Introduction

Photocatalysis is an intensively investigated alternative method for the treatment of contaminated water and air, and is effective for the removal of the majority of organic pollutants [1], in contrast with the most commonly applied microbial water treatment procedure, which is unable to eliminate some industrial compounds [1, 2] (e.g. certain pharmaceuticals [3–6], dyes [2, 7–15], or toxic compounds such as pesticides [8, 16, 17] or herbicides [18] ).

Photocatalysts applied in suspensions must be separated after their use for water treatment [1, 10, 19–21]. However, in consequence, the very small grain size (5–500 nm), the filtration step demands considerable energy, which hinders their economic applicability [9, 18, 20, 22–25]. The very important challenge has therefore arisen to immobilize the photocatalyst particles in the form of photocatalytically active surfaces which can be applied in fixed-bed flow reactors [22, 25]. These methods can result in economic applications, with the utilization of solar light for the activation of the photocatalyst, without any separation costs.

The simplest method of immobilization is dip coating. The support (mainly glass) is dipped in an aqueous (or alcoholic) suspension of the photocatalyst and this is followed by a drying step. After several repetitions of the dipping/drying procedure, the final cleaning step is usually preceded by calcination [4, 11, 19, 26–29]. As described by Behnajady et al. [11], in the course of the calcination the OH groups on the catalyst surface react with the support to form oxygen bridges through water loss, and this ensures the fixing [12]. However the calcination step is difficult to carry out on large surfaces, demands the investment of considerable energy and can change the properties of the precrystallized photocatalysts.

Another possibility is to embed the photocatalytically active nanoparticles into polymers [25]. Noorjahan [30] applied a UV light-resistant acrylic polymer to make TiO2-based thin films. Polyethylene films [31] or sheets [13], polyvinyl acetate [32], polyvinyl chloride [33] and polystyrene beads [14] have also been investigated. However, organic polymers can be damaged by activated photocatalysts during utilization, which may result in mobilization of the nanoparticles.

Electrophoretic deposition is a further means of immobilization, which allows the production of thin films with controlled thickness [34]. Dunlop et al. [34] immobilized Degussa P25 TiO2 on conducting supports with the utilization of voltage, but also applied a calcination step.

Sol–gel processes, the most widely investigated methods of in situ photocatalyst generation [15, 35, 36], apply Ti compounds (mainly alkoxides or TiCl4) as precursors: an amorphous titanium hydroxide layer is produced on a support by hydrolysis and finally the photocatalytically active TiO2 surface is formed via crystallization in a calcination step. Since the photocatalytically active particles are formed during the fixing procedure, the properties of the resulting TiO2 are limited.

Special, complicated methods of synthesis to obtain highly efficient (or visible light- active) crystalline TiO2 with desired properties have been discussed in thousands of papers. The main goal of the present study was to develop a simple, stable immobilization method for the fixing of highly efficient precrystallized (or even doped) TiO2 in a UV-resistant layer which is not sensitive to the degrading action of activated TiO2 nanoparticles.

Experimental

Materials

The applied ceramic paper, produced by COTRONIC Corporation, is a highly pure Al2O3-based, non-woven, 1.6-mm-thick ceramic paper (Catalog No.: 300-040-1).

For the immobilization procedure, Ti(OEt)4 (ABCR Gmbh. & Co. Kg), which contains 10–15 % Ti(OiPr)4 and ethanol (Spektrum 3D, absolute ethanol) was applied.

The immobilized commercial photocatalysts were Aeroxide P25 (pure TiO2: ~90 m/m% anatase, ~10 m/m% rutile, Danatase ~25 nm, Drutile ~40 nm, aS BET~49 m2 g−1) produced by Evonik Industries, and VLP7000 (doped TiO2: ~100 m/m% anatase, Danatase ~7.8 nm, aS BET~297 m2 g−1) produced by Kronos Titan Gmbh. More details about the investigated photocatalysts are published elsewhere [37].

High-purity nitrogen gas (Messer, >99.995 %) was used to spray the materials onto the surfaces.

Water treatment experiments were carried out by applying phenol (Spektrum 3D; analytical grade), oxalic acid (GyKV, Hungary; analytical grade) and a pesticide (monuron; Sigma-Aldrich; analytical grade) as model contaminants, which were dissolved in MilliQ water. The pH values were ~ 6 in case of phenol and monuron solutions, while in the case of oxalic acid, the initial pH value was ~ 2.8, which increased during the decomposition.

Methods and instrumentation

Scanning electron microscopy (SEM) was used on a Hitachi S-4700 Type II FE-SEM instrument equipped with a cold field emission gun operating in the range 5–15 kV.

Changes in phenol, monuron and oxalic acid concentrations were followed with an Agilent 1100 series HPLC system. For phenol and monuron determination, it was equipped with a Lichrospher RP 18 column and a methanol/water mixture was used as eluent (detection at 210 nm). For oxalic acid measurements, a GROM-RESIN ZH column was used, with 19.3 mM aqueous sulfuric acid as eluent (λD = 206 nm).

The produced TiO2-coated ceramic sheets were used in a home-made, continuous-flow recirculating reactor, illustrated in Fig. 1. The phenol solution (500 mL, 0.1 mM) was placed into a thermostated (25 °C) reservoir, while the liquid flow was maintained by a peristaltic pump (200 mL min−1) on the upper part of a tray in which a photocatalytically active impregnated ceramic sheet (10 cm × 30 cm) was placed. To activate the TiO2, UV fluorescent lamps (LIGHTECH; UV-A; 4 × 40 W), or in other cases, halogen floodlight reflectors (JEN CE-82, 2 × 500 W) were placed above the flow reactor. Fig. 2 depicts the spectra of the applied modes of irradiation.

Fig. 1
figure 1

Photograph of laboratory-scale flow reactor applying immobilized TiO2 photocatalyst, equipped with visible light-emitting halogen floodlight reflectors (a) and UV-A fluorescent tubes (b)

Full size image
Fig. 2
figure 2

Emission spectra of the two forms of artificial irradiation applied

Full size image

In the case of solar photocatalytic experiments, the UV dose was measured by a hand-held UV radiometer (Optix Tech, Inc; UVTEX a + b idm).

Results and discussion

Immobilization of TiO2 on ceramic paper

A sheet of ceramic paper was perfused with ethanol to wet the surface, and was then impregnated with Ti(OEt)4 by dragging it between two plastic rollers, the lower of which was partially immersed in pure Ti(OEt)4 as shown in Fig. 3. After this step, precrystallized photocatalysts were immediately sprayed with a handheld airbrush onto the surface in suspension (ethanol) form. The concentration of the suspension was 20 g L−1, and the sprayed amount was 0.1 mL cm−2. The impregnated ceramic sheets were dried in air at room temperature for 24 h. During this step, the Ti(OEt)4 was hydrolyzed by the humidity in air, and the forming amorphous Ti(IV) oxide-hydroxide fixed the photocatalytically active particles onto the surface (Fig. 4). After the drying step, the impregnated ceramic sheets were washed with distilled water to eliminate non-immobilized nanoparticles, and were finally cleaned from possible organic surface contaminants by irradiation with UV-A light for 24 h. This immobilization method results in a titania layer which contains UV and VIS-resistant inorganic fixing material. The layer components therefore remain inert during the photocatalytic processes.

Fig. 3
figure 3

Schematic figure of the TiO2 immobilization method

Full size image
Fig. 4
figure 4

SEM images of the semiconductor nanoparticles immobilized on the fibers of the ceramic sheet

Full size image

Reproducibility of the decomposition on impregnated ceramic paper

Three different ceramic papers were produced by the described impregnation method to investigate the reproducibility. Photocatalytic activities were estimated for these three ceramic papers through phenol degradation, followed during 80 min under UV irradiation. The immobilized Aeroxide P25 photocatalyst coatings displayed very similar catalytic activities (Fig. 5). The measured differences in the amounts of degraded phenol for the three different immobilized TiO2-coated ceramic papers were <1 %, i.e. the reproducibility of the preparation process was very good.

Fig. 5
figure 5

Reproducibility: Phenol (V = 500 mL; cphenol = 0.1 mM) degradation on 3 different ceramic sheets (size is 10 × 30 cm) coated by Aeroxide P25 (applying UV irradiation)

Full size image

Persistence of the catalyst on impregnated ceramic paper

To demonstrate the persistence of the immobilized layer, the photocatalytic decomposition of phenol during a 2-h period was repeated five times on the same immobilized TiO2 (Aeroxide P25)-coated ceramic paper. Between the experiments, the sheet was irradiated for 24 h with UV light for the complete decomposition of the adsorbed phenol and oxidation by-products. For the five cycles, the calculated initial degradation rates were identical within the experimental error (Fig. 6). These results confirmed the stability of the coating, which was also observed visually. No detached particles were found in the phenol solutions during these experiments.

Fig. 6
figure 6

Stability of the photocatalytic efficiency of the immobilized TiO2-coated ceramic sheet: phenol (V = 500 mL; cphenol = 0.1 mM) degradation on the same TiO2 coated ceramic sheet (size is 10 × 30 cm) for five 2-h cycles (applying UV irradiation)

Full size image

Phenol decomposition by-products

An 8-h photocatalytic experiment was carried out under UV irradiation, during which phenol was totally decomposed (its final concentration was below the detection limit). During the photodegradation of phenol, three main oxidation by-products were formed: hydroquinone, pyrocatechol and resorcinol (dihydroxybenzenes). These compounds were also degraded by this treatment during the 8-h irradiation (Fig. 7).

Fig. 7
figure 7

Phenol degradation by-products during the photocatalytic decomposition of phenol (V = 500 mL; cphenol = 0.1 mM) by immobilized Aeroxide P25 (applying UV irradiation)

Full size image

Experiments with different photocatalysts under visible light irradiation

Phenol decomposition experiments were carried out on immobilized Aeroxide P25 and Kronos VLP7000 TiO2 under visible light irradiation. Because of the slow degradation, the phenol concentrations were followed for 5 h (Fig. 8). The decay curves indicated a much higher activity in the case of the Kronos VLP7000-coated ceramic paper. This result confirms the possibility of the creation of photocatalytically active surfaces with higher performance through the demonstrated immobilization technique, with the application of doped, precrystallized TiO2.

Fig. 8
figure 8

Photocatalytic degradation of phenol (V = 500 mL; cphenol = 0.1 mM) under visible light irradiation by two different forms of commercial TiO2 (Aeroxide P25 and Kronos VLP7000)

Full size image

Waste decomposition by solar irradiation on a semi-pilot scale

A large ceramic sheet (28 cm × 120 cm), coated with Aeroxide P25 by the described immobilization method was applied in a semi-pilot-scale flow photoreactor. Fig. 9 demonstrates that this apparatus has a very similar structure to that of the laboratory-scale flow reactor. In this system 10–50 L of model waste water could be purified. This reactor was applied to confirm that the scaling-up of the developed immobilization technique can easily be accomplished.

Fig. 9
figure 9

Photograph of pilot-scale flow reactor with (a) and without (b) artificial irradiation (semiconductor particles were activated by solar radiation)

Full size image

Outdoor solar experiments were carried out in the above-described flow reactor to demonstrate that organic contaminants can be decomposed by solar light with immobilized TiO2. The photocatalytic decontamination of 10 L of model waste water containing phenol, oxalic acid or a pesticide (monuron) was investigated on sunny days in the summer (in Szeged, Hungary). The presented results (Fig. 10) revealed that the model compounds (0.1–1 mM) were decomposed to an extent of 60–95 % after solar irradiation for 4 h.

Fig. 10
figure 10

Solar photocatalytic experiments in a pilot-scale (V = 10 L; cphenol = 0.1 mM, cmonuron = 0.1 mM, coxalic acid = 1.0 mM) photoreactor (in blank experiments, the flow reactor was equipped with non-coated ceramic papers during solar irradiation)

Full size image

Conclusions

In this study, a simple immobilization technique was developed to fix highly efficient precrystallized (or even doped) semiconductor nanoparticles on surfaces via an amorphous Ti(IV) oxide-hydroxide layer, which is not sensitive to the reactive radicals produced by activated TiO2 nanoparticles. The persistence was confirmed in photocatalytic experiments in which the rate of phenol decomposition remained constant throughout five 2-h cycles on the same TiO2-coated ceramic paper. The potential for application on an industrial scale was demonstrated by a pilot-plant-scale flow reactor (containing a 28 cm × 120 cm impregnated ceramic sheet), in which environmental contaminants were decomposed by solar irradiation. This immobilization method is a simple technique that can be used to investigate the long-term efficiency of novel TiO2 samples, or can even be applied in real air/water treatment processes. It seems that this method is also suitable to fix a photocatalyst on other surfaces (e.g. on the walls).