Synthesis of Bimetallic Zn–Fe Sulphide Nanoheterostructures with Improved Solar Energy Conversion

INTRODUCTION
The synthesis of metal sulfide semiconductors has received much attention because they possess tunable optical, electrical and catalytic characteristics [1-4]. Thus, they are potential materials for use in photocatalysis, optoelectronics and energy storage applications. Among the metal sulfides, zinc sulfide (ZnS) and iron sulfide (FeS) have been extensively studied because of their low cost, natural abundance and environmental friendliness [2,4].
Zinc sulfide (ZnS) is an II-VI compound semiconductor and has a broad band gap between 3.5 and 3.9 eV [5]. It mainly possesses two common crystal structures which include cubic sphalerite (zinc blende) and hexagonal wurtzite, and these two crystalline structures demonstrate differing optical and electronic properties [3,5]. It has been recognized that control over the phase and morphology (e.g., nanoparticles, nanorods, hierarchical nanostructures) of ZnS plays a role in improving its properties for applications in light-emitting diodes, sensors and photocatalysts [1,3,6,7].
The metal sulfide iron sulfide (FeS) exists in various allotropes including troilite (FeS), pyrite (FeS), marcasite (FeS) and greigite (FeS), and they possess different physical, chemical and magnetic properties due to various crystal structures [8,9]. FeS and their derivatives have potential for battery, supercapacitors, catalysts and remediation applications because of their good electric conductivity and redox characteristics [9,10]. In these structures, different crystalline phase, conductivity and stability have also led to different performance [9]. The synthetic methods have been proved to have great influence over phase and morphology of FeS and hence their applications [8,10].
The composite made up of ZnS and FeS have recently become a growing field and these composites represent a new materialsystem trying to integrate the optical and wide band gap property of ZnS with the electrical conductivity and catalytic activity of iron sulfide [11], [12]. Through formation of heterojunction interfaces of ZnS and FeS, electron-hole recombination can be suppressed efficiently which could lead to improved efficiency for photocatalytic and electrochemical applications [11]. Thus they have the advantages over the individual compound sulfides in performance and stability [12,13].
Several methods of synthesis of ZnS, FeS and their composites including co-precipitation, solvothermal and hydrothermal, and chemical bath deposition have been explored [2,7,14]. Precise control over the phase, particle size and morphology has been achieved by tuning synthetic parameters and these factors significantly influence the properties. Hence the synthesis of ZnS, FeS and their composites remains an important area for high-performance functional materials.

MATERIALS AND METHODS
ZnS, FeS and Zn-FeS composites were synthesized by using Na2S as the precipitating agent through a chemical precipitation process. Firstly, the aqueous precursors were prepared by dissolving the needed metal salts in deionized water. For ZnS, 2M ZnCl2 solution was prepared by dissolving 27.26g ZnCl2 in 100mL deionized water. For FeS, 1M FeCl2 solution was prepared by dissolving 16.29g FeCl2 in 100mL deionized water. For the composite sample, mixed solution containing 2M ZnCl2 and 1M FeCl2 was prepared in 100mL deionized water. All the solutions were heated on a hot plate for 30min under constant stirring in order to get uniform solution. In a separate container, 2M Na2S solution was prepared by dissolving 31.22g Na2S in 200mL deionized water and heated for 30min for full dissolution. Dropwise addition of Na2S solution using burette to above solutions (ZnCl2 solution, FeCl2 solution and mixed ZnCl2-FeCl2 solution) was performed until precipitation occurred. PH of the resulting mixture was 2.91, 6.76 and 6.43 for ZnS, FeS and Zn-FeS composite respectively. After precipitaion, the mixture was stirred on a magnetic stirrer at 60 o C for 3 h to ensure that the precipitation is complete. The precipitates obtained were filtered and washed many times with deionized water to remove impurities and the obtained products were dried in oven at 100 o C for 3 h for ZnS, FeS and Zn-FeS composite powder formation.

RESULTS AND DISCUSSION
The XRD diffractograms for the synthesised nanomaterials are shown in Fig. 1. It can be seen that zinc sulfide had 4 diffraction reflections and a peak maximum at 2 = 28.6646 (d = 3.11432). Iron sulfide on the other hand had 7 diffraction reflections, and a peak maximum at 2 = 26.1031 (d = 3.41384). Zinc-iron sulfide had 4 reflections and a peak maximum at 2 = 29.2014 (d = 3.05829). The diffraction shown by these samples is typical for powder diffraction of crystalline solids [15-18].
The identified peaks match with triclinic ZnS, monoclinic FeS and monoclinic zinc-iron sulfide composite phases, following the crystallographic database values [23]. XRD results illustrate the utilization of sodium sulfide (NaS) as sulfide source to obtain the metal sulfides through precipitation-a commonly practiced route of synthesized sulfide nanoparticles [22-24]. Zinc and iron sulfides form crystals in the triclinic structure (space group P1) while the composite material crystallizes in monoclinic (P2/m). This structural transformation is a result of the strong interaction between Zn and Fe ions, causing the crystal lattice to re-organize to an even higher symmetric crystalline phase [22].
Lattice parameters further demonstrate this change. ZnS and FeS shows triclinic shape having no angle 90, while the composite crystal lattice form a monoclinic crystal having two angles 90, consistent to crystallographic laws [22,24]. It shows the lattice distortion because of incorporation of ions in the lattice. The unit cell volume calculated was 311.45, 489.50 and 366.85 for ZnS, FeS and composite, suggesting that the resulting composite material has a mixed lattice rather than physical mixture [17,22] as clearly seen in Tab 1.
The determined densities of ZnS, FeS and composite (4.365 g/cm, 4.175 g/cm and 4.857 g/cm respectively) prove that the atomic arrangement is more packed and stable for the composite system-a typical property of mixed metal sulfides [22,24]. All these parameters indicate that zinc-iron sulfide nanocomposite have been successfully synthesized having modifications in crystal structure, lattice parameters and improved packing, thus potentially affecting the properties of the composite [15,17].
The crystal structures of nanomaterials are shown in Fig. 2. This figure Shows how the atoms are ordered in the nanomaterial. By seeing how the atoms are ordered it would tell you more about the physical and chemical nature of the material. The order in relation to crystallite size and sintering temperature for nanomaterials is key in the synthesis and characterization of nanomaterials [21,22].
FESEM images were also observed to check the morphological features (shape, distribution and size of particle) of zinc sulfide, iron sulfide and the Zn-Fe sulfide composite. Fig. 3 displays the morphology of zinc sulfide nanomaterials; the particles are nearly spherical with cauli flower-like cluster morphology. The size of the primary particle is from 20-60 nm and some agglomerate can reach to a few hundreds nanometer. Particles are closely distributed and agglomerate tightly because of the large surface energy [25,26], morphology is also relatively homogeneous among the field.
But the structure of FeS presents an interconnected pore network instead of separate particles. There are no uniform and regular parts in the porous FeS. The whole is irregular ligament network with cavity structures (30-80 nm in thickness) and distribution heterogeneity is relatively high (Fig. 3) ponge-like structure of is commonly resulted from anisotropic growth under synthesis in solution system [27,28]. As can be seen from Fig. 3, the Zn-Fe sulfide composite takes the shape of hybrid structures: well-dispersed ZnS nanoparticles (25-70 nm) covered on/among a porous FeS matrix structure. The well-dispersed feature suggests strong interfacial contact and mixing ability but moderate heterogeneity still exists since hybrid structure has two different parts. These hybrids greatly promote interfacial area and active sites and hence enhance the function properties [29,30].
EDS also proves the elemental composition of ZnS (Zn, S), FeS (Fe, S) and the composite (Zn, Fe, S). As trace amount of Au peaks in Fig. 3 is result of carbon coating, small fluctuation of stoichiometry should be ascribed to sulfur vacancies or oxidation [31]. Summary, according to FESEM-EDS, ZnS is homogeneous accumulated nanoparticles, FeS is irregular porous network, whereas composite is hybrid structure with favorable interfacial contact.
Zinc sulfide has two typical first order phonon modes at 274 cm-1 (TO) and 347 cm-1 (LO) attributed to the Zn-S vibrations in cubic zinc blende structure as supported by XRD (PDF 01-077-2100; 00-002-0564). Since this structure possesses high symmetry (F-43m) only TO and LO modes are Raman active, additional broad peaks arise from confinement of phonons and defects (related to nanocrystallinity and sulfur vacancies) [32,33]. Iron sulfide has multiple Raman peaks in the region of 250-450 cm-1 attributed to Fe-S vibrations; corresponding to troilite/pyrrhotite structure confirmed by XRD (PDF 00-003-0822; 00-049-1632) [34]. Its low symmetry and the deviation in the stoichiometry (FeS) provide more Raman active modes and the additional peaks between 640-720 cm-1 may arise due to multiphonon or defect related phenomena (Fig. 4).
With addition of iron, in the mixed metal sulfides Zn-Fe-S, XRD indicates the presence of an additional mixed metal sulfide phase thus showing a ternary or solid-solution type structure. Raman spectrum of the Zn-Fe-S shows shifts and broadening in the peaks corresponding to Zn-S and Fe-S vibrations and some new peaks. These indicate lattice distortion, replacement of cations (crossover effect) and reduced symmetry causing relaxed selection rules. The defect related and multiphonon scattering is enhanced and facilitated by fast sodium sulfide precipitation, promoting cation mixing in the ternary structure.
Fig. 5 also presents the images from atomic force microscopy (AFM) for surface morphology and roughness of zinc sulfide, iron sulfide and zinc-iron sulfide composite produced using sodium sulfide. A difference in surface morphology, grains distribution and roughness of samples can clearly be seen from the figures. The surface of the zinc sulfide is relatively homogeneous and densified with nanograins spread densely. The roughness is quite small and values for Ra=1.01nm, Rq=1.30nm and Rt=12.38nm is observed in Table 2 that show very smooth nanosurface.
The small Gaussian height distribution is indicative of controlled growth and uniform nucleation.Smooth morphologies were previously correlated with stable growth conditions of sulfide nanomaterials [35,36]. In the case of iron sulfide, however, the surface is much rougher and less uniform. The large values of Ra (8.77nm), Rq (10.93nm) and Rt (76.69nm) show considerable variations and undulations of the surface and between features. AFM images indicated large particles agglomerated together with uneven distributions, suggesting rapid nucleation and growth in the alkaline sulfide environment favoring particle growth and agglomeration [37,38].
Intermediate behavior of the zinc-iron sulfide composite as can be observed by the roughness values Ra=7.53 nm, Rq=9.25 nm, Rt=55.89 nm between Zinc and Iron sulfide. And by the morphology a more uniform particle distribution and less aggregation than Iron Sulfide. The growth of iron sulfide seems to be hindered by the incorporation of zinc which promotes a more uniform structure. This behavior has been observed in mixed metal sulfides where cation substitution affects growth kinetics [39, 40]. It suggests that zinc might be acting as a stabilizing factor which is preventing large growth and excessive aggregation. Smooth surfaces of ZnS are good for optoelectronics applications, whererough surface of FeS will be beneficial for catalytic properties [36, 38]. A balance between the uniformity and roughness make it suitable for multifunctional applications. This shows AFM confirms that the use of sodium sulfide as the precipitating agent allowed to get surfaces of controlled roughness, and that the incorporation of Zn leads to a significantly improved uniformity and maintainance of nanoscale roughness.
Fig. 6 presents UV-Vis electronic spectra of Zinc sulfide, Iron sulfide and (Zn-Fe) sulfide composite that was synthesized by sodium sulfide. The absorption edges for ZnS, FeS and the composite (Zn-Fe) Sulfide were 461 nm, 954 nm and 700 nm and correspond to band gaps of 2.69, 1.30 and 1.77 eV respectively. The band gaps are calculated using ( Eg = 1240/lambda). Zinc sulfide have a rather wide band gap (2.69 eV) with the absorption confined to the UV region. This is smaller compared to bulk Zinc sulfide which is ~ 3.6 eV. The lower band gap value could be related to the formation of defects, sulfur vacancies or the quantum effects at nanostructures which is related to chemical synthesis of the particles [41, 42]. Optical transitions are attributed to transitions between the S 3p valence band and the Zn 4s conduction band, and d-d transitions are forbidden due to the presence of filled Zn 3d orbitals. A sharp absorption edge indicates direct transition and presence of few defect states.
Iron sulfide possesses a very narrow band gap (1.30 eV) with the absorption spanning over the entire range from UV to the visible and even to near-infrared region. This is attributed to electronic states of Fe 3d orbitals and includes charge-transfer transitions (S 3p Fe 3d) d-d transitions and defect states [43, 44]. The existence of a broad absorption tail or Urbach tail implies highly disordered structure with large defect density, therefore suitable for solar energy conversion or photocatalysis applications. The (Zn-Fe) sulfide composite shows intermediate band gap (1.77 eV), and there is a strong interaction between the two sulfides on electronic level. The absorption red shifts with respect to ZnS as the incorporation of Fe causes the band gap narrowing and Fe 3d levels come to contribute to electronic band states in impurity levels [42, 45]. It is due to incorporation of Fe atoms and hence forming heterojunction between ZnS and FeS, or hybridization of d-orbitals or Fe substitution with Zn ions [45]. The reduction of the band gap helps it absorb a wider spectrum of visible light and also enhance charge separation, which would improve the performance as a photocatalyst. The precipitating agent sodium sulfide led to quick nucleation but has the drawbacks of high concentration of sulfur vacancies, lattice strain leading to localized electronic states, thus to band gap narrowing [46].
Fig. 7 compares theTG-DTA thermograms of zinc sulfide, iron sulfide and (zinc-iron) sulfide composite. Sodium sulfide is a source of S2 ions directly; so the formation of metal sulfides was very fast. The TG-DTG profiles were mainly correlated to the lost adsorbed water, subtle lattice restructure and stability as previously reported in the sulfides synthesized by inorganic sulfur sources [47,48]. TheZnS sample hasa total mass loss of about13.7% in range of 30-160 °C, which attributed to adsorbed water and solvent. A DTG peak located at 110-120 °C means mass loss from surface adsorbed specie (low bond strength). From 160 °C to 600 °C the mass shows an extremely stable plateau which represents the solid nature of the strongly bound Zn-S linkage and high stability which is characteristic of well-crystallized zinc sulfide nano particles [49]. The FeS also reveals a significant mass loss (about24.7%) in two steps (30-200 °C for moisture lose, 200-320 °C for structural rearrangement and loosely bound S lose). A higher mass loss and a main peak around 280-300 °C may mean more defects or other metastable phases exist in the structure. After 350 °C the FeS became quite stable as reported previously [50]. The least mass loss (~3.36%) for the (Zn-Fe)S composite shows the greatest thermal stability. A slightly lower mass loss below 100 °C corresponds to the loss of moisture, the second smaller peak between 170-300 °C relates to some slight lattice rearrange; above 300 °C it also remains very stable. It may result from the enhancement of stability due to interaction of ZnS and FeS phases [51].
TGA data reveal the kinetic and thermodynamic behavior of zinc sulfide, iron sulfide, and their composite as demonstrated in Table 3. Iron sulfide shows the highest thermal stability with ΔEa of 77,425 J/mol, followed by zinc sulfide (61,682 J/mol), while the composite has the lowest (49,790 J/mol) due to synergistic interactions, defects, and better heat transfer that ease decomposition [52]. Enthalpy follows suit: iron sulfide and zinc sulfide need more heat, but the composite’s low ΔH (7,614 J/mol) reflects structural disorder and lattice strain [53].
All samples have positive S, which represents an increase in the degree of disorder due to phase transition or gas evolution, with ZnS slightly higher than the others. Positive G means non-spontaneous reaction (both ZnS and FeS are positive), the negative G value in the composite (49,486J/mol) confirms the decomposition reaction (Fig. 5) is spontaneous [54]. It indicates that the composite would consume less energy, be more reactive and easier to be used for catalysis and energy applications [55].
The carrier types of the samples fabricated in the presence of NaS are n-type, the negative Hall coefficient was tested for the three samples and recorded in Table 4. It shows a sulfur-rich growth condition is employed, creating donor-type defects, for example sulfur interstitials and metal vacancies, so that the electrons are the dominant carriers [56,57].
The carrier concentrations and the relatively low mobility of the individual sulfide shows that the carrier scattering exists strongly at the grain boundaries and defect sites. Compared with single phase ZnS and FeS, the composite material exhibits an enhancement of conductivity and carrier mobility, which is mainly due to better carrier percolation and lower interfacial resistance of heterostructure network. There are many literatures reported the transport properties of sulfide composite materials prepared by rapid precipitation similar improvement [58,59].
The solar cell applications of the synthesized nanomaterials were tested as shown in Fig, 8. The J–V characteristics reveal clear differences in the photovoltaic performance of zinc sulfide, iron sulfide, and the (Zn–Fe) sulfide composite. Iron sulfide shows the highest efficiency (4.8%), followed by the composite (4.2%), while ZnS has the lowest efficiency (1.8%) as explained in Table 5. This trend is mainly due to variations in short-circuit current density (Jsc) and fill factor (FF).
The iron sulfide and the composite demonstrate high values for Jsc (19 and 18mA/cm2) which are expected values for narrow-bandgap materials with high absorption and charge generation [60, 61] whereas zinc sulfide demonstrates very low Jsc values (5.5mA/cm2) and low absorption. Zinc sulfide had the highest Voc (0.82V) which is attributed to the wide bandgap and low recombination within the material [62]. Fill factor was highest for iron sulfide (0.49) meaning less internal losses and lowest for zinc sulfide (0.40) as the material has higher resistive losses. Based on Jsc, Voc and FF values calculated and graphed efficiency is highest for iron sulfide due to the high current and the good FF value, Iron sulfide is the most efficient absorber material, composite has average absorption material qualities whereas zinc sulfide is more of an effective window or buffer layer rather than an absorption layer [63, 64].

CONCLUSION
We successfully prepared a Zn-Fe sulfide nanocomposite via simple chemical precipitation which showed good interaction between Zn and Fe ions, thus forming a mixed phase structure. A typical hybrid morphology which composed of spherical ZnS nanoparticles embedded into a porous FeS matrix was observed, which resulted in better surface area and interfacial contact. From optical properties, the band gap was narrowed down to 1.77 eV indicating enhanced visible light absorption, and electrical measurements reveal the n-type conductivity and charge transport were enhanced. Enhanced thermal stability also existed in composite as comparing to individual sulfides. Photovoltaic results indicated that the power conversion efficiency reached 4.2% which could be attributed to the enhanced separation efficiency and suppressed recombination. TheZn-Fe sulfide nanocomposite could be an ideal candidate as a cheap and high-efficiency material for photovoltaic application.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interests regarding the publication of this manuscript.