───Processes and equipment of food productions───
Two-stage whey treatment by
nanofiltration and reverse osmosis
1 1
Ivanna Kyrychuk , Valerii Myronchuk ,
1 2
Yurii Zmiievskіi , Serhii Holiachuk
1 – National university of food technologies, Kyiv, Ukraine
2 – Lutsk national technical university, Lutsk, Ukraine
Abstract
Keywords: Introduction. Treatment of whey, which is a by-product
of cheese manufacturing process, is of great interest for dairy
Whey
industry due to its high content of valuable and nutritional
Dairy compounds. Although the development of complex whey
Nanofiltration treatment by membrane processes and use of its derivative
Reverse osmosis products is still current.
Permeate Materials and methods. The raw whey and permeate
after nanofiltration of whey were used for the study. The
experiments were carried out in the pressure-driven
Article history: laboratory setup of dead-end type using nanofiltration
OPMN-P and reverse osmosis NanoRo membranes.
Received 21.09.2015 Results and Discussion. Considering high lactose content
Received in revised in whey and results obtained during lactose solutions
form 14.11.2015 filtration using OPMN-P membrane, it was proved that whey
Accepted 24.12.2015 should be concentrated by nanofiltration to total solids of 20-
22%. During whey concentration it was observed two stages
of permeate flux decrease: rapid decrease of flux at the
beginning of the process and further gradual flux decrease.
The first is caused by membrane fouling and the latter is
attributed to concentration polarization, formation and
growth of the cake layer.
From the analysis of the obtained permeate flux-pressure
and retention-pressure curves for reverse osmosis membrane
Corresponding NanoRO it was found that the rational value of pressure for
author: the concentration of nanofiltration whey permeate is 3.0
MPa. At this pressure, permeate flux decreased twice with
Ivanna Kyrychuk
increase of solution concentration from 6 to 40 g/L, while
E-mail:
average salt and lactose retention was 96.0% and 97.5%
[email protected]
respectively. Based on the obtained results, the scheme of
two-stage whey treatment was developed.
Conclusions. The obtained results of the study on two-
stage whey treatment by nanofiltration and reverse osmosis
can be used in the technology of complex whey processing in
the dairy plant. It allows using all the whey components,
obtaining the purified water for reuse and partially
eliminating problem of environmental pollution by dairy
plants.
638 ───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ──
─── Processes and equipment of food productions───
Introduction
Whey is a by-product of cheese production which is rich in valuable components. Up
to 1960’s it was considered to be a waste of cheese manufacture, until the beginning of
application of membrane processes in the dairy industry. Whey contains lactose (4.0-5.0%),
proteins (0.6-1.0%), minerals (0.5-0.9 %), e.g., calcium, magnesium, phosphorus, vitamins,
and milkfat in small concentrations [1-3]. It cannot be discharged to the environment or
released into wastewaters because of its high content of organic compounds, high volume
(often 90% of the mass of milk used) and its extremely high biological and chemical
oxygen demand.
Nowadays, whey treatment by membrane processes is of great interest for researchers
as via their using it is possible to recover useful products and to alleviate the pollution
problem. The advance of whey treatment is caused by three main factors: an increase in
costs of its release, the emergence of new technologies in extraction of whey protein and
scientific researches, due to which valuable nutritional and biological properties of this
product was found.
Among the membrane processes nanofiltration is the most suitable process for pre-
concentration and partial demineralization of whey at the same time [4-6]. Due to high
permeability of nanofiltration membranes for monovalent salts (such as NaCl, KCl) it is
possible to remove them from whey. Monovalent ions (sodium, potassium and chloride) are
undesirable components of food products due to their salty taste and negative health
impacts [7, 8]. Moreover, such pre-concentration by nanofiltration is desirable before
further whey treatment by electrodialysis for deep demineralization [9]. It can be explained
by increasing whey conductivity and reducing its volume that results in lowering of load on
electrodialysis equipment and increasing its efficiency [10]. However, nanofiltration
membranes have low permeability for organic compounds with molecular weight less than
300 Da. That’s why nanofiltration permeate may contain some lactose (up to 0.3%) that
mainly depends on the nanofiltration membranes properties [11-14].
Whey permeate, which volume is approximately 65% of treated whey, is not usually
used and is discharged to the waste. Considering the current demands to the composition
and concentration of wastewaters its chemical oxygen demand must not exceed 500 mg
О2/dm3. Although, according to the literature data, chemical oxygen demand of
nanofiltration whey permeate can reach up to 3000 mg О 2/dm3 [5, 15], mainly because of
the lactose. Therefore, it must be pretreated before be released. The most appropriate
method for its purification is reverse osmosis since reverse osmosis membranes allow
concentrating and removing all the solutes presented in the feed and obtaining water for
reuse.
The aim of this work was to study two-stage whey treatment by nanofiltration and
reverse osmosis for the development of the whey processing technology. The choice of NF
in the first stage was based on higher water flux at lower pressure. Reverse osmosis was
chosen for the second stage due to high lactose and minerals rejection.
───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ── 639
───Processes and equipment of food productions───
Materials and methods
The raw whey was used for the experiments. Its composition is presented in the Table
1. “Edible” lactose was used for preparation of model solutions of lactose.
Table 1
Composition of whey
Parameter Fat Protein Lactose Mineral salts Dry matter
Concentration, % 0.35 1.0 3.5 0.7 6.0
A pressure-driven laboratory setup of dead-end type (Fig. 1) with membrane effective
area of 1.38∙10–3 m2 was used for the study of membrane separation of whey and its
nanofiltration permeate. It consists of gas cylinder (not shown), membrane cell and
magnetic stirrer 4. The membrane cell include two covers 1, 2 and metal cylinder 10. The
porous support 6 and membrane 5 were placed in its bottom part and pressed by metal
cylinder 10. Stir bar 3 impelled by magnetic stirrer 4 was put over the membrane 5. The
special hole was made in the bottom cover 2 of the membrane cell for collecting permeate
through the tube 11. With open fittings 7 and 8, a feed solution was introduced through one
of them into the working chamber. Pressure gauge 9 was attached to fitting 8 for
monitoring pressure in the middle of the unit and fitting 7 was connected to a pressure
regulator mounted on the inert gas cylinder (not shown). The working pressure in the cell
was created by opening the valves on the gas cylinder and the pressure regulator 12. The
temperature of solutions during the experiments was in the limits 20±2°C. The temperature
inside the membrane cell was measured controlled by thermal couple 13.
Nanofiltration membrane OPMN-P (ZAO STC “Vladipor”, Russian Federation) was
used for whey concentration. Reverse osmosis membrane NanoRO (ZAO “RM Nanotech”,
Russian Federation) was used for separation of nanofiltration whey permeate. Before
separation, each membrane was soaked in deionized water for at least 12 h. Then they were
compacted at pressure of 2.0 MPa for nanofiltration membrane and 4.0 MPa for reverse
osmosis membrane by filtering distilled water through them until a steady flux was
established.
The chemical composition of feed, retentate and permeate was determined by standard
methods. Dry matter content was measured by a refractometer URL-1. Lactose
concentration was determined by iodometric method. The mineral salts content was
measured by a conductivity meter (HANNA Instruments DIST 1). Ion content of Ca 2+,
Mg2+ was determined by atomic absorption (Pye Unicam 8800 UV/VIS, Philips). K +, Na+
ion content was measured by flame photometer (PFM-U4.2, Analitpribor).
640 ───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ──
─── Processes and equipment of food productions───
Fig. 1. Schematic diagram of the dead-end laboratory setup:
1, 2 – covers; 3 – stir bar; 4 – magnetic stirrer; 5 – membrane; 6 – porous support; 7, 8 – fittings; 9 –
pressure gauge; 10 – metal cylinder; 11 – tube for permeate outlet; 12 – pressure regulator; 13 –
thermal couple.
Permeate flux J (L/(m-2∙h-1) is the volume of permeate V (L) collected per unit
2
membrane area S (m ) per unit time t (s):
J 3600 V . (1)
St
The membrane retention R of any feed component was calculated as:
CP
R 1 100% , (2)
CR
where CP and CR are the permeate and the retentate concentrations respectively.
Volume reduction ratio (VRR) vs. time was calculated as:
V
Vf t f t
VRR VR t Vf t VP t , (3)
where Vf t , VR t , VP t is the feed, retentate and permeate volume at time t ,
respectively.
───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ── 641
───Processes and equipment of food productions───
Results and discussion
Analyzing the composition of whey (Table 1) [1, 2] it can be seen that lactose is up to
70% of its total solids. It is very important to concentrate all the lactose while whey
processing because of its high chemical oxygen demand. That’s why lactose rejection of
nanofiltration membrane OPMN-P must be high. Therefore the separation characteristics of
OPMN-P membrane were previously studied during filtration of lactose solutions.
Lactose filtration was carried out at pressure of 2.0 MPa. The obtained results are
presented in Fig. 2. It can be seen that increase of lactose concentration from 5 to 18% leads
to permeate flux decrease approximately by 10 times while lactose retention is very high
and remains almost constant within 98%. The decrease in permeate flux is caused by
concentration polarization and increase in osmotic pressure of the solution near the
membrane surface. The osmotic pressure of lactose solution at concentration of 5% and
18% is 0.4 and 1.28 MPa respectively. Besides, the initial lactose crystallization may occur
at high lactose concentration. Saturation of aqueous solution with lactose happens at
concentration of 19.2 g /100 g H2O at temperature of 20 °C [16]. Thus whey should be
concentrated to dry matter content of 20-22%. High retention of OPMN-P membrane is
probably caused by formation of dynamic membrane on membrane surface that additionally
prevents lactose penetration through nanofiltration membrane into permeate. This
phenomenon was discussed in the previous paper during reverse osmosis of lactose
solutions [17].
60 100
%
R,
J, L/(m2 ·h)
50 98
40
96
30
94
20
10 92
0 90
5 10 15 20 5 10 15 20
C, % С, %
a b
Fig. 2. Dependence of permeate flux J (a) and retention R (b) of OPMN-P membrane on
concentration C of lactose solution (ΔP=2.0 MPa, t=20 °C)
Whey concentration by nanofiltration was performed at 2.0 MPa that corresponds to
high permeate flux and retention as was established in paper [12]. Appliance of higher
pressure can cause severe membrane fouling and pore blocking [18]. Lower pressure is
insufficient because of the low permeate flux and large membrane area needed for
separation.
In Fig. 3 permeate flux is plotted vs. dry matter content during the concentration of
whey. It can be observed that flux decreases with concentration mainly due to the increase
in the osmotic pressure. The curve at the Fig. 3 can be divided into two parts: the rapid
642 ───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ──
─── Processes and equipment of food productions───
permeate flux decrease at the beginning of the filtration and further gradual flux decrease.
At the first stage, flux reduced almost twice, when the concentration of 8% was reached.
The reason of this can be fouling of the membrane by whey components caused by
adsorption of proteins on membrane surface [19]. Due to large molecular size of proteins,
i.e. α-lactalbumin (3.0 nm), β-lactalbumin (4.0 nm), caseins (25-130 nm), its low mobility
and small pore size of nanofiltration membrane (in the range of 0.1-1.0 nm) they deposit on
membrane surface and form a dynamic membrane. At the second stage, the further decrease
of permeate flux is attributed to concentration polarization, formation and growth of the
cake layer [11, 19]. A cake layer is formed mainly of salts (calcium and phosphate ions)
and partially of lactose. This layer creates an additional resistance to permeate flow.
35
J, L/(m2·h)
30
25
20
15
10
5
0
4 7 10 13 16 19 Dry matter
content, %
Fig. 3. Relationship of the permeate flux as a function of the dry matter content for OPMN-P
membrane during whey concentration
(ΔP=2.0 MPa, t=20 °C)
The whey concentration was carried out until dry matter content reached 18-19%. At
this point the permeate flux was less than 5 L/(m 2∙h) (Fig. 3). Retention of OPMN-P
membrane was high for macromolecular substances (fat and protein). Lactose retention was
in the range of 93-96% during whey concentration, and minerals retention was 56-62%.
The composition of obtained permeate is shown in Table 2. As it can be seen, nanofiltration
whey permeate consists of lactose (50% of total solids) and minerals (50%) presented by
multivalent ions Ca2+, Mg2+ and monovalent ions Na+, K+, Cl-. In the complex processing
technology for whey this permeate should be concentrated by reverse osmosis.
Table 2
Composition of nanofiltration whey permeate
Parameter Total Lactose Minerals Ca2+ Mg2+ Na+ K+ Cl-
solids
Value, g/L 6,0±0,1 3,0±0,1 2,9±0,1 0,015 0,009 0,264 1,248 1,364
───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ── 643
───Processes and equipment of food productions───
The dependence of the permeate flux and retention for NanoRO membrane on
pressure is shown in the Fig. 4 during filtration of whey permeate. It can be observed
almost proportional rise of permeate flux J to the increase of the pressure P to 4 MPa (Fig.
4 curve 1). Retention R for dry substances (Fig. 4 curve 2) increases with pressure up to a
transmembrane pressure of 3.0 MPa. It increases slowly from 3.0 to 4.0 MPa, and
obviously it will remain constant with further pressure rise. Based on this, pressure of 3.0
MPa was chosen to minimize energy consumption and to perform concentration of
nanofiltration whey permeate by reverse osmosis.
90 98
J, L/(m2 ·h)
80 97
R, %
2
70 96
60
95
50
1 94
40
93
30
92
20
10 91
0 90
0 1 2 3 4 5
ΔP, MPa
Fig. 4. Experimental data on separation of nanofiltration whey permeate for reverse osmosis
membrane NanoRO:
1 – relationship of the permeate flux J as a function of the pressure P;
2 – relationship of retention R as a function of the pressure P.
During concentration of nanofiltration whey permeate, flux decreased continuously
with increasing VRR, i.e. feed concentration factor (Fig. 4 a). It reduced in 15 L/(m 2∙h)
when VRR 1.5 was reached. At concentration factor higher than 1.5, the flux decreased
gradually. This fact can be explained as a consequence of the concentration polarization
layer formation on the membrane surface and membrane pore blocking by solution
components. The pore blocking increases the membrane resistance while the retained
particles on membrane creates an additional layer of resistance to the permeate flow. It also
leads to the raise of osmotic pressure of the solution. At the end of the filtration, the
concentration of nanofiltration whey permeate was 40 g/L including 26.2 g/L of lactose and
13.8 g/L of minerals.
Retention for lactose and minerals decreased gradually with VRR increase (Fig. 4 b).
It can be explained as follows: due to formation of concentration polarization layer on
membrane surface the filtration through the membrane occurs from the layers with the
enhanced concentration. Thus the permeate concentration increases and retention decreases.
The average salt retention was 96.0% and lactose retention was 97.5%. As the result,
permeate contained 0.21 g/L of lactose and 0.12 g/L of mineral salts.
644 ───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ──
─── Processes and equipment of food productions───
70 98,0
·h)
2
97,5
60
(m
2
J, L/
97,0
50 96,5
96,0
40
95,5
1
95,0
30
94,5
20 94,0
1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5
VRR VRR
a b
Fig. 4. Permeate flux J (a) and retention R (b) of NanoRO membrane during the concentration
of nanofiltration whey permeate (ΔP=3.0 MPa):
1 – minerals; 2 – lactose.
The scheme of two-stage whey treatment was developed based on the results of the
study (Fig. 5). It includes nanofiltration at the first stage and reverse osmosis at the second.
The obtained whey retentate after nanofiltration can be further concentrated up to 50% total
solids by evaporation or demineralized by electrodialysis. The retentate after reverse
osmosis can be used in non-lactose milk production to recover the mineral salt content [RU
Patent No. 2305196, 2007]. Reverse osmosis permeate with low lactose and salt content
can be discharged or used for cleaning, pre-rinsing, for washing floors and the outside of
plant and vehicles. Such a two-stage membrane treatment allows all the whey components
to be completely used and up to 90% of purified water on the amount of treated permeate
after nanofiltration of whey to be received that can be reused in the dairy plant.
Fig. 5. Material flows on the process flow diagram of two-stage whey treatment by
nanofiltration and reverse osmosis:
V – volume of the solution; C – concentration of total solids.
───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ── 645
───Processes and equipment of food productions───
Conclusions
Based on the results of the study it can be concluded that whey should be concentrated
to total solids of 20-22% by nanofiltration according to the results of lactose solutions
separation by nanofiltration. It was found that nanofiltration whey permeate contains 50%
of organic compounds (lactose) and 50% of inorganic components. Due to its high
chemical oxygen demand it should be previously treated before be discharged. The rational
pressure for its concentration by reverse osmosis is 3.0 MPa. The benefit of using NF+RO
cascade treatment of whey is full use of its components and the recovery of water suitable
for reuse in the dairy plant.
References
1. Myronchuk V. H., Zmiievskyi Yu. H. (2013), Membranni protsesy v tekhnolohii
kompleksnoi pererobky molochnoi syrovatky, NUKhT, Kyiv.
2. Akpinar-Bayizit A., Ozcan T., Yilmaz-Ersan L. (2009), Membrane processes in
production of functional whey components, Mljekarstvo, 59 (4), pp. 282-288.
3. Suárez E., Lobo A., Álvarez S., Riera F A., Álvarez R. (2006), Partial
demineralization of whey and milk ultrafiltration permeate by nanofiltration at pilot-
plant scale, Desalination, 198, pp. 274–281.
4. Bidhendi G. N., Nasrabadi T. (2006), Use of nanofiltration for concentration and
demineralization in the dairy industry, Pakistan Journal of Biological Sciences, 9(5),
pp. 991-994.
5. Yorgun M. S., Balcioglu I. A., Saygin O. (2008), Performance comparison of
ultrafiltration, nanofiltration and reverse osmosis on whey treatment, Desalination,
229, pp. 204–216.
6. Turan M. (2004), Influence of filtration conditions on the performance of
nanofiltration and reverse osmosis membranes in dairy wastewater treatment,
Desalination, 170, pp. 83-90.
7. Rice G., Barber A., O’Connor A., Stevens G., Kentish S. (2009), Fouling of NF
membranes by dairy ultrafiltration permeates, Journal of Membrane Science, 330,
pp. 117–126.
8. Rice G., Barber A. R., O’Connor A. J., Stevens G. W., Kentish S. E. (2011),
Rejection of dairy salts by a nanofiltration membrane, Separation and Purification
Technology, 79, pp. 92–102.
9. Simova H., Kysela V., Cernin A. (2010), Demineralization of natural sweet whey by
electrodialysis at pilot-plant scale, Desalination and Water Treatment, 14, pp. 170-
173.
10. Varivoda A. A. (2014), Molochnaya syvorotka membrannoy obrabotki v tekhnologii
plavlenykh syrov, Mezhdunarodnyy nauchno-issledovatel'skiy zhurnal,#2 (21), pp.
80-84.
11. Cuartas-Uribe B., Alcaina-Miranda M. I., Soriano-Costa E., Bes-Pia A. (2007),
Comparison of the behavior of two nanofiltration membranes for sweet whey
demineralization, Journal of Dairy Science, 90, pp. 1094–1101.
12. Zmiievskyi Yu. H., Kyrychuk I. I., Myronchuk V. H. (2013), Porivnialnyi analiz
protsesiv nanofiltratsii ta zvorotnoho osmosu pry rozdilenni molochnoi syrovatky,
Naukovi pratsi Odeskoi natsionalnoi akademii kharchovykh tekhnolohii, 2(43), pp.
137-141.
646 ───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ──
─── Processes and equipment of food productions───
13. Gong Y.-W., Zhang H.-X., Cheng X.-N. (2012), Treatment of dairy wastewater by
two-stage membrane operation with ultrafiltration and nanofiltration, Water Science
& Technology, 65.5, pp. 915-919.
14. Luo J., Ding L., Qi B., Jaffrin M. Y., Wan Y. (2011), A two-stage ultrafiltration and
nanofiltration process for recycling dairy wastewater, Bioresource Technology, 102,
pp. 7437–7442.
15. Balannec B., Vourch M., Rabiller-Baudry M., Chaufer B. (2005), Comparative
study of different nanofiltration and reverse osmosis membranes for dairy effluent
treatment by dead-end filtration, Separation and Purification Technology, 42, pp.
195–200.
16. Caric M., Akkerman J. C., Milanovic S., Kentish S. E., Tamime A. Y. (2009),
Technology of Evaporators, Membrane Processing and Dryers, in:Dairy Powders
and Concentrated Products, Blackwell Publishing Ltd, pp. 99-148.
17. Zmievskii, Yu. G., Kirichuk I. I., Mironchuk V. G. (2014), Membrane treatment of
wastewater obtained after the whey processing, Journal of Water Chemistry and
Technology, 36(6), pp. 309-316.
18. Myronchuk V. G., Grushevskaya I. O., Kucheruk D. D., Zmievskii Yu. G. (2013),
Experimental study of the effect of high pressure on thee of Whey nanofiltration
processu an OPMNP membrane, Petroleum Chemistry, 53(7), pp. 439–443.
19. Kyrychuk I. I., Myronchuk V. H., Zmiievskyi Yu. H. (2013), Vyznachennia
mekhanizmu zabrudnennia nanofiltratsiinykh membran pry rozdilenni molochnoi
syrovatky, Visnyk Kharkivskoho natsionalnoho tekhnichnoho universytetu silskoho
hospodarstva imeni Petra Vasylenka «Suchasni napriamky tekhnolohii ta
mekhanizatsii protsesiv pererobnykh ta kharchovykh vyrobnytstv», 140, pp. 147-152.
───Ukrainian Food Journal. 2015. Volume 4. Issue 4 ── 647
Copyright of Ukrainian Food Journal is the property of National University of Food
Technologies and its content may not be copied or emailed to multiple sites or posted to a
listserv without the copyright holder's express written permission. However, users may
print, download, or email articles for individual use.
