Sonocatalytic degradation of caffeine using CeO2 catalyst: parametric and reusability studies

This work examined sonocatalytic degradation of caffeine in the presence of CeO2 prepared by hydrothermal method. Several characterization techniques were used to study the CeO2 including TEM, XRD and BET method. Effects of various parameters such as pH, initial concentration of caffeine and dosage of CeO2 were investigated. This study also examined the reusability of CeO2. Results showed that the CeO2 had mixed shapes of crystallites consisting of rods and cubes with the specific surface area (Sbet) and pore volume of 126.63 m2/g and 0.4898 cm3/g, respectively. About 95.5% of caffeine degradation was achieved under the best parameter conditions i.e. at pH of 7.5, 5.0 mg/L of initial concentration of caffeine and 1.0 g/L of CeO2 dosage within 150 min. It was also revealed that the operating parameters played major roles in caffeine degradation efficiency. In addition, the CeO2 were quite stable since the efficiency of caffeine degradation achieved in the third cycle was 81.4%.


Introduction
Caffeine is commonly employed as a stimulant in pharmaceutical and personal care products (PPCPs) [1] and as an ingredient in beverages such as coffee, tea, caffeinated soft drinks, chocolate and numerous food products [2]. Caffeine as one of the most frequently detected of PPCPs, has gained environmental concern due to its potential adverse effect on the aquatic organisms and environment [3]. Caffeine is recognized as a micropollutant and its presence in surface water, groundwater (from the wastewater treatment plants) [4] and drinking water [5] which can be detected at trace concentration level in the matter of a few µgL -1 or ngL -1 [6]. However, at high concentration, caffeine is known to be toxic [7].
Advanced oxidation processes (AOPs) have been recognized as an attractive technology to remove persistent organic pollutant from contaminated water. It involves the generation of reactive species mainly hydroxyl radicals (•OH) which could oxidize a broad range of organic pollutants that present in water. Sonocatalysis as one of AOPs potential process could treat caffeine or other organic micropollutants in wastewater [89]. This process utilizes ultrasonic irradiation in the presence of suitable semiconductor catalyst. The semiconductor materials that have been reported in previous studies are such as TiO2 [910], ZnO [11], MnO2 [12], FeCeOx [13] and many more. Among other catalysts, cerium oxide (CeO2) or ceria has received much attention recently due to its high temperature stability and non-toxic in catalytic applications [14]. Besides, CeO2 or cerium-based materials have gained increasing attention in photocatalysis application for the past decade as alternative of conventional photocatalyst, TiO2 [14]. Therefore, this material is hardly studied for its sonocatalytic properties and only a few studies are available in the literature regarding the sonocatalytic activity of CeO2 or ceria based [15][16].
In the present work, effects of various parameters such as pH, initial concentration of caffeine and catalyst dosage on the sonocatalytic degradation of caffeine under ultrasonic irradiation were studied to evaluate the performance of the catalyst. In addition, the reusability potential of catalyst was also demonstrated.

Preparation of catalyst
Cerium oxide was prepared using a hydrothermal method according to the published procedure [17]. An amount of Ce(NO3)3·6H2O was dissolved in 5 mL of deionized water. Then, 75 mL of NaOH solution (6.4 M) was added drop wise into the ceria precursor solution under vigorous stirring for 30 min. The above mixture was then transferred into a Teflon-lined stainless-steel autoclave and heated up to 100C for 24 h. The resultant precipitant was then washed with two-litter deionized water and separated using centrifuge for three times. Then, the precipitate was dried in an oven at 60C overnight.

Characterization of catalyst
X-ray powder diffraction (XRD) pattern of CeO2 was recorded on an X-ray diffractometer (Bruker D8 Advance, Germany) using monochromatic Cu Kα as the radiation source (λ = 1.5406 Ǻ), with an operating voltage of 40 kV and emission current of 40 mA. The diffractogram was measured from 10 to 70 (2) at step size of 0.020. Then, the diffraction peaks of crystalline phases were compared with the standard compound in JCPDS data file. The morphology of CeO2 was examined by transmission electron microscope (TEM) (ZEISS LIBRA 120 microscope) with an accelerating voltage of 100 kV. The textural properties of CeO2 was estimated via adsorption-desorption of N2 at -195.87°C using a Micromeritics instrument model ASAP 2020. The sample (0.1444 g) was degassed under vacuum at 250C for four hours. The specific surface area was calculated using the standard BrunauerEmmetTeller (BET) method while the pore volume was calculated by Barret-Joyner-Halenda (BJH) method for mesopores analysis.

Experimental set-up and sonocatalytic degradation of caffeine
The sonolytic and sonocatalytic degradation reaction of caffeine solution were carried out in a 250 mL Erlenmeyer flask as a reaction vessel under ultrasonic irradiation. The ultrasonic bath (Elmasonic S 60 H, Germany) was used for this purpose with a frequency and an output power of 37 kHz and 150 W, respectively. The sonocatalytic system was also equipped with two pumps, thermometer and holding tank. Figure 2 shows the schematic diagram of this experimental set-up system. Throughout the process, the temperature of water bath was kept constant at 25±3°C by adding ice cubes in the holding tank in order to avoid the temperature rise in the ultrasonic bath.
For a typical catalytic process, a certain quantity of CeO2 was added to a 100 mL of caffeine aqueous solution with a certain concentration. Prior to the catalyst addition, the pH of the aqueous solution was adjusted using NaOH (0.1 M) or HCl (0.1 M). Then, the suspension was agitated for 30 min in the dark condition to achieve the adsorption-desorption equilibrium. Subsequently, the suspension was added with 20 mM of H2O2 as the oxidant and was sonicated for 150 min in the ultrasonic bath. During the process, 4 mL of sample was withdrawn and filtered through 0.22 µm syringe filter (PVDF) at 30 min of interval time. The absorbance of filtrate was then determined at the maximum wavelength (λmax) of 273.5 nm using Lambda 25 UV-VIS Spectrophotometer (Perkin Elmer, USA). Then, the concentration of sample was determined based on the standard calibration curve. Meanwhile, the removal performance for caffeine aqueous solution was calculated as follows: (Ci -C)/ Ci, where Ci is the initial concentration of caffeine aqueous solution and C is the concentration of caffeine aqueous solution after sonication.

Reusability of CeO2
The reusability of CeO2 was examined for up to three catalytic runs at obtained optimal parameter with similar procedure as stated in Subsection 2.4. After each run, the catalysts were washed, filtered using centrifuged (6000 rpm, 30 min) and dried in an oven at 60°C for six hours. Then, the catalyst was reused with a fresh caffeine aqueous solution for the next run.

Characterization of CeO2
A typical XRD pattern of CeO2 materials for CeO2 sample is shown in figure 3(a). This catalyst showed intense and sharp diffraction peaks of fluorite cubic structure with Fm3m space group (JCPDF no. 00-034-0394) [18].    18.7 nm and 12.1  3.5 nm, respectively. The findings of this study do not support most of the previous works which only single shape of CeO2 was supposedly to be formed; rods. However, Piumetti et al. [19] clarified that at a low aging temperature (<150°C) of hydrothermal method, it would lead to the formation of mixed shapes of catalyst. Besides, at high base concentration (6.4 M of NaOH), the Ce(OH)3 nuclei could become metastable and were oxidized to CeO2 cube shape which was more stable [17,20].

Figure 3(a). XRD pattern and (b) TEM image of CeO2.
The nitrogen adsorption-desorption isotherm curves were also obtained as shown in figure 4(a) for CeO2. It depicts a typical Type IV physisorption isotherm based on IUPAC that indicates typical mesoporous materials (pore width 2 -50 nm) [21]. The CeO2 demonstrates a narrow hysteresis loop reflecting the presence of more porous structure [22]. From the hysteresis loop also, the shape of CeO2 can be correlated with shape of cylindrical pores (Type A) [2324]. Meanwhile, the specific surface area (SBET) and pore volume of the catalyst are 126.63 m 2 /g and 0.4898 cm 3 /g, respectively. The result of SBET clearly shows that the mixed structural shapes of CeO2 higher compared to single shape of CeO2 rod and cube [17]. Therefore, it indicates that the mixed shapes of CeO2 increased the availability of more active sites on the catalyst surface for adsorption of caffeine molecules.    Figure 4(b) shows sonocatalytic degradation of caffeine using CeO2 at three different pH values. The sonocatalytic activity of caffeine was enhanced at pH 7.5 in which caffeine was degraded by about 91.5% and it decreased significantly at low pH (pH 3.5) [25]. The effect of pH on degradation of caffeine depend on the point zero charge (pZC) of CeO2 and the ionization of caffeine molecules in aqueous solution. Since the pZC of CeO2 is 6.95 [26], the catalyst surface is positively charged (adsorption of anionic molecules is favorable) when the caffeine solution was at pH < 6.95 and is negatively charged (adsorption of cationic molecules is favorable) when the pH > 6.95. In addition, the molecules of caffeine are fully protonated at pH<10.4 (pKacaffeine) [5]. Therefore, the best sonocatalytic degradation of caffeine could be observed at pH 7.5 due to the adsorption enhancement of caffeine molecules on the CeO2 that was favored at pH between 6.95 and 10.4. In addition, this result also clearly demonstrated that the effect of pH was significant in near-neutral solution in degrading the caffeine.

Effect of initial concentration of caffeine.
Effect of initial concentration of caffeine (in the range of 5 -30 mg/L) on the sonocatalytic degradation using CeO2 was investigated and the result is shown in figure 5(a). Three initial caffeine concentrations of 5, 10 and 15 mg/L experienced the significant degradation within 150 min especially for 5 mg/L (91.5%). However, the caffeine degradation decreased drastically to 19.8% when the initial concentration was 30 mg/L. The sonocatalyic activity was observed ineffective when the concentration of caffeine is higher than 15 mg/L [8]. This can be explained by the fact that at high initial caffeine concentration, more caffeine molecules were adsorbed on the surface of CeO2 to result in fewer active sites for adsorption of hydroxyl ions. Thus, the generation of OH radicals (reactive species) would be decreased leading to low degradation efficiency of caffeine [27]. From this finding, it was demonstrated that the sonocatalytic degradation of caffeine was influenced by its initial concentration. Overall, the highest activity of sonocatalytic degradation of caffeine was found at initial concentration of 5 mg/L which was in agreement with several previous studies which they applied low range of initial caffeine concentration (2.5 -30 mg/L) in their studies [5,8,27].

Effect of CeO2 dosage.
The influence of CeO2 dosage on degradation of caffeine was examined by varying the catalyst amount from 0.5 to 2.0 g/L ( figure 5(b)). From this figure, the optimal dosage for degradation of caffeine was 1.0 g/L (95.5%) which was relatively higher compared with other dosages. Better dispersion of CeO2 was achieved due to the turbulence generated by ultrasonic induced cavitation at this dosage [28]. This result also suggested that the other CeO2 dosages (lesser or more than 1.0 g/L) had comparatively lower degradation efficiency with not more than 50% of degradation. This can be explained when the dosage was insufficient, lesser nuclei would be generated in the cavitation effect which lead to the fewer OH radicals generated by the CeO2. On the other hand, when the CeO2 was in excess amount, it would lead to the shielding effect among the particles and interference to the ultrasonic irradiation. Thus, the generation of OH radicals would be affected [8] leading poorer degradation efficiency of caffeine. Figure 6(a) shows the caffeine degradation as a function of irradiation time. The degradation of caffeine was insignificant for the absorption process which almost no degradation achieved within 150 min. Meanwhile, when the ultrasonic radiation was applied, the degradation increased to 13.1% and increased further to 24.1% when H2O2 was added as the oxidant [8]. This result demonstrated that there was a slight phenomenon of sonolysis occurred when this irradiation was used in both systems. As expected, the caffeine degradations further increased when CeO2 was applied simultaneously in conjunction with ultrasonic radiation either with (95.5%) or without oxidant (86.3%). This could be related to the hydroxyl free radicals (OH) production as the reactive species [16]. As consequent, sonolysis alone with or without the oxidant was not sufficient to degrade caffeine.

Reusability of CeO2
After three cycles, a negligible decrease in degradation of caffeine was observed ( figure 6(b)), highlighting that the CeO2 exhibited high sonocatalytic activity even in the third cycle (81.4%). The reused CeO2 showed a similar efficiency to the fresh catalyst, indicating the stability and reusability potential of CeO2. This result might indicate the future potential application of CeO2 in industrial process.

Conclusions
In summary, CeO2 was successfully synthesized using the hydrothermal method. The results indicated that the sonocatalytic degradation of caffeine was obviously affected by pH, initial concentration of caffeine and dosage of CeO2. In general, the degradation efficiency was found to increase with a decrease in initial concentration of caffeine and dosage of CeO2 at near-neutral solution. The highest degradation of caffeine (95.5%) was achieved in 150 min at a pH of 7.5, 5.0 mg/L of initial concentration of caffeine and 1.0 g/L of CeO2 dosage. The sonocatalytic process using CeO2 showed great potential in treatment of caffeine. In addition, CeO2 could be applicable to the practical application since it showed high stability and reusability.