ORIGINAL RESEARCH Blackwell Oxford, International IDT Society 1364-0307 54 of UK Dairy Publishing, Journal Technology of Ltd. Dairy2005 Technology
O RI GI NA L RESEA RCH
ORIGINAL RESEARCH
Comparative study of the efficacy of three coagulants in treating dairy factory waste water A HAMDANI,1 * M M OUNT ADAR 2 and O AS S OB HE I 1 1
Laboratory of Applied Microbiology and Biotechnology and 2Department of Analytical Chemistry, Faculty of Sciences, PO Box 20, El Jadida, 24000, Morocco
The treatment of dairy factory waste water by coagulation and decantation has shown that calcium hydroxide at a weak dose of 0.49–0.63 g provides the highly efficient removal of suspended matter (SM) (94%) and total phosphorus (TP-P) (89%) accompanied by an average elimination of chemical oxygen demand (COD), total Kjeldahl nitrogen (TKN-N), faecal coliforms (FC) and faecal streptococci (FS). This is within the Moroccan limits for the first two parameters (SM and TP-P). The dose necessary to obtain optimal removal is 0.8–1.2 g when using aluminium sulfate and 0.6–0.75 g with iron chloride and the percentage elimination of chemical and bacteriological pollutants is not substantial. However, coagulation by calcium hydroxide induced less sludge (0.93 g/L) than either aluminium sulfate (1.21 g/ L) or iron chloride (1.38 g/L). In terms of cost, the price of treating 1 m3 of dairy effluent by using calcium hydroxide is lower (approximately 25 times less expensive) than when using the other two coagulants. Keywords Chemical–physical treatment, Coagulation–decantation, Comparative study, Dairy effluent, Pollution.
I N T RO D U C T I O N
*Author for correspondence. E-mail:
[email protected] © 2005 Society of Dairy Technology
Although the composition of the effluents resulting from the dairy industry is variable according to the type of activity and depends on the manufacturing process implemented, their common characteristic is a mixture of organic matter, nitrogen, phosphorus and bacteria (Khoudir et al. 1997; Longhurst et al. 2000; Garrido et al. 2001; Hamdani et al. 2001). The treatment of this polluted waste water has always been one of the major concerns of industrialists, both for the protection of the environment as required by legal standards and for the reuse and recycling of purified water. In the majority of cases, the recommended treatments can be chemical– physical and/or biological, and require the removal of a large part of the organic load, often accompanied by the elimination of nitrogen and phosphorus (Rotereau 1969; Moletta and Torrijos 1999; Hamdani 2002). The experimental tests carried out in this study lie within the scope of pretreatment of dairy effluent, which aims to eliminate the particulate and colloidal components of the polluting load by using aluminium sulfate, iron chloride and calcium hydroxide, the most common coagulants (Degremont 1989; Baptiste 1994; Desjardin et al. 1996). The performances of the three coagulants were compared in
terms of removal efficiencies of parameters of pollution, the doses of reagents added to the effluent, the quantity of sludge produced, and the costs involved. The effects of pH and duration of agitation on the coagulation–decantation process were also studied. M AT E R I A L S A N D M E T H O D S
Sampling Coagulation–decantation tests were carried out on an average sample of 50 L, representative of 24 h of ejected waste water according to the flow. Samples of effluent were taken from the drain that received the total liquid ejected by the dairy unit. Chemical–physical analyses of dairy effluent before and after treatment were carried out according to the methods described in the AFNOR standard (Association Francaise de Normalisation 1986). The following parameters were measured: pH, suspended matter (SM), chemical oxygen demand (COD), total Kjeldahl nitrogen (TKN-N) and total phosphorus (TP-P). Bacteriological analyses comprised the enumeration of faecal streptococci (FS) on Bartley and Slanetz agar (Biokar Diagnostic, Beauvais, France) at 37°C, and the numeration of faecal coliforms (FC) by the method of the most probable number
Vol 58, No 2 May 2005 International Journal of Dairy Technology
83
Vol 58, No 2 May 2005
Table 1 Characteristics of coagulants used and parameters analysed
Crude product used
Chemical formula
Coagulation agent
Concentration (%)
Parameter analysed
White granulated powder (17% alumina)
Al2(SO4)3·18H2O
Al3+
2
Liquid solution (1 L contains 600 g of FeCl3)
FeCl3·6H2O
Fe3+
2
Limestone
Ca(OH)2
CaO
2
pH COD SM TKN-N TP-P FC FS
Table 2 Quality of dairy effluent submitted to coagulation–decantation tests
pH
SM Total COD (mg/L) (mg/L)
Value 7.1 900 Moroccan project of norms 6.5–8.5 50
5000 500
Soluble COD (mg/L)
TKN-N (mg/L)
TP-P FC FS (mg/L) (cfu/mL) (cfu/mL)
4100 —
123 30
27 10
2.7 × 104 4.5 × 104 — —
after culture on Brilliant Green Lactose Bile Broth (Biokar, France) and incubation at 44.5°C.
Treatment tests Treatment tests were carried out at laboratory temperature (24 ± 2°C) using a Jar-test (Model JF/ 6, Isco, Italy) with the following conditions: fast speed of 150 rev/min for 3 min, followed by slow speed of 30 rev/min for 20 min, and finally decantation for 90 min. The coagulants used and their characteristics are summarized in Table 1. Effectiveness of treatment was assessed analytically by following the rate of abatement of SM, COD, TKNN, TP-P, FC and FS. Chemical–physical sludge was determined by measurement of the SM of the particles flocculated for each reagent tested. SM concentrations were measured according to the AFNOR standards mentioned above. R E S U LT S A N D D I S C U S S I O N
Characterization of dairy effluent before treatment Table 2 shows the values obtained for each parameter analysed and reveals that the effluent studied had a high concentration of organic matter, nitrogen and phosphorus, and indicates bacteria from faecal contamination. The contents of total COD, TKN-N and TP-P were, respectively, 5000, 123 and 27 mg/ L. The bacterial load for FC and FS was 2.7 × 104 and 4.5 × 104 cfu/mL, respectively. The ratio of FC/ FS < 1 proves the animal origin of faecal contamination of dairy effluent (Borrego et al. 1982). The quality of this contamination exceeds the limits fixed by Moroccan standards, reinforcing 84
© 2005 Society of Dairy Technology
Figure 1 Efficiency of removal of parameters of pollution as a function of coagulant added to dairy effluent.
the necessity for treatment before ejecting it in the area, or possibly reusing it.
Treatment by coagulation–decantation Abatement of chemical–physical parameters Figure 1 reveals that the rates of abatement of COD are about 39% with the use of aluminium sulfate and 35% with iron chloride. With regard to SM, although rates of 89% and 91% were obtained, respectively, for the aluminium-based coagulant and the ferric chloride, the Moroccan standard of 50 mg/ L was exceeded for this parameter. Figure 1 also shows that coagulation by aluminium sulfate produced an abatement of 28% for total nitrogen (Kjeldahl) and 33% for total phosphorus. For iron chloride, these rates were 27% for both parameters. Compared with the two other coagulants, calcium hydroxide achieved higher rates of abatement of pollution (Figure 1). Doses of 0.63, 0.56 and 0.49 g/ L of calcium hydroxide added to the dairy effluent
Vol 58, No 2 May 2005
Figure 2 Efficiency of removal of parameters of pollution as a function of dose of coagulant added to dairy effluent.
led to abatement of 94, 89, 40 and 30%, respectively, for SM, TP-P, COD and TKN-N (Figures 1 and 2). As the effect of coagulation relates primarily to the separation of fine or superfine particles and of colloids of the interstitial phase by precipitation (Guettier et al. 1994), the weak rates of abatement of COD by all the coagulants can be explained by the predominance of the soluble fraction, which represents 82% of the total COD (see Table 2). In addition, the beneficial action on the removal of TP-P by calcium hydroxide results from the calcic formation of precipitated hydroxyapatite, especially at high pH values (8–12) (Degremont 1989): Ca(PO4H)2 + Ca(OH)2 → Ca2(PO4)2 + 2H2O Abatement of bacteriological parameters Figure 1 show that the three coagulants contribute to a partial reduction of the bacterial density contained in the dairy effluent. This is because the flocs formed, once eluted, take with them bacterial colonies. Rates of abatement of 33–47% and 32– 43% were recorded for FC and FS, respectively.
Effect of pH pH is known to influence the rate of abatement of pollution contained in waste water (Robert and Sheldon 1996). In our study, optimal pH values for the elimination of the SM, COD, TKN-N and TP-P ranged between 6 and 6.5 for the aluminic coagulant (Figure 3) and between 6.5 and 7 for the ferric coagulant, except for the abatement of SM obtained at pH 8 (Figure 4). These values of pH are close to 6, which allows an optimal abatement, according to Taha et al. (1995). The latter worked on the optimization of the coagulation–decantation of white water of dairies while varying the pH and the amount of added coagulants (‘acid cracking’). However, these pH values are higher than those noted in other studies using slightly acid solutions (Lefebre and Legube 1990; Desjardin et al. 1996). In addition, we noted that the progressive addition of the two coagulants with crude dairy waste water caused a fall in the pH from 7 to 6.1 for the © 2005 Society of Dairy Technology
Figure 3 Removal of parameters of pollution by aluminium sulfate.
Figure 4 Removal of parameters of pollution by iron chloride.
aluminium-based coagulant and to 6.4 for the ferric. This fall in pH can be explained by the fact that the addition of iron or aluminium salts to water involves a release of H+ ions according to the following hydrolysis reaction: Al3+ + nH2O 7 Al(OH)n + nH+ Fe3+ + nH2O 7 Fe(OH)n + nH+ These values are in the range of the optimal pH of coagulation obtained for these two coagulants, thus not requiring any correction of the pH of the effluent. Contrary to the two preceding coagulants, and as Figure 6 shows, the progressive addition of calcium hydroxide increased the pH from 7 to 12. It should be noted that the rate of abatement of SM and the speed of precipitation of phosphorus increased beyond pH 8 and the optimum was obtained with pH close to 11 (Figure 5).
Effect of duration of agitation In seeking optimization of the pH and the amount of coagulant, we also examined the effect of different durations of agitation. The principal results are presented in Table 3, which shows that the maximum abatement is obtained after 20 min agitation. Beyond 20 min, the rate of abatement of organic matter, nitrogen and phosphorus stops increasing and then stabilizes. This duration of 20 min would thus be the minimum time necessary for the calcium hydroxide, iron or aluminium flocs to adsorb a fraction of the 85
Vol 58, No 2 May 2005
Table 3 Effect of duration of agitation at 30 rev/min on percentage rate of abatement of COD, SM, TKN-N, TP-P, FC and FS with coagulants Al3+, Fe3+ and CaO SM COD TKN-N TP-P FC FS Duration 3+ 3+ 3+ 3+ 3+ 3+ 3+ 3+ 3+ 3+ (min) Al Fe CaO Al Fe CaO Al Fe CaO Al Fe CaO Al Fe CaO Al3+ Fe3+ CaO 3 5 10 15 20 25 30
39 68 77 90 91 87 89
25 58 72 91 91 92 89
42 69 85 94 96 95 —
12 21 29 35 37 36 35
10 16 24 31 33 32 31
12 19 27 38 41 40 —
10 18 26 32 38 37 40
07 14 22 30 32 31 30
9 17 24 30 31 29 —
06 14 21 29 33 32 30
04 08 13 17 25 23 22
15 37 59 90 93 91 —
9 15 26 32 40 41 40
8 13 20 30 34 35 33
12 21 35 43 50 48 —
10 15 28 33 39 39 38
8 14 22 31 36 34 35
11 18 33 39 45 45 —
pH values are fixed at 6.5, 8 and 11, respectively, for the use of aluminium sulfate, iron chloride and calcium hydroxide Dose of coagulating agent: 1200 mg of aluminium sulfate; 750 mg of iron chloride; 630 mg of calcium hydroxide
Table 4 Concentration of SM produced by coagulation–decantation with three coagulants tested Al3+ Fe3+ CaO
Dose of coagulant (g /L) Concentration of SM (g/L) Dose of coagulant (g /L) Concentration of SM (g/L) Dose of coagulant (g /L) Concentration of SM (g/L)
0 — 0 — 0 —
0.2 0.41 0.15 0.52 0.07 0.3
0.4 0.58 0.3 0.79 0.14 0.45
0.8 0.9 0.6 1.28 0.28 0.74
1 1.15 0.75 1.38 0.35 0.81
1.2 1.21 0.9 1.47 0.42 0.88
— — — — 0.49 0.93
— — — — 0.56 0.96
— — — — 0.63 0.98
Figure 5 Removal of chemical and bacteriological parameters by calcium hydroxide.
Figure 6 pH as a function of the calcium hydroxide dose added to diary effluent.
pollutant expressed in terms of COD, SM, TKN-N, TP-P, FC and FS.
coagulants (1.21 g/ L for aluminium sulfate and 1.38 g/ L for ferric chloride). The large amounts of the chemical reagents based on aluminium or containing iron can pose problems for the management of these coagulums and consequently increase the cost of treatment.
Chemical–physical production of sludge Coagulation–decantation of the dairy effluent generates a sludge containing the suspended particles, organic matter, nitrogen and phosphorus that are removed from dairy waste water, and also a quantity of the chemical reagents used in the form of precipitates. The results obtained are presented in Table 4 and show that the concentration in SM-flocculated particles increases gradually following the addition of the coagulants to dairy waste water. Concerning the maximum optimal amounts found, we note that coagulation–decantation by calcium hydroxide induced less sludge (0.98 g of SM per litre of dairy effluent) than the two other 86
0.6 0.76 0.45 0.95 0.21 0.67
© 2005 Society of Dairy Technology
Use of the chemico-physical sludge produced Given its composition (the presence of fertilizing and mineral elements, the absence of toxic and pathogenic elements), agricultural use can be considered for this type of sludge. It embodies a distinct agronomic richness that could be exploited either as an organic soil conditioner or as a mineral fertilizer. Vidou (1984) indicated that the solid phase of dairy sludge is rich in organic nitrogenated compounds whose mineralization in the ground is
Vol 58, No 2 May 2005
Table 5 Estimation of cost of three coagulants used to treat the dairy effluent
Coagulant 3+
Al Fe3+ CaO
Cost*
Dose (kg/m3 treated water) 1.2 0.75 0.63
DH/m3 treated water
DH/kg eliminated COD
2.040 2.475 0.090
1.080 1.412 0.038
*One euro is equivalent to 11 ± 1 DH (Dirham)
Table 6 Principal results of comparative study of the three coagulants Coagulating agent 3+
Al Fe3+ CaO
Optimal dose (g/L)
Optimal pH
0.8–1.2 0.6 –0.75 0.49–0.63
6–6.5 6.5–8 11–12
Rate of abatement (%) SM
COD
TKN-N
TP-P
FC
FS
Quantity of sludge (g SM/L)
89 91 94
39 35 40
28 27 30
33 27 89
36 33 47
38 32 43
1.21 1.38 0.98
Cost (DH*/m3 treated water) 2.040 2.475 0.090
*One euro is equivalent to 11 ± 1 DH (Dirham)
extensive, and in which 60–90% of the nitrogen transforms into mineral nitrogen that can be rapidly assimilated by the plants. In addition, contrary to certain sludges of urban or industrial origin, those coming from the liquid waste processing of dairies is devoid of toxic substances, or they are present at concentrations lower than the limits of standard AFNOR U44-041 (Vidou 1984; Agence de Bassin Loire Bretagne 1989; Sachon 1990; Guettier et al. 1994). The absence of phytotoxic effects favours agricultural use, which seems to be the most obvious application of this type of sludge, allowing for limitations implied by restrictions of spreading of sludge. According to Hassen et al. (1994), the bacterial load contained in waste water is always eliminated and concentrated in sludge, which represents a potential risk for a possible use in agriculture. In this respect, it is fortunate that the incidence of calcium hydroxide in the treated effluent causes a rise in the pH that stops bacterial growth and inhibits the fermentable capacity of sludge. Another final treatment process that could be applied (Hamdani et al. 2004) is incineration, but the problems of cost and implementation are a disadvantage.
Cost of treatment by coagulation–decantation The cost of purification can be expressed on the basis of 1 L of milk processed or the quantity of eliminated pollution. However, the variability of the dairy unit studied led to important changes in the polluting loads of the waste water produced by each workshop (the load generated by 1 L of fresh milk is not same as that resulting from 1 L of cream, for example). Given this situation, and so that the base of comparison is significant, we estimated the cost © 2005 Society of Dairy Technology
price of the chemical reagents in terms of m3 of water treated and of kg of polluting load (COD) eliminated. The results are summarized in Table 5 and indicate that to treat 1 m3 of the total dairy effluent, the cost of calcium hydroxide is approximately 25 times less than that of the other two coagulants. CONCLUSION The tests of total dairy processing waste water by coagulation–decantation using aluminium sulfate, ferric chloride and calcium hydroxide showed that the last coagulent determined at the optimum conditions of dose (0.49–0.63 g/ L), pH (11–12) and duration of agitation (20 min) to obtain an effluent in conformity with Moroccan standards for two parameters, SM and TP-P, with respective abatements of 94% and 89%. Compared to the aluminium-based and ferric coagulants, as indicated in Table 6, calcium hydroxide thus provides the best results: • greater reduction of the polluting load; • minimal dosage of the coagulant; • low production of sludge; • much lower cost. Despite the good results recorded for the elimination of TP-P and SM by calcium hydroxide, this type of treatment remains a partial treatment that produces only a limited elimination of the organic and nitrogenated pollution, whose polluting potential is high. To ensure complete treatment, coagulation–decantation should be followed by biological treatment to reduce the pollution load to the minimum and to satisfy the limits fixed by Moroccan standards (Hamdani et al. 2004). 87
Vol 58, No 2 May 2005
REFERENCES Agence de Bassin Loire Bretagne (1989) Traitement des eaux usées de l’industrie agro-alimentaire. In Qualité de l’Environnement et Productions Alimentaires, pp 167– 188. France: Ministère de l’Agriculture. Association Française de Normalisation (1986) Eaux, méthodes d’essai. Recueil de normes françaises, Paris, France. Baptiste Ph (1994) Les traitements d’épuration avant admission sur une station collective. Rapport de stage sur l’épuration des effluents agro-alimentaires, Agence de l’Eau Adour Garonne, Toulouse, France. Borrego J J, Arrabal F and Romero P (1982) Study of the microbiological pollution of a Malaga littoral area. II. Relationship between faecal coliforms and faecal streptococci. In Vièmes Journées sur l’Étude des Pollutions. 2–4 Décembre, Cannes, France, pp 561–564. Degremont (1989) Memento technique de l’eau. Tome I. Techniques et Documentation Lavoisier, Paris, France. Desjardin R, Lengo K and Ndiong S (1996) Etude comparative de la performance de divers coagulants préhydrolysés sur l’enlèvement de la matière organique dissoute. Vecteur Environnement 29 37–45. Garrido J M, Omil F, Arrojo B, Mendez R and Lema J M (2001) Carbon and nitrogen removal from a wastewater of an industrial dairy laboratory with a coupled anaerobic filter-sequencing batch reactor systems. Water Sciences and Technology 43 249–256. Guettier Ph, Iwema J P, Mathian R, Raby D, Vachon A and Vidou P (1994) L’assainissement des agglomérations: Techniques d’épuration actuelles et évolutions. Etude inter agences numéro 27. France: Agences de l’eau. Hamdani A (2002) Caractérisation et essais de traitement des effluents d’une industrie laitière: aspects microbiologiques et physico-chimiques. Thèse de doctorat. Faculté des Sciences d’El Jadida, Maroc. Hamdani A, Assobhei O and Mountadar M (2001) Caractérisation et essais de dénitrification biologique d’un effluent de laiterie située dans la ville d’El Jadida (Maroc). Eau Industrie Nuisances 242 50–54. Hamdani A, Chennaoui M, Assobhei O and Mountadar M (2004) Dairy effluent characterisation and treatment by
88
© 2005 Society of Dairy Technology
coagulation–decantation. Dairy Sciences and Technology 84 317–328. Hassen A, Jedidi N, Kallali H, Ferchichi M, Gharbi A, Chebbi F, Chefai A, Saidi N, Shayeb H and Ennabli M (1994) Elimination des indicateurs bactériens et des métaux lourds au cours d’un traitement des eaux usées domestiques dans une station pilote semi-industrielle. Sciences et Techniques de l’Eau 27 34–41. Khoudir A, Lamari H and Louehli S (1997) Traitement biologique en litfixe d’une eau usee de laiterie. Eau Industrie Nuisances 203 37–39. Lefebre E and Legube B (1990) Coagulation par Fe(III) de substances humiques extraites d’eaux de surface: effet du pH et de la concentration en substances humiques. Water Research 24 591–606. Longhurst R D, Roberts A H C and O’Connor M B (2000) Farm dairy effluent: a review of published data on chemical and physical characteristics in New Zealand. New Zealand Journal of Agronomic Research 43 7–14. Moletta R and Torrijos M (1999) Traitement des effluents de la filière laitière. Techniques de l’Ingénieur F-1501 1–21. Robert J S and Sheldon J B D (1996) Coagulation and precipitation of a mechanical pulping effluent. I: Removal of carbon, colour and turbidity. Water Research 30 1169– 1178. Rotereau J-C (1969) Considérations générales sur l’industrie laitière et le problème des eaux résiduaires. Industries Alimentaires et Agricoles 6 1563–1577. Sachon G (1990) Traitement des eaux résiduaires d’industries agro-alimentaires. Actes du Colloque ‘Productions Alimentaires et Environnement’, 11 déc 1990, Paris, France. Taha S, Tremaudan G and Dorange G (1995) Etude comparative entre les techniques de coagulation–décantation et d’ultrafiltration frontale pour l’élimination du COT des eaux blanches. In Les procédés d’épuration des effluents et déchets des industries biologiques et alimentaires, pp 55–60. Cergy–Pontoise Editions Technique et Documentation Paris, France: Lavoisier. Vidou P (1984) Epandage des boues de stations d’épuration de laiteries. Etude publiée par l’Agence de Bassin Loire Bretagne, Orléans, France.
