Novel Photocatalytic NH₃ Synthesis by NO₃⁻ Reduction over CuAg/TiO₂

Abstract

The highly effective reaction system was investigated for the photocatalytic ammonia synthesis from the reduction of nitrate ions by using the semiconductor photocatalyst, Cu and Ag doped on TiO₂ (CuAg/TiO₂) at room temperature under UV light irradiation (max. 352 nm). In this study, CuAg/TiO₂ gave the high efficiency and the selectivity for the ammonia synthesis by the photoreduction of nitrate in the presence of methanol as a hole scavenger. For the evaluation of the photocatalytic activity over CuAg/TiO₂, various TiO₂ samples, such as standard TiO₂, Cu/TiO₂, and Ag/TiO₂, were evaluated in the same procedure. The chemical properties were investigated by XRD, TEM, XPS, PL, and DRS. We examined the optimum conditions for the experimental factors and the important issues, including the effect of the molar ratio of Cu and Ag onto TiO₂, the optimization of the CuAg amount loaded on TiO₂, the influence of the loading amount of the catalyst on the reduction of nitrate ions, the exploration of the optimum hole scavenger, and the reusability of the optimum photocatalyst. The very efficient conversion of nitrate ions (95%) and the highest selectivity (86%) were achieved in the reaction with the optimum conditions. Here, we reported the process that nitrate ions can efficiently be reduced, and ammonia can be selectively synthesized over CuAg/TiO₂.

1. Introduction

Ammonia (NH₃) has recently been recognized as one of the most important chemical products in the present world [1]. It is used for the chemical fertilizers, dyes, medicines, and so on. Also, it is expected to utilize the energy careers with the hydrogen production and substitute to the CO₂ free fuel [2]. Therefore, it clearly shows that the demand on the NH₃ production expands, as the population increases in the future. However, there are some serious problems in the present NH₃ production. In the current NH₃ synthesis, NH₃ is directly synthesized by N₂ and H₂ over an iron-based catalyst (Haber-Bosch process). However, this method needs the conditions at high temperature (400–500 °C) and high pressure (150–250 bar), which is not a sustainable process in the NH₃ production for a long time. In addition to this disadvantage, a large amount of fossil fuels is consumed, and its production simultaneously releases a huge amount of CO₂ [3]. Therefore, these issues may create significantly negative effects for the environment on the earth. Many studies were evaluated for the NH₃ synthesis with the operation using the low energy without using fossil fuel [4]. As the method for the substitution of Haber Bosch process, various ammonia production processes were investigated such as following methods; the plasma catalysis [5], the electrochemical system [6], the production from biomass [7].

The nitrate ion (NO₃⁻) reduction by using the photocatalysts should be noted as the method to synthesize NH₃. Various photocatalysts were applied into the new approaches for the photocatalytic NH₃ synthesis are as follows: M/TiO₂ (M = Ru, Rh, Pd, Pt) [8], Ru/C [9], Au/SrTiO₃ [10], and BiOCl [11] as the photocatalysts for synthesizing NH₃ at a lower reaction temperature and pressure. For the application to the industrial process, it is suggested that NO₃⁻ is separated from wastewater by the ion exchange method, and the catalyst is added to this treated water including NO₃⁻. After the photocatalytic reduction of NO₃⁻ in the treated water, ammonium ions are generated in the solution. This can be collected as ammonia by the ammonia stripping process. In this process, ammonia is released in the gas phase by adjusting to the basic pH condition. However, the above catalysts did not indicate the high reactivity in the reduction processes because of the poor photocatalytic activity.

Titanium dioxide (TiO₂) is one of the well-known semiconductor photocatalysts, which absorb the UV light and can cause the photocatalytic reaction. Also, TiO₂ has some advantages such as the very high stability in the solution with the various pH, the low cost, and the high practicality due to easy handling [12]. Therefore, the reaction system of the photocatalytic NO₃⁻ reduction to NH₃ over metal-doped TiO₂ may be significant as a green process for the photocatalytic NH₃ synthesis by the reduction of NO₃⁻ in the lower energy without using fossil fuel. As the source of N₂, NO₃⁻ has attracted attention by the recent studies because it is the very toxic pollutant in the waste water and gives the very harmful effect to human health [13]. Therefore, the NH₃ synthesis by using NO₃⁻ as a raw material may enable to contribute to the improvement of the environmental issues by the decontamination of waste water including NO₃⁻ and the development of the excellent ecological approach for the NH₃ production.

However, the reduction efficiency of NO₃⁻ and the selectivity of NH₃ are still very poor in the photocatalytic reduction of NO₃⁻ [14]. Methanol (CH₃OH) may be applied into the reduction of NO₃⁻ to improve this problem. Generally, CH₃OH was used as a hole scavenger and it was oxidized to formaldehyde by holes. Also, the reduction simultaneously produces reactive hydrogen [15], which results in the more efficient reduction of NO₃⁻. Thus, CH₃OH may become very useful for the photocatalytic NO₃⁻ reduction.

In the present study, we used Cu and Ag as doping metals for the preparation of photocatalysts. Cu is one of the well-known metals and it is widely applied to the photocatalysts because it has several advantages of the high electro conductivity and the low cost. Noble metals, such as Au, Ag, and Pt show the localized surface plasmonic resonance (LSPR), which has the ability to absorb the visible light [16]. It can improve the photoabsorption by its light trapping effects [17]. Also, it is indicated that these metals nanoparticles can perform as an electron donor to promote the electron transfer from metal to semiconductor [18,19]. In addition to this function, these metals can suppress the recombination of surface electrons and holes [20,21]. In the photocatalytic reaction over these metals doped TiO₂, Kamat et al. proposed that the excited electrons were transferred in these metals to TiO₂. Therefore, the Cu and Ag doped TiO₂ photocatalyst minimized the recombination of photoelectron-hole and gave the high photoabsorption and the proper bandgap for the photocatalytic reaction [22]. Here, we investigated the method for the selective NH₃ synthesis by the photocatalytic reduction of NO₃⁻ over Cu and Ag doped TiO₂ (CuAg/TiO₂). The reaction was operated in the presence of 50 ppm of NO₃⁻ aqueous solution including 10 vol% CH₃OH at room temperature under UV irradiation (λ: max. 352 nm). In the present work, we examined in detail the photocatalytic activity, the optimization of the experimental conditions, and the reusability of the optimum catalyst. This reaction system would give the excellent approach for the selective NH₃ synthesis by photocatalytic NO₃⁻ reduction over the optimized photocatalyst.

2. Materials and Methods

2.1. Materials

Titanium dioxide (TiO₂-P25, 99.50%) was purchased from Degussa Co. Cu 1000 ppm standard solution Cu(NO₃)₂ in 0.1 mol/L HNO₃ was obtained from FUJIFILM Wako Pure Chemical Corp. Ag 1000 ppm standard solution (AgNO₃ in 0.1 mol/L HNO₃) and potassium nitrate (KNO₃) were obtained from Nacalai Tesque Inc. All of the reagents and the materials used in this research were analytical grade and were used without further purification. KNO₃ and distilled water were used for the preparation of the simulated waste water including NO₃⁻. Methanol (CH₃OH) was added to the simulated waste water as a hole scavenger. For the evaluation of the optimum hole scavenger, ethanol (C₂H₅OH), isopropyl alcohol ((CH₃)₂CHOH), and t-butyl alcohol ((CH₃)₃COH) were applied in this study.

2.2. Preparation of TiO₂ Samples

Cu²⁺ 1000 ppm standard solution, Ag⁺ 1000 ppm standard solution and 300 mg of TiO₂ were added in 10 vol% CH₃OH aqueous solution. The total amount of metals (Cu and Ag) were 1 wt% for the amount of TiO₂. The mixed solution was stirred under blacklight irradiation (λ: max 352 nm) for an hour. After the centrifugation, the supernatant was removed, and the photocatalyst was dried under vacuum overnight. The different TiO₂ samples were obtained in this preparation method (Table S1). The obtained products were used for NH₃ synthesis by NO₃⁻ reduction.

2.3. Characterization

X-ray diffraction (XRD) was performed by Ultima Ⅳ, RIGAKU with Cu-Kα radiation, equipped with the graphite monochromator on the diffracted beam. The morphology of the catalysts was operated by transmission electron microscope (TEM) using JEM-1011, HITACHI. X-ray photoelectron spectroscopy (XPS) was measured by PHI Quantera SXM, ULVAC-PHI. The C1s peak at 284.6 eV was used to calibrate all of the binding energy. The photoluminescence spectra (PL) were given by a fluorescence spectrophotometer (RF-5300PC, SHIMADZU) at an excitation wavelength of 330 nm using D₂ lamp at ambient temperature. UV-Vis diffuse reflectance spectra (DRS) were operated on V-750 spectrophotometer, JASCO in the range of 200–800 nm using D₂ lamp.

2.4. Photocatalytic Reduction of NO₃⁻ to NH₃

The photocatalytic reduction of NO₃⁻ was performed in the Pyrex reactor. Firstly, 50 ppm of NO₃⁻ aqueous solution (45 mL) including 10 vol% CH₃OH was prepared, and 30 mg of the photocatalyst was added to the solution. The suspension was stirred with a magnetic stirrer at room temperature under UV irradiation for 3 hours. As the light source, black light (max: 352 nm) was used for NO₃⁻ reduction. The measurements of the concentration of ions (NO₃⁻, NO₂⁻, and NH₄⁺) in the solution after the reaction were performed by ion chromatography (Metrohm Compact IC 761).

3. Results and Discussion

3.1. Calculation of the NO₃⁻ Conversion and Selectivity

The NO₃⁻ conversion and the selectivity of the products were calculated by the following equations:

$${\mathrm{C}}_{{\mathrm{NO}}_3{}^{-}}=\frac{{\left[{\mathrm{n}}_{{\mathrm{NO}}_3{}^{-}}\right]}_0-{\left[{\mathrm{n}}_{{\mathrm{NO}}_3{}^{-}}\right]}_{\mathrm{t}}}{{\left[{\mathrm{n}}_{{\mathrm{NO}}_3{}^{-}}\right]}_0}$$

(1)

$${\mathrm{S}}_{{\mathrm{NO}}_2{}^{-}}=\frac{{\left[{\mathrm{n}}_{{\mathrm{NO}}_2{}^{-}}\right]}_{\mathrm{t}}}{{\left[{\mathrm{n}}_{{\mathrm{NO}}_3{}^{-}}\right]}_0-{\left[{\mathrm{n}}_{{\mathrm{NO}}_3{}^{-}}\right]}_{\mathrm{t}}}$$

(2)

$${\mathrm{S}}_{{\mathrm{NH}}_4{}^{+}}=\frac{{\left[{\mathrm{n}}_{{\mathrm{NH}}_4{}^{+}}\right]}_{\mathrm{t}}}{{\left[{\mathrm{n}}_{{\mathrm{NO}}_3{}^{-}}\right]}_0-{\left[{\mathrm{n}}_{{\mathrm{NO}}_3{}^{-}}\right]}_{\mathrm{t}}}$$

(3)

$${\mathrm{S}}_{{\mathrm{N}}_2\mathrm{product}}{=100-(\mathrm{S}}_{{\mathrm{NO}}_2{}^{-}}{+\mathrm{S}}_{{\mathrm{NH}}_4{}^{+}})$$

(4)

where [n]₀ is the amount of material (mmol) at time = 0 and [n]_t is the amount of material (mmol) at time = t.

3.2. Characterization of Photocatalysts

X-ray diffraction (XRD) spectra of various TiO₂ samples were shown in Figure 1. The TiO₂ P25 particle is consisted of anatase phase and rutile phase. The structure ratio of anatase phase and rutile phase are 80% and 20%, respectively. The characteristic patterns of anatase phase were clearly found at 2θ = 25.2° (101), 37.0° (103), 37.9° (200), 48.0° (200), 53.9° (105), 55.0° (211), 62.8° (204), and 75.1° (215). The typical patterns of the rutile phase were observed at 2θ = 27.5° (110), 36.0° (101), 41.3° (111), 56.5° (220), 62.0° (002), 64.2° (311), 69.0° (301), and 70.3° (112) [23]. These sharp peaks showed that the crystallinity of all TiO₂ samples was very high. However, the peaks of Cu and Ag were not confirmed due to the relatively low amount of metals loaded on TiO₂. Also, the XRD results of other TiO₂ samples used in this study were summarized in Figure S1.

The TEM images of the photocatalysts were shown in Figure 2. As a result, the existence of Cu and Ag was not confirmed, as the amount of Cu and Ag was very low compared with the amount of TiO₂. On the other hand, it was obviously confirmed that all of the TiO₂ samples had the same morphology. Therefore, it was indicated that the morphology of photocatalyst could not be changed by doping Cu and Ag. TEM images of other TiO₂ samples used in this study were shown in Figure S2. All of the TiO₂ samples were nanoparticles and the size were roughly 20–30 nm. Cu and Ag particles were not observed because they are too small to be detected by TEM.

For the identification of the composition of Cu and Ag on TiO₂, X-ray photoelectron spectroscopy (XPS) was applied to the Cu_0.9Ag/TiO₂ composite photocatalyst (Figure 3). The peak of C1s was used for the calibration, which was attributed to the contaminated carbon [24,25].

In the O 1s signals, the presence of Ti-O of titania at about 530 eV could be confirmed, and the peak at 528.9 eV was attributed to CO (Figure 3a) [26]. The spectrum region corresponding to Ti 2p revealed that two peaks at 464.1 eV and 458.6 eV were due to Ti 2p_1/2 and Ti 2p_3/2, respectively. As the results, we assigned it with 4+ oxidation number by analyzing the intensity of Ti 2p_3/2 of this sample. These peaks were highly sharp and intense, so that it indicated that this photocatalyst consists only of Ti⁴⁺ species [27]. This result shows that Cu_0.9Ag/TiO₂ was properly prepared and its structure was not changed by the deposition of Cu and Ag under UV irradiation.

The narrow scan of Cu 2p of Cu_0.9Ag/TiO₂ showed two peaks at 950.9 eV and 931.9 eV and these peaks were attributed to Cu 2p_1/2 and Cu 2p_3/2, respectively (Figure 3c). However, it cannot easily be distinguished between Cu⁰ and Cu²⁺, because the difference of the binding energy between Cu⁰ and Cu²⁺ is only 0.1 eV. In addition, Ag 3d peaks at 373.2 eV and 367.1 eV were corresponding to Ag 3d_3/2 and Ag 3d_5/2, respectively (Figure 3d) [28]. These characteristic peaks indicated that it was existed as metal Ag [29].

As a result, the existence of Cu and Ag was confirmed in the samples, and all of elements including O, Ti, Cu, and Ag were observed. It clearly showed that Cu and Ag were successfully formed on TiO₂. In addition to these results, XPS spectra of other TiO₂ samples used in this study were shown in Figures S3–S6.

UV-Vis diffuse reflectance spectra (DRS) were evaluated to identify the optical absorption properties (Figure 4). The absorption edge of the standard TiO₂ was about 380 nm. All of TiO₂ samples were confirmed to show the light absorption in the UV region, and the intensity was enhanced by loading Cu and Ag doped on TiO₂. In this research, the band-gap energy was calculated by following formula [30]:

E_g (eV) = 1240/λ_g (nm)

(5)

where E_g refers to the band-gap energy and λ_g is the wavelength at the absorption edge [16]. The band-gap energies E_g of TiO₂ and Cu_0.9Ag/TiO₂ were estimated to 3.26 eV and 2.92 eV, respectively.

Photoluminescence (PL) spectra were measured to examine the recombination of electron-hole over the photocatalysts (Figure 5). All of the PL spectra were obtained in the range of 350–600 with an excitation wavelength at 330 nm. The comparison of TiO₂ and metals-doped TiO₂ samples was investigated in Figure 5a. Compared with standard TiO₂, metal-doped TiO₂ samples showed the low intensity. It revealed that loading metals gave the decrease of the PL intensity, and the recombination of photogenerated electron and hole was suppressed by doping metals on TiO₂. This reason may be considered it that the electrons were efficiently trapped by CuAg bimetal. The PL spectra of the TiO₂ samples with the different amount of the CuAg doped on TiO₂ were shown in Figure 5b. As the amount of CuAg increased, the PL intensity decreased. Therefore, the doping of Cu and Ag gave the good effect on the recombination of charge careers, which showed the high photocatalytic activity.

3.3. Photocatalytic Activity in NO₃⁻ Reduction to NH₃

Firstly, the evaluation for the photocatalytic activity in the reduction of NO₃⁻ to NH₃ was performed over TiO₂ alone, Cu/TiO₂, Ag/TiO₂, and CuAg/TiO₂ in the presence of CH₃OH under UV irradiation for three hours (Figure 6). The amount of photocatalyst was 30 mg in all of test cases. In the photocatalytic NO₃⁻ reduction, NH₄⁺, NO₂⁻, and N₂ seem to be mainly generated as the products. In addition, CH₃OH was oxidized to HCHO by a photocatalytic reaction as a hole scavenger and the reduction produced hydrogen, which played an important role in NO₃⁻ reduction by following the equations:

2H⁺ + 2e⁻ → H₂

(6)

NO₃⁻ + H₂ → NO₂⁻ + H₂O

(7)

NO₃⁻ + 4H₂ → NH₄⁺ + H₂O + 2OH⁻

(8)

NO₃⁻ + 5/2H₂ → 1/2N₂ + 2H₂O + OH⁻

(9)

As a result, Cu_0.9Ag/TiO₂ gave the very efficient NO₃⁻ conversion (96%) and the highest NH₃ selectivity (85%), while the NO₃⁻ conversion and the NH₃ selectivity over TiO₂ alone were only 44.9% and 33.6%, respectively. Also, the reaction had not occurred without the photocatalyst. In addition to these results, the comparison of the photocatalytic activity with other catalysts was performed (Table S2). Consequently, it clearly showed that loading the optimum amount of Cu and Ag onto TiO₂ gave the excellent photocatalytic activity in the reduction of NO₃⁻ to NH₃. Hence, CuAg/TiO₂ was very useful to photocatalytic NO₃⁻ reduction rather than other TiO₂ samples.

3.4. Optimization of Structure and Loading Amount of CuAg/TiO₂

We optimized the molar ratio of Cu and Ag doped on TiO₂ (Figure 7). The total amount of Cu and Ag was controlled to be 1 wt% of TiO₂ amount. As shown in the result, the highest NO₃⁻ conversion and the NH₃ selectivity were achieved over Cu_0.9Ag/TiO₂ (Cu:Ag = 0.9:1). Consequently, both the NO₃⁻ conversion and NH₄⁺ selectivity was respectively almost same when Cu_0.3Ag/TiO₂ and Cu_0.9Ag/TiO₂ were used. Silver is one of the precious metals and it is so expensive for the application to the chemical industry. This means that the catalyst with the lower Ag amount (Cu:Ag = 0.9:1) is better than the catalyst with higher Ag amount (Cu:Ag = 0.3:1). Therefore, we selected Cu:Ag = 0.9:1 as an optimum molar ratio in this study.

In addition to the optimization of the molar ratio, the loading amount of Cu_0.9Ag on TiO₂ was investigated (Figure 8). The NO₃⁻ conversion and the NH₃ selectivity were improved as the loading amount of Cu_0.9Ag increased. When the excess amount of the catalyst was used, the photocatalytic activity decreased. The highest NO₃⁻ conversion and the NH₃ selectivity were given by 1 wt% of Cu_0.9Ag doped on TiO₂. Therefore, 1 wt% Cu_0.9Ag/TiO₂ was used as an optimum photocatalyst in this study.

After the optimization to the structure of the photocatalyst, we investigated the optimum dosage of the photocatalyst in the reduction of NO₃⁻ to NH₃ over 1 wt% Cu_0.9AgTiO₂ (Figure 9). This experiment was performed in the presence of CH₃OH under UV irradiation for nine hours, and the photocatalysts in the range of 10–40 mg were used. Loading 30 mg and 40 mg of the catalyst gave the significantly high NO₃⁻ reduction rate (Figure S7). Thus, we selected 30 mg as an optimum dosage of the photocatalyst in the NO₃⁻ reduction, because it can perform the reduction of NO₃⁻ with the lower catalyst amount. In this study, we fixed 30 mg of 1 wt% Cu_0.9Ag/TiO₂ as the optimal conditions for the reaction system.

3.5. Effect of Hole Scavenger on NO₃⁻ Reduction to NH₃

The hole scavenger was also the very important factor in the reaction with the photocatalyst. Figure 10 shows the effect of the hole scavenger on the reduction of NO₃⁻ to NH₃. As a result, CH₃OH indicated the most efficient NO₃⁻ conversion and the NH₃ selectivity. In the photocatalytic NO₃⁻ reduction, CH₃OH was turned to HCHO with h⁺, and then H₂ was simultaneously released. Generated H₂ on the photocatalyst surface was useful to reduce NO₃⁻ and NO₂⁻ (Equations (6)–(11)).

NO₂⁻ + 3H₂ → NH₄⁺ + 2OH⁻

(10)

2NO₂⁻ + 2H₂ → N₂ + 4OH⁻

(11)

On the other hand, it is considered that C₂H₅OH, (CH₃)₂CHOH, and (CH₃)₃COH were not so reactive, because of the larger steric hindrance. Therefore, CH₃OH could give more H₂ than those with other hole scavengers, and CH₃OH was selected as an optimal hole scavenger in this study.

3.6. Effect of pH

We checked the pH effect in the solution during the NO₃⁻ reduction under UV irradiation for three hours (Figure 11). The initial pH value was about six without the pH adjustment. The NO₃⁻ conversion and the selectivity of NH₃ decreased under pH = 5. This reason may be owing to it that the protons in the solution were not sufficiently reacted with NO₃⁻ adsorbed on the catalyst surface, as the protons repelled to the positive charges on the photocatalyst surface. Under alkaline conditions, the NO₃⁻ conversion was kept high, while the NH₃ selectivity decreased compared with the case of pH = 6. There are two proposed reasons; i) some •OH radicals were generated in the solution, which promoted the oxidation of methanol. Hence, the reduction rate of NO₃⁻ was simultaneously enhanced and more NO₂⁻ ions and N₂ products were generated; ii) the H⁺ concentration in the solution decreased under alkaline conditions, which leads to inhibit further reductions of NO₂⁻ to NH₄⁺. On the other hand, the high NO₃⁻ conversion and the NH₃ selectivity were achieved without the pH adjustment (pH = 6). This result may be shown that it was given by the sufficiently high H⁺ concentration and the good adsorption of NO₃⁻ on the catalyst surface under pH = 6.

3.7. Reusability of Photocatalyst in NO₃⁻ Reduction to NH₃

We investigated the reusability of 1 wt% Cu_0.9Ag/TiO₂ photocatalyst in the reduction of NO₃⁻ to NH₃ under UV irradiation for three hours. Five repeated experiments were operated to evaluate the stability of the optimum photocatalyst. As shown in Figure 12, the NO₃⁻ conversion and the NH₃ selectivity were changed for each cycle. It may be owing to the loss of the catalyst amount in the operation for sample collection. Therefore, it is considered that the photocatalytic activity was not degraded by five cycles. It clearly shows that the stability of 1 wt% Cu_0.9Ag/TiO₂ in this research.

3.8. Reaction Mechanism of Photocatalytic NO₃⁻ Reduction to NH₃

Scheme 1 shows the proposed reaction mechanism of CuAg/TiO₂ in this study. First, CuAg/TiO₂ is stimulated by UV light, and electrons and holes are generated. The photogenerated electrons in conduction band are transferred to valence band, and then are moved to CuAg alloy. NO₃⁻ is reduced to NO₂⁻ with the electrons, and NO₂⁻ is reduced to NH₃ with the electrons and protons. The hole scavenger, CH₃OH is oxidized to HCOOH with holes, and H₂ was released, which gives further reduction of NO₃⁻. The oxidation of methanol to formaldehyde was confirmed by the experiment shown in Figure S7.

The reaction processes can be expressed as following formulas:

$$\mathrm{Photocatalyst}\stackrel{\mathrm{h}\mathsf{\nu },\mathrm{cat}.}{\to }{\mathrm{h}}^{+}+{\mathrm{e}}^{-}$$

(12)

NO₃⁻ + 2H⁺ +2e⁻ → NO₂⁻ + H₂O

(13)

NO₃⁻ + 10H⁺ +8e⁻ → NH₄⁺ + 3H₂O

(14)

NO₃⁻ + 6H⁺ +5e⁻ → 1/2N₂ + 3H₂O

(15)

4. Conclusions

The significantly effective photocatalyst CuAg/TiO₂ was prepared for the highly selective NH₃ synthesis by photocatalytic reduction of NO₃⁻. Cu to Ag molar ratio of 0.9:1 and 1 wt% Cu_0.9Ag doped TiO₂ indicated the excellent NO₃⁻ conversion and the NH₃ selectivity. These results were brought by the advantages of the CuAg alloy and the presence of CH₃OH in the reduction of NO₃⁻. Loading Cu and Ag onto TiO₂ gave the effective separation of electron-hole pairs and enhanced the light absorption in UV region. CH₃OH was used as the optimum hole scavenger for the source of H₂. From the result of the cycle test, the stability of CuAg/TiO₂ was indicated by five cycles.