Au-Ag/TiO 2 Thin Films Preparation by Laser Ablation and Sputtering Plasmas for Its Potential Use as Photoanodes in Electrochemical Advanced Oxidation Processes (EAOP)

: Titanium dioxide (TiO 2 ) is widely used, studied, and synthesized using different method-ologies. By a modiﬁcation of the material, it can be applied to wastewater treatment. A combined sputtering-laser ablation setup was used to deposit TiO 2 thin ﬁlms modiﬁed, individually and simultaneously, with gold (Au) and silver (Ag). To investigate the effect of the metal incorporation in titanium and its impact on the photocatalytic activity, with dye discoloration as a pollutant compound model, the deposited ﬁlms were characterized by UV–Vis, photoluminescence, and Raman spectroscopies, as well as by parallel beam X-ray diffraction. The results showed that ﬁlms with different Au and Ag loads, and an 18 nm average crystallite size, were obtained. These metals have an essential effect on the deposited ﬁlm’s compositional, structural, and optical properties, directly reﬂected in its photocatalytic activity. The photocatalytic test results using UV-Vis showed that, after 1 h of applying a 4.8 V electric voltage, a discoloration of up to 80% of malachite green (MG) was achieved, using ultraviolet (UV) light.


Introduction
The preparation of thin films using plasma-based deposition systems represents an innovative route to prepare materials on a nanometric scale, with potential applications in different scientific research fields, such as renewable energy, photocatalysis, and water treatment [1,2]. Plasmas produced by laser ablation and magnetron sputtering have been widely used for thin film deposition, using different experimental setups. Nanomaterials, including thin films, have been studied in recent years because of their properties, which can differ significantly from the same material's properties in a bulk form. It is possible to prepare different thin films with thicknesses from a few, to hundreds, of nanometers by varying the deposition parameters. These techniques, based on the use of plasmas in some cases, have distinctive characteristic; for example, laser ablation has the advantage of congruent transfer, with which it is not necessary to use other reagents or chemical elements to obtain complex oxides with more than two cations in thin film form. On the other hand, the main advantages of the sputtering technique are high deposition rates, good uniformity, and easy control of the properties of the deposited films [3]. TiO 2 has been widely applied as a photocatalyst to mineralize non-biodegradable and recalcitrant chemical compounds into CO 2 and water under UV-light irradiation [4]. This

Microstructural Characterization
Films with thicknesses around 48.4 ± 9.6 nm (average of seven measurements), as is shown in Figure 1, were obtained. This proved that the deposited materials consisted of thin films.

Optical Characterization
To determine the bandgap of the thin films and to know the nature of the metallic particles, UV-Vis measurements in the absorbance and transmittance modes were carried out. Figure 6a shows the absorbance spectra corresponding to the different thin films. For the samples Au/TiO2 and Ag/TiO2, the presence of absorbance bands peaking at 557 nm and 441 nm is clearly observed, attributed to the surface resonance plasmons of Au and Ag, and indicating the presence of nanoparticles with spherical shapes and sizes in the nanometer scale; ranging 15-20 nm for Ag and 20-30 nm for Au. The spectrum corresponding to the sample with Au-Ag does not show defined absorbance band peaks, probably due to the lack of spherical nanoparticle formation [11,33,43,44]. (

Optical Characterization
To determine the bandgap of the thin films and to know the nature of the metallic particles, UV-Vis measurements in the absorbance and transmittance modes were carried out. Figure 6a shows the absorbance spectra corresponding to the different thin films. For the samples Au/TiO 2 and Ag/TiO 2 , the presence of absorbance bands peaking at 557 nm and 441 nm is clearly observed, attributed to the surface resonance plasmons of Au and Ag, and indicating the presence of nanoparticles with spherical shapes and sizes in the nanometer scale; ranging 15-20 nm for Ag and 20-30 nm for Au. The spectrum corresponding to the sample with Au-Ag does not show defined absorbance band peaks, probably due to the lack of spherical nanoparticle formation [11,32,42,43]. surface of the thin film [21,41]. In the case of the film with silver, the signal broadening was lower than the corresponding signal of the films with Au ( Figure 5a).

Optical Characterization
To determine the bandgap of the thin films and to know the nature of the metallic particles, UV-Vis measurements in the absorbance and transmittance modes were carried out. Figure 6a shows the absorbance spectra corresponding to the different thin films. For the samples Au/TiO2 and Ag/TiO2, the presence of absorbance bands peaking at 557 nm and 441 nm is clearly observed, attributed to the surface resonance plasmons of Au and Ag, and indicating the presence of nanoparticles with spherical shapes and sizes in the nanometer scale; ranging 15-20 nm for Ag and 20-30 nm for Au. The spectrum corresponding to the sample with Au-Ag does not show defined absorbance band peaks, probably due to the lack of spherical nanoparticle formation [11,32,42,43]. ( Figure 6b shows the transmittance spectra of the thin films. A red shift of the absorption edge for the Au-Ag/TiO2 is observed. The bandgap was determined using the Tauc method, assuming indirect transitions due to the nature of the semiconductor ( Figure 6c); this was plotted (αhv) 1/2 as a function of the photon energy. Eg values were obtained by a linear fit of the linear portion of the curve, determining its intersection with the photon energy axis as the quotient of the intercept to the slope. The obtained Eg values were 3.72 eV for the thin film with Au, and 3.76 eV for the film with Ag. The higher bandgap values of metal-modified TiO2 thin films compared to the bulk TiO2 bandgap value can be attributed to the thickness of the thin film, which modifies the optical properties, and it is not possible to observe well-defined interference patterns in the transmittance spectrum ( Figure 6b); therefore, the thin film behaves as a kind of electrical insulator. Consequently, the photon effect over the surface causes a bandgap shift, requiring higher energy to be activated [21].
The presence of Ag and Au in the form of nanoparticles was corroborated through transmission electron microscopy (TEM); due to the joint between the thin film and the substrate meaning they could not be analyzed together, the Ag and Au samples were deposited directly on copper TEM grids under the same laser ablation conditions mentioned in the experimental section. Figure 7a corresponds to the Ag nanoparticles, showing quasi-spherical shapes with an average particle size close to 13 nm. Figure 7b corresponds to the Au nanoparticles with a completely different shape, which could have been formed by quasi-spherical particles forming linear agglomerates. The observed sizes are in good agreement with the wavelength position of the surface plasmons discussed in the UV-Vis results ( Figure 6a). Therefore, it can be inferred that the thermal process did not modify the metallic particles deposited by ablation over the thin films. [32,43,44].  Figure 6b shows the transmittance spectra of the thin films. A red shift of the absorption edge for the Au-Ag/TiO 2 is observed. The bandgap was determined using the Tauc method, assuming indirect transitions due to the nature of the semiconductor ( Figure 6c); this was plotted (αhv) 1/2 as a function of the photon energy. Eg values were obtained by a linear fit of the linear portion of the curve, determining its intersection with the photon energy axis as the quotient of the intercept to the slope. The obtained Eg values were 3.72 eV for the thin film with Au, and 3.76 eV for the film with Ag. The higher bandgap values of metal-modified TiO 2 thin films compared to the bulk TiO 2 bandgap value can be attributed to the thickness of the thin film, which modifies the optical properties, and it is not possible to observe well-defined interference patterns in the transmittance spectrum ( Figure 6b); therefore, the thin film behaves as a kind of electrical insulator. Consequently, the photon effect over the surface causes a bandgap shift, requiring higher energy to be activated [21].
The presence of Ag and Au in the form of nanoparticles was corroborated through transmission electron microscopy (TEM); due to the joint between the thin film and the substrate meaning they could not be analyzed together, the Ag and Au samples were deposited directly on copper TEM grids under the same laser ablation conditions mentioned in the experimental section. Figure 7a corresponds to the Ag nanoparticles, showing quasi-spherical shapes with an average particle size close to 13 nm. Figure 7b corresponds to the Au nanoparticles with a completely different shape, which could have been formed by quasi-spherical particles forming linear agglomerates. The observed sizes are in good agreement with the wavelength position of the surface plasmons discussed in the UV-Vis results ( Figure 6a). Therefore, it can be inferred that the thermal process did not modify the metallic particles deposited by ablation over the thin films [32,43,44]. Figure 8 shows the photoluminescence emission spectra obtained for the thin films prepared in this work. In general terms, it is observed that the spectra have the same shape and broadening for each thin film. Four major features can be distinguished in these spectra. The shoulder in the wavelength interval from 375 to 400 nm (3.3 to 3.1 eV) may be attributed to band to band transitions related directly to the bandgap [2,45]. On the other hand, the signal between 430 and 450 nm (2.9 to 2.7 eV) may be associated with the production of excitons (electron-hole pairs) generated by the excitation energy.  Figure 8 shows the photoluminescence emission spectra obtained for the thin films prepared in this work. In general terms, it is observed that the spectra have the same shape and broadening for each thin film. Four major features can be distinguished in these spectra. The shoulder in the wavelength interval from 375 to 400 nm (3.3 to 3.1 eV) may be attributed to band to band transitions related directly to the bandgap [2,45]. On the other hand, the signal between 430 and 450 nm (2.9 to 2.7 eV) may be associated with the production of excitons (electron-hole pairs) generated by the excitation energy.
Finally, the emissions observed in the wavelength interval between 450-475 nm (2.7 to 2.6 eV) and 550-575 nm (2.2 to 2.1 eV) can be attributed to the surface states of oxygen vacancies in TiO2 in its anatase phase [21,46]. Additionally, the differences in emission intensity reveals that metal modification changes the recombination rate of the electronhole pairs. The thin film with Au shows a lower recombination rate, while the thin film with the simultaneous presence of Au and Ag, seems to favor the recombination process.    Figure 8 shows the photoluminescence emission spectra obtained for the thin film prepared in this work. In general terms, it is observed that the spectra have the same shap and broadening for each thin film. Four major features can be distinguished in these spec tra. The shoulder in the wavelength interval from 375 to 400 nm (3.3 to 3.1 eV) may b attributed to band to band transitions related directly to the bandgap [2,45]. On the othe hand, the signal between 430 and 450 nm (2.9 to 2.7 eV) may be associated with the pro duction of excitons (electron-hole pairs) generated by the excitation energy.
Finally, the emissions observed in the wavelength interval between 450-475 nm (2. to 2.6 eV) and 550-575 nm (2.2 to 2.1 eV) can be attributed to the surface states of oxygen vacancies in TiO2 in its anatase phase [21,46]. Additionally, the differences in emission intensity reveals that metal modification changes the recombination rate of the electron hole pairs. The thin film with Au shows a lower recombination rate, while the thin film with the simultaneous presence of Au and Ag, seems to favor the recombination process Finally, the emissions observed in the wavelength interval between 450-475 nm (2.7 to 2.6 eV) and 550-575 nm (2.2 to 2.1 eV) can be attributed to the surface states of oxygen vacancies in TiO 2 in its anatase phase [21,46]. Additionally, the differences in emission intensity reveals that metal modification changes the recombination rate of the electronhole pairs. The thin film with Au shows a lower recombination rate, while the thin film with the simultaneous presence of Au and Ag, seems to favor the recombination process. Figure 9 shows the absorbance spectrum corresponding to the malachite green dye at 50 mg/L concentration. The bands peaking at 617, 427, and 317 nm are characteristic of malachite green dye [47]. The band responsible for the color is the most intense, located at 617 nm in the spectrum [48,49].  Figure 9 shows the absorbance spectrum corresponding to the malachite green dye at 50 mg/L concentration. The bands peaking at 617, 427, and 317 nm are characteristic of malachite green dye [47]. The band responsible for the color is the most intense, located at 617 nm in the spectrum [48,49].  Figure 10 shows the obtained results from the electro and photo-electrocatalytic experiments for each of the thin films (photoanodes). The experiments were carried out with and without light irradiation (PEC and EC, respectively). The degradation of malachite green dye using the three photoanodes was satisfactory. In the EC experiments the lowest color removal, close to 45%, was obtained with the Ag/TiO2 film; followed by the Au/TiO2 film, with 60% of color removal; the maximum of 70% being obtained with the Ag-Au/TiO2 film in a total reaction time of 60 min. Concerning the photo-electrocatalytic experiment (PEC), the Ag/TiO2 and Ag-Au/TiO2 films reached approximately the same discoloration degree, close to 75%, while the thin film of TiO2-Au achieved a maximum degradation of 84% at the same time. These results for the percentage of MG discoloration are directly related to the obtained PL results, because the thin film with Au showed a lower recombination rate of electron-hole pairs, contributing to the TiO2 mechanism for organic molecule degradation. [4,21,29].  Figure 10 shows the obtained results from the electro and photo-electrocatalytic experiments for each of the thin films (photoanodes). The experiments were carried out with and without light irradiation (PEC and EC, respectively). The degradation of malachite green dye using the three photoanodes was satisfactory. In the EC experiments the lowest color removal, close to 45%, was obtained with the Ag/TiO 2 film; followed by the Au/TiO 2 film, with 60% of color removal; the maximum of 70% being obtained with the Ag-Au/TiO 2 film in a total reaction time of 60 min. Concerning the photo-electrocatalytic experiment (PEC), the Ag/TiO 2 and Ag-Au/TiO 2 films reached approximately the same discoloration degree, close to 75%, while the thin film of TiO 2 -Au achieved a maximum degradation of 84% at the same time. These results for the percentage of MG discoloration are directly related to the obtained PL results, because the thin film with Au showed a lower recombination rate of electron-hole pairs, contributing to the TiO 2 mechanism for organic molecule degradation [4,21,29]. The results showed that an electric voltage applied directly affects the discoloratio process when using a semiconductor as anode [21,31], due to the fact that in the photolysi process no discoloration of the solution was observed, as expected. On the other hand, th  The results showed that an electric voltage applied directly affects the discoloration process when using a semiconductor as anode [21,31], due to the fact that in the photolysis process no discoloration of the solution was observed, as expected. On the other hand, the irradiation with UV light facilitated the EC experiments, by contributing to the color disappearance process, in which, as a result, better discoloration was obtained than when the experiments were performed without the application of UV light.

Photoelectrocatalytic Experiments
The three bands corresponding to malachite green at 617, 427, and 317 nm where noticeably diminished in their intensity with the electric voltage (Figure 11a) and, in the presence of electromagnetic radiation in the form of ultraviolet light combined with the electric voltage (Figure 11b), with the three photoanodes. More kinetic studies and measurements of the mineralization of MG are needed to confirm if these photoanodes are capable of achieving complete CO 2 and H 2 O formation from the dye MG molecules. Nevertheless, as can be observed from the discoloration test, the Au-TiO 2 photoanode showed remarkable properties and can be considered a functional photoanode for advanced electrochemical oxidation processes, leading the prospect of having a novel formation route for this supported catalyst. The results showed that an electric voltage applied directly affects the discoloration process when using a semiconductor as anode [21,31], due to the fact that in the photolysis process no discoloration of the solution was observed, as expected. On the other hand, the irradiation with UV light facilitated the EC experiments, by contributing to the color disappearance process, in which, as a result, better discoloration was obtained than when the experiments were performed without the application of UV light.
The three bands corresponding to malachite green at 617, 427, and 317 nm where noticeably diminished in their intensity with the electric voltage (Figure 11a) and, in the presence of electromagnetic radiation in the form of ultraviolet light combined with the electric voltage (Figure 11b), with the three photoanodes. More kinetic studies and measurements of the mineralization of MG are needed to confirm if these photoanodes are capable of achieving complete CO2 and H2O formation from the dye MG molecules. Nevertheless, as can be observed from the discoloration test, the Au-TiO2 photoanode showed remarkable properties and can be considered a functional photoanode for advanced electrochemical oxidation processes, leading the prospect of having a novel formation route for this supported catalyst. The data shown in Figure 12 were used to plot ln (A/A0) vs. time. A pseudo-firstorder kinetic model was fit to these plots [50]. Figure 12a,b show good linear fits for all samples, with constant rates from 0.00987 to 0.2595 min −1 (Table 1). For the EC experiments it can be seen that the load of silver and gold in the Ag-Au-TiO 2 photoanode facilitated the discoloration process compared to the single metal TiO 2 photoanode modification. Nevertheless, for the PEC experiments, the Au-TiO 2 photoanode result in the highest kinetic reaction constant (0.02867 cm −1 ), but like the rest of the photoanode materials, further studies regarding the electric properties of the photoanodes are proposed, to fully understand how the incorporation of a single or combined metallic nanoparticle in the TiO 2 thin film could modify the donor concentration, due to the oxygen vacancies, or the interfacial electron transfer [51]. Nevertheless, for the PEC experiments, the Au-TiO2 photoanode result in the highest kinetic reaction constant (0.02867 cm −1 ), but like the rest of the photoanode materials, further studies regarding the electric properties of the photoanodes are proposed, to fully understand how the incorporation of a single or combined metallic nanoparticle in the TiO2 thin film could modify the donor concentration, due to the oxygen vacancies, or the interfacial electron transfer [51].
(a) (b) Figure 12. Pseudo-first-order kinetic fit of (a) EC and (b) PEC discoloration test of 50 ppm MG at 60 min.  The degree of electrochemical enhancement (E) and the degree of process synergy (S) were calculated using Equations (1) and (2), as reported by [52,53]. where KPEC, KEO, and KPC are the apparent rate constants for the photo-electrocatalytic, electrochemical, and photocatalytic degradation of MG. For all photoanodes, the process synergy was (S) > 0, confirming that the improved performance obtained with PEC degradation is more than a simple addition of the individual EC and photocatalytic degradation processes. For Ag-TiO2, Au-TiO2, and Ag-Au-TiO2 thin films, the S values were 0.01, 0.39, and 0.18, respectively. The calculated E values were 40% for the Ag-TiO2 film, whereas for the Au-TiO2 and Ag-TiO2 in both cases they were >90%. These results are in good agreement with those observed in the kinetic analysis and discoloration tests.
To the best of our knowledge, there has been no previous work that used a hybrid system of laser ablation and sputtering for the formation of thin films of TiO2 modified with noble metals such as Au and Ag. The use of the proposed hybrid deposition system takes advantage of the qualities of each of these techniques for the formation of thin films, with The degree of electrochemical enhancement (E) and the degree of process synergy (S) were calculated using Equations (1) and (2), as reported by [52,53].
where K PEC , K EO , and K PC are the apparent rate constants for the photo-electrocatalytic, electrochemical, and photocatalytic degradation of MG. For all photoanodes, the process synergy was (S) > 0, confirming that the improved performance obtained with PEC degradation is more than a simple addition of the individual EC and photocatalytic degradation processes. For Ag-TiO 2 , Au-TiO 2 , and Ag-Au-TiO 2 thin films, the S values were 0.01, 0.39, and 0.18, respectively. The calculated E values were 40% for the Ag-TiO 2 film, whereas for the Au-TiO 2 and Ag-TiO 2 in both cases they were >90%. These results are in good agreement with those observed in the kinetic analysis and discoloration tests.
To the best of our knowledge, there has been no previous work that used a hybrid system of laser ablation and sputtering for the formation of thin films of TiO 2 modified with noble metals such as Au and Ag. The use of the proposed hybrid deposition system takes advantage of the qualities of each of these techniques for the formation of thin films, with potential applications in photo-electrocatalytic processes (PEC), as revealed in the results presented in this work.
There have been several different works where thin films were obtained with similar methods and used in PEC processes. Nevertheless, only the photocatalytic process was evaluated, with high-power lamps (compared to that of the present work), wavelengths close to the UVC range (greater than the range handled in this work), a greater electrode work area, and also with relatively low concentrations of pollutant in the water (up to a couple of orders of magnitudes less than in the present work) and longer reaction times [2,3,21,[54][55][56][57].

Thin Films Preparation
TiO 2 thin films deposited by magnetron sputtering were modified by the incorporation of Au and Ag, using the laser ablation technique. This experimental setup had been used in previous works with interesting results [1,2,21]. The base pressure was approximately 5 × 10 −5 Torr, and the working pressure was set at 2 × 10 −2 Torr of argon of high purity. A titanium target (purity 99.99%, Kurt J. Lesker Company, Jefferson Hills, PA, USA 2" diameter and 0.25" thickness, was sputtered at a power of 150 W. Films were deposited on ITO (indium and tin oxide) substrates (1 × 1 inch, prepared using a ITO target, purity 99.99%, Kurt J. Lesker Company, Jefferson Hills, PA, USA) kept at room temperature. The target to substrate distance was 12.5 cm, and the deposition time was 30 min. Figure 13 shows schematically the setup used to deposit the metals by laser ablation. In this case, Au and Ag targets (high purity, 99.99%, Kurt J. Lesker Company, Jefferson Hills, PA, USA), 1" diameter and 0.25" thick, were used. The third harmonic (355 nm) of a Nd:YAG laser (Q-Smart 850, Quantel, Lannion, France) was used as an energy source, focusing the laser beam using a 30-cm focal length plano-convex lens. The target to substrate distance was 7 cm. The target was rotated with an electric motor to avoid depletion of material at any given spot. After deposition of the Ti film, the vacuum chamber was evacuated to the base pressure to incorporate simultaneously the gold (Au/TiO 2 ), silver (Ag/TiO 2 ) and Au-Ag (Au-Ag/TiO 2 ) nanoparticles on the Ti film. The ablation conditions were energy per laser pulse of 18 mJ (1000 pulses) for gold and 25 mJ (7000 pulses) for silver. After deposition, the films were subjected to a thermal treatment at 450 • C for two hours, to promote the growth of the anatase crystalline phase through thermal oxidation. We expected to obtain the anatase phase after the thermal treatment, which is well known as a catalyst for reduction-oxidation reactions for organic chemical compound removal.

Thin Films Characterization
The microstructure of the films was studied using Raman spectroscopy; the spectra were obtained using a micro-Raman system (LabRam HR 800, Horiba Jobin Yvon, France), equipped with a confocal microscope (Olympus BX40, Olympus Corporation, Pennsylvania, USA); the samples were analyzed using a wavelength of 532 nm at a power close to 1 mW. The crystalline structure of the deposited thin films was determined by a X-ray diffraction system (Rigaku Ultima IV Diffractometer, Rigaku Corporation, Texas, USA), with the Cu Kα radiation line (λk = 1.5406 Å) operated at 40 kV and 30 mA. The diffraction patterns were recorded using a parallel beam in a 2θ configuration. The incident parallel beam was fixed at 2° from the sample surface, acquiring data from 2θ = 5 to 80° with a Solid-State Detector (D/teX-ULTRA, Rigaku Corporation, Texas, USA) at a speed of 1°/min and a sampling time of 0.02 sec. The photoluminescence properties of the thin films were studied using PL spectroscopy, using a spectrofluorometer (FluoroMax 4, Horiba Jobyn Ivon, France) equipped with a 150 W Xenon lamp as excitation source. The optical absorbance and transmittance of the thin films were measured with a Spectrophotometer (Lambda 35, PerkinElmer, Waltham, MA, USA), from 200 to 800 nm.

Thin Films Characterization
The microstructure of the films was studied using Raman spectroscopy; the spectra were obtained using a micro-Raman system (LabRam HR 800, Horiba Jobin Yvon, France), equipped with a confocal microscope (Olympus BX40, Olympus Corporation, Pennsylvania, USA); the samples were analyzed using a wavelength of 532 nm at a power close to 1 mW. The crystalline structure of the deposited thin films was determined by a X-ray diffraction system (Rigaku Ultima IV Diffractometer, Rigaku Corporation, Texas, USA), with the Cu Kα radiation line (λ k = 1.5406 Å) operated at 40 kV and 30 mA. The diffraction patterns were recorded using a parallel beam in a 2θ configuration. The incident parallel beam was fixed at 2 • from the sample surface, acquiring data from 2θ = 5 to 80 • with a Solid-State Detector (D/teX-ULTRA, Rigaku Corporation, Texas, USA) at a speed of 1 • /min and a sampling time of 0.02 sec. The photoluminescence properties of the thin films were studied using PL spectroscopy, using a spectrofluorometer (FluoroMax 4, Horiba Jobyn Ivon, France) equipped with a 150 W Xenon lamp as excitation source. The optical absorbance and transmittance of the thin films were measured with a Spectrophotometer (Lambda 35, PerkinElmer, Waltham, MA, USA), from 200 to 800 nm.

Photoanodes Preparation and Photoelectrocatalytic Experiments
Photoanodes were formed by cutting a piece of the thin film (1 cm 2 ) and attaching it to a copper wire, using silver paint as electrical contact. The electrical connection between the film and the copper wire was coated with epoxy resin to avoid the copper wire from reacting with the solution inside the photoelectrochemical cell. The photo-electrocatalytic experiments were carried out in a photoelectrochemical cell, which consisted of a Teflon container of 15 cm 3 with a quartz window to allow the UV light (404 nm) to focus directly on the photoanode. During the experiments, air bubbling flow (1 L/min) and constant stirring were used. Graphite was used as cathode for the experiments.
The photo-electrocatalytic activity of the thin films (Au/TiO 2 , Ag/TiO 2 , and Au-Ag/TiO 2 ) was evaluated by monitoring the discoloration of malachite green (MG) dye at different reaction times (5,10,15,20,30, and 60 min). A MG solution of 50 mg/L was prepared with Na 2 SO 4 (0.5 mM) as electrolyte in an acid environment (pH = 1 adjusted with H 2 SO 4 , 95-98% JT Baker). The discoloration was evaluated by applying an electric voltage of 4.8 V with a Triple output DC power supply 0-6 V, 5A/0 ± 25 V,1A (Keysight E3631A, Keysight Technologies, California, USA) in the presence and absence of light using a UV-lamp (365 nm), and with a distance between the window cell and the light source of 10 cm [21,58]. Photolysis and photocatalytic reactions were performed to compare the discoloration obtained for the electro-and photo-electrocatalytic systems. Dye degradation was followed by a decrease of the characteristic MG absorption band at 616 nm, as a function of the reaction time.

Conclusions
Using an alternative deposition method, combining two different plasma deposition techniques, it was possible to obtain semiconductor thin films based on TiO 2 modified with Au, Ag, and Au-Ag, with potential applications in advanced electrochemical oxidation processes (EAOP), to remove non-biodegradable and emerging organic pollutant compounds. The physical and chemical characterization of the prepared materials showed a crystalline material with a suitable phase to carry out the photocatalysis process. The UV-Vis results and TEM images confirmed the presence of metallic Au and Ag nanoparticles on the TiO 2 thin films, with sizes close to 13 nm for the Ag particles. The photoluminescence spectroscopy results revealed different emission bands corresponding to exciton generation, the oxygen surface states, and the bandgap in the anatase phase in the thin film form. In the malachite green dye photo-electrocatalysis tests, it was possible to achieve a maximum 84% color removal after 60 min under UV radiation, applying a 4.8 V electric potential. From the kinetic analysis and the calculation of the degree of electrochemical enhancement (E) and the degree of process synergy (S), it was possible to assess the color removal results and to set a precedent for subsequent studies regarding the electronic properties of thin films.
Furthermore, as future work, we plan to study the mineralization and degradation of the MG molecule, comparing the photo-electrocatalytic process with other EAOP, such as the electro-Fenton (EF) and photoelectro-Fenton (PEF) methods.