|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
DEVELOPMENT OF TiO2/ACTIVATED CARBON COMPOSITE
PHOTOCATALYST FOR THE REMOVAL OF METHANOL AND HYDROGEN
SULFIDE FROM PAPER MILLS
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
To my parents, Bang-He Tao and Ping-Fen Li, who raised me up. Your love is always in
my heart. To my twin sister, Hui Tao. Without your encouragement and support, I could
not have this achievement.
I would like to give my sincere gratitude to Dr. Chang-Yu Wu for his support and
guidance in my Ph.D. study. His advice and encouragement were precious for my
graduate study. My special thanks go to my committee, Dr. David W. Mazyck, Dr. Jean
M. Andino, and Dr. Wolfgang Sigmund, who gave me continuous comments and
valuable advice on my research.
I also thank the staff in ERC and MAIC for teaching and allowing me to use many
instruments such as XRD, SEM, and BET. I would like to thank our past and present
group members. Special thanks go to Jennifer, Ameena, Yu-Mei, A-Chuan, Ying, and
Finally I would like to give my sincere gratitude to my family. Without their care
and encouragement, I could not have achieved this goal.
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ................................................................................................. iv
LIST O F TA BLE S ........... .............. .... ............. .............. .. vii
L IST O F F IG U R E S .......................................................................... ..... viii
A B ST R A C T ................. .......................................................................................... x
1 GENERAL INTRODUCTION ............................................................................1
2 PHOTOCATALYTIC REGENERATION ............................ ............... 10
In tro du ctio n ...............10..............................................
E x p erim ental Section ......................................................................... ................... 14
C atalyst and C hem icals ............................................... ............................ 14
Characterization ................. .................. ...... .. .................. 14
M ethanol R em oval Evaluation............................................. ......... ... ............... 15
R results and D discussion .......... .. .................... ................ ............ .... ..... 19
TiO 2/A C C haracterization ............................................................... ..................19
Methanol Removal by Adsorption and Adsorption/Photocatalytic Oxidation....20
E effect of W after ............................................. .. ..................... 2 1
Regeneration Performance ....................................................... 25
C o n c lu sio n s........................................................................................................... 2 8
3 PREPARATION OF TiO2/AC COMPOSITE PHOTOCATLYST BY DRY
IM PR E G N A T IO N ..................................................................... ..... ......................30
Introduction ........................................................................................................ 30
E xperim ental Section ........... ...... ...................................... .............. .. .... .. .. 32
M materials ................................................................................................... ....... 32
M material Preparation .................................................. .............................. 32
C characterization .................................................................. 35
Photocatalytic A activity Evaluation.................................... ...................... 35
R results and D iscu ssion ................................................. ... .. ........ .... ............36
Effect of Calcination Conditions .......................................................................36
Effect of Hydrolysis Conditions.......... .. .................................. ...............37
E effect of TTIP C concentration ...................................................................... ...43
Effect of M oisture in Carbon...................................................... ... ........... 44
Photocatalytic A activity ............................................... ............................. 46
C conclusion ...................................................................................................... ....... 49
4 H2S REMOVAL BY A TiO2/AC COMPOSITE PHOTOCATALYST
PREPARED BY DRY IMPREGNATION ..................................... ...............50
Introduction ..................................... ............ 50
Properties of H2S and Regulations .........................................................50
H2S Adsorption/Oxidation on AC ............................................ ............... 52
E xperim ental Section ........... ...... ...................................... .............. .. .... ...... 55
Materials ...................... .. ...............55
Photocatalytic A activity Evaluation.................................... ...................... 55
R results and D discussion ..................... .. .. .......................... ..... ..... 57
C o n c lu sio n s........................................................................................................... 6 3
5 MICROWAVE-ASSISTED PREPARATION OF TiO2/AC COMPOSITE
P H O T O C A T A L Y ST ......................................................................... ...................64
In tro d u ctio n ............................................................................................6 4
E xperim ental Section ........... ...... ...................................... .............. .. .... ...... 69
Materials .............. ......... ........ ....... ...............69
T iO 2/A C P reparation ................................................... .. ........ ...... ............70
Characterization ........... ... .. .. ....... .. .......................... ....... .. 71
Photocatalytic A activity Evaluation................................... ....................... 71
R esu lts an d D iscu ssion ....................................................... ............................ .....72
Carbon Weight Loss and TiO2 Loading ................................... ...............72
TiO2/AC Characterization .................................. .......................... 76
M ethanol rem oval testing ........................................................ ............... 79
C conclusions ................................................ 81
6 CONCLUSIONS AND RECOMMANDATIONS.......................... .....................83
L IST O F R E F E R E N C E S ........................................................................ .....................86
B IO G R A PH IC A L SK E TCH ..................................................................... ..................93
LIST OF TABLES
1-1 Pollution control techniques for gaseous and particulate matter emissions from
K raft pulp m ill sources ....................................... .............. .. .. ........ .. ..
1-2 H2S and total reduced sulfur (TRS) compounds emissions ......................................5
2-1 The experimental conditions of methanol removal ................................................18
..2-2 BET surface area and pore size distribution of AC and TiO2/AC ........................20
3-1 E effect of hydrolysis conditions* ................................................... ..................... 33
3-2 Effect of calcination conditions........................................ ............................ 34
3-3 TTIP concentration and carbon moisture* .......... .........................................34
4-1 Surface pH of AC and TiO2/AC ........... ........... ............... 59
5-1 Penetration depth of microwaves (2.45 GHz) ......... ..................................... 67
5-2 Preparation conditions and characterization of TiO2/AC...................................71
5-3 Ash content of F400 AC before and after microwave process ..............................74
LIST OF FIGURES
1-1 Simplified process flow chart of the Kraft pulp mill ...............................................2
1-2 Schematic photoexcitation in a photocatalyst followed by deexciatation events ......7
2-1 Overview of available techniques for AC regeneration.................................. 11
2-2 Setup of experimental apparatus .............. .... ............................ ............. 17
2-3 SEM images for the TiO2/AC composite ............... ................ 20
2-4 The relative effluent methanol concentration profiles for the virgin AC and
T iO 2/A C ............................................................................22
2-5 The relative methanol effluent concentration of 7 g TiO2/AC composite ...............22
2-6 The relative effluent methanol concentration for Sets 1, 2 and 3 ...........................24
2-7 The adsorption breakthrough curves of methanol from fresh and regenerated
beds of TiO2/AC of Set 4 .................................... ......... ...............26
2-8 The adsorption capacity of each cycle of Set 4, 5 and 6 .......................................26
2-9 The methanol adsorption amount in adsorption, methanol desorption and
formaldehyde formation amount in regeneration of each cycle of Sets 7 and 8 ......27
3-1 Schematic of the hydrolysis reactor ........ .... ................................. ............... 34
3-2 XRD patterns of samples calcined under different conditions.............................37
3-3 XRD patterns of samples prepared under different hydrolysis conditions ..............38
3-4 SEM images of samples prepared under different hydrolysis conditions ..............39
3-5 The LaMer diagram related to nucleation and growth mechanism........................42
3-6 SEM im age of sample 2's inner surface........................................ ............... 43
3-7 SEM im ages of samples 9-10 ......... .................. ................... ...................... 44
3-8 SEM im age of sam ple 11 .................................. ........................................ 45
3-9 X RD pattern of sam ple 11 .............................................. .............................. 45
3-10 M ethanol effluent concentration profiles ...................................... ............... 47
3-11 Average methanol removal efficiency of samples ................................................47
4-1 Proposed pathway of H2S oxidation on unmodified AC in the presence of water ..54
4-2 Experim ental set-up for H2S rem oval ........................................... ............... 57
4-3 Outlet H 2S concentration profiles ........................................ ........................ 58
4-4 H2S removal efficiency and SO4 conversion efficiency..............................59
4-5 H2S removal efficiency and SO4 conversion efficiency of AC...............................61
4-6 Outlet H2S concentration passing empty reactor..........................................62
5-1 Schematic description of a microwave). ...................................... ............... 65
5-2 Rotation of molecules with microwave............ ............................ .............65
5-3 F400 AC Weight Loss under medium level MW irradiation...............................74
5-4 Weight loss curves of TiO2/AC samples....................... ...... ...............76
5-5 SEM images of TiO2/AC samples............................ ...................... 77
5-6 Cross-section of Sam ple 12............................................. ............................. 77
5-7 Region 1 and Region 2 in Figure 5-6 ............................................ ............... 78
5-8 XRD patterns of different sam ples..................................... .......................... 79
5-9 M ethanol effluent concentration profiles ...................................... ............... 80
5-10 Average methanol removal efficiencies ............ .............................................81
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
DEVELOPMENT OF TiO2/ACTIVATED CARBON COMPOSITE
PHOTOCATALYST FOR THE REMOVAL OF METHANOL AND HYDROGEN
SULFIDE FROM PAPER MILLS
Chair: Chang-Yu Wu
Major Department: Environmental Engineering Sciences
The objective of this study was to investigate the technical efficacy of in-situ
treatment of pulp and paper emissions via adsorption and photocatalytic regeneration.
Firstly, activated carbon was coated with a commercially available photocatalyst by
a spray desiccation method. The spent TiO2/AC was regenerated by UV light irradiation.
The photocatalytic regeneration is ascribed to both desorption from AC and
photocatalytic degradation on TiO2.
In order to improve the photocatalytic degradation rate, the synthesis of TiO2/AC
composites by dry impregnation method was developed. The composites prepared using
various hydrolysis and calcination conditions were evaluated. High hydrolysis
temperature resulted in rough particulate coating layers with higher surface area. The
TiO2 loading was positively correlated with the precursor concentration although the
TiO2 loading in the study range (2 8 wt%) was not critical to the photocatalytic
performance. The moisture in carbon was beneficial for the hydrolysis of precursor
(titanium tetra-isopropoxide, TTIP) and improved the composite performance. Under
proper preparation conditions, the TiO02/AC composite outperformed the composite
prepared by spray desiccation at removing methanol.
The BioNuchar AC support itself was a good H2S remover. After coating TiO2 by
dry impregnation, H2S removal efficiency of TiO02/AC decreased compared with the
virgin AC due to the change of surface pH. Under UV light irradiation, H2S removal
efficiency of TiO02/AC composite doubled, and its sulfate conversion efficiency was
higher than that of AC. The formation of sulfate is preferred for water regeneration.
Ti02/AC composite photocatalyst was also prepared by a novel microwave-assisted
impregnation method and was employed for the removal of methanol from humid air
streams. A commercial microwave oven (800 W) was used as the microwave source.
Under 2450 MHz microwave irradiation, TTIP was quickly hydrolyzed and anatase TiO2
was formed in a short time (< 20 minutes). Due to the volumetric heating and selective
heating of microwave, the solvent and by-products were quickly removed which reduced
energy consumption and processing time. The formed submicron TiO2 particles were
mainly deposited on the external surface of carbon and had photocatalytic activity.
The pulp and paper industry deserves special attention for both its energy and
environmental impact. It is the fourth largest consumer of electricity and fuels and the
third largest consumer of fresh water in the United States (Sittig, 1977). During the 4
years spanning 1997-2000, the pulp and paper industry spent $6 billion per year on
energy, or about 4 percent of its net sales and averaged over $800 million per year on
environmental protection capital, or about 14 percent of the average annual capital
invested on equipment (Department of Energy [DOE], 2003).
To produce paper or paperboard, the wood is pulped at first. Pulps are made from
wood chips, whole tree chips, sawmill residues, or logs. Pulps can be prepared through
chemical and/or mechanical means. The pulp may then be bleached to various degrees of
brightness. Finally, bleached or unbleached pulp is processed into paper board or paper.
The dominant wood pulping process today is the Kraft process (Buonicore and Davis,
1992). Figure 1-1 shows a simplified process flow sheet of the pulp mill (Buonicore and
Pulp and paper mill effluents are complex mixtures. The characteristics of each
effluent are dependent on numerous factors including wood furnish and process
technology (including washing, cooking, bleaching, prebleaching, etc.), as well as final
effluent treatment (Serros, 1996). The atmospheric emissions from the Kraft process
include both gaseous and particulate materials. The major gaseous emissions are
malodorous reduced sulfur compounds referred to as total reduced sulfur (TRS), such as
hydrogen sulfide (H2S), methyl mercaptan (CH3SH), dimethyl sulfide ((CH3)2S), and
dimethyl disulfide ((CH3)2S2); organic nonsulfur compounds; oxides of sulfur; and
oxides of nitrogen The particulate emissions are primarily sodium sulfate and sodium
carbonate. Historically, odor and visible particulate emissions from Kraft pulp mills have
received considerable attention. A summary of the major control techniques for gaseous
and particulate emissions from specific kraft pulp mill sources are presented in Table 1-1.
"- J II -
.......if c t o te K TOI' p lr
.n i 19 -2 ). [E
Sai Ctr and Stream Imn----
Lfl-iQU ------*L,- i T ,,-i
and Davis, 1992)
Inc. (NCASI) conducted a study to characterize the emissions of volatile organic
sources (NCASI, 1994). Seven lime kilns, four smelt dissolving tanks, and a number of
TO BLEACH PLANT '- ~ -~ ]
OR PAPPfl NL NMI
Figure 1-1 Simplified process flow chart of the Kraft pulp mill (adapted from Buonicore
and Davis, 1992)
In 1992, the National Council of the Paper Industry Air and Stream Improvement,
Inc. (NCASI) conducted a study to characterize the emissions of volatile organic
compounds (VOCs) and hazardous air pollutants (HAPs) from chemical pulp mill
sources (NCASI, 1994). Seven lime kilns, four smelt dissolving tanks, and a number of
Table 1-1 Pollution control techniques for gaseous and particulate matter emissions from
Kraft pulp mill sources
Emission Source Gaseous Control Particulate Control
Digester gases Incineration NA(not applicable)
Washer vent Incineration NA
Evaporator gases Incineration NA
Condensate water Steam stripping NA
Condensate Stripper Vent Incineration NA
Black Liquor Oxidation Incineration NA
Tall Oil Vent Scrubbing NA
Recovery Furnance Scrubbing Precipitators
Smelt Tank Scrubbing Scrubbing
Lime Kiln Scrubbing Precipitators
Slaker Vent NA Scrubbing
Bleach Plant Scrubbing NA
Paper Machine Incineration NA
Power Boiler NA Cyclones
(Adapted from Sittig, 1996)
vents in the causticizing areas of eight mills were tested. Their results showed the average
HAPs and EPA method 25A VOC emissions from causticizing areas vents, including
lime kilns, smelt dissolving tanks, and miscellaneous causticizing area vents, were 0.66
lb/ADTP (pounds/air dried tons of pulp) and 0.48 lb C/ADTP ( lb as carbon (Method 25
A)/ADTP). The smelt dissolving tank contributed over 75% of the total HAP emissions.
The total HAP emissions of the four tested smelt dissolving tank scrubbers vent averaged
0.50 lb C/ADTP, and methanol was the major HAP which contributed on average 98% to
the total HAPs. The average concentration of methanol in the four tested smelt dissolving
tank scrubbers' vent ranged from around 10 to 1100 ppm. The flow rates ranged from
5000 to 16000 DSCFM (dry standard cubic feet per minute).
Although odor emissions from pulp and paper mill have received considerable
attention, the control and abatement of them are difficult tasks to achieve in the air
pollution problems afflicting the paper and pulp mills. This is due to the very low human
olfactory threshold levels of TRS gases. Two main TRS sources are recognized to occur
in paper and pulp mills: 1) the LVHC (low volume high concentration) gases, usually
dealt with by incineration, emerge from brownstock washers, digester, and evaporator
systems, etc.; 2) the HVLC (high volume low concentration) gases, dealt with by alkaline
amine wet scrubbing, are released by recovery furnaces and lime kilns. However, these
control devices can not remove all the pollutants. Bordado and Gomes (2001)
characterized the reduced sulfur compounds emitted from a Kraft pulp mill, which was
the biggest Portuguese plant producing bleached eucalyptus elemental chlorine free pulp
and its output was 430,000 t/yr. Table 1-2 shows part of their results (Bordado and
Gomes, 2001). It is clear that there is considerable amount of TRS gases in the emissions
from incinerators and scrubbers.
In 1998, the U.S. Environmental Protection Agency (EPA) promulgated the
"cluster rules" for the pulp and paper industry. The Maximum Achievable Control
Technology (MACT) portion of the April 1998 Cluster Rule specifies the control of
hazardous air pollutant emissions. For the pulp and paper industry, this is primarily
methanol. These rules require compliance by April 15, 2001. Additional controls will be
required by 2006 for HVLC streams (Springer, 2000). New regulations for air emission
have stimulated researchers to search for a coat-effective technique for in-situ treatment
of these pollutants. A limited number of Kraft mills currently have HVLC collection and
incineration systems in place (Varma, 2003). Considering the operating cost and the
formation of noxious by-products, incineration is unfavorable in the long run. Therefore,
an alternative technique for the treatment of the HVLC emission is worthy of exploration.
Table 1-2 H2S and total reduced sulfur (TRS) compounds emissions
source H2S Q H2S Emission TRS Emission
I(ppm) (mgNm3) (m3/h) (kg/h) (kg/h)
Recovery boiler scrubber <5 <7.2 183100 <1.318 16.5
Smelt tank 70 100.8 5000 0.504 0.754
Incinerator exhaust <5 <7.2 5000 <0.036 0.679
Incinerator scrubber 48 69.12 5000 0.346 0.346
Lime kiln 17 24.48 21200 0.519 0.519
Pulp cooking and Washing section
Bin hopper 46 66.24 3900 0.258 7.793
Pulp bleaching section
Oxygen reactor <5 <7.2 2662.9 <0.019 0.24
(Adapted Bordado and Davis, 2001)
Granular activated carbon (GAC) adsorption is a commonly used technique for the
removal of various VOCs and HAPs. The major limitation of activated carbon adsorption
is its nondestructive character. Once exhausted, activated carbon is either disposed of, or
treated to destroy the adsorbed pollutants (regeneration). Moreover, the spent carbon
itself may have to be handled as a hazardous waste when disposed of (Liu et al., 1996).
Regeneration of spent carbon is a very critical step to the effective use of carbon and thus
lowers the cost.
Recently, heterogeneous photocatalysis and its application in environmental
cleanup have been one of the most active research topics. Scientific research in
heterogeneous photocatalysis started in 1970s. Fujishima and Honda (1972) discovered
the photocatalytic splitting of water on TiO2 electrodes, which marked the beginning of a
new era in heterogeneous photocatalysis. Photocatalysts are invariably semiconductors,
which can promote reactions in the presence of light and are not consumed in the overall
reactions. Nanosized titanium dioxide (TiO2) under the form of anatase has been found as
an excellent photocatalyst since it is able to utilize near UV light, is safe, and is
inexpensive. It has been shown to be effective for a variety of inorganic and organic
compounds, for the destruction of microorganisms, for the inactivation of cancer cells,
for odor control, and for the clean-up of oil spills. (Hoffmann et al., 1995; Linsebigler et
al., 1995; Peral et al., 1997; Bhatkhande et al., 2001)
Figure 1-2 shows the mechanism of photocatalysis. Semiconductors possess a void
energy region, which extends from the top of the filled valence band (VB) to the bottom
of the vacant conduction band (CB). This void region is called band gap. By light
absorption with energy equal to or greater than the band gap, an electron in the VB is
excited to the CB (the enlarged section of Figure 1-2). The highly reactive electron-hole
pair undergoes de-excitations in several pathways, as shown in Figure 1-2. A
photoinduced electron can migrate to the surface and reduce the electron acceptor
(usually oxygen in air or an aerated solution). A photoinduced hole can migrate to the
surface and oxidize the donor species. In this way, highly reactive radical species, such
as hydroxyl radical OH* and superoxdie ion 0O are formed. Pollutant degradation may
occur indirectly via the surface-bound hydroxyl radical or directly via the VB hole before
it is trapped (Hoffmann et al., 1995; Linsebigler et al., 1995; Peral et al., 1997;
Bhatkhande et al., 2001).
In practice, the separation of nanosized photocatalyst from the treated fluid limits
its industrial applicability (Chen et al., 2001; El-Sheikh et al., 2004). To overcome this
drawback, the powder photocatalyst can be dispersed onto a support. The photocatalyst
coated on nonporous support has a limited contact area, thus low efficiency (El-Sheikh et
al., 2004). Porous support has been used to enhance the efficiency (El-Sheikh et al.,
2004). Activated carbon is chemically inert at low temperature and hence is a suitable
support for photocatalysts (Torimoto et al., 1997; Harada et al., 1999; Khan, 2003;Tao et
hv CB hv
e- + h+
A- 4 e- + h+ ht D+
Figure 1-2 Schematic photoexcitation in a photocatalyst followed by deexciatation
events. (Linsebigler et al., 1995)
On the other hand, in a heterogeneous photocatalysis system, photoinduced
molecular transformations take place at the surface of a catalyst (Linsebigler et al., 1995).
The destruction rate is dependent upon the ability of the pollutants to diffuse to the
catalyst. When the pollutants' concentrations are low, the degradation rates are low.
Malato and coworkers (2001) studied the photocatalytic oxidation of 2,4-Dichlorophenol
(DCP) in water using Degussa P25 TiO2 (BET specific surface area 50 m2 g1, mean
primary particle diameter z 20 nm, density = 4 g cm-3, 80% anatase phase/20% rutile
natase phase/20% rutile
phase) suspension under solar radiation. Their result showed that mineralization of DCP
was slow when its concentration was low. They proposed a combination of photocatalytic
degradation with granular AC treatment. The effluent from the photocatalytic process
was filtered through a granular AC adsorber. The total cost was reduced considerably by
using AC adsorption as the last step of treatment (Malato et al., 2001). However, the
saturated AC must be disposed of or regenerated.
Immobilizing TiO2 on AC can result in a synergistic combination of both
adsorption and photocatalysis. On one hand, AC works as the support of nanosized TiO2
photocatalyst and concentrates the pollutants and intermediates around the TiO2; on the
other hand, the photocatalyst can destroy the pollutants thus regenerating the AC in situ.
In traditional thermal or microwave regeneration, part of the pollutants is simply
desorbed. Regeneration on the TiO2/AC composite is achieved by photocatalytic
oxidation of pollutants; thus no post-treatment is needed for the desorbed pollutants.
Besides the advantage of low-temperature in situ regeneration, this novel composite
possesses synergistic functions of simultaneous adsorption and oxidation that are greater
than in the cases when either carbon or TiO2 irradiated with ultraviolet (UV) light is
employed alone (Herrmann and Guillard, 2000; Matos et al., 2001; Khan, 2003).
In summary, the TiO2/AC composite photocatalyst for air pollution control of pulp
and paper mills is a technique worthy of exploration. In Chapter 2, the technical efficacy
of photocatalytic regeneration was studied. Methanol was chosen as the model pollutant.
The effects of humidity and purge air on regeneration were investigated. In Chapter 3, a
dry impregnation method was developed to prepare TiO2/AC composite photocatalyst in
order to improve its performance. The impacts of preparation conditions on the
photocatalytic activity (methanol as the model pollutant) of prepared TiO2/AC were
further studied. In Chapter 4, the removal performance of hydrogen sulfide, which was
chosen as the representative of TRS gases, was tested. In Chapter 5, a novel microwave-
assisted impregnation was developed to prepare TiO2/AC composite photocatalyst. In
Chapter 6, conclusion of this work and recommendations are provided.
One of the major technologies for the abatement of low concentrations of toxic
organic compounds in air is adsorption. Activated carbon (AC) is by far the most
frequently used adsorbent. The term "activated" refers to the increased internal and
external surface area imparted by special treatment processes. Any carbonaceous
materials, such as coconut shells, bones, wood, coal, petroleum coke, lignin, and lignite,
can be converted to AC. AC is manufactured by first dehydrating and carbonizing the
carbonaceous raw material. Activation is completed by heating the carbonized materials
in the presence of an oxidizing gas (usually CO2 or H20) during a controlled oxidation
step. AC is tailored for special end use by both raw material selection and control of the
activation process. ACs typically have a surface area in the range from 600 to 1400 m2/g,
an internal porosity from 55% to 75%. Most of the pore volume is distributed over a
narrow range of pore diameters, usually ranging from 4 to 30 angstroms (Cooper and
As discussed in Chapter 1, successful regeneration is critical to a wider
application of AC adsorption processes. The first significant commercial-scale granular
AC regeneration was the burning of spent AC in sugar refineries around 1828.
Regeneration in modern sense is aimed to restore the adsorption capacity without much
loss of carbon and without much altering the surface of AC. A variety of regeneration
techniques have been suggested, evaluated and applied. These methods are based either
on desorption or decomposition. Figure 2-1 shows an overview of available techniques
for the regeneration of spent AC adsorbents (Sheintuch and Mataov-meytal, 1999). These
regeneration methods have their advantages and disadvantages.
generation of spent AC
Thermal Nonthermal Microbal
Inert gas Solvent extraction
Steam Surfactant enhanced
Hot water Supercntical fluid extraction
Figure 2-1 Overview of available techniques for AC regeneration (Sheintuch and
Thermal regeneration is the most commonly used regeneration method which refers
to processes of drying, thermal desorption and high temperature reactive treatment (700 -
1000 C) in the presence of inert gas or limited quantities of oxidizing gases such as
water vapor of flue gas. The thermal regeneration behavior of the AC loaded with various
compounds has been studied (Sheintuch and Matatov-Metal, 1999). Spent AC undergoes
the following scenario with increasing temperature: drying and loss of highly volatile
compounds occurs at temperatures below 200 C, vaporization and decomposition of
unstable compounds takes place at 200 < T < 500 C and pyrolysis of nonvolatile
adsorbates to form char occurs at 500 < T < 700 C followed by oxidation of the residue
at higher temperatures. Exposure to temperatures of 750 980 C leads to oxidation of
Catalytc HDC T hermal
the residual material as well as that of the carbon itself. The pore structure may be altered
in the latter steps where small pores (< 2 nm) are lost while large pores are created.
There are some disadvantages of thermal regeneration. Firstly, it is usually not
conducted in situ, requiring special regeneration units such as multiple hearth furnaces or
rotary kilns (Sheintuch and Mataov-meytal, 1999; Khan, 2003). Secondly, it typically
results in a continuous loss of 5 -15 % per cycle in adsorption capacity and in surface
area due to the high-temperature (Sheintuch and Mataov-meytal, 1999). The adsorption
capacity may even drop to zero after few cycles. Thirdly, the cost of regeneration is high,
accounting for nearly 50% of the entire treatment technique expenditures (Khan, 2003).
Generally speaking, thermal regeneration is applicable to all. However, it is economically
feasible only for large systems that use more than 500,000 tons of granular AC per year
(Sheintuch and Mataov-meytal, 1999). Thermal regeneration depends both on thermal
desorption and thermal oxidation. The degree of desorption and oxidation depends on the
nature of the adsorbent and the adsorbate and the rate of the process (Sheintuch and
Matatov-Metal, 1999). Some undestroyed pollutants and harmful byproducts may get into
the environment. Therefore, another disadvantage of thermal regeneration is that some
adsorbates, especially highly volatile organic compounds, just desorb. Post-treatment
devices may be needed.
Several other regeneration methods have been suggested in recent years, such as
solvent extraction, supercritical fluid extraction, surfactant enhanced regeneration, and
chemical oxidation using various oxidants (chlorine, chlorine dioxide, peroxide, ozone,
and potassium permanganate). However, these methods have not proven technically
feasible for continuous operation, nor economically viable (Sheintuch and Mataov-
Photocatalytic oxidation is one recently suggested method that can be used to
regenerate spent AC adsorbent and destroy organic adsorbates simultaneously
(Crittenden et al., 1993; Liu et al., 1996). The advantages of heterogeneous
photocatalysis over other regeneration methods include:
1. It can destroy a wide variety of organic compounds (Hoffmann et al., 1995);
2. No post-treatment is needed since the organic pollutants can be mineralized into
nontoxic by-products such as H20, C02, and mineral acids (Hoffmann et al., 1995;
Alberici and Jardim 1997) ;
3. The process can be performed at low temperature (Hoffmann et al., 1995; Alberici
and Jardim 1997; Pitoniak et al., 2003);
4. It can be promoted by solar radiation, resulting in low energy cost (Crittenden et
al., 1997; Malato et al., 2001);
5. On-site regeneration of spent adsorbent and destruction of adsorbed organic
material is provided (Crittenden et al., 1993);
6. The loss of adsorbents due to attrition and burn-off which occurs in thermal
regeneration is less (Crittenden et al., 1993).
Several works were carried out in applying AC adsorption and photocatalytic
regeneration in water treatment. Crittenden and coworkers studied the removal of
trichlorethene (TCE) and p-dichlorobenzene (DCB) from water by Pt-TiO2 coated AC
(Filtrasorb-400). Based on this study, the photocatalytic regeneration process was found
to be limited by reaction rate at the beginning of the regeneration cycle and then by
desorption of the adsorbates from the interior of the AC. The photocatalytic regeneration
for water treatment was a very long process which makes it unsuitable for this practical
application (Crittenden et al., 1996). The diffusibility of molecules in air is faster than
that in water. The gas phase application of photocatalytic regeneration may be suitable.
The objective of this chapter was to study the technical efficacy of photocatalytic
regeneration in gas phase application. The model pollutant, methanol, was removed from
humid air stream by using combined AC adsorption and photocatalytic regeneration. To
achieve this objective, TiO02/AC composite was prepared by a low-cost spray desiccation
method. The removal of methanol and the regeneration performance were tested. The
effect of humidity on adsorption and photocatalytic oxidation and the effect of purge air
on photocatalytic regeneration were further studied.
Catalyst and Chemicals
The TiO2 photocatalyst used in this study was Degussa P25 titanium dioxide.
Research on the use of Degussa P25 has been reported in numerous articles (Lu et al.,
1999; Pozzo et al., 2000; Bhatkhande et al., 2001; Liu et al., 2004; Jeong et al., 2004).
AC used in this study was MeadWestvaco BioNuchar 120 (a wood based chemically AC,
8 12 mesh). The BioNuchar 120 was selected because it possesses the best methanol
adsorption capacity among various carbons tested, according to a previous study (Stokke,
2003). TiO02/AC composite is prepared by a spray desiccation method. The P25 slurry
was sprayed on AC and then dried in a rotary kiln. The methanol/air mixture cylinder
(1000 ppm) was purchased from the Praxair Company.
The specific surface area and pore size distribution of the carbon and TiO02/AC
samples were obtained by N2 adsorption/desorption isotherms performed at 77 K (NOVA
1200, Quantachrome). All samples were dried at 110 oC for 2 h prior to measurement.
The specific surface area was determined by multipoint BET (Brunauer, Emmett, and
Teller) method using the adsorption data in the relative pressure (P/Po) range of 0.05-
0.30. The isotherms were used to determine the pore size distribution using the Barrett,
Joyner, and Halenda (BJH) method. The surface morphology of TiO2/AC composites was
characterized by Scanning Electron Microscopy (JSM6330F, JEOL).
The amount of P25 TiO2 and AC added in this procedure depended on the
experimental conditions. The TiO2 loading on the AC was determined by ash content
analysis. Both virgin AC and TiO2/AC composite were combusted in an oven at 600 oC
for 4 h. After combustion in the oven at 600 oC for two or more hours, the mass of the
material did not further change. Based on the condition that the ash content of carbon
remains the same for both the AC and TiO2/AC, the loading of TiO2 can then be
determined by the following equation,
W02 R, ash (2-1)
where, Wash (%) is the ash content of AC (i.e. the residual mass of virgin AC after
combustion); WR,T (%) is the residual mass of TiO2/AC after combustion; WTio2 (%) is
the TiO2 loading.
Methanol Removal Evaluation
In order to simulate the emissions from paper mills, a low methanol concentration
and high humidity were chosen. The experimental set up is shown in Figure 2-2. One
gram of TiO2/AC was placed on the frit in the reactor which was equipped with an 8 W
black light UV lamp (peak wavelength at 365 nm) at the center of the reactor. The
distance between the UV lamp and the inner wall of the reactor is 9 mm. The outer
diameter of the reactor is 48 mm. The methanol concentration in the influent and effluent
of the reactor during the experiment was measured according to the NCASI chilled
impinger test method (Method CI/SG/PULP-94.02). The methanol in the air flow was
e air flow was
first collected by drawing it through two midget impingers (Analytical Research Systems,
Inc.) in series which were filled with 10 mL of water. The impingers were kept in an ice
water bath (around 2 C) during sampling to enhance collection efficiency. The sampling
time was 1 h. The methanol concentration in the impingers was analyzed by direct
injection into a gas chromatograph (Clams 500, Perkin Elmer) equipped with a flame
ionization detector (GC/FID). Cyclohexanol solution (3 mg L1) was used as an internal
standard. The methanol calibration curve is liner (correlation coefficient grater than 0.99)
throughout the range of the calibration curve (0.5 100 mg L1).The methanol removal
performance of different samples was evaluated by the concentration of methanol in the
effluent. Formaldehyde is one of the possible intermediate products of methanol
photocatalytic oxidation (Arafia et al., 2004). The NCASI chilled impinger method
(Method CI/WP-98.01) was used to measure formaldehyde concentration. A 2.0 mL
aliquot of the impinger sample was mixed thoroughly with 2.0 mL of acetylacetone
reagent and reacted in a water bath at 60 C for 10 min. After cooling to room
temperature, the absorbance of the solution at 412 nm was measured by a
spectrophotometer (DR/4000U, HACH). Formaldehyde concentration was calculated
according to a standard calibration curve. The formaldehyde calibration curve is liner (
correlation coefficient grater than 0.99) throughout the range of the calibration curve (0.5
- 10 mg L1).
Due to the high humidity in the atmospheric emission of pulp and paper mills,
before testing 1.00 g TiO2/AC sample was prehumidified until equilibrium was
established at a constant stream humidity (RH z 80% at 298 K, water vapor concentration
was 19 mgL-1) which was achieved by bubbling the carrier gas (air) through the vessel
with water at the rate of 0.4 Lmin1. Humid methanol laden air was then passed through
the fixed bed of TiO02/AC with or without UV light for 6 h. With UV light irradiation, the
temperature in the reactor rose to 328 K due to the heat release from the UV lamp and the
relative humidity dropped to 16%. The bed depth was 0.4 cm and the empty bed contact
time (EBCT) was about 0.35 s. EBCT is determined by dividing the volume of the carbon
bed (L) by the airflow rate (L min-'). Note that the actual contact time is less than the
EBCT because the carbon fills much of the bed volume. Air flows through only the void
space that is smaller than the entire bed volume. EBCT is used in the study because the
actual contact time is difficult to measure. In order to investigate the effect of EBCT,
methanol removal performance in 7.00 g TiO02/AC (prehumidified) column with and
without UV irradiation was also tested. The bed depth was 2.8 cm and the EBCT was
2 6 8
Figure 2-2 Setup of experimental apparatus: 1. Methanol/Air mixture cylinder (1000
ppmv MeOH); 2. Mass flow controller; 3. Water bubble bottle; 4.
Photocatalytic reactor (equipped with an 8 W black light UV lamp); 5.
Impingers in ice bath; 6. TiO02/AC; 7. Frit; 8. Thermocouple
In order to evaluate the effect of water on methanol adsorption and photocatalytic
oxidation, experiments were also carried out without prehumidification and/or in dry
airstreams. Table 2-1 lists the experimental conditions using 1.00 g TiO02/AC to evaluate
the effect of water on adsorption and photocatalytic oxidation (Set 1-3). In order to
reduce the operation cost, adsorption followed by periodic photocatalytic regeneration of
Ti02/AC was tested. Experiments were carried out following the similar procedure used
in Sets 1-3. Methanol laden air was passed through the fixed bed of 1.00g TiO02/AC
without UV light (T = 298 K) for 6 h. The flow was then cut off and the UV light was
turned on to regenerate the spent carbon for 3 to 9 h (T = 328 K). The adsorption-
regeneration cycle was repeated four times. The methanol adsorption capacity of each
cycle was determined. The effect of purge air flow rate in regeneration was also
Table 2-1 The experimental conditions of methanol removal*
Set Prehumidification RH Operation Regeneration
1 Yes 80% 6 h, with or
2 No 80% 6 h, with or
3 No 0% 6 h, with or
4 Yes 80% 6 h without 9 h between the 1st and the 2nd cycles; 6
UV light h between the 2nd and the 3rd cycles; 3 h
between the 3rd and the 4th cycles; all
without purge air
5 No 80% 6 h without 3 h, without purge air
6 No 0% 6 h without 3 h, without purge air
7 No 0% 6 h without 3 h, with 0.1 L/min purge air
8 No 0% 6 h without 3 h, with 0.2 L/min purge air
*EBCT= 0.35 s
Table 2-1 also lists the experimental conditions using 1.00 g TiO2/AC to evaluate
the effect of water on regeneration (Set 4-6) and the effect of purge air flow rate on
regeneration (Set 6-8).
Results and Discussion
Figure 2-3 shows the SEM images for the TiO2/AC composite prepared by the
described method. The SEM image of the external surface indicates the P25 TiO2
particles were coated on the AC surface by the described method. Although the primary
particle size of P25 is about 20 nm (Degussa AG, TI 1234), the P25 nanoparticles on the
carbon surface were agglomerated. The difference between the SEM images of the
external and internal surface of the TiO2/AC particle indicates that the TiO2 was mainly
coated on the outer surface of the AC particles. The TiO2 loading was 7.61 + 0.20 wt %
measured by the ash analysis method described earlier. Table 2-2 lists the specific surface
area and pore size distribution of AC before and after TiO2 coating. The BET surface area
measurement of one F400 AC (coal based thermally activated, Calgon) sample was
repeated three times. The results were 733, 742, and 745 m2/g. The reproducibility was
0.84%. Because of the inhomogeneity of AC, the reproducibility of different AC samples
(same type of AC) is even higher. The BET surface area measurement of F400 AC
(different samples) was repeated five times. The results were 618, 653, 709, 740, and 745
m2/g. The reproducibility was 8.03%. Therefore, the TiO2 coating didn't significantly
change the BET surface area of the AC which indicated that the TiO2 particles just coated
on the external surface of AC.
Table 2-2 BET surface area and pore size distribution of AC and TiO2/AC
BET Total Pore Volume Micropores Mesopores
(2gm-) (ccg1 -1 (cc-1 (cc-1
AC 1472 1.45 0.46 0.84
TiO2/AC 1380 1.38 0.54 0.79
Figure 2-3 SEM images for the TiO2/AC composite: (a) external surface; (b) internal
Methanol Removal by Adsorption and Adsorption/Photocatalytic Oxidation
The efficiency of the virgin AC and TiO2/AC composites for methanol removal
was evaluated in the presence and absence of UV light. The relative effluent methanol
concentration profiles for the virgin AC and TiO2/AC are shown in Figure 2-4. It is
apparent from Figure 2-4 that the effluent methanol concentration increased quickly
when treated by the virgin AC with and without UV light. The effluent concentration
was higher when treated with UV light probably due to the heat released by UC lamp.
When treated by TiO2/AC without UV light irradiation, a similar adsorption profile was
observed, and the methanol adsorption capacities for the virgin AC and TiO2/AC
composite were similar. However, when methanol was treated by the TiO2/AC with UV
light irradiation, the methanol concentration didn't reach saturation for the duration of the
experiment. The effluent methanol concentration increased during the first 2 h and then
remained almost constant (-35% removal). Figure 2-5 displays the methanol effluent
concentration when using 7.00 g TiO2/AC composite. As shown, increasing the EBCT
increased the methanol removal. Under UV irradiation, the methanol removal efficiency
remained stable around 90% for 12 h. These two sets of experiments show that
photocatalytic oxidation can be used to destroy methanol adsorbates simultaneously and
to extend the AC's usage life.
Effect of Water
In order to investigate the effect of water on methanol adsorption and
photocatalytic oxidation, several experiments were carried out as described in Table 2-1
(Sets 1-3). Figure 2-6 shows the relative effluent methanol concentration.
It is well known that the adsorption of organic vapors on AC can be disturbed by
the presence of water vapor because of the molecular interactions that account for the
various nonidealities exhibited during the coadsorption of the mixture. Typically, the
adsorption of organic compounds exhibit a type I isotherm (Taqvi et al., 1999;
Finqueneisel et al., 2005). As a result, much of the pore volume is filled at low relative
o U.6 ... .
0.2..... 0...O... TiO2/AC/UV
0 1 2 3 4 5 6
Figure 2-4 The relative effluent methanol concentration profiles for the virgin AC and
TiO2/AC (Co= 31.0 ppm, EBCT=0.35 s).
S.. ... .
...... O .
-*- With UV
O 0- Without UV
4 6 8 10 12
Figure 2-5 The relative methanol effluent concentration of 7 g TiO2/AC composite
(Co=36.4 ppm, EBCT = 2.45 s)
pressures. Water dose not interact strongly with carbonaceous solids and exhibits a type
V (S-shaped) isotherm on AC (Taqvi et al., 1999; Finqueneisel et al., 2005). At higher
relative humidity values ( >40 to 50%) the moisture adsorption increases sharply due to
capillary condensation. The adsorbed water fills the small pores in the adsorbent and can
interfere with the adsorption capacity of organic compounds (Noll, 1999). Methanol is a
polar molecule and is miscible in water. Taqvi et al. (1999) reported water promoted the
adsorption of methanol on BPL AC by the ability of alcohol to form H bond with water.
Friqueneisel et al. (2005) developed a model of adsorption isotherms of methanol/water
vapor mixture on microporous AC (hydrophobic surface). This model predicted that the
amount of methanol adsorbed at RH = 42% is equal to the amount of methanol adsorbed
in dry conditions. Below this value the amount of methanol adsorbed is lower than in dry
conditions and above this value the amount is higher than in dry conditions. Gubkina et
al. (2003) studied the adsorption of gas phase methanol on a humidified AC (T = 293 K,
RH = 75%). Within their studied concentration range (methanol < 0.2 mg L-1), the
adsorption for methanol on humidified AC in the presence of water vapor (T = 293 K,
RH = 75%) has a weakly concave shape and can be considered to be linear.
Our results of methanol removal without UV light (Figure 2-6 a) show that high
humidity greatly hindered the methanol adsorption on TiO2/AC. The methanol adsorption
capacity of Set 1 and Set 2 was much lower than that of Set 3. This result is different
form the results of Taqvi et al. (1999) and Friqueneisel et al. (2005). The difference is due
to the different conditions tested. Their results were based on equilibrium adsorption and
our results were based on fixed-bed breakthrough experiments and the bed depth was
short. For Set 1, TiO2/AC was humidified. Compared with Set 2, the methanol adsorption
was higher in the beginning due to the formation of hydrogen bond with water molecules.
However, the difference diminished later because the adsorbed water filled the pores
resulting in less space available for adsorption.
08 ............ _____ 08-
8__ -- --^4 81
04. -- --- .. .-'
02-- Set 1 -- Set 1
02 Set2 02- G.o Set2
-T- Set3 -V- et 3
0 1 2 3 4 5 6 0 1 2 3 4 5 6
Time (h) Time (h)
Figure 2-6 The relative effluent methanol concentration for Sets 1, 2 and 3: (a) without
UV light; (b) with UV light. (Co=31.0 ppm)
With UV light irradiation, the temperature rose to 328 K and the RH decreased to
16%. When adsorption is an exothermic process, the adsorption of both methanol and
water decreased with the increase of temperature. Therefore, the photocatalytic oxidation
is the main mechanism of methanol removal under UV light irradiation. The results of
methanol removal with UV light (Figure 2-6 b) show that methanol oxidation on
Ti02/AC was improved in dry condition. This result is in concord with the results of Kim
and Hong (2002) that the photocatalytic degradation rate of methanol was relatively high
in lower water vapor concentration and that high humidity (RH > 10.6%, T = 318 K)
hindered the photocatalytic degradation of methanol. The influence of water vapor on the
photocatalytic oxidation has been reported by many researchers. Although water plays an
important role in the formation of the hydroxyl radicals, adsorbed water is an effective
electron-hole recombination center leading to less photocatalytic activity (Linsebigler et
n-hole recombination center leading to less photocatalytic activity (Linsebigler et
al., 1995; Chang et al., 2005). Hence, high concentration of water vapor reduces the
adsorption of organic vapor and leads to the inhibition of methanol oxidation.
The advantages of the integration of adsorption and photocatalytic regeneration are
that the photocatalytic oxidation on TiO2/AC can be accelerated by the high concentration
of pollutants eluted from the adsorbent and can reduce the UV irradiation time, thus
improving the economy of the process. Moreover, the regeneration process can be
operated in-situ at ambient conditions. Figure 2-7 shows the adsorption breakthrough
curves of methanol from fresh and regenerated beds of TiO2/AC of Set 4. Figure 2-7
shows the adsorption capacity of each cycle of Sets 4, 5 and 6.
For Set 4, Figures 2-7 and 2-8 show that around 60% of the virgin capacity of the
TiO2/AC was regenerated and increasing the regeneration time did not increase the
regeneration capacity. Therefore, 3 h regeneration was used in Set 5 and Set 6. For Set 5,
the virgin capacity was slightly higher than that of Set 4. This result is consistent with the
result of Figure 2-6a. The regeneration capacity of cycles 2, 3 and 4 of Set 5 was around
80%, 74% and 60% of the virgin capacity, respectively. In the first 3 cycles of adsorption,
the TiO2/AC was not saturated with water vapor. That resulted in a higher capacity of Set
5 compared to Set 4. After 3 cycles of adsorption, the TiO2/AC was almost saturated with
water vapor. Therefore the regeneration capacity of the 4th cycle of Set 5 was similar to
that of Set 4. For Set 6, Figure 2-8 shows that the virgin capacity was 1.8 times of the
virgin capacity of Set 4. This result was consistent with the result of Figure 2-6a.
However, the regeneration capacity of Set 6 was quite similar to that of Set 4. That
indicated just the outer layer of the TiO2/AC was regenerated where the UV light can
reach and where the TiO2 coating is. This is consistent with the finding reported by
Crittenden et al. (1997) that the photocatalytic regeneration process was limited by the
desorption of the adsorbate from the interior of the carbon.
Figure 2-7 The adsorption breakthrough curves of methanol from fresh and regenerated
beds of TiO2/AC of Set 4 (Co =31.0 ppm).
Cycle 1 Cycle 2 Cycle 3 Cycle 4
Figure 2-8 The adsorption capacity of each cycle of Set 4(Co =31.0 ppm), 5 (Co:
34.3ppm) and 6 (Co =27.6 ppm)
A possible way to increase the regeneration capacity is to increase the desorption
rate through heating or purging. Therefore, purge air was used to increase the desorption
rate and the effect of purge air flow was investigated. During regeneration, the methanol
and formaldehyde in reactor effluents were collected and measured. Figure 2-9 shows the
amount of methanol adsorbed, the amount of methanol desorbed in purge air and the
amount of formaldehyde formed in regeneration in each cycle of Sets 7 and 8.
When using 0.2 L min- purge air (Set 8), around 77% of the original capacity was
regenerated after 3 h regeneration. Around 52% regeneration capacity resulted from
direct desorption. When using 0.1 L min- purge air (Set 7), around 80% of the original
capacity was regenerated after 3 h regeneration. Around 24% regeneration capacity
resulted from direct desorption. Without purge air (Set 6), only 40% of the original
capacity was regenerated after 3 h UV irradiation.
Set 7 Set 8
I I Desorption
S 1.5 -
E 1.0 -
1-Set7 2-Set7 3-Set7 4-Set7 1-Set8 2-Set8 3-Set8 4-Set8
Figure 2-9 The methanol adsorption amount in adsorption, methanol desorption and
formaldehyde formation amount in regeneration of each cycle of Sets 7 and 8
(Co =23.7 ppm)
Photocatalytic regeneration of AC is ascribed to both desorption from AC and
photocatalytic degradation on TiO2. Therefore, both the desorption rate and degradation
rate affect the regeneration efficiency. Without purge air, desorption rate from AC was so
low that the methanol adsorbed on the carbon could not be effectively transferred to the
TiO2 photocatalyst. Hence, the photocatalytic regeneration process was limited by the
desorption rate. With 0.1 L min- purge air, the regeneration efficiency was greatly
increased. If the desorption rate was higher than the degradation rate, part of the methanol
directly desorbed without degradation as demonstrated by the comparison of Sets 7 and 8.
With 0.2 L min- purge air, the desorption rate was greatly enhanced although the
regeneration efficiency decreased because of the decreased degradation rate resulting
from the reduction in contact time. Furthermore the formaldehyde formation of Set 7 was
lower than that of Set 8. This further indicates that the contact time was not enough for
complete photocatalytic oxidation. Without purge air (Set 6), there should be no
incomplete degradation product formed because the contact time was much longer. Based
on the above mentioned results, the photocatalytic degradation should reach maximum
when the rate of degradation and the rate of desorption match with each other.
Photocatalytic oxidation can be used to regenerate spent adsorbent (Bio-Nuchar
AC) and destroy methanol simultaneously. The photocatalyst loaded onto the AC has no
significant impact on the adsorption capacity of the AC. Increasing the EBCT from 0.35 s
to 2.45 s can significantly increase the efficiency of the simultaneous adsorption and
photocatalytic oxidation. High humidity can reduce the effectiveness of methanol
adsorption and simultaneous adsorption and photocatalytic oxidation on TiO2/AC. The
regeneration process is limited by desorption of adsorbate from the interior surface of the
carbon. Increasing desorption rate can significantly increase the regeneration capacity.
However, when the rate of desorption is greater than the rate of photocatalytic oxidation,
part of the methanol directly desorbs without degradation.
PREPARATION OF TiO2/AC COMPOSITE PHOTOCATLYST BY DRY
Recently, several works were carried out on the preparation and application of
TiO2/AC composite photocatalyst. Few works prepared TiO2/AC during the activation of
carbon, such as carbonization of a mixture of coal and TiO2 or TiO2 precursor (Aikyo et
al., 1996; Przepiorski et al., 2001). Most works used commercially available AC as the
raw material. The deposition of TiO2 nanophotocatalyst on commercial AC can be
categorized into chemical and physical methods. The chemical methods mainly rely on
hydrolysis of titanium alkoxides such as chemical vapor deposition (El-Sheikh et al.,
2004), impregnation (Harada et al., 1999) and sol-gel method (Capio et al., 2005). The
physical methods use commercially available photocatalyst. Examples include spray
desiccation (Lu et al., 1999), mechano-fusion (Khan, 2003), dip-coating (Jeong et al.,
2004), and spray desiccation technique (used in Chapter 2). However, the best methods
and experimental conditions of carrying out the process are not yet clear because of the
high porosity and non-homogeneous nature of AC. The inhomogeneity makes it difficult
to produce homogeneous distribution of TiO2 on the surface of AC (El-Sheikh et al.,
2004). Furthermore, UV light cannot penetrate into pores rendering TiO2 nanoparticles
deposited inside the pores useless. Therefore, TiO2 nanoparticles deposited on the outer
surface of activated carbon is desired.
Impregnation is a commonly used method in supported catalyst preparation.
Impregnated catalysts are usually obtained from preformed supports by impregnation
with the active phase. The impregnation method involves three steps: (1) contacting the
support with the impregnating solution for a certain period of time, (2) drying the support
to remove the imbibed liquid and (3) activating the catalyst by calcination, reduction or
other appropriate treatment. Two methods of contacting, wet impregnation and dry
impregnation, may be distinguished, depending on the total amount of solution (Perego
and Villa, 1997). The principle of dry impregnation is that the volume of the precursor
solution used in the impregnation is equal to the pore volume of the support, which
results in a better distribution of the solute on the support surface (Huang et al., 2002).
The advantages of impregnation method include:
* The process can be performed continuously in industry.
* It produces uniform coating with good reproducibility and adhesion;
* It controls crystal structure and surface morphology of the TiO2 by controlling the
The spray desiccation coating method used in this Chapter 2 can effectively coat
commercial TiO2 on AC. However, the nanoparticles of TiO2 were agglomerated on the
carbon surface which reduced the photocatalytic efficiency. Besides, this coating method
is a physical method. The TiO2 particles were coated on the carbon surface by weak
physical force. In order to improve the photocatalytic efficiency and the adherence
between TiO2 photocatalyst and carbon support, dry impregnation method was used to
prepare TiO2/AC composite photocatalyst in this chapter. Although impregnation method
has been adopted in TiO2/AC preparation before, very few studies have been carried out
to prepare TiO2/AC composite using titanium tetra-isopropoxide precursor and
information about the preparation details is very limited. This chapter focused on the
understanding of the effects of various preparation parameters hydrolysiss temperature,
hydrolysis time, water vapor concentration, precursor concentration, and moisture content
of AC) of the synthetic method. The prepared TiO2/AC composite photocatalyst was also
evaluated by removing low concentration methanol from humid air stream.
All chemicals were reagent grade or better: titanium tetra-isopropoxide (TTIP)
(Ti(OC3H7)4, 98+%, Fisher); 2-proponal (99.9%, Fisher); methanol/air mixture (1000
ppmv, Praxair); AC (Bio-Nucharl20, wood based chemically activated carbon, 8-16
mesh, pore volume of 1.45 cc/g, MeadWestvaco), commercial TiO2 photocatalyst
The TiO2/AC composites were prepared by dry impregnation using a TTIP solution
(with 80 vol % of 2-propanol) followed by hydrolysis and calcination. The TTIP solution
was stored in sealed container. No hydrolysis in the solution was observed. Before
coating, the AC was heated at 105 C for 4 h to remove moisture. The carbon (3.5 g) was
then mixed with 5 mL of TTIP solution and immediately placed in a hydrolysis reactor,
as shown in Figure 3-1. The diameter of the reactor is 1 inch, and the pore size of the
support frit to better distribute the air flow is 25-50 tm. Air saturated with water from a
humidifier was passed through the reactor at 0.75 L/min for 2 or 24 h. The empty bed
contact time was 0.8 s. The temperature of the humidifier was maintained at 25 or 90 C
by a hot plate with a temperature controller to determine the effect of moisture content.
The temperature of the reactor was controlled at 25, 90, or 175 C, depending upon the
desired synthesis strategy, by heating tape wrapped around the reactor. The various
hydrolysis conditions are summarized in Table 3-1. After hydrolysis, the samples were
dried at 105 C for 4 h to evaporate the adsorbed 2-propanol and then calcined at 300 C
for 2 h in air. This calcination condition was chosen according to Col6n et al. (2002) who
reported that TiO2/AC samples (prepared by means of sol-gel precipitation from TTIP)
calcined under this condition showed higher photon efficiency. Other different
calcination conditions were also investigated for their effects on product properties.
Considering that carbon will partly gasify during calcination in air, calcination in N2 was
also performed. Table 3-2 lists the calcination conditions. Samples 6-8 listed in Table 3-2
were prepared by 20 vol % TTIP solution and hydrolyzed in an open vessel under humid
air (RH = 47 %) for 24 h. BioNuchar AC coated with 9 wt % P25 photocatalyst by spray
desiccation method (used in Chapter 2) was used as the baseline for photocatalytic
Table 3-1 Effect of hydrolysis conditions*
Sample Water bath Hydrolysis Hydrolysis Specific TiO2
No. Temp. Temp. Time Surface Area Amount
(C) (C) (hour) (m2/g) (%wt)
1 25 25 2 1113 6.38
2 25 25 24 1150 5.53
3 25 90 2 1140 5.09
4 25 175 2 1018 2.38
5 90 90 2 1270 8.33
* TTIP concentration was 20 vol % in 2-propanol; calcination conditions were the same
as sample 7.
The effect of TTIP concentration on the prepared TiO2/AC was also tested. The
TTIP concentration varied from 20 vol % to 5 vol % in 2-propanol. Table 3-3 listed the
preparation conditions. Carbon moisture is another important factor that affects the
properties of the prepared TiO02/AC. Sample 11 was prepared under the same conditions
as sample 2 except that the carbon was not dried before preparation.
Table 3-2 Effect of calcination conditions
Sample Specific Temp. Atmosphere Time Total Micropores Mesopores
Surface (oC) (h) Pore (<20A) (200-
Area Volume (cc/g) 500A)
AC 1472 1.45 0.46 0.84
AC 1338 300 air 2.0 NA NA NA
6 1265 300 air 0.5 1.35 0.47 0.84
7 1238 300 air 2.0 1.32 0.46 0.78
8 1479 300 N2 2.0 NA NA NA
* BioNuchar AC calcined at 300 oC in air
* Pore volume distribution wasn't measured because the anatase phase was not present
Figure 3-1 Schematic of the hydrolysis reactor: 1. mass flow controller; 2. water bubble
bottle with water bath and hot plate (with temperature controller); 3. heating
tape; 4. temperature controller; 5. porous frit; 6. carbon loaded with TTIP
solution; 7.hydrolysis reactor
Table 3-3 TTIP concentration and carbon moisture*
Sample Carbon TTIP Conc. TiO2 Amount
(vol %) (wt %)
9 Dry 10 4.27
10 Dry 5 2.01 0.17
11 With moisture 20 5.59 0.33
*Other conditions were the same as sample 2.
The samples were analyzed by X-ray diffraction (XRD 3720, Philips) for
identification of crystalline species in the continuous-scan mode (scanning speed:
0.005o/sec, scanning range: 200 to 500). The major anatase (101) peak at 20 = 25.4 was
analyzed. The specific surface area and pore size distribution of the carbon and TiO2/AC
samples were obtained by N2 adsorption/desorption isotherms performed at 77 K (NOVA
1200, Quantachrome). All samples were dried at 105 C for 2 h prior to measurement.
The specific surface area was determined by multipoint BET (Brunauer, Emmett and
Teller) using the adsorption data in the relative pressure (P/Po) range of 0.05-0.30. As
discussed in Chapter 2, the inhomogeneity of AC resulted in the high deviation of the
measured BET surface area. The isotherms were used to determine the pore size
distribution using the Barrett, Joyner, and Halenda (BJH) method with cylindrical pore
size. The surface morphology of TiO2/AC composites was characterized by Scanning
Electron Microscopy (JSM6330F, JEOL).
The TiO2 loading on the TiO2/AC composites were determined using
thermogravimetric analysis (TGA, STA 449, NETZSCH). Approximately 20 mg of
material was heated up to 1200 C at 10 oC/min under 50 mL min1 air flow. By
comparing the resulting ash content of the AC with that of each TiO2/AC composite, the
TiO2 loading in each composite was calculated according to Equation 2-1 in Chapter 2.
Photocatalytic Activity Evaluation
The same experimental set up used in Chapter 2 (shown in Figure 2-2) was used to
evaluate the photocatalytic activity of the composite material thus prepared. One gram of
TiO2/AC was used each time. The TiO2/AC sample was prehumidified until saturated by
passing humid air through the reactor at the rate of 0.4 L/min for 16 h. Methanol laden air
(RH = 80%) was then passed through the fixed bed of TiO2/AC with and without UV
light for 6 h. The empty bed contact time was about 0.35s. The methanol concentration in
the influent and effluent of the reactor during the experiment was measured by the same
method in Chapter 2. The photocatalytic activity of different samples was evaluated by
comparing the concentration of methanol in the effluent. Each test was repeated at least
Results and Discussion
Effect of Calcination Conditions
In order to optimize the formation of anatase, samples prepared by hydrolysis at
ambient (25 C, RH = 47%) for 24 h and dried at 105 C for 4 h were calcined under
different conditions (Table 3-2). Figure 3-2 shows the XRD patterns of these samples. By
comparing the XRD patterns of samples 6 and 7, it can be seen that the anatase phase was
formed during the heat treatment, and the degree of TiO2 crystallinity increased with
longer treatment time. The comparison of the XRD patterns of samples 7 and 8 shows
that the anatase phase did not form to the same extent when calcined in N2 at the same
The specific surface areas for the samples prepared utilizing different calcination
conditions are also listed in Table 3-2. When calcined in air, carbon was partially
oxidized and resulted in the decrease of surface area. The specific surface area of
BioNuchar AC after calcination (Table 3-2) was 1338 m2/g, and the weight loss was 4.3
wt %. The surface area of sample 8 did not change a lot from the virgin carbon because
the porous structure of the carbon didn't incur serious damage during calcination in N2.
Compared with that of the un-coated Bionuchar, the volume of micropores and
mesopores of samples 6 and 7 didn't change significantly. This showed that TiO2
particles mainly deposited on the outer surface and only slightly in the macropores. This
is further confirmed by SEM analyses to be discussed later.
[Iounts. ) I Anaase
Figure 3-2 XRD patternsSample 7
In summary, the anatase phase was formed during cSaple 8
to a te 2 8 6 e e e e a e
Figure 3-2 XRD patterns of samples calcined under different conditions
In this study, hydrolysis reaction immediately started after mixing the precursor
solution with the carbon. The precursor solution might not have sufficient time to
penetrate the pores of the carbon. Therefore, Ti02 just formed on the outer layer of the
carbon particle. Since UV light can not penetrate into the carbon, the formation of Ti02
formation on the outer layer of carbon particles is desired.
In summary, the anatase phase was formed during calcination at 300 TC for 2 h in
air (sample 7) with minimal carbon loss. This calcination condition was used in the
Effect of Hydrolysis Conditions
Five hydrolysis conditions were tested (Table 3-1) in order to investigate how these
changes effected anatase formation. The specific surface areas for those samples after
calcination and the Ti02 loading derived from TGA measurement described above are
also listed in Table 3-1. Figure 3-3 shows the XRD patterns of samples prepared under
different hydrolysis conditions. Before calcination, there was no anatase peak on the
XRD pattern; i.e. the as-prepared TiO2 without calcination was amorphous. Regarding
the morphology, Figure 3-4 shows the SEM images of their outer surface.
2 sanpl 4
eemlp la 4 hb Fa rr c allc ttim r,
said- I- PM _P1_ I'l
z. Z1 L 30n q1' I" .e 1 so
Figure 3-3 XRD patterns of samples prepared under different hydrolysis conditions
Two simultaneous reactions hydrolysis and polycondensation take place when
TTIP reacts with water. The overall reaction of TTIP occurs as (Seto et al., 1995):
TTIP + 2H20 TiO2 + 4C3H70H (3-1)
The vapor phase hydrolysis rate constant k is given by Seto et al. (1995):
k=3.0 X 1015 exp(-8.43KJ.mol-/RT)
Although the hydrolysis of TTIP in vapor phase is very fast, the hydrolysis of TTIP
adsorbed on the carbon is much slower due to mass transfer limitation. There are two
possible routes for the adsorbed TTIP to be hydrolyzed: 1) the TTIP desorbs first and
then hydrolyzes in vapor phase; 2) the water vapor directly hydrolyzes the adsorbed
vapor directly hydrolyzes the adsorbed
(C ) (d)
Figure 3-4 SEM images of samples prepared under different hydrolysis conditions
(Magnification: 10,000): (a) sample 1; (b) sample 2; (c) sample 3; (d) sample
4; (e) sample 5; (f) 9%P25/AC.
Hydrolysis is the key step for TiO2 formation in this system. Hydrolysis
temperature, hydrolysis time and reactant concentration all effect the hydrolysis reaction.
Kinetically, the higher the temperature, the faster the reaction. According to Kim et al.
(2000), when the hydrolysis reaction is incomplete, a large amount of unhydrolyzed
alkyls remain on the powder. The presence of these alkyls prevents crystallization, thus
keeping TiO2 in the amorphous phase (Kim et al., 2000). Therefore, the higher the degree
of hydrolysis reaction is, the higher the degree of TiO2 crystallinity. Comparing the XRD
patterns for samples 1, 3, and 4, it is clearly demonstrated that the formation of the
anatase phase was favored at high hydrolysis temperature. The result concurs with the
notion that the hydrolysis reaction is faster at higher hydrolysis temperatures and it agrees
with the trend explained by Kim et al. (2000). At the same hydrolysis temperature and
reactants concentration, increasing the reaction time didn't significantly increase the
degree of TiO2 crystallinity as demonstrated by the similar XRD patterns for samples 1
and 2. In addition to temperature, a higher reactant concentration also results in a faster
reaction. Therefore, when the water concentration was higher (sample 5 vs. sample 3),
the reaction rate and the major anatase (101) peak at 20 = 25.4 was higher.
Moreover, the hydrolysis conditions affect the TiO2 loading and the morphology of
TiO2. Table 3-1 shows that the TiO2 loading decreased with the increase in hydrolysis
temperature due to the increase of desorption of TTIP (boiling point: 220 C). The TiO2
loadings for sample 1, sample 3, and sample 4 were 6.38%, 5.09%, and 2.38%
respectively. Although the vapor pressure of TiO2 is extremely low, TiO2 molecules
formed by hydrolysis need to diffuse to carbon surface and then nucleate. Part of the TiO2
molecules may be transported by the purge air. Under the same hydrolysis conditions, the
TiO2 loadings for sample 1 and sample 2 were 6.38% and 5.53% respectively. The TiO2
loading decreased with the increase of hydrolysis time because the loss of TTIP and/or
TiO2 mainly caused by the purge air. The unhydrolyzed TTIP left on the carbon can be
hydrolyzed and/or thermally decomposed in the subsequent processes. The TiO2 thus
formed has higher probability to deposit on the carbon surface because of no purge air.
Desorption of TTIP should work the same way under the same hydrolysis temperature for
sample 3 and sample 5. The TiO2 loading of sample 5 (8.33%), however, is higher than
that of sample 3 (5.09%). The comparison of these two conditions evidences that the
TiO2 loading positively correlates with the rate of hydrolysis when desorption of TTIP
works the same. It should be noted that the theoretical TiO2 loading (assuming all the
TTIP on the carbon is converted to TiO2) is 8.24%. That indicates all TTIP can be
converted to TiO2 under proper hydrolysis conditions.
Figure 3-4 shows the SEM images of different samples. TiO2 formed a rough
particulate coating layer on the sample 4, and dense coating layers on the other samples.
The results presented in the previous paragraph demonstrated that TTIP desorption into
the gas phase followed by vapor phase hydrolysis was the main route for the formation of
hydrolyzed titanium compounds. The subsequent nucleation of hydrolyzed titanium
compounds follows a LaMer type nucleation process (Figure 3-5) (Giesche, 1998).
Heterogeneous nucleation doesn't start until the hydrolyzed titanium compounds
concentration reaches above the saturation vapor pressure. If the critical nucleation
concentration is never exceeded, the system is forced to follow a heterogeneous
nucleation and growth mechanism (Curve I in Figure 3-5). Dense coating layers with a
uniform thickness are formed (Giesche, 1998). When the critical nucleation concentration
is exceeded, homogeneous nucleation is preferred. A lot of nuclei are formed which then
relieve the supersaturation (Curve II). Part of those nuclei may also deposit on the carbon
surface due to heterocoagulation. When the concentration of hydrolyzed titanium
compounds falls below the critical nucleation concentration, heterogeneous nucleation is
preferred again and results in the subsequent uniform growth on the existing nuclei
(Giesche, 1998). Therefore rough particulate coating layers formed on sample 4 because
the TTIP desorption rate was high enough for the formed hydrolyzed titanium
compounds concentration to exceed critical nucleation concentration. Dense coating
layers formed on other samples because the critical nucleation concentration wasn't
exceeded. The SEM image for the sample prepared by spray desiccation (9% P25/AC)
was also shown in Figure 3-4. Although the primary particle size of P25 is about 20 nm,
coating thus prepared consisted of agglomerated particles on the carbon surface. The
distribution of P25 on TiO2/AC prepared by spray desiccation was not well dispersed
compared to the TiO2 on the samples prepared by the described method.
i, Critical nucleation conc. | Nucleation
U // I
Saturation vapor pressure
Figure 3-5 The LaMer diagram related to nucleation and growth mechanism
Figure 3-6 shows the SEM image of sample 2's inner surface. Compared with the
SEM image of sample 2's outer surface (Figure 3-4), it indicates TiO2 was mostly coated
on the outer layer of the AC. This is consistent with the pore size distribution
In summary, the hydrolysis reaction rate positively correlated with the hydrolysis
temperature and the reactant concentration. Low hydrolysis temperature and shorter
hydrolysis time can reduce the vaporization loss of TTIP. Regarding the TiO2
morphology, high hydrolysis temperature results in particulate coating layer with higher
surface area than dense coating layer. Although sample 4 satisfied the requisites for
higher photocatalytic activity, large surface area and high crystallinity (Kominami et al.,
1997), the hydrolysis conditions of sample 2 were used in the following sections. This is
because: 1) carbon is flammable and heating in air poses a hazard of self-ignition; 2) high
TTIP loss of sample 4; 3) low energy consumption of sample 2.
Figure 3-6 SEM image of sample 2's inner surface
Effect of TTIP Concentration
Three different TTIP concentrations (samples 2, 9, 10) were used to evaluate the
effects of this parameter. Figure 3-7 shows the SEM images of the outer surface of
samples 9 and 10. The SEM image of the outer surface of sample 2 is in Figure 3-4. The
measured TiO2 loading amount are listed in Table 3-1 for sample 2 and Table 3-3 for the
The results show that dense coating layers formed on all three samples. Therefore,
TTIP concentration has no direct impact on the morphology of TiO2. The TiO2 loading
increased with the increase of TTIP concentrations. Assuming all of the TTIP on the
carbon is converted to TiO2, the theoretical TiO2 loading values are 8.24, 4.12, and 2.06
wt %, for 20, 10, and 5 vol % TTIP solution. Different from sample 2, the TiO2 loadings
of samples 9 and 10 were very close to the theoretical values. All TTIP converted into
TiO2 on samples 9 and 10. This is because, as discussed previously, that the hydrolysis
rate was not quick enough to hydrolyze all TTIP in gas phase when the TTIP
concentration was 20 vol %.
Figure 3-7 SEM images of samples 9-10. (a) sample 9; (b) sample 10.
Effect of Moisture in Carbon
Activated carbon is a good adsorbent for moisture. The AC used in samples 1- 10
was dried to remove moisture before coating. Sample 11 was prepared under the same
MAIC St I Y'OkV X10.000 1pril WO 13 91 .... i
conditions of sample 2 except that the AC wasn't dried before coating. After heating at
105 C for 2 h, the weight loss of the used AC was 4.73%. Because the moisture content
of AC depends on the storage conditions, its value varies a lot. Figure 3-8 shows the SEM
image of the outer surface of sample 11. The measured TiO2 loading is listed in Table 3-
3. Figure 3-9 shows the XRD pattern of sample 11.
Figure 3-8 SEM image of sample 11
28 25 35
Figure 3-9 XRD pattern of sample 11
Even though the moisture hadn't been removed before impregnation, TiO2 particles
were not formed immediately after mixing with the TTIP solution. The reason is that the
adsorbed water has lower reaction activity than water vapor. The TiO2 loading of sample
2 and sample 11 were similar. Although dense coating layers formed on sample 11
(Figure 3-8), there were quite a few particles as shown. Those particles deposited on the
surface of the formed dense coating layer. This indicated two kinds of hydrolysis of
TTIP: 1) hydrolyzed by adsorbed water; 2) hydrolyzed by water vapor in the air streams.
The TTIP hydrolyzed by adsorbed water formed the dense coating layer. The surplus
TTIP hydrolyzed by water vapor in airstreams and deposited on it. Compared with the
XRD patterns of sample 2 and sample 11, the moisture in the carbon increased the degree
of TiO2 crystallinity. So, it is not necessary to remove the moisture before impregnation.
Methanol removal by the original AC and various TiO2/AC composites with and
without UV light was carried out in order to compare their ability to remove methanol via
simultaneous adsorption and oxidation. The effluent methanol concentration profiles are
shown in Figure 3-10 for the virgin AC and sample 4 as a representative. The effluent
methanol concentration increased quickly when treated by the virgin AC with and
without UV light. When treated by sample 4 without UV light irradiation, a similar
adsorption profile was observed, and the methanol adsorption capacities for the virgin
AC and sample 4 composite were similar. However, when methanol was treated by
sample 4 with UV light irradiation, the methanol concentration didn't reach saturation for
the duration of the experiment. The methanol concentration increased during the first 2 h
and then maintained at about 58% removal. Figure 3-11 shows the methanol removal
efficiency for the different TiO2/AC samples based on the measurements of the last four h
(i.e., after the value did not vary more than 2 ppm), wherein, 9% P25/AC was Bio-
Nuchar AC coated with 9 wt % P25 by spray desiccation.
Figure 3-10 Methanol
1 2 3 4
effluent concentration profiles (Co:
1 2 3 4 5 9 10 11
Average methanol removal efficiency of samples (Co:
Because of the low surface area of dense coating layer compared with particulate
coating layer, photocatalytic activity of samples 1, 2, 3 and 5 was lower than that of
sample 4. The TiO2 loading also played an important role. For those samples with a dense
coating layer, the photocatalytic activity was negatively correlated to the TiO2 loading. It
is well known that the photocatalytic activity decreases with the increase of particle size
due to quantum size effect (Alberici et al., 1997). Although sample 5 had a higher degree
of anatase crystallinity than did samples 1, 2 and 3, its methanol removal efficiency was
the lowest because of the increase in the film thickness. Furthermore, it can be
conjectured that the photocatalytic activity can be further improved by reducing the TiO2
loading to a critical value. Below this critical value, the photocatalytic activity would
decrease because the TiO2 loading is too low to cover all surfaces.
The methanol removal performances of sample 2 (20 vol % TTIP) and sample 9
(10 vol % TTIP) were similar. The methanol performance of sample 10 (5 vol % TTIP)
was better because of lower TiO2 loading than those two samples. Under the same
hydrolysis and calcination conditions, the TTIP concentration only influence the TiO2
loading and had no significant effect on the TiO2 morphology. As mentioned above,
dense coating layers were formed on these three samples and photocatalytic activities
were similar. Considering the effective usage of TTIP, lower concentration was preferred.
The photocatalytic activity of sample 11 was better than that of sample 2 because of
the degree of TiO2 crystallinity. Comparing sample 4 and sample 11, their methanol
removal efficiencies were similar. Sample 11 was therefore used in Chapter 4 due to its
high performance and simplicity in preparation procedure.
In comparison, the methanol removal efficiency of the base line (9% P25/AC) was
lower than samples 2, 3, 4, 9, 10, and 11. The better distribution of TiO2 particles
prepared by the dry impregnation contributes to the better performance. The results
demonstrate that the dry impregnation can be an effective method for preparing the
photocatalytic TiO2/AC composites that have a higher photocatalytic activity for
TiO2/AC composites were successfully prepared by the described dry
impregnation. TTIP was effectively converted to the anatase form of TiO2 by hydrolysis
and calcination. The hydrolysis conditions influenced the loading and morphology of the
formed TiO2. The favorable hydrolysis condition to form particulate coating layer was
high temperature that would result in a concentration exceeding the critical nucleation
concentration. The favorable calcination condition was 300 C in air for 2 h. The TiO2
was mostly coated on the outer surface and macropores of AC. High hydrolysis
temperature results in particulate coating layer and high TiO2 crystallinity. Under the
same hydrolysis and calcination conditions, the TTIP concentration mainly affected the
TiO2 loading and had little influence on the photocatalytic activity. It is not necessary to
remove moisture from carbon before coating. The moisture in the carbon increased the
crystallinity of TiO2. Under proper preparation conditions, the TiO2/AC composite
prepared by dry impregnation outperformed the TiO2/AC composite prepared by spray
H2S REMOVAL BY A TiO2/AC COMPOSITE PHOTOCATALYST PREPARED BY
Properties of H2S and Regulations
Hydrogen sulfide (H2S) is a highly toxic air pollutant which has been identified in
the list of 190 air toxic substances in Title III of the 1990 Clean Air Act Amendments
(Yang, 1992). It is almost as toxic as hydrogen cyanide (HCN). Human exposure to low
concentration of H2S in air can cause headaches, nausea, and eye irritation, and higher
concentration can cause paralysis of the respiratory system, which results in fainting and
possibly death. Concentrations of the gas approaching 0.2% (2000 ppmv) are fatal to
humans after exposure for a few minutes. Hydrogen sulfide is an odorous gas, and its
presence at low concentrations is easily perceived and recognized due to its characteristic
odor of rotten eggs. H2S is perceptible to most people at the concentration in excess of
0.5 parts per billion (ppb) in air. However, as the level of H2S increases, a person's ability
to sense dangerous concentration by smell is quickly lost. If the concentration is high
enough, unconsciousness will come suddenly, followed by death if there is no prompt
rescue (Yang, 1992). The Occupational Safety and Health Administration (OSHA)
Permissible Exposure Limit (PEL) for General Industry as follows:
Exposure shall not exceed 20 ppm (ceiling) with the following exception: if no
other measureable exposure occurs during the 8-hour work shift, exposures may
exceed 20 ppm, but not more than 50 ppm (peak), for a single time period up to 10
The National Institutes of Occupational Safety and Health (NIOSH) recommends a
single 10-ppm 10-minute ceiling PEL for this substance. Control of H2S emissions is
essential to the protection of public health and welfare as well as to the mitigation of
vegetation and material damage problems (Yang, 1992).
Hydrogen sulfide is one of the major pollutants in air emissions from kraft pulp
mills. It is a colorless gas, slightly heavier than air, and moderately soluble in water. The
solubility of H2S decreases with increasing temperatures. The dissolved H2S dissociates
in accordance with the following reversible ionization reactions (Yang, 1992):
H2S -> HS + H (4-1)
HS < S2 + H (4-2)
The dissociation of H2S is effected by pH. The concentration of HS- species is
insignificant when pH values are less than 6, and S2- may not occur at all when pH values
are less than 6.
Processes that have been used to remove H2S from waste gas streams involve either
physical treatment or chemical oxidation. The means of removing H2S depend on the
concentration. At high concentrations, the Claus reaction process is commonly used
(Bouzaza et al., 2004). At lower concentrations, several processes have been developed,
such as transition-metal oxides catalytic oxidation (Li et al., 1997), bifunctional redox
scrubbing process (Petre and Larachi, 2005), and AC adsorption/oxidation (Katoh et al.,
1995; Bagreev et al., 2001; Bandosz, 2002; Le Leuch et al., 2003; Bouzaza et al., 2004).
Some methods require additions of chemicals, and energy expenditure is usually
necessary for physical treatment. Additional environmental problems are encountered
with chemical additives, where the resulting products and by-products require further
treatment and disposal (Yang, 1992).
H2S Adsorption/Oxidation on AC
Activated carbons used for the removal of H2S are generally impregnated with
caustic materials such as NaOH or KOH, or are otherwise modified. Air currents around
odor-generating facilities are initially washed in scrubbers, during which they intake high
levels of humidity, and are then blown through the AC vessels. The carbon bed is mostly
used as a passive support for the caustic material. The disadvantages of caustic carbons
include: 1) the impregnation decreases the ignition temperature of the carbon and poses a
hazard of self-ignition; 2) the oxidation product is elemental sulfur which cannot be
removed from carbon by washing with water; 3) the activity of caustic carbons toward
H2S oxidation ends when the caustic is consumed and the carbon pores are blocked by
sulfur and sodium or potassium salts; 4) its cost is much higher than virgin AC (Bandosz,
The application of virgin unimpregnatedd) AC for the removal of H2S from air has
been investigated. Many papers published in the literature indicate a good removal
efficiency of H2S on ACs (Katoh et al., 1995; Bagreev et al., 2001; Bandosz, 2002; Le
Leuch et al., 2003; Bouzaza et al., 2004). The structural parameters and surface chemistry
of ACs, which are governed by the presence of heteroatoms (such as oxygen, hydrogen,
nitrogen, sulfur, and phosphours), are important for their performance as a H2S remover.
The surface area and pore volume are not the crucial factors that determine the final
performance of carbons as H2S removers (Bandosz, 2002).
The pH of carbon surface has a significant effect on the efficiency of H2S
dissociation and its oxidation to various sulfur species. A moderately low average pH of
the carbon surface is expected to suppress the dissociation of H2S and the creation of HS-
ions. Hydrosulfide ions, when present in low concentration on carbon surface are
oxidized to sulfur oxides which subsequently form sulfuric acid. When the pH value is
very low (< 4.5), only physical adsorption can occur and the concentration of dissociated
H2S ions is negligible. On the other hand, a pH in the basic range promotes the
dissociation of H2S. This results in a high concentration of HS- ions, which are then
oxidized to sulfur. For the efficacy of regeneration of the spent carbon, the formation of
sulfuric acid is preferred. Water is another important factor in the process of H2S
removal. The data collected by Bandosz clearly show that the H2S breakthrough
capacities of prehumidified carbons are about two to six times higher than those of the as
received carbons (Bandosz, 2002). The presence of humidity in airstreams increased
strongly the equilibrium capacity (in batch reactor) of ACs compared to dry atmosphere
(Le Leuch et al., 2003). However, an excess of water flooding was reported as a factor
that decreased adsorption of H2S on AC. The optimal relative humidity for WWP3 AC
cloth was around 60% (Le Leuch et al., 2003).
When H2S is adsorbed onto unmodified ACs at elevated temperatures, elemental
sulfur is the main product. Elemental sulfur also deposits when caustic carbons are used.
At ambient temperature, however, sulfuric acid is formed and results in a significant
decrease in the pH of carbon (Bandosz, 2002). The mechanism of H2S oxidation on AC
proposed by Hidden et al. (1976) requires the presence of a water film, which enables the
dissociation of H2S molecules to HS- ions. The following figure shows the proposed
pathway of H2S oxidation on unmodified AC (Bandosz, 2002).
Under dry air, Le Leuch et al. (2003) proposed the mechanism of H2S oxidation on
AC as follows:
2C, + 0, 2C(O) (4-3)
H2S + C(O) -> S, + H20+Cf (4-4)
where Cf is free active site, maybe an oxygen-containing site or radical C formed at the
carbon surface. The first mechanism is the adsorption of oxygen from air by the carbon.
The second mechanism is the oxidation of H2S by the adsorbed oxygen.
H2S + H20 H5- + H30O
C, + 0.5 02 = C(O)
(Medium strength) Acidic pH -- Strong Basic
HS(ads) + C(O) C(S*) + H20 2 HS-(ads)) +C(O) C(SSH) + H20
C(S*) + 02-, SO (ads)+ Cf C(SSH) + 2HS- C(S3SH) + H20
SO2(ads) + 0.5 02 S03(ads)
SO3(ads) H20 (ads) H2504(ads)
H2SO4+ HS S +x H2O C(SJ)
Strong Acidic pH
H2S H2S(os) Cf- free active sites
Figure 4-1 Proposed pathway of H2S oxidation on unmodified AC in the presence of
water (Bandosz, 2002)
Bandosz (2002) investigated the regeneration of the exhausted carbon by water
washing or heat treatment. The study of the water regeneration showed that after the first
adsorption run around 60% capacity was lost irreversibly. This was the result of the
deposition of elemental sulfur in carbon micropores and the strong adsorption of sulfuric
acid leading to the low pH. The regeneration of spent carbon by heating at 573 K resulted
in 70% capacity loss.
The H2S destruction in gas-phase using TiO2 photocatalyst has been studied by
Canela et al. (1998). These authors observed high degradation yields of 99% at
concentration of 33 to 855 ppmv. Oxygen was shown to be necessary for the
photodestruction of H2S. The photocatalytic process using TiO2/UV-VIS showed that at
217 ppmv, there was no loss of the activity over extended operation periods (3 h). The
main product in the photocatalytic destruction of H2S in gas-phase was sulfate ions.
The objective of this chapter was to evaluate the performance of the TiO2/AC
composite photocatalyst developed in the previous chapter as a H2S remover. Although
H2S is among the pollutants effectively removed by AC (Bagreev et al., 2001), TiO2/AC
composite was used to increase the oxidation of H2S to sulfate.
All chemicals were reagent grade or better: H2S/air mixture (1000 ppmv, Praxair);
DI water (> 17.9 MQ/cm); air (breath grade, Praxair). AC used in this study was
MeadWestvaco Bio-Nuchar 120 (a chemically activated wood based carbon, 8 12
mesh). The TiO2/AC composite photocatalyst used was Sample 11 prepared by dry
impregnation method as described in Chapter 3.
Photocatalytic Activity Evaluation
Figure 4-2 shows the experimental set-up. The H2S (1000 ppm in air) from a
cylinder was mixed with air and diluted to around 40 ppm. The reactor was the same as in
Chapters 2 and 3. The temperature of the reactor was measured by the thermal couple
inserted in the reactor. Each time 1.00 g adsorbent was used. The flow rate was 0.42
L/min and the EBCT was 0.35 s. The H2S removal was tested with UV light and without
UV light (at room temperature). Because the reactor has no cooling system, the
temperature rises to 328+2 K when the UV light is turned on. The H2S removal by AC at
328 K was also carried out to investigate the effect of temperature (the UV lamp was
covered by Al foil and turned on so that there was no UV light in the reactor but the
temperature was the same as the run with UV light). For convenience, the labels RT (at
room temperature, without UV), 328 K (at 328 K, without UV), and UV (under UV Light
irradiation) were used to indicate the experimental conditions.
A Jerome 631-X H2S analyzer (Arizona Instrument LLC) was used to measure H2S
concentrations in the sample gas. The detection range of the analyzer is 0.003 to 50 ppm
and the precision is 5% relative standard deviation. H2S concentration at the inlet of the
reactor was measured when the H2S laden gas bypassed the reactor. By comparing the
outlet H2S concentration with the inlet level, the H2S removal efficiency can be obtained.
Finally, the gas stream passed through an alkaline trap before it was exhausted into the
fume hood. The amount of sulfate formed on the adsorbents was measured according to
the following method. The used sample was taken out and added to 100 ml DI water (>
17.9 MQ/cm) and the suspension was stirred overnight to reach equilibrium (Bandosz,
2002). In the following morning, the sample was filtered and the sulfate concentration in
the filtrate was measured by ion chromatography (ICS-1500, DIONEX).
The H2S removal efficiency and sulfate conversion efficiency were calculated
according to the following equations:
n, = (4-5)
(c, -C) AtxQ
n = 2 (4-6)
r, --n x 100% (4-7)
77= n x100% (4-8)
where, nt (mol) is the total H2S amount entering the reactor; Ci (ppm) is the inelt
concentration of H2S; t (min) is the total time of experiment; Q (L/min) is the gas flow
rate; 24.5 (L/mol) is the gas constant at 25 C and 1 atm; nr (mol) is the removed H2S
amount during the experiment; Ct is the outlet concentration of H2S at time t; At (min) is
the time interval between two data points; ns (mol) is the amount of sulfate formed on the
adsorbent; fir is the H2S removal efficiency; rl, is the sulfate conversion efficiency.
HS Analyzer Vent
Chamber La .1
MFC 2 Three Way On/Off Valves Alkaline
Va le Trap
Figure 4-2 Experimental set-up for H2S removal
Results and Discussion
Figure 4-3 shows the outlet concentration profiles of virgin BioNuchar AC and
TiO2/AC in dry condition. From bottom to top, the six lines are AC/UV, AC/328K,
AC/RT, TiO2/AC/UV, TiO2/AC/328K, and TiO2/AC/RT respectively. In all cases, the
outlet H2S concentration increased with time and was lower than the inlet concentration
during the entire operation period (120 min), indicating that the adsorbents were not
saturated. Hence, the removal amount calculated by Eq. 4- 6 was not the saturation
capacity. The outlet concentrations of TiO2/AC were higher than that of AC in all three
conditions. Figure 4-4 shows the H2S removal efficiency and sulfate conversion
efficiency in each situation. For AC, the removal efficiencies were 91.40 + 1.84 % (RT),
96.84 0.65 % (328 K), and 97.97 0.10 % (UV); the sulfate conversion efficiencies
were 15.01 + 1.93 % (RT), 12.57 1.66 % (328 K), and 14.10 2.15 % (UV). For
TiO2/AC, the removal efficiencies were 45.70 5.64 % (RT), 53.78 0.32 % (328 K),
and 86.62 1.93 % (UV); the sulfate conversion efficiencies were 13.45 4.37 % (RT),
12.11 0.39 % (328 K), and 18.73 + 1.24 % (UV). Details of the effect of each
parameter are discussed below.
30 .---- AC/UV
E0 25 TiO2/AC/328K
H -o- TiO2/AC/UV
0 20 40 60 80 100 120 140
Figure 4-3 Outlet H2S concentration profiles (inlet concentration was 40.5 ppm, RH =
Surface pH As shown, BioNuchar AC is a better H2S remover than TiO2/AC. This
may result from the changes of the surface properties during the preparation. As
discussed earlier, the surface chemistry is the dominant factor in the H2S removal
performance of carbons. The pH of carbon surface has a significant effect on the removal
capacity of H2S. To verify the effect of this parameter, the pH of the virgin BioNuchar
AC and the TiO2/AC composite were measured. 0.4 g of dry sample was added to 20 ml
of DI water ( > 17.9 MQ/cm) and the suspension was stirred overnight to reach
equilibrium (Bandosz, 2002). The sample was then filtered and the pH of the filtrate was
measured by Accumet AP71 pH meter. The results are listed in Table 4-1.
4 Sure val Conversion
Figure 4-4 H2S removal efficiency and S04 conversion efficiency (inlet concentration
was 40.5 ppm, RH = 0%)
Table 4-1 Surface pH of AC and T2/AC2/AC/RT
Sample Surface pH
BioNuchar AC 6.840.13
Figure virgin BioNuchar AC hasiency a moderate pH which is good for both high capacity
and sulfuric acid formation (Bandosz, 2002). After coating TiO, the surface pH was
decreased to acidic range which is not desired for H2S removal.
Table 4-1 Surface High temperature improved the HS removal on AC and T2/AC/AC.
Sample Surface pH
BioNuchar AC 6.84+0.13
The virgin BioNuchar AC has a moderate pH which is good for both high capacity
and sulfuric acid formation (Bandosz, 2002). After coating TiO2, the surface pH was
decreased to acidic range which is not desired for H2S removal.
Temperature High temperature improved the H2S removal on AC and TiO2/AC.
Generally, chemical reaction proceeds faster at higher temperature. The oxidation of H2S
on AC was faster at 328 K than at room temperature. Therefore the removal efficiency
increased. The sulfate conversion efficiencies at 328 K decreased compared to at room
temperature. As mentioned above, elemental sulfur is the main product of H2S oxidation
on unmodified AC at elevated temperature (Bandosz, 2002).
UV light At the same temperature (at 328 K), UV light irradiation improved the
H2S removal on TiO2/AC from 53.78 % to 86.62%. Furthermore, sulfate conversion
efficiency, which is important for regeneration using water wash, was also increased.
This proved that H2S was photocatalytically degraded on TiO2/AC composite. Compared
to the results of Canela et al. (1998), the H2S removal efficiency and sulfate conversion
efficiency of TiO2/AC with the presence of UV light are lower. This is mainly because
the EBCT was much shorter (0.35 s) in the current system. The EBCT used by Canela
and coworkers was 48 s.
UV light also improved the removal of H2S on AC. In Figures 4-3, the outlet H2S
concentration of AC/UV was lower than that of AC (328 K). The difference between
AC/UV and AC (328 K), however, was not clear because the removal efficiencies in both
cases were high. To clarify the effect of UV light, H2S was increased to 59 ppm. It
should be noted that the inlet concentration was calculated according to the flow rates of
the two mass flow controller and the H2S concentration in the cylinder because the inlet
concentration was out of the measuring range of 63 IX H2S analyzer. Figure 4-5 shows
the removal and conversion efficiencies.
With the increase of inlet H2S concentration to 59 ppm, the removal efficiencies
decreased to 85.03 5.07 % (RT), 88.77 3.78 % (328 K), and 94.75 1.96 % (UV)
respectively. The difference between AC at 328 K and under UV irradiation was
increased. This proved that UV light improved the AC performance. The sulfate
conversion efficiencies were 15.72 1.87 % (RT), 20.03 + 1.19 % (328 K), and 15.14 +
0.12 % (UV). Except at 328 K without UV irradiation, the sulfate conversion efficiencies
were almost the same as the case of 40.5 ppm inlet concentration. The peak wavelength
of the used UV lamp (black light lamp) is 365 nm. Photolysis of H2S occurs in light with
wavelength shorter than 300 nm (Khriachtchev et al., 1998). However, Nozawa et al.
(2001) observed the photolysis of H2S under black light lamp (365 nm). This was
inconsistent with the results of Khriachtchev and co-workers (1998). Figure 4-6 shows
the relative outlet concentration of H2S under UV light irradiation without AC or
Ti02/AC adsorbent (i.e. empty reactor).
60 -- AC/328K
0E 40 g
Figure 4-5 H2S removal efficiency and SO04 conversion efficiency of AC (inlet
concentration was 59 ppm, RH = 0%)
The outlet concentration was lower than the inlet concentration. This might result
from the H2S adsorption on the reactor (reactor wall, porous frit, and tubing). With the
presence of UV light, the outlet concentration was not lower than that without UV light.
After 30 minutes, the outlet concentration with the presence of UV light was higher than
that without UV light because of the rise of temperature and decrease of adsorption. The
results proved there was no photolysis under the black light lamp used (365 nm).
Therefore, photolysis is not responsible for the improved H2S removal on AC under the
UV light in the current system (365 nm). There are two possible mechanisms for the
enhanced removal and conversion of AC under UV. It is possible that UV light induces
heterogeneous photolysis of H2S adsorbed to the external surface of the carbon. Another
possible mechanism is the functional groups of AC, such as carbonyl group, are excited
under UV light irradiation. The mechanism wasn't further explored due facility
limitation. To the best of the author's knowledge, this phenomena and mechanism have
not been reported
1.0 ................--^ ^
0.6 -*- Without UV
O-- With UV
0 10 20 30 40 50 60
Figure 4-6 Outlet H2S concentration passing empty reactor (inlet concentration was 45
ppm, RH = 0%)
BioNuchar AC itself is a good H2S remover. Without UV irradiation, a fraction of
adsorbed H2S was already oxidized to sulfate. The presence of UV light improved H2S
removal efficiency in dry airstreams. After coating TiO2, the H2S removal efficiency of
TiO2/AC decreased due to the surface change. Under UV light irradiation, H2S removal
efficiency and sulfate conversion efficiency of TiO2/AC composite increased. The
sulfate conversion efficiency of TiO2/AC composite was higher than that of AC which is
preferred for water regeneration.
MICROWAVE-ASSISTED PREPARATION OF TiO2/AC COMPOSITE
Microwave is a form of energy that falls at the lower frequency end of the
electromagnetic spectrum. It is defined in the 300 to 300,000 megahertz (MHz) frequency
range. A microwave is comprised of two elements: an electric field and a magnetic field
(Figure 5-1). Microwave was first developed primarily for communications. Around 40
years ago, the heating effect of microwave was discovered. Microwave, as a heating
method, is a dielectric heating method using dipole rotation and ionic conduction,
wherein the applied energy is converted into heat by mutual interaction between media.
In contrast to all other commonly used heating methods, microwave allows volumetric
heating of materials. Processes based on microwave heating find many industrial
applications such as cooking, tempering and thawing, and curing of wood and rubber
The electromagnetic energy of microwaves is dissipated by substances through
three different mechanisms (Bathen, 2003):
* Magnetic losses in ferromagnetic materials;
* Ohmic losses in conducting materials;
* Electric losses caused by electromagnetic inhomogeneities like dipoles and ions.
The following explains the mechanism of electric loss. The electric field of
microwave interacts with molecules through either a dipole or ionic conductivity in the
molecules themselves. As the electric field changes from positive to negative, the positive
and negative ends of the dipole (or the positive and negative ions) seek to align with the
opposite field. This causes molecular rotation (Figure 5-2). The rotational motion of the
molecule as it tries to orient itself with the field results in transfer of energy. The
coupling ability of this mechanism is related to the polarity of the molecules and their
ability to align with the electric field. There are a number of factors that will ultimately
determine the dipole rotation coupling efficiency; however, any polar species (solvent
and/or substrate) present will encounter this mechanism of energy transfer.
E = electric field
H1 = magnetic field
S=- wavelengrh (12,2 cm for 2450 MHz)
c = speed of light (300,000 kmis)
Figure 5-1 Schematic description of a microwave (Hayes, 2002).
Microwave Electric Field Interaction With Water Molecule
E E E
SWater molecule ns
t=o t (.3ns
S ter molecule
\H H/ Water molecule
8: Water molecule 6; H H
The water dipole moment The dipole moment rotates in The dipole moment rotates in the opposite
is aligned with the field. an attempt to follow the field. direction trying to align with the field.
Figure 5-2 Rotation of molecules with microwave (Hayes, 2002).
molecules with microwave (Hayes, 2002).
Of the four frequencies available for industrial, scientific, or medical applications
(915 MHz, 2450 MHz, 5800 MHz, and 22,125 MHz), 2450 MHz is the most commonly
used. At this frequency, the energy supplied to the system is equal to 0.037 kcal/mol.
Even the weakest intramolecular bonds (Van der Waals forces) are 48 kcal/mol.
Therefore, microwave only serves to rotate the molecules, and will neither break nor
form any additional bonds. (Hayes, 2002)
Materials like glass, octane or nitrogen (with none of the properties mentioned
earlier), are transparent to microwave. Microwave energy transfer is not dependent on
thermal conduction and is very rapid, e.g. heating water in a beaker. With conventional
(conductive) heating, as a hotplate is slowly heated, thermal energy is slowly transferred
to the beaker. That heat must then be transferred again through each layer of water until
the temperature is uniform. Microwave transfers energy differently. Glass vessels are
transparent to microwave energy, enabling the water molecules to absorb the energy
directly, meaning more efficient energy transfer. As a consequence, thermal energy is
produced only as a byproduct (Hayes, 2002).
The property that describes how well a material absorbs microwave energy and
converts it into heat is the dielectric loss factor ( ). In general, the heat-up rate of a
material in an applied electric field is proportional to the dielectric loss factor, frequency,
and the square of the strength of the electric field inside the material. The penetration
depth (6), defined as the distance from the surface of the bed at which the power decay to
1/e of its value at the surface, is inversely proportional to the dielectric loss factor
(Bathen, 2003) as
8 A= (5-1)
where Xo is the wavelength of the radiation in free space, F is the dielectric constant, and
F is the dielectric loss factor. This equation suggests that a material with a high loss-
factor, such as activated carbon, exhibits short penetration depths, making it difficult to
uniformly heat a large bed of the material. Table 5-1 lists some calculated values
according to the above equation (Bathen, 2003).
Table 5-1 Penetration de th of microwaves (2.45 GHz)
Compound Temperature Penetration depth
Air 25 infinity
Water (distilled) 25 1.42
Water with 29 g/1NaCl 25 0.38
Methanol 25 0.64
Tetrachlorocarbon 25 3210
Zeolite NaX 20 15.71
Zeolite DAY 20 >100
Aluminum 25 0.02
Aluminum Oxide 25 656
Advantages of microwave heating include (Bykov et al., 2001):
7. Reducing energy consumption and process time: microwave energy transforms into
heat inside the material, which results in significant energy savings and reduction in
8. Rapid and controllable heating: the heat is transmitted directly inside the sample,
and the heating rate is controlled by the power of the microwave source.
9. Inverse temperature profile: the heat is dissipated to the environment through the
surface. Therefore, the temperature inside the sample is always higher than on the
surface (including the case when a constant temperature is maintained). In order to
reduce this temperature difference the sample can be surrounded by a layer of a
thermally insulating material which reduces heat loss from the surface of the
10. Selectivity: microwave heating is based on the capacity of a material to absorb the
electromagnetic energy. The selective absorption of microwave irradiation opens a
way to implementation of selective heating.
11. Possibility of surface processing: for a given material and frequency, if the
microwave penetration depth is small enough, microwave heating takes place only
in the near surface layer of the material.
Besides these advantages, there are many experimental observations that suggest
non-thermal microwave effects that accelerate reaction rates, alter reaction pathways, and
enhance mass transport and result in unique properties in polymers, ceramics and
composites. Still, the specific mechanism of this effect needs further theoretical and
experimental investigation (Bykov et al., 2001).
There are several reports in the literature of TiO2 powder preparation by microwave
processing. Ramakvishnan (1999) synthesized titania ceramic powders from TTIP in
methanol under microwave irradiation. Ayll6n et al. (2000) prepared anatase TiO2
powder from fluorine-complexed titanium aqueous solution using microwave irradiation.
Crystalline anatase TiO2 powders with submicron size were obtained in a short time, at
low temperature and atmospheric pressure. However, the obtained materials did not show
photocatalytic activity. Wilson et al. (2002) reported microwave hydrothermal treatment
of colloidal TiO2 suspensions. Their results showed that the microwave treatment allowed
for rapid heating rate and rapid recrystalization. This resulted in highly nanocrystalline
material and reduced process time and energy with respect to the conventional
hydrothermal treatment. Yamamota et al. (2002) studied hydrolysis and polycondensation
of TTIP in alkanedial solvent under microwave irradiation to obtain anatase TiO2
nanocrystallite. Hart et al. (2004) synthesized anatase TiO2 by a sol-gel method followed
by microwave heat treatment (silicon carbide was used as a microwave susceptor). Using
microwave processing, crystallization of TiO2 was significantly faster and occurred at
lower temperature than by conventional furnace treatments. While there have been many
studies on the preparation of TiO2 powder by microwave processes, there are only few
literature relevant to the preparation of TiO2/AC by microwave process. Lee et al. (2004)
reported the preparation of TiO2/AC by a modified sol-gel method followed by drying
and microwave heat treatment. The prepared material successfully degraded microcystin-
LR in water. However, they simply used microwave in the heat treatment to accelerate
crystallization without taking advantage of enhanced chemical reactions.
The objective of this chapter was to prepare dispersed TiO2 photocatalyst on AC
via microwave-assisted impregnation method for the treatment of HVLC emissions from
paper and pulp mills. Methanol was chosen as the target pollutant because it is the major
pollutant in the HVLC emissions. With microwave heating, the microwave supply energy
to the carbon particles themselves. Some carbons have free electrons whose displacement
is restricted by grain boundaries. When these carbons are subjected to the
electromagnetic field of microwave, space charge polarization takes place. Entire
macroscopic regions of the material become either positive or negative synchronizing
their orientation with the field. This mechanism is often called the Maxwell-Wagner
effect. At low frequency the polarization synchronizes its orientation with the field, but as
the frequency increases there is a phase lag between the polarization of energy and
heating of the carbon particles (Carrott et al., 2001). One advantage of microwave-
assisted preparation of TiO2/AC is the use of AC as the microwave susceptor. In this
way, heat-enhanced processes such as desorption, hydrolysis, and crystallization are
favored, eventually leading to quick formation of crystallized TiO2.
All chemicals were reagent grade or better: Titanium tetra-isopropoxide (TTIP)
(Ti(OC3H7)4, 98+%, Fisher); 2-proponal (99.9%, Fisher); methanol/air mixture (1000
ppmv, Praxair); DI water; F400 AC (coal based thermally activated carbon, effective
particle size 14-20 mesh, Calgon). F400 AC was chosen because the thermal stability of
BioNuchar AC used in previous chapters was not as good as F400 AC.
A commercial microwave oven (SHARP CAROUSEL II) was used as the
microwave source. The capacity of the oven is 0.8 ft3, the full power is 800W, and the
microwave frequency is 2450 MHz. It should be noted that the microwave power output
in a common microwave oven can not be adjusted. When the power level is adjusted, it
merely regulates the fraction of synchronic irradiation time. At high, medium, medium
low, and low power level, the fractions of synchronic irradiation time are 100%, 60%,
40%, and 20% respectively.
Before experiments, AC was dried at 383 K for 2 hours. A typical procedure of the
preparation of TiO2/AC by microwave-assisted impregnation was as follows: F400 AC
(1.00 g) was impregnated with 1 mL 50% TTIP 2-propanol solution in a 20 mL glass vial
and then stood for overnight to allow the solution to penetrate the pores of carbon. A
designated amount of water was added into the vial and the vial was immediately
exposed to microwave irradiation. The hydrolysis of TTIP produced a supersaturated
vapor of TiO2, which nucleated and grew to form TiO2 particles on the AC surface. After
a designated period of time, the sample was taken out and weighed. The effects of
microwave power level, H20/TTIP ratio and irradiation time on the activity of TiO2/AC,
which were characterized by the methanol removal efficiency, were explored. Table 5-2
lists the preparation conditions.
Table 5-2 Preparation conditions and characterization of Ti02/AC
Sample MW Power H20/TTI Irradiation TiO2 BET Total
No. P Ratio Time Loading Surface Pore
(wt %) Area Volume
12 Medium 28 20 min 8.590.05 753.06 0.4091
13 Medium Low 28 20 min 8.780.21 723.52 0.3928
14 Low 28 20 min -- -- --
15 Medium 4 20 min 9.420.52 887.48 0.4561
16 Medium 28 30 min 9.270.12 905.19 0.4652
17 Medium 4 30 min 8.940.46 891.83 0.4504
The specific surface area and pore size distribution of the carbon and TiO2/AC
samples were obtained by N2 adsorption/desorption isotherms performed at 77 K (NOVA
2200, Quantachrome). All samples were outgassed at 393 K on the outgas station
overnight prior to measurement. The specific surface area was determined by multipoint
BET using the adsorption data in the relative pressure (P/Po) range of 0.005-0.20. The
isotherms were used to determine the total pore volume of different samples. The surface
morphology of TiO02/AC composites was characterized by Scanning Electron Microscopy
(JSM 6330F, JEOL). The samples were also analyzed by X-ray diffraction (XRD 3720,
Philips) for identification of crystalline species in the continuous-scan mode. The
scanning speed was 0.05 osec'l or 0.005 osec'l and the scanning range was from 200 to
500. The major anatase (101) peak at 20 = 25.40 was analyzed. The TiO2 loading on the
AC was estimated by ash content analysis as described in Chapter 2.
Photocatalytic Activity Evaluation
As used in Chapters 2 and 3, low methanol concentration and high humidity were
chosen. The experimental set up is shown in Figure 2-2. 2.00 g of Ti02/AC was placed
on the frit in the reactor. The methanol concentration in the influent and effluent of the
reactor during the experiment was measured by the same method used in Chapters 2 and
The TiO2/AC sample was humidified until equilibrium was established at a
constant stream humidity (RH= 80% at 298 K, water concentration was 19 mgL-1).
Humidification was achieved by bubbling the carrier gas (air) through the vessel with
water at the rate of 0.4 L/min. Humid methanol laden air (water concentration was 19
mgL-1) was then passed through the fixed bed of TiO2/AC with or without UV light for 6
hours. With UV light irradiation, the temperature of the reactor rose to 328 K from room
temperature due to the heat release from the UV lamp and the relative humidity was 16%
at 328 K. The bed depth was 0.4 cm and the empty bed contact time (EBCT) was
Results and Discussion
Carbon Weight Loss and TiO2 Loading
In order to investigate the stability of AC under microwave irradiation, 1.00 g
virgin F400 AC was placed in a 20 mL glass vial and then exposed to microwave
irradiation. After a designated period of time, it was taken out and weighed. Figure 5-3
shows the weight loss curves of F400 AC under different power levels of microwave
The carbon did not start to combust under the tested conditions. However, the
results (Figure 5-3) show that under high power level F400 activated carbon continue to
lose weight in 30 minutes. Under medium power level F400 AC just lost weight in the
first 10 minutes and the weight loss was much less than that under high power level.
Because of the instability of F400 AC under high power level, this level was not used in
subsequent study and only power levels equal to or less than medium were used. The
weight loss of F400 AC is likely due to the removal of some surface functional groups
and/or the oxidation of carbon. The ash content of F400 AC before and after microwave
process listed in Table 5-3 proved that the carbon lost some flammable composition in
the microwave process. The pH of the carbon suspension, which provides some
information about the average acidity/basicity of carbons, was measured following the
procedures described in Chapter 4. The pH of F400 AC before and after microwave
process are also listed in Table 5-3. The results showed the pH of carbon surface
increased after microwave process. This may result from the decomposition of
oxygenated surface groups. The carboxylic and lactonic functional groups on AC surface
are less stable, which decompose at temperatures as low as 570 K and evolve CO2 (Li et
al., 2003). The phenol, quinine, and carbonyl groups are fairly stable, which only
decompose at temperature above 770 K and evolve CO (Li et al., 2003). Li and co-
wokers (2003) observed the starting decomposition of surface functional groups of AC in
He flow as low as 400 K. Li et al. (2003) also reported that AC treated in air at 693 K
resulted in the increase of surface area and pore volume. This proved that AC may be
oxidized by air at elevated temperature.
Table 5-3 lists the ash content, BET surface area, and total pore volume of virgin
F400 AC and the F400 AC after 20 minutes medium level microwave irradiation. As
shown, the microwave process did not significantly affect the specific surface area and
total pore volume of F400 AC. This also proved that the F400 AC was stable under
medium power level irradiation. Because carbon and 2-propanol solvent are flammable, it
may not be safe to run the microwave process in air, if vapor pressure is high. The
explosion limits of 2-propanol are 2.0 vol % (lower) and 12.7 vol % (upper). In each
experiment, just 1 g of carbon and 1 mL of 50% TTIP solution were used. The volume
percentage of 2-propanol vapor in the used microwave oven was less than 1 vol %.
Hence, it was safe under medium or lower power level.
--- Medium Level
0.92 -0- High Level
0.90 i I I I
0 5 10 15 20 25 30 35
Irradiation Time (min)
Figure 5-3 F400 AC Weight Loss under medium level MW irradiation
Table 5-3 Ash content of F400 AC before and after microwave process
Ash Content BET Surface Total Pore Surface pH
(wt %) Areal Volume
Virgin F400 5.540.03 1005.10 0.5290 6.180.02
MW F400* 5.690.06 1025.19 0.5289 6.83 0.03
* MW F400 is virgin F400 carbon exposed to medium level MW irradiation for 20
Figure 5-4 shows the weight loss in the preparation of samples 12- 15. The results
showed that the higher the microwave power was, the faster the weight loss occurred.
Considering that heat was dissipated to the environment through the surface
1 The BET surface area of F400 AC listed in this table is higher than that listed in Chapter 2. This is
because the relative pressure ranges of the adsorption data used in the multipoint BET calculation were
different. The relative pressure range used here was 0.005 -0.20. The relative pressure range used in
Chapter 2 was 0.05 -0.30.
continuously, the bulk temperature and heating rate of the sample under different
microwave power levels should be different. Although the bulk temperature wasn't
measured because of the facility limitation, it can be conjectured that the bulk
temperature and heating rate increased with the power level. The weight loss of sample
15 was faster than that of sample 12 although the microwave power levels used were the
same. The evaporation of water and solvent was responsible for the difference. Due to
the different heat capacities of these two materials, the heat dissipation rate was also
different, which resulted in bulk temperature of sample 15 being higher than that of
sample 12. Assuming that all the TTIP was converted into TiO2, 0.13 g TiO2 would be
generated. The weights of sample 12-15 were 1.13 g, 1.13 g, 1.32 g and 1.10
respectively. Clearly, the low power level was not enough to vaporize the chemicals and
there was still volatile material (solvent, by-products, and/or unhydrolyzed TTIP)
adsorbed on sample 14. Hence, low power level was not considered in subsequent
experiments. Considering the weight loss of virgin AC under microwave irradiation (0.03
g), the expected weight was 1.10. Therefore, the TTIP conversion of samples 12 and 13
perhaps were also incomplete.
The TiO2 loading, BET surface area, and total pore volume of TiO02/AC sample are
also listed in Table 1. The ash content of F400 AC after microwave process was used to
calculate the TiO2 loading. The various preparation conditions listed in Table 5-2 did not
exhibit significant influences on the TiO2 loading, specific surface area, and total pore
volume of samples 15, 16 and 17. Compared with samples 15, 16 and 17, these properties
of samples 12 and 13 were lower. This further supports that the TTIP conversion of
samples 12 and 13 was not complete. Compared with virgin F400 AC, however, the
specific surface area and total pore volume of samples 15 17 were lower which resulted
from the TiO2 deposited on the carbon surface that blocked the pores.
2.2 \ Sample 12
........ 0 ....... Sam ple 13
S2.0 ------- Sample 14
S\. ----v .. Sample 15
"1. C 'T-- -- -
\ ...... ..
0 5 10 15 20
Irradiation Time (min)
Figure 5-4 Weight loss curves of TiO2/AC samples
Figure 5-5 shows the SEM images of TiO2/AC prepared by the described method
and the virgin carbon. It demonstrates that TiO2 particles were formed on the carbon
surface. The preparation conditions didn't significantly affect the TiO2 morphology
because the actual power output rate at each condition was the same; just the irradiation
time was changed. Figure 5-6 shows the SEM image of the cross-section of sample 12.
Figure 5-7 shows the SEM images, EDS spectra, and the EDS mapping of Ti element on
section 1 (external surface) and section 2 (internal surface) in Figure 5-6. Obviously, the
formed TiO2 was rich on the external surface of carbon. The deposition of TiO2 on the
external surface is preferred since UV light cannot penetrate into inner pores. Because
water was added later, thermal reaction of TTIP inside pores could yield TiO2 on the
Figure 5-5 SEM images of TiO2/AC samples: (a) Sample 15; (b) Virgin Carbon
Figure 5-6 Cross-section of Sample 12
internal surface. Under microwave irradiation, however, part of TTIP inside pores could
also be desorbed and/or evaporated out before decomposition. These TTIP molecules
reacted with water and then deposited on outer surface.
o 5 10 15 20
(I c) (2c)
Figure 5-7 Region 1 and Region 2 in Figure 5-6: (a) SEM images; (b)EDS spectra;
(c)EDS mapping of Ti element.
Figure 5-8 shows the XRD patterns of different samples. Fast scanning speed, 0.05
/sec was used initially. If any clear peaks were detected, slow scanning speed, 0.005
/sec was used to verify the result. No significant peak was detected on samples 12 and
13. However, some peaks were detected on samples 15 -17. Therefore slow scanning
0 5 10 15 20
speed was used to rerun samples 15-17. The results revealed that anatase phase was
formed on samples 15 and 17, and rutile phase was formed on samples 16 and 17 that
resulted from the high bulk temperature. The formation of anatase TiO2 is important due
to its photocatalytic performance, which will be explained in the next section.
16088 I0i^ f Sample 17
0 .. . ...- .. ..... ,:,,. ....,,.
20 25 38 35 48 45 [201 50
Figure 5-8 XRD patterns of different samples (scanning speed: 0.05 o/sec for samples 12
and 13; 0.005 /sec for samples 15 -17)
Methanol removal testing
Methanol removal by the original F400 AC and TiO2/AC composites with and
without UV light was carried out in order to compare their ability. The inlet methanol
concentration was 22.4 ppm. The relative effluent methanol concentration profiles for the
virgin AC and TiO2/AC are shown in Figure 5-9. It is apparent from Figure 5-9 that the
effluent methanol concentration increased quickly when treated by the virgin AC with
and without UV light. When treated by TiO2/AC without UV light irradiation, a similar
adsorption profile was observed, and the methanol adsorption capacities for sample 12
composite was actually lower than that of the virgin carbon due to the lower surface area.
However, with UV light irradiation, the methanol concentration did not reach saturation
for the duration of the experiment. Another thing that should be mentioned was that
acetone and 2-propanol were detected by GC in the impinger samples of samples 12 and
13 (with UV light). Acetone is the product of reaction between TTIP and the carbonyl
groups on activated carbon surface (Tatsuda et al., 2005). Because acetone and 2-
propanol weren't the products of methanol degradation, their presence further proved that
the TTIP conversion of samples 12 and 13 was incomplete. Regarding sample 15 (with
UV light), the effluent methanol concentration increased during the first 2 hours and then
maintained at about 53% removal. Therefore, the average of the last four data was used to
calculate the average methanol removal efficiency for subsequent analysis.
SI I ,1 ---
0.8 -- -- ---
---_ ---- Sample 12
0.2 --...--v- Sample 12/UV
---- -- Sample 15/UV
0 1 2 3 4 5 6
Figure 5-9 Methanol effluent concentration profiles: inlet methanol concentration was
22.4 ppm, inlet water concentration was 19 mgL-1, EBCT was 0.35 s
Figure 5-10 shows the average methanol removal efficiencies of samples 15 -17
with UV irradiation. No acetone and 2-propanol were detected by GC in the impinger
samples of these samples, and their average methanol removal efficiencies were similar.
This revealed that increasing the irradiation time can not further increase the
photocatalytic activity once the TTIP conversion was completed.
E 0.2 -
S 0.1 -
Sample 15 Sample 16 Sample 17
Figure 5-10 Average methanol removal efficiencies: inlet methanol concentration was
38.8 ppm, inlet water concentration was 19 mgL1, EBCT was 0.35 s
Under medium level of 800 W microwave irradiation, anatase TiO2 was quickly
formed from TTIP precursor in a short time, at atmospheric pressure. F400 AC was stable
under this energy level, and the formed submicron TiO2 particles were rich on the
external surface of carbon. When the TTIP conversion was completed, the irradiation
time and water/TTIP ratio would no longer pose any significant impact on the final
product. The prepared TiO2/AC composite photocatalyst showed lower adsorption
showed lower adsorption
capacity for methanol than virgin carbon due to pore blockage by the newly formed TiO2
particles. Photocatalytic oxidation of methanol from humid air was successfully
accomplished by the composite, and the material did not reach saturation for the duration
of the experiment.
CONCLUSIONS AND RECOMMENDATIONS
Pulp and paper mills rank in the ten largest industrial activities in the United States.
More and more stricter air emission regulations urged researchers and engineers to
explore advanced control technologies. This study was carried out to develop TiO2/AC
composite photocatalyst as an alternative technique for the treatment of high volume low
concentration air emissions from pulp and paper mills.
TiO2/AC composite prepared by conventional methods as well as microwave
methodology was characterized and evaluated by activity tests for the degradation of
methanol and H2S, two important pollutants of HVLC air emissions from pulp and paper
First, the technical efficacy of photocatalytic regeneration of the spent carbon was
investigated. The model pollutant, methanol, was removed from airstreams using
TiO2/AC composite prepared by spray desiccation method. The spent adsorbent was
regenerated under UV light irradiation. Photocatalytic regeneration of TiO2/AC is
ascribed to both desorption from AC and photocatalytic degradation on TiO2. Increasing
desorption rate by using purge air greatly increased the regeneration capacity. However,
when the desorption rate was greater than the photocatalytic oxidation rate, part of the
methanol was directly desorbed without degradation. So, improving the photocatalytic
degradation rate is important for the application of photocatalytic regeneration in the gas
Dry impregnation was developed to improve the photocatalytic activity of TiO2/AC
composite. Titanium tetra-isopropoxide (TTIP), the precursor, was effectively converted
to the anatase form of TiO2 on the carbon surface after hydrolysis and calcination. Under
proper preparation conditions, the TiO2/AC composite outperformed the composite
prepared by spray desiccation method in removing methanol. Another model pollutant,
hydrogen sulfide, also photocatalytically degraded on TiO2/AC composite. The sulfate
conversion efficiency of TiO2/AC composite was higher than that of AC. A high sulfate
conversion efficiency is desired since the formed sulfate can be easily washed away.
Finally, a novel microwave-assisted impregnation was developed for TiO2/AC
composite photocatalyst preparation. Due to the volumetric heating and selective heating
of microwave, the solvent and by-products were quickly removed which required less
energy and shorter processing time. The formed submicron TiO2 particles were mainly
deposited on the external surface of carbon.
Based on the conclusions presented above and the experience gained in this
research, recommendations can be made to help further advance the application of
TiO2/AC composite photocatalyst.
1. The moisture in carbon is beneficial to the hydrolysis of TTIP. The effect of
moisture content on the performance of final product should be further studied.
There will be a relationship between the moisture content and the hydrolysis
degree. When the moisture reaches a critical value, it is possible that no extra water
is needed for the hydrolysis of TTIP. An additional benefit is that the hydrolysis
and drying can be combined into one step. However, there is a critical issue of how
to control the moisture content of the carbon and its distribution to deliver optimal
2. The UV light (peak wavelength is 365 nm) irradiation improved the H2S removal
and sulfate conversion efficiency on virgin BioNuchar AC. The mechanism has not
been reported in prior work. Two possible mechanisms can be hypothesized: 1) the
adsorbed H2S and/or the oxidation products of H2S on AC (mainly elemental
sulfur) could be excited under the irradiation of UV light which may create a more
adsorbable species; 2) the surface of AC can be modified by UV light irradiation.
These hypotheses are worthy of exploration.
3. The application of the suggested microwave-assisted impregnation can be further
studied. This method may be applied to preparation of supported catalyst. Due to
the facility limitation, the microwave power and sample temperature couldn't be
controlled. This limited further investigation of the suggested method. Inert
atmosphere could be used to protect the carbon and avoid any possible combustion
and explosion. Thermally insulating material, temperature sensors, and automatic
power control systems could be used to improve the uniformity of microwave
heating and avoid overheating. Further understanding of how microwave energy
interacts with materials is the key.
4. Investigated the effect of different types of AC (coal based or wood based) on
photocatalytic activity of TiO02/AC composite.
LIST OF REFERENCES
Aikyo, H., Suzuki, M., Matsumura, A. I. K., Okuyama, K., 1996. The European Caubon
Conference Carbon 96, Newcastle, UK, (Extended Abstracts), 296.
Alberici, R. M., Jardim, W. F., 1997. Photocatalytic destruction of VOCs in the gas-phase
using titanium dioxide. Appl. Catal. B: Env. 14, 55-68.
Ao, C.H., Lee, S.C., 2003. Enhancement effect of TiO2 immoblized on activated carbon
filter for the photodegradation of pollutants at typical indoor air level. Appl. Catal.
B: Env. 44, 191-205.
Arafia, J., Dofia-Rodriguez, J.M., Garriga I Cabo, C., Gonzalez-Diaz, 0., Herrera-Melian,
J.A., Perez-Pefia, J., 2004. FTIR study of gas-phase alcohols photocatalytic
degradation with TiO2 and AC-TiO2. Appl. Catal. B: Environ. 53, 221-232.
Ayll6n. J. A., Peir6, A. M., Saadoun, L., Vigil, E., Domenech, X., Peral, J., 2000.
Preparation of anatase powders from fluorine-complexed titanium (IV) aqueous
solution using microwave irradiation. J. Mater. Chem. 10, 1911-1914.
Bandosz, T., 2002. On the adsorption/oxidation of hydrogen sulfide on activated carbons
at ambient temperatures. J. Colloid Interface Sci. 246, 1-20.
Bargreev, A., Adib, F., Bandosz, T., 2001. pH of activated carbon surface as an
indication of its suitability for H2S removal from moisture air streams. Carbon 39,
Bathen, D. Physical waves in adsorption technology-an overview. Sep. Purif. Technol.
2003, 33, 163.
Bhatkhande, D. S., Pangarkar, V. G., Beenackers A. ACM, 2001. Photocatlytic
degradation for environmental applications a review. J. Chem. Technol.
Biotechnol. 77, 102-116.
Bordado, J. C. M., Gomes, J. F. P., 2002. Atmospheric emissions of Kraft pulp mills.
Chem. Eng. Processing 41, 667-671.
Bouzaza, A., Laplanche, A., Marsteau, S., 2004. Adsorption-oxidation of hydrogen
sulfide on activated carbon fibers: effect of the composition and the relative
humidity of the gas phase. Chemosphere 54, 481-488.
Buonicore, A. J., Davis, W. T. Air pollution engineering manual. AWMA. Van Nostrand
Reinhold, New York.
Bykov, Y. V., Rybakov, K. I., Semenov, V. E., 2001. High-temperature microwave
processing of materials. J. Phys. D: Appl. Phys. 34, 55-75.
Canela, M. C., Alberici, R. M. Jardim, W. F., 1998. Gas-phase destruction of H2S using
TiO2/UV-VIS. J. Photoch. Photobio. A: Chem. 112, 73-80.
Capio, E., Zuihiga, P., Ponce, S., Solis, J., Rodriguez, J., Estrada, W., 2005. Photocatlytic
degradation of phenol TiO2 nanocrystals supported on activated carbon. J. Mol.
Catal. A 228, 293-298.
Carrott, P. J. M., Nabais, J. M. V., Ribeiro Carrott, M. M. L., Pajares, J. A., 2001.
Preparation of activated carbon fibres from acrylic textile fibres. Carbon 39, 1543-
Chang, C., Chen, J., Lu, M., Yang, H., 2005. Photocatalytic oxidation of gaseous DMF
using thin film TiO2 photocatalyst. Chemosphere 58, 1071-1078.
Chen, A., Lu, G., Tao, Y., Dai, Z., Gu, H.,2001. Novel photocatlyst immobilized on
springs and packed bed photoreactor. Mater. Phys. Mech. 4,121-124.
Col6n,G., Hidalgo, M. C., Macias, M., Navio, J. A., Dofia, J. M., 2003. Influence of
residual carbon on the photocatalytic activity of TiO2/C samples for phenol
oxidation. Appl. Catal. B: Environ.43,163.
Crittenden, J. C., Notthakun, S., Hand, D. W., Perram, D. L., 1993. Regeneration of
adsorbents using advanced oxidation. US Patent, No. 5,182,030.
Crittenden, J. C., Suri, R. P. S., Perram, D. L., Hand, D. W., 1997. Decontamination of
water using adsorption and photocatalysis. War. Res. 31, 411-418.
Degussa AG, Titanium dioxide P 25. Technical Bulletin No. 80, September 2003.
Ding, Z., Hu, X., Yue, P. L., Gao, Q. L., Greenfield, P. F., 2001. Synthesis of anatase
TiO2 supported on. Porous solids by chemical vapor deposition. Catal. Today 68,
DOE (Department of Energy ). Forest products annual report, 2003. Washington DC
El-Sheikh, A. H., Newman, A. P., Al-paffaee, H., Phall, S., Cresswell, N., York, S., 2004.
Deposition of anatase on the surface of activated carbon. Surf. Coat. Technol. 187,
Finqueneisel, G., Zimny, T., Weber, J. V., 2005. On the prediction of adsorption
isotherms of methanol/water vapour mixtures on microporous activated carbon.
Carbon 43, 1093-1095.
Giesche, H., 1998. Preparation and applications of coated powders in ceramics and
related fields. J. Disper. Sci. Technol. 19, 249 (Ml)
Gubkina, M. L., Larin, A. V., Polyakov, N. S., 2003. Measurement of adsorption
isotherm of methanol on a humidified activated carbon by the dynamic method.
Colloid J. 65, 155-158.
Harada, M., Honda, M., Yamashita, H., Anpo, M., 1999. preparation of titanium oxide
photocatalysts loaded on activated carbon and their photocatalytic reactivity for the
degradation of 2-propanol diluted in water. Res. Chem. Intermed. 25, 757-768.
Hart, J. N., Cervini, R. Cheng, Y.-B., Simon, G.P., Spiccia, L., 2004. Formation of
anatase TiO2 by microwave processing. Sol. Energy Mater. Sol. cells 84, 135-143.
Hayes, B. L., 2002. Microwave Synthesis: Chemistry at the Speed of Light, CEM
Publishing: Matthews, NC.
Herrmann, Jean-Marie, Guillard, Chantal. 2000, Photocatalytic degradation of pesticides
in agricultural used waters. Comptes Rendus de l'Academie des Sciences, Serie
IIc: Chimie C.R.Acad. Sci. Paris, Serie IIc, Chimie 3, 417-422.
Hoffmann, M. R., Martin, S. T., Choi, W., Bahnemann, D. W., 1995. Environmental
applications of semiconductor photocatalysis. Chem. Rev. 95, 69-96.
Huang, Z., Zhu, Z., Liu, Z., 2002. Combined effect of H20 and SO2 and V205/AC
catalysts for NO reduction with ammonia at Lower Temperatures. Appl. Catal. B:
Environ. 39, 361.
Jeong, J., Sekiguchi, K., Sakamoto, K., 2004. Phtochemical and photocatalytic
degradation of gaseous toluene using short-wavelength UV irradiation with TiO2
catalyst: comparison of three UV sources. Chemosphere 57, 663-671.
Khan, A. M., 2003. Titanium Dioxide Coated Activated Carbon: A Regenerative
Technology for Water Recovery. Master's Thesis, University of Florida.
Khan, A.Y., Mazyck, D. W., Wu, C., 2002. TiO2 coated activated carbon: a regenerative
technology for water recovery. NAA ICE Conference.
Khriachtchev, L., Pettersson, M., Isoniemi, E., Rasanen, M., 1998. 193 nm photolysis of
H2S in rare-gas matrices: Luminescence spectroscopy of the product. J. Chem.
Phys. 108, 5747- 5754.
Kim, J. S., Ki C., Pratsinis, S. E., 2000. The effect of hydrolysis temperature on the
synthesis of bimodally nanostructured porous titania. J. Nanopart. Res. 2, 419.
Kim, S. B., Hong, A. C., 2002. Kinetic study for photocatalytic degradation of volatile
organic compounds in air using thin film TiO2 photocatalyts. Appl. Catal. B: Env.
Kirkbir, K. and Komiyama, H., 1998. Continuous production of fine TiO2 powder by
vapor-phase hydrolysis of titanium tetraisopropoxide. Adv. Ceram. Mater. 3, 511.
Kong, Y., Cha, C.Y., 1996. Reduction of NOx adsorption on char with microwave energy.
Carbon 34, 1027.
Le Leuch, L., Subrenat, A., Le Cloirec, P., 2003. Hydrogen sulfide adsorption and
oxidation onto activated carbon cloths: application to odorous gaseous emission
treatment. Langmuir 19, 10869-10877.
Lee, D.K., Kim, S.C., Kim, S.J., Chung, I.S., Kim, S.W., 2004. Photocatalytic oxidation
of microcystin-LR with TiO2-loaded activated carbon. Chem. Eng. J. 102, 93-98.
Lei, L., Chu, H. P., Hu, X., Yue, P.,1999. Preparation of heterogeneous photocatalyst
(TiO2/Alumina) by metallo-organic chemical vapor deposition. Ind. Eng. Chem.
Res. 38, 3381-3385.
Li, Y. H., Lee, C. W., Gullet, B.K., 2003. Importance of activated carbon's oxygen
surface functional groups on elemental mercury adsorption. Fuel 82, 451-457.
Linsebigler, A. L., Lu, G., Yates, J. T., 1995. Photocatalysis on TiO2 surfaces: principles,
mechanusms, and selected results. Chem. Rev. 95, 735-758.
Liu, J., Crittenden, J. C., Hand, D. W., Perram, D. L., 1996. Regeneration of adsorbents
using heterogeneous photocatalytic oxidation. J. Environ. Eng. 8, 707-713.
Liu, S. X., Sun, G. L., Zhang, S. R., 2004. Photocatalytic regeneration of exhausted
activated carbon saturated with phenol. Bull. Environ. Contam. Toxicol. 73, 1017-
Lu, M., Chen, J., Chang, K., 1999. Effect of adsorbents coated with titanium dioxide on
the photocatalytic degradation of propoxur. Chemosphere 38, 617-627.
Malato, S., Blanco, J., Fernandez-lbafiez, P., Caceres, J., 2001. Treatment of 2,4-
dichlorophenol by solar photocatalysis: comparison of coupled photocatalytic-
active carbon vs. active carbon. Trans. ASME 123, 138-142.
Matos, J., Laine, J., Herrmann, J.-M. 2001 Effect of the type of activated carbons on the
photocatalytic degradation of aqueous organic pollutants by UV-irradiated titania.
J. Catal. 200, 10-20.
Mergel, D., Buschendorf, D., Eggert, S., Grammes, R. And Samset, B., 2000. Thin Solid
Films, 371, 218.