Photocatalysis in Elongated Titanates and TiO2

During the last three decades, titanium dioxide has been comprehensively studied as a wide band gap photocatalyst for the oxidation of organic compounds.90 The best TiO2-based catalysts are usually characterised by a highly crystalline structure (which can reduce the recombination of photogenerated carriers); a high specific surface area (for acceleration of the interfacial reaction rate); and an abundance of surface -OH groups (which are required for the generation of OH radicals during photocatalytic reactions). All of these features are intrinsic to elongated titanates and anodised arrays of TiO2.

Photocatalysis in elongated titanates

The optical properties of titanate nanotubes have been recently studied by various methods. The absorption threshold determined from diffuse reflectance spectra is usually very close to the band gap of titanium dioxide,91 which is ca. 3.2 eV. However, more accurate studies of diluted colloidal solutions of TiNT, which allowed the errors caused by elastic light scattering to be avoided, estimated the nanotube band gap as ca. 3.87 eV (ref. 92). This indicates that the band gap of titanate nanotubes is wider than that of TiO2 and closer to that of titanate nanosheets (3.84 eV; ref. 93). Photoluminescence measured at —196 °C from powder-state samples usually shows a band at 2.4 eV (ref. 94), whereas the spectrum from nanotubes dispersed in water92 shows a multiple-line spectrum with several characteristic bands at 3.99, 3.77, 3.54, 3.09, 2.94, 2.51, 2.38, 2.16, 2.08 and 1.99 eV. Discrepancies between these two methods are probably due

Table 5.4 Reported application of nanostructured titanates and Ti02 as supercapacitor.

Catalyst

Method of

Particle size/

Loading

Process reaction

Activity or

Benefits

Ref.

Formula

preparation

um

performance

Ru02/TiNT

Precipitation

n/a

4-21 wt%

Ru0, + H,0 + e" RuÖOH + OH"

230 Fg-1

Early reports

85,86

YOv/TiNF

Adsorption

n/a

YÖ2+ + H2O

n/a

Capacitance is higher than for bulk V205

84

Co(OH)2/

Precipitation

n/a

25-75 wt%

Co(OH), + OH"

229 Fg"1

Capacitance is

88

TiNT

CoOOH + H20 + e~

higher than for

250 Fg-1

bulk Co(OH)2

RUl_vCrv02/

Co-

n/a

4-23 wt%

Rui_,,Crv O, + H,0 + e"

Lower cost than

87

TiNT

precipitation

Ru'i_vCrvÖOH + OH"

631 Fg_1

pure Ru02

Co(OH), +

Co-

n/a

60 wt%

(CoNi')(OH), + OH"

High capacitance

89

Ni(OH),/

precipitation

(C0Ni)00H + H,0 + e"

TiNT

to the strong scattering of light from the solid powder samples, which can mask the photoluminescence signal.

Absorption of light by titanate nanotubes results in the generation of charge carriers, which can eventually relax into a single electron trapped oxygen vacancy (SETOV)95 or be trapped by Ti41 ions forming Ti31 centres, which can cause visible light absorption (see Chapter 3). Transient studies of photogenerated charged carriers in titanate and TiO2-(B) nanotubes have revealed that the lifetime of trapped electrons is longer than that of TiO2 nanoparticles, suggesting an improved charge separation due to the elongated morphology.96

In the presence of oxygen and organic molecules, the photogenerated carriers undergo relaxation processes following the routes shown in Figure 5.8a. Photogenerated electrons diffuse onto the surface of nanotubes and usually reduce oxygen molecules to form peroxo-species. Photogenerated holes also diffuse to the surface and react with surface -OH groups forming OH radicals, which further react with any organic molecules present.90 Conventional TiO2 photocatalysts are operated based on this principle.

It is well known that Na1 impurities can significantly decrease the photo-catalytic activity of TiO2 acting as a recombination centre.90 The high level of sodium ions retained in titanate nanostructures after alkaline hydrothermal synthesis, can also significantly reduce the photocatalytic activity of titanates. This was recently confirmed by observation of the negative effect of sodium content on the photocatalytic activity of titanate nanotubes during of oxidation of dyes.97'98 In contrast, the removal of sodium ions by the protonation of titanate nanotubes results in luminescence quenching,91 indicating that the centres of radiative recombination are associated with sodium sites. This is consistent with observation of the negative effect of sodium ions on the pho-tocatalytic activity of TiNT.

The photocatalytic activity of prepared titanate nanotubes was found to be smaller (though not zero) than that of the standard P25 catalyst in the reaction

Figure 5.8

Photocatalytic processes during the oxidation of organics: a) initial photocatalytic reactions and b) the process of photochemical water splitting on nanotubular TiO2 and titanates.

Figure 5.8

Photocatalytic processes during the oxidation of organics: a) initial photocatalytic reactions and b) the process of photochemical water splitting on nanotubular TiO2 and titanates.

of NH3 oxidation,99 as well as in the reaction of dye oxidation in aqueous suspensions.100 This can be either attributed to impurities of sodium or to the moderate crystallinity of the 'as-prepared' titanate nanotubes. Further improvements in activity are focussed on the transformation of initial nano-tubes to the more photocatalytically active forms of TiO2.

Two methods for H-TiNT transformation have been reported, namely heat and heat/acid treatments. The anatase nanoparticles101 or nanorods102,103 produced by the calcination of H-TiNT at 400 °C, were characterised by an improved initial H-TiNT photocatalytic activity during the reaction of various organic molecules or dye oxidation. The increase in photocatalytic activity was accompanied by an improvement of nanostructure crystallinity. A further increase in calcination temperature resulted in a lower photocatalytic activity, due to stripping of the surface -OH groups and a reduction in the surface area of nanostructures. In contrast, the acid hydrothermal treatment of H-TiNT with residues of HCl at 200 °C (ref. 104), or H-TiNF with HNO3 (0.1 mol dm-3; pH 0-7) at 180 °C (ref. 105), resulted in the formation of nanostructured anatase with a fibrous or particulate morphology. This showed a good photo-catalytic activity for the oxidation of model organic dyes. The method of acid hydrothermal transformation allows a reduction in synthesis temperatures and helps to keep surface -OH groups intact.

An additional improvement in charge separation in elongated titanates can be achieved using the recently discovered synergetic effect that takes place in mixed-phase nanocomposites. The effect occurs between two crystalline forms of TiO2 (anatase and rutile) as a result of the small difference in flat band potentials, which stimulates a spatial separation of carriers thereby reducing their recombination.106 The bi-crystalline mixture of TiO2-(B) nanotubes-anatase nanoparticles (33 : 67%), prepared by calcination of H-TiNT, is characterised by an improved photocatalytic activity when compared with P25 TiO2 for the hydrogen evolution from ethanol.107 This approach has also demonstrated the versatility of this method for the easy preparation of mixed-phase nanocomposites. Titanate nanofibres108 and nanotubes109 decorated with TiO2 (anatase) nanoparticles, deposited by hydrolysis of TiF4 in the presence of H3BO3, are also characterised by good surface adhesion, increased surface area and improved photocatalytic activity.

Although TiO2-based materials can be very active in photocatalytic reactions, their major drawback, delaying their widespread industrial use, is the relatively short wavelength of light necessary to participate in the photo-catalytic reactions. The sensitisation of the photocatalyst to visible light is the ''Holy Grail''. Several approaches to the sensitisation of elongated titanate and TiO2 nanostructures have been recently explored.

One of the successful ways to sensitise TiO2 to visible light is to dope it with nitrogen, forming additional levels in the forbidden zone of wide band gap TiO2. Several methods have been reported for doping titanate nanotubes with nitrogen. The first method involves the ion-exchange of ammonium (NH4+) ions with protons in H-TiNT, followed by calcination of the sample in air at 400 °C to form N-doped TiO2- (B) nanotubes.110,111 The second method employs the alkaline hydrothermal treatment of preliminary doped TiO2 with nitrogen for the formation of N-doped titanate nanotubes.112 The third method involves the nitration of TiNT with gaseous NH3 at temperatures of 400700 °C113 at atmospheric pressure. This last method provides control over the level of N-doping to a nitrogen surface concentration of up to 12%. All N-doped elongated nanostructures show good photocatalytic activity under the visible range of incident light. The nature of photoactive centres in N-doped elongated titanates and TiO2 is being actively studied. Recent XPS and ESR data show that nitrogen is most likely to be present in the form NO, occupying the interstitial positions between oxygen vacancies and Ti4+-forming visible light absorption centres, and providing energy levels positioned within TiO2 band gap. The photocatalytic activity in the visible range correlates with a concentration of these centres which is higher for N-doped anatase nanorods obtained by nitration of H-TiNT, than N-doped P25 prepared at 600 °C.113

Another method to sensitise TiO2 to visible light is implantation with Cr(III) ions,114 where Cr31 ions occupy the positions of Ti41 in the lattice and form electron levels in the forbidden zone. Photocatalytic activity under visible light is observed for catalysts with isolated Cr31 ions inside the TiO2 lattice. The agglomeration of chromium ions at higher chromium loading results in the appearance of recombination centres and a decrease in activity. There are several reports of chromium doping of titanate nanotube photocatalysts. Low levels of chromium doping (0.5 wt%) in titanate nanotubes can be achieved by the alkaline hydrothermal treatment of preliminary Cr-doped anatase.115 The photocatalyst showed some activity under visible light during dye oxidation. The acid-assisted hydrothermal transformation of H-TiNT to anatase, in the presence of HNO3 (0.1 moldm-3) and chromium(iii) at 240 °C over 24 h, results in the formation of Cr-doped anatase nanoparticles.116 The level of chromium in these Cr-doped anatase nanoparticles can be varied up to 10 wt%. However, the most active catalyst for photoelectrochemical water splitting was found to be a catalyst having a loading, of 3 wt%, in which no Cr2O3 phase formation was observed.

Several attempts to decorate titanate nanotubes with the narrow band gap semiconductor nanoparticles CdS 117 119 and ZnS120 by ion-exchange, followed by H2S treatment, have also been reported. Such heterojunction photocatalysts showed a moderate activity for dye oxidation, but the photocorrosion of sulfides is a major problem in such systems. Sensitisation of titanate nanofibres with NiO nanoparticles121 or tin porphoryrin (Sn-TTP) complexes intercalated between layers of titanates122 also resulted in photocatalytic activity of the composite material in the visible range. Femtosecond studies of the latter showed an effective charge separation of visible light photoinduced carriers. The photocatalytic oxidation of methyl orange showed a synergistic enhancement of activity using both UV and visible illumination.122

In conclusion, elongated titanate and TiO2 nanostructures have been considered for photocatalytic processes, including: the oxidation of organic wastes in both air and water;101,102,123 the splitting of water;116,124 and the generation of hydrogen using sacrificial hole scavengers107,125,126 (see Table 5.5). A comparison of elongated morphologies125 for hydrogen evolution from methanol indicated that photocatalytic activity followed the trend: anatase NF > TiO2-(B) NF » H-TiNF. More extensive and systematic studies of all elongated morphologies, including: nanotubes, nanosheets and nanofibres transformed from each other by calcination or hydrothermal treatments, are needed.

Glass surfaces coated with photocatalytically active TiO2 are already being commercially used as self-cleaning surfaces, due to their anti-fogging and super hydrophilic properties under UV light.90 The elongated H-TiNT and TiO2 anatase nanorods127,128 demonstrate an even better surface wettability, as their tubular morphology results in an increased surface roughness, which can be beneficial in achieving a smaller contact angle between the surface coating and water droplets. A similar effect is also observed on anodic TiO2 nanotube array surfaces, attributed to both the porous structure of the nanotubes and their high photocatalytic activity in the oxidation of hydrophobic molecules on the surface.129

Photochemical Water-splitting and Photocatalytic Oxidation on TiO2 Nanotube Arrays

Anodic TiO2 nanotube arrays have recently been thoroughly studied as a promising electrode for the photoelectrolysis of water and are reviewed else-where.130 The principle of photochemical water-splitting is illustrated in Figure 5.8b. Absorption of light in nanotubular TiO2 results in the generation of the main charge carriers: electrons and holes. Photogenerated holes migrate to the surface of the nanotubes and oxidise the water molecules, generating gaseous oxygen, whereas photogenerated electrons are collected to the external circuit and reduce the protons on the platinum counter electrode, generating gaseous hydrogen.

Nanotubular TiO2 electrodes demonstrate an enhanced activity in the photo-electrolysis of water.131 Such an enhancement can be attributed to significant improvements in light absorption at wavelengths near the band edge (375400 nm) by trapping the light using photonic principles and the tubular morphology of material.130 As a result, the near band-edge light can pass through the layer of TiO2 several times, allowing the layer thickness to be decreased whilst maintaining a high level of light absorption. According to Figure 5.9, the photolysis of water is characterised by approximately -0.8 V vs. Ag/AgCl open-circuit potentials, a short circuit current density of > 10 mA cm-2 and a photoconversion efficiency,z, of approximately 12-15%. The latter value can be estimated as fol-

lows:130

where 1.229 V is the standard cell voltage required for the electrolysis of water (the difference between the standard redox potentials of oxygen and hydrogen

Table 5.5 Reported photocatalytic processes using nanostructured titanates and Ti02.

Catalyst Formula

Method of preparation

Particle size/ nm

Loading

Process reaction

Activity or performance

Benefits

Ref

Pt/(Ti02)-(B)

Calcination at

10 nm anatase

1 wt% Pt 33%

CH3CH2OH + hv

20% higher

Facile method to

107

NT-anatase

400 °C

Ti02-(B)

than Pt/P25

bi-crystalline

NP

CH,CHO + H,

catalyst

Cr(m)/Ti02 NP

Hydrothermal acid of Cr-TiNT

5 nm

3 wt%

H,0 + hv h2 + o2

n/a

Novel route for ion implantation

116

Ti02 NP

Calcination H-

n/a

Photocatalytic oxi

Higher than

Novel route for

101

(anatase)

TiNT

dation of organics

P25

anatase nanoparticles

Ti02 NR

Calcination H-

10 x 100 nm

n/a

Photocatalytic oxi

Higher than

Novel route for

102

(anatase)

TiNT

dation of organics

P25

anatase nanorods

Pt/TiNT

Photodeposition

n/a

1 wt%

CH3CH2OH + hv CH3CHO + H2

dehydrogenation

126

CdS/TiNT

Ion-exchange surface reaction

6nm

n/a

Dye oxidation

n/a

Photosensitization of TiNT

117,118

N-doped TiO,

Calcination with

10 nm

1% (N)

2 C3H6 + 9 02 -

3 times higher

Photosensitization

113

NP

NH3 (gas)

6 C02 + 6 H20

than N-P25

of VIS light

Q

NiO/TiNF TiNF

Impregnation

n/a

0.2 wt%

CHC13 oxidation

Higher than P25

Accommodation of Ni in tunnel structure

121

Ti02TiNF

TiO, NF/Ti02 NP

Pt/TiNT

SnTTP/TiNF

TiNF, Ti02-(B) NF, anatase NF Bi2Ti207 NT

Microwave n/a

Epitaxial growth 10-50 nm by precipitation Acid treatment n/a of TiNF at 180 °C sputtering

In situ intercalation Calcination n/a n/a n/a n/a n/a n/a n/a n/a

Ti02 NT array Anodisation n/a n/a

Ti02 NT array Anodisation n/a n/a

TiNT: titanate nanotubes. TiNF: titanate nanofibres. NR: nanorods. NP:

NH4+ + O,

less than P25

Good adsorption

99

NO, NO;

of NH3

Dye oxidation

Higher than

Support for

108

TiNF

photocatalyst

Dye oxidation

Comparable

Novel route for

105

with P25

nanostrucuted

anatase

H,0 + hv

Higher than

Early data showing

124

h2 + o2

fio2

water splitting

Dye oxidation

n/a

Synergy in using

122

UV/YIS light"

CH,OH + hv

Higher than

Systematic com

125

HCHO + H2

P25

parison of elon

gated structures

Dye oxidation

Hiaher than

Effect of dimension

123

bulk

on the activity of

Bi,Ti,07

catalyst

H,0 + hv

12-16 %

Perspective elec

131

h2 + o2

trode with high

performance

Dye oxidation

Higher than

Novel

136

P25

photocatalyst

-1.5 -1 -0.5 0 0.5 1 1.5 b) Potential, Evs. Ag/AgCI / V

Figure 5.9 Graphs showing: a) photocurrent density and b) the corresponding photoconversion efficiency of a TiO2 nanotube array electrode with nanotubes of 205 nm outer diameter and 30 mm length. Electrodes were annealed at the indicated temperatures for 1 h in oxygen. Photolysis of water occurred in KOH (1moldm-3) electrolyte under UV illumination (98mWcm-2, 320-400 nm range). (Images are reproduced with kind permission from ref. 131).

evolution under standard conditions), Vbias is the applied potential, Ip is the current density responsible for generation of hydrogen and oxygen, and Pt is the power density of illumination (W m~2).

Figure 5.9 shows that a higher calcination temperature for anodised TiO2 nanotubes results in an improvement in the photoconversion efficiency and an increase in photocurrent. This is probably due to an improvement in the crystallinity of the nanotube walls leading to a reduction of the amorphous regions and the borders of grain boundaries, which reduces the number of charge-carrier recombination centres. Further increases in calcination temperature, however, result in a deterioration of electrode performance, due to the growth of a barrier layer of oxide which reduces electrical conductivity and a peeling of the nanotubes from the substrate.

Despite relatively good conversion efficiency, TiO2 nanotube arrays can utilise light of only short wavelength. Another problem with TiO2 nanotubes is their low catalytic activity in the reaction of oxygen evolution. Both of these problems can be tackled by employing composite materials based on TiO2 nanotube arrays.

Examples of such composites include Cu-Ti-O (ref. 132) or Fe-Ti-O (ref. 133) ternary oxide nanotube arrays, which can be fabricated by anodic oxidation in a fluoride-containing electrolyte of Cu-Ti and Fe-Ti films, deposited previously by simultaneous co-sputtering. The internal pores of nanotubes can be filled or conformally coated with some narrow band-gap semiconductors leading to heterojuction-type nanotubular composites, including: CdS/TiO2 NT (ref, 134) and CdTe/TiO2 NT (ref. 135). All of these composites demonstrate a successful sensitisation of composite electrodes to the visible wavelength of incident light, allowing better utilisation of solar radiation.

Anodised TiO2 nanotube arrays also demonstrate an activity in the photo-catalytic oxidation of organic compounds which exceeds the performance of the standard P25 Degussa photocatalyst.136 This activity can even be further accelerated by decorating the surface of nanotubes with suitable metallic nanoparticles.

Was this article helpful?

0 0

Post a comment