ARTICLE IN PRESS
Available online at www.sciencedirect.com
Chemosphere xxx (2008) xxx–xxx
www.elsevier.com/locate/chemosphere
Technical Note
Photocatalytic oxidation of gas-phase elemental mercury
by nanotitanosilicate fibers
Seok Ho Jeon, Yujin Eom, Tai Gyu Lee *
Department of Chemical Engineering, Yonsei University, 134 Sinchon-dong, Seodaemun-gu, Seoul 120-749, Republic of Korea
Received 14 September 2007; received in revised form 21 November 2007; accepted 21 November 2007
Abstract
Photocatalytic fibers were generated from the continuous evaporation of titanium tetraisopropoxide with tetraethyl orthosilicate
through a flame burner. The morphology, the crystal form, and the components of the nanotitanosilicate fibers were analyzed by Raman
spectroscopy, Field emission-scanning electron microscope, X-ray diffraction, and Brunauer–Emmett–Teller surface area analysis. The
nanotitanosilicates prepared by three different carrier gases (air, N2, and Ar) were tested for their photocatalytic ability to remove/oxidize
gas-phase elemental mercury. Under UV black light, the Hg0 capture efficiencies were 78%, 86%, and 85% for air, N2, and Ar, respectively. For air, the value was close to 90%, even under household fluorescent light. The Hg0 capture efficiency by nanotitanosilcate was
measured under fluorescent light, UV black light, and sunlight.
Ó 2007 Elsevier Ltd. All rights reserved.
Keywords: Photocatalytic fiber; Diffusion flame burner; Nanotitanosilicate; Mercury removal
1. Introduction
In recent years, various applications of TiO2 in many
different types of catalytic devices, solar cells, and other
optoelectronic devices have been demonstrated (Taramasso
et al., 1983; Regan and Grätzel, 1991; Hoffmann et al.,
1995). The effectiveness of theses systems depends on the
photocatalytic properties of TiO2, which are in turn
strongly related to its crystal structure and grain size. In
most cases, TiO2 in the anatase phase appears to be most
useful, due to its high photocatalytic activity and physicochemical stability (Rao et al., 1980; Nishimoto et al., 1985).
Nanocrystalline TiO2 grains grow at relatively lower temperatures and undergo phase transformation, which limits
their application in many situations accompanying high
temperature (Anderson et al., 1988; Bhave, 1991). On the
other hand, the fibrous membrane with high thermal stability allows for gas separation at high temperatures. In combination with a chemical reaction, the membrane acts as a
*
Corresponding author. Tel.: +82 2 2123 5751; fax: +82 2 312 6401.
E-mail address: [email protected] (T.G. Lee).
catalyst, as well as a selective barrier to remove one of the
formation components (Ezzo et al., 1979; Brink and
Scherer, 1990; Makishima et al., 1990; Bhave, 1991). Therefore, as with the titania membrane systems, it is necessary to:
(1) generate the TiO2-compounds at certain conditions to
prevent any further phase transformations once they have
been formed and (2) to prepare the compounds in fibrous
form with a relatively high surface area. Recent studies suggest that silica-modified titania exhibit high thermal stability
(Iwamoto et al., 2000). Also, several researchers have
reported the fabrications of dense TiO2 in various configurations such as: (1) titania nanoribbons with a size range of
30–200 nm under hydrothermal conditions, from amorphous titania or titanium precursors in a basic medium
(Yuan et al., 2002), (2) amorphous titania nanofibers, electrospun from titanium tetraisopropoxide (TTIP) mixed with
acetic acid and high molecular weight polyvinylpyrrolidone
(Li and Xia, 2003), and (3) titania films with micrometer
pore size, by electrostatic sol-spray deposition using TTIP
(Nomura et al., 2003). Little has been reported regarding
the morphogenesis of column-shaped nanotitanosilicate,
prepared using a flame reactor.
0045-6535/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2007.11.050
Please cite this article in press as: Jeon, S.H. et al., Photocatalytic oxidation of gas-phase elemental mercury ..., Chemosphere (2008),
doi:10.1016/j.chemosphere.2007.11.050
ARTICLE IN PRESS
2
S.H. Jeon et al. / Chemosphere xxx (2008) xxx–xxx
Meanwhile, many researchers have been seeking effective ways to capture mercury from combustion flue gases
(Korpiel and Vidic, 1997; Hsi et al., 2001; Jurng et al.,
2002), due to their high toxicity, tendency to bio-accumulate, and difficulty to control. In previous studies (Wu
et al., 1998; Lee et al., 2001), a very high Hg0 capture efficiency by commercially available TiO2 powders, under various light sources have been reported, using UV light
(black and sterilizing, >99%), fluorescent light (home-style,
>99%), and visible blue light (>400 nm, 85%). These
studies showed that a far stronger bond formed between
mercury and TiO2 under photocatalytic oxidation. For
the elemental mercury photo-oxidation process by TiO2,
the mechanistic illustration has been introduced (Lee and
Hyun, 2006).
This study introduces a novel method for preparing
photocatalytic fibers by the controlled growth of columnshaped nanotitanosilicate. Also, we investigate their physical and chemical properties, specifically their ability to
photo-oxidize and capture gas-phase elemental mercury
in a well-controlled gas system (Hg0 + air + carrier gas).
2. Experimental
2.1. Nanotitanosilicate fiber preparation
A schematic of the experimental setup for the preparation of the nanotitanosilicate fiber is shown in Fig. 1a. A
diffusion flame reactor was designed to generate the nanotitanosilicate particles, and a slide glass plate (75 25 mm)
was used to collect and grow the photocatalytic fibers. Different ratios of TTIP and tetraethyl orthosilicate (TEOS)
mixture (2:1, 4:1, 10:1, and 12:1, by weight) were introduced into the flame system by a variably selected carrier
gas. Flow rate of fuel (LPG), oxidant (O2), sheath gas
(N2), and carrier gas were precisely controlled at 250,
400, 500, and 2000 sccm, respectively. In Fig. 1a, precursor
solution was vaporized and carried with carrier gas (air,
N2, and Ar) through the nozzle located in the center. The
sheath gases, O2 and N2, were passed through the nozzles
located in the 2nd and 3rd layer from the center, respectively. LPG was passed through the space between nozzles.
The temperature of the water bath containing TTIP was
maintained at 50 °C. The mixture of precursors was
dynamically stirred at room temperature for 20 min. The
glass plate was rotating horizontally at the speed (22–
23 rpm) enough to avoid melting and/or deforming, passing through the center of the flame (1450–1550 °C). The
nanotitanosilicate particles were collected on the lower surface of the glass plate, only 3 cm above the top of the burner. In order to collect 40 mg, about 90 h was required.
2.2. Characterization of nanotitanosilicate fiber
The crystallinity of the samples was checked with both a
Rigaku X-ray diffractometer (XRD) with a Cu Ka radiation source (k = 0.15405 nm) and Raman spectroscopy
(T64000, Jobin-Yvon, France) with a 514.532 nm laser
Ar-ion (2 mW on sample). Analysis of the morphology
and size of the prepared nanotitanosilicate fibers was conducted by SEM (JSM5410LV, JEOL, Japan), after coating
the collected samples with a gold-layer (20 nm) in a
cathodic evaporator. The BET surface area of nanotitanosilicate fibers was determined from nitrogen adsorption–
desorption obtained at 77 K by a MicromeriticsÒ ASAP
2020.
2.3. Photoreactivity (Hg0 capture efficiency) of
nanotitanosilicate fibers
Two nanotitanosilicate-coated glass plates, prepared by
using the diffusion flame reactor (Fig. 1a), were placed side
Fig. 1. Schematic diagram of (a) a diffusion flame reactor and (b) a photochemical reaction system.
Please cite this article in press as: Jeon, S.H. et al., Photocatalytic oxidation of gas-phase elemental mercury ..., Chemosphere (2008),
doi:10.1016/j.chemosphere.2007.11.050
ARTICLE IN PRESS
S.H. Jeon et al. / Chemosphere xxx (2008) xxx–xxx
by side in a quartz-covered flat-type photochemical reaction cell (inner dimension; 75 50 mm). A complete reaction system is shown in Fig. 1b.
Mercury vapor was introduced into the system through
a photochemical reaction cell by passing N2 at a precisely
controlled flow rate (MKS Mass-Flo Controller, MKS
Instruments, Inc.) above liquid elemental mercury
(99.9%, Aldrich), contained in a temperature controlled
gas-washing bottle. The flow rate of N2 through the Hg
containing gas-washing bottle in the water bath was
50 sccm; water bath temperature was 20 °C. The range of
Hg inlet concentration tested was 400–500 lg m3.
The gas-phase elemental mercury concentration from
the photochemical reaction cell was constantly monitored
in real-time by the on-line Hg0 analyzer (VM3000, Mercury
Instruments, Germany). The light source was supplied only
after the mercury concentration reading had stabilized, and
the outlet mercury concentration was measured.
Wavelength distribution and the corresponding intensity
were measured for the fluorescent light (2 FL40D,
Osram-Korea, Korea) and UV black light (2 TLD36w/
08, Philips, Netherlands) at a distance of photochemical
reaction cell from the light source.
3
ratios, are shown in Fig. 3. A crystal form was only
observed for the TTIP/TEOS ratio of 10:1 and 12:1, and
not for the TTIP/TEOS ratio of 2:1 and 4:1, indicating
the crucial dependence of crystallite growth on TTIP/
TEOS ratio. Five distinct Raman peaks, located at 142.2,
196.1, 396.3, 518.5, and 638.1 cm1, were identified for
TTIP:TEOS of 10:1 and 12:1. However, for TTIP:TEOS
of 2:1 and 4:1, the samples did not show crystal structures
due to the increased amount of silica.
Fig. 4 shows the field emission-scanning electron microscope (FE-SEM) images of the fibers prepared with three
different carrier gases (air, N2, and Ar). The diameters of
the fibers are in the range of 50–100 nm, and their lengths
are typically up to several tens of lm. A relatively rough
and coarse surface was observed, probably caused by the
volume shrinkage during the collection process due to further heating and densification of the readily deposited fiber
by the high temperature flame. The results in Fig. 4 suggested that the surface areas of the fiber samples correspond to the exterior surfaces of nanotitanosilicate fibers
and that no micropores were formed in the interlayer.
The measured BET surface areas of nanotitanosilicate
fibers corresponding to the carrier gas used (air, N2, and
Ar) were 45, 78, and 126 m2 g1, respectively (Table 1).
3. Results and discussion
3.1. Characterization of nanotitanosilicate fibers
3.2. Photoreactivity (Hg0 capture efficiency) of
nanotitanosilicate fibers
Nanotitanosilicate fibers are synthesized by the evaporating of TEOS with TTIP by atomization method at
1450–1550 °C, and the proposed mechanism is shown in
Fig. 2. Raman spectroscopy results of the samples, all prepared using air as a carrier gas at various TTIP to TEOS
In Fig. 5, the XRD spectra of column-shaped nanotitanosilicates show the existence of both anatase and rutile
phases. Three different carrier gases (air, N2, and Ar) were
used for comparison, maintaining the TTIP/TEOS ratio at
12–1. This ratio was selected after considering the Raman
Fig. 2. Proposed mechanism for the controlled growth of the nanotitanosilicate fiber.
Please cite this article in press as: Jeon, S.H. et al., Photocatalytic oxidation of gas-phase elemental mercury ..., Chemosphere (2008),
doi:10.1016/j.chemosphere.2007.11.050
ARTICLE IN PRESS
4
S.H. Jeon et al. / Chemosphere xxx (2008) xxx–xxx
Table 1
Hg0 removal efficiency by the nanotitanosilicate fibers under different light
sources
Carrier gas used for
nanotitanosilicate fibers
(TTIP:TEOS = 12:1)
Air
N2
Ar
Light sources
fluorescent
(%)
UV
black
The
sun
88
9
8
78
86
85
51
17
15
BET surface
area
(m2 g1)
45
78
126
Fig. 3. Raman spectra of nanotitanosilicate fibers using air as a carrier gas
(TTIP:TEOS ratios of: (a) 2:1, (b) 4:1, (c) 10:1, and (d) 12:1).
Fig. 5. X-ray diffraction patterns of nanotitanosilicate fibers (TTIP:TEOS
ratios of 12:1 carried by (a) air, (b) N2, and (c) Ar; A: antase, R: rutile).
analysis results previously discussed (Fig. 3). Such crystallinity strongly affected the structural and electronic properties of the photocatalyst.
The samples were then tested for their ability to oxidize
Hg0, generally referred to as Hg0 capture efficiency, in a
specially designed nanotitanosilicate photochemical reactor. The Hg0 capture efficiency (g) was based on inlet/outlet
gas-phase Hg0 concentrations; g ¼ ðHg0in Hg0out Þ=Hg0in .
A description of the reaction between the columnshaped nanotitanosilicate particle and Hg0 has been suggested by previous studies (Wu et al., 1998; Lee et al.,
2001). The positive holes may result in OH radical forma-
tion by adsorption of O2 or H2O, which then oxidizes mercury, thus creating a lower volatile form (HgO) that is
retained on the titania particle surface (Linsebigler et al.,
1995; Lee et al., 2001).
Ti–OH þ Hg ! Ti–HgOH þ HO ! Ti–HgO þ H2 O
Fig. 6 shows the plot of real-time Hg0 concentration at
the exit of the reaction cell of nanotitanosilicate fibers prepared with three different carrier gases, air, N2, and Ar
under various light sources: (a) fluorescent, (b) UV black,
Fig. 4. FE-SEM images of nanotitanosilicate fibers (TTIP/TEOS carried by (a) air, (b) N2, and (c) Ar).
Please cite this article in press as: Jeon, S.H. et al., Photocatalytic oxidation of gas-phase elemental mercury ..., Chemosphere (2008),
doi:10.1016/j.chemosphere.2007.11.050
ARTICLE IN PRESS
S.H. Jeon et al. / Chemosphere xxx (2008) xxx–xxx
(c) the sun. All fiber samples were prepared under the same
conditions except for the type of carrier gas used. The ratio
of TTIP to TEOS was maintained at 12–1, and the amount
of sample tested was always 40 mg. To verify the fact that
elemental mercury was actually removed by chemi-sorption
onto the photocatalytic fiber surface and not simply dis-
5
charged into the gas-phase in the oxidized form, the simulated flue gas out of the photochemical reactor was
sampled using the Ontario Hydro Method and then analyzed. No oxidized form of mercury was detected.
The results in Table 1 show the following order of
decreasing photoreactivity for the carrier gases tested: air,
N2, and Ar. Photocatalytic fibers prepared with air as the
carrier gas exhibited the highest Hg0 capture efficiency of
88% under the fluorescent light (2 36 W). However, in
the case of N2 and Ar as carrier gases, the Hg0 capture efficiency was less than 10% under the same fluorescent light.
On the other hand, under UV black light with wavelengths
of 300–400 nm, the Hg0 capture efficiencies were relatively
high for all three carrier gases (air: 78%, N2: 86%, and Ar:
85%). The low Hg0 capture efficiencies of N2- and Arderived nanotitanosilicate under the fluorescent light
(which was also observed under the sunlight) are due to
the lack of a rutile phase. This is in contrast to the case
for air (Fig. 5), since the rutile phase better corresponds
to light sources with longer wavelengths (Lee et al.,
2004). It also explains the relatively high Hg0 capture efficiency of air-derived nanotitanosilicate, exhibiting both
anatase and rutile phases under fluorescent light and sunlight. Values for Hg0 capture efficiency, listed in Table 1,
could be increased (as high as 100%) by increasing the
amount of nanotitanosilicate used in the photochemical
reactor (Lee et al., 2004). Values in Table 1 were used in
this study only to compare the relative Hg0 removal effectiveness between the tested samples. A relatively high Hg0
capture efficiency of amorphous TiO2 was observed under
UV light probably due to large moisture contents (relative
humidity of more than 90% during particle collection). It
was previously reported that the hydrated forms of amorphous titania, TiO2 nH2O, should be noted as a significant
factor in promoting the greater concentrations of Ti3+ at
the surface and higher photoreactivity rates for H2 evolution (Zhang and Maggard, 2007). In addition, according
to the previous reports (Kaneko and Okura, 2002; Zhang
et al., 2007; Zhang and Maggard, 2007), amorphous TiO2
has greater band gap energy of 3.27 eV than rutile and anatase TiO2, which have band gap energy of 3.0 and 3.2 eV,
respectively. Since UV black light can still provide enough
energy for the band gap of the amorphous nanotitanosilicate fibers, much higher Hg0 removal efficiency was
observed even for the photocatalytic fibers prepared with
N2 and Ar as carrier gases.
4. Conclusions
Fig. 6. Gas-phase Hg0 concentration vs. time for three different carrier
gases (air, N2, and Ar) under (a) fluorescent light, (b) UV black light, and
(c) sunlight.
In summary, we developed a new method for the preparation of photocatalytic fibers from column-shaped nanotitanosilicate by a diffusion flame reactor, using a mixture
of TTIP and TEOS. First, the mixture of TTIP and TEOS
is vaporized and carried into the flame system at 1450–
1550 °C in the presence of fuel (LPG), oxidants (O2),
sheath gas (N2), and different types of carrier gas (air,
N2, and Ar). Vapors are then condensed and particles are
Please cite this article in press as: Jeon, S.H. et al., Photocatalytic oxidation of gas-phase elemental mercury ..., Chemosphere (2008),
doi:10.1016/j.chemosphere.2007.11.050
ARTICLE IN PRESS
6
S.H. Jeon et al. / Chemosphere xxx (2008) xxx–xxx
formed by nucleation. The particles are then self-assembled
to form nanotitanosilicate fibers during which carbon dioxide and hydrogen molecules are removed. The resulting
fibers have high thermal stability and a large number of terminal oxygens, which makes them suitable for photocatalytic materials.
Furthermore, the photocatalytic activities of various
nanotitanosilicate samples were evaluated by observing
the Hg0 oxidation capacity, generally referred to as the
Hg0 capture efficiency. The Hg0 capture efficiency was
obtained for all three types of carrier gas (air, N2, and
Ar). The highest Hg0 capture efficiency (88%) was achieved
for the case of air under household fluorescent light, which
is low-cost and easily maintainable.
Acknowledgement
This work was supported by Ministry of Environment
as ‘‘The Eco-technopia 21 project” (No. 2007-010030059-0).
References
Anderson, M.A., Giesemann, M.J., Xu, Q., 1988. Titania and alumina
ceramic membranes. J. Membrane Sci. 39, 243–258.
Bhave, R.R., 1991. Inorganic membranes synthesis. In: Characteristics
and Applications. Van Nostrand Reinhold, New York.
Brink, C.J., Scherer, G.W., 1990. Sol–gel Science: The Physics and
Chemistry of Sol–gel Processing. Academic Press, San Diego.
Ezzo, E.M., Elnabarawy, T.H., Youssef, A.M., 1979. Studies of the mixed
oxide catalyst Al2O3–Cr2O3: I. Surface properties. Surf. Technol. 9,
111–118.
Hoffmann, M.R., Martin, S.T., Choi, W., Bahnemann, D.W., 1995.
Environmental applications of semiconductor photocatalysis. Chem.
Rev. 95, 69–96.
Hsi, H.C., Rood, M.J., Rostam-Abadi, M., Chen, S., Chang, R., 2001.
Effects of sulfur impregnation temperatures on the properties and
mercury adsorption capacities of activated carbon fibers (ACFs).
Environ. Sci. Technol. 35, 2785–2791.
Iwamoto, S., Tanakulrungsank, W., Inoue, M., Kagawa, K., Praserthdam, P., 2000. Synthesis of large-surface area silica-modified titania
ultrafine particles by the glycothermal method. J. Mater. Sci. Lett. 19,
1439–1443.
Jurng, J., Lee, T.G., Lee, G.W., Lee, S.J., Kim, B.H., Seier, J., 2002.
Mercury removal from incineration flue gas by organic and inorganic
adsorbents. Chemosphere 47, 907–913.
Kaneko, M., Okura, I., 2002. Photocatalysis, Science and Technology.
Springer-Verlag, Berlin Heidelberg New York.
Korpiel, J.A., Vidic, R.D., 1997. Effect of sulfur impregnation method on
activated carbon uptake of gas-phase mercury. Environ. Sci. Technol.
31, 2319–2325.
Lee, T.G., Hedrick, E., Biswas, P., 2001. Comparison of Hg0 capture
efficiencies of three different in situ generated sorbents. AIChE J. 47,
954–961.
Lee, T.G., Hyun, J.E., 2006. Structural effect of the in situ generated
titania on its stability to oxidize and capture the gas-phase elemental
mercury. Chemosphere 62, 26–33.
Lee, Y.G., Park, J.-W., Kim, J.-H., Min, B.R., Jurng, J., Kim, J., Lee,
T.G., 2004. Comparison of mercury capture efficiency from a
simulated exhaust gas by several types of TiO2 under various light
sources. Chem. Lett. 33, 36–37.
Li, D., Xia, Y.N., 2003. Fabrication of titania nanofibers by electrospinning. Nano Lett. 3, 555–560.
Linsebigler, A.L., Lu, G., Yates Jr., J.T., 1995. Photocatalysis on TiO2
surfaces: principles, mechanisms, and selected results. Chem. Rev. 95,
735–758.
Makishima, A., Asami, M., Wada, K.J., 1990. Preparation and properties
of TiO2–CeO2 coatings by sol–gel process. Non-Cryst. Solids 121, 310–
314.
Nishimoto, S., Ohtani, B., Kajiwara, H., Kagiya, T., 1985. Correlation of
the crystal structure of TiO2 prepared from titanium tetra-2-propoxide
with the photocatalytic activity for redox reactions in aqueous propan2-ol and silver salt solutions. J. Chem. Soc. Faraday Trans. 81, 61–68.
Nomura, M., Meester, B., Schoonman, J., Kapteijn, F., Moulijn, J.A.,
2003. Improvement of thermal stability of porous titania films
prepared by electrostatic sol-spray deposition (ESSD). Chem. Mater.
15, 1283–1288.
Rao, M.V., Rajeshwar, K., Pal Vernerker, V.R., Dubow, J., 1980.
Photosyntlhetic production of H2 and H2O2 on semiconducting oxide
grains in aqueous solutions. J. Phys. Chem. 84, 1987–1991.
Regan, O.B., Grätzel, M., 1991. A low-cost, high-efficiency solar cell based
on dye-sensitized colloidal TiO2 films. Nature 35, 737–740.
Taramasso, M., Perego, G., Notari, B., 1983. Preparation of porous
crystalline synthetic material comprised of silicon and titanium oxides.
US Patent 4, 410, pp. 501–508.
Wu, C.Y., Lee, T.G., Tyree, G., Arar, E., Biswas, P., 1998. Capture of
mercury in combustion systems by in situ generated titania particles
with UV irradiation. Environ. Eng. Sci. 15, 137–148.
Yuan, Z., Colomer, J., Su, B., 2002. Titanium oxide nanoribbons. Chem.
Phys. Lett. 363, 362–366.
Zhang, M., Lin, Guoqaing, Dong, C., Wen, L., 2007. Amorphous TiO2
films with high refractive index deposited by pulsed bias arc ion
plating. Surf. Coat. Technol. 201, 7252–7258.
Zhang, Z., Maggard, P.A., 2007. Investigation of photocatalytically-active
hydrated forms of amorphous titania, TiO2 nH2O. J. Photochem.
Photobiol. A 18, 8–13.
Please cite this article in press as: Jeon, S.H. et al., Photocatalytic oxidation of gas-phase elemental mercury ..., Chemosphere (2008),
doi:10.1016/j.chemosphere.2007.11.050