The Effects of Chlorine Doping on the Thermoelectric Properties of Bi₂O₂Se

Abstract

In this study, we investigate the effects of chlorine doping on the structural, electronic, and thermoelectric properties of Bi₂O₂Se by employing density functional theory combined with semiclassical Boltzmann transport theory. It is shown that chlorine doping has significant effects on the electronic structure and thermoelectric properties of Bi₂O₂Se. As chlorine is incorporated into the selenium sites in Bi₂O₂Se, additional electrons are acquired, thereby inducing metallic properties in chlorine-doped Bi₂O₂Se. Meanwhile, Cl doping leads to an increase in the electrical conductivity of Bi₂O₂Se at room temperature by 25 times (from 358.59 $\mathrm{S}/\mathrm{c}\mathrm{m}$ to 9390 $\mathrm{S}/\mathrm{c}\mathrm{m}$), and the power factor is enhanced by a factor of 2.12 (from 4.04 mW/mK² to 12.59 mW/mK²). This study demonstrates that chlorine doping is an effective method to modify the physical properties of Bi₂O₂Se.

1. Introduction

In recent years, thermoelectric (TE) materials have received extensive attention because of their environmentally friendly and sustainable characteristics. Thermoelectric materials can directly convert heat into electricity, thus reducing carbon dioxide and greenhouse gas emissions [1,2,3,4,5,6]. The energy conversion efficiency of thermoelectric materials can be evaluated using the ZT ($ZT=\mathrm{T}\frac{S^2\mathsf{\sigma }}{\kappa }$) value [7,8], where $\sigma$, S, T, $\kappa$ and $S^2\sigma$ are the electrical conductivity, Seebeck coefficient, absolute temperature, thermal conductivity, and power factor, respectively [9,10,11,12]. Excellent thermoelectric materials should have a high Seebeck coefficient, high electrical conductivity, and low thermal conductivity [5]. However, these parameters are strongly coupled; e.g., the electrical conductivity is proportional to the thermal conductivity, while the carrier concentration is inversely proportional to the Seebeck coefficient. Thus, it is difficult to further improve the ZT value of TE materials.

Bismuth oxyselenide (Bi₂O₂Se) is a potential thermoelectric material due to its low thermal conductivity; however, its ZT value (0.05~0.5) is lower than that of other thermoelectric materials (~1) [13,14,15,16,17,18]. In the past decades, many techniques have been employed to improve the ZT value of Bi₂O₂Se. Experimental studies performed by Zhan et al. have shown that Bi defects significantly influence the thermoelectric performance of Bi₂O₂Se. Compared to the ZT value of 0.05 for pure Bi₂O₂Se, Bi_1.9O₂Se exhibits a 130% improvement in ZT value, reaching 0.12 at 773 K [19]. Guo et al. investigated the effect of Nb doping on the ZT value of Bi₂O₂Se using ball milling and hot-pressing sintering. They found that Nb-doped Bi₂O₂Se has a ZT value of 0.195 at 823 K, which is 325% higher than the value of 0.045 for pure Bi₂O₂Se [20]. Hong et al. studied the influence of Ce⁴⁺ doping on the thermoelectric properties of Bi₂O₂Se employing spark plasma sintering and found that Ce⁴⁺-doped Bi₂O₂Se has a ZT value of 0.26 at 773 K, which is 1.27 times higher than the value of 0.11 for pure Bi₂O₂Se [21]. Song et al. studied the influence of Ti doping on the thermoelectric properties of Bi₂O₂Se and found that the ZT value of Ti-doped Bi₂O₂Se is 0.56 at 773 K, which is 20% higher than the ZT value of 0.47 for the original Bi₂O₂Se [22]. Chen et al. investigated the effect of Sb doping on the thermoelectric properties of Bi₂O₂Se and found that the ZT value of Sb-doped Bi₂O₂Se reaches 0.59 at 773 K, which is 80% higher than that of 0.33 for the original Bi₂O₂Se [23]. Pan et al. synthesized polycrystalline Bi₂O_2−xS_xSe (x = 0, 0.01, 0.02, and 0.03) via a high-temperature solid-state reaction and demonstrated that the substitution of a small amount of S for O can effectively improve the thermoelectric performance of Bi₂O₂Se. The doped Bi₂O₂Se exhibits a ZT value of 0.29, which is 3.2 times higher than that of 0.09 for pristine Bi₂O₂Se at 793 K [24]. Fu et al. synthesized Bi_2−xZr_xO₂Se (x = 0, 0.02, 0.04, 0.06) via a combination of high-energy ball milling and cold pressing. It was found that Zr doping has a significant impact on the thermoelectric performance of Bi₂O₂Se. The peak ZT value of the Bi_1.96Zr_0.04O₂Se sample is approximately 0.27, which is 2.4 times higher than that of the undoped sample (~0.11) [25]. Song et al. obtained Bi₂O₂Se with Bi₂Te_2.7Se_0.3 through a liquid-assisted shear exfoliation-reassembly (LASE-R) process and found a significant improvement in its thermoelectric performance. At 772 K, the addition of 0.3 mol% Bi₂Te_2.7Se_0.3 results in a ZT value of 0.7, several times higher than the pristine sample (~0.1) [26]. Despite the above studies, the ZT value of Bi₂O₂Se remains low and requires further improvement.

In recent years, chlorine (Cl) has been widely used to improve the thermoelectric properties of materials [27,28,29]. Tan et al. experimentally studied the effect of Cl doping on the thermoelectric performance of AgBi₃S₅. They found that the ZT value of Cl-doped AgBi₃S₅ reaches 1 at 800 K, which is five times that of the original AgBi₃S₅ (0.2) [28]. Zhang et al. systematically studied the effect of Cl doping on the thermoelectric properties of AgPb₁₈SbSe₂₀ using experimental methods. It was shown that the ZT value of Cl-doped AgPb₁₈SbSe₂₀ is 1.3 at 873 K, which is 766% higher than that of the original AgPb₁₈SbSe₂₀ (0.15) [29]. Furthermore, Wang et al. carried out experimental investigations of the effect of Cl doping on the thermoelectric properties of BiSbSe₃ and reported that the ZT value of Cl-doped BiSbSe₃ is 1.0 at 800 K, which is five times higher than that of the original BiSbSe₃ (0.2) [27]. These investigations demonstrate that Cl doping is an effective method to improve the thermoelectric properties of materials.

In this work, the effects of Cl doping on the structure, electronic, and thermoelectric properties of Bi₂O₂Se are investigated by combining the DFT method and semi-classical Boltzmann transport theory. The relaxation time $\tau$, electrical conductivity $\sigma$, Seebeck coefficient S, power factor (PF), thermal conductivity $\kappa$, and dimensionless figure of merit ZT are provided. The results show that Cl doping can introduce additional electrons, causing Bi₂O₂Se to exhibit metallic character, thus enhancing the electrical conductivity and power factor. Therefore, this study provides a theoretical basis for experimental observation and reveals the potential mechanism of Cl-doping-induced improvement in the thermoelectric properties of Bi₂O₂Se.

2. Computational Details

The density functional theory (DFT) method in the Vienna Ab initio Simulation Package (VASP) code (Vienna, Austria) [30] is used to perform geometrical optimization and electronic structure calculations. The projector augmented-wave approach (PAW) [31] pseudopotential is used to describe the interaction between electrons and ions. In addition, the exchange-correlation potentials among electrons are described by the Perdew–Burke–Ernzerhof (PBE) functional within the generalized gradient approximation (GGA) [32] and the Heyd–Scuseria–Ernzerhof (HSE06) hybrid functional [33]. In this work, the Monkhorst–Pack scheme [34] with a 6 × 6 × 6 k-point sampling for the Brillouin zone is used for structural optimization and electronic structure calculations. The cut-off energy of the plane waves is set as 500 eV. The electronic transport coefficients are calculated by semi-classical Boltzmann theory using the BOLTZTRAP code [35] and the Brillouin zone is sampled with an 8 × 8 × 8 k-point grid. The lattice thermal conductivity ${\mathsf{\kappa }}_{\mathrm{l}}$ is calculated by using the Slack equation [36], and the energy and force convergence criteria are set to be 1 × 10⁻⁵ eV/atom and 1 × 10⁻² eV/Å, respectively. For chemical element doping, a 2 × 2 × 1 supercell containing 40 atoms is used to build the structural model for Bi₂O₂Se_0.875Cl_0.125, with Cl dopants substituting for the Se atoms. It should be pointed out that the doping concentration in this work is 1.31 × 10²¹ e/cm³, which may exceed the saturation limit. The dissolution limit can be raised by tuning experimental conditions, such as the experimental temperature and powder size of the samples [37,38,39]. On the other hand, in the literature, Liu et al. have demonstrated that the carrier concentration of Ge-doped Bi₂O₂Se can reach 10²¹–10²² cm⁻³ by controlling experimental conditions [40]. Therefore, the doping concentration of 1.31 × 10²¹ e/cm³ can probably be achieved experimentally.

3. Results and Discussions

3.1. Structural Properties of Pristine and Cl-Doped Bi₂O₂Se

The bent [Bi₂O₂]²⁺ layers and [Se]²⁻ layers are alternately stacked along the c-axis through weak electrostatic interactions to form Bi₂O₂Se [41], in which the [Bi₂O₂]²⁺ layer is an insulating layer while the [Se]²⁻ layer is a conducting layer. Bulk Bi₂O₂Se has a tetragonal I4/mmm (No. 139) crystal structure, and its unit cell contains ten atoms. The optimized geometrical structures of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 are shown in Figure 1, and both structures contain 40 atoms.

Table 1. Comparison of lattice constants a₀ and c₀ (Å), volume (ų), and bond length (Å) of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 with experimental and other theoretical results.

		a0	c0	Volume	<Bi-O>	<Bi-Se>
Bi2O2Se	Our cal.	3.917	12.357	189.592	2.337	3.311
	Exp. [42]	3.88	12.16	183.06	---	---
	Other cal. [43]	3.90	12.39	188.45	---	---
Bi2O2Se0.875Cl0.125	Our cal.	3.946	12.287	191.320	2.339	3.336

shows the lattice constants, volume, and bond lengths of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 calculated by us and from other literature sources [42,43]. The lattice constants of Bi₂O₂Se are a₀ = b₀ = 3.917 Å and c₀ = 12.357 Å, which are in good agreement with experimental results (a₀ = b₀ = 3.88 Å and c₀ = 12.16 Å) [42] as well as calculated results (a₀ = b₀ = 3.90 Å and c₀ = 12.39 Å) [43] reported in the literature. Additionally, the calculated <Bi-O> and <Bi-Se> bond lengths of Bi₂O₂Se are 2.337 Å and 3.331 Å, respectively, which agree well with the calculated values of 2.312 Å and 3.272 Å reported by Wu et al. [44]. Compared with Bi₂O₂Se, the calculated lattice constants a₀ and b₀ (3.946 Å) of Bi₂O₂Se_0.875Cl_0.125 increased by 0.74%, respectively, while the c₀ (12.287 Å) decreased by 0.57%, resulting in a 0.76% increase in volume. The calculated <Bi-O> bond length (2.339 Å) of Bi₂O₂Se_0.875Cl_0.125 increased by 0.09%, and <Bi-Se> bond length (3.336 Å) increased by 0.76%. As Cl substitutes for Se in Bi₂O₂Se, despite the smaller ionic radii of Cl (1 Å) compared to Se (1.15 Å), the stronger electronegativity (3.16) of the Cl atom than Se (2.55) results in the attraction of electrons from its neighboring atoms. Consequently, the interactions between Bi atoms and O as well as Se atoms are weakened, as indicated by the increased bond lengths of <Bi-O> and <Bi-Se> by 0.09% and 0.76%, respectively (see Table 1). This probably eventually leads to the volume expansion of 0.9% caused by Cl doping in Bi₂O₂Se. Considering that Cl and Se atoms have 7 and 6 valence electrons, respectively, when Cl is used to substitute for Se, it is likely that one additional electron is introduced.

3.2. The Influence of Cl Doping on the Electronic Structure of Bi₂O₂Se

The density of state (DOS) distribution of Bi₂O₂Se before and after Cl doping is first studied using standard and hybrid DFT methods. Figure 2a,b show the total and projected density of state distribution of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 around the Fermi level obtained by the standard DFT method. For Bi₂O₂Se, the valence bands from 4 to 5.53 eV are mainly composed of Se 4p orbitals hybridized with O 2p and Bi 6s orbitals, while the conduction bands from 6.02 to 8 eV are mainly composed of Bi 6p and O 2p orbitals. The bandgap between the maximum value of the valence band and the minimum value of the conduction band is 0.49 eV, which is similar to the theoretical values of 0.41 eV [45], 0.43 eV [46], and 0.472 eV [47]. Figure 2b shows the DOS distribution of Bi₂O₂Se_0.875Cl_0.125, revealing that many electrons are distributed at the Fermi level. These features indicate that Bi₂O₂Se_0.875Cl_0.125 has metallic properties that differ from the results reported by Tan et al. [48]. This is mainly due to the fact that our doping concentration of 2.5% is far greater than the solubility limit of 1.5% in the experimental literature.

As we know, the standard DFT method severely underestimates the bandgap of materials. In order to obtain a more accurate bandgap, we use the hybrid DFT method to further calculate the DOS distribution of Bi₂O₂Se before and after Cl doping. Figure 2c,d show the DOS distribution of Bi₂O₂Se before and after Cl doping, respectively. For Bi₂O₂Se, the calculated band gap between the maximum of the valence band and the minimum of the conduction band is 1.05 eV, which is similar to other calculated hybrid DFT results of 0.9 eV [49], 0.99 eV [50] and 1.01 eV [51]. For Bi₂O₂Se_0.875Cl_0.125, it can be seen that as compared with the standard DFT method, the hybrid DFT method obtains a more delocalized electron distribution, and a certain number of electrons still distribute on the Fermi level, i.e., the doped Bi₂O₂Se exhibits metallic properties.

3.3. Thermal Transport Properties of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125

3.3.1. Relaxation Time and Electrical Conductivity of Pure and Cl-Doped Bi₂O₂Se

When calculating thermoelectric properties such as the Seebeck coefficient, electrical conductivity, and thermal conductivity, two variables, i.e., carrier concentration and relaxation time, must be determined first, both of which are functions of temperature. Considering the phenomenon that Cl doping results in metallic behavior in Bi₂O₂Se similar to that of W doping [52], we refer to the carrier concentration of W-doped Bi₂O₂Se for our calculation. Therefore, the carrier concentration of pure and Cl-doped Bi₂O₂Se is n = 1.36 × 10¹⁹ cm⁻³ and n = 1.56 × 10¹⁹ cm⁻³, respectively. The relaxation time $\tau$ for Bi₂O₂Se at 300 K can be obtained with the following formula [53]:

$$\tau =\frac{\mu m^{\ast }}{e},$$

(1)

where m* is the effective mass and $\mu$ is the carrier mobility. Employing m* = 0.33 m₀ and $\mu =8$0 ${\mathrm{c}\mathrm{m}}^2{\mathrm{V}}^{-1}{\mathrm{s}}^{-1}$ at 300 K, as reported by Gao et al. [52], we obtain τ = 1.5 × 10⁻¹⁴ s at 300 K for Bi₂O₂Se, which is comparable with the experimental result of 1.2 × 10⁻¹⁴ s reported by Pan et al. [54]. As for Bi₂O₂Se_0.875Cl_0.125, referring to the relaxation time of W-doped Bi₂O₂Se (1.9 × 10⁻¹⁴ s) [52] and other metals (0.31 × 10⁻¹⁴ s~2.2 × 10⁻¹⁴ s) [55], an approximate relaxation time τ of 1 × 10⁻¹⁴ s at 300 K is employed.

The relaxation time $\tau$ for Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 at different temperatures can be calculated by the following equation [56]:

$$\tau =\frac{C}{T{n}^{\frac{1 }{3}}}$$

(2)

where n is the carrier concentration, T is the temperature, and C is a constant. According to the experimental literature [52], carrier concentrations are fixed at 1.36 × 10¹⁹ ${\mathrm{c}\mathrm{m}}^{-3}$ for Bi₂O₂Se and 1.56 × 10¹⁹ ${\mathrm{c}\mathrm{m}}^{-3}$ for Bi₂O₂Se_0.875Cl_0.125, which are also employed in other investigations [23,56,57,58,59]. Taking the relaxation time $\tau$ calculated at 300 K into Equation (2), the constants C are determined to be 1.78 × 10⁻¹⁰ $\mathrm{s}\mathrm{K}{\mathrm{c}\mathrm{m}}^{-1}$ for Bi₂O₂Se and 8.21 × 10⁻¹⁰ $\mathrm{s}\mathrm{K}{\mathrm{c}\mathrm{m}}^{-1}$ for Bi₂O₂Se_0.875Cl_0.125. Then, by substituting the constants C and n into Equation (2), the relaxation time $\tau$ at different temperatures is obtained. The calculated results are plotted in Figure 3a. It is shown that the relaxation time $\tau$ is dependent on the temperature and becomes smaller with increasing temperature.

Based on the determined relaxation time and carrier concentration, we further calculate the electrical conductivity $\sigma$ of Bi₂O₂Se before and after Cl doping. Figure 3b shows the temperature-dependent electrical conductivity $\sigma$ of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125. For Bi₂O₂Se, the calculated electrical conductivity $\sigma$ decreases with increasing temperature, which is consistent with the experimental literature [52]. However, our results differ from those of Rurova et al. [60], due to the reason that the carrier concentration is fixed in our calculations while dependent on the temperature in the work of Rurova et al. As for Bi₂O₂Se_0.875Cl_0.125, the electrical conductivity $\sigma$ is as high as 9390 $\mathrm{S}/\mathrm{c}\mathrm{m}$ at 300 K, which is 26 times larger than the value of 358.5 $\mathrm{S}/\mathrm{c}\mathrm{m}$ for Bi₂O₂Se. Experimentally, the Bi₂O₂Se sample ($\sigma$= 175 S/cm) was prepared by the exfoliation technique [52], which may produce defects and influence its physical properties [61]. In addition, the dissolution limit of Cl in Bi₂O₂Se reported in the literature is 1.5% [48], while the dopant concentration considered in this work is much higher, i.e., 2.5%. Although the dopant concentration considered in this work is higher than the experimental dissolution limit, the dissolution limit can be raised by techniques such as tuning the experimental temperature and powder size of the samples [37,38,39]. Therefore, the presented results will provide a theoretical reference for further related experimental research.

3.3.2. Seebeck Coefficient and Power Factor of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125

The Seebeck coefficient S is an essential parameter for predicting the thermoelectric performance of materials. It is mainly determined by the thermal electromotive force ($△V$) and the temperature difference ($△T$) and can be expressed as [62]:

$$S=\frac{△V}{△T}$$

(3)

Figure 4a presents the variation of the Seebeck coefficient S with temperature for Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125. It is shown that the Seebeck coefficients S of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 are both negative, indicating that electrons are the dominant charge carriers. Due to the increase in free electron concentration, Cl doping leads to a significant decrease in the absolute value of the Seebeck coefficient S of Bi₂O₂Se. In addition, as the temperature increases, the absolute Seebeck coefficient S increases for both Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125. For Bi₂O₂Se, the absolute Seebeck coefficient S increases from 87.6 μV/K at room temperature to 177 μV/K at 800 K; for Bi₂O₂Se_0.875Cl_0.125, the absolute Seebeck coefficient S increases from 23.8 μV/K at room temperature to 59.7 μV/K at 800 K.

Based on the calculated electrical conductivity $\sigma$ and Seebeck coefficient S, we further calculate the power factor PF of Bi₂O₂Se before and after Cl doping using the following formula:

$$PF=S^2\sigma$$

(4)

Figure 4b shows the temperature-dependent power factors (PFs) of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125. It is evident that the PF values for Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 increase with increasing temperature. In addition, Cl doping significantly enhances the PF of Bi₂O₂Se. For example, at a temperature of 800 K, the power factor PF of Bi₂O₂Se_0.875Cl_0.125 is 12.59 mW/m K², which is 3.04 times that of pure Bi₂O₂Se (4.14 mW/m K²).

3.3.3. Thermal Conductivity and Figure of Merit ZT of Pure and Cl-Doped Bi₂O₂Se

The thermal conductivity of a crystal is determined by the combined effects of electronic thermal conductivity and lattice thermal conductivity, i.e.,

$$\kappa ={\kappa }_e+{\kappa }_l$$

(5)

The Wiedemann–Franz law [35,63,64] can be applied to calculate the electronic thermal conductivity of materials, i.e.,

$${\kappa }_e=L\sigma T$$

(6)

Here, L represents the Lorenz constant (2.4 × 10⁻⁸ $\mathrm{W}\mathsf{\Omega }{\mathrm{K}}^{-2}$). The temperature-dependent behavior of electronic thermal conductivity is shown in Figure 5a. It is observed that the electronic thermal conductivity of Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 increases with increasing temperature. Moreover, the electronic thermal conductivity of Bi₂O₂Se_0.875Cl_0.125 is significantly higher than that of Bi₂O₂Se. For instance, at a temperature of 800 K, the calculated value of Bi₂O₂Se_0.875Cl_0.125 is 6.78 W/mK, which is 27.12 times larger than that of Bi₂O₂Se (0.25 W/mK). This can be attributed to the remarkable improvement in the electronic conductivity of Bi₂O₂Se caused by Cl doping.

The Slack equation [36] can be used to calculate the lattice thermal conductivity ${\kappa }_l$:

$${\kappa }_{\iota }=A·\frac{\overline{M}{\Theta }^3\delta }{{\gamma }^{2}T{n}^{\frac{2}{3}}}$$

(7)

The Slack model has been demonstrated to provide lattice thermal conductivity results that agree well with experimental measurements and has been extensively used in the calculation of lattice thermal conductivity for materials [49,65,66,67]. Here, A is a constant, and the calculation formula is given by:

$$A=2.43\ast \frac{{10}^{-6}}{1-\frac{0.514}{\gamma }+\frac{0.228}{{\gamma }^2}}$$

(8)

In this equation, $\overline{M}$ represents the average atomic mass, ${\delta }^3$ is the volume of each atom, γ is the Grüneisen parameter, Θ is the Debye temperature, and T is the absolute temperature. Figure 5b shows the temperature-dependent behavior of the lattice thermal conductivity ${\kappa }_l$. It is observed that the lattice thermal conductivity of both Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 decreases with increasing temperature. For Bi₂O₂Se, the calculated lattice thermal conductivity ${\kappa }_l$ at 300 K is 1.68 W/mK, which agrees well with the experimental data of κ = 1.8 W/mK and other calculated results of κ = 1.2 W/mK [51]. Furthermore, Cl doping reduces the lattice thermal conductivity of Bi₂O₂Se. For example, the calculated lattice thermal conductivity of Bi₂O₂Se_0.875Cl_0.125 at 300 K is 1.29 W/mK, which is 23.21% smaller than that of Bi₂O₂Se (1.68 W/mK). This is mainly due to the low elastic constants of Bi₂O₂Se_0.875Cl_0.125. The reduced lattice thermal conductivity will be beneficial for suppressing heat transfer and minimizing thermal energy loss.

Combining the calculated electronic thermal conductivity ${\kappa }_e$ and lattice thermal conductivity ${\kappa }_l$ into Equation (5), the thermal conductivity of Bi₂O₂Se before and after Cl doping can be obtained, as depicted in Figure 5c. It is shown that Cl doping results in an increase in the thermal conductivity of Bi₂O₂Se, due to a substantial increase in the electronic thermal conductivity ${\kappa }_e$.

The ZT value of Bi₂O₂Se before and after Cl doping can be obtained by the following formula [7]:

$$ZT=\frac{S^2\sigma T}{\kappa }$$

(9)

as shown in Figure 5d. At 300 K, the calculated ZT value for Bi₂O₂Se is 0.04, which is in good agreement with the reported values of 0.05 by Gao et al. [52] and 0.045 by Song et al. [22]. Furthermore, the ZT values of both Bi₂O₂Se and Bi₂O₂Se_0.875Cl_0.125 increase with increasing temperature. However, the ZT value of Bi₂O₂Se_0.875Cl_0.125 is lower than that of Bi₂O₂Se due to the introduction of Cl dopants. This is mainly due to the decrease in the Seebeck coefficient caused by the introduction of Cl dopants. The presented results thus suggest that Cl doping is not beneficial for improving the thermoelectric properties of Bi₂O₂Se.

4. Conclusions

In summary, the effects of Cl doping on the structural, electronic, and thermoelectric properties of Bi₂O₂Se are investigated by combining the DFT method and semi-classical Boltzmann theory. The results indicate that Cl doping leads to a 0.76% volume expansion of Bi₂O₂Se and a significant shift of the Fermi level into the conduction band, resulting in metallic properties. Additionally, Cl doping can introduce extra electrons, resulting in a higher electrical conductivity (σ), a lower absolute Seebeck coefficient (S), and a higher power factor (PF). These findings suggest that Cl doping can optimize the thermoelectric properties of Bi₂O₂Se. However, the introduction of Cl dopants in Bi₂O₂Se results in a significant increase in the absolute value of the Seebeck coefficient, leading to a notable decrease in its ZT value. In conclusion, Cl doping can affect the thermoelectric properties of Bi₂O₂Se significantly, providing insights for further related experimental and theoretical studies.