1. Introduction
In the last decades, transition metals experienced a renewal of interest due to their excellent electrical, transport, magnetic, and optical properties [
1,
2,
3]. In general,
pyrite is an ideal material for the fabrication of solar cells and photovoltaic devices [
4,
5,
6,
7,
8,
9] due to its high absorption coefficient
[
5], its small band gap (about 0.95 eV) [
10], high photocurrent quantum efficiency (
) [
6], and low material cost [
4,
5,
6,
7,
8,
9,
10,
11].
pyrite was one of the first crystal structures that resulted from Bragg, in 1914 [
12], with his XRD system. It has a simple cubic structure similar to that of rock salt.
pyrite is a good option for thin film photovoltaic. Considering its potential and current importance [
4,
5,
6,
7,
8,
9], many experimental [
4,
5,
6,
7,
8,
9,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28,
29,
30,
31,
32,
33,
34,
35,
36,
37,
38,
39,
40] and theoretical works [
41,
42,
43,
44,
45,
46,
47,
48,
49,
50,
51,
52,
53,
54,
55,
56,
57,
58,
59,
60] have been interested to
. Schlegel et al. [
33] determined the transition and reflectivity spectrum of a single crystal of
. They showed that
pyrite has an empty 3d e
g band at 300 K, a completely filled 3 d t
2g, and an indirect band gap equal to 0.95 eV. Kou et al. [
25] found a band gap at 297 K of about 0.84 eV. Karguppikar et al. [
40] reported that pyrite can be an indirect semiconductor from its conductivity properties, Hall effect data, and optical gap of 0.92 eV. Sun et al. [
37] determined an
pyrite thin film by sulfurizing oxide precursor films. From their UV-vis absorbance spectroscopy and X-ray photoelectron spectroscopy (XPS), they showed a direct band gap of about 0.75 eV, an indirect band gap of about 1.19 eV, and a high absorption efficiency (
). Yu et al. [
39] used the chemical bath deposition (CBD) method. They reported that the band gap of
can be increased from 0.86 to 1.31 eV when doped by Mn.
Many other preparation methods, such as spray pyrolysis, metal organic chemical vapor deposition (MOCVD), and ion beam sputtering have been declared for the synthesis of nanocrystals, nanowires, and crystallites of .
Mostly all experiments found a band gap between 0.84 and 1.03 eV. For theoretical study, Bullet [
42] used the first principle local density approximation (LDA) calculation to investigate the optical properties of iron pyrite. He found an indirect band gap of 0.4 eV for marcasite and 0.7 eV for pyrite. This value is smaller by 0.25 eV when compared to the experimental indirect band gap (0.95 eV). Zhoa et al. [
59] performed the self-consistent linear combination of atomic orbital (LCAO) formalism to determine the electronic properties of iron pyrite. Their smallest theoretical direct band gap was about 0.64 eV, and they found an indirect band gap of 0.59 eV. Additionally, Opahlele et al. [
53,
54] determined the electronic properties of
utilizing an LDA potential parameterized by the Perdew-Zunger (LDA-PZ). Their calculated band gap was about 0.85 eV. Muscat et al. [
52] employed the periodic LCAO method with the CRYSTAL 98 package and pseudopotential technique with CASTEP Software package.
Wadia et al. [
61,
62] showed a complete research study a few years ago and investigated 23 potential materials for photovoltaics and found that
pyrite was the best one, beating all materials in terms of cost. It was confirmed that the extraction cost of silicon was 57 times more than that of
(USD 1.7 for silicon compared to USD 0.03 for
). Additionally, the silicon energy output for extractions was 12 times bigger than that of
(24 KWh kg
−1 vs. 2 KWh kg
−1 for 24 KWh kg
−1). Rahman et al. [
62] showed that
is much more cost-effective than silicon if they are produced with taxation and the same regulations in the same country. All these beneficial and interesting features make
an excellent candidate for photovoltaic performance.
We included the prepared structure to show the real effect of the structure on band gap for devices that are not photovoltaic. The biggest dilemma for iron pyrite is attributed to the structure of pyrite, for which we complete studied to include the impact of the nanoparticles of iron pyrite on photovoltaic performance. This work aimed to improvise a new progress in the use of iron pyrite in photovoltaics.
3. Band Structure
3.1. Computational Details
Density–Functional Theory [
72,
73] was used within the linear muffin-tin orbital method in the atomic-sphere approximation (LMTO-ASA). The LMTO ASA method was explained in detail in several reports [
74,
75,
76]. In our calculations, we employed the self-consistent band calculations because they are the first principles of calculations utilizing density functional theory (see [
72]), utilizing the local density approximation (see [
77]), and utilizing numerical techniques based on the treatment of electron ion interaction in the pseudopotential approximation [
78]. Moreover, the Hamiltonian Atomic Spheres Approximation is totally specified by the potential parameters. It generates moments from the eigenvectors of the Hamiltonian. Regarding specified potential, there is an individual correspondence between the energy
of the wave function
and the logarithmic derivative
at the sphere radius. In essence, it is possible to specify either one. The potential P becomes simple because [
79,
80]:
where
,
, and
are the “potential parameters” that parameterize P.
defines the band “center of gravity”,
is the “band width” parameter, which correlates with the bandwidth of l channel if it were uncoupled from the other channels, and
is the “band distortion parameter”, which describes the deformations relative to a universal shape. Generally, small parameterization is a perfect method to study band structure.
First, we obtained the potential parameters for all atomic spheres. The muffin-tin potential constant
was the crossing point of muffin-tin potential around
and
, and it is listed in
Table 4. We had 24 symmetry operations. The initial sphere packing was equal to
, and it was scaled to
. The role of these empty spheres is to reduce the number of iterations in this system and to reduce the overlap between the spheres centered at
and
.
3.2. Pyrite Crystal Structure
pyrite has a cubic crystal structure and the space group number 205 (with space- group
). In these structure, there are eight S atoms located in eight positions and four Fe atoms in four positions. The lattice parameters for
pyrite are listed in
Table 5.
In pyrite, each S atom is coordinated with three Fe atoms, for which the dimer pairs S_S are in tetrahedral sites, and each Fe atom is coordinated with six S atoms in octahedral sites.
Moreover, these structures contain pairs of sulfur S
2 molecules, contrary to individual S atoms presented in the
Figure 5a image of the overlapped unit cell of
pyrite. To study the deviation from tetrahedral and octahedral geometries, we describe the correlation of the S-S bond length and cubic lattice. The relationships between our cell parameters are presented in
Table 5. The structure
of the pyrite is between 0.10 and 0.13 [
81]. Our work demonstrated that the value of
ranges between 0.111 and 0.113, showing significant effects of increased temperature conditioning
pyrite. We noticed that the Fe sites had a small trigonal distortion, for which the S-Fe-S bond angles were
at 350 °C and
at 400 °C, and the three Fe-S-Fe bonds were between
and
for which the S sites were distorted from tetrahedral symmetry.
3.3. Energy Bands of Pyrite
The DFT energy bands for samples at different temperatures are shown in
Figure 5b,c. The two figures provide the band structure on a form of energy that shows a general electronic structure. These figures present the region around the Fermi energy, which clearly depicts the details of a low conduction band and the highest valence band.
However, our band structure indicates that has a band gap semiconductor. We had four units in each unit cell and accommodated 40 occupied valence bands. The minimum conduction band was in , and the maximum valence band was at . The direct transition was observed at and had the values of o.87 eV and o.90 eV.
Our obtained results are different than those of Zhoa et al. [
59] and Temmerman [
58], who found 0.59 eV and 0.64 eV. Their gap was smaller than the experimental gap of 0.95 eV. Moreover, we were successful because our obtained gaps were significantly consistent with our optical gaps.
The calculated gap, optical gap, minimum band conduction CB
max, maximum valence band VB
max, and Fermi energy are summarized in
Table 6. The bands are relative to bonding and antibonding pairs of S
2 orbitals. In the range between −14 and 8 eV corresponding S3s, the structure of sulfur and these S3 states are predominant. The S 3p state is presented with a small addition of an Fe 3d function, starting at approximately −6 eV. Basically, these bands have Fe t
2g hybridized with p orbitals, and they are below the Fermi energy. The lowest conduction band contained a combination of Fe e
g orbitals. They were above Fermi energy. S3p Fe 3d is a preeminent character on the conduction band.
Figure 6a,b presents the density of the states calculated, and it shows the states above and below the Fermi level of iron pyrite. It discloses the importance of hybridization between Fe and S states and the effect of temperature on the fabrication of pyrite.
For the two graphs, level t2g lies between −0.1 Ry and 0, below the Fermi level, but for FeS2 pyrite prepared at 400 °C, it is near 0 and near Fermi level, which implies the good crystallite and electronic properties of iron pyrite prepared at 400 °C.
For both graphs, we noted that the conduction band was made entirely of Fe eg with Sp, marking that the conduction band was pure Sp while the valence bands were completely derived from the Fe t2g.
4. Pyrite in Photovoltaics: Modeling the ITO/ZnO/FeS2/ MoO3/Au/Ag Device
The synthesized
pyrite samples were evaluated for the application of photodetector devices. We modeled ITO/ZnO/
/
/Au/Ag to study the improvement in solar cell characteristics realized by the increase of temperature in two cases of preparation of
pyrite. We chose this application because it holds numerous benefits due its ability to be prepared at mild conditions, its low cost of chemicals, its mechanical flexibility, its better tuning, and due to it being a suitable alternative to silicone-based solar cells [
82,
83].
The device structure is presented in
Figure 7a. ITO film was cut mechanically to obtain a 2.5 cm × 2.5 cm substrate. All substrates were cleaned in isopropanol, water, and soap for 15 min. For layer parameters, a washed indium tin oxide (ITO) glass substrate was managed by ultraviolet-ozone for 15 min. The ZnO layers were spin coated with 60 mg ml
−1 ZnO/CHC
3 solution annealed at 250 °C for 15 min in the air to form a ZnO layer of 100 nm. The MoO
3 (20 nm) Au (30 nm) Ag (90 nm) layers were placed successively by thermal evaporation.
In recent years, ZnO has become the prime candidate for organic photovoltaic cells [
84] since its efficient improvement in stability. Here, we used the n-p layer heterojunction p-type
pyrite solar cell using an n-type window layer. Additionally, MoO
3 thin film can react as an effective electron-blocking layer or hole transporting to reduce the recombination of holes and electrons [
85].
The schematic illustration characterization by exercising voltage from −1 to 2 V under a dark current using the modeled ITO/ZnO/
/
/Au/Ag structure, as presented in
Figure 7b. The reported I-V characteristics and calculations for p-type
pyrite at 350 °C were similar [
86], principally below the onset voltage and the I-V curves above 1.5 V. However, the I-V curves for p-type
pyrite at 400 °C showed a large difference above 1.4 V, for which the onset voltage was around 1.3 eV. This result corresponds well with the fabrication of
pyrite, indicating that the nanostructures composed of
pyrite at 400 °C are excellent in building n-p junction ZnO/
pyrite.
We concluded that, to obtain high-performing photovoltaic cells on ITO/ZnO///Au/Ag, it is necessary to focus on the quality of nanoparticles and nanostructures of pyrite, which improved by increasing the temperature of preparation (400 °C). We also mentioned to the effect of sulfur position, distance sulfur–sulfur, and temperature to band gap of pyrite.
5. Conclusions
This work was an inclusive study of pyrite. Our study reported on the increase of temperature of preparation, characterization, and calculation of band gap of pyrite. Our experiment demonstrated the effect of temperature of preparation on the favorable optical and electrical properties of pyrite. Our results confirmed that p-type pyrite is a good choice for fabricating solar cells. We also found the gap energy and sulfur-sulfur distance for samples with different temperatures of preparation, which were prepared by spray pyrolysis. We proved the correlation between growth parameters and the calculated band structure. The optical gap energy obtained in this work is in good agreement with the gap energy calculated by the LMTO-ASA method.
Our findings proved a significant powerful dependency between gap energy and distance sulfur–sulfur. Moreover, we concluded that the excellent crystallinity, nanoparticles, and nanostructures of pyrite confirm a more efficient photovoltaic application.
Finally, it is important to note that the high-performing photovoltaic cells on ITO/ZnO///Au/Ag positively improved the quality of nanoparticles and nanostructures of pyrite.