Photocatalytic Decomposition of Gas-Phase Chlorobenzene with Transition Metal–Doped KLaTi 2 O 6 Under Visible Light Irradiation

Abstract

Photocatalytic degradation of volatile organic compounds under visible light is an important issue for use of solar energy. Perovskite-type oxide, KLaTi₂O₆, was prepared by the sol-gel method. Photocatalytic activities of catalysts for chlorobenzene decomposition were investigated. Four types of transition metal (Co, Ni, Zn, Mn) were doped into KLaTi₂O₆ to improve photocatalytic activity for gas-phase chlorobenzene degradation. Photocatalysts were characterized by X-ray diffraction, energy-dispersive X-ray spectroscopy, ultraviolet–visible spectroscopy, and nitrogen adsorption analysis. Results showed that under visible light irradiation, the photocatalytic activity of metal-doped KLaTi₂O₆ was higher than undoped KLaTi₂O₆. We found that the predominant species accumulating on the catalyst surface is benzaldehyde. This report proposes the pathway of benzaldehyde formation during photocatalytic degradation of chlorobenzene under visible light irradiation.

Introduction

Volatile organic compounds (VOCs) are considered to be significant atmospheric pollutants since they are malodorous, toxic, carcinogenic, and mutagenic (Wang et al., 2008). Among the wide variety of VOCs, halogenated VOCs are refractory organics. Ultraviolet (UV) photocatalysis is widely applicable for degradation of these organic compounds (Macwan et al., 2011; Park et al., 2011). However, visible light accounts for 45% of energy from solar radiation, whereas UV light is less than 3–4% (Hsieh et al., 2010). The widely used photocatalyst TiO₂ only absorbs UV light due to its relatively wide band gap. Therefore, developing new types of photocatalysts with a visible light response attracts increasing interests.

Recently, perovskite-type oxides for photocatalytic performance have been studied intensively (Wu et al., 2007; Osterloh, 2008). However, most reports focused on splitting water to produce hydrogen; there are a few reports on photocatalytic degradation of halogenated VOCs, especially under visible light irradiation.

In this study, we synthesized solid solution KLaTi₂O₆ and metal-doped KLaTi₂O₆. Photocatalytic performance for degradation of chlorobenzene was investigated under visible light irradiation. We also discuss the intermediates generated on the catalyst surface and analyze the mechanism of chlorobenzene photodegradation by gas chromatography/mass spectroscopy (GC-MS) and in situ Fourier-transform infrared (FTIR) spectroscopy.

Experimental Details

Preparation of photocatalysts

KLaTi₂O₆ powders were prepared by the sol-gel process. Tetrabutyl titanate (0.2 mol) and citric acid (0.3 mol) were thoroughly mixed with severe agitation. KNO₃ (0.3 mol) and La(NO₃)₃ (0.1 mol) were dissolved in the above solution. The solution pH 1 was adjusted by nitric acid. The homogeneous and transparent sol obtained after reaction for 2 h was dried at 80°C. After grinding, the resultant gel was calcined at 300°C for 2 h and recalcined at 800°C for 5 h. The powders gained were KLaTi₂O₆.

Four types of aqueous solution containing 0.2 M metal ions [Zn(CH₃COO)₂·2H₂O, Co(CH₃COO)₂·4H₂O, NiSO₄·7H₂O, and MnSO₄·H₂O] were prepared for the doping. The parent KLaTi₂O₆ powders (1 g) were added into each of the above aqueous solutions (100 mL) and the white color of the powder changed instantaneously to other colors (Fig. 1). The solutions were maintained at 60°C for 3 days. The samples after washing with distilled water to remove \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $${ \rm SO}_4^{2 -}$$ \end{document} anions were dried at room temperature and calcined at 300°C for 2 h to remove organics. The samples obtained were referred to as metal-doped KLaTi₂O₆ (metal=Zn, Co, Ni, and Mn).

FIG. 1.

(a) KLaTi₂O₆ and Zn(CH₃COO)₂·2H₂O solution; (b) KLaTi₂O₆ and Co(CH₃COO)₂·4H₂O solution; (c) KLaTi₂O₆ and NiSO₄·7H₂O solution; (d) KLaTi₂O₆ and MnSO₄·H₂O solution.

Characterization

X-ray diffraction (XRD) patterns of the powders were obtained on a Rigaku D/max 2500 X-ray diffraction analyzer. The crystallite size can be calculated by the following equation (Gu et al., 2004): \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} \frac { \beta \cos \theta } { \lambda } = \frac { 1 } { L } + \frac { \eta \sin \theta } { \lambda } \tag { 1 } \end{align*} \end{document} where λ, θ, and β are the X-ray wavelength (0.15418 nm for Cu-Ka), Bragg diffraction angle, and full width at half maximum (FWHM) attributable to lattice strain and crystallite size, respectively. The crystallite size L and lattice strain η can be obtained from the y-intercept and the gradient of the approximation line, respectively. UV–visible absorption spectra of samples were collected using Shimadzu Corporation UV-2540. The specific surface area and pore size distribution of samples were measured by a BELSOROP-MINI II apparatus. Metal contents of the samples were measured with the EDX spectroscopy (Hitachi S4700).

Photocatalytic experiments

The photocatalytic reaction was carried out in a cylindroid glass reactor using a 350 W xenon lamp (ShenZhen AnHongDa Opto Technology Co., Ltd.) along with a filter to cut down infrared (IR) and UV radiation to permit only visible light (400 nm≤λ≤700 nm) radiation transmission (Fig. 2). A large cuboid box was used to avoid interference from other light sources. A constant temperature was maintained with a cooled water system outside the reactor (not shown). A typical experiment was carried out as follows: A suspension containing 50 g/L of catalyst in distilled water was prepared. The surface of a frosted glass plate (13.5 cm²) was cleaned with acetone. The suspension was poured on the glass plate and was then dried at 50°C. The deposition process was performed four times to obtain a catalyst loading density of 1.4×10⁻² g/cm², as estimated by weighting the plate before and after the deposition process. The plate was placed at the center of the 250 mL reactor. Next, 0.5 μL of chlorobenzene liquid was injected into the sealed reactor filled with air and allowed to evaporate with a small magnetic stirring fan. Illumination was initialized after adsorption equilibrium was reached. The small fan guaranteed the gas circulation inside the reactor. A small amount of gas (500 μL) was withdrawn from the reactor periodically and analyzed by gas chromatography. The photocatalytic activity of samples was evaluated by measuring the degradation ratio of chlorobenzene. \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}Dr = \frac { C_o - C_t } { C_o } \times 100 \% \tag { 2 } \end{align*} \end{document} where Dr is the degradation ratio of chlorobenzene, C_o is the initial concentration of chlorobenzene, C_t is the concentration of chlorobenzene at time t, t is the irradiation time.

FIG. 2.

Device for photocatalytic reaction.

Analytical methods

Identification of products was verified using GC-MS and in situ FTIR spectroscopy.

For GC-MS analysis, 18.4 μL of chlorobenzene liquid was injected into the reactor and the other experiment conditions were the same as those in photocatalytic experiments. After irradiation, the intermediates adsorbed on the photocatalyst were extracted with methanol. The extract was filtered through a 0.45-μm filter and stored at 4°C until instrumental analysis was conducted. The GC-MS (Agilent 7890A-5975C) was equipped with a DB-5 ms capillary column (30 m, 0.25 mm inner diameter, 0.25 μm film thickness) and electron ionization (EI) detector. Mass spectra were compared with the NIST database for compound identification.

For the in situ FTIR analysis, the quartz IR cell with IR-transparent CaF₂ windows was used as the reaction cell. The catalyst was pressed into a pellet and placed at the center of the cell. Then, it was pretreated in N₂+10% O₂ stream at 400°C. After the catalyst sample was cooled to 25°C, chlorobenzene gas was introduced to the cell for chlorobenzene adsorption. Subsequently, a photoreaction was performed under visible light irradiation in a closed system. All infrared spectra were collected using Nicolet 380 with a resolution of 4 cm⁻¹ and 32 scans in the region of 4000–1000 cm⁻¹.

Results and Discussion

Characterization

Figure 3 and Table 1 show the XRD patterns and lattice parameters of the powders. It was found that the powder XRD pattern of the undoped sample as shown in Fig. 3 agrees well with the KLaTi₂O₆ crystal phase (JCPDS No. 01-089-4930). According to the X-ray diffraction data, the lattice parameters of KLaTi₂O₆ were found to be a=b=c=0.3897 nm and the crystallite size was 9.7 nm.

FIG. 3.

X-ray diffraction patterns of (1) undoped KLaTi₂O₆, (2) Ni-doped KLaTi₂O₆, (3) Zn-doped KLaTi₂O₆, (4) Mn-doped KLaTi₂O₆, and (5) Co-doped KLaTi₂O₆.

Table 1.

Influence of Doped Metal Ions on the Lattice Parameters of KLaTi₂O₆ Photocatalyst

	Catalyst type
	Undoped	Ni	Zn	Mn	Co
Lattice parameter a, b, c (nm)	0.3897	0.3894	0.3908	0.3911	0.3900
Crystallite size (nm)	9.7	5.7	6.2	8.7	6.2

Figure 3 and Table 1 also show that metal doping does not change the crystal structure and all the samples consist of single phase KLaTi₂O₆. However, metal doping changes the lattice parameters of KLaTi₂O₆, indicating that metal ions are inserted into the crystal of KLaTi₂O₆, which is consistent with the published literature (Yang et al., 2009a, 2009b). The metal-doped samples exhibit broadening peaks and the peak intensity decreases, which indicate a decrease in the crystallinity due to metal doping (Kumar et al., 2011).

From the quantitative analysis of EDX, the metal contents were obtained by averaging the contents of five random points on the sample. The metal contents for metal-doped samples are estimated at ∼2–3 at% (Table 2). Pore size distribution determined by the BJH model shows that it is broader in metal-doped samples than undoped KLaTi₂O₆ (Fig. 4). The average pore size for all samples is ∼5 nm. However, the specific surface area of metal-doped samples increases by ∼20 m²/g over undoped KLaTi₂O₆, ranging from 96 to 102 m²/g (Table 2).

FIG. 4.

BJH pore size distribution of samples.

Table 2.

Specific Surface Area, Band Gap, and Metal Content of Samples

	Catalyst type
	Undoped	Ni	Zn	Mn	Co
Surface area (m²/g)	77	99	96	102	96
Band gap (eV)	3.39	3.36	3.30	2.91	2.89
Metal content (at%)	0	2.47	2.64	2.35	2.67

The UV-visible absorption spectra of the samples are shown in Fig. 5. In the present study, the sharp UV absorption of undoped KLaTi₂O₆ is identified near 360 nm. However, metal doping causes a red shift of the absorption edge and the metal-doped photocatalysts show different absorption behaviors in visible light range from 400 to 700 nm. The undoped KLaTi₂O₆ is white, while four different colors appear for metal-doped samples. For a semiconductor, the optical absorption near the band edge follows the equation (Huang et al., 2010): \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}ahv = A ( hv - Eg ) ^{n / 2} \tag{3}\end{align*} \end{document} where a, v, Eg, and A are absorption coefficient, light frequency, band gap, and a constant, respectively.

FIG. 5.

Ultraviolet–visible light absorption spectra of different samples.

The value of n for the direct band gap material should be fixed at 1 and the experimental band gaps of the samples are listed in Table 2. The band gaps narrowed by metal doping ranged 2.89–3.36 eV versus 3.39 eV for undoped KLaTi₂O₆. This means that the metal-doped samples can obtain more photons to excite the electron from the valence band to the conduction band (Huang et al., 2010). The optical band gaps are found to follow the order: undoped KLaTi₂O₆>Ni-doped KLaTi₂O₆>Zn-doped KLaTi₂O₆>Mn-doped KLaTi₂O₆>Co-doped KLaTi₂O₆.

Photocatalytic degradation of chlorobenzene

Figure 6 shows the degradation ratio of chlorobenzene for the samples, along with the photolysis of chlorobenzene without a catalyst under visible light irradiation. It is evident that the chlorobenzene photodecomposition is negligible without the catalyst. Metal doping can improve the photoactivity of KLaTi₂O₆ and the decomposition ratio order of chlorobenzene is Co-doped KLaTi₂O₆>Mn-doped KLaTi₂O₆>Zn-doped KLaTi₂O₆>Ni-doped KLaTi₂O₆. The optimum doping metal is found to be Co and the degradation ratio is more than 50%. This means that metal-doped KLaTi₂O₆ is excited more effectively than undoped KLaTi₂O₆ under visible light irradiation, which might be attributed to the formation of a new localized energy level (Yang et al., 2009a, 2009b) and increase of specific surface area (Li et al., 2011) (Table 2).

FIG. 6.

Comparison of photocatalytic activities under visible light irradiation (1) without a catalyst, (2) with undoped KLaTi₂O₆, (3) with Ni-doped KLaTi₂O₆, (4) with Zn-doped KLaTi₂O₆, (5) with Mn-doped KLaTi₂O₆, and (6) with Co-doped KLaTi₂O₆.

Photocatalytic decomposition pathway of chlorobenzene

Figure 7 illustrates the GC-MS analysis for chlorobenzene decomposition with Zn-doped KLaTi₂O₆. The five dominant peaks at 7.98, 8.15, 8.45, 12.15, and 8.70–9.10 min are detected, and the associated mass spectra (Fig. 8) identify the corresponding intermediates as benzaldehyde, o-chlorophenol, m-chlorophenol, p-chlorophenol, and dichlorobenzene, respectively. Further confirmation was made by comparing the retention times of the corresponding pure samples.

FIG. 7.

Gas chromatogram of filter liquid after the photocatalytic decomposition of high concentration of chlorobenzene with Zn-doped KLaTi₂O₆.

FIG. 8.

Mass spectra of the corresponding intermediates.

To investigate the source of carboxyl group of benzaldehyde, in situ FTIR was utilized during the photo-oxidation process in the absence of CO₂. The results are shown in Fig. 9. The IR spectrum displays the characteristic bands of Zn-doped KLaTi₂O₆ [(1) in Fig. 9]. After chlorobenzene is introduced, the new bands at 3085, 1586, 1481, 1120, 1085, and 1016 cm⁻¹ appear. They are the same as those characteristic vibration bands for chlorobenzene in the published literature (Zhou et al., 1993; Liu et al., 2002) (Table 3).

FIG. 9.

Infrared spectra of (1) Zn-doped KLaTi₂O₆ photocatalyst before adsorption of chlorobenzene, (2) chlorobenzene adsorbed on catalyst before visible light irradiation, and (3) chlorobenzene adsorbed on catalyst after exposure to visible light for 180 min. The spectra of the reaction cell full of N₂+10% O₂ are utilized as the background.

Table 3.

Fundamental Vibrational Frequencies of Chlorobenzene on the Zn-Doped KLaTi₂O₆ Photocatalyst and Their Assignments

No.	Freq. (cm⁻¹)	Assignment
1	3085	Ar-H stretching
2	1586	Aromatic ring skeletal vibration
3	1481	C-H rocking
4	1120	C-H bending
5	1085	C-Cl stretching
6	1016	Ring deformation

Upon irradiation, a slight decrease in the chlorobenzene bands at 3085, 1120, and 1085 cm⁻¹ is observed and the band at 1016 cm⁻¹ almost disappears, while new weak bands appear at 1740, 1680, 1648, 1540, 1515, and 1417 cm⁻¹. The bands at 1680, 1648, 1586, and 1515 cm⁻¹ prove the presence of benzaldehyde (Augugliaro et al., 1999; Li et al., 2011). Bands at 1740 and 1680 cm⁻¹ are characteristic of compounds containing the C=O group (Li et al., 2003). The bands at 1740, 1540, and 1417 cm⁻¹ show the formation of formaldehyde (Lochař, 2006; Sun et al., 2010). FTIR analysis shows that benzaldehyde and formaldehyde exist during the photocatalytic process. Obviously, the carboxyl group of benzaldehyde could come from formaldehyde formed by aromatic ring cleavage during the photo-oxidation of chlorobenzene.

A distinct pathway might occur during the photocatalytic degradation of chlorobenzene, which is the formation of benzaldehyde. Some chlorobenzene molecules are first cleaved, then dechlorinated, and ultimately converted to simple organic acid and aldehyde. Others are first dechlorinated and then converted to benzaldehyde by reaction with aldehyde radicals formed earlier. The pathway is depicted in Fig. 10.

FIG. 10.

Proposed degradation process of chlorobenzene.

Although benzaldehyde and other byproducts adsorbed on the catalysts were detected, no byproducts were detected by GC-MS in the gas phase during any of the experiments.

Conclusions

In summary, metal-doped KLaTi₂O₆ photocatalysts were successfully prepared. The Co-doped KLaTi₂O₆ photocatalyst showed a significantly higher photocatalytic activity for chlorobenzene degradation under visible light irradiation. GC-MS analysis and in situ FTIR analysis indicated that benzaldehyde was the predominant species during the photocatalytic degradation of gas-phase chlorobenzene under visible light irradiation.

Footnotes

Acknowledgment

This research was financially supported by “the Double Hundred Plan.”

Author Disclosure Statement

No competing financial interests exist.

References

Augugliaro

, Coluccia

, Loddo

, Marchese

, Martra

, Palmisano

, and Schiavello

(1999). Photocatalytic oxidation of gaseous toluene on anatase TiO₂ catalyst: Mechanistic aspects and FT-IR investigation. Appl. Catal. B, 20, 15.

, Wang

S.F.

, Lü

M.K

, Zhou

G.J.

, Xu

, and Yuan

D.R.

(2004). Photoluminescence properties of SnO₂ nanoparticles synthesized by sol-gel method. J. Phys. Chem. B, 108, 8119.

Hsieh

C.T.

, Lai

M.H.

, and Pan

(2010). Synthesis and visible-light-derived photocatalysis of titania nanosphere stacking layers prepared by chemical vapor deposition. J. Chem. Technol. Biot., 85, 1168.

Huang

Y.F.

, Wei

Y.L.

, Cheng

S.H.

, Fan

L.Q.

, Li

Y.B.

, Lin

J.M.

, and Wu

J.H.

(2010). Photocatalytic property of nitrogen-doped layered perovskite K₂La₂Ti₃O₁₀. Sol. Energy Mat. Sol. C, 94, 761.

Kumar

, Govind , and Uma

(2011). Investigation of cation (Sn²⁺) and anion (N³⁻) substitution in favor of visible light photocatalytic activity in the layered perovskite K₂La₂Ti₃O₁₀. J. Hazard. Mater., 189, 502.

X.Y.

, Zhu

Z.R.

, Zhao

Q.D.

, and Liu

S.M.

(2011). FT-IR study of the photocatalytic degradation of gaseous toluene over UV-irradiated TiO₂ microballs: Enhanced performance by hydrothermal treatment in alkaline solution. Appl. Surf. Sci., 257, 4709.

Y.X.

, Lu

G.X.

, and Li

S.B.

(2003). Photocatalytic production of hydrogen in single component and mixture systems of electron donors and monitoring adsorption of donors by in situ infrared spectroscopy. Chemosphere, 52, 843.

Liu

, Wu

W.C.

, Guan

Y.J.

, Ying

P.L.

, and Li

(2002). FT-IR spectroscopic study of the oxidation of chlorobenzene over Mn-based catalyst. Langmuir, 18, 6229.

Lochař

(2006). FT-IR study of methanol, formaldehyde and methyl formate adsorption on the surface of Mo/Sn oxide catalyst. Appl. Catal. A, 309, 33.

10.

Macwan

D.P.

, Dave

P.N.

, and Chaturvedi

(2011). A review on nano-TiO₂ sol-gel type syntheses and its applications. J. Mater. Sci., 46, 3669.

11.

Osterloh

F.E.

(2008). Inorganic materials as catalysts for photochemical splitting of water. Chem. Mater., 20, 35.

12.

Park

J.H.

, Seo

Y.S.

, Kim

H.S.

, and Kim

I.K.

(2011). Photodegradation of benzene, toluene, ethylbenzene and xylene by fluidized bed gaseous reactor with TiO₂/SiO₂ photocatalysts. Korean J. Chem. Eng., 28, 1693.

13.

Sun

, Ding

J.J.

, Bao

, Gao

, Qi

Z.M.

, and Li

C.X.

(2010). Photocatalytic oxidation of gaseous formaldehyde on TiO₂: An in situ DRIFTS study. Catal. Lett., 137, 239.

14.

Wang

, Xi

J.Y.

, and Hu

H.Y.

(2008). Chemical identification and acute biotoxicity assessment of gaseous chlorobenzene photodegradation products. Chemosphere, 73, 1167.

15.

J.H.

, Cheng

Y.H.

, Lin

J.M.

, Huang

Y.F.

, Huang

M.L.

, and Hao

S.C.

(2007). Fabrication and photocatalytic properties of HLaNb₂O₇/(Pt, Fe₂O₃) pillared nanomaterial. J. Phys. Chem. C, 111, 3624.

16.

Yang

Y.H.

, Chen

Q.Y.

, Yin

Z.L.

, and Li

(2009a). Study on the photocatalytic activity of K₂La₂Ti₃O₁₀ doped with vanadium (V). J. Alloy. Compd., 488, 364.

17.

Yang

Y.H.

, Chen

Q.Y.

, Yin

Z.L.

, and Li

(2009b). Study on the photocatalytic activity of K₂La₂Ti₃O₁₀ doped with zinc (Zn). Appl. Surf. Sci., 255, 8419.

18.

Zhou

X.F.

, Fogarasi

, Liu

R.F.

, and Pulay

(1993). Building a database of force constants based on scaled ab initio (SQM) results. I. chlorobenzenes. Spectrochim. Acta. A, 49, 1499