Effect of tin on bandgap narrowing and optical properties of ZnO–Zn2SnO4 electrospun nanofibre composite

The quest for improved visible light absorption in ZnO-based photocatalysts and photodetectors is a prevailing research problem. One method of solving this problem is by narrowing the band gap of this material well into the range of visible light energy. In this work, we have demonstrated considerable band narrowing into the visible region in our electrospun ternary ZnO–Zn2SnO4 nanofibres. We also showed that band narrowing is highly dependent on the amount of Sn in the synthesized composite which could be attributed to the creation of localized states in the ZnO band and the attendant structural changes. However, beyond 15%wt doping there were observed changes in the trend of the bandgap reduction, crystallite sizes, and width of the localized states in the ternary ZnO–Zn2SnO4 band. The details of experimental procedures and discussion of results have been included in this paper.


Introduction
ZnO is one of the most explored materials for photocatalysis and photodegradation applications [1][2][3][4][5]. However, the bandgap energy of pure zinc oxide makes it only favourable for ultraviolet-assisted photocatalytic and photodegradation despite its versatility. Hence, further research and enhancement have given way to a better-developed ternary like Zn 2 SnO 4 which has emerged as a new compound belonging to II 2 -IV-VI 4 semiconductor family [6]. ZnO and Zn 2 SnO 4 are nontoxic transparent wide bandgap direct semiconductor materials with excellent functional properties such as high photocatalytic activity and electron mobility [7]. These properties have made these materials to be extensively used in diverse applications such as a dyesensitised solar cell [8,9], gas sensor [10], photovoltaic [11] and optoelectronics [12]. Zn2SnO4 has been synthesized by several methods; mechanochemical [13], thermal evaporation [14] hyrothermal [15] and sputtering [16]. Few researchers have attempted to investigate different compositions and ratio of Zn and Sn in the ZnO-Zn 2 SnO 4 composite; Bai et al. [17] used molar ratio 2: 1of Zn to Sn, Wang and Wang [18] have also tried weight per cent ratio of 9: 1, Chen et al. [19] also used a molar per cent of 75% Zn to 25% Sn, while Iglesias et al. [20] have also used molar concentration of 0.1, 1 and 10 mol per cent of Sn in the ZnO-Zn 2 SnO 4 composite.
The compositions of semiconductor material along with its morphology play an essential role in its properties [11,21,22]. Aside the well-studied binary semiconductor oxides such as ZnO, TiO 2 , SnO 2, ternary oxides such as Zn 2 SnO 4 semiconductor-semiconductor heterojunction or coupling of metal oxides are well fascinating due to their modified optical and electronic properties [23][24][25]. For instance, the suitability of binary-based ternary oxide as inorganic-organic pollutant removal had been reported by Yiang et al. and Qian et al. [26,27]. The use of core-shell nanostructure of Au-ZnO-Pd ternary had been utilized to synergistically improve the electron co-trapping catalyst and plasmonic effect for the enhancement of photocatalytic performance [28].
Since it had been established that semiconductor material composition in ternary semiconductors plays a vital role in their final structural and optical properties; hence this work sought to investigate the effect of tin (Sn) concentration on the structural and optical properties of the ternary structure of ZnO-Zn 2 SnO 4 . The choice of Sn 4+ as dopant is due to the formation of inverse spinel structure of Zn 2 SnO 4 (whereby Zn 2+ and Sn 4+ share the octahedral site) which is known to be responsible for the Zn 2 SnO 4 extended conduction band and its high electronic conductivity. However, to the best of our knowledge. very few works have been done to investigate how the Sn 4+ concentration affect the energy band properties.

Materials and method
Poly(vinyl acetate) (PVAc) (Mw = 500,000 g by GPC), tin (ii) chloride dihydrate (SnCl 2 2H 2 O) salt ≥ 98% assay and zinc acetate dihydrate (Zn(CH 3 COO) 2 2H 2 O) salt ≥ 99% assay from Sigma-Aldrich and anhydrous N-N-dimethylformamide ≥ 99.8% assay from Scharlau were used as purchased as precursor material. The coupled oxide of ZnO-Zn 2 SnO 4 was synthesized from (Zn (CH 3 COO) 2 2H 2 O) and (SnCl 2 2H 2 O). Two (2) gram mixture of zinc and tin salts were prepared using weight percentage of 95% to 5%, 90% to 10%, 85% to15% 80% to 20% and 75% to 25%, respectively. The mixed salt was then dissolved in the polymer solution containing 1.6 g of PVAc dissolved in 10 ml of N, N, dimethylformamide (DMF) and was stirred until it was homogeneous. The glass substrates used were pre-cleaned using dilute ethanol, hydrochloric (HCl) acid and distilled water. The solution was electrospun for about 3 h to have a thick fibre formation using a DC voltage of capable of reaching 25 kV developed at the Centre for Energy Research & Development, Obafemi Awolowo University, Ile-Ife. The spinneret to substrate distance was maintained at 20 cm with the fibres electrospun at 12 kV. The deposited fibres were calcined at 700°C in a tubular furnace in air. The as-spun and calcined fibre were later analysed for their chemical, structural, optical and electrical properties.
TESCAN scanning electron microscope equipped with Oxford Instrument X-Max (EDS) was used for the surface morphology and elemental composition studies of the as-spun and calcined fibres. 1.7 MeV Ion Beam Accelerator was used for the Rutherford backscattering analysis to determine the elemental concentration   and the thickness of the deposited fibre. The Fourier transform infrared (FTIR) analyses were carried out using NICOLET S5 from Thermo Scientific. Stellanet Ultraviolet Spectrometer EP2000 (UV-VIS-NIR) was used for the optical characterization of the electrospun fibres. Fibre diameters were determined using ImageJ software. The crystal structure analyses were carried out using Brucker D8 high-resolution X-ray diffractometer.

Morphology
Figures 1 and 2 show the samples of the SEM images of the as-spun and calcined ZnO-Zn 2 SnO 4 nanofibre respectively. The mean diameter of the bead free non-woven nanofibre was found to reduce from 484.25-280.72 nm after calcination. This is given in Table 1. The reduction in the fibre diameter after calcination is attributed to the decomposition of the PVAc polymer that serves as a carrier for the metal salt.

Chemical and elemental analysis
The energy dispersive X-ray (EDX) analysis was carried out on some of the calcined samples to confirm the presence of the deposited metal such as Zn and Sn and also to determine the possible impurities that may be present in the samples. Figures 3-5 show the EDX spectra of the calcined samples and Tables 2-4 show the elementary compositions in the sample. It is evident from the spectra that the selected fibres contained the desired elements. A little amount of chlorine observed in Figure 3 could be because the chlorine atom has not entirely evaporated for the sample calcined at 600°C while total evaporation of the chlorine was observed for the samples calcined at 700°C. Asokan et al. [29] observed that for annealing at the temperature range of 200-575°C,       [30] observed an exothermic peak at 734°C, but no other peak was experienced beyond 750°C. The observed curve by Song and Dong [30] in the range of 530-750°C was attributed to the decomposition of the chloride and acetate from the tin and zinc salts respectively. The absence of peak beyond 750°C signifies no further decomposition of the organic components or other unwanted constituents. From the above findings, it shows that the required temperature for total decomposition of the unwanted constituents varies, depending on so many parameters, which may include among others the starting precursor, electrospinning process and optimization of the calcination process. In this work traces of chlorine was still observed in the samples calcined at 600°C but were not presents in the samples calcined at 700°C. With this, it shows that the required temperature for the total decomposition of the organic constituents and the chlorine and acetate from the tin and zinc salt precursor occurred between 600°C and 700°C. The presence of silicon in all the calcined samples of  can be attributed to the substrate. The Rutherford backscattered (RBS) result for selected calcined samples at 700°C is shown in Figures 6 and 7. The result is to complement the EDX studies. The distinct nature of the peaks from the spectra suggests that the electrospun nanofibre adheres firmly to the substrate. The RBS spectra suggest the decomposition of the starting precursor material to form the oxides of the metals present in the salt used. The spectra also show little or no trace of impurities from the organic constituent like chlorine or acetate from the precursor salt used indicating that the peaks are mainly for the metal oxide deposited. It could also be observed from the spectra that the peaks corresponding to each element of Zn, Sn is well separated, this can be attributed mainly to the mass difference of the element that made up of the nuclei of the targeted atoms such as zinc and tin [31].

Fourier transform infrared
The FTIR analysis of the studied ZnO-Zn 2 SnO 4 electrospun nanofibre for the as-spun and calcined fibres  are given in Figure 8(a,b), respectively. The broad peak observed at 3499 cm −1 in Figure 8

Structural characterisation
The X-ray diffraction patterns of the calcined samples were studied to confirm the crystallinity of the samples and investigate their crystal structure. The ZnO-Zn 2 SnO 4 composite electrospun nanofibre is presented in Figure 9 showing polycrystalline nature consisting of several diffraction peaks.
The  [32,33]. The ZnO has a hexagonal wurtzite structure, while the Zn 2 SnO 4 is a cubic structure. Rietveld refinement converges for the samples with the chi-square ranging from 6.2-9.3 and the  Bragg's weighted average ranging from 98 to 99. The concentration of the phases as given by the match3 software for Rietveld refinement shows that the amount of the Zn 2 SnO 4 increases from 51.5% to 61.2% and the ZnO decreases from 48.5% to 38.8% as the concentration of the zinc decreases and tin increases in the sample. In our earlier work [34], we reported the preferred growth direction for ZnO at [100], the shift in the 2θ position for the ZnO in the ZnO-Zn 2 SnO 4 is attributed to the interaction between the ZnO and the Zn 2 SnO 4 . The calculated crystallite size for the diffraction peaks using Scherrer relationship given Equation (1) is 14.78, 13.29, 12.31, 13.57 and 12.69 nm for 5% Sn, 10% Sn, 15% Sn, 20% Sn and 25% Sn respectively. This show that the crystallite size decreases as the Sn contents increase up to 15% Sn before it dropped and then increases.

Optical properties -UV-Visible spectrometric
The transmittance and absorption plot for the ZnO-Zn 2 SnO 4 nanofibre presented in Figure 10 shows an increase in the percentage transmittance between 0.8% and 18% to between 66% and 98% for the as-spun and calcine fibres, respectively. A decrease in the percentage transmittance was observed for 5% Sn to 15% Sn, which later slightly increases from 20% Sn up to 25% Sn. Rakshit et al. [22] while studying the effect of SnO 2 on the properties of ZnO-SnO 2 observed that the Sn is majorly doped into the lattice of ZnO up to 15% concentration of SnO 2 in the composite using X-ray photoelectron spectroscopy. They further observed that the phases are segregated for concentration above 15% Sn Similarly, Jooyoung et al. [35] also observed that increasing the concentration of SnO 2 in the SnO 2 /ZnO composite changes the composite from its crystalline state to an amorphous state. It could be established from these findings that increasing the Sn concentration in the SnO 2 -ZnO composite leads to changes in the phase of SnO 2 -ZnO composite, which will also lead to changes in the other properties of the material such as the optical and electrical properties. The observed changes in the transmittance pattern above 15% Sn in ZnO-Zn 2 SnO 4 phase could be attributed to some structural changes and variations in the degree of the crystallinity of the ZnO-Zn 2 SnO 4 composite material.

Refractive index and extinction coefficient
The refractive index of semiconducting materials plays an essential role in their electrical and optical properties [36]. The refractive index of the material was determined using Equations (2) and (3) and the plot is given in Figure 11.
where R is the reflectance, T is the transmittance, A is the absorptance and n is the refractive index. The range of the refractive index is in agreement with those observed for the studied metal oxides by various researchers, such as in [37][38][39]. It can be seen that anomalous behaviour was observed in some of the refractive index spectra of 15%Sn and 20%Sn which could be attributed to change in phase of the material [22,35].
Extinction coefficient k, given by Equation (4) is a parameter that defines how strongly a material absorbs light at any given wavelength. The extinction coefficient is also related to the refractive index of the material. This relation is given by Gadallah and Nahass [40] in Equation (5). The extinction coefficient of the material was determined using Equation (5).
where k is the extinction coefficient, α is the absorption coefficient and λ is the wavelength.
The plot of the extinction values obtained using Equation (5) is given in Figure 12. It could be observed  that the values of the extinction coefficient of the material increases as the Sn concentration increases up to 15%Sn before it starts suddenly decreases. This result is consistent with the transmittance pattern and the value obtained for the bandgap (Figure 13).

Optical band gap determination
The optical bandgap of materials is known as the minimum energy required by semiconductor material to excise a photon. To determine the optical bandgap of the materials deposited, we used the Tauc plot given in Equation (6).
A plot of (αhv) n againsthv gives the Tauc plot for the determination of the bandgap. The intersection of the linear fit with the hv axis at (αhv) n = 0 gives the optical bandgap, E g . Since both ZnO and Zn 2 SnO 4 are direct bandgap semiconductor material, the value of n is given to be 2. A plot of this is shown in Figure 14; the two possible band gap values obtained suggested that the composite of ZnO-Zn 2 SnO 4 had transited to an indirect bandgap material. The result is further confirmed by the plot of (αhv) 1/2 againsthv in Figure 14 given only one    [41,42]. This result indicates that the wide bandgap of the semiconductor ZnO and Zn 2 SnO 4 has transformed to a narrow bandgap material. The extrapolated results from the indirect bandgap plot presented in Table 5 further shows the bending of the optical bandgap of the composite ZnO-Zn 2 SnO 4 from the ultraviolet to a visible region. It could be observed that the bandgap values were increasing up to 15%Sn content before a reversal in the phase.

Dielectric properties
The real and imaginary part of the dielectric constant, and the optical conductivity of the nanofibre materials were calculated from the values of the extinction coefficient, and the refractive index using Equations (7)-(9), respectively, are presented in Figure 15-17.
where i , r , k, n, α and c are the imaginary dielectric constant, real dielectric constant, extinction coefficient, refractive index, absorption coefficient and speed of light, respectively. Figure 13 shows that the conductivity increases towards the high photon energy and increases sharply around the value of the bandgap energy. We also observed that the conductivity increases with the increase in the Sn concentration up to 15%W of Sn in the region of the lower photon energy. The graph of the real dielectric shows the same trend as that of the conductivity. The observation in the real dielectric plot is inverse to that obtained for the imaginary part. The sharp increase in the conductivity around the bandgap can be attributed to the strong interaction between the photon and electron. The results obtained are similar to that obtained by [43,40], for the dielectric property of ZnO.

Tail width of localized states
According to [44] "The exponential tail appears because disordered and amorphous materials produced localized states extended in the bandgap." The spectral dependence of the absorption edge in the low photon energy is assumed to obey the Urbach rule given in Equation (10) where α o is a constant, hν is the photon energy and E denotes "the width of the tail of localized states in the bandgap" (Urbach energy). By plotting ln (α) against hν, the inverse of the slope from the linear fit of the linear portions of the curves gives the Urbach tail width energy of the localized state of the bandgap. The results obtained are presented in Figure 18 and Table 6. It is evident from table 4.18 that the degree of the disorderliness increases in the ZnO-Zn 2 SnO 4 from 5% wt of Sn up to 15% wt of Sn before it starts to decreases again. Since the disorderliness in materials decreases their crystallinity, it could be said that the crystallinity of the materials reduces with the addition of Sn up to 15% wt of Sn before it starts to increase again. This observation is an agreement with the observation of [35]. The result confirms that the observed sudden increase in the transmittance pattern in Figure 10 after 15% wt of Sn is as a result of a change in the degree of crystallinity.

Conclusion
The variation of Sn in the nanofibre composite has been observed to affect the optical and structural properties of the ternary ZnO-Zn 2 SnO 4 . It was also observed that there is a turning effect in the properties of the composite material around the 15% Sn. The narrowing of the optical bandgap from the ultraviolet to visible region suggests its usage as a narrow bandgap material in applications such as visible light photocatalysis, photodetection and other light-harvesting applications.

Disclosure statement
No potential conflict of interest was reported by the author(s).