Investigation of the structural and optical properties of Sb₂S₃/PVA/PVP nanocomposites blended sheets for optoelectronic applications

Sb₂S₃ powder has been prepared by the chemical co-precipitation method as follows (Salem and Selim 2001). In a 100 mL beaker; 650 mg of SblC₃ dissolved in 2.5 mL of acetone, followed by adding 25 mL of 1 M sodium thiosulfate (Na₂S₂O₃·5H₂O) maintained at 10 °C, after that adding 70 mL of double distilled water also kept at 10°C. A clear solution of pH 5 was obtained, stirred well for a few minutes and maintained in a refrigerator for the next day. A granulated powder of a red–orange color precipitated at the bottom of the beaker, was collected, washed well with distilled water several times, and dried at a temperature of \({60}^{o}\) until a constant weight was obtained. The chemical equations reaction for this process is as follows:

Separate PVA and PVP solutions were prepared by dissolving 6 wt% of each in distilled water and stirred using a magnetic stirrer at 90 °C for 2 h until a clear solution was obtained. Stock solution of PVA/PVP was prepared by mixed PVA: PVP of ratio 2:1 at room temperature and stirred continuously for an hour to get a clear homogenous solution. Separate solutions of the above mixed PVA/PVP were filled by different 1, 3, and 5 wt% of Sb₂S₃ powder prepared according to Eq. (1). Each solution was ultra-sonicated for 20 min, then poured onto a petri dish, and kept in an oven for 24 h at 40 °C to prepare polymeric sheets. Figure 1a, b shows the schematic diagram for the preparation of precipitated Sb₂S₃ powder (a), and Sb₂S₃ NPs/PVA/PVP polymeric sheets (b).

The structural characterization of the prepared Sb₂S₃ powder at room temperature was performed using X-ray diffraction (Type a Bruker D-8 advance X-ray diffractometer). The diffractometer worked at 40 kV and 40 mA using Cuk_α1 radiation of wavelength λ = 0.1541 nm in the 2θ range 10°–75°. The microstructure of the powder was investigated using a transmission electron microscope (Type JEOL-JEM 1230) operating at 120 kV. Transmission electron micrograph of a thin polymer sheet of thickness 30 nm for PVA/PVP sample filled with Sb₂S₃ by a ratio of 5 wt%, prepared by cutting a piece of polymer sheet using a glass knife of ultra-microtome (Type Leica EM UC6) was investigated using the TEM device. The Fourier transform (FTIR) measurement of the prepared composite sheets has been performed to detect the various types of bond stretching and vibrations in the prepared composite sheets using a spectrometer VERTEX 80 (Bruker Corporation, Germany) in conjunction with a Platinum Diamond ATR, which uses a diamond disc as the internal reflection element in the wavenumber range from 4000 to 500 cm⁻¹. The transmission and reflection spectra of the prepared composite sheets were carried out at room temperature using a double-beam UV–visible–NIR spectrophotometer (Type Jasco, V-570, UV–Vis–NIR) in the wavelength range from 190 to 2500 nm.

Figure 2 shows the X-ray diffraction of the Sb₂S₃ powder prepared by the co-precipitation method. The figure depicts that the prepared powder exhibits a polycrystalline structure corresponding to Sb₂S₃ as a major orthorhombic phase (JCPDS cared No: 87–1135). However, three minor additional peaks were noticed; these peaks are indexed as (113), (313), and (402) corresponding to Sb_9.8S₁₅ (JCPDS cared No: 77–2467), besides two other faint peaks indexed as (111), and (022) corresponding to antimony oxide Sb₂O₃ as a minor orthorhombic phase (JCPDS cared No: 74–1725). The 2θ values of the observed peaks and interplaner distance as well as the corresponding (hkℓ) are listed in Table 1. The lattice parameters a, b, and c for the present major orthorhombic phase have been calculated using the reflection (022), (700), and (004) planes via the equation (Cullity 1978):where \(h, k, and\mathcal{l}\) are the Miller indices of the diffracting planes. These calculated values were found to be 1.4202, 1.159, and 0.7474 nm, for a, b, and c, respectively, with a unit cell volume of 123.0229 nm, which are consistent with the standard diffraction pattern. The average crystallite size of the Sb₂S₃ powder was calculated using the Gaussian fit of the main observed (022) plane according the well-known Scherrer’s formula.where D is the crystallite size, λ is the wavelength of the X-ray, β is full width at half maximum of the X-ray plane, and θ is the diffraction angle. The calculated crystallite size using the Gaussian fit was found in a range of 24 nm. The transmission electron microscope and the corresponding histogram for the particle size determination are shown in Fig. 3a, b. The figure depicts a regular distribution for particles of the same shape. The TEM micrograph has been analyzed using ImageJ software to determine an average particle size of 40.66 nm. It was observed that there is a disagreement between the particle size calculated from both XRD and TEM. This can be attributed to the fact that XRD calculates the crystallite size while TEM determines the particle size which includes multiple crystallites (El-naggar et al. 2022a; El-Gamal and Sayed 2019).

Figure 4 illustrates the X-ray diffraction patterns of the prepared Sb₂S₃/PVA/PVP NPs polymeric sheets in the range of 2θ from 10 to 80. The figures exhibit a PVA/PVP characteristic broadening peak around 2θ = 19.5°, attributed to the semi-crystalline nature of the PVA (Aziz et al. 2017a; Mohammed et al. 2018), and may be due to the presence of strong intra-molecular hydrogen bonding in isolated monomer units of PVA and inter-molecular hydrogen bonding between other monomer units (Mahendia et al. 2011). This characteristic peak is clearly also observed for all Sb₂S₃/PVA/PVP polymer sheets. Notably, there is also a slight peak observed at 2θ = 40.6° for all Sb₂S₃/PVA/PVP curves which are related also to the semi-crystalline nature of PVA polymer (Aziz et al. 2017a; Mohammed et al. 2018; Mahendia et al. 2011). The other peaks observed at 2θ = 28.53°, and 33.74° correspond to Sb₂S₃ peaks of (022), and (501), respectively.

Figure 5 represents the transmission electron micrograph of 5 wt.% Sb₂S₃ NPs/PVA/PVP polymer sheet. It is observed that the high concentration of Sb₂S₃ NPs aggregate in the PVA/PVP matrix, resulting in intermolecular and intramolecular hydrogen bonding of the polymer chains (Aziz 2016). So, as a result of the aggregation of NPs in a polymer matrix, an increase in the sample's crystallinity occurs as illustrated by XRD analysis. The inset of Fig. 5 represents the region that has been zoomed as assigned by the arrow where the grains of NPs are observed obviously.

Figure 6 shows the FTIR transmittance spectra of pure PVA/PVP polymer blend besides that doped with different doping weight ratios percentages (1, 3, and 5 wt%) of Sb₂S₃ nanoparticles. The figure depicts the presence of a broad absorption peak centered at 3283 cm⁻¹ for pure PVA/PVP polymeric sheet, which corresponds to the O–H stretching vibration of hydroxyl groups of the PVA/PVP polymeric blend (Ali et al. 2018). The two bands observed at 2857 cm⁻¹ and 2263 cm⁻¹ were corresponding to C–H asymmetric stretching (Bhajantri et al. 2009; Abdullah et al. 2016), and C–H symmetric stretching (Zidan et al. 2019; Bhajantri et al. 2009). The band at 1645 cm⁻¹ is due to the C=O stretching of PVA and PVP (Zidan et al. 2016), thereby confirming the intermolecular interaction between the OH groups in PVA and carbonyl groups in PVP (Rajesh et al. 2019b; Gökmeşe et al. 2013). It was found that the intensity of this band increases as a result of adding the Sb₂S₃ NPs powders to the PVA/PVP as a host matrix. The band at 1420 cm⁻¹ was assigned to the symmetric bending of CH₂ vibration (Elashmawi et al. 2013). The band at 1327 cm⁻¹ in pure PVA/PVP was shifted to 1286 cm⁻¹ due to the S–S vibration in nanoparticles of SbS structure and enhanced after PVA/PVP polymeric was melded with Sb₂S₃ NPs powders. The observed increase of the band at 1645 cm⁻¹ besides the shift of the band from 1327 to 1286 cm⁻¹, reflects the success of the interaction between the filler SbS material with the PVA/PVP as a host polymeric matrix. The bands at 1083 and 917 were attributed, respectively, to C=O stretching (Zidan et al. 2019), and CH₂ rocking (Abdelghany et al. 2016). Other bands at 836 and 577, were attributed respectively, to C–C stretching (Omkaram et al. 2007) and N–C=O bending vibrations (Safo et al. 2019; Kamaruddin et al. 2017; Rezeki et al. 2019; Bryaskova et al. 2011). The FTIR spectra show also a weak absorption band observed at 655 cm⁻¹ which is related to the stretching vibration of Sb-S.

Figure 7 illustrates the optical transmittance and reflectance spectra of Sb₂S₃/PVA/PVP NPs polymer sheets in the wavelength range 250–2500 nm. The figure depicts that a pure PVA/PVP sheet exhibits high transparency reaching a value of 82% which agrees quite well with that reported previously (Badawi 2020). the transmittance spectra of Sb₂S₃/PVA/PVP NPs polymer sheets showed a drastic decrease with increasing Sb₂S₃ wt%. The transmittance spectra of the samples can be divided into two different regions. The 1st region is related to the fundamental absorption region in the optical wavelength range 250–1000 nm at which the polymer sheets strongly and gradually absorb the incident radiation. It was observed that the fundamental absorption edge of the pure PVA/PVP and those filled with 1 and 3 wt% Sb₂S₃ NPs shift to a higher wavelength side (red-shift). However, with the increase of the filing Sb₂S₃ NPs concentration in the PVA/PVP matrix to 5 wt%, the fundamental absorption edge shifts to the shorter wavelength (blue shift), reflecting a variation of the optical band gap in a reverse manner. These variations reflect an active incorporation of the filed Sb₂S₃ NPs in the PVA/PVP host matrix. The variation of the optical band gap energy against the variation of the Sb₂S₃ NPs filling percent in the PVA/PVP matrix will be illustrated below. The 2nd region within the wavelength range 1450–2320 nm, in which the transmission spectra contain five multi-oscillations absorption bands located, respectively, at 1482, 1746, 1958, 2100, and 2320 nm, which are related to the first overtone of the –CH group (Sultanova et al. 2012). Similar behavior of the transmission spectra was reported elsewhere (Badawi 2020; AlAbdulaal et al. 2022). It was also observed that the reflectance spectra of the investigated samples increase with increasing the embedded Sb₂S₃ NPs in the PVA/PVP host matrix in a reverse manner to the behavior of the transmittance spectra. The optical constants (real and imaginary parts of the refractive index, n, and k) were calculated as functions of the incident wavelength from the recorded transmittance and reflectance spectra via the relations (Mohammed et al. 2018; Subramanian and C.l. P, P.P. D, 2010):

where \(\alpha\) is the absorption coefficient (\(\alpha =4\pi k/\lambda\)), d is the sheet thickness, T is the transmittance, R is the reflectance and \(\lambda\) is the wavelength of the incidence wavelength. The Spectral variations of the absorption coefficient, \(\alpha\) as a function of the photon energy, \((h\nu )\) using Eq. 3 are shown in Fig. 8. The figure depicts an exponential dependence of the absorption coefficient on the photon energy. At lower energy < 2 eV, the absorption coefficient is slightly low, this may be attributed to the fact that the energy of the incident photon energy is not sufficient to transmit the electrons from the valence band to the conduction band. While at higher energies, the incident photon energy is sufficient to free electrons to high levels, which reflects an increase in the absorption coefficient. On the other hand, the absorption coefficient of nanocomposite polymer sheets increases with increasing concentrations of the NPs Sb₂S₃ wt% in the PVA/PVP host matrix which reflects increases in the number of charge carriers. The absorption coefficient in the fundamental absorption region can be described by Tauc’s relation (Mohammed et al. 2018):where Β is a constant and n is an exponent characterizing the nature of the optical transition process in the k-space causing the optical absorption. Plots \((\alpha h\nu {)}^{2}\) vs. (hν) for a direct optical transition as shown in Fig. 9, and extrapolating the linear part of the curves to zero energy one can estimate the values of the energy band gap, which are listed in Table 2. The value of the optical band gap energy for pure PVA/PVP sheet is 4.68 eV in context with the values of 4.95 eV, for PVA/PVP blended with the ratio 50/50 (Badawi 2020), with the value of 4.66 eV for 40/60 PVA/PVP (El-Mahalawy et al. 2023), 4.88 eV for 50/50 PVA/PVP (El-naggar et al. 2021), 5.26 eV for 70/30 PVA/PVP (El-naggar et al. 2022a), and 4.97 eV for 70/30 PVA/PVP (Heiba et al. 2021b). The value of the energy band gap decreases from 4.68 eV for pure PVA/PVP to the value of 4.21 and 3.03 eV, respectively, due to increasing Sb₂S₃ NPs filling in the PVA/PVP blend matrix to 1%, and 3 wt%, respectively. This decrease in the E_g value can be attributed to an increase in the degree of the disordered of the PVA/PVP blend matrix due to filling with Sb₂S₃ NPs (Abdullah et al. 2016, 2015b), hence creating a localized state in the band gap as a result of Sb₂S₃ doping. However, the determined band gap value in the case of the PVA/PVP sheet filled with 5% increased as the opposite of expected to 3.19 eV rather than a decrease as in the two other cases. The results of the optical band gap of the Sb₂S₃/PVA/PVP polymer sheets make them effective candidates for optoelectronics applications and optical devices. For instance, 5 wt% of Sb₂S₃-filled PVA/PVP polymer blend sheet which possesses a band gap of 3.19 eV, corresponding to λ ≈ 389 nm, could be used as an effective UV blocker filter (Badawi 2020).

The variation of the optical band gap energy could be also calculated as a result of doping Sb₂S₃ NPs wt% in the PVA/PVP host matrix, using the absorption spectra of the investigated samples shown in Fig. 10a, based on the absorption spectrum fitting model (ASF). According to this model, the absorption coefficient in Eq. (5) can be expressed as a function of wavelength by El-Mahalawy et al. (2023); Nasrallah and Ibrahim 2022; Bhogi et al. 2022):where \({\lambda }_{g}\), h and c are the wavelength conformable to the optical band gap, plank’s constant and velocity of light, respectively. The Eq. (6) will be rewritten on the basis of Beer-Lambert’s law as follows:where \({{\text{D}}}_{1}=(B{(hc)}^{n-1}d/2.303)\) and D₂ is a constant that is considered to be a reflection. For the direct optical transition, the extrapolation of the linear part in (A/λ)² vs. (1/λ) plot as shown in Fig. 10b provides the value of \(1/{\lambda }_{g}\), which can then be used to determine the optical band gap as follows:

The calculated \({{\text{E}}}_{g}^{ASF}\) values for the investigated samples are listed in Table 2. It can be noticed that the values of \({{\text{E}}}_{{\text{g}}}^{{\text{ASF}}}\) agree very well with the \({E}_{g}^{d}\) values calculated using Tauc’s relation and have the same trend. As the doping content of Sb₂S₃ NPs increases in the PVA/PVP blend up to 3 wt%, the energy gap reduces. However, for the PVA/PVP blend filled with 5 wt% Sb₂S₃ NPs, the energy gap increases. The increase in the optical band energy of the PVA/PVP sample filled with 5 Sb₂S₃ wt% can be interpreted as the following. At low concentrations of Sb₂S₃ NPs, the grains of NPs are randomly and uniformly distributed throughout the PVA/PVP matrix (Dissanayake et al. 2003). The presence of these grains can create additional favorable conducting pathways near their surface, where the number of them increases with increasing surface area of NPs added to the PVA/PVP matrix. Therefore, the charge carriers can transfer between these conducting pathways leading to an increase in the conduction and, consequently a decrease in the optical band gap. Further increase of Sb₂S₃ NPs may result in the aggregation of the NPs in the PVA/PVP matrix, which leads to the immobilization of longer polymer chains (El-Gamal and Sayed 2019; Dissanayake et al. 2003; Aziz et al. 2017b). This, in turn, reduces the surface area of the NPs, causing a decrease in the conduction and, consequently an increase in the optical band gap. The aggregation of the grains has been confirmed earlier by TEM.

The spectral variations of the refractive index, n against the wavelength for pure PVA/PVP and PVA/PVP polymeric sheets filled with different wt% of Sb₂S₃ NPs are shown in Fig. 11. The figure depicts that the investigated samples exhibit a peak occurring for all samples observed in the wavelength range 500 and 850 nm, which may be due to the electronic transition from bonding to antibonding molecular orbital (Elrasasi et al. 2022). The refractive index variation vs. wavelength for each of the investigated samples shows un-systematic behavior within the wavelength range 800–1850 nm (slightly up and down) which may be attributed to the presence of some of the absorption bands in the transmission and/or reflection spectra, while beyond 1850 nm, it begins to decrease gradually with the wavelength. Whereas, beyond the wavelength 850 nm, the refractive index notably increases with increasing the addition of Sb₂S₃ content into the PVA/PVP as a host matrix. This implies that the density of the PVA/PVP melded polymer increased with an increase in the amount of Sb₂S₃ wt% (Muhammad et al. 2015).

The refractive index can be analyzed according to Wemple DiDomenico single oscillator model (Wemple and DiDomenico 1971; Teleb et al. 2021). According to the model, the refractive index n is related to the oscillator energy, E_o, and the dispersion energy E_d via relations:i.e.where \(h\nu\) is the photon energy. Experimental fulfillment of this relation can be obtained by plotting \(({n}^{2}-1{)}^{-1}vs. (h\nu {)}^{2}\) in the photon energy range 0.24–0.94 eV² as shown in Fig. 12, for natural behavior. Linear fits of the function to the smaller energy data yield straight lines equations as shown in Fig. 12 allowing us to determine the values of (E_o E_d)⁻¹ and (E_o/E_d) from the slope and intercept, respectively. The determined oscillator parameters values E_o, E_d, the static refractive index, \({n}_{o}^{2}=(1+{E}_{d}/{E}_{o})\) where,\({n}_{o}\left(0\right)=n at hv\to 0\), the oscillator strength f, (\(f={E}_{o}{E}_{d})\) as well as the value of the static high-frequency dielectric constant \({\varepsilon }_{s}={n}_{o}^{2}\) are listed in Table 2. It was seen from the listed data that the trend of variation in the oscillator energy values, E_o calculated from the Wemple DiDomenico model follows the same variation trend of the calculated optical band gap values for each of the Tauc’s or ASF models, which indicates the applicability of the applied Wemple DiDomenico model in this work. Where the values of the single oscillator energy, E_o, decrease from 4.04 to 3.15 eV, and dispersion energy, E_d decreases from 3.77 to 3.55 eV, as the Sb₂S₃ NPs filler increases in the PVA/PVP polymer matrix from 1 to 3 wt%.

Further analysis of the refractive index data at long wavelength region (\(1500-2500 nm)\) was also performed on the basis of the Drude free electron model to elucidate the effects of filling Sb₂S₃ NPs to the PVA/PVP as a host matrix on the variation the carrier concentration to the effective mass ratio, \({\text{N}}/{{\text{m}}}^{*}\) as well as the lattice high dielectric constant, \({\varepsilon }_{\infty }\) using the relations (Ammar et al. 2015; El-Gendy 2017):where e is the electron charge, \({\varepsilon }_{o}\) the permittivity of the free space (\(=8.854\times {10}^{-12}\) F/m), \({\varepsilon }_{\infty }\) the high lattice dielectric constant, and \(N/{m}^{*}\) is the carrier concentration to the effective mass ratio. The values of \(N/{m}^{*}\) as well as the lattice high-frequency dielectric constant, \({\varepsilon }_{\infty }\) could be determined, respectively, from the slope and intercept of the plots \({\varepsilon }_{r}\) vs. \({\lambda }^{2}\) shown in Fig. 13. Those values are listed in Table 2. It is evident from those values that, for all of the investigated samples \({\varepsilon }_{L}>{\varepsilon }_{s}\) which are mainly attributed to the contribution of the free carriers-concentration to the lattice vibration in the dispersion modes (El-Gendy 2017; Qasrawi and Shukri Ahmad 2006). It was noticed that the calculated N/m^* values increase with the increasing addition of Sb₂S₃ in the polymeric mixture, which indicates an increase in charge carriers.

The optical conductivity is related to the absorption coefficient and the refractive index of the prepared samples by the relation (Abdullah et al. 2016; Abd El-Ghany et al. 2022):where α is absorption coefficient, n is the refractive index and c is the light speed. Figure 14 shows the variation of optical conductivity with photon energy. Obviously, the optical conductivity increases with the addition of Sb₂S₃. This means that the concentration of charge carriers increases in the PVA/PVP matrix due to the dispersion of Sb₂S₃ in the interstitial spaces of the polymer chains (Badawi 2020; Abdullah et al. 2015b). Additionally, the increase in the optical conductivity at the high photon energy is due to the high absorption of ultraviolet radiation by the prepared samples (Abdullah et al. 2016).