Phenol wastewater remediation: advanced oxidation processes coupled to a biological treatment

A. Rubalcaba*, M.E. Suarez-Ojeda*, F. Stuber*, A. Fortuny**, C. Bengoa*, I. Metcalfe***, J. Font*, J. Carrera**** and A. Fabregat*
*Departament dEnginyeria Quimica, Universitat Rovira i Virgili, Tarragona, Spain ` **Departament dEnginyeria Quimica, Universitat Politecnica de Catalunya, Vilanova i la Geltru, Barcelona, Spain ***School of Chemical Engineering and Advanced Materials Department, University of Newcastle upon Tyne, Newcastle upon Tyne NE1 7RU, UK ` ****Departament dEnginyeria Quimica, Universitat Autonoma de Barcelona, Barcelona, Spain Abstract Nowadays, there are increasingly stringent regulations requiring more and more treatment of industrial efuents to generate product waters which could be easily reused or disposed of to the environment without any harmful effects. Therefore, different advanced oxidation processes were investigated as suitable precursors for the biological treatment of industrial efuents containing phenol. Wet air oxidation and Fenton process were tested batch wise, while catalytic wet air oxidation and H2O2promoted catalytic wet air oxidation processes were studied in a trickle bed reactor, the last two using over activated carbon as catalyst. Efuent characterisation was made by means of substrate conversion (using high liquid performance chromatography), chemical oxygen demand and total organic carbon. Biodegradation parameters (i.e. maximum oxygen uptake rate and oxygen consumption) were obtained from respirometric tests using activated sludge from an urban biological wastewater treatment plant (WWTP). The main goal was to nd the proper conditions in terms of biodegradability enhancement, so that these phenolic efuents could be successfully treated in an urban biological WWTP. Results show promising research ways for the development of efcient coupled processes for the treatment of wastewater containing toxic or biologically non-degradable compounds. Keywords Activated carbon; Fenton process; hydrogen peroxide; phenol; WAO

Water Science & Technology Vol 55 No 12 pp 221227 Q IWA Publishing 2007

Introduction

Traditionally, the attention received by biological treatment far exceeds that of other remediation processes, in part due to its low cost and versatility. However, the presence of toxic or inhibitory chemicals dramatically reduces the viability of biological processes. In this case, advanced oxidation processes (AOPs) can be a useful option (Esplugas et al., 2002) to prepare the efuents before the biological remediation. The objective of this work is to nd a suitable oxidation technology among wet air oxidation (WAO), catalytic wet air oxidation (CWAO), H2O2-promoted CWAO (H2O2-CWAO) the last two over activated carbon (AC) and Fenton process in order to design an integrated chemical biological system for phenolic wastewater treatment. To follow this idea, the knowledge of the physical, chemical and biological properties of the oxidation intermediates and the extension of phenol degradation in each oxidation process are key points (Scott and Ollis, 1995). AC was used as catalyst since its low cost and high adsorption properties make it a suitable material for wastewater remediation through chemical oxidation processes (Stuber et al., 2005). Respirometric techniques have been applied to assess the biodegradability, toxicity or inhibitory effect of industrial wastewaters since they allow evaluation of the behaviour of an activated sludge against different substrates (Guisasola et al.,
doi: 10.2166/wst.2007.412
221

2003). Phenol has been chosen as model pollutant because it is one of the most prevalent forms of toxic and weakly biodegradable chemical pollutants from industrial activities.
Methods
Materials

Aldrich provided analytical grade phenol. Deionised water was used to prepare all the solutions. Feed concentrations were taken as 5 g L21 (11.9 g COD L21) of phenol. To prepare high performance liquid chromatography (HPLC) mobile phases, HPLC grade methanol (Aldrich) and ultra-pure water (Millipore Direct-Q system) were used. The synthetic air used as oxidant in CWAO and H2O2-CWAO experiments was 99.995% pure (Carburos Metalicos, Spain). Hydrogen peroxide used in the Fenton and H2O2-CWAO processes was bought from Panreac as a 30% w/v aqueous solution. Ferrous sulphate heptahydrate from Panreac was used as homogeneous catalyst in the Fenton experiments. The AC used as catalyst in CWAO and H2O2-CWAO processes was supplied by Merck (reference #102518) in the form of 2.5 mm pellets. Some characteristics of this AC, prior to use, can be found elsewhere (Suarez-Ojeda et al., 2005).
Experimental set-up and procedure

A. Rubalcaba et al.
222

WAO and Fenton processes were tested batch wise. Both batch experiments were conducted for 1 h. WAO tests were done at 215, 240 and 265 8C using 2 bar of oxygen partial pressure (PO2). Fenton oxidation was studied at 30, 50 and 85 8C, using the stoichiometric amount of H2O2 for phenol mineralisation and 500 mg L21 of Fe2 as homogeneous catalyst. CWAO and H2O2-CWAO experiments were performed in a trickle bed reactor for 72 h in continuous mode. The air ow rate was adjusted to guarantee excess oxygen and the liquid hourly space velocity was calculated according to the weight of the catalytic bed to give 8.2 h21. Total operating pressure was selected to provide a PO2 of 2 bar while temperature was set to 140 and 160 8C. For the H2O2-CWAO tests, 20% of the stoichiometric amount of H2O2 needed for phenol mineralisation was directly added in the feed tank. Details of WAO and CWAO reaction systems can be found elsewhere (Abu-Hassan et al., 2005; Suarez-Ojeda et al., 2005). Efuent characterisation was done by HPLC using a method previously developed (Suarez-Ojeda et al., in preparation). According to the oxidation pathway proposed in the literature for phenol (Devlin and Harris, 1984; Joglekar et al., 1991; De et al., 1999; Alnaizy and Akgerman, 2000), a set of proposed reaction intermediates were calibrated and analysed by HPLC. These intermediates were: glyoxilic acid, oxalic acid, formic acid, malonic acid, acetic acid, maleic acid (and its isomer fumaric acid), succinic acid, acrylic acid, propionic acid and muconic acid (both stereoisomers), hydroquinone, catechol and p-benzoquinone. In addition, 4-hydroxybenzoic acid (4-HB), its isomer salicylic acid (2-HB) and resorcinol were also calibrated. All calibration curves were performed by means of external standards. Chemical oxygen demand (COD) and total organic carbon (TOC) were done following the standard methods 5220D and 5310B, respectively (AWA, 1999). Hydrogen peroxide concentration in samples from Fenton and H2O2-CWAO experiments was measured by iodometric titration. Biodegradation parameters of the AOPs efuents were determined by respirometric tests (Guisasola et al., 2003; Suarez-Ojeda et al., 2007). For these tests, an LFS respirometer (Spanjers et al., 1989) in which the oxygen dissolved in the liquid phase (So) is monitored was used. When a substrate pulse is added, microorganisms consume this So while metabolising the substrate. Then, the oxygen balance in the liquid phase is solved and the oxygen uptake rate (OUR) and the oxygen consumption (OC) are obtained as indicators

of the biodegradability, toxicity or inhibitory effect of the studied substrate. More detailed information about the OUR and the OC calculation can be found in the literature (Spanjers et al., 1989; Guisasola et al., 2003). Each respirometric test was done with a pulse of 20 mg COD L21 from an AOPs efuent. The respirometer temperature was xed at 30 ^ 0.5 8C using a thermostatic bath, the biomass had an average concentration of volatile suspended solids (SSV) of 2,100 ^ 320 mg VSS L21 and the pH was kept at 7.5 ^ 0.5. The activated sludge used in the respirometric tests came from an urban biological WWTP in Tarragona (Catalonia, Spain) and was not adapted to metabolise phenolic compounds.
Results

A. Rubalcaba et al.

The results and discussion are divided into two sections. In the rst one, the performance of each AOP is discussed by means of the phenol conversion (Xphenol), COD removal (XCOD) and TOC abatement (XTOC). Also, the main detected intermediates in each process are examined. In the second section, the determination of the efuents biodegradability parameters is presented.
Phenol removal: Comparison of the selected treatments

The efuent characteristics at the end of the WAO and Fenton experiments are given in Table 1, whereas Figure 1 shows the results from CWAO and the H2O2-CWAO tests, at 2 bar of PO2 and two different temperatures, 140 and 160 8C. For WAO and Fenton experiments (Table 1), it can be observed that WAO, even using higher temperatures than Fenton process, did not reach adequate oxidation. On the contrary, Fenton experiments at low or even close to ambient temperature, led to nearly complete mineralisation. However, it has to be pointed out that H2O2 was used in high amounts and also this process would need a post-treatment in order to remove the homogeneous catalyst from the efuent, both factors increasing the operational costs. In the case of WAO samples, the main partial oxidation products were light carboxylic acids such as oxalic, acetic and succinic acids in concentrations between 80 and 665 mg L21. Malonic, maleic, fumaric (isomer of maleic acid), 4-HB and 2-HB acids were detected in amounts higher than 5 mg L21 but below 100 mg L21. Finally, hydroquinone and catechol were detected in concentrations up to 275 mg L21 whereas p-benzoquinone was detected in concentration as low as 6 mg L21. No other possible intermediates were detected. The presence of 4-HB and 2-HB suggest a reaction pathway where either phenol or dihydric-phenols (hydroquinone and catechol) can be oxidised via two routes, the rst and dominating pathway leads to the formation of benzoquinones and more oxidised products. The second route should proceed through the reaction with malonic acid to give 4-HB, 2-HB and acetic acid. To the best of our knowledge only two previous works reported the presence of 4-HB as reaction intermediate but, in CWAO of phenol with AC as catalyst (Stuber et al., 2005; Santos et al., 2006).
Table 1 Phenol wastewater treatment through 1 h WAO and Fenton process
Process T (8C) Xphenol (%) XCOD (%) XTOC (%)

WAO

Fenton

215 240 260 30 50 85

18 21 25 100 100 100

11 11 14 86 91 92

10 15 16 nm nm nm
223

nm: not measured

A. Rubalcaba et al.
224

Figure 1 CWAO (solid symbols) and H2O2-CWAO (open symbols) at (a) 140 8C and (b) 160 8C. Symbols: (V) phenol conversion, (O) COD conversion, (B) TOC conversion. Lines only show trends

TOC data for Fenton experiments is not available. However, due to the high COD elimination it could be inferred that nearly total mineralisation was achieved. Moreover, the HPLC analyses showed that the main oxidation product was oxalic acid and that catechol and hydroquinone were found only in concentrations lower than 2 mg L21. In the case of CWAO and H2O2-CWAO tests (Figure 1), three different zones related to the use of AC as catalyst could be distinguished in all the cases, as previously found (Suarez-Ojeda et al., 2005). Firstly, an adsorption-dominating period resulted in an apparent total conversion. Secondly, the equilibrated zone progresses throughout the catalytic bed as the run proceeds, so that a transient period was observed, which resulted in a rapid conversion fall. Finally, conversion almost attained steady state. Since the amount adsorbed over an adsorbent decreases as temperature increases (Do Duong, 1998), the adsorption-dominating period for 160 8C tests is longer than for 140 8C tests. For example, the increase in the adsorption-dominating period in the CWAO at 160 8C compared to the CWAO at 140 8C was up to 4 h. Other simultaneous effects as oxidation of phenol, oxidative coupling of phenol or AC surface modication can also account for the adsorption-dominating period. As could be expected, a temperature rise from 140 to 160 8C led to a rise in the conversion. This effect was enhanced when H2O2 was used as an oxidation promoter, which makes easy to expect also a higher biodegradability enhancement. Regarding the reaction intermediates, once quantied, the theoretical oxygen demand of each one was calculated using the empirical correlations obtained by Baker et al. (1999). Partial oxidation products were grouped into remaining phenol, carboxylic acids, quinone-like products, condensation products (4-HB and 2-HB) and non-identied products. Later, the contribution of each group was compared to the total experimental COD obtained in each CWAO and H2O2-CWAO test. Figure 2 summarises these results. From Figure 2 in CWAO tests, it can be observed that the carboxylic acids fraction increases with the increase in temperature. In the process using H2O2, although having a higher fraction of non-identied compounds than CWAO tests, general trends show how using higher temperatures lead to an efuent with higher concentration of carboxylic acids and less quinone-like products.

Biodegradability enhancement

Once the biodegradability parameters (OUR and OC) are obtained, the fraction of readily biodegradable COD (%CODRB) can be calculated (Suarez-Ojeda et al., 2007). The total biodegradable COD is the sum of the readily and the slowly (CODS) biodegradable fractions, therefore CODRB is always lower than the total biodegradable COD. The determination of CODS requires different experiments that are extremely time-consuming;

A. Rubalcaba et al.

Figure 2 Distribution of oxidation products in the CWAO and H2O2-CWAO tests

therefore the CODRB can be used as a fast method to compare the biodegradability enhancement reached from the different oxidation treatments with respect to the initial phenol solution, which has 0% of CODRB. Figure 3 shows the % CODRB obtained from the respirometric tests performed to each efuent coming from WAO, CWAO and H2O2-CWAO experiments. Due to the high phenol and COD removal in the Fenton process, it was decided not to perform the respirometric tests. However, based on the main intermediates found, mainly oxalic acid, a % CODRB estimation has been made and included in Figure 3. As can be seen in Figure 3, for WAO and Fenton efuents, the %CODRB (which is related to the generation and intermediates distribution) is very low (, 4%), whatever the temperature applied in the treatment. However, it has to be taken into account that these apparently similar results are a consequence of different effects. In WAO tests, the temperature, pressure and especially the residence time used were not sufcient to oxidise all initial phenol and quinone-like intermediates (the highest XCOD obtained was only 14%) to carboxylic acids and eventually CO2 and water. On the contrary, in Fenton efuents, the oxidation progressed so deeply (XCOD was always close to 90%) that there were not enough biodegradable intermediates or substrate as to be consumed by the microorganisms in the respirometer. Therefore, WAO efuents at the conditions tested were not suitable for a subsequent biological degradation due to their high concentration of poorly biodegradable compounds and the Fenton process was not a good choice to make a coupling with a urban biological WWTP, because its high removal of organic matter makes a following treatment unnecessary.

Figure 3 CODRB obtained from respirometric tests

225

In contrast, CWAO and H2O2-CWAO experiments gave better results in terms of biodegradability. From respirometric results, the %CODRB was between 5 and 36%, being the main readily biodegradable intermediates carboxylic acids (fumaric, propionic, acetic, succinic, formic and glyoxylic acids) and fewer quinone-like compounds.

Conclusions

A. Rubalcaba et al.
226

WAO, CWAO, H2O2-CWAO (last two over AC as catalyst) and Fenton processes have been tested in order to nd a suitable technology among them for an integrated chemical biological system for phenolic wastewater treatment. Biodegradability enhancement with WAO and Fenton process was below 4%, because Fenton process yields total mineralisation of phenol and reaction intermediates, making a biological post-treatment unnecessary whereas the selected conditions for WAO were not enough to oxidise all initial phenol and quinone-like intermediates to more biodegradable intermediates. In contrast, CWAO and H2O2-CWAO experiments gave better results in terms of biodegradability, the readily biodegradable fraction of these efuents being between 5 and 36%. Ongoing research is focused on optimising the oxidation conditions among WAO, CWAO, H2O2-CWAO and Fenton process for other model compounds in order to achieve similar biodegradability enhancements to those obtained in this study for phenol wastewater. Also, efforts will be focused on establishing the kinetic model of phenol oxidation, including both the intermediates and biodegradability enhancement prediction.

References
Abu-Hassan, M.A., Mantzavinos, D. and Metcalfe, I.S. (2005). Wet air oxidation and ultrasound for the removal of linear alkylbenzene sulfonates from wastewater: the benecial role of catalysis. Top. Catal., 33, 141 148. Alnaizy, R. and Akgerman, A. (2000). Advanced oxidation of phenolic compounds. Adv. Environ. Res., 4, 233 244. American Water Association (1999). Standard Methods for the Examination of Water and Wastewater, Washington American Public Health Association, Washington. Baker, J., Milke, M.W. and Mihelcic, J.R. (1999). Relationship between chemical and theoretical oxygen demand for specic classes of organic chemicals. Water Res., 33, 327 334. Devlin, H.R. and Harris, I.J. (1984). Ind. Eng. Chem. Fundam., 23, 387. De, A.K., Chaudhuri, B. and Bhattacharjee, S. (1999). A kinetic study of the oxidation of phenol, o-chlorophenol and catechol by hydrogen peroxide between 298K and 333K: The effect of pH, temperature and ratio of oxidant to substrate. J. Chem. Technol. Biotechnol., 74, 162 168. Do Duong, D. (1998). Adsorption Analysis: Equilibria and Kinetics, Imperial College Press, London, Chapter 2. Esplugas, S., Gimenez, J., Contreras, S., Pascual, E. and Rodrguez, M. (2002). Comparison of different advanced oxidation processes for phenol degradation. Water Res., 36, 1034 1042. Guisasola, A., Baeza, J.A., Carrera, J., Casas, C. and Lafuente, J. (2003). An off-line respirometric procedure to determine inhibition and toxicity of biodegradable compounds in biomass from an industrial WWTP. Water Sci. Technol., 48(11), 267 275. Joglekar, H.S., Samant, S.D. and Joshi, J.B. (1991). Kinetics of wet air oxidation of phenol and substituted phenols. Water Res., 25, 135 145. Santos, A., Yustos, P., Gomis, S., Ruiz, G. and Garcia-Ochoa, F. (2006). Reaction network and kinetic modeling of wet oxidation of phenol catalyzed by activated carbon. Chem. Eng. Sci., 61, 2457 2467. Scott, J.P. and Ollis, D.F. (1995). Integration of chemical and biological oxidation processes for water treatment: review and recommendations. Environ. Prog., 14, 88 103. Spanjers, H., Vanrolleghem, P.A., Olsson, G., Dold, P.L. (1998). Respirometry in Control of the Activated Sludge Process (Scientic and Technical Report No 7), IAWQ, London. Stuber, F., Font, J., Fortuny, A., Bengoa, C., Eftaxias, A. and Fabregat, A. (2005). Carbon materials and catalytic wet air oxidation of organic pollutants in wastewater. Top. Catal., 33, 3 50.

 Suarez-Ojeda, M.E., Fabregat, A., Stuber, F., Fortuny, A., Carrera, J. and Font, J. (In preparation). CWAO of substituted phenols using activated carbon as catalyst: effect of temperature and pressure. Suarez-Ojeda, M.E., Guisasola, A., Baeza, J.A., Fabregat, A., Stuber, F., Fortuny, A., Font, J. and Carrera, J. (2007). Integrated catalytic wet air oxidation and aerobic biological treatment in a municipal WWTP of high-strength o-cresol wastewater. Chemosphere, 66, 2096 2105. Suarez-Ojeda, M.E., Stuber, F., Fortuny, A., Fabregat, A., Carrera, J. and Font, J. (2005). Catalytic wet air oxidation of substituted phenols using activated carbon as catalyst. Appl. Catal. B, 58, 105114.

227

A. Rubalcaba et al.