Process Safety and Environmental Protection 1 0 1 ( 2 0 1 6 ) 27–33

Contents lists available at ScienceDirect

Process Safety and Environmental Protection

journal homepage: www.elsevier.com/locate/psep

Recovery of whey proteins and lactose from dairy

waste: A step towards green waste management

Bipasha Das a , Santanu Sarkar a , Ankur Sarkar a , Sangita Bhattacharjee b ,

Chiranjib Bhattacharjee a,∗
a Department of Chemical Engineering, Jadavpur University, Kolkata, India
b Department of Chemical Engineering, Heritage Institute of Technology, Kolkata, India

a r t i c l e i n f o a b s t r a c t

Article history: Whey is produced as a by-product during cheese and casein manufacture containing some
Received 13 March 2015 important components such as lactose and protein. In this study, ultrafiltration process was
Received in revised form 14 May adopted to separate lactose and protein with high yield and purity from whey using hol-
2015 low fiber module. Ultrafiltration in a diafiltration mode was used in order to improve the
Accepted 18 May 2015 yield of protein in the retentate, which was then freeze-dried to get the end product in
Available online 27 May 2015 dried form. Nanofiltration of the permeate stream from ultrafiltration was done to concen-
trate the lactose part and was similarly freeze-dried. To assure that after freeze–drying the
Keywords: quality of both the protein and lactose was not affected, FTIR analysis was done. The per-
Whey formance of both ultrafiltration and nanofiltration was characterized in terms of permeate
Hollow fiber membranes flux. The influence of transmembrane pressure on both ultrafiltration and nanofiltration
Diafiltration membranes was studied. The effect of fouling was also studied on both ultrafiltration and
Milk proteins nanofiltration membranes and it was observed that the fouling effect was less in case of
Lactose both the membranes and as such the membranes could be re-used for several times for
Freeze–drying effective separation of those components. The quantitative measurements were done for
lactose and protein and it was observed that up to 90% of lactose and 80% of protein recovery
could be achieved using the advanced separation technology. Therefore, the present article
represents a novel approach for separation whey proteins and lactose from dairy waste to
meet the socio economic requirements as well as to mitigate waste disposal problem.
© 2015 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction a number of important nutrients such as lactose (4.5–5% w/v),

Whey is the principal by-product of dairy industry which is
produced during the manufacture of cheese and casein from
milk during coagulation process. Due to the huge production
of whey and its high organic content, exhibiting a biochem-
ical oxygen demand (BOD) ranging from 30 to 50 g L−1 and a
chemical oxygen demand (COD) ranging from 60 to 80 g L−1 ,
this liquid whey is commonly regarded as the “environmental
problem” and is creating a great difficulty for the dairy indus-
try in their disposal (Domingues et al., 1999). Whey contains
∗
Corresponding author. Tel.: +91 98364 02118/+91 33 2457 2699.
E-mail addresses: cbhattacharyya@chemical.jdvu.ac.in,
c.bhatta@gmail.com (C. Bhattacharjee) .
http://dx.doi.org/10.1016/j.psep.2015.05.006
0957-5820/© 2015 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
soluble proteins (0.6–0.8% w/v), lipids (0.4–0.5% w/v) and min-
eral salts (8–10% of dried extract) (Gonzfilez Siso, 1996), so it
can be exploited as a resource of a number of valuable end
products rather than as a waste stream.
Since, whey is a waste product; so it can be utilized as a
cheap source of lactose and protein and those can be used
in food, dairy and pharmaceutical industries. Lactose can
be either directly fermented or it can be hydrolysed to pro-
duce glucose and galactose, whereas proteins are widely used
in food and pharmaceutical products, because they possess
high nutritional value and versatile functional properties
(Jayaprakasha and Yoon, 2005). Therefore, the recovery of lac-
tose and protein can help to reduce the BOD and COD loading
of whey and can help in solving the problem of environmental
pollution being caused by the disposal of whey.
28 Process Safety and Environmental Protection 1 0 1 ( 2 0 1 6 ) 27–33
In last few decades, membrane filtration is routinely used
for a number of applications within the dairy industry. The
recovery of proteins can be done efficiently via ultrafiltration
with high yield and purity (Atra et al., 2005). Microfiltration is
a process by which whey is being pre-treated before ultrafil-
tration to remove suspended fat and casein particles to reduce
the fouling of the ultrafiltration membrane (Cancino et al.,
2006). Nanofiltration is used to remove ions from the feed
solution and can be used to concentrate valuable components
of whey. Usually, the flow pattern through the membranes
can be classified into two categories; dead-end filtration and
cross-flow or tangential flow filtration. However, in tangen-
tial flow, the feed flows parallel to the membrane surface and
is more attractive of the two because it causes less deposi-
tion of solutes on the membrane and helps in alleviating the
particles deposited on the membrane surface (Baker, 2005).
Ultrafiltration in a diafiltration mode using water as a buffer
is a useful technique which helps to increase the concentra-
tion of solutes such as proteins, lactose or other biomolecules
and has an advantage of economical permeate flux (Chollangi
and Hossain, 2007; Cuartas-Uribe et al., 2009). Among several
types of membrane modules, hollow fiber membrane (HFM)
is more acceptable than other modules, due to its high mem-
brane packing density, structural integrity, construction and
thus it can withstand high permeate backpressure (Kuriyel,
2000; Cheryan, 1998; Nielsen, 2000). Hollow fiber membrane
modules have several significant advantages over other con-
ventional membrane modules considering its recovery of flux
after cleaning of the membranes (Cheryan, 1998; Nielsen,
2000). Moreover, being a compact module, its structural con-
figuration allows a high membrane surface area resulting in
the increase of the output through the process, while utiliz-
ing minimal space, with low power consumption (Nielsen,
2000; Winston Ho and Sirkar, 1992). Whey contains different
proteins with different molecular weight and can be eas-
ily separated through ultrafiltration with different molecular
weight cut-off of membranes (Spălăţelu, 2012).
The present experimental work is an attempt to develop an
effective process for separation of proteins and lactose with a
high yield and purity, from whey using hollow fiber module.
The present article also includes ultrafiltration with diafiltra-
tion mode to enhance the yield as well as the performance
of ultrafiltration. In order to concentrate the lactose solutions
obtained during the diafiltration process, nanofiltration (in
diafiltration mode) was carried out where almost clear water
along with ions were obtained in the permeate and lactose was
obtained in the retentate. During microfiltration, ultrafiltra-
tion and nanofiltration, the effect of transmembrane pressure
on permeate flux was studied. Freeze–drying of the concen-
trated protein and lactose solutions was performed to ensure
the better quality of the end products, which can be directly
utilized commercially for different purposes. Overall, the cur-
rent research work enlightens the recovery and purification of
the valuable components from dairy waste by-product, whey,
resulting in a reduction of pollution load on the environment
due to its disposal.
2. Materials and methods
2.1. Chemicals
The whey sample was a gift from a local sweet industry, Hin-
dustan Sweets Pvt. Ltd, Jadavpur, India. Lactose was purchased
from Loba Chemie Pvt. Ltd, Mumbai, India. The chemicals:
Sodium hydroxide (NaOH, MERCK, Mumbai, India), Sodium
hypochlorite (NaOCl, MERCK, Mumbai, India), orthophos-
phoric acid (MERCK, Mumbai, India), sulphuric acid (H2 SO4 ,
MERCK, Mumbai, India), phenol (MERCK, Mumbai, India) were
of analytical grade.
2.2. Experimental set-up
The hollow fiber membrane module was used for our exper-
imental work. The whole setup was purchased from local
manufacturer. Hollow fiber microfiltration membrane with
average pore diameter of 0.45 ␮m and ultrafiltration mem-
brane with MWCO (molecular weight cut-off) of 8 kDa having
average surface area of 0.028 m2 were used. The solution was
fed to the system using a pump and the inlet and outlet
pressures were adjusted using globe valves. The schematic
representation of the whole setup has been shown in Fig. 1.
2.3. Experimental method
The whey sample was collected and cold centrifuged at
11,000 rpm for 10 min to remove residual casein particles and
fat. Hollow fiber membranes are extruded in an in-house
extrusion machine. The centrifuged whey was filtered using
hollow fiber microfiltration (MF) membrane to further remove
the suspended casein and fat particles to minimize the fouling
of membrane during ultrafiltration. The microfiltration study
was done at different transmembrane pressure e.g. 0.5, 0.75
and 1 bar. In the subsequent stage the microfiltered whey was
subjected to ultrafiltration using the hollow fiber membrane
module with 8 kDa molecular weight cut-off (MWCO). The
experimental runs were conducted in a system where perme-
ate was collected in a separate tank while the retentate from
the membrane is recycled back to the feed. The schematic rep-
resentation of experimental methodology has been shown in
Fig. 2.
The ultrafiltration was carried out by collecting the perme-
ate stream in a separate tank until the volume concentration
factor (VCF) becomes 2, i.e. feed solution volume was reduced
to 50% of its initial volume. The VCF was determined as the
ratio between the initial (Vinitial ) and final volumes (Vfinal ) in
the feed tank as shown in the following equation:
Vinitial
VCF = (1)
Vfinal
In the subsequent stage, equal amount of distilled water
was added to the ultrafiltration retentate and the first stage
of diafiltration was carried out. The process was stopped until
Hollow fibre membrane P
P
Retentate
Back pressure
regulator
Permeate
Flow meter
By-pass Valve
Feed
Tank
Feed pump
Fig. 1 – Schematic representation of hollow fiber module
setup.
 Process Safety and Environmental Protection 1 0 1 ( 2 0 1 6 ) 27–33 29

Fig. 2 – Flow diagram representing the experimental methodology.

VCF 2 has been achieved. Similarly, a second stage diafiltration
was carried out and added water was removed as permeate.
Permeate or membrane flux was calculated according to the
following equation:
1 dV
J= (2)
A dt
where J is the permeate flux (L/m−2 h−1 ), A the area of the
membrane (m2 ), V the filtrate volume (L) and t is the unit time
(h).
Since, 8 kDa ultrafiltration hollow fiber membrane was
used, so most of the protein was retained and the lactose with
some salts passed through the membrane and was collected as
permeate. The lactose permeate obtained in ultrafiltration was
passed through nanofiltration membrane having molecular
weight cut-off of 0.2 kDa so as to obtain concentrated lactose
in the retentate. Nanofiltration was also carried out in a diafil-
tration mode. To determine the fouling effect, water runs were
performed before and after the filtration process. After com-
pletion of separation process, those membranes were cleaned
with sodium hydroxide and sodium hypo chloride solution
according to washing protocol of the membranes.
Thereafter, the protein enriched stream obtained after
ultrafiltration and lactose enriched stream obtained after
nanofiltration, was previously frozen for drying under vac-
uum in a freeze dryer at a temperature of 20 ◦ C and pressure
5.5 mm of Hg. Freeze–drying (lyophilization) is a dehydra-
tion process which involves the removal of water or other
solvent from a frozen product by sublimation and desorp-
tion. The freeze–drying process consists of three stages:
freezing, primary drying and secondary drying (Roy and
Gupta, 2004). The structural changes because of freeze–drying
have been investigated, especially by FTIR (Fourier-transform
IR, IRAffinity-1, Shimadzu) spectroscopy. In general, drying
results in a decrease of alpha-helix and random struc-
ture and an increase in beta-sheet structure in proteins
recovered.
Cary 50Bio, Part No. EL07113760). Using a standard curve and
the absorbance values, the lactose content in the sample was
measured.
For determination of protein concentration, 1 ml of the
sample was taken in a test tube. 5 ml of the Bradford dye
was added to the tube and mixed by vortex or inversion. It
was allowed to stand for 5 min and then the absorbance of
the mixed sample was measured at 595 nm with UV–vis spec-
trophotometer (Chollangi and Hossain, 2007). The unknown
concentration of the protein in the sample was determined
using the BSA standard curve.
4. Result and discussion
4.1. Fouling effect on microfiltration
The suspended particles as well as fats were separated
through centrifugal separation process from the whey sam-
ple, which helps to reduce the load on the microfiltration
process to some extent. Microfiltration was carried out under
different transmembrane pressure of 0.50, 0.75 and 1.00 bar.
During the process, for a fixed transmembrane pressure, the
permeate flux was declining in nature with time due to depo-
sition of suspended particles on the membrane surface, which
resulted in an increase of the thickness of deposited layer due
to the increase of the concentration polarization. The varia-
tion of permeate flux with time for different TMP was shown
in Fig. 3. Moreover, the rate of decrement of the flux is less sig-
nificant because the flow was tangential for which the effect
of fouling reduced to some extent. This pre-treatment pro-
cess was performed to minimize the fouling effect as well as

3. Analytical method

The quantification of lactose and protein was one of the major

tasks for the present study. For determining the lactose con-
centration, 1 ml of sample is taken into a test tube and 1 ml
of 5% phenol in water is added. Then 5 ml of concentrated
sulphuric acid is added rapidly directed against the liquid sur-
face rather than the wall of the test tube in order to obtain
a good mixing (Dubois et al., 1956). The tubes are allowed
to stand for 10 min, and then they are shaken and placed
in a water bath at about 25 ◦ C–30 ◦ C for 10 to 20 min. The
absorbance of the cooled sample was taken at a wavelength Fig. 3 – Flux variation during microfiltration with different
of 490 nm with the help of UV–vis Spectrophotometer (Varian TMP.
30 Process Safety and Environmental Protection 1 0 1 ( 2 0 1 6 ) 27–33

Fig. 4 – The flux variation during diafiltration (UF) of casein whey at different TMP (inset showing the initial and after run
pure water flux at corresponding TMP).

Fig. 5 – The flux variation during diafiltration (NF) of casein whey at different TMP (inset showing the initial and after run
pure water flux at corresponding TMP).

to attain maximum flux in protein separation process through
ultrafiltration.
After completion of microfiltration, membrane was
cleaned thoroughly following the washing protocol and water
flux was measured and compared with initial pure water flux
prior to the microfiltration process. It was observed that both
the values were almost same. Therefore, it can be stated that
in real scenerio the hollow fiber microfiltration membrane can
be used for a long time with proper maintenance.
4.2. Ultrafiltration of casein whey
Ultrafiltration was carried out to separate lactose and protein
from casein whey using hollow fiber membrane with molecu-
lar weight cut-off of 8 kDa. As all the proteins content in whey
are of molecular weight more than the MWCO of the mem-
brane, these proteins were retained in feed solution whereas
lactose and other salts will pass through the membrane as
permeate stream.
4.2.1. Effect of transmembrane pressure on ultrafiltration
of casein whey
For any pressure driven membrane process, TMP is the main
driving force during separation process. The effect of trans-
membarne pressure on the permeate flux was thus studied
and are shown in Fig. 4. It is quite obvious that after first
stage diafiltration and second stage diafiltration, the permeate
flux will increase with increasing transmembrane pressure.
This was expected as the increase in transmembrane pressure
increases the driving force on the membrane surface resulting
a higher permeate flux value. The diafiltration was carried out
at VCF 2, calculated using Eq. (1). For a constant pressure, in
each stage of diafiltration the flux was improved (Fig. 4) due to
addition of the fresh water which resulted in the dereament of
concentration polarization on membrane surface. Although,
it was not possible to recover the flux to its initial value, as
some sort of fouling will occur during the filtration process
which was not recoverable. Two stage diafiltration was per-
formed for better separation of proteins and lactose from the
whey sample.
 Process Safety and Environmental Protection 1 0 1 ( 2 0 1 6 ) 27–33
Fig. 6 – FTIR chromatogram of UF retentate and freeze-dried retentate.
4.2.2. Flux recovery during ultrafiltration
It is very important to regain the original flux of a mem-
brane after the separation process and to do so a thorough
cleaning of membranes is required using suitable protocol to
ensure the reusability of the membrane for a longer period of
time. After performing ultrafiltration process, the hollow fiber
memebrane module was cleaned and water flux was verified
with its initial one. The comparison of the flux recovery has
been shown in the inset of Fig. 4. It has been observed that
upto 99% flux recovery was possible which ensured very less
parmanent fouling of the membrane. Thus, a single hollow
fiber membrane module can be used several times to separate
protein and lactose from casein whey effectively.
4.3. Nanofiltration of casein whey
Lactose is one of the major components of casein whey.
Nanofiltration (NF) of the UF permeate is thus required as it is
highly efficient in retarding organic compounds with molecu-
lar weights greater than 300 Da. The NF membranes are highly
permeable for monovalent salts (as NaCl and KCl) and for the
present study, NF membranes with 200 Da MWCO were cho-
sen for the purpose of separation and purification of lactose
from the salts retained in ultrafiltration permeates.
4.3.1. Effect of transmembrane pressure on nanofiltration
of casein whey
The effect of transmembrane pressure on the recovery of lac-
tose from the ultrafiltration permeate was studied and shown
31
in Fig. 5. The two stage discontinuous diafiltration was car-
ried out by adding water to the reduced feed as mentioned in
Section 2.3. Generally, discontinuous diafiltration was used to
enhance the permeation of salts through the membrane and
improve the total yield of lactose recovery.
The figure showed that there was a linear correlation
between the transmembrane pressure difference and the per-
meate flux and it increased with increased TMP. However,
higher range of TMP value showed better flux values than its
lower ranges. For example, with increase of TMP value from
6 bar to 8 bar, the average steady state permeate flux increased
only 12% whereas it enhanced almost 26% when TMP value
increased from 8 bar to 10 bar.
4.3.2. Flux recovery during nanofiltration
Similar to the ultrafiltration process, after every nanofiltra-
tion run, the membrane was thoroughly washed and the pure
water flux was measured and compared with its initial one.
The results were shown in the inset of Fig. 5. The better flux
recovery was obtained when nanofiltration was performed
with low TMP values. It was found that for 6, 8, and 10 bar,
the flux recovery were 98.5%, 97% and 95.5%, respectively.
It was obvious because of the increased fouling rate on the
membrane surface with increased TMP values, which cause
some permanent blockage on the membrane surfaces. So,
with increased TMP values, the total output of the process
might be enhanced with increased permeate flux. However,
simultaneously, it can cause more damages on the membrane
surface reducing its total life cycle.
32 Process Safety and Environmental Protection 1 0 1 ( 2 0 1 6 ) 27–33
Fig. 7 – FTIR chromatogram of NF retentate and freeze-dried retentate.
4.4. Protein (from UF retentate) and lactose (from NF
retentate) recovery in dried form using freeze drying
Protein enriched UF retentate was freeze dried to ensure the
use of whey proteins commercially. Freeze–drying is a very
essential part of the present study to reduce the volume of UF
retentate and produce it in dried form. For quantitative deter-
mination of protein in the freeze dried product, Bradford assay
was used and it was observed that up to 80% of the whey pro-
tein could be recovered using the present advanced separation
strategy.
The quality of the protein was determined with the help
of FTIR. The N H stretches of amines are in the region
3300–3000 cm−1 . These bands are weaker and sharper than
the O–H stretches of alcohol, which appear in the same region.
The N H bending vibration of primary amines is observed in
the region 1650–1580 cm−1 . The comparison between protein
present in UF retentate and freeze-dried protein has been por-
trait in Fig. 6 and it has been assured that the quality of the
protein has not been affected during freeze–drying. Therefore,
the protein recovery in the present system has been estab-
lished as better alternative.
NF retentate contained mainly the lactose of the origi-
nal whey. The NF retentate was freeze dried to produce it
in dried form and the qualitative and quantitative determi-
nations were done. For quantitative determination of lactose
in the freeze-dried form, phenol-sulphuric acid method was
used. Lactose recovery of approximately 90% was achieved
using this advanced separation technology.
The quality of the lactose in freeze-dried form was deter-
mined using FTIR. The chromatogram (Fig. 7) shows similar
results like its initial counterpart for protein determination,
which confirms that the quality of lactose after freeze–drying
was not affected.
5. Conclusion
The effectiveness of the developed protocol has been well
established to recover proteins as well as lactose from whey.
The main advantage of the present experimental study is
fouling tendency of the microfiltration, ultrafiltration as well
as nanofiltration has been removed to a great extent using
HFM configuration due to cross flow action. At higher TMP
the flux is more without hampering the performances of
the membrane such as flux recovery for both ultrafiltration
and nanofiltration process. Moreover, from the FTIR analy-
sis quality of freeze-dried protein was found to be same as
before freeze–drying and at the same time lactose recovery
is a highlighted area of this work. Using the present separa-
tion methods, approximately 90% of lactose and 80% of protein
could be recovered in the form of freeze-dried products. The
recovered whey proteins and lactose can further be converted
high value end products such as whey protein concentrate
(WPC) and bioethanol, respectively. Therefore, both the goals
have been achieved through the present scheme, first one is
the effective separation of lactose as well as protein and later
one is to lower down of dreadful impact on environment due
to discharge from the dairy industry. In future, the scale up of
 Process Safety and Environmental Protection 1 0 1 ( 2 0 1 6 ) 27–33
the process should be the area of research interest for possible
commercialization of developed methodology.
Acknowledgements
The work reported in this article is part of an Indo-Norwegian
Cooperation Programme 2014(INCP) (vide sanction letter no.
58-3/2014(IC) dated December 26, 2014), entitled “Research
and education within advanced hybrid separation techniques
in industrial wastewater treatment”, funded in India by
University Grants Commission (UGC, Government of India).
Accordingly, the contribution of UGC (India) is gratefully
acknowledged.
References
Atra, R., Vatai, G., Bekassy-Molnar, E., Balint, A., 2005.
Investigation of ultra- and nanofiltration for utilization of
whey protein and lactose. J. Food Eng. 67, 325–332.
Baker, R.W., 2005. Membrane Technology, Kirk-Othmer
Encyclopedia of Chemical Technology. John Wiley & Sons, Inc,
New Jersey, pp. 1–66.
Cancino, B., Espina, V., Orellana, C., 2006. Whey concentration
using microfiltration and ultrafiltration. Desalin 200, 557–558.
Cuartas-Uribe, B., Alcaina-Miranda, M.I., Soriano-Costa, E.,
Mendoza-Roca, J.A., Iborra-Clar, M.I., Lora-García, J., 2009. A
study of the separation of lactose from whey ultrafiltration
permeate using nanofiltration. Désalin 241, 244–255.
33
Cheryan, M., 1998. Ultrafiltration and Microfiltration Handbook.
Technomic Pub. Co, Lancaster, pp. 527.
Chollangi, A., Hossain, Md.M., 2007. Seperation of proteins and
lactose from dairy wastewater. Chem. Eng. Process.: Process
Intesfication 46, 398–404.
Domingues, L., Dantas, M.M., Lima, N., Teixeira, J.A., 1999.
Continuous ethanol fermentation of lactose by a recombinant
flocculating Saccharomyces cerevisiae strain. Biotechnol Bioeng.
64, 692–697.
Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F.,
1956. Colorimetric method for determination of sugars and
related substances. Anal. Chem. 28, 350–356.
Gonzfilez Siso, M.I., 1996. The biotechnological utilization of
cheese whey: a review. Bioresour. Technol. 57, 1–11.
Jayaprakasha, H.M., Yoon, Y.C., 2005. Production of functional
whey protein concentrate by monitoring the process of
ultrafiltration. Asian-Aust. J. Anim. Sci. 18 (2005),
433–438.
Kuriyel, R., 2000. Ultrafiltration, Kirk-Othmer Encyclopedia of
Chemical Technology. John Wiley & Sons, Inc, New Jersey, pp.
1–22.
Nielsen, W.K. (Ed.), 2000. Membrane Filtration and Related
Molecular Separation Technologies. APV Systems, Denmark,
p. 223.
Roy, I., Gupta, M.N., 2004. Freeze–drying of proteins: some
emerging concerns. Biotechnol. Appl. Biochem. 39, 165–177.
Spălăţelu (Vicol), C., 2012. Biotechnological valorisation of whey.
Innovative Rom. Food Biotechnol. 10, 1–8.
Winston Ho, W.S., Sirkar, K.K., 1992. Membrane Handbook.
Chapman & Hall, New York, NY, pp. 954.