Ligand exchange induced crystal structure and morphology evolution of copper–tin–sulfur binary and ternary compounds

Abstract

In this manuscript, a simple one-pot heat-up method has been used to prepare multi-component copper–tin–sulfur nanomaterials, including binary Cu_1.94S, ternary Cu₄SnS₄, and Cu_1.94S/Cu₄SnS₄ nanocrystals by varying the reaction temperature, reaction time, and the type of copper source. Post-synthetic ligand exchange (LE) has further been introduced to replace the long-chain ligands originating from 1-dodecanethiol. It has been found that the LE process not only changes the surface ligands but also significantly affects the crystal structure and optical properties of nanocrystals. After LE, the crystal structures of Cu_1.94S and Cu₄SnS₄ transformed to Cu₇S₄ and Cu₃SnS₄, respectively, with the Cu_1.94S/Cu₄SnS₄ nanocrystals showing the same trend. This phenomenon could be ascribed to the loss of Cu⁺ originating from the strong complexation of Cu⁺ and ammonia with the formation of [Cu(NH₃)_n]²⁺ ions under aerobic conditions. Proton nuclear magnetic resonance (¹H NMR) has been used to characterize the ligands on the surface before and after LE, which further demonstrated that the –SH was replaced during LE. Meanwhile, the band gaps of the obtained nanocrystals after LE show an obvious shift in the near-infrared region due to the evolution of crystal structures. This study will provide useful guidance for the LE of nanocrystals and the application of copper-based sulfide nanomaterials in optoelectronics and other fields.

---

1 Introduction

Copper-based binary and ternary compounds, such as Cu_xS, and Cu–Sn–S, have been widely applied in the fields of solar cells, Li-ion batteries,^1,2 super-capacitors,³ photocatalysis,^4,5etc. There have been considerable efforts to synthesize corresponding nanocrystals by various methods, such as solvothermal,⁶ hydrothermal,⁷ and colloidal chemistry methods.^8,9 Among them, the colloidal chemistry method can yield nanocrystals with a controllable crystal structure, uniform morphology, and high quality.¹⁰ This is mainly due to the organic ligands, which can coordinate with the surface of nanocrystals to lower the surface energy, resulting in the formation of nanocrystals. Thiols are one of the typical ligands which can act both as sulfur source and reaction solvents.^11–13 They are also especially advantageous for the synthesis of copper based nanocrystals with a wurtzite structure.^14–16 However, the so-called “crystal-bond” thiols are reported to sit in the high coordination sites of the crystal, which will suppress the surface reactivity of nanocrystals.¹⁷ Meanwhile, the obtained particles capped with a final layer of intact thiols will prevent efficient electronic charge transport between neighbouring nanocrystals and conductivity of nanocrystals,^18,19 thus hindering the application in most biomedical, sensor, and photocatalytic fields.

To solve this problem, ligand exchange (LE) has been proposed to be an effective way to remove the surface long-chain ligand which can hinder the electron transfer.^18–23 At present, LE mainly involves the exchange of functional groups on a crystal surface. Arnold et al.¹⁹ have exchanged the as-synthesized dodecanethiol ligands with short ethanedithiol or ethylenediamine ligands, which can effectively enhance the carrier mobility and concentration of Cu₂S quantum dots. The improvements in the electrical conductivity of Cu_2−xS nanocrystals via LE can be ascribed to the reduction of the inter-nanocrystal separation in the films, as indicated by Pereira et al.²⁴ Zhang et al.²⁵ reported that the short organic aromatic ammonium ligands can partly replace the long-chain OA⁻ and OAm⁺ ligands in a quantum dot solution, and the CsPbI₃ quantum dot solar cells showed enhanced power conversion efficiency with improved electronic coupling. The complete removal of the organic ligand in PbS nanocrystal films can also be achieved using (NH₄)₂S methanol solution.²⁶ The LE process has also been conducted between ruthenium organo metallic cofactor precursors and protein scaffolds to generate new enzymes.²⁷ However, there are few studies on the influence of LE on the crystal structure, morphology evolution and the corresponding optical properties.

Previously, we observed an unexpected crystal structure evolution of Cu₄SnS₄ after LE during the application of Cu₄SnS₄-rGO nanocrystals.²⁸ In this study, we have systematically investigated the crystal structure and morphology evolution of binary Cu_1.94S, ternary Cu₄SnS₄, and Cu_1.94S/Cu₄SnS₄ nanocrystals. The changes of the surface ligands of Cu_1.94S/Cu₄SnS₄ have further been confirmed by proton nuclear magnetic resonance (¹H NMR). Meanwhile, the optical properties also showed significant changes after LE. These findings will provide useful guidance for the modification and application of copper-based sulfide nanomaterials in optoelectronics and other fields.

2 Experimental section

2.1 Materials

Copper(II) acetylacetonate (Cu(acac)₂, Aladdin, 97%), copper(II) chloride (CuCl₂, 99%), tin(II) chloride dihydrate (SnCl₂·2H₂O, Aladdin, analytical grade), 1-dodecanethiol (1-DDT, Aladdin, 98%), oleylamine (OLA, Aladdin, 80–90%), oleic acid (OA, Aladdin, AR), ammonium sulfide ((NH₄)₂S, Macklin, 14%), formamide (FA, 99%), sulfur powder (Sinopharm Chemical Reagent Co., Ltd, 99.999%), sodium sulfide (Na₂S, Energy Chemical, 60%), and sodium hydroxide (NaOH, Sinopharm Chemical Reagent Co., Ltd, AR) were used without further purification. Other chemicals were all of the analytical grade.

2.2 Synthesis of Cu_1.94S, Cu_1.94S/Cu₄SnS₄ and Cu₄SnS₄ nanocrystals

All the experiments were taken out via a simple one-pot heat-up method in a four-neck flask, under a continuous N₂ flow.

2.2.1 Synthesis of Cu_1.94S and Cu₄SnS₄ nanocrystals. Cu_1.94S nanocrystals were prepared according to our previous reports with necessary modification.²⁸ CuCl₂ (0.1076 g) and SnCl₂·2H₂O (0.09 g) were dissolved in 1-DDT (20 mL) and then heated to 120 °C for 30 min. Subsequently, the solution was further heated to 200 °C for 30 min and finally heated to 220 °C for 30 min to synthesize Cu_1.94S nanocrystals. Cu₄SnS₄ nanocrystals can be prepared following the above procedure, while the final reaction temperature was changed to 250 °C. Different reaction temperatures such as 230 °C and 240 °C were also utilized.

2.2.2 Synthesis of Cu_1.94S/Cu₄SnS₄ nanocrystals. Cu_1.94S/Cu₄SnS₄ nanocrystals were prepared using Cu(acac)₂ as a copper source with the reaction temperature and reaction time to be controlled at 220 °C and 20 min, respectively. The other procedures were maintained as above.

2.3 Ligand exchange (LE)

LE was conducted based on the reported procedure.²⁹ The Cu_1.94S nanoplates (40 mg) were dissolved in 4 mL of n-hexane (10 mg nanoplates per mL n-hexane). A solvent mixture consisting of 1 mL of (NH₄)₂S and 10 mL of FA was added to the solution and stirred for 30 min, leading to a complete phase transfer of Cu_1.94S nanoplates from n-hexane to FA. The mixture in the FA phase was washed thoroughly with ethanol. The LE of Cu₄SnS₄ and Cu_1.94S/Cu₄SnS₄ nanocrystals was conducted by the same method.

2.4 Characterization

The crystal structure of the products was characterized using X-ray diffraction (XRD, Rigaku-Smart Lab Advance) with Cu Kα radiation (λ = 1.5408 Å) as the X-ray source. The morphology of the nanocrystals was studied using field emission scanning electron microscopy (FESEM, HITACHI S-4800) and transmission electron microscopy (TEM, JEM-2100F). The scanning TEM elemental mapping was conducted on an FEI Talos F200X G2. The UV-vis-NIR absorption spectra were collected using a UV-vis spectrometer (Varian Cary 5000). The proton nuclear magnetic resonance (¹H NMR) was used to analyze the nanocrystals before and after LE, with chloroform-d (CDCl₃) and dimethyl sulfoxide (DMSO) being used as solvents. To determine the content of Cu element in the (NH₄)₂S and FA layer after LE, a sample of 2 mL of liquid was taken after centrifugation with ethanol. The liquid was diluted for 5 times, and the Cu content was determined using an inductively coupled plasma spectrometer (ICP) with a model of Thermo Fisher iCAP PRO.

3 Results and discussion

The crystal structures of products synthesized at different reaction temperatures were investigated by XRD, as shown in Fig. 1. When the reaction temperature was 220 °C, the XRD pattern shows diffraction peaks at 2θ = 27.7°, 37.7°, 46.4°, and 48.6°, corresponding to the (−2 1 4), (8 4 2), (10 5 2), and (−2 8 2) crystal planes of Cu_1.94S (JCPDS # 34-0660), indicating the formation of monoclinic Cu_1.94S at a low temperature. At reaction temperatures from 230 °C to 250 °C, the products have well-defined peaks at 2θ = 26.7°, 27.8°, 46.6°, and 50.6°, corresponding to the (2 2 0), (0 0 2), (4 3 1), and (2 2 3) crystal planes of orthorhombic Cu₄SnS₄ (JCPDS # 27-0196). This corroborated a fast phase transformation from Cu_1.94S to Cu₄SnS₄ with the incorporation of Sn at the reaction temperature of 230 °C and above. A time series analysis of the crystal structures during the reaction process was conducted to more accurately understand the phase transition process from Cu_1.94S to Cu₄SnS₄ nanocrystals. Experiments were conducted at 230 °C with different reaction times (0, 10, 20, and 30 minutes). XRD (Fig. S1†) analysis reveals that at initial reaction times of 0 to 10 min, the products formed were pure Cu_1.94S nanocrystals. With increased reaction time to 20 min, an Sn precursor was gradually incorporated into the Cu_1.94S nanocrystals, and the phase transition from Cu_1.94S to Cu₄SnS₄ occurred. Meanwhile, a complete formation of pure Cu₄SnS₄ nanocrystals could be obtained at 30 min. This further demonstrates that the Sn precursor can further react with initially-formed Cu_1.94S nanocrystals at reaction temperatures higher than 230 °C with the required duration of time.

Fig. 1 XRD patterns of Cu–Sn–S nanocrystals prepared at different temperatures.

The XPS spectra were analyzed to investigate the surface electronic states and composition of Cu_1.94S nanocrystals. The survey XPS spectrum (Fig. S2a†) confirms the presence of Cu, Sn and S on the surface of Cu_1.94S nanocrystals (Fig. S2a†). The binding energies of Cu 2p_3/2 and Cu 2p_1/2 (Fig. S2b†) are 932.4 eV and 952.2 eV, respectively, corresponding to Cu⁺.³⁰ Moreover, Sn 3d_5/2 and Sn 3d_3/2 (Fig. S2c†) obtain peaks at 486.3 eV and 494.7 eV, corresponding to the value of Sn⁴⁺. For the high-resolution XPS spectrum of S 2p (Fig. S2d†), the peaks at 161.6 and 162.7 eV are attributed to the 2p_3/2 and 2p_1/2 core energy levels of S²⁻. The XPS analysis shows the existence of Sn(IV) element for Cu_1.94S nanocrystals. This observation is consistent with a previous report,³¹ which claimed that Sn can exist as tin sulfide (Sn–X) complex binding to the surface of Cu₂S nanoplates.

In order to further investigate the morphology and microstructure of Cu_1.94S and Cu₄SnS₄ nanocrystals, SEM and TEM measurements have been performed. As shown in Fig. 2a and b, the morphology of Cu_1.94S is mainly thick nanoplates with a lateral size of 27.8 ± 0.7 nm and a thickness of 9.8 ± 0.1 nm (Fig. 2d). The lattice distance is found to be 0.322 nm, which corresponds to the (−2 1 4) crystal plane of Cu_1.94S (Fig. 2c). As shown in Fig. 2e–h, the morphology of Cu₄SnS₄ nanocrystals is mainly cylindrical with a height of about 40.7 ± 1.0 nm and a size of 29.7 ± 0.8 nm. The lattice fringe of 0.323 nm, corresponds to the (0 0 2) crystal plane of Cu₄SnS₄ (Fig. 2g).

Fig. 2 SEM (a and e), TEM (b and f) and HRTEM (c and g) images of (a–c) Cu_1.94S and (e and f) Cu₄SnS₄, and size distribution histograms of (d) Cu_1.94S and (h) Cu₄SnS₄ nanocrystals.

We further investigated the effect of LE on the crystal structure of the as-prepared Cu_1.94S and Cu₄SnS₄ nanocrystals. It is well known that organic ligands can effectively control the size, shape, and composition of nanocrystals during synthesis, but the weak polarity of long-chain hydrocarbon ligands makes nanocrystals poorly dispersible in water. What's more, the bulky organic ligands will limit the conductivity between particles and affect the practical application of materials in the field of photocatalysis and optoelectronics.²⁰ Thus, the LE process has been applied, which can effectively remove the long-chain functional groups on the surface of nanomaterials and improve the conductivity of the materials. Previously, most works have mainly studied the influence of the LE process on hydrophilicity and photoelectric properties.²⁶ However, research on the influence of the LE process on the crystal structure of the material is very rare. We occasionally observed the phase transformation of Cu₄SnS₄ to Cu₃SnS₄ after LE in our previous work.²⁸ Herein, we systematically studied the XRD evolution of binary Cu_1.94S and ternary Cu₄SnS₄ before and after LE. It can be seen from Fig. 3a that after LE, the overall diffraction peaks of Cu_1.94S right-shifted to a higher angle. The peaks corresponding to the (0 16 0) and (8 8 6) crystal planes of Cu₇S₄ formed, indicating that the crystal structure has changed to Cu₇S₄ (JCPDS # 23-0958) after LE. As shown in Fig. 3b, the characteristic diffraction peaks of Cu₄SnS₄ right shifted. The peaks located at 27.3°, 30.9°, 48.3° and 51.8° corresponded to the (2 0 0), (1 2 6), (0 0 20) and (2 0 18) planes of Cu₃SnS₄ (JCPDS # 36-0217).

Fig. 3 XRD patterns of (a) Cu_1.94S and (b) Cu₄SnS₄ before and after LE.

In order to confirm the oxidation states of Cu and the surface composition of Cu₄SnS₄ and Cu₃SnS₄, the XPS spectra were analyzed. As shown in Fig. 4, both Cu₄SnS₄ and Cu₃SnS₄ are composed of three elements – Cu, Sn, and S. In Cu₄SnS₄ (Fig. 4b), the binding energies of Cu 2p_3/2 and Cu 2p_1/2 are 932.4 eV and 952.2 eV, respectively, with a peak splitting of 19.8 eV, indicating the existence of Cu in the form of Cu⁺.³⁰ In Cu₃SnS₄ (Fig. 4f), except for peaks located at 932.4 eV (Cu 2p_3/2) and 952.2 eV (Cu 2p_1/2) corresponding to Cu⁺, there are satellite peaks at 934.6 eV for Cu 2p_3/2, and 953.6 eV for Cu 2p_1/2. The characteristic peaks suggests the presence of both Cu⁺ and Cu²⁺ ions within the Cu₃SnS₄ compound.³⁰ For the high-resolution XPS spectra of Sn3d in Cu₄SnS₄ (Fig. 4c), the binding energies of Sn3d_5/2 and Sn3d_3/2 are 486.2 and 494.6 eV, respectively, consistent with the reported Sn⁴⁺ values in the literature.³⁰ In Cu₃SnS₄, the Sn 3d_5/2 peak at 486.2 and the Sn 3d_3/2 peak at 494.7 eV are assigned to Sn(II), and the Sn 3d_5/2 peak at 487.4 and the Sn 3d_3/2 peak at 495.7 eV are associated with Sn(IV). This suggests Sn exists in both Sn(II) and Sn(IV) oxidation states.^30,32 For the high-resolution XPS spectra of S 2p (Fig. 4d and h) for both Cu₄SnS₄ and Cu₃SnS₄, the peaks at 161.6 and 162.7 eV are attributed to the 2p_3/2 and 2p_1/2 core energy levels of S²⁻.³³ We notice that during the phase transformation of Cu₄SnS₄ to Cu₃SnS₄via LE, the valence of Cu has changed from +1 for Cu₄SnS₄ to +1 and +2 for Cu₃SnS₄. This indicates partial oxidation of Cu during the LE process.

Fig. 4 XPS spectra of (a–d) Cu₄SnS₄ and (e–h) Cu₃SnS₄ composites. (a and e) Survey spectra, and the high-resolution core level spectra of (b and f) Cu2p, (c and g) Sn3d and (d and h) S2p.

During the heating up of the reaction precursors, we notice that just pure Cu_1.94S and Cu₄SnS₄ can be obtained. It is still unknown whether the crystal structure evolution shows a similar trend for Cu_1.94S/Cu₄SnS₄ nanocrystals. We then synthesized Cu_1.94S/Cu₄SnS₄ nanocrystals using Cu(acac)₂ as the copper source and reacted at 220 °C for 20 minutes. For the as-prepared sample, distinguished peaks corresponding to both Cu_1.94S and Cu₄SnS₄ can be obtained, indicating the successful formation of Cu_1.94S/Cu₄SnS₄ nanocrystals, as shown in Fig. 5a. This indicates that Cu(acac)₂ has higher reactivity than CuCl₂, which can partially transform to Cu₄SnS₄ at a lower reaction temperature of 220 °C. After LE, the crystal structures match well with Cu₇S₄ and Cu₃SnS₄, respectively, which indicates the formation of Cu₇S₄/Cu₃SnS₄ nanocrystals. This trend is in accordance to pure Cu_1.94S and Cu₄SnS₄ (Fig. 3). Fig. 5b shows the illustration of the LE process and corresponding morphology evolution. First, the Cu_1.94S/Cu₄SnS₄ nanocrystals were uniformly dispersed in n-hexane (upper layer), and the lower layer was a mixture of (NH₄)₂S and FA. After LE, the nanocrystals were transferred from the upper layer to the lower layer, with the formation of Cu₇S₄/Cu₃SnS₄ nanomaterial. The digital micrographs of Cu_1.94S/Cu₄SnS₄ before and after LE are shown in Fig. 5c and d. A rapid phase transformation happened, thus the lower layer turned to be turbid as they were mixed (Fig. 5c). The phase transformation from hexane to FA indicates that the hydrophobic nanocrystals have been successfully transferred into hydrophilic ones by LE.

Fig. 5 (a) XRD patterns of Cu_1.94S/Cu₄SnS₄ and Cu₇S₄/Cu₃SnS₄ nanocrystals (after LE), and (b) illustration of the LE process and digital micrographs of Cu_1.94S/Cu₄SnS₄ (c) before and (d) after LE.

The morphologies of Cu_1.94S/Cu₄SnS₄ nanocrystals before and after LE have also been investigated by TEM measurements. As shown in Fig. 6a, the Cu_1.94S/Cu₄SnS₄ nanocrystals are mainly composed of hexagonal nanoplates and larger cylinders. In the HRTEM images (Fig. 6b and c), obvious lattice fringes can be observed, and the d-spacings are 0.360 and 0.341 nm, corresponding to the (8 0 0) plane of Cu_1.94S and (2 2 0) plane of Cu₄SnS₄, respectively. After LE, the edge of nanoplates tends to be inapparent (Fig. 6d) and both the particle size and thickness are comparable to the nanoplates without LE. The slight change in the morphology may be due to the loss of copper during the LE process. From the side-view of nanoplates, the measured d-space of lattice fringe is 0.343 nm, corresponding to the (16 0 0) crystal plane of Cu₇S₄ (Fig. 6e). The HRTEM image from the front view shows a lattice fringe with d spaces of 0.323 nm, referring to the (0 2 6) plane of Cu₃SnS₄. This indicated the formation of Cu₇S₄/Cu₃SnS₄ nanocrystals after LE, which is in accordance with the XRD results. Fig. 6g shows the high-angle annular dark-field (HAADF) image of Cu_1.94S/Cu₄SnS₄ nanocrystals. Fig. 6h–j present the scanning TEM elemental mapping results of Cu_1.94S/Cu₄SnS₄ nanocrystals. In these images, Cu and S are evenly distributed in all nanocrystals (Fig. 6h and j). However, the Sn element shows an obvious difference depending on the morphologies (Fig. 6i). Most Sn is distinctly observed on the large cylinders, while the hexagonal nanoplates exhibit only Cu and S. This observation indicates variations in the elemental distribution among nanocrystals with different morphologies. From the EDS mapping results, it can be concluded that the large cylinders indicate the initial formation of Cu₄SnS₄ nanocrystals, while the hexagonal nanoplates are still maintained as Cu_1.94S (Fig. 6h–j).

Fig. 6 (a) TEM and (b and c) HRTEM images of Cu_1.94S/Cu₄SnS₄ nanomaterials synthesized using Cu(acac)₂ as a copper source, (d) TEM and (e and f) HRTEM images of Cu₇S₄/Cu₃SnS₄ nanocrystals after LE, and (g) high-angle annular dark-field (HAADF) image of Cu_1.94S/Cu₄SnS₄ nanomaterials and (h–j) its corresponding EDS elemental mappings of Cu, Sn, and S.

The change of surface ligands induced via LE has further been confirmed by proton nuclear magnetic resonance (¹H NMR).³⁴Fig. 7 shows the ¹H NMR spectra of Cu_1.94S/Cu₄SnS₄ nanocrystals before and after LE. For the as-prepared Cu_1.94S/Cu₄SnS₄, the characteristic peak was around 1.306 ppm, corresponding to –SH in 1-DDT. After LE, the characteristic peak of –SH was greatly weakened, indicating the successful removal of the long-chain ligand of thiol. Research has shown that metal-free inorganic ions such as S²⁻, HS⁻, Se²⁻etc. can be used to replace the organic capping ligands for nanocrystals and provide electrostatic stabilization. Concentrated aqueous solutions of (NH₄)₂S, K₂S, and Na₂S can be used as a S²⁻ source to carry out the ligand exchange.³⁵ Therefore, the functional group should be S²⁻ after LE, with most of the –SH in 1-DDT being replaced by S²⁻ from (NH₄)₂S.

Fig. 7 ¹H NMR spectra of Cu_1.94S/Cu₄SnS₄ nanomaterials before and after the LE.

To clarify the presence and existing state of copper in the FA and (NH₄)₂S solution after LE, the lower layer of liquid after LE was separated and centrifuged to exclude any nanocrystals for further analysis. The colour of the original liquid is light yellow without any precipitate in it (Fig. 8a, left). We then dropped NaOH and Na₂S solutions into the sample to observe the colour changes. When NaOH solution was dropped in, there was a light change of colour and the solution turned to be slightly turbid (Fig. 8a, middle). However, some dark brown precipitates appeared within several minutes when Na₂S solution was dropped in (Fig. 8a, middle). It is well known that only free Cu²⁺ can react with OH⁻ to form Cu(OH)₂ precipitates, while S²⁻ has the ability to capture Cu²⁺ in the [Cu(NH₃)_n]²⁺ complex to form CuS precipitates.³⁶ The observation demonstrates the existence of copper and indicates that copper in the FA and (NH₄)₂S solution after LE mainly exists in the form of [Cu(NH₃)_n]²⁺, rather than free Cu²⁺. At the same time, the FA and (NH₄)₂S solution before LE were also tested by dropping the Na₂S solution (Fig. 8a, right), and no significant change in the solution was observed. Such phenomena further manifested that [Cu(NH₃)_n]²⁺, which further reacted with Na₂S to form the dark brown precipitate, was originated from the process of LE. In addition, we also conducted the UV-vis measurement on the FA and (NH₄)₂S solution before and after LE, as shown in Fig. 8b. There was a broad absorption in the range of 380–440 nm for the pure solution of FA and (NH₄)₂S. However, a broad and strong absorption peak at about 420 nm was observed for the sample after LE, which further indicates the presence of [Cu(NH₃)_n]²⁺ in the solution. The UV-vis spectra predicted by DFT calculations also show a wide absorption peak in the visible light region for [Cu(NH₃)₄]²⁺ ions.³⁷ The content of Cu element in the lower layer of FA and (NH₄)₂S solution after LE was determined using ICP, with a concentration of 43.4 mg L⁻¹ (Table S1†).

Fig. 8 (a) Digital images of the FA and (NH₄)₂S solution after LE, in the original state, and after titration using NaOH and Na₂S; the FA and (NH₄)₂S solution before LE titration by using Na₂S, and (b) UV-vis spectra of the FA and (NH₄)₂S solution before and after LE.

We have summarized the crystal structures of Cu_1.94S, Cu₇S₄, Cu₃SnS₄, and Cu₄SnS₄, as shown in Table 1. Interestingly, both Cu_1.94S and Cu₇S₄ are in the monoclinic crystal structure; meanwhile, Cu₄SnS₄ and Cu₃SnS₄ obtained the same crystal structure of an orthorhombic crystal. The similarity of the crystal structures may contribute to the crystal transformation during LE. It has been claimed that the Cu atoms in Cu_1.94S behave like “fluid” due to the reason that only partial copper sites are occupied by sulfur atoms.²⁸ We propose that during the LE process, partial loss of Cu atoms will result in the rearrangement of the copper and sulfur atoms with a decrease of the symmetry, thus the a and b values of Cu₇S₄ are about twice the value of Cu_1.94S. The phase transformation from Cu₄SnS₄ to Cu₃SnS₄ may also contribute to a similar reason.

Table 1 Lattice parameters of the four crystal structures

Formula	Structure	Space group	a/Å	b/Å	c/Å	Angle/°
Cu1.94S	Monoclinic	P21/n (14)	26.897	15.745	13.465	90.0 × 90.13 × 90.0
Cu7S4	Monoclinic	C2/m (12)	53.79	30.90	13.36	90.0 × 90.0 × 90.0
Cu3SnS4	Orthorhombic	Pmn21 (31)	6.525	7.523	37.662	90.0 × 90.0 × 90.0
Cu4SnS4	Orthorhombic	Pnma (62)	13.57	7.69	6.42	90.0 × 90.0 × 90.0

---

We have further studied the optical properties of the three samples before and after LE, as shown in Fig. 9. The optical absorption spectrum of the Cu_1.94S and Cu₄SnS₄ nanocrystals exhibits a broad range of absorbance from the UV to near-infrared (NIR) region. The Cu_1.94S/Cu₄SnS₄ nanocrystals also combine the optical properties of Cu_1.94S and Cu₄SnS₄ throughout the band. Upon comparing the UV-vis spectra of the samples before and after LE, it is observed that the spectrum generally undergoes a blue shift following the LE. The band gap (E_g) of the material is obtained by plotting the function of absorption coefficient and photon energy ((αhν)² ∼ hν), where α is the absorbance, h is the Planck constant, and ν is the radiation frequency. By extrapolating the linear part of the spectrum to intercept the intersection with the x-axis, the forbidden bandwidth of the material can be obtained. Fig. 9 shows that the band gaps of Cu_1.94S, Cu_1.94S/Cu₄SnS₄, and Cu₄SnS₄ are 1.17, 1.90, 1.74 eV, and 1.63, 1.62, 1.56 eV for the corresponding nanocrystals after LE, respectively. The nanomaterial combines the characteristics of two pure phases. Considering the slight change of the morphologies of nanocrystals after LE, the change of band gaps should be due to the change of crystal structures and elemental compositions.

Fig. 9 (a, d and g) UV-vis-NIR absorbance spectra and the plots of (αhν)²vs. hν of (b and c) Cu_1.94S, (e and f) Cu_1.94S/Cu₄SnS₄ nanomaterials, and (h and i) Cu₄SnS₄ nanocrystals before and after the LE.

Table 2 presents the specific elemental composition of Cu_1.94S, Cu_1.94S/Cu₄SnS₄ and Cu₄SnS₄ nanocrystals before and after LE. Among them, the Cu : S ratios of Cu_1.94S and Cu₇S₄ are approximately 1.96 : 1 and 1.74 : 1, respectively, which are almost consistent with the stoichiometry. The contents of Sn element in nanomaterial materials are relatively small, and the ratio of Sn in Cu_1.94S/Cu₄SnS₄ : Cu₇S₄/Cu₃SnS₄ is about 2 : 1. The Cu contents in Cu₇S₄/Cu₃SnS₄ are higher than the expected values, which indicate the copper-rich composition of each material. It may be due to the copper ion originated from the LE process, which was still bound to the surface of the existing nanocrystals.

Table 2 EDS of Cu_1.94S, Cu_1.94S/Cu₄SnS₄, and Cu₄SnS₄ before and after LE

Composite	Elements (at%)	Cu : Sn : S
		Cu	Sn	S
Cu1.94S	65.86	0.46	33.68	1.96 : 0.01 : 1
Cu7S4	62.93	0.90	36.17	1.74 : 0.02 : 1
Cu1.94S/Cu4SnS4	51.70	6.43	41.87	8.04 : 1 : 6.51
Cu7S4/Cu3SnS4	65.41	3.09	31.4	21.16 : 1 : 10.2
Cu4SnS4	47.07	12.01	40.92	3.92 : 1 : 3.41
Cu3SnS4	39.61	13.34	47.05	2.97 : 1 : 3.53

---

In this study, we further investigated whether the Cu₇S₄ nanocrystals obtained after LE can further react with SnCl₂·2H₂O to form ternary Cu–Sn–S compounds. A fixed amount of Cu₇S₄ nanocrystals of and SnCl₂·2H₂O were dispersed in different solvents of 1-DDT, OLA, and OA, respectively. Experiments were conducted at 230 °C for 30 min. The products were characterized using XRD, as shown in Fig. 10a. Interestingly, despite the difference of solvents, the same product of Cu₄SnS₄ nanocrystals can be obtained. This discovery suggests the possibility of further reaction between Cu₇S₄ and Sn sources to form Cu₄SnS₄ nanocrystals even without any sulfur source. However, when S powder dissolved in OLA was added to the OLA solvent containing SnCl₂·2H₂O, the resulting product exhibits mixed phases of Cu_1.8S, Cu₂SnS₃ and Cu₄SnS₄. This observation provides us the possibility to further dope Cu₇S₄ with Sn and form some possible Cu₇S₄-based heterostructures, while the intrinsic reason for the phase transformation is still under further investigation and will be reported elsewhere.

Fig. 10 XRD patterns of the products prepared with (a) different solvents of OLA, OA and 1-DDT, and (b) using S powder as an S source and OLA as a solvent.

4 Conclusions

In this study, a simple one-pot heat-up method was used to prepare multi-component copper-based sulfide nanomaterials. By adjusting the reaction temperature and time, the binary Cu_1.94S and ternary Cu₄SnS₄ nanomaterials can be successfully obtained. Besides, a consistent phase transformation was observed, revealing the partial loss of copper after LE, with the crystal structure of Cu_1.94S and Cu₄SnS₄ changing to Cu₇S₄ and Cu₃SnS₄. The phase transformation can mainly be ascribed to the loss of copper due to the strong complexation of Cu⁺ and ammonia with the formation of [Cu(NH₃)_n]²⁺ ions under aerobic conditions. Meanwhile, the edge of the as-prepared Cu_1.94S/Cu₄SnS₄ nanoplates changed to be inapparent with a slight decrease of crystal size after LE. Furthermore, UV-vis shows that the band gaps of all materials show an obvious change due to the change of crystal structures.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (No. 51425202 and 21506089).

References

1. Y. Chen, J. Wang, Y. Hong, Y. Yang, L. Tan, N. Li, C. Ma, J. Wang, X. Fan and Y. Zhu, J. Mater. Chem. A, 2023, 11, 6168–6180.
2. Z. Wen, H. Ren, S. Yao, C. Gu, S. W. Joo and J. Huang, J. Alloys Compd., 2023, 932, 167644.
3. X. Li, J. Cao, L. Yang, M. Wei, X. Liu, Q. Liu, Y. Hong, Y. Zhou and J. Yang, Dalton Trans., 2019, 48, 2442–2454.
4. D. Ren, R. Shen, Z. Jiang, X. Lu and X. Li, Chin. J. Catal., 2020, 41, 31–40.
5. J. Wang, S. Lin, N. Tian, T. Ma, Y. Zhang and H. Huang, Adv. Funct. Mater., 2021, 31, 2008008.
6. Z. Li, H. Li, S. Wang, F. Yang and W. Zhou, Chem. Eng. J., 2022, 427, 131830.
7. L. Zhou, Z. Liu, Z. Guan, B. Tian, L. Wang, Y. Zhou, Y. Zhou, J. Lei, J. Zhang and Y. Liu, Appl. Catal., B, 2020, 263, 118326.
8. A. C. Lokhande, K. V. Gurav, E. Jo, C. D. Lokhande and J. H. Kim, J. Alloys Compd., 2016, 656, 295–310.
9. M. Sun, X. Fu, K. Chen and H. Wang, ACS Appl. Mater. Interfaces, 2020, 12, 46146–46161.
10. J. Park, J. Joo, S. G. Kwon, Y. Jang and T. Hyeon, Angew. Chem., Int. Ed., 2007, 46, 4630–4660.
11. S.-H. Choi, K. An, E.-G. Kim, J. H. Yu, J. H. Kim and T. Hyeon, Adv. Funct. Mater., 2009, 19, 1645–1649.
12. D. Mott, J. Yin, M. Engelhard, R. Loukrakpam, P. Chang, G. Miller, I.-T. Bae, N. C. Das, C. Wang, J. Luo and C.-J. Zhong, Chem. Mater., 2009, 22, 261–271.
13. M. Kruszynska, H. Borchert, A. Bachmatiuk, M. H. Ruemmeli, B. Buechner, J. Parisi and J. Kolny-Olesiak, ACS Nano, 2012, 6, 5889–5896.
14. Y.-H. A. Wang, X. Zhang, N. Bao, B. Lin and A. Gupta, J. Am. Chem. Soc., 2011, 133, 11072–11075.
15. A. Singh, H. Geaney, F. Laffir and K. M. Ryan, J. Am. Chem. Soc., 2012, 134, 2910–2913.
16. X. Zhang, G. Guo, C. Ji, K. Huang, C. Zha, Y. Wang, L. Shen, A. Gupta and N. Bao, Sci. Rep., 2014, 4, 5086.
17. M. J. Turo and J. E. Macdonald, ACS Nano, 2014, 8, 10205–10213.
18. A. J. Nozik, M. C. Beard, J. M. Luther, M. Law, R. J. Ellingson and J. C. Johnson, Chem. Rev., 2010, 110, 6873–6890.
19. A. S. Brewer and M. S. Arnold, Thin Solid Films, 2014, 567, 91–95.
20. C. H. M. van Oversteeg, F. E. Oropeza, J. P. Hofmann, E. J. M. Hensen, P. E. de Jongh and C. de M. Donega, Chem. Mater., 2019, 31, 541–552.
21. J. Li, T. Jiu, G. H. Tao, G. Wang, C. Sun, P. Li, J. Fang and L. He, J. Colloid Interface Sci., 2014, 419, 142–147.
22. Q. Liang, J. Nanopart. Res., 2016, 18, 161.
23. Q. Tian, F. Jiang, R. Zou, Q. Liu, Z. Chen, M. Zhu, S. Yang, J. Wang, J. Wang and J. Hu, ACS Nano, 2011, 5, 9761–9771.
24. W. Aigner, G. K. Nenova, M. A. Sliem, R. A. Fischer, M. Stutzmann and R. N. Pereira, J. Phys. Chem. C, 2015, 119, 16276–16285.
25. X. Zhang, H. Huang, Y. M. Maung, J. Yuan and W. Ma, Chem. Commun., 2021, 57, 7906–7909.
26. H. Zhang, B. Hu, L. Sun, R. Hovden, F. W. Wise, D. A. Muller and R. D. Robinson, Nano Lett., 2011, 11, 5356–5361.
27. G. S. Biggs, O. J. Klein, S. L. Maslen, J. M. Skehel, T. J. Rutherford, S. M. V. Freund, F. Hollfelder, S. R. Boss and P. D. Barker, Angew. Chem., Int. Ed., 2021, 60, 10919–10927.
28. Y. Tang, M. Liu, H. Xu, M. Xu, X. Zhang and N. Bao, Nanotechnology, 2022, 33, 295701.
29. X. Zhang, Y. Tang, Y. Wang, L. Shen, A. Gupta and N. Bao, CrystEngComm, 2020, 22, 1220–1229.
30. J. Wang, T. Bo, B. Shao, Y. Zhang, L. Jia, X. Tan, W. Zhou and T. Yu, Appl. Catal., B, 2021, 297, 120498.
31. X. Li, H. Shen, Z. Niu, S. Li, Y. Zhang, H. Wang and L. Li, J. Am. Chem. Soc., 2010, 132, 12778–12779.
32. N. Sharma, T. Das, S. Kumar, R. Bhosale, M. Kabir and S. Ogale, ACS Appl. Energy Mater., 2019, 2, 5677–5685.
33. H. Liu, Z. Chen, Z. Jin, Y. Su and Y. Wang, Dalton Trans., 2014, 43, 7491–7498.
34. P. Behera, S. Karunakaran, J. Sahoo, P. Bhatt, S. Rana and M. De, ACS Nano, 2023, 17, 1000–1011.
35. A. Nag, M. V. Kovalenko, J. S. Lee, W. Liu, B. Spokoyny and D. V. Talapin, J. Am. Chem. Soc., 2011, 133, 10612–10620.
36. K. Li, Z. Chen, J. Li, X. Sun, F. Xu and L. Xu, Mater. Corros., 2018, 69, 1597–1608.
37. Z. Liu, X. Wu, T. Gao, M. Zhang, X. Min, W. Chen and C. Tang, Chem. Eng. J., 2024, 481, 148692.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4dt00309h

‡ These authors contributed equally to this work.

---