You are on page 1of 8

Renewable Energy 83 (2015) 455e462

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Enzymatic hydrolysis and anaerobic biological treatment of sh


industry efuent: Evaluation of the mesophilic and thermophilic
conditions
J.G. Duarte b, L.L.S. Silva a, D.M.G. Freire b, M.C. Cammarota a, *, M.L.E. Gutarra a, c
ria, Centro de Tecnologia, Bl. E, Sl. 203,
Department of Biochemical Engineering, School of Chemistry, Federal University of Rio de Janeiro, Cidade Universita
~o, 21941-909 Rio de Janeiro, Brazil
Ilha do Funda
b
ria, Centro de Tecnologia, Bl. A, Sl. 549,
Department of Biochemistry, Institute of Chemistry, Federal University of Rio de Janeiro, Cidade Universita
~o, 21949-900 Rio de Janeiro, Brazil
Ilha do Funda
c
Federal University of Rio de Janeiro, Campus Xer
em, Estrada de Xer
em, 27, Xer
em, 25245-390 Duque de Caxias, Brazil
a

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 23 July 2014
Accepted 23 April 2015
Available online 15 May 2015

Enzymatic hydrolysis and anaerobic treatment of efuent similar to that generated in the sh processing
industry were evaluated at 30  C and 50  C. Hydrolysis used lipase produced by fungus Penicillium
simplicissimum in solid state fermentation with babassu cake as substrate, which has optimal activity at
50  C. Hydrolysis kinetics was conducted with mixtures of efuent (containing 1500 mg oils and greases/
L) and different lipase activities (0e0.67 U/ml of efuent), verifying that with 0.16 U/ml of efuent,
9.69 mmol/ml of free acids were produced after 4 h at 50  C. Anaerobic biodegradation assays were
conducted with efuent submitted to three different treatments: thermophilic (hydrolysis and anaerobic
treatment at 50  C), mesophilic (hydrolysis and anaerobic treatment at 30  C) and hybrid (hydrolysis at
50  C and anaerobic treatment at 30  C). The best results (97.5% of chemical oxygen demand [COD]
removal and 105.4 ml CH4/g CODremoved) were obtained with the hybrid treatment in only 68 h. The
thermophilic hydrolysis not only reduced the amount of enzyme and the hydrolysis time but also
reduced the time and the cost of mesophilic anaerobic treatment, favoring the application of this
treatment on an industrial scale.
2015 Elsevier Ltd. All rights reserved.

Keywords:
Anaerobic treatment
Efuent
Enzymatic hydrolysis
Fish industry
Mesophilic
Thermophilic

1. Introduction
Fish industry efuents have high levels of chemical oxygen demand (COD), temperature, oil and grease, and total suspended
solids (TSS). They are generally discharged into water bodies after
receiving treatment that is incompatible with their degree of
pollution, causing serious environmental damage [1].
Although aerobic processes are traditionally used in the treatment of industrial efuents, anaerobic processes are gaining
prominence due to the advantages they bring, such as the removal
of organic matter and methane generation from high-load efuents
(COD > 4000 mg/L), such as those produced by the sh industry
[2e4].

* Corresponding author. Tel.: 55 21 3938 7568; fax: 55 21 3938 7567.


E-mail address: christe@eq.ufrj.br (M.C. Cammarota).
http://dx.doi.org/10.1016/j.renene.2015.04.056
0960-1481/ 2015 Elsevier Ltd. All rights reserved.

However, anaerobic processes are adversely affected by the high


oil and grease concentrations in these efuents, which can form
agglomerates or pellets in the sludge ocs, hindering sludge sedimentation and reducing the efciency of the treatment [1,5,6].
Alternatives to the pretreatment of efuents rich in fats to
reduce their oil and grease content and aid the subsequent biological treatment should therefore be investigated. One such potential alternative is enzymatic pre-hydrolysis with lipases
(glycerol ester hydrolase, E.C. 3.1.1.3), enzymes that generally act in
the aqueouseorganic interface, catalyzing the hydrolysis of triacylglycerols [7e11].
Enzymatic pretreatment is a way of improving the activity of the
microbial population in the subsequent biological treatment, since
it prevents the accumulation of fats in the sludge. In addition, it
permits the conversion of complex organic compounds in the form
of fats and proteins, which would be discarded as problematic solid
waste, into methane, which can be used as a source of energy in the
same industry.

456

J.G. Duarte et al. / Renewable Energy 83 (2015) 455e462

The application of these enzymes is growing steadily because of


their ability to catalyze a wide range of reactions, including the
hydrolysis of oils and greases in efuents from sh industry [7].
However, for the enzymatic pre-hydrolysis step in efuent treatment to be economically feasible on an industrial scale, less costly
ways of producing these enzymes, such as solid state fermentation
(SSF), should be adopted.
SSF can use agroindustrial waste as a support and substrate for
the growth of microorganisms that not only reduce the total cost of
the process, but are also biodegradable. Fermented solids containing enzymes called solid enzyme preparations (SEP), obtained
by SSF, can be used directly in the treatment of efuents, bypassing
the enzyme extraction and/or recovery steps. Thus, SSF assigns
values to a residue that would be discarded into the environment,
transforming it into a byproduct of great interest to various industries [11,12].
This study evaluated the anaerobic biological treatment of a sh
industry efuent at 30  C and 50  C, with and without preliminary
hydrolysis at 30  C and 50  C, using a pool of enzymes rich in
thermophilic hydrolase obtained by SSF.
2. Material and methods
2.1. Synthetic efuent
In view of the extreme variability of the composition of the industrial efuent used in a prior study [27], we chose to use a synthetic efuent whose composition was similar to the industrial one.
The synthetic efuent was prepared following a methodology
 n [1] using 160 g heads, viscera, bones, and tails of
proposed by Rollo
fresh sardines (Sardinella sp.) and 400 ml distilled water. The
mixture was homogenized in a blender at the slowest speed for
1 min, then sieved (1 mm average pore size) to obtain a concentrate. An aliquot of this concentrate was removed to characterize its
COD and oil and grease content, with the remainder being stored
at 20  C. At the time of use, the concentrate was diluted in
distilled water to obtain the desired oil and grease concentration,
and had its pH, COD, BOD5, total Kjeldahl nitrogen, total phosphorus, oil and grease, and total solids determined.
2.2. Microorganism and propagation medium for the production of
enzymes
The enzyme preparation for use in the hydrolysis was produced
using a strain of Penicillium simplicissimum as an inoculum, selected
for its capacity to produce lipases in different semi-synthetic mediums [13] and in a medium composed of babassu cake [14]. The
strain was propagated at 30  C for 7 days in a medium with the
following composition (% w/v): soluble starch, 2.0; MgSO4.7H2O,
0.025; KH2PO4, 0.05; (NH4)2SO4, 0.5; CaCO3, 0.5; yeast extract, 0.1;
olive oil, 1.0; and agar, 2.5. After 7 days of growth, spores were
scraped and suspended in a sterile phosphate buffer (50 mM, pH
7.0), forming a spore suspension whose concentration was determined by counting in a Neubauer chamber.
2.3. Solid state fermentation (SSF)
The raw material used as a fermentation medium was babassu
cake, a byproduct from the production of babassu oil (provided by
Tocantins Babau S.A.), which was supplemented with 6.25% (w/w)
molasses (waste product from sugar production). The cake obtained was ground in a Wiley mill to yield particles with diameters
of up to 3 mm, although only particles of <1.18 mm were used in the
production of the enzyme preparation. Fermentations were conducted in cylindrical tray-type reactors (15 cm height and 10 cm

diameter). Each tray contained 15 g of the babassu cake forming a


1 cm-deep layer to achieve a good aeration and heat transfer between the cake and the surrounding space. The reactors were
inoculated with 107 spores/g of dry babassu cake and incubated in a
chamber with controlled temperature (30  C) and humidity (90%)
for 72 h. Part of the fermented cake was sampled to quantify protease and lipase activity, and the remainder e the solid enzyme
preparation (SEP) e was stored under vacuum until use.
2.4. Submerged fermentation
Submerged fermentation was performed in 1000 ml Erlenmeyer
asks containing 240 ml of one of the two medium compositions.
The synthetic medium (C:N ratio of 8.9:1) had the following
composition (w/v): 1% meat peptone, 0.5% yeast extract, 0.05%
MgSO4, 0.05% KCl, 0.001% FeSO4$7H2O, 1% glucose, 0.3% olive oil,
diluted in 0.1 M sodium phosphate buffer at pH 7.6 [13]. The submerged fermentation with babassu cake used 2.5% (w/v) babassu
cake with particle size <1.18 mm suspended in distilled water. In
both fermentations, the medium was inoculated with
2  104 spores/ml, and the Erlenmeyer asks were incubated in a
shaker at 170 rpm and 30  C [15]. Fermentation was monitored by
measuring lipase activity, and was discontinued after 96 h. The
fermented broth was ltered through Whatman lter paper (11 mm
pore size) for biomass separation. An aliquot was used to determine
lipase and protease activity, and the remainder (liquid enzyme
preparation) was stored at 4  C until use.
2.5. Enzymatic hydrolysis
Enzymatic hydrolysis was carried out in 500 ml Erlenmeyer
asks containing 250 ml synthetic efuent with an initial concentration of around 1500 mg oil and grease/L, agitated in a shaker at
150 rpm and maintained at 50  C (thermophilic hydrolysis) or 30  C
(mesophilic hydrolysis). Initial tests were conducted with lipase
activity of 0.67 U/ml of efuent at 50  C with and without the
addition of 1% (w/v) sodium azide, an anti-microbial substance that
acts as a respiratory chain uncoupler. The azide was added to prevent the growth of microorganisms and the consumption of the
free acids produced so as to distinguish between the action of
enzymatic catalysis and that of microbial biodegradation. The kinetics of the thermophilic hydrolysis was evaluated by determining
the production of free acids over time by withdrawing 10 ml aliquots every 4 h up to 24 h. Having selected the best hydrolysis
condition (at 50  C), this was repeated without the addition of
sodium azide for subsequent use in anaerobic biodegradability
tests. Mesophilic hydrolysis was conducted for 8 h, as described by
Alexandre et al. [7], for subsequent use in anaerobic biodegradability tests.
2.6. Anaerobic biodegradability tests
The enzymatically pretreated efuent and the crude efuent
(without enzymes) were used in the anaerobic biodegradability
tests. The tests were conducted in 100 ml penicillin asks with
90 ml working volume of a mixture composed of anaerobic sludge
and raw or pretreated efuent. The amount of sludge used in each
biodegradability test was calculated to maintain an initial efuent
COD-to-sludge volatile suspended solids (VSS) ratio of 1:1. The
asks were sealed with rubber stoppers and aluminum seals and
incubated at 30  C or 50  C up to biogas volume stabilization.
The sludge was collected from a mesophilic upow anaerobic
sludge blanket reactor in operation at a poultry processing plant,
and was adapted to the synthetic efuent simulating the efuent
generated in sh industry. Sludge adaptation was conducted at

J.G. Duarte et al. / Renewable Energy 83 (2015) 455e462

30  C for mesophilic tests and 50  C for thermophilic tests. For this,


approximately 1/3 of the working volume of 1 L bottles was lled
with sludge, and were then topped up with a low-fat-content
synthetic efuent (COD 300 mg/L). The bottles were sealed with
rubber stoppers with a single outlet for the collection and measurement of the biogas produced by liquid displacement, and
incubated at 30  C or 50  C. When biogas production stabilized
(which took anything from 3 to 7 days), the supernatant was
replaced by fresh diluted efuent, whose COD was gradually
increased up to a maximum of 2000 mg/L.
The sludge was characterized for its total volatile solids content
(TVS e 12,000 mg/L) and specic methanogenic activity
SMA e 0:173 g CODCH4 =g VSS:d. SMA can be dened as the
maximum methane production capacity of a consortium of anaerobic microorganisms held under controlled conditions. It was
determined in 100 ml penicillin asks, with 10% headspace, without
stirring, at 30  C for 14 days, with an initial anaerobic sludge concentration of 2 g TVS/L and a substrate (mixture of volatile organic
acids e acetic, propionic and butyric) with 2.0 g COD/L, according
Chernicharo [3].
Biodegradability was measured by COD removal efciency and
biogas production, as assessed by the displacement of the plunger
of 20 ml graduated plastic syringes connected to the asks until the
complete stabilization of biogas production. After stabilization, the
volume of biogas in all the syringes was measured and the total
volume of each ask was used to calculate the biogas. The biogas
produced was collected and injected into a Varian Micro GC 4900
chromatograph to determine the percentage of methane. The asks
were then opened to remove samples for quantication of pH and
nal COD. Each condition was evaluated in ve replicates and the
results are presented as means and standard deviations.
2.7. Analytical methods
All the analyses required to characterize the efuent and sludge
and monitor the biodegradability tests were performed as
described in the Standard Methods [16]. Lipase activity was
measured by a spectrophotometric method based on the formation
of a chromophore product obtained from the lipase-catalyzed hydrolysis of p-nitrophenyllaurate [15]. One lipase unit was dened as
the amount of enzyme that released 1 mmol of p-nitrophenol under
assay conditions. Protease activity was measured by spectrophotometry as described by Bendicho et al. [17]. One protease activity
unit was dened as the unit difference in absorbance between the
reaction blank and the sample per minute under assay conditions.
The production of free fatty acids during hydrolysis was monitored
in an automatic titrator (end point pH 11) using 0.04 M NaOH as the
titrant.
3. Results and discussion

457

Table 1
Characterization of synthetic efuent and industrial efuent.
Parameter

Synthetic efuent

Industrial
efuent [27]

pH
COD (mg/L)
BOD (mg/L)
Total Kjeldahl nitrogen (mg/L)
Total phosphorus (mg/L)
Oil and grease (mg/L)
Total solids (mg/L)
Total xed solids (mg/L)
Total volatile solids (mg/L)

6.5e6.9
6000e15,767
2122
123.3
56.8
2111e6317
2271
208
2063

5.5e7.2
1313e12,333
967e2900
5.3e499
51e1400
78e3656
848e1485
140e230
618e1345

3.2. Enzymatic hydrolysis


SSF with babassu cake results in enzymatic preparations with
varying lipase activities that yield different free fatty acid levels in
the hydrolysis step [19]. Thus, in order to compare the results obtained for the synthetic and industrial efuent, hydrolysis assays
were conducted with different amounts of enzyme preparation,
measured by lipase activity per volume of efuent. Pre-hydrolysis
of a sh industry efuent containing 1500 mg oil and grease/L
with 0.2%, 0.5%, and 1.0% (w/v) P. simplicissimum solid enzyme
preparation (SEP) with lipase activity of 134 7 U/g at 30  C was
monitored for up to 18 h. The hydrolysis condition that resulted in
the highest methane volumes in the subsequent anaerobic treatment step was the one that used 0.5% SEP (or 0.67 U/ml of efuent)
for 8 h of hydrolysis with an output of 4.7 mmol/ml of free acids [20].
Thus, initial enzymatic hydrolysis tests (Test 1) were run at 50  C,
with and without the use of sodium azide, maintaining lipase activity of 0.67 U/ml of efuent. The results are shown in Table 2.
There was an increase in the concentration of free fatty acids
produced by the action of the SEP in both conditions (with and
without azide) and for both reaction times. Furthermore, it was
observed that similar concentrations of free acids were produced in
24 h by 0.67 U/ml of efuent with or without azide, indicating that
its use was not necessary for determining the optimal time and
enzyme activity for anaerobic biodegradability tests. Differing from
the ndings of Alexandre et al. [7], in the hydrolysis conducted at
30  C, a decline in the free acid content was observed due to the

Table 2
Production of free acids in the enzymatic hydrolysis of the synthetic efuent, containing 1500 mg oil and grease/L and different solid enzymatic preparation (SEP)
concentrations with lipase activity of 62 U/g after 4 h (DT4) and 24 h (DT24) of hydrolysis at 50  C. The control experiments were conducted with efuent without the
addition of SEP.
Condition

3.1. Efuent characterization


The synthetic efuent was prepared with viscera and parts of
heads, tails and bones of Sardinella sp., as shes' heads and viscera
contain the highest oil and fat content [18]. Due to the presence of
blood, mainly from the heads, the synthetic efuent had efuent
had a reddish color and high COD and oil and grease values, as can
be seen in Table 1. Nevertheless, except for total volatile solids
(TVS), these efuent's values fell within the ranges obtained for the
sh industry efuents. The relatively high TVS concentration of the
synthetic efuent was due to the inclusion of virtually all the suspended material produced when it was made, since sieving
removes only very coarse solids, while in industrial efuents, such
material is separated during sh processing.

Test 1 e Hydrolysis with and without


SEP and sodium azide
Efuent without SEP (control)
Efuent 0.67 U/mlb
Efuent 0.67 U/mlb 1% (w/v) azide
Test 2 e Hydrolysis with different levels
of lipase activity
Efuent without SEP (control)
Efuent 0.16 U/mlb
Efuent 0.32 U/mlb
Efuent 0.67 U/mlb

Free acid production


(mmol/mL)a

DT4

DT24

2.57
17.81
16.02

8.34
26.86
25.93

3.50
9.69
16.43
17.81

7.30
12.02
19.59
26.86

a
DTt Ct  C0, where Ct concentration of free acids at time t;
C0 concentration of free acids in the initial time (data not shown).
b
U/ml of efuent. Lipase activity obtained by addition of different quantities of
SEP containing average lipase activity of 62 U/g.

458

J.G. Duarte et al. / Renewable Energy 83 (2015) 455e462

aerobic microorganisms'use of the acids produced as a substrate.


This behavior was not observed in experiments carried out in the
presence of sodium azide, where the acid content continued to
increase throughout the assay [7].
As the synthetic efuent consisted only of water and sardines,
most microorganisms were part of the sh biota. The parts used to
produce the efuent were responsible for most of the bacterial
ora, which is usually concentrated in the intestines and gills of sh
[21]. The sardines used in the synthetic efuent were from the
Brazilian coast, i.e., sh from tropical waters with moderate temperatures. Most of the bacteria present in these organisms were
therefore mesophilic, which explains the low microbial activity
observed during hydrolysis at 50  C.
The concentrations of free acids obtained after 4 h hydrolysis at
50  C (17.81 and 16.02 mmol/ml for the SEP with and without azide)
were higher than those obtained by several authors under their
best treatment conditions and highest subsequent methane production, which indicates that lipase activity in hydrolysis at 50  C
could be reduced to achieve lower concentrations of free fatty acids
and thus better results.
Therefore, further hydrolysis tests were conducted with 0.32
and 0.16 U/ml of efuent by measuring the concentration of free
acids up to 24 h (Test 2). The results obtained at 4 and 24 h of hydrolysis can be seen in Table 2.
Using 0.16 U/ml of efuent, the production of free acids was
9.69 mmol/ml after 4 h of hydrolysis at 50  C. In a previous study, a
maximum variation of 7.3 mmol/ml of free acids was obtained after
22 h hydrolysis at 35  C with 1% w/v (0.21 U/ml) of the enzyme
preparation from P. restrictum in the treatment of poultry slaughterhouse efuents with 1200 mg oil and grease/L [11]. Hydrolysis of
efuents from the dairy industry with 1000 mg oil and grease/L and
0.1% w/v (0.02 U/ml of efuent) of the same enzyme preparation
resulted in an 8.0 mmol/ml increase in the production of free acids
after 24 h of hydrolysis at 35  C [8]. In other study, 0.5% w/v of the
enzyme preparation from P. simplicissimum (0.67 U/ml of efuent)
resulted in 4.7 mmol/ml after 8 h hydrolysis at 30  C [20]. All these
results were reported as the best conditions for the production of
methane in anaerobic treatment subsequent to hydrolysis. Based
on the free acid concentrations, which ranged from 4.7 to 8 mmol/
ml for various times and types of efuents, it was found that the
best activity to be used in hydrolysis prior to anaerobic biodegradability tests would be 0.16 and 0.32 U/ml of efuent and 4 h
hydrolysis, since this time showed the highest yield.
3.3. Anaerobic biodegradability tests in sequencing batches after
hydrolysis
Fig. 1 and Table 3 show biogas production over time and the
results obtained in anaerobic biodegradability tests, respectively,
using three sequential batches with non-hydrolyzed efuent
(control) and efuent previously hydrolyzed with 0.16 or 0.32 U/ml
of efuent, with both hydrolysis and anaerobic treatment conducted at 50  C. In the rst and third batches, biogas production
stabilized after 72 h, while in the second batch it took 96 h to
stabilize. This time is much shorter than the 12 days previously
reported for complete biogas production stabilization at 30  C [20].
There was no lag phase for any of the three batches, indicating that
the sludge was well adapted to the synthetic efuent's constituents
and to a temperature of 50  C (data not shown).
Data presented in Table 3 show that in the rst contact of the
sludge with efuent containing 1500 mg oil and grease/L, similar
COD removal rates and methane percentages were obtained under
all the conditions. The second contact of the sludge with the
efuent yielded a sharp drop in the biogas volume in the control
(reduction of 71%), contrasting with what occurred when the

Fig. 1. Biogas production (50  C, 1 atm) after different contact times between efuent
and anaerobic sludge in the control experiments (non-hydrolyzed efuent) and with
efuent previously hydrolyzed with 0.16 or 0.32 U/ml of efuent, with both hydrolysis
and anaerobic treatment conducted at 50  C.

efuent was pretreated with 0.16 and 0.32 U/ml of efuent, where
the reductions in the biogas volume were smaller (20% and 3%).
This result was expected, since in the control condition there was a
much higher adsorption of non-hydrolyzed fats in the microbial
biomass, which impaired the metabolism of the organic matter in
the efuent [1], resulting in a 14% reduction in COD removal. When
the efuent was hydrolyzed, although there was an accumulation
of non-hydrolyzed fats and a decrease in COD removal, it was
considerably lower (9% and 8%).
In the third batch, although biogas production was lower under
the control conditions, it actually decreased under all three conditions evaluated. One explanation for this sharp drop in biogas
production is that the accumulation of fat exceeded the adaptation
capacity of the sludge, hampering the activity of microorganisms

J.G. Duarte et al. / Renewable Energy 83 (2015) 455e462

459

Table 3
Results of anaerobic biodegradability tests with raw efuent (control conditions) and efuent after hydrolysis with 0.16 and 0.32 U/ml of efuent, both at 50  C.
Condition
1st batch
Control
0.16c
0.32c
2nd batch
Control
0.16c
0.32c
3rd batch
Control
0.16c
0.32c
a
b
c

CODi (mg/L)

CODf (mg/L)

COD
removal (%)

Biogas
volume (ml)a

% CH4

SMP
(ml CH4/g CODremoved)a,b

6580 10
7220 200
8065 250

284 8
318 10
341 10

95.3
95.0
95.2

67.3 5.8
92.5 10.0
86.0 9.0

74.0 0.3
68.0 0.2
72.0 0.2

88
101
89

5865 10
7780 110
8600 100

1040 5
1045 10
1060 10

82.3
86.6
87.7

19.8 1.0
74.3 4.5
83.5 4.7

43.5 0.1
93.0 0.1
82.0 0.2

20
114
101

7845 15
7590 210
7770 160

2454 20
1972 35
2440 25

68.72
74.02
68.60

4.3 1.5
9.3 3.5
9.5 0.5

33.3 0.5
53.7 0.7
40.4 0.4

3
10
8

At 50  C/1 atm.
Specic methane production.
U/ml of efuent, with SEP containing average lipase and protease activity of 75 U/g and 70 U/g at 50  C.

in COD removal and methane production. Contrasting behavior


was observed at 30  C, as the batches were conducted with the
microbial biomass in adaptation and increasing methane production [11].
Another explanation is that the proteases present in the SEP
were acting to reduce the microbial biomass. The reduction in the
sludge concentration was greater in the presence of proteases
secreted by Brevibacillus sp. and minimized by the use of protease
activity inhibitors [22].
The high temperature is another factor that may have affected
methane production, although the sludge was previously adapted to
50  C. Hydrolysis at 50  C may have released long-chain fatty acids
into the medium, known inhibitors of methanogenic archaea [23].
3.4. Effect of proteases on anaerobic biodegradability
In order to assess whether the presence of proteases in the SEP
was harmful in the subsequent stage of methane production,
enzyme preparations derived from three different fermentation
systems were used: submerged fermentation with synthetic medium, submerged fermentation with babassu cake, and solid state
fermentation (SSF). Each of these enzyme preparations had
different levels of protease activity with respect to lipase activity.
The ratio of enzyme activity (protease: lipase) was higher in the SEP
obtained from SSF (1.16), intermediate in the liquid enzyme preparation obtained from submerged fermentation with babassu cake
(0.4), and null in the liquid enzyme preparation from SSF. Thus,
maintaining lipase activity at 0.16 U/ml of efuent for all the
enzyme preparations evaluated, hydrolysis was carried out at 50  C
for 4 h. The hydrolyzed efuents were then submitted to anaerobic
biodegradability tests at 50  C. The results are shown in Table 4.
In a previous biodegradability test (Table 3), three sequential
batches were performed and a signicant decrease in the parameters evaluated was observed for the third batch. In this experiment
(Table 4), two sequential batches were performed, followed by one
week of sludge incubation at 50  C, and two more sequential
batches. The sludge incubation period was designed to assist in the
solubilization/hydrolysis of fats adhered to the biomass. In general,
the COD removal was similar for all the conditions, dropping along
the rst three batches (90.5 2.4, 86.6 4.3, 74.2 4.9%) and
stabilizing in the fourth batch (79.2 4.9%). Thus, differences
among the conditions can best be evaluated by specic methane
production, which, under the control condition, yielded much
lower values in the rst three batches and recovered in the fourth
batch, probably due to the adaptation of the microorganisms,
which started producing and excreting hydrolytic enzymes in the

reaction medium, and also due to the sludge recovery period, which
enabled the assimilation of fat accumulated before the third
efuent exchange.
Overall, the differences between one hydrolysis condition and
another were not great enough to justify the use of the enzyme
preparation produced by SSF without proteases. Even so, the hydrolysis condition with the highest protease activity (SSF) did not
hamper the anaerobic treatment. Interestingly, sh industry efuents contain a signicant portion of biodegradable organic matter
primarily composed of proteins and lipids [6,24]. Thus, it is
important to have proteases present in the hydrolysis of proteins in
the efuent; the proteases present in the enzyme prepared from
pig pancreas help the treatment of efuents rich in fats and
proteins [25].
The hydrolysis condition with 0.16 U/ml of efuent using the
SEP from SSF for 4 h at 50  C was selected to continue the work.
Although the specic methane production was very low (compared
with the theoretical value of 414 ml CH4/g CODremoved at
50  C/1 atm), this condition reduced treatment costs. In a previous
study, the best condition obtained in the tests e 8 h hydrolysis at
30  C with 0.5% (w/v) SEP e yielded 216 ml CH4/g at 30  C [20]. This
value corresponds to 230 ml CH4/g CODremoved at 50  C, or
approximately 56% of the maximum expected theoretical value.
This difference may be due to the higher temperature used in the
hydrolysis and biodegradability tests, or because the efuent used
in this work was synthetic, while in the previous study an industrial
efuent was used.
3.5. Anaerobic biodegradability test under mesophilic and
thermophilic conditions after hydrolysis at 50  C
The high energy consumption involved in maintaining the
temperature at 50  C for anaerobic treatment may limit the feasibility of the process on an industrial scale. However, using this
temperature for only 4 h of hydrolysis may be feasible, since
industrial efuents are generated at high temperatures. Thus, hydrolysis at 50  C followed by anaerobic treatment at 30  C and 50  C
was evaluated.
To carry out the anaerobic biodegradability tests, sludge previously adapted to thermophilic and mesophilic conditions was submitted to three consecutive batches with efuent hydrolyzed at
50  C for 4 h or 30  C for 8 h (condition used in Ref. [20]) to obtain
COD removal over 80%. Thus, the biogas production results under the
three conditions (thermophilic, mesophilic and hybrid) shown in
Fig. 2 correspond to the fourth contact of the sludge with the efuent
containing 1500 mg oil and grease/L hydrolyzed in each condition.

460

J.G. Duarte et al. / Renewable Energy 83 (2015) 455e462

Table 4
Results of anaerobic biodegradability tests with raw efuent (without addition of enzyme e control) and with efuent after hydrolysis with 0.16 U/ml of different fermentations, both at 50  C.
Enzimatic
pool (P:L)a
1st batch
No addition
SFd(0)
SFBe(0.4)
SSFf(1.16)
2nd batch
No addition
SF (0)
SFB (0.4)
SSF (1.16)
3rd batch
No addition
SF (0)
SFB (0.4)
SSF (1.16)
4th batch
No addition
SF (0)
SFB (0.4)
SSF (1.16)
a
b
c
d
e
f

CODi (mg/L)

CODf (mg/L)

COD
removal (%)

Biogas
volume (mL)b

% CH4

SMP
(ml CH4/g CODremoved)b,c

3723
4267
3477
5107

40
64
207
57

307
440
433
359

5
12
48
14

91.8
89.7
87.5
93.0

14.3
87.0
43.3
97.5

3.5
8.4
12.2
6.4

85.1
88.9
89.3
86.6

0.1
0.2
0.1
0.1

40
225
141
198

5463
4673
3917
5907

200
74
112
337

615
606
767
582

30
31
19
3

88.7
87.0
80.4
90.1

46.0
62.8
41.9
60.7

7.1
8.1
13.7
5.5

89.5
89.3
89.5
85.9

0.05
0.06
0.03
0.04

94
153
132
109

4552
6433
4373
6075

69
115
326
171

1219
1998
1147
1174

81
30
32
60

73.2
68.9
73.8
80.7

8.0
50.5
56.5
80.5

0
6.8
1.9
1.0

82.9
80.1
84.4
81.3

0.07
0.09
0.05
0.01

22
101
164
148

5822
6372
5128
6517

42
25
67
85

1199
1656
1151
931

14
18
5
7

79.4
74.0
77.6
85.7

66.5
61.3
52.0
68.5

4.8
3.1
2.9
6.4

84.8
85.2
85.3
84.4

0.07
0.56
0.09
0.02

136
123
124
115

P protease; L lipase.
At 50  C/1 atm.
Specic methane production.
SF submerged fermentation.
SFB submerged fermentation with babassu cake.
SSF solid state fermentation.

It was found that in the hybrid and thermophilic treatments,


biogas production stabilized after around 68 h (2.8 days), while
under mesophilic conditions it took about 180 h (7.5 days) to
stabilize. In another study, biogas production took 12 days to stabilize under the same conditions [20], which is probably related to
the use of a more complex industrial efuent. The higher biogas
yield under the hybrid conditions (376.1 ml biogas/L.d) was owing
to the maintenance of a high initial production rate (51.1 ml/d) for
up to 18 h, while under mesophilic conditions the yield was much
lower (139.4 ml biogas/L.d) because the initial production rate

(49.0 ml/d) dropped as of 22 h and remained at an average of


5.3 ml/d until it stabilized after 180 h. The thermophilic conditions
did not produce good results, yielding only 20.9 1.2 ml biogas, or
18.8 ml methane.
Under the hybrid conditions, despite the considerably shorter
hydrolysis time (only 4 h), the temperature used increased the
hydrolysis reaction speed, allowing the fats present in the efuent
to break down into free fatty acids and glycerol (which are more
easily degraded by the microbial consortium) in a shorter space of
time, accelerating the subsequent anaerobic biological treatment

Fig. 2. Biogas production (STP) over time for efuent with 1500 mg oil and grease/L submitted to hydrolysis/anaerobic treatment under mesophilic (hydrolysis and anaerobic
treatment at 30  C), thermophilic (hydrolysis and anaerobic treatment at 50  C), and hybrid (hydrolysis at 50  C and anaerobic treatment at 30  C) conditions.

J.G. Duarte et al. / Renewable Energy 83 (2015) 455e462

461

Fig. 3. Anaerobic biodegradability tests with efuent containing 1500 mg oil and grease/L submitted to hydrolysis/anaerobic treatment under hybrid conditions: specic production
of methane (STP) and COD over time.

process. Moreover, the anaerobic digestion process is facilitated


when treatment occurs after thermophilic hydrolysis [26].
Fig. 2 shows that in the rst 24 h of degradation, the efuents
hydrolyzed at 30  C and 50  C maintained the same biogas production rate (about 2.3 ml/h). However, the efuents hydrolyzed at
30  C yielded a gradually declining biogas production rate until
stabilization at 180 h, while the efuents hydrolyzed at 50  C
maintained the same biogas production rate for 20 h and then stabilized. This behavior indicates that hydrolysis at 50  C releases more
products capable of being assimilated more quickly by the microbial
population, which is of the utmost importance for the anaerobic
treatment process. The hybrid condition yielded the best biogas
production results: 106.0 1.7 ml, and 95.5 ml methane. Thus, the
higher hydrolysis temperature yielded biogas production rates as
high as those obtained under mesophilic conditions e 104.1 6.2 ml
of biogas or 93.8 ml methane e but in a much shorter time.
A comparison of the specic methane production (SMP) at
standard temperature and pressure (STP) under the three conditions revealed that the thermophilic conditions yielded SMP of
19.5 ml CH4/g CODremoved, far below the theoretical value of
350 ml CH4/g CODremoved at STP. This could be because there was a
concentration of methanogenic archaea adapted to 50  C in the
sludge collected from the reactor operating under mesophilic
conditions. Under mesophilic conditions, the highest SMP of
125 ml CH4/g CODremoved was obtained after 180 h. A sh industry
efuent yielded 194.6 ml CH4/g CODremoved under mesophilic
conditions [20]. This indicates that the industrial efuent may
contain components that aid the conversion of organic material
into methane.
The maximum SMP under hybrid conditions (Fig. 3) was
110.6 ml CH4/g CODremoved after 228 h. However, in only 68 h it had
reached 105.4 ml CH4/g CODremoved, differing from the mesophilic
conditions, under which it took 96 h to attain values close to the
maximum (98.6 ml CH4/g CODremoved). As to COD removal, in only
4 h of test, residual COD reached 25% of the original, which means
COD was reduced by 75%. By 28 h, COD removal had reached 93.3%;
in 116 h, it reached 97.9%, and by 228 h it had reached 98.2%. After
68 h, COD removal reached 97.5% and SMP reached levels close to
the maximum obtained at the time of the test. Thus, 68 h is long
enough for the efuent to reach the desired levels.
These results suggest that the factor limiting the reduction of
the treatment time is not thermophilic anaerobic biodegradability,
but thermophilic hydrolysis. The latter probably promotes the
release of components that aid the biodegradation of the efuent
and methane production, while reducing the quantity of enzyme

required [26]. Hydrolysis at 50  C reduces enzyme activity and reaction time and may be the key to accelerating the subsequent
mesophilic anaerobic biological treatment, which, conducted at
intermediate temperatures, would not increase process costs.
Anaerobic treatment can be conducted at room temperature in
most parts of Brazil because high temperatures can be guaranteed
most of the year. Even if the hydrolysis temperature has to be
maintained, the higher energy consumption could possibly be
offset by the shorter treatment time, which makes it an important
target for the scaling up of the enzymatic/biological process.
4. Conclusions
The enzymatic hydrolysis of fats present in the efuent at 50  C for
4 h led to concentrations of free acids that were compatible with the
subsequent anaerobic biological treatment and with smaller quantities of enzyme (0.32 and 0.16 U/mL of efuent) than the hydrolysis
conducted at 30  C. When hydrolysis and anaerobic biological
treatment were both conducted at 50  C, it was found that the
smaller quantity of enzyme evaluated (0.16 U/mL of efuent) yielded
higher specic methane production than in the control assays
(without hydrolysis). Even so, the results obtained under thermophilic conditions were lower than those obtained at 30  C. When we
evaluated the hybrid conditions, with hydrolysis at 50  C and
anaerobic biological treatment at 30  C, the results obtained were the
same as those obtained when both stages were run at 30  C, but in a
much shorter time (68 h). This indicates that these conditions (hydrolysis at 50  C and biological treatment at 30  C) could be adopted
for industrial scale combined (enzymatic/biological) treatments.
Acknowledgments
This work was supported by project funds from the Brazilian
National Council for Research and Development (CNPq) Process no
557194/2010-5, Carlos Chagas Filho Foundation for Research Support in the State of Rio de Janeiro (FAPERJ) Process no E-26/100.647/
2007, and Petrobras Contract no 0050.0045686.08.2.
References
 n AP. Anaerobic digestion of sh processing wastewater with special
[1] Rollo
emphasis on hydrolysis of suspended solids [Ph.D thesis]. Wageningen, The
Netherlands: Wageningen Agricultural University; 1999. p. 123.
[2] Chan YJ, Chong MF, Law CL, Hassel DG. A review on anaerobic-aerobic treatment of industrial and municipal wastewater. Chem Eng J 2009;155:1e18.

462

J.G. Duarte et al. / Renewable Energy 83 (2015) 455e462

[3] Chernicharo CAL. Biological wastewater treatment series. Anaerobic reactors.


1st ed., vol. 4. London: IWA Publishing; 2007.
ve J, Helsen L, Lievens B, Willems K, et al. Anaerobic
[4] Appels L, Lauwers J, Degre
digestion in global bio-energy production: potential and research challenges.
Renew Sustain Energy Rev 2011;15:4295e301.
[5] Chowdhury P, Viraraghavan T, Srinivasan A. Biological treatment processes for
sh processing wastewater e a Review. Bioresour Technol 2010;101:439e49.
 n AP, Zeeman G, Lubberding HJ, Lettinga G, Alaerts GJ. Treatment of sh
[6] Rollo
processing wastewater in a one- or two-step upow anaerobic sludge blanket
(UASB) reactor. Water Sci Technol 2002;45:207e12.
[7] Alexandre VMF, Valente AM, Cammarota MC, Freire DMG. Performance of
anaerobic bioreactor treating sh-processing plant wastewater prehydrolyzed with a solid enzyme pool. Renew Energy 2011;36:3439e44.
[8] Leal MCMR, Freire DMG, Cammarota MC, Sant'Anna Jr GL. Effect of enzymatic
hydrolysis on anaerobic treatment of dairy wastewater. Proc Biochem
2006;41:1173e8.
[9] Mendes AA, Castro HF, Pereira EB, Furigo Jr A. Application of lipases for
wastewater treatment containing high levels of lipids. Qum Nova 2005;28:
296e305.
[10] Sharma R, Chisti Y, Banerjee UC. Production, purication, characterization and
applications of lipases. Biotechnol Adv 2001;19:627e62.
[11] Vallad~
ao ABG, Freire DMG, Cammarota MC. Enzymatic pre-hydrolysis applied
to the anaerobic treatment of efuents from poultry slaughterhouses. Int
Biodeter Biodegr 2007;60:219e25.
[12] Krishna C. Solid-State fermentation systems- an overview. Crit Rev Biotechnol
2005;25:1e30.
[13] Freire DMG, Teles EMF, Bon EPS, Sant'Anna Jr GL. Lipase production by Penicillium restrictum in a bench scale fermenter: media composition, agitation
and aeration. Appl Biochem Biotechnol 1997;63:409e21.
[14] Gutarra MLE, Cavalcanti EDC, Freire DMG, Castilho LR, Sant'anna Jr GL. Lipase
production by solid-state fermentation: cultivation conditions and operation
of tray and packed-bed bioreactors. Appl Biochem Biotechnol 2005;121:
105e16.
[15] Gutarra MLE, Godoy MG, Castilho LR, Freire DMG. Lipase production and
Penicillium simplicissimum morphology in solid-state and submerged fermentations. Biotechnol J 2009;4:1450e9.
[16] Greenberg AE, Clesceri LS, Eaton AD. Standard methods for the examination of
water and wastewater. 21st ed. Washington, DC: American Public Health

[17]
[18]
[19]

[20]

[21]
[22]

[23]

[24]
[25]

[26]

[27]

Association, American Water Works Association, Water Pollution Control


Federation; 2005.
Bendicho S, Mart G, Hern
andez T, Martin O. Determination of proteolytic
activity in different milk Systems. Food Chem 2002;79:245e9.
Eiroa M, Costa JC, Alves MM, Kennes C, Veiga MC. Evaluation of the biomethane potential of solid sh waste. Waste Manag 2012;32:1347e52.
~o ABG, Freire DMG, Cammarota MC. Evaluation of different preVallada
hydrolysis times and enzyme pool concentrations on the biodegradability of
poultry slaughterhouse wastewater with a high fat content. Water Sci Technol
2009;60:243e9.
-hidro
lise
Valente AM, Alexandre VM, Cammarota MC, Freire DMG. Pre
enzim
atica da gordura de euente da indstria de pescado objetivando o
~o de metano [Enzymatic hydrolysis of fat from sh inaumento da produa
dustry efuents aimed at increasing methane production]. Cienc Tecnol
Aliment 2010;30:483e8.
Jay JM. Modern food microbiology. 6th ed. Maryland: Aspen Publishers; 2000.
Maeda AT, Yoshimura AT, Garca-Contreras BR, Ogawa IH. Purication and
characterization of a serine protease secreted by Brevibacillus sp. KH3 for
reducing waste activated sludge and biolm formation. Bioresour Technol
2011;102:10650e6.
Hatamoto M, Imachi H, Ohashi A, Harada H. Identication and cultivation of
anaerobic, syntrophic long-chain fatty acid-degrading microbes from mesophilic and thermophilic methanogenic sludges. Appl Environ Microbiol
2007;73:1332e40.
Gonzalez JF. In: Wastewater treatment in the shery industry, vol. 355. Rome
(Italy): Fisheries Dept; 1996. FAO Fisheries Technical Paper.
Mendes AA, Pereira EB, Castro HF. Effect of the enzymatic hydrolysis pretreatment of lipids-rich wastewater on the anaerobic biodigestion. Biochem
Eng J 2006;32:185e90.
Zhou Y, Takaoka M, Wang W, Liu X, Oshital K. Effect of thermal hydrolysis pretreatment on anaerobic digestion of municipal biowastes: a pilot scale study
in China. J Biosci Bioeng 2013;116:101e5.
~o de enzimas hidrolticas no tratamento biolo
gico
Valente AM. Aplicaa
bio de euente de indstria de pescado [Application of hydrolytic enanaero
zymes on anaerobic biological treatment of efuents from shing industry]
[Ph.D thesis]. Brazil: Federal University of Rio de Janeiro; 2009.