Milk Filtration Research Articles

The effect of heat treatment of PES and PVDF ultrafiltration membranes on morphology and performance for milk filtration

The flat sheet polyethersulfone (PES) and poly(vinylidene fluoride) (PVDF) membranes were prepared by immersion precipitation technique. The influence of hot air and water treatment on morphology and performance of membranes were investigated. The membranes were characterized by AFM, SEM, cross-flow filtration of milk and fouling analysis. The PES membrane turns to a denser structure with thick skin layer by air treatment at various temperatures during different times. This diminishes the pure water flux (PWF). However the milk permeation flux (MPF) was considerably improved at 100 °C air treatment for 20 min with no change in protein rejection. The smooth surface and slight decrease in surface pore size for air treated PES membrane at 100 °C compared to untreated membrane may cause this behavior for the membrane. The water treatment of PES membranes at 55 and 75 °C declines the PWF and MPF and increases the protein rejection. This is due to slight decrease in membrane surface pore size. The treatment of PES membrane with water at higher temperature results in a porous structure with superior performance. The fouling analysis of 20 min treated membrane indicates that the surface properties of 100 °C air treated and 95 °C water treated PES membranes are improved compared to untreated membrane. The SEM observation depicts that the morphology of air and water treated PVDF membranes was denser and smoother with increasing the heat treatment temperature. The 20 min air treated PVDF membranes at 100 °C and water treated at 95 °C exhibited the highest performance and antifouling properties.

Impact of zeta potential and size of caseins as precursors of fouling deposit on limiting and critical fluxes in spiral ultrafiltration of modified skim milks

Caseins, the main proteins of milk, are known to be responsible of flux limitations in tangential filtration of skim milk. These proteins are well known to be versatile with the pH. Their initial form, corresponding to the so-called “native casein micelles” is due to a complex aggregation of monomers through calcium phosphate bridges that can be disrupted by small pH variations. This paper deals with the determination of both critical and limiting fluxes during ultrafiltration of natural and modified skim milks. pH ranging from 3.7 to 11.5 allowed to consider skim milk ultrafiltration (UF) for production purpose (pH 6.6–6.7) as well as acidic and alkaline concentrated effluents. The physico-chemical characteristics of caseins, as diameter and zeta potential, were measured according to the pH variations obtained by the addition of HCl and NaOH. Then, both limiting and critical fluxes variations were correlated to these characteristics. The critical and limiting fluxes varied in a quite complicated way with the pH, leading to four different regions. At acidic pH, zeta potential was null and fluxes were controlled by large particles of caseins that precipitated in the solution. Around neutral pH, the diameter of casein micelles remained constant whereas the zeta potential varied strongly. In this region the fluxes increased linearly with the zeta potential of caseins. Then at alkaline pH, the zeta potential of caseins remained constant whereas the diameter varied showing successively partial destructuration and partial “restructuration” of new types of micelles, both different from native casein micelles. In this region the fluxes varied with the rate of large micelles. Finally, retention of lactose (main neutral solute of milk) and whole proteins, mainly responsible of COD, were measured and discussed in relationship with above-mentioned parameters.

High performance polyethersulfone UF membrane for manufacturing spiral wound module: preparation, morphology, performance, and chemical cleaning

AbstractLarge sheet asymmetric polyethersulfone (PES) ultrafiltration membranes were prepared via phase inversion process in a continuous conveyor system with addition of PVP to the casting solution. Dimethylacetamide (DMAc) and mixture of water and isopropyl alcohol (70/30 v%) were used as solvent and non‐solvent respectively. The prepared membrane was 0.96 m wide and 3 m long. The pieces of the membrane were selected randomly for characterization in terms of performance using cross flow filtration for milk concentration, image analysis, scanning electron microscopy (SEM), and cleaning procedures. It was found that the prepared membrane has high porosity and high water flux during milk filtration. In addition, cleaning experiments indicated that NaOH/HCl/NaOH sequence is an effective procedure for cleaning the fouled membrane during milk concentration. Copyright © 2007 John Wiley & Sons, Ltd.

Differential analysis of deposition layers from micellar casein and milk fat globule suspensions onto ultrafiltration and microfiltration membranes

Particle deposition behaviour on membrane filters is of the utmost importance for the flux and selectivity. For milk filtration not only the particle behaviour as such but also the behaviour in time is a factor to take into account. Therefore we applied a differential analysis to the flux decline during dead-end filtration of latex, casein micelles and milk fat globules. Based on our analysis for latex we distinguished two regions, one best described by pore blocking and one by cake filtration. Besides these two regions, no other time dependent behaviour was observed, as was expected for latex. However, for casein micelles and milk fat globules various other time dependent effects were observed and quantified. Pore blocking of an UF membrane by micellar casein was relatively small, as concluded from the values of the pore blocking parameter of 0.1–0.2. In contrast, the cake resistance was relatively high, 35 times higher than expected from the Carman–Kozeny theory. This may be attributed to a decrease of the size of the casein micelles, leading to the formation of very compact layers. The specific cake resistance of micellar casein was lower for a MF membrane where depth filtration played a role. Milk fat globules generate layers with a very high cake resistance, both for liquid and for partly crystallised fat. For liquid fat, spreading of milk fat globules onto the membrane and deformation of milk fat globules seem to be the main cause for the high cake resistance. When a large part of the milk fat was crystallised ( T = 5 °C), partial coalescence of milk fat globules in the cake layer seems to play a role. Differential cake analysis proved to be a valuable method for the evaluation of concentrated particle layers, resulting in better knowledge about the specific (time dependent) particle behaviour in these layers. In this way, parameters are identified that are essential for the design of microfiltration or ultrafiltration processes.

Membrane fractionation of milk: state of the art and challenges

Separation of milk into well-defined fractions will lead to a more optimal use of milk components (milk fat, casein, serum proteins) and their functional properties. In principle, membrane separation technology is well capable of large-scale fractionation of milk. Membrane processes for the isolation of serum proteins from whey, and the reduction of bacteria and spores in skimmed milk have already been adopted by the dairy industry. Other separation steps, such as the separation and/or fractionation of cream and the concentration of casein micelles, could become feasible in the future. In this paper, we give an overview of the current use of membranes in the fractionation of milk and discuss recent developments in membrane technology. Besides these different separation steps, we focus on the problem of fouling, which is considered the limiting factor in milk filtration. Different strategies to avoid and to decrease fouling are discussed, as well as their suitability in a fractionation process for milk. Further, new developments on computer modeling are discussed. Both analytical models and computational fluid dynamics (CFD) provide insight in the mechanisms of fouling and can be used to evaluate different methods to control fouling and to optimize process parameters. For the rational design of a milk fractionation process, it is essential that research in various fields, such as new membranes, module design, fouling control and modeling be carried out as part of an integrated approach toward a radical new process.

Membrane fouling during filtration of milk––a microstructural study

The surface morphology and internal microstructure of a membrane are the result of membrane manufacturing processes and subsequent use during fluid processing in industry. Both these structural factors have a great effect on fouling and filtration performance. In this study, scanning electron microscopy (SEM), atomic force microscopy (AFM) and X-ray photoelectron spectroscopy were used to comprehensively characterise the surface of unused microfiltration and ultrafiltration membranes, the fouled layer on the surface of membranes used for milk filtration and the internal fouling within the used membranes. A simple and effective sample preparation method was developed to study the internal membrane structure using high-resolution field emission SEM with low-accelerating voltage. Various methods of structural characterisation were compared and the results showed that for flat sheet membranes AFM is an appropriate and convenient technique for examining the surface topography of membranes. In contrast SEM is a very appropriate technique for examining the cross-sectional and internal structure of a membrane, either unused or fouled. Using these complimentary techniques it has been shown that internal fouling, during filtration of skim milk, proceeds by protein–polymer and protein–protein interactions. A gel layer forms on the surface of the membrane and has been imaged using SEM. This layer is slightly compressible and densifies as it grows. Fouling initiation commences after a very short filtration time.

Critical stability conditions in skimmed milk crossflow microfiltration: impact on operating modes

There exists a critical ratio (convection towards the membrane/erosion) in crossflow micro- filtration, MF, below which there is no marked fouling by colloidal particles and above which per- formance are altered: sharp increase of fouling, reduced operating time, large decrease in perme- ability and solute transmission. This paper outlines the impact of the critical ratio on skimmed milk MF processing (separation of casein micelles from the soluble proteins) and gives objective ele- ments (taking into account the critical ratio) for the selection of start-up procedures and stationary modes of operation: controlled transmembrane pressure or permeation flux; static or dynamic counter- pressure, the latter achieved by circulating the permeate co-current to the retentate to maintain an equal transmembrane pressure profile along the filtering path. Results showed that the controlled trans- membrane pressure mode of operation was appropriate for conducting skimmed milk filtration, since it prevented from sharp increase of the hydraulic resistance of fouling and maintained nearly constant performances (permeability, selectivity) over the course of the time. In order to determine the more appropriate counter pressure mode and to predict skimmed milk performances, a four step method was proposed : i) assessment of the critical ratio; ii) selection of the critical flux at the required wall shear stress; iii) determination of the corresponding critical transmembrane pressure in dynamic counter-pressure; iv) calculation of the MF performance in static mode taking into account the gradient of transmembrane pressure induced by the retentate pressure drop. hydrodynamics / crossflow microfiltration / operating mode / critical permeation flux / milk

Microfiltration and Ultrafiltration of Milk: Some Aspects of Fouling and Cleaning

Milk filtration using membranes is one of the most effective ways of separating fat, proteins and many types of microbial cells without the use of heat. This investigation was undertaken to determine the main factors responsible for the decline in permeate flux during milk processing and to study the fouling phenomena. There are a variety of fouling mechanisms, which may or may not be significant depending upon the combined characteristics of the membrane, feed stream and operating conditions. In this study, the performance of a microfiltration membrane was compared with an ultrafiltration membrane in terms of fouling and cleaning characteristics using a cross flow membrane rig. Atomic Force Microscopy was employed to explore the fouling phenomena. Factors of importance for membrane cleaning such as temperature, transmembrane pressure, circulation time and cleaning performance were also investigated.

Cloning and sequencing of the central region of the flagellin gene from the Gram-positive bacterium Clostridium tyrobutyricum ATCC 25755.

The purpose of this study was to sequence the central part of the coding region of the Clostridium tyrobutyricum fiagellin gene to improve the immunoenzymatic counting of cells after milk filtration. The coding region was amplified by PCR, and the amplified products were cloned. A DNA sequence analysis of positive clones gave us 1,131 nucleotides with a partial calculated flagellin molecular mass of 40,143 Da. The flagellar filament protein sequence exhibited high levels of homology to sequences of flagellin protein from other bacteria in both N- and C-terminal parts, but little homology in the central domain. A PCR-restriction fragment length polymorphism analysis of amplified C. tyrobutyricum flagellin gene products confirmed the variability of the central domain. The flagellin mRNA was determined to be 1.1 kb in size, which suggests a monocistronic mRNA. Furthermore, the deduced protein flagellin contains eleven potential N-glycosylation sites and one sequence rich in serine, which could be modified by O-glycosylation.

Monoclonal antibody detection of Clostridium microcolonies directly on membrane used for milk filtration

The normal procedure for bacterial colony detection requires a nitrocellulose transfer step after membrane filtration and culture to prevent the development of a high background during the immunodetection. In this paper, we describe a modification of the basic protocol that omits the transfer step and reduces the risk of background. Previous observations indicated that interactions between milk components (principally cream) and membrane are responsible for the high non-specific staining observed. Experiments were performed to remove lipid components or to block the membrane binding sites before milk filtration. Samples of milks of different origin (collected at different times of the year) and different membranes were tested. The results obtained showed that removing lipids did not significantly improve the test but, on the contrary, led to an antigen diffusion. Incubation of the membrane in 0.1% (w/v) of Tween 20 in phosphate-buffered saline before milk filtration prevented non-specific binding, and allowed performance of the detection without any noticeable background.

Monoclonal antibody detection of Clostridium microcolonies directly on membrane used for milk filtration

Monoclonal antibody detection of Clostridium microcolonies directly on membrane used for milk filtration

The concentration of bovine IgG in human breast milk measured using different methods.

The bovine IgG concentration in human breast milk was measured using three different methods: competition radioimmunoassay (RIA), competition enzyme-linked immunosorbent assay (ELISA) and sandwich ELISA. In 36 samples of human breast milk taken from mothers on diets including cow's milk, the concentration of bovine IgG was significantly higher when measured by either competition RIA or ELISA (mean 324.9 +/- 97.4 ng/ml and 396.7 +/- 58.7 ng/ml, respectively) than by sandwich ELISA (3.44 +/- 0.48 ng/ml). In 7 samples of human breast milk taken from mothers on cow's-milk-free diets of more than one month, the concentration of bovine IgG was still within a detectable range (112.3 +/- 41.4 ng/ml) using competition ELISA, whereas it was not detected when using sandwich ELISA, suggesting that competition ELISA might measure some unknown substances in addition to bovine IgG. Gel filtration of pooled human breast milk revealed three large fractions. The concentration of bovine IgG in every fraction was significantly higher using competition compared with sandwich ELISA, suggesting that the difference between these assays does not result from measuring different quantities of bovine IgG, or of fragments of this protein. The time course of the bovine IgG level was different between each human breast milk sample after taking cow's milk. We showed that human breast milk contains bovine IgG at a concentration of approximately 3 ng/ml when analysed using sandwich ELISA. This level is comparable to the quantity of either casein or of beta-lactoglobulin in human breast milk. Bovine IgG might be an important allergen in cow's milk allergy due to the quantity in human breast milk.

A rapid spectrophotometric method for quantitative determination of lactulose in heated milk and milk products

Summary - The reddish-brown colour produced on heating protein-free filtrate of heated milks containing less than 1.0% lactose and free of ions in 6.4 N sulphuric acid was used for quantitative determination of lactulose. The absorptivity of the coloured solution of lactulose at 330 nm was 14fold more than that obtained from 62.S-fold more concentration of lactose and other aldoses. By following the optimum conditions established in this method, a minimum of 10 mg/100 ml concentration of laetulose can be determined in high heated unsweetened milk and milk products. The method was found to be very useful and sufficiently sensitive under dairy industry conditions where the ion removed step can be omitted if raw milk filtrate is taken for blank preparation.

Identification and partial characterization of parathyroid hormone-related protein in human and bovine milk

Parathyroid hormone-related protein (PTHrP) was measured in human and bovine milk by radioimmunoassay (RIA) and bioassay, and the molecular forms characterized by gel chromatography and immunoblotting of affinity-purified PTHrP. Mean immunoreactive PTHrP(1-34) concentrations were 23 and 87 micrograms/l in human and bovine milk respectively. Bioactive (BIO) PTHrP concentrations determined by cyclic AMP production by ROS 17/2.8 cells correlated significantly (P less than 0.001) with those obtained by RIA (BIO = 1.04RIA--3.4, r = 0.939). Gel filtration of human and bovine milk identified several peaks with immunoactivity and bioactivity. Immunoblotting of affinity-purified PTHrP revealed multiple molecular species including components with mobilities similar to those of PTHrP and its subfragments. These studies confirm the presence of immuno- and bioactive PTHrP in milk and suggest that post-translational processing is complex and variable.

Membrane Filtration of Milk for Counting Spores of Clostridium tyrobutyricum

Estimation of Clostridium tyrobutyricum spores in milk by most probable number, after anaerobic incubation in a liquid medium, is not rapid, sensitive, nor reproducible.Membrane filtration appears to be the only technique able to concentrate microorganisms in milk sufficiently for detection of Clostridium tyrobutyricum spores at the high sensitivity required by the spoilage capacity of this microorganism in Emmenthal-type cheese.Incubation of milk with .072% trypsin at 55°C for 15min, followed by filtration under three bars (30.6 Mg/m2) pressure through .8-μm porosity membrane filtration, permitted filtration of volumes larger than 300ml, although with high standard deviation and important seasonal variation. The test of other enzymes gave lower results.Viability was not impaired, and spore recovery was good under these conditions of treatment and filtration. This technique seems practical for enumeration of Clostridium tyrobutyricum spores in milk.

The passage of maternal dietary proteins into human breast milk.

Samples of breast milk and serum were taken from 29 women at various stages of lactation before and after they had ingested 1 raw egg and half a pint of cow's milk. The samples were analysed for cow's milk and egg proteins using a solid-phase radioimmunoassay. beta-Lactoglobulin, ovalbumin (OA) and ovomucoid were detected in breast milk from 10 out of 19, 13 out of 22 and 7 out of 9 women, respectively, in concentrations ranging from 110 pg/ml to 6.4 ng/ml. Maximum levels in breast milk were attained 4 or 6 h after ingestion and in serum 1-2 h earlier. The OA in breast milk was of normal molecular size (43,500 daltons) and indistinguishable from native OA in the radioimmunoassay. Gel filtration of whole breast milk or high molecular weight fractions at pH 2.6 failed to reveal additional OA, suggesting that immune complexes were not present. In serum, OA was detected both in its native form and in immune complexes. The possible significance of these antigens in the suckling infant is discussed.

High Performance Liquid Chromatographic Determination of Bifidobacterium bifidum Growth Factors in Human Milk

An isocratic high performance liquid chromatography (HPLC) method has been developed for the determination of Bifidobacterium bifidum growth factors in human milk. The method involves the gradual addition of 3 volumes of ethanol to the milk sample, filtration, and analysis of the growth factors in the filtrate by HPLC. The HPLC system consisted of a carbohydrate analysis column, a water-acetonitrile (70 + 30) solvent system, a flow rate of 1.0 mL/min, and a refractive index detector. The method is simpler and requires less time than the present microbiological method. Moreover, it revealed for the first time the presence of 2 separable growth factors in all human milk samples tested. The HPLC method developed is sensitive and can be used to monitor the type and the amount of growth factors in mothers' milk during lactation.

Identification of Yellow Material Remaining on Disks After Filtration of Milk

Identification of Yellow Material Remaining on Disks After Filtration of Milk

Correlation of physical and biological properties of mouse mammary tumor agent.

Biophysical procedures have been used to determine the size and structure of the biologically active agent responsible for the transmission, through milk, of mouse mammary adenocarcinoma. Filtration of milk from RIII high-breast-cancer mice through gradocol membranes with decreasing pore sizes indicated that a minimum of activity passed through intermediate pore sizes (100 to 160 mmicro). Filtrates through smaller pores were significantly active. Milk treated with small doses of deuteron irradiation produced more tumors than the control, unirradiated milk; larger doses indicated a particle size much less than 100 mmicro. Free diffusion experiments indicated that the activity was associated with particles of two different sizes. Altogether the data denoted the presence of a large agent about 100 mmicro in diameter and a small agent 20 to 30 mmicro in diameter or possibly smaller. Furthermore, the presence in the milk of an inhibitor 40 to 60 mmicro is indicated by the results of all three approaches. The complex nature of the milk agent disclosed by the physical measurements agrees with the picture of one of the structures revealed by electron microscopy as well as with seemingly conflicting measurements reported in the literature. The large agent defined by these indirect methods corresponds to the whole particle seen in the electron microscope and the small agent corresponds to its internal core or nucleoid. It is suggested that the nucleoid is essentially a nucleic acid which may, in the absence of the "inhibitor," retain its activity after being stripped of its outer membrane or sac.

Filtration and Clarification of Milk

Filtration and Clarification of Milk