4.1. Removal of VOCs by Photocatalytic Oxidation
Photocatalytic oxidation (PCO) has attracted more attention because of its unique characteristics on the removal of chemicals. In recent years, PCO has been perceived as a technology to remove indoor VOCs. Titanium dioxide (TiO2) is known as the most extensive studied photocatalyst due to its excellent stability, high photo-activity, and suitable band gap structure. Low cost and non-toxicity are also the main advantages for its application.
The basic mechanism of photocatalytic degradation is that organics would be oxidized to H
2O, CO
2 or any inorganic harmless substances with •OH or superoxide (•O
2−) radicals, which are generated on the surface of photocatalyst (e.g., TiO
2) under ultra-violet (UV) light irradiation [
77]:
In the heterogeneous reaction system, TiO
2 is excited by the absorption of a photon with energy greater than or equivalent to the band gap energy of the semiconductor, resulting in the electron transition from the valence band to the conduction band. The radiation could consequently produce electrons and holes (e
−/h
+) in conduction band and valence band, respectively. Following the irradiation, the electrons and holes can undergo redox reactions with the adsorbed reactants on the photocatalyst’s surface that lead to the formation of intermediates and products. The reaction series are the so-called complete mineralization. Besides VOCs degradation, the reactions can be used as a method of disinfection and sterilization [
78,
79].
PCO of VOCs consists of a chain of stepwise reactions; that is, they take more than one elementary step to complete.
Scheme 1 shows a series of PCO reaction mechanisms for
o-xylene. Besides the final oxidized products, the steps also yield different oxidation states of the intermediates such as aldehydes, ketones or organic acids [
80]. These compounds can be qualified by real-time or offline monitoring and analytical methods such as gas chromatography/flame ionization detection (GC/FID), GC/mass spectrometry (GC/MS), high pressure liquid chromatography (HPLC), and Fourier-transform infrared spectroscopy (FTIR) [
81,
82].
Table 3 lists the intermediates formed in the PCO of VOCs (e.g., benzene, toluene and xylene) shown in the literature.
Table 3.
Summary on the intermediates formed in photocatalytic oxidation of typical indoor VOCs.
Table 3.
Summary on the intermediates formed in photocatalytic oxidation of typical indoor VOCs.
Target VOC | Concentration (ppm) | Light Source | Main Intermediates | Chemical | Analytical Method | Ref. |
---|
Benzene | 3000–6000 | 4000 W Xe lamp | Benzaldehyde, benzoic acid | - | GC/MS | [83] |
| 614 | White fluorescent lamp | Phenol | Hydroquinone, 1,4-benzoquinone | GC/MS | [84] |
| - | - | Phenol, hydro-quinone, benzoic acid | Malonic acid, benzoquinone | GC/MS/FTIR | [85] |
Toluene | 10 | Black light lamp | Benzaldehyde, benzoic acid | Benzyl alcohol | FTIR | [86] |
| 50–800 | 365 nm UV | Acetone, acetaldehyde, formaldehyde | Acrolein, butanone | TDS-GC/MS/FID, HPLC/UV/FTIR | [87] |
| 370 | >400 nm | Benzaldehyde, benzoic acid | - | DRIFTS | [88] |
Xylene | 3000–6000 | 4000 W Xe lamp | Benzaldehyde, Methyl-benzaldehydes | 2,5-Furandione, 1,3-isobenzofurandione | GC/MS | [83] |
| 25–75 | UV | o-Tolualdehyde, o-toluic acid, benzoate ion | - | FTIR | [89] |
For instance, the highly stable aromatic ring of toluene usually remains intact while its active methyl group can be oxidized step-by-step to benzoic acid. The formation of the carbonyl group even makes the phenyl ring more inert because the conjugation effect reduces its electron density. The complete oxidation products such as CO
2 and H
2O would be generated from any of the intermediates until the phenyl ring is broken. However, if PCO are conducted at room temperature, the active sites on the photocatalyst’s surface could be gradually occupied by irreversibly chemisorbed intermediates, which retard the reactions. For example, during the photocatalytic oxidation processes for toluene over TiO
2 catalysts, it was found that the toluene photooxidation behavior was strongly affected by the formation and oxidation behavior of intermediate compounds [
90]. The study carried out by Nakajima
et al., showed that H
2SO
4 treatment of TiO
2 surface provides higher photocatalytic removal efficiency on toluene which can be ascribed to the fast decomposition of intermediates by surface strong acid itself [
91]. Moreover, the progresses of the research carried out into TiO
2-based photocatalysts were summarized by several recent reviews [
21,
92].
Scheme 1.
The PCO reaction mechanism for o-xylene.
Scheme 1.
The PCO reaction mechanism for o-xylene.
Anatase and rutile, two crystalline phases of TiO
2, have been shown their feasibilities for PCO of indoor air pollutants under UV light irradiation. The band-gap energies of anatase and rutile are 3.23 and 3.02 eV, respectively. Anatase has shown better performance in PCO processes than that of rutile because of its more favorable conduction band configuration and stable surface peroxide groups. In general, TiO
2 is fixed on some substrate, such as hollow tubes, silica gel, beads, and woven fabric. These catalysts can be obtained using the methods such as electrochemical [
93], plasma deposited [
94], dip coating and sol-gel method [
95].
Table 4 summarizes potential photocatalysts used for removal of indoor VOCs. Different single or combined photocatalysts have particular removal rates and efficiencies in PCO. Most TiO
2-based catalysts have optimized performance on near-UV light region because of its large energy band gap between electron-hole pairs of ~3.2 eV. A light source at a wavelength (λ) of <387 nm is required to overcome the gap, understanding that the PCO can only uptake
ca. 3% of the sunlight [
96]. Therefore, a limited number of TiO
2 catalysts can exhibit high degradation activity under a visible light. A lot of works have been thus done on the improvement of TiO
2 photocatalytic efficiency, such as doping with nonmetals and metals and coupling with other supports. TiO
2 doping with a nonmetal atom can enhance the photo-response in a practical application [
97]. The nonmetal can substitute the oxygen on TiO
2 lattice and lead to a band gap narrowing, resulting in activation at far-visible light region. The common photocatalysts are primarily metal oxides, which can be doped with elements such as carbon (C), nitrogen (N) or transition metal ions. For instance, the nitrogen-doped catalysts can be activated more efficiently because of higher energy level of the valence band of N2p than O2p. The fluorescence-assisted TiO
2−xN
y can decompose pollutants such as acetaldehyde through gaseous phase photocatalytic reaction [
98]. CaAl
2O
4: (Eu, Nd)/TiO
2−xN
y composite is able to store and release energy to continuously maintain the visible-light response to TiO
2−xN
y, even in the darkness. Such a property allows the fluorescence-assisted photocatalysts to function at night without supplying extra light sources.
Table 4.
Summary on potential photocatalysts applied for indoor VOCs removal.
Table 4.
Summary on potential photocatalysts applied for indoor VOCs removal.
Photocatalyst | Preparation/Coating Method | Configuration | Compounds | Light Source | ηremoval (%) | Ref. |
---|
TiO2 | Sol-gel | F | Acetone, toluene p-xylene | UV lamp, 254 nm | 77–62 (3 L/min) | [95] |
TiO2 | Electrochemical | F | Acetaldehyde | UV | 99+ (110 min) | [93] |
TiO2 | Sol-gel | F | Toluene | Black light | 52 (3.6 L/min) | [86] |
TiO2 | Plasma deposited | F | m-Xylene | UV lamp | 99+ (30 min) | [94] |
TiO2−xNx | Calcination | P | Toluene | Visible light | 99+ (3000 min) | [82] |
TiO2−xNx | Hydrothermal | P | Acetaldehyde | Fluorescence | - | [98] |
C-TiO2 | Hydrothermal | P | Toluene | Visible light | 60+ (120 min) | [106] |
C-TiO2 | Hydrothermal | P | Toluene | Visible light | 20 (120 min) | [107] |
CNT-TiO2 | Hydrothermal | P | Styrene | UV-LED, 365 nm | 50 (20 mL/min) | [108] |
Pt/TiO2 | Photo-deposition | P | Benzene | Black light, 300–420 nm | 100 (100 mL/min) | [99] |
Ln3+-TiO2 | Sol-gel | P | Benzene, toluene, ethylbenzene, o-xylene | UV, 365 nm | 22–79 | [109] |
Ce-TiO2 | Sol-gel | F | Toluene | Visible light | 90 | [110] |
Fe-TiO2 | Sol-gel | P | p-Xylene | Visible light—LED | 22 (5 min) | [111] |
Fe-TiO2 | Sol-gel | P | Toluene | Visible light | 99+ (120 min) | [88] |
In(OH)3 | Ultrasound radiation | P | Acetone, Benzene, Toluene | UV lamp, 254 nm | 99+ (5 h) | [104] |
β-Ga2O3 | Chemical deposition | P | Benzene | UV-lamp, 254 nm | 60 (20 mL/min) | [105] |
Ag4V2O7/Ag3VO4 | Hydrothermal | P | Benzene | White fluorescent lamp | 99+ (120 min) | [84] |
Pt/WO3 | Photo-deposition | P | DCA, 4-CP, TMA | Visible light, >420 nm | 99+ (3 h) | [112] |
Pd/WO3 | Calcination | P | Acetaldehyde, toluene | Fluorescence/visible light | 99+ (3 h) | [26] |
The TiO
2-Pt/TiO
2 hybrid catalyst system allows a complete oxidation of benzene to CO
2 at ambient temperature [
99]. TiO
2 after doping with Pt has an increased number of active sites, which convert the intermediate form of carbon monoxide (CO) into CO
2. Pt/TiO
2 is thus the most useful catalyst for the purification of VE gases containing benzene. Doping with lanthanide ions can promote the formation of oxygen vacancies which have relatively high solubility compared with other oxygen species [
100]. In particular, cerium (Ce) is a low cost photocatalyst that has the ability to migrate between Ce
4+ and Ce
3+ through oxidization and reduction reactions. Ce doped with TiO
2 can decompose toluene under a visible light source.
Aside from TiO
2-based photocatalysts, other semiconductors can be also applied in the removal of VOCs such as ZnO [
101], ZnS [
102], SnO
2 [
103], In(OH)
3 [
104], and β-Ga
2O
3 [
105]. Nano-sized porous In(OH)
3 and porous Ga
2O
3 have high activity and long-term durability for photocatalytic decomposition of acetone, benzene, toluene and other aromatic derivatives under ambient conditions.
4.3. Influencing Factors
Photocatalytic reaction rate, additional with the reaction kinetic and adsorption coefficients, are direct tools to evaluate the efficiency of a photocatalyst in removal of VOCs.
Table 6 shows kinetic parameters and PCO conversion efficiency for the common VOCs. There are critical factors such as light source and intensity, pollutant concentration, RH, temperature, and deactivation and reactivation which can control the photocatalytic reaction rate. In order to study the PCO processes, many kinetic experiments for removal of common pollutants (e.g., benzene, toluene, xylene, and formaldehyde) have been thus conducted in optimal reactors. Here we summarize and review these factors.
Light source and intensity. The electron-hole pairs of a photocatalyst must be firstly excited for the subsequent VOCs degradation. The common catalysts (e.g., TiO
2) usually require an UV wavelength equivalent energy source for the excitation. Medium pressure mercury lamps, xenon lamps, and UV lights are common light sources for PCO. The light intensity is usually represented by units of light-irradiation (energy per unit area) or photon flux on the catalyst’s surface. Theoretically, the reaction rate of PCO is proportional to the intensity of the light supply. The reaction rate of PCO is regulated by the first order of consumption rate of electron hole pairs and a half order of their recombination rate [
129]. Thus there is no doubt that the light intensity can directly control the first-order of reaction [
95]. In addition, the internal structures of photocatalysts can affect the adsorption rate of the photons and consequently impact the conversion rate [
130]. Bahnemann and Okamoto [
131] investigated the relationships between UV light intensity and photocatalytic reaction rate with TiO
2. A linear correlation was found in the low intensity range whereas the degradation rate is proportional to square root of the light intensity under the moderate intensities. When light intensity is greater than 6 × 10
−5 Einstein L
−1·S
−1, the VOC degradation rate is not further enhanced subject to any changes.
As UV light is harmful to human and potentially leads to produce secondary pollutants (e.g., more strong oxidizing substances) in indoor air, more attention is being paid to applying visible light stimulating catalytic reaction for the removal of VOCs. However, the influences of light intensity are seldom studied with visible light sources. The formaldehyde removal rates with N-doped TiO
2 photocatalyst were enhanced linearly form 25.5% to 59.6%, and stabilized thereafter, when the intensity increased to 30,000 lux with an initial concentration of 0.98 mg/m
3 [
132].
Table 6.
Kinetic parameters and PCO conversion efficiency (%) for the common VOCs.
Table 6.
Kinetic parameters and PCO conversion efficiency (%) for the common VOCs.
Pollutants | Reactor Design | Initial Reaction Conditions | Deactivation | Ref. |
---|
RT | Photocatalyst | [VOC] Gas (ppm) | PW(nm)/I (mW·cm−2) | RH (%) | T (°C) |
---|
Styrene | CR | CNT-TiO2 | 25 ± 1.5 | 365/70 | - | - | Y | [108] |
Benzene | CR | Pt/TiO2 | 80 | 300–420/- | 65 | Ambient | n.r. | [99] |
CR | In(OH)3 | 920 | 245/- | - | 25 | n.r. | [104] |
Acetone | CR | In(OH)3 | 420 | 245/- | - | 30 ± 1 | n.r. | [104] |
Toluene | CR | TiO2 | 10 | >300/0.7 | 0–40 | Ambient | Y | [86] |
CR | TiO2 | 17–35 | 365/2.34 | 47 | 25 | n.r. | [104] |
CR | P25 | 50–800 | 365/10 ± 1 | 0–50 | 25 | n.r. | [87,104] |
CR | Ce-TiO2 | 0.15–0.6 | Visible/- | <3–75 | 42 | n.r. | [110] |
CR | Fe-TiO2 | 370 | >400/- | 60 | 25 | Y + N | [88] |
CR | Ln3+-TiO2 | 23 ± 2 | 365/0.75 | - | - | n.r. | [109] |
CR | In(OH)3 | 1220 | 245/- | - | 25 | n.r. | [104] |
CR | TiO2 fibers | 200 | 365/9 | 20–60 | - | n.r. | [133] |
Xylene | CR | P25 | 25–75 | UV/1.5 | 30–90 | - | Y | [89] |
Pollutant concentration. The concentration levels of pollutants can influence the photocatalytic performance in terms of the reaction rate. In the PCO process, the mass flux between the surface of photocatalyst and inlet can be accounted by the convective mass transfer [
134]:
where N
A is mass flux, k
A is convective mass transfer coefficient and ∆C
A is the concentration difference of transfer substance between the interface and the inlet. Eventually, the pollutant concentration over the photocatalyst’s surface varies from that in the inlet; however, it is difficult to accurately monitor the surface concentration by any means of measurement techniques. As a result, the use of inlet concentration for the computation of kinetic parameters may contain different degrees of errors. In order to decrease the concentration disparities, it is necessary to increase the airflow rate for improving the convective mass transfer [
135].
Pollutant concentration (C) and photocatalytic reaction rate (r) are the two kinetic parameters for reaction model computation. The Langmuir-Hinshelwood (L-H) model has been widely applied to establish the relationship between C and r in the PCO process for many VOCs such as acetone, benzene, toluene, and xylene [
136,
137]. In general, the degradation rate decreases when the pollutant concentration increases [
87,
95]. However, only a few investigations on the photocatalytic kinetics for indoor VOCs are reported. Among those, most have conducted the tests at an extremely high concentration (e.g., ppmv level). This demonstration concentration for a VOC could even cause instant headaches, irritation, and discomfort to humans [
138]. The results cannot reflect the realistic situation in most indoor environments (
i.e., pptv to sub-ppbv level). Ce-doped TiO
2 had a decrease in degradation efficiency while the formaldehyde levels increased from 0.1 to 0.5 mg/m
3 [
121]. In addition, in a concentration range of 0.1–1.0 mg/m
3, the degradation efficiency of formaldehyde was up to 80.8% with a photocatalyst from the 3M Company, but was sharpely reduced to 52.9% when the concentration raised to 2.0 mg/m
3 [
139].
Relative Humidity. Hydroxyl groups can be generated from water molecules adsorbed on the photocatalyst during the PCO processes, which can be captured by photo-generated charge carriers to produce reactive radicals (e.g., •OH) to further oxidize the indoor organic pollutants. Therefore, water vapor either from indoor air or generated from the mixed reactions plays a significant role in the photo-degradation process [
99]. In the absence of water vapor, the photo-degradation of VOCs (e.g., toluene) is seriously retarded since the mineralization could not completely occur. At the initial stage of the photocatalytic reaction, hydroxyl groups are expended due to the reactions between water vapor and organics on the photocatalysts’ surfaces. However, the presence of water vapor would lead to electron-hole recombination [
140]. There is also an adsorption competition between water and organics when RH is excessive. The water molecules can hide the active sites of the photocatalyst surface and so reduce the VOCs degradation rate and the photocatalytic activity. A typical breakthrough curve was obtained to demonstrate the competitive adsorption of water and toluene in the TiO
2 photocatalytic reactions [
86]. The result indicated that the photocatalyst is more sensitive to RH under a low hydrophobic condition. The indoor RH is usually regulated by ventilation (e.g., air-conditioning) or humidifiers, thus the competitive adsorption between water and trace contaminants has a strong impact on the oxidation rates [
135].
RH is also the key factor for formaldehyde degradation, which has been demonstrated with the photocatalytic performances of Zr
xTi
1−xO
2 at different RHs of 50% ± 5%, 65% ± 5%, 85% ± 5%, respectively [
127]. The work reported that the activity is the highest at RH of 50% ± 5%, representing that the photocatalytic reaction can be suppressed at humid environments. Similar observations were found for TiO
2-C coated and TiO
2-CN coated photocatalysts at a RH range of 20%–90% [
141]. The effect of RH on the degradation is negligible at a formaldehyde concentration of 3.3 ± 0.3 ppmv; while at a higher concentration level (8.6 ± 0.5 ppmv), the degradation efficiency significantly dropped at a RH of 90%. It is necessary to note that the impacts of water vapor on the removal efficiency for VOCs and formaldehyde were inconsistent for different photocatalysts. For this reason, an optimized working RH must be investigated when different systems are applied.
Deactivation and reactivation. Lifetime of a photocatalyst is an important parameter for the real application in removal of indoor pollutants. This should include the consideration of deactivation, regeneration, reactivation, or replacement. The gas-solid photocatalytic activity decreases with time while the number of effective active sites on the catalyst surface decreases at the same time. Deactivation thus occurs due to the accumulation of such partially-oxidized intermediates which occupy the active sites on the photocatalyst’s surface. Many kinetic studies indicate that the adsorption of poisonous intermediates during the initial stage of the photocatalytic reactions is almost irreversible. The initial oxidation rate is proportional to the effective surface area of catalyst. For instance, acetic acid and formic acid are the two main detectable intermediates formed in the photocatalytic degradation of acetaldehyde by TiO
2. Even though trace amounts of these intermediates could possibly discharge into the airs, these polar organic compounds have a stronger affinity to be accumulated on the photocatalyst’s surface until they can be decomposed by further steps of PCO. In some extent, a complete deactivation of the photocatalyst occurred after 20 consecutive PCO reactions due to the fully occupation of the active sites by the intermediates [
89]. Mendez-Roman investigated the relationship between the formation of surface species and catalyst deactivation during the photocatalytic oxidation of toluene, and their results showed that the accumulation of benzoic acid on the surface resulting in the catalyst deactivation [
142]. Recovery of photocatalytic activity requires a regeneration technique. The adsorbed polar intermediates such as benzaldehyde and benzoic acid can be removed completely with a heat treatment at 653 K for 3 h [
107]. However, such reactivation of the photocatalysts is a practically difficult since it consumes high energy or requires working with a furnace.
Other potential factors. Rather than the above, the loading amount of noble metal, content of the photocatalyst, and gas flow rate can also affect the photocatalytic activity. These multiple parameters can either advance or suppress the PCO subject to the kind of photocatalysts applied for the VOCs removal.