Amines free environmentally friendly rapid synthesis of Cu2SnS3 nanoparticles

Amines free environmentally friendly rapid synthesis of Cu2SnS3 nanoparticles

Optical Materials 58 (2016) 268e278 Contents lists available at ScienceDirect Optical Materials journal homepage: www.elsevier.com/locate/optmat Am...

3MB Sizes 5 Downloads 14 Views

Optical Materials 58 (2016) 268e278

Contents lists available at ScienceDirect

Optical Materials journal homepage: www.elsevier.com/locate/optmat

Amines free environmentally friendly rapid synthesis of Cu2SnS3 nanoparticles A.C. Lokhande a, S.A. Pawar a, Eunjin Jo a, Mingrui He a, A. Shelke b, C.D. Lokhande b, Jin Hyeok Kim a, * a Optoelectronic Convergence Research Centre, Department of Materials Science and Engineering, Chonnam National University, Gwangju 500-757, South Korea b Thin Film Physics Laboratory, Department of Physics, Shivaji University, Kolhapur 416004, M.S, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 February 2016 Received in revised form 17 March 2016 Accepted 17 March 2016

Cubic and tetragonal structured Cu2SnS3 (CTS) nanoparticles are rapidly synthesized within a short reaction time of 5 min using low cost amine free octadecene (ODE) solvent by hot injection technique. The effects of precursor concentration, sulfur source and reaction time on the CTS nanoparticle synthesis are studied. The crystal structure, size, phase purity, atomic composition, oxidation state and optical properties of these nanoparticles are studied in detail by X-ray diffraction (XRD), transmission electron microscopy (TEM), Raman, energy dispersive X-ray spectroscopy (EDX), X-ray photoelectron spectroscopy (XPS) and UVevisible spectroscopy techniques. Spherical shaped particles in the size range of 25e30 nm are obtained with atomic composition in close agreement with the stoichiometry of Cu2SnS3. The reaction mechanism of CTS nanoparticle formation is proposed. The synthesized cubic and tetragonal structured CTS nanoparticles exhibit optimal band gaps of 1.23 eV and 1.45 eV, respectively, suitable for use in photovoltaic applications. © 2016 Elsevier B.V. All rights reserved.

Keywords: Cu2SnS3 Nanoparticles Band gap Hot injection Sulfur source

1. Introduction Solar energy conversion has received great attention in the recent years because it is a promising source of clean, sustainable and renewable energy that is sufficiently abundant to meet the projected world needs [1]. Earth abundant elements such Cu, Zn, Sn and S have been investigated for use in photovoltaic applications. For example, Kesterite Cu2ZnSnS4 (CZTS) has been used for photovoltaic applications because of its favorable band gap and high optical absorption coefficient [2]. CZTS based solar cells have demonstrate power conversion efficiencies (PCE) beyond 10.2% [3]. However, the reliability of solar devices based on CZTS based absorber compound is limited. The synthesis of pure phase kesterite CZTS absorber compound is difficult and challenging owning to the formation of secondary phases [2]. Additionally, from the standpoint of economics, use of a higher number of elements in the absorber compound leads to higher costs in solar cell production. Hence, attempts have been made to replace the CZTS absorber

* Corresponding author. E-mail address: [email protected] (J.H. Kim). http://dx.doi.org/10.1016/j.optmat.2016.03.032 0925-3467/© 2016 Elsevier B.V. All rights reserved.

compound by a ternary semiconductor compound [1,4]. Cu2SnS3 (CTS), a ternary ‘P’ type semiconductor compound is an emerging candidate for the possible replacement of the quaternary CZTS compound. CTS is composed of earth abundant, non-toxic, and inexpensive elements such as Cu, Sn and S. CTS exhibits a high optical absorption coefficient (105 cm1) and a tunable band gap (1.2e1.4 eV) [5,6]. Chalcopyrite compounds such as CuInS2 and CuInSe2 also exhibits advantages and optimal optical properties like CTS. However due to the scarcity of In, it is not preferred as a absorber material for solar cell [3]. Avellaneda et al. [7] reported that the theoretical power conversion efficiency (PCE) of CTS thin film solar cell can reach upto to 30%. KuKu et al. reported a PCE of 0.11% for the first time with CTS films deposited by vacuum evaporation technique [8]. Recently, Nakashima et al. [9] reported 4.63% PCE of CTS solar cell by vacuum evaporation method. CTS films have been synthesized by various physical and chemical methods. The physical methods include co-sputtering [10], co-evaporation [11], PLD (pulsed laser deposition) [12], sulfurization [13] and ball milling [14]. Physical techniques are costly due to the use of vacuum and complex equipments. Compared to the physical techniques, the chemical techniques are simple, inexpensive and do not require vacuum or complex equipment [15]. Chemical techniques such as

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

solvothermal [16], hydrothermal [17], hot injection [18], electrodeposition [19], SILAR (successive ionic layer adsorption and reaction) [20], spin coating [6], spray pyrolysis [21], microwave irradiation [22] and sonochemical [23] have been employed for the synthesis of CTS nanoparticle. Amongst the above techniques, hot injection (HI) is usually preferred for nanoparticle synthesis due to good control over the size, structure and composition [1]. Development of monocrystalline solar cells based on earth abundant and non-toxic elements is one approach for reducing the cost of solar cell device fabrication. The nano size of the particles enables rapid charge carrier collection and eliminates the need for high material purity and its high associated cost [24]. Recently CTS nanoparticles have been synthesized using the HI method. CTS nanoparticles were synthesized using cationic deficient Cu31S16 seeds for reaction time periods of 4 h to form the CTS nanoparticles [25]. Okano et al. [26] synthesized tetragonal CTS nanoparticles by sequentially injecting the sulfur and the tin followed by sulfur in oleylamine (OLA) solvent. Chang et al. [27] synthesized CTS cubic phase nanoparticles in the octadecene (ODE) solvent using copper iodide and tin acetate by conducting the reaction for a longer period (overnight). Liu et al. [18] synthesized cubic CTS nanoparticles by injecting sulfur into the Cu þ Sn complex at 240  C in the OLA solvent for reaction time periods more than 60 min. These results show that CTS nanoparticle synthesis is complicated and time consuming. In hot injection technique, reaction time is a very influential parameter in nanoparticle synthesis. Sufficient reaction time favors the complete chemical reaction and hence produces pure phase compounds without secondary phases. A longer reaction time produces a highly crystalline and thermodynamically stable phase. On the other hand, a short reaction time may lead to the formation of secondary phases due to the insufficient chemical reaction between the reactants [1]. Solvents with strong coordinating property are preferred for nanoparticle synthesis. OLA solvent has strong coordinating ability. However, OLA is costly and toxic. On the other hand, the octadecene (ODE) solvent is nontoxic and cheap compared to OLA and can be employed for CTS nanoparticle synthesis [1]. To the best of our knowledge, as of now, there are no reports on the rapid synthesis of cubic and tetragonal CTS nanoparticles using single metal precursor source (chloride) in noncoordinating ODE solvent with short reaction time. In the present article, simple and rapid synthesis of cubic and tetragonal phase CTS nanoparticles by hot injection method using the nontoxic ODE solvent is reported. The effects of precursor concentration, sulfur source and reaction time on the CTS nanoparticle synthesis are studied. The various properties of CTS nanoparticles are evaluated and the potential application of CTS nanoparticles in solar cell is suggested. The possible mechanism of CTS nanoparticle formation and growth has been elaborated.

269

toluene, followed by centrifugation at 5000 rpm for 10 min. The supernatant was then discarded and the precipitation step was repeated two more times to remove the remaining solvent content and the CTS nanoparticle powder was collected. The crystal structure of the CTS nanoparticles was studied using high resolution powder X-ray diffraction (PXRD) with Ni-filtered Cu-Ka radiation of 1.5405 Å (X'pert PRO, Philips, Eindhoven, Netherlands). The Raman spectra of the CTS nanoparticles were recorded in the range of 150e500 cm1 using a micro-Raman spectrometer (Via Reflex UV Raman microscope, Renishaw, U.K. at KBSI Gwangju center) employing a HeeNe laser source with an excitation wavelength of 488 nm and a resolution of 1 cm1 at 15 mW laser power. The surface morphology of the CTS powder was examined by field emission scanning electron microscope (FESEM; S-4700, Hitachi) with an attached energy dispersive X-ray spectroscope (Varian, CARY, 300 Conc.). The high-resolution transmission electron microscopy image of CTS particles was obtained using a high resolution JEOL-3010 microscope. The chemical bonding was examined by X-ray photoelectron spectroscopy (XPS), VG Multilab 2000, Thermo VG Scientific, UK) with monochromatic Mg-Ka (1253.6 eV) radiation source. An optical absorption study of the nanoparticles was carried out in the wavelength range of 200e1000 nm using a UVeViseNIR spectrophotometer (Cary 100, Varian, Mulgrave, Australia). 3. Results and discussion 3.1. Structural study of CTS nanoparticles Precursor concentration plays an important role in nanoparticle synthesis. The Cu:Sn molar concentration ratio was varied from 2.2 to 1.6 and the crystal structure of CTS nanoparticles was determined using PXRD study. Fig. 1(aed) shows the PXRD diffraction patterns of the CTS nanoparticles for various Cu:Sn molar concentration ratios. In all cases, cubic CTS phase is formed. For the Cu:Sn molar concentration ratio of 2.2 (Fig. 1(a)), CTS is formed together with a CuS secondary phase. As the Cu:Sn ratio is reduced to 2, the secondary CuS peaks are reduced (Fig. 1(b)). As the Cu:Sn ratio is further reduced to 1.8 (Fig. 1(c)) and 1.6 (Fig. 1(d)), the CuS secondary phase content is reduced to a greater extent and is then completely eliminated. As seen in Fig. 1(d), all (111), (200), (220) and (311) diffraction peaks at 28.44 , 32.9 , 47.3 and 56.12 , respectively could be

2. Experimental details and characterization All chemicals were used as received without further purification. Copper chloride (CuCl2$2H2O), tin chloride (SnCl2$2H2O), sulfur powder, thiourea and 1-octadecene (ODE) were purchased from SigmaeAldrich. In the typical synthesis of CTS nanoparticles, 2 mM copper chloride and 1 mM tin chloride were mixed in 20 ml ODE solvent in the three neck round bottom flask. The mixture was purged with argon for 10 min and then the reaction was initiated by raising the temperature to 140  C and later aged for 10 min to form the (Cu þ Sn) complex. The temperature was subsequently raised to 220  C and 3 mM sulfur was injected into the flask. The reaction was aged in the time range of 5e15 min to enable the formation of the CTS nanoparticles. After cooling to room temperature, the reaction mixture was precipitated with acetone and dispersed in 5 ml

Fig. 1. PXRD diffraction patterns of CTS cubic nanoparticles synthesized from varied Cu:Sn molar concentration ratio as (a) 2.2, (b) 2, (C) 1.8 and (d)1.6.

270

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

assigned to the cubic CTS phase (JCPDS 01-089-2877). The absence of secondary peaks in the diffraction pattern indicates that pure phase CTS nanoparticles have been synthesized. The diffraction pattern clearly indicates that the synthesized CTS nanoparticles are highly polycrystalline with a preferred growth orientation along (111) plane and are rather pure. Control of precursor reactivity is vital in nanoparticle formation because the difference in the reactivity may lead to formation of different separate compounds, such reactivity control can be achieved by controlling the precursor concentrations. The use of copper under stoichiometry avoids the precipitation of copper and the formation of copper sulfide. Hence, it can be concluded that the use of copper under stoichiometry avoids secondary CuS phase formation in CTS nanoparticles. Additionally as observed in the XRD spectra, CuS peaks are observed for the Cu:Sn molar concentration ratios of 2.2,2 and 1.8 due to excessive copper content causing copper precipitation and also due to the preferential reaction between copper (soft acid) and sulfur (soft base) rather than between tin (hard acid) and sulfur based on the principle of hard and soft acids and bases (HSAB) [1]. Therefore, CuS is initially formed during the CTS nanoparticle formation. The crystallite size, D, was calculated for the (111) plane at full width half maximum (FWHM) using the Scherrer's formula and is given as



0:9l b cosq

formation of sulfur polyhedra in a cubic lattice. The refined lattice parameters (a, b and c ¼ 5.4325 A ), bond length (Cu-Sn ¼ 1.35 A ) and unit cell volume (210 A ) are calculated from the refinement model. Cu, Sn and S atoms occupy the 4b, 4a and 24g sites, respectively. The sulfur polyhedra contribute the formation of equivalent positions of Cu and Sn at the A and B sites as observed from the structural model. The equivalent positions of the Cu and Sn provide structural stability and sulfur polyhedra contribute the structural polarization, i.e., any distortion in the sulfur polyhedra produces changes in the charge distribution and hence the structure can be polarized. 3.2. Reaction time variation and EDX study of CTS nanoparticles The effect of reaction time variation on the crystal structure of CTS has been evaluated. The Cu:Sn ratio of 1.6 was used in the study to evaluate the effect of reaction time. Fig. 3(aec) shows the PXRD spectra of the CTS nanoparticles at reaction time of 5, 10 and 15 min, respectively, after sulfur injection. As observed from the spectra, in all cases, a pure CTS phase is formed without any secondary phase. As the reaction time increases from 5 to 15 min, the

(1)

where, b is the broadening of the diffraction line measured at half maximum intensity (radians), l ¼ 1.5404 Å is the wavelength of the CuKa X-ray and q is the Bragg's angle. The size of the synthesized CTS nanoparticles using (111) plane is found to be approximately 24.02 nm. Fig. 2(a) shows the Rietveld analysis of the X-ray diffraction (XRD) pattern of the cubic CTS nanoparticles obtained for the Cu:Sn ratio of 1.6. The quantitative Rietveld analysis of this XRD pattern was carried out using the FULLPROF software. The experimental and calculated curves are well fitted in the F-43m space group of cubic structure. The pattern refinement was performed using the Gaussian and Lorentzian functions. The observed goodness of fit of the refinement is 1.33. The corresponding structure model is shown in Fig. 2(b). The model depicts a cubic arrangement with the

Fig. 3. PXRD spectra of CTS cubic nanoparticles synthesized with reaction times of (a) 5, (b) 10 and (c) 15 min, after sulfur injection.

Fig. 2. (a) Rietvield analysis of X-ray diffraction pattern of CTS cubic nanoparticles synthesized from Cu:Sn molar concentration ratio of 1.6 in FULLPROF program and (b) corresponding structural model drawn in Vesta.

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

intensity of the prominent (111) peak increases and the peak becomes narrow indicating the increased size of the nanoparticles due to the crystallization of the CTS nanoparticles with the increased reaction time (Table 1). The atomic compositions of the CTS nanoparticles were determined from the EDX study. Table 1 shows the results of the EDX analysis of the samples synthesized at reaction times from 5 to 15 min. The samples clearly indicate the presence of copper, tin and sulfur in stoichiometry close to that of the cubic CTS with a slightly deficiency in the copper content. Hence, the results of PXRD and EDX study show that the CTS nanoparticles can be synthesized in a short reaction time of 5 min with the stoichiometric composition and reaction time does not have significant effect on the composition of the compound. 3.3. FESEM and TEM analysis of CTS nanoparticles The surface topography of the CTS nanoparticles obtained using the Cu:Sn ratio of 1.6 with reaction time of 5 min was examined by FESEM. Fig. 4(a) shows the FESEM image of the CTS nanoparticles. As seen in the image, non-uniform sized spherical shaped CTS nanoparticles are densely packed forming a cluster of grains. Smaller grains are aggregated with each other to form larger grains. The grain size is in the100e120 nm range. The aggregation is attributed due to the use of the strong reactive chloride metal precursor sources [1]. This leads to the chemical reactions occurring at higher uncontrolled rates. Hence, nucleation and grain growth processes occur simultaneously at a very fast rate. The newly formed nuclei acting as the seeds for the growth of the particles absorb the newly generated monomers and grow rapidly and freely in all directions. The grain growth rate exceeds the nucleation rate, so that the particles in the grain growth stage are unstable due to their high free surface energy. To maintain the stability, the particles seek to lower their free surface energy by reducing the surface area by merging of the particles together with each other resulting in aggregation. Fig. 4(b) shows the TEM image of the CTS nanoparticles. The CTS nanoparticles are polydispersed with an average particle size of 20 nm, consistent with the XRD results. The TEM image clearly reveals the aggregation of smaller particles forming the larger particles. The HRTEM image in Fig. 4(c) clearly shows the lattice fringes with the interplaner lattice spacing of 0.31 nm, that can be assigned to the (111) plane of cubic CTS [28]. Fig. 4(d and e) shows the FESEM images of CTS nanoparticles synthesized with reaction times of 10 and 15 min, respectively. The images clearly indicate the aggregation of the particles and the increase in the grain size with increased reaction time, providing evidence for the highly crystalline nature of the particles as revealed by the XRD study. 3.3.1. Mechanism of nanoparticle formation The mechanism of nanoparticle formation is explained in the schematic illustrations presented in Fig. 5. The growth of the nanoparticles is associated with three important stages of (a) precursor to monomer conversion, (b) nucleation and (c) particle formation. Initially, when the precursors are mixed in the solvent and the reaction is initiated, the precursors are converted into

Table 1 Crystallite size and EDX analysis of CTS cubic nanoparticles synthesized from varied reaction time. Reaction time (min)

Crystallite size (nm)

Cu at%

Sn at%

S at%

5 10 15

25 28 32

30 29 27

17 16 16

56 51 54

271

monomers. The rate of the precursor into monomer conversion depends upon the reaction temperature and is increased with increasing temperature. The newly formed monomers form metastable nuclei due to their high free energy. The nuclei attempt to maintain their stability by combining with the monomers and forming stable nuclei. The newly formed stable nuclei further grow due to the diffusion of the monomers at the particle surface interface. With time, these monomers diffuse to a greater extend and finally are incorporated into the particle leading to its growth. As the reaction continues, these newly formed particles achieve the metastable state because of the high free energy at the particleeliquid interface. To achieve stability, the particles lower their free energy by lowering their surface area. Hence, the particles combine with each other and eventually reduce their free energy. The combining of the particles together with other particles is called the aggregation (d). The aggregation of particles is reaction time dependent and hence the particle size increases with longer reaction time and grain growth takes place (e). 3.3.2. Reaction mechanism of CTS nanoparticles The formation of CTS nanoparticles is related to sequential stages namely, the formation of binary sulfides and then the formation of ternary sulfides. The metal chlorides (copper chloride and tin chloride) were added in the solvent and the reaction was initiated by heating the solution. As the reaction temperature (140  C) is increased, the metal chlorides form Cu-Sn complex. Then as the reaction temperature is raised (220  C), sulfur is swiftly injected into the complex solvent. The sulfur then reacts with the metal complex and form binary CuS and SnS sulfides simultaneously. These binary sulfides then further react with each other to give the final ternary sulfide (Cu2SnS3) compound. In some cases, these binary sulfides appear as secondary phases in the final compound due to the incomplete reaction and excess concentration of the precursors [1,2]. 3.4. Raman analysis of CTS nanoparticles The XRD results cannot decisively confirm the crystal structure of the CTS nanoparticles because cubic and tetragonal CTS structures exhibit similar XRD spectra [26,27]. These CTS structures also resemble the zinc blende structure of CZTS and ZnS and therefore are indistinguishable. Hence, Raman analysis study was conducted because it is sensitive to lattice vibrations and can easily differentiate the crystal structures [1,29]. Fig. 6 shows the Raman spectrum of the CTS nanoparticles. The peaks at 291.32 and 334.15 cm1 confirm the formation of a cubic CTS structure and matches well with the previously reported results [22]. The peak at 291.32 cm1 can be attributed to the A1 mode of the CTS [29]. The Raman peaks of the tetragonal CTS are reported to be located at 335 and 350 cm1 [30]. Hence, crystal structures can be easily distinguished from the Raman analysis. The secondary phase peaks of SnS and CuS at 221 cm1 and 470 cm1, respectively are not observed in the spectra, indicating that pure cubic CTS nanoparticles have been synthesized. 3.5. XPS analysis of CTS nanoparticles The oxidation states of the elements and bonding changes in CTS were confirmed from the XPS study. The X-ray photoelectron spectra of Cu(2p), Sn(3d) and S(2p)are shown in Fig. 7(aec),. Fig. 7(a) shows the double peak feature in the Cu(2p) spectrum for CTS, suggesting the presence of ‘Cu’ in CTS. However, to precisely determine the features of double peaks of Cu(2p1/2) and Cu(2p3/2), the Cu(2p) XPS spectra were deconvoluted using Voigt curve fitting. The deconvolution resulted in a perfect fit for only the two peaks

272

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

Fig. 4. (a) FESEM image, (b) TEM image and (c) HRTEM image of CTS cubic nanoparticle synthesized from 5 min reaction time. And the FESEM images of CTS nanoparticles synthesized from (d) 10 min and (e) 15 min reaction time.

Fig. 5. Schematic mechanism of CTS nanoparticle formation.

located at the binding energies of 952.12 and 931.07 eV, corresponding to Cu(2p1/2) and Cu(2p3/2) core levels of the Cuþ state in CTS crystal structure, respectively. The energy separation between these two peaks is 19.05 eV Fig. 7(b) shows the double peak feature in the Sn(3d) spectrum for CTS, suggesting the presence of ‘Sn’ in CTS. The deconvolution resulted into perfect fit for only the two peaks located at the binding energies of 495.13 and 487.09 eV

corresponding to the Sn(3d3/2) and Sn(3d5/2) core levels of the Sn4þ state in the CTS crystal structure, respectively. The energy separation between these two peaks is 8.04 eV. The core level spectrum of sulfur S(2p3/2)at 162.8 eV indicates the presence of sulfur in sulfide form. The absence of the peak at 169 eV that is attributed to the presence of the local surface oxidation and impurities implies that the sulfur exists in S2 state. The presence of carbon and oxygen

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

273

performance of a material. Fig. 8(a) shows the optical absorption spectrum of the CTS nanoparticles. CTS nanoparticles exhibit optical absorption in the visible region with along absorption edge at 850 nm. The corresponding band gap of the CTS nanoparticles was evaluated by extrapolating straight the line of the plot of (ahv)2 as a function of the photon energy hv (a ¼ absorption coefficient, h ¼ Plank's constant and v ¼ frequency)(Fig. 8(b)). The estimated optical band gap of 1.23 eV, shows that the CTS nanoparticles can be efficiently used as an absorber material for thin film solar cell application. 3.7. Effect of sulfur sources

Fig. 6. Raman analysis spectra of CTS cubic nanoparticles synthesized from 5 min reaction time.

peaks in the survey spectrum (Fig. 7(d)) is due to use of ODE solvent and hydrous metal precursor sources, respectively. Hence, the XPS study confirms Cuþ, Sn4þ, and S2 states, respectively in CTS. 3.6. Optical study of CTS nanoparticles Optical properties play decisive role in the photovoltaic

The effect of using different sulfur sources on the synthesis of CTS nanoparticles is studied. DDT, saccharin, sodium sulfide and thiourea are used as sulfur sources instead of the elemental sulfur powder and all other experimental parameters are kept same as mentioned above. Fig. 9(aec) shows the XRD spectra of the nanoparticles obtained by using DDT, saccharin and sodium sulfide for the CTS nanoparticle synthesis. As observed from the XRD spectra, the CTS phase is not formed for any of the sulfur sources indicating the unsuitability of these sulfur sources to produce CTS synthesis. The incapability of these sulfur sources to produce CTS nanoparticles can be due to the low reactivity of these sulfur sources and the uncontrolled S2 ion release rate. However, when thiourea was used as sulfur source, formation of tetragonal CTS phase is observed (Fig. 10). The molar concentration of thiourea was varied from 3 to 2.4. With 3 M concentration of thiourea, the tetragonal CTS phase is formed along with the CuS secondary phase (Fig. 10 (a)). As the

Fig. 7. XPS photoelectron spectra of (a) Cu, (b) Sn and (c) S and (d) survey spectrum of CTS cubic nanoparticles synthesized from 5 min reaction time.

274

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

Fig. 8. (a) Optical absorption spectra of CTS cubic nanoparticles and (b) The corresponding band gap plot of the CTS nanoparticles synthesized from 5 min reaction time.

thiourea concentration is reduced, the CuS phase is found to be reduced as well and is finally eliminated indicating the phase pure synthesis of tetragonal CTS phase (Fig. 10 (d)). All (112), (200), (220) and (312) diffraction peaks are assigned to the tetragonal CTS phase (JCPDS 01-089-4714). The crystallite size calculated for the (112) plane was found to be 27.25 nm. When 3 M concentration thiourea was used, CuS was precipitated. However, on the other hand when 2.4 M concentration of thiourea was used, the reaction was controlled by dynamics [1,16] and pure phase and highly crystalline CTS nanoparticles were formed. Thus, the use of thiourea under stoichiometry resulted in the synthesis of pure tetragonal CTS nanoparticle. Hence, it can be concluded that using a highly reactive sulfur source (elemental sulfur powder) produced the cubic phase at low temperature while the use of thiourea produced the tetragonal phase. Thus, the phase of the CTS nanoparticles depends on the type of sulfur source used. There is also possibility of the formation of Cu rich Cu3SnS4 at higher Cu/Sn ratio. In our case, the synthesized material is Cu2SnS3 and is in stoichiometric composition. The nanoparticles are synthesized and used without annealing treatments, hence there is no tin loss from Cu2SnS3 compound to form Cu rich Cu3SnS4 compound [J Mater Sci: Mater Electron (2013) 24:1490e1494]. The Cu3SnS4 and Cu2SnS3 have similar diffraction pattern but can be

distinguished easily by Raman analysis study. The reported Raman peaks for Cu3SnS4 at 295, 318 and 350 cm1 [2] are absent in our case indicating the absence of Cu3SnS4 phase. Fig. 11(a) shows the results of the Rietvield analysis of the XRD pattern of the CTS tetragonal nanoparticles obtained using the 2.4 M thiourea concentration. The experimental and calculated curves are matched in the I-42m space group of tetragonal crystal structure. The observed goodness of fit is 1.32. The unit cell structural model of CTS with atomic coordination is shown in Fig. 11(b). The Cu and Sn atoms occupy the equivalent positions forming bonds to the S atom. Inspection of the structural model shows that every metal ion is linked by four S atoms and forms a distorted tetrahedron The S atom is tetrahedrally bonded to two M1, one M2 and Cu atom and there is no S-S bonding. The Cu and Sn atoms occupy 43.6(2) at% Sn and 56.4(2) at% Cu at M1 site and 46.3(3) at% Sn and 53.7(3) at% Cu at M2 site. During the refinement, the Cu, Sn and S occupy the 2a, 2b and 8f site symmetries of tetragonal structure. The refined lattice parameters (a, b ¼ 5.4146 A and c ¼ 10.8122 A ), bond length (CueSn ¼ 2.44 A ) and unit cell volume (316.446 A ) are calculated from the refinement model.

Fig. 9. (a) PXRD diffraction patterns of nanoparticles synthesized from varied different sulfur sources as (a) DDT, (b) saccharin, (C) Na2S.

Fig. 10. (a) PXRD diffraction patterns of CTS tetragonal nanoparticles synthesized from varied thiourea molar concentration as (a) 3, (b) 2.8, (C) 2.6 and (d) 2.4.

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

275

Fig. 11. (a) Rietvield analysis of X-ray diffraction pattern of CTS tetragonal nanoparticles synthesized from thiourea molar concentration of 2.4 in FULLPROF program and (b) corresponding structural model drawn in Vesta.

3.7.1. Variation of reaction time and EDX analysis The effect of reaction time variation after thiourea injection was evaluated similar to the investigations of the cubic CTS nanoparticles described above. Fig. 12(aec) shows the PXRD spectra of CTS tetragonal nanoparticles produced with reaction times of 5, 10 and 15 min, respectively. As observed from the spectra, CTS is formed without any secondary phases in all cases. Furthermore, the intensity of (112) plane reflection is increased and narrowed with increasing reaction time (10 and 15 min). The EDX analysis was conducted and the atomic composition was found to be in stoichiometry with sulfur poor content. Table 2 shows the compositions of the CTS nanoparticles synthesized using the various reaction times. Based on the data in Tables 1 and 2, it can be concluded that the precursor concentration has a direct effect on the composition of the final synthesized compound. 3.7.2. FESEM and TEM analysis of CTS tetragonal nanoparticles Fig. 13(aec) shows the corresponding FESEM, TEM and HRTEM images of tetragonal CTS nanoparticles synthesized using reaction time of 5 min. As seen in the FESEM image, the particles are

aggregated. The grain size is in the range of 120e150 nm Fig. 13(b) shows the TEM image of the CTS nanoparticles clearly indicating the aggregation of the particles by merging of the particles together. The average particle size was found to be 30 nm consistent with the XRD results presented in Fig. 10. The HRTEM image (Fig. 13(c)) shows the lattice fringes with the interplaner lattice spacing of 0.3125 nm, which is assigned to the (112) plane of tetragonal CTS [33].

3.7.3. Raman and XPS analysis of CTS tetragonal nanoparticles Fig. 14(a) shows the Raman analysis for the CTS nanoparticles. The peaks at 295.12 and 360.04 cm1 confirm the formation of tetragonal CTS structure. No peaks of secondary phases are observed in the spectra indicating phase pure tetragonal CTS nanoparticle synthesis. The Raman peaks obtained for cubic and tetragonal CTS are different due to different arrangement of atoms (structural geometry) in the structure as seen from Rietvield analysis. The Raman profile of tetragonal CTS seems likely to be the mixture of cubic and tetragonal CTS. The reported Raman peaks for tetragonal CTS are at 301, 335 and 350 cm1 wavenumbers [30,32]. These slight deviation in the Raman peak values with that of the reported values is maybe due to the presence of strain or defects in the synthesized compound [31]. Fig. 14(aec) shows the photoelectron spectra of Cu(2p), Sn(3d) and S(2p), respectively. The Cu binding energy peaks Cu(2p1/2) at 953.08 eV and Cu(2p3/2) at 933.8 eV confirms the existence of copper in Cuþ state. The shake up peak at 942.4 eV also indicates the presence of copper in the Cu2þ state. The Sn binding energy peaks at Sn(3d3/ 2)(495.11 eV) and Sn(3d5/2) (487.25 eV) confirms the presence of tin in the Sn4þ state. The sulfur binding energy peaks S(2p3/2) at 162.20 eV and S(2p1/2) at 163.41 eV with peak splitting of 1.21 eV represents the presence of sulfur in the S2 state. The peak at 168.81 eV indicates the presence of surface impurities and local surface oxidation.

Table 2 EDX analysis of CTS tetragonal nanoparticles synthesized from varied reaction time.

Fig. 12. PXRD spectra of CTS tetragonal nanoparticles synthesized from varied reaction time of (a) 5, (b) 10 and (c) 15 min, after thiourea injection.

Reaction time (min)

Crystallite size (nm)

Cu at%

Sn at%

S at%

5 10 15

34 31 37

32 38 30

19 17 19

49 45 51

276

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

Fig. 13. (a) FESEM image, (b) TEM image and (c) HRTEM image of CTS tetragonal nanoparticle synthesized from 5 min reaction time using thiourea sulfur source.

3.7.4. Optical study of CTS tetragonal nanoparticles Fig. 15(a) shows the optical absorption spectra of tetragonal CTS nanoparticles. CTS nanoparticles exhibit optical absorption in the visible region. The corresponding optical band gap plot of the CTS nanoparticles is shown in Fig. 15(b). The optical band gap was found to be 1.45 eV. Hence, due to favorable optical properties, these synthesized CTS tetragonal nanoparticles can be regarded as a possible candidate for solar applications. 4. Conclusions Rapid synthesis of cubic and tetragonal CTS nanoparticles by hot

injection method in non-coordinating and nontoxic ODE solvent is reported. Synthesis parameters such as precursor concentration, reaction time and sulfur source, play an important role in CTS nanoparticle synthesis. The use of sulfur and thiourea under stoichiometry favors pure cubic and tetragonal CTS nanoparticle synthesis, respectively and the phase formation in CTS nanoparticles depends on the type of sulfur source used. The atomic composition of the compound is determined at the nucleation stage and is independent of the reaction time. The synthesized nanoparticles are polycrystalline with crystallite size in the range of 25e30 nm. Reaction time of 5 min was sufficient to produce both cubic and tetragonal CTS nanoparticles in stoichiometry close to that of

Fig. 14. (a) Raman analysis spectra and the XPS photoelectron spectra of (b) Cu, (c) Sn and (d) S of CTS tetragonal nanoparticles synthesized from 5 min reaction time using thiourea sulfur source.

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

277

Fig. 15. (a) optical absorption spectra of CTS tetragonal nanoparticles and (b) The corresponding band gap plot of the CTS nanoparticles synthesized from 5 min reaction time using thiourea sulfur source.

Cu2SnS3 compound. The Raman analyses indicate pure phase nanoparticles were synthesized. The XPS study confirms the presence of copper, tin and sulfur in Cuþ, Cu2þ, Sn4þ and S2 states, respectively. The estimated band gaps of 1.23 and 1.45 eV of the synthesized cubic and tetragonal CTS nanoparticles respectively, are optimal for solar cell application. Acknowledgment This work was supported by the Human Resources Development program (No. 20124010203180) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) Grant funded by the Korea government Ministry of Trade, Industry and Energy and supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (NRF2015R1A2A2A01006856). References [1] A.C. Lokhande, K.V. Gurav, Eunjin Jo, C.D. Lokhande, Jin Hyeok Kim, Chemical synthesis of Cu2SnS3 (CTS) nanoparticles: a status review, J. Alloys Compd. 656 (2016) 295e310. [2] A.C. Lokhande, K.V. Gurav, Eunjin Jo, Mingrui He, C.D. Lokhande, Jin Hyeok Kim, Towards cost effective metal precursor sources for future photovoltaic material synthesis: CTS nanoparticles, Opt. Mater. 54 (2016) 207e216. [3] D.B. Mitzi, O. Gunawan, T.K. Todorov, K. Wang, S. Guha, The path towards a high- performance solution -processed Kesterite solar cell, Sol. Energy Mater. Sol. Cells 95 (2011) 1421e1436. [4] S. Siebentritt, S. Schorr, Kesteritesda challenging material for solar cells, Progress Photovolt. Res. Appl. (2012), http://dx.doi.org/10.1002/pip.2156. [5] B.A. Andersson, Materials availability for large-scale thin-film photovoltaics, Prog. Photovolt. Res. Appl. 8 (2000) 61e76. [6] G.M. Ilari, C.M. Fella, C. Ziegler, A.R. Uhl, Y.E. Romanyuk, A.N. Tiwari, Cu2ZnSnSe4 solar cell absorbers spin-coated from amine-containing ether solutions, Sol. Energy Mater. Sol. Cells 104 (2012) 125e130. [7] D. Avellaneda, M.T.S. Nair, P.K. Nair, Cu2SnS3 and Cu4SnS4 Thin Films via Chemical Deposition for Photovoltaic Application, J. Electrochem. Soc. 157 (6) (2010) D346eD352. [8] T.A. Kuku, O.A. Fakolujo, Photovoltaic characteristics of thin films of Cu2SnS3, Sol. Energy Mater. 16 (1987) 199e204. [9] M. Nakashima, J. Fujimoto, T. Yamaguchi, M. Izaki, Cu2SnS3 thin-film solar cells fabricated by sulfurization from NaF/Cu/Sn stacked precursor, Appl. Phy Express 8 (2015) 042303. [10] R. Bodeux, J. Leguay, S. Delbos, Influence of composition and annealing on the characteristics of Cu2SnS3 thin films grown by cosputtering at room temperature, Thin Solid Films 582 (2015) 229e232. [11] T.S. Reddy, R. Amiruddin, M.C. Santhosh Kumar, Deposition and characterization of Cu2SnS3 thin films by co-evaporation for photovoltaic application, Sol. Energy Mater. Sol. Cells 143 (2015) 128e134.

[12] R.B. Ettlinger, A. Cazzaniga, S. Canulescu, N. Pryds, J. Schou, Pulsed laser deposition from ZnS and Cu2SnS3 multicomponent targets, Appl. Surf. Sci. 336 (2015) 385e390. [13] P.A. Fernandes, P.M.P. Salom, A.F. Cunha, A study of ternary Cu2SnS3 and Cu3SnS4 thin films prepared by sulfurizing stacked metal precursors, J. Phys. D. Appl. Phys. 43 (2010) 215403e215413. [14] Q. Chen, X. Dou, Y. Ni, S. Zhuang, Study and enhance the photovoltaic properties of narrow bandgap Cu2SnS3 solar cell by p-n junction interface modification, J. Colloid Interface Sci. 376 (2012) 327e330. [15] A.C. Lokhande, J.S. Bagi, Studies on enhancement of surface mechanical properties of electrodeposited Ni-Co alloy coatings due to saccharin additive, Surf. Coat. Technol. 258 (2014) 225e231. [16] F. Chen, J. Zai, M. Xu, X. Qian, 3D-hierarchical Cu3SnS4 flowerlike microspheres: controlled synthesis, formation mechanism and photocatalytic activity for H2 evolution from water, J. Mater. Chem. A 1 (2013) 4316e4326. [17] Z. Zhang, Y. Fu, C. Zhou, J. Li, Y. Lai, EDTA-Na2-assisted hydrothermal synthesis of Cu2SnS3 hollow microspheres and their lithium ion storage performances, Solid State Ion. 269 (2015) 62e66. [18] Q. Liu, Z. Zhao, Y. Lin, P. Guo, S. Li, D. Pan, X. Ji, Alloyed (ZnS)x(Cu2SnS3)1-x and (CuInS2)x (Cu2SnS3)1-x nanocrystals with arbitrary composition and broad tunable band gaps, Chem. Commun. 47 (2011) 964e966. [19] D.M. Berg, R. Djemour, L. Gutay, G. Zoppi, S. Siebentritt, P.J. Dale, Thin film solar cells based on the ternary compound Cu2SnS3, Thin Solid Films 520 (2012) 291e6294. [20] Z. Su, K. Sun, Z. Han, F. Liu, Y. Lai, J. Li, Y. Liu, Fabrication of ternary Cu-Sn-S sulfides by a modified successive ionic layer adsorption and reaction (SILAR) method, J. Mater. Chem. 22 (2012) 16346e16352. [21] M. Bouaziz, M. Amlouk, S. Belgacem, Structural and optical properties of Cu2SnS3 sprayed thin films, Thin Solid Films 517 (2009) 2527e2530. [22] W. Wang, H. Shen, J. Li, Rapid synthesis of hollow CTS nanoparticles using microwave irradiation, Mater. Lett. 111 (2013) 5e8. [23] J. Park, M. Song, W.M. Jung, W.Y. Lee, H. Kim, Y. Kim, C. Hwang, W. Shim, Syntheses of Cu2SnS3 and Cu2ZnSnS4 nanoparticles with tunable Zn/Sn ratios under multi bubbles on luminescence conditions, Dalton Trans. 42 (2013) 10545e10550. [24] Katherine Plass, Development of earth abundant mixed metal sulfide nanoparticles for use in solar energy conversion, United States Patent, Jul. 4, 2013, Pub. No.: US 2013/0171056 A1. [25] L. Yi, D. Wang, M. Gao, Synthesis of Cu3SnS4 nanocrystals and nanosheets by using Cu31S16 as seeds, Chem. Eng. Commun. 14 (2012) 401e404. [26] S. Okano, S. Takeshita, T. Isobe, Formation of Cu2SnS3 nanoparticles by sequential injection of tin and sulfur oleylamine solutions into Cu1.8S nanoparticle dispersion, Mater. Lett. 145 (2015) 79e82. [27] J. Chang, E.R. Waclawik, Controlled synthesis of CuInS2, Cu2SnS3 and Cu2ZnSnS4 nano-structures: insight into the universal phase selectivity mechanism, Chem. Eng. Commun. 15 (2013) 5612e5619. [28] S. Rabaoui, H. Dahman, N.B. Mansour, L.E. Mir, Structural, optical and electrical properties of Cu2SnS3nanoparticles synthesized by simple solvothermal technique, J. Mater. Sci. Mater. Electron 26 (2015) 1119e1124. [29] Y. Dong, J. He, X. Li, W. Zhou, Y. Chen, L. Sun, P. Yang, J. Chu, Synthesis and optimized sulfurization time of Cu2SnS3 thin films obtained from stacked metallic precursors for solar cell application, Mater. Lett. 160 (2015) 468e471. [30] D. Tiwari, T.K. Chaudhuri, T. Shripathi, U. Deshpande, V.G. Sathe, Structural and optical properties of layer-by-layer solution deposited Cu2SnS3 films, J. Mater. Sci. Mater. Electron 25 (2014) 3687e3694.

278

A.C. Lokhande et al. / Optical Materials 58 (2016) 268e278

[31] H. Guan, H. Shen, C. Gao, X. He, Structural and optical properties of Cu2SnS3 and Cu3SnS4 thin films by successive ionic layer adsorption and reaction, J. Mater. Sci. Mater. Electron 24 (2013) 1490e1494. [32] S.A. Vanalakar, G.L. Agawane, S.W. Shin, H.S. Yang, P.S. Patil, J.Y. Kim, J.H. Kim, Non-vacuum mechanochemical route to the synthesis of Cu2SnS3 nano-ink

for solar cell applications, Acta Mater. 85 (2015) 314e321. [33] W. Li, X. Han, Y. Zhao, L. Liu, J. Wang, S. Yang, T. Tanaka, Cu2ZnSnS4 alloys synthesized from [email protected] nanoparticles via a facile hydrothermal approach, Mater. Lett. 125 (2014) 167e170.