PII: ELSEVIER

S0269-7491(97)00062-6

Environmental Pollution, Vol. 97 No. 1 2, pp. 147 152, 1997 ,~ 1997 Elsevier Science Ltd All rights reserved. Printed in Great Britain 0269-7491/97 $17.00+0.00

DETOXIFICATION OF T A N N E R Y WASTE LIQUORS WITH A N ELECTROLYSIS SYSTEM

Apostolos G. Vlyssidesa* and Cleanthes J. Israilides b
aNational Technical University of Athens, Laboratory of Organic and Chemical Technology, 9, Heroon Polytechniou Street, Zographou 157 00, Greece blnstitute of Technology of Agricultural Products, National Agricultural Research Foundation, 1, So./: Venizelou Street, Lycovrissi 141 23, Greece
(Received 14 February 1997; accepted 23 April 1997)

Abstract This paper describes an electrochemical treatment and detoxification of tannery waste liquors ( T W L ) . In this technique, T W L was passed through an electrolytic cell using a Ti/Pt anode and a stainless steel304 cathode. Owing to the strong oxidizing potential of the chemicals produced (chlorine, oxygen, hydroxyl radicals and other oxidants) the organic and inorganic pollutants (ammonia, sulfides and chromium) were wet oxidized to carbon dioxide, nitrogen oxides and sulfur dioxide. In addition, chromium was precipitated a s C r 2 ( S 0 4 ) 3. Experiments were run in a batch, laboratory-scale, pilot-plant, and the results are reported herein. After 30 min and 3 h of electrolysis at 0.26 A cm -2, 45C and p H 9, total chemical oxygen demand (COD) was reduced by 52 and 83% and biochemical oxygen demand (BODs) was reduced by 35 and 66%, respectively. Additionally, total suspended solids ( TSS) were reduced by 8.6 and 26%, total phenolic compounds were reduced by 95.6 and 99.4%. Ammonia, sulfides and soluble chromium were reduced by 100% in both cases, while the mean anode efficiency was 81gh -1 A - l m -2 and 1.9gh -1A - l m - e . Also, the mean energy consumption was 4.8kwh kg -1 of COD reduced and 200kwh kg -1 of COD reduced for 0.5 and 3 h, respectively. These results strongly indicate that this electrolytic method of total oxidation of T W L cannot be cost effective for wide use. However, it can be used as an effectivepretreatment stage for detoxification of the wastewater, owing to great efficiency especially with respect to COD and toxicity (phenolics ) reduction. 1997 Elsevier Science Ltd Keywords: Tannery waste liquors, electrochemical treatment, oxidation of phenols, chromium(Ill) removal.

INTRODUCTION
Environmental pollution owing to tannery waste liquors (TWL) is an important problem and causes great concern to the leather industries owing to the strain it creates on their resources. The T W L are among the most *To whom correspondence should be addressed. Fax: 30 17723163. 147

difficult wastes to treat due to their high organic load. Tanneries are producing about 7-15m 3 of T W L per tonne of rough leather. An average composition of T W L is given on Table 1 (ESCAP, 1982). Among the organic compounds which are present in these wastes are tannins (polyphenolic compounds) which are difficult to break and are considered to be highly toxic pollutants. Among the inorganic compounds, which are present in T W L in high concentrations, .are ammonia (Szpryokowicz et al., 1995; Naumczyk et al., 1996), sulfides and chromium (Sykes and Corning, 1987) that are strong inhibition factors for any kind of biological treatment. There is not any universally accepted method for an effective treatment of T W L at worldwide level. A thorough review of tannery wastewater treatment was given by Tsotsos (1986). Existing technologies for the treatment of T W L include the use of ozone gas, ozone and UV, and UV and hydrogen peroxide. In spite however, of improvements resulting in less expensive ozone production, the cost of ozone generation is still one of the main drawbacks when it is compared with other technologies The most widely used technology, the aerobic treatment of the sludge after a primary removal of proteins by sedimentation is not considered cost effective in small and medium size industries like most of those which occur in Greece. In general, biological treatment methods whether aerobic or anaerobic, even though they lead to a significant removal of chemical oxygen demand (COD) and biochemical oxygen demand (BOD), still leave unacceptable levels for direct discharge. Moreover, full nitrification cannot be established (Grenschow and Hegemann, 1993). Oxidation technologies for the treatment of industrial wastewaters have been an attractive alternative to biological methods, especially where biorecalcitrant and toxic wastes are to be treated (Mantzavinos et al., 1996). In the case of TWL, electrochemical treatment has been reported as a feasible process (Szpryokowicz and Zillio Grande, 1994). However, some researchers are in favour of a combined oxidative and biological treatment for tannery wastewaters (Szpryokowicz and Zillio Grande, 1994, 1995; Jochimsen et al., 1996; Kaul and Szpryokowicz, 1996; Naumczyk et al., 1996).

148

A. G. Vlyssides, C. J. Israilides
of chloride--chlorine-chloride is reduced owing to the production of C103- that is a stable chlorate anion. Therefore, at low pHs the chlorides are reduced during the electrolysis process producing free chlorine, while in high p H the chlorides are oxidised producing chlorates. Only in moderate alkaline to neutral conditions, the initial concentrations of chlorides remain stable (Vlyssides et al., 1996) during the continuous production of free hydroxyl radicals. Table 2 shows the electrochemical and chemical reactions (Israilides et al., 1997) that can take place in a mixture of brine solution and wastewater in an electrolysis cell using Ti/Pt anode.

Table 1. Mean composition of tannery waste liquors (TWL)

Characteristics BOD5 COD Total suspended solids Total chromium Sulfides Chlorides Total phenolic compounds Ammonium nitrogen Kjehdahl nitrogen Fats and oils pH Range (mg litre- 1) 210-4300 180-27 000 925-36 000 3-350 1-500 1500-28 000 0.4-100 17-380 90-630 49-620 1-13

This study deals with a physicochemical approach to the treatment of TWL, based on the principle of oxidative degradation of all its organic compounds in an electrolysis system.

M A T E R I A L S AND M E T H O D S

Theoretical approach The time of oxidation depends upon the stability and concentration of compounds; concentration of NaCI used; temperature; pH of the solution; time of recirculation; size of the anode; and current and voltage applied (Vlyssides et al., 1996). The electrochemical reactions, which take place during the electrolysis of a brine solution, are complicated and not entirely known. For the time being only assumptions can be made, based on the products that can be determined (C12, CIO2, 03, O H , O , C1OH, H202, 02, H2, CO2). The radicals O H , O and C1OH have a very short life owing to their high oxidation potential and they are either decomposed to other oxidants (C12, 02, C102, 03, and H202) or they oxidise organic compounds (direct oxidation). The primary (C12, 02) and secondary (C102, 03, and H202) oxidants that are produced from the destruction of radicals have quite a long life and are diffused into the area away from the electrodes continuing the oxidation process (indirect oxidation). The direct electro-oxidation rate of organic pollutants is dependent on the catalytic activity of the anode; on the diffusion rates of organic compounds in the active points of the anode; and on the applied current density (Antropov, 1977; Prentice, 1991). The indirect electrooxidation rate is dependent on the diffusion rate of secondary oxidants into the solution, temperature and the pH. An effective pollutant degradation is based on the direct electrochemical process because the secondary oxidants are not able to completely convert all organics into water and carbon dioxide. F r o m previous investigations (Wilk et al., 1987; Vlyssides et al., 1996), in acid solutions, oxygen, free chlorine and maybe some amounts of ozone and chlorine oxides are the main secondary oxidants as byproducts of the direct oxidation process. In moderate alkaline solutions a cycle of chloride-chlorine-hypochlorite--chloride takes place, which produces OCI-, oxygen and some amounts of hydrogen peroxide and maybe ozone. In strong alkaline solutions the cycle

The laboratory pilot plant The experimental plant consisted (Fig. 1) of the electrolytic cell, the recirculation system, the pH-correction system and the cooling system for the wastewater. The electrolytic cell The electrolytic cell consisted of the cathode, which was a stainless steel 304 cylinder of 20 cm diameter, and of the anode, which was located in the centre of the cylinder and which was made of titanium alloy, measuring 48 cm in length and 2.54 cm in diameter, covered by platinum alloy foil approximately 0.25 mm thick. The electrodes were operated at 15 volts D.C. and 100 amperes. The recirculation system The recirculation system included a vessel of 15 litres, which contained the wastewater to be treated, and a Table 2. Electrochemical reactions
Anode Primary electrochemical reactions n 2 0 + M [ ] + Cl- ~ M [CIOH-] + H + + 2eH20 + M [ ] ~ M [OH-] + H + + eSecondary electrochemical reaction 6OC1- + 3 H 2 0 - 6 e - ~ 3 / 2 0 2 + 6 H + 4 C 1 - +2C103Cathode Primary electrochemical reactions 2 H30 + + 2e- ~ H2 + 2 H20 2 H20 + 2e----~ H2 + 2 OHSecondary electrochemical reaction OC1- + H20 + 2 e - ~ C l ~ + 2 O H Oxidation reactions In a closed anode area R + M[OH-]---~ M [] + RO + H + + eR + M [ C I O H ] ~ M [ ] + RO + H + + CI- + eH20 + M[OH-]---*M[] + 02 + 3H + + 3 e H20 + M[C1OH-] + C1- ~ C12 + M[] + 02 + 3H + 4 e Away from electrodes R + 02 + 2H ~ R O + H 2 0 - 2 e R + C12--* RCI + C I - - e R + OH----~ RO + H + 2 e RC1 + O H - ~ R O + C1- + H + + eEquilibrium reactions C12 + 2 0 H - ~ H20 + OCI- + CI-

Detoxification of tannery waste liquors

149

+ anode Pt/Ti

cathode SS-304 cooling jacket

pp

I -NaOH C
vE
F

S~-------~-'\ -1

cooling water

surge vessel

sampling

Fig. 1. Experimental laboratory pilot plant. C = electrolytic cell; E = electromagnetic valve; P = recirculating pump; TICR = temperature indicator controller recorder; pHIR = pH indicator recorder; RIR = redox indicator recorder; PP = peristaltic pump. centrifugal pump (P) of 40 litre min -I flow rate, which continuously recirculated the reactor contents into the electrolytic cell.
Material The composition of the T W L that was used in all experiments is shown in Table 3. Methodology The efficiency of the electrolytic cell was studied when pH and temperature remained stable at 9.0 and 45C, respectively, during the experiment. Four experiments under the same conditions were run, and the results presented in this work are the average of the four separate measurements. Each experiment was of batch operation and its duration was 3 h. Every 5 rain samples were taken from the reactor. Soluble COD, BODs, total suspended solids (TSS), sulfides, ammonia, total Kjeidahl nitrogen (TKN), total soluble chromium, pH, redox, chlorides, chlorine and total oxidants were measured. Additionally, samples were taken for total phenolic compound determination. The COD, BODs, TSS, T K N , ammonia, sulfides and free chlorine analysis were carried out according to the Standard Methods for the Examination of Water and Wastewater ( A P H A - A W W A - W P C F , 1989). The determination of ozone and the total oxidants except for chlorine was carried out according to Wilk (1989). The

The pH correction system

The pH correction system consisted of a pH indicatorrecorder (pHIR) which continuously measures the pH content of the surge vessel and with the help of the dosing pumps PP-1 and PP-2 which supply H C L and NaOH, respectively, the pH is kept constant at 9.0.

Cooling system
The cooling system included a temperature indicator controller recorder (TICR), an electric valve for the water and a cooling water jacket which was located in the reactor. The cylinder wall and the cooling jacket constituted the cathode. Whenever the temperature of the water exceeded 45C, cooling water was circulated in the jacket, until the temperature returned to the desired value. The change of redox potential in the reactor was continuously recorded by a redox indicator-recorder (RIR). The samples for analysis were drawn from the recirculation reactor.
Table 3. Composition of tannery waste liquor (TWL) used in experiments

Characteristics pH BOD5 COD Total suspended solids Total dissolved solids Total chromium Sulfides Chlorides Total phenolic compounds Ammonium nitrogen Kjehdahl nitrogen Alkalinity as CaCO3 Fats and oils

Value (mg litre-I) 4 1720 8540 6200 22 400 67 385 18 500 91 288 520 1770 280

9000 8000 7000 6000 E 5000 4000 8 3000 2000 1000 0 time, minutes Fig. 2. Reduction of total suspended solids (--F-I--), COD ( - - 0 - - ) and BODs ( - - ~ - - ) during the electrolysis treatment.

150 100 90 ~ 80 e., = 70 ~. 60 o o 40 30

A. G. Vlyssides, C. J. Israilides
400 350 30O o 250 r 200 ~o 150 o 10o 50 0 time, m i n u t e s

10

--It-~.~.-7.~.~.7..7.~.7.7.'-117.7.7.~2 . . . . . . . time, m i n u t e s

Fig. 3. Concentration of total phenolic compounds during electrolysis treatment as a function of time.

Fig. 5. Sulfides (--I-1--) and total soluble chromium (--@--) reduction during electrolysis treatment. The T K N was almost completely degraded (98% reduction) during the 3 h of electrolysis (Fig. 4). The same figure shows that an increase of nitrates was observed which started to be produced after 30min of electrolysis, when the ammonium nitrogen was depleted and reached its highest concentration (21 mglitre -~) at 120 min. Then it was gradually reduced to 11 mg litreat the end of the 3 h of electrolysis. There was also a very rapid reduction of the sulfides (Fig. 5) which were reduced from 385mglitre -1 to nil, during the first 25 min of the electrolysis. In the environment, chromium is found in the form of a trivalent and a much more toxic six-valent atom. In the T W L chromium is found in the trivalent form. Its salts and oxides (CrC13, Cr2(SO4)3, Cr203 ) are easily removed due to their low solubility in water, in contrast with the six-valent chromium salts and oxides which are water soluble. During the strong oxidative conditions of electrolysis the trivalent chromium was converted to six-valent. As a result, it remains in solution while the toxicity of the solution increases. There was an impressive decrease in the concentration of soluble chromium which was nullified during the first 10-15 rain of electrolysis. The reduction of chromium could probably be connected with the parallel reduction of sulfides according to the following reactions (Jorgensen and Johnsen, 1989):

total phenolic compounds were measured by the FolinCiocalteu method (Slinkard and Singleton, 1977). Chromium was determined by atomic absorption using a Perkin Elmer (model 2380) analyser.

R E S U L T S AND D I S C U S S I O N As shown in Fig. 2 the TSS were reduced by only 26% (from 6200 to 4560mglitre -~) during the 3h of electrolysis. This is an indication of the very slow rate of hydrolysis, at least during the first hours of electrolysis. On the other hand, there was a great decrease of BOD5 (Fig. 2), from 1720 to 590mglitre - l (65% during the 3 h of electrolysis). The COD was reduced even further from 8540 to 1370 mg litre -I, as shown in Fig. 2, a reduction of about 84% during the 3 h of electrolysis. The C O D / B O D ratio which is an index of toxicity during biological treatments was improved considerably and dropped from 5 to 2.3. Phenolic compounds were destroyed very quickly, as shown in Fig. 3. They were reduced by 95.5% during the first 30min of the electrolysis. The reduction of ammonium nitrogen was also very fast. As shown in Fig. 4, it was reduced by 99% during the first 60 min of the electrolysis. 600 - 5OO 400 j 3OO 8 20O 100 0 ......... _ .,,. ~ . . . jL. . JL__...._ ::;.~..~t._L
~O
oo

--

20000 18000 16000 14000 12000 10000 8000 4000 20OO

0 +-H-H H--~ ~ ~ ~ +~--'~c-t I I t I I I ~-H-++-~-++q-++-

time, minutes Fig. 4. Total Kjeldahl nitrogen (--D--), ammonia ( - - Q - - ) reduction and nitrates ( - - A - - ) production during electrolysis treatment. time, minutes Fig. 6. Chlorides (--I7--), total chlorine (--@--) and other oxidants (--/k--) changes during electrolysis treatment.

Detoxification of tannery waste liquors 90 80 < 70

60

151

,~ 50
m 40 o 30 ~ 20 10

to appear (Fig. 5) reaching 5 mg iitre -t after 3 h of electrolysis. The resolubilisation of chromium was probably due to the oxidation of precipitants and to an inability to reduce Cr(VI) to Cr(III) due to the absence of sulfides.
Production of oxidants

time, minutes Fig. 7. Anode efficiency during electrolysis treatment. 250 ..~ 200
O

Figure 6 shows the concentrations of chloride ions and total chlorine, as well as of other oxidants as a whole. The concentration of chlorides was reduced in the first 10 min from 18 540 mg litre-I to 12 300 mg litre- 1 and was stabilised at this concentration during the whole process of electrolysis. The chlorine was increased during the first 10rain from zero to 6300mglitre -I and remained stable thereafter. This proves the existence of the theoretical cycle, chlorides-chlorine-chlorides. The other oxidants, except for chlorine, increased rapidly in the first 10rain from zero to 5840mglitre -1 and then continued to increase with a rate of 6.8 mg litre-I rain -~. The presence of 03 and H 2 0 : was also detected.
Anode efficiency

100
50

o time, minutes Fig. 8. Energy consumption during electrolysis treatment.

S - + 02 --~ S02 + eS02 + 1120 --~ H2S03 3H2S03 + 2H2Cr04 --~ Cr2(S04)3 + 51120 The above considerations may be confirmed by:

Figure 7 shows the anode efficiency of the electrode measured in Kg COD removed per hour per m 2 of anode surface and per ampere applied (kg COD~ h -I m - Z A - I ) . With the start of the electrolysis the efficiency was 81 kg CODr h -1 m -2 A -I which is considered very satisfactory (Comninellis, 1992), but in the next 15min it was reduced to 6.8kg COOt h -1 m - 2 A -1. Then it continued to be reduced reaching 1.8kg COD~ h J m - 2 A -1 at the end of the 3h. These results cannot be considered satisfactory, after the 15min of electrolysis. This is clearly seen in Fig. 8, where the demands in energy consumption are shown. The gradual decrease of electrode efficiency indicates the increasing difficulty for oxidation of residual organics in the waste.
Energy consumption

(a) repetition of the experiment under the same conditions without the presence of sulfides which were removed by nitrogen stripping. Only very little chromium was removed (15%) and the chromium was present in the solution in its main form Cr(VI); (b) analysis of the precipitant showed that chromium was present as Cr2(504)3; (c) after 1 h of electrolysis soluble Cr(VI) has started

Figure 8 shows the energy demands versus time in the electrolysis system. During the first 10min of electrolysis the energy consumption was fairly low (from 4.8 to 11.0 Kwh per kg CODr) but then increased abruptly to 58 Kwh per kg CODr. Later, although at a slower rate, the energy consumption continued to increase reaching 200 Kwh per kg CODr rendering unfavorable the application of the electrolytic oxidation for the complete treatment of TWL.

Table 4. Efficiency of electrolysis for the first 30 min of electrolysis

Duration of electrolysis (min) 5 l0 15 20 25 30

Soluble COD reduction (%) 20.0 38.7 47.0 48.73 50.4 52.0

Total phenolic compounds reduction (%) 27.4 40.6 67.0 80.2 93.4 95.6

Ammonium nitrogen reduction (%) 19.4 27.0 34.7 42.0 49.6 64.5

Sulfides reduction (%) 28.5 63.1 84.4 98.7 100 100

Soluble chromium reduction (%) 85.0 98.5 100 ----

COD/BOD ratio 4.96 4.00 4.00 4.00 3.84 3.70

152 CONCLUSIONS

A. G. Vlyssides, C. J. Israilides

The electrolytic oxidation as an integrated solution for the treatment o f T W L does not seem to be feasible due to high energy demands. However, the application of electrolysis as a pretreatment and for residence time up to 30min seems to be very efficient, mainly for the detoxification of TWL. In the first 30 min of electrolysis the following could be achieved (Table 4): 1. 2. 3. 4. 5. 6. reduction of phenolic compounds by 95%; reduction of ammonium nitrogen by 64.5%; complete oxidation of sulfides; complete sedimentation of chromium; reduction of COD by 52%; improvement of biodegradability index (COD/ BOD) from 4.96 to 3.7.

The pretreatment of T W L is favoured by the fact that it contains high amounts of NaCI: 0.25-4.6%.

REFERENCES APHA-AWWA-WPCF (1989) Standard Methods for the Examination of Water and Wastewater, 17th edn. American Public Health Association, Washington, DC. Antropov, L. (1977) Theoretical Electrochemistry, 1st edn. Mir, Moscow. Comninellis, C. (1992) The electrochemical treatment of wastewater, gwa 11,792-797. ESCAP (1982) Report on Industrial Pollution Control Guidelines Part VII." Tanning Industry. United Nations, Bangkok. Grenschow, E. and Hegemann, W. (1993) Anaerobic treatment of tannery wastewater, gwf Wasser Abwasser 134, 262-268. Israilides, C. J., Vlyssides, A. G., Mourafeti, V. N. and Karvouni, G. (1997) Olive oil wastewater treatment with the use of an electrolysis system. Bioresource Technology (in press). Jochimsen, J. C., Jekel, M. R. and Hegemann, W. (1996) Combined oxidative and biological treatment for separated streams of tannery wastewater. Proceedings of the 2nd Specialized Conference on Pretreatment of Industrial Wastewaters. IAWQ, Athens, pp. 404-412.

Jorgensen, S. E. and Johnsen, I. (1989) Removal of heavy metals. In Principles of Environmental Science and Technology, eds E. S. Jorgensen and I. Johnsen, 2nd edn, pp. 385386. Elsevier. Kaul, S. N. and Szpryokowicz, L. (1996) Biological process as a pretreatment for electrochemical oxidation. A case study for cluster of tanneries from a common effluent treatment plant. Proceedings of the 2nd Specialized Conference on Pretreatment of Industrial Wastewaters. IAWQ, Athens, pp. 386-395. Mantzavinos, D., Lauer, E., Hellenbrand, R., Livingston, A. G. and Metcaife, I. S. (1996) The wet oxidation as a pretreatment method for wastewaters contaminated by bioresistant organics. Proceedings of the 2nd Specialized Conference on Pretreatment of Industrial Wastewaters. IAWQ, Athens, pp. 209-216. Naumczyk, J., Szpyrkowicz, L., De Faveri, D. M. and Zillio Grandi, F. (1996) Electrochemical treatment of tannery wastewater containing high strength pollutants. Trans IChemE 74(Part B), 59-68. Prentice, G. (1991) Electrochemical Engineering Principles, 1st edn. Prentice Hall, NJ. Slinkard, K. and Singleton, V. (1977) Total phenol analysis: automation and comparison with manual methods. American Journal of Enol. Vitic. 2,8, 49-55. Sykes, R. U and Corning, D. R. (1987) Pollution abatement and control in leather industry, industry and environment. UNEP/IEO 10(2), 9-22. Szpryokowicz, L. and Zillio Grandi, F. (1994) Application of electrochemical processes for tannery wastewater treatment. Toxicology and Envirnonmental Chemistry 44, 183-220. Szpryokowicz, L. and Zillio Grandi, F. (1995) Electrochemical treatment of tannery wastewater using Ti/Pt and Ti/Pt/Ir electrodes. Water Research 29(2), 517-524. Tsotsos, D. (1986) Tanneries. A short survey of the methods applied for wastewater treatment. Water Science and Technology 18, 69-76. Vlyssides, G. A., Israilides, C. J., Loizidou, M., Karvouni, G. and Mourafeti, V. (1996) Electrochemical treatment of vinasse from beet molasses. Proceedings of the 2nd Specialised Conference on Pretreatment of Industrial Wastewaters. IAWQ, Athens, pp. 522-529. Wiik, I. J., Altmann, R. S. and Berg, J. D. (1987) Antimicrobial activity of electrolyzed saline solutions. The Science of the Total Environment 63, 191-197. Wilk, I. J. (1989) Economic advantages of an electrolysis system for pulp bleaching. Proceedings of the International Symposium on Wood and Pulping Chemistry. TAPPI, Atlanta, GA, pp 135-137.