Environment International 31 (2005) 297 – 304

www.elsevier.com/locate/envint

Treatment of olive mill effluents

Part II. Complete removal of solids by direct flocculation
with poly-electrolytes
Roika Sarika, Nicolas Kalogerakis, Dionissios Mantzavinos*
Department of Environmental Engineering, Technical University of Crete, Polytechneioupolis, GR-73100 Chania, Greece

Available online 5 November 2004

Abstract

The pre-treatment of three different olive oil processing effluents by means of direct flocculation (i.e. without prior coagulation) was
investigated. Four cationic and two anionic poly-electrolytes were tested and most of them were found capable of removing nearly
completely total suspended solids (TSS) as well as reducing considerably the concentration of chemical (COD) and biochemical oxygen
demand (BOD5) without altering solution pH. Flocculant dosage was crucial to achieve effective separation. For three cationic and one
anionic poly-electrolytes, the minimum dosage required to initiate separation was about 2.5–3 g/L. The remaining two poly-electrolytes failed
to cause separation even at dosages as high as 7 g/L. Lime and ferric chloride were also tested as reference coagulants and found quite
effective in terms of TSS removal although the degree of COD reduction was generally lower than that with poly-electrolytes. However, lime
treatment would require greater dosages and longer treatment times than that with poly-electrolytes and would also increase considerably
solution pH. A preliminary cost analysis showed that lime treatment for complete solids removal was generally less costly than that with
poly-electrolytes presumably due to its low market price. Nonetheless, cost–benefits may be defied by several drawbacks associated with the
use of lime.
D 2004 Elsevier Ltd. All rights reserved.

Keywords: Coagulation; Flocculation; Olive mill effluents; Poly-electrolytes; Treatment

1. Introduction the dissolved organic matter of OME. As OME also contain

significant amounts of suspended and colloidal matter at
Olive mill effluents (OME) constitute a serious environ- concentrations as high as 190 g/L (Mulinacci et al., 2001;
mental problem in the Mediterranean Sea region due to the Zouari, 1998), successful treatment may require a pre-
unique features of this type of agro-waste discussed in detail conditioning stage to remove the suspended and colloidal
elsewhere (Mantzavinos and Kalogerakis, 2004). fractions which typically consist of pectins, proteins, oils
Several treatment technologies based on biological and and tannins. This can be done by means of separation
advanced oxidation processes have been proposed to deal processes and several case studies have been reported in the
with OME and an overview of them is given in the first part literature. Turano et al. (2002) developed a two-stage
of this communication (Mantzavinos and Kalogerakis, centrifugation–ultrafiltration process capable of reducing
2004). These processes are employed either alone or in the total suspended solid (TSS) and concentration of
various combinations and aim at degrading predominantly chemical (COD) contents of OME by 80% and 90%,
respectively. Interestingly, fats could easily be recovered as
they were completely rejected by the membrane. Al-Malah
* Corresponding author. Tel.: +30 28210 37797; fax: +30 28210
et al. (2000) described a three-stage process comprising
37847. centrifugation, filtration and adsorption on bentonite clay to
E-mail address: [email protected] (D. Mantzavinos). treat OME. Centrifugation and filtration resulted in nearly
0160-4120/$ - see front matter D 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.envint.2004.10.006
298 R. Sarika et al. / Environment International 31 (2005) 297–304

complete TSS removal followed by only about 25% COD 2. Experimental and analytical
and 60% phenols removal. However, clay adsorption was
capable of decreasing the residual COD and phenolic 2.1. Materials
contents by as much as 70% and 80%, respectively. The
use of sodium polyacrylate super-absorbent polymers to OME from two three-phase olive mills (W1 and W2) as
detoxify OME has been reported in a recent study (Davies et well as the wash-waters (W3) from the final centrifugation
al., 2004). The absorbent was capable of retaining the of a two-phase mill (all three mills were located in the region
polyphenolic fraction inside the polymeric chains, while of Chania, Crete) were collected and stored at 20 8C. The
large molecules such as proteins were concentrated in the major effluent properties are shown in Table 1.
residual liquid phase whose phyto-toxicity was substantially Four cationic (FO-4700-SH, FO-4490-SH, FO-4350-
reduced. SHU and FO-4190-SH) and two anionic (FLOCAN 23
Coagulation with materials such as lime, alum, ferric and AN 934-SH) poly-electrolytes manufactured by SNF
chloride and ferrous sulphate has also been extensively Floerger were provided by ChemFlo-Hellas. In all cases, the
employed in water/wastewater treatment. Beccari et al. appropriate mass of polymer was dissolved in de-ionized
(1999) proposed a process comprising OME pre-treatment water to give 1% w/v feedstock solutions. This was done
by means of lime coagulation and adsorption on bentonite since addition of polymers directly in the effluent proved
followed by anaerobic digestion. Pre-treatment was capable difficult due to their moderate solubility in water. Calcium
of removing oleic acid, polyphenols and COD by as much oxide (N96%) and ferric chloride (~97%) were supplied by
as 99.5%, 43% and 55%, respectively. The resulting effluent Fluka and used as received.
was subject to 1:1.5 dilution and then treated anaerobically;
for the same dilution level, the original un-treated OME was 2.2. Experimental procedure
not amenable to anaerobic degradation. In further studies
(Aktas et al., 2001), the impact of lime pre-treatment on Experiments were carried out in a jar-test apparatus (Velp
total solids, COD and polyphenols removal for 17 different Scientifica) equipped with six beakers of 1 L volume.
OME was evaluated. OME treatment with lime or ammo- Effluent samples were removed from the freezer, thawed to
nium iron sulphate resulted in about 40–70% COD and ambient conditions and analyzed with respect to their initial
color removal depending on the coagulant dosage employed TSS, dissolved COD and dissolved BOD5 concentrations.
(Zouari, 1998). The resulting effluent was more readily Prior to COD and BOD5 analyses, samples were filtered
degradable anaerobically than the original un-treated efflu- through 0.45-Am disposable filters. The samples were then
ent. Treatment of OME subject to lime or alum coagulation thoroughly shaken, for re-suspension of possible settled
has also been demonstrated by Flouri et al. (1996) and Lolos solids and 300 ml of the sample was transferred to the
et al. (1994). Other recent studies report the use of electro- beaker. The appropriate dosage of flocculant or coagulant
coagulation over iron and aluminium electrodes (Inan et al., was then added while stirring at 200 rpm for 5 to 60 min,
2004) and lime coagulation coupled with the Fenton’s after which separation was complete and sample from the
reagent (Vlyssides et al., 2003) for OME treatment. supernatant was taken and analyzed with respect to the
Although lime is an inexpensive and readily available concentration of TSS, COD and BOD5.
material, lime coagulation generally has several drawbacks, For the runs carried out with poly-electrolytes, the
namely: (i) increase of solution pH and hardness, (ii) low dosage was adjusted using a measured volume of the
COD removal capacity typically between 20–40%, (iii) need respective 1% solution. It should be mentioned that feed-
for large dosages leading to the generation of excessive stock solutions at various concentrations greater than 1%
volumes of sludge which is still quite phyto-toxic and and up to 5% were also prepared but their use proved
cannot be readily disposed of and (iv) need for coupling impractical and, therefore, it was discarded. The 1%
with flocculation to improve efficiency (Tatsi et al., 2003; solution, although slightly viscous, was still easily manage-
Vlyssides et al., 1996). able, while as concentration increased to 2%, the flocculant
The objective of this work was to study for the first time
the destabilization of OME by means of direct flocculation
(i.e. without the addition of coagulants) using various Table 1
Physico-chemical properties of the effluents studied in this work
cationic and anionic poly-electrolytes. Effluents taken from
two-phase and three-phase olive mills were treated to assess Properties W1 W2 W3
the effect of parameters such as flocculant type and dosage, Type of mill Three-phase Three-phase Two-phase
Type of olive Koroneiki Koroneiki Tsounati
treatment time and stirring speed on the removal of TSS,
processed and Liani
COD and biochemical oxygen demand (BOD5). In addition, TSS, mg/L 42 400 101 850 540
lime and ferric chloride were used as reference coagulants COD, mg/L 89 200 57 100 4500
and compared to poly-electrolytes. A preliminary cost BOD5, mg/L 22 800 23 200 1000
analysis was also carried out to evaluate treatment costs BOD5/COD 0.26 0.41 0.22
pH 5.4 5.5 6.7
associated with the materials in question.
 R. Sarika et al. / Environment International 31 (2005) 297–304 299

formed a gel and at concentrations greater than 5% the

solution turned virtually into a solid-like gel. Solutions at
concentrations below 1% were not tested since their use
would unnecessarily increase the effluent volume.
For the runs with lime or ferric chloride, the appropriate
mass of coagulant was added directly in the sample.
Changes in solution pH were measured with a GLP-21
Crison pH-meter.

2.3. Analytical techniques

TSS concentration was determined filtering a well-mixed

sample through a Pall (type A/E) glass-fiber filter and then
drying the residue retained on the filter at 103 8C for 60 min.
The increase in weight of the filter corresponds to TSS.
COD was determined by the closed reflux, colorimetric
method. The appropriate amount of sample was introduced
into commercially available digestion solution (Hach
Europe, Belgium) and the mixture was then incubated
for 120 min at 150 8C in a COD reactor (Model 45600-
Hach, USA). COD concentration was measured colori-
metrically using a DR/2010 spectrophotometer (Hach,
USA).
BOD5 was assessed by measuring the oxygen uptake in a
sample over a period of 5 days at 208C, in the dark using the
BODTrak apparatus (Hach, USA). To maintain the con-
trolled conditions required, tests were performed in a BOD
incubator (Hach, USA). One milliliter of sample was added
to the BOD bottle and diluted with 99 ml of un-seeded
aerated water. As dissolved oxygen was consumed, air Fig. 1. Effect of treatment time and stirring on separation. (a) Treatment of
pressure in the bottle dropped and pressure changes were 300 ml W2 with 4.33 g/L FO-4700-SH at 45 rpm; (b) treatment of 300 ml
continuously recorded by the instrument’s sensors and W1 with 3 g/L FO-4350-SHU for 5 min.
converted (taking into account sample dilution) to BOD5
concentrations. practically no effect on separation efficiency expressed in
Imhoff settling cones of 1-L capacity were used to terms of TSS and COD removal (data for 15 min are not
measure the volume of the resulting liquid and solid phases shown). In further experiments with W1 treated with 3 g/L
following separation. The settling time was 5 and 60 min for FO-4350-SHU which corresponds to 90 ml of flocculant
the runs with flocculants and coagulants, respectively. It solution for 5 min, the effect of stirring conditions was
should be noted that, for the runs with poly-electrolytes, assessed with the beaker contents being stirred either at
TSS, COD and BOD5 values of the liquid after treatment 200 rpm or through bubbling air in the beaker. As seen in
were corrected for the dilution due to the addition of the Fig. 1b, no effect of stirring on separation was recorded.
flocculant solution and the adjusted volumes were used to The rationale behind the use of air is that aeration may
compute the respective removals. prove beneficial if flocculation pre-treatment is to be
followed by aerobic biodegradation. Fig. 1 also shows
the ratio (S/T) of sludge generated following separation
3. Results and discussion over the total initial volume (i.e. effluent sample and
flocculant solution); this ratio remains practically
3.1. Experiments with poly-electrolytes unchanged with varying treatment time or stirring con-
ditions. In all subsequent runs with poly-electrolytes,
Preliminary runs were performed to screen the effect of treatment time was set for 5 min while stirring at 200 rpm.
treatment time and stirring conditions on separation and Figs. 2 and 3 show the effect of flocculant dosage on
representative results for certain samples and flocculants separation efficiency. Samples W1 and W2 were treated
are shown in Fig. 1. Experiments with W2 treated with with various dosages of FLOCAN 23 and FO-4490-SH,
4.33 g/L FO-4700-SH (this corresponds to 130 ml of respectively, and the results are shown in Fig. 2. As seen,
flocculant solution) for 5, 15 and 30 min (shown in Fig. there appears to be a minimum flocculant dosage of about
1a) suggest that increasing treatment from 5 to 30 min had 2–2.5 g/L below which no separation occurred. However,
300 R. Sarika et al. / Environment International 31 (2005) 297–304

Fig. 4 shows the relative efficiency of various poly-

electrolytes to treat all three effluents. For the runs with W1
and W2, the flocculant dosage was about 3 g/L, while for the
runs with W3 the dosage was about two orders of magnitude
lower (namely, 0.033 g/L with FO-4490-SH and FLOCAN
23, 0.066 g/L with FO-4700-SH and 0.17 g/L with FO-4350-
SHU) since the concentration of W3 is far less than that of
W1 or W2. In all cases, the pH of the resulting liquid phase
was practically identical to that of the original sample. As
seen, nearly complete TSS removal was recorded in all cases;
however, the extent of COD and BOD5 removal as well as
the volume of sludge generated were found to depend on the
type of flocculant used. FO-4350-SHU appears to be the
most promising flocculant concerning the treatment of W2
since complete TSS removal was accompanied by about
55% COD and 23% BOD5 reduction. Furthermore, the
sludge generated was only about 20% of the initial volume
and this was well below the amount generated by other
flocculants; minimization of sludge production is important
considering the cost implications associated with subsequent
sludge treatment and disposal. It is also interesting to notice
that the extent of COD decrease after treatment of W2 with
poly-electrolytes other than FO-4350-SHU was always
lower than that of BOD5 decrease, thus implying that the
resulting liquid phase was less biodegradable aerobically (in
terms of the BOD5/COD ratio) than the original sample.
Conversely, the BOD5/COD ratio of the W2 sample treated
with FO-4350-SHU increased from 0.41 prior to treatment to
0.69 after treatment. This also occurred (although to a lesser
Fig. 2. Effect of flocculant dosage on separation. (a) Treatment of 300 ml extent) for sample W1 treated with FO-4350-SHU (Fig. 1b),
W2 with FO-4490-SH; (b) treatment of 300 ml W1 with FLOCAN 23. where the BOD5/COD ratio increased from its initial value of
0.26 to 0.29 after treatment. This may be an important
for concentrations greater than this value, TSS removal was consideration if the resulting liquid phase is to be post-
nearly complete regardless the dosage used. Moreover, treated biologically.
increasing dosage generally increased the extent of COD or
BOD5 removal. For instance, treatment of W1 with 3.33 g/L 3.2. Experiments with inorganic coagulants
FLOCAN 23 led to about 40% COD removal, twice as
much as that with 2.66 g/L of flocculant. Likewise, In further studies, calcium oxide and ferric chloride were
treatment of W2 with 3.33 g/L FO-4490-SH resulted in used to treat the effluents in question (without flocculants).
about 45% BOD5 removal, three times as much as that with
2.33 g/L of flocculant. Several experiments with W2 were
also performed during which the flocculant dosage was
increased stepwise (typically 0.33 g/L each time) in order to
evaluate the minimum amount required to achieve separa-
tion. Results for all six poly-electrolytes are shown in Fig. 3.
As seen, four of them (namely, FO-4700-SH, FO-4490-SH,
FO-4350-SHU and FLOCAN 23) exhibited similar behavior
with the minimum dosage ranging between about 2.3 and 3
g/L. On the contrary, FO-4190-SH and AN 934-SH failed to
cause separation at dosages even as high as 5 and 7 g/L,
respectively. No further attempts were made to determine
minimum dosage for these two poly-electrolytes since their
use in OME treatment is likely to be too expensive given
that their market price (as shown in Table 2) is comparable Fig. 3. Treatment of 300 ml W2 with various poly-electrolytes. Black bar:
with that of the rest; consequently, neither FO-4190-SH nor minimum dosage needed for separation; hatched bar: maximum dosage
AN 934-SH were tested further. tested for which no separation occurred.
 R. Sarika et al. / Environment International 31 (2005) 297–304 301

Table 2
Market price for bulk sales of flocculants and coagulants
Material FO-4700 FO-4490 FO-4350 FO-4190 FLOCAN AN 934 Lime FeCl3
Price, o/ton 2980 2800 2800 2800 2500 2550 130 450
With the exception of lime, prices are those quoted by ChemFlo-Hellas. For lime, the price was taken from a local kiln.

Screening experiments showed that, unlike the runs with increased to 15–60 min depending on the coagulant and
poly-electrolytes, treatment for 5 min was not sufficient to the type of effluent.
achieve successful separation. Therefore, the time was Fig. 5 shows data regarding treatment of W2 and W3 at
various CaO dosages. Unlike experiments with poly-
electrolytes, the use of lime expectedly increases the pH
of the liquid phase which, in turn, appears to affect treatment
efficiency. The pH range in which coagulation occurs may
be the single most important factor affecting separation and
optimal values are dependent on the type and dosage of
coagulant as well the effluent characteristics in question. At
the conditions employed in this study, pH effect was more
pronounced for the runs with W2; pH values between 11
and 12 (this corresponds to lime dosages between 6.7 and
13.3 g/L) resulted in successful separation. However,
increasing lime dosage to 20 g/L had a detrimental effect
on the process, possibly due to re-stabilization of the
colloids. Talaat et al. (2002) who studied the lime

Fig. 4. Relative efficiency of various poly-electrolytes to treat 300 ml of Fig. 5. Effect of CaO dosage on separation. (a) Treatment of W2 for 60 min;
(a) W1; (b) W2; (c) W3. (b) treatment of W3 for 15 min.
302 R. Sarika et al. / Environment International 31 (2005) 297–304

coagulation of agricultural drainage water with a TSS

concentration of 93 mg/L, reported that the extent of TSS
removal gradually decreased with increasing lime dosage
from 60 to 80 to 100 mg/L. Similar results were reported by
Zouboulis et al. (2004) who treated biologically stabilized
leachates (initial COD of 0.75 g/L) with alum at a constant
pH=5.5. Increasing alum dosage from about 0.2 to 0.5 g/L
improved COD removal from 15% to 50%; nonetheless,
further dosage increases up to 1.4 g/L decreased removal to
about 30%. Although lime treatment at pHc11 also serves
as a means of disinfection, treatment with poly-electrolytes
appears to be advantageous over lime since (i) treatment
time and dosage with poly-electrolytes are considerably
lower than those with lime to achieve comparable removals,
(ii) addition of lime changes the pH of the liquid phase and
(iii) lime generates excessive volumes of sludge. For
instance, the sludge generated after treatment of W2 with
lime (Fig. 5a) is about 200 ml, while the respective value
with FO-4350-SHU (Fig. 4b) is only 90 ml.
In a final set of experiments, ferric chloride was used as
coagulant to treat W1 and W3 and the results are
summarized in Fig. 6. Treatment of W1 with 3.7 g/L of
FeCl3 for 30 min led to 86% TSS removal. However,
increasing the dosage to 6.7 or 8.3 g/L had an adverse effect
on the process as FeCl3 failed to cause separation. At the
conditions under consideration, complete TSS removal was
achieved for W3 at a relatively low coagulant dosage;
however, this was accompanied by a moderate reduction of
COD. It should be pointed out that prior to adding the
coagulant, the pH of the original W1 or W3 sample was
adjusted from its ambient value to 5 with a measured
amount of KH2PO4. This was done since pH values in the
range 4.5–5.5 are known to favor FeCl3 performance
(Genovese and González, 1998).

3.3. Cost evaluation

A preliminary analysis was performed to evaluate the

cost associated with the various flocculants and coagulants
employed in this study. For all materials but lime, cost

Fig. 7. Cost of various materials for treatment of (a) W1; (b) W2; (c) W3.

estimates were based on the current market price (bulk

sales) in Greece as quoted by a major supplier (Table 2). For
lime, the market price was taken from a lime kiln located in
Chania, Western Crete. For all three effluent streams, the
cost of poly-electrolytes per kg TSS removed was calculated
according to the data shown in Fig. 4. For the streams
treated with inorganic coagulants (W2 and W3), calcula-
tions were based on the minimum dosage needed to
maximize TSS removal according to the results of Figs. 5
and 6. In addition, the annual cost was calculated according
to the following assumptions: the milling season in the
island of Crete typically lasts between November/December
Fig. 6. Treatment of W1 and W3 for 30 min with FeCl3. and February/March during which an average mill processes
 R. Sarika et al. / Environment International 31 (2005) 297–304 303

about 1500–1700 tons of olives. Given that olive processing treatment with polymers. In light of this, an optimum
usually generates about 1.5 m3 of wastewaters per ton of combination of coagulants and flocculants might be
olive processed, this would amount to about 2500 m3 OME. needed to improve process economics.
Cost estimates are shown in Fig. 7.
Treatment of OME (streams W1 and W2) with various Although flocculation appears to be a promising technol-
poly-electrolytes would cost no more than a few cents per ogy for OME pre-conditioning, it is evident that special
kg TSS removed. Of the various polymers tested, FLOCAN attention must be paid to ensure that the resulting liquid and
23 appears to be marginally better than the rest. Nonethe- solid phases are suitable for further treatment, disposal or use.
less, lime and ferric chloride are still substantially less Mineralization of the liquid phase would require some form
expensive than polymers presumably due to their lower of treatment possibly by means of advanced oxidation or
market price. The fact that treatment of W1 with polymers biological processes or a combination of them. In this view,
costs nearly twice as much as treatment of W2 (per kg TSS we are currently investigating the use of high power
removed) is due to the fact that a common flocculant dosage ultrasound for the post-treatment of the liquid phase. On
of about 3 g/L was used for both streams although TSS the other hand, it is well-documented that the solid fraction of
concentration in W1 is only half that in W2. various agro-wastes may be composted to yield soil fertilizers
The low initial TSS concentration of wash-waters W3 (Manios, 2004). In a recent study (Garcı́a-Gómez et al.,
explains the relatively low annual cost of treatment. 2003), sludge from the flocculation pre-treatment of two- and
However, it should be emphasized that data for the various three-phase OME (using a commercial poly-electrolyte) was
materials shown in Fig. 7c are not directly comparable to subject to composting with olive leaves, thus resulting in the
each other since experiments were carried out at different formation of biologically stabilized composts.
flocculant or coagulant dosages. This would explain the
rather unexpectedly high cost associated with lime or ferric
chloride since the dosage on which calculations were based Acknowledgments
(i.e. 2.5 g/L for lime and 0.67 g/L for ferric chloride) might
be well over the stoichiometric amount required for The authors wish to thank Mr. A. Hatzopoulos of
complete removal. ChemFlo-Hellas for donating the poly-electrolytes.

4. Conclusions References

To the best of our knowledge, this is the first report Aktas ES, Imre S, Ersoy L. Characterization and lime treatment of olive
regarding pre-treatment of OME by direct flocculation with mill wastewater. Water Res 2001;35:2336 – 40.
poly-electrolytes. The results are encouraging and key Al-Malah K, Azzam MOJ, Abu-Lail NI. Olive mills effluent (OME)
wastewater post-treatment using activated clay. Sep Purif Technol
conclusions can be summarized as follows: 2000;20:225 – 34.
Beccari M, Majone M, Riccardi C, Savarese F, Torrisi L. Integrated
(i) Poly-electrolytes are generally capable of removing treatment of olive oil mill effluents: effect of chemical and physical
completely the TSS fraction as well as reducing the pretreatment on anaerobic treatability. Water Sci Technol 1999;40:
organic matter in the aqueous phase. Coagulation with 347 – 55.
Davies LC, Novais JM, Martins-Dias S. Detoxification of olive mill
lime or ferric chloride, although nearly equally wastewater using superabsorbent polymers. Environ Technol 2004;25:
effective in terms of solids removal, results in lower 89 – 100.
organic matter reduction. Flouri F, Sotirchos D, Ioannidou S, Balis C. Decolorization of olive oil mill
(ii) Unlike lime coagulation which is strongly pH liquid wastes by chemical and biological means. Int Biodeterior
sensitive, flocculation with poly-electrolytes can Biodegrad 1996;38:189 – 92.
Garcı́a-Gómez A, Roig A, Bernal MP. Composting of the solid fraction of
operate at OME ambient pH (which is typically olive mill wastewater with olive leaves: organic matter degradation and
slightly acidic). biological activity. Bioresour Technol 2003;86:59 – 64.
(iii) Flocculation does not alter the pH of the resulting Genovese CV, González JF. Solids removal by coagulation from fisheries
liquid; on the contrary, lime treatment generates waste waters. Water SA 1998;24:371 – 2.
Inan H, Dimoglo A, Simsek H, Karpuzcu M. Olive oil mill wastewater
highly alkaline effluents and pH adjustment would
treatment by means of electro-coagulation. Sep Purif Technol 2004;
be needed if the stream was to be post-treated 36:23 – 31.
biologically. Lolos G, Skordilis A, Parissakis G. Polluting characteristics and lime
(iv) Poly-electrolytes are likely to generate less sludge precipitation of olive-mill waste-water. J Environ Sci Health, A 1994;
than lime. 29:1349 – 56.
(v) However, the operational cost associated with the use Manios T. The composting potential of different organic solid wastes:
experience from the island of Crete. Environ Int 2004;29:1079 – 89.
of poly-electrolytes should not be disregarded. A Mantzavinos D, Kalogerakis N. Treatment of olive mill effluents: Part I
preliminary analysis showed that treatment with lime Organic matter degradation by chemical and biological processes—an
or ferric chloride would be less expensive than overview. Environ Int 2004;31:289 – 95 (this issue).
304 R. Sarika et al. / Environment International 31 (2005) 297–304

Mulinacci N, Romani A, Galardi C, Pinelli P, Giaccherini C, Vincieri FF. Vlyssides AG, Bouranis DL, Loizidou M, Karvouni G. Study of a
Polyphenolic content in olive oil waste waters and related olive demonstration plant for the co-composting of olive-oil-processing
samples. J Agric Food Chem 2001;49:3509 – 14. wastewater and solid residue. Bioresour Technol 1996;56:187 – 93.
Talaat HA, Sorour MH, Rahman NA, Shaalan HF. Pretreatment of Vlyssides A, Loukakis H, Israilides C, Barampouti EM, Mai S.
agricultural drainage water (ADW) for large-scale desalination. Detoxification of olive mill wastewater using a Fenton process. In:
Desalination 2002;152:299 – 305. Kalogerakis N, editor. 2nd European Bioremediation; 2003. p. 531 – 4.
Tatsi AA, Zouboulis AI, Matis KA, Samaras P. Coagulation–flocculation Zouari N. Decolorization of olive oil mill effluent by physical and chemical
pretreatment of sanitary landfill leachates. Chemosphere 2003; treatment prior to anaerobic digestion. J Chem Technol Biotechnol
53:737 – 44. 1998;73:297 – 303.
Turano E, Curcio S, de Paola MG, Calabro V, Iorio G. An integrated Zouboulis AI, Chai XL, Katsoyiannis A. The application of bioflocculant
centrifugation–ultrafiltration system in the treatment of olive mill for the removal of humic acids from stabilized landfill leachates.
wastewater. J Membr Sci 2002;209:519 – 31. J Environ Manag 2004;70:35 – 41.