Tannery wastewater characterization and toxicity effects on Daphnia ...

Tannery Wastewater Characterization and Toxicity Effects on Daphnia spp. K. Cooman,1 M. Gajardo,1 J. Nieto,1 C. Bornhardt,2 G. Vidal1 1

Environmental Science Centre EULA–Chile, University of Concepcioń, P.O. Box 160-C, Concepcioń, Chile 2

Department of Chemical Engineering, University of La Frontera, P.O. Box 54-D, Temuco, Chile

Received 2 June 2002; revised 30 June 2002; accepted 8 August 2002 ABSTACT: Tannery wastewater contains large quantities of organic and inorganic compounds, including toxic substances such as sulfides and chromium salts. The evaluation of wastewater quality in Chile nowadays is based on chemical specific measurements and toxicity tests. The goal of this research was to characterize tannery wastewater and to relate its physical/chemical parameters with its acute toxicity effect on Daphnia pulex. To distinguish the most important toxic compounds, physical/chemical techniques were applied to a grab sample of a final effluent based on the Phase I toxicity identification evaluation (TIE) procedure. In addition, the toxicity of a beamhouse effluent after an activated sludge reactor treatment was investigated on Daphnia magna (introduced species) and D. pulex (native species). Effluent from different tannery processes (soaking, beamhouse, tanning and final) demonstrated high values of chemical organic demand (COD; 2840 –27 600 mg L⫺1), chloride (1813–16 500 mg L⫺1), sulfate (230 –35 200 mg L⫺1), and total solids (8600 – 87 100 mg L⫺1). All effluents showed extremely toxic effects on D. pulex, with 24-h mean lethal values (LC50) ranging from 0.36% to 3.61%. The Phase I TIE profile showed that toxicity was significantly reduced by air stripping, filtration, and a cationic exchange resin, with toxicity reductions ranging between 46% and 76%. The aerobically treated beamhouse effluent showed significantly less toxicity for both species (43%–74%). The chemical parameters demonstrated that the remaining toxicity of the treated beamhouse effluent was associated with its ammonia (120 mg NONH3 L⫺1) and chloride (11 300 mg Cl⫺ L⫺1) contents. © 2003 Wiley Periodicals, Inc. Environ Toxicol 18: 45–51, 2003.

Keywords: daphnids; tannery wastewater; acute toxicity; TIE; activated sludge reactor

Correspondence to: Dr. Gladys Vidal; e-mail: [email protected] Contract grant sponsor: European Union. Contract grant number: INCO-DC N° ERB IC18-CT98-0286. Contract grant sponsor: Direccioń de Investigacioń of University of Concepcioń (Chile). Contract grant number: DIUC 201096054-1.0. Contract grant sponsor: Direccioń de Investigacioń of University of La Frontera (Chile). Contract grant number: EP-2101. Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/tox.10094 © 2003 Wiley Periodicals, Inc.

45

46

COOMAN ET AL.

INTRODUCTION The leather tannery industry is a well-known to have a severe negative impact on the environment. In this industry animal hides are transformed into leather in a succession of many complex stages, consuming high quantities of water and using large amounts of such chemicals as lime, sodium sulfide, ammonium sulfate, sodium chloride, bactericides, vegetable tannins, and chrome salts. Tannery wastewaters are mainly characterized by high salinity, high organic loading, and specific pollutants such as sulfide and chromium (Tu¨nay et al., 1999; Song et al., 2000). The beamhouse and tanning processes are the most contaminating: the former mainly because of its high organic load and sulfide content, the latter principally because of inorganic salts of chloride, ammonia, chromium, and sulfate (Meneńdez and Diaz, 1998). Chilean tannery industry survival and growth depend not only on process efficiency but also on its environmental protection efforts. There are about 30 small- to mediumsized leather tannery enterprises in Chile, most of which use conventional chromium-tanning processes and thus are inefficient in their water use (Rivela et al., 2002). To transform 1 ton of wet salted hides (1 twsh) into 500 kg of leather product, they use about 55– 63 m3 fresh water and about 442 kg of chemicals, generating, depending on their technology, wastewater at the rate of 55–100 m3/twsh and solid wastes at the rate of 477– 696 kg/twsh (Rivela et al., 2002; Konrad et al., 2002). Effluent treatment plants are excessively costly to construct and operate and produce a lot of sewage sludge. So, for these small- to medium-sized factories cleaner production strategies that focus on reducing their impact along the entire lifecycle of the product, could be a good alternative. However, after the introduction of cleaner production strategies, tannery wastewater still needs pretreatment before being discharged into public sewers and is not acceptable for direct discharge into rivers (Song et al., 2000; Konrad et al., 2002). Since 2001 the evaluation of the quality of wastewater in Chile has been based on chemical-specific measurements as well as the use of toxicity tests with Daphnia spp. (D. magna or D. pulex). However, detecting toxicity is only a starting step, and further efforts have to be undertaken to reduce toxicity. To assist in this process, techniques have been developed that couple physical/chemical manipulations with toxicity tests and chemical analyses to isolate and to identify toxicants or classes of toxic compounds. This approach, commonly referred as toxicity identification evaluation (TIE), can help industries develop cleaner production strategies. The specific aims of this study were (1) to relate the physical/chemical characteristics distinct to tannery process wastewater with its toxic effect on D. pulex in order to indicate the toxic compounds; (2) to distinguish the major groups of important compounds of a final effluent that are

toxic to D. pulex using the Phase I TIE approach; and, finally, (3) to evaluate the performance of an activated sludge reactor by the reduction in a beamhouse effluent’s toxicity of on D. magna (introduced species) and D. pulex (native species).

MATERIALS AND METHODS Industrial Wastewater Grab samples (10 L) were taken from the soaking, beamhouse (unhairing–– deliming), tanning process wastewater and of final effluent at the time that all batch processes were running. Samples were transported on ice in polyethylene containers and immediately stored at approximately 4°C. When the bioassays could not be executed within 48 h, samples were frozen (⫺6°C).

Acute Toxicity Testing Acute toxicity was determined by exposing D. magna and D. pulex juveniles (⬍24 h) to each sample for 24 – 48 h and recording mortality at the end of exposure, which was defined as lack of organism mobility when the vessel was shaken. Organisms were obtained from in-house cultures that were fed three times weekly with a suspension of baker’s yeast, trout chow, and alfalfa with an equivalent carbon content of 7.2 mg C L⫺1 on Monday and Wednesday and 10.8 mg C L⫺1 on Friday. The culture medium was changed before feeding, and neonates were removed within 24 h. Cultures and animal exposures were conducted at 20 ⫾ 2°C in a photoperiod of 16h light:8 h dark. The dilution and control water was reconstituted moderately hard water prepared according to USEPA (1993). The solutions were not renewed and the organisms were not fed during the experiments. Dissolved oxygen (DO, HANNA, HI 9142), pH (Schott, CG 825) and conductivity (HANNA, HI8733) were measured at the beginning and end of each test. When DO was below 2 mg L⫺1 the sample was aerated before executing the test to provide enough oxygen to the test organisms (USEPA, 1993). Extreme pHs (⬍4 and ⬎9) were adjusted to be in the tolerance interval of the test organisms before testing by adding dropwise proanalysis HCl and NaOH of different normality values (0.1, 1.0, 2.0; p.a., Merck). The 24-h and 48-h median lethal concentrations (LC50) were calculated using the probit or Spearman– Karber methods (Finney, 1971, 1978). Estimation of the theoretical toxicity of the samples was based on chemical analysis of the individual potential toxicants. Therefore, to express the relative toxicity of a certain compound, acute toxic units (TUtheor,i) for this compound were calculated as: (nominal compound concentration in test solution)/LC50. Assuming these toxicants were the real causative agents, then, according to the concentration–addition model of

TANNERY WASTEWATER CHARACTERIZATION AND TOXICITY EFFECTS ON DAPHNIA SPP.

Anderson and Weber (1975), the observed toxicity (TUobserved) should equal the sum of the theoretical expected toxicity caused by the suspected toxicants: TUexpected ⫽ ¥TUtheor,i. If no interaction between toxicants is expected, then the observed sample toxicity should equal the theoretical toxicity of the toxicant with the greatest effect.

Phase I Toxicity Identification Evaluation The baseline test, the pH/adjustment tests (pH3/ad and pH11/ad), the graduated pH tests (pH6.9/grad and pH8.9/ grad), and the pH adjustment/aeration tests (pH3/aer, pHi/ aer, and pH11/aer) were performed as described by Norberg-King et al. (1991). The other fractionations were done based on procedures described by Van Sprang and Janssen (1997). The concentrations of EDTA (Titriplex III, p.a., Merck) and Na2S2O3 (p.a., Merck) in the dilution water and effluent sample were based on the 24-h 10% lethal concentrations (LC10) in D. pulex (Cooman, unpublished data) and were conducted at final EDTA and Na2S2O3 concentrations of 100 and 500 mg L⫺1, respectively. For the pH adjustment/filtration tests (pH3/fil, pHi/fil, pH11/fil), 2.5 ⫻ 25– cm glass columns were filled with synthetic hydrophobic filter wool (Diprolab MR). Additional column tests consisted of an anion (10 g, Amberlite IRA-440, Merck), a cation (10 g, amberlite IR-120, Merck), and a mixed exchange resin (1:1 mixture of mentioned resins); an 8- to 20-mesh activated carbon (10 g, Sigma) column test and a zeolite (10 g, Steelhead Specialty Minerals) column test also were incorporated. Baseline toxicity tests were duplicated, but TIE manipulations were not. Adjustment and readjustment of the pH were accomplished by adding dropwise proanalysis HCl and NaOH of different normality values (0.1, 1.0, and 2.0 N; p.a., Merck). For all manipulations, the 24-h LC50 values were converted to toxic units (TUfractionation ⫽ 100/24-h LC50) and were compared with the toxicity of the baseline test (TUbaseline). The percentage toxicity reduction (%TR) because of the applied fractionation was calculated as %TR ⫽ 100 ⫻ 关1 ⫺ 共TUfractionation ⫼ TUbaseline兲兴

(1)

Activated Sludge System (AS) An aerobic (1.8 L) reactor fed with beamhouse wastewater without any nutrient supply was operated during the 180 days of testing. The organic load rate (OLR) varied from 0.3 to 6.5 g COD L⫺1 d⫺1, and the hydraulic retention time (HRT) from 16.6 to 1 days. The sludge was periodically recycled, and the excess was withdrawn from the settling unit (6 L) in order to obtain a 30-day aged sludge. Oxygen concentration inside the reactor was maintained above 6 mg L⫺1 by blowing air with a pump through a diffuser system, which also provided its mixing. Toxicity tests were performed with grab samples of 0.45 ␮m of filtrated (Advantec

47

GC 50) influent and effluent coming from the reactor after 45 and 127 days of operation time.

Analytical Methods Analysis of chemical oxygen demand (COD), biological oxygen demand (BOD5), total solids (TS), volatile solids (VS), total suspended solids (TSS), and volatile suspended solids (VSS); and of phosphate (P-PO3⫺ nitrate 4 ), ⫺ (NONO⫺ ), nitrite (NONO ), ammonia (NONH ), sulfide 3 2 3 ⫺ (S2⫺), sulfate (SO2⫺ 4 ), chloride (Cl ), chromium (total Cr and Cr6⫹), and fats was carried out according to standard methods (APHA-AWWA-WPCF, 1985). Total Kjeldahl nitrogen (TKN) was determined by sample digestion with sulfuric acid (p.a., Merck) and selenium (Kjeldahl tablets, Merck) reagent, with a Kjeldahl apparatus used for distillation and titration of samples with hydrochloric acid (p.a, Riedel-de Hae¨ n). Samples for measurements of COD, ⫺ ⫺ 2⫺ BOD5, P-PO3⫺ , SO2⫺ 4 , NONO3 , NONO2 , NONH3, S 4 , ⫺ Cl , and chromium were filtered through a 0.45-␮m membrane (Advantec GC 50) prior to analysis.

RESULTS AND DISCUSSION Physical–Chemical and Toxicological Data of Tannery Wastewater The physical– chemical and toxicological data of the different tannery effluents are summarized in Table I. The beamhouse wastewater was characterized by an alkaline pH (12.3) and the tanning effluent a very acid pH (3.7), whereas the pH of the final effluent was nearly neutral pH (7.9). Effluents with similar pH characteristics were found by Mene´ ndez and Dıáz (1998). The COD values of the final effluent, the soaking wastewater, and the tanning wastewater were found to be 2840; 12 650 and 7150 mg L⫺1, respectively. As reported by Tu¨ nay et al. (1999), in their study beamhouse wastewater accounted for the highest COD (27 600 mg L⫺1) and the lowest DO (O2 concentration: 0.1 mg L⫺1). The effluents were all characterized by high concentrations of chloride (Cl⫺ concentration: 1813– 16 500 mg L⫺1) and total solids (8600 – 87 100 mg L⫺1). A high concentration of sulfide was only detected (S2⫺ concentration: 2150 mg L⫺1) in the beamhouse effluent, whereas the tanning wastewater contained an important amount of total chromium (Cr concentration: 4950 mg L⫺1). All process wastewaters were found to have an extremely toxic effect on D. pulex, with 24-h LC50 values ranging from 0.36% to 3.61%. These results are similar to the 24-h LC50 values for D. pulex (0.15% and 3.36%) of the chromium tanning and beamhouse effluents, respectively, reported by Yatrabi and Nejmeddine (2000). Similar results

48

COOMAN ET AL.

TABLE I. Physical– chemical results and acute toxicity of tannery effluents to D. pulex Parameter*

Soaking

Beamhouse

Tanning

Final

pH DO COD BOD5 TS VS TSS VSS Total N NONH3 NONO⫺ 2 NONO⫺ 3 POPO3⫺ 4 S2⫺ SO2⫺ 4 Cl⫺ Total Cr Cr6⫹ Fats 24h LC50b

8.1 5.4 12 650 4000 33 950 6100 3760 2860 680 180 n.d.a n.d. 20 2.7 430 13 000 n.d. n.d. 1600 0.36 (—)

12.3 0.1 27 600 6550 34 200 14 200 12 600 8100 2860 120 n.d. n.d. 0.24 2150 230 11 300 n.d. n.d. 600 2.04 (1.76–2.57)

3.7 5.2 7150 1900 87 100 16 900 770 450 830 245 n.d. n.d. 0.33 0.022 35 200 16 500 4950 0.30 90 2.63 (2.07–3.51)

7.9 4.9 2840 1359 8600 1840 203 159 390 255 0.13 0.32 0.06 ⬍0.003 2400 1813 n.d. n.d. n.d. 3.32 (3.03–3.61)

24-h LC50 values for D. magna found in literature, assuming total bioavailability of the individual compound and similar responses (same order of magnitude) of both daphnids to the individual compounds. The TUtheor values could only be used as a rough guide. Indeed, as TUexpected values differed from the TUobserved values by 1 or 2 orders of magnitude, there was no way to indicate the multiple toxicants and whether they were additive, partially additive, or independent. This analysis also showed that interactions of other known and unknown organic and inorganic compounds have to be considered. Tisˇler and Zagorc-Konç an (2001) found that in addition to ammonia and chloride in tannery wastewater exceeding toxicity thresholds of D. magna, organic compounds and undetected substances were also responsible for wastewater toxicity. As also indicated by Yatrabi and Nejmeddine (2000), we have concluded that it is too difficult to correlate the LC50 values obtained with those of the physical– chemical parameters, as the tested effluents are very complex matrices. For this reason, we note the need to combine fractionation techniques and chemical analysis with toxicity tests in order to indicate the real toxic compound or group of compounds for each wastewater stream.

* All parameters are expressed in mg L⫺1 except for pH and 24-h LC50 (%). a n.d.: not determined. b Median lethal concentration calculated by the Spearman–Karber method. Value in parentheses is the 95% confidence interval.

Phase I Toxicity Identification Evaluation of a Final Tannery Effluent Because a final effluent is effectively discharged without any previous treatment and combines distinct wastewater streams, in our study only the final effluent was subjected to Phase I TIE procedures in an attempt to distinguish the compound or group of compounds toxic to D. pulex. The toxicity characterization profile of the final effluent is summarized in Figure 1. The effluent exhibited acute toxicity on D. pulex of 30.12 TU (27.70 –33.00; Table II). A pH drift of ⫾0.1 units after completion of the bioassays was observed

with Vibrio fischeri for tanning and beamhouse effluents were reported by Catarino et al. (2001). To explain the individual toxic effects on D. pulex of every measured potential compound, we compared the acute toxic units of each individual compound (TUtheor) with the observed toxicity of the whole sample (TUobserved, Table II). The TUtheor values were calculated based on the

TABLE II. Expected toxic units of individual compounds of tannery effluents based on acute toxicity (24-h LC50) to D. magna Tannery Wastewater Soaking Parameter

mg L⫺1

NONH3 S2⫺ SO2⫺ 4 Cl⫺ Cr3⫹ Cr6⫹ TUexpected TUobserved

180 2.7 430 13,000 b n.d. n.d. — —

a

Beamhouse

Tanning

TUtheor

mg L⫺1

TUtheor

mg L⫺1

TUtheor

mg L⫺1

TUtheor

Reference

2.84 0.38 0.10 3.36 n.d. n.d. 6.68 277.78

120 2,150 230 11,300 n.d. n.d. — —

9.54 303.82 0.05 2.92 n.d. n.d. 316.33 49.02

245 0.022 35,200 16,500 4,950 0.30 — —

0.67 0.003 8.27 4.26 c 531.11 0.86 545.17 38.03

255 ⬍0.003 2,400 1,813 n.d. n.d. — —

2.79 0.00 0.56 0.47 n.d. n.d. 3.82 30.12

USEPA (1985) Bringmann and Kuhn (1977) Mount et al. (1997) Mount et al. (1997) Chapman et al. (1980) Fargasˇova´ (1994) This work This work

a

TUtheor calculations based on 24-h LC50 values for D. magna found in literature. n.d.: not determined. c TUtheor calculations based on 48-h LC50 value for D. magna. b

Final


Fig. 1. Percent toxicity reduction for D. pulex (24 h) of a final tannery effluent resulting from Phase I TIE fractionation techniques.

for both the baseline and fractionation tests. To cope with what had been a major pH drift, ⫾0.5 units, from the baseline for the zeolite, pHi/aer, and pH3/aer tests, they were compared with the most appropriate graduate test, resulting in a pH difference of ⫾0.1 units. As shown in Figure 1, the absence of %TR by EDTA and Na2S2O3 addition, by activated carbon, and pH11/fil indicated that toxicity was not a result of bivalent cationic metals or oxidative chemicals (Hockett and Mount, 1996). The toxic effect was reduced by air stripping, with %TR ranging between 64% and 76%. During and after air stripping precipitation was observed at pH 3 and pH 11, but bioassays were performed with the supernatant only. Filtration at pH 3 resulted in removal of 57% of the toxicity, suggesting that the %TR in the pH3/aer test was from removal of toxicants by precipitation resulting from pH change alone. In contrast, the toxicity reduction mechanism in the pH11/aer and pHi/aer tests was volatilization or oxidation, as its respective filtration tests showed no toxicity reduction (%TR ⫽ 0; Fig. 1). Toxicity reductions of 46% and 29% were found by passing the samples on the amberlite cation and the zeolite exchange resin, respectively, suggesting an inorganic cation as toxicant. This finding was strengthened by the lack of toxicity reduction of the anion

49

exchange resin and the 50% TR of the mixed exchange resin, probably because of the %TR of the cation exchange resin only. The toxicity of the carbon-treated effluent was not reduced, reinforcing our suspicion that the toxicant was inorganic, although some inorganic chemicals can also be adsorbed to carbon (Mount and Hockett, 2000). Because chemical analyses were lacking after completion of the fractionations, we compared our TIE profile with profiles of chromium, sulfide, ammonia, or major ions like sulfates and chlorides, which are typical inorganic compounds for tannery wastewaters. Given that Cr6⫹ is exchanged by anionic exchange resins (Mount and Hockett, 2000) and easily adsorbed by activated carbon (Linstedt et al., 1971), we concluded that Cr6⫹ was not present in toxic amounts. If Cr3⫹ were present in toxic amounts, significant toxicity reductions by pH 11 filtration (Mene´ ndez and Dıáz, 1998) and also a poor %TR by the strong sulfonic cation resin (Tiravanti et al., 1997) would have been observed. From the Phase I TIE profile, we concluded that chromium (Cr3⫹, Cr6⫹) did not account significantly for the observed toxicity. The amount of S2⫺ in the final effluent (⬍ 0.003 mg L⫺1) suggested that sulfide was not responsible for the observed toxicity. A pH-dependent toxicity, a considerable reduction of toxicity by air stripping at high pH (pH ⫽ 11) and by passing the samples through cation exchange resins, zeolite and activated carbon indicated ammonia as the responsible toxicant (Van Sprang and Janssen, 1997). According to Dickerson et al. (1996) toxicity by total dissolved solids (TDS) should be considered when conductivity exceeds 3000 ␮S/cm at the LC50 concentration. Although the final effluent demonstrated a mean conductivity of 1520 ␮S/cm (data not shown), the TDS could be responsible for the observed toxicity according to the Phase I TIE profile (Fig. 1). Thus, the significant %TR, ranging between 46% and 76%, suggested that the toxic compound or group of compounds consisted of a volatile or easily oxidized, filterable compound that precipitates or degrades at extreme pH and/or an inorganic cation. Given the Phase I TIE profile and the physical– chemical characterization shown in Table I, we conclude that ammonia and/or TDS were the main causes of the observed toxicity. The combination of chemical separation methods and biotesting without final chemical analyses provides information about the nature of toxic compounds, although it does not allow the identification of the toxic agents so other TIE phases are needed.

Characterization and Toxicity of an Aerobic Treated Beamhouse Effluent Results from the AS system showed that the efficiency of both BOD5 and COD removal decreased as the OLR increased from 0.3 to 6.0 g COD L⫺1 d⫺1 (Table III). In the

50

COOMAN ET AL.

TABLE III. Physical– chemical results and acute toxicity of beamhouse effluent to Daphnia spp. before and after activated sludge treatment Influent

Effluent Operation Time

Parameter* pH COD BOD5 TS TVS TDS NONH3 Cl⫺ S2⫺ SO2⫺ 4 D. magna 24-h LC50 D. magna 48-h LC50 D. pulex 24-h LC50 D. pulex 48-h LC50

45 Days 7.7 3780 1460 7230 2895 3030 24 2950 0.06 31 26.32b 14.46b 20.47b 12.25b

(22.08–34.38) (11.20–16.66) (16.38–25.48) (9.43–15.39)

127 Days 8.4 2070 1070 n.d.a n.d. 1510 39 2500 n.d. 66 8.66c 5.76c 8.66c 8.66c

(—) (4.94–6.72) (—) (—)

45 Days 8.0 186 13 5060 805 3440 123 2140 0.01 1100 68.30c 60.87c 70.30c 53.66c

(63.84–73.08) (53.41–69.37) (—) (42.91–67.11)

127 Days 7.6 1270 35 5470 2110 5470 111 1900 0.21 1125 65.98c 65.98c 68.30c 67.65c

(60.12–72.41) (60.12–72.41) (63.94–73.08) (61.14–74.85)

* All parameters are expressed in mg L⫺1 except for pH and 24-h and 48-h LC50 (%). a Not determined. b Probit method. c Trimmed Spearman–Karber method. Value in parentheses is 95% confidence interval.

best case (less than 100 days of operation), BOD5 removal in the AS was greater than 99%, whereas in COD removal was more than 60% contained a fraction recalcitrant to biological treatment. The bioassays done with D. magna and D. pulex revealed that both species responded in the same manner to the influent and effluent and that the toxicity of the biologically treated beamhouse effluent could be reduced only partially. The 24-h to 48-h LC50 values of the treated effluent ranged between 54% and 68%. Klinkow et al. (1998) reported that an anaerobic–aerobic treatment of tannery effluents reduced the 1/EC50 values for Vibrio fischeri 50-fold compared with the untreated sample. They also reported that the remaining organic compounds, mainly high-molecular-weight proteins, were unable to penetrate through the cell walls of the outer membrane or the slime layers surrounding bacteria and that therefore their toxic contribution should not be considered. Considering the physical– chemical characterization of the treated effluent, we suggested that the remaining toxicity was caused by inorganic compounds such as ammonia. Indeed, the effluent generated at OLR of 0.5 g COD L⫺1 d⫺1 demonstrated a ratio of TUtheor(NONH3)/TUobserved ranging between 0.97 and 1.13 (data not shown), suggesting no interaction between the identified compounds. However, the effluent generated at OLR of 1.9 g COD L⫺1 d⫺1 demonstrated a ratio of about 1.5; in this case, the toxicity could be explained by ammonia and chloride, indicating concentration additivity. In conclusion, the different process stages separately generate strong wastewater. The Phase I TIE results showed

that toxicity of the final effluent was caused by ammonia and/or TDS. Through additional fractionations and chemical analyses in combination with bioassays, the real toxic cause has to be further confirmed. Applying the TIE technique to indicate the group of toxic compounds of each process wastewater can help the tannery industry to develop cleaner production procedures. The AS reactor of the beamhouse effluent significantly reduced the COD and BOD5 loads and partially removed the toxicity for D. magna and D. pulex. The reduced COD load simplifies later toxicant identification. Comparing the observed toxicity with the expected toxicity, inorganic compounds such as ammonia and chloride were indicated as the toxic compounds of the beamhouse-treated effluent for both daphnids. This study demonstrates that further investigations have to be done to see if the remaining toxicity to daphnids of aerobically treated tannery wastewater is a result of inorganic compounds. Author KC sincerely thanks the Flemish Cooperative Agency (VVOB) for financial support.

REFERENCES Anderson PD, Weber LJ. 1975. The toxicity to aquatic populations of mixtures containing heavy metals. Proceedings International Conference on Heavy Metals in the Environment, Toronto, Ontario, Canada; 1975, October 27–31; p 933–953.


APHA-AWWA-WPCF. 1985. Standard methods for examination of water and wastewater, 16th ed. Washington, DC. Bringmann G, Kuhn R. 1977. The effects of water pollutants on D. magna. Z Wasser-Abwasser-Forsch 10:161–166. Catarino J, Lanç a A, Silva L, Mendonç a E, Picado A, Fernandes A, Peneda M, Pinho MN. 2001. New environmental technologies in leather industry wastewater decontamination. Proceedings of 11th Annual Meeting of SETAC Europe, Madrid, Spain; May 6 –10; 158 p. Dickerson KK, Hubert WA, Bergman HL. 1996. Toxicity assessment of water from lakes and wetlands receiving irrigation drainwater. Environ Toxicol Chem 15:1079 –1101. Fargasˇova´ A. 1994. Toxicity of metals on Daphnia magna and Tubifex tubifex. Ecotoxicol Environ Saf 27:210 –213. Finney DJ. 1971. Probit analysis. Cambridge, U.K.: Cambridge University Press. 333 p. Finney DJ. 1978. Statistical method in biological assay. London: Charles Griffin & Co. Ltd. 508 p. Hockett JR, Mount DR. 1996. Use of metal chelating agents to differentiate among sources of acute aquatic toxicity. Environ Toxicol Chem 15:1687–1693. Klinkow N, Oleksy-Frenzel J, Jekel M. 1998. Toxicity-directed fractionation of organic compounds in tannery wastewater with regard to their molecular weight and polarity. Water Res 32: 2583–2592. Konrad C, Lorber KE, Me´ ndez R, Lo´ pez J, Mun˜ oz M, Hidalgo D, Bornhardt C, Torres M, Rivela B. 2002. System analysis of material fluxes at tanneries—an international comparison. J Soc Leather Technol Chem 86:18 –25. Lindstedt KB, Houck CP, O’Connor JT. 1971. Trace element removals in adversed wastewater treatment process. Water Poll Control Fed 43:1507–1513. Mene´ ndez C, Dıáz M. 1998. Tannery wastewater treatment (in Spanish). Tecnologıá de agua 183:27–32. Mount DR, Gulley DD, Hockett JR, Garrison TD, Evans JM. 1997. Statistical models to predict the toxicity of major ions to Ceriodaphnia dubia, Daphnia magna and Pimephales promelas (fathead minnows). Environ Toxicol Chem 16:2009 –2019. Mount DR, Hockett JR. 2000. Use of toxicity identification evaluation methods to characterize, identify, and confirm hexavalent

51

chromium toxicity in an industrial effluent. Water Res 34:1379 – 1385. Norberg-King TJ, Mount DI, Durhan EJ, Ankley GT, Burkhard LP, Amato JR, Lukasewycz MK, Schubauer-Berigan MK, Anderson-Carnahan L. 1991. Methods for aquatic toxicity identification evaluations: Phase I toxicity characterization procedures for samples exhibiting acute and chronic toxicity, EPA/ 600/6-91/003. Washington, DC: U.S. Environmental Protection Agency. Chapman GA, Ota S, Recht F. 1980. Effects of water hardness on the toxicity of metals to Daphnia magna. Corvalis, OR: U.S. Environmental Protection Agency. 17 p. Rivela B, Bornhardt C, Vidal G, Me´ ndez R. 2002. Towards a cleaner production in developing countries: a case study in a Chilean tannery. Waste Manage Res (submitted). Song Z, Williams CJ, Edyvean RGJ. 2000. Sedimentation of tannery wastewater. Water Res 34:2171–2176. Tiravanti G, Petruzzelli D, Passino R. 1997. Pretreatment of tannery wastewaters by an ion exchange process for Cr(III) removal and recovery. Wat Sci Tech 36:197–207. Tisˇler T, Zagorc-Konç an J. 2001. Toxicity potential of substances in the tannery wastewater. Proceedings of 11th Annual Meeting of SETAC Europe, Madrid, Spain, May 6 –10. 6 p. Tu¨ nay O, Kabdas¸li I, Orhon D, Cansever G. 1999. Use and minimization of water in leather tanning processes. Water Sci Tech 40:237–244. U.S. Environmental Protection Agency [USEPA]. 1985. Ambient Water Quality Criteria for Ammonia. EPA/440/5-85-001. Washington, DC: U.S. Environmental Protection Agency. USEPA. 1993. Methods for measuring the acute toxicity of effluents and receiving waters to freshwater and marine organisms. EPA-600/4-90-027F. Cincinnati, OH: U.S. Environmental Protection Agency. Van Sprang PA, Janssen CR. 1997. Identification and confirmation of ammonia toxicity in contaminated sediments using a modified toxicity identification evaluation approach. Environ Toxicol Chem 16:2501–2507. Yatrabi A, Nejmeddine A. 2000. Fractionnement et mobilite´ des me´ taux lourds dans un sol en amont des eaux use´ es de tanneries. Rev Sci Eau 13:203–212.