Facile synthesis of composition and morphology modulated quaternary CuZnFeS colloidal nanocrystals for photovoltaic application

Amit Daluia, Ali Hossain Khana, Bapi Pradhana, Jayita Pradhana, Biswarup Satpatib and Somobrata Acharya*a
aCentre for Advanced Materials, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700032, India. E-mail: camsa2@iacs.res.in
bSurface Physics and Material Science Division, Saha Institute of Nuclear Physics, 1/AF Bidhannagar, Kolkata 700064, India

Received 6th September 2015 , Accepted 4th November 2015

First published on 9th November 2015


Abstract

Quaternary semiconductor CuZnFeS (CZIS) nanocrystals (NCs) with controlled size, shape and composition composed of earth abundant elements have been successfully synthesized using the colloidal synthesis method. The size, shape and composition of the NCs have been controlled by tuning the reaction parameters to obtain NCs in the form of dots, triangles, hexagons, sheets, rods and wires. These quaternary CZIS NCs show high light absorbing properties towards visible to near infrared light with a high absorption coefficient suitable for photovoltaic applications. Utilizing layer-by-layer deposition of CIZS NCs films, heterojunction devices consisting of ITO/PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS/CZIS NCs/Al are fabricated for photovoltaic performance. The devices exhibit excellent rectification behavior (rectification ratio of ∼150) and good photoresponsitivity (on/off ratio of ∼55). The broad range of absorptions with strong extinction coefficient properties of CZIS NCs has been utilized to fabricate quantum dot sensitized solar cells (QDSSCs). Our synthesis protocol marks an advance for chalcopyrite NCs based solar cells and offers a possible template for the synthesis of other ternary and quaternary NCs with robust photoelectric properties.


Introduction

Solution processed semiconductor NCs consisting of abundant and accessible elements in the earth's crust are of central interest for photosensitive applications.1,2 Binary, ternary or quaternary NCs such as FeS2, Cu2S and copper zinc tin sulfide (CZTS) have shown potential for photovoltaic applications owing to their band alignment, low material requirements and cost.2 Multinary copper chalcogenide semiconductors (I–III–VI2, I2–II–IV–VI4) are promising alternatives impacting various technologies such as solar harvesting, catalysis and biomedicines.3–15 The optimum band gap located in the solar spectrum (0.9–2.4 eV), high radiation stability, low toxicity and high optical absorption coefficient (105 cm−1) of multinary copper chalcogenide NCs make them suitable for a variety of applications.3–7 Quaternary I–II–III–VI2 NCs are emerging as technologically viable material due to the composition tunable band gap over a wide range (1.1–2.4 eV) and high thermal stability.5–7 Colloidal dispersion of these NCs is attractive for thin film device fabrication using wet manufacturing processes such as spin-coating, doctor-blending, roll-to-roll printing or inkjet printing.16–20 Since the size and shape of the NCs have significant effect on the optoelectronic properties, it is important to control the size and shape distribution of NCs to create printable colloidal solutions (inks)19 for photovoltaic applications. In addition, the control of the size, shape and composition of NCs is expected to tune optoelectronic properties in a selective way which can be explored in low-cost, solution-processed photovoltaic applications.21–29

Replacing expensive and not-environmental friendly components of solar cells with green and abundant materials is an important scientific challenge.30–32 Various carbonaceous materials, polymers and composites were explored for energy harvesting applications. However, natural materials obtained directly from surroundings are abundant, accessible and environmentally friendly. Several naturally derived materials with unique structures have been developed for advanced applications in energy harvesting, storage and photovoltaic.30–32 Among various types of solar cells, the CuInxGa1−xS(Se)2 (CIGS) thin-film solar cells are of great interest for its high power conversion efficiency and stability.11,19 However, the low availability of indium or gallium increases the production costs and creates difficulties for the development of CIGS thin-film solar cells. Cu2ZnSnS4 (CZTS) NCs, which is structurally similar to CIGS, has drawn much attention because of its earth-abundant composition. The abundance of In in the upper continental crust is estimated to be 0.05 ppm compared with abundances of 25, 71 and 5.5 ppm for copper, zinc and tin, respectively.33–35 Hence, intense research is carried out to find a suitable substitute for CIGS or CZTS that avoids using costly elements.11,19,34,35 Unlike binary compounds, the synthesis of ternary and quaternary materials involve reactions between multiple precursors with different reactivities, therefore, suffer from inherently poor control over NC nucleation and growth.36 Shapes and structures can be strongly influenced by various factors, such as the nature of reactant and solvents, reactant concentration, the reaction temperature and time.37 This inevitably leads to broad variations in the size, morphology and composition of the NCs.36,37

Here, we report on the colloidal synthesis of a new chalcopyrite phase CuZnFeS NCs (CZIS NCs) composed of more earth abundant and environment friendly elements. Size, shape and composition of the NCs have been successfully controlled by tuning the reaction parameters. Utilizing the benefit of the ligands and precursor used in the synthesis, we have carried out systematic study to obtain different shaped CZIS NCs in the form of dots, triangles, hexagons, sheets, rods and wires. These NCs can be well dispersed in non-polar solvents and simply create thin film by direct spin coating method. Devices fabricated using these NCs show excellent rectification behavior (rectification ratio of ∼150) and good photo-switching property (on/off ratio of ∼55). CZIS NCs are also found to be useful for fabrication of quantum dot sensitized solar cells (QDSSCs) due to broad range of absorption with strong extinction coefficient. Our rational synthesis route may open up the further possibilities for in-depth study of the other multi-component alloyed NCs with excellent photoelectric properties with an ease.

Experimental section

Materials

Zinc(II) chloride (anhy., 99.9%, Aldrich), cuprous chloride (Cu2Cl2, 97%, Loba chemie), iron(III) acetylacetonate (97%, Aldrich), oleylamine (OLA, tech., 70%, Aldrich), sulfur (99.9%, Aldrich), 1-octadecene (ODE, tech., 90%, Aldrich), oleic acid (OA, tech., 90%, Aldrich), n-tetradecylphosphonic acid (TDPA, 97%, Aldrich), 1-dodedcanethiol (1-DDT, >98%, Aldrich), 3-mercaptopropanoic acid (MPA), PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS solution (Aldrich), titanium tetrachloride (TiCl4, Sigma), titanium tetraisoproxide (TTIP), anatase TiO2 nanoparticles (size-20 nm, Sigma), acetic acid, α-terpeneol, ethyl cellulose. Indium tin oxide (ITO) and fluorine doped tin oxide (FTO) substrates were purchased from Aldrich.

Synthesis procedure of CZIS NCs

Here we have studied the size, shape and composition controlled synthesis of CZIS NCs. In a typical synthesis of CZIS NCs,38 0.1 mmol copper(I) chloride, 0.1 mmol zinc(II) chloride, 0.1 mmol iron(III) acetylacetonate, 0.5 ml oleylamine (OLA), 0.5 ml oleic acid (OA) and 6 ml of 1-octadecene (ODE) were degassed with nitrogen at 120 °C for 30 minutes to get rid of moisture and oxygen. Then it was slowly heated to 200 °C. During this time the solution color changes from dark green to light yellow. At this temperature, 0.6 mmol of sulfur dissolved in 3 ml ODE (S-ODE) was swiftly injected. The reaction was annealed at 200 °C for a fixed time and then allowed to cool to room temperature. Upon injection the sulfur precursor, the reaction solution immediately turned into magenta black indicating the rapid nucleation of the CZIS NCs and finally turns into black upon annealing. The as-prepared NCs were purified by several precipitation–dispersion cycles using chloroform and ethanol. Then it was dried in air to get black powdered solid. Different size CZIS NCs were synthesized under same reaction condition but changing the reaction temperature in the range of 130°–250 °C. Composition tunable CZIS NCs were synthesized by varying the amount of zinc and iron precursor but retaining the copper amount same. Morphology controlled CZIS NCs have been synthesized at 200 °C by proper choice of capping ligand and sulfur precursors. Rod shaped CZIS NCs was prepared by either injecting or one pot non injecting synthesis method using 1-DDT as sulfur source instead of S-ODE. Whereas hexagonal disk shaped NCs was prepared by replacing capping ligand oleic acid with TDPA and 1-DDT as sulfur source. Wire shaped CZIS NCs was obtained by introducing TOPO as capping ligand and 1-DDT as sulfur source.

Ligand exchange procedure

As synthesized oleylamine (OLA) and oleic acid (OA) capped CZIS NCs are exchanged with 3-mercaptopropanoic acid (MPA) to get M-CZIS NCs.39 In a typical procedure, 50 mg of OA and OLA capped CZIS NCs were dissolved in 3 ml chloroform. Then aqueous solution of MPA (pH of the solution adjusted to greater than ten by adding sodium hydroxide) is added drop-wise under vigorous stirring. After 30 minutes a separation of two distinct layers of chloroform and NCs were observed. The NCs were separated out by centrifugation and washed two times with chloroform to remove un-exchanged OA and OLA. The NCs were further dispersed in water and precipitated out with acetone to remove excess MPA. Then the resultant NCs were used for further application.

Photoconductivity measurements

A thin film of MPA capped CZIS (M-CZIS) NCs has been fabricated by spin-casting of NCs ink on ITO coated glass substrate containing PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS film. ITO coated glass substrates (Aldrich, 15–25 Ω square per inches) were first cleaned with detergent, ultrasonicated in acetone and isopropyl alcohol and subsequently dried overnight in an oven. A thin film of PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS film was deposited on top of the ITO by spin casting the PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS solution at 5000 rpm for 60 seconds. Then it was heated at 150 °C for 1 hour. After cool down, as synthesized NCs solution (30 mg ml−1) in toluene is spin coated (2000 rpm) on top of that film. Then capping ligands are exchanged using methanolic solution of MPA (5%). Then wash with methanol to remove excess MPA and followed by washing with toluene. Next layer of NCs is deposited using the same protocol and 10–12 layers of NCs were deposited. Finally, deposited film was heated at 130 °C for 1 hour. The aluminum electrode, which serves as back electrode, has been fabricated by vacuum deposition technique. Current (I)–voltage (V) characteristics of the devices were measured using a Keithley-2420 source meter. Photo IV measurements were done by illuminating with xenon arc light source from Newport at energy 100 mW cm−2.

Quantum dots sensitized solar cell

TiO2 electrode preparation. Fluorine-doped tin oxide (FTO) glasses were cleaned with detergent solution followed by sonication in milli-Q water, acetone and isopropanol sequentially for 20 minutes each. Then it was dried in hot air oven. First, a thin TiO2 layer was deposited on FTO glass by treating the FTO glass in 40 mM TiCl4 solution at 70 °C for 30 minutes. Then a compact layer of TTIP in slightly acidic iso-propanol was spin casted and dried at 120 °C for 1 hour. TiO2 paste was then deposited over it by doctor blade technique. The TiO2 paste was prepared according to the method reported in the literature.40 The TiO2 electrode was heated at 450 °C for 30 minutes. The scattering layer of TiO2 was deposited on top of the active layer by treating the TiO2 film with 40 mM TiCl4 solution at 70 °C for 30 minutes. Then it was heated again at 450 °C for 30 minutes.
Fabrication of solar cells. CZIS NCs sensitized solar cell is fabricated according to the previously reported method.41 CZIS NCs was immobilized on TiO2 film by depositing the M-CZIS NCs aqueous dispersion onto the TiO2 film and staying for drying the solvent. After deposition, the M-CZIS NCs sensitized TiO2 films were coated with ZnS for two cycles by alternate dipping into the methanolic solution of 0.2 M Zn(OAc)2·2H2O and 0.2 M Na2S solution. The device was fabricated by assembling the brass-based Cu2S counter electrode and M-CZIS NCs sensitized TiO2 electrode using a 50 μm thick scotch spacer with a binder clip. Polysulfide aqueous solution is used as electrolyte, consisting of 2.0 M Na2S, 2.0 M S and 0.2 M KCl.

Characterization

Transmission electron microscopy (TEM) and high resolution TEM (HRTEM) investigations were carried out using a JEOL JEM-2010 microscope operating at 200 kV. The sample was prepared by drop casting the NCs from chloroform/ethanol solution onto a lacey carbon-coated gold grid (300 meshes) prior to measurements. Elemental analysis was carried out using X-ray photo-electron spectroscopy (XPS) measurements with an Omicron X-ray photoelectron spectrometer. High-angle annular dark field scanning transmission electron microscopy (STEM-HAADF) measurements were carried out using a FEI, TF30-ST microscope operating at 300 kV equipped with a scanning unit and an HAADF detector from Fischione (model 3000). The TF30 is also equipped for electron energy loss spectroscopy with a post-column Gatan Quantum SE (model 963). Compositional analysis was performed by energy dispersive X-ray spectroscopy (EDX) attached to the TF30. Powder X-ray diffraction (XRD) patterns were recorded with a Bruker D8 Advanced diffractometer using Cu-kα radiation (λ = 1.5405 Å). UV-vis-NIR absorption spectra were recorded with a Varian-cary-5000 spectrophotometer. Photoluminescence (PL) spectra were recorded using a Jobin Yvon nanolog spectrophotometer by dispersing the NCs in tetrachloroethylene/water. Current (I)–voltage (V) characteristics of the devices were measured using a Keithley-2420 source meter.

Results and discussion

We have adopted facile colloidal synthesis method to prepare quaternary CZIS NCs. Synthesis methodology involves taking cationic precursor, capping agent and high boiling point solvent together to form homogeneous mixture and injection of sulfur precursor at high temperature while annealing at a particular temperature to grow the NCs. We have used 1-DDT as one of the sulfur precursor to synthesize NCs in one-pot reaction methods by taking all metal and DDT together and heated to high temperature to form CZIS NCs (details are given in the Experimental Section). This one pot synthesis method can yield NCs in gram scale level at a time indicating the applicability of this method for the large-scale production of quaternary semiconducting CZIS NCs. As synthesized NCs can be dispersed homogeneously in a variety of organic solvent like; chloroform, toluene, hexane or chloro-benzene to form NCs ink predicting its prospect in the field of thin film device preparation (Fig. 1a, reaction schematic). TEM images (Fig. 1b) reveal that CZIS NCs have faceted spherical morphology with diameter around ∼12 ± 3 nm. Additionally, triangular shaped NCs with larger sizes are also evidenced. High resolution TEM images show the well resolved inter-planar lattice spacing corresponding to the chalcopyrite crystal phase as in the case of CuFeS2 (Fig. S1, ESI).42
image file: c5ra18157g-f1.tif
Fig. 1 (a) Schematic presentation of the reaction involved in the synthesis of CZIS NCs. As synthesized NCs can be dissolve in chloroform to produce CZIS NCs ink. (b) Low-resolution TEM images of CZIS NCs capped with oleic acid and oleylamine. (c) STEM-HAADF image of CZIS NCs and (d–g) chemical map of the constituent elements Cu, Zn, Fe and S respectively which are obtained using STEM-HAADF technique.

Chemical mapping of the NCs using STEM-HAADF technique reveals the homogeneous distribution of the Cu, Zn, Fe and S all-over the NCs (Fig. 1c–g).43 Energy dispersive X-ray (EDX) spectroscopy of the NCs taken from STEM-HAADF technique suggest the elemental composition of Cu[thin space (1/6-em)]:[thin space (1/6-em)]Zn[thin space (1/6-em)]:[thin space (1/6-em)]Fe[thin space (1/6-em)]:[thin space (1/6-em)]S = 26[thin space (1/6-em)]:[thin space (1/6-em)]5[thin space (1/6-em)]:[thin space (1/6-em)]26[thin space (1/6-em)]:[thin space (1/6-em)]43 within the CZIS NCs (Fig. S2, ESI). FTIR spectrums suggest that NCs surface are capped mostly with oleic acid compared to oleylamine as carbonyl stretching vibration appears with higher intensities and lower in energy in the FTIR spectrum of the NCs (Fig. S3, ESI). This is also evidenced from the X-ray photoelectron spectroscopy (XPS) measurements where the peak intensity corresponds to C–O binding energy dominates over C–N peak intensity (Fig. S4, ESI). Inspection of the spectrum also reveals the presence of carbon, nitrogen and oxygen from the ligands and trace amounts of chlorine from the reaction precursors. High resolution XPS spectrum suggests an elemental composition of Cu+, Zn2+, Fe3+ and S2− within the CZIS NCs.44,45 Spectral region corresponds to specific element Cu, Zn, Fe and S are shown in Fig. 2a–d. Copper shows doublet 2p peak of 2p1/2 and 2p3/2 at binding energy 952.1 eV and 932.5 eV respectively with a peak splitting of 19.6 eV indicating the presence of monovalent Cu (Fig. 2a). The doublet peaks (2p1/2 and 2p3/2) appears at binding energy of 1044.9 eV and 1022.3 eV with a peak separation of 22.6 eV confirms the presence of divalent Zn (Fig. 2b). The two peaks centered at 724.4 eV and 710.7 eV with a peak separation of 13.7 eV can be attributed to presence of trivalent Fe (Fig. 2c). The peaks at 162.9 eV and 161.9 eV are consistent with those observed for sulfur in metal sulfides (Fig. 2d).44,45


image file: c5ra18157g-f2.tif
Fig. 2 High resolution XPS spectrums of the constituent elements present within CZIS NCs (a) Cu 2p, (b) Zn 2p, (c) Fe 2p, (d) S 2p respectively.

Influence of reaction temperature

To investigate the influence of reaction temperature on the formation of CZIS NCs, we have synthesized CZIS NCs at different reaction temperature of 130 °C to 250 °C keeping the other reaction parameter identical. Size and shape evolution of the NCs at different reaction temperature have been examined using TEM (Fig. 3). The NCs are mostly triangular in shape with an average size of ∼7 nm at a reaction temperature of 150 °C (Fig. 3a). Triangular NCs with average size of ∼11 nm along with some faceted spherical NCs are obtained at 200 °C (Fig. 3b). Larger triangular NCs with an average size of ∼23 nm along with few pyramidal NCs are observed at reaction temperature of 250 °C (Fig. 3c).
image file: c5ra18157g-f3.tif
Fig. 3 (a–c) Low resolution TEM images of different sized CZIS NCs obtained at variation of reaction temperature 150 °C, 200 °C and 250 °C. (d–f) High resolution TEM images of the different sized CZIS NCs where interplanar distance corresponds to the chalcopyrite crystals phase. (g–i) Selected area electron diffraction pattern of the CZIS NCs showing the diffraction pattern corresponds to (112), (312) and (204) planes of chalcopyrite crystals phase.

HRTEM images (Fig. 3d–f) show well-defined lattice planes for all the NCs. Interplanar distances of 0.34 ± 0.02 nm corresponds to (112) planes and 0.184 ± 0.02 nm corresponds to (312) planes of chalcopyrite crystallographic phase were observed for all three different sizes of the NCs (Fig. 3d–f and S1, ESI).42 Selected area electron diffraction (SAED) of NCs obtained at three different reaction temperatures show diffraction spots corresponding to (112), (204), (312) planes of the chalcopyrite crystals structure (Fig. 3g–i). Elemental compositions of the as synthesized NCs at different reaction temperature were investigated by using EDX spectrum attached with scanning electron microscope (SEM). EDX spectrums have been taken from the thin film of the sample which shows the presence of constituent elements Cu, Zn, Fe and S within the NCs (Table S1, ESI).

Powder X-ray diffraction (XRD) was used to characterize the crystal structure of the NCs. The XRD patterns of the as synthesized CZIS NCs show (Fig. 4a) diffraction peaks corresponding to (112), (103), (220), (204), (312), (116) planes of chalcopyrite crystals structure (JCPDS #370471).42 The (112) peak become sharper and intense with the increase of reaction temperature from 130 °C to 250 °C owing to the increase in NC size and crystallinity at higher reaction temperature. Notably, the intensity of the X-ray diffraction peaks matches with the SAED patterns (Fig. 3g–i). In addition to the major diffraction peaks, additional XRD peaks are observed at reaction temperature below 200 °C. The diffraction peak at 2θ = 27.8° may index to the (122) peak of Cu1.81S (JCPDS#410959). At lower reaction temperature reaction rate is relatively slow. As a result of which it may appear that greater percentage copper sulfide portion remain within the CZIS NCs which contributed the diffraction peak at 2θ = 27.8°. At higher reaction temperature reactivity of the different cationic component increases forming homogeneous chalcopyrite CZIS NCs (JCPDS#370471). During the reaction at different temperature, the yellow solution at the initial stage gradually changed to magenta black color, which finally turned to black color. When reaction temperature was set at 250 °C, the initial solution changed color to black within 10 seconds. However, the reaction turned into black color within 5 minutes, when the reaction temperature is low (150 °C) indicating that growth rate of the NCs depends on the reaction temperature. In addition, we anticipate from the XRD measurements (Fig. 4a) that NCs at lower temperature are more faceted compared to at higher reaction temperature. CZIS NCs show absorption features in the UV-vis-NIR region (Fig. 4b). The absorption maxima is found to be shifted from 490 nm to 550 nm with the change in the reaction temperature from 150 °C to 250 °C indicating the increase of NCs size at higher reaction temperature. We have calculated the band gaps of NCs from Tauc plot by plotting (αhυ)2 versus hυ (α = molar absorbance coefficient, h = Planck's constant and υ = frequency of light) and extrapolating the linear portion of the spectrum in the band edge region (Fig. S5, ESI). The band gap threshold is calculated to be 0.8 eV for the NCs synthesized at 200 °C. CZIS NCs have high molar absorption coefficient of ∼104 in the photon energy range of 1.5 to 4 eV (Fig. 4c). Although, the CZIS NCs show broad absorption feature in the UV-vis-NIR range, however, photoluminescence was not detectable from the NCs. Hence, CZIS NCs with broad range absorption with high molar extinction coefficient may be found useful for photocatalysis and photovoltaics applications.


image file: c5ra18157g-f4.tif
Fig. 4 (a) Powder X-ray diffraction pattern of the CZIS NCs synthesized at different reaction temperature (T = 130 °C to 250 °C). XRD pattern have been compared with the standard JCPDS data file (JCPDS #370471) of the chalcopyrite crystals as in the case of CuFeS2 NCs. (b) UV-vis-NIR absorption spectrums of the CZIS NCs obtained at different reaction temperature. (c) Plot of molar absorption coefficient with respect to photon energy.

Influence of reaction time

We have investigated the effect of annealing time at a fixed temperature during the growth of CZIS NCs. The evolutions of the absorption spectrums of the NCs with different reaction times are monitored by collecting aliquot during the annealing at 200 °C and 250 °C (Fig. S6, ESI). The absorption spectra were measured using NCs dispersion in tetrachloroethylene. CZIS NCs exhibit broad absorption feature with a peak in the visible region. The absorption edge of NCs gradually shifts toward longer wavelengths with the reaction annealing time. At reaction time of 2 hours the visible range absorption feature become flat indicating that the longer annealing time leads to broad size distribution of the CZIS NCs. This is also evidenced from the time-dependent morphology evolution of the NCs at 250 °C using TEM (Fig. S7, ESI). At the beginning of the reaction in 20 seconds, just after the injection of sulfur precursor, the dominant products are triangular NCs with average sizes of ∼7 nm. At a reaction time of 2 minutes, triangular NCs increase in size to ∼15 nm. With further increase of the reaction annealing time to 20 minutes, bigger triangular NCs with average sizes of ∼30 nm were observed. Elemental compositions of the NCs obtained at different reaction temperature have been investigated using ICP-AES (Table S2, ESI). All NCs show elemental ratios of the constituent elements suggesting homogeneous alloy formation of the as synthesized CZIS NCs.

Influence of elemental composition

To verify the effect of elemental composition on the NCs morphology, we have prepared composition tunable CZIS NCs with various iron and zinc ratios keeping the copper content constant. We have termed different NCs with various composition as CZS NCs (CuZnS, 0 mmol Fe + 0.2 mmol Zn), CZIS (CuZnFeS) NCs with different iron and zinc content as CZIS-1 (0.025 mmol Fe + 0.175 mmol Zn), CZIS-2 (0.05 mmol Fe + 0.15 mmol Zn), CZIS-3 (0.075 mmol Fe + 0.125 mmol Zn) and CZIS-4 (0.09 mmol Fe + 0.11 mmol Zn) respectively. The NCs without Zn is termed as CIS (CuFeS2, 0.2 mmol Fe + 0 mmol Zn). All the CZIS NCs exhibit XRD pattern corresponding to (112), (220), (204), (312) and (116) planes of chalcopyrite crystal structure (Fig. 5a). The major peaks are shifted to lower angles for CZS and CZIS NCs in comparison to the CIS NCs. The changes of lattice constant due to composition variation can be obtained from analysis of XRD patterns. For tetragonal unit cell, lattice parameter ‘a’ and ‘c’ can be calculated using following equations:
 
image file: c5ra18157g-t1.tif(1)
 
image file: c5ra18157g-t2.tif(2)

image file: c5ra18157g-f5.tif
Fig. 5 (a) Composition dependent XRD patterns of NCs which are obtained at reaction temperature 200 °C by varying the amount of copper, zinc and iron precursor and S-ODE was used as sulfur precursor. Concentration of the copper (0.1 mmol) kept fixed for all cases. CZS NCs (CuZnS, 0 mmol Fe + 0.2 mmol Zn), CZIS (CuZnFeS) NCs with different iron and zinc content CZIS-1 (0.025 mmol Fe + 0.175 mmol Zn), CZIS-2 (0.05 mmol Fe + 0.15 mmol Zn), CZIS-3 (0.075 mmol Fe + 0.125 mmol Zn), CZIS-4 (0.09 mmol Fe + 0.11 mmol Zn), CIS (CuFeS2, 0.2 mmol Fe + 0 mmol Zn). (b) Table showing the ratios of the different constituent elements taken for the reaction and the ratios of the different constituent elements present within the NCs. The amounts of the different constituent elements are obtained from the ICP-AES analysis of the NCs. (c) TEM image of copper sulfide, (d) TEM image of copper–zinc–sulfide (CZS) NCs and (e) TEM images of copper–iron–sulfide (CIS) NCs respectively.

We found that the lattice constant ‘c’ changes from 9.67 Å to 10.29 Å while replacing Fe by Zn to convert CuFeS2 (c = 9.67 Å) into CuZnS NCs (c = 10.29 Å) (Fig. S8, ESI). The cations Cu+, Fe3+ and Zn2+ have ionic radii of 91 pm, 69 pm and 88 pm respectively.46 Hence, the replacement of smaller sized Fe3+ with larger sized Zn2+ results in shifting of diffraction peaks towards lower angles and subsequently increases the lattice constant. We have probed the elemental composition within the resultant NCs by ICP-AES measurements and compared with the value with the amount of precursors taken for reaction (Fig. 5b). The analyses suggest that the ratio of elemental composition follows the same trend in both the cases. Importantly, we can design binary, ternary or quaternary NCs by proper choice of the precursor using the same reaction parameters demonstrating the versatility of our synthetic protocol. Fig. 5c–e shows the TEM images of the as-synthesized copper sulfide, copper–zinc– sulfide and copper–iron–sulfide NCs by proper control of precursors. As-synthesized NCs are of 6 nm, 8 nm and 11 nm in sizes for copper sulfide, copper–zinc–sulfide and copper–iron–sulfide respectively (Fig. 5c–e). These observations suggest that our synthetic protocol of designing quaternary, ternary or binary NCs is facile with precise control over elemental composition.

Influence of capping ligand and sulfur source

We have investigated the effect of capping ligand, cationic precursor and sulfur precursor on the formation of CZIS NCs at a reaction temperature of 200 °C. CZIS NCs have been synthesized by varying the capping ligands and sulfur source and cationic precursors. When oleic acid (OA) and oleylamine (OLA) mixture are used as capping ligand and S-ODE is used as sulfur source, CZIS NCs are mostly triangular in shape with average size of 12 nm (Fig. 6a). By changing the capping ligands with TDPA and TOPO while retaining S-ODE as sulfur source, triangular shaped NCs with average size of 18 nm was obtained (Fig. 6b and c). However, replacement of S-ODE by DDT while using oleic acid and oleylamine as capping ligands, CZIS nanorods (NRs) was obtained (Fig. 6d). The TEM image reveals that the NRs are of 25–30 nm in lengths and ∼6 nm in width. When DDT was retained as sulfur source and oleic acid and oleylamine is replaced by TDPA, hexagonal shaped NCs with average size of 35–40 nm were obtained (Fig. 6e). Earlier report on CuInxGa1−xS2 synthesis includes phosphonic acid, which is reported to be a very strong passivating ligand for {002} facets allowing the growth only along the {220} facets.47 Hexagonal disks shaped NCs have been reported in the presence of phosphonic acids in CuInS2, where the phosphonic acids changes in the reaction environments by lowering the monomer activity and the chemical potential of the reaction.48 This ultimately results in the formation of thermodynamically more stable faceted crystals, which crystallizes in hexagonal wurtzite phase. When TOPO was used as capping agent and DDT as sulfur source, supercrystalline CZIS nanowires with diameter of 1.5 nm and lengths of 50–80 nm was obtained. Elemental composition of nanowires obtained from ICP-AES analysis shows that nanowires contain 90% copper content whereas iron and zinc contents are remaining within 10% (Fig. 6f and S9, ESI). In case of Cu2ZnSnS4 (CZTS) NCs it is reported that highly reactive elemental S yields fast formation of kesterite CZTS phase, while DDT results in a slow and gradual formation of wurtzite CZTS phase.49 Li et al. suggested that H2S generated from the reduction of S by ODE was the real S precursor in the formation of CdS nanocrystals and in comparison to saturated alkanes, octadecene was found to be substantially more active as a reductant for elemental sulphur.50 In our reaction system, injection of S-ODE precursor, the colour of solution mixture turned dark immediately. Thus, H2S produced by S-ODE precursor reacts rapidly with Cu-, Zn-, Fe-OLA and OA complex, which leads to fast formation of CZIS NCs. On the other side, the thiol group in DDT provides S for the formation of CZIS NCs where S is covalently bonded to one C and one H atom, making this a rather stable molecule compared to H2S.49 In presence of DDT, the reaction mixture turned gradually from yellow to black. This reaction undergoes a thiolate formation stage that involves the metal ions which decompose into corresponding sulfides at certain temperature.51
image file: c5ra18157g-f6.tif
Fig. 6 (a–c) TEM images of the CZIS NCs obtained using S-ODE as sulfur source but changing the capping ligand OA and OLA, TDPA and TOPO. (d–f) TEM images of the CZIS NCs obtained using 1-DDT as sulfur source and OA, TDPA and TOPO as capping ligand.

Also, due to the strong coordination with the metal cations exposing on the surface of nanocrystal, dodecanethiol along with the coordinating ligand may assists the formation different shaped CZIS NCs.51 When only DDT was use as capping agent as well as sulfur source, CZIS NCs contains very less iron content and a high copper content. In this reaction system, blue colored precipitate of iron thiolate complex is major product and lesser amount CZIS NCs are observed during purification process of the product. Ryan et al. have systematically studied the shape development of CuInxGa1−xS2 NCs using TOPO, DDT and amine.47 They have shown the influence of variable chain length organic amines on controlling the morphology and aspect ratio variation of NRs. The CuInxGa1−xS2 NCs formation get starts with formation of spherical copper sulfide NCs at 210 °C and then insertion of In at 230 °C to form CuInS and finally incorporation of Ga at 270 °C it form quaternary CuInxGa1−xS2 NCs.47 However, in contrary of their result, for our system, we have obtained a homogeneous quaternary CZIS NCs from the starting of the reaction (just in 30 seconds) indicating a different reaction pathways and reactivity of the precursor towards the formation of CZIS NCs. When TDPA and oleylamine were used as capping agent but S-ODE act as sulfur source, mixture of two different sizes NCs were observed. TEM images contains larger sized NCs with size of 20 nm and smaller NCs of 2–3 nm (Fig. S10, ESI). When oleic acid was used as capping agent and 1-DDT as sulfur source, CZIS nanosheet was obtained with micrometer lengths. Thin nanosheet was trend to stack together (Fig. S11, ESI). We have also investigated the formation of CZIS NCs using TMS as sulfur source and TDPA as capping agent. TEM images show that as synthesized NCs have mostly spherical morphology with diameter around ∼10 nm (Fig. S12, ESI).

Photo-responsive properties

To test the applicability of CZIS NCs for photoelectric applications, we fabricated thin film device consisting of ITO/PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS/M-CZIS NCs/Al. The PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS film was spin-coated at 5000 rpm for 1 minute onto a pre-cleaned ITO/glass substrate, which was then annealed at 150 °C for 1 hour. Then, CZIS NCs film was deposited on top of the PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS film using spin coating method. CZIS NCs were deposited by layer-by-layer fashion. Ligand exchange protocol with MPA was applied after each layer of deposition (details procedure is given in the experimental section). Following CZIS NCs deposition, a ∼150 nm Al layer was deposited using thermal evaporation method. The cross-sectional scanning electron microscope (SEM) image of the device (Fig. S13, ESI) used for the measurement reveals the different active components present within the device. Fig. 7a outlines the flat-band energy level diagram corresponding to the each component of the device. Valence band (VB) and conduction band (CB) energy levels of M-CZIS NCs have been extracted using scanning electron microscopy and ultra-violet photoelectron spectroscopy.38,52 M-CZIS NCs have VB energy level at −5.05 eV and CB at −4.0 eV which is suitably positioned with respect to the energy levels of the PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS and ITO facilitating the transfer of photo generated holes to the ITO (Fig. 7a). On the other side energy level of Al is positioned below the CB of CZIS NCs indicating the favorable electron transfer pathways from CB of CZIS NCs to the Al. Fig. 7b shows the IV curves of a ITO/PEDOT[thin space (1/6-em)]:[thin space (1/6-em)]PSS/M-CZIS NCs/Al device measured in the dark and under 100 mW cm−2 AM 1.5G illumination. Interestingly, CZIS NCs based device exhibits a strong photoconductive effect, where the observed current at +2 V (28.9 mA cm−2) under AM 1.5 G illumination is much larger than in the dark (0.14 mA cm−2). This increase in current is likely due to the photo generated electron–hole pair in the depleted M-CZIS NCs layer. We observed that the M-CZIS NCs heterojunction device exhibits excellent rectification behavior with a rectification ratio of ∼150. Similar rectification behavior was also observed for other earth-abundant nanocrystalline semiconductor devices such as thin film FeS2 NCs-based photodiodes.53,54 The photoresponse of the M-CZIS NCs heterojunction device at zero bias voltage under alternating cycles of illumination and dark condition is shown in Fig. 7c. Upon switching on the illumination, the current rises sharply and decreases rapidly when the illumination is turned off. The photo current switching is repeatable over multiple on-off switching cycles. The average on-off switching response time was ∼5 s, which is faster compared to thin film FeS2 NCs photodiodes.53 We further investigated the wavelength-dependent spectral response of the device from the incident photon-to-current efficiency (IPCE) action spectrum measurement. The IPCE spectrum (Fig. S14, ESI) of the device exhibits a broad photoresponse range from 350 to 750 nm wavelength similar to the UV-vis absorption spectrum of the CZIS NCs.
image file: c5ra18157g-f7.tif
Fig. 7 (a) Schematic representation of the energy levels of various layers involved in the layered heterojunction photovoltaic device (b) current (I)–voltage (V) characteristics of the CZIS NCs device with irradiation (red line) and in dark (blue line) in the bias voltage range −2 V to +2 V. Rectification ratio of the device was calculated from the current ratio at 2 V and −2 V in the light. Inset: schematic illustration of the device used in the measurement. (c) On/off switching of the device at zero bias voltage. (d) Current–voltage characteristic of the QDSSCs using M-CZIS NCs as sensitizer where black curve represent the dark current and red curve indicates the IV under light illumination. Inset showing the schematic representation of the device where energy levels59 of the different component and possible carrier transfer pathways have mentioned.

Quantum dot sensitized solar cells

The major requirement for QDSSCs is broad range of absorption with strong absorption coefficient, efficient separation of photo-excited electrons and holes due to the proper alignment of energy level with respect to the TiO2.55 CZIS NCs exhibit broad absorption in the UV-vis-NIR region with strong absorption coefficient which meets the criteria for QDSSCs fabrication. We have fabricated device using CZIS NCs and ligand exchanged with MPA prior to the loading on to the TiO2 matrix. This ex-situ post-synthesis ligand exchange approach has been found to be effective for high loading NCs on a mesoporous TiO2 film.41,56 The UV-vis absorption, XRD and TEM of MPA-capped CZIS NCs shows no change in comparison to the as synthesized NCs suggesting that the optical and structural properties of the CZIS NCs remains the same after ligand exchange protocol (Fig. S15, ESI). The M-CZIS NCs were then tethered on TiO2 electrodes by dropping the CZIS NCs dispersion in water on the film.41 ZnS passivation layer was then deposited onto the film by dipping it into the Zn2+ and S2− aqueous solutions sequentially following the literature method.57 Sandwich-type cells were constructed by assembling a M-CZIS-sensitized TiO2 film electrode and Cu2S counter electrode (brass) using binder clips, with filling a drop of polysulfide electrolyte in between two electrodes.58 Cross-sectional SEM image of the photoanode (Fig. S16, ESI) sensitized with M-CZIS NCs shows essential components required for the QDSSCs. Fig. 7d, inset describes the energy band alignment of the component involve in the CZIS NCs sensitized solar cell. MPA-capped CZIS NCs have conduction band energy level at −4.0 eV which is higher than TiO2 (−4.2 eV) and FTO (−4.4 eV).59 So, photoexcited electron of M-CZIS NCs can easily transfer to photoanode and holes will transfer to sulfide/polysulfide electrolyte to get neutralized, generating photo current to the external circuit. Photoexcited electrons are injected into a large band gap semiconductor such as TiO2, while photoexcited holes are collected by a redox couple.59–61 IV characteristic of the device shows open circuit voltage (Voc) around 0.3 V and short circuit current of 5.5 mA cm−2 upon illumination of light (100 mW cm−2) whereas without light no such characteristic was observed (Fig. 7d). These results indicate that CZIS NCs can be used as sensitizer in QDSSCs and the device performance can be improved by proper adjustment of electrolyte and counter electrodes.

Conclusions

In summary, we have reported on the successful synthesis of CZIS NCs using earth abundant elements in a variety of shapes. UV-vis absorption spectra revealed that CZIS NCs have broad range of absorption with high extinction coefficient indicating its potential application as a solar light absorber. We have shown the ability to vary the size and shapes of the NCs using the same reaction protocol. Composition of elements within the NCs can be engineered precisely by controlling the amount of the precursor components. Key factors for the formation of various shaped CZIS NCs is capping ligand as well as sulfur source. Photoelectric measurement showed that CZIS NCs possess rectification behavior and good photo-response properties over multiple cycles. We have also shown that CZIS NCs can be used as sensitizer for QDSSCs. The synthesis strategy developed in this work may be used as a general process for the shape controlled synthesis of other multinary chalcogenide alloyed NCs and may have great potential for high efficiency, yet low cost photosensitive devices.

Acknowledgements

The authors acknowledge SERB, DST India for financial support. B. Pradhan and J. Pradhan acknowledge CSIR and DST-INSPIRE research fellowship of India.

References

  1. D. B. Mitzi, Solution Processing of Inorganic Materials, Wiley-Interscience, Hoboken, NJ, 2009 Search PubMed.
  2. C. Wadia, A. P. Alivisatos and D. M. Kammen, Environ. Sci. Technol., 2009, 43, 2072–2077 CrossRef CAS PubMed.
  3. T. Unold and H. W. Schock, Annu. Rev. Mater. Res., 2011, 41, 297–321 CrossRef CAS.
  4. C. Persson, J. Appl. Phys., 2010, 107, 053710–053718 CrossRef.
  5. K. Ramasamy, M. A. Malik, N. Revaprasadu and P. O'Brien, Chem. Mater., 2013, 25, 3551–3569 CrossRef CAS.
  6. D. Aldakov, A. Lefrançois and P. Reiss, J. Mater. Chem. C, 2013, 1, 3756–3776 RSC.
  7. F.-J. Fan, L. Wu and S.-H. Yu, Energy Environ. Sci., 2014, 7, 190–208 CAS.
  8. A. Singh, S. Singh, S. Levcenko, T. Unold, F. Laffir and K. M. Ryan, Angew. Chem., Int. Ed., 2013, 52, 9120–9124 CrossRef CAS PubMed.
  9. T. H. Nguyen, W. Septina, S. Fujikawa, F. Jiang, T. Harada and S. Ikeda, RSC Adv., 2015, 5, 77565–77571 RSC.
  10. C. J. Stolle, T. B. Harvey, D. R. Pernik, J. I. Hibbert, J. Du, D. J. Rhee, V. A. Akhavan, R. D. Schaller and B. A. Korgel, J. Phys. Chem. Lett., 2014, 5, 304–309 CrossRef CAS PubMed.
  11. Q. Guo, G. M. Ford, H. W. Hillhouse and R. Agrawal, Nano Lett., 2009, 9, 3060–3065 CrossRef CAS PubMed.
  12. B.-L. Wu, H.-J. Chao, C.-P. Chen, C.-H. Yang and J.-Y. Chang, RSC Adv., 2015, 5, 36605–36613 RSC.
  13. M. Ibánez, R. Zamani, A. LaLonde, D. Cadavid, W. Li, A. Shavel, J. Arbiol, J. R. Morante, S. Gorsse, G. J. Snyder and A. Cabot, J. Am. Chem. Soc., 2012, 134, 4060–4063 CrossRef PubMed.
  14. S. N. Guin and K. Biswas, Chem. Mater., 2013, 25, 3225–3231 CrossRef CAS.
  15. P. Sonström, M. Adam, X. Wang, M. Wilhelm, G. Grathwohl and M. Bäumer, J. Phys. Chem. C, 2010, 114, 14224–14232 Search PubMed.
  16. J. Perelaer and U. S. Schubert, J. Mater. Res., 2013, 28, 564–573 CrossRef CAS.
  17. F. Padinger, C. J. Brabec, T. Fromherz, J. C. Hummelen and N. S. Sariciftci, Opto-Electron. Rev., 2000, 8, 280–283 CAS.
  18. S. Srivastava and N. A. Kotov, Acc. Chem. Res., 2008, 41, 1831–1841 CrossRef CAS PubMed.
  19. M. G. Panthani, V. Akhavan, B. Goodfellow, J. P. Schmidtke, L. Dunn, A. Dodabalapur, P. F. Barbara and B. A. Korgel, J. Am. Chem. Soc., 2008, 130, 16770–16777 CrossRef CAS PubMed.
  20. Q. Guo, S. J. Kim, M. Kar, W. N. Shafarman, R. W. Birkmire, E. A. Stach, R. Agrawal and H. W. Hillhouse, Nano Lett., 2008, 8, 2982–2987 CrossRef CAS PubMed.
  21. X. S. Fang, T. Y. Zhai, U. K. Gautam, L. A. Li, L. M. Wu, B. Yoshio and D. Golberg, Prog. Mater. Sci., 2011, 56, 175–287 CrossRef CAS.
  22. X. J. Xu, L. F. Hu, N. Gao, S. X. Liu, S. Wageh, A. A. Al-Ghamdi, A. Alshahrie and X. S. Fang, Adv. Funct. Mater., 2015, 25, 445–454 CrossRef CAS.
  23. N. Pradhan, S. Acharya, K. Ariga, N. S. Karan, D. D. Sarma, Y. Wada, S. Efrima and Y. Golan, J. Am. Chem. Soc., 2010, 132, 1212–1213 CrossRef CAS PubMed.
  24. W. D. Xiang, X. Ma, L. Luo, W. Cai, C. P. Xie and X. J. Liang, Mater. Chem. Phys., 2015, 149, 437–444 CrossRef.
  25. R. Xie, M. Rutherford and X. Peng, J. Am. Chem. Soc., 2009, 131, 5691–5697 CrossRef CAS PubMed.
  26. Y. H. Xie, C. Y. Zhang, F. Yue, Y. Zhang, Y. T. Shi and T. L. Ma, RSC Adv., 2013, 3, 23264–23268 RSC.
  27. M. Shtein, R. Nadiv, M. Buzaglo, K. Kahil and O. Regev, Chem. Mater., 2015, 27, 2100–2106 CrossRef CAS.
  28. V. N. Kumar, R. Suryakarthick, S. Karuppusamy, M. Gupta, Y. Hayakawa and R. Gopalakrishnan, RSC Adv., 2015, 5, 23015–23021 RSC.
  29. Y. Xie, L. Carbone, C. Nobile, V. Grillo, S. D'Agostino, F. D. Sala, C. Giannini, D. Altamura, C. Oelsner, C. Kryschi and P. D. Cozzoli, ACS Nano, 2013, 7, 7352–7369 CrossRef CAS PubMed.
  30. J. C. Lopez, G. Quijano, R. Perez and R. Munoz, Chem. Eng. Trans., 2014, 40, 211–216 Search PubMed.
  31. C.-L. Wang, J. Y. Liao, S.-H. Chung and A. Manthiram, Adv. Energy Mater., 2015, 5, 1401524–1401527 Search PubMed.
  32. M. Imperiyka, A. Ahmad, S. A. Hanifah and F. Bella, Phys. B, 2014, 450, 151–154 CrossRef CAS.
  33. D. B. Mitzi, O. Gunawan, T. K. Todorov and D. A. R. Barkhouse, Philos. Trans. R. Soc., A, 2013, 371, 1–22 CrossRef PubMed.
  34. Q. Guo, H. W. Hillhouse and R. Agrawal, J. Am. Chem. Soc., 2009, 131, 11672–11673 CrossRef CAS PubMed.
  35. Z. F. Tong, Z. H. Su, F. Y. Liu, L. X. Jiang, Y. Q. Lai, J. Li and Y. X. Liu, Mater. Lett., 2014, 121, 241–243 CrossRef CAS.
  36. M. J. Thompson, T. P. A. Ruberu, K. J. Blakeney, K. V. Torres, P. S. Dilsaver and J. Vela, J. Phys. Chem. Lett., 2013, 4, 3918–3923 CrossRef CAS.
  37. X. Zhang, G. Guo, C. Ji, K. Huang, C. Zha, Y. Wang, L. Shen, A. Gupta and N. Bao, Sci. Rep., 2014, 4, 5086–5093 CAS.
  38. A. Dalui, U. Thupakula, A. H. Khan, T. Ghosh, B. Satpati and S. Acharya, Small, 2015, 11, 1829–1839 CrossRef CAS PubMed.
  39. Z. Pan, K. Zhao, J. Wang, H. Zhang, Y. Feng and X. Zhong, ACS Nano, 2013, 7, 5215–5222 CrossRef CAS PubMed.
  40. S. Ito, P. Chen, P. Comte, M. K. Nazeeruddin, P. Liska, P. Péchy and M. Grätzel, Prog. Photovoltaics, 2007, 15, 603–612 CAS.
  41. Z. Pan, I. Mora-Seró, Q. Shen, H. Zhang, Y. Li, K. Zhao, J. Wang, X. Zhong and J. Bisquert, J. Am. Chem. Soc., 2014, 136, 9203–9210 CrossRef CAS PubMed.
  42. D. Liang, R. Ma, S. Jiao, G. Pang and S. Feng, Nanoscale, 2012, 4, 6265–6268 RSC.
  43. R. Kore, R. Srivastava and B. Satpati, ACS Catal., 2013, 3, 2891–2904 CrossRef CAS.
  44. L. Ai and J. Jiang, J. Mater. Chem., 2012, 22, 20586–20592 RSC.
  45. M. D. Regulacio, C. Ye, S. H. Lim, M. Bosman, E. Ye, S. Chen, Q.-H. Xu and M.-Y. Han, Chem.–Eur. J., 2012, 18, 3127–3131 CrossRef CAS PubMed.
  46. R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751–767 CrossRef.
  47. C. Coughlan, A. Singh and K. M. Ryan, Chem. Mater., 2013, 25, 653–661 CrossRef CAS.
  48. M. Kruszynska, H. Borchert, J. Parisi and J. Kolny-Olesiak, J. Am. Chem. Soc., 2010, 132, 15976–15986 CrossRef CAS PubMed.
  49. Z. Li, A. L. K. Lui, K. H. Lam, L. Xi and Y. M. Lam, Inorg. Chem., 2014, 53, 10874–10880 CrossRef CAS PubMed.
  50. Z. Li, Y. Ji, R. Xie, S. Y. Grisham and X. Peng, J. Am. Chem. Soc., 2011, 133, 17248–17256 CrossRef CAS PubMed.
  51. X. Lu, Z. Zhuang, Q. Peng and Y. Li, Chem. Commun., 2011, 47, 3141–3143 RSC.
  52. A. Dalui, B. Pradhan, U. Thupakula, A. H. Khan, G. S. Kumar, T. Ghosh, B. Satpati and S. Acharya, Nanoscale, 2015, 7, 9062–9074 RSC.
  53. D.-Y. Wang, Y.-T. Jiang, C.-C. Lin, S.-S. Li, Y.-T. Wang, C.-C. Chen and C.-W. Chen, Adv. Mater., 2012, 24, 3415–3420 CrossRef CAS PubMed.
  54. E. J. Luber, M. H. Mobarok and J. M. Buriak, ACS Nano, 2013, 7, 8136–8146 CrossRef CAS PubMed.
  55. J. Huang, B. Xu, C. Yuan, H. Chen, J. Sun, L. Sun and H. Ågren, ACS Appl. Mater. Interfaces, 2014, 6, 18808–18815 CAS.
  56. F. Huang, D. Chen, Y. Chen, R. A. Caruso and Y.-B. Cheng, J. Mater. Chem. C, 2014, 2, 1284–1289 RSC.
  57. N. Guijarro, J. M. Campiña, Q. Shen, T. Toyoda, T. Lana-Villarreal and R. Gómez, Phys. Chem. Chem. Phys., 2011, 13, 12024–12032 RSC.
  58. V. Chakrapani, D. Baker and P. V. Kamat, J. Am. Chem. Soc., 2011, 133, 9607–9615 CrossRef CAS PubMed.
  59. L.-H. Lai, L. Protesescu, M. V. Kovalenko and M. A. Loi, Phys. Chem. Chem. Phys., 2014, 16, 736–742 RSC.
  60. I. Mora-Seró, S. Giménez, F. Fabregat-Santiago, R. Gómez, Q. Shen, T. Toyoda and J. Bisquert, Acc. Chem. Res., 2009, 42, 1848–1857 CrossRef PubMed.
  61. H. J. Yun, T. Paik, M. E. Edley, J. B. Baxter and C. B. Murray, ACS Appl. Mater. Interfaces, 2014, 6, 3721–3728 CAS.

Footnote

Electronic supplementary information (ESI) available: ESI file contains figures related to materials characterizations and controlled experiments. See DOI: 10.1039/c5ra18157g

This journal is © The Royal Society of Chemistry 2015
Click here to see how this site uses Cookies. View our privacy policy here.