Photo-degradation of yperite over V, Fe and Mn-doped titania-silica photocatalysts

Article abstract of DOI:10.1039/b810200g

The photocatalytic decomposition of yperite (bis(2-chloroethyl)sulfide), a chemical warfare agent, was achieved by using titania-silica catalysts doped with several transition metal ions. The preparation of these catalysts was achieved by impregnation of a titania-silica mixed oxide previously synthesized using a sol-gel route with salts of the doping elements (vanadium, iron, manganese). The above catalysts were characterized using several spectroscopic techniques: FTIR, Raman, DR-UV-Vis, and XPS. The band gap energy was measured for each photocatalytic system. The reaction was carried out in two different types of reactors, i.e. naturally aerated and a closed quartz tube aerated under a constant flow, and using two types of irradiation, UV-Vis and Vis. The investigated systems proved to be extremely active, leading to an almost complete degradation of yperite in 2 h of irradiation. An excellent correlation between the photocatalytic performances and the band gap has been found. Based on the characterization data and on the temporal evolution of the reaction products, a reaction mechanism has been suggested. This mechanism considers two distinct pathways for the decomposition of yperite, namely the C-S bond cleavage and the S oxidation.

Full text of DOI:10.1039/b810200g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Photo-degradation of yperite over V, Fe and Mn-doped titania–silica
photocatalysts
a
a
b
b
ˆ
S- tefan Neat-u, Vasile I. Parvulescu,* Gabriel Epure, Emilia Preda,
Vasile S- omoghi,b Alessandro Damin,c Silvia Bordigac and Adriano Zecchinac
Received 17th June 2008, Accepted 24th July 2008
First published as an Advance Article on the web 23rd September 2008
DOI: 10.1039/b810200g
The photocatalytic decomposition of yperite (bis(2-chloroethyl)sulfide), a chemical warfare agent,
was achieved by using titania–silica catalysts doped with several transition metal ions. The
preparation of these catalysts was achieved by impregnation of a titania–silica mixed oxide
previously synthesized using a sol-gel route with salts of the doping elements (vanadium, iron,
manganese). The above catalysts were characterized using several spectroscopic techniques: FTIR,
Raman, DR-UV-Vis, and XPS. The band gap energy was measured for each photocatalytic
system. The reaction was carried out in two different types of reactors, i.e. naturally aerated and a
closed quartz tube aerated under a constant flow, and using two types of irradiation, UV-Vis and
Vis. The investigated systems proved to be extremely active, leading to an almost complete
degradation of yperite in 2 h of irradiation. An excellent correlation between the photocatalytic
performances and the band gap has been found. Based on the characterization data and on the
temporal evolution of the reaction products, a reaction mechanism has been suggested. This
mechanism considers two distinct pathways for the decomposition of yperite, namely the C–S
bond cleavage and the S oxidation.
Heterogeneous photocatalysis is one of the so-called
1. Introduction
advanced oxidation processes (AOP) that can be carried out
at ambient conditions, with the specific advantage that no
chemical oxidants have to be added.6 A photocatalytic reac-
tion proceeds on the surface of a semiconductor in this order:
(1) production of electron–hole pairs by irradiating the semi-
conductor with light, having an energy content higher than its
band gap; (2) separation of photogenerated electrons and
holes trapping the species adsorbed by the semiconductor;
(3) redox reactions between trapped electrons and holes and
adsorbates; and (4) desorption of reaction products and
reconstruction of the surface.7 The catalysts regularly used
in photocatalytic reactions are constituted by crystalline
titanium dioxide, mainly in the anatase form, such as the
well-known material Degussa P25.8 To maximize TiO2 photo-
activity, the particles should be small enough to offer a high
specific surface area. Unfortunately, for applications in the
liquid phase, such as small particles means high filtration costs
to remove the catalyst once the reaction is finished. These
problems have motivated the development of supported
photocatalysts in which TiO2 has been immobilized on diverse
materials.9 Another problem using TiO2 is its limited efficiency
under visible irradiation. Doping the semiconductor with
various transition metal ions may lead to an increase in the
photocatalytic efficiency of titania. The main objective of
doping is to induce a bathochromic shift (i.e., a decrease of
the band gap or introduction of intra-band gap states, which
results in more visible light absorption).10
Chemical weapons are still causing international concern. There
are still countries possessing such compounds and their produc-
tion is simple enough to allow them to be prepared in many parts
of the world. According to the Chemical Weapons Convention
of the Organization for the Prohibition of Chemical Weapons, all
chemical weapons should be destroyed.1 Among many types of
chemical weapons, the agent which has been produced and
stored in large quantities is bis(2-chloroethyl)sulfide, also known
as yperite or mustard gas.
One of the methods used by the US Army for the destruc-
tion of chemical weapons is incineration.2 However, chemical
neutralization by nucleophilic substitution and oxidation
seems to be a more suitable alternative method for yperite
disposal,2 but using the suggested nucleophile (monoethanol-
amine) leads to the formation of large quantities of organic
products that have to be disposed.3 Furthermore, the oxida-
tion using oxone as agent may result in equally toxic products,
such as bis(2-chloroethyl)sulfoxide and divinyl sulfone.4 The
bioremediation process is another method for degradation of
yperite.5 These techniques can be more harmful than helpful,
therefore new techniquies that are more efficient must be
discovered.
a University of Bucharest, Faculty of Chemistry, Department of
Chemical Technology and Catalysis, Bd. Regina Elisabeta 4–12,
030018 Bucharest, Romania. E-mail: v_parvulescu@chem.unibuc.ro;
Fax: +4021 4100241; Tel: +4021 4100241
b Chemical Centre for Defence NBC and Ecology, Sos. Oltenitei 225,
041309 Bucharest, Romania
Some works have already been reported in literature,
concerning the degradation of mustard11 as well as its simu-
lants using TiO2 as a photocatalyst.12,13 In this paper we
c University of Turin, Department of Chemistry IFM and NIS-Centre
of Excellence, Via P. Giuria 7, 10125 Turin, Italy
ꢀc
This journal is the Owner Societies 2008
6562 | Phys. Chem. Chem. Phys., 2008, 10, 6562–6570

View Article Online
present the results of the photodegradation of yperite using
titania–silica and more complex titania–silica doped with Mn,
Fe or V photocatalysts under both UV light and ambient
irradiation conditions, also the correlation between the
electronic properties of the catalysts and their behavior in
such reactions.
value of Eg, the band gap of the semiconductor, can be derived
from the spectra by plotting (F(R)Áhn)2 against hn.15
Fourier transform infrared (FTIR) spectra were measured
with a Nicolet 4700 spectrometer. For IR measurements, the
sample material was mixed with KBr in a 5/95 ratio mix. The
IR spectra were scanned in the region of 4000–400 cmÀ1 with a
resolution of 4 cmÀ1. For achieving higher sensitivity,
accumulation of 100 scans was used. Raman spectra were
collected on a Renishaw Raman Microscope Spectrometer
equipped with a Renishaw HPNIR laser emitting at 785 nm.
The laser power adopted during the measurements was set just
to guarantee the best compromise between the best S/N ratio
and the stability (verified by means of several test measure-
ments) of the samples under the laser beam. In the case of a V
containing sample, 10% of the total laser power was adopted;
for Fe and Mn containing samples only 1% of the total laser
power was employed. The photons scattered by the sample
were dispersed by an 1200 lines mmÀ1 grating monochromator
and simultaneously collected on a CCD camera, the collection
optic was set at 20Â objective. Each spectrum corresponds to
the sum of 50 acquisitions, each of them during 10 s.
2. Experimental
2.1 Photocatalysts preparation
The synthesis of the catalysts focused on pre-hydrolysis of
Si(OC2H5)4 as an essential step in the preparation of
titania–silica mixed oxides. The pre-hydrolysis of Si(OC2H5)4
has been used to adjust the sol-gel reactivities of titanium and
silicon alkoxides. The preparation of the catalysts followed a
two-stage sol-gel procedure: (1) Pre-hydrolysis of the silicon
source started from a solution of tetraethylorthosilicate
(Si(OC2H5)4, 98 wt%, Aldrich; TEOS) in ethanol (molar ratio
ethanol/TEOS = 2) that was hydrolyzed in the presence of a
0.05 N HCl solution (molar ratio H2O/TEOS/HCl = 1/1.2/
0.05) at room temperature for 90 min. (2) Hydrolysis of the
transition metal and titanium source was carried out simulta-
neously in an ice-water bath. Initially, a solution of titanium
tetraisopropoxide (Ti(i-OC3H7)4, 97 wt%, Aldrich; TTIP) in
ethanol (molar ratio ethanol/TTIP = 1) was stirred for
60 min. To this solution, 15 wt% ethanol solution of the
transition metal salt (Fe(III)acetylacetonate, Mn(II)acetylace-
tonate, VO(OC3H7)3) was added. Then the TTIP and the
metal salt mixture were added dropwise to the silica sol. (3)
The condensation of the gel was done by slowly increasing the
pH of the mixture by adding drops of 1 N NH3 solution. The
resulted material was then dried at room temperature for 48 h,
followed by increasing the temperature at 373 K for 2 h. The
removal of residual organics and thermal stabilization of the
structure was achieved by calcination for 5 h at 773 K in air.
By this procedure the following named-catalysts were ob-
tained: 10%TiO2–SiO2, 5%M–10%TiO2–SiO2 (M = V, Mn,
and Fe), 1% V–10%TiO2–SiO2, and 5%V–20%TiO2–SiO2.
The XPS spectra were recorded at room temperature using a
SSX-100 spectrometer, Model 206 from Surface Science
Instrument. The pressure in the analysis chamber during the
analysis was 1.33 mPa. Monochromated Al Ka radiation
(hn = 1486.6 eV) was used, generated by bombarding the Al
anode with an electron gun operated with a beam current of
12 mA and an acceleration voltage of 10 kV. The spectrometer
energy scale was calibrated using the Au 4f7/2 peak centered at
83.98 eV. Charge correction was made with the C 1s signal of
adventitious carbon (C–C or C–H bonds) located at 284.8 eV.
The composite peaks were decomposed by a fitting routine
included in the ESCA 8.3 D software. The atomic surface
compositions were calculated using the sensitivity factors
provided with the ESCA 8.3 D software, applied to the surface
below the corresponding fitted XPS signals. An estimated
error of Æ0.1 eV can be assumed for all measurements.
2.3 Photocatalytic tests
2.2 Photocatalysts characterization
The catalytic tests have been carried out in two types of quartz
reactors: (i) a tubular reactor where the amount of oxygen was
kept constant and (ii) a closed tube where a constant flow of
air was introduced (50 cm3 minÀ1) under both UV and visible
light irradiation. The amount of oxygen was monitored with
an Alpha Omega Series 2000 instrument. The difference in the
oxygen consumed was automatically added using a home-
made system. Under UV conditions, as irradiation source, a
125 W high-pressure Hg UV lamp (HQV 125W, Osram,
Germany) with a maximum emission at 365 nm and 3.0 W
UVA radiated power was used. Photocatalysts were also
tested under Vis light using a 200 W Ne lamp (F74-765,
Tungsram, Hungary) with a maximum emission at 600 nm.
The emission spectra of the two lamps are given in Fig. 1 and
2. Light intensity at the distance where the sample was placed
was 810 Lx for the UV lamp and 4300 Lx for the neon lamp.
Light intensity measurements were performed using an 840006
Speer Scientific luxmeter. In a typical experiment we have
used 20 mg of photocatalyst with 5 mg of yperite in CH2Cl2
The characterization of the photocatalysts was carried out
using several techniques. The powder X-ray diffraction (XRD)
patterns were recorded on a Siemens D 500 diffractometer
using Cu Ka radiation. The diffuse reflectance DR-UV-Vis
spectra were recorded at room temperature on a Cary 300
spectrophotometer from Varian with a Varian DRA for
reflectance measurements, along with Spectralon was used as
the reflectance standard. DR-UV-Vis spectra of the catalysts
were recorded in reflectance units and were transformed in
Kubelka–Munk remission function F(R). The calculation of
the band gap was performed using the formula:
E = hc/l (eV)
where h is the Planck constant (4.135 Â 10À15 eVs), c is the
speed of light (3 Â 108 msÀ1), and l the absorption threshold
value for each sample.
The Kubelka–Munk function, F(R), can be considered
proportional to radiation absorption.14 On this basis, the
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 6562–6570 | 6563

View Article Online
air into the equipment (GC/MS). The evolution of the reaction
was also followed by taking the content of the vat at the
required reaction time, washing the catalyst with CH2Cl2,
concentrating and analyzing the resulting solution using the
same GC/MS equipment.
3. Results and discussion
3.1 Photocatalysts characterization
All the prepared photocatalysts were amorphous as they
resulted from the XRD patterns. Except the boss in the range
2y 25–30 assigned to the amorphous oxide no other reflection
plane has been identified.
Structural characterization tried to give answers to several
questions concerning: (i) the location of titanium in the
titania–silica catalysts; (ii) the interaction of the dopant
element with titanium; and (iii) the values of the band gap
energy as a consequence of the points (i) and (ii).
Fig. 1 Emission spectrum of the 125 W high-pressure Hg UV lamp
used in the photocatalytic study.
DRIFT spectra confirmed the presence of Si–O–Ti linkages
in the TiO2–SiO2 framework, which is obvious for a material
prepared by sol-gel.
Fig. 3 shows the DRIFT spectra of SiO2, pure and doped
Fe, V, or Mn–TiO2–SiO2 samples. No clear differences can be
seen from these spectra and showing that the dopants are
incorporated in a similar way in the titania–silica matrix. The
DRIFT spectra of all TiO2–SiO2 systems display in the low
frequency region bands at 1115, 807 and 467 cmÀ1 due to
asymmetric stretching, symmetric stretching, and bending
modes of bulk Si–O–Si, respectively.16 The peak at 960 cmÀ1
is generally attributed to the stretching vibrations of a SiO4
tetrahedron perturbed by the presence of a metallic group
ensemble, i.e. –(SiO3)Si–OTi.17,18 The intensity of this band
increased with the increase of Ti loading, and might be a
suggestion for the incorporation of titanium into the silica
framework. The substraction of the spectrum of SiO2 from
that 10%TiO2–SiO2 (Fig. 3, A) makes this behaviour more
clear. A sharp strong band at 3648 cmÀ1 due to stretching
Fig. 2 Emission spectrum of the 200 W Ne lamp used in the
photocatalytic study.
(a 1700 ppm solution) added. (Caution: sulfur mustard is a
potent vesicant and must be handled only by trained personnel
using safety procedures in a closed system or under a well
ventilated hood.) The photocatalysts and sulfur mustard solu-
tion were placed into an open polyethylene vat that has been
introduced in the above reactors. The ensemble was placed in
the reactor, perpendicular to the light propagation direction.
Samples of the gas phase inside the closed reactor were taken
at regular intervals with a gas-tight syringe and analyzed using
a GC/MS equipment Saturn 2100 T from Varian. The chro-
matograph was calibrated for yperite (m/z 27 (25.2%), 45
(13.2%), 47 (15.8%), 59 (9.5%), 63 (34.7%), 65 (11.2%), 109
(100%), 111 (38.3%), 158 (23.8%), 160 (16.2%), carbon
dioxide (m/z 16 (5.9%), 28 (7.1%), 44 (100%), sulfur dioxide
(m/z 16 (5.2%), 32 (10.7%), 48 (49.8%), 64 (100%), 66 (4.9%),
bis(2-chloroethyl)disulfide (m/z 27 (100%), 45 (23.5%), 59
(13.5%), 63 (77.6%), 64 (21.7%), 65 (26.1%), 92 (16.5%),
128 (13.5%), 190 (31.0%), 192 (22.7%), acetaldehyde (m/z 14
(10.8%), 15 (36.4%), 16 (56.0%), 26 (54.0%), 27 (35.0%), 29
(100%), 42 (13.7%), 43 (47.4%), 44 (82.5%), and chloroethyl-
ene (m/z 25 (13.7%), 26 (34.0%), 27 (100%), 35 (8.7%), 47
(42.0%), 62 (76.6%), 64 (24.2%). The calibration was per-
formed by injection of the mixtures of the respective gases with
Fig. 3 DRIFT spectra of SiO2 (a), 10%TiO2–SiO2 (b), 5%Mn–
10%TiO2–SiO2 (c), 5%Fe–10%TiO2–SiO2 (d), 1%V–10%TiO2–SiO2
(e), and 5%V–20%TiO2–SiO2 (f). The spectrum A represents the
subtraction of the DRIFT SiO2 spectrum from that of 10%TiO2–SiO2.
ꢀc
This journal is the Owner Societies 2008
6564 | Phys. Chem. Chem. Phys., 2008, 10, 6562–6570

View Article Online
vibrations of the isolated (non-interacting) surface silanol and
a very small shoulder at 3615 cmÀ1, preceded by a tail in the
range 3500–3300 cmÀ1 were observed. The bands located at
1635 and 3430 cmÀ1, respectively, are assigned to the OH
binding and stretching vibration of water.
Fig. 4 shows the Raman spectra for the samples doped with
V, Fe and Mn, respectively. The spectrum of vanadia contain-
ing samples (Fig. 4a) shows the presence of the vanadium in
different coordinations: i.e. as terminal O3(VQO) and
polymeric V–O–V species. Thus, the peaks at 995, 974, and
1029 cmÀ1 can be assigned to microcrystalline or amorphous
V2O5, V–O–V, and terminal VQO stretching vibrations of the
isolated VO4 species.19 Except for these bonds, Raman spectra
also show two peaks at 248 and 648 cmÀ1 which are assigned
to the bending and stretching modes of Ti–O–V bridging
bonds,20 most probably as OQV(ÀOÀTiÀOÀSi) complexes.
These bands are indicative of the direct interaction of vana-
dium with titania in the support. Fig. 4b shows the Raman
spectrum of the Fe–titania–silica photocatalyst. This spectrum
contains only peaks due to the presence of isolated iron oxide
(hematite) particles in the titania–silica matrix. It presents no
evidence of any direct bond with titania in the support. Similar
to iron, the Raman spectrum of the sample containing man-
ganese (Fig. 4c) contains only peaks that can be assigned to an
oxidic Mn3O4 phase.
The DR-UV-Vis spectra provided additional information
about the doping effect of these elements. Fig. 5 gives the
typical DR-UV-Vis spectra of the prepared samples. Table 1
summarizes the band gap energy (Eg) and the corresponding
Fig.
5
DR-UV-Vis spectra of the reference materials (a)
10%TiO2–SiO2, (b) 1%V–10%TiO2–SiO2, (c) 5%V–20%TiO2–SiO2,
(d) 5%Mn–10%TiO2–SiO2 and (e) 5%Fe–10%TiO2–SiO2.
Table 1 Eg and the corresponding wavelength (lEg) and absorption
threshold values
Material
Eg/eV
lEg/nm
l/nm
10%TiO2–SiO2
4.7
4.3
4.07
4.08
3.2
215
239
253
257
291
264
290
305
304
387
5%Mn–10%TiO2–SiO2
1%V–10%TiO2–SiO2
5%Fe–10%TiO2–SiO2
5%V–20%TiO2–SiO2
wavelength (lEg) and absorption threshold values calculated
for each material.
The spectrum recorded for pure 10%TiO2–SiO2 is very similar
to that of titanium silicalite showing an absorption threshold at
264 nm (4.7 eV) and an O2À - Ti4+ charge-transfer band with
a maximum at 215 nm; according to the literature data these
values account for such a high distribution of titanium in the
silica that the majority of titanium is in the matrix.21
The doping of titania–silica samples with transitional metals
caused a shift in the band gap transition, which can be
explained by the introduction of energy levels of the transition
metal ions into the band gap of TiO2. The presence of the
energy levels below the conduction band edge and above the
valence band edge is expected to influence the photoreactivity
of TiO2, since the metal ions can act as electron (or hole) traps
altering the electron–hole pair recombination rate. All doped
titania–silica samples showed a red shift in the band gap
transition caused by the incorporation of V, Fe and Mn ions
into the titania–silica matrix.
Fig. 4 Raman spectra for titania–silica systems containing (a) V,
(b) Fe, and (c) Mn.
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 6562–6570 | 6565

View Article Online
Doping 10%TiO2–SiO2 with vanadium led to a more
pronounced shift, which paralleled the metal loading. In
addition to the information collected from the Raman spectra,
this might be another indication of the presence of V in close
contact with titanium. Thus, 1%V–10%TiO2–SiO2 showed an
absorption at 305 nm (4.07 eV) while 5%V–20%TiO2–SiO2 at
387 nm (3.2 eV). Under these conditions the 3d orbital of
vanadium might change the band gap of titania making these
catalysts able to absorb visible light.22
in the crystalline network and the average crystal size of the
semiconductor can also influence the band gap.24 In accor-
dance, the calculation of the band gap should consider these
aspects and different groups for recommended different proce-
dures. For example, some authors calculate the Eg values by
direct extrapolation of the F(R) spectrum whereas others
reported the wavelength corresponding to the maximum
absorption.25 On the other hand, the magnitude of the UV-
Vis absorption intensity depends on various factors such as
crystal structures, band gap energy, thickness, and the presence
and absence of defects and foreign elements in concerned
semiconducting materials.26 As a consequence of the different
methodologies, different Eg values were reported for rutile and
anatase. For anatase-based materials different threshold
wavelength values were measured: 370 nm,27 380 nm,28 387
nm,29 393 nm30 and 403 nm.31 Using these data the authors
calculated band gaps in the range from 3.08 to 3.35 eV. In this
study we calculated the values of Eg, using the methodology
proposed by Lassaleta et al.15 Using this methodology the
band gap energy calculated values for the synthesized materials
were between 3.2 eV for 5%V–20%TiO2–SiO2 and 4.7 eV for
10%TiO2–SiO2 (Table 1), respectively, demonstrating the po-
sitive role of the doping. Table 2 compiles the surface Ti/Si, M/
Ti, and M/Si (M = Fe, Mn, V) XPS atomic ratios and the
corresponding binding energies of the Si2p, Ti 2p3/2, O1s, and
M 2p3/2 (M = Fe, Mn, and V) levels obtained. For all samples
the binding energy of the Ti 2p3/2 level corresponds to the Ti4+
oxidation state.32–35 The binding energies of the doped metals
correspond to Fe3+ in Fe2O3, Mn4+ in MnO2, and V5+ in
V2O5, respectively.36 For the doped photocatalysts, besides the
band of O 1 s at B532.9 eV, which is due to the oxygen of the
SiO2 support as was verified from the analysis of pure silica, a
second band arose at B530.1 eV, and its percentage increased
with increasing metal oxide loading. This new band is ascribed
to the oxygen associated to the dopant species.37 Table 2
compiles the percentage in these species. The XPS atomic
ratios given in the Table 2 correlate very well with the chemical
analysis of these compounds. Chemical analysis indicates a Fe/
Ti ratio of 0.42, a Mn/Ti ratio of 0.42, and a V/Ti ratio of 0.28
for catalysts with 5 wt% V and of 0.14 for those with 1 wt% V.
These values demonstrate an excellent homogeneity of these
catalysts.
The spectra of iron and manganese doped samples show a
different picture as compared with vanadium, and exhibit two
distinct features. The first one is the presence of the shift of the
threshold that is likely to indicate a close interaction with the
framework. The second one is connected to the presence of a
broad and complex absorption covering the visible range. The
later absorption is associated to the existence of separate iron
and manganese oxide phases, which is again in a good con-
cordance with the Raman spectra. The sample containing 5 wt%
Fe shows an absorption threshold at 304 nm (4.0 eV) and the
O2À - Ti4+ charge-transfer band can be observed at ca 220 nm.
Previous studies attributed the red shift of the absorption
edge of Fe doped titania–silica sample to the excitation of 3d
electrons of Fe to the titania conduction band (charge-transfer
transition).23 Doping titania–silica with manganese resulted in
a red shift corresponding to adsorption at 290 nm. The 3d
electron of Mn also changed the band-gap of titania, but this
was less effective than in the case of Fe or V. For Mn the band
gap energy is higher, i.e. 4.3 eV.
The spectral data demonstrate that the different dopants
used in this study can affect the photoproperties of TiO2 in
different ways.
The band gap values usually reported in the literature for
pure anatase and rutile are 3.2 and 3.03 eV, respectively.24
However, as these authors stressed, these values are influenced
by the synthesis procedure. This includes both the preparation
route, such as sol-gel method, hydrothermal processing, micro-
emulsion method, flame pyrolysis process, etc, along with the
various parameters such as the ratio of the amount of water to
that of Ti containing compound, the amount of acid, reaction
time and temperature. The interrelationships between these
parameters are complex, and the analysis of this system to
optimize the factors is complicated. The existence of impurities
Table 2 Surface composition and binding energies of core electrons for the reference materials
B.E./eV
Ti/Si XPS
atomic ratio
Metal/Ti XPS
atomic ratio
Metal/Si XPS
atomic ratio
Sample
Si 2p
Ti 2p3/2
O 1s
M 2p3/2
SiO2
10%TiO2–SiO2
5%Fe–10%TiO2–SiO2
103.4
103.9
103.8
532.9
532.9
0.029
0.035
460.0
459.9
0.291
0.502
0.555
0.500
0.015
0.035
0.052
0.050
532.5 (91)a
530.1 (9)
532.7 (95)
530.2 (5)
532.4 (90)
530.1 (10)
532.4 (95)
530.1 (5)
530.7
710.7
642.2
517.5
517.5
5%Mn–10%TiO2–SiO2
5%V–20%TiO2–SiO2
1%V–10%TiO2–SiO2
0.035
0.042
0.032
103.7
103.2
103.3
460.0
458.9
458.8
458.5
TiO2
a
The values in parentheses indicate the percentage of the XPS concentration of the two O1s species.
ꢀc
This journal is the Owner Societies 2008
6566 | Phys. Chem. Chem. Phys., 2008, 10, 6562–6570

View Article Online
3.2 Photocatalytic behaviour
Blank experiments proved that the decomposition of yperite is
a photocatalyzed process. Before switching the lamp on, no
decomposition of yperite in the reactor effluent was found.
This demonstrates that no dark reaction under the investi-
gated catalysts and temperature conditions occurs. Also,
photo-exposure of yperite in the absence of the catalyst was
not accompanied by any decomposition. Fig. 6 and 7 show the
results of photodegradation of yperite in the presence of the
investigated photocatalysts after 1 and 2 h UV irradiation
exposure. Compared to the parent 10%TiO2–SiO2 photo-
catalyst, doping with the investigated metals induced a positive
effect. It can be seen that all materials are effective catalysts.
Under the investigated conditions the yield of degradation of
yperite using the parent 10%TiO2–SiO2 catalyst was approxi-
mately 50%. This yield was preserved irrespective of the
reactor being used.
Fig. 7 Photocatalytic degradation percentage of yperite under nor-
mal atmosphere correlated with the band gap energy (catalyst 20 mg;
UV irradiation; irradiation time 1 and 2 h in the photoreactor with
constant oxygen concentration).
Other products were distributed between the gas and the
surface according to their volatility.
Fig. 6 and 7 also present the band gap energy for each
investigated photocatalysts. As can be observed, in the case of
the reactor with constant oxygen concentration, to the most
active catalyst, 5%V–20%TiO2–SiO2 it corresponds to the
smallest band gap energy. Actually, in this case, after 2 h,
the activity in yperite degradation over the titania–silica
based-catalysts decreased in the series 5%V–20%TiO2–
All photocatalysts were also tested under Vis light irradia-
tion (Fig. 8 and 9). Under these conditions, after 2 h exposure,
the conversion was between 10–15% smaller than under UV
irradiation, which is indeed remarkable. This behaviour can
also be associated with the higher emission intensity of the
neon lamp. It is well known that there are two regimes of the
photocatalytic reaction with respect to the photon flux, i.e.
(i) a first-order regime for fluxes up to about 25 mW cmÀ2 and
(ii) a half-order regime for higher intensities, which is the case
of the neon lamp. In the first case, the electron–hole pairs are
consumed more rapidly by chemical reactions than by recom-
bination reactions, whereas in the half-order regime, the
recombination rate is dominant.38,39 The variation of the
reaction rate as a function of the used wavelength follows
the adsorption spectrum of the catalyst with a threshold
corresponding to its band energy. This is in a perfect con-
cordance with the modification of the band gap energy by
promoting with this element. As mentioned, the distribution of
the reaction products remained unchanged under these
conditions.
SiO2 Z 1%V–10%TiO2–SiO2 4 5%Fe–10%TiO2–SiO2
B
5%Mn–10%TiO2–SiO2 4 10%TiO2–SiO2, while the band gap
energy increased in the same order.
In the case of the reactor with controlled air flow, the obtained
data show that this setup is as efficient as the other one.
Photocatalytic decomposition of yperite resulted in the
formation of the same gaseous products irrespective of the
investigated catalyst. The gaseous products comprised chloro-
ethylene, carbon dioxide, sulfur dioxide, and bis(2-chloro-
ethyl)disulfide. The extract in dichloromethane from the
surface of the catalysts operated for 2 h at room temperature
in yperite decomposition containing only bis(2-chloroethyl)-
disulfide. Acid products and salts (such as sulfates), which are
usually produced at the final stage of the oxidation of sulfur in
yperite were not detected because of their extremely low
volatility. Only bis(2-chloroethyl)disulfide was detected both
as gaseous and as surface product.
Fig.
6
Photocatalytic degradation percentage of yperite under
Fig. 8 Photocatalytic degradation percentage of yperite under nor-
mal atmosphere (air flux 50 cm3 minÀ1 air; catalyst 20 mg; Vis
irradiation; irradiation time 1 and 2 h in the photoreactor with a
controlled air flow).
normal atmosphere correlated with the band gap energy (air flux
50 cm3/min air; catalyst 20 mg; UV irradiation; irradiation time 1 and
2 h in the photoreactor with a controlled air flow).
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 6562–6570 | 6567

View Article Online
Fig. 9 Photocatalytic degradation percentage of yperite under nor-
mal atmosphere (catalyst 20 mg; Vis irradiation; irradiation time 1 and
2 h in the photoreactor with constant oxygen concentration).
Fig. 11 Temporal evolution of (a) yperite, (b) bis(2-chloroethyl)-
disulfide and (c) chloroethylene.
Kinetic experiments are presented in Fig. 10. The temporal
analysis of the degradation of yperite on the
5%Fe–10%TiO2–SiO2 (d) and 5%Mn–10%TiO2–SiO2 (e)
Fig. 11 presents the temporal evolution of yperite, chloro-
ethylene, and bis(2-chloroethyl)disulfide over 160 min of UV
irradiation on 5%Mn–10%TiO2–SiO2 photocatalyst. As can
be seen, the concentration of yperite decreases rapidly in the
first 80 min, while the concentration of both chloroethylene
and bis(2-chloroethyl)disulfide increases after 20 min of reac-
tion, attaining a maximum after 40 min. After 2 h of reaction
almost all the yperite, chloroethylene and bis(2-chloroethyl)-
disulfide were completely photodegraded.
photocatalysts follows
10%TiO2–SiO2 photocatalyst the yperite degradation
accounts to pseudo-zero order kinetics. The
a first order kinetics. For the
a
5%V–20%TiO2–SiO2 catalyst showed a completely different
behaviour. The degradation process proceeded through two
stages, i.e. initial low degradation for the first 60 min, followed
by rapid degradation in the next hour. The analyses we carried
out showed that in the initial step yperite had mostly been
adsorbed on the catalysts surface, and only after reaching a
certain superficial concentration of the warfare compound did
the photodegradation occur with a high rate.
The different temporal evolution of the yperite degradation on
vanadia–titania–silica catalysts is connected to its superficial con-
centration in the first hour. This behaviour has been proven by the
chemical analysis of the compounds separated by extraction.
The reproducibility of the runs was checked in two aspects:
(i) all the catalysts were tested three times in successive runs,
and the reproducibility in the temporal evolution was 98%;
and (ii) the catalysts on which the conversion was almost total
(1%V–10%TiO2–SiO2 and 5%Mn–10%TiO2–SiO2) were
washed with CH2Cl2 and re-tested for another three times.
The photoactivity of these catalysts within the three cycles
varied in a range of conversion of 4%.
The data presented in Fig. 10 show that the differences
among the different dopants are not so remarkable. The main
differences are induced by the amount of the inserted dopant.
These results are in line with previous observations for the
degradation of acetone using other dopants.40,41
Kinetic evolution of the detected intermediate species
suggests that the decomposition of yperite over titania–silica
systems follows two main reaction pathways, i.e. (i) C–S bond
splitting and (ii) S atom oxidation. However, typical for a
photocatalytic process, the first stage consists in the generation
of free charge carriers, electrons and holes (Scheme 1).
Scheme 2 presents the possible mechanism of the yperite
photo-degradation process over titania–silica photocatalysts.
Active consumption of yperite starts only after switching on
the ultraviolet illumination of the catalyst. This shows that
dark adsorption of yperite was low and adsorption is
stimulated by irradiation. Low dark adsorption most likely
results from strong chemisorption of water on the surface of
titania–silica (eqn (1)).
Fig. 10 Kinetics of photocatalytic degradation of yperite under UV
irradiation on 10%TiO2–SiO2 (a), 5%V–20%TiO2–SiO2 (b),
1%V–10%TiO2–SiO2
(c),
5%Fe–10%TiO2–SiO2
(d)
and
5%Mn–10%TiO2–-SiO2 (e) photocatalysts.
Scheme 1
ꢀc
This journal is the Owner Societies 2008
6568 | Phys. Chem. Chem. Phys., 2008, 10, 6562–6570

View Article Online
Scheme 2
The resultant TiIII sites (eqn (2)) can be restored to highly
adsorptive TiIV sites by reaction with holes. Adsorbed yperite
is probably not desorbed by this mechanism. Yperite is
oxidized by holes to form a sulfide radical cation (eqn (4)).
The formation of a sulfur radical cation as an intermediate
in photosensitized oxidation of sulfides is commonly accepted
and proved.13,42,43 The redox potential of sulfide is lower than
the redox potential of holes in TiO2. We could speculate that
singlet oxygen might also mediate oxidation as it is in homo-
geneous photooxidation,44 but studies by Davidson and
Pratt45 and Fox and Abdel-Wahab46 demonstrated that this
is not the case for TiO2-assisted photooxidation. Further
transformations of the radical cation can follow different
pathways. The formation of products like 2-chloroethanol
and chloroethylene clearly testifies for the contribution of
the C–S bond cleavage which did not happen in the dark.
The cleavage can occur via a hydrolysis route, in which
2-chloroethanol is liberated (eqn (5)) or with liberation of
chloroethylene and H+ (eqn (6)).
As has been demonstrated by the characterization data, the
doping changes the electronic properties of the titanium active
site, thus enhancing photocatalytic activity. We did not find
changes in the selectivity of the products we have quantified,
but only changes in the photoactivity.
4. Conclusion
This study has revealed the photocatalytic degradation on
titania–silica systems under UV light radiation. In the absence
of UV radiation yperite decomposition reaction does not take
place. Also, the yperite decomposition reaction does not take
place when the non-active materials and UV light have been
used. Comparing the two types of reactors in which the photo-
catalytic degradation was carried out, the decomposition rate is
generally lower for the reactor with controlled air flow. Such
behaviour was assigned to the fact that the oxygen is consumed
rapidly and the flow is too low to compensate this consumation.
Photocatalytic tests confirmed the characterization data
concerning the promoter role played by the dopant ions. In
the presence of these photocatalysts, the decomposition rate
was higher than on the parent titania–silica catalyst, in all the
cases. Water adsorption on the external surface of the photo-
catalysts also plays a role in this reaction.
Following the step described by eqns (5) and (6), the
dimerization of surface [ClCH2CH2S] radical species will take
place, resulting in bis(2-chloroethyl)disulfide (eqn (7)).
Bis(2-chloroethyl)disulfide is very stable and remains on the
catalyst surface.
The second obvious route of yperite photocatalytic decom-
position is the oxidation of the sulfur atom. Due to their low
vapor pressure, the products of this oxidation pathway (sulf-
oxide and sulfone) resided on the surface of the catalyst, where
they could be further oxidized leading to the formation of
inorganic compounds. The second branch of oxidation results
in sulfoxide and sulfone. During the photocatalytic degrada-
tion these routes can intersect each other. Finally, after 2 h,
photocatalytic oxidation results in inorganic oxides, H2O,
CO2, and SO2.
The product distribution suggests that the two main pathways
of the degradation are the oxidation of sulfur and carbon atoms
and the cleavage of the C–S bond. The products of the complete
oxidation (SO2 and CO2) were detected in very small quantities.
Acknowledgements
The authors kindly acknowledge NATO’s Scientific Affairs
Division in the framework of the Science for Peace
Programme Sfp 981476 for the financial support. This research
ꢀc
This journal is the Owner Societies 2008
Phys. Chem. Chem. Phys., 2008, 10, 6562–6570 | 6569

View Article Online
has also been partially supported by the CNCSIS through the
PN-II-RU-TD-2007-1 project (contract number 20/2007)
supervised by S- tefan Neat-u.
19 G. Deo and I. E. Wachs, J. Catal., 1991, 129, 137.
20 F. Hardcastle and I. E. Wachs, J. Phys. Chem., 1991, 95, 5031.
21 X. Gao and I. E. Wachs, Catal. Today, 1999, 51, 233.
22 J. C.-S. Wu and C.-H. Chen, J. Photochem. Photobiol., A, 2004,
163, 509.
23 Y. H. Zhang and A. Reller, J. Mater. Chem., 2001, 11, 2537; H.
Cheng and J. Ma, Chem. Mater., 1995, 7, 663.
24 Heterogeneous Photocatalysis, ed. M. Schiavello, Wiley,
Chichester, 1997, and references herein.
25 R. J. Davis and Z. Liu, Chem. Mater., 1997, 9, 2311.
26 D. B. Hamal and K. J. Klabunde, J. Colloid Interface Sci., 2007,
311, 514.
27 D. F. Ollis, Catal. Technol., 1998, 2, 149.
28 K. Kosuge and S. Singh, J. Phys. Chem. B, 1999, 18, 3563.
29 A. A. Belhekar, S. V. Awate and R. Anand, Catal. Commun., 2002,
3, 453.
30 B. J. Aronson, C. F. Blanford and A. Stein, Chem. Mater., 1997, 9,
2842.
31 S. Cheng, S. Tsai and Y. Lee, Catal. Today, 1995, 26, 87.
32 P. Wauthoz, M. Ruwet, T. Machej and P. Grange, Appl. Catal.,
1991, 69, 149.
33 G. B. Raupp and J. A. Dumesic, J. Phys. Chem., 1985, 89, 5240.
34 S. Mukhopadhyay and S. Garofalini, J. Non-Cryst. Solids, 1990,
126, 202.
References
1 Convention on the Prohibition of the Development, Production,
Stockpiling and Use of Chemical Weapons and on their
Destruction, available online on http://www.opcw.org.
2 Y.-C. Yang, Chem. Ind., 1995, 9, 334.
3 I. N. Luganskii, V. V. Sheluchenko, I. N. Krotovich, V. V.
Chebotaev, V. A. Kholodeva and O. A. Papkova, Zh. Ross. Khim.
Ob-va im. D. I. Mendeleeva, 1994, 38, 34.
4 R. I. Tilley and D. R. Leslie, Aust. J. Chem., 1995, 48, 1781.
5 H. Wariishi, N. Itoh, M. Yoshida and H. Tanaka, Biotechnol.
Lett., 2002, 24, 501.
6 W. H. Glaze, in Chemical Oxidation: Technologies for the nineties,
Proceedings of the third International Symposium: Chemical Oxida-
tion: Technology for the Nineties, ed. W. W. Eckenfelder, A. R.
Bowers and J. A. Roth, Technomic Publishing Company,
Lancaster, 1994, vol. 3, pp. 1–9.
7 X. Fu, W. A. Zetner and M. A. Anderson, in Semiconductor
nanoclusters, ed. P. V. Kamat and D. Meisel, Studies in Surface
Science and Catalysis, Elsevier Science, Amsterdam, 1996,
pp. 445–461.
35 Z. Wei, Q. Xin, X. Guo, E. Sham, P. Grange and B. Delmon, Appl.
Catal., 1990, 63, 305.
36 Handbook of X-Ray photoelectron Spectroscopy, ed. C. D. Wagner,
W. M. Riggs, L. E. Davis, J. F. Moulder and G. E. Muilenberg,
Perkin-Elmer Corporation, Minnesota, MN, 1979, pp. 70, 74 and
76.
8 M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann,
Chem. Rev., 1995, 95, 69.
9 R. L. Pozzo, M. A. Baltanas and A. E. Cassano, Catal. Today,
1997, 39, 219.
37 A. Y. Stakheev, E. S. Shpiro and J. Apijok, J. Phys. Chem., 1993,
97, 5668.
10 W. Choi, A. Tormin and M. R. Hoffman, J. Phys. Chem., 1994, 98,
13669.
38 D. F. Ollis, E. Pelizzetti and N. Serpone, Environ. Sci. Technol.,
1991, 25, 1522.
39 W. A. Jacoby, D. M. Blake, R. D. Noble and C. A. Koval, J.
Catal., 1995, 157, 87.
11 Y.-C. Yang, Photocatalytic Oxidation of Mustard Using Semicon-
ductor Oxides, Chemical Research Development and Engineering
Center, 1989, pp. 83–89.
12 M. A. Fox, Y.-S. Kim and A. A. Abdel-Wahab, Catal. Lett., 1990,
5, 369.
40 C. M. Visinescu, R. Sanjines, F. Le
Catal., B, 2005, 60, 155.
41 C. M. Visinescu, R. Sanjines, F. Le
´
vy and V. I. Parvulescu, Appl.
13 A. V. Vorontsov, E. V. Savinov, L. Davydov and P. G. Smirniotis,
Appl. Catal., B, 2001, 32, 11.
´
vy, V. Marcu and V. I.
Parvulescu, J. Photochem. Photobiol., A, 2005, 174, 106.
42 N. Soggiu, H. Cardy, J. L. H. Jiwan, I. Leray, J. P. Soumillion and
S. Lacombe, J. Photochem. Photobiol., A, 1999, 124, 1.
43 E. Baciocchi, C. Crescenzi and O. Lanzalunga, Tetrahedron, 1997,
53, 4469.
14 C. Anderson and A. J. Bard, J. Phys. Chem. B, 1997, 101, 2611.
15 G. Lassaleta, A. Fernandez, J. P. Espinos and A. R.
Gonzalez-Elipe, J. Phys. Chem., 1995, 99, 1848.
16 A. N. Lazarev, Vibrational Spectra and Structure of Silicates,
Consultants Bureau, 1972, pp. 165.
44 S. M. Bonesi, M. Freccero and A. Albini, Phys. Org. Chem., 1999,
12, 703.
17 Astorino, J. B. Peri, R. J. Willey and G. Busca, J. Catal., 1995, 157,
482.
45 R. S. Davidson and J. E. Pratt, Tetrahedron Lett., 1983, 52, 5903.
46 M. A. Fox and A. A. Abdel-Wahab, Tetrahedron Lett., 1990, 31,
4533.
18 A. Zecchina, S. Bordiga, C. Lamberti, G. Ricchiardi, C. Lamberti,
G. Ricchiardi, D. Scarano, G. Petrini, G. Leofanti and M.
Mantegazza, Catal. Today, 1996, 32, 97.
ꢀc
This journal is the Owner Societies 2008
6570 | Phys. Chem. Chem. Phys., 2008, 10, 6562–6570