaulding, E. A.; Lin, H.; Paik, T.; Diroll, B. T.; Muramoto, S.; Murray, C. B.; Kagan, C. R. Designing High-Performance PbS and PbSe Nanocrystal Electronic Devices through Stepwise, Post-Synthesis, Colloidal Atomic Layer Deposition. Nano Lett. 2014, 14, 1559−1566. (14) Tang, J.; Kemp, K. W.; Hoogland, S.; Jeong, K. S.; Liu, H.; Levina, L.; Furukawa, M.; Wang, X.; Debnath, R.; Cha, D.; Chou, K. W.; Fischer, A.; Amassian, A.; Asbury, J. B.; Sargent, E. H. ColloidalQuantum-Dot Photovoltaics Using Atomic-Ligand Passivation. Nat. Mater. 2011, 10, 765−771. (15) Chuang, C.-H. M.; Brown, P. R.; Bulović, V.; Bawendi, M. G. Improved Performance and Stability in Quantum Dot Solar Cells Through Band Alignment Engineering. Nat. Mater. 2014, 13, 796− 801. (16) Chen, S.; Walsh, A.; Gong, X.-G.; Wei, S.-H. Classification of Lattice Defects in the Kesterite Cu2ZnSnS4 and Cu2ZnSnSe4 EarthAbundant Solar Cell Absorbers. Adv. Mater. 2013, 25, 1522−1539. (17) Suehiro, S.; Horita, K.; Kumamoto, K.; Yuasa, M.; Tanaka, T.; Fujita, K.; Shimanoe, K.; Kida, T. Solution-Processed Cu2ZnSnS4 Nanocrystal Solar Cells: Efficient Stripping of Surface Insulating Layers Using Alkylating Agents. J. Phys. Chem. C 2014, 118, 804−810. (18) Kriegel, I.; Jiang, C.; Rodríguez-Fernández, J.; Schaller, R. D.; Talapin, D. V.; da Como, E.; Feldmann, J. Tuning the Excitonic and Plasmonic Properties of Copper Chalcogenide Nanocrystals. J. Am. Chem. Soc. 2012, 134, 1583−1590. (19) Riha, S. C.; Johnson, D. C.; Prieto, A. L. Cu2Se Nanoparticles with Tunable Electronic Properties Due to a Controlled Solid-State Phase Transition Driven by Copper Oxidation and Cationic Conduction. J. Am. Chem. Soc. 2011, 133, 1383−1390. (20) Riha, S. C.; Parkinson, B. A.; Prieto, A. L. Compositionally Tunable Cu2ZnSn(S1−xSex)4 Nanocrystals: Probing the Effect of SeInclusion in Mixed Chalcogenide Thin Films. J. Am. Chem. Soc. 2011, 133, 15272−15275. (21) Comin, A.; Manna, L. New Materials for Tunable Plasmonic Colloidal Nanocrystals. Chem. Soc. Rev. 2014, 43, 3957−3975.
(22) Dorfs, D.; Härtling, T.; Miszta, K.; Bigall, N. C.; Kim, M. R.; Genovese, A.; Falqui, A.; Povia, M.; Manna, L. Reversible Tunability of the Near-Infrared Valence Band Plasmon Resonance in Cu2−xSe Nanocrystals. J. Am. Chem. Soc. 2011, 133, 11175−11180. (23) Ji, S.; Shi, T.; Qiu, X.; Zhang, J.; Xu, G.; Chen, C.; Jiang, Z.; Ye, C. A Route to Phase Controllable Cu2ZnSn(S1‑xSex)4 Nanocrystals with Tunable Energy Bands. Sci. Rep. 2013, 3, 2733. (24) Méar, F. O.; Essi, M.; Sistat, P.; Guimon, M. F.; Gonbeau, D.; Pradel, A. Structural Characterization of a Cu(II) Thin-Film Aging in a Cu-Nitrate Solution. Appl. Surf. Sci. 2009, 255, 6607−6611. (25) Li, M.; Zhou, W.-H.; Guo, J.; Zhou, Y.-L.; Hou, Z.-L.; Jiao, J.; Zhou, Z.-J.; Du, Z.-L.; Wu, S.-X. Synthesis of Pure Metastable Wurtzite CZTS Nanocrystals by Facile One-Pot Method. J. Phys. Chem. C 2012, 116, 26507−26516. (26) Mainz, R.; Singh, A.; Levcenko, S.; Klaus, M.; Genzel, C.; Ryan, K. M.; Unold, T. Phase-Transition-Driven Growth of Compound Semiconductor Crystals from Ordered Metastable nanorods. Nat. Commun. 2014, 5, 3133. (27) Chen, S.; Wang, L.-W.; Walsh, A.; Gong, X. G.; Wei, S.-H. Abundance of CuZn+SnZn and 2CuZn+SnZn Defect Clusters in Kesterite Solar Cells. Appl. Phys. Lett. 2012, 101, 223901. (28) Kobyakov, P. S.; Moore, A.; Raguse, J. M.; Swanson, D. E.; Sampath, W. S. Deposition and Characterization of Cd1−xMgxTe Thin Films Grown by a Novel Cosublimation Method. J. Vac. Sci. Technol., A 2014, 32, 021511.
4917
DOI: 10.1021/acsami.5b11037 ACS Appl. Mater. Interfaces 2016, 8, 4911−4917
