J Food Sci Technol
DOI 10.1007/s13197-014-1425-4
ORIGINAL ARTICLE
Aqueous two phase partitioning of fish proteins:
partitioning studies and ATPS evaluation
Viswanatha H. Nagaraja & Regupathi Iyyaswami
Revised: 19 April 2014 / Accepted: 26 May 2014
# Association of Food Scientists & Technologists (India) 2014
Abstract A suitable Aqueous Two Phase System (ATPS) has
been identified for the partitioning of crude fish proteins from
fish processing industrial effluent. A detailed study has been
performed to analyze the influence of various parameters on
protein partitioning such as molecular weight of PEG, effect
of different salts (MgSO4, K2HPO4, Na3C6H5O7, Na2SO4,
(NH4) 2PO4, K3C6H5O7) and their concentrations, pH, tem-
perature, Tie Line Length (TLL), effluent loading and volume
ratio. PEG 2000 - sodium sulphate ATPS was found to be a
most favourable system among the selected ATPS for higher
partition coefficient of proteins. The binodal curve and equi-
librium characteristics of PEG 2000 - sodium sulphate were
established and fitted to empirical equations. The equilibrium
compositions (tie line) were correlated using OthmerTobias
and Bancroft equations.
Keywords Fish proteins . Aqueous two-phase system .
Partitioning coefficient . Phase diagram
Abbreviations
ATPS Aqueous two-phase systems
TLL Tie line length (w/w%)
Vbp Volume of bottom phase (ml)
Vtp Volume of top phase (ml)
Wp Weight fraction of polymer
Ws weight fraction of salt
VR volume ratio
K capartition coefficient
V. H. Nagaraja : R. Iyyaswami (*)
Department of Chemical Engineering, National Institute of
Technology Karnataka, Surathkal, Mangalore, India
e-mail: regupathi@nitk.ac.in
Introduction
The effluent generated from fish process industry contains a
high organic load and it should be treated/recovered before
discharging, in order to prevent a negative impact on the
environment. On the other hand, this effluent contains
a large amount of potentially valuable proteins. The degree
of effluent generation from fish processing industry depends
on type of operation involved and amount of seafood proc-
essed. Typically, a plant of 100 t/h fish processing capacity
plant generates 1040 m3/h effluent with protein loads of 0.5
20 g/l (Mateusz and Daniela 2009; Maria and Rodrigo 2002;
Maria et al. 2004). Research has been carried out in order to
develop appropriate methods to convert these wastes into
useful products, like fish protein hydrolysates. The fish pro-
tein hydrolysates have been reported to exhibit various bio
functionalities such as antioxidative, antibacterial, anti hyper-
sensitive and anticancer properties etc. (Kim and Mendis
2006). Recovered protein molecules have vital importance in
several pharmaceutical industries as well as feed formulations
apart from providing value addition to fish processing indus-
trial waste. Conventional methods used for recovery and
purification of proteins from fish processing industrial effluent
include enzymatic hydrolysis, centrifugation (Guerrerof et al.
1998), filtration, flocculation, precipitation (Taskaya and
Jaczynski 2009), pH shifting (Helgi and Ingrid 2009), mem-
brane filtration (Mateusz and Daniela 2009; Maria and
Rodrigo 2002) and emerging technologies such as sub-
critical water hydrolysis, supercritical fluid extraction
and ohmic heating (Kobsak et al. 2008). However the
proteins purified through these conventional techniques
may lead to poor functionality and nutritive value (Sanmartin
et al. 2009) and high production cost. Therefore, it is essential
to design a reliable separation strategy that would improve
yield, selectivity, economy and feasibility of the proteins
purification process.

Aqueous two-phase systems (ATPS) have been successful-
ly used for separation and purification of proteins, metal ions,
nanoparticles and dyes from various sources such as waste
water, fermentation broth, animal and plant cell organelles
(Selvaraj et al. 2011). ATPS are formed by adding two
water-soluble polymers or polymer and salt together above a
critical concentration. The ATPS have advantages over other
conventional separation techniques due to high water content
in both the phases, offering high biocompatibility and low
interfacial tension which results high mass transfer and min-
imal degradation of biomolecules. ATPS also provides high
yield with relatively high handling capacity. Polymer/salt
ATPS are more preferred over polymer/polymer systems since
it is easy to separate the phases due to high density difference
exist between the phases. In addition, the system can be easily
scaled up. ATPS could be a good alternative for a first purifi-
cation step since such systems allow removal of several con-
taminants by a simple and low-cost process (Albertsson 1986;
Raghavarao et al. 1995 and 1998; Rajni 2001; Rosa et al. 2010
and 2011).
The mechanism governing the partition of biomolecules
from a particular source in ATPS is still not fully understood
and largely unknown. In general, protein partitioning is
driven by Van der Waals, hydrophobic, hydrogen bond,
and ionic interactions between the biomolecules and the
surrounding phase. Hence the partition may be influenced
by the concentrations and molecular weight of phase-forming
polymer, type and concentration of phase-forming salt, and
pH. Several studies have resolute on the influences of the
above mentioned parameters on the separation of various
biomolecules. Tubio et al. (2007) has studied the effect of
different factors such as PEG molecular weight, pH, TLL,
temperature and the presence of an inorganic salt on the
protein partition coefficient. Chia-Kai and Been (2006) has
studied extraction behaviour of lysozyme from chicken
egg white by PEG - sulphate system and found that
the maximum partitioning was occurred at a temperature
of 25.8 C and pH 10. Shahbaz and Omidinia (2008)
reported the influence of PEG molecular weight and its
concentrations, pH, ammonium sulphate concentrations and
TLL on recombinant enzyme phenylalanine dehydrogenase
partitioning. Silva et al. (2007) studied the influence of PEG
1500 with various type of salts (potassium phosphate, sodium
citrate, lithium sulphate or sodium sulphate) and temperature
(545 C) on the caseinomacropetide partition coefficient and
the thermodynamic parameters (Ho, So and Go) were 4000, PEG 6000 and PEG 8000) was purchased from
calculated for PEG1500 and sodium citrate ATPS at different
PEG concentrations. The crude proteins (Zea mays malt) were
partitioned from maize malt using PEG 6000/CaCl2 system
and the influence of phase composition and pH on the parti-
tion coefficient were studied and optimized through statistical
method (Ferreira et al. 2007). Further the - and -amylase
enzymes from Zea mays malt were purified 130-fold in the
J Food Sci Technol
salt rich phase by continuous extraction in a PEG/CaCl2 ATPS
(Biazus et al. 2007).
Nowadays, ATPS have been utilized to recover valuable
bio-molecules from various industrial wastewaters. Saravanan
et al. (2007) studied the recovery of proteins from tannery
wastewater using PEG 4000magnesium sulphate. The same
research group (2006) has also reported the recovery of
value-added globular proteins from tannery wastewater
using PEG 6000 - sodium sulphate salt ATPS. Perumalsamy
and Murugesan (2012) developed an environmentally benign
PEG-sodium citrate ATPS for the recovery of value added
cheese whey proteins from dairy effluents. Recently,
Regupathi et al. (2012) developed and studied a new PEG/
lithium citrate salt based ATPS and applied to the fish industry
effluent to partition the proteins. The ATPS formed with PEG
(4000, 6000 and 8000) with salts (tri-potassium citrate, tri-
sodium citrate, sodium sulphate and lithium citrate) were
studied and found that the PEG 4000 - lithium citrate ATPS
resulted with maximum partition coefficient of 7.82 at TLL
38 %. However the usage of lithium salts at higher concen-
trations has been toxic (Aral and Vecchio-Sadus 2008).
Lithium in water is a strong base and reacts slowly with
water and liberates hydrogen, a flammable gas and generates
the lithium hydroxide in the solution. Lithium compounds are
generally pyrophonic and require special handling. Therefore
the salt rich bottom phase need to be treated before its disposal
to environment.
Accordingly the present work was formulated to identify
an environmentally suitable, best ATPS to partition the crude
proteins from the fish processing industry effluent by consid-
ering PEG (2000) - salts (MgSO4, K2HPO4, Na3C6H5O7,
Na2SO4, (NH4) 2PO4, K3C6H5O7) ATPS systems and varying
their concentrations at different pH. The effect of PEG mo-
lecular weight, temperature, effluent loading and volume ratio
were studied. The binodal curve and tie lines were developed
and studied at 25 C for the PEG2000- Na2SO4.
Materials and methods
MgSO4, K2HPO4, Na3C6H5O7, Na2SO4, (NH4) 2PO4 and
K3C6H5O7 salts of purity analytical standard (PAS) were
obtained from Merck, Mumbai, India. Analytical grade
Polyethylene Glycol (PEG) of average molecular weight
1.9 kDa, 3.85 kDa, 5.8 kDa and 7.75 kDa (PEG 2000, PEG
Sigma Chem. Co., USA. All the reagents were of analytical
grade. Double distilled water was used for the present
work. Bradford reagent, (Sigma Aldrich Inc., Germany)
was used to estimate the protein content of the equilib-
rium phases and effluent. The fish effluent used in the
experiments was obtained from a fishmeal processing industry
located at Mangalore, India.
J Food Sci Technol
Preparation of ATPS
The PEG molecular weight their respective concentration and
salt concentration were chosen based on the phase diagrams.
To study the effect on partitioning of protein from fish pro-
cessing effluent, different salts like MgSO4, K 2HPO 4,
Na3C6H5O7, Na2SO4, (NH4) 2PO4, and K3C6H5O7 and their
concentrations were mixed with varying PEG (2000, 4000,
6000 and 8000), fish processing industry effluent 50 % (w/w)
and distilled water were used to attain 100 % (w/w) of the
system. The mixtures were dissolved completely using vortex
mixer. Phase separation was achieved by centrifugation for
10 min at 3,500 rpm. The samples were incubated to attain
equilibrium in a thermostat maintained at a constant temper-
ature of 25 C for 24 h with an uncertainty of0.05 C (Lab.
Companion, Model RW-0525G, Korea). The top phase was
carefully separated using a pipette. Volumes of the separated
phases were measured and aliquots from each phase were
taken for further analysis.
Measurement of polymer and salt concentrations
Determination of PEG concentration was performed by mea-
suring refractive index using Digital refractometer (RX-500,
ATAGO Co. ltd, Japan). Calibration curves were prepared
with known concentrations of PEG and Salt in the homoge-
neous regions of binodal diagram of the individual ATPS. The
relation between the refractive index (nD) and the mass frac-
tion of polymer (wp) and salt (ws) is given by Eq 1.
nD a0 a1 wP a2 wS 1 The equilibrium characteristic of the system was analysed
Where a0, a1 and a2 are the fitting parameters.
Salt concentrations in the top and bottom phase were
measured using flame photometer. The phases were diluted
appropriately to the required concentrations that are suitable
for the analysis in flame photometer. Salt calibration curve
was prepared with known concentration of salt with suitable
dilutions.
Protein estimation
Bradford method was used to estimate total protein concen-
tration. The samples were read at 595 nm against the blanks
with the same compositions of PEG and salt. Assays were
performed in triplicate and their respective averages were
used in the calculations. The results are expressed in
terms of bovine serum albumin (BSA) equivalents.
Partition coefficient (K) was calculated by (Eq. 2);
Ctp
K 2
Cbp
Partition coefficient is defined as the ratio of protein con-
centration in top phase, Ctp (mg/ml) to bottom phase,
Cbp (mg/ml). Phase volume ratio (VR) is defined as the ratio
of volume of top phase, Vtp (ml) to that of bottom phase,
Vbp (ml) (Eq. 3).
Vtp
VR 3
Vbp
The percentage yield of protein in the top phase is calcu-
lated using Eq 4.
Ctp  Vtp
Yield %  100 4
C0  V0
Where, Co is initial concentration of protein (mg/ml), Vo =
volume of feed sample (ml).
Phase diagram
Binodal curve for PEG 2000 and sodium sulphate ATPS was
developed through could point method. The experiment was
conducted in 250 ml jacketed vessel at 25 C with an uncer-
tainty of0.05 C. A known concentration of salt solution was
titrated against to polymer solution or vice versa, until the
clear solution turn out to be turbid. The phase composition at
this point was estimated through weighing method using a
higher precision (0.01 mg) analytical balance. Further
known quantity of water was added until the turbidity get into
clear solution. This procedure was repeated to obtain different
bimodal points (Regupathi et al. 2011 and 2012).
through tie lines. Pre weighted quantity of varied PEG 2000,
sodium sulphate and water were added into a centrifuge tube
of 15 ml to obtain 10 g system. The samples were thoroughly
mixed and centrifuged at 3500 rpm for 10 min. The mixture
was allowed to settle for 24 h at 25 C in a temperature bath,
resulted in to two equilibrated phases. The individual phase
volumes were measured and subsequently separated using
micro pipette. The concentrations of PEG 2000 and sodium
sulphate in both top and bottom phase were determined using
refractive index and flame photometer, respectively, which
represents the tie line composition. Further the tie line length
was calculated using the equilibrium compositions by Eq. (5).
h 1=2

2
2 i
TLL W PT W PB W S T W S B 5
Results and discussion
Selecting a suitable ATPS for the higher partitioning of pro-
teins in a specific effluent is a complex process since it is

highly depends on the characteristics of the solute such as
hydrophobicity, molecular size, electrochemical property, mo-
lecular conformation and bio specificity as well as the ATPSs
properties such as phase forming polymer or salt, pH, temper-
ature, TLL, effluent loading and volume ratio. Initially the
suitable salt (at fixed concentration) was identified by
conducting the experiments at different PEG2000 concentra-
tion and varying salt concentration at fixed PEG2000 concen-
tration. Further the effect of pH, PEG molar mass, tempera-
ture, and volume ratio were studied for the selected system.
Effect of PEG 2000 concentrations with different salts
The partitioning of soluble crude proteins from fish processing
industrial effluent was studied by varying the PEG2000 con-
centration at a fixed salt concentration of 22.5 % (w/w). Five
different salts as MgSO4, K2HPO4, Na3C6H5O7, Na2SO4,
(NH4) 2PO4 and K3C6H5O7 were considered for the present
work (Fig. 1). From the figure it was observed that the sodium
sulphate and potassium di-hydrogen phosphate ATPSs
showed maximum partitioning coefficient in the range of 1.7
to 2.3. The partitioning coefficient found to increases with
increasing PEG2000 concentration for all the selected salts
due to salting out effect except ammonium phosphate and
resulted in movement of proteins from bottom phase to top
phase. At higher PEG concentration, the degree of protein
interaction with PEG found to increases due to the hydrophilic
nature of the system. Saravanan et al. (2006 and 2007) has
observed similar effect for tannery waste contained proteins.
However, there is a limitation in the solubility of protein in the
top phase of PEG2000 - (NH4) 2PO4 system at higher PEG
concentration; hence the partitioning coefficient was found to
decrease with increasing PEG concentration. However
MgSO4 ATPS showed maximum precipitation at the inter
phase. The mixed effect was observed probably due to the
combination of hydrophobic, ionic change and free volume
2.5
2
Partition coefficient
1.5
1
0.5
0 0
10 12.5 15 17.5
PEG 2000 % (w/w)
Fig. 1 Effect of PEG2000 concentrations with different salts ( Na2SO4,
K2HPO4, (NH4) 2PO4, K3C6H5O7 and Na3C6H5O7) of fixed
concentration 22.5 % (w/w) on the partition coefficient at 25 C
J Food Sci Technol
interactions exist with the proteins. The affordable partition
coefficient was observed at a PEG2000 concentration of
15 % (w/w) in all the selected salts forming ATPS. Mokhtarani
et al. (2008) also conducted similar experiments for the
partitioning of Ciprofloxacin and reported that the salt con-
centration had significant effect than the PEG concentration.
Selection of phase forming salt
The experiments were carried out at constant PEG2000 con-
centration of 15 % (w/w) with different salts and their concen-
trations to select the suitable ATPS for efficient fish protein
partitioning (Fig. 2). The ATPSs are formed for the said
purpose with 50 % (w/w) of fish processing industrial effluent.
From the figure it was observed that the partitioning coeffi-
cient was increasing with increasing salt concentration due to
the decrease in the solubility of protein in bottom phase. The
free water volume available in the bottom phase for the protein
dissolution may decreases as the salt concentration increase.
However the top PEG phase shows stronger affinity towards
the proteins, due to the corresponding increase in the PEG
concentration in the top phase. Further, the partition coeffi-
cient may vary for different salts as shown in Fig. 2 due to the
nature of the salt cation and anion and the systems net
charges. The influence of salt on the partitioning is caused
by the nonuniform distribution of the salt ions in the upper and
lower phases and by the difference in the electric potential,
which improves the movement of the protein to the other
phase through electrostatic repulsion/attraction. The said
effect may be explained through the Hofmeister series
(Gupta et al. 2002), which is reported as, for anions SO42 >
HPO42 > CH3COO > Cl > Br > I > SCN, and for cations
Li+ > Na+ > K+ > NH4+ > Mg2+. At higher concentrations of
the salt, the ions to the left of the series decrease the protein
2.5
2
Partition coefficient
1.5
1
0.5
20 12.5 15 17.5 20 22.5 25
Salt % (w/w)
Fig. 2 Effect of salts ( Na2SO4, K2HPO4, (NH4) 2PO4,
K3C6H5O7 and Na3C6H5O7) concentration on partition coefficient at
15 % (w/w) PEG 2000 concentration at 25 C
J Food Sci Technol
solubility (salting-out effect) in the salt rich bottom phase.
Further the hydrophobic interaction between the protein and
polymer phase increases due to the hydration effect of the salt
molecule surrounding the protein and lead to the aggregation
of proteins in the top phase. The present result shows the
similar effect for the selected salt systems. Among the salt
systems potassium phosphate and sodium sulphate showed
better partition when compared to ammonium phosphate,
sodium citrate and potassium citrate salts. A similar
behaviour had been reported earlier for various proteins in
different systems (Saravanan et al. 2006; Kohler et al. 1991;
Perumalsamy and Batcha 2011).
Effect of pH on protein partitioning
In general, the pH affects the partitioning of protein by chang-
ing the net charge of the phases, consequently the hydropho-
bic and hydrophilic forces exist in the system may be altered
and which promotes the partitioning of the proteins towards a
particular phase. However the change in electrostatic force
with pH may vary significantly for different salts. Hence in the
present study the pH was altered in the range of 5 to 10 for all
the ATPSs (with different salts) to understand the partitioning
behaviour of the systems with constant PEG2000 and salt
concentration (Fig. 3). From the figure it was observed that a
small change in pH values had a stronger effect on partition
coefficient in all the ATPS. Significant increase in partition
coefficient as well as top phase yield was observed for all the
systems around the pH values of 5 to 6. However the maxi-
mum partition coefficient of 13.14 was observed at pH 5 for
PEG 2000 - sodium sulphate ATPS. At lower pH, the hydro-
phobic attraction between PEG and proteins were predomi-
nant due to the net negative charge gained by the protein
molecules and resulted to higher partitioning coefficient.
Conversely the partitioning coefficient was found to decrease
with increasing pH since the bottom phase was enriched with
sulphate which reduces the net hydrophobic interaction. The
14
12
Partition coefficient
10
8 10
6
4
2 2
0 0
4 5 6 7 8 9 10
pH
Fig. 3 Effect of pH on partitioning coefficient for different salts (
Na2SO4, K2HPO4, (NH4) 2PO4, K3C6H5O7 and Na3C6H5O7)
with 15 % (w/w) PEG concentration at 25 C
ATPS with sulphate as the phase-forming salt shown higher
partition since the sulphate anion has the ability to promote
hydrophobic difference between the phases through salting-
out phenomena (Chia-Kai and Been 2006). Similar behaviour
was observed in the literature for other systems (Patil and
Raghavara 2007; Ferreira et al. 2007; Yucekan and Onal
2011; Perumalsamy and Batcha 2011). The partitioning coef-
ficient of proteins from Shrimp waste increases when the pH
increases from 6 to 8 and maximum partition coefficient of
2.96 was achieved at pH 8 (Ramyadevi et al. 2012). The
results revealed that the fish proteins gained maximum nega-
tive charge at the pH of 5. From the observations, it was
decided to carry out the further experiments at 15 % (w/w)
of PEG 2000 and 22.5 % (w/w) sodium sulphate and at an
optimised pH of 5 for the purification of proteins from fish
processing industrial effluent.
Effect of PEG molecular weight on protein partitioning
Molecular weight of PEG alters the partition coefficient of
biomolecules, since the hydrophobic interaction and effective
excluded volume of the ATPS greatly depends on it. Figure 4
shows the effect of molecular weight on the partitioning of
soluble proteins from fish processing industrial effluent in
15 % (w/w) of PEG and 22.5 % (w/w) sodium sulphate
ATPS at 5 pH. The partition coefficient of proteins decreased
with increasing PEG molecular weight. The increase in PEG
molecular weight increases the hydrophobicity and effective
excluded volume of polymer rich top phase in the systems
which favours the partitioning of proteins in the PEG phase
however the reduction in top PEG rich phase free volume for
biomolecule salvation reduces the protein partitioning in top
phase. Similar observations were reported for the partitioning
of BSA in PEG-tri sodium citrate/ tri potassium citrate system
(Perumalsamy and Batcha 2011; Kalaivani and Regupathi
2013) [13, 15]. However, the increase of PEG molecular
weight associated with increase in the hydrophobicity of
16 100
14
80
Partition coefficient
12
60
Yield
8
6 40
4
20
0
11 0 2000 4000 6000 8000 10000 12000
PEG molecular weight
Fig. 4 Effect of PEG molecular weight on protein partitioning on
PEG 15 % (w/w) - sodium sulphate 22.5 % (w/w) at 25 C. () K and
() yield %

PEG, does not favour the partitioning of the fish protein in to
the top PEG rich phase which indicates that the Effective
Excluded Volume of PEG predominates the hydrophobic
effect. Further, at higher PEG molecular weight the proteins
start to precipitate at the interface due to the significant reduc-
tion in free volume for biomolecule solubility in both the
phases. The maximum partition coefficient 13.14 and
84.45 % yield were obtained at 15 % (w/w) PEG 2000
22.5 % (w/w) sodium sulphate ATPS at pH 5. Saravanan
et al. (2007) has also reported a similar result for the recovery
of proteins from tannery wastewater in PEG-magnesium sul-
phate system. Ramyadevi et al. (2012) observed that the
partition coefficient and extraction yield of proteins from
Shrimp waste was decreased with the increase of PEG molec-
ular weight from 4000 to 10,000.
Effect of temperature on protein partitioning
To select the optimal temperature for the process study, dif-
ferent temperatures of 2050 C at constant pH 5 were studied
and the results are shown in Fig. 5. The change in the partition
coefficients of the biomolecules with temperature may be
attributed towards variation in the phase compositions. In
the present study, the obtained results indicated that increase
in temperature leads to decrease in partition coefficient of
protein. The maximum protein partition 14.13 was observed
at 25 C with yield of 84.45 %. At higher temperature, PEG
structure becomes more extended and as a result the preferen-
tial interaction between proteins and PEG decreases, conse-
quently the partition coefficient and yield of proteins extracted
also decreases. On the other hand, increasing the temperature
the proteins denature and lose its activity. Similar temperature
effect were found for various biomolecules such as globular
proteins from tannery wastewater (Saravanan et al. 2006
tannase from fermentation broth (Naidu et al. 2008), alkaline
phosphatase from Bacillus licheniformis (Pandey and Banik
2010), bromelain from pineapple (Ketnawa et al. 2010), BSA
(Perumalsamy and Batcha 2011) and -glucosidase from
Trichoderma reesei (Gautam and Simon 2006).
16 100
K
14
Yield 80
Partition coefficient
12
10 60
Yield
8 8
6 40
4 4
20
2 2
0 0
15 20 25 30 35 40 45 50 55
Temperature (C)
Fig. 5 Effect of temperature on protein partitioning in 15 % (w/w) PEG
200022.5 % (w/w) sodium sulphate. () K and () yield %
J Food Sci Technol
Effect TLL on protein partitioning
In general PEG and salt concentrations present in the ATPS
were mainly responsible for the two phase formation.
However, the type and nature of the salt present in the system
also contributed towards protein partitioning. The TLL %,
which in turn represents the equilibrium of the ATPS, was
utilized to study the combined effect of the weight fraction of
PEG and salt. The effect of TLL on partition and yield in PEG
2000 - sodium sulphate at 25 C are shown in Fig. 6. It was
observed that as the TLL increases the partition coefficient
also increases. At larger TLL %, PEG concentration in
the top phase and salt concentration in the bottom phase
increases. Due to increase in PEG concentration in top
phase, the number of polymer-protein hydrophobic
interactions may increase. On the other hand, the
biomolecules may be excluded from the bottom phase
to top phase due to the solubility limits of the biomolecule in
salt phase through effective salting out phenomena. It
promotes the partition of protein from the bottom phase to
top phase. The similar phenomenon has been reported
by Patil and Raghavara (2007) and Ferreira et al. (2007).
Ramyadevi et al. (2012) also reported that the partition
coefficient of proteins from Shrimp waste was increased
with increasing TLL and found maximum of 2.96 at a TLL
of 36.
Effect of fish effluent loading on protein partitioning
As described in the earlier sections, fish processing effluent
load of 50 % (w/w) was used. Higher effluent loads would be
advantageous since larger feed volumes could be processed in
single stage ATPS. Fish processing industrial effluent loading
is considered to be one among the important variable in the
selection of partitioning system and was used to examine the
processing capacity of ATPS by varying effluent loading.
20 100
18 90
K
16 80
Yield
Partition coefficient
14 70
Yield (%)
12 60
10 50
40
6 30
20
10
0 0
30 35 40 45 50 55
TLL (%)
Fig. 6 Effect of TLL on protein partitioning coefficient and yield in PEG
2000 - sodium sulphate system
J Food Sci Technol

16 100
Effluent loading of 1060 % (w/w) was used and their respec- 90
14
tive loading effects are shown in Fig. 7. As the effluent loading 80

Partition coefficient
12
increased from 2050 % (w/w) gradually, partition coefficient 70

Yield (%)
10 60
increases from 2.8 to 10.62 and yield from 61.0585.28 %
8 50
respectively. Further loading at 60 % (w/w) the partition K
6 40
coefficient decreased to 3.36. This is due to increased protein Yield 30
4
precipitation at interphase and due to insufficient free volume 20
available in the phases. Similar observations of precipitation 2 10
and phases free volume at higher loading of proteins to the 0 0
0 1 2 3 4
system were reported in the case of biomass containing fer- Volume Ratio (VR)
mentation broths (Patil and Raghavara 2007; Selvakumar Fig. 8 Effect of volume ratio on protein partitioning coefficient and yield
et al. 2012). in PEG 2000 - sodium sulphate system

Effect of volume ratio on protein partitioning
The important step of any purification process should com-
bine in lower volume and high yield with a partition of the
target molecule. In ATPS this could be easily achieved by
altering volume of the phases, where the target molecule is
partitioned into one particular phase. Volume ratio was varied
by changing the volume of the individual phases at constant
total volume in which the phase compositions (PEG and salt)
of the top and bottom phases remain same with respect to their
tie tiles. The differential partitioning coefficient of fish pro-
teins at different phase volume ratios are shown in Fig. 8. The
maximum salvation of the proteins in the individual phases
will be dictated by the solubility limit of the target compound
in the accommodating phases. At lower volume ratios the
protein solubility limits was reached in the bottom phase.
However the remaining proteins try to move towards the top
phase and dissolved in the excess free volume available,
which in turn leads to the increase in the partitioning coeffi-
cient. Moreover, it was suggested that as the volume ratio
increases, the protein partition coefficient increases wherein
the desired molecule gets transferred to the polymer rich top
phase as described in other literatures; betalains partition
coefficient in PEG 6000 - ammonium sulphate (Chethana
et al. 2007), chymosin and pepsin partition coefficient in
PEG - phosphate (Spelzini et al. 2006) and penicillin acylase
partition coefficient in PEG - sodium citrate (Marcos et al.
1998) ATPS. The increase in partitioning coefficient was
noticed till the saturation limits attained by the PEG top phase,
there after the proteins were precipitated at the inter-phase.
Hence the yield was observed to increase till the volume
ratio of 2.5 and further remains constant with volume ratio
(Fig. 8).
Phase diagram for PEG 2000 - sodium sulphate
The overall partitioning behaviour was primarily de-
pends on the equilibrium characteristics of the ATPS.
The phase composition and their equilibrium were
expressed through the phase diagram of the ATPS, which
includes the binodal curve and tie lines. The phase diagram
describes the boundary between two-phase and single phase
region and provide the equilibrium concentrations of the
phase components at the given feed composition. Hence the

12 100 0.5

11 K 90 0.45 34.65%
10 Yield (%) 80 0.4 36.23%
Partition coefficient

9 70 0.35 39.11%
Wp % (w/w)

0.3
Yield (%)

8 60 45.63%
7 50 0.25 50.29%
6 40 0.2
Binodal
5 30 0.15
4 20 0.1
3 10 0.05
2 0 0
10 20 30 40 50 60 70 0 0.1 0.2 0.3
Effluent loading % (w/w) Ws % (w/w)
Fig. 7 Effect of fish effluent loading on protein partitioning coefficient Fig. 9 Binodal curve and tie lines for the PEG 2000 - sodium sulphate
and yield in PEG 2000 - sodium sulphate system system at 25 C
 J Food Sci Technol

Table 1 Literature empirical correlations and their constants for PEG 2000 - sodium sulphate system at 25 C

Correlations a b c D R2

Wp =a+bWs0.5 +cWs (Cheluget et al. 1994 & Regupathi et al. 2012) 0.8472 3.903 4.443 0.9964
Wp =a+bWs0.5 +cWs +dWs2 (Regupathi et al. 2011& Hu et al. 2004) 0.7756 3.189 2.451 3.570 0.9979
InWp =a+bWs0.5 +cWs3 (Regupathi et al. 2011& Maria et al. 2001) 1.2027 12.93 26.52 0.8987
1
Ws a bW 0:5
p cW p (Regupathi et al. 2011& Graber et al. 2001)
27.74026 155.67200 327.53700 0.88610

phase diagram was developed at the maximum protein Phase Equilibrium

partitioning conditions (25 C and pH 5) for PEG 2000 -
sodium sulphate ATPS.
Binodal curve
The binodal curve for the selected PEG 2000 - sodium sul-
phate ATPS was obtained through could point method
(Fig. 9). Further, the experimental binodal data were fitted
by modifying the constants and coefficients of the expressions
available in the literature equations (Cheluget et al. 1994;
Zafarani-Moattar and Hamidi 2003; Regupathi et al. 2011).
The constants and coefficients with standard deviation
were reported in Table 1. Among the equations proposed
in literature, the best fit was obtained with the nonlinear
Eq 6.
Wp a bW0:5
s cWs dWs
2
The phase equilibrium studies of the selected ATPS are nec-
essary for effective development of real system. Equilibrium
compositions of the phases are responsible for the
partitioning and maintaining the surface properties of
targeted biomolecule in ATPS. The equilibrium phase
compositions (tie line compositions) and tie line lengths
for different feed conditions were obtained at 25 C and
reported in Table 2. The phase diagram, which includes the
binodal curve and tie lines for the PEG 2000 - sodium sulphate
ATPS, is shown in Fig. 9. The equilibrium phase com-
positions for different tie lines were correlated using litera-
ture empirical correlations of Other Tobias (eq. 7) and
Bancroft (eq. 8).
!  n
1Wtp 1Wbs
K 7
Wtp Wbs
6

Where, Wp and Ws are the mass fractions of PEG   !r

Wbw Wtw
2000 - sodium sulphate, respectively. On the basis of K1 8
Wbs Wtp
the obtained standard deviation (0.997), it was conclud-
ed that above mentioned nonlinear equation can be satis-
factory used to predict the binodal compositions for the inves- where K, n, K1, and r are the fitting parameters. Super-
tigated system. scripts t and b represent the polymer-rich phase (top
phase) and the salt-rich phase (bottom phase), respec-
tively. Subscripts p, s and w stand for PEG, salt and
Table 2 Equilibrium Phase composition of PEG 2000 - sodium sulphate
water, respectively. The fitting parameters and the corre-
system 25 C
sponding regression coefficient were reported in Table 3,
Feed composition Top PEG rich Bottom salt rich TLL % which can be used to predict the equilibrium composition of
Phase phase (w/w) the PEG 2000 - sodium sulphate ATPS.
Wp % Ws % Wp % Ws % Wp % Ws %
(w/w) (w/w) (w/w) (w/w) (w/w) (w/w)
Table 3 Othmer-Thobias and Bancroft equation constants for the PEG
15 9 32.5 0.2 2.7 15.4 34.65 2000 - sodium sulphate system at 25 C
15 10 35.3 1.3 2.9 15.6 36.23
Othmer-Thobias equation Bancroft equation
15 12 37.0 0.6 3.4 17.7 39.11
15 15 43.6 1.4 3.7 21.0 45.63 K n R2 K1 r R2
15 18 49.2 3.1 3.9 24.1 50.29 0.24595 1.25467 0.99110 3.02629 0.81171 0.96788
J Food Sci Technol
Conclusion
Aqueous Two Phase Extraction process was experimented to
recover soluble proteins from fish processing industrial efflu-
ent. The experimental results revealed that the PEG 2000 -
sodium sulphate ATPS is a suitable system and maximum
partitioning coefficient and yield of fish proteins (14.53 and
97.75 %, respectively) can be obtained at the following con-
ditions: 15 % (w/w) PEG 200022.5 % (w/w) sodium sul-
phate, pH 5, temperature 25 C, effluent loading 50 % (w/w)
and volume ration 3. Further the equilibrium characteristic of
the selected ATPS was studied and the respective phase dia-
gram also developed. The obtained results demonstrate the
partitioning potential of the selected ATPS, which can be
considered as a first step for the crude fish protein purification
from the bulk industrial effluent.
Acknowledgment The authors acknowledge the grant (Scheme num-
ber: 01(2339)/09/EMR-II) from the Council of Scientific and Industrial
Research (CSIR), Government of India, for this research work.
References
Albertsson PA (1986) Partition of proteins in liquid polymer/polymer
two-phase systems in partition of cell particles and macromolecules,
3rd edn. Wiley, New York, pp 709771
Aral H, Vecchio-Sadus A (2008) Toxicity of lithium to humans and
the environmenta literature review. Ecotoxicol Environ Saf
70(3):34956
Biazus JPM, Santana JCC, Souza RR, Tambourgi EB, Jordo E (2007)
Continuous extraction of - and -amylases from Zea mays malt in
a PEG4000/CaCl2 ATPS. J Chromatogr B 858:227233
Cheluget EL, Gelines S, Vera JH, Weber ME (1994) Liquid-liquid equi-
librium of aqueous mixtures of poly (propylene glycol) with NaCl. J
Chem Eng Data 39:127130
Chethana S, Nayak CA, Raghavarao KSMS (2007) Aqueous two phase
extraction for purification and concentration of betalains. J Food
Eng 81:679687
Chia-Kai S, Been HC (2006) Partitioning and purification of lysozyme
from chicken egg white using aqueous two-phase system. Process
Biochem 41:257263
Ferreira GB, Evangelista AF, Severo Junior JB, Souza RR, Santana JCC,
Tambourgi EB, Jordo E (2007) Partitioning Optimization of pro-
teins from Zea mays malt in ATPS PEG 6000/CaCl2. Braz Arch
Biol Technol 50:557564
Gautam S, Simon L (2006) Partitioning of -glucosidase from
Trichoderma reesei in poly (ethylene glycol) and potassium phos-
phate aqueous two-phase systems: influence of pH and temperature.
Biochem Eng J 30:104108
Graber TA, Maria ET, Asenjo JA, Andrews BA (2001) Influence of
molecular weight of the polymer on the liquid liquid equilibrium
of the poly (ethylene glycol) + NaNO3 + H2O system at 298.15 K. J
Chem Eng Data 46:765768
Guerrerof L, Omil F, Mendez R, Lema JM (1998) Protein recovery during
the overall treatment of waste water from fishmeal factories.
Bioresour Technol 63:221229
Gupta V, Sunil N, Subhash C (2002) Role of water structure on phase
separation in polyelectrolytepolyethyleneglycol based aqueous
two-phase systems. Polymer 43:33873390
Helgi N, Ingrid U (2009) The acid and alkaline solubilisation process for
the isolation of muscle proteins: state of the art. Food Bioproc
Technol 2:127
Hu M, Zhai Q, Jiang Y, Jin L, Liu Z (2004) Liquid Liquid and liquid
liquidsolid equilibrium in PEG + Cs2SO4 + H2O. J Chem Eng Data
49(5):14401443
Kalaivani S, Regupathi I (2013) Partitioning studies of -lactalbumin
in environmental friendly poly (ethylene glycol) - citrate salt
aqueous two phase systems. Bioprocess Biosyst Eng 36:1475
1483
Ketnawa S, Rawdkuena S, Chaiwutb P (2010) Two phase partitioning
and collagen hydrolysis of bromelain from pineapple peel Nang Lae
cultivar. Biochem Eng J 52:205211
Kim SK, Mendis E (2006) Bioactive compounds from marine processing
by-products - a review. Food Res Int 39:383393
Kobsak K, Tipaporn Y, Suvit T, Pisit WN (2008) Protein removal from
fish minces washwater using ohmic heating. Songklanakarin J Sci
Technol 30(3):41419
Kohler K, Veide A, Enfors SO (1991) Partitioning of -galactosidase
fusion proteins in PEG/potassium phosphate aqueous two-phase
systems. Enzym Microb Technol 13:204209
Marcos JC, Fonseca LP, Ramalho MT, Cabral JMS (1998) Variation of
penicillin acylase partition coefficient with phase volume ratio in
poly (ethylene glycol)sodium citrate aqueous two-phase systems. J
Chromatogr B 711:295299
Maria DA, Rodrigo B (2002) Nanofiltration of wastewaters from the fish
meal industry. Desalination 151:131138
Maria DA, Javier F, Rodrigo B (2004) An economic assessment of
proteins recovery from fish meal wastewaters by ultrafiltration.
Trends Food Sci Technol 15:506512
Maria ET, Rocha OA, Graber TA (2001) Liquid-liquid and solidliquid
equilibria of the poly (ethylene glycol) + sodium sulfate + water
system at 298.15 K. J Chem Eng Data 46:308311
Mateusz K, Daniela S (2009) Application of microfiltration and ceramic
membranes for treatment of salted aqueous wastewaters from fish
processing. Desalination 241:227235
Mokhtarani B, Karimzadeh R, Amini MH, Manesh SD (2008)
Partitioning of ciprofloxacin in aqueous two-phase system of
poly (ethylene glycol) and sodium sulphate. Biochem Eng J
38:241247
Naidu RB, Saisubramanian N, Selvakumar D, Janardhanan S,
Puvanakrishnan R (2008) Partial Purification of tannase from asper-
gillus foetidus by aqueous Two phase extraction and its characteri-
zation. Curr trends biotechnol pharm 2(1):201207
Pandey SK, Banik RM (2010) Extractive fermentation for enhanced
production of alkaline phosphatase from bacillus licheniformis
MTCC 1483 using aqueous two-phase systems. Bioresour Technol
102(5):42264231
Patil G, Raghavara KSMS (2007) Aqueous two phase extraction for
purification of C-phycocyanin. Biochem Eng J 34:156164
Perumalsamy M, Batcha MI (2011) Synergistic extraction of bovine
serum albumin using polyethylene glycol based aqueous biphasic
system. Process Biochem 46:494497
Raghavarao KSMS, Martin RG, Paul T (1998) Recent developments in
aqueous two-phase extraction in bio processing. Sep Purif methods
27(1):149
Raghavarao KSMS, Rastogi GK, Karanth NG (1995) Aqueous two-
phase extraction for downstream processing of enzymes/ proteins.
Adv Appl Microbiol 41:97171
Rajni HK (2001) Aqueous two-phase systems methods and protocols.
Humana Press, New Jersey, pp 3246

Ramyadevi D, Subathira A, Saravanan S (2012) Potential recovery of
protein from shrimp waste in aqueous two phase system. Res J
Chem Sci 2(7):4752
Regupathi I, Srikanth KC, Sindhu N (2011) Liquidliquid equilibrium of
poly (ethylene glycol) 2000 + diammonium hydrogen citrate+water
system at different temperatures. J Chem Eng Data 56(9):3643
3650
Regupathi I, Girish B, Prasanna B, Vishwanatha HN (2012) Development
and evaluation of PEG-lithium citrate salt based aqueous two phase
system and its application in partitioning of proteins from fish
industry effluent. Sep Sci Technol 47(4):591598
Rosa PAJ, Ferreira IF, Azevedo AM, Aires-Barros MR (2010) Aqueous
two-phase systems: a viable platform in the manufacturing of
biopharmaceuticals. J Chromatogr A 1217:22962305
Rosa PAJ, Azevedo AM, Sommerfel S, Backer W, Aires-Barros MR
(2011) Aqueous two-phase extraction as a platform in the
biomanufacturing industry: economical and environmental sustain-
ability. Biotechnol Adv 29(6):559567
Sanmartin E, Carlos J, Arboleya VM, Moreno FJ (2009) Recent advances
in the recovery and improvement of functional proteins from fish
processing by-products: use of protein glycation as an alternative
method. Compr rev food sci and food saf 8:332344
Saravanan S, Raghavarao J, Murugesan T, Balachandran UN, Ramasami
T (2006) Recovery of value-added globular proteins from tannery
wastewaters using PEG salt aqueous two-phase systems. J Chem
Technol Biotechnol 81:18141819
Saravanan S, Raghavarao J, Murugesan T, Balachandran UN, Ramasami
T (2007) Partition of tannery wastewater proteins in aqueous two-
phase poly (ethylene glycol)-magnesium sulfate systems: effects of
molecular weights and pH. Chem Eng Sci 62:969978
J Food Sci Technol
Selvakumar P, Chuan Ling T, Walkera S, Lyddiatt A (2012) Recovery of
glyceraldehyde 3-phosphate dehydrogenase from an unclarified
disrupted yeast using aqueous two-phase systems facilitated by
distribution analysis of radiolabelled analytes. Sep Purif Technol
85:2834
Selvaraj R, Vytla RM, Varadavenkatesan T, Vinayagam R, Vinayagam R
(2011) Aqueous two-phase systems for the recovery of biomole-
cules a review. Sci Technol 1(1):716
Shahbaz MH, Omidinia E (2008) Development and application of aque-
ous two-phase partition for the recovery and separation of recombi-
nant phenylalanine dehydrogenase. Iran J Chem Eng 27(2):119127
Silva CAS, Coimbra JSR, Rojas EG, Minima LA, Silva LHM (2007)
Partitioning of caseinomacropeptide in aqueous two-phase systems.
J Chromatogr B 858:205210
Spelzini D, Pico G, Farruggia B (2006) Dependence of chymosin and
pepsin partition coefficient with phase volume and polymer
pausidispersity in polyethyleneglycolphosphate aqueous two-
phase system. Colloids Surf B: Biointerfaces 51:8085
Taskaya L, Jaczynski J (2009) Flocculation-enhanced protein recovery
from fish processing by-products by isoelectric solubilization/
precipitation. Food Sci Technol 42:570575
Tubio G, Nerli B, Pico G (2007) Partitioning features of bovine trypsin
and -chymotrypsin in polyethylene glycol-sodium citrate aqueous
two-phase systems. J Chromatogr B 852:244249
Yucekan I, Onal S (2011) Partition of invertase from tomato in poly
(ethylene glycol)/sodium sulphate aqueous two phase systems.
Process biochem 46:226232
Zafarani-Moattar MT, Hamidi AA (2003) Liquid-liquid equilibria of
aqueous two-phase poly (ethylene glycol)-potassium citrate system.
J Chem Eng Data 48:262265