Fabrication of crystalline $\hbox {Sb}_{2}\hbox {S}_{3}$ sheaf structure composed of nanorods by a hot-injection method

Abstract

Herein, we demonstrate a hot-injection method towards the fabrication of $\hbox {Sb}_{2}\hbox {S}_{3}$. The evolution from amorphous $\hbox {Sb}_{2}\hbox {S}_{3}$ nanoparticles to a sheaf of $\hbox {Sb}_{2}\hbox {S}_{3}$ nanorods cross-linked together occurring with the increase of reaction temperature and time is studied. The structural, compositional and morphological features of $\hbox {Sb}_{2}\hbox {S}_{3}$ products indicate the formation of crystalline $\hbox {Sb}_{2}\hbox {S}_{3}$ with orthorhombic phase and high purity. As the reaction proceeds, it can be observed that the individual nanorod grows along the elongated direction (c-axis), and it is noteworthy that its round cross-section could develop into the rectangular cross-section subsequently. In this study we also propose a possible formation mechanism for the growth process of $\hbox {Sb}_{2}\hbox {S}_{3}$, and it reveals that the present hot-injection could provide an ideal growth environment for synthesizing the nanostructured materials by optimizing the experimental parameters.

1 Introduction

The newly distinct one-dimensional (1-D) nanostructure materials are promising research area, owing to their extraordinary optical, electronic, catalytic and mechanical properties. $\hbox {Sb}_{2}\hbox {S}_{3}$ (stibnite) is a promising semiconductor with abundant elemental storage and excellent stability, which has been particularly appealing as a new kind of light absorbing material in high-performance solar cells [1,2,3], owing to its suitable bandgap (${\approx }1.7\,\hbox {eV}$) and strong absorption coefficient ($1.8 \times 10^{5}\,\hbox {cm}^{-1})$. Meanwhile, it has a range of potential applications in photocatalysis [4] and energy storage [5]. Up to date, $\hbox {Sb}_{2}\hbox {S}_{3}$ 1-D materials including nanorods [6, 7], nanoribbons [8,9,10] and nanotubes [11, 12] have been successfully synthesized. Pursuing convenient method to fabricate 1-D nanostructure with controllable size and shape is an interesting research work for the wider application fields. A solution-phase method for preparing $\hbox {Sb}_{2}\hbox {S}_{3}$ 1-D materials, which is considered to be more promising, with the advantages of low cost and easy fabrication, was seldom reported in the past few years [13,14,15]. For example, Chen et al [16] obtained complex $\hbox {Sb}_{2}\hbox {S}_{3}$ hierarchical nanostructure by a hydrothermal reaction at $180^{\circ }\hbox {C}$ for 12 h, with the assistance of polyvinyl pyrrolidone (PVP). Wu et al [17] fabricated the double cauliflower-like $\hbox {Sb}_{2}\hbox {S}_{3}$ by using hydrothermal method for a much longer time of 30 h. So these reported methods usually require the longer reaction time to complete the crystallization process.

Herein, high-purity $\hbox {Sb}_{2}\hbox {S}_{3}$ sheaf composed of nanorods (1.2–$1.6\,\upmu \hbox {m}$ in length and 112 nm in width) is reported to be produced by using a simple hot-injection synthesis with several minutes. The uniformity in morphology and the copiousness in quantity suggest an ideal growth environment for the fabrication of crystalline $\hbox {Sb}_{2}\hbox {S}_{3}$ with this method. The formation mechanism for the growth process was also investigated in this work. It could be proved that the present strategy is an effective route to facilitate the nucleation-growth process and manipulate the microstructure of nanomaterials.

Fig. 1

(a) XRD patterns and (b) Raman spectra of $\hbox {Sb}_{2}\hbox {S}_{3}$ nanorods synthesized at different temperatures and times.

2 Experimental

Antimony trichloride ($\hbox {SbCl}_{3}$, 99.9%), 1-octadecene (ODE, 90%) and oleylamine (OAm, 80–90%) were purchased from Aladdin. Sulphur powder (S, 99.99%) was obtained from Sigma-Aldrich. All the materials were used directly without further purification.

In a typical process, 0.16 g sulphur powder was initially added into a 50 ml three-neck round-bottom flask containing 5 ml of OAm, followed by degassing with a flow of nitrogen. The solution was heated and stirred magnetically at $90{^\circ }$ until the sulphur powder was completely dissolved. Then, 0.0456 g $\hbox {SbCl}_{3}$ was dissolved in 3 ml of OAm and 2 ml of ODE. The antimony precursor solution was heated up to a higher temperature (in the range of 180–$210^{\circ }\hbox {C}$), under nitrogen flow and strong magnetic stirring. Then the as-prepared sulphur precursor solution was injected into the reaction system and remained for 5–20 min. Finally, the precipitation was collected through centrifugation.

The crystalline structure was measured by X-ray diffraction (XRD, BRUKER D8 ADVANCE) using Cu radiation operated at 40 kV and 40 mA.  The morphologies and compositions were obtained by using scanning electron microscope (SEM, FEI Nova Nano SEM450) and the energy-dispersive X-ray (EDX) spectroscopy, respectively.  The high-resolution transmission electron microscope (HRTEM) and the corresponding selected-area electron diffraction (SAED) were performed with a PHILIPS CM200 microscope operating at 200 kV accelerating voltage. X-ray photoelectron spectroscopy (XPS) was measured by Thermo SCIENTIFIC ESCALAB 250Xi X-ray photoelectron spectrometer with Al $\hbox {K}\!\upalpha$ radiation (1486.6 eV). Raman spectrum analysis was performed on a HR Evolution Raman spectrometer, using an Ar-ion laser excitation with a wavelength of 532 nm. Absorption spectrum was recorded using a UV–Visible spectrophoto meter (Perkin Elmer Lambda 950).

3 Results and discussion

The XRD patterns of $\hbox {Sb}_{2}\hbox {S}_{3}$ samples synthesized at different temperatures are shown in figure 1a. When the hot-injection is performed at $180^{\circ }\hbox {C}$ for 5 min, no obvious diffraction peaks of the sample are found, indicating that the sample is essentially amorphous. Peaks become sharper as the temperature increases to $190^{\circ }\hbox {C}$, although the intensity and shape of the diffraction peaks reveal that the sample is still not well crystallized. With further elevating the temperature up to $210^{\circ }\hbox {C}$ and increasing the reaction time up to 20 min, the peak intensities steadily become stronger, demonstrating an enhancement of the crystallization. All the clear peaks can be well indexed to the orthorhombic phase of stibnite $\hbox {Sb}_{2}\hbox {S}_{3}$ (cell constants: $a = 11.24\,{\AA }$, $b = 11.31\,{\AA }$, $c = 3.84\,{\AA }$; JCPDS 42-1393). No characteristic peaks corresponding to impurities are detected, indicating that the as-prepared samples possess a highly pure phase. Raman spectrum was also performed to further investigate the formation of crystalline $\hbox {Sb}_{2}\hbox {S}_{3}$, as shown in figure 1b. For the sample synthesized at $180^{\circ }\hbox {C}$, the broad single peak at 293 $\hbox {cm}^{\mathrm {-1}}$ owing to the symmetric vibrations of $\hbox {SbS}_{3}$ pyramid [12, 18] implies the weak crystallization. Then the peak is found to be splitted into two peaks at 280 and $305\,\hbox {cm}^{-1}$, which become much clearer as the temperature increases up to $210^{\circ }\hbox {C}$, suggesting the enhanced crystallinity of stibnite $\hbox {Sb}_{2}\hbox {S}_{3}$ [1, 19]. These results are in accordance with the XRD.

Fig. 2

A typical XPS of $\hbox {Sb}_{2}\hbox {S}_{3}$ synthesized at $210^{\circ }\hbox {C}$ for 5 min: (a) survey spectrum, (b) Sb3d and (c) S2p.

Fig. 3

TEM images of $\hbox {Sb}_{2}\hbox {S}_{3}$ synthesized at 180–$200^{\circ }\hbox {C}$ for 5 min: (a) 180, (b) 190 and (c) $200^{\circ }\hbox {C}$; the inset in (a) is the corresponding SAED pattern; the insets in (b) and (c) are the corresponding HRTEM images, respectively.

Figure 2a–c shows the typical XPS of $\hbox {Sb}_{2}\hbox {S}_{3}$ synthesized at $210^{\circ }\hbox {C}$. The survey spectrum confirms the presence of Sb and sulphur elements. The two peaks at 529.12 and 538.56 eV can be assigned to Sb $3\hbox {d}_{\mathrm {5/2}}$ and Sb $3\hbox {d}_{\mathrm {3/2}}$, respectively. The peak at 161.21 eV is assigned to the binding energy of S $2\hbox {p}_{\mathrm {3/2}}$. The results are consistent with the reported values in the literature [20]. The weak intensity of S $2\hbox {p}_{\mathrm {1/2}}$ signal in 162.31 eV could be ascribed to a small amount of oxidation state of sulphur [21].

The morphology and structure of the products were also examined using TEM and SAED. When the reaction temperature is set at $180^{\circ }\hbox {C}$, the amorphous orange-red particles with the irregular shape in sizes ranging from 60 to 80 nm are immediately formed (figure 3a). It is worthwhile to note here that some adjacent nanoparticles join with each other, forming the connected clusters, as indicated by the dotted oval in the image. In figure 3b, it could be found that the small disordered bundles of nanorods are produced at $190^{\circ }\hbox {C}$ for 5 min. Moreover, the obtained sample still consists of a large amount of amorphous nanoparticles and a small amount of short nanorods, ascribed to the incomplete reaction. As shown in the inset of figure 3b, the ambiguous lattice fringes demonstrate the poor crystallinity for this sample as well. It is inferred that the formation of bundle is probably due to the growth of each particle seed in the cluster. At $200^{\circ }\hbox {C}$, the longer bundles with simple splitting are formed (figure 3c). The fringe spacing of 0.79 nm closely matches the interplanar spacing of the (110) plane of the orthorhombic phase of $\hbox {Sb}_{2}\hbox {S}_{3}$.

Fig. 4

TEM images, HRTEM images and SAED patterns of $\hbox {Sb}_{2}\hbox {S}_{3}$ nanorods synthesized at $210^{\circ }\hbox {C}$ for different reaction times: (a–c) 5, (d–f) 10 and (g) 20 min.

Fig. 5

SEM images of samples synthesized at (a) 190 and (b) $200^{\circ }\hbox {C}$; (c) EDX spectrum corresponding to (b); SEM images of samples synthesized at $210^{\circ }\hbox {C}$ for (d) 5, (e) 10 and (f) 20 min.

Fig. 6

Schematic of growth mechanism for the as-prepared $\hbox {Sb}_{2}\hbox {S}_{3}$ with different morphologies.

In order to further evaluate the influence of reaction time on the morphology of product in hot-injection synthesis, figure 4 presents TEM images of the sample synthesized at $210^{\circ }\hbox {C}$ for 5–20 min. In figure 4a, it clearly indicates that a lot of rod bundles appear as a sheaf of nanorods cross-linked in the middle, forming a sheaf-like microstructure. In the inset of figure 4a, the lattice spacing is about 0.30 nm corresponding to the (121) plane of a stibnite $\hbox {Sb}_{2}\hbox {S}_{3}$ structure. The angle between the (121) plane and growth direction is $52.9^{\circ }$, and thus it can be inferred that the nanorod grows along the c-axis, i.e., [001] direction. Figure 4b and c are the HRTEM image and SAED pattern, corresponding to an isolated nanorod, taken from the sample presented in figure 4a. As expected, both the lattice spacing and the SAED pattern can be assigned to the orthorhombic phase of $\hbox {Sb}_{2}\hbox {S}_{3}$, which could be evidenced by the literature with the similar reported data for $\hbox {Sb}_{2}\hbox {Se}_{3}$ nanoribbon [10]. Figure 4d shows a TEM image of $\hbox {Sb}_{2}\hbox {S}_{3}$ sample synthesized at $210^{\circ }\hbox {C}$ for 10 min, where the inset is a corresponding HRTEM of an individual nanorod. A careful inspection indicates that the crystalline $\hbox {Sb}_{2}\hbox {S}_{3}$ sheaf structure is made up of a lot of closely packed and well-defined prismatic nanorods with uniform widths (112 nm) and lengths (1.2–$1.4\,\upmu \hbox {m}$). The discernible lattice fringes could reveal that the high-quality $\hbox {Sb}_{2}\hbox {S}_{3}$ product is obtained, without lattice defects (e.g., dislocation). Figure 4e indicates that the lattice spacings of 0.56 and 0.36 nm in the individual nanorod match well with the interplanar spacing of the (020) and (101) planes of stibnite $\hbox {Sb}_{2}\hbox {S}_{3}$, respectively, and the angle of $71.2^{\circ }$ is in good agreement with the theoretical value. The corresponding SAED pattern (figure 4f) taken from [$\overline{2}02$] zone axis provides a further evidence to support an orthorhombic-structured $\hbox {Sb}_{2}\hbox {S}_{3}$. The length of nanorods becomes much longer ($1.6\,\upmu \hbox {m}$), while the reaction time increases up to 20 min, as shown in figure 4g.

Figure 5a presents the SEM image of $\hbox {Sb}_{2}\hbox {S}_{3}$ sample fabricated at $190^{\circ }\hbox {C}$ for 5 min, clearly demonstrating that it exhibits rough morphology, except for very few bundles. The compact bundles of nanorods with uniform lengths ($1.1\,\upmu \hbox {m}$) and diameters (80 nm) are clearly observed for $\hbox {Sb}_{2}\hbox {S}_{3}$ product prepared at a higher temperature of $200^{\circ }\hbox {C}$ (figure 5b). Both Sb and sulphur elements are observed in the EDX spectrum, as shown in figure 5c. The atom ratio of Sb:sulphur is 1:1.45, close to 2:3, suggesting an appropriate chemical composition of $\hbox {Sb}_{2}\hbox {S}_{3}$ in the nanorods. Further increasing the reaction temperature to $210^{\circ }\hbox {C}$, a large quantity of sheaf composed of uniform and slippery $\hbox {Sb}_{2}\hbox {S}_{3}$ nanorods can be observed, as illustrated in figure 5d. Prolonging the time subsequently could greatly reduce the amorphous phase, yielding well-crystallized and much longer nanorods with rectangular cross-section (figure 5e and f). The phenomenon is generally in agreement with that observed in TEM.

Here, we propose a possible formation mechanism for the growth process of $\hbox {Sb}_{2}\hbox {S}_{3}$, as described in figure 6. At the beginning of reaction, amorphous $\hbox {Sb}_{2}\hbox {S}_{3}$ nanoparticles are formed quickly. The collision of adjacent nanoparticles could result in the formation of agglomerated cluster while the dissolution and nucleation are going on. With the increase in temperature, the growth of each nanoparticle along the oriented c-axis direction produces the nanorod. Therefore, these primary nanorods tend to form the bundle-like structure. Further increasing the temperature up to a much higher temperature, higher energy will be involved in the reaction, inducing a self-assembly formation of sheaf-like structure made up of several nanorods with round cross-section. Subsequently, it could develop into the longer prismatic nanorods with rectangular cross-section, with prolonging the reaction time, due to the enhancement of crystallinity. In general, the uniformity in morphology and the copiousness in quantity of the crystalline $\hbox {Sb}_{2}\hbox {S}_{3}$ product fabricated by using the present hot-injection approach, imply an ideal growth environment in the reaction system. Furthermore, UV–Vis absorption spectra in figure 7 demonstrate a red shift of the absorption onset ranging from 722 to 783 nm. Thus, the bandgap value (1.58–1.72 eV) of the as-prepared $\hbox {Sb}_{2}\hbox {S}_{3}$ is close to its optimum value for photovoltaic application, which is the focus in our forthcoming work.

Fig. 7

UV–Vis absorption spectra of $\hbox {Sb}_{2}\hbox {S}_{3}$ products.

4 Conclusions

In the present work, we succeeded in synthesizing high-quality $\hbox {Sb}_{2}\hbox {S}_{3}$ 1-D material using a hot-injection method. The composition and purity of the as-prepared $\hbox {Sb}_{2}\hbox {S}_{3}$ sample were examined by XRD, Raman spectrum and XPS analysis. All the results reveal that the well-crystallized stibnite $\hbox {Sb}_{2}\hbox {S}_{3}$ with orthorhombic phase and high purity could be formed under the proper reaction conditions. The optical bandgap is calculated to be 1.58–1.72 eV. Under certain hot-injection condition, the evolution from amorphous $\hbox {Sb}_{2}\hbox {S}_{3}$ nanoparticles to a sheaf of $\hbox {Sb}_{2}\hbox {S}_{3}$ nanorods occurs with the increase in reaction temperature and time. At early stage of the nucleation process, the agglomerated cluster of $\hbox {Sb}_{2}\hbox {S}_{3}$ amorphous particles can be formed via the collision among the adjacent nanoparticles. Then, the growth of each nanoparticle along the oriented c-axis direction produces the nanorod. These $\hbox {Sb}_{2}\hbox {S}_{3}$ nanorods develop into a bundle structure made up of nanorods with the round cross-section and the length of ${\sim }1.1\,\upmu \hbox {m}$, in subsequent growth stages, and complete their development as a sheaf of longer prismatic nanorods (1.2–$1.4\,\upmu \hbox {m}$ in length), owing to the enhanced crystallinity. It would be expected to extend this reaction system for growing the other high-quality nanocrystals with novel morphologies.