Vacuum Electrodeposition of Cu(In, Ga)Se₂ Thin Films and Controlling the Ga Incorporation Route

Abstract

The traditional electrochemical deposition process used to prepare Cu(In, Ga)Se₂ (CIGS) thin films has inherent flaws, such as the tendency to produce low-conductivity Ga₂O₃ phase and internal defects. In this article, CIGS thin films were prepared under vacuum (3 kPa), and the mechanism of vacuum electrodeposition CIGS was illustrated. The route of Ga incorporation into the thin films could be controlled in a vacuum environment via inhibiting pH changes at the cathode region. Through the incorporation of a low-conductivity secondary phase, Ga₂O₃ was inhibited at 3 kPa, as shown by Raman and X-ray photoelectron spectroscopy. The preparation process used a higher current density and a lower diffusion impedance and charge transfer impedance. The films that were produced had larger particle sizes.

1. Introduction

CIGS thin films are one of the most promising photovoltaic materials for second-generation solar cells [1]. They are composite semiconductor materials with chalcopyrite structures [2] with band gaps of 1.0 eV to 1.7 eV [3]. Co-evaporation [4], magnetron sputtering [5], spray pyrolysis [6], solution gel method [7], and electrodeposition [8] are the main methods to prepare CIGS thin films.

Electrodeposition has the advantages of a low preparation cost, high efficiency, large area, and continuous preparation and is likely to be used in future industrialized production methods [9]. However, it has some inherent flaws when used to prepare metal materials. For example, hydrogen is easily generated on the cathode surface and is difficult to remove and the deposition rate can be slow [10,11].

Generally, a vacuum environment induces rapid off-gassing [12] and lower oxidation and contamination [13]. Based on these properties, researchers have conducted studies on the electrodeposition of metal thin films in a vacuum environment. In the 1940s, RCA Company [14] electrodeposited Fe and Mn in a vacuum environment and found that hydrogen quickly escaped from the films, and the oxidation of the film was avoided. This resulted in the preparation of smooth and bright metal films. In 1984, E Muttilainen et al. [15] electrodeposited Cr in a vacuum environment and found that a low-pressure environment increased the current efficiency and reduced the porosity and roughness of the coating. They concluded that low pressure was one of the most important factors affecting the quality of the coating. In a following study, S. E. Nam [16,17], R. Su [18], and P. Ming et al. [19,20] prepared metal thin films with excellent properties using vacuum electrodeposition.

In the application of electrodeposition to prepare CIGS thin films, the Ga₂O₃ phase is also easily generated, along with the generation of hydrogen on the cathode surface [21,22]. The entry of Ga₂O₃ into the films can decrease the conductivity [23,24]. Therefore, in this article, electrodeposition was applied to prepare CIGS thin films in a vacuum environment. The effects of the vacuum environment on the deposition process and the composition, morphology, and phases in the CIGS thin films were studied.

2. Materials and Methods

The electrolyte solution contained 200 mL deionized water with 4 mM CuCl₂, 10 mM InCl₃, 10 mM GaCl₃, 8 mM H₂SeO₃, and a supporting electrolyte (100 mM LiCl, 100 mM NH₄Cl, and 60 mM NH₂SO₃H). The pH of the solution was adjusted to 1.8 by adding concentrated hydrochloric acid dropwise.

Electrochemical experiments, including linear scanning voltammetry (LSV), potentiostatic polarization, electrochemical impedance spectroscopy (EIS), and electrodeposition were carried out on CIGS thin films in a three-electrode system with a SnO₂/glass (1 × 1.5 cm) as the working electrode, a Pt electrode as the counter electrode, and a saturated calomel electrode (SCE) as the reference electrode.

A CorrTest CH310H electrochemical workstation was used for the LSV, potentiostatic polarization, EIS, and electrodeposition CIGS thin film studies. The LSV curves were measured at potentials between 0.1 V to −1.0 V (vs. SCE) with a scanning rate of 10 mV/s. The potentiostatic polarization curves were measured at a potential of −0.60 V (vs. SCE) and a polarization time of 30 min. EIS was performed at a potential of −0.60 V (vs. SCE) with a test frequency range of 0.01 Hz to 100 kHz. On electrodeposited CIGS thin films, the potential range was −0.1 V to −0.9 V (vs. SCE), and the polarization time was 30 min. When the experiment was carried out at 3 kPa, a vacuum pump was used to extract the air from the experimental system for 30 min. Figure 1 illustrates the electrodeposition process of CIGS thin films.

Scanning electron microscopy (SEM, TM3030Plus, Hitachi, Tokyo, Japan) was used to measure the morphology of CIGS thin films, and energy-dispersive spectroscopy (EDS, TM3030Plus, Hitachi, Tokyo, Japan) was utilized to characterize the elemental composition of the CIGS thin films. X-ray diffraction (XRD, Rigaku Ultima IV, Tokyo, Japan), Raman spectroscopy (Raman, DXRxi, Thermo Scientific, MA, USA), and X-ray photoelectron spectroscopy (XPS, Escalab Xi⁺, Thermo Scientific, MA, USA) were used to characterize the phases in the CIGS thin films.

3. Results and Discussion

Figure 2 illustrates a schematic of how the vacuum environment affects the electrochemical behavior of Ga³⁺. In the cathode region, the vacuum environment inhibited the Volmer reaction of H⁺ to a hydrogen atom (H_ads), leading to lower H⁺ consumption on the cathode surface. In the anode region, the oxygen evolution reaction (OER) of H₂O to O₂ was promoted, which produced additional O₂. At the same time, the rapid release of O₂ bubbles agitated the electrolyte solution, which led to a uniform H⁺ distribution in the electrolyte solution. Thus, the inhibition of the Volmer reaction and the promotion of the OER together stabilized the pH of the cathode region. This finally interrupted the route of Ga³⁺ entering the films in the form of Ga₂O₃ by the hydrolytic reaction of Ga³⁺.

3.1. LSV Analyses

Figure 3a illustrates the effects of the vacuum environment on the hydrogen evolution potentials. Evidence of an inhibited reaction of H⁺ to H_ads was found in a vacuum environment. Peak C₁ corresponding to the reaction of H⁺ to H_ads (Equation (1)) did not appear at 3 kPa, possibly due to a decrease in the H₂ partial pressure. Because of the reduced partial pressure of H₂, the overpotential of H⁺/H_ads on the electrode surface was increased. This phenomenon has also been observed on Au and Pt electrodes [25,26].

$${\mathrm{H}}^{+}+{\mathrm{e}}^{-}\to {\mathrm{H}}_{\mathrm{ads}}$$

(1)

Figure 3b illustrates the LSV curve of the Cu unary solution at 3 kPa. Peaks C₁, C₂, and C₃ corresponded to the reaction of Cu²⁺ to Cu⁺ (Equation (2)), Cu⁺ to Cu (Equation (3)), and H⁺ to H_ads (Equation (1)), respectively [27]. Peak C₁ was indistinguishable at 80 kPa and 3 kPa, indicating that pressure did not affect the reaction of Cu²⁺ to Cu⁺. Peak C₂ was reduced at 3 kPa, indicating that the formation of Cu was promoted in a low-pressure environment. Peak C₃ did not appear at 3 kPa, indicating that the reaction of H⁺ to H_ads was inhibited in a vacuum environment.

$${\mathrm{Cu}}^{2+}+{\mathrm{e}}^{-}\to {\mathrm{Cu}}^{+}$$

(2)

$${\mathrm{Cu}}^{+}+{\mathrm{e}}^{-}\to \mathrm{Cu}$$

(3)

Figure 3c illustrates the LSV curve of the In unary solution at 3 kPa. Peak C₁ corresponded to the reaction of In³⁺ to In (Equation (4)) [28]. This peak was reduced at 3 kPa, showing that the formation of In was promoted in a vacuum environment.

$${\mathrm{I}\mathrm{n}}^{3+}+3{\mathrm{e}}^{-}\to \mathrm{In}$$

(4)

Figure 3d illustrates the LSV curve of the Ga unary solution at 3 kPa. Peak C₁ corresponds to the reaction of H⁺ to H_ads [27] (Equation (1)). This peak did not appear at 3 kPa, indicating that the reaction of H⁺ to H_ads was inhibited in a vacuum environment. The reaction of Ga³⁺ to Ga did not occur within the range of 0.1 V to −1.0 V because the reaction required a higher potential [24].

Figure 3e illustrates the LSV curve of the Se unary solution at 3 kPa. Peaks C₁, C₂, C₃, and C₄ corresponded to the reactions of H₂SeO₃ to Se (Equation (5)), H₂SeO₃ to H₂Se (Equation (6)), H⁺ to H_ads (Equation (1)), and Se to H₂Se (Equation (7)), respectively [27]. The current densities of peaks C₁, C₂, and C₄ were greater at 3 kPa, indicating that the formation of H₂Se and Se was promoted. Peak C₃ did not appear at 3 kPa, indicating that the reaction of H⁺ to H_ads was inhibited in a vacuum environment.

$${\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3+4{\mathrm{H}}^{+}+4{\mathrm{e}}^{-}\to \mathrm{Se}+3{\mathrm{H}}_2\mathrm{O}$$

(5)

$${\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3+6{\mathrm{H}}^{+}+6{\mathrm{e}}^{-}\to {\mathrm{H}}_2\mathrm{Se}+3{\mathrm{H}}_2\mathrm{O}$$

(6)

$$\mathrm{Se}+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to {\mathrm{H}}_2\mathrm{Se}$$

(7)

Figure 4a illustrates the LSV curves of Cu-Se binary solutions at 3 kPa. Peaks C₁ and C₂ corresponded to the formation of Cu⁺ and Se, respectively (Equations (2) and (5)). Peaks C₃ and C₄ corresponded to the reaction of H₂SeO₃ to H₂Se and Se to H₂Se, respectively (Equations (6) and (7)). Since the solution contained both Cu⁺ and Cu²⁺, the two ions were induced by H₂SeO₃ or H₂Se to form Cu₂Se or CuSe, respectively [26] (denoted as Cu₃Se₂ phase) (Equations (8)–(11)). The current densities of peaks C₃ and C₄ were greater at 3 kPa, indicating that the formation of Cu₃Se₂ was promoted in a vacuum environment.

$$2{\mathrm{Cu}}^{+}{+\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3+4{\mathrm{H}}^{+}+6\mathrm{e}\to {\mathrm{Cu}}_2\mathrm{Se}+3{\mathrm{H}}_2\mathrm{O}$$

(8)

$${\mathrm{Cu}}^{2+}{+\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3+4{\mathrm{H}}^{+}+6\mathrm{e}\to \mathrm{Cu}\mathrm{Se}+3{\mathrm{H}}_2\mathrm{O}$$

(9)

$$2{\mathrm{Cu}}^{+}{+\mathrm{H}}_2\mathrm{Se}\to {\mathrm{Cu}}_2\mathrm{Se}+2{\mathrm{H}}^{+}$$

(10)

$${\mathrm{Cu}}^{2+}{+\mathrm{H}}_2\mathrm{Se}\to \mathrm{Cu}\mathrm{Se}+2{\mathrm{H}}^{+}$$

(11)

Figure 4b illustrates the LSV curves of the In-Se binary solutions at 3 kPa. Peaks C₁ and C₂ corresponded to the reaction of H₂SeO₃ to H₂Se and Se to H₂Se, respectively (Equations (6) and (7)). H₂SeO₃ reacted with H₂Se to form Se (Equation (12)). Because In₂Se₃ has a high standard Gibbs energy of formation (−386 kJ/mol [29]), H₂Se induced In³⁺ to form In₂Se₃ [30] (Equation (13)). The current densities of peaks C₁ and C₂ were greater at 3 kPa, indicating that the formation of H₂Se, In₂Se₃, or In was promoted in a vacuum environment.

$$2{\mathrm{H}}_2\mathrm{Se}{+\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3\to 3\mathrm{Se}+3{\mathrm{H}}_2\mathrm{O}$$

(12)

$$3{\mathrm{H}}_2\mathrm{Se}+2{\mathrm{In}}^{3+}+6{\mathrm{e}}^{-}\to {\mathrm{In}}_2{\mathrm{Se}}_3$$

(13)

Figure 4c illustrates the LSV curves of Ga-Se binary solutions at 3 kPa. The peaks C_1, C_2, and C₃ correspond to the reaction of H₂SeO₃ to Se, H₂SeO₃ to H₂Se, and Se to H₂Se, respectively (Equations (5)–(7)). Because Ga₂Se₃ has a high standard Gibbs energy of formation (−418 kJ/mol [31]), Ga³⁺ was induced by H₂Se to form Ga₂Se₃ [32] (Equation (14)). The current densities of the peaks C₂ and C₃ are greater at 3 kPa, indicating that the formation of H₂Se or Ga₂Se₃ was promoted in a vacuum environment.

$$3{\mathrm{H}}_2\mathrm{Se}+2{\mathrm{Ga}}^{3+}\to {\mathrm{Ga}}_2{\mathrm{Se}}_3+6{\mathrm{H}}^{+}$$

(14)

Figure 4d illustrates the LSV curves of Cu-In-Se ternary solutions at 3 kPa. Peaks C_1, C₂, C₃, C₄, and C₆ correspond to the formation of Cu⁺, Se, Cu₃Se₂, H₂Se, and In₂Se₃ or In, respectively (Equations (2), (5), (8)–(11), (6), (13) or (4)). In³⁺ was induced by Cu₃Se₂ + H₂SeO₃ or Cu₃Se₂ + H₂Se to produce more stable CuInSe₂ [32] (Equations (15) and (16)) at peak C₅. As the polarization potential increased, the current density became greater at 3 kPa, indicating that the formation of CIS was promoted in a vacuum environment.

$$3{\mathrm{In}}^{3+}+{\mathrm{Cu}}_3{\mathrm{Se}}_2+{4\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3+16{\mathrm{H}}^{+}+25\mathrm{e}\to 3\mathrm{CuIn}{\mathrm{Se}}_2+12{\mathrm{H}}_2\mathrm{O}$$

(15)

$$3{\mathrm{In}}^{3+}+{\mathrm{Cu}}_3{\mathrm{Se}}_2+{4\mathrm{H}}_2\mathrm{Se}+\mathrm{e}\to 3\mathrm{CuIn}{\mathrm{Se}}_2+8{\mathrm{H}}^{+}$$

(16)

Figure 4e illustrates the LSV curves of Cu-Ga-Se ternary solutions at 3 kPa. Peaks C_1, C₂, C₃, and C₄ corresponded to the formation of Cu⁺, Se, Cu₃Se₂, and Ga₂Se₃, respectively (Equations (2), (5), (8)–(11) and (14)). Ga³⁺ was induced by Cu₃Se₂ + H₂SeO₃ or Cu₃Se₂ + H₂Se to produce CuGaSe₂ [31] (Equations (17) and (18)) at peak C₄. As the polarization potential increased, the current density became greater at 3 kPa, indicating that the formation of CGS was promoted in a vacuum environment.

$$3{\mathrm{Ga}}^{3+}+{\mathrm{Cu}}_3{\mathrm{Se}}_2+{4\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3+16{\mathrm{H}}^{+}+25\mathrm{e}\to 3\mathrm{Cu}\mathrm{Ga}{\mathrm{Se}}_2+12{\mathrm{H}}_2\mathrm{O}$$

(17)

$$3{\mathrm{Ga}}^{3+}+{\mathrm{Cu}}_3{\mathrm{Se}}_2+{4\mathrm{H}}_2\mathrm{Se}+\mathrm{e}\to 3\mathrm{Cu}\mathrm{Ga}{\mathrm{Se}}_2+8{\mathrm{H}}^{+}$$

(18)

Figure 4f illustrates the LSV curves of Cu-In-Ga-Se ternary solutions at 3 kPa. Peaks C₁, C₂, and C₃ corresponded to the formation of Cu⁺, Se, and Cu₃Se_2, respectively (Equations (2), (5), (8)–(11)). Ga and In are homotopic elements, and their reduction processes at the cathode are similar (noted as (In, Ga)³⁺ phase). Meanwhile, both In³⁺ and Ga³⁺ were induced by Cu₃Se₂ + H₂SeO₃ or Cu₃Se₂ + H₂Se into the films (Equations (15)–(18)). Therefore, In³⁺ and Ga³⁺ competed for reduction at the cathode. According to previous conclusions, the reduction potential of In and Ga in the CIGS thin films was −0.6 V. Therefore, peak C₄ corresponded to the formation of CuInSe₂, CuGaSe₂, or Cu(In, Ga)Se₂ (Equations (19) and (20)). Peak C₅ corresponded to the formation of In₂Se₃, Ga₂Se₃, or In (Equations (13), (14), or (4)). The current density of peak C₄ was greater at 3 kPa, indicating that the formation of CIGS was promoted in a vacuum environment.

$$3{(\mathrm{In}, \mathrm{Ga})}^{3+}+{\mathrm{Cu}}_3{\mathrm{Se}}_2+{4\mathrm{H}}_2\mathrm{Se}{\mathrm{O}}_3+16{\mathrm{H}}^{+}+25\mathrm{e}\to 3\mathrm{Cu}(\mathrm{In}, \mathrm{Ga}){\mathrm{Se}}_2+12{\mathrm{H}}_2\mathrm{O}$$

(19)

$$3({\mathrm{In}, \mathrm{Ga})}^{3+}+{\mathrm{Cu}}_3{\mathrm{Se}}_2+{4\mathrm{H}}_2\mathrm{Se}+\mathrm{e}\to 3\mathrm{Cu}(\mathrm{In}, \mathrm{Ga}){\mathrm{Se}}_2+8{\mathrm{H}}^{+}$$

(20)

3.2. Potentiostatic Polarization and EIS Analyses

According to the previous conclusion, the main reduction potential of In³⁺ and Ga³⁺ in the CIGS thin films was −0.6 V. Therefore, the potentiostatic polarization potential was set to −0.6 V. Figure 5a illustrates the potentiostatic polarization curve of the CIGS thin films at 3 kPa. The current density was always greater at 3 kPa, indicating that the resistance during the preparation of CIGS thin films was lower in a vacuum environment.

Changes in the diffusion impedance and charge-transfer impedance during the preparation of CIGS thin films were investigated by EIS at a low pressure. Figure 5b illustrates the impedance Nyquist plot of the CIGS thin films deposition process at 3 kPa. The semicircle diameter of the curve was smaller, and the slope of the straight line was larger at 3 kPa, indicating that the electroreduction and diffusion impedances were lower, respectively. In addition, the ohmic impedance was smaller. Thus, the resistance was lower when applying vacuum electrodeposition. This corresponds to the conclusion of the current density-time curves. Both curves in Figure 5b were semicircular with straight lines, indicating that electroreduction and diffusion controlled the CIGS thin films deposition [33].

In this study, the impedance test data were combined in order to fit the equivalent circuit of the CIGS thin films deposition process. The fitted EIS data of the CIGS thin films deposition process at different pressures are shown in

Table 1. Fitting data of CIGS thin films deposition impedance at 3 kPa (Ω).

Pressure	Ws	Rct	C	Rs
80 kPa	417.6	42.39	2.292 × 10−4	18.77
3 kPa	129.1	29.83	3.546 × 10−4	19.23

. W_s is the Warburg diffusion impedance, R_ct is the charge-transfer impedance between the interface of cathode and electrolyte solutions, C is the electrical double-layer capacitance on the electrode surface, and R_s is the electrolyte solution impedance.

When the system pressure was reduced from 80 kPa to 3 kPa, W_s fell by 69.1%, R_ct fell by 29.6%, C rose by 35.4%, and R_s remained constant. Figure 6 illustrates the OER on the anode surface during the preparation of the CIGS thin films. At 3 kPa, the number of bubbles at the Pt electrode was higher, and the volume was larger. Therefore, the formation of O₂ was promoted under a vacuum. The promoted formation of O₂ indicates that the electrode reaction proceeded more smoothly. The smooth occurrence of the electrode reaction was also one of the reasons for the higher current density at 3 kPa. In addition, according to Boyle’s law, the bubble volume should be larger in a vacuum environment. This is consistent with the phenomenon in Figure 6. The rapid release of O₂ bubbles enhanced the diffusion of the electrolyte solution. Therefore, the W_s value fell. Since the reaction of H⁺ to H_ads was inhibited, and the rapid dehydrogenation occurred in the electrolyte at 3 kPa, the contact area increased between the interface of the cathode and electrolyte solution, which decreased the R_ct value. Because the reaction of H⁺ to H_ads was inhibited, the deposition of major elements increased. Thus, the thickness of the CIGS thin films increased, which increased the C value at 3 kPa. Because it was determined by the electrolyte solution, the R_s value remained constant.

3.3. EDS Analysis

Figure 7 illustrates the EDS composition curves of Cu, In, Ga, and Se in CIGS thin films at polarization potentials from −0.1 V to −0.9 V. Overall, changing the pressure had little effect on the element deposition pattern but had a significant impact on the elemental content. Combined with Figure 3, the first reduction potentials of Cu and Se were 0 V and −0.2 V, respectively. Therefore, Cu ions were more easily reduced when the reduction potential was −0.1 V. As shown in Figure 7a, the content of Cu in the films was highest at this time. The formation of Cu and Cu₃Se₂ was promoted in a vacuum environment, which in turn led to a high content of Cu in the films at 3 kPa. When the reduction potential was −0.2 V, the reduction of Se became easier. As shown in Figure 7d, the content of Se in the films increased. As the potential increases, the Cu and Se content tended to be relatively stable. As a whole, the content of Cu and Se in the CIGS thin films was slightly higher at 3 kPa, indicating that the vacuum environment was beneficial for the formation of Cu and Se.

As shown in Figure 7b, the potential of In in the CIGS thin films was −0.2 V at 3 kPa. However, this potential was −0.4 V at 80 kPa. The reason may be that the formation of Cu₃Se₂ was promoted in a vacuum environment and In was induced by the excess Cu₃Se₂ to produce a more stable CuInSe₂ phase, which in turn could enter the films at a lower potential. Meanwhile, the content of In in the films was higher at 3 kPa. As a result, the formation of In was promoted in a vacuum environment.

As shown in Figure 7c, when the potential was lower, the formation of Ga was largely unaffected due to the lower In content in the films. Because the In content in the films increased as the potential increased, and the formation of In and Ga had a competitive relationship, the formation of Ga was inhibited. This was the first reason for the low concentration of Ga in the CIGS thin films at 3 kPa. Upon increasing the potential, the production of H₂ at the cathode surface increased at 80 kPa. In turn, the consumption of H⁺ in the cathode region increased. Along with the consumption of H⁺, the pH at the cathode region increased, which led to the hydrolysis of Ga³⁺ to Ga₂O₃ [24]. Ga³⁺ entered the films in the form of Ga₂O₃ [23]. Because the reaction of H⁺ to H_ads was inhibited and the OER was promoted at 3 kPa, the pH remained relatively stable at the cathode region, which prevented Ga³⁺ from entering the films in the form of Ga₂O₃. This was the second reason for the low Ga concentration in the CIGS thin films at 3 kPa. In summary, the formation of Ga was inhibited in a vacuum environment.

3.4. Morphologic Analyses

SEM images showing the morphologies of the CIGS thin films prepared at various deposition potentials and pressures are displayed in Figure 8. Overall, the particles [29] in the films were larger at 3 kPa. The first reason is that the reaction of H⁺ to H_ads was inhibited, which in turn inhibited H₂ production. Another reason is that the vacuum environment quickly removed H₂ from the surface of the CIGS thin films. The amount of H₂ adsorbed on the cathode surface decreased, and the space available for the deposition of major elements increased.

3.5. Phase Analyses

As shown in Figure 9, in the CIGS thin films, there are three main diffraction peaks due to (101), (112), and (220) planes corresponding to CuIn_0.7Ga_0.3Se₂. The positions of these diffraction peaks are quite similar. In addition, the difference in pressure changes the preferred crystallographic orientation. When the pressure is 80 kPa, the CuIn_0.7Ga_0.3Se₂ phase possesses a (101) preferred crystallographic orientation at approximately 17.27°. However, under 3 kPa, the preferred crystallographic orientation is a (112) plane located at approximately 26.90°. For the CuIn_0.7Ga_0.3Se₂ phase, a preference for the (112) plane is more favorable. Therefore, the main CuIn_0.7Ga_0.3Se₂ phase can be electrochemically deposited at the pressure of 80 kPa and even at 3 kPa, although with different preferred crystallographic orientations.

The surface phases were characterized by Raman spectroscopy, as shown in Figure 10 and Figure 11. According to Figure 10a, the vibrational peaks of the Ga₂O₃ phase (200 cm⁻¹ [34]) appeared in the potential range from −0.1 V to −0.2 V at both 80 kPa and 3 kPa. This was because the reaction of H₂SeO₃ to Se consumed H⁺ in the cathode region, which in turn hydrolyzed Ga to Ga₂O₃ in the films [35]. Meanwhile, little O₂ formed on the anode surface due to the low potential and could not increase the mass transfer rate of the solution at 3 kPa. This was why the Ga₂O₃ phase appeared at a low potential at 3 kPa.

When the potential was −0.3 V, the main phase of the films changed from the Cu-Se phase (260 cm⁻¹ [36]) to the CIGS A1 phase (175 cm⁻¹ [37]) at 3 kPa. The formation of the CIGS A1 phase led to a decrease in the consumption of H⁺. Increasing the potential led to a gradual increase in the O₂ production, which in turn led to a gradual increase in the mass transfer rate, resulting in a stable pH in the cathode region. Ga³⁺ hydrolysis was inhibited, and the Ga₂O₃ phase disappeared. However, at 80 kPa, when the potential was −0.4 V, only the main phase of the films began to transform. In correspondence with the higher potential, the hydrogen evolution reaction began to occur. Therefore, the vibrational peak of the Ga₂O₃ phase was always present when the potential was in the range of −0.3 V to −0.5 V.

As shown in Figure 11, Ga entered the films as Ga-Se phase (214 cm⁻¹ [38]) at 3 kPa. Combined with the above conclusions, the formation of Ga₂O₃ was inhibited at 3 kPa.

Since poorly-crystallized Ga₂O₃ did not produce a Raman response, this article determined the phase of Ga from the Ga 3d binding energy. The XPS spectra of the CIGS thin films are presented in Figure 12. According to Figure 11, at 80 kPa, the binding energy of 19.73 eV and 20.51 eV are related to Ga of CIGS thin films, and at 3 kPa, 19.07 eV and 19.87 eV are related to Ga of CIGS thin films. According to the literature, 19.07 eV corresponds to GaSe (19.00 eV, Ga 3d) [39], and 19.73 eV and 19.87 eV correspond to Ga₂Se₃ (19.70 eV or 19.90 eV, Ga 3d5/2) [40]. In addition, 20.51 eV is close to Ga₂O₃ (20.50 eV, Ga 3d5/2) [41]. In summary, the phases of the films were Ga₂Se₃ and Ga₂O₃ at 80 kPa, while GaSe and Ga₂Se₃ were the phases at 3 kPa. That is to say, the formation of Ga₂O₃ must be inhibited in a vacuum environment.

4. Conclusions

In this article, CIGS films without a low-conductivity Ga₂O₃ phase were prepared using vacuum electrodeposition, as supported by the conclusions of Raman and XPS. There were two main reasons for the inhibited Ga₂O₃ formation. The first reason is the inhibition of the H⁺ to H_ads reaction, which decreased the consumption of H⁺ on the cathode surface, as confirmed by the disappearance of the H_ads generation peak in the LSV curve. The second reason is that the OER reaction was promoted by a smoother electrode reaction and a decrease in the electrochemical impedance during deposition. The larger current density in the current density-time curves and the smaller semicircle diameter in the impedance Nyquist plots confirmed this. The inhibition of the H⁺ to H_ads reaction and the promotion of the OER together stabilized the pH in the cathode region, which in turn inhibited Ga³⁺ hydrolysis. The path of Ga³⁺ into the films in the form of Ga₂O₃ was blocked.

In a vacuum environment, the content of Cu, In, and Se in the films increased, the current density of the preparation process increased, and the resistance decreased. The increase in the deposition space of the main elements produced larger particles in the films.