Skip to main content
NCBI home page
Journal of Food Science and Technology logoLink to Journal of Food Science and Technology
. 2021 Jan 21;58(10):3676–3688. doi: 10.1007/s13197-020-04958-6

Applications of reverse osmosis in dairy processing: an Indian perspective

Gaurav Kr Deshwal 1, Akshit 1, Saurabh Kadyan 2, Heena Sharma 1, Ashish Kumar Singh 1, Narender Raju Panjagari 1, Ganga Sahay Meena 1,✉
PMCID: PMC8357910  PMID: 34471292

Abstract

The Indian dairy industry is highly diversified in terms of milk production, collection, processing and waste disposal. Membrane processing allows dairy sector to manufacture high quality nutritive dairy products at lower costs with minimum water use and product losses. Compared to prevailing traditional methods of milk concentration, reverse osmosis (RO) is still evolving, finding newer applications in dairy processing because of its potential benefits. A brief overview of RO, membranes, process variables, fouling, merits and demerits along with potential suppliers and membrane utilizing dairy plants in India are systematically presented in this review. Different applications of RO in dairy industry including concentration of liquid dairy streams, further utilization of RO retentate in formulation of ice-cream, dahi, traditional Indian dairy products, cheese and dried powders is also included. RO can play a prominent role in Indian dairy sector for simplifying the process automation, product diversification and efficient waste utilization.

Keywords: Reverse osmosis, Concentration, Dairy products, Fouling, Effluent treatment

Introduction

Reverse osmosis (RO), nano filtration (NF), ultrafiltration (UF) and micro filtration (MF) are the four main pressure-driven membrane processes. Among these processes, RO membrane has the smallest pore size (10–4–10–3 µm), hence it demands highest operating pressure. RO is a well-established dewatering or concentration process that selectively separates the solutes of ≥150 Daltons molecular weight from a solvent using a suitable semipermeable, non-porous membrane when applied external pressure exceeds than the osmotic pressure of the solvent. Small solutes exhibit higher osmotic pressure; hence, 35–100 bar pressure is applied in RO process to overcome their high osmotic pressure (Cheryan 1998). The high throughput and efficiency of RO process is governed by the desirable characteristics of membrane module such as superior selectivity for solvent, high physicochemical stability, durability and less fouling (Henning et al. 2006).

RO finds more promising applications in concentration of dairy streams such as milk, whey and buttermilk without altering retentate composition (de Boer and Nooy 1980). Further, such concentrate could be frozen, stored, thawed and converted into desired dairy products. With continuous improvement in membrane materials, modules and their selectivity, recovery of growth factors from whey is now a reality. Nevertheless, the key challenge faced by whole membrane industry is to reduce membrane fouling without compromising key operating parameters such as permeate flux, rejection capacity and trans-membrane pressure (TMP). So far, a systematic review on various applications and significance of RO process in terms of process efficiency, quality improvement and product diversification in dairy industry is unavailable. Hence, this review aims to explain the basic concepts of reverse osmosis, its effect on milk components and applications in dairy industry. Further, future challenges and opportunities in harnessing the complete potential of reverse osmosis in dairy sector has also been discussed.

History and basics of reverse osmosis

RO membrane was accidently manufactured by Srinivasa Sourirajan and Sidney Loeb at the University of California, Los Angeles during 1958–1960. Commercially available cellulose acetate membranes were subjected to heating treatment under water to widen their pores and increase the flux. However, exactly opposite happened; surprisingly membrane pores got contracted after heating (Cheryan 1998). Subsequently, commercially available UF membranes were subjected to similar heat treatment. This shrunk their pores that noticeably increased the salts rejection and flux values to the level never achieved earlier. The ultrastructure of UF membranes was changed by a phenomenon of ‘anisotropy or asymmetry’ due to applied heating or annealing process. The asymmetric membrane by Loeb and Sourirajan was able to withstand 75 bar pressure, rejected 99% NaCl with a 14.5 L/m2h flux rate and set the corner stone for current RO process. Since, this discovery of RO membrane in early 1960s by Loeb and Sourirajan, RO as a membrane filtration process is finding applications in treating brackish and seawater, desalination and biochemical sectors. Its role in food and beverage processing has also increased with time, preferably in processing of heat labile molecules and recovery of value-added food components (Wenten and Khoiruddin 2016).

Van’t Hoff equation is used to determine the osmotic pressure of solutions and expressed as π = n × R × T = C × R × T/M; where, π, n, R, T, C and M are osmotic pressure, molar concentration of solute (moles/L), universal gas constant, absolute temperature (°K), concentration of solute (g/L of solution) and molecular weight of solute, respectively (Nielsen 2000). This equation clearly indicates that at a given concentration, the osmotic pressure of a solution is directly proportional to temperature and indirectly proportional to its molecular weight. Scientifically, when semi-permeable membrane separates two liquids of different concentrations in a vessel; water will pass through the membrane (because salt is rejected) to achieve similar chemical potential on both sides of the membrane by equalizing the concentration (Cheryan 1998). The flow of water molecules through membrane continues until the pressure exerted by rise in its level, becomes equal to the osmotic pressure. However, the direction of this water movement through membrane can be reversed by applying external pressure higher than the osmotic pressure. Thus, a process in which direction of osmotic flow across the semi-permeable, non-porous membrane, separating two liquids of different osmotic pressures, is reversed by applying an external pressure higher than their osmotic pressure difference is known as reverse osmosis (Nielsen 2000). Basically, RO is a pressure driven filtration process that involves mass transfer through a dense, semi- permeable, non-porous membrane via solution-diffusion phenomena. It is also called hyperfiltration or dewatering technique. A typical pilot scale RO unit and its components are illustrated in Fig. 1.

Fig. 1.

Fig. 1

Open in a new tab

Images of the reverse osmosis unit a lab scale batch type RO unit and, b illustrative diagram of RO membrane plant

First and second-generation membranes are used in a RO plant. RO membranes are manufactured using cellulose acetate (CA), cellulose triacetate (CTA), cellulose acetate/-triacetate blend (CA/ CTA), polyamide (PA), polybenzimidazole (PBI), polybenzimidazolone (PBIL), polyimide (PI), polyacrylic acid + zirconium oxide and, polyethyleneimin + toluene diisocyanate thin film composite (TFC) membrane (Nielsen 2000). Several suppliers and fabricators dealing with RO units for dairy sector are outlined in Table 1. Dissolved ions viz. chlorides, calcium and molecules greater than 150 Dalton molecular weight (sugars, amino-acids etc.) are retained by the membrane. RO membranes can retain sodium chloride concentrations up to 90% under appropriate running conditions, however, maintenance of high flux rates with such systems is often a challenge. Additionally, these membranes need to be operated under 45 °C due to low glass transition temperature as operation at high temperature results in membrane compaction and low permeation (Donnelly et al. 1974). These membrane materials can be configured into all four well-established membrane configurations namely, tubular, plate and frame, hollow fiber and spiral wound. Table 2 highlights the key merits and demerits of RO process.

Table 1.

Suppliers and fabricators of reverse osmosis units for dairy processing plants

Firm name Head office Websites Indian contact point/address
GEA Dusseldorf, Germany www.gea.com Block No. 8, P.O.-Dumad, Savli Road, Vadodara, India-391740
Siemens Ltd Munich, Germany www.siemens.com Birla Aurora, Level 21, Plot No. 1080, Annie Berant Road, Worli, Mumbai, India-40030
SPX Flow Tech. Pvt. Ltd Charlotte, North California, US www.spxflow.com Survey No. 275, Odhav Road, Odhav, Ahmedabad, India-382415
Tetrapak Limited Pully, Switzerland www.tetrapack.com DLF Cyber Terrace, Building no. 5, 16th Floor, DLF Cyber City, DLF phase III, Gurugram, India-122002
Merck Pvt. Ltd Darmstadt, Germany www.merckmillipore.com 8th Floor, Godrej One, Pirojshanagar, Eastern Express Highway, Vikhroli, Mumbai, India-400079
IDMC limited – www.idmc.com Plot No. 124–128, GIDC Estate, Vitthal Udyognagar, Gujarat, India-388121
MSS India Pvt. Ltd – www.membranesystem.co.in S2,127, Vedant commercial complex, Kores road, Vartaknagar, Thane, India-400606
Permionics Membrane Pvt. Ltd Vadodara, India www.permionics.com 3/29–4/19, BIDC, Gorwa, Vadodara, India-390016
Rushi Ahmedabad, India www.rushiprojects.com 140, Tribhuvan Estate, Road no. 11, Kathwada GIDC, Odhav industrial estate, Ahmedabad, Gujarat, India-382430
Thermax Pune, India www.thermaxglobal.com D-13, MIDC Industrial Area, R D Aga Road, Chinchwad, Pune, India-411019
DuPont Wilmington, Delaware, United States www.dupont.com DLF Cyber Greens, 7th Floor, Tower C, Sector-25A, DLF Phase 2, Gurugram, Haryana, India-122002
Sumitomo Tokyo, Japan www.sumitomocorp.com 302 and 303, 3rd floor, World Mark 2, Asset NO. 8, Aerocity Hospitality District, Aerocity, New Delhi, India-110037
Lenntech Delfgauw, The Netherlands www.lenntech.com Distributieweg 3, 2645 EG Delfgauw, The Netherlands
Osmonics Wisconsin, USA www.osmonicsrosystem.com 851 W Main Street, Twin Lakes, Wisconsin, USA-53181
DSS Silkeborg, Denmark www.sepco.dk –
Toray Tokyo, Japan www.toraywater.com Nihonbashi Mitsui Tower 2–1-1 Nihombashi Muromachi, Chuo-ku, Tokyo, Japan, 103–8666
Alfa Laval Lund, Sweden www.alfalaval.in Office No. 301, Global Port Building, Survey No. 45/1–10, Mumbai Bangalore Highway Baner, Pune, Maharashtra, India-411045
Open in a new tab

Table 2.

Advantages and disadvantages of reverse osmosis (RO) process

Advantages Disadvantages
A simple dewatering process without any phase change Upper limit for concentration of skim milk and whole milk by RO is 28% and 38%, respectively
Less energy intensive and cheaper than single, multiple effect evaporation, boiling and freeze concentration Shorter life span of membranes and suffers from membrane fouling
Preserves and concentrates milk components in natural state as operated at ≤ ambient temperatures Narrow range of operating parameters (flow rate, temperature, pH and pressure)
Huge initial capital investments are not required as compared to multiple effect evaporation systems Efficiency decreases as feed concentration increases
No requirement of complicated set up, only electrical energy is desired to operate high pressure pumps Membrane replacement cost are very high and contributes 40% of the total plant cost
Open in a new tab

Critical parameters during RO processing

In dairy industry, RO is mainly used to concentrate liquid dairy streams (via dewatering) at desired temperature (10–40 °C) without any phase change. The performance and efficiency of RO is governed by different factors such as feed type and its characteristics (total solids, pH, viscosity, salts and protein content), processing temperature, pressure, degree of concentration, membrane type, membrane configuration, concentration polarization and fouling. Collective effect of these parameters is studied in terms of membrane flux (D'souza and Mawson 2005). During RO concentration, specific gravity of the retentate or permeate volume indicates the degree of concentration at any point of time (Syrios et al. 2011).

Along with desired rejection of target components, higher flux rate is always a prerequisite for economic plant operation and better process efficiency. The maximum flow rate for a particular module is decided by its mechanical design and membrane material. The osmotic pressure of liquid dairy streams is governed by their chemical composition, wherein, lactose involvement is maximum (D'souza and Mawson 2005). The energy requirements for RO is mainly contributed by high pressure pumps and repetitive membrane replacement cost. During RO operation, the flux i.e. permeate flow per unit membrane per unit time decreases gradually due to phenomenon of concentration polarization which is accumulation and adherence of retentate components on membrane and forms a boundary layer. This in turn increases the osmotic pressure difference across membranes. Polarization effects should be mitigated to ensure longer operating runs. This can be improved using membranes in cross-flow filtration assemblies and maintaining turbulent flow velocities of 3–6 ft/s. The lifespan of RO membrane generally ranges between 6–15 months and restricted by time–temperature conditions of cleaning-in-place system and compaction effect (Beaton 1979). Donnelly et al. (1974) found direct correlation in membrane flux during whey concentration whereas increased flux rate was observed with rise in processing temperature from 10 to 35 °C. However, high temperature decreased the life of membrane and favored microbial proliferation. Boer and Nooy (1980) created higher turbulence using glass beads in the form of fluidized bed (<3 mm) over membrane surface to decrease fouling and increase flux. This strategy was restricted by lower separation efficiency due to enlargement of pores by glass beads. Thus, aversion of concentration polarization and membrane fouling are prominent for economic RO plant operation and better process efficiency.

Applications in dairy industry

Compositional changes in dairy fluids during reverse osmosis

During RO process, most of the components of dairy fluids except water, some salts, a portion of non-protein nitrogen (NPN), low molecular weight sugars and lactic acid, are retained and concentrated (Donnelly et al. 1974). The compositional changes occurring after RO concentration of dairy fluids including non-bovine milk are summarized in Table 3. Pal (2007) reported that retention coefficient (fraction of dissolved substances retained by membrane) of RO process is 0.999, 0.90–0.98 and 0–90 for larger molecules (such as sugar, protein, fat etc.), sodium (like smaller ions) and other organic compounds with lower molecular weight, respectively. Syrios et al. (2011) reported no losses of ionic calcium and total solids (TS) in RO permeate. Further, Barbano and Bynum (1984) observed that urea content of RO permeates (12.7 mg/100 mL) was similar to milk (13.1 mg/100 mL). Shenana et al. (2007) reported increase in ash content of RO retentate with proportional increase in its protein content. The pH of skim milk RO retentate (6.30 ± 0.04) got reduced with increasing TS due to proportional increase in calcium, while permeate had pH (6.62 ± 0.19) similar to skim milk (6.64 ± 0.03).

Table 3.

Changes in composition of dairy fluids after concentration by Reverse Osmosis

Composition ↓ Dairy fluid →  Skim milk Whole milk Sheep milk Goat milk Cheese whey Chhana whey Sweet cream buttermilk
Fat (%) – 3.68 6.4 3.27 – 0.5–0.7 0.75
Protein (%) 3.12 2.98 5.88 3.04 0.7 0.12–0.13 2.66
Lactose (%) 4.90 3.9 4.72 3.6 4.7 4.5–4.78 –
Ash (%) 0.76 0.67 0.89 0.8 0.7 0.52–0.57 –
Total solids (%) 8.70 11.8 17.8 11.35 6.5 5.9–6.17 8.26
Composition of RO retentate
Fat (%) – 7.54 12.00 3.35 – – 2.02
Protein (%) 8.76 6.60 10.37 3.89 2.8 0.12–0.13 7.05
Lactose (%) 14.38 8.20 8.36 4.86 18.6 4.32–4.58 10.87
Ash (%) 1.50 0.92 1.40 0.97 2.5 0.37–0.41
Total solids (%) 24.98 25.62 32.07 13.94 25 5.5–5.7 21.94
References Donnelly et al. (1974) Agbevavi et al. (1983) Voutsinas et al. (1996) Abrahamsen and Holmen (1981) Donnelly et al. (1974) Jindal and Grandison (1992) Govindasamy-Lucey et al. (2007)
Open in a new tab

Liquid milk processing

RO acts as prospective tool for concentrating unpasteurized milk at farm level especially at bulk milk collection units or chilling centers. This on-site milk concentration is profitable since it diminishes the initial bulk and curtails the cost involved in further transportation, cooling, heating and overall handling of milk (Sorensen et al. 2016; Tamime 2013). RO is capable of removing 70% water present in milk at farm level. Whole milk concentrated by RO at farm level demonstrated benefits in terms of less carbon dioxide emission and efficient utilization of clear permeate (water) in other miscellaneous application in plant or for direct cattle feeding (Sorensen et al. 2016). Skim milk was economically concentrated to 22–25% TS using RO (Glover 1971; Fenton-May et al. 1972). Pal (2007) reported that buffalo milk can be successfully concentrated to twofolds without any significant changes in its quality parameters at 4 ± 1 °C up to 48 h. It was also evaluated that installation of RO plant at chilling center had a payback period of only 10 months. The concentration of buffalo milk to 1.5- and 2-folds had processing cost of 3.64 and 4.64 paise/kg of milk, respectively. Subsequently, flexible and energy-efficient RO process can also supplement the evaporation process. For instance, RO was cheaper in concentrating skim milk from 8.8 to 25% TS as compared to multistage evaporation, whilst, combination of both in increasing TS to 45% showed less energy requirements than evaporation alone (Stabile 1983). The upper limit of whole and skim milk concentration by RO is 38% and 28% TS, respectively but flux becomes negligible at these levels. Therefore, the reported economical levels of RO concentration for skim milk and whole milk were 22% and 30% TS, respectively (Pal 2007).

Theoretically, a twofold increase in coliforms and total bacterial counts is expected after RO concentration of milk, however, it mainly depends on initial microbial counts of raw milk and sanitation of RO plant (Voutsinas et al. 1996). Hence, RO process is often accompanied by pre-treatment such as thermization, pasteurization or microfiltration to preserve bacteriological quality of milk. Twofold whole milk RO retentate, stored for 24 h at 5 °C, diluted to natural milk composition followed by pasteurization, homogenization and flavor addition was converted to different dairy products. The taste panel (106 out of 111) could not find any difference in organoleptic quality of diluted milk and products manufactured from it. The reconstituted whole milk preserves well for 9–12 days under refrigerated storage (Pal 2007). However, its shelf life is negatively influenced by high pressure (>3 MPa) and turbulent flow conditions which damage milk fat globules and liberates free fatty acids (FFA) making it prone to flavor defects during storage. Lipase mediated microbial deterioration can be subsided by thermal treatment in the range of 60–70 °C. A decrease in FFA content to 0.22 mmol/100 g fat and enhancement of proteolysis by 13.4% upon increasing the feed pressure from 2 to 3 MPa during RO concentration of whole milk at farm level was reported (Sorensen et al. 2016). FFA and proteolysis increased to 0.24 mmol/100 g fat and by 23.4%, respectively when processing temperature was raised from 4 to 10 °C along with expansion of membrane spacer thickness from 30 to 48 mm. Barbano et al. (1983) revealed reduction in FFA level on reducing the diameter of retentate pipe from 2.54 to 0.95 cm during whole milk concentration using RO. In another study, 24% TS containing RO retentate was distributed for household evaluation in Australia. The 67, 22 and 11% of the respondents reported as favorable, neutral and negative opinion on product attributes (Pal 2007).

The storage stability of RO concentrate is critical due to protein concentration and sedimentation of milk ingredients. RO retentates possess higher acidity and buffering capacity (Hinrichs 2000). Whole sheep’s milk RO concentrate with 32.1% TS and skim milk RO concentrate standardized to 35.3% TS using cream were stable for 6–8 months at −20 °C. Physical stability, fat oxidation and milk lipolysis of RO concentrates as represented by sedimentation value, peroxide value and acid degree value, respectively, had no significant changes during their storage. The total bacterial and coliform count were higher but decreased subsequently during frozen storage due to rupture of bacterial membrane by ice crystals (Voutsinas et al. 1996). RO process can be utilized for preparing long shelf life UHT treated milk concentrates (fivefold) without any additives. The heat stability of RO processed UHT milk concentrate was found to be negatively affected by fat and ash content, while lactose content had no affect (Hinrichs 2000).

Whey and whey products

Whey is the major by-product of dairy industry with an estimated volume of 190 × 106 t/year worldwide and creates huge pollution if not disposed efficiently. Conversion of whey obtained from cheddar cheese, co-precipitates, casein, channa and paneer containing only 6–7% TS into lactose powder, whey powder and demineralized whey powder demands huge energy. Economical operations of RO can facilitate whey concentration to recover its valuable ingredients with subsequent minimization of waste-disposal problem (Yorgun et al. 2008). Pal (2007) reported that whey concentration employing batch and multi-stage recycle operation of RO plants achieved 18% and 26–28% TS in RO retentate. During whey concentration, flux declines rapidly as a function of increase in lactose concentration and precipitation of residual protein fractions causing fouling of membrane. Prior to RO concentration, denaturation of whey protein by heating (>50 °C) avoided fouling. Rektor and Vatai (2004) suggested that for small sized plants, whey should be defatted and sterilized using microfiltration before its concentration by RO. Jindal and Grandison (1992) produced channa whey powder by spray drying defatted RO channa whey retentate with 21% TS.

According to Boer et al. (1977) utility requirements for evaporation of whey was 55% of total costs while this was only 25% for RO during twofold concentration of Gouda cheese whey. Compared to UF-RO combination, use of forward osmosis and RO was reported as the best combination for 77.4% water removal and production of whey powder with a payback period of only 0.8 years. The combined NF-RO treatment of whey not only recovers protein and lactose separately but also reduces COD by approximately 94% (Yorgun et al. 2008). However, Aydiner et al. (2014) designated UF-RO as most effective whey pre-treatment system.

Condensed and dried milks

Liquid dairy streams (such as whole milk, skim milk, buttermilk etc.) are concentrated by various techniques like boiling, single or multi-stage evaporation, freeze concentration and reverse osmosis to diminish their volume and enhance their shelf-life. Concentrated milks are either consumed directly or used as an ingredient in the formulation of different products. Concentration of liquid dairy streams is also necessary to improve drying efficiency and achieve desired wettability, sinkability, dispersibility and solubility in resultant powders (Balde and Aider 2017). However, based on different factors such as capital investment, product quality, nutrient retention, throughput and process economy, RO is considered as the most economical dewatering process (Pal 2007).

RO concentration of milk circumvents fore-warming (105–108 °C for no hold or 85–90 °C for 15–20 min) step thus averting nutrient degradation, protein denaturation and Maillard browning. This also retains desirable organoleptic attributes in concentrated milk since RO concentration occurs at ambient temperature (Gupta and Pal 1993). However, lack of preheating leads to lower heat stability in RO concentrates. Calcium plays a significant role in deciding the heat stability, gel strength and zeta potential of milk powder, but calcium content increases during RO concentration (Balde and Aider 2017). Abbot et al. (1979) reported no significant changes between the heat stability of RO and conventional concentrates. Syrios et al. (2011) concentrated the skim milk using RO and NF, retentates thus obtained were spray dried and powders were reconstituted to 25% TS solutions. Reconstitution of RO concentrate powder showed lower thermal stability at sterilization temperature (115 °C for 15 min) compared to reconstituted solution of NF concentrate powder. However, addition of 0.2% trisodium citrate and 0.1% disodium hydrogen phosphate improved the sterilization stability of both the retentates. Thus, RO milk concentrates could be successfully used for production of heat stable skim milk powder (SMP). Moreover, SMP obtained from RO concentrate is whiter as RO concentration decreases the extent of Maillard reaction. Balde and Aider (2017) reported cryo-concentration followed by RO and vacuum evaporation as the best suitable method for concentration of milk before drying. According to Sorensen et al. (2017), SMP manufactured using twofold RO retentate had 12 months shelf-life without any adverse effect on any of the powder properties. Whole milk powder prepared from RO retentate showed higher free fat content (57 mg/g of powder) and unacceptable lipolytic taints due to rupture of fat globule during passage of RO retentate through pressure relief valve and pump. Nevertheless, thermal inactivation of lipase prior to RO treatment was effective in averting high FFA liberation (Abbot et al. 1979). However, such powder could be used in products where free fat is desired.

RO concentration could be instrumental in increasing the SNF content of sweet cream buttermilk (SCBM) and whey buttermilk (Morin et al. 2006). Irrespective of the type, buttermilks have higher proportion of phospholipids than skim milk enabling them to act as ideal ingredient for functional foods (nutraceuticals if purified) or as natural emulsifier in liquid and dried form in selected dairy products such as ice-cream (Rao 2002). Pre-concentration of SCBM by RO can be effectively used for buttermilk powder production as Govindasamy-Lucey et al. (2007) explored RO process for efficient concentration of SCBM from 8.2 to 21% TS. However, Rao (2002) observed decline in flux with increasing pH of buttermilk from 6.6 to 8.0 while rate of fouling exacerbated at lower pH values.

Fermented dairy products

During manufacturing of yoghurt or similar fermented products, milk is standardized to desired TS (12–14%) by adding liquid RO/UF retentates, concentrated milks or dried milk powders to increase SNF/TS. It acts as a pre-requisite for achieving desirable sensory attributes such as sliceable texture and better starter growth. RO had been used for increasing the SNF in milk base during production of yoghurt and Koumiss (Tamime and Robinson 2007). At similar TS content, yoghurt prepared using ordinary milk and RO retentate containing 12.5–15% TS, exhibited similar culture growth, acid production, acetaldehyde concentration, viscosity and sensory scores (Davies et al. 1977). However, the pH decline in yoghurt produced from RO retentate was slower due to its higher buffering capacity (Ozer et al. 1998). According to Tamime (2013), products manufactured from RO retentate/powder offer better sensorial characteristics as compared to thermally concentrated milk or dried powder owing to drastic reduction in Maillard browning and insoluble particles.

Dahi, an indigenous fermented milk product of India is synonymous to yoghurt. Dahi manufactured from cow milk forms weak gel network due to low TS (particularly lower protein and calcium content) and encounters defects like higher syneresis (wheying off), loose body, soft texture and reduced consumer acceptability as compared to buffalo milk Dahi (Ranganadham et al. 2016). Dahi prepared using 1.5–2 folds RO cow milk retentate possessed attributes similar to buffalo milk Dahi. However, slow acidification rate due to higher buffering capacity and lower levels of acetyl methyl and other volatile acids production were observed (Kumar and Pal 1994b). Additionally, Dahi produced using twofold concentrated milk had an extended shelf-life of 4–6 days at refrigerated temperature due to reduced post-acidification (Reddy et al. 1986). Chakka or de-wheyed Dahi, produced using 2.5-fold RO retentate increased the yield to 35.5% as compared to 28.3% in conventional process (Ranganadham et al. 2016). In nutshell, RO can be successfully employed for the overall quality improvement of set-fermented dairy products such as Dahi and yoghurt.

Cheese and cheese-based products

The yield of cheese drives the economics and efficiency of a cheese plant and increases with higher fat and casein retention in cheese curd. A few researchers observed enhanced yield of cheese manufactured from RO retentates due to better retention of whey proteins within the casein network along with simultaneous reduction in handling volume of whey (Barbano and Bynum 1984; Hydamaka et al. 2000). RO is beneficial in cheese manufacturing due to its ability to concentrate protein, fat and potential for developing a continuous (automatic) cheese manufacturing unit (Modler 1988). Yield of direct acidified cheese manufactured from UF retentate (VCR-4:1) was markedly higher than obtained from RO concentrate (2.5:1), however the percent TS recovery was 5.11, 58 and 70.5 in control, UF and RO cheese, respectively. Also, milk concentration significantly changed the textural attributes of cheese except cohesiveness and springiness (Hydamaka et al. 2000). Agbevavi et al. (1983) reported that as compared to non-concentrated milk, cheddar cheese prepared using whole milk RO retentate (25% TS) required 50% and 60% lesser starter cultures and rennet along with 2–3% higher cheese yield. RO retentate based cheddar cheese had higher lactose and ash content over control cheese. Barbano and Bynum (1984) reported that cheddar cheese produced from RO retentate concentrated to 5–20% VCR had higher fat retention and higher starter activity, however, showed similar proteolysis pattern during initial 3-months ripening. Homogenization of milk after RO concentration led to better interaction of casein micelles with integrated milk fat globule membrane and enhanced fat recovery. It was suggested that for the production of RO based cheddar cheese, 0.68–0.70 casein to fat ratio and 15% volume reduction should be maintained. Tamime (2013) reported 5% increase in yield for cottage cheese after 8% volume reduction of skimmed milk by RO. However, Mistry et al. (2004) reported that RO finds limited application in cheese industry because of high fat losses in whey, induced lipolysis, higher buffering capacity due to concentration of minerals and higher lactose retention which collectively led to altered acidification rate during cheese ripening and also causes variation in lipolysis and proteolysis kinetics during cheese maturation. Paneer, an Indian counterpart of cottage cheese, was prepared using 1.5–2.0-fold concentrated RO milk, exhibited 2–3% more yield without adversely affecting its sensorial attributes (Kinjal et al. 2015). The texture of Paneer produced from RO retentate was brittle and granular due to whey entrapment and lactose retention (Gupta and Pal 1995). Moreover, Hydamaka et al. (2000) successfully produced direct acidified soft varieties of cheese such as Mozzarella cheese, cottage cheese etc. using RO concentrate.

Ice-cream

According to Garcia et al. (1989), RO was more energy efficient than UF and vacuum concentration during preparation or standardization of ice-cream mixes. Savello (1998) patented the process for production of RO concentrated and UHT sterilized (138 °C/2 s) ice-cream mixes for storage at ambient temperature. In ice-cream mix, 50–100% replacement of milk solids not fat (MSNF) content by RO retentate didn’t change TS and fat, but increased protein, ash, viscosity, specific gravity, freezing point and resistance to melting (higher viscosity resulted from higher protein hydration); correspondingly, stabilizer requirement markedly decreased as protein in RO retentate had higher water binding capacity (Shenana et al. 2007). However, complete SNF replacement with RO retentate led to sandiness and coarse texture defects in ice-cream due to high lactose content (Lee and White 1991). To overcome these defects, a lactose to protein ratio (1.25–1.45) had been recommended in ice-cream mixes (Hofi 1989).

Traditional Indian dairy industry

Several researchers have utilized RO as a dewatering technique during manufacturing of traditional Indian dairy products (TIDPs) including khoa (Pal and Cheryan 1987), khoa powder (Rizvi et al. 1987), basundi, chakka, shrikhand, rabri and kheer (Patel et al. 2005). Compared to traditional energy intensive, cumbersome boiling of milk in open pan, dewatering of milk employing RO suits better for the development of economical and mechanized manufacturing processes for heat-desiccated indigenous products such as Khoa. Buffalo and cow milk concentration (1.5 and 2–2.5 folds) using RO for Khoa preparation had been successfully demonstrated. However, Khoa had slightly higher moisture but its organoleptic quality was acceptable (Kumar and Pal 1994a; Pal and Cheryan 1987). RO has also been exploited for continuous production of satisfactory quality khoa by concentrating milk to 30% TS and its subsequent feeding to scraped surface heat exchangers (SSHE) for further desired concentration which resulted in a net saving of 335–430 kcal energy per kg milk. As per Aggarwal et al. (2018), concentration of one kg milk to 65% TS using initial RO concentration needs only 20 kcal energy however, 136 kcal is required for its concentration in open pan boiling to similar TS level. Rizvi et al. (1987) produced Khoa powder from RO retentate (20% TS) that was heated (144–150 °C in SSHE) and cooled to 83 °C prior to roller drying. The obtained Khoa powder had comparable reconstitutional properties, better shelf-life and almost half energy consumption but had detectable burnt flavour.

For the production of Rabri (synonymous to condensed milk), milk was concentrated (~ 24% TS) employing RO followed by sugar addition (5–6% of milk), heating (95 °C) and desiccation in SSHE up to 50% TS with addition of very thin paneer slices towards the end which yielded Rabri with desirable attributes (Ranganadham et al. 2016). A brief overview of applications of reverse osmosis in dairy industry is presented in Fig. 2.

Fig. 2.

Fig. 2

Open in a new tab

An overview of reverse osmosis applications in dairy industry

Miscellaneous applications of RO in dairy

Suarez et al. (2014) studied the feasibility of recovering condensate generated during UHT and flash cooler for its reuse as boiler water. Conductivity, chemical oxygen demand (COD) and pH were used to determine the permeate quality. During RO concentration of condensate at VCR of 6, exactly 98.2% and 97.8% decline in conductivity and COD were achieved. However, pH of permeate water produced was lower than required for boiler water. Non-uniformity of feed parameters such as conductivity and COD while processing these condensates was encountered as major difficulties. RO-ACC (activated carbon column) has been reported to minimize the load of organic matter and meet the requirements of boiler water. Similarly, Vourch et al. (2008) utilized RO for treating dairy wastewater. The treated water could be used for heating, cooling and cleaning purposes. Even, drinking water could be produced if RO is combined with polishing (RO + NF) process. Storage stability of dairy wastewater at 25 °C and 4 °C indicated that dairy wastewater should be stored at 4 °C prior to its treatment in order to check further breakdown of lactose to ethanol and lactate due to natural acidification. Owing to higher rejection rates, these compounds decreases the performance of RO membrane. It was reported that RO can recover 90–95% milk solids at an average flux of 11 L/m2h of permeate. Kyrychuk et al. (2014) also conducted similar studies on dairy effluents using NF and RO membranes. RO membrane retained almost all components and produced permeate with low minerals that can be used for washing floors and vehicles. Better quality permeates were observed by RO due to low content of minerals. Retentates of NF and RO resembled skim milk and whole milk in composition, hence, such retentates can be used in non-food applications (animal feed supplement) after making them safe for animal consumption.

Fouling during RO processing of dairy fluids

Fouling is mainly caused by accumulation or aggregation of colloidal materials (fat, bacterial cell, protein or inorganic salts) within pores or over the membrane surface facing feed at high pressure that ultimately decreases separation efficiency and throughput by lowering permeate flux (Koh et al. 2013). Heating of cheese whey to 90–100 °C for precipitating and separating curd fines followed by injection of supernatant whey as feed to RO plant exhibited lesser fouling while retaining similar nutritional value (Yorgun et al. 2008). Maintenance of high cross-sectional flow velocities, proper process parameters and efficient cleaning were found instrumental in reducing fouling and improving permeate flux (Lim et al. 1971; Hiddink et al. 1980).

Lim et al. (1971) studied the role of various protein fractions in cottage cheese whey during RO concentration and fouling. It was observed that initially flux was restricted by internal membrane resistance and concentration polarization, while at later stages, flux declined due to membrane fouling. Casein, β-lactoglobulin, α-lactalbumin and non-protein fractions were reported as major protein foulants with lower diffusion coefficient of casein. Hiddink et al. (1980) studied various liquids including Gouda cheese whey (pH-6.6, 6.0, 4.6), skim milk, desalted whey (pH-6.6), lactose solution, UF permeate and decalcified whey to evaluate the effect of their composition and processing variables on permeate flux. Osmotic pressure and fouling components of feed were reported as the two major flux limiting parameters. Skim milk showed fouling due to casein, while Gouda cheese whey at pH 6.6 showed fouling due to Ca-phosphate complex. At 10 °C, the fouling in desalted whey was due to lowered stability of whey proteins by 95% desalting and their tendency to aggregate over membrane surface. At 30 °C, Gouda cheese whey (pH 6.0) showed lower fouling due to solubilization of Ca-phosphate complex, while at isoelectric point of protein (pH 4.6), fouling was attributed to aggregation and accumulation of protein over membrane surface (Hiddink et al. 1980).

Cleaning process of RO membrane plant

Cleaning process involves removal of undesired deposits/soils from the membrane and its auxiliary components (Koh et al. 2013). During operation, the foulants get deposited over/into membrane surface and form a gel-like film. Membrane systems are cleaned regularly using prescribed cleaning-in-place (CIP) procedures and chemicals at high velocity ensuring sufficient turbulence to dislodge the foulants. Several arrays of cleaning solutions are used depending on membrane material, module configuration and type of deposits. Non-ionic detergents or surfactants are used to remove the fat deposits, while acid and alkaline detergents are employed to clean inorganic salts and proteins, respectively (D'souza and Mawson 2005). Cellulose acetate (CA) RO membranes are generally operated below 40 °C in the pH range of 3–8, hence, cleaned using the combination of enzymes and detergents. CA membranes are sensitive to high chlorine concentration (> 50 ppm) unlike UF membranes that can easily tolerate 100–200 ppm of available chlorine (Nielsen 2000). RO systems usually require 3–4 h of CIP because of cleaning through enzyme soaking. The extent of membrane cleaning is generally ascertained by measuring the water flux rates during rinsing. For best results, efficient separation and longer membrane life, all the membrane irrespective of the process, make or material, must be properly cleaned and sanitized prior and after each run as per the instructions of the membrane manufacturers. The generalized flow diagram related to operation and cleaning of RO plant for dairy fluids with critical parameters is depicted in Fig. 3.

Fig. 3.

Fig. 3

Open in a new tab

Generalized flow diagram for starting, operating, cleaning and shutting down a reverse osmosis plant for dairy fluids with critical parameters

Conclusion and future considerations

RO is the most economical milk dewatering technique at farm level owing to reduction in volume, transportation cost and carbon footprints. Although, many Indian dairy plants are now utilizing membrane technique(s) for production of different products and waste treatment (Table 4), still, much work needs to be done for its wider application and commercialization in diverse field of Indian dairy sector. The problem of membrane fouling needs major attention as it reduces productivity and lifespan of membranes. Research in development of newer membrane modules; improvement in process designs; better understanding of fouling mechanism and optimization of cleaning protocols can overcome membrane fouling. Novel membrane modules with superior performance in terms of high flux rates, reduced fouling, mechanically resistant and better tolerance to cleaning solutions (especially chlorine) can greatly reduce capital and operational costs of industrial RO plants. To minimize the payback period for RO plant, this technology can be exploited to concentrate and recover high value-added bioactive components such as glycomacropeptide (GMP), lactoferrin etc. from whey. In nutshell, with the advancement in membrane technological processes, reverse osmosis can certainly have a big role in Indian dairy sector by simplifying the process automation and production of diverse food formulations at curtailed cost.

Table 4.

Utilization of membrane processing for manufacturing dairy products in India

Company Location Membrane process Utilization/products
AMUL Amulfed, Gandhinagar Reverse osmosis Water purification
Parag Ghaziabad Ultrafiltration microfiltration diafiltration Whey protein isolate
AMUL Kaira, Anand Ultrafiltration Greek yoghurt (10% protein)
Milky mist Erode, Chennai Nanofiltration Whey powder
Govardhan Pune Microfiltration Cold sanitation of cheese brine
Milky mist Erode, Chennai
AMUL Khatraj, Gujarat Ultrafiltration Whey powder
Hatsun Chennai
Nestle Moga, Punjab Reverse osmosis For concentration of skim milk (23–24% T.S.)
Milky mist Erode, Chennai Reverse osmosis For CIP water production
Schreiber Dynamix Baramati, Maharashtra Ultrafiltration Whey powder
Parag Palamaner, Andhra Pradesh Reverse osmosis For CIP water production
Open in a new tab

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Abbot J, Glover FA, Muir DD, Skudder PJ. Application of reverse osmosis to the manufacture of dried whole milk and skim-milk. J Dairy Res. 1979;46(4):663–672. doi: 10.1017/S0022029900020720. [DOI] [Google Scholar]
  2. Agbevavi T, Rouleau D, Mayer R. Production and quality of Cheddar cheese manufactured from whole milk concentrated by reverse osmosis. J Food Sci. 1983;48(2):642–643. doi: 10.1111/j.1365-2621.1983.tb10809.x. [DOI] [Google Scholar]
  3. Aggarwal D, Raju PN, Alam T, Sabikhi L, Arora B. Advances in processing of heat desiccated traditional dairy foods of Indian sub-continent and their marketing potential. J Food Nutr Res. 2018;3:172. [Google Scholar]
  4. Aydiner C, Sen U, Topcu S, Sesli D, Ekinci D, Altınay AD, Keskinler B. Techno-economic investigation of water recovery and whey powder production from whey using UF/RO and FO/RO integrated membrane systems. Desalin Water Treat. 2014;52(1–3):123–133. doi: 10.1080/19443994.2013.786655. [DOI] [Google Scholar]
  5. Balde A, Aider M. Effect of cryoconcentration, reverse osmosis and vacuum evaporation as concentration step of skim milk prior to drying on the powder properties. Powder Technol. 2017;319:463–471. doi: 10.1016/j.powtec.2017.07.016. [DOI] [Google Scholar]
  6. Barbano DM, Bynum DG. Whole milk reverse osmosis retentates for Cheddar cheese manufacture: cheese composition and yield. J Dairy Sci. 1984;67(12):2839–2849. doi: 10.3168/jds.S0022-0302(84)81644-6. [DOI] [Google Scholar]
  7. Barbano DM, Bynum DG, Senyk GF. Influence of reverse osmosis on milk lipolysis. Int J Dairy Sci. 1983;66(12):2447–2451. doi: 10.3168/jds.S0022-0302(83)82109-2. [DOI] [Google Scholar]
  8. Beaton NC. Ultrafiltration and reverse osmosis in the dairy industry: an introduction to sanitary considerations. J Food Prot. 1979;42(7):584–590. doi: 10.4315/0362-028X-42.7.584. [DOI] [PubMed] [Google Scholar]
  9. Boer RD, Wit JD, Hiddink J. Processing of whey by means of membranes and some applications of whey protein concentrate. Int J Dairy Technol. 1977;30(2):112–120. doi: 10.1111/j.1471-0307.1977.tb01184.x. [DOI] [Google Scholar]
  10. Cheryan M. Ultrafiltration and microfiltration handbook. Boca Raton: CRC Press; 1998. [Google Scholar]
  11. Davies FL, Shankar PA, Underwood HM. Recent developments in yoghurt starters: the use of milk concentrated by reverse osmosis for the manufacture of yoghurt. Int J Dairy Technol. 1977;30(1):23–28. doi: 10.1111/j.1471-0307.1977.tb01161.x. [DOI] [Google Scholar]
  12. De Boer R, Nooy PFC. Concentration of Raw whole milk by reverse osmosis and its influence on fat globules. Desalination. 1980;35:201–211. doi: 10.1016/S0011-9164(00)88611-7. [DOI] [Google Scholar]
  13. Donnelly JK, O'Sullivan AC, Delaney RAM. Reverse osmosis—concentration applications. Int J Dairy Technol. 1974;27(3):128–140. doi: 10.1111/j.1471-0307.1974.tb01687.x. [DOI] [Google Scholar]
  14. D'souza NM, Mawson AJ. Membrane cleaning in the dairy industry: a review. Crit Rev Food Sci Nutr. 2005;45(2):125–134. doi: 10.1080/10408690490911783. [DOI] [PubMed] [Google Scholar]
  15. Fenton-May RI, Hill CG, Amundson CH, Lopez MH, Auclair PD. Concentration and fractionation of skim milk by reverse osmosis and ultrafiltration. J Dairy Sci. 1972;55(11):1561–1566. doi: 10.3168/jds.S0022-0302(72)85720-5. [DOI] [Google Scholar]
  16. Garcia A, Medina B, Verhoek N, Moore P. Ice cream components prepared with ultrafiltration and reverse osmosis membranes. Biotechnol Prog. 1989;5(1):46–50. doi: 10.1002/btpr.5420050111. [DOI] [Google Scholar]
  17. Glover FA. Concentration of milk by ultrafiltration and reverse osmosis. J Dairy Res. 1971;38(3):373–379. doi: 10.1017/S0022029900019506. [DOI] [Google Scholar]
  18. Govindasamy-Lucey S, Lin T, Jaeggi JJ, Martinelli CJ, Johnson ME, Lucey JA. Effect of type of concentrated sweet cream buttermilk on the manufacture, yield, and functionality of pizza cheese. J Dairy Sci. 2007;90(6):2675–2688. doi: 10.3168/jds.2006-681. [DOI] [PubMed] [Google Scholar]
  19. Gupta S, Pal D. Studies on concentration of buffalo whole milk for bulk transportation employing reverse osmosis. Ind J Dairy Sci. 1993;46:440. [Google Scholar]
  20. Gupta S, Pal D. Suitability of reverse osmosis concentrated milk for the manufacture of Paneer. J Food Sci Technol. 1995;32:166–168. [Google Scholar]
  21. Henning DR, Baer RJ, Hassan AN, Dave R. Major advances in concentrated and dry milk products, cheese, and milk fat-based spreads. J Dairy Sci. 2006;89(4):1179–1188. doi: 10.3168/jds.S0022-0302(06)72187-7. [DOI] [PubMed] [Google Scholar]
  22. Hiddink J, De Boer R, Nooy PFC. Reverse osmosis of dairy liquids. J Dairy Sci. 1980;63(2):204–214. doi: 10.3168/jds.S0022-0302(80)82915-8. [DOI] [Google Scholar]
  23. Hinrichs J. UHT processed milk concentrates. Le Lait. 2000;80(1):15–23. doi: 10.1051/lait:2000102. [DOI] [Google Scholar]
  24. Hofi MA. The use of ultrafiltration on ice cream making. Egypt J Dairy Sci. 1989;17:27–34. [Google Scholar]
  25. Hydamaka AW, Wilbey RA, Lewis MJ. Manufacture of direct acidified cheese from ultrafiltration and reverse osmosis retentates. Int J Dairy Technol. 2000;53(3):120–124. doi: 10.1111/j.1471-0307.2000.tb02673.x. [DOI] [Google Scholar]
  26. Jindal AR, Grandison AS. Preparation and composition of chhana whey powders using membrane processing techniques. J Sci Food Agric. 1992;58(4):511–517. doi: 10.1002/jsfa.2740580409. [DOI] [Google Scholar]
  27. Kinjal D, Patel HG, Gokhale AJ (2015) Advancements in manufacture of paneer. In: National seminar on Indian dairy industry opportunities and challenges, 11th alumni convention, Gujarat, India
  28. Koh LLA, Ashokkumar M, Kentish SE. Membrane fouling, cleaning and disinfection. In: Tamime AY, editor. Membrane processing: dairy and beverage applications. 1. New Jersey: Blackwell Publishing Ltd; 2013. pp. 73–106. [Google Scholar]
  29. Kumar S, Pal D. Production of khoa from buffalo milk concentration by reverse osmosis process. Indian J Dairy Sci. 1994;47:211–214. [Google Scholar]
  30. Kumar S, Pal D. Quality of dahi (curd) manufactured from buffalo milk concentrated by reverse osmosis. Indian J Dairy Sci. 1994;47:766–769. [Google Scholar]
  31. Kyrychuk I, Zmievskii Y, Myronchuk V. Treatment of dairy effluent model solutions by nanofiltration and reverse osmosis. Ukr Food J. 2014;3(2):280–287. [Google Scholar]
  32. Lee FV, White CH. Effect of ultrafiltration retentates and whey protein concentrates on ice cream quality during storage. J Dairy Sci. 1991;74:1170–1180. doi: 10.3168/jds.S0022-0302(91)78270-2. [DOI] [Google Scholar]
  33. Lim TH, Dunkley WL, Merson RL. Role of protein in reverse osmosis of cottage cheese whey. J Dairy Sci. 1971;54(3):306–311. doi: 10.3168/jds.S0022-0302(71)85834-4. [DOI] [Google Scholar]
  34. Mistry VV, Maubois JL. Application of membrane separation technology to cheese production. In: Fox PF, McSweeney PLH, Cogan TM, Guinee TP, editors. Cheese chemistry, physics and microbiology. London: Elsevier; 2004. pp. 261–285. [Google Scholar]
  35. Modler HW. Development of a continuous process for the production of Ricotta cheese. J Dairy Sci. 1988;71:2003–2009. doi: 10.3168/jds.S0022-0302(88)79775-1. [DOI] [Google Scholar]
  36. Morin P, Pouliot Y, Jimenez-Flores R. A comparative study of the fractionation of regular buttermilk and whey buttermilk by microfiltration. J Food Eng. 2006;77:521–528. doi: 10.1016/j.jfoodeng.2005.06.065. [DOI] [Google Scholar]
  37. Nielsen WK. Membrane filtration and related molecular separation technologies. Denmark: APV Systems; 2000. [Google Scholar]
  38. Ozer BH, Robinson RK, Grandison AS, Bell AE. Gelation properties of milk concentrated by different techniques. Int Dairy J. 1998;8(9):793–799. doi: 10.1016/S0958-6946(98)00115-0. [DOI] [Google Scholar]
  39. Pal D. Application of reverse osmosis for concentration of milk. In: Khatkar BS, editor. Food science and technology. 1. New Delhi: Daya Publishing House; 2007. pp. 120–131. [Google Scholar]
  40. Pal D, Cheryan M. Application of reverse osmosis in the manufacture of khoa: process optimization and product quality. J Food Sci Technol. 1987;24:233–238. [Google Scholar]
  41. Patel HG, Upadhyay KG, Prajapati JP. Influence of methods of milk concentration on physico-chemical and sensory properties of basundi. Indian J Dairy Sci. 2005;58:85–89. [Google Scholar]
  42. Ranganadham M, Satish Kumar MH, Devraja HC, Garg FC (2016) Traditional dairy products, ICAR e-course book. http://www.agrimoon.com/traditional-dairy-products
  43. Rao HGR. Mechanisms of flux decline during ultrafiltration of dairy products and influence of pH on flux rates of whey and buttermilk. Desalination. 2002;144:319–324. doi: 10.1016/S0011-9164(02)00336-3. [DOI] [Google Scholar]
  44. Reddy DC, Rama Rao M, Reddy CR. Manufacture of long-life dahi and yogurt from pre-concentrated milk. Indian Dairyman. 1986;38:437. [Google Scholar]
  45. Rektor A, Vatai G. Membrane filtration of Mozzarella whey. Desalination. 2004;162:279–286. doi: 10.1016/S0011-9164(04)00052-9. [DOI] [Google Scholar]
  46. Rizvi SSH, Mann RS, Ali SI. A case study of appropriate technology transfer: development of an automated, continuous khoa powder manufacturing process. Indian Dairyman. 1987;39(2):63–67. [Google Scholar]
  47. Savello PA (1998) Method for manufacturing ice cream mixes by reverse osmosis concentration and ultra-high temperature processing. In: US Patent No. 5,753,294. Washington, DC: US Patent and Trademark Office
  48. Shenana ME, Sania MA, Hassaan HM, El-Alfy MB, Nasr WI. Utilization of UF-and RO-retentates in ice cream making. Egypt J Dairy Sci. 2007;35(2):243. [Google Scholar]
  49. Siebert JW, Alejandro L, Sung-Yong K. The commercial potential of new dairy products from membrane technology. J Food Distrib Res. 2001;3:24–33. [Google Scholar]
  50. Sorensen I, Jensen S, Ottosen N, Neve T, Wiking L. Chemical quality of raw milk retentate processed by ultra-filtration or reverse osmosis at the dairy farm. Int J Dairy Technol. 2016;69(1):31–37. doi: 10.1111/1471-0307.12296. [DOI] [Google Scholar]
  51. Sorensen I, Neve T, Ottosen N, Larsen LB, Dalsgaard TK, Wiking L. Storage stability of whole milk powder produced from raw milk reverse osmosis retentate. Dairy Sci Technol. 2017;96(6):873–886. doi: 10.1007/s13594-016-0309-y. [DOI] [Google Scholar]
  52. Stabile RL. Economics of reverse osmosis and multistage evaporation for concentrating skim milk from 8.8 to 45% solids. J Dairy Sci. 1983;66(8):1765–1772. doi: 10.3168/jds.S0022-0302(83)82004-9. [DOI] [Google Scholar]
  53. Suarez A, Fidalgo T, Berdasco MA, Riera FA. UHT condensate recovery by reverse osmosis: a pilot-plant study. Ind Eng Chem Res. 2014;53(39):15237–15244. doi: 10.1021/ie5026166. [DOI] [Google Scholar]
  54. Syrios A, Faka M, Grandison AS, Lewis MJ. A comparison of reverse osmosis, nanofiltration and ultrafiltration as concentration processes for skim milk prior to drying. Int J Dairy Technol. 2011;64(4):467–472. doi: 10.1111/j.1471-0307.2011.00719.x. [DOI] [Google Scholar]
  55. Tamime AY, editor. Membrane processing: dairy and beverage applications. New Jersey: Blackwell Publishing Ltd; 2013. [Google Scholar]
  56. Tamime AY, Robinson RK. Tamime and robinson’s yoghurt: science and technology. Cambridge: Woodhead Publishing; 2007. [Google Scholar]
  57. Vourch M, Balannec B, Chaufer B, Dorange G. Treatment of dairy industry wastewater by reverse osmosis for water reuse. Desalination. 2008;219(1–3):190–202. doi: 10.1016/j.desal.2007.05.013. [DOI] [Google Scholar]
  58. Voutsinas LP, Katsiari MC, Pappas CP, Mallatou H. Production of yoghurt from sheep's milk which had been concentrated by reverse osmosis and stored frozen. 2. Compositional, microbiological, sensory and physical characteristics of yoghurt. Food Res Int. 1996;29(3–4):411–416. doi: 10.1016/0963-9969(96)83273-X. [DOI] [Google Scholar]
  59. Wenten IG, Khoiruddin, Reverse osmosis applications: prospect and challenges. Desalination. 2016;391:112–115. doi: 10.1016/j.desal.2015.12.011. [DOI] [Google Scholar]
  60. Yorgun MS, Balcioglu IA, Saygin O. Performance comparison of ultrafiltration, nanofiltration and reverse osmosis on whey treatment. Desalination. 2008;229(1–3):204–216. doi: 10.1016/j.desal.2007.09.008. [DOI] [Google Scholar]

Articles from Journal of Food Science and Technology are provided here courtesy of Springer

RESOURCES