- Email: [email protected]
Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation
Journal of Cleaner Production xxx (2015) 1e8
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation Chin Kui Cheng a, b, c, *, Mohd Rizauddin Deraman c, Kim Hoong Ng c, Maksudur R. Khan c a
Rare Earth Research Centre, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia Centre of Excellence for Advanced Research in Fluid Flow, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia c Faculty of Chemical & Natural Resources Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 November 2014 Received in revised form 4 June 2015 Accepted 24 June 2015 Available online xxx
This paper reports on the photocatalytic degradation of pre-treated palm oil mill effluent (POME) over titania loaded with photocatalyst. The argentum loading on the titania was varied from 0.25 to 1.0 wt% via wet-impregnation technique. X-ray diffraction characterization of all the photocatalysts showed that the photo-active rutile phase was still intact after the photocatalyst synthesis. In addition, the UVeVis diffuse reflectance measurements indicate an improved visible light energy absorption and that the band gap energy was significantly reduced (averaging 2.50 eV) when titania was loaded with argentum, compared to the pristine titania that recorded a reading of 3.20 eV. The 0.50 wt% argentum/titania photocatalyst offered the most effective degradation of pre-treated POME under the irradiation of 100 W of UV light (25.0%) and also visible light (16.0%), respectively, over a loading of 0.2 g/L. Significantly, the maximum photocatalyst loading determined from the current work was 1.0 g/L. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Argentum Photocatalysis Silver POME Titania Wastewater
1. Introduction Palm oil mill effluent (POME) is a thick, brownish slurry waste produced from a burgeoning oil palm industry. Malaysia is currently the world's largest exporter of palm oil although it is the second-largest producer of the oil after the neighbouring country Indonesia. It has planted area of 4.9 million hectares with a production of 17 million tonnes of palm oil. The palm oil industry remains a significant contributor to the Malaysian economy with a reported export earnings of about RM60 billion annually (Sri and Dompok, 2011). In terms of composition, POME is a complex liquor comprised of amino acids, inorganic nutrients such as sodium, potassium, calcium, magnesium, short fibres, organelles, nitrogenous constituents, free organic acids and a mixture of carbohydrates ranging from hemicelluloses to simple sugars, etc. (Santosa, 2008). It has been estimated that approximately 1.5 m3 of water are required to process a ton of fresh fruit bunch whereby half of the amount becomes POME (Kongnoo et al., 2012). In
* Corresponding author. Faculty of Chemical & Natural Resources Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia. Tel.: þ60 9 549 2896; fax: þ60 9 549 2889. E-mail address: [email protected] (C.K. Cheng).
general, the chemical oxygen demand (COD) and biochemical oxygen demand (BOD) in the POME are in the range of 15,000e100,000 mg/L and 10,000e43,750 mg/L, respectively. Due to its acidic nature (pH 3.4e5.2), high values of COD and also biochemical oxygen demand (BOD), POME can inflict considerable environmental problems if discharged without effective treatment (Borja et al., 1996). Over the past decades, several economicallyviable technological solutions have been proposed for the treatment of POME, viz. simple skimming devices (Roge and Velayuthan, 1981; Ng et al., 1988), land disposal (Ma and Ong, 1986), chemical coagulation and flotation (Badri, 1984; Othman et al., 2014; Chin et al., 1987; Karim and Hie, 1987), aerobic (Abdul et al., 1989) and anaerobic biological processes (Harsono et al., 2014). Among traditional technologies, anaerobic treatment followed by aerobic biological processes are most widely used because of their particular advantages, such as low biomass yield, low nutrient requirement, and high volumetric organic loading (Spanjers and Lier, 2006). Unfortunately, some of these methods are land-intensive while others require long hydraulic retention time (Tabassum et al., 2015). Alternatively, the destruction of organic pollutants could be carried under light source employing semiconductor-based materials as the heterogeneous catalysts. Specifically, two types of light spectrum have been explored viz. UV spectrum
http://dx.doi.org/10.1016/j.jclepro.2015.06.104 0959-6526/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
2
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
(200e400 nm) and visible light spectrum (400e700 nm). Due to the different wavelengths, material selection is critical in ensuring that light energy can be absorbed by the photocatalyst. Titania (TiO2) photocatalyst has been widely used to degrade organic compound, referenced herein (Wang et al., 2013; Virkutyte et al., 2012; Kasanen et al., 2011; Giri et al., 2008). Nevertheless, owing to the wide band gap (~3.3 eV), TiO2 can only effectively absorb UV light, which however, only represents about 5% of our solar energy. In contrast, the visible light spectrum accounts for circa 46%, hence represents a more viable source. Consequently, two different strategies have been undertaken viz. to employ new materials to substitute the TiO2 (Shahid et al., 2013; Chen et al., 2014, 2013; Shen et al., 2013; Mahmoodi, 2011) or to incorporate dopants into the TiO2 matrix to extend the light spectrum into the visible region (Zalas, 2014; Ramchiary and Samdarshi, 2014; Quesada-Cabrera et al., 2014; Chakrabortty and Gupta, 2013; Ao et al., 2010). Specifically, this work focuses on latter approach by doping TiO2 with the silver (Ag) for the photocatalytic degradation of POME. The current work will serve as an important benchmark in studying the use of photocatalytic technique to treat the POME. 2. Methodology 2.1. Catalyst preparation and characterization A 0.1 M aqueous solution of AgNO3 was mixed with TiO2 powder (Sigma Aldrich) of different weight percentage, viz. 0.25, 0.5 and 1.0 wt% of Ag/TiO2 in a 500 mL beaker. The mixture was then stirred using an ultrasonic water bath for 8 h at 328 K. Subsequently, it was dried in an oven at 378 K for 8 h. The resulting powder was milled and washed repeatedly by distilled water to remove the impurities. The powder was re-dried for 8 h before re-milled. It was finally calcined at 723 K for catalyst characterization and photocatalytic reaction studies. BET multipoint analysis was carried out using Thermo-Scientific Surfer to determine the BET specific surface area of the fresh catalysts. N2 with a cross-sectional area of 16.2 Å2 was used as the adsorbate in the analysis with analysis temperature at 77 K. The crystalline structure was obtained via XRD diffraction analysis. The samples were irradiated by Ni-filtered CuKa with a wavelength (l) of 1.542 Å at 40 mA and 45 kV, and scanning from 10 to 80 at 4 min1 in a Rigaku Miniflex II instrument. In addition, band gap energy of all the synthesized catalysts was spectroscopicallydetermined from the UVevis diffuse reflectance measurements provided by a Jasco V-550 spectrophotometer equipped with an integrating sphere. The instrument measured the absorption spectrum of a sample at wavelengths in the range of 190e900 nm. The light source used was a deuterium lamp and a halogen (WI) lamp. In addition, FTIR spectroscopy was also applied primarily to study the chemical structure. The infrared spectra of the catalyst samples were measured at wavelengths ranging from 1000 to 4000 cm1.
2.3. Photocatalytic degradation of pre-treated POME For the studies of photocatalytic degradation of POME, the fresh POME was filtered to eliminate the solid suspension. Subsequently, it was further diluted with deionized water (Millipore). The filtration and dilution steps were necessary as fresh POME in its existing form was impenetrable to the light source. In order to prevent the wastewater from undergoing microbial-biodegradation that would change the content of the POME, the diluted POME was preserved at below 4 C in a chiller (Thermo) (Bello et al., 2013). By using a similar source of pre-treated POME, it would prevent the variation in the POME composition. For the photocatalytic degradation studies, a 1-L quartz photoreactor with separable quartz compartment housing the light source was procured from Shanghai Sunny Scientific Collaboration Co. Ltd. Fig. 1 shows the setup of photoreaction whereby the temperature during the reaction was consistently-maintained at 298 K. For the reaction studies, 300 mL of the pre-treated POME was mixed with a pre-determined catalyst amount in the quartz reactor and rigorously-stirred. A 100 W UV lamp and a 100 W Xenon lamp (Shanghai Sunny Scientific Collaboration Co. Ltd.) were employed as the light sources for the current work. The technical specification provided by the manufacturer has indicated that the visible light system was free from UV interference. For the sampling purpose, around 5.0 mL of sample was withdrawn from the reactor for chemical oxygen demand (COD) measurement. For the current study, the model of COD reactor used was Hach RB-200. 3. Results and discussion 3.1. Characterization of photocatalyst Previous works have indicated that TiO2 is mainly comprised of anatase or rutile phase with the former being more active towards UV-absorption whilst the latter is inactive (Chuong et al., 2008). Moreover, the crystalline phase could transform with calcination temperature. Hence, in the current work, the pristine TiO2 and calcined TiO2 samples were subjected to XRD scanning to determine a suitable calcination temperature in order to prevent rutile phase formation. Fig. 2 shows the resulting XRD diffractogram.
2.2. Wastewater collection and characterization Raw POME was collected from Felda palm oil industries located in Kuantan, Malaysia. The temperature of raw POME at the discharge point was around 80e90 C. The sample was transported to the laboratory in sterile 1000 mL Schott bottles placed in ice and stored at 4 C until use. The wastewater characteristics such as chemical oxygen demand (COD), biochemical oxygen demand (BOD), total solids, total suspended solids, ammoniacal nitrogen, nitrate nitrogen and total dissolved solids were analyzed by standard methods (Andrew et al., 2005).
Fig. 1. The setup of photo-reactor.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
3
Table 1 Crystal size at the diffracted peak 2q of 25.5 anatase phase (1 0 1).
Fig. 2. XRD diffractogram of uncalcined and calcined pure TiO2 samples.
Obviously, the uncalcined and calcined TiO2 samples exhibited multiple sharp peaks symptomatic of high crystallinity. The presence of anatase phase can be positively identified via 2q ¼ 25.5 , 37.2 , 38.0 , 38.8 , 48.2 , 54.0 , 55.0 and some other minor peaks, i.e. 62.0 , 70.0 and 76.0 . Moreover, the phase of the uncalcined and calcined TiO2 samples showed similar diffracted spectrums indicating that the formation of rutile phase was avoided under calcination temperature of 573 and 1173 K. Therefore, a calcination temperature of 723 K was set in the current work in lieu of the aforementioned results and also to adopt a similar calcination temperature as reported by the previous researchers (Liu et al., 2007; Sun and Li, 2003; Ma et al., 2003; Tsai and Teng, 2004). In addition, the synthesized catalysts were also subjected to the same XRD peak scanning and the results are shown in Fig. 3. Significantly, it can be observed that the crystallite species associated with Ag was undetected by the XRD diffractogram and that the diffractogram representing the Ag-loaded samples was identical with the pristine TiO2. It may be inferred that the Ag metal mixed and dispersed very well within the TiO2 matrix. In addition, Table 1 lists the crystal size associated with 2q at 25.5 (1 0 1), the sharpest peak
Fig. 3. XRD diffractogram of Ag/TiO2 samples.
Sample
Crystallite size (nm)
Uncalcined TiO2 Calcined TiO2 0.25 wt% Ag 0.50 wt% Ag 1.0 wt% Ag
45.5 45.7 46.3 45.5 46.5
recorded from the current analysis as a representative of the sample. It can be seen that the crystal size before and after doping remained nearly invariant (45.0e46.0 nm) suggesting that the metal introduction and high calcination temperature did not cause neither phase change nor sintering of crystal cluster, further validating the earlier proposition. Figs. 4 and 5 show the FESEM images of the calcined TiO2 and a few selected Ag/TiO2 samples, viz. 0.50 and 1.0 wt% of Ag, respectively. Clearly, the particles were in spherical shape and the size of the particles although varied, showed a dominance of smaller particle size. Indeed, the estimations of the size showed particle dominance with the size below 100 nm with an occasional presence of larger particles, above 150 nm (inset). Moreover, it can be observed that no marked differences in size or appearance can be found for the Ag/TiO2 solid sample across the different metal loadings. This observation was also consistent with the XRD pattern that showed no marked differences. It can be observed from Table 2 that the calcined TiO2 has a rather low BET specific surface area (8.73 m2 g1). Upon the incorporation with Ag metal, it can be seen that the BET specific surface area showed a slight increment (averaged 10.0 m2/g). The increase in BET specific surface area can be attributed to the deposited Ag that has endowed surface ruggedness to the synthesized catalyst. In addition, Table 2 also shows that the pore diameter values fall into the mesopores range (11.0e23.0 nm). Fig. 6 shows the diffuse reflective UVevis spectrum of TiO2 and also the Ag/TiO2. The results obtained have indicated that the pure TiO2 exhibited strong absorption in the UV region only, corresponding to its band gap energy. The absorption weakened as the spectrum transitioned into the visible light region. Significantly, the optical absorption edge has obviously shifted to the visible range with the increasing in Ag/TiO2 ratio. The Ag/TiO2 exhibited absorption in the short wavelength region of visible
Fig. 4. FESEM images of the calcined TiO2.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
4
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
Fig. 6. Diffuse reflectance UVeVis spectra of the photocatalysts.
3.2. Characterization of POME The wastewater characteristics were analyzed by using standard methods (Andrew et al., 2005) and the results are shown in Table 4. The COD and BOD values were 60,000 and 24,000 mg/L, respectively, with total solids and total suspended solids averaging 24,000 and 10,000 mg/L, respectively. These values are typical of POME discharge from the first pond. 3.3. Photocatalytic degradation of POME
Fig. 5. FESEM images of (a) 0.50 wt% Ag/TiO2 and (b) 1.0 wt% Ag/TiO2.
light from 403 to 423 nm in comparison with the calcined TiO2 (394 nm). The maximum absorption of visible light was observed for 1.0 wt% Ag/TiO2. Moreover, the band gap energy (E) of all the photocatalysts can be estimated from E ¼ 1240=l ðnmÞ and the obtained results are summarized in Table 3. Result indicated that the band gap energy values of Ag/TiO2 were between 3.15 and 2.4 eV, considerably narrower than that of the pure TiO2 which was 3.15 eV. Furthermore, it was also demonstrated that visible light absorption was enhanced with Ag percentage. The improvement in visible light absorption may have resulted from: (i) Ag atoms replace a portion of O atoms in TiO2 crystallite, forming a new energy level on top of the valence band to generate a red shift; and (ii) the specific 3d electronic configuration of Ag plays crucial role in generating electronehole pairs to improve the visible light response. Table 2 BET specific surface area and pore volume of the catalysts. Catalyst
BET specific surface area (m2 g1)
BJH average pore diameter (nm)
Calcined TiO2 0.25 wt% Ag 0.5 wt% Ag 1.0 wt% Ag
8.73 10.86 9.23 10.73
22.37 15.00 22.90 11.10
3.3.1. Effects of Ag metal doping To determine the effects of Ag/TiO2 photocatalysts towards the photocatalytic degradation of POME irradiated by 100 W of UV and visible light sources, respectively, a volume of 300 mL of filtered POME was diluted to the concentration of circa 650e700 ppm (BOD of 450) and mixed with 0.06 g of photocatalyst (loading of 0.2 g L1) at the onset of experiments. These pre-treatment of POME steps were necessary to ensure the employed light source can penetrate the reaction medium and activate the photocatalyst. Due to the reason that the reaction in the current work involved a chemical pathway, therefore the determination of degradation efficiency, or also known conversion was measured in terms of COD parameter only. It was estimated as:
X ð%Þ ¼
1
CA CAo
100
(1)
whereby CA and CAo denote the COD concentration of POME at time (t) and the initial COD concentration of the POME, respectively. A blank run was carried out beforehand, viz. in the presence of solid catalyst under the dark condition. It was found that after 10 h of rigorous stirring, the COD of POME was unchanged indicating that the adsorption of POME was small, most likely owing to the small BET specific surface area. Subsequently, for all the remaining experiments, the slurry solution was stirred in the dark for 2 h to
Table 3 Band gap energy of the solid samples. Photocatalyst
Band gap energy (eV)
Calcined TiO2 0.25 wt% Ag/TiO2 0.5 wt% Ag/TiO2 1.0 wt% Ag/TiO2
3.20 2.85 2.50 2.40
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8 Table 4 Characterization of raw palm oil mill effluent (POME). Parameters
Values (mg/L)
COD BOD Total solids Total suspended solids Ammoniacal nitrogen Nitrate nitrogen Total dissolved solids
60,600 24,000 24,050 10,040 23 160 12,900
obtain good dispersion and establish adsorptionedesorption equilibrium before photocatalytic reaction commenced. Fig. 7(a) and (b) shows the transient pre-treated POME photo-degradation under the UV-irradiation and visible light-irradiation, respectively, for 80 min in each runs to gauge the initial photo-catalytic reaction trend. The organic compounds in the POME have decomposed upon irradiation. Consequently, the concentration always dropped with the irradiation time due to the dwindling of the organic compounds in the POME sample attributed to the photocatalytic decomposition. It can be observed from the transient conversion profiles depicted in Fig. 8 that the blank run (with light source and an absence of photo-catalyst) yielded POME photo-degradation conversion of less than 1% under the 100 W UV and also 100 W visible
Fig. 7. Transient concentration profiles for the pre-treated POME at 0.2 g L1 photocatalyst loading under the 100 W of (a) UV light and (b) visible light.
5
light sources, respectively. In the presence of TiO2 however, the degradation improved considerably to attain 6.5% and 2.0% conversions at the 80th min for both the UV light and visible light sources, respectively. This has indicated that photocatalytic has indeed occurred to accelerate the degradation rate of the pretreated POME which serves as an important finding from the current work. Significantly, the Ag/TiO2 photocatalyst consistently recorded improved photocatalytic degradation activity compared to the pristine TiO2 photocatalyst under both the UV and also visible light irradiations indicating the superiority of the Ag inclusion. This may be due to the enhanced absorption wavelength associated with Ag/TiO2 photocatalyst indicative of the easiness of the Ag/TiO2 catalyst to function effectively under the two different types of light sources. Moreover, across the different metal dopant, 0.5 wt% loading of Ag dopant has yielded the maximum conversion for both the UV (25.0% conversion) and visible light (17.0% conversion) irradiations. In addition, under the UV light, a conversion of 10% was achieved by the 1.0 wt% Ag/TiO2 whilst for the 0.25 wt% Ag/TiO2, the conversion was 6.0%. When irradiated with visible light, the 0.25 wt% and 1.0 wt% Ag/TiO2 photocatalysts yielded almost similar conversion, at 4.0%. This showed that the UV irradiation still yielded the highest degradation efficiency compared to the visible light. This is unsurprising considering that the matrix of the Ag/TiO2 photocatalysts was still comprised of primarily TiO2 material which is a proven active material under the UV source. Nonetheless, the interesting part is that when TiO2 was incorporated with Ag metal, the synthesized photocatalyst showed significant activity towards the photocatalytic degradation of POME
Fig. 8. Photocatalytic degradation of pre-treated POME over the 0.2 g L1 of different photocatalysts under the irradiation of 100 W of (a) UV light and (b) visible light.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
6
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
under the visible light. The superiority of 0.5 wt% Ag/TiO2 may be ascribed to a better Ag metal distribution on the surface of titania, hence enabling rapid transfer of electron and consequently, preventing a recombination of proton (hþ) and electron (e) from taking place. Furthermore, the said photocatalyst also exhibited low band gap energy of 2.50 eV (cf. Table 3). Based on the concept of photocatalytic degradation, the following mechanism is proposed:
Ag=TiO2 þ hv/Ag=TiO2
þ e cb þ hvb
(M1)
Ag=TiO2 þ H2 Oads /Ag=TiO2 þ OH þ Hþ
(M2)
e cb þ O2;ads /O2
(M3)
· hþ vb þ OH / OH
(M4)
·
(M5)
OH þ Organic complexes ðPOMEÞ/H2 O; CO2 ; etc:
According to the LangmuireHinshelwood (LeH) rate law model for the irreversible and surface reaction-limited that occurs on a single site, the following LeH rate law can be employed to describe the reaction pattern:
dC kCA A ¼ ðrA Þ ¼ dt 1 þ KA CA
(2)
At low concentration of reactant, in the current study represented by the organic content of the pre-treated POME, the expression for the denominator can be further simplified into (1 [ KACA) z 1. Consequently, the POME decomposition behaviour can be simply described by a common Power Law model. For a batch reactor with photo-catalytic kinetics adhering to the pseudo first-order reaction kinetics, it follows that:
dC A ¼ kCA dt
(3)
Upon integration, the following expression is produced:
C Ln Ao ¼ kt CA
(4)
whereby CAo ¼ Initial concentration (mg L1), CA ¼ Concentration at time t (mg L1), t ¼ time (min), k ¼ apparent specific reaction rate (min1). When the concentration profiles were fitted to the pseudo firstorder kinetics model, it can be observed that the photocatalytic degradation trend was convincingly-captured (cf. Fig. 9). Consequently, the specific reaction rate constants were estimated. Fig. 9 shows the linearized plot of the transient concentration data from where the values of k were determined from the linear slope. Table 5 shows the corresponding modelling results. The apparent specific reaction rate (k) as summarized in Table 5 shows that the 0.50 wt% Ag metal loading yielded the highest values at 3.60 103 min1 for the UV-light irradiation and 2.45 103 min1 for the visible light irradiation, respectively. Hence, this explained the highest conversions attained compared to their counterparts and even the pure TiO2. It is not surprising to note that the blank run yielded the lowest k values at 0.092 103 and 0.120 103 min1, respectively. 3.3.2. Effects of different photocatalyst loadings As determined previously, the 0.50 wt% Ag/TiO2 showed the highest performance in photocatalytic degradation of pre-treated
Fig. 9. The apparent first-order kinetics of pre-treated POME degradation over 0.2 g L1 of different photocatalysts when irradiated with 100 W of (a) UV source and (b) visible light.
POME under the irradiation of both UV and visible lights. Hence, effects of different photocatalyst loadings to the photocatalytic degradation of a pre-treated POME were further sought. In order to capture the catalytic performance in a wider time spectrum, all the reactions were examined over duration of 8 h. Various loadings viz. 0.2, 0.5, 1.0, 1.5 and 2.0 g L1 of 0.50 wt% Ag/TiO2 were therefore employed for the ensuing investigation. The resulting transient concentration profiles are shown in Fig. 10. It can be observed that the degradation efficiency seems to follow a hyperbolic trend. This Table 5 Kinetics parameters obtained from the simple power law modelling. Photocatalyst
Apparent specific reaction rate, k (103 min1)
Regression coefficient, R2
100 W UV source Blank TiO2 only 0.25 wt% Ag/TiO2 0.50 wt% Ag/TiO2 1.0 wt% Ag/TiO2
0.092 1.045 1.036 3.582 1.502
0.95 0.85 0.85 0.96 0.92
100 W visible light Blank TiO2 only 0.25 wt% Ag/TiO2 0.50 wt% Ag/TiO2 1.0 wt% Ag/TiO2
0.120 0.284 0.730 2.440 0.579
0.98 0.94 0.96 0.97 0.96
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
7
degradation of pre-treated POME judging by the highest conversion recorded, around 85.0% at the 8th hour, over the 1.0 g/L. In contrast, the visible light system only recorded 60.0% conversion at the 8th hour, as the highest conversion attained over the 1.0 g/L photocatalyst, which was however, significantly higher than the pure TiO2. Even at the same photocatalyst loading of 0.2 g/L, Ag/TiO2 photocatalyst yielded a higher conversion compared to the TiO2 only. Once again, the superiority of Ag/TiO2 photocatalyst compared to the pure TiO2 has been underlined for both types of system. 4. Conclusions Titania loaded with Ag was employed as a photocatalyst for photocatalytic degradation of POME, an organic liquid discharge from oil palm mills. XRD characterization showed that the final form of the entire Ag/TiO2 photocatalysts (0.25e1.0 wt% Ag) still has retained its nascent anatase phase with crystal diameter ranged from 42.0 to 45.0 nm. Interestingly, UVeVis spectrum seems to indicate that the Pt inclusion successfully narrowed the band gap to 2.50 eV (for the 0.5 wt% Ag/TiO2) from 3.20 eV as recorded for the pure TiO2 photocatalyst. Moreover, the absorption of light in the visible light spectrum has noticeably-improved whilst almost similar absorption capacity for the UV region was also retained. Photocatalytic degradation of the pre-treated POME showed that Ag/TiO2 has significantly functioned more effectively compared to the pure TiO2 when irradiated by both UV and also visible light sources, respectively. This improvement can be largely credited to the narrower band gap energy and improved visible light absorption. In addition, Ag may have also enhanced the charge separation by rapidly-transferring the e away from the positive hþ charge on the TiO2 surface, thus minimizing charge recombination. It was also determined from this work that the 0.50 wt% Ag/TiO2 yielded the best photocatalytic degradation whilst a loading of 1.0 g/L of photocatalyst was the most preferred. Acknowledgements Fig. 10. Transient conversion profiles of pre-treated POME degradation over a 0.50 wt% Ag-loaded TiO2 at different photocatalyst loadings under the 100 W irradiation of (a) UV light and (b) visible light source.
may indicate that deactivation of photocatalyst has occurred, most likely via an adsorption of intermediate organic species on active sites, consequently the number of active sites dropped with irradiation time. For the case of UV-light irradiation, as in Fig. 10(a), it can be observed that generally, Ag/TiO2 showed a far better facilitation of POME degradation compared to the pure TiO2 at a similar photocatalyst loading viz. 0.2 g/L. In addition, when the photocatalyst loading was increased for Ag/TiO2, the POME photocatalytic degradation was even more profound. This may be because the presence of higher concentration of photocatalyst would produce more OH radical. Consequently, the POME photocatalytic degradation would have increased with the photocatalyst loading. Nonetheless, beyond a certain photocatalyst loading limit, the degradation performance would be negatively-affected, i.e. for Ag/ TiO2 system, the 1.5 and 2.0 g/L loadings showed a markedly lower performance compared to the 1.0 g/L. This can be ascribed to the excessive entry of photocatalyst into the reaction media that has caused the penetration of light being reduced. Consequently, the degradation efficiency has been adversely-affected. For the visible light system (cf. Fig. 10(b)), the same trend was observed. Blank run as well as 0.2 g/L of pure TiO2 yielded the lowest degradation efficiency judging by the concentration profile. Furthermore, the UV source still offered the best light spectrum for the photocatalytic
Authors would like to acknowledge the Science Fund Grant (RDU130501) from the Ministry of Science, Technology and Innovation Malaysia. MRD appreciates the Universiti Malaysia Pahang for a provision of studentship (GRS120355). References Abdul, K., Mohamed, I., Kamil, A., 1989. Biological treatment of palm oil mill effluent using Triychodermu viride. Biol. Waste 27, 143e152. Andrew, D., Lenore, S., Eugene, W., Mary Ann, H., 2005. Standard Methods for the Examination of Water and Wastewater. American Public Health Association. Ao, Y., Xu, J., Zhang, S., Fu, D., 2010. A one-pot method to prepare N-doped titania hollow spheres with high photocatalytic activity under visible light. Appl. Surf. Sci. 256, 2754e2758. Badri, M.A., 1984. Identification of heavy metal toxicity levels in solid wastes by chemical speciation. Conserv. Recycl. 7, 257e269. Bello, M., Nourouzi, M., Abdullah, L.C., Choong, T.S., Koay, Y., Keshani, S., 2013. POME is treated for removal of color from biologically treated POME in fixed bed column: applying wavelet neural network (WNN). J. Hazard. Mater. 262, 106e113. nchez, E., 1996. Anaerobic treatment of palm oil mill effluent Borja, R., Banks, C.J., Sa in a two-stage up-flow anaerobic sludge blanket (UASB) system. J. Biotechnol. 45, 125e135. Chakrabortty, D., Gupta, S.S., 2013. Photo-catalytic decolourisation of toxic dye with N-doped titania: a case study with acid blue 25. J. Environ. Sci. 25, 1034e1043. Chen, Z., Xu, J., Ren, Z., He, Y., Xiao, G., 2013. Low temperature synthesis of ZnIn2S4 microspheres as a visible light photocatalyst for selective oxidation. Catal. Commun. 41, 83e86. Chen, W., Chen, Z., Liu, T., Jia, Z., Liu, X., 2014. Fabrication of highly visible light sensitive graphite-like C3N4 hybridized with Zn0.28Cd0.72S heterojunctions photocatalyst for degradation of organic pollutants. J. Environ. Chem. Eng. 2, 1889e1897. Chin, K., Ng, W., Ma, A., Wong, K., 1987. Treatability studies of palm oil refinery wastewaters. Water. Sci. Technol. 19, 23e29.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
8
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
Chuong, T.V., Dung, L.Q.T., Khieu, D.Q., 2008. Synthesis of nano titanium dioxide and its application in photocatalysis. J. Kor. Phys. Soc. 52, 1526e1529. Giri, R.R., Ozaki, H., Takanami, R., Taniguchi, S., 2008. A novel use of TiO2 fiber for photocatalytic ozonation of 2, 4-dichlorophenoxyacetic acid in aqueous solution. J. Environ. Sci. 20, 1138e1145. Harsono, S.S., Grundmann, P., Soebronto, S., 2014. Anaerobic treatment of palm oil mill effluents: potential contribution to net energy yield and reduction of greenhouse gas emissions from biodiesel production. J. Clean. Prod. 64, 619e627. Karim, M.I.A., Hie, L.L., 1987. The use of coagulating and polymeric flocculating agents in the treatment of palm oil mill effluent (POME). Biol. Waste 22, 209e218. Kasanen, J., Salstela, J., Suvanto, M., Pakkanen, T.T., 2011. Photocatalytic degradation of methylene blue in water solution by multilayer TiO2 coating on HDPE. Appl. Surf. Sci. 258, 1738e1743. Kongnoo, A., Suksaroj, T., Intharapat, P., Promtong, T., Suksaroj, C., 2012. Decolorization and organic removal from palm oil mill effluent by Fenton's process. Environ. Eng. Sci. 29, 855e859. Liu, W., Gao, J., Zhang, F., Zhang, G., 2007. Preparation of TiO2 nanotubes and their photocatalytic properties in degradation methylcyclohexane. Mater. Trans. 48, 2464. Ma, A., Ong, A., 1986. Palm oil processingdnew development in effluent treatment. Water. Sci. Technol. 18, 35e40. Ma, R., Bando, Y., Sasaki, T., 2003. Nanotubes of lepidocrocite titanates. Chem. Phys. Lett. 380, 577e582. Mahmoodi, N.M., 2011. Photocatalytic ozonation of dyes using copper ferrite nanoparticle prepared by co-precipitation method. Desalination 279, 332e337. Ng, W., Goh, A.C., Tay, J., 1988. Palm oil mill effluent treatmentdliquidesolid separation with dissolved air flotation. Biol. Waste 25, 257e268. Othman, M.R., Hassan, M.A., Shirai, Y., Baharuddin, A.S., Ali, A.M.M., Idris, J., 2014. Treatment of effluents from palm oil mill process to achieve river water quality for reuse as recycled water in a zero emission system. J. Clean. Prod. 67, 58e61. Quesada-Cabrera, R., Mills, A., O'Rourke, C., 2014. Action spectra of P25 TiO2 and a visible light absorbing, carbon-modified titania in the photocatalytic degradation of stearic acid. Appl. Catal. B-Environ. 150, 338e344.
Ramchiary, A., Samdarshi, S., 2014. Ag deposited mixed phase titania visible light photocatalyst e superiority of Agetitania and mixed phase titania co-junction. Appl. Surf. Sci. 305, 33e39. Roge, W., Velayuthan, A., 1981. Preliminary trials with Westfalia-3-phase decanters for palm oil separation. In: Palm Oil Prod. Technol. Eighties, Rep. Proc. Int. Conf., Inc. Sot. Plant. Kuala Lumpur, Malaysia, pp. 327e334. Santosa, S.J., 2008. Palm oil boom in Indonesia: from plantation to downstream products and biodiesel. Clean Soil Air Water 36, 453e465. Shahid, M., Jingling, L., Ali, Z., Shakir, I., Warsi, M.F., Parveen, R., Nadeem, M., 2013. Photocatalytic degradation of methylene blue on magnetically separable MgFe2O4 under visible light irradiation. Mater. Chem. Phys. 139, 566e571. Shen, Y., Wu, Y., Li, X., Zhao, Q., Hou, Y., 2013. One-pot synthesis of MgFe2O4 nanospheres by solvothermal method. Mater. Lett. 96, 85e88. Spanjers, H., Lier, J., 2006. Instrumentation in anaerobic treatment research and practice. Water. Sci. Technol. 53, 63e76. Sri, T., Dompok, B., 2011. Palm Oil and Rubber. Sun, X., Li, Y., 2003. Synthesis and characterization of ion-exchangeable titanate nanotubes. Chem. Eur. J. 9, 2229e2238. Tabassum, S., Zhang, Y., Zhang, Z., 2015. An integrated method for palm oil mill effluent (POME) treatment for achieving zero liquid discharge e a pilot study. J. Clean. Prod. 95, 148e155. Tsai, C.-C., Teng, H., 2004. Regulation of the physical characteristics of titania nanotube aggregates synthesized from hydrothermal treatment. Chem. Mater. 16, 4352e4358. Virkutyte, J., Jegatheesan, V., Varma, R.S., 2012. Visible light activated TiO2/microcrystalline cellulose nanocatalyst to destroy organic contaminants in water. Bioresour. Technol. 113, 288e293. Wang, J., Li, C., Zhuang, H., Zhang, J., 2013. Photocatalytic degradation of methylene blue and inactivation of Gram-negative bacteria by TiO2 nanoparticles in aqueous suspension. Food Control 34, 372e377. Zalas, M., 2014. Synthesis of N-doped template-free mesoporous titania for visible light photocatalytic applications. Catal. Today 230, 91e96.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation Chin Kui Cheng a, b, c, *, Mohd Rizauddin Deraman c, Kim Hoong Ng c, Maksudur R. Khan c a
Rare Earth Research Centre, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia Centre of Excellence for Advanced Research in Fluid Flow, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia c Faculty of Chemical & Natural Resources Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 November 2014 Received in revised form 4 June 2015 Accepted 24 June 2015 Available online xxx
This paper reports on the photocatalytic degradation of pre-treated palm oil mill effluent (POME) over titania loaded with photocatalyst. The argentum loading on the titania was varied from 0.25 to 1.0 wt% via wet-impregnation technique. X-ray diffraction characterization of all the photocatalysts showed that the photo-active rutile phase was still intact after the photocatalyst synthesis. In addition, the UVeVis diffuse reflectance measurements indicate an improved visible light energy absorption and that the band gap energy was significantly reduced (averaging 2.50 eV) when titania was loaded with argentum, compared to the pristine titania that recorded a reading of 3.20 eV. The 0.50 wt% argentum/titania photocatalyst offered the most effective degradation of pre-treated POME under the irradiation of 100 W of UV light (25.0%) and also visible light (16.0%), respectively, over a loading of 0.2 g/L. Significantly, the maximum photocatalyst loading determined from the current work was 1.0 g/L. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Argentum Photocatalysis Silver POME Titania Wastewater
1. Introduction Palm oil mill effluent (POME) is a thick, brownish slurry waste produced from a burgeoning oil palm industry. Malaysia is currently the world's largest exporter of palm oil although it is the second-largest producer of the oil after the neighbouring country Indonesia. It has planted area of 4.9 million hectares with a production of 17 million tonnes of palm oil. The palm oil industry remains a significant contributor to the Malaysian economy with a reported export earnings of about RM60 billion annually (Sri and Dompok, 2011). In terms of composition, POME is a complex liquor comprised of amino acids, inorganic nutrients such as sodium, potassium, calcium, magnesium, short fibres, organelles, nitrogenous constituents, free organic acids and a mixture of carbohydrates ranging from hemicelluloses to simple sugars, etc. (Santosa, 2008). It has been estimated that approximately 1.5 m3 of water are required to process a ton of fresh fruit bunch whereby half of the amount becomes POME (Kongnoo et al., 2012). In
* Corresponding author. Faculty of Chemical & Natural Resources Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia. Tel.: þ60 9 549 2896; fax: þ60 9 549 2889. E-mail address: [email protected] (C.K. Cheng).
general, the chemical oxygen demand (COD) and biochemical oxygen demand (BOD) in the POME are in the range of 15,000e100,000 mg/L and 10,000e43,750 mg/L, respectively. Due to its acidic nature (pH 3.4e5.2), high values of COD and also biochemical oxygen demand (BOD), POME can inflict considerable environmental problems if discharged without effective treatment (Borja et al., 1996). Over the past decades, several economicallyviable technological solutions have been proposed for the treatment of POME, viz. simple skimming devices (Roge and Velayuthan, 1981; Ng et al., 1988), land disposal (Ma and Ong, 1986), chemical coagulation and flotation (Badri, 1984; Othman et al., 2014; Chin et al., 1987; Karim and Hie, 1987), aerobic (Abdul et al., 1989) and anaerobic biological processes (Harsono et al., 2014). Among traditional technologies, anaerobic treatment followed by aerobic biological processes are most widely used because of their particular advantages, such as low biomass yield, low nutrient requirement, and high volumetric organic loading (Spanjers and Lier, 2006). Unfortunately, some of these methods are land-intensive while others require long hydraulic retention time (Tabassum et al., 2015). Alternatively, the destruction of organic pollutants could be carried under light source employing semiconductor-based materials as the heterogeneous catalysts. Specifically, two types of light spectrum have been explored viz. UV spectrum
http://dx.doi.org/10.1016/j.jclepro.2015.06.104 0959-6526/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
2
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
(200e400 nm) and visible light spectrum (400e700 nm). Due to the different wavelengths, material selection is critical in ensuring that light energy can be absorbed by the photocatalyst. Titania (TiO2) photocatalyst has been widely used to degrade organic compound, referenced herein (Wang et al., 2013; Virkutyte et al., 2012; Kasanen et al., 2011; Giri et al., 2008). Nevertheless, owing to the wide band gap (~3.3 eV), TiO2 can only effectively absorb UV light, which however, only represents about 5% of our solar energy. In contrast, the visible light spectrum accounts for circa 46%, hence represents a more viable source. Consequently, two different strategies have been undertaken viz. to employ new materials to substitute the TiO2 (Shahid et al., 2013; Chen et al., 2014, 2013; Shen et al., 2013; Mahmoodi, 2011) or to incorporate dopants into the TiO2 matrix to extend the light spectrum into the visible region (Zalas, 2014; Ramchiary and Samdarshi, 2014; Quesada-Cabrera et al., 2014; Chakrabortty and Gupta, 2013; Ao et al., 2010). Specifically, this work focuses on latter approach by doping TiO2 with the silver (Ag) for the photocatalytic degradation of POME. The current work will serve as an important benchmark in studying the use of photocatalytic technique to treat the POME. 2. Methodology 2.1. Catalyst preparation and characterization A 0.1 M aqueous solution of AgNO3 was mixed with TiO2 powder (Sigma Aldrich) of different weight percentage, viz. 0.25, 0.5 and 1.0 wt% of Ag/TiO2 in a 500 mL beaker. The mixture was then stirred using an ultrasonic water bath for 8 h at 328 K. Subsequently, it was dried in an oven at 378 K for 8 h. The resulting powder was milled and washed repeatedly by distilled water to remove the impurities. The powder was re-dried for 8 h before re-milled. It was finally calcined at 723 K for catalyst characterization and photocatalytic reaction studies. BET multipoint analysis was carried out using Thermo-Scientific Surfer to determine the BET specific surface area of the fresh catalysts. N2 with a cross-sectional area of 16.2 Å2 was used as the adsorbate in the analysis with analysis temperature at 77 K. The crystalline structure was obtained via XRD diffraction analysis. The samples were irradiated by Ni-filtered CuKa with a wavelength (l) of 1.542 Å at 40 mA and 45 kV, and scanning from 10 to 80 at 4 min1 in a Rigaku Miniflex II instrument. In addition, band gap energy of all the synthesized catalysts was spectroscopicallydetermined from the UVevis diffuse reflectance measurements provided by a Jasco V-550 spectrophotometer equipped with an integrating sphere. The instrument measured the absorption spectrum of a sample at wavelengths in the range of 190e900 nm. The light source used was a deuterium lamp and a halogen (WI) lamp. In addition, FTIR spectroscopy was also applied primarily to study the chemical structure. The infrared spectra of the catalyst samples were measured at wavelengths ranging from 1000 to 4000 cm1.
2.3. Photocatalytic degradation of pre-treated POME For the studies of photocatalytic degradation of POME, the fresh POME was filtered to eliminate the solid suspension. Subsequently, it was further diluted with deionized water (Millipore). The filtration and dilution steps were necessary as fresh POME in its existing form was impenetrable to the light source. In order to prevent the wastewater from undergoing microbial-biodegradation that would change the content of the POME, the diluted POME was preserved at below 4 C in a chiller (Thermo) (Bello et al., 2013). By using a similar source of pre-treated POME, it would prevent the variation in the POME composition. For the photocatalytic degradation studies, a 1-L quartz photoreactor with separable quartz compartment housing the light source was procured from Shanghai Sunny Scientific Collaboration Co. Ltd. Fig. 1 shows the setup of photoreaction whereby the temperature during the reaction was consistently-maintained at 298 K. For the reaction studies, 300 mL of the pre-treated POME was mixed with a pre-determined catalyst amount in the quartz reactor and rigorously-stirred. A 100 W UV lamp and a 100 W Xenon lamp (Shanghai Sunny Scientific Collaboration Co. Ltd.) were employed as the light sources for the current work. The technical specification provided by the manufacturer has indicated that the visible light system was free from UV interference. For the sampling purpose, around 5.0 mL of sample was withdrawn from the reactor for chemical oxygen demand (COD) measurement. For the current study, the model of COD reactor used was Hach RB-200. 3. Results and discussion 3.1. Characterization of photocatalyst Previous works have indicated that TiO2 is mainly comprised of anatase or rutile phase with the former being more active towards UV-absorption whilst the latter is inactive (Chuong et al., 2008). Moreover, the crystalline phase could transform with calcination temperature. Hence, in the current work, the pristine TiO2 and calcined TiO2 samples were subjected to XRD scanning to determine a suitable calcination temperature in order to prevent rutile phase formation. Fig. 2 shows the resulting XRD diffractogram.
2.2. Wastewater collection and characterization Raw POME was collected from Felda palm oil industries located in Kuantan, Malaysia. The temperature of raw POME at the discharge point was around 80e90 C. The sample was transported to the laboratory in sterile 1000 mL Schott bottles placed in ice and stored at 4 C until use. The wastewater characteristics such as chemical oxygen demand (COD), biochemical oxygen demand (BOD), total solids, total suspended solids, ammoniacal nitrogen, nitrate nitrogen and total dissolved solids were analyzed by standard methods (Andrew et al., 2005).
Fig. 1. The setup of photo-reactor.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
3
Table 1 Crystal size at the diffracted peak 2q of 25.5 anatase phase (1 0 1).
Fig. 2. XRD diffractogram of uncalcined and calcined pure TiO2 samples.
Obviously, the uncalcined and calcined TiO2 samples exhibited multiple sharp peaks symptomatic of high crystallinity. The presence of anatase phase can be positively identified via 2q ¼ 25.5 , 37.2 , 38.0 , 38.8 , 48.2 , 54.0 , 55.0 and some other minor peaks, i.e. 62.0 , 70.0 and 76.0 . Moreover, the phase of the uncalcined and calcined TiO2 samples showed similar diffracted spectrums indicating that the formation of rutile phase was avoided under calcination temperature of 573 and 1173 K. Therefore, a calcination temperature of 723 K was set in the current work in lieu of the aforementioned results and also to adopt a similar calcination temperature as reported by the previous researchers (Liu et al., 2007; Sun and Li, 2003; Ma et al., 2003; Tsai and Teng, 2004). In addition, the synthesized catalysts were also subjected to the same XRD peak scanning and the results are shown in Fig. 3. Significantly, it can be observed that the crystallite species associated with Ag was undetected by the XRD diffractogram and that the diffractogram representing the Ag-loaded samples was identical with the pristine TiO2. It may be inferred that the Ag metal mixed and dispersed very well within the TiO2 matrix. In addition, Table 1 lists the crystal size associated with 2q at 25.5 (1 0 1), the sharpest peak
Fig. 3. XRD diffractogram of Ag/TiO2 samples.
Sample
Crystallite size (nm)
Uncalcined TiO2 Calcined TiO2 0.25 wt% Ag 0.50 wt% Ag 1.0 wt% Ag
45.5 45.7 46.3 45.5 46.5
recorded from the current analysis as a representative of the sample. It can be seen that the crystal size before and after doping remained nearly invariant (45.0e46.0 nm) suggesting that the metal introduction and high calcination temperature did not cause neither phase change nor sintering of crystal cluster, further validating the earlier proposition. Figs. 4 and 5 show the FESEM images of the calcined TiO2 and a few selected Ag/TiO2 samples, viz. 0.50 and 1.0 wt% of Ag, respectively. Clearly, the particles were in spherical shape and the size of the particles although varied, showed a dominance of smaller particle size. Indeed, the estimations of the size showed particle dominance with the size below 100 nm with an occasional presence of larger particles, above 150 nm (inset). Moreover, it can be observed that no marked differences in size or appearance can be found for the Ag/TiO2 solid sample across the different metal loadings. This observation was also consistent with the XRD pattern that showed no marked differences. It can be observed from Table 2 that the calcined TiO2 has a rather low BET specific surface area (8.73 m2 g1). Upon the incorporation with Ag metal, it can be seen that the BET specific surface area showed a slight increment (averaged 10.0 m2/g). The increase in BET specific surface area can be attributed to the deposited Ag that has endowed surface ruggedness to the synthesized catalyst. In addition, Table 2 also shows that the pore diameter values fall into the mesopores range (11.0e23.0 nm). Fig. 6 shows the diffuse reflective UVevis spectrum of TiO2 and also the Ag/TiO2. The results obtained have indicated that the pure TiO2 exhibited strong absorption in the UV region only, corresponding to its band gap energy. The absorption weakened as the spectrum transitioned into the visible light region. Significantly, the optical absorption edge has obviously shifted to the visible range with the increasing in Ag/TiO2 ratio. The Ag/TiO2 exhibited absorption in the short wavelength region of visible
Fig. 4. FESEM images of the calcined TiO2.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
4
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
Fig. 6. Diffuse reflectance UVeVis spectra of the photocatalysts.
3.2. Characterization of POME The wastewater characteristics were analyzed by using standard methods (Andrew et al., 2005) and the results are shown in Table 4. The COD and BOD values were 60,000 and 24,000 mg/L, respectively, with total solids and total suspended solids averaging 24,000 and 10,000 mg/L, respectively. These values are typical of POME discharge from the first pond. 3.3. Photocatalytic degradation of POME
Fig. 5. FESEM images of (a) 0.50 wt% Ag/TiO2 and (b) 1.0 wt% Ag/TiO2.
light from 403 to 423 nm in comparison with the calcined TiO2 (394 nm). The maximum absorption of visible light was observed for 1.0 wt% Ag/TiO2. Moreover, the band gap energy (E) of all the photocatalysts can be estimated from E ¼ 1240=l ðnmÞ and the obtained results are summarized in Table 3. Result indicated that the band gap energy values of Ag/TiO2 were between 3.15 and 2.4 eV, considerably narrower than that of the pure TiO2 which was 3.15 eV. Furthermore, it was also demonstrated that visible light absorption was enhanced with Ag percentage. The improvement in visible light absorption may have resulted from: (i) Ag atoms replace a portion of O atoms in TiO2 crystallite, forming a new energy level on top of the valence band to generate a red shift; and (ii) the specific 3d electronic configuration of Ag plays crucial role in generating electronehole pairs to improve the visible light response. Table 2 BET specific surface area and pore volume of the catalysts. Catalyst
BET specific surface area (m2 g1)
BJH average pore diameter (nm)
Calcined TiO2 0.25 wt% Ag 0.5 wt% Ag 1.0 wt% Ag
8.73 10.86 9.23 10.73
22.37 15.00 22.90 11.10
3.3.1. Effects of Ag metal doping To determine the effects of Ag/TiO2 photocatalysts towards the photocatalytic degradation of POME irradiated by 100 W of UV and visible light sources, respectively, a volume of 300 mL of filtered POME was diluted to the concentration of circa 650e700 ppm (BOD of 450) and mixed with 0.06 g of photocatalyst (loading of 0.2 g L1) at the onset of experiments. These pre-treatment of POME steps were necessary to ensure the employed light source can penetrate the reaction medium and activate the photocatalyst. Due to the reason that the reaction in the current work involved a chemical pathway, therefore the determination of degradation efficiency, or also known conversion was measured in terms of COD parameter only. It was estimated as:
X ð%Þ ¼
1
CA CAo
100
(1)
whereby CA and CAo denote the COD concentration of POME at time (t) and the initial COD concentration of the POME, respectively. A blank run was carried out beforehand, viz. in the presence of solid catalyst under the dark condition. It was found that after 10 h of rigorous stirring, the COD of POME was unchanged indicating that the adsorption of POME was small, most likely owing to the small BET specific surface area. Subsequently, for all the remaining experiments, the slurry solution was stirred in the dark for 2 h to
Table 3 Band gap energy of the solid samples. Photocatalyst
Band gap energy (eV)
Calcined TiO2 0.25 wt% Ag/TiO2 0.5 wt% Ag/TiO2 1.0 wt% Ag/TiO2
3.20 2.85 2.50 2.40
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8 Table 4 Characterization of raw palm oil mill effluent (POME). Parameters
Values (mg/L)
COD BOD Total solids Total suspended solids Ammoniacal nitrogen Nitrate nitrogen Total dissolved solids
60,600 24,000 24,050 10,040 23 160 12,900
obtain good dispersion and establish adsorptionedesorption equilibrium before photocatalytic reaction commenced. Fig. 7(a) and (b) shows the transient pre-treated POME photo-degradation under the UV-irradiation and visible light-irradiation, respectively, for 80 min in each runs to gauge the initial photo-catalytic reaction trend. The organic compounds in the POME have decomposed upon irradiation. Consequently, the concentration always dropped with the irradiation time due to the dwindling of the organic compounds in the POME sample attributed to the photocatalytic decomposition. It can be observed from the transient conversion profiles depicted in Fig. 8 that the blank run (with light source and an absence of photo-catalyst) yielded POME photo-degradation conversion of less than 1% under the 100 W UV and also 100 W visible
Fig. 7. Transient concentration profiles for the pre-treated POME at 0.2 g L1 photocatalyst loading under the 100 W of (a) UV light and (b) visible light.
5
light sources, respectively. In the presence of TiO2 however, the degradation improved considerably to attain 6.5% and 2.0% conversions at the 80th min for both the UV light and visible light sources, respectively. This has indicated that photocatalytic has indeed occurred to accelerate the degradation rate of the pretreated POME which serves as an important finding from the current work. Significantly, the Ag/TiO2 photocatalyst consistently recorded improved photocatalytic degradation activity compared to the pristine TiO2 photocatalyst under both the UV and also visible light irradiations indicating the superiority of the Ag inclusion. This may be due to the enhanced absorption wavelength associated with Ag/TiO2 photocatalyst indicative of the easiness of the Ag/TiO2 catalyst to function effectively under the two different types of light sources. Moreover, across the different metal dopant, 0.5 wt% loading of Ag dopant has yielded the maximum conversion for both the UV (25.0% conversion) and visible light (17.0% conversion) irradiations. In addition, under the UV light, a conversion of 10% was achieved by the 1.0 wt% Ag/TiO2 whilst for the 0.25 wt% Ag/TiO2, the conversion was 6.0%. When irradiated with visible light, the 0.25 wt% and 1.0 wt% Ag/TiO2 photocatalysts yielded almost similar conversion, at 4.0%. This showed that the UV irradiation still yielded the highest degradation efficiency compared to the visible light. This is unsurprising considering that the matrix of the Ag/TiO2 photocatalysts was still comprised of primarily TiO2 material which is a proven active material under the UV source. Nonetheless, the interesting part is that when TiO2 was incorporated with Ag metal, the synthesized photocatalyst showed significant activity towards the photocatalytic degradation of POME
Fig. 8. Photocatalytic degradation of pre-treated POME over the 0.2 g L1 of different photocatalysts under the irradiation of 100 W of (a) UV light and (b) visible light.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
6
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
under the visible light. The superiority of 0.5 wt% Ag/TiO2 may be ascribed to a better Ag metal distribution on the surface of titania, hence enabling rapid transfer of electron and consequently, preventing a recombination of proton (hþ) and electron (e) from taking place. Furthermore, the said photocatalyst also exhibited low band gap energy of 2.50 eV (cf. Table 3). Based on the concept of photocatalytic degradation, the following mechanism is proposed:
Ag=TiO2 þ hv/Ag=TiO2
þ e cb þ hvb
(M1)
Ag=TiO2 þ H2 Oads /Ag=TiO2 þ OH þ Hþ
(M2)
e cb þ O2;ads /O2
(M3)
· hþ vb þ OH / OH
(M4)
·
(M5)
OH þ Organic complexes ðPOMEÞ/H2 O; CO2 ; etc:
According to the LangmuireHinshelwood (LeH) rate law model for the irreversible and surface reaction-limited that occurs on a single site, the following LeH rate law can be employed to describe the reaction pattern:
dC kCA A ¼ ðrA Þ ¼ dt 1 þ KA CA
(2)
At low concentration of reactant, in the current study represented by the organic content of the pre-treated POME, the expression for the denominator can be further simplified into (1 [ KACA) z 1. Consequently, the POME decomposition behaviour can be simply described by a common Power Law model. For a batch reactor with photo-catalytic kinetics adhering to the pseudo first-order reaction kinetics, it follows that:
dC A ¼ kCA dt
(3)
Upon integration, the following expression is produced:
C Ln Ao ¼ kt CA
(4)
whereby CAo ¼ Initial concentration (mg L1), CA ¼ Concentration at time t (mg L1), t ¼ time (min), k ¼ apparent specific reaction rate (min1). When the concentration profiles were fitted to the pseudo firstorder kinetics model, it can be observed that the photocatalytic degradation trend was convincingly-captured (cf. Fig. 9). Consequently, the specific reaction rate constants were estimated. Fig. 9 shows the linearized plot of the transient concentration data from where the values of k were determined from the linear slope. Table 5 shows the corresponding modelling results. The apparent specific reaction rate (k) as summarized in Table 5 shows that the 0.50 wt% Ag metal loading yielded the highest values at 3.60 103 min1 for the UV-light irradiation and 2.45 103 min1 for the visible light irradiation, respectively. Hence, this explained the highest conversions attained compared to their counterparts and even the pure TiO2. It is not surprising to note that the blank run yielded the lowest k values at 0.092 103 and 0.120 103 min1, respectively. 3.3.2. Effects of different photocatalyst loadings As determined previously, the 0.50 wt% Ag/TiO2 showed the highest performance in photocatalytic degradation of pre-treated
Fig. 9. The apparent first-order kinetics of pre-treated POME degradation over 0.2 g L1 of different photocatalysts when irradiated with 100 W of (a) UV source and (b) visible light.
POME under the irradiation of both UV and visible lights. Hence, effects of different photocatalyst loadings to the photocatalytic degradation of a pre-treated POME were further sought. In order to capture the catalytic performance in a wider time spectrum, all the reactions were examined over duration of 8 h. Various loadings viz. 0.2, 0.5, 1.0, 1.5 and 2.0 g L1 of 0.50 wt% Ag/TiO2 were therefore employed for the ensuing investigation. The resulting transient concentration profiles are shown in Fig. 10. It can be observed that the degradation efficiency seems to follow a hyperbolic trend. This Table 5 Kinetics parameters obtained from the simple power law modelling. Photocatalyst
Apparent specific reaction rate, k (103 min1)
Regression coefficient, R2
100 W UV source Blank TiO2 only 0.25 wt% Ag/TiO2 0.50 wt% Ag/TiO2 1.0 wt% Ag/TiO2
0.092 1.045 1.036 3.582 1.502
0.95 0.85 0.85 0.96 0.92
100 W visible light Blank TiO2 only 0.25 wt% Ag/TiO2 0.50 wt% Ag/TiO2 1.0 wt% Ag/TiO2
0.120 0.284 0.730 2.440 0.579
0.98 0.94 0.96 0.97 0.96
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
7
degradation of pre-treated POME judging by the highest conversion recorded, around 85.0% at the 8th hour, over the 1.0 g/L. In contrast, the visible light system only recorded 60.0% conversion at the 8th hour, as the highest conversion attained over the 1.0 g/L photocatalyst, which was however, significantly higher than the pure TiO2. Even at the same photocatalyst loading of 0.2 g/L, Ag/TiO2 photocatalyst yielded a higher conversion compared to the TiO2 only. Once again, the superiority of Ag/TiO2 photocatalyst compared to the pure TiO2 has been underlined for both types of system. 4. Conclusions Titania loaded with Ag was employed as a photocatalyst for photocatalytic degradation of POME, an organic liquid discharge from oil palm mills. XRD characterization showed that the final form of the entire Ag/TiO2 photocatalysts (0.25e1.0 wt% Ag) still has retained its nascent anatase phase with crystal diameter ranged from 42.0 to 45.0 nm. Interestingly, UVeVis spectrum seems to indicate that the Pt inclusion successfully narrowed the band gap to 2.50 eV (for the 0.5 wt% Ag/TiO2) from 3.20 eV as recorded for the pure TiO2 photocatalyst. Moreover, the absorption of light in the visible light spectrum has noticeably-improved whilst almost similar absorption capacity for the UV region was also retained. Photocatalytic degradation of the pre-treated POME showed that Ag/TiO2 has significantly functioned more effectively compared to the pure TiO2 when irradiated by both UV and also visible light sources, respectively. This improvement can be largely credited to the narrower band gap energy and improved visible light absorption. In addition, Ag may have also enhanced the charge separation by rapidly-transferring the e away from the positive hþ charge on the TiO2 surface, thus minimizing charge recombination. It was also determined from this work that the 0.50 wt% Ag/TiO2 yielded the best photocatalytic degradation whilst a loading of 1.0 g/L of photocatalyst was the most preferred. Acknowledgements Fig. 10. Transient conversion profiles of pre-treated POME degradation over a 0.50 wt% Ag-loaded TiO2 at different photocatalyst loadings under the 100 W irradiation of (a) UV light and (b) visible light source.
may indicate that deactivation of photocatalyst has occurred, most likely via an adsorption of intermediate organic species on active sites, consequently the number of active sites dropped with irradiation time. For the case of UV-light irradiation, as in Fig. 10(a), it can be observed that generally, Ag/TiO2 showed a far better facilitation of POME degradation compared to the pure TiO2 at a similar photocatalyst loading viz. 0.2 g/L. In addition, when the photocatalyst loading was increased for Ag/TiO2, the POME photocatalytic degradation was even more profound. This may be because the presence of higher concentration of photocatalyst would produce more OH radical. Consequently, the POME photocatalytic degradation would have increased with the photocatalyst loading. Nonetheless, beyond a certain photocatalyst loading limit, the degradation performance would be negatively-affected, i.e. for Ag/ TiO2 system, the 1.5 and 2.0 g/L loadings showed a markedly lower performance compared to the 1.0 g/L. This can be ascribed to the excessive entry of photocatalyst into the reaction media that has caused the penetration of light being reduced. Consequently, the degradation efficiency has been adversely-affected. For the visible light system (cf. Fig. 10(b)), the same trend was observed. Blank run as well as 0.2 g/L of pure TiO2 yielded the lowest degradation efficiency judging by the concentration profile. Furthermore, the UV source still offered the best light spectrum for the photocatalytic
Authors would like to acknowledge the Science Fund Grant (RDU130501) from the Ministry of Science, Technology and Innovation Malaysia. MRD appreciates the Universiti Malaysia Pahang for a provision of studentship (GRS120355). References Abdul, K., Mohamed, I., Kamil, A., 1989. Biological treatment of palm oil mill effluent using Triychodermu viride. Biol. Waste 27, 143e152. Andrew, D., Lenore, S., Eugene, W., Mary Ann, H., 2005. Standard Methods for the Examination of Water and Wastewater. American Public Health Association. Ao, Y., Xu, J., Zhang, S., Fu, D., 2010. A one-pot method to prepare N-doped titania hollow spheres with high photocatalytic activity under visible light. Appl. Surf. Sci. 256, 2754e2758. Badri, M.A., 1984. Identification of heavy metal toxicity levels in solid wastes by chemical speciation. Conserv. Recycl. 7, 257e269. Bello, M., Nourouzi, M., Abdullah, L.C., Choong, T.S., Koay, Y., Keshani, S., 2013. POME is treated for removal of color from biologically treated POME in fixed bed column: applying wavelet neural network (WNN). J. Hazard. Mater. 262, 106e113. nchez, E., 1996. Anaerobic treatment of palm oil mill effluent Borja, R., Banks, C.J., Sa in a two-stage up-flow anaerobic sludge blanket (UASB) system. J. Biotechnol. 45, 125e135. Chakrabortty, D., Gupta, S.S., 2013. Photo-catalytic decolourisation of toxic dye with N-doped titania: a case study with acid blue 25. J. Environ. Sci. 25, 1034e1043. Chen, Z., Xu, J., Ren, Z., He, Y., Xiao, G., 2013. Low temperature synthesis of ZnIn2S4 microspheres as a visible light photocatalyst for selective oxidation. Catal. Commun. 41, 83e86. Chen, W., Chen, Z., Liu, T., Jia, Z., Liu, X., 2014. Fabrication of highly visible light sensitive graphite-like C3N4 hybridized with Zn0.28Cd0.72S heterojunctions photocatalyst for degradation of organic pollutants. J. Environ. Chem. Eng. 2, 1889e1897. Chin, K., Ng, W., Ma, A., Wong, K., 1987. Treatability studies of palm oil refinery wastewaters. Water. Sci. Technol. 19, 23e29.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104
8
C.K. Cheng et al. / Journal of Cleaner Production xxx (2015) 1e8
Chuong, T.V., Dung, L.Q.T., Khieu, D.Q., 2008. Synthesis of nano titanium dioxide and its application in photocatalysis. J. Kor. Phys. Soc. 52, 1526e1529. Giri, R.R., Ozaki, H., Takanami, R., Taniguchi, S., 2008. A novel use of TiO2 fiber for photocatalytic ozonation of 2, 4-dichlorophenoxyacetic acid in aqueous solution. J. Environ. Sci. 20, 1138e1145. Harsono, S.S., Grundmann, P., Soebronto, S., 2014. Anaerobic treatment of palm oil mill effluents: potential contribution to net energy yield and reduction of greenhouse gas emissions from biodiesel production. J. Clean. Prod. 64, 619e627. Karim, M.I.A., Hie, L.L., 1987. The use of coagulating and polymeric flocculating agents in the treatment of palm oil mill effluent (POME). Biol. Waste 22, 209e218. Kasanen, J., Salstela, J., Suvanto, M., Pakkanen, T.T., 2011. Photocatalytic degradation of methylene blue in water solution by multilayer TiO2 coating on HDPE. Appl. Surf. Sci. 258, 1738e1743. Kongnoo, A., Suksaroj, T., Intharapat, P., Promtong, T., Suksaroj, C., 2012. Decolorization and organic removal from palm oil mill effluent by Fenton's process. Environ. Eng. Sci. 29, 855e859. Liu, W., Gao, J., Zhang, F., Zhang, G., 2007. Preparation of TiO2 nanotubes and their photocatalytic properties in degradation methylcyclohexane. Mater. Trans. 48, 2464. Ma, A., Ong, A., 1986. Palm oil processingdnew development in effluent treatment. Water. Sci. Technol. 18, 35e40. Ma, R., Bando, Y., Sasaki, T., 2003. Nanotubes of lepidocrocite titanates. Chem. Phys. Lett. 380, 577e582. Mahmoodi, N.M., 2011. Photocatalytic ozonation of dyes using copper ferrite nanoparticle prepared by co-precipitation method. Desalination 279, 332e337. Ng, W., Goh, A.C., Tay, J., 1988. Palm oil mill effluent treatmentdliquidesolid separation with dissolved air flotation. Biol. Waste 25, 257e268. Othman, M.R., Hassan, M.A., Shirai, Y., Baharuddin, A.S., Ali, A.M.M., Idris, J., 2014. Treatment of effluents from palm oil mill process to achieve river water quality for reuse as recycled water in a zero emission system. J. Clean. Prod. 67, 58e61. Quesada-Cabrera, R., Mills, A., O'Rourke, C., 2014. Action spectra of P25 TiO2 and a visible light absorbing, carbon-modified titania in the photocatalytic degradation of stearic acid. Appl. Catal. B-Environ. 150, 338e344.
Ramchiary, A., Samdarshi, S., 2014. Ag deposited mixed phase titania visible light photocatalyst e superiority of Agetitania and mixed phase titania co-junction. Appl. Surf. Sci. 305, 33e39. Roge, W., Velayuthan, A., 1981. Preliminary trials with Westfalia-3-phase decanters for palm oil separation. In: Palm Oil Prod. Technol. Eighties, Rep. Proc. Int. Conf., Inc. Sot. Plant. Kuala Lumpur, Malaysia, pp. 327e334. Santosa, S.J., 2008. Palm oil boom in Indonesia: from plantation to downstream products and biodiesel. Clean Soil Air Water 36, 453e465. Shahid, M., Jingling, L., Ali, Z., Shakir, I., Warsi, M.F., Parveen, R., Nadeem, M., 2013. Photocatalytic degradation of methylene blue on magnetically separable MgFe2O4 under visible light irradiation. Mater. Chem. Phys. 139, 566e571. Shen, Y., Wu, Y., Li, X., Zhao, Q., Hou, Y., 2013. One-pot synthesis of MgFe2O4 nanospheres by solvothermal method. Mater. Lett. 96, 85e88. Spanjers, H., Lier, J., 2006. Instrumentation in anaerobic treatment research and practice. Water. Sci. Technol. 53, 63e76. Sri, T., Dompok, B., 2011. Palm Oil and Rubber. Sun, X., Li, Y., 2003. Synthesis and characterization of ion-exchangeable titanate nanotubes. Chem. Eur. J. 9, 2229e2238. Tabassum, S., Zhang, Y., Zhang, Z., 2015. An integrated method for palm oil mill effluent (POME) treatment for achieving zero liquid discharge e a pilot study. J. Clean. Prod. 95, 148e155. Tsai, C.-C., Teng, H., 2004. Regulation of the physical characteristics of titania nanotube aggregates synthesized from hydrothermal treatment. Chem. Mater. 16, 4352e4358. Virkutyte, J., Jegatheesan, V., Varma, R.S., 2012. Visible light activated TiO2/microcrystalline cellulose nanocatalyst to destroy organic contaminants in water. Bioresour. Technol. 113, 288e293. Wang, J., Li, C., Zhuang, H., Zhang, J., 2013. Photocatalytic degradation of methylene blue and inactivation of Gram-negative bacteria by TiO2 nanoparticles in aqueous suspension. Food Control 34, 372e377. Zalas, M., 2014. Synthesis of N-doped template-free mesoporous titania for visible light photocatalytic applications. Catal. Today 230, 91e96.
Please cite this article in press as: Cheng, C.K., et al., Preparation of titania doped argentum photocatalyst and its photoactivity towards palm oil mill effluent degradation, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.104