photocatalytic activity of doped and undoped titanium ... › pdf › 32-43-pmr-jan13.pdf ·...

� •Platinum Metals Rev., 2013, 57,�(1),�32–43•

32 © 2013 Johnson Matthey

Photocatalytic Activity of Doped and Undoped Titanium Dioxide Nanoparticles Synthesised by Flame Spray PyrolysisPlatinum-doped TiO2 composites show improved activity compared to commercially available product

http://dx.doi.org/10.1595/147106713X659109 http://www.platinummetalsreview.com/

Irene E. Paulauskas and Deena R. Modeshia

Bio Nano Consulting Ltd, 338 Euston Road, London NW1 3BT, UK,

and London Centre for Nanotechnology, University College London, 17–19 Gordon Street, London WC1H 0AH, UK

Tarek T. Ali, Elsayed H. El-Mossalamy, Abdullah Y. Obaid and Sulaiman N. Basahel

Chemistry Department, Faculty of Science, King Abdulaziz University, PO Box 80203, Jeddah 21589, Saudi Arabia

Ahmed A. Al-Ghamdi

Physics Department, Faculty of Science, King Abdulaziz University, PO Box 80203, Jeddah 21589, Saudi Arabia

Felicity K. Sartain*

Bio Nano Consulting Ltd, 338 Euston Road, London NW1 3BT, UK

*Email: [email protected]

The photocatalytic activities of a series of titanium dioxide (TiO2) based nanoparticles, synthesised via flame spray pyrolysis (FSP), have been investigated and compared with the commercially available Evonik Aeroxide® TiO2 P 25 (P 25). The effects of metal ions aluminium, tin and platinum, respectively, on the physical and chemical properties of the TiO2 nanoparticles are reported. The set of six samples were characterised by X-ray powder diffraction (XRD), transmission electron microscopy (TEM), inductively coupled plasma-mass spectrometry (ICP-MS) and ultraviolet-visible (UV-vis) diffuse reflectance spectroscopy. Specific surface areas were determined using nitrogen adsorption and desorption measurements. Subsequent photocatalytic studies of the degradation of methyl orange (MO) dye under UV irradiation demonstrated that addition of Al and Sn had a negative effect on catalytic performance, whereas the addition of ≤0.7 at% Pt to each sample enhanced photocatalytic activity. Most interestingly, the Pt-doped composite samples (TiO2-Sn/Pt and TiO2-Al/Pt) both showed a significantly higher rate of degradation of MO, when compared to P 25. All Pt-doped samples show an increased visible photon absorption capacity. The relationships between the physical and chemical characteristics are discussed in relation to photocatalytic performance.

1. IntroductionThe use of semiconductors as materials for photocatalytic applications has been intensively studied since Fujishima and Honda reported their water electrolysis work in 1972 (1). Research into the development of photocatalysts for environmental applications was subsequently initiated by Frank and Bard (2) who investigated the use of TiO2 powder for the decomposition of highly polluting cyanide ions in water. Following these studies, photocatalysis using TiO2 for the degradation of wastewater pollutants has attracted a lot of attention due to its inherent

http://dx.doi.org/10.1595/147106713X659109� •Platinum Metals Rev., 2013, 57,�(1)•


photoactivity, chemical and biological stability, non-toxicity, low cost and well known synthesis methods (3). Numerous approaches have been applied to improve the effectiveness of TiO2, and one, which is now well established, is the use of nanoparticles rather than a bulk powder.

The functional performance of nanoparticles is directly dependent on their particle size, nanostructure and morphological organisation (4, 5). Thus, it is clear that the method of synthesis will strongly influence the photocatalytic activity of the final compound. A number of techniques for the synthesis of TiO2 have been reported, including hydrothermal methods (6), sol–gel methods (7) and flame synthesis (FSP) (8). Of these, FSP represents an ideal synthetic method since its versatility allows for fast and cost-effective mass production of homogenous nanoparticles. Its use for the production of undoped and doped TiO2 has been widely reported in the literature (9–11) and, when compared with more traditional wet chemistry techniques, it provides a methodology that facilitates the synthesis of homogeneously doped materials (12).

Composite/coupling formation and doping of TiO2 are examples of other approaches used to improve efficiencies, where generally the aim is to increase the separation between the photogenerated electrons and holes and thus improve upon the low quantum yield (13). Addition of such elements tends to alter the aggregate size and morphology of the bulk TiO2, whilst also conferring phase changes within the crystal structure, which in turn alter the band gap of the photocatalyst. Anatase, rutile and brookite are the three distinct crystallographic forms of TiO2, of which anatase is widely accepted to be most photocatalytically active (14). It should be noted that efficiencies of TiO2-based catalysts are not solely influenced by one pure phase but by the ratios of different phases present; an excellent example of this is Evonik Aeroxide® TiO2 P 25 (P 25), which consists of a mixture of two phases (~75% anatase and ~25% rutile) (8, 14, 15), and has been shown to have higher photocatalytic activity than pure anatase (10, 15). To this end, several doping elements and oxides have been incorporated into TiO2 nanoparticles; these include Sn (10, 16), Al (17, 18), Pt (19, 20), iron(III) oxide (Fe2O3) (21), tungsten trioxide (WO3) (22), and zinc oxide (ZnO) (23, 24). Furthermore, it has also been reported that introducing two metal ions onto nanocrystalline TiO2 particles, rather than mono-doping them, leads to an improved photocatalytic performance (25).

The addition of either Sn (16) or Al (17) to TiO2 nanoparticles has been reported to result in an improved photocatalytic performance for the degradation of dye molecules, when compared to undoped TiO2 samples. Similarly, the synthesis of Pt-doped TiO2 in a flame aerosol reactor has been reported to have a positive effect on the degradation rate of MO with a Pt loading of 0.5% when compared with undoped samples of similar surface area (20). However, the combined effect of these elements in TiO2 and their effect on the photocatalytic properties for the degradation of organic compounds have been less well investigated.

In this study we compare the activity of undoped TiO2, Sn-doped TiO2, Al-doped TiO2 and Pt-doped TiO2 nanoparticles with undoped TiO2 and Pt-doped composite materials synthesised via the FSP technique, as well as with the commercially available P 25. We discuss the general morphology, specific surface area, phase composition, absorption range and photocatalytic behaviour of these particles as a function of the degradation rate of MO, a model organic pollutant/dye, in the presence of UV irradiation.

2. Experimental2.1 Sample PreparationSamples were prepared using a FSP technique, where a liquid precursor feed – metal precursor(s) dissolved in a solvent – is sprayed with an oxidising gas into a flame zone. The spray is combusted and the precursor(s) are converted into nano-sized metal or metal oxide particles, depending on the metal and the operating conditions (26). The technique is flexible and allows the use of a wide range of precursors, solvents and process conditions, thus providing control over particle size and composition. All the samples produced were TiO2-based nanoparticles. The liquid precursor feeds consisted of appropriate mixtures of platinum acetylacetonate (Alfa Aesar), titanium ethoxide (Aldrich), tin 2-ethylhexanoate (Alfa Aesar), aluminium(III) tri-sec-butoxide (Alfa Aesar) and xylene (Fisher). The concentration of the reactants was 0.5 mol l–1. The Al:Ti and Sn:Ti molar ratios were kept at 0.14 and 0.7, respectively, while the Pt loading was kept at ca. 0.4 at% in all doped catalysts.

In a typical run, a flow of 5 ml min–1 of liquid precursor solutions was delivered to the nozzle using a syringe pump (Kd Scientific, KDS200). The precursor solutions were then atomised by 5 l min–1

of dispersant O2 while maintaining a pressure drop between 1–1.5 bar at the nozzle tip. The combustion of



the dispersed droplets by the surrounding supporting methane/oxygen (1.5 l min–1/3.2 l min–1) formed the main core flame. A sintered metal plate ring provided an additional O2 sheath flow (5 l min–1) surrounding the supporting flame. Calibrated mass flow controllers (MKS Instruments) were used to monitor all gas flows. Product particles were collected on a glass microfibre filter (Whatman GF/A grade, 15 cm in diameter) with the aid of a vacuum pump (Busch Seco SV 1025 C pump from West Technology Systems).

The following six samples were synthesised: FSP-TiO2, TiO2-Al, TiO2-Sn, TiO2/Pt, TiO2-Al/Pt and TiO2-Sn/Pt. In order to obtain a reference for effectiveness of these samples, commercially available Evonik Aeroxide®

TiO2 P 25 (donated for this study) was investigated in parallel.

2.2 Sample CharacterisationTEM analysis was performed to determine basic morphology and the average particle size of the TiO2 nanoparticles. The instrument used was a high-resolution JEOL 2010 TEM microscope. The samples were prepared by suspending 10 mg of the mixed oxide in 5 ml ethanol followed by ultrasonication for 5 min. 0.5 ml of the suspension was dropped on top of a copper mesh for TEM imaging and dried in air.

Elemental analysis was carried out as follows: samples were prepared by dissolving 10 mg of the TiO2 nanoparticles in 4 ml sulfuric acid at 523 K for up to 1 h. The solution was subsequently allowed to reach room temperature and any undissolved matter was removed by using a 0.2 mm PTFE Millipore membrane filter. ICP-MS analysis was conducted using a Perkin Elmer/Sciex ELAN® Dynamic Reaction Cell (DRCPlus) coupled with a Perkin Elmer AS-93plus Autosampler with Elan v. 3.3 software for data collection. A crossflow nebuliser was used with a Scott spray chamber and the basic settings consisted of a RF power of 1100 W, a plasma gas flow of 15 l min–1, a nebuliser flow of 0.97 l min–1 and a sample flow rate of 0.4 ml min–1.

XRD analysis of all the samples was determined at room temperature using an X’Pert PRO powder diffractometer (PANalytical BV, The Netherlands) with a silver anode X-ray source and a scintillating crystal detector. The samples were scanned over a 2θ range of 25º–70º in steps of 0.01º.

The surface area analysis was made using a Tristar 3000 Micromeritics instrument, and the software used was the Tristar 3000 v. 6.04. This analyser uses physical adsorption and capillary condensation principles to obtain information about the surface area and porosity of a solid material. More specifically, the adsorption

and desorption isotherms of N2 were carried out at 77 K. Specific surface areas were calculated according to the Brunauer, Emmet and Teller (BET) equation from the adsorption isotherm. The pore-size distribution of the samples was calculated from desorption branch using the Barrett, Joyner and Halenda (BJH) method.

UV-vis diffuse reflectance measurements were carried out at room temperature and performed on a Perkin-Elmer Lambda 950 spectrophotometer equipped with an integrating sphere to test absorption shifts in the wavelength range from 300 to 500 nm.

2.3 Photocatalysis StudyPhotocatalytic activity measurements were carried out in a custom-built reactor, details of which have been previously reported (24). The model dye solution selected for this study was methyl orange (MO) and a standard solution of 10 mg l–1 was used. In a typical experiment, 50 mg of the TiO2-based nanoparticles to be tested were added to 100 ml of the MO solution and mixed in an ultrasonic bath for 5 min, then placed into the reactor. Once in the reactor, the solutions were stirred at 300 revolutions per minute (rpm) for 30 min in the dark at room temperature to allow for chemisorption to occur. Once this time had elapsed, a sample of the solution was taken, and the remaining reaction solution was irradiated with UV light. Further samples were taken every 5 to 10 min until all of the MO had degraded.

Using the UV-vis spectrophotometer, the absorbance peak of each sample solution was measured at 465 nm to determine the MO concentration. The reaction rate constant (k) was then estimated assuming first order kinetics using Equation (i):

(i)

where C0 is the initial concentration, C is the concentration at time t, and k is the reaction rate constant. This procedure was repeated three times with fresh nanoparticles each time to determine the standard deviation of the measurement.

3. Results and Discussion3.1 CharacterisationThe morphologies of the TiO2-based catalysts were investigated by TEM (see Figure 1) and the common issue of agglomeration of nanoparticles was observed in each sample, which meant that it was not straightforward to find single monolayers.

0 –Cln ktC

=



The FSP-TiO2 particles were found to be regular in shape (polyhedral) with a size distribution between 5 and 20 nm, comparable to the other samples except for those containing Al, which on average showed a larger particle size. For the TiO2-Al and TiO2-Sn samples the particles appeared to become more rounded, however there was no visible addition of the respective metal ions observed (data not shown). For the TiO2/Pt sample the TiO2 crystalline planes were easily observed, but they were found to have poor homogeneity with very irregular

shapes and the size distribution was determined to be between 5 and 20 nm. Although the Pt doping particles were too small to be characterised by chemical analysis techniques such as energy dispersive X-ray spectroscopy, the presence of small dark particles on the surface of the TiO2 (Figure 1(d)) can be observed. This result is also consistent with previous observations found in the literature (20), and indicates that unlike the Sn and Al ions the Pt ions remain on the surface of the TiO2 rather than being incorporated into the crystal lattice.

(a)

50 nm 20 nm

(b) (c) 20 nm

(d) 5 nm (e) 20 nm (f) 5 nm

(g)

20 nm

(h)

5 nm

Fig. 1. TEM images of doped and undoped TiO2 nanoparticles synthesised by FSP: (a) and (b) FSP-TiO2; (c) and (d) TiO2/Pt; (e) and (f) TiO2-Sn/Pt; and (g) and (h) TiO2-Al/Pt



Interestingly, the TiO2-Sn/Pt nanoparticles were found to be more rounded than the TiO2/Pt, having a comparable platelet size with the distribution also ranging between 5 and 20 nm. In this sample no significant feature was depicted that could be attributed to the Pt dopant. The TiO2-Al/Pt nanoparticles were also more circular, but with less uniform edges and had a larger size distribution ranging between 5 and 30 nm. Additionally, in this case some dark particles comparable to those observed in the TiO2/Pt sample were also seen (Figure 1(h)).

Results from the ICP elemental analysis revealed that there was only a fractional deviation between the theoretical and actual percentages for each of the TiO2 samples synthesised that contained additional metal ions (see Table I). The atomic percentages of Al and Sn were larger – 14 at% and 7 at% respectively – than for the Pt ion (≤0.7 at%), so only Pt can be truly considered as a dopant in these samples. The rationale for making the Al and Sn composites with varied at% was that from previous studies it had been found that at each of these respective percentages the balance of anatase to rutile phases shifted such that the rutile phase in TiO2 was dominant (data not shown).

The XRD pattern of the FSP-TiO2 sample (see Figure 2) showed that the anatase polymorph

dominates (19, 27, 28). Smaller peaks were observed at 2θ values of 27º and 36º (27) and are indicative of the rutile polymorph. The addition of Al, Sn and the Pt dopant resulted in a variation in the ratio of rutile and anatase across all of the samples (Figure 2). Synthesis of TiO2 with Al and Sn respectively significantly altered the phase ratio, with the rutile phase being the dominant phase in each sample, as anticipated. The TiO2/Sn sample showed a higher proportion of rutile phase to anatase, despite having a lower at% of the composite metal compared to the TiO2/Al sample. This can, in part, be explained by comparing the valence and ionic radii of the respective composite metals with those of Ti4+. Substitution of the Ti4+ (ionic radius: 0.61 Å) ions by other ions of similar ionic radius induces a shift in phase from anatase to rutile (Al3+ = 0.53 Å, Sn4+ = 0.69 Å) (10, 19). Thus, the matching valence and comparable atomic radii of Sn4+ and Ti4+ suggests that substitution between the two can occur readily and induce the observed alteration in phase ratio.

Given the high at% of the samples with Al and Sn, it is plausible to assume the presence of their respective metal oxides; however, for the TiO2-Al sample the Bragg reflections associated with such phases have not been observed via XRD. In the TiO2-Sn diffractogram, the sharp peak at ~33º is consistent

Table I

Characterisation Results for the Synthesised and Commercial Titanium Dioxide Nanoparticles

(Doped and Undoped)

Nanoparticles Preparation, at% (Sn or Al/Ti), (Pt/Ti)

ICP, at%(Sn or Al/Ti), (Pt/Ti)

Surface area,m2 g–1

Particle size,nma

Reaction rate constant,k, × 10–3 s–1

P 25 n/a n/a 58 n/a 0.90 ± 0.20

FSP-TiO2 n/a n/a 88 5–20 0.36 ± 0.02

TiO2/Pt 0.4% 0.4% 81 5–20 0.57 ± 0.07

TiO2-Sn 7% 7% 80 5–20 0.18 ± 0.03

TiO2-Sn/Pt 7%, 0.4% 7%, 0.3% 74 5–20 2.30 ± 0.20

TiO2-Al 14% 15% 42 10–30 0.18 ± 0.02

TiO2-Al/Pt 14%, 0.4% 14%, 0.7% 90 5–30 2.75 ± 0.60

a From TEM images



with flame-made Ti-Sn oxides (10) that exhibited a similar peak at 33.9º corresponding to the (101) plane of SnO2. In this instance, the SnO2 particles within the structure act as seed nuclei favouring the formation of rutile TiO2 (10). In the case of Al, where substitution of the dopant also infers a change in valence, it has been reported that doping can create oxygen vacancies in the TiO2 matrix, which favours the anatase to rutile transformation (12).

The addition of Pt to the samples had less impact on the ratio between the phases, with the proportion of rutile increasing in the TiO2-Al/Pt sample, and also in the TiO2/Pt sample but to a lesser extent compared to their respective undoped samples. The SnO2 was lost in the TiO2-Sn/Pt sample. The promotion of the rutile phase in the Pt-doped TiO2 and TiO2-Al samples could be attributed to the generation of the dopant metal oxide during synthesis, which is possible through the substitution of a Ti4+ ion with Pt4+ which has a similar ionic radius of 0.63 Å. This in turn may also provide a nucleation site for rutile TiO2 (10, 19). The addition of the Pt to the composite samples resulted in each of them being composed of anatase to rutile phases in an approximate ratio of at least 20:80, despite the

variances in at% of the Sn4+ and Al3+. Additionally, for both the TiO2-Al and TiO2-Sn composites after doping with Pt there was a reduction in crystallite size, as indicated by the broader peaks recorded for each powder sample.

The surface area of the FSP-TiO2, determined by BET, was found to be ~88 m2 g–1, which is larger than that recorded for the commercially available P 25 (~58 m2 g–1). The addition of Sn resulted in a slight decrease in surface area, whereas addition of Al caused a significant decrease to 42 m2 g–1. The observed decrease may be attributed to the creation of oxygen vacancies upon Al3+ substitution with Ti4+. In turn, this enhances oxygen diffusion, which increases the sintering rate and thus particle size (previously observed by TEM analysis) (10). The addition of the Pt dopant to the TiO2-Al sample resulted in a recovery of the surface area to 90 m2 g–1; whilst upon Pt doping the TiO2 and TiO2-Sn samples, a slight decrease was observed compared to TiO2 and TiO2-Sn, respectively (Table I). The surface area for all the Pt-doped samples was larger than that of the commercially available P 25. The porosity of the samples was also calculated from the adsorption branch of the isotherms using the BJH formula. For all

25 30 35 40 45 50 55 60 65

Diffraction angle, º2θ

Diff

ract

ed in

tens

ity, a

rbitr

ary

units

FSP-TiO2

TiO2-Al

TiO2-Sn

TiO2/Pt

TiO2-Al/Pt

TiO2-Sn/Pt

P 25

A

R R RR RAA A

A AAA

Fig. 2. XRD diffraction patterns for the six TiO2 based FSP synthesised materials and the commercially available P 25 sample for reference. Peaks representative of the anatase (A) and rutile (R) phases are labelled accordingly



of the samples, nitrogen adsorption isotherms showed a hysteresis loop typical of mesoporous materials (type IV isotherm according to the International Union of Pure and Applied Chemistry (IUPAC) classification). This is indicative of porosity between particles rather than within individual particles (11).

Finally, the band gap of each of the samples was determined using UV-vis diffuse reflectance measurements (see Figure 3). The band edges for all samples were found to be in the UV region, which suggests that they will all be photocatalytically active under UV irradiation (350–400 nm). Interestingly, the absorption range of the FSP-TiO2 was almost identical to that of P 25 and a red-shift was observed in the absorption range of both the TiO2-Al and TiO2-Sn when compared to the undoped TiO2. Previously it has been reported that such a shift is indicative of the incorporation of metal ions into the TiO2 framework (17) and, additionally, of the presence of the rutile phase within the structure (14). The addition of Pt dopant to the samples resulted again in slight changes in the absorption range; in the case of TiO2/Pt a small red-shift was seen, but for the composite samples a slight blue-shift was observed. It should be noted that by comparison with the FSP-TiO2, the absorption range for all of the other samples was red-shifted. Interestingly, all the Pt samples showed that they have an increased visible photon absorption capacity, something which distinguishes them from all the other samples tested.

3.2 Photocatalytic Degradation of Methyl Orange DyeThe photocatalytic activity of each sample was investigated by monitoring the photodegradation of MO dye and in all cases the initial concentration of the MO solution was 10 mg l–1. Since in most cases an exponential decay was recorded, the photoactivity profile of each TiO2 sample was fitted assuming first order kinetics, as described in Section 2.3 and by Equation (i). The commercially available P 25 TiO2 sample was also investigated, to provide a benchmark against which the catalysts synthesised using FSP could be measured. The intial rate of reaction using the FSP-TiO2 was approximately half of that determined for the P 25 sample (Figure 4 and Table I), despite the former having a significantly larger surface area. This difference is likely to be due to a combination of factors, including the slight variance in the ratio between the anatase and rutile phases, and the activity of the FSP-TiO2 sample may be reduced due to the presence of residual carbon generated during

the combustion process. Additionally, P 25 is known to have a unique microstructure, which enables intimate contact between the phases and, in turn, increases the efficiency of the electron-hole separation (19).

Comparing the FSP-TiO2 with the TiO2-Al and TiO2-Sn samples, both the latter samples showed a reduced photocatalytic performance compared with the undoped sample (Figures 4 and 5). The BET analysis indicates that there was little difference between the surface areas of the TiO2-Sn and TiO2 samples, whereas the TiO2-Al sample had a surface area almost half that of the former two samples. This observed reduction in surface area for the TiO2-Al sample provides an explanation for its slower rate of reaction. However, the TiO2-Sn sample had the slowest rate constant of all the samples investigated, despite having a comparable surface area to the more reactive samples. This reduced rate of reaction may be due to having a significant fraction of the surface occupied by SnO2. The small red-shift in the absorption bands and the XRD diffractograms recorded for the composite samples may provide some rationale for these results. The addition of the metal ions Al or Sn resulted in a significant shift in the ratios of the crystal phases present in the structure, with an increase in the amount of rutile being observed compared to the undoped TiO2.

The addition of the Pt dopant to all the samples was shown to improve reaction rates and in the case of TiO2-Al and TiO2-Sn approximately a 10-fold and 5-fold enhancement in photocatalytic activity, respectively, was recorded. In fact the TiO2-Al/Pt and TiO2-Sn/Pt samples had the highest MO degradation rates compared to the other samples, including P 25 (Figure 4). The addition of Pt in all cases conferred a slight increase in the content of rutile within the TiO2 nanoparticles, a phenomenon previously observed when similar compounds were synthesised by FSP (19). Bickley et al. (14) have discussed at length the multiphasic nature of P 25 TiO2, and suggest that it is the complex relationship between anatase and rutile phases that contributes to its high photocatalytic activity. Here, the addition of the metal ions, in their respective ratios, appears to generate such biphasic TiO2-based materials. This, along with the comparably large specific surface areas and enhanced visible light absorption observed for all of the FSP samples after Pt doping, may explain the enhanced activity measured.

Additionally, it has been reported that doping TiO2 nanoparticles with Pt improves photocatalytic activity (19, 20). However, it is important to control



1

0.8

0.6

0.4

0.2 0

Abs

orba

nce

300 350 400 450 500 550 600

Wavelength, nm

TiO2-Sn

TiO2-Pt/Sn

TiO2-Al

TiO2-Al/Pt

(b)

P 25

1

0.8

0.6

0.4

0.2

0

Abs

orba

nce

300 350 400 450 500 550 600Wavelength, nm

FSP-TiO2

TiO2/Pt

TiO2-Al

TiO2-Sn

(a)

Fig. 3. UV-vis diffuse reflectance spectra of the TiO2-based samples: (a) P 25, FSP-synthesised TiO2, TiO2-Al, TiO2-Sn and TiO2/Pt; (b) Pt-doped TiO2 nano-composites, TiO2-Al and TiO2-Sn samples are also shown for comparison. Shaded area indicates the irradiance range of the UV light source used for the photocatalytic activity studies



500 1000 1500

1.2 1 0.8 0.6 0.4 0.2

(b)

Time, seconds

P 25

FSP:TiO2

TiO2/Pt

TiO2-Al

TiO2-Sn

–Ln(

C/C o

)

0

Met

hyl o

rang

e co

ncen

trat

ion,

%

1000 2000 3000 4000

P 25

FSP-TiO2

TiO2/Pt

TiO2-Sn

(a)

100

80

60

40

20

Time, seconds

TiO2-Al

0

Fig. 4. Photocatalytic degradation of the methyl orange solution by the P 25 FSP-synthesised TiO2, TiO2-Al, TiO2-Sn and TiO2/Pt: (a) Change in the methyl orange concentration as a function of time; (b) Estimation of the initial reaction rate constant based on Equation (i)

the concentration of Pt used since photocatalytic properties tend to decrease as the Pt concentration increases above 1 at% (19). From the ICP analysis

it was found that 0.4, 0.3 and 0.7 at% Pt had been added to the TiO2, TiO2-Sn and TiO2-Al respectively, and each gave an improved performance over their



500 1000 1500

(b)

Time, seconds

TiO2-Sn

TiO2-Sn/Pt

TiO2-Al

TiO2-Al/Pt

–Ln(

C/C o

)

1.8

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0

Met

hyl o

rang

e co

ncen

trat

ion,

%

1000 2000 3000 4000

TiO2-Sn

TiO2-Sn/Pt

TiO2-Pt/Al

(a)

100

80

60

40

20

Time, seconds

TiO2-Al

0

Fig. 5. Photocatalytic degradation of the methyl orange solution by the Pt-doped TiO2 nano-composites, TiO2-Al and TiO2-Sn samples are also shown for comparison: (a) change in the methyl orange concentration as a function of time; (b) estimation of the reaction rate constant based on Equation (i)

undoped equivalents. The TEM analysis clearly showed the presence of Pt on the surface of the TiO2/Pt and TiO2-Al/Pt samples, though this feature was less apparent than that observed in the TiO2-Sn/Pt sample. The presence of such particles on the surface can lead to increased efficiency of electron-hole separation by trapping or removing electrons from the TiO2 surface

(19, 20), and this in turn can result in enhanced photocatalytic activity.

4. ConclusionsSix TiO2-based nanocatalysts were synthesised using a FSP technique and it was found that the addition of Al and Sn to the TiO2, at a ratio of 0.14 and 0.7, respectively,



resulted in significant phase changes within the crystal structure. In the former case there was a large decrease in surface area which could be attributed to an increase in oxygen vacancies, which in turn increases the particle size. For each of these composite samples a reduction in photocatalytic activity was observed when compared to the pristine TiO2. Conversely, the addition of Pt as a dopant in all samples resulted in an enhancement in photocatalytic activity, with both the TiO2-Al/Pt and TiO2-Sn/Pt samples having a higher reactivity than the commercially available P 25 in degrading methyl orange.

The effects of surface area, ratios of crystal structure, and the metal dopants have been discussed to provide rational explanations for the photocatalytic activities observed. Due to the complex nature of these multiphasic and doped materials it is not straightforward to determine the precise relationship between the surface and bulk properties which result in the enhancements seen, and further studies are required to fully understand the interactions between the chemical and structural properties of these materials.

AcknowledgementsThe authors thank Peter Bishop, Benedicte Thiébaut, Weiliang Wang, Gregory Goodlet and the Analytical Department (Johnson Matthey Technology Centre (JMTC), Sonning Common, UK); Junwang Tang (University College London, UK); David Sarphie, Sanjay Santhasivam and Ainara Garcia Gallastegui (Bio Nano Consulting, UK) for valuable discussions and experimental assistance; Andrew Cakebread (King’s College London, UK) for ICP analysis; Zlatko Saracevic (Department of Chemical Engineering and Biotechnology, University of Cambridge, UK) for BET measurements; and the Deanship of Scientific Research at King Abdulaziz University, Saudi Arabia, for the support of this project (T/80/429).

References 1 A. Fujishima and K. Honda, Nature, 1972, 238,

(5358), 37

2 S. N. Frank and A. J. Bard, J. Am. Chem. Soc., 1977, 99, (1), 303

3 A. Fujishima and X. Zhang, C. R. Chim., 2006, 9, (5–6), 750

4 D. Barreca, G. Carraro, V. Gombac, A Gasparotto, C. Maccato, P. Fornasiero and E. Tondello, Adv. Funct. Mater., 2011, 21, (14), 2611

5 R. Van de Krol, Y. Liang and J. Schoonman, J. Mater. Chem., 2008, 18, (20), 2311

6 Y. Zhao, J. Liu, L. Shi, S. Yuan, J. Fang, Z. Wang and M. Zhang, Appl. Catal. B: Environ., 2010, 100, (1–2), 68

7 E. Sánchez, T. López, R. Gómez, X. Bokhimi, A. Morales and O. Novaro, J. Solid State Chem., 1996, 122, (2), 309

8 G. P. Fotou, S. Vemury and S. E. Pratsinis, Chem. Eng. Sci., 1994, 49, (24), Part 2, 4939

9 R. Mueller, L. Mädler and S. E. Pratsinis, Chem. Eng. Sci., 2003, 58, (10), 1969

10 S. Vemury and S. E. Pratsinis, J. Am. Ceram. Soc., 1995, 78, (11), 2984

11 R. Strobel and S. E. Pratsinis, Platinum Metals Rev., 2009, 53, (1), 11

12 W. Y. Teoh, R. Amal, L. Mädler and S. E. Pratsinis, Catal. Today, 2007, 120, (2), 203

13 C. Wang, C. Shao, X. Zhang and Y. Liu, Inorg. Chem., 2009, 48, (15), 7261

14 R. I. Bickley, T. Gonzalez-Carreno, J. S. Lee, L. Palmisano and R. J. D. Tilley, J. Solid State Chem., 1991, 92, (1), 178

15 T. Ohno, K. Sarukawa, K. Tokieda and M. Matsumura, J. Catal., 2001, 203, (1), 82

16 C. Xiong and K. J. Balkus, Jr., J. Phys. Chem. C, 2007, 111, (28), 10359

17 S. Liu, J. Lu, Q. Feng and W. Tang, Chin. J. Chem. Eng., 2011, 19, (4), 674

18 S. Liu, G. Liu and Q. Feng, J. Porous Mater., 2010, 17, (2), 197

19 W. Y. Teoh, L. Mädler, D. Beydoun, S. E. Pratsinis and R. Amal, Chem. Eng. Sci., 2005, 60, (21), 5852

20 V. Tiwari, J. Jiang, V. Sethi and P. Biswas, Appl. Catal. A: Gen., 2008, 345, (2), 241

21 B. Pal, M. Sharon and G. Nogami, Mater. Chem. Phys., 1999, 59, (3), 254

22 T. Ohno, F. Tanigawa, K. Fujihara, S. Izumi and M. Matsumura, J. Photochem. Photobiol. A: Chem., 1998, 118, (1), 41

23 M. R. Vaezi, J. Mater. Process. Technol., 2008, 205, (1–3), 332

24 M. S. Hamdy, P. Nickels, I. H. Abd-Elmaksood, H. Zhou, E. H. El-Mossalamy, A. O. Alyoubi, S. Lynch, A. Nathan and G. Thornton, J. Photochem. Photobiol. A: Chem., 2012, 228, (1), 1

25 P. Yang, C. Lu, N. Hua and Y. Du, Mater. Lett., 2002, 57, (4), 794

26 B. Thiébaut, Platinum Metals Rev., 2011, 55, (2), 149

27 B. Li, X. Wang, M. Yan and L. Li, Mater. Chem. Phys., 2003, 78, (1), 184

28 K. R. Lee, S. J. Kim, J. S. Song, J. H. Lee, Y. J. Chung and S. Park, J. Am. Ceram. Soc., 2002, 85, (2), 341

29� F.� Sayılkan,� M.� Asiltürk,� P.� Tatar,� N.� Kiraz,� S.� Sener,� E.� Arpaç� and�H.� Sayılkan,�Mater. Res. Bull., 2008, 43, (1), 127



The Authors

Irene Paulauskas is a Materials Science Engineer with a PhD obtained from the University of Tennessee, USA. She has experience in the development, evaluation and characterisation of alloys and semiconductors for a variety of green energy applications such as fuel cell components, hydrogen generation and water purification.

Deena Modeshia’s research interests are focused on the synthesis of oxide materials as depollution catalysts and for water purification. These are synthesised via a CVD or solvothermal route to form pgm doped oxides both in situ and ex situ. Other interests include the inclusion of ceria into a titanate pyrochlore structure to increase the efficiency of the material for use as a catalyst in both solid oxide fuel cells and low temperature water-gas shift catalysis.

Tarek Ali’s research interests include acid-base catalysis, oxidation catalysts, photocatalysis, supported metal catalysts and the preparation of nanocomposite materials by novel methods. He is currently Assistant Professor of Physical Chemistry at King Abdulaziz University, Saudi Arabia.

Professor Elsayed El-Mossalamy investigates the synthesis, characterisation and applications of experimental charge transfer complexes in organic and inorganic materials, nanoparticles and nanocomposites. More recently his work has focused on the study of optical and electrical properties of nanomaterials.

Professor Abdullah Obaid is a Professor of Physical Chemistry at King Abdulaziz University. His main research interest is in the field of thermal analysis and physicochemical characterisation of inorganic and organic materials. He has also worked on the synthesis and characterisation of platinum-doped and undoped titanium dioxide catalysts.

Professor Sulaiman Basahel is a Professor of Physical Chemistry at King Abdulaziz University. His research interests cover heterogeneous catalysis, nanomaterials, advanced materials, nanoparticles and the synthesis and characterisation of metal oxide-supported platinum catalysts.

Felicity Sartain is a Chemist with Bio Nano Consulting, UK, whose current research interests are in the development and implementation of nanocatalysts for cleantech applications, specifically carbon capture and water purification.

Professor Ahmed Al-Ghamdi is a Professor of Solid State Physics at King Abdulaziz University. His research interests lie in the preparation, characterisation and application of bulk and thin film organic and inorganic materials in the solid state. He has been working on nanocomposites for eight years, and is now working on the development of optical and electrical devices utilising semiconductor nanomaterials.

photocatalytic activity of doped and undoped titanium ... › pdf › 32-43-pmr-jan13.pdf ·...

Documents