Photocatalytic CO₂ reduction is one of ways to have potential to recycle and reuse CO₂. It is reported that CO₂ could be converted into fuel species such as CO, CH₄, CH₃OH, etc., by photocatalyst [1-3]. TiO₂ is the most popular photocatalyst performing with ultraviolet light (UV) which is used for CO₂ reduction [1-3]. However, we have to consider that pure TiO₂ can only work under UV light illumination accounting for only 4% in sunlight [4]. The visible light (VIS) and infrared light (IR) account for 44 % and 52 % of solar energy reaching the earth, respectively [4]. If VIS and IR could be utilized, the CO₂ reduction performance of TiO₂ photocatalyst would be improved.

As to the photocatalyst to extend the absorption of light wavelength from UV to VIS, many approaches have been tried [5-14]. We can introduce a metal doping as one of the popular attempts. Cu is a popular metal dopant. Cu/TiO₂ has exhibited the absorption of light whose wavelength is ranged from 400 nm to 800 nm and has produced CO of 0.5 µmol/g and H₂ of 4 µmol/g [5]. Cu₂O/TiO₂ has produced 80 mmol/g of CO under the Xe lamp illumination whose wavelength of light is 320 – 780 nm [6]. Cu ultrathin TiO₂ absorbing the light whose wavelength is ranged from 400 nm to 800 nm has produced CO of 7 µmol/g [7]. Cu₂O/TiO₂ heterostructures absorbing the light whose wavelength is ranged from 300 nm to 650 nm has produced CO of 2 mmol/g [8]. Pd is also popular as a metal dopant. Pd/TiO₂ nanowire has exhibited the absorption of light whose wavelength is 350 – 700 nm, which has produced CH₄ yield of 26.7 mmol/g and CO yield of 50.4 mmol/g [9]. Pd/TiO₂ (3 wt% of Pd) extending the absorption limit up to 700 nm has produced CH₄ of 4.2 mmol/g and CO of 2.1 mmol/g [10]. Zn and Pd co-modified TiO₂ has performed CH₄ yield of 53.3 mmol/g under the illumination condition of 500 W Xe arc lamp whose wavelength of light is 290 – 800 nm [11]. Pt is another candidate as a metal dopant. Graphene-wrapped Pt/TiO₂ has exhibited the light absorption from 300 nm to 750 nm, resulting in CO production of 320 mmo/g and CH₄ production of 45 mmol/g [12]. Pt/TiO₂ synthesized by thermal hydrolysis of two different precursors has performed the light absorption from 200 nm to 700 nm and produced CH₄ of 0.73 mmol/g and CO of 0.17 mmol/g [13]. Nanocrystals-supported PtRu/TiO₂ has performed the light absorption ranged from 300 nm to 750 nm and CH₄ of 300 mmol/g [14].

Regarding the photocatalyst to extend the absorption of light wavelength up to IR, there are some reports [15-18]. W₁₈O₄₉/g-C₃N₄ composite has performed the production of CO of 45 mmol/g and CH₄ of 28 mmol/g under the illumination condition whose wavelength is ranged from 200 nm to 2400 nm [15]. WS₂/Bi₂S₃ nanotube has performed the absorption of VIS and near IR light (wavelength: 420 nm – 1100 nm), which has produced CH₃OH of 28 mmol/g and C₂H₅OH of 25 mmol/g [16]. CuInZnS decorated g-C₃N₄ has exhibited the absorption performance of light whose wavelength is ranged from 200 nm to 1000 nm, producing CO of 38 mmol/g [17]. Hierarchical ZnIn₂S₄ nanorods has prepared by solvothermal method, which has produced CO of 54 mmol/g and CH₄ of 9 mmol/g [18].

The authors’ previous studies [19,20] have prepared P₄O₁₀/TiO₂ which could extend the absorbed wavelength up to IR. Under IR light illumination condition, the largest molar quantity of CO per unit weight of photocatalyst for P₄O₁₀/TiO₂ film in the case of CO₂/H₂O is 2.36 mmol/g, while that in the case of CO₂/NH₃ is 33.4 mmol/g [19,20].

Enhancing the gas movement around photocatalyst is another way to improve the CO₂ reduction performance further. According to some reports, the membrane reactor which separates the product from the reaction surface of photocatalyst promotes the photocatalytic CO₂ reduction [21-23]. It is clarified from the authors’ calculation that the mass transfer time of 10⁵ s to 10^-1 s is slower than the photo reaction time of 10^-9 s to 10^-15 s [23]. Therefore, the mass transfer is thought to be the inhibition factor to speed up photocatalytic reaction. Another reason causing the low reforming rate of photocatalytic CO₂ reduction is the re-organization of the products. Since the reaction surface is covered by products, the movement of the reactants to the reaction surface is prevented and the reverse reaction, i.e. re-oxidization, which reproduces CO₂ from products such as CO and CH₄ is occurred. Consequently, it is desirable that CO and CH₄ are removed from the reaction surface as soon as they are produced. As a result, the reactants, i.e. CO₂ and water vapor or NH₃ can continue to react on the reaction surface [23]. Fuel production can be sustained by maintaining the non-equilibrium. One of new approach to enhance the gas movement around the photocatalyst is the natural thermosiphon movement of gasses around TiO₂ photocatalyst using black body material which has been conducted by the authors [24]. When the heat capacity of black body material is set to be adequate, the CO₂ reduction performance is improved by the natural thermosiphon movement of gasses around TiO₂ photocatalyst due to the temperature rise of gas in the reactor [24]. The maximum concentration of formed CO using black body materials has performed 2 to 5 times as large as that without using black body material [24]. Though there are some previous studies to extend the absorption range to IR [15-20] and to enhance the gas movement around the photocatalyst by the natural thermosiphon movement of gasses around TiO₂ photocatalyst using black body material [24] in order to improve the CO₂ reduction performance of photocatalyst, there is no study investigating the combination effect of these two trials on the CO₂ reduction performance of photocatalyst. If the synergy effect on the CO₂ reduction performance is confirmed, it is a new finding which can propose the novel approach to improve the CO₂ reduction performance of photocatalyst.

The purpose of this study is to investigate the coupled effect of P₄O₁₀/TiO₂ photocatalyst that absorbs the light of wavelength up to IR and the gas movement enhanced by introduced black body. This study also investigates the impact of molar ratio of CO₂/NH₃ on the CO₂ reduction characteristics of P₄O₁₀. Although many previous studies investigated the CO₂ reduction reacting with H₂O or H₂ [25,26], it is thought that NH₃ is superior to H₂O and H₂ due to having 3H⁺. The reaction scheme to reduce CO₂ with NH₃ can be shown as follows [27,28]:

<Photocatalytic reaction>

TiO₂ + hn -> h⁺ + e^-                                                                                                                                                  (1)

<Oxidization reaction>

2NH₃ -> N₂ + 3H₂                                                                                                                                                    (2)

H₂ -> 2H⁺ + 2e^-                                                                                                                                                        (3)

<Reduction reaction>

H⁺ + e^- -> ·H                                                                                                                                                            (4)

CO₂ + e^- -> ·CO₂^-                                                                                                                                                     (5)

CO₂^- + H⁺ + e^- -> HCOO^-                                                                                                                                        (6)

HCOO^- + H⁺ -> CO + H₂O                                                                                                                                     (7)

CO₂ + 8H⁺ + 8e^- -> CH₄ + 2H₂O                                                                                                                           (8)

We investigate the impact of the natural thermosiphon movement of gasses around P₄O₁₀/TiO₂ photocatalyst created by black body material on the CO₂ reduction performance of P₄O₁₀/TiO₂ changing the wavelength of illuminated light of UV + VIS + IR, VIS + IR, and IR only. In addition, this study also clarifies the optimum molar ratio of CO₂/NH₃ for the CO₂ reduction performance of P₄O₁₀/TiO₂ to understand the photochemical reaction under the condition promoting the mass transfer surrounding the photocatalyst by black body material.

The Preparation Procedure of P₄O₁₀/Tio₂ Film

The TiO₂ film used in this study was prepared by sol-gel and dip-coating process [19,20,29]. [(CH₃)₂CHO]₄Ti (purity: 95 wt%, producer: Nacalai Tesque Co., Kyoto, Japan) of 0.3 mol, anhydrous C₂H₅OH (purity: 99.5 wt%, producer: Nacalai Tesque Co., Kyoto, Japan) of 2.4 mol, distilled water of 0.3 mol, and HCl (purity: 35 wt%, producer: Nacalai Tesque Co., Kyoto, Japan) of 0.07 mol were mixed to prepare the TiO₂ sol solution. The TiO₂ film was coated on a netlike glass fiber (SILLIFGLASS U, producer: Nihonmuki Co., Tokyo, Japan) via sol-gel and dip-coating processes. The glass fiber with the diameter of about 10 mm, which is weaved as a net, is assembled to be the diameter of about 1 mm. According to the specification on netlike glass fiber, the porous diameter of glass fiber and the specific surface area is approximately 1 nm and 400 m²/g, respectively. The netlike glass fiber consists of SiO₂ of 96 wt%. The netlike glass fiber has the opening space of about 2 mm × 2 mm. The netlike glass fiber has a porous characteristic, resulting that the netlike glass fiber can trap the TiO₂ film easily via sol-gel and dip-coating processes. In addition, it can be expected that CO₂ and reductant such as NH₃ are more easily absorbed by the prepared photocatalyst since the netlike glass fiber has a porous characteristic. The netlike glass fiber is cut to be the disc form with the diameter of 50 mm and the thickness of 1 mm. The dipping speed of the netlike glass disc into TiO₂ sol solution was controlled at 1.5 mm/s and the speed of drawing up was fixed at 0.22 mm/s. Then, the film was dried out and fired by controlling a firing temperature (FT) and a firing duration time (FD), resulting that the TiO₂ film was fastened on the base material. The FT and FD were set at 623 K and 180 s, respectively.

In this study, P₄O₁₀ which had been identified by XPS analysis [19] was made from the red P by a mechanical synthesis [30]. The red P (average diameter: 75 mm; producer: Nacalai Tesque Co., Kyoto, Japan) was charged in a ball mill crusher (AV-1, producer: Asahi Rika Factory, Chiba, Japan) with Al₂O₃ ball whose diameter of 3/8 inch (HD-10, producer: NIKKATO CORPORATION, Osaka, Japan). The weight ratio of Al₂O₃ balls to red P particles in the ball mill crusher was set at 20 [30]. Rotation with the speed of 600 rpm was kept for 12 hours, after that the P₄O₁₀ was prepared.

The prepared P₄O₁₀ particles were put into TiO₂ sol solution and mixed with TiO₂ sol solution by a magnetic stirrer for 60 min. After that, the netlike glass disc was immersed into this mixed solution. The following process was same as explained above. The weight ratio of P₄O₁₀ to TiO₂ was 2.0 wt%, which was confirmed by EPMA analysis quantitatively.

The Preparation Procedure of Black Body Material

This study prepared a black body material by spraying the black body spray (TA410KS, produced by ICHIKEN TASCO Corp., Osaka, Japan) on both surfaces of the Cu disc. The emissivity of black body spray was 0.94. The Cu solid disc with the diameter of 50 mm and the thickness of 1.4 mm (Iwasaki syoten Corp., Osaka, Japan) was adopted as a base material for spraying black body spray. The Cu solid disc had the diameter of 50 mm which was equal to the inside diameter of the reactor, as explained below. The emissivity of the polished surface of Cu was 0.01 [31]. The used Cu solid disc had a purity of 99.90 % [32]. The specific heat, thermal conductivity, and thermal diffusivity of Cu at 30 ? were 0.386 kJ/(kg·K), 398 W/(m·K), and 117 mm²/s, respectively [31]. Three Cu solid discs sprayed by the black body spray were installed in the reactor in this study, referring to the authors’ previous study [24]. According to the authors’ previous study [24], the promotion of CO₂ reduction performance was not obtained by one Cu solid disc but three Cu solid discs. Figure 1 displays not only the black body material prepared by this study but also and the Cu solid disc before spraying black body spray [24].

Figure 1: Photo of black body material prepared by this study and Cu solid disc before playing black body spray.

The Characterization Procedure of P₄O₁₀/Tio₂ Film

This study evaluated the characteristics of the external and crystal structure of P₄O₁₀/TiO₂ film by SEM (JXA-8530F, producer: JEOL Lt., Tokyo, Japan) and EPMA (JXA-8530F, producer: JEOL Ltd., Tokyo, Japan) [19,20,29]. The netlike glass disc which was used for a base material to coat TiO₂ film cannot conduct electricity, resulting that we deposited the vaporized Pt by means of the Pt coating device (JEC-1600, producer: JEOL Ltd., Tokyo, Japan) on the surface of the TiO₂ film before the characterization. The deposited Pt has the thickness of 15 nm. The electrode emitted the electrons to the sample by setting the acceleration voltage and the current at 15 kV and 3.0 × 10^-8 A respectively, to analyze the external structure of TiO₂ film by means of SEM. We analyzed the character X-ray by means of EPMA at the same time, resulting that the amount of chemical element was estimated based on the relationship between the character X-ray energy and the atomic number. The space resolution of SEM and EPMA is 10 µm. The structure of prepared P₄O₁₀/TiO₂ photocatalyst was analyzed by the EPMA.

In addition, this study also evaluated the chemical composition state of P₄O₁₀/TiO₂ film by XPS (PHI Quantera SXM^TM, producer: ULVAC. PHI. Inc., Chigasaki, Japan) [19]. This procedure uses X-ray to analyze the characterization. The X-ray is emitted from the probe whose diameter is 100 mm to the sample by setting the acceleration voltage of 15 kV. In this study, XPS analysis is conducted to identify the type of P loaded on TiO₂ film.

The Experimental Procedure of CO₂ Reduction

Figure 2 illustrates the experimental apparatus. The reactor consists of a stainless tube with the scale of 100 mm (H.) × 50 mm (I.D.), P₄O₁₀/TiO₂ film which is coated on the netlike glass disc with the scale of 50 mm (H.) × 50 mm (D.), a quartz glass disc having the scale of 84 mm (D.) × 10 mm (t.), a sharp cut filter removing the wavelength of light which is below 400 nm (SCF-49.5C-42L, producer: SIGMA KOKI CO LTD., Tokyo, Japan) or 800 nm (ITF-50C-85IR, producer: SIGMA KOKI CO LTD., Tokyo, Japan), a 150 W Xe lamp (L2175, producer: Hamamatsu Photonics K. K.), mass flow controller and CO₂ gas cylinder (purity: 99.995 vol%) and NH₃ gas cylinder (purity: 99.99 vol%). The reactor size for charging the gases was 1.25 × 10^-4 m³. The three black body materials were located under the P₄O₁₀/TiO₂ film coated on a netlike glass disc. The light of the Xe lamp positioned on the stainless tube was illuminated toward P₄O₁₀/TiO₂ film. As the netlike glass disc had the aperture area of the net, which is 4 mm² for each, the light can reach the black body material. The light of Xe lamp located on the stainless tube was illuminated toward P₄O₁₀/TiO₂ film passing the sharp cut filter and the quartz glass disc positioned on the top of the stainless tube. The wavelength of light illuminated from Xe lamp was distributed from 185 nm to 2000 nm. The sharp cut filter can remove the UV and VIS from the Xe lamp, providing the wavelength of light illuminating P₄O₁₀/TiO₂ film ranged from 401 nm to 2000 nm or 801 nm to 2000 nm [33]. Figure 3 exhibits the light transmittance data of sharp cut filter cutting the wavelength below 400 nm to clarify the light illumination condition as an example [20]. The mean light intensity of light illuminated from Xe lamp from 185 nm to 2000 nm was 70.4 mW/cm², that from 401 nm to 2000 nm was 60.7 mW/cm², and that from 801 nm to 2000 nm was 46.6 mW/cm².

After filling the CO₂ gas with a purity of 99.995 vol% and NH₃ with a purity of 99.99 vol%, they were controlled by a mass controller and introduced into the reactor pre-vacuumed by a vacuum pump for 15 min. The valves installed at the inlet and the outlet of the reactor were closed during CO₂ reduction with NH₃. After that, this study confirmed the pressure of 0.1 MPa and the gas temperature at 298 K in the reactor. Due to the heat of IR light components illuminated by the Xe lamp, the temperature of the gas in the reactor rose. The temperature of the experimental room was controlled and set at 293 K by an air conditioner. The molar ratio of CO₂/NH₃ was changed by 1:0.5, 1:1, 1:2, 1:4, 3:2, and 3:8. The reacted gas filled in the reactor was extracted by gas syringe via a gas sampling tap and it was analyzed by an FID gas chromatograph (GC353B, produced by GL Science) and a methanizer (MT221, produced by GL Science). The FID gas chromatograph and methanizer have a minimum resolution of 1 ppmV. The temperature of the gas in the reactor was measured by a thermocouple installed in the tap, which was located 1 mm above the P₄O₁₀/TiO₂ film coated on a netlike glass disc. The CO₂ reduction experiment was conducted up to 8 hours. Gas sampling and temperature measurements were carried out from the start of the experiment until 8 hours by 2 hours.

Figure 2: Schematic diagram of CO₂ reduction experimental apparatus. The reactor consists of stainless pipe, P₄O₁₀/TiO₂ film photocatalyst positioned on Teflon cylinder, black body materials, a quartz glass disc, a 150 W Xe lamp, mass flow controller, CO₂ gas cylinder, and NH₃ gas cylinder.

Figure 3: Light transmittance data of sharp cut filter.

The Characterization of P₄O₁₀/TiO₂ Film

Figure 4 shows SEM (Scanning Electron Microscope) and EPMA (Electron Probe Microanalyzer) images of P₄O₁₀/TiO₂ film which is coated on netlike glass disc. We obtained the black and white SEM image whose magnification was 1500 times. It was also used for EPMA analysis. EPMA images show the concentration distribution of each chemical element in observation area exhibited by the diverse colors. If the amount of chemical element is small, dark colors such as black and blue are adopted. On the other hand, light colors such as white, pink, and red are adopted when the amount of chemical element is large.

Figure 4: SEM and EPMA images of P₄O₁₀/TiO₂ film coated on netlike glass disc.

It is seen from Figure 4 that it can be seen TiO₂ film with a teeth-like shape coated on the netlike glass fiber. It is thought that the temperature distribution of TiO₂ solution which was adhered on the netlike glass disc was not even during firing process since the thermal conductivity of Ti and SiO₂ at 600 K are 19.4 W/(m·K) and 1.82 W/(m·K), respectively [31]. Since the thermal expansion and shrinkage around the net like glass fiber might be occurred, a thermal crack formed within TiO₂ film [29]. Consequently, TiO₂ film on the netlike glass fiber was teeth-like shape. In addition, it can be seen that P is detected in the area where Ti is detected. Since this study has prepared P₄O₁₀/TiO₂ film by sol-gel and dip-coating processes, the fine particles of P₄O₁₀ are mixed in TiO₂ sol solution. Therefore, it is thought P₄O₁₀ is adhered with TiO₂ on netlike glass disc.

Figure 5 shows XPS (X-ray Photoelectron Spectroscopy) data of P₄O₁₀/TiO₂ film coated on netlike glass disc [19]. We can see from Figure 5 that the 2P_p3/2 spectrum shows the peaks with binding energies (BE) of 134 eV which indicates P₄O₁₀ [30]. Therefore, we can confirm that the P ins the film exists in the form of P₄O₁₀.

Figure 5: XPS analysis result on P₄O₁₀.

The CO₂ Reduction Performance of P₄O₁₀/TiO₂ with and without Black Body Material under the Illumination Condition with UV + VIS + IR

Figures 6 and 7 show comparison of concentration of formed CO among different molar ratios with and without black body material, respectively. The results under the illumination condition with UV + VIS + IR are shown in these figures. The other fuels were not detected. Regarding a blank test, this study conducted the same experiment under no Xe lamp illumination condition as a reference case before the experiment. We detected no fuel during the blank test as we hoped. As to the reproducibility of experiments, this study displays the data averaging three times experiments. After three experiments, the changes of surface structure can not be observed by the naked eye. Moreover, we have tried to touch the surface of photocatalyst, resulting that the degradation of surface has not been confirmed. The weight of P₄O₁₀/TiO₂ film is 0.04 g, which is the average value of 12 samples.

It is seen from Figure 6 that the highest concentration of formed CO without black body material is obtained in the case of CO₂:NH₃ = 3:2, whose concentration is 410 ppmV. This concentration is converted into into the molar quantity of CO per unit weight of photocatalyst, resulting that 49.6 µmol/g. This molar ratio accords with the theoretical molar ratio to produce CO according to the reaction scheme shown by Eqs. (1) – (8). Though P₄O₁₀ is loaded, the reaction scheme to reduce CO₂ with NH₃ for P₄O₁₀/TiO₂ follows the reaction scheme to reduce CO₂ with NH₃ for TiO₂. In addition, the formation rate of CO decreases from the start of illumination of Xe lamp to 8 hours, attaining approximately 0 µmol/h, this study thinks the concentration of formed CO would be saturated.

It is seen from Figure 7 that the highest concentration of formed CO with black body materials is obtained in the case of CO₂:NH₃ = 3:2, whose concentration is 461 ppmV. This concentration is converted into the molar quantity of CO per unit weight of photocatalyst, resulting that 55.1 µmol/g. This molar ratio accords with the theoretical molar ratio to produce CO according to the reaction scheme shown by Eqs. (1) – (8). Though P₄O₁₀ is loaded and the CO₂ reduction experiment has been conducted with black body materials, the reaction scheme to reduce CO₂ with NH₃ for P₄O₁₀/TiO₂ follows the reaction scheme to reduce CO₂ with NH₃ for TiO₂ without black body material. In addition, the formation rate of CO decreases from the start of illumination of Xe lamp to 8 hours, attaining approximately 0 µmol/h, this study thinks the concentration of formed CO would be saturated. Moreover, comparing the highest concentration of formed CO in the case of CO₂:NH₃ = 3:2 with black body materials to that without black body material, the concentration of formed CO increases by 51 ppmV. Therefore, the effect of the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst created by black body materials on the CO₂ reduction performance is obtained. The mechanism of this phenomenon is discussed in the following section.

Figure 6: Comparison of concentration of formed CO among different molar ratios under the illumination condition with UV + VIS + IR without black body material.

Figure 7: Comparison of concentration of formed CO among different molar ratios under the illumination condition with UV + VIS + IR with black body materials.

The CO₂ Reduction Performance of P₄O₁₀/TiO₂ with and without Black Body Material under the Illumination Condition with VIS + IR

Figures 8 and 9 show comparison of concentration of formed CO among different molar ratios with and without black body material, respectively. The results under the illumination condition with VIS + IR are shown in these figures. The other fuels were not detected. Regarding a blank test, this study conducted the same experiment under no Xe lamp illumination condition as a reference case before the experiment. We detected no fuel during the blank test as we hoped. As to the reproducibility of experiments, this study displays the data averaging three times experiments. After three experiments, the changes of surface structure can not be observed by the naked eye. Moreover, we have tried to touch the surface of photocatalyst, resulting that the degradation of surface has not been confirmed. The weight of P₄O₁₀/TiO₂ film is 0.04 g, which is the average value of 12 samples.

It is seen from Figure 8 that the highest concentration of formed CO without black body material is obtained in the case of CO₂:NH₃ = 3:2, whose concentration is 205 ppmV. This concentration is converted into the molar quantity of CO per unit weight of photocatalyst, resulting that it is 25.4 mmol/g. It is found from this result that P₄O₁₀ supports to absorb VIS + IR. In addition, this molar ratio accords with the theoretical molar ratio to produce CO according to the reaction scheme shown by Eqs. (1) – (8). Though P₄O₁₀ is loaded, the reaction scheme to reduce CO₂ with NH₃ for P₄O₁₀/TiO₂ under the illumination condition with VIS + IR follows the reaction scheme to reduce CO₂ with NH₃ for TiO₂ under the illumination condition with UV [27, 28]. Moreover, the formation rate of CO decreases from the start of illumination of Xe lamp to 8 hours, attaining approximately 0 mmol/h, this study thinks the concentration of formed CO would be saturated.

It is seen from Figure 9 that the highest concentration of formed CO with black body materials is obtained in the case of CO₂:NH₃ = 3:2, whose concentration is 224 ppmV. This concentration is converted into the molar quantity of CO per unit weight of photocatalyst, resulting that 27.7 mmol/g. It is found from this result that P₄O₁₀ supports to absorb VIS + IR. This molar ratio accords with the theoretical molar ratio to produce CO according to the reaction scheme shown by Eqs. (1) – (8). Though P₄O₁₀ is loaded and the CO₂ reduction experiment has been conducted with black body materials, the reaction scheme to reduce CO₂ with NH₃ for P₄O₁₀/TiO₂ under the illumination condition with VIS + IR follows the reaction scheme to reduce CO₂ with NH₃ for TiO₂ without black body material [27,28]. In addition, the formation rate of CO decreases from the start of illumination of Xe lamp to 8 hours, attaining approximately 0 mmol/h, this study thinks the concentration of formed CO would be saturated. Moreover, comparing the highest concentration of formed CO in the case of CO₂:NH₃ = 3:2 with black body materials to that without black body material, the concentration of formed CO increases by 19 ppmV. Therefore, the effect of the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst created by black body materials on the CO₂ reduction performance is obtained. The mechanism of this phenomenon is discussed in the following section.

Figure 8: Comparison of concentration of formed CO among different molar ratios under the illumination condition with VIS + IR without black body material.

Figure 9: Comparison of concentration of formed CO among different molar ratios under the illumination condition with VIS + IR with black body materials.

The CO₂ Reduction Performance of P₄O₁₀/TiO₂ with and without Black Body Material under the Illumination Condition with IR

Figures 10 and 11 show comparison of concentration of formed CO among different molar ratios with and without black body material, respectively. The results under the illumination condition with IR only are shown in these figures. The other fuels were not detected. Regarding a blank test, this study conducted the same experiment under no Xe lamp illumination condition as a reference case before the experiment. We detected no fuel during the blank test as we hoped. As to the reproducibility of experiments, this study displays the data averaging three times experiments. After three experiments, the changes of surface structure can not be observed by the naked eye. Moreover, we have tried to touch the surface of photocatalyst, resulting that the degradation of surface has not been confirmed. The weight of P₄O₁₀/TiO₂ film is 0.04 g, which is the average value of 12 samples.

Figure 10: Comparison of concentration of formed CO among different molar ratios under the illumination condition with IR without black body material.

Figure 11: Comparison of concentration of formed CO among different molar ratios under the illumination condition with IR with black body materials.

It is seen from Figure 10 that the highest concentration of formed CO without black body material is obtained in the case of CO₂:NH₃ = 3:2, whose concentration is 53 ppmV. This concentration is converted into the molar quantity of CO per unit weight of photocatalyst, resulting that it is 6.65 mmol/g. It is found from this result that P₄O₁₀ supports to absorb IR. In addition, this molar ratio accords with the theoretical molar ratio to produce CO according to the reaction scheme shown by Eqs. (1) - (8). Though P₄O₁₀ is loaded, the reaction scheme to reduce CO₂ with NH₃ for P₄O₁₀/TiO₂ under the illumination condition with IR only follows the reaction scheme to reduce CO₂ with NH₃ for TiO₂ under the illumination condition with UV [27, 28]. Moreover, the formation rate of CO decreases from the start of illumination of Xe lamp to 8 hours, attaining approximately 0 mmol/h, this study thinks the concentration of formed CO would be saturated.

It is seen from Figure 11 that the highest concentration of formed CO with black body materials is obtained in the case of CO₂:NH₃ = 3:2, whose concentration of CO is 66 ppmV. This concentration is converted into the molar quantity of CO per unit weight of photocatalyst, resulting that it is 8.20 mmol/g. It is found from this result that P₄O₁₀ supports to absorb IR. This molar ratio accords with the theoretical molar ratio to produce CO according to the reaction scheme shown by Eqs. (1) – (8). Though P₄O₁₀ is loaded and the CO₂ reduction experiment has been conducted with black body materials, the reaction scheme to reduce CO₂ with NH₃ for P₄O₁₀/TiO₂ under the illumination condition with IR follows the reaction scheme to reduce CO₂ with NH₃ for TiO₂ without black body material [27,28]. In addition, the formation rate of CO decreases from the start of illumination of Xe lamp to 8 hours, attaining approximately 0 mmol/h, This study thinks the concentration of formed CO would be saturated. Moreover, comparing the highest concentration of formed CO in the case of CO₂:NH₃ = 3:2 with black body materials to that without black body material, the concentration of formed CO increases by 13 ppmV. Therefore, the effect of the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst created by black body materials on the CO₂ reduction performance is obtained. The mechanism of this phenomenon is discussed in the following section.

Relationship between Temperature Rise of Gas in Reactor and Concentration of Formed CO

Figures 12, 13 and 14 show comparison of the relationship between the maximum concentration of formed CO and the temperature rise in the reactor with and without black body material under the illumination condition with UV + VIS + IR, VIS + IR and IR only, respectively. In these figures, the temperature rise from the temperature at the start of CO₂ reduction experiment to the temperature when the concentration of formed CO attains the maximum value is shown.

It is seen from Figure 12 that the temperature in the reactor under the illumination condition with UV + VIS + IR rises more due to black body materials by 3.8 ? – 6.2 ? compared to that without black body material. The maximum temperature rise is 6.2 ?. In addition, the maximum concentration of formed CO increases due to black body materials by 21 ppmV – 71 ppmV compared to that without black body material. It is thought that the mass transfer surrounding the P₄O₁₀/TiO₂ photocatalyst is promoted by the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst created by black body materials under the illumination condition with UV + VIS + IR [24].

Figure 12: Relationship between the maximum concentration of formed CO and the temperature rise in the reactor under the illumination condition with UV + VIS + IR (a): without black body material, b): with black body materials).

It is seen from Figure 13 that the temperature in the reactor under the illumination condition with VIS + IR rises more due to black body materials by 1.0 ? – 2.9 ? compared to that without black body material. In addition, the maximum concentration of formed CO increases due to black body materials by 19 ppmV – 40 ppmV compared to that without black body material. The maximum value is 40 ppmV. It is thought that the mass transfer surrounding the P₄O₁₀/TiO₂ photocatalyst is promoted by the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst created by black body materials under the illumination condition with VIS + IR.

Figure 13: Relationship between the maximum concentration of formed CO and the temperature rise in the reactor under the illumination condition with VIS + IR (a): without black body material, b): with black body materials).

It is seen from Figure 14 that the temperature in the reactor under the illumination condition with IR only rises more due to black body materials by 2.0 ? – 3.1 ? compared to that without black body material. In addition, the maximum concentration of formed CO increases due to black body materials by 4 ppmV – 13 ppmV compared to that without black body material. It is thought that the mass transfer surrounding the P₄O₁₀/TiO₂ photocatalyst is promoted by the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst created by black body materials under the illumination condition with IR only.

Figure 14: Relationship between the maximum concentration of formed CO and the temperature rise in the reactor under the illumination condition with IR (a): without black body material, b): with black body materials).

From the investigation under the illumination condition with UV + VIS + IR, VIS + IR and IR only, it is confirmed that the mass transfer promotion by black body material is effective for the improvement in CO₂ reduction performance irrespective of the illumination condition of Xe lamp. This study investigates the heat balance between black body materials and the mixed gas of CO₂ and NH₃ in the reactor. The heat capacity of Cu solid disc used for black body material is 0.189 J/K per one disc is 30 ? [31], resulting that the heat capacity of three Cu discs used in this study is 0.567 J/K at 30 ?. On the other hand, the volume of gas in the reactor is 1.25 × 10⁵ mm³. When the reactor is charged with the mixed gas of CO₂ and NH₃, e.g. whose molar ratio is 3:2, the density of mixed gas is 1.346 kg/m³ at 30 ? [31]. The specific heat of CO₂ and NH₃ at 30 ? is 0.8546 kJ/(kg·K) and 2.171 kJ/(kg·K), respectively [31]. Considering them, the heat capacity of mixed gas whose molar ratio of CO₂:NH₃ = 3:2 is 0.255 J/K at 30 ?. Comparing the heat capacity of the mixed gas of CO₂ and NH₃ with the heat capacity of three Cu solid discs, the heat capacity of three Cu solid discs is over that of mixed gas of CO₂ and NH₃. Therefore, it is thought that the black body materials used in this study can absorb the heat from IR light to provide the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst [24]. Figure 15 illustrates the image that the black body materials absorb IR light and the heat is transferred to the photocatalyst by thermal conduction [24]. Then, the temperature of the photocatalyst rises, resulting that the natural thermosiphon movement of gases around the P₄O₁₀/TiO₂ photocatalyst occurs due to the radiation. Comparing the heat capacity of mixed gas of CO₂/NH₃, the largest heat capacity which is 0.252 J/K is estimated in the case of CO₂:NH₃ = 1:1, while the smallest capacity which is 0.238 J/K is estimated in the case of CO₂:NH₃ = 1:4. They does not match the molar ratio of CO₂:NH₃ = 3:2 which provides the highest concentration of formed CO in this study irrespective of light illumination condition of Xe lamp. It is thought that the optimum molar ratio of CO₂:NH₃ is decided by not the heat capacity of mixed gas of CO₂/NH₃ but the reaction scheme to reduce CO₂ with NH₃.

According to the investigation of this study, the highest performance of CO₂ reduction under the illumination condition with IR is obtained in case of CO₂:NH₃ = 3:2 with black body materials, whose molar quantity of CO per unit weight of photocatalyst, it is 8.20 mmol/g. As to the photocatalyst to extend the absorption of light wavelength up to IR, W₁₈O₄₉/g-C₃N₄ composite has performed the production of CO of 45 mmol/g and CH₄ of 28 mmol/g under the illumination condition whose wavelength is ranged from 200 nm to 2400 nm [15]. WS₂/Bi₂S₃ nanotube has performed the absorption of VIS and near IR light (wavelength: 420 nm – 1100 nm), which has produced CH₃OH of 28 mmol/g and C₂H₅OH of 25 mmol/g [16]. CuInZnS decorated g-C₃N₄ has exhibited the absorption performance of light whose wavelength is ranged from 200 nm to 1000 nm, producing CO of 38 mmol/g [17]. Hierarchical ZnIn₂S₄ nanorods has prepared by solvothermal method, which has produced CO of 54 mmol/g and CH₄ of 9 mmol/g [18]. Comparing the previous reports with this study, the CO₂ reduction performance exhibited by this study is lower. However, the investigation on the combination effect of two trials, i.e. to extend the absorption range to IR and to enhance the gas movement around the photocatalyst by the natural thermosiphon movement of gasses around photocatalyst using black body material on the CO₂ reduction performance of photocatalyst is a novel approach to improve the CO₂ reduction performance of photocatalyst.

Considering the investigation by this study, it is necessary to promote the CO₂ reduction performance of P₄O₁₀/TiO₂ photocatalyst with the natural thermosiphon movement of gasses around P₄O₁₀/TiO₂ photocatalyst created by black body material more. The reactor design to consider the heat balance between the base material for black body material and mixed gas of CO₂ and NH₃ is thought to be one approach. If the black body material emits the larger radiation, it is expected that the temperature of gasses around P₄O₁₀/TiO₂ photocatalyst increases, resulting in the promotion of mass transfer. To select the optimum base material is the future work.

Figure 15: Image of mass transfer promotion by natural thermosiphon movement of gasses around TiO₂ photocatalyst created by black body materials.

In this study, the prepared P₄O₁₀/TiO₂ photocatalyst works under the illumination condition with IR. Under this condition, the wavelength of light illumination from Xe lamp is ranged from 801 nm to 200 nm. The band gap energy can be calculated by the following equation:

E = hc/l      (9)

where E is a band gap energy [J], h is Planck’s constant (= 6.62607×10^-34) [J·s], c is a light velocity (= 2.998×10⁸) [m/s] and l is a wave length of light [m]. 1 eV = 1.60219×10^-19 J. Here, l = 800 nm is substituted into the above equation, resulting that E = 2.483×10^-19 J (= 1.55 eV) is obtained.

This study has investigated the impact of the natural thermosiphon movement of gasses around P₄O₁₀/TiO₂ photocatalyst created by black body material on CO₂ reduction performance of P₄O₁₀/TiO₂. This study also has investigated the impact of molar ratio of CO₂/NH₃ on the CO₂ reduction characteristics of P₄O₁₀. As a result, the following conclusions are drawn:

• It is observed that TiO₂ film on the netlike glass fiber is teeth-like shape and P is detected in the area where Ti is detected.
• The highest concentration of formed CO with and without black body material is obtained in the case of CO₂:NH₃ = 3:2 irrespective of illumination condition of Xe lamp. The optimum molar ratio of CO₂/NH₃ obtained in this study matches the theoretical molar ratio to produce CO according to the reaction scheme to reduce CO₂ with NH₃.
• It is found that the CO₂ reduction performance could be improved by enhanced gas movement created by a black body material. Under the illumination condition of UV + VIS + IR, the highest concentration of formed CO is obtained with CO₂:NH₃= 3:2 and black body materials. The concentration of formed CO is found to increase due to black body materials by 21ppmV – 71ppmV compared to that without black body material. The temperature in the reactor rises more due to black body materials by 3.8 ? – 6.2 ? compared to that without black body material.
• Under the illumination condition of VIS + IR, the highest concentration of formed CO is obtained with CO₂:NH₃ = 3:2 and black body materials. The concentration of formed CO is found to increase due to black body materials by 19 ppmV – 40 ppmV compared to that without black body material. The temperature in the reactor rises more due to black body materials by 1.0 ? – 2.9 ? compared to that without black body material.
• Under the illumination condition with IR only, the highest concentration of formed CO is obtained with CO₂:NH₃ = 3:2 and black body materials. The concentration of formed CO is found to increase due to black body material by 4 ppmV – 13 ppmV compared to that without black body material. The temperature in the reactor rises more due to black body materials by 2.0 ? – 3.1 ? compared to that without black body material.

Acknowledgments

The authors acknowledge JSPS KAKENHI Grant Number JP21K04769.

Funding

This research was funded by Mie University and JSPS KAKENHI Grant Number JP21K04769.

1. Jesic D, Jurkovic LD, Pohar A, Suhadolnik L, Likozar B. Engineering photocatalytic and photoelectrocatalytic CO₂ reduction reactions: Mechanisms, intrinsic kinetics, mass transfer resistances, reactors and multi-scale modeling simulations. Chem Eng J. 2021; 407.
2. Razzaq A, Ali S, Asif M, In SI. Layered Double Hydroxide (LDH) based photocatalysts: An outstanding strategy for efficient photocatalytic CO₂ Catalysts. 2020; 10.
3. Matavos-Aramyan S, Soukhakian S, Jazebizadeh MH, Moussavi M, Hojjati MR. On Engineering strategies for photoselective CO₂ reduction – A through review. Appl Mat Today. 2020; 18.
4. Remiro-Buenamanana S, Garcia H. Photoassisted CO₂ conversion into fuels. Chem Cat Chem Minirev. 2019; 11: 342-356.
5. Zhu S, Chen X, Li Z, Ye X, Liu Y, Chen Y, et al. Cooperation between inside and outside of TiO₂: Lattice Cu⁺ accelerates carrier migration to the surface of metal copper for photocatalytic CO₂ Applied Catalysis B: Environmental. 2020; 264.
6. Jeong S, Kim GM, Kang GS, Kim C, Lee H, Kim WJ, et al. Selectivity modulated by surface ligands on Cu₂O/TiO₂ catalysts for Gas-Phase photocatalytic reduction of carbon dioxide. J Physical Chem C. 2019; 123: 29184-29191.
7. Jiang Z, Sun W, Miao W, Yuan Z, Yang G, Kong F, et al. Living atomically dispersed Cu ultrathin TiO₂ nanosheet CO₂ reduction photocatalyst. Advanced Sciences. 2019; 6.
8. Aguirre ME, Zhou R, Eugene AJ, Guzman MI, Grela MA. Cu₂O/TiO₂ heterostructures for CO₂ reduction through a direct Z-Scheme: Protecting Cu₂O from photocorrosion. Applied Catalysis B: Environmental. 2017; 217: 485-493.
9. Su KY, Chen CY, Wu RJ. Preparation of Pd/TiO₂ nanowires for the photoreduction of CO₂ into renewable hydrocarbon fuels. J Taiwan Institute Chem Eng. 2019; 96: 409-418.
10. Camarillo R, Toston S, Martinez F, Jimenez C, Rincon J. Enhancing the photocatalytic reduction of CO₂ through engineering of catalysts with high pressure technology: Pd/TiO₂ J Supercritical Fluids. 2017; 123: 18-27.
11. Yu Y, Lan Z, Guo L, Wang E, Yao J, Cao Y. Synergetic effects of Zn and Pd species in TiO₂ towards efficient Photo-Reduction of CO₂ into CH₄. New J Chem. 2018; 42: 483-488.
12. Zhao Y, Wei Y, Wu X, Zheng H, Zhao Z, Liu J, et al. Graphene-wrapped Pt/TiO₂ photocatalysts with enhanced photogenerated charges separation and reactant adsorption for high selective photoreduction of CO₂ to CH₄. Applied Catalysis B: Environmental. 2018; 226: 360-372.
13. Toston S, Camarillo R, Martinez F, Jimenez C, Rincon J. Supercritical synthesis of Platinum-Modified titanium dioxide for solar fuel production from carbon dioxide. Chinese J Catalysis. 2017; 38: 636-650.
14. Wei Y, Wu X, Zhao Y, Wang L, Zhao Z, Huang X, Liu J, Li J. Efficient photocatalysts of TiO₂ Nanocrystals-Supported PtRu alloy nanoparticles for CO₂ reduction with H₂O: Synergistic effect of Pt-Ru. Applied Catalysis B: Environmental. 2018; 236: 445-457.
15. Hong LF, Guo RT, Yuan Y, Ji XY, Lin ZD, Gu JW, et al. Urchinlike W₁₈O₄₉/g-C₃N₄ Z-Scheme heterojunction for highly efficient photocatalytic reduction of CO₂ under full spectrum light. Energy Fuels. 2021; 35: 11468-11478.
16. Dai W, Yu J, Luo S, Hu X, Yang L, Zhang S, et al. WS₂ Quantum dots seeding in Bi₂S₃ Nanotubes: A novel Vis-NIR light sensitive photocatalyst with Low-Resistance junction interface for CO₂ Chem Eng J. 2020; 389.
17. Gan J, Wang H, Hu H, Su M, Chen F, Xu H. Efficient synthesis of tunable Band-Gap CuInZnS decorated g-C₃N₄ hybrids for enhanced CO₂ Photocatalytic reduction and Near-Infrared-Triggered photordegradation performance. Appl Surface Sci. 202; 564.
18. Yu M, Lv X, Idris AM, Li S, Lin J, Lin H, et al. Upconversion nanoparticles coupled with hierarchical ZnIn₂S₄ nanorods as a Near-Infrared Responsive Photocatalyst for Photocatalytic CO₂ J Colloid and Interface Sci. 2022; 612: 782-791.
19. Nishimura A, Mae H, Kato T, Hu E. Utilization from ultraviolet to infrared light for CO₂ reduction with P₄O₁₀/TiO₂ Phy Astro Int J. 2022; 6: 145-154.
20. Nishimura A, Mae H, Hannyu R, Hu E. Impact of Loading Amount of P₄O₁₀ on CO₂ reduction performance of P₄O₁₀/TiO₂ with H₂O extending absorption range from ultraviolet to infrared light. Phy Astro Int J. 2022; 6: 186-194.
21. Baniamer M, Aroujalian A, Sharifnia S. Photocatalytic membrane reactor for simultaneous separation and photoreduction of CO₂ to methanol. Int J Energy Research. 2020; 45: 2353-2366.
22. Molinari R, Lavorato C, Argurio P. The evolution of photocatalytic membrane reactors over the last 20 years: A state of the art perspective. Catalysts. 2021; 11.
23. Nishimura A, Komatsu N, Mitsui G, Hirota M, Hu E. CO₂ reforming into fuel using TiO₂ photocatalyst and gas separation membrane. Catalysis Today. 2009; 148: 341-349.
24. Nishimura A, Kato T, Mae H, Hu E. Impact of black body material enhanced gas movement on CO₂ photocatalytic reduction performance. Catalysts. 2022; 12.
25. Nahar S, Zain MFM, Kadhum AAH, Abu HH. Advances in photocatalytic CO₂ reduction with water: A review. Materials. 2017; 10: 629.
26. Tahir M, Amin NS. Advances in visible light responsive titanium oxide based photocatalysts for CO₂ conversion to hydrocarbon fuels. Energy Convers Manag. 2013; 76: 194-214.
27. Lo CC, Hung CH, Yuan CS, Wu JF. Photoreduction of carbon dioxide with H₂ and H₂O over TiO₂ and ZrO₂ in a circulated photocatalytic reactor. Sol Energy Mater Sci. 2007; 91: 1765-1774.
28. Nemoto J, Goken N, Ueno K. Photodecomposition of ammonia to dinitrogen and dihydrogen on platinized TiO₂ nanoparticles in an aqueous solution. J Photochem Photobiol A Chem. 2007; 185: 295-300.
29. Nishimura A, Sakakibara Y, Koshio A, Hu E. The impact of amount of Cu on CO₂ reduction performance of Cu/TiO₂ with NH₃ and H₂ Catalysts. 2021; 11.
30. Japan Science and Technology Agency, J-STORE. P2009-184861A. 2022.
31. Japan Society of Mechanical Engineering. Heat Transfer Hand Book, 1st ed.; Maruzen: Tokyo. Japan. 1993.
32. Tahir M, Amin NS. Photocatalytic reduction of carbon dioxide with water vapors over montmorillonite modified TiO₂ Appl Catal B Environ. 2013; 142-143: 512-522.
33. Nishimura A, Mitsui G, Nakamura K, Hirota M, Hu E. CO₂ reforming characteristics under visible light response of Cr- or Ag-doped TiO₂ prepared by Sol-gel and Dip-coating process. Int J Photoenergy. 2012.