ARTICLE IN PRESS

Water Research 39 (2005) 5055–5063

www.elsevier.com/locate/watres

Development of a reliable alkaline wastewater treatment

process: optimization of the pre-treatment step
M. Prisciandaroa, G. Mazziotti di Celsob, F. Vegliòa,
a
Department of Chemistry, Chemical Engineering and Materials, University of L’Aquila, Monteluco di Roio, L’Aquila, Italy
b
Department of Food Science, University of Teramo, Mosciano S. Angelo, Teramo, Italy
Received 15 February 2002; received in revised form 3 March 2004; accepted 21 December 2004

Abstract

Alkaline waters produced by caprolactam plants polymerizing the fibres of nylon-6 are characterized by a very high
alkalinity, salinity and COD values, in addition to the presence of recalcitrant organic molecules. These characteristics
make alkaline wastewaters very difficult to treat; so the development of the suitable sequence to carry out in a
depuration process appears of great interest.
The proposed general process consists of three main steps: first, pre-treatment for the acidification of the polluted
stream, second, a successive extraction of the bio-recalcitrant compound (noted as cycloexanecarboxysulphonic acid
(CECS)) and a final biological treatment. In particular, this paper deals with the pre-treatment step: it consists of an
acidification process by means of sulphuric acid with the concomitant precipitation of black slurries in the presence of
different substances, such as solvents, CaCl2, bentonite, several flocculants and coagulants.
The aim of this study is to set an experimental procedure, which could minimize fouling problems during sludge
filtration. The use of additives like bentonite seems to give the best results, because it allows good COD reductions and
a filterable precipitate, which avoids excessive fouling problems of the experimental apparatus.
r 2005 Elsevier Ltd. All rights reserved.

Keywords: Alkaline wastewaters; Bio-recalcitrant compounds; Mud precipitation; Pre-treatment

1. Introduction The detection of the appropriate treatment sequence

Industrial wastewater treatment has to be a suitable
arrangement of chemical, physical and biological
processes, to minimize the inhibitory, toxic or recalci-
trant effect of mineral and organic contaminants,
particularly on the microbiological agents of wastewater
treatment plant, in order to guarantee the role of
ultimate step to the biological stage (Donati et al., 1968;
Tempesti et al., 1976).
Corresponding author. Tel.: +39 862 434223;
fax: +39 862 434203.
E-mail address: veglio@ing.univaq.it (F. Vegliò).
plays a major role in those processes in which waste-
waters are particularly difficult to treat with a conven-
tional procedure, since they contain contaminants very
hard to neutralize (Hashim et al., 1992). This is the case of
alkaline wastewaters (AWW), like those produced by
caprolactam plants (Fantauzzi et al., 1998; Baccella et al.,
2000). On the other side, only effective solutions of
problems of increasing the profitability of petrochemical
plants makes it possible to produce competitive products.
A possible way is to raise the degree of processing of the
initial feedstock by recycling the byproducts, which may
create considerable problems for the environment, as in
the case of caprolactam plants (Semenov et al., 1997).

0043-1354/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2004.12.025
 ARTICLE IN PRESS
5056 M. Prisciandaro et al. / Water Research 39 (2005) 5055–5063
Alkaline waters produced by the caprolactam plants
polymerizing the fibres of nylon-6 are characterized by a
very high alkalinity, salinity and COD values, in
addition to the presence of recalcitrant organic mole-
cules (Baccella et al., 2000). These characteristics hinder
the industrial development of those processes, which
produce liquid wastes like AWW. This is a problem
encountered in the South of Italy, where about 150,000
tons of AWW have been stored in tanks, waiting to be
treated with an appropriate process.
In a previous work, Baccella et al. (2000) devised a
treatment process of AWW based on a pre-treatment
with CaCl2 and a subsequent aerobic biological stage by
means of selected bacterial strains, carried out in shake
flasks and in a laboratory-scale fermentor. These
preliminary experiments led to interesting results and
suggestions for the development of an enhanced
chemical–biological treatment method; in particular, it
was concluded that the pre-treatment with CaCl2
ameliorated AWW susceptibility to microbial treatment,
furthermore reducing waste COD.
It is worth noting that the main organic constituents
of AWW produced by caprolactam plants contain
carboxylic groups, and namely Na salts of cycloexane-
carboxysulphonate isomers and derivatives and Na salt of
2-aminocapronic acid, whose concentration in solution is
referred to as carboxylic group concentration (CGC).
This parameter was evaluated by alkaline titration after
previous acidification of AWW (Baccella et al., 2000).
Experiments reported in this work verified that among
carboxylic groups, cycloexanecarboxysulphonic acid
(CECS) is bio-recalcitrant as expected, while other
CECS extraction
to
regeneration
Ammine
CECS extraction
AWW
H2SO4
Black mud
precipitation
Co-precipitator
Ca(OH)2
Solid waste
Precipitate
treatment
Acidification Filtration
Fig. 1. Scheme of the complete AWW treatment process.
constituents, such as e-caprolactam and related poly-
mers, 6-aminocaproic acid and similar compounds can
be biologically removed, so the COD reduction experi-
mentally obtained during the biological treatment (in
aerobic conditions) is mainly due to the consumption of
these last organic compounds.
Based on previous results (Baccella et al., 2000), a
renewed process, which provides for extraction and
subsequent recycling of CECS to production process, has
been devised. Conventionally, since in AWW CECS is
present as an anion, the extraction takes place by means
of protonated secondary ammines (Brandani et al.,
1998a, b). Therefore, a preliminary acidification step is
required, in order to create the required experimental
conditions for the subsequent liquid–liquid extraction.
On the other hand, due to the acidification, a black mud
precipitate (e.g. benzoic acid) is produced, that is
impossible to remove by filtration. This problem can be
solved carrying out the acidification step with a suitable
solvent, which could not only reduce the precipitation of
black slurries, but also make them easier to filter.
The proposed reliable AWW treatment process is
reported in its complete flowsheet in Fig. 1.
Essentially, the process is divided into four stages:
acidification step with black mud precipitation, filtrate
precipitation, CECS extraction by means of ammine,
biological treatment of residual wastewaters with a
previous neutralization treatment. In particular, the pre-
treatment stage consists of an acidification step by using
sulphuric acid with the concomitant precipitation of black
slurries in the presence of different substances, i.e. solvents,
CaCl2, bentonite, different flocculants and coagulants.
Biogas
to
Anaerobic Aerobic discharge
Neutralization
degradation degradation
Calcium sulfate Slurry Slurry
Biological treatment
 ARTICLE IN PRESS
M. Prisciandaro et al. / Water Research 39 (2005) 5055–5063 5057

The present paper describes the experimental study of
the process pre-treatment step, aimed at establishing the
more suitable experimental procedure which makes
easier the sludge filtration, with a less fouling of all
subsequent equipments.
2. Materials and methods
AWW samples have been taken from an industrial
storage tank located in the South of Italy; their average
chemical compositions and properties are reported in
Table 1 (Baccella et al., 2000). Measurements of CGC,
COD, pH, sulphate have been carried out. The specific
protocol expressly devised to determine CGC is reported
in detail elsewhere (Baccella et al., 2000); COD is
measured as absorbance at 620 nm of the clear centri-
fugation supernatants after reaction with K2Cr2O7
(Baccella et al., 2000); supernatant pH is obtained with
a pH-meter Crison/Micro TT 2050; sulphate concentra-
tion is evaluated by turbidity measurements using a
UV–visible Cary 1E/Varina spectrophotomer at 410 nm
(Baccella et al., 2000). Moreover, sludge samples have
been analysed by means of a Philips SEM 505.
Filtration tests have been carried out by using
filtration equipment (filters of 0.8 mm diameter-
Millipore) connected to a vacuum pump; precipitate
weight has been calculated as a difference between
filter weight, alone and after filtration. Measurements
have been carried out both on wet precipitate (after
filtration) and on dry precipitate, i.e. after drying at
80 1C for 24 h.
Filter specific resistance has been experimentally
measured by means of the Buchner Funnel Test, which
allows a good estimate of sludge performance, by
evaluating free drainage filtrate volume, filtrate clarity
and sludge cake characteristics.
In particular, the following balance equation—
derived from the elementary theory of filtration,

Table 1
Composition of the alkaline wastewaters (from Baccella et al., 2000)

Range

PH 10.0–12.5
Density (kg/m3) 1.1–1.2
Dry weight at 105oC (%) 22–34
Water (%) 66–78
Cycloexanecarboxylic salt (%)a 0.2–2.0
N-Hexahydro benzoyl-5-aminocaproic acid sodium salt (%)a 0.1–1.8
2-Aminocaproic acid sodium salt (%)b 0.3–4.0
Cycloexanecarboxysulphonate (CECS) bisodium salt (%)b 8.0–12.5
Caprolactamec 0.02–1.60
Benzoic acid sodium salt (%)a 0.01–0.5
N-Hexahydrobenzoic 5-amino valeric acid sodium salt (%) 0.01–0.3
Adipic acid sodium salt (%)a 0.01–0.25
1-Cyclometilhexancarboxylic acid sodium salt (%)a 0.03–0.25
5-Ethyl di-hydro 2 (3H0 furanone (%)c 0.01–0.25
5-Methyl valero lactone (%)c 0.01–0.25
Other organic products (excluded PCB-dioxines-pesticides-hydrocarbon aromatic polycycles) (%) 0.005–0.020
Sodium sulphate (%) 0.7–1.5
Sodiumchloride (%) 2.0–4.5
Ammonia (%) 0.02–0.15
Total alkalinity expressed as Na2CO3 (%) 2.5–9.0
Total phenols (mg/l) 60–150
Monochloro phenol (mg/l) 1–20
2,4-Dichlorophenol (mg/l) 0.01–2.0
2,4,6-Trichlorophenol (mg/l) 0.01–1.0
Toluene (mg/l) 0.05–10.0
Benzene (mg/l) 0.01–0.50
N-Hexane (mg/l) 0.01–0.50
Nitrites (mg/l) 50–100
Nitrates (mg/l) 150–350
COD (mg/l) 325,000–350,000
a
Neutral components.
b
No extractable components.
c
Acid components.
 ARTICLE IN PRESS
5058 M. Prisciandaro et al. / Water Research 39 (2005) 5055–5063

assuming laminar flow of filtrate liquid through the
cake—has been used (Bailey and Ollis, 1986):
dV DPA2
¼ ,
dt mðaW þ rAÞ
where DP is the pressure drop across the filter, A is the
area of filtering surface, m is filtrate viscosity, a the
average specific cake resistance, V is filtrate volume, r is
the resistance coefficient of filter medium and W the
mass of accumulated dry cake solids ¼ [rw/(1
mw)]V, r
is filtrate density, w is the mass fraction of solids in the
slurry and m is the ratio of wet to dry-cake mass.
Eq. (1) can be solved assuming that the cake is
incompressible and a is constant. In these hypotheses the
following solution is obtained:
t marw mr
¼ Vþ ¼ bV þ b1 .
V 2DPA2 ð1
mwÞ DPA
Eq. (2) shows a linear relation between the ratio t/V
and V, whose slope b allows to determine the value of a
(Bailey and Ollis, 1986).
Experimental runs have been organized into three series,
in order to investigate about the precipitation of black
slurries in the presence of different substances, namely
 ethyl acetate as a solvent (I series);
Two sample types, e.g. wastewater coming from the
storage tanks as is and concentrated wastewater, have
been tested with a CECS concentration equal to 0.5 and
0.8 mol/l, respectively, the concentrated AWW has been
(1)
obtained by means of an evaporation plant with two
effects and stored in the industrial area. Samples have
been tested to measure the amount of precipitate, its
nature and filterability.
Table 2 reports the complete I series of experiments,
they have been carried out using H2SO4 as acidification
agent with and without the organic solvent. In
particular, Fig. 2 shows the weight variations in
precipitates after solvent adding for four experimental
runs, in particular, a great reduction of precipitates in
those samples added with the solvent (runs 3 and 4) is
observable, with a removal of about 70% in weight.
(2) Moreover, it is well clear that wet precipitates behave in
the same manner as dry precipitates. No relevant
differences are noticeable between the concentrated
sample (runs 2 and 4) and the AWW as is sample (runs
1 and 3).
As for filtration results, Table 2 shows the filtration
velocity after the treatment with ethyl acetate in H2SO4.
40
wet
Precipitate weight kg/m3 AWW

 CaCl2 (II series); 35

dry
 bentonite, with and without coagulants and floccu- 30
lants (III series).
25

20

3. Results and discussion 15

3.1. I Series 10

5
Black slurries are mainly made by organic polymers,
so they can be dissolved by a solvent, simplifying the 0
1 2 3 4
acidification step. During the first experimental series
Runs
the effect of ethyl acetate on the precipitation of black
slurries has been analysed. Fig. 2. Slurry precipitation in the presence of ethyl acetate.

Table 2
Precipitation and filtration experimental results with ethyl acetate

Run no. AWW Ethyl acetate H2SO4 (kg/ pH Wet Dry Filtration Notes
(m3/m3 m3 AWW) precipitate precipitate rate (103 m3/
AWW) (kg/m3) (kg/m3) h)

1 Norm. 0 98.1 1.6 34.36 21.34 0.037

2 Conc. 0 98.1 2.7 32.56 20.71 0.037
3 Norm. 1 147.2 1.6 9.36 3.03 0.400
4 Conc. 1 147.2 2.8 11.85 3.81 0.400
5 Conc. 1 147.2 2.8 — — 0.400 Acetonea
6 Conc. 1 147.2 2.8 — — 0.400 Silicab
a
0.2 m3/m3 AWW.
b
40 kg/m3 AWW.
 ARTICLE IN PRESS
M. Prisciandaro et al. / Water Research 39 (2005) 5055–5063 5059

When ethyl acetate is added in solution, reaction tanks
appears less affected by fouling problems, especially in
case of non-concentrated AWW sample (run no. 3 and
run no. 4); this behaviour is due to the solution of some
substances in the solvent.
Results indicate that when ethyl acetate is added in
solution, particularly to non-concentrated AWW sam-
ple, reaction tanks appear cleaner as a result of solvent
adding, i.e. part of the substances present in the
wastewater sample has been dissolved in the solvent.
This is an important aspect from the practical point of
view, because preliminary precipitation tests, carried out
in an industrial-scale stirred tank, have shown the
unfeasibility of this process for its dirty characteristics.
Moreover, the adding of the solvent improves the
filterability rate.
In summary, results of the first experimental series
suggest that solvent addition has a positive outcome,
as it lowers precipitate amount, increases filtration
rate and finally reduces fouling problems on reaction
vessels.
3.2. II Series
(b) CaCl2 and H2SO4 have been added in sequence to
the wastewater sample (CaCl2 at first, then H2SO4);
(c) CaCl2 and H2SO4 have been simultaneously added
to the wastewater sample.
Figs. 3i and ii and 4 show the results for case a, when
only H2SO4 is added in AWW, in particular Fig. 3i and
ii reports the effects of an increasing pH, due to a raised
amount of 10% sulphuric acid in weight (about 1 M) in
solution. Moreover, as the quantity of sulphuric acid in
solution increases from 15 to 75 kg/m3 AWW (Fig. 3ii),
pH value decreases from 8 to 1 (Fig. 3i), while the
amount of precipitates grows from 10 to 35 kg/m3
AWW. This behaviour is probably due to the strong pH
drop off, which allows the formation of insoluble
compounds (C6H5COOH) and makes some substances,
already present in AWW such as caprolactam dimers or
trimers, insoluble.
20
H 2 SO 4 2.5M
Precipitate weight (kg/m3)

H 2 SO 4 5.9M

During the second experimental series mud precipita-

15
tion has been analysed adding in solution 2 M calcium
chloride as a co-precipitator. Ca2+ ions combines with
SO2
4 ones produced by sulphuric acid dissociation, to
crystallize at a specific concentration as calcium sulphate 10
dihydrate (CaSO4  2H2O, gypsum), a sparingly soluble
substance (Prisciandaro et al., 2001). Gypsum precipi-
tates together with slurry, leading to a precipitate
globally easier filterable and less adhering to vessel walls. 5
Second series can be divided into three experimental 0 10 20 30 40 50 60
cases H2SO4 (kg/m3 AWW)

(a) H2SO4 and CaCl2 have been separately added to the Fig. 4. Effect of different sulphuric acid concentration on
wastewater sample; precipitation.

35
Precipitate weight (kg/m3)

30

25

20

15

10
0 2 4 6 8 10 20 30 40 50 60 70 80
pH H2SO4 (kg/m3 AWW)

Fig. 3. (i and ii) Sulphuric acid and pH effects on slurry precipitation.

 ARTICLE IN PRESS
5060 M. Prisciandaro et al. / Water Research 39 (2005) 5055–5063

To analyse the effect of sulphuric acid on the (iii)

Wet
precipitate formation, two different molar concentra- 500
Dry
tions have been used, namely 2.5 and 5.9 M. Results in 400
Fig. 4 show that a more concentrated acid produces
300
lower quantities of precipitate. This is an important issue
of the acidification step, because it allows to minimize 200

the solid waste disposal of the purification process; 100

anyway this effect is not so relevant mainly due to the

Precipitate weight (kg/m3)

low amount of H2SO4. Wet (ii)
500 Dry
Table 3 shows the results for case a, when only CaCl2
is added in AWW, as for H2SO4, the more CaCl2 is 400
added, the more precipitate phase is produced. How- 300
ever, in this case the black mud seems to be more 200
granulated and filterable than the one achieved during
100
the H2SO4 case, which appeared more gluing and
adhesive to reactor walls. SEM analysis has shown the
CaCl2 (i)
presence of CaSO4 on this black sludge, where gypsum 500 H2SO4
formations is well visible.
400
These results have suggested to carry out the
experiments by precipitating the largest slurry amount 300

with calcium chloride, in order to limit the formation of 200

black mud in the acidification step.
Experiments belonging to case b have been performed
carrying out precipitation with CaCl2 followed by
H2SO4 acidification, subsequently. Considering the
results above, concentrated sulphuric acid (96%) has
been used. These tests have been performed to verify if
precipitates produced with CaCl2 are the same of those
ones produced by acidification. The results achieved
have been reported in Fig. 5i, as expected, the greatest
amount of precipitates has been obtained after CaCl2
adding, thus improving the conditions for the subse-
quent acidification. Moreover the slurry produced seems
to be more manageable, probably due to its coexistence
with a gypsum precipitate. Anyway the next acidifica-
tion step has not showed a significant amount of
precipitates, so, it is possible to conclude that both
CaCl2 and sulphuric acid can be suitable to precipitate
the insoluble organic compounds.
Further experiments reported in Fig. 5ii aimed to
verify the possibility to complete the precipitation of all
slurries, by adding increasing calcium chloride quanti-
ties. However, results have demonstrated that black mud
precipitation in the acidification step cannot be avoided.
As a matter of fact, the same figure shows that
Table 3
Precipitation results with calcium chloride.
Run no. CaCl2 2 M Precipitate weight
(kg/m3 AWW) (kg/m3 AWW)
1 1.1 7.9
2 11.0 14.1
3 33.0 25.7
4 55.0 39.6
100
0
50 60 70 80 90 100 110 120 130 140
CaCl2 (kg/m3 AWW)
Fig. 5. Precipitation results: (i) precipitate weight as a function
of CaCl2 concentration before and after acidification with
sulphuric acid; (ii) precipitate weight after CaCl2 treatment; (iii)
precipitate weight after simultaneous CaCl2 and H2SO4
addition.
precipitate quantity increases with calcium chloride
until the value of about 100 kg/m3 AWW, where the
curves for both wet and dry precipitate reach a plateau.
Finally, simultaneous addition of calcium chloride
and sulphuric acid (case c) has been carried out, to
perform the acidification of AWW, aimed to the next
liquid–liquid extraction of CECS, and to make the
precipitate treatable with usual solid–liquid operation
units, too. These results have been reported in Fig. 5iii,
which shows that at least for wet precipitate, its amount
is lower than the one obtained with the separate pre-
treatments (CaCl2 and H2SO4 in sequence).
This circumstance can be explained considering that,
as a matter of fact, in the presence of CaCl2 in solution,
the higher alkalinity of wastewater causes also CaCO3 to
precipitate in addition to gypsum; when two steps are
carried out simultaneously, the acidity of the solution
allows CO2 to escape, thus inhibiting carbonate pre-
cipitation. This behaviour is proved by the abundant
foaming in the reactor, due to the gaseous CO2
formation according to the following reaction:
CO2
þ
3 þ 2H ! H2 O þ CO2 " . (3)
 ARTICLE IN PRESS
M. Prisciandaro et al. / Water Research 39 (2005) 5055–5063 5063

treatment of alkaline industrial waste waters. Process
Biochem. 35, 595–602.
Bailey, J.E., Ollis, D.F., 1986. Biochemical Engineering
Fundamentals. McGraw-Hill Int., Nw York.
Brandani, S., Brandani, V., Vegliò, F., 1998a. On the
purification of b-naphtalensulfonic acid from dilute acqu-
eous solutions containing sulfuric acid. Ind. Eng. Chem.
Res. 37 (11), 4528–4530.
Brandani, S., Brandani, V., Vegliò, F., 1998b. Extraction of
anions from aqueous solutions using secondary amines. Ind.
Eng. Chem. Res. 37 (1), 292–295.
Donati, G., Sioli, M., Taverna, M., 1968. Caprolattame da toluene:
il processo SNIA-viscosa. Chim. Ind. (Milan) 50, 997–1001.
Fantauzzi, E., Baccella, S., Chiarini, M., Vegliò, F., Toro, L.,
Cerichelli, G., Lepidi, A., 1998. Development of a chemical
and biological treatment of alkaline industrial waste waters:
a preliminary study. Fresen. Environ. Bull. 7, 934–950.
Hashim, M.A., Kulundai, J., Hassan, R., 1992. Biodegrad-
ability of branched alkylbenzene sulphonates. J. Chem.
Technol. Biotechnol. 54, 207–214.
Naseem, R., Tahir, S.S., 2001. Removal of Pb(II) from
aqueous/acidic solutions by using bentonite as an adsor-
bent. Water Res. 35 (16), 3982–3986.
Prisciandaro, M., Lancia, A., Musmarra, D., 2001.
Gypsum nucleation in NaCl solutions. AIChE J. 47,
929–934.
Semenov, M.V., Litvintsev, I.Y., Komarov, A.G., 1997.
Recycling of the aqueous waste of caprolactam production
with the isolation of commercial adipic acid. Petrol. Chem.
37, 531–536.
Tempesti, L., Giuffrè, G., Buzzi, F., Serena, E., 1976.
Monomeri. An investigation in to the kinetics of reaction
between cyclohexane-carboxilic acid and oleum. Chim. Ind.
(Milan) 58, 247–251.