Effective Carbon Dioxide Photoreduction over Metals ( Fe-, Co-, Ni-, and Cu-) Incorporated TiO 2 / Basalt Fiber Films

Mineralogical basalt fibers as a complementary adsorbent were introduced to improve the adsorption of CO 2 over the surfaces of photocatalysts. TiO 2 photocatalysts (M-TiO 2 ) incorporated with 5.0mol.% 3d-transition metals (Fe, Co, Ni, and Cu) were prepared using a solvothermal method and mixed with basalt fibers for applications to CO 2 photoreduction. The resulting 5.0mol.%M-TiO 2 powders were characterized by X-ray diffraction, scanning electronmicroscopy, ultraviolet-visible spectroscopy, photoluminescence, Brunauer, Emmett, and Teller surface area, and CO 2 -temperature-programmed desorption. A paste composed of two materials was coated and fixed on a Pyrex plate by a thermal treatment. The 5.0mol.% M-TiO 2 /basalt fiber films increased the adsorption of CO 2 significantly, indicating superior photocatalytic behavior compared to pure TiO 2 and basalt fiber films, and produced 158∼360 μmol gcat −1 L CH 4 gases after an 8 h reaction. In particular, the best performance was observed over the 5.0mol.% Co-TiO 2 /basalt fiber film. These results were attributed to the effective CO 2 gas adsorption and inhibition of photogenerated electron-hole pair recombination.


Introduction
To mitigate the greenhouse effect, there has been increasing interest in converting the greenhouse gas, carbon dioxide (CO 2 ), to useful molecules, such as carbon monoxide (CO) [1,2], methane (CH 4 ) [3,4], formic acid (HCOOH) [5], formaldehyde (HCHO) [6], or methanol (CH 3 OH) [7,8], via chemical routes.On the other hand, because CO 2 is a chemically stable compound owing to its carbon-oxygen double bonds, its conversion to some hydrocarbons requires substantial energy input for bond cleavage [9].Solar energy provides a readily available and continuous energy supply that would be suitable to drive this conversion process.In CO 2 photocatalysis to synthesize some fuels, semiconductor materials with an appropriate band-gap (the energy region extending from the bottom of the empty conduction band to the top of the occupied valence band) are required as photocatalysts.When a semiconductor photocatalyst absorbs light in a typical photocatalytic process, an electron is excited from the fully occupied valence band of the semiconductor to a higher energy empty conduction band, forming an electron-hole pair [10,11].These charge carriers can also recombine on the surface or in the bulk before reacting with adsorbed species, dissipating the energy as heat or light.The photoreactions occur continually with electron accepting or donating species adsorbed on the surface of the semiconductor photocatalyst.Therefore, electron-hole recombination must be minimized for photocatalytically induced redox reactions to take place [12][13][14].
The TiO 2 semiconductor has been assessed for CO 2 photoreduction because of its chemical stability and natural abundance.Although TiO 2 has several unique features, its use has been limited by its large band-gap (3.2 eV), meaning it can only be activated by ultraviolet light, which comprises 2-5% of sunlight [15,16], and the relatively fast recombination between the electron and holes [17,18].Therefore, the photocatalyst should have a lower band-gap and an increased lifetime of the photogenerated electrons and holes through effective charge carrier separation and the suppression of electron-hole recombination.The photocatalytic activity for visible light can be increased by coupling semiconductors of different energy levels or doping with metals or nonmetals 2 International Journal of Photoenergy to suppress the recombination rate and increase the quantum yield [19,20]. Considerable efforts have been made to extend the adsorption of reactive species and the utilization of the incident light properties of catalysts to improve their photocatalytic efficiency.The approaches of combining TiO 2 with a support, such as porous materials, polycrystalline fibers, or tube materials (e.g., CNTs or polymers, magnetic materials, and minerals), to achieve charge separation and the adsorption of reactive species, have decreased the recombination rate and maintained excellent catalytic activity [21,22].
Therefore, this study examined whether a mineral, basalt fiber [23] as a complementary material, would be sufficient to facilitate both electron-hole charge separation and CO 2 adsorption.Basalt is one of the most common rock types in the world and is found widely in large igneous provinces.Basalt contains MgO, CaO, Fe 2 O 3 , TiO 2 , Al 2 O 3 , and SiO 2 and can be used as a substrate mineral to adsorb carbon dioxide.Another objective in this study was to examine the effects of 3d-transition metals as a cocatalyst in the TiO 2 photocatalytic system.In general, 3d-transition metals with excellent oxidation-reducing power are often used as thermal catalysts in some redox reactions.In particular, Ni, Co, Fe, and Cu have been applied widely as the main catalytic species from CO 2 or CO to methanol or hydrocarbons, so-called Fischer-Tropsch synthesis [24][25][26].Consequently, this study examined the synergistic effects of a basalt mineral absorbent (YJC com., Hampyeong, Jeonnam, Republic of Korea) and adding transition metals (Fe, Co, Ni, and Cu) to the TiO 2 anatase framework on its properties as a photosensitizer for the photoreduction of CO 2 to CH 4 .

Experimental
2.1.Preparation and Characterization of Nanosized 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 Powders and 5.0 mol.%M-TiO 2 /Basalt Fiber Films.Before the design of the 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 /basalt fiber films, four types of 5.0 mol.%M-TiO 2 powders were prepared using a solvothermal treatment [27].Titanium tetraisopropoxide (TTIP, 99.95%, Junsei Chem.Co., Japan) and metal nitrates (M(NO 3 ) 2 H 2 O, M = Fe, Ni, Co, and Cu, 99.9%, Junsei Chem.Co., Japan) were used as the precursors for titanium and other metals, respectively, and absolute ethanol was used as the solvent.After adding 0.095 M of TTIP to 250 mL of absolute ethanol in a 500 mL beaker, 0.4 M of distilled water was dropped carefully and slowly into the solution with stirring for 1 h to avoid aggregation by rapid hydrolysis, and 0.005 M (5.0 mol.%) of the metal precursors was then added to the solution.The final solution was stirred continuously for 2 h at room temperature and moved to an autoclave for thermal treatment at 200 ∘ C for 8 h under a nitrogen atmosphere at a pressure of approximately 10.0 atm.The resulting precipitate was washed with distilled water until the pH was 7.0 and dried at 50 ∘ C for 24 h.The pure TiO 2 nanoparticles were also obtained using the same method.The prepared TiO 2 and 5.0 mol.%M-TiO 2 were used as the main photocatalytic species for CO 2 reduction.
For the design of 5.0 mol.%M-TiO 2 /basalt fiber films for application to the CO 2 photoreduction reaction, the basalt fibers were ground to a small size, and 70-100 mesh-sized powders were filtered and selected for use as the support substrate.The paste was composed of 2.0 g of 5.0 mol.%M-TiO 2 powders and basalt fiber pieces with a mixture containing 5.0 g of -terpineol, 0.5 g of cellulose, and 20 mL of ethanol.The paste was then sonicated for 24 h at 1,200 Wcm −3 .Subsequently, a Pyrex plate was coated with the mixed paste with the M-TiO 2 powders and basalt fiber pieces using a squeeze printing technique.The size of the formulated 5.0 mol.%M-TiO 2 /basalt fiber films was 8.0 mm × 5.0 mm.The 5.0 mol.%M-TiO 2 /basalt fiber films were then heat treated at 450 ∘ C for 30 minutes to remove the additives.
The synthesized 5.0 mol.%M-TiO 2 powders were examined by X-ray diffraction (XRD, MPD, PANalytical) using nickel-filtered CuK radiation (30 kV, 30 mA).The reflectance UV-vis spectra of the 5.0 mol.%M-TiO 2 powders were obtained using a Cary 500 spectrometer with a reflectance sphere in the range 200∼800 nm.The Brunauer, Emmett, and Teller (BET) surface areas of the 5.0 mol.%M-TiO 2 powders were measured using a Belsorp II instrument.The recombination tendency of the photogenerated electronhole pairs (e − /h + ) in 5.0mol.%M-TiO 2 powder was estimated by photoluminescence (PL, Photoluminescence Spectrometer, PerkinElmer) spectroscopy using a He-Cd laser source at a wavelength of 325 nm.The adsorption of CO 2 on the 5.0 mol.%M-TiO 2 powders was measured from CO 2 -TPD (temperature-programmed desorption) experiments in the same manner used for thermogravimetric analysis (TGA, Sinco Co., Republic of Korea).
The morphology of the 5.0 mol.%M-TiO 2 /basalt fiber films was observed by scanning electron microscopy (SEM, JEOL 2000EX) and the atomic compositions of the films were determined by energy-dispersive X-ray spectroscopy (EDAX, EX-250, Horiba).

Photocatalytic
Activities over 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 /Basalt Fiber Films.A batch type photoreactor was designed in the laboratory, as shown in Figure 1.The reactor consisted of a rectangular quartz cell with a total volume of 60.0 mL.The photocatalytic activity was examined using the formulated 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 /basalt fiber films (8.0 × 5.0 mm 2 size) fixed uniformly to the bottom of the reaction chamber.A 1.0 mm thick quartz glass window cover was placed on the top of the reactor to enable the effective transfer of irradiation from two 6.0-W/cm 2 mercury lamps with a 365 nm wavelength.The reactor chamber and lamp were covered with aluminum foil to ensure that all the radiation that participated in the reaction had passed through the quartz window.The reactor, which had been checked for leakage at atmospheric pressure for several hours, was purged with helium carrier gas.The reaction temperature and pressure were maintained at 313 K and 1.0 atm, respectively.Before starting the experiment, the reactor was purged for one hour with a mixture of CO 2 and helium.The CO 2 : H 2 O ratio was fixed to 1 : 2. During the photocatalysis process, the product mixture was sampled off-line using a gas tight syringe (Agilent, 250 m) with the same volume and analyzed by gas chromatography (GC, iGC7200, Donam Co., Republic of Korea) equipped with a thermal conductivity detector (TCD) and a flame-ionized detector (FID).First, the gaseous products produced by the in situ system were flowed into the TCD detector that was connected to a Carboxen 1000 (Young Lin Instrumentals Co., Republic of Korea) column to analyze the light gases (H 2 , O 2 , CO, CO 2 , CH 3 OH, and CH 3 COOH).The extracted gases were then inserted into the FID detector to separate the C 1 (methane)-C 3 light hydrocarbons.The selectivity of the product was calculated using the following equation: C  (%) = C  moles in product/total moles of C produced × 100.
Figures 3(a) and 3(b) show the UV-visible absorption spectra and their Tauc's plots of the prepared 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 powders.An absorption band for the anatase structured TiO 2 (Figure 3(a)) was observed in the UV-region around 378 nm when extrapolated, which is similar to the absorption wavelength reported elsewhere [31].According to the addition of transition metal species, there are different absorption band shifts, which were shifted to higher wavelengths compared to the absorption band of TiO 2 .The 5.0 mol.%M-TiO 2 samples showed broad curves for the metal oxides in the visible region; the maximum absorption was observed at 490, 580, 700 (too broad), and 650 (too broad) nm for 5.0 mol.%Fe-, Co-, Ni-, and Cu-TiO 2 , respectively.These bands can convert to the following absorption terms using Tanabe-Sugano's energy absorption [32]: 5 T 2g → 5 E g for d 6 -FeO, 4 T 1g → 4 T 2g for d 7 -CoO, 3 A 2g → 3 T 2g for d 8 -NiO, and 2 E g → 2 T 2g for the d 9 electron configuration CuO.Generally, the band-gap in a semiconductor material is closely related to the wave range absorbed: the larger the absorption wavelength, the narrower the band-gap [33].Using Tauc's equation [34], the band-gaps for the absorption of TiO 2 and 5.0 mol.%Fe-, Co-, Ni-, and Cu-TiO 2 in the UV-region were estimated to be 3.18 and 2.80, 2.48, 2.98, and 3.10 eV, respectively, as shown in Figure 3(b).The inserted transition metals could alter the band-gap of TiO 2 significantly, leading to easy absorption and eventually efficient photocatalysis.suggests that the electrons in the valence band are transferred to the conduction band and are stabilized by photoemission.In general, the PL intensity increases with increasing number of electrons emitted, resulting from recombination between the excited electrons and holes, and a consequent decrease in photoactivity [35].Therefore, there is a strong relationship between the PL intensity and photoactivity.In particular, the PL intensity decreases significantly when a metal can capture excited electrons or exhibit conductivity, which is known as the relaxation process.The 5.0 mol.%M-TiO 2 samples exhibited a PL signal with a similar curve shape, demonstrating the presence of TiO 2 .Pure TiO 2 exhibited a strong PL signal in the range 400-550 nm, with a maximum excitation wavelength of 445 nm, whereas the 5.0 mol.%M-TiO 2 curve intensities were weakened dramatically.The decreasing tendency was observed in the following order: TiO 2 > 5.0 mol.%Fe-TiO 2 > Cu-TiO 2 > Ni-TiO 2 > Co-TiO 2 , which was possibly due to the new oxygen vacancies produced by metal-doping.The photogenerated electrons in the conduction band initially reached the vacant space and then recombined with the photogenerated holes in the valance band to produce fluorescence emission.The reduced PL intensities of 5.0 mol.%M-TiO 2 might be due to defects generated by the transition metal ions inserted into the TiO 2 structures, which accelerate electron transfer and hinder electron-hole pair recombination on the 5.0 mol.%M-TiO 2 surface.
Figure 5 shows the adsorption-desorption isotherm curves of N 2 at 77 K for the as-synthesized 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 particles.With the exception of the 5.0 mol.%Fe-TiO 2 and Cu-TiO 2 samples, all isotherms were close to type II according to the IUPAC classification [36], indicating the lack of pores in the particles.Otherwise, the isotherms of the 5.0 mol.%Fe-TiO 2 and Cu-TiO 2 samples were attributed to type IV, indicating bulk pores by aggregation between each nanoparticle.The hysteresis slopes were observed at intermediate and relative pressures greater than / 0 = 0.7 in all samples.
Table 1 lists the specific surface areas of the 5.0 mol.%M-TiO 2 samples.The specific surface areas were located in the range 68∼165 m 2 g −1 .In particular, the surface area was the largest at 165 m 2 g −1 in 5.0 mol.%Fe-TiO 2 .In general, the surface area increases with decreasing particle size [37].These results showed some degree of reliability and were well matched to their calculated crystallite sizes (Figure 2).From the results of the average bulk pore diameter for the samples, it is believed that the 5.0 mol.%Fe-TiO 2 and Cu-TiO 2 may contain bulk mesopores, approximately 15∼24 nm, due to aggregation between their nanoparticles.Otherwise, the pore volumes of the samples were varied from 0.17 to 0.64 cm 3 g −1 .The atomic compositions calculated the energy-dispersive Xray spectra that are also included in this table.The TiO 2 surface showed only two elements, Ti and O, whereas three elements were observed in the M-TiO 2 samples.In contrast to expectations, the amount of doped metals was much higher than the Ti amount in all samples, and the M : Ti ratios were approximately 1 : 10.This was calculated from EDAX analysis, and the values were different from the actual amount.The amount of metal doped in TiO 2 was reduced to the order of Fe > Co > Ni > Cu.
During the CO 2 photoreduction reaction, CO 2 is adsorbed onto the surface of the photocatalysts in the first step, and the photoreduction reaction progresses.Therefore, the photocatalytic performance depends on the adsorption capacities of CO 2 .Accordingly, the CO 2 adsorption abilities of the 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 samples need to be determined.The CO 2 desorption profiles were obtained over the range 50∼900 ∘ C, as shown in Figure 6.The curve intensity for CO 2 desorption was higher in 5.0 mol.%M-TiO 2 than in pure TiO 2 , which means considerably more CO 2 molecules had been adsorbed on the surfaces of the 5.0 mol.%M-TiO 2 samples.In addition, TiO 2 generally has hydrophilic properties [38], which suggests that, during the CO 2 photoreduction reaction, CO 2 and H 2 O will be adsorbed preferentially on the metal ions and TiO 2 , respectively.The adsorption abilities were observed in the following order: 5.0 mol.%Co-TiO 2 > Ni-TiO 2 > Fe-TiO 2 > TiO 2 > Cu-TiO 2 samples.In general, a rapid catalytic reaction occurs when many reactants are welladsorbed over the catalyst.The presence of transition metals in the 5.0 mol.%M-TiO 2 samples most likely caused the relative increase in the number of CO 2 and H 2 O molecules adsorbed compared to pure TiO 2 .This synergy contributed significantly to improving the catalytic performance of the 5.0 mol.%M-TiO 2 samples.The regular array of composites could be observed clearly through both the side and top directions.The TiO 2 /basalt fiber and 5.0 mol.%Fe-TiO 2 /basalt fiber composites did not cover the Pyrex plate surface completely, and bulk pores were observed.On the other hand, the coating parts were quite uniform and fine particles were well-dispersed.In contrast, the surfaces of the films fabricated from 5.0 mol.%Co-TiO 2 /basalt fiber, 5.0 mol.%Ni-TiO 2 /basalt fiber, and 5.0 mol.%Cu-TiO 2 /basalt fiber composites were quite dense, despite containing some cracks.In particular, the surface was the most compact over the 5.0 mol.%Co-TiO 2 /basalt fiber film.The thickness of the films in all samples was approximately 30∼35 m with the exception of the pure basalt fiber film (20 m).

Characteristics and CO
Energy-dispersive X-ray spectroscopy (EDAX) confirmed the presence of metals on the surfaces of the basalt fiber and 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 /basalt fiber films, as shown in Figure 8, and the table below lists the atomic compositions determined by EDAX.For the basalt fiber, various metal oxides were observed, such as Na, K, Ca, Mg, Al, Si, Fe, and Ti.Pure basalt fiber exhibited a composition of 19.28 wt.% Si, 3.14 wt.% total alkali metals, 1.66 wt.% Ti, 11.77 wt.% Fe, and 5.61 wt.% Al.As a CO 2 adsorbent, the Ca and Mg contents were 5.31 and 2.09 wt.%, respectively.Ti contained on the 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 /basalt fiber films decreased in the following order: 5.0 mol.%Ni-TiO 2 (47.74 wt.%) > Co-TiO 2 (42.68 wt.%) > TiO 2 (40.94 wt.%) > Fe-TiO 2 (33.44 wt.%).The amounts of transition metal species were in the range 2.69 to 3.02 wt.%, with the exception of copper (0.98 wt.%).These values did not appear to be perfectly quantitative: EDAX is a very good surface analytical method but it is prone to error because the composition can vary according to the location.In particular, the variation is large when the sample is nonuniform [39].
The efficiencies of the photogenerated electron-hole production in the basalt fiber and 5.0 mol.%M (Fe-, Co-, Ni-, International Journal of Photoenergy and Cu-) TiO 2 /basalt fiber films were measured from the photocurrent response under solar light irradiation at an applied potential of 0.7 V versus SCE. Figure 9 shows the typical real time photocurrent response of the films when the light source is switched on and off, exhibiting a rapid photocurrent rise and decay.In semiconductor systems, when irradiation provides an energy higher than the bandgap of a semiconductor, the energy excites an electron from the valence band to the conduction band, leaving a hole in the valence band.The electron-hole pair is responsible for the photocurrent.When the light was turned on, a rapid increase in the photoreduction current was observed, and the photocurrent then turned to a steady state after a few seconds.When the light was turned off, the photocurrent decreased instantaneously to almost zero [40].When the light was turned on, the maximum photocurrents obtained for the 5.0 mol.%Co-TiO 2 /and Ni-TiO 2 /basalt fiber films were 310.49 and 300.75 mA cm −2 , respectively, which were more than four times higher than that achieved on pure basalt fiber (71.23 mA cm −2 ) and TiO 2 /basalt fiber films (74.01 mA cm −2 ).The maximum photocurrent of the 5.0 mol.%Cu-TiO 2 /basalt fiber film was too small due to trapping of excited electrons by copper with strong oxidation agency.In addition, the TiO 2 /basalt fiber, 5.0 mol.%Cu-TiO 2 /basalt fiber, and 5.0 mol.%Ni-TiO 2 /basalt fiber films showed no current transients in both the light-on and the light-off regions in the samples falling off with time in 50 s to a steady state, which indicates that few surface recombination processes had occurred.Clearly, some electron-hole recombination was observed in the 5.0 mol.%Fe-TiO 2 /basalt fiber and Co-TiO 2 /basalt fiber films, but it was negligible compared to the total current value.Therefore, the 3d-transition metal ingredients have a beneficial effect on the photocurrent; it plays a role as an intermediate for the efficient separation of photogenerated hole-electron pairs.Figure 10 [42].As shown in this figure, almost no methane was observed when only the lamp was turned on.On the other hand, the amounts of CO and O 2 generation increased, and a small amount of hydrogen was observed.This means that the light does not induce photoreduction processes alone.On the other hand, there are different product distributions according to the loaded metal species: excessive methane is generated over the catalysts loaded with Ni and Co, and the quantities of CO and CH 3 OH gases are increased over the catalysts loaded with Cu and Fe.The M-TiO 2 /basalt fiber films showed higher CH 4 production from CO 2 photoreduction than the pure basalt fiber and TiO 2 /basalt fiber films.70.0 mol g cat −1 L −1 CH 4 was emitted over the pure basalt fiber film, suggesting that it contains a small amount of TiO 2 .On the other hand, CH 4 production was 112.1 mol g cat −1 L −1 over the TiO 2 /basalt fiber film.In contrast, the maximum  CH 4 yields over the 5.0 mol.%Co-TiO 2 /basalt fiber and Ni-TiO 2 /basalt fiber films were 360.5 and 307.3 mol g cat −1 L −1 , respectively.The differences in yields were attributed to their band-gaps, electron-hole recombination tendencies, and the gas adsorption abilities of the 5.0 mol.%M-TiO 2 /basalt fiber films.In particular, this study confirmed that the 5.0 mol.%Co-TiO 2 particles show synergistic performance when the basalt fiber is mixed.
The reduction of CO 2 requires a multiple electron transfer and leads to production of a variety of products, depending on the number of transferred electrons, which determine the final oxidation state of the carbon atom.The standard redox potentials of the CO 2 reduction half-reactions vary from CO 2 /HCOOH [43] to CO 2 /CH 4 [44]: CO 2 adsorbed on a photocatalyst surface can be reduced to • CO 2 − anion radical, which can react with H + and e − for forming HCOOH or HCOO − or which can go to the postulated disproportionation reaction of two • CO 2 − anion radicals into CO (after all to be C radical) and CO 3 2− to react with H radical.This study seems to follow the latter.Scheme 1 presents a model for CO 2 photoreduction over the M-TiO 2 /basalt fiber films based on the relationships between the optical properties of the photocatalysts and the catalytic activities.Photon excitation in M-TiO 2 over the M-TiO 2 /basalt fiber films will begin rapidly because M-TiO 2 has a shorter band-gap than pure TiO 2 , and the excited electrons can be transferred efficiently to CO 2 molecules.In addition, CO 2 molecules are adsorbed preferentially and easily on the basalt fiber surface.The positive holes on the valance band of M-TiO radicals by electrons, and the H radicals then react with C and CO radicals formed from the reduction of CO 2 − over M-TiO 2 , particularly over Co-or Ni-TiO 2 semiconductors, resulting methane production [45].The mixed films of basalt fibers and M-TiO 2 can also promote the separation of photogenerated electron-hole pairs (e − /h + ) on M-TiO 2 .Therefore, the synergetic effects of the basalt fibers and M-TiO 2 in the M-TiO 2 /basalt fiber films achieved higher CO 2 reduction efficiency.

Conclusions
The 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) incorporated TiO 2 photocatalysts obtained by the solvothermal method were coated densely with basalt fibers on a Pyrex plate using a squeeze technique and applied to the photoreduction of CO 2 to CH 4 .The CO 2 adsorption abilities were improved over the basalt fiber, and CH 4 generation was enhanced dramatically over 5.0 mol.%M-TiO 2 /basalt fiber films with a threefold higher yield compared to the pure basalt fiber and TiO 2 /basalt fiber films.In particular, the photoreduction of CO 2 with H 2 O revealed a remarkable increase in CH 4 generation over the 5.0 mol.%Co-TiO 2 /basalt fiber film to 360.5 mol g cat −1 ⋅L −1 after an 8 h reaction.

Figure 7 presents
SEM images (top and side views) of the six samples of pure basalt fiber and TiO 2 films and 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 /basalt fiber films.The basalt fiber in the films was approximately 15 m in diameter with different lengths and the surfaces were smooth.The 5.0 mol.%M-TiO 2 materials were mixed with basalt fibers and were fabricated as thick films by coating on a Pyrex plate.

Figure 8 :
Figure8: Energy-dispersive X-ray spectra of the surfaces of the pure basalt fiber and 5.0 mol.%M-TiO 2 /basalt fiber films and the atomic compositions.

Figure 10 :
Figure 10: Photoreduction abilities of CO 2 with H 2 O vapor to CH 4 over the pure basalt fiber film and 5.0 mol.%M-TiO 2 /basalt fiber films after 8 h.

International 2 2 M-TiO 2 (
Scheme 1: Model for CH 4 production from the photoreduction of CO 2 with H 2 O vapor over the M-TiO 2 /basalt fiber films.
presents the photoreduction abilities of CO 2 with H 2 O vapor to CH 4 over the pure basalt fiber film and 5.0 mol.%M (Fe-, Co-, Ni-, and Cu-) TiO 2 /basalt fiber films The photogenerated electrons on the photocatalysts by UV-radiation induce CO 2 reduction to produce CO 2 radicals, whereas the holes react with the adsorbed H 2 O molecules to perform oxidation.The intermediate photogenerated species undergo different reactions to produce CO, CH 3 OH, and CH 4 .The production of CH 4 from methyl radicals ( • CH 3 ) was confirmed, which are dependent directly on the formation of the intermediate product, CO