Enhanced TiO₂-Based Photocatalytic Volatile Organic Compound Decomposition Combined with Ultrasonic Atomization in the Co-Presence of Carbon Black and Heavy Metal Nanoparticles

Abstract

Volatile organic compounds (VOCs) are representative indoor air pollutants that negatively affect the human body owing to their toxicity. One of the most promising methods for VOC removal is photocatalytic degradation using TiO₂. In this study, the addition of carbon black (CB) and heavy metal nanoparticles (NPs) was investigated to improve the efficiency of a TiO₂-based photocatalytic VOC decomposition system combined with ultrasonic atomization and ultraviolet irradiation, as described previously. The addition of CB and Ag NPs significantly improved the degradation efficiency. A comparison with other heavy metal nanoparticles and their respective roles are discussed.

1. Introduction

Volatile organic compounds (VOCs) are representative indoor air pollutants that negatively affect the human body owing to their toxicity and carcinogenicity. These pollutants are closely related to the sick building syndrome, including mucous membrane irritation, headaches, and fatigue [1,2]. Photocatalytic degradation is one of the most promising methods for VOC removal [3,4,5,6]. TiO₂-photocatalyzed systems have been extensively studied because of their potential for operation under ultraviolet (UV, typically UV-A) irradiation without external heating. The presence of water molecules in the gas phase is effective in maintaining catalytic activity, and the formation of OH radicals is the key to achieving high degradation efficiency. However, hydrophobic VOCs such as toluene are less reactive than hydrophilic VOCs in TiO₂-photocatalyzed systems. A combined ozone oxidation and photocatalytic degradation system has also been studied [7,8,9,10,11]. Despite the effective degradation of gaseous organic compounds, these systems still suffer from several problems related to their strong dependence on relative humidity [12], the interference of intermediate degradation products with the degradation process [13,14], and catalyst deterioration owing to the adsorption of intermediates [13,15]. In addition, the handling of ozone has some complexity due to its toxicity under high concentration conditions. Although a combined system using shorter wavelength UV-C has also been studied to generate OH radical effectively [16,17,18], the in situ formation of ozone is concerned. Therefore, developing other methods to effectively accelerate photocatalytic reactions is still desirable.

Sekiguchi et al. proposed a method for the decomposition of VOCs using mists containing TiO₂ particles generated by the ultrasonic (US) irradiation of a TiO₂ particle suspension [19,20]. Ultrasonically generated mists range from nanometer to micrometer and allow the transportation of solid materials in the gas phase depending on the size of the mists [21,22]. This method is applicable to the continuous decomposition of toluene as a model hydrophobic VOC by increasing the gas–liquid interphase owing to atomization. A possible application of this system is a humidifying air purifier with degradation functionality. In this system, ultrasonically generated mists react with gas-phase toluene to produce water-soluble intermediates that are dissolved in the TiO₂ suspension phase and then further decomposed. The incorporation of the water-soluble intermediate products into the droplet is the key to maintaining a stable photocatalytic decomposition rate.

In this study, we aimed to improve the degradation efficiency of the aforementioned systems by combining them with additives such as carbon black (CB) and heavy metal nanoparticles (NPs). Carbonaceous materials are well-known adsorbents for capturing hydrophobic organic molecules, possibly incorporating the intermediate product in a water phase [3], and heavy metal NPs have the potential to improve the photocatalytic efficiency of TiO₂ systems [23,24]. The addition of both CB and Ag NPs was the best combination for improving the degradation efficiency in the gas and water phases. The roles of CB and Ag NPs are discussed.

2. Results and Discussion

Figure 1 shows the toluene degradation efficiency, E_d, obtained in the absence (denoted as TiO₂ only) and presence of heavy metal NPs (denoted as Ag, Pt, Pd, and Au). The addition of a very small amount of metal NPs enhanced the E_d from 40% to 48–60%. The E_d values increased in the following order: Ag > Pd > Pt > Au > TiO₂ only. Next, CB was added to the TiO₂ suspension (denoted as CB) or the Ag- or Pd NP-containing suspension (denoted as Ag+CB and Pd+CB, respectively). In all cases, the addition of CB increased the E_d by approximately 5% (from 40%, 60%, and 55% to 56%, 65%, and 59%, respectively; Figure 2). The addition of both heavy metal NPs and CB effectively improved the E_d. The lower effectiveness of metal NPs in the co-presence of CB compared (approximately 5% increase) might be due to the lower toluene concentration at the high-level conversion region and resulting lower mass transfer. The XRD measurements of TiO₂ (as received) before and after the addition of Ag NPs and under UV irradiation or not were performed. In all cases, only the XRD pattern derived from anatase and rutile was observed (Figure S1).

To gain insight into the degradation reaction and the effects of CB and Ag NPs, the formation of water-soluble organic compounds (WSOCs) was monitored under four sets of reaction conditions: TiO₂ only, TiO₂+CB, TiO₂+Ag, and TiO₂+CB+Ag. Toluene vapor was fed from 0 to 3 h while monitoring the concentration of the WSOCs, C_WSOC. After 3 h of reaction, toluene feeding was stopped, while the C_WSOC was continuously monitored. Figure 3 shows the time course of the C_WSOC in the TiO₂ suspension with and without additives (CB or Ag). Without any additives (TiO₂ only), after 3 h, the C_WSOC was approximately 0.9 mg-C/L and then slightly increased to approximately 1.0 mg-C/L after stopping toluene vapor feeding. In the presence of CB (TiO₂+CB), after 3 h, the C_WSOC was approximately twice as high as that without any additives (approximately 1.8 mg-C/L), and the C_WSOC monotonically increased from approximately 1.8 to 2.4 mg-C/L despite the absence of toluene vapor feeding. Toluene was adsorbed on the CB during vapor feeding and then decomposed into water-soluble intermediates, increasing the C_WSOC after stopping toluene feeding. When Ag NPs were added to the above two reaction systems (TiO₂+Ag and TiO₂+CB+Ag), the C_WSOC decreased after stopping toluene vapor feeding, resulting in much lower C_WSOC values (below 0.5 mg-C/L) compared to the cases in the absence of Ag (TiO₂ only and TiO₂+CB). These results clearly show that Ag NPs accelerate the TiO₂ photocatalytic degradation of toluene, especially the successive decomposition of water-soluble intermediates into mineralized products.

We also calculated the toluene decomposition rate normalized by carbon numbers (R_d) during toluene vapor feeding (0–3 h), as well as the formation rate of WSOC (R_g__WSOC) under toluene feeding (0–3 h) and after stopping the feeding (3–6 h). The R_d and R_g__WSOC were compared for four reaction conditions: TiO₂ only, TiO₂+CB, TiO₂+CB+Ag, and TiO₂+Ag (Figure 4). The addition of CB increased the R_g__WSOC. The ratio of the summed R_g__WSOC (determined by the summation of the values at 0–3 h and 3–6 h) to the R_d was also calculated (denoted as f_WSOC). The f_WSOC value for TiO₂+CB was calculated as 98.8%, which is much higher than that for TiO₂ alone (63.3%), confirming that the TiO₂ suspension promotes the degradation of water-soluble intermediates, whereas the main role of CB is likely the adsorption of toluene rather than the promotion of decomposition reactions. The addition of Ag NPs to TiO₂ and TiO₂+CB systems decreased the f_WSOC from 63.3% to 11.6% and from 98.8% to 23.4%, respectively, which supports the view that Ag NPs enhanced the degradation of water-soluble intermediates to mineralized products. In conclusion, the enhanced degradation efficiency of the TiO₂+CB+Ag system was ascribed to the combination of toluene adsorption by CB and the acceleration of the TiO₂-photocatalyzed degradation of water-soluble intermediates.

To investigate the adsorption properties of the series of TiO₂ suspensions, UV-vis measurements were performed. Figure 5 shows the UV-vis spectra for a series of TiO₂ suspensions, including one containing Ag NPs, after US and UV irradiation. Regardless of the presence or absence of Ag NPs, US irradiation induced an increase in absorbance in the UV-vis spectra owing to the enhancement of dispersiveness. UV irradiation further increased the absorbance, especially for the TiO₂ suspension containing Ag NPs (TiO₂+Ag), indicating that the oxidized surface of the Ag NPs was photoreduced by UV irradiation [25]. The UV-vis spectra of the TiO₂ suspensions are shown in Figure S2. The addition of heavy metal NPs resulted in enhanced absorbance in the UV-vis spectra. The absorbance at 360 nm increased in the following order: Ag > Pd > Au > Pt > TiO₂. This order is slightly different from that of the E_d (Ag > Pd > Pt > Au > TiO₂).

We also performed X-ray photoelectron spectroscopy (XPS) measurements to study the surface chemical states of TiO₂ before and after UV irradiation. Figure 6a–d show the XPS spectra for the Ti 2p region for a series of TiO₂ samples. The spectra for the O 1s are also shown in Figure S3. All the XPS spectra exhibited peaks at approximately 464 and 458 eV assignable to the Ti 2p_1/2 and 2p_3/2 states, respectively, which are derived from the Ti⁴⁺ species on TiO₂. For TiO₂ without metal NPs, the peak position and full width at half maximum (WFHM) values for Ti 2p_2/3 were nearly the same, regardless of the presence or absence of UV irradiation (0.741 and 0.740 eV for TiO₂ and TiO₂-UV, respectively, Figure 7). In the case of the Ag NP-containing TiO₂, the WFHM value increased from 0.744 to 0.763 eV upon UV irradiation, indicating that the Ti 2p_3/2 peak broadened. Ag NPs capture electrons generated from TiO₂ by UV irradiation and inhibit the recombination of electrons and holes [26]. In addition, previous reports have found that the Ti 2p_3/2 peaks were broadened in the XPS spectra of Ag-loaded TiO₂ prepared by photoreduction because some of the surface Ti⁴⁺ species were reduced to Ti³⁺ by trapped electrons [25]. In our case, the added Ag NPs were immobilized on the TiO₂ surface, promoting the reduction in the surface Ti⁴⁺ species in a similar manner.

We also performed a similar XPS analysis for other metal (Pd, Au, and Pt) NP-containing TiO₂. A similar increase in the WFHM value was observed after UV irradiation, although the degree of increase was lower than that for Ag NPs (Figure S4). The WFHM values increased in the same order as the E_d (i.e., Ag > Pd > Pt > Au, Figure 8). This result differs from the discussion of the UV-vis measurements (see above). To further investigate the correlation between the photocatalytic degradation performance and the surface chemical state, the WFHM value was plotted as a function of the toluene degradation rate, R_d. Good correlation was found between the WFHM value and R_d (Figure 9, R² = 0.97), indicating that capturing the electrons generated from TiO₂ by UV irradiation is the key to enhancing the degradation efficiency. Holes in TiO₂ generated by UV irradiation are known to react with H₂O to give OH radicals, which further react with organic molecules to promote degradation reactions [27]. The added Ag NPs effectively inhibited electron–hole recombination, serving as the most effective heavy metal NPs (Figure S5). The generated OH radicals effectively decompose toluene and WSOC intermediates captured by CB, resulting in the enhanced degradation efficiency.

3. Materials and Methods

3.1. Photocatalytic Degradation of Toluene

The experimental setup for the degradation of toluene (as a model VOC) using ultrasonically generated mists containing TiO₂ is shown in (Figure 10). The experimental reactor, which consisted of poly(methyl methacrylate) resin (Figure S6), was equipped with an ultrasonic transducer (Honda Electronics, Toyohashi, Japan, HM-303N). Ultrasonically generated mists containing TiO₂ were generated inside the reactor when a TiO₂ suspension in deionized water was irradiated with 2.4 MHz ultrasound waves. Degussa P-25 TiO₂ (Nippon Aerosil, Tokyo, Japan) was used in all the experiments because the P-25 TiO₂ particles can generate a sufficient amount of OH radicals in the liquid phase [28]. The crystal structure of the P-25 TiO₂ particles was approximately 80% anatase and 20% rutile, and the average particle diameter was approximately 30 nm. The submicrometer-sized aggregation of TiO₂ was observed by scanning electron microscopy (SEM) (Figure S7). The surface area of TiO₂, as measured with a BET surface analyzer (Micromeritics, Norcross, GA, USA, Flowsorb III-2305), was 50 m²g⁻¹. The TiO₂ concentration was fixed at 1.5 g L⁻¹. A black light blue (BLB) lamp with a maximum light intensity output of 365 nm (UV₃₆₅; Sankyo Denki, Hiratsuka, Japan, FL4BLB) was used as the light source. Detailed emission spectra of the lamps are presented elsewhere [19]. The observed relative humidity was over 95%, which was outside of the measurement range for all conditions, indicating that the mist in the reactor could be maintained under stable conditions.

Dry air containing 5 ppm of toluene vapor was introduced into the reactor at 3.0 L min⁻¹ from the bottom side of the reactor. Before the reaction, TiO₂ suspension was irradiated by a US wave under black light irradiation at 35 °C to remove the pollutants on TiO₂ surfaces and/or in aqueous phase. Toluene vapor was then introduced for 30 min to obtain a stable toluene concentration. After stabilizing the toluene concentration at 5 ppm, the degradation reaction was started (t = 0 min).

In this study, CB (Mitsubishi Chemical Corporation, Tokyo, Japan, MA100) and/or heavy metal NPs, such as Ag (particle size: 5–30 nm), Pt (1–6 nm), Au (1–4 nm), and Pd (2–7 nm) (Renaissance Energy Research Co., Ltd., Osaka, Japan, as a dispersed solution), were introduced into the reaction system. The experimental schedule is shown in Figure S8. To compare the E_d after the addition of the additives with that before the addition of the additives, an experiment was first conducted using only a suspension of TiO₂ particles, and then a similar experiment was repeated with the addition of the additives. The amounts of CB and metal NPs were 25 mg/L and 20 μmol/L, respectively. The E_d was determined as follows:

$$E_{\mathrm{d}}\left[\mathrm{\%}\right]=(1-{\displaystyle \frac{C_{\mathrm{o}\mathrm{u}\mathrm{t}}\left[\mathrm{p}\mathrm{p}\mathrm{m}\right]}{C_{\mathrm{i}\mathrm{n}}\left[\mathrm{p}\mathrm{p}\mathrm{m}\right]}})\times 100$$

(1)

3.2. Determination of the Amount of WSOCs

Ultrapure water treated by UV irradiation (ADVANTEC, Tokyo, Japan, RFU464CC) was used as the suspension to reduce the TOC concentration (approximately 5 ppb). The suspension (5 mL) in the reaction tank was collected from the sampling tube using a syringe (TERUMO, Tokyo, Japan, SS-10SZP) through a cartridge filter (ADVANTEC; 25HP045AN). The reaction conditions were the same as those for the photocatalytic degradation experiments.

The suspension for WSOC concentration measurement was collected using a syringe through a cartridge filter every 1 h during UV irradiation for 6 h (t = 0–6 h). The air supply containing toluene was stopped at t = 3 h, whereas samples were collected every 1 h in the same manner until t = 6 h. As a pretreatment procedure, 5 mL of each water sample in the suspension was diluted to 50 mL with ultrapure water and analyzed using a TOC analyzer. When the experiments were conducted in the presence of additives, the experiments were first conducted with a TiO₂ particle suspension only. After confirming the stabilization of degradation efficiency, additives were added, and the experiments and sample collection were performed in the same manner. The schedule of these experiments is shown in Figure S9.

For comparison of the amount of carbon in the decomposed toluene with the amount of WSOCs present in the suspension as determined by a TOC meter, the R_d [μg-C/s] and R_{g_WSOC} [μg-C/s] were calculated using the following equations.

$$R_{\mathrm{d}}\left[\mathsf{\mu }\mathrm{g}-\mathrm{C}/\mathrm{s}\right]=E_{\mathrm{d}}\times C_{\mathrm{i}\mathrm{n}}\left[\mathrm{g}/{\mathrm{m}}^3\right]\times Q\left[{\mathrm{m}}^3/\mathrm{s}\right]\times \left({\displaystyle \frac{M_{\mathrm{C}7} [\mathrm{g}-\mathrm{C}/\mathrm{m}\mathrm{o}\mathrm{l}]}{M_{\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}} \left[\mathrm{g}/\mathrm{m}\mathrm{o}\mathrm{l}\right]}}\right)\times {10}^6$$

(2)

$$R_{\mathrm{g}\_\mathrm{W}\mathrm{S}\mathrm{O}\mathrm{C}}\left[\mathsf{\mu }\mathrm{g}-\mathrm{C}/\mathrm{s}\right]=\left({\displaystyle \frac{{\overline{C}}_{\mathrm{W}\mathrm{S}\mathrm{O}\mathrm{C}}\left[\mathrm{g}/\mathrm{L}\right]\times V\left[\mathrm{L}\right]}{t \left[\mathrm{s}\right]}}\right)\times {10}^6$$

(3)

where M_C7, M_toluene, ${\overline{C}}_{\mathrm{W}\mathrm{S}\mathrm{O}\mathrm{C}}$, and V denote the molar mass of seven carbons, molar mass of toluene, average WSOC concentration in the suspension at t = 0–3 h, and suspension volume, respectively. The f_WSOC was then calculated from the R_d and R_{g_WSOC} as follows:

$$f_{\mathrm{W}\mathrm{S}\mathrm{O}\mathrm{C}}[\mathrm{\%}]={\displaystyle \frac{R_{\mathrm{g}\_\mathrm{W}\mathrm{S}\mathrm{O}\mathrm{C}} [\mathsf{\mu }\mathrm{g}-\mathrm{C}/\mathrm{s}]}{R_{\mathrm{d}} [\mathsf{\mu }\mathrm{g}-\mathrm{C}/\mathrm{s}]}}\times 100$$

(4)

3.3. XPS Analysis of the TiO₂ Suspension

Approximately 70 mL of the TiO₂ suspension was transferred from the photocatalytic reactor to a filter (ADVANTECH, GC-50) to separate the water by vacuum filtration. The obtained solid sample was dried in a desiccator for at least 24 h and used for XPS analysis. Heavy metal NP-containing TiO₂ samples were prepared in the same manner. For the UV-irradiated samples, the TiO₂ suspension was irradiated with BLB lamp in the reactor, and UV irradiation was performed immediately before the XPS sample preparation (Figure S10). XPS spectra were measured on a JEOL (Tokyo, Japan) JPS-9030 spectrometer having a modified UHV chamber employing Mg Kα radiation. Charge correction was performed using the O 1s peak at 532.0 eV.

4. Conclusions

In conclusion, CB and Ag NPs are effective additives for the TiO₂-photocatalytic decomposition of toluene under ultrasonic atomization. The R_d was enhanced by a factor of approximately two in the co-presence of CB and Ag NPs. A detailed analysis of the WSOC concentration indicates that the roles of CB and Ag NPs in improving the degradation efficiency are different. CB effectively accumulated WSOC intermediates in the aqueous phase, whereas Ag NPs accelerated the decomposition of WSOCs. The XPS analysis of TiO₂ after the addition of different heavy metal NPs followed by UV irradiation revealed a reduction in the surface species, possibly owing to the electrons generated from TiO₂ by UV irradiation, as indicated by the increase in the WHFM of the Tip_2/3 peak. This value increased in the same order as the E_d as follows: Ag > Pd > Pt > Au > TiO₂ (without NPs), implying that Ag NPs effectively capture the electrons derived from the UV irradiation of TiO₂, thus suppressing the electron–hole recombination and improving the photocatalytic degradation efficiency.