Photocatalytic Activity and Durability of Commercial Ti[O.sub.2] Photocatalysts for Indoor Air Purification.
Poor indoor air quality (IAQ) in non-industrial buildings can be largely attributed to the presence of several classes of VOCs (e.g. alcohols, aromatics, aldehydes, alkanes, ketones, etc) which enter the indoor environment via ventilation and/or emissions from indoor sources. These compounds can cause serious health issues including irritations of upper respiratory system, sinus infection, allergic reaction, asthma, headache, nausea, and dizziness (Wang et al. 2007). Typically, ventilation is employed to control the level of VOCs in indoor environment. However, the effectiveness of this strategy is greatly affected by the outdoor air quality and, except in the mildest of climates, conditioning (e.g. heating/cooling and humidifying/dehumidifying) the outdoor air to reach thermal comfort level requires a lot of energy. In this regard, improvement in IAQ and curtailment in energy consumption of HVAC system can be simultaneously realized by applying air cleaning devices (Haghighat et al. 2008; Hodgson et al. 2007; Bahri and Haghighat 2014). Among various approaches developed so far, UV-PCO air cleaners showed good results for abatement of VOCs at low concentrations (Mamaghani et al. 2017). Basically, in UV-PCO, a UV-illuminated semiconductor photocatalyst decomposes the VOCs adsorbed on its surface. When a semiconductor (e.g. Ti[O.sub.2]) is irradiated by photons having energy greater than semiconductor's band gap, due to the excitation of electrons from the valence band to the conduction band, electron-hole pairs ([e.sup.-]-[h.sup.+]) are produced. These charge carriers react with adsorbed water, surface hydroxyls, and [O.sub.2] and produce reactive species such as hydroxyl radical (*OH). PCO reactions between these reactive radicals and VOCs result in degradation of VOCs to light intermediates/by-products and final inorganic products (C[O.sub.2] and water) (Nakata and Fujishima 2012). UV-PCO offers a number of advantages: room temperature operation, low pressure drop, medium to high activity towards most VOCs, and low energy consumption. Nevertheless, it is widely accepted that the generation of harmful by-products and deactivation of photocatalyst are critical shortcomings of this technology which need to be closely investigated. In this regard, several research groups have reported that a variety of by-products such as aldehydes and ketones can be formed during PCO reactions and enter the recirculated air stream (Hodgson et al. 2007; Mamaghani et al. 2018b; Mo et al. 2009; Wisthaler et al. 2007). This becomes a serious problem when the quantity of generated by-products is high and/or by-products pose a greater health risks than the parent compounds. On the other hand, the deactivation of photocatalyst during prolonged operations not only results in decrement in VOC removal efficiency, but also increases the amount of byproducts. Despite its great importance from the practical point of view, there are only a handful of articles which investigated the by-products generation during PCO air purification (Hodgson et al. 2007; Zhong et al. 2013) and the issue of deactivation (Zhong et al. 2016) under realistic operating conditions.
In this article we aim to present our experimental findings on the photocatalytic activity and durability of three Ti[O.sub.2] photocatalysts for treating a mixture of VOCs at low concentrations. Photocatalysts and the titania-coated air filer were characterized to be able to explain the trends in VOC removal efficiency and by-products generation. Health risk index was used to assess the risk associated with the gaseous by-products and the reactants in the outlet stream.
Experimental set-up and procedure
The experimental set-up for PCO tests is depicted in Figure 1. The coated filter is installed perpendicular to the airflow inside the photoreactor (made of aluminum with 10 cm x 10 cm (3.94 x 3.94 inch) inner cross section area and 1.3 m (51.18 inch) length) and two UV lamps (Philips, TUV PL-S 5W/4P) are placed at each side of the filter. Details regarding the operation of the system, and air sampling procedure and analyzing can be found in (Mamaghani et al. 2018b). The challenge compounds and their properties are listed in Table 1. The test procedure is as follows: after installing the titania-coated Ni filter in the reactor, air with target relative humidity and flow rate is introduced. A dark adsorption step is employed prior to PCO test to bring titania catalyst to adsorption equilibrium with the pollutants. Afterwards, the injection of VOC mixture with desired VOCs concentration is initiated. After 5 min of VOC injection, all the UV lamps are switched on and the samplings for determining the concentrations of VOCs (using GC-MS (PerkinElmer Clarus[R] 500)) and by-products (using HPLC (PerkinElmer Flexar)) are started. The experimental conditions are summarized in Table 2.
PCO filter preparation and characterization
P25 Aeroxide from Evonik[R], PC500 from Cristal Global Companies, and UV100 from Hombikat[R] were tested in this study. Titania-filters were prepared by a pipetting method in which 1 wt% aqueous solution of Ti[O.sub.2] powder is placed drop by drop on a nickel foam filter (Shanghai Tankii Alloy Material) until the solution covers the entire surface. Filter is dried at 80[degrees]C (176 [degrees]F) for 12 h and subsequently weighed to calculate the coating density. The morphology of the coated filters was studied by SEM (Hitachi S-4700). FTIR analyses were conducted using a Nicolet 6700 FT-IR. The crystalline properties were measured by X-ray diffraction (Bruker, D8 advance). Surface area was measured with a nitrogen adsorption apparatus (AUTOSORB-1, Quantochrome).
Photocatalytic system performance evaluation
The VOC single-pass removal efficiency ([eta]) at different times is calculated as follows:
[[eta].sub.t](%) = 100x[([C.sup.t.sub.up]-[C.sup.t.sub.down])/[C.sup.t.sub.up]] (1)
[C.sub.up.sup.t] and [C.sub.down.sup.t] are the upstream and downstream concentrations of each VOC (in ppb) at time t (min). By-product generation is defined as the difference between the amounts of detected compound in the outlet and inlet air streams:
By - products generation (ppb) - [G.sup.t.sub.t,up][G.sup.t.sub.t,dow] (2)
[G.sup.t.sub.1, up] and [G.sup.t.sub.i, down] are the upstream and downstream concentrations of by-product i in ppb at time t (min), respectively. Among the detected by-products, formaldehyde and acetaldehyde are extremely dangerous to human's health and are regarded as carcinogens. Consequently, it is important to evaluate the risk level to human health of each photocatalytic air filter during operation. To do so, a health-related index (HRI) was determined based on the concentrations of by-products (C) and the recommended exposure limit (REL) values published by NIOSH (National Institute for Occupational Safety and Health) given in Table 3 (NIOSH 2018). [HRI.sub.i] and [HRI.sub.tot] are defined as:
[HRI.sub.wt]=[summation][HRl.sub.i] where [HRl.sub.i] = [[C.sub.t]/[REL.sub.t]] (3)
RESULTS AND DISCUSSION
Some of the key characteristics of selected Ti[O.sub.2] samples are reported in Table 4. [N.sub.2] adsorption-desorption results showed that UV100 and PC500 have considerable amount of micro and mesoporosity while P25 can be regarded as a non-microporous material. As noted, surface areas of UV100 and PC500 are much larger (roughly 7 times) than that of P25. XRD analysis revealed that P25 has a much higher degree of crystallinity with respect to the other samples. To a have a clearer comparison, relative anatase crystalinity was calculated by dividing the intensity of the anatase (101) diffraction peak of each sample to that of P25. As can be judged by the values of relative crystallinity and also crystal sizes listed in Table 4, UV100 and PC500 have very close crystalline qualities while P25 by far excels. It is noteworthy that anatase is the sole crystal phase in PC500 and UV100 while P25 is a mixture of anatase and rutile. Surface OH groups play a central role in the adsorption of VOCs on the surface of titania; thus, their population and type can have a great effect on the photocatalytic performance. FTIR analysis was performed in the OH spectral region (38002600 [cm.sup.-1]) to investigate the nature and concentration of surface OH groups on various titania samples. All photocatalysts showed broad and strong IR absorption between 3600-2800 [cm.sup.-1] which is assigned to the presence of different OH groups. Specifically, bands at 3625, 3663, and 3696-3728 [cm.sup.-1] are respectively ascribed to acidic, basic, and isolated hydroxyl groups (Mamaghani et al. 2018a).
The morphology of the titania-coated nickel filters were studied via SEM imaging. The SEM images of filters coated with P25, PC500, and UV100 are presented in Figure 2. It can be seen that nanoparticles mainly attached to the skeleton and walls of the filter while in some areas particles filled the void spaces of the nickel foam. Compare to UV100 and PC500, P25 coating was more uniform and a better dispersion of nanoparticles was achieved.
Photocatalytic performance and by-products generation
VOC removal efficiency strongly depends on the properties of photocatalyst (e.g. degree of crystalinity, crystal size, surface area, porosity, and population of surface hydroxyl groups) and the operating conditions (e.g. relative humidity, residence time, light intensity, and VOC concentration). Since in all experiments same operating parameters were applied (Table 2) and the fact that our titania samples have quite different features (Table 4), it is possible to pinpoint those features of photocatalyst that influence the activity the most.
Figure 3a, 4a, and 5a display the trends in VOCs removal efficiency with time over different titania samples. As could be expected, for all VOCs, an almost steady downward trend for removal efficiency with time of operation can be seen. This progressive decline in the photocatalytic performance after only 60 h of operation can stem from a number of facts. One of the main reasons of photocatalyst deactivation is the adsorption of heavy/poisonous reaction intermediates/by-products on active sites (Mo et al. 2013). Due to the fact that a mixture of VOCs undergoes PCO reactions, a great number of by-products could form on the surface of titania. In the case of aromatics (i.e. toluene and o-xylene), it has been reported that the strong adsorption of benzaldehyde, benzene, and benzoic acid on the surface can lead to severe deactivation (Mo et al. 2013; Wang and Ku 2003). For alcohols, formation and adsorption of carboxylic acids (i.e. acetic acid, butanoic acid, etc) were accounted for the deactivation of titania (Tang and Yang 2012). We observed that the color of P25-coated filter changed from white to light yellow after 60 h of testing, which could result from the adhesion of carbonaceous residues to the surface (Weon and Choi 2016). Depletion of surface hydroxyl groups can also result in photocatalyst deactivation in dry condition or when photocatalyst is incapable of replenishing the OH groups by dissociation of adsorbed water (Geng et al. 2010).
With some exceptions, the removal efficiency for various VOCs on all photocatalysts follows the order: 2-propanol~1-butanol > MEK > o-xylene > toluene > octane > n-hexane. Apparently, the removal efficiency is closely connected to the adsorption affinity of titania towards each VOC. The main intermolecular forces between titania and VOCs are: van der Waals and hydrogen bonding for alcohols, dipole-dipole interactions for ketones, dispersion forces for alkanes, and van der Waals interactions for aromatics (Zhong et al. 2013). The observed removal efficiency sequence agrees well with strength of intermolecular forces. Moreover, taking into account that at a high relative humidity (60%) a well-organized multi-layer water film forms on the surface of titania and also the hydrophilicity of Ti[O.sub.2], the water-solubility and polarity of VOC can be very influential. Consequently, polar and water-soluble VOCs (Table 1) such as 2-propanol and MEK have far greater adsorption capacities than non-polars (i.e. alkanes and aromatics). Another factor that could to some extent explain the order of removal efficiencies in each VOC class is the reaction rate with hydroxyl radical. It is widely acknowledged that *OH is the main oxidizing agent during the PCO of VOCs in gas phase. Therefore, considering the short residence time applied here (0.012 s), high reaction rate with hydroxyl radicals can significantly contribute to superior removal efficiency. As can be noted in Table 1, the compound with higher molecular weight in each family has a greater reaction rate with *OH. This corresponds well with presented results in Figure 3a, 4a, and 5a, showing [[eta].sub.1-Butanol] > [[eta].sub.2-propanol], [[eta].sub.o-xylene] > [[eta].toluene], and [[eta].sub.octane] > [[eta].sub.n-hexane]. As shown in Figures 3b, 4b, and 5b, the main gaseous by-products were acetone, acetaldehyde, formaldehyde, and propionaldehyde. Moreover, in some experiments trace amounts of benzaldehyde and crotonaldehyde were also found in the gas phase. Detecting only a small number of by-products could be an indicator of high degree of mineralization or may be attributed to the fact that many of by-products stay on the surface (or in the water film) instead of becoming airborne. An interesting observation is that in general the amounts of by-products decrease with reaction time which seems peculiar considering gradual photocatalyst deactivation. This can be explained by the sharp and continuous reduction in VOC removal efficiencies for all photocatalysts. By-products are generated due to the partial oxidation of VOCs over titania. Since with time smaller amounts of VOC adsorb on the surface and participate in PCO reactions, it is reasonable to have lower quantities of by-products as well.
Figures 3-5 clearly reveal that in terms of VOC removal efficiency and amounts of generated by-products UV100 and PC500 outperform P25, especially at longer exposure times. In order to justify these results, it is important to highlight the connections between the key properties of photocatalysts and PCO processes. Crystalinity and crystal size directly affect [e.sup.-]-[h.sup.+] pairs generation, separation, and migration to the surface which is the prerequisite for reactive radicals formation. In this regard, P25 has the advantage of high degree of crystallinity (roughly three times UV100 and PC500) and the co-presence of two crystalline phases which could greatly help photoactivity (Mamaghani et al. 2017, 2018b). However, as can be judged by the removal efficiencies, this positive aspect of P25 was clearly overpowered by the relatively large surface areas of UV100 and PC500 (almost six times P25). In heterogeneous photocatalysis adsorption of pollutants on the surface of titania and effective contact between the VOC molecules and active species are crucial. Large surface area of UV100 and PC500 offer a greater number of active sites while their high porosity makes pollutants access/diffusion to the interior of catalyst (and reaction sites) easier. Furthermore, surface OH groups density exerts a great influence on the adsorption of VOCs and also the photocatalytic reactions since they react with photogeneraated holes and produce *OHs. As can be calculated by the data given in Table 4, the total number of surface OH on P25 is roughly one-third of that on PC500 and UV100. Figure 6 illustrates the variations in by-products-HRI and reactants-HRI (unreacted VOCs exist in the outlet air) with time using different photocatalytic systems. It should be highlighted that since in our experiments none of the VOCs in the mixture are carcinogenic (for safety reasons) and have very larges RELs (i.e. high health risk), the inlet HRI is obviously much smaller compared to the outlet air which contains for instance formaldehyde. Based on HRI values one may argue that the PCO air cleaner, in fact, makes the air more harmful to humans. This is not the case in actual application since the indoor air normally contains many toxic pollutants such as formaldehyde, acetaldehyde, benzene, etc. As a result, it can be expected that when we apply the PCO air cleaners in HVAC systems, the HRI of the treated air would be smaller than the inlet HRI. By-products HRIs for all photocatalysts gradually decrease with time which is understandable considering the smaller amounts of generated by-products because of decrement in removal efficiencies at longer exposures. In contrast, reactants HRIs steadily grow due to the deactivation of photocatalyst which causes a larger portion of challenge VOCs leaves the PCO filter unreacted.
Photocatalytic activity and durability of three commercial titanium dioxides (PC500, UV100, and P25) were investigated for degradation of a mixture VOCs at low concentrations. Experiments were conducted in a single-pass continuous flow photoreactor under realistic operating conditions (residence time, relative humidity, concentration, and light intensity) for 60 h. VOC removal efficiencies sharply diminished with time on all titania samples which could be originated from the accumulation of intermediates/by-products on the surface and/or dehydroxylation of the surface. Removal efficiencies can be ranked: 2-propanol~1-Butanol > MEK > o-xylene > toluene > octane > n-hexane. This order could be rationalized based on the adsorption affinity between these compounds and titania and the reaction rate with *OH. Detected gaseous by-products were acetone, acetaldehyde, formaldehyde, propionaldehyde, benzaldehyde and crotonaldehyde. UV100 and PC500 noticeably outperformed P25 in VOC removal efficiency and amount of by-products which could be due to the large surface area, high porosity, good crystallinity, and large concentration of surface OHs. UV100 possessed the smallest values of HRIs in most time intervals while P25 showed comparatively high values of HRI (>0.35), especially in the beginning of operation.
The authors would like to express their gratitude to Concordia University for the support through the Concordia Research Chair--Energy & Environment.
Geng, Q., Guo, Q., and X. Yue. 2010. Adsorption and Photocatalytic Degradation Kinetics of Gaseous Cyclohexane in an Annular Fluidized Bed Photocatalytic Reactor. Industrial & Engineering Chemistry Research 49: 4644-4652.
Haghighat, F., Lee, C.-S., Pant, B., Bolourani, G., Lakdawala, N., and A. Bastani. 2008. Evaluation of various activated carbons for air cleaning--Towards design of immune and sustainable buildings. Atmospheric Environment 42: 81768184.
Hodgson, A.T., Destaillats, H., Sullivan, D.P., and W.J. Fisk. 2007. Performance of ultraviolet photocatalytic oxidation for indoor air cleaning applications. Indoor Air 17: 305-316.
Mamaghani, A.H., Haghighat, F., and C.-S. Lee. 2017. Photocatalytic oxidation technology for indoor environment air purification: The state-of-the-art. Applied Catalysis B: Environmental 203: 247-269.
Mamaghani, A.H., Haghighat, F., and C.-S. Lee. 2018a. Gas phase adsorption of volatile organic compounds onto titanium dioxide photocatalysts. Chemical Engineering Journal 337: 60-73.
Mamaghani, A.H., Haghighat, F., and C.-S. Lee. 2018b. Photocatalytic degradation of VOCs on various commercial titanium dioxides: Impact of operating parameters on removal efficiency and by-products generation. Building and Environment 138: 275-282.
Bahri, M., and F. Haghighat. 2014. Plasma-Based Indoor Air Cleaning Technologies: The State of the Art-Review. CLEAN--Soil, Air, Water 42: 1667-1680.
Mo, J., Zhang, Y., and Q. Xu. 2013. Effect of water vapor on the by-products and decomposition rate of ppb-level toluene by photocatalytic oxidation. Applied Catalysis B: Environmental 132-133: 212-218.
Mo, J., Zhang, Y., Xu, Q., Zhu, Y., Lamson, J.J., and R. Zhao. 2009. Determination and risk assessment of by-products resulting from photocatalytic oxidation of toluene. Applied Catalysis B: Environmental 89: 570-576.
Nakata, K., and A. Fujishima. 2012. Ti[O.sub.2] photocatalysis: Design and applications. Journal of Photochemistry and Photobiology C: Photochemistry Reviews 13: 169-189.
NIOSH. 2018. National Institute for Occupational Safety and Health (NIOSH), http://www.cdc.gov/niosh/.
Tang, F., and X. Yang. 2012. A "deactivation" kinetic model for predicting the performance of photocatalytic degradation of indoor toluene, o-xylene, and benzene. Building and Environment 56: 329-334.
Wang, S., Ang, H.M., and M.O. Tade. 2007. Volatile organic compounds in indoor environment and photocatalytic oxidation: State of the art. Environment International 33: 694-705.
Wang, W., and Y. Ku. 2003. Photocatalytic degradation of gaseous benzene in air streams by using an optical fiber photoreactor. Journal of Photochemistry and Photobiology A: Chemistry 159: 47-59.
Weon, S., and W. Choi. 2016. TiO2 Nanotubes with Open Channels as Deactivation-Resistant Photocatalyst for the Degradation of Volatile Organic Compounds. Environmental Science & Technology 50: 2556-2563
Wisthaler, A., Strom-Tejsen, P., Fang, L., Arnaud, T.J., Hansel, A., Mark, T.D., and D.P. Wyon. 2007. PTR-MS assessment of photocatalytic and sorption-based purification of recirculated cabin air during simulated 7-h flights with high passenger density. Environmental Science and Technology 41: 229-234.
Zhong, L., Haghighat, F., Lee, C.-S., and N. Lakdawala. 2013. Performance of ultraviolet photocatalytic oxidation for indoor air applications: Systematic experimental evaluation. Journal of Hazardous Materials 261: 130-138.
Zhong, L., Lee, C.-S., Haghighat, F., and A. Bahloul. 2016. Deactivation and ultraviolet C-induced regeneration of photocatalytic oxidation air filters. Science and Technology for the Built Environment 22: 576-585.
Student Member ASHRAE
Fariborz Haghighat, PhD
Chang-Seo Lee, PhD
Associate Member ASHRAE
Alireza Haghighatmamaghani is a PhD student, Fariborz Haghighat is a professor and Chang-Seo Lee is a research associate in the Department of Building, Civil and Environmental Engineering, Concordia University, Montreal, Canada.
Table 1: Physical and chemical properties of the selected VOCs. Compound VOC Chemical formulae Dielectric family constant 2-Propanol Alcohols [C.sub.3][H.sub.8]O 18.6 1-Butanol [C.sub.4][H.sub.10]O 16.68 n-Hexane Alkanes [C.sub.6][H.sub.14] 1.88 Octane [C.sub.8][H.sub.18] 1.94 Toluene Aromatics [C.sub.7][H.sub.8] 2.38 O-Xylene [C.sub.8][H.sub.10] 2.57 MEK Ketones [C.sub.4][H.sub.8]O 18.51 Compound k ([cm.sup.3]/molecule/s) 2-Propanol 5.7x[10.sup.-12] 1-Butanol 7.8x[10.sup.-12] n-Hexane 5.2x[10.sup.-12] Octane 8.11x[10.sup.-12] Toluene 6x[10.sup.-12] O-Xylene 1.68x[10.sup.-11] MEK 1.1X[10.sup.-12] Table 2. PCO tests experimental conditions (mean value [+ or -] 95% confidence interval). Parameter Value Unit Total 108 [+ or -] 3.80 ppb VOC inlet concentration Relative 56.9 [+ or -] 2.20 % humidity Volumetric 50 (1.76) L/min ([ft.sup.3]/min) flow rate Residence 0.012 s time Light 50 (322.6) mW/[cm.sup.2] intensity (mW/[in.sup.2]) Temperature 23.1 [+ or -] 0.8 [degrees]C ([degrees]F) (73.58 [+ or -] 1.44) Ti[O.sub.2] 1[+ or -] 0.03 mg/[cm.sup.2] concentration ((1.422[+ or -]0.043)X10-5) (lb/[in.sup.2]) Table 3. Health-related information of tested VOCs and generated by-products Pollutnat REL (ppm) REL data source IARC carcinogenic classification 2-Propanol 400 NIOSH Group 3, not classifiable as to its carcinogenicity to humans 1-Butanol 50 NIOSH REL for - isobutanol n-Hexane 50 NIOSH - Octane 75 NIOSH - Toluene 100 NIOSH Group 3, not classifiable as to its carcinogenicity to humans O-xylene 100 NIOSH Group 3, not classifiable as to its carcinogenicity to humans MEK 200 NIOSH - Formaldehyde 0.016 NIOSH Group 1, carcinogenic to humans Acetaldehyde 0.078 OEHHA (*) Group 2B, possibly carcinogenic to humans Acetone 250 NIOSH - Propionaldehyde - Not established - (*) Office of Environmental Health Hazards Assessments, California Environmental Protection Agency, U.S. Table 4. Characteristics of several commercial photocatalysts Properties BET Mean pore Crystal size Crystalline phase ([m.sup.2]/g) diameter (nm) (nm) UV100 330 <5 9.1 Anatase PC500 345 6.1 10.4 Anatase P25 52 Non-porous 25.8 Anatase (81.3%), Rutile (18.7%) Properties Relative anatase Surface OH crystalinity density (/n[m.sup.2]) UV100 0.31 2.5 PC500 0.35 2.4 P25 1 4.8
|Printer friendly Cite/link Email Feedback|
|Author:||Haghighatmamaghani, Alireza; Haghighat, Fariborz; Lee, Chang-Seo|
|Date:||Jan 1, 2019|
|Previous Article:||Quantifying the Importance of Measured Metabolic Rates and Clothing Levels for Effectively Using a Predicted Mean Vote Model: A Global Sensitivity...|
|Next Article:||Capabilities and Limitations of Commercially Available Wireless Indoor Environment Sensors.|