Indoor Formaldehyde Removal by Thermal Catalyst: Kinetic

ARTICLE pubs.acs.org/est

Indoor Formaldehyde Removal by Thermal Catalyst: Kinetic Characteristics, Key Parameters, and Temperature Influence Qiujian Xu,† Yinping Zhang,*,† Jinhan Mo,†,‡ and Xinxiao Li† †

Department of Building Science and ‡Department of Chemistry, Tsinghua University, Beijing 100084, P. R. China

bS Supporting Information ABSTRACT: Thermal catalytic oxidation (TCO) technology can continuously degrade formaldehyde at room temperature without added energy. However, there is very little knowledge on the TCO kinetic reaction mechanism, which is necessary in developing such air cleaners and in comparison with other air cleaning techniques. This paper addresses the problem of a novel TCO catalyst, Pt/MnOx
CeO2. The experiments measuring the outlet concentrations of formaldehyde and other possible byproducts were conducted at temperatures of 25, 40, 60, 100, and 180 °C and at a series of inlet formaldehyde concentrations (280
3000 ppb). To measure the concentrations precisely and real timely, proton transfer reaction-mass spectrometry (PTR-MS) was used. We found the following from the experimental results: (1) no byproducts were detected; (2) the bimolecular L
H kinetic model best described the catalytic reaction rate; (3) the activation energy of the oxidation was about 25.8 kJ mol
1; (4) TCO is most energy efficient at room temperature without auxiliary heating; (5) compared with photocatalytic oxidation (PCO) which needs ultraviolet light radiation, the reaction area of TCO can be much larger for a given volume so that TCO can perform much better not only in formaldehyde removal efficiency but also in energy saving.

’ INTRODUCTION Formaldehyde is one of the most prevalent pollutants in indoor air.1 It is harmful to human health, causing adverse respiratory effects and even cancer.2 The WHO guideline concentration for indoor formaldehyde is about 80 ppb (0.1 mg/m3).3 However, due to tight building and high emission materials, indoor formaldehyde level often exceeds this threshold and causes health problems.4 To remove formaldehyde from indoor air, several purification techniques such as physical adsorption,5 chemical adsorption,6 photocatalytic oxidation (PCO),7,8 plasma catalytic oxidation,9 and biotechnical decomposition 10 have been studied. However, there are some shortcomings which limit their practical indoor application: (1) adsorbents have limited capacities and need regular replacement; (2) harmful byproduct may be generated by PCO and plasma when purifying indoor chemicals;11,12 (3) the life cycle of microorganisms in biotechnical decomposition is still a problem which may affect the purification performance. Recently, thermal catalytic oxidation (TCO) by specific catalysts was reported to remove formaldehyde efficiently. TCO does not require ultraviolet (UV) light nor plasma, and the catalytic reaction occurs in the entire bulk of the catalyst rather than only on the surface activated by UV. Such advantages make it more suitable than other techniques for indoor formaldehyde removal. Various TCO catalysts have been investigated for their formaldehyde removal performance, including metal oxides,13 supported non-noble metals,14 and supported noble metals on metal oxides.15,16 Peng et al.14 investigated activities of a series of metals (Pt, Pd, Rh, Cu, Mn) supported on TiO2 for the catalytic r 2011 American Chemical Society

oxidation of formaldehyde. It was found that supported noble metals on metal oxides have much better low-temperature activeness, and the activities were correlated with the dispersion of the noble metal on the supporting material. Foster and Masel17 studied the formaldehyde oxidation rate on nickel oxide at 220 °C using the rate equation derived from Conner and Bennett’s mechanism. Imamura et al.18 found that Ru/CeO2 completely oxidized formaldehyde at 200 °C. The complete removal temperature for formaldehyde and methanol on Pd
manganese oxide catalyst was found to be about 88 °C.19 These studies showed high formaldehyde removal performance; however, their operating temperatures were much higher than the normal room-temperature range. Recently, the Pt/TiO2 catalyst by Zhang et al.20 and the Pt/ MnOx
CeO2 catalyst by Tang et al.21 were proved to be efficient for formaldehyde decomposition at room temperature. Formaldehyde of 102 ppm through the Pt/TiO2 catalyst was found to be totally oxidized into CO2.20 Tang et al. found that formaldehyde of 30 ppm was decomposed on Pt/MnOx
CeO2 completely into CO2 and H2O without deactivation over a 120 h reaction period21 and that the humidity has a negligible influence on the activity of a MnOx
CeO2 catalyst.22 However, there were few studies concerned with the reaction’s kinetic mechanism, the Received: March 24, 2011 Accepted: May 31, 2011 Revised: May 29, 2011 Published: June 13, 2011 5754

dx.doi.org/10.1021/es2009902 | Environ. Sci. Technol. 2011, 45, 5754–5760

Environmental Science & Technology

ARTICLE

Table 1. Physical Parameters of the Catalyst particle parameters

bed parameters

particle size (mesh)

40
60

weight (mg)

26.2

average pore diameter (nm)a

24.6

bed depth (mm)

4.2

BET area (m2/g)a particle porositya

60.6 0.48

bed diameter (mm) bulk density (kg m
3)

3 874.4

particle density (kg m
3)b

1287

a

Measured by the nitrogen adsorption BET method. b Measured by the mercury porosimeter method.

relevant parameters, and the influencing factors for TCO, especially for concentrations below 1 ppm.13,20,21 This limited the design and development of TCO air cleaners for removing indoor formaldehyde. The objectives of this work are to determine (1) the kinetic equation, the key parameters in it, and the factors influencing these parameters of a Pt/Mn
CeOx catalyst with concentrations of 280
3000 ppb in a temperature range 25
180 °C, (2) the performance and energy cost of the TCO technique, and (3) the advantages of applying TCO over PCO for indoor formaldehyde purification.

’ METHOD Catalyst. The catalyst Pt/MnOx
CeO2 was prepared by a conventional impregnation process.21 The ratio of Mn and Ce in the catalyst was 1:1. The platinum loading was about 1.0 wt %. Excess water was removed in a rotary evaporator at 50 °C until dry. The resulting sample was dried at 110 °C for 12 h and then further thermally processed at 400 °C for 4 h. The catalyst was characterized by pore structure analysis. The nitrogen adsorption isotherm was obtained at
196 °C on an analyzer for surface area and pore size, Quadrasorb SI (Quantachrome, USA). Macrostructure information was obtained by the mercury intruding porosimetry test on a mercury porosimeter AutoPore IV 9500 (Micromeritics, USA). The physical properties of the catalyst are listed in Table 1. Experimental Setup. The experiments were performed in a stainless steel once-through fixed-bed reactor (Figure S1 in the Supporting Information). The catalyst was fixed in the reactor by 100 mesh screens. The parameters of the reactor and catalyst bed are also shown in Table 1. The reactor was located in a steel chamber with temperature control in the range of 20
200 °C. The reaction temperature was measured by thermocouple and recorded using an HP 34970A data logger (Agilent, USA). Gaseous formaldehyde was generated by bubbling a formaldehyde solution using a nitrogen gas flow. The flow rate was controlled by a mass flow controller. The gaseous formaldehyde was diluted by synthetic air (O2 20 vol %, N2 balance). The total flow rate through the reactor was 2.5 L/min, and the gas hourly space velocity (GHSV) was 4.43  106 h
1. The inlet formaldehyde concentration was controlled by adjusting the flow ratio of formaldehyde gas. The concentration range was between 280 and 3000 ppb, which was a compromise for minimizing the test error and close to the indoor formaldehyde concentration level. The water vapor in the air flow was controlled at 7500 ppm by a water bubbling bottle. A standard type of PTR-MS (Ionicon Analytik, Austria) was used to measure the concentration of formaldehyde online. It was reported that the humidity has a significant influence on the

formaldehyde measurement by PTR-MS.23,24 However, formaldehyde quantification on PTR-MS is possible by determining the sensitivity as a function of the humidity.25 In this study, all reactions were at the same humidity, so the formaldehyde can be monitored by the PTR-MS with linear calibration. The m/z 31 signal measured by the PTR-MS was calibrated by the MBTH (3-methyl-2-benzothiazolinonehydrazone hydrochloride) spectrophotometry method, and the measurement accuracy of formaldehyde was 5%. Some other M þ 1 mass peaks such as m/z 21, 33, 39, 41, 43, 45, 46, 47, 93, 107 were also monitored during the formaldehyde oxidation process. A mass scan was made by the PTR-MS both at the inlet and at the outlet to examine possible gaseous byproduct. The scan range was from m/z 21 to m/z 120. Formation of CO2 was not examined in this study, because the concentrations were lower than the quantification limit (e.g., 3 ppm by a photoacoustic analyzer). Experimental Procedure. The catalyst was fixed in the reactor and assembled in the pipeline. The leakage was checked by a soap film flow meter GL-103A (Jiesidayi Analytic, China). The concentrations of the inlet and outlet were monitored by the PTR-MS. A steady-state condition was achieved after several minutes to several hours depending on the temperature. The final outlet concentration was determined under steady-state conditions. A fresh catalyst was used for each experiment. Data Analysis. The once-through formaldehyde conversion, ε, is written as follows ε¼

Cin
Cout Cin

ð1Þ

where Cin and Cout are the steady-state inlet and outlet formaldehyde concentrations, ppb, respectively. The average reaction rate was obtained from the mass balance in the reactor r ¼

GðCin
Cout Þ ABET

ð2Þ

where r was the average reaction rate per BET surface area, ppb m s
1, G the gas flow rate, l min
1, and ABET the total BET surface area, m2. The apparent reaction factor Kapp, m s
1, was defined as the reaction rate divided by the surface concentration r ð3Þ Kapp ¼ Cs Generally, the reaction takes place on the catalyst surface, so the mass transfer from bulk air to the surface needs to be considered.26 At the steady-state condition, the reaction rate in the reactor equals to the external mass transfer rate Kapp Cs ðxÞ ¼

hm As ΔCðxÞ ABET

ð4Þ

where hm was the mean external mass transfer coefficient, m s
1, ΔC the concentration difference between the bulk flow and the surface, ppb, As the total external surface area, m2, and x the location in flow direction, m. hm can be estimated by an empirical correlation.27 hm and Kapp were assumed as lumped parameters along the reactor. By deriving eq 4, the following equation was obtained Cs Cs ð0Þ Cs ðLÞ ¼ ¼ Cin
Cs ð0Þ Cout
Cs ðLÞ ΔC 5755

ð5Þ

dx.doi.org/10.1021/es2009902 |Environ. Sci. Technol. 2011, 45, 5754–5760

Environmental Science & Technology

ARTICLE

Figure 1. Once-through formaldehyde conversion at different temperatures with GHSV 4.43  106 h
1; O2 20 vol%, N2 balance; water vapor concentration 7500 ppm.

Figure 2. Apparent reaction coefficient in the reactor: impact of external mass transfer; 700 ppb inlet formaldehyde; O2 20 vol%, N2 balance; water vapor concentration 7500 ppm; reaction temperature 60 °C.

where ΔC was the logarithmic mean concentration difference between the bulk flow and the surface ΔC ¼

ðCin
Cs ð0ÞÞ
ðCout
Cs ðLÞÞ lnðCin
Cs ð0ÞÞ
lnðCout
Cs ðLÞÞ

ð6Þ

The average surface concentration was obtained by solving eqs 4
6 Cs ¼

Cin
Cout fm lnðCin Þ
lnðCout Þ 3

ð7Þ

where fm was the external mass transfer correction factor fm ¼

hm As hm As þ Kapp ABET

ð8Þ

The value of fm stands for the ratio of the reaction resistance in the series process of external mass transfer and the surface reaction. There are three unknown parameters (Kapp, fm, and Cs) in eqs 3, 7, and 8. By solving these three equations, Kapp and fm can be obtained. To avoid the influence of external mass transfer, a high flow rate condition is necessary, which results in fm being closer to 1. The reaction kinetics was studied by analyzing the variation of the reaction rate with the average surface concentration.

’ RESULTS Once-Through Formaldehyde Conversion. Figure 1 shows the results of ε at different temperatures. At inlet formaldehyde concentrations of 280
500 ppb, ε achieved 35.4%, 39.5%, 55.4%, 79.0%, and 96.3% at 25, 40, 60, 100, and 180 °C (GHSV 4.43  106 h
1), respectively. The formaldehyde conversion showed a significant decrease with decreasing temperature and increasing concentration. The mass scan result of PTRMS and some other monitored signals are shown in Figures S2
S4 of the Supporting Information. The results showed that no significant increases of any signals were detected by PTR-MS in the effluent air. Impact of External Mass Transfer. The impact of external mass transfer was studied by measuring the reaction rates at flow rates between 1.0 and 3.0 L/min. The measured apparent reaction coefficient and the external mass transfer correction factor are shown in Figure 2. From the results, fm was more than 0.95 when GHSV reached 4.43  106 h
1 (flow rate 2.5 L/min). On the basis

Figure 3. Reaction rate at different surface formaldehyde concentrations at temperatures of 25, 40, 60, 100, and 180 °C. GHSV 4.43  106 h
1; O2 20 vol%, N2 balance; water vapor concentration 7500 ppm: (points) experimental data; (solid lines) nonlinear regression by bimolecular L
H model.

of eq 7, the external mass transfer effect can be neglected, which was necessary to obtain the kinetic mechanism. Effect of Temperature and Concentration. The reaction rate is related to the surface formaldehyde concentration. Figure 3 shows the reaction rate of different surface formaldehyde concentrations under different reaction temperatures. The reaction rate increased with increasing temperature and increasing concentration (below ppm levels). As the formaldehyde concentration kept increasing, the reaction rate peaked and began to decrease. The concentration at the turning point was higher at higher temperatures. To determine the influence of the oxygen concentration on the oxidation reaction rate, the experiment was conducted using two different atmospheres with oxygen concentrations of 20% and 2% (by volume), respectively. The results show that the influence of oxygen concentration on the reaction rate was not very significant (Figure S5 of the Supporting Information). The decrease of oxygen concentration by a factor of 10 changes the reaction rate by less than 40%.

’ DISCUSSION Kinetic Study. The form of kinetic model for the reaction is determined by two factors: the rate-determining step and the 5756

dx.doi.org/10.1021/es2009902 |Environ. Sci. Technol. 2011, 45, 5754–5760

2.04  10
6

1.55  10
2

7.4  10
4 1.45  10
2

3.5  10
3 4.2  10
3

r = k0 Cs

gaseous formaldehyde reaction

formaldehyde and O2

reaction model

31

5 first-order

Krevelen mechanism

dissociation adsorbed formaldehyde reacted with adsorbed O2, with competitive adsorption

electronic balance between r = k0 Cs/1 þ k0 ACs

30

4 Mars
van

Hinshelwood mechanism

7

2 bimolecular Langmuir


Hinshelwood mechanism

3 dissociated adsorption Langmuir
r = kKCs1/2/(1 þ KCs1/2)2 Hinshelwood mechanism 29

O2, with competitive adsorption

adsorbed formaldehyde reacted with adsorbed r = kKCs/(1 þ KCs)2

no competitive adsorption

adsorbed formaldehyde reacted with O2, 1 unimolecular Langmuir


26

r = kKCs/1 þ KCs

principle and rate-determine step form of reaction rate kinetic model no.

Table 2. Experimental Data Fitting Using Different Kinetic Models

a Values were estimated at 25 °C. k is the reaction constant, ppb 3 m 3 s
1. K in unimolecular and bimolecular L
H mechanism, the adsorption constant, ppb
1. K in dissociated adsorption L
H mechanism, the adsorption constant, ppb
0.5. k0 is the reaction constant, m 3 s
1. A is the constant of the Mars
van Krevelen mechanism, ppb
1 3 m
1 3 s. b Mean fitting result of five temperatures.

3.79  10
6


0.30

0.83 239.8 1.44  10
5

1.19  10
2 (ppb
0.5)

(ppb )

k (ppb 3 m 3 s )

ARTICLE

a


1

4.42  10
7

0.73 9.91  10
6

0.92 3.07  10
7

0.83 4.42  10
7

2 2
2 2b 3 s) (ppb 3 m 3 s ) R

3m (m 3 s ) (ppb


1
1

a

k0 Ka


1

Aa


1

variance b

Environmental Science & Technology

Table 3. Parameters of Formaldehyde Catalytic Oxidation in Bimolecular L
H Form temperature (°C)

k (ppb 3 m 3 s
1)

25

(1.45 ( 0.14)  10
2

(7.4 ( 1.8)  10
4

40 60


2

(1.82 ( 0.13)  10 (3.32 ( 0.14)  10
2

(7.6 ( 1.5)  10
4 (4.7 ( 0.5)  10
4

100

(9.53 ( 0.37)  10
2

(3.1 ( 0.2)  10
4

180

K (ppb
1)

3.66 ( 14.58

(2.0 ( 7.0)  10
5

phase of the reaction compounds in that step. A recent study on Au/Co3O4
CeO2 catalyst supposed a reaction route of formaldehyde oxidation,28 where HCOOH was found to be an intermediate on the catalyst surface during oxidation of formaldehyde into CO2. According to our result, no gaseous HCOOH yields (m/z 47) were detected (Figures S2
S4 of the Supporting Information). It indicated that the rate of HCOOH further oxidizing into CO2 is higher than that of its generation. This may due to a low product generation rate caused by a low inlet concentration. At this condition the effect of product desorption rate also can be ignored. In addition, according to Figure 3, the variation of the reaction rate with the formaldehyde concentration is significant, which indicates that the rate-determining step is related to the formaldehyde reaction. To determine the reaction compounds’ phase in the ratedetermining step, several kinetic models7,26,29
31 (See Table 2 for the details) were assumed for the reaction. The experimental data were fit into those models to regress the kinetic constants at each reaction temperature. The fitting results are also shown in Table 2. From these results, it is apparent that the model with bimolecular L
H kinetics, based on the competitive adsorption mechanism, provided the best fit to the data (with the highest R2 value). The fitting parameters for bimolecular L
H kinetics are shown in Table 3. The bimolecular L
H kinetics implies that two reactants in the rate-determining step compete with one another for adsorption sites. According to the reaction mechanism, the reactants were probably formaldehyde and oxygen (both in adsorbed phase). Tang et al. reported the importance of the surface active oxygen in the reaction.21 In addition, from the experimental result (Figure S5 of the Supporting Information), there was little impact of gaseous oxygen concentration on the overall reaction rate. It indicated the surface active oxygen in the rate-determining step was not directly from the air oxygen but from the metal oxide. CeO2 is reported as a good oxygen storage material, which may help to store and provide the surface active oxygen.32 Under conditions of high formaldehyde concentration, adsorbed formaldehyde prevented surface active oxygen from transferring to the adsorption sites, so that the reaction rate decreased. At the lower formaldehyde concentrations typically encountered in indoor environments (usually lower than 500 ppb), the effect of competitive adsorption appears to be negligible (Figure 3) and the model also can be reduced to a unimolecular L
H model by neglecting the square term of concentration in the denominator. Temperature Impact. The temperature dependence of a specific reaction coefficient and adsorption coefficient can be described by the Arrhenius equation ln kArr ¼ ln AArr
E=RT

ð9Þ

where AArr is frequency factors for the specific reaction coefficient, same units as kArr, E the activation energy, J mol
1, T the 5757

dx.doi.org/10.1021/es2009902 |Environ. Sci. Technol. 2011, 45, 5754–5760

Environmental Science & Technology

ARTICLE

Figure 4. Predicted versus experimental reaction rates for the catalytic oxidation of formaldehyde.

absolute temperature, K, and R the universal gas constant, J mol
1 K
1 . By applying the Arrhenius equation to correlate the reaction coefficient and adsorption coefficient at different temperatures, the bimolecular L
H kinetics were expressed as    
E1
E2 k0 exp K0 exp Cs RT RT r ¼  ð10Þ   2
E2 1 þ K0 exp Cs RT where k0 and K0 are frequency factors for the reaction coefficient and the adsorption coefficient, ppb m s
1 and ppb
1, respectively, E1 is the activation energy of the reaction, J mol
1, and E2 is the activation energy of adsorption, J mol
1. The activation energy is the minimum energy required to overcome the potential barrier to start the reaction. In a thermal catalytic oxidation reaction, the energy of the reactant molecule is expressed as the reaction temperature. The activation energy reflects the influence of temperature on the reaction rate. When the activation energy is higher, the influence of temperature is larger. The parameter values and the fitting results are shown in Figure 4. The coefficient of determination R2 is 0.97, which indicates the model fit well with the experimental data. The apparent activation energy of oxidation was estimated as 25.8 kJ mol
1. It is clear that pollutant-removal efficiency and energy costs are taken into account in practical air cleaners. Through eq 10, the performance and energy cost of the TCO technique can be evaluated when it is used in a practical honeycomb-type reactor. A simulated honeycomb reactor of 100 000 channels, channel size 1 mm  1 mm, length 2 cm, catalyst coating thickness 10 μm, and flow rate 360 m3/h, was used in this calculation. The reactor was assumed to be thermally insulated, so the auxiliary heating energy was equal to the heating energy removed by the air flow. A heat and mass transfer correlation for honeycomb reactors33 was applied for the calculation. The environment temperature was 20 °C, and the formaldehyde concentration was 200 ppb. The performance of an air reactor is evaluated by its clean air delivery rate (CADR), which is defined as the volume of fresh air generated per hour.34 CADR can be expressed by the following equation CADR ¼ εG

ð11Þ

Figure 5. Influence of temperature on clean air delivery rate (CADR) and clean air delivery rate per power input (CADR/P). Honeycomb reactor with 100 000 channels, length 2 cm, channel size 1 mm  1 mm, catalyst thickness 10 μm, flow rate 360 m3/h, formaldehyde 200 ppb, environmental temperature 20 °C.

The clean air delivery rate per power input (CADR/P) has been considered as part of the present analysis to evaluate the energy consumption of air purifiers with a certain CADR, where P is the energy consumption rate, including fan power and any auxiliary heating rate. The energy of auxiliary heating is zero when the reaction occurs at room temperature. Figure 5 shows that the CADR value of this reactor increases with temperature almost linearly, while CADR/P decreases with temperature exponentially. Thus, the catalyst needs much more energy input to increase the same amount of output at a higher temperature. In a real indoor environment, when the reactor is connected to an indoor heating device (such as the heating radiation surface or system), the heating can help to increase the efficiency of the catalyst; otherwise, it needs extra heating energy to first heat the air (or heat the catalyst) and then cool the air. In addition, based on the kinetic parameters from Figure 4, such a simulated reactor has a CADR of about 118 m3/h at room temperature, which shows good performance compared with other commonly used indoor air cleaners (1.4
40 m3/h).35 Thus, operating at room temperature without auxiliary heating is the favorable mode for catalyst operation. Comparison of TCO and PCO for Indoor Air Purification. Catalytic oxidation reactors are suitable for solving chronic emission problems such as formaldehyde in indoor air since they are expected to have a longer effective life than various adsorption methods. To predict the performance of TCO in indoor air applications, a comparison was made between TCO and PCO for removal of formaldehyde. Since the reaction with PCO also has bimolecular L
H kinetics, the parameters in the equation are comparable. For the conditions of temperature at 25 °C, a water vapor concentration of 7500 ppm, and a UV intensity of 330 μW/cm2, k and K of PCO were estimated to be 138 ppb m s
1 and 4.2  10
4 ppb
1, respectively.7 The value of k represents the reaction rate per reaction area. The result indicates that k for PCO was 4 orders of magnitude larger than that for TCO (Table 3). However, these k’s have different respective reaction areas. The reaction area of PCO is limited to its irradiated surface (about 0.1 m2/g),36 while the reaction area of TCO is the total BET surface (60.6 m2/g). Accounting for the reaction area, by coating more catalyst to the surface within reasonable thickness limits, the 5758

dx.doi.org/10.1021/es2009902 |Environ. Sci. Technol. 2011, 45, 5754–5760

Environmental Science & Technology reaction rate of TCO would be larger than PCO. The value of the equivalent adsorption constant K represents the adsorption ability of the catalyst. The K of TCO is about 1.8 times that of PCO, so it tends to perform better than PCO at low formaldehyde concentration. Although the cost will be greater when using noble metals in the TCO catalyst, it can be made acceptable by distributing the catalyst more efficiently in the reactors. The good performance shown by the honeycomb reactor calculated above (Figure 5) only needs about 100 g of the TCO catalysts (including 1 g Pt). This is not too costly and is economically feasible when compared with other air cleaners. In real indoor environments, the influence of humidity, particles, and other VOCs on the performance of the TCO needs to be considered. In addition, more real operational data are required to compare with other air purification methods such as PCO. This is an interesting topic for further study. However, the present analysis is an important first step in evaluating the suitability of thermal catalytic oxidation for formaldehyde removal from indoor air.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional details on the experimental setup (Figure S1), mass scan result of PTR-MS (Figures S2
S4), and influence of oxygen concentration (Figure S5). This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: þ86 10 6277 2518; fax: þ86 10 6277 3461; e-mail: [email protected].

’ ACKNOWLEDGMENT We acknowledge Dr. Shoufang Kang for his work preparing the catalyst and Professor Charles J. Weschler for suggestions on the manuscript. This work was supported by the National Nature Science Foundation of China (Grant Nos. 50725620, 51006057) and project of State Key Laboratory of Subtropical Building Science (Grant No. 2008KA08). ’ REFERENCES (1) Salthammer, T.; Mentese, S.; Marutzky, R. Formaldehyde in the Indoor Environment. Chem. Rev. 2010, 110 (4), 2536–2572. (2) International Agency for Research on Cancer (IARC). Formaldehyde, 2-Butoxyethanol and 1-tert-Butoxy-2-propanol. IARC Monographs of the evaluation of Carcinogenic Risks to Humans; World Health Organization: Lyon, France, 2006; Vol. 88. (3) World Health Organization (WHO). Air quality guidelines for Europe, Air quality guidelines for Europe, 2nd ed.; European Series No. 91; WHO Regional Publications: 2000 (http://www.euro.who.int/en/ what-we-do/health-topics/environment-and-health/air-quality/publications/ pre2009/air-quality-guidelines-for-europe). (4) Zhang, L. Z.; Niu, J. L. Modeling VOCs emissions in a room with a single-zone multi-component multi-layer technique. Build. Environ. 2004, 39 (5), 523–531. (5) Fang, L.; Zhang, G.; Wisthaler, A. Desiccant wheels as gas-phase absorption (GPA) air cleaners: evaluation by PTR-MS and sensory assessment. Indoor Air 2008, 18 (5), 375–385.

ARTICLE

(6) Saeung, S.; Boonamnuayvitaya, V. Adsorption of formaldehyde vapor by amine-functionalized mesoporous silica materials. J. Environ. Sci. (China) 2008, 20 (3), 379–384. (7) Obee, T. N. Photooxidation of sub-parts-per-million toluene and formaldehyde levels an titania using a glass-plate reactor. Environ. Sci. Technol. 1996, 30 (12), 3578–3584. (8) Wang, H. Q.; Wu, Z. B.; Zhao, W. R.; Guan, B. H. Photocatalytic oxidation of nitrogen oxides using TiO2 loading on woven glass fabric. Chemosphere 2007, 66 (1), 185–190. (9) Park, J. H.; Byeon, J. H.; Yoon, K. Y.; Hwang, J. Lab-scale test of a ventilation system including a dielectric barrier discharger and UVphotocatalyst filters for simultaneous removal of gaseous and particulate contaminants. Indoor Air 2008, 18 (1), 44–50. (10) Chen, W. H.; Zhang, J. S. S. Zhang, Z. B. Ashrae Performance of air cleaners for removing multiple volatile organic compounds in indoor air. ASHRAE Trans. 2005, 111, 1101–1114. (11) Mo, J. H.; Zhang, Y. P.; Xu, Q. J.; Zhu, Y. F.; Lamson, J. J.; Zhao, R. Y. Determination and risk assessment of by-products resulting from photocatalytic oxidation of toluene. Appl. Catal., B 2009, 89 (3
4), 570–576. (12) Ding, H. X.; Zhu, A. M.; Yang, X. F.; Li, C. H.; Xu, Y. Removal of formaldehyde from gas streams via packed-bed dielectric barrier discharge plasmas. J. Phys. D: Appl. Phys. 2005, 38 (23), 4160–4167. (13) Sekine, Y. Oxidative decomposition of formaldehyde by metal oxides at room temperature. Atmos. Environ. 2002, 36 (35), 5543–5547. (14) Peng, J. X.; Wang, S. D. Performance and characterization of supported metal catalysts for complete oxidation of formaldehyde at low temperatures. Appl. Catal., B 2007, 73 (3
4), 282–291. (15) Zhang, C. B.; He, H.; Tanaka, K. Catalytic performance and mechanism of a Pt/TiO2 catalyst for the oxidation of formaldehyde at room temperature. Appl. Catal., B 2006, 65 (1
2), 37–43. (16) Shen, Y. N.; Yang, X. Z.; Wang, Y. Z.; Zhang, Y. B.; Zhu, H. Y.; Gao, L.; Jia, M. L. The states of gold species in CeO2 supported gold catalyst for formaldehyde oxidation. Appl. Catal., B 2008, 79 (2), 142–148. (17) Foster, J. J.; Masel, R. I. Formaldehyde oxidation on nickeloxide. Ind. Eng. Chem. Res. 1986, 25 (4), 563–568. (18) Imamura, S.; Uematsu, Y.; Utani, K.; Ito, T. Combustion of formaldehyde on ruthenium cerium(Iv) oxide catalyst. Ind. Eng. Chem. Res. 1991, 30 (1), 18–21. (19) O’Shea, V.; Alvarez-Galvan, M. C.; Fierro, J. L. G.; Arias, P. L. Influence of feed composition on the activity of Mn and PdMn/Al2O3 catalysts for combustion of formaldehyde/methanol. Appl. Catal., B 2005, 57 (3), 191–199. (20) Zhang, C. B.; He, H.; Tanaka, K. Perfect catalytic oxidation of formaldehyde over a Pt/TiO2 catalyst at room temperature. Catal. Commun. 2005, 6 (3), 211–214. (21) Tang, X. F.; Chen, J. L.; Huang, X. M.; Xu, Y.; Shen, W. J. Pt/ MnOx-CeO2 catalysts for the complete oxidation of formaldehyde at ambient temperature. Appl. Catal., B 2008, 81 (1
2), 115–121. (22) Tang, X. F.; Li, Y. G.; Huang, X. M.; Xu, Y. D.; Zhu, H. Q.; Wang, J. G.; Shen, W. J. MnOx-CeO2 mixed oxide catalysts for complete oxidation of formaldehyde: Effect of preparation method and calcination temperature. Appl. Catal., B 2006, 62 (3
4), 265–273. (23) Hansel, A.; Singer, W.; Wisthaler, A.; Schwarzmann, M.; Lindinger, W. Energy dependencies of the proton transfer reactions H3Oþ þ CH2O f CH2OHþ þ H2O. Int. J. Mass Spectrom. Ion Processes 1997, 167
168, 697–703. (24) de Gouw, J.; Warneke, C. Measurements of volatile organic compounds in the earth’s atmosphere using proton-transfer-reaction mass spectrometry. Mass Spectrom. Rev. 2007, 26 (2), 223–257. (25) Herbig, J.; Seger, M.; Kohl, I.; Mayramhof, G.; Titzmann, T.; Preinfalk, A.; Winkler, K.; Dunkl, J.; Pfeifer, B.; Baumgartner, C.; Hansel, A. Online breath sampling with PTR-MS-A setup for large screening studies. In Proceedings of 4th International Conference on Proton Transfer Reaction Mass Spectrometry and its Applications, Obergurgl, Austria, 2009; pp 46
50. 5759

dx.doi.org/10.1021/es2009902 |Environ. Sci. Technol. 2011, 45, 5754–5760

Environmental Science & Technology

ARTICLE

(26) Yang, R.; Zhang, Y.; Xu, Q.; Mo, J. A mass transfer based method for measuring the reaction coefficients of a photocatalyst. Atmos. Environ. 2007, 41 (6), 1221–1229. (27) Wakao, N.; Funazkri, T. Effect of fluid dispersion coefficients on particle-to-fluid mass-transfer coefficients in packed-beds - correlation of Sherwood numbers. Chem. Eng. Sci. 1978, 33 (10), 1375–1384. (28) Ma, C. Y.; Wang, D. H.; Xue, W. J.; Dou, B. J.; Wang, H. L.; Hao, Z. P. Investigation of Formaldehyde Oxidation over Co3O4-CeO2 and Au/Co3O4-CeO2 Catalysts at Room Temperature: Effective Removal and Determination of Reaction Mechanism. Environ. Sci. Technol. 2011, 45 (8), 3628–3634. (29) Zhou, L. B.; Akgerman, A. Catalytic-oxidation of ethanol and acetaldehyde in supercritical carbon-dioxide. Ind. Eng. Chem. Res. 1995, 34 (5), 1588–1595. (30) Chuang, K. T.; Zhou, B.; Tong, S. M. Kinetics and mechanism of catalytic-oxidation of formaldehyde over hydrophobic catalysts. Ind. Eng. Chem. Res. 1994, 33 (7), 1680–1686. (31) Zhang, Y. P.; Yang, R.; Zhao, R. Y. A model for analyzing the performance of photocatalytic air cleaner in removing volatile organic compounds. Atmos. Environ. 2003, 37 (24), 3395–3399. (32) Campbell, C. T.; Peden, C. H. F. Chemistry - Oxygen vacancies and catalysis on ceria surfaces. Science 2005, 309 (5735), 713–714. (33) Hawthorn, R. D. Afterburner catalysts - effects of heat and mass transfer between gas and catalyst surface. AIChE Symp. Ser. 1974, 70 (137), 428–438. (34) ANSI/AHAM AC-1988, Portable household electric cordconnected room air cleaners. In Association of home appliance manufacturers: Chicago, 1988. (35) Chen, W. H.; Zhang, J. S.; Zhang, Z. B. Performance of air cleaners for removing multiple volatile organic compounds in indoor air. ASHRAE Trans. 2005, 111, 1101–1114. (36) Mo, J. H.; Zhang, Y. P.; Xu, Q. J.; Lamson, J. J.; Zhao, R. Y. Photocatalytic purification of volatile organic compounds in indoor air: A literature review. Atmos. Environ. 2009, 43 (14), 2229–2246.

5760

dx.doi.org/10.1021/es2009902 |Environ. Sci. Technol. 2011, 45, 5754–5760