Preparation and photoelectric property of a Cu2FeSnS4 nanowire array

Liang Shi* and Yanan Li
Department of Chemistry, University of Science and Technology of China, Hefei 230026, P. R. China. E-mail: sliang@ustc.edu.cn; Fax: +86-551-3607402; Tel: +86-551-3607234

Received 18th June 2014 , Accepted 9th September 2014

First published on 9th September 2014


Abstract

A single crystalline Cu2FeSnS4 nanowires array has been prepared via a convenient solution approach. Porous anodic aluminum oxide was used as a morphology directing template and played a significant role in the formation of single crystalline Cu2FeSnS4 nanowires. The as-prepared Cu2FeSnS4 nanowires are uniform with a [110] growth direction. Structure, morphology, composition and optical absorption properties of the as-prepared samples were characterized with X-ray powder diffraction, transmission electron microscopy, energy dispersive X-ray spectrometry, scanning electron microscopy and UV-Vis spectrophotometry. The formation mechanism of Cu2FeSnS4 nanowires array has been discussed. Thin films prepared from Cu2FeSnS4 single crystalline nanowires displayed an obvious photoelectric response, suggesting their potential application as low cost solar absorber materials.


1. Introduction

Cu2FeSnS4 is an important copper-based quaternary chalcopyrite semiconductor and has attracted increasing attention in recent years because it is one of the promising low cost alternatives to conventional solar cell materials. Cu2FeSnS4 has a high absorption coefficient and its band gap (around 1.5 eV) matches the desired absorption range of the solar spectrum.1,2 It has been theoretically demonstrated that the presence of tetrahedrally coordinated copper atoms is a critical feature for the exhibition of good photovoltaic properties of chalcogenide absorbers.3 Among various copper-based chalcopyrite solar cell absorber materials, Cu(In,Ga)Se2 (CIGS) has been currently most successful because of its high power conversion efficiency (20.3%) and good stability.4 However, the limited supply and increasing price of rare metals, including In and Ga, have triggered a need to find alternative solar materials with high abundance and low cost. As one of possible substitutes to CIGS, Cu2FeSnS4 has a special advantage due to its containing naturally abundant elements Fe and Zn as well as its relatively low toxicity. This enables the large-scale commercial application of toward creating low cost and sustainable thin film solar cells. Over the past several years, many studies have been performed on nanostructured chalcopyrite materials because the shape and size of these nanoscale semiconductors may exert a significant influence on their optoelectronic function and device performance, and sometimes induce unique physical and chemical properties different from their bulk counterparts. Some novel features have been found in the nanoscale semiconductor systems, such as an obvious enhancement of solar energy conversion efficiencies of photovoltaic devices covered with nanocrystalline-based semiconductor absorber layers.5,6 Recently, a few research investigations on nanoscale Cu2FeSnS4 have been conducted. Cu2FeSnS4 nanoparticles, microspheres and porous nanotubes composed of nanoparticles have been prepared successfully by various methods,1,2,7–10 including solvothermal synthesis, microwave nonaqueous strategy and hot-injection approach. However, to the best of our knowledge, preparation of single crystalline Cu2FeSnS4 nanowires in a well-aligned array configuration has not been reported. Such chalcopyrite semiconductors in the nanowire morphology can offer a well-defined nanoscale domain with clearly identifiable grain boundaries, where an energy barrier could better avoid the charge carrier recombination.11,12 Furthermore, continuous charge carrier transport pathways without dead ends may exist in the well-aligned nanowires. These favorable characteristics might lead to increase in conversion efficiency in nanowires photovoltaic devices. In spit of these excellent features, it is usually difficult to control the nucleation and growth of Cu2FeSnS4 for fabrication of well-aligned nanostructures in part because quaternary Cu2FeSnS4 exhibits very complicated phase diagrams and nanocrystals show greater phase complexity than their corresponding bulk materials. To date, it remains still a challenge for synthesis of single crystalline Cu2FeSnS4 nanowires array. The template method assisted by porous anodic aluminum oxide (AAO) is one of the most efficient methods for the fabrication of highly ordered well-distributed one-dimensional nanostructures, especially for polynary compound nanowires.13–15 AAO has many advantages including monodisperse size distribution, high pore density, nearly parallel porous structures and easily controlled pore diameter.16,17 Most importantly, AAO templates are thermally and mechanically stable; hence they can be employed in rigorous reaction conditions. Here we report a convenient solution approach with AAO as a morphology directing template for preparation of array of ordered single crystalline Cu2FeSnS4 nanowires. In addition, a film made of as-synthesized Cu2FeSnS4 nanowires was found to have obvious photo-electric response, indicating the potential use of them in solar energy conversion systems, such as the fabrication of photovoltaic devices.

2. Experimental section

All reagents are analytical grade and used without further purification. Anhydrous Cu(I)Cl, anhydrous SnCl2, FeCl2·4H2O, elemental sulfur and anhydrous ethylenediamine (En) were used as starting materials. AAO templates (Whatman Co., U.K.) with pore sizes of 200 nm in diameter were used in the experiments. In a typical procedure, for the fabrication of single-crystalline Cu2FeSnS4 nanowires array within the AAO template, 1.5 mmol anhydrous CuCl, 0.75 mmol anhydrous SnCl2, 0.75 mmol FeCl2·4H2O and 3 mmol elemental sulfur were added to a 25 mL flask in air. Then 16 mL anhydrous ethylenediamine were added into the above flask with mild magnetic stirring. Afterward, an AAO template was added into the flask and immersed in the liquid. The above liquid mixture in the flask was treated by sonication for 5 min to remove air in the pore of AAO template and fill the pore with liquid. The flask was attached to a Schlenk line and purged of oxygen and water by pulling vacuum for 10 min, followed by nitrogen bubbling for 10 min with mild magnetic stirring. The evacuation and N2 bubbling process was cycled for 3 times at room temperature. The sonication and multi-evacuation and bubbling process was important to enhance the filling efficiency of the AAO template pores with mixture of reaction solution. As a result, a uniform and high density of nanowires sample can be obtained. The above solution in the flask was then transferred into a 20 mL stainless steel teflon-lined autoclave. The autoclave was sealed and the temperature was maintained at 200 °C for 40 hours before being cooled down to room temperature. The AAO template containing the product was taken out, thoroughly washed with ethanol and distilled water, air-dried for characterization.

The overall crystallinity of the product is examined by X-ray diffraction (XRD, Rigakau RU-300 with CuKα radiation). The general morphology of the products was characterized using scanning electron microscopy (FESEM QF400). Detailed microstructure analysis was carried out using transmission electron microscopy (TEM Tecnai 20ST). The chemical composition and the spatial distribution of the compositional elements in the product were examined using an energy dispersive X-ray (EDX) spectrometer and Gatan image filtering (GIF) system, attached to the same microscope. For the SEM measurements, several drops of 1 M NaOH aqueous solution were added onto the sample to dissolve some part of the AAO template. The residual solution on the surface of the template was rinsed with distilled water. For the TEM and HRTEM measurements, the template was completely dissolved in 2 M NaOH aqueous solution. The product was centrifuged, thoroughly washed with distilled water to remove residual NaOH and then rinsed with absolute ethanol. The UV-Vis spectrum of the product was recorded in a UV-Vis spectrophotometer (UV-1601PC, Shimadzu Corporation). IV curves were measured by linear sweep voltammetry on CH Instrument 660C electrochemical analyzer. A photoresponse device structure was fabricated to study the optoelectronic properties of the as-synthesized nanowires. The fabrication details of the device structure are as follows: a insulating quartz substrate was used as the substrate and thoroughly washed with a mixed solution of deionized water, acetone, and 2-propanol under sonication for 30 min. RF sputtering of Au electrodes were then completed on insulating quartz substrates. The Cu2FeSnS4 film was fabricated by dripping the concentrated nanowires methanol dispersion on the electrodes and the substrate, and then drying it for 5 hours at room temperature under ambient conditions. A post-anneal process was conducted at 300 °C in an Ar atmosphere to improve substrate adhesion.

3. Results and discussion

The product was firstly characterized by X-ray diffraction (XRD) to obtain information on crystal structure and phase composition. A typical XRD pattern of the as-prepared Cu2FeSnS4 product is shown in Fig. 1. All diffraction peaks can be indexed to the stannite structured Cu2FeSnS4 in the tetragonal space group I[4 with combining macron]2m. After refinement, the lattice constant, a = 5.450 Å and c = 10.741 Å, is obtained, which matches well to the reported value for Cu2FeSnS4 crystal (JCPDS card, no. 44-1476). The broadening of the XRD peaks suggests that the grain sizes of the product are on a nanometer scale. No other phases are found in the product.
image file: c4ra05878j-f1.tif
Fig. 1 A representative XRD pattern of Cu2FeSnS4 product.

Fig. 2 shows a SEM image of the as-prepared sample with the AAO template partially removed. A highly ordered array of well-distributed nanowires are grown in a large area. These nanowires are continuous, smooth and parallelly aligned. The size distribution of the as prepared nanowires is uniform over the entire area and the average diameter of the nanowires is 200 nm ± 10 nm, which is consistent with the pore size dimension of the AAO template.


image file: c4ra05878j-f2.tif
Fig. 2 A SEM image of the as-prepared Cu2FeSnS4 product.

The microstructure and chemical composition of Cu2FeSnS4 nanowires have been studied with TEM studies accompanied by selected area electron diffraction (SAED) and energy dispersive X-ray spectrometry (EDX). A representative TEM image in Fig. 3a indicates that the as-prepared Cu2FeSnS4 nanowires are straight with smooth surface. These nanowires have an average diameter of 200 nm, this is consistent with the pore size dimension of the AAO template. The selected-area electron diffraction (SAED) pattern of the nanowires is displayed in the inset of Fig. 3a, which shows clear symmetrical diffraction spots, disclosing the single crystalline nature of the Cu2FeSnS4 nanowires. A high resolution TEM (HRTEM) image shown in Fig. 3b reveals clear lattice spacing of 0.385 nm which is calculated from the line profile in Fig. 3c. This value of 0.385 nm corresponds well to the d spacing of the (110) planes in tetragonal stannite structured Cu2FeSnS4, confirming the well-crystallized nanowires. The inset of Fig. 3b shows a two-dimensional Fourier transform pattern of the lattice resolved image, which can be indexed to the [[1 with combining macron]11] zone of tetragonal stannite Cu2FeSnS4, demonstrating the [110] growth direction and single crystallinity of the Cu2FeSnS4 nanowires. The EDX spectrum (Fig. 3d) taken from the sample reveals intense peaks of Cu, Fe, Sn and S, indicating the chemical composition of Cu, Fe, Sn and S, only. The gold and carbon signals come from the supporting TEM grid. EDX quantitative analysis gives an average Cu/Fe/Sn/S composition (%) of 25.18 13.12[thin space (1/6-em)]:[thin space (1/6-em)]13.27[thin space (1/6-em)]:[thin space (1/6-em)]48.43, being close to molar ratio of 2[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]4 and in accordance with the stoichiometry of Cu2FeSnS4.


image file: c4ra05878j-f3.tif
Fig. 3 (a) A typical TEM image of the as-prepared single-crystalline Cu2FeSnS4 nanowires, inset: a SAED pattern; (b) a typical HRTEM image of the as-prepared Cu2FeSnS4 nanowires, the inset is a two-dimensional Fourier transform pattern of the lattice resolved image; (c) line profile from the area marked with the rectangular frame in (b); (d) a typical EDX spectrum of the as-prepared Cu2FeSnS4 nanowires.

STEM-EDX elemental mapping can provide information on the spatial distribution of different compositional elements in the Cu2FeSnS4 nanowires. Fig. 4 shows the dark field image of a portion of the Cu2FeSnS4 single nanowire and gives the elemental maps of Cu, Fe, Sn and S, respectively. The uniform spatial distribution of different compositional elements in Cu2FeSnS4 nanowires is illustrated evidently in the elemental maps.


image file: c4ra05878j-f4.tif
Fig. 4 Dark field image of a portion of a Cu2FeSnS4 nanowire and EDX elemental maps of Cu, Fe, Sn S, respectively.

Dark black color of the as-prepared Cu2FeSnS4 nanowires is observed, suggesting its strong photon absorption in the entire visible range of light. Fig. 5 shows the room temperature UV-Vis absorption spectrum for the as-prepared Cu2FeSnS4 nanowires sample, indicating broad and strong optical absorption in the UV-visible region. Estimation on the optical band gap (Eg) of the Cu2FeSnS4 nanowires can be obtained by plotting (αhν)2 as a function of the photon energy (in the inset of Fig. 5), with α being the absorption coefficient, h Planck's constant, and ν the frequency. The Eg value is calculated to be 1.42 eV based on the intersection of the extrapolated linear portion, being consistent with that of reported value (1.28–1.54 eV) for Cu2FeSnS4.1,2 Such band gap value is desirable for the potential photovoltaic applications.


image file: c4ra05878j-f5.tif
Fig. 5 A typical room-temperature UV-visible absorbance spectrum of the as-prepared single-crystalline Cu2FeSnS4 nanowires, inset: plotting (αhν)2, the square of the absorption coefficient (α) multiplied by the photon energy (), versus photon energy ().

It has been found that the porous AAO template plays a decisive role in the synthesis of single-crystalline Cu2FeSnS4 nanowires. If AAO template was not used with other reaction conditions unchanged, Cu2FeSnS4 nanoparticles, instead of nanowires array, can be obtained. Fig. 6a shows a typical XRD pattern of Cu2FeSnS4 nanoparticles. All diffraction peaks can be indexed to the stannite structured Cu2FeSnS4. The obviously broadening of XRD peaks suggests that the as-prepared Cu2FeSnS4 particles are of very small sizes. Based on the Scherrer equation, D = (0.89λ)/β(cos[thin space (1/6-em)]θ), here λ is the wavelength for the Kα1 (1.54056 Å), β is the peak width at half-maximum in radians and θ is the Bragg's angle, the average particle size was calculated to be 35 nm. The particle size result is consistent with later TEM analysis. A TEM image in Fig. 6b indicates that the as-prepared product is composed of a lot of aggregated nanoparticles with size in the range of 20–50 nm. The diffraction rings of the selected-area electron diffraction (SAED) pattern taken from these nanoparticles, as displayed in the inset of Fig. 6b, reveal polycrystalline nature of Cu2FeSnS4 sample and can be indexed to (112), (220) and (312) reflections, consistent with the expected tetragonal crystal lattice. The EDX spectrum (Fig. 6c) taken from the sample reveals intense peaks of Cu, Fe, Sn and S, displaying the chemical composition of Cu, Fe, Sn and S, only. EDX quantitative analysis gives an average Cu/Fe/Sn/S molar ratio of nearly 2[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]1[thin space (1/6-em)]:[thin space (1/6-em)]4, in accordance with the stoichiometry of Cu2FeSnS4.


image file: c4ra05878j-f6.tif
Fig. 6 (a) A representative XRD pattern of Cu2FeSnS4 nanoparticles; (b) a typical TEM image and corresponding SAED pattern (inset) of the as-prepared Cu2FeSnS4 nanoparticles; (c) a typical EDX spectrum of the as-prepared Cu2FeSnS4 nanoparticles.

The observation from above controlled experiment suggests that random polycrystalline nanoparticles should formed during the initial stage of reaction process. In the case with the use of AAO template, many random nanocrystals should be produced confined within the pores of AAO template, and then, a subsequent crystal growth process continued. It is known that nanocrystals have high surface energy and they tend to contact and grow together or fuse into larger particles to decrease the surface energy. Oriented attachment mechanism can be used to explain the crystal growth in which smaller particles with common crystallographic orientations directly combine together to form larger ones by crystallographic fusion at the planar interface. During the process of oriented attachment, random particles with no common crystallographic orientation tend to rotate into an orientation to obtain the structural accord at the interface. Then, a coherent grain–grain boundary is formed and disappeared when a single larger nanocrystal is produced finally. Since our present crystal growth occurs in long confined AAO pores, oriented attachment serves a most possible mechanism due to its not involving of diffusion of the grain boundary over large distance scales. In the long time reaction process, misoriented Cu2FeSnS4 grains rotate or reorient to obtain a structural accord at the interface and induced further crystal growth along the preferred growth axis [110] direction. As a result of the confinement of the AAO pores, one-dimensional single crystalline Cu2FeSnS4 nanowires with diameter determined by the pore size in the AAO template were generated, which is evidenced by the SEM and TEM characterization of our product.

To investigate the photoelectric properties, the current–voltage (IV) measurements for the Cu2FeSnS4 nanowires thin films were performed. Fig. 7 shows the IV curves of the films tested in the darkness and under an illumination intensity of 100 mW cm−2 from a 150 W Xenon lamp (Bentham IL7), which was measured in a 5 V bias range. It is found that an increase of about 29% in photocurrent at 5 V by under light irradiation relative to the dark state. In the present case, the energy from Xe light irradiation excites electrons in the Cu2FeSnS4 semiconductor from valance band to the conduction band and then increases the holes in the film. As a result, the current is increased obviously. This obvious photoelectric response suggests that the as-synthesized Cu2FeSnS4 nanowires would be a potential candidate in the fabrication of photovoltaic devices.


image file: c4ra05878j-f7.tif
Fig. 7 The current–potential (IV) curve of the Cu2FeSnS4 film tested in the darkness (black line) and under illumination (red line).

4. Conclusions

In conclusion, single-crystalline Cu2FeSnS4 nanowires have been synthesized through a simple solution strategy within AAO pores. Oriented attachment mechanism has been used to explain the growth of Cu2FeSnS4 nanowires. UV-Vis absorption spectra revealed that the as-prepared Cu2FeSnS4 nanowires have strong optical absorption in the visible region and the band gaps match well with that of the corresponding bulk materials, disclosing their suitability for the photovoltaic application. The Cu2FeSnS4 nanowires showed obvious photoelectric response under light irradiation. The as-prepared Cu2FeSnS4 nanowires may be promising for use as light-absorption layers in solar cells as well as other nanoelectronic devices.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 21371163).

References

  1. C. Yan, C. Huang, J. Yang, F. Liu, J. Liu, Y. Lai, J. Li and Y. Liu, Chem. Commun., 2012, 48, 2603–2605 RSC.
  2. X. Y. Zhang, N. Z. Bao, K. Ramasamy, Y. H. Wang, Y. F. Wang, B. P. Lin and A. Gupta, Chem. Commun., 2012, 48, 4956–4958 RSC.
  3. C. Domain, S. Laribi, S. Taunier and J. F. Guillemoles, J. Phys. Chem. Solids, 2003, 64, 1657–1663 CrossRef CAS.
  4. P. Jackson, D. Hariskos, E. Lotter, S. Paetel, R. Wuerz, R. Menner, W. Wischmann and M. Powalla, Prog. Photovoltaics, 2011, 19, 894–897 CAS.
  5. Y. Wu, C. Wadia, W. Ma, B. Sadtler and A. P. Alivisatos, Nano Lett., 2008, 8, 2551–2555 CrossRef CAS PubMed.
  6. I. Gur, N. A. Fromer, M. L. Geier and A. P. Alivisatos, Science, 2005, 310, 462–465 CrossRef CAS PubMed.
  7. L. H. Ai and J. Jiang, J. Mater. Chem., 2012, 22, 20586–20592 RSC.
  8. C. An, K. Tang, G. Shen, C. Wang, L. Huang and Y. Qian, Mater. Res. Bull., 2003, 38, 823–830 CrossRef CAS.
  9. Z. Gui, R. Fan, X. Chen, Y. Hu and Z. Wang, Mater. Res. Bull., 2004, 39, 237–241 CrossRef CAS PubMed.
  10. L. Li, X. Y. Liu, J. Huang, M. Cao, S. Chen, Y. Shen and L. Wang, Mater. Chem. Phys., 2012, 133, 688–691 CrossRef CAS PubMed.
  11. M. J. Hetzer, Y. M. Strzhemechny, M. Gao, M. A. Contreras, A. Zunger and L. J. Brillson, Appl. Phys. Lett., 2005, 86, 162105 CrossRef PubMed.
  12. H. Peng, D. T. Schoen, S. X. Meister, F. Zhang and Y. Cui, J. Am. Chem. Soc., 2007, 129, 34–35 CrossRef CAS PubMed.
  13. W. B. Choi, B. H. Cheong, J. J. Kim, J. Chu and E. Bae, Adv. Funct. Mater., 2003, 13, 80–84 CrossRef CAS.
  14. O. Rabin, P. R. Herz, Y. M. Lin, A. I. Akinwande, S. B. Cronin and M. S. Dresselhaus, Adv. Funct. Mater., 2003, 13, 631–638 CrossRef CAS.
  15. T. A. Crowley, K. J. Ziegler, D. M. Lyons, D. Erts, H. Olin, M. A. Morris and J. D. Holmes, Chem. Mater., 2003, 15, 3518–3522 CrossRef CAS.
  16. T. E. Mallouk, Science, 2001, 291, 443–444 CrossRef CAS.
  17. Y. Mao and S. S. Wong, J. Am. Chem. Soc., 2004, 126, 15245–15252 CrossRef CAS PubMed.

This journal is © The Royal Society of Chemistry 2014
Click here to see how this site uses Cookies. View our privacy policy here.