M. Bingham and A. Mills
JournalofPhotochemistry&PhotobiologyA:Chemistryxxx(xxxx)xxxx
limit severely the latter's use in producing electricity using a fuel cell
[24]. Following on from this work [23], this paper explores the pho-
tonic efficiency and selectivity of a number of different M/TiO2 pho-
tocatalysts, where M (the co-catalyst) is Pt, Pd, Au and Ag, for reaction
(1) at 100 °C in a continuous gas phase flow reactor.
2. Experimental
Fig. 1. A series of digital images of the 0.2 wt% metal deposited TiO2 photo-
catalyst powders.
2.1. Materials
Unless stated otherwise, all chemicals were purchased from Sigma
Aldrich and used as received. The Aeroxide® P25 TiO2 powder was a
gift from Evonik (formerly Degussa). All gases were purchased from
BOC and certified to be of 99.999% purity.
obtained for each 0.2 wt% M/TiO2 photocatalysts and the results are
illustrated in Fig. 2, with examples of the metal islands highlighted by a
red ring. All the images obtained by TEM reveal an average TiO2 par-
ticle diameter of ca. 30 nm, which is consistent with a measured BET
surface area of ca. 50 m2 g−1. In contrast, the size of the metal particles,
although largely consistent for each metal, varied significantly from
metal to metal. Thus, from the TEMs, the average metal particle dia-
meter for the 0.2 wt% M/TiO2 photocatalysts were: 1 nm, 6 nm, 4 nm
and 3 nm for M = Pt, Pd, Au and Ag, respectively. The TEMs of the
photocatalysts before and after use showed no obvious change, with the
metal particle sizes remaining unchanged.
Note that in a previous paper [23], using ostensibly the same Pt/
TiO2 photocatalyst, aggregated Pt particles of 2−3 nm were observed
using HAADF STEM and this might appear to suggest some lack or re-
liability in the method of production, the incipient wetness method –
which would be surprising given its widespread use [23,26–29].
However, a closer inspection of the HAADF STEM image reveals that
whilst some aggregated particles were present, the bulk of the Pt par-
ticles appeared to be ca. 0.8 nm in diameter, which is reasonably con-
sistent with our measured value of 1 nm, especially give the slight
differences in preparation methods, such as the level of H2 used in the
reduction step, i.e. 5% in [23] cf. 100% here.
2.2. Photocatalyst preparation
All M/TiO2 photocatalysts, where M = Pt, Pd, Au and Ag, were
prepared using the incipient wetness method [9,23,25,26]. Thus, for
the typical photocatalyst Pt/TiO2 (0.2 wt% Pt), 58 mg of the metal salt
(H2PtCl6.6H2O) were dissolved in 80 cm3 of water and the resulting
solution then added dropwise to 10 g of the P25 TiO2 powder which
was agitated continuously. The resulting dispersion was dried at 110 °C
overnight before being calcined at 500 °C for 2 h and then sieved to an
aggregated particle size between 250−400 μm. The powder was sub-
sequently reduced at 200 °C for 2 h in a tubular furnace under a steady
flow of pure H2 (50 cm3 min−1), thereby yielding the 0.2 wt% Pt/TiO2
photocatalyst. Note: the incipient wetness method, with its final re-
duction step, is often used by the catalyst community to produce na-
noparticulate metal dispersions on a wide variety of inert substrates
[23,26–29]. In this work, as noted elsewhere, the conversion of the
cream colour PtO2 coated TiO2 powder particles by the hydrogen pro-
duces a striking and rapid colour change to a grey powder, commonly
assumed to be due to the reduction of the metal oxide to the metal, as
illustrated by the before and after photographic images illustrated in
the Fig. S2 in the SI section; no further colour change is effected by
increasing the temperature of reduction even to 500 °C. Temperature
programme reduction studies of the Pt/TiO2 catalyst produced using
this technique show that the reduction is readily effected at 200 °C
2.4. Photocatalytic activity testing: in the gaseous phase
In the study of reactions (1) and (2) in the gas phase, using different
M/TiO2 photocatalyst samples, a serpentine ‘S-bend’ borosilicate glass
reactor with internal and external diameters 3.75 mm and 6 mm, re-
spectively, was loaded with ca. 2.4 g (
2%) of each photocatalyst
The above method was also used to prepare the other M/TiO2
photocatalysts, where M = Pd, Au and Ag, at 0.2 wt%, using the fol-
lowing metal salts: PdCl2, HAuCl4.3H2O and AgNO3, respectively,
which produced coloured products consistent with the generation of
nanometre sized metal particles [14]. The prompt generation of H2
observed upon illumination of any of the M/TiO2 photocatalysts sug-
gest little or no oxide layer exists on these pre-reduced metal particles.
powder under test; the powder was held in place with small plugs of
glass wool at the top and bottom of the serpentine reactor. Once loaded,
the 'S-bend' reactor was inserted into the center of the heated flow re-
actor. The heated flow reactor [23,26] consisted of a semi-cylindrical
ceramic fiber heater (diameter: 14.7 cm; height: 35.7 cm) surrounded
by a quartz cylinder (diameter: 19.7 cm; height of 38.5 cm). The re-
mainder of the quartz cylinder was packed with a thermal insulation
blanket to ensure a constant reactor temperature (ca. 100 °C), although
a small window was left, so as to allow the UV radiation of the ser-
pentine photocatalyst-containing reactor vessel at the center of the flow
reactor, with 365 nm UV light from a 10 W 365 nm narrow band UV
LED (RS Components, LZ1-10UV00). Unless stated otherwise, the irra-
diance was fixed at 9.5 mW cm−2, generated using an applied voltage
of 0.7 V, and placed 10 cm from the reactor. Here, and in all this work,
the UV irradiance was confirmed using a calibrated spectroradiometer
(Gooch & Housego, OL 756). In experiments where the irradiance of the
UV LED was varied, this was achieved by varying the applied voltage.
Further details concerning the photoreactor have been reported else-
2.3. Photocatalyst characterisation
The weight loadings of each prepared catalyst were confirmed using
an Inductively Coupled Plasma-Optical Emission Spectrometer, ICP-
OES, (Agilent, 5110). The photocatalyst powders were also analysed
using: (i) X-ray powder diffraction (XRD) (Panalytical, X’Pert) to con-
firm the crystal structure of the TiO2 and (ii) Brunauer-Emmett-Teller
(BET) (Micromeritics, Tristar 3020) surface area analysis. From these
studies the TiO2 appeared to be an 80:20 mix of anatase and rutile with
a specific surface area of ca. 50 m2 g−1, before and after metal de-
position and annealing.
A digital photographic image was obtained of each of the M/TiO2
photocatalysts as well as images and the results are illustrated in Fig. 1.
Thus, the 0.2 wt% M/TiO2 photocatalyst powders were: dark grey, light
grey, blue and dark red/brown for the Pt, Pd, Au and Ag metals de-
posited powders, respectively.
In a typical reaction run, a stream of Ar containing H2O (10% by
volume, 4.16 mM) and MeOH (5 % by volume, 2.08 mM) was passed
over the photocatalyst powder in the 'S' glass reactor, for 4 h, at a flow
rate of 5 cm3 min−1. After the first 1.25 h, the Pt/TiO2 photocatalyst
sample under test was UV irradiated for 1.5 h, and then returned to the
dark for the remaining 1.25 h. The gas stream was sampled every 0.5 h
The same powders were analysed using Transmission Electron
Microscopy, TEM, (Jeol, JEM-1400) and typical micrographs were
during this 4 h period and subjected to analysis using
a gas
2