TREATMENT OF PETROLEUM REFINERY WASTEWATER

Abstract

Petroleum industries transform crude oil into useful products such as kerosene, gasoline, and other petrochemical feedstocks. This process consumes large amount of water, consequently significant amount of wastewater gets generated during processing of crude oil in refineries. Wastewater generated in refinery process is about 0.4-1.6 times the crude oil processed [Coelho et al., 2006]. Petroleum refinery wastewater (PRW) contains phenolic compounds (such as phenol, resorcinol, catechol, etc.), oil and grease, and other toxic compounds which can cause serious environmental problem if discharged to various aquatic bodies. Refineries follow traditional treatment technologies that consist of different mechanical, physico-chemical and biological treatment methods. Refractory toxic compounds present in PRW reduce the efficiency of biological treatment processes. Among physico-chemical methods, adsorption as a wastewater treatment process has aroused considerable interest during recent years. Granulated activated carbon is generally used as a commercial adsorbent. However, because of its high cost, industrial and agricultural wastes such as rice husk ash (RHA), bagasse fly ash, etc. have gained a lot of attention as potential low-cost adsorbents [Kumar et al., 1987; Sharma et al., 2010; Alam et al., 2007]. However, only few studies [McKay and Al- Duri, 1991; Allen et al., 2004; Suresh et al., 2012a,b] have been reported in the open literature for the multi-component adsorption of phenolic compounds on various types of adsorbents. Since, actual effluents generated in industries such as petroleum refineries contain several phenolic compounds; therefore, equilibrium adsorption data iii for closely related binary and ternary compounds is of utmost importance for the design of adsorption systems. Sequential batch reactor (SBR) is a biological treatment method which has gained more importance as compared to other biological treatment methods because of its inherent flexibility in treatment time. SBR treatment system consists of five operations in sequence: fill phase in which reactor is fed with wastewater; react phase in which the organic matter degradation occurs; settle phase in which biomass is allowed to settle; draw phase in which wastewater is withdrawn from reactor to retain the biomass and idle phase [Chan and Lim, 2007; Thakur et al., 2013a]. Survey of the open literature for treatment of petroleum wastewater by SBR [Viero et al., 2008; Shariati et al., 2011; Kutty et al., 2011] shows that previous researchers did not use fill time as an operating parameter, and kinetic study was not performed in any of the previous studies for the treatment of PRW in SBR. Recently electrochemical treatment using various types of electrodes has been used for the treatment of various type of industrial wastewaters [Sengil et al., 2009; Verma et al., 2013]. Only a few studies are available in open literature for the electrochemical treatment of PRW. Yavuz et al. [2010] performed electrochemical oxidation of PRW using boron doped diamond anode, ruthenium mixed metal oxide electrode, and electrofenton and electrocoagulation using iron electrode. Yan et al. [2011] treated PRW using electrochemical process with three dimensional multi-phase porous graphite plate electrodes. Thus, the most common iron/stainless steel (SS) electrodes have not been used earlier for the electrocoagulation (EC) treatment of PRW. More over in previous studies, the sludge was not characterized from disposal point of view. iv The present study has been undertaken with endeavor to study treatment of wastewater containing phenolic compounds and actual PRW by following methods: 1. Individual and simultaneous adsorptive removal of phenol, catechol and resorcinol from aqueous solution by RHA. 2. Treatment of synthetic phenolic wastewater containing phenol, catechol and resorcinol by SBR. 3. Treatment of actual PRW by SBR. 4. Treatment of actual PRW by EC method using SS as an electrode. Adsorptive removal of phenolic compounds like phenol, catechol and resorcinol was studied by means of adsorption onto low-cost RHA [Thakur et al., 2013b,c]. Physico-chemical characterization including surface area, X-ray diffraction analysis, scanning electron microscopy (SEM), and Fourier transform infrared spectroscopy (FTIR) of the RHA before and after adsorption have been done to understand the adsorption mechanism. Effect of adsorbent dose and contact time was studied out by varying the dosages in the range of 2-20 g/L at natural pH. All the batch experiments were carried out at 30±1ºC. Optimum adsorbent dose for all the compounds was found to be 20 g/L. Contact time for adsorption equilibrium was found to be 16 h. The pseudosecond- order kinetics represented the adsorption process well for all the adsorbates. Redlich–Peterson isotherm model was found to best-represent the individual equilibrium data for all the three compounds by RHA. The adsorption of phenol, catechol and resorcinol from the binary and ternary solutions onto RHA has been generally found to be antagonistic in nature. Equilibrium isotherms for the binary and ternary adsorption have been analyzed by using non-modified Langmuir, modified Langmuir, extended- Langmuir, extended-Freundlich and Sheindorf–Rebuhn–Sheintuch models. Extended v Langmuir model which assumes overlapping of sites for different adsorbate better represented the binary isotherm data for all three phenol-catechol, catechol-resorcinol and phenol-resorcinol systems and the ternary system data. Treatment of synthetic wastewater containing phenol, resorcinol and catechol was studied in a SBR [Thakur et al., 2013a]. Activated sludge was collected from sewage treatment plant located in Rishikesh, Uttarakhand, India. The sludge was first screened for the removal of coarse and bigger particles and then it was aerated for 1-2 d and acclimatized for treatment of phenol containing wastewater. The reactor was composed of Plexi glass, with the dimension of 7.37 cm×40.64 cm (radius × height) and having 5 litre working volume. Aeration in the SBR was achieved by an aquarium-type air pump with sintered-sand diffusers at the bottom of the reactor. Mixed liquor suspended solid (MLSS) concentration was controlled between 1200 and 2200 mg/L with a sludge age of ≈20 d. Excess sludge, which grew during the aeration stage, was drawn out at the end of every operating cycle, in order to maintain proper MLSS concentration. The reactors were operated in an isothermal chamber with temperature at 30±1oC. Parameters such as hydraulic retention time (HRT) and filling time have been optimized to increase the phenol, resorcinol, catechol and chemical oxygen demand (COD) removal efficiencies. The optimum HRT value was found to be 1.25 d whereas optimum fill time was found to be 1.5 h. More than 99% phenol, 95% resorcinol and 96% catechol and 89% COD removal efficiencies have been obtained at optimum conditions. EDAX analysis shows the increased carbon content and utilization of nutrients in the sludge after the treatment of wastewater in the SBR. The heating value of the activated sludge was found to be 12.09 MJ/kg. The filtered sludge can be dried and fired as fuel in the furnaces/incinerators for its heat recovery. The bottom ash can vi be used for blending with organic manure for its use in agriculture/horticulture or can be blended with clay/coal fly ash for use in making bricks/ceramic tiles for the building industry [Thakur et al., 2013a]. SBR was also used for the treatment of actual PRW in terms of COD and total organic carbon (TOC) removal. Actual PRW was collected from a nearby petroleum refinery and was characterized for its various physico-chemical characteristics. In this study, reactor was operated on a fill-and-draw basis, with a cycle time of 8 h, while, settle time, decant time and idle time were 1, 0.5 and 0.5 h, respectively. React time was varied according to the fill phases strategies to degrade the carbonaceous materials and nitrogen present in PRW. For this, two phases study was carried out. In the phase-I of SBR operation, instantaneous filling strategy was implemented by varying the volume exchange ratio (VER) and HRT in the range of 0.10-0.60 d and 0.56-3.33 d, respectively. Aeration was given during the react phase only. Maximum COD and TOC removal of 77% and 79% was observed at HRT=0.83 d. Further experiments were performed by varying the fill time from 0.5-2 h at HRT=0.83 d. Maximum COD and TOC removal of 79.7% and 83.5%, respectively, was achieved when the fill time was 2 h. FTIR and UV-visible analysis were done so as to understand the degradation mechanism [Thakur et al., 2013e]. Performance of a batch EC treatment of actual PRW was studied using SS electrode [Thakur et al., 2013d]. A rectangular perspex glass lab-scale batch reactor having a working volume of 1.5 l (110 mm×110 mm×125 mm) was used for the EC experiments. Four SS electrodes of 3 mm thickness and 80 mm×80 mm dimension were used in the present study. The inter-electrode gap was varied in the range of 0.5- 2.5 cm. A digital direct current power supply (0–20 V, 0–5 A) was used to maintain vii constant current in the range of 1-5 A during the run. Solution was agitated with magnetic stirrer during the experimental run so as to achieve proper mixing. Full factorial central composite (CCD) design was used to study the effect of four key process parameters on the reduction of COD and TOC. The parameters used in this study are initial pH (pHo): 2–10; current density (j): 39.06–195.31 A/m2; inter-electrode distance (g): 1–2.5 cm and electrolysis time (t): 30–150 min. Quadratic models were developed in terms of input parameters such as pHo, j, g and t by carrying out experiments as per the CCD design. These quadratic models were further used for determining parametric condition for maximum COD and TOC removal. COD and TOC removal efficiencies of 61% and 51.6%, respectively, were obtained at optimized operating condition of pHo=6.0, j=182 A/m2, g=1.5 cm and t=145 min. Mechanism of EC treatment was studied by carrying out zeta potential measurement, UV-visible and FTIR analysis of PRW before and after treatment. Overall treatment of PRW was because of the charge neutralization of colloids, adsorption onto generated iron hydroxides and by degradation of the organic pollutants by oxidizing agents generated in situ during EC treatment. Sludge generated has also been characterized by FTIR, SEM, energy dispersive X-ray analysis and thermo-gravimetric analysis so as to evaluate its disposal aspect [Thakur et al., 2013d]. For treatment of synthetic wastewater containing phenolic compounds like phenol, catechol and resorcinol, SBR was found to show better treatment efficiency as compared to adsorption with RHA. Comparison of various treatment methods for actual PRW shows that treatment in SBR removes highest amount of COD, however, it requires higher treatment time as compared to EC treatment