Clean Technologies and Environmental Policy

, Volume 6, Issue 4, pp 288–295

Petrochemical wastewater treatment by means of clean electrochemical technologies


    • Gebze Institute of Technology
  • H. Y. Akbulut
    • Gebze Institute of Technology
  • F. Cihan
    • Gebze Institute of Technology
  • M. Karpuzcu
    • Gebze Institute of Technology
Original Paper

DOI: 10.1007/s10098-004-0248-9

Cite this article as:
Dimoglo, A., Akbulut, H.Y., Cihan, F. et al. Clean Techn Environ Policy (2004) 6: 288. doi:10.1007/s10098-004-0248-9


The removal of chemical oxygen demand (COD), turbidity, phenol, hydrocarbon and grease from petrochemical wastewater (PCWW) was experimentally done by using electroflotation (EF) and electrocoagulation (EC). In the EF unit, a graphite anode and a stainless steel mesh as cathode were used. In the EC unit, iron and aluminium were used simultaneously as materials for two blocks of alternating electrodes. The reactor voltage was 12 V, current density (CD) was varied from 5 to 15 mA cm−2, and the residence time varied in the limits of 2–20 min for EF and 1–10 min for EC. The results have shown that EC removes the mentioned contaminants from PCWW more effectively than EF. Turbidity removal in the process of PCWW purification was estimated as 83% for EF and 88% for EC. The yields of phenol, hydrocarbon and grease removal by EC were examined under different values of residence time, CD, and with iron and aluminium as materials for electrodes.


Petrochemical wastewaterElectroflotationElectrocoagulationContaminants removalAl/Fe electrode


The problem of petrochemical wastewater (PCWW) purification is a challenging issue (see Wise and Fahrenthold 1981, Wong 2000, Sponza 2003 and references cited therein). The authors are exploiting different physicochemical, mechanical, and biological approaches along with other contemporary methods of purification.

The traditional treatment of effluents in refineries is based on the mechanical and physicochemical methods such as oil–water separation and coagulation, followed by biological treatment within the integrated activated sludge treatment plant (Wong 2000). Some physicochemical methods used in the process of the petrochemical manufacturing wastewater purification are mentioned below.

One of the papers considers the wastewater disposal from dimethylformamide solvent regeneration tower in a butadiene extraction plant of a petrochemical manufacturing (Kardasz et al. 1999). After wet peroxide oxidation, more than 70% of organic contaminants were removed, and, as the result, the odourless wastewater could be discharged to the sewage system of the industrial plant. To treat highly concentrated toxic wastewater from the chemical and petrochemical industries, an observer-based time-optimal control strategy was implemented on a sequencing batch reactor (SBR) by the use of respirometric techniques (Vargas et al. 1999).

The PCWW was also treated with coagulants (alum, ferric chloride, ferrous sulfate and lime) and with some clays in order to see their effects in clarifying the wastewater before its bio-purification (Demirci at al. 1998). A novel electro-Fenton method was developed to treat high strength hexamine- and petrochemical-containing wastewater (Chou et al. 1999; Huang et al. 1999). Under this approach, a ferrous ion produced at the anode is the catalyst for H2O2. The approach was applied to the bio-effluent of the petrochemical manufacturing wastewater treatment.

A new treatability test based on a direct far UV photo-oxidation of the sample and coupled with an UV spectrophotometric survey of the waste’s quality (Castillo et al. 1999) has been applied to different samples from chemical and petrochemical industries. The results were compared to those obtained with classical tests, for example, biodegradation tests using either air or pure oxygen. For most of the samples, quite good correlation was observed between the photo-oxidation test and biodegradation. The UV treatability test is currently being used for checking up on petrochemical wastewater and chemical sewage.

The potential advantages of the UV/H2O2 process becomes evident when the process is used as the direct pre-treatment of petrochemical wastewater (Juang et al. 1997). The results of the direct pre-treatment with the UV/H2O2 process revealed that the recalcitrant compounds found in raw wastewater could be destroyed to small molecules and might reduce some degree of activity inhibition to bioculture. In the detoxified investigation of spiking aromatic compounds, this process could obtain good removal efficiency, including high level of COD removal and the complete detoxification. The treatment of petrochemical wastewater from oil industry drilling by using a coalescer, an advanced oxidation process (UV/ozone), and a membrane treatment technology based on the use of fractionator were studied by Patino (1999).

There is a tendency to design biological units for chemical or petrochemical wastewater treatment. Some of the treatment plants are even used sometimes for the degradation of any external industrial sewage transported for this purpose. Here, the decision to accept or refuse the wastes must be rapid and sure. A conducted batch and continuous upflow fixed bio-film reactor was used for the biodegradation of some organic compounds in PCWW (Acuna-Askar et al. 1999). The PCWW was effectively biodegraded under oxidizing environmental conditions in the presence of an acclimated mixed culture isolated from a petrochemical bio-treater.

A methanogenic consortium (anaerobic process) was used to degrade phenol and ortho-cresol from a specific effluent of a petrochemical plant in a continuous fixed-film anaerobic reactor (Charest et al. 1999). The application of laccase and manganese-dependant peroxidase from Trametes versicolor to facilitate removal of aromatic hydrocarbons from a petrochemical industrial effluent was investigated by Edwards et al. (2002). The system was applied to an industrial petrochemical-based effluent and compared with the synthetic make-up effluent in terms of “defouling” efficiency. High concentrations of fluoride at the membrane interface during the ultra-filtration of the petrochemical-based effluent contribute significantly to the inhibition of the immobilized enzyme suite.

The efficiency of purifying wastewater from suspended contaminants and oil products in that number is influenced by both dispersion state of the environment and the choice of an appropriate method. To remove colloidal particles from wastewaters, more effective new technologies are to be used in addition to present treatment technologies. One of these technologies is electrochemical treatment of water (Rajeshwar and Ibanez 1997). It must be emphasized that recently the volume of research related to the application of electrochemical methods to environment protection has greatly increased (Müller 1992; Sequeira 1996; Yousuf et al. 2001; Mills 2000; Jiang et al. 2002). The researchers consider the treatment of wastewaters containing oil and grease (Matteson et al. 1995; Israilides et al. 1997; Chen et al. 2000; Longhi et al. 2001), dyeing and textile industry wastewater (Xiong et al. 2001; Kim et al. 2002), and also the purification of potable water (Vik et al. 1984) and urban wastewaters (Persin and Rumean 1992; Pouet and Grasmick 1995).

In this content, further researches on the utility of electroflotation (EF) and electrocoagulation (EC) are considered to be of utmost relevance and necessity. We have reported earlier on the combined application of EF and EC to the purification of wastewaters from agro-industry (Karpuzcu et al. 2002), dyeing trade (Romanov et al. 2000a) and galvanic baths (Karpuzcu et al. 2000). Now the results of the application of EF and EC to the cleaning of oil refinery wastewater are presented. The present study uses EF and EC for the treatment of PCWW. The methods entail no secondary contamination, are easy in exploitation and use a minimal volume of materials and reagents. In this study, removal of COD, turbidity, phenol, hydrocarbons and grease from the PCWW was examined as a treatment step prior to the biological treatment.

Materials and methods

Characterization of PCWW

The petrochemical wastewater under study was obtained from a distillery plant (TUPRAS Refinery, Izmit, Turkey), which uses a modern production process. The wastewater arriving in the drainage tank was sampled. Chemical characteristics of the raw wastewater are given in Table 1. A spectrophotometer was used for the photometric COD analysis, and the standard method (APHA, AWWA, WPCF 1995) was employed. A ratio turbidimeter was used for the turbidity measurements in the nephelometric turbidity unit (NTU). For all experiments, statistical analysis data were expressed as the mean of three or more independent sample measurements ±SD (standard deviation). Standard error estimations for the parameters under study are given in Tables 1 and 3.
Table 1

Chemical characteristics of wastewater from the TUPRAS refinery



COD (mg O2 /l)


Turbidity (NTU)


Phenol (mg/l)


Hydrocarbons (mg/l)


Grease (mg/l)




Reactors characterization

Two laboratory-scale reactors (their schemes given in Fig. 1) were used for EF and EC. All characteristics of the reactors are given in Table 2. In the experiments, EF and EC units made of organic glass were used (they are units with horizontal arrangement of blocks of alternating electrodes). Iron/aluminium were used in the EC process as materials for anodes/cathodes or vice versa. Before the experiments, the electrodes were immersed in 1% HCl for 8 h (Scott 1995).
Fig. 1

The EF(a) and EC (b) reactors used in the laboratory experiments

Table 2

Characteristics of the reactors used


EC cell

EF cell


 Material of anode/cathode

Al/Fe; Fe/Al

Graphite/stainless steel mesh




 Surface area (cm2)



 Number of electrodes

Anode=8, cathode=8

Anode=1, cathode=1

Reactor characteristics

 Dimensions (cm)



 Distance between anode and cathode (mm)



 Volume of cell (ml)



EF and EC were studied with a voltage of 12 V applied across the electrodes. The current density (CD) was taken as 5, 10 and 15 mA cm−2; the duration varied in the limits of 2.5–20 min for EF and 1–10 min for EC. Temperature (21–22 °C) and pH=7.6 were stable during the experiments. Before the flotation, 50 mg L–1 of coagulant (aluminium sulfate) was added to the EF-cell.

The mechanisms of reactions that take place in the electrochemical units have been studied well enough (Antropov 1977; Scott 1995). Some chemical reactions that occur on electrodes and in the bulk wastewater are shown below.

EF cell (Fig. 1a)

In this study, graphite anodes and stainless steel mesh as cathode were used in the EF process. The flocks formed settle in the water and are then transported to the surface by the bubbles of gases (Н2, О2, etc.) produced in the process of electrolysis in the EF cell.

Reactions on the anode:
$$ \begin{aligned} & {\text{4OH}}^{ - } - {\text{4}}\overline{{\text{e}}} = {\text{2H}}_{{\text{2}}} {\text{O}} + {\text{O}}_{{{\text{2 }}{\left( {\text{g}} \right)}}} \\ & {\text{2H}}_{{\text{2}}} {\text{O}} - {\text{4}}\overline{{\text{e}}} = {\text{O}}_{{{\text{2 }}{\left( {\text{g}} \right)}}} + {\text{4H}}^{ + } \\ \end{aligned} $$
Reactions on the cathode:
$$ \begin{aligned} & {\text{2H}}_{{\text{2}}} {\text{O}} + 2\overline{{\text{e}}} = {\text{H}}_{{{\text{2 }}{\left( {\text{g}} \right)}}} + 2{\text{OH}}^{ - } \\ & {\text{O}}_{{\text{2}}} + 2{\text{H}}_{{\text{2}}} {\text{O}} + {\text{4}}\overline{{\text{e}}} = 4{\text{OH}}^{ - } \\ \end{aligned} $$

An essential moment of the process is getting bubbles of the same size that are evenly distributed in the EF cell volume. Under CD variations in the range of 10–40 mА cm−2, the average diameter of hydrogen bubbles is 15–30 μm, while for oxygen bubbles it is 45–60 μm. Normally, small bubbles are preferable because they possess a greater total surface and remain in the solution for a long time.

EC cell (Fig. 1b)

Wastewater enters the EC unit (Fig. 1b). It passes by aluminium (iron) electrodes and, being enriched with Al3+ (Fe2+) ions, passes into section 2, where flotation of PCWW coagulated on Al(OH)3 (or Fe(OH)2) happens.

Reactions on the Al anode:
$$ \begin{aligned} & {\text{Al}}_{{{\left( {\text{s}} \right)}}} - {\text{3}}\overline{{\text{e}}} = {\text{Al}}^{{{\text{3}} + }} _{{{\left( {{\text{aq}}} \right)}}} \\ & {\text{Al}}^{{{\text{3}} + }} _{{{\left( {{\text{aq}}} \right)}}} + {\text{3H}}_{{\text{2}}} {\text{O}} = {\text{Al}}{\left( {{\text{OH}}} \right)}_{{\text{3}}} + {\text{3H}}^{ + } \\ \end{aligned} $$
Reactions on the Fe anode:
$$ \begin{aligned} & {\text{Fe}}_{{{\left( {\text{s}} \right)}}} - {\text{2}}\overline{{\text{e}}} = {\text{Fe}}^{{{\text{2}} + }} _{{{\left( {{\text{aq}}} \right)}}} \\ & {\text{Fe}}^{{{\text{2}} + }} _{{{\left( {{\text{aq}}} \right)}}} + {\text{2H}}_{{\text{2}}} {\text{O}} = {\text{Fe}}{\left( {{\text{OH}}} \right)}_{{\text{2}}} + {\text{2H}}_{{\text{2}}} {\text{O}} \\ & {\text{Fe}}^{{{\text{2}} + }} - \overline{{\text{e}}} = {\text{Fe}}^{{{\text{3}} + }} \\ & {\text{Fe}}^{{{\text{3}} + }} + {\text{3H}}_{{\text{2}}} {\text{O}} = {\text{Fe}}{\left( {{\text{OH}}} \right)}_{{\text{3}}} + {\text{3H}}^{ + } \\ \end{aligned} $$

Put through the flotation, particles of PCWW are removed from the water surface by means of a vacuum pump, and purified water pours into the collector. In the process of the EC, the absorption volumes of the metal hydroxides formed are very high. Coagulated particles attract and absorb micro-colloidal particles and ions from the wastewater.

The specific electrical energy consumption of the reactors for the PCWW treatment is quite low (~1.3–2.2 kWh m−3 for current densities of 5–15 mA cm−2).

Results and discussion

Relationships between the PCWW removal parameter and EF time/CD

One of contemporary methods applied to the purification of wastewater from hydrophobic contaminants is the EF process. Its effect depends on the size of bubbles issued, which determines the properties of air phase, and on the EF procedure. In the figures below (Figs. 2, 3, 4, 5 and 6), variations of the PCWW parameterswith the time of PCWW treatment in the EF cell are shown. In addition, the CD i was varied in the range 5–15 mA cm−2 for the temporal series under study as well.

As seen from Fig. 2, in the first stages of the EF process (2.5 min) the percentage COD removal is small (15–25%). An increase in the EF time leads to better values of the COD parameter. Thus, the wastewater processed during 20 min (i=5 mA cm−2) exhibits a 45% improvement in the COD index. A CD of 15 mA cm−2 leads to a higher COD value (72%). When i=10 mA cm−2, the COD index is 53%. This can be explained by the growth in the number of bubbles of electrolytic gases produced on the electrodes with the growth of the CD. The bubble growth favours the more intensive removal of suspended particles from the solution. The changes in clarity of the solution studied are shown in Fig. 3. Increasing the duration of EF decreases the muddiness of the solution. At i=15 mA cm−2 the muddiness decreases to 5 NTU cm−2 after 20 min of treatment.
Fig. 2

COD removal ranges in the EF process

Fig. 3

Turbidity removal ranges in the EF process

The content of phenol and hydrocarbons in wastewater is an important index of the degree of decontamination. The main constituents of the raw wastewater were n-alkanes, isoalkanes, cyclic alkanes, aromatic hydrocarbons, and phenols. According to Gulyas and Reich (1995), flotation was found to be a suitable pre-treatment step for removing the majority of alkanes, aromatics and phenols. Flotation protects the biological stages against the inhibitory effects of phenols, for example. Since volatile aromatics (such as toluene, xylenes, ethylbenzene) were also detected in the raw wastewater, it is recommended to collect and treat the off-gas from flotation units. In Fig. 4 the relationship between the phenol concentration and the duration of EF is shown.
Fig. 4

Phenol removal ranges in the EF process

The greatest removal of phenol is attained under certain optimal values of the duration of EF (20 min) and i=15 mA cm−2. However, the growth in CD has a small influence on the degree of phenol removal, as seen from Fig. 4. The curves are parallel and close to each other.

Figures 5 and 6 shows the percentage of hydrocarbons and oil (grease) removal as a function of the EF duration and CD. As seen from the curves, rapid removal of hydrocarbons and oils happens in the initial stage of the EF process.
Fig. 5

Hydrocarbon removal ranges in the EF process

Fig. 6

Grease removal ranges in the EF process

After 20 min the changes of the corresponding indices are insignificant, and the efficiency of EF declines. The maximal degree of hydrocarbon and oil removal (67% and 61%, correspondingly) is reached for an EF duration of 20 min and i=15 mA cm−2.

Figure 7 presents the resulting graph, where the common wastewater cleaning effect is expressed as a percentage and is shown as a function of CD and the duration of EF. As seen from the figure, the best EF index (85%) for different CD values is attained for muddiness, and the worst (56%) is attained for phenol. Data analysis shows that the main factor that influences the wastewater cleaning effect under EF is the duration of the EF process. CD influences the changes in the wastewater characteristics studied, but to a lesser extent.
Fig. 7

Summary of EF performance—percentage removal

It is worth noting that in the process of EF the synthesis of decontaminants happens. Their biocidal activity against all forms of microorganisms (and spores) exceeds hundred times analogous indices for such chemicals as hypochlorite solution, peroxides, glutaric aldehyde etc. As shown by Romanov et al. (1998, 2000b), intensive oxidation of liquid media under direct electrolysis conditions can promote the improvement of both flotation indices and the quality of the system being processed (decrease in BOD, control of the redox potential of the environment, deodouration, decontamination, disinfection, etc).

The statistical results obtained were calculated from a simple linear model for the EF process (see Table 3). The output shows the results of fitting the linear model to the relationships between the PCWW removal parameter and EF time description. The equations of the fitted model is
Table 3

The results of statistical processing of experimental EF data for the linear model




Correlation coefficient

Standard error of estimation


 i=5 mA cm−2





 i=10 mA cm−2





 i=15 mA cm−2






 i=5 mA cm−2





 i=10 mA cm−2





 i=15 mA cm−2






 i=5 mA cm−2





 i=10 mA cm−2





 i=15 mA cm−2






 i=5 mA cm−2





 i=10 mA cm−2





 i=15 mA cm−2






 i=5 mA cm−2





 i=10 mA cm−2





 i=15 mA cm−2





$$ {\text{\% Removal }}{\left( {{\text{EF}}} \right)} = a + bT{\text{ }}{\left( {T - {\text{time}}} \right)} $$
Since the P value in the ANOVA table is less than 0.05, there is a statistically significant relationship between % removal and time at the 95% confidence level. The average correlation coefficient equals 0.91–0.95, indicating a relatively strong relationships between the variables. The standard error of the estimate shows the standard deviation of the residuals as equal to 6.000.

Relationships between removal parameter and EС time/CD

The possibility of using alternative methods of wastewater decontamination (namely, EC combined with further EF) was also studied (see Fig. 1b). The results of the study relative to five wastewater indices are given in Figs. 8, 9, 10, 11 and 12 for Fe and Al anodes.

As seen from Fig. 8, the maximal COD removal in the process of EC (80%) is reached after 10 min of wastewater processing for both types of anodes. Stabilization follows the first 5 min of intensive cleaning of the solution (~75%). Analogous curves are observed for the turbidity (Fig. 9). When comparing results for Fe and Al anodes, it becomes evident that better cleaning of wastewater happens with Al anodes. Thus, the value of turbidity after 10 min of EC processing at i=15 mA cm−2 equals 6.7 NTU for Fe anodes, while for Al anodes its value is determined as 2.5 NTU. An analogous situation happens when phenol is to be removed (see Fig. 10): application of Al anodes gives better results than Fe anodes.
Fig. 8

COD removal ranges in the EC process

Fig. 9

Turbidity removal ranges in the EC process

Fig. 10

Phenol removal ranges in the EC process

Somewhat different results are found in the case of hydrocarbon removal (see Fig. 11). For low values of CD (i=5–10 mA cm−2) both groups of anodes demonstrate approximately equal level of hydrocarbon removal (57%). However, when the CD is 15 mA cm−2, the Fe anode gives better results (80% hydrocarbon removal) than the Al anode (68%).
Fig. 11

Hydrocarbon removal ranges in the EC process

In Fig. 12, the dependence of grease removal on the duration of coagulation is shown for different values of CD. For both types of electrodes (Fe, Al), the indices of grease removal are approximately the same (~80%). A distinctive feature is that at i=10 mA cm−2 the index of grease removal is better for the Fe anode.
Fig. 12

Grease removal in the EC process

In Fig. 13, the dependence of the five indices studied on the CD is shown for both types of anodes. As seen from the figure, for both Fe and Al anodes the wastewater treatment is the most effective relative to the second index (turbidity). The data comparison shows that for the rest of indices the best results are obtained for Al anodes.
Fig. 13

Summary of EC performance—percentage removal. a Fe anode. b Al anode


In this study, the efficiency of electrolytic processes (EF and EC) applied to the purification of oil refinery wastewater from different contaminants is investigated. With the help of these processes, a high and stable level of extraction of contaminants can be attained. The results of the EF applied have shown that it is possible to separate suspended particles (oils, greases, oil-fuel products) with a density close to that of water, which cannot normally be separated under ordinary flotation.

Application of the EC method needs no chemical reagents and makes the treatment of oil refinery wastewater easy for regulation and automation.

A constructional optimisation of the EF and EC apparatus for the treatment of PCWW, which allows passing from laboratory-scale models to industrial specimens possessing the same level of technological characteristics, is possible and is now under development.


The authors are grateful for the financial support to the Scientific Fund of Gebze Institute of Technology.

Copyright information

© Springer-Verlag 2004