Transient vacuum drops in the milkline during one ordinary milking were recorded as a measure of vacuum stability in milking installations on 184 Norwegian dairy farms in the summer of 1997. An association between the frequency of vacuum drops and technical data including milkline diameter, length and slope, number of milking units and effective reserve was demonstrated. The direct connection between the transient vacuum drop and its cause could be established for most drops during milking observations. A high frequency of transient vacuum drops in the milkline was associated with a high level of mastitis and a high new infection rate as inferred from changes in somatic cell counts for individual cows. The frequency of vacuum drops during one milking is only a rough indication of the long-term vacuum stability in a specific installation, and must be interpreted with care. The dimensions and slope of the milkline and the effective reserve probably give as good information about the installation's ability to maintain a stable vacuum.

Thirty Holstein cows in mid-lactation (158±20 DIM) were given a total mixed ration based on grass silage, maize silage and rolled barley. After a preliminary period of 1 week, this diet was supplemented with nothing (control), unprotected fish oil (3.7% of dry matter, DM), or two levels of glutaraldehyde-protected microcapsules of fish oil (1.5% and 3.0% of DM, respectively). Unprotected and protected supplements contained, respectively, 74% and 58% of DM as lipids. Cows given the unprotected supplement reduced their feed intake by >25%. Consequently, these cows lost body weight and produced less milk. DM intake, body weight, and milk yield were unaffected by protected fish oil. Fish oil reduced both milk fat and protein percentages, and decreased the proportion of short-chain fatty acids, stearic, and oleic acids in milk fat. Milk trans C18[ratio ]1 fatty acids increased in cows given both unprotected and protected fish oil. Milk fat content of very-long-chain n3 polyunsaturated fatty acids, including C20[ratio ]5 and C22[ratio ]6, increased with fish oil in the diet. Accordingly, the peroxide index increased and a taste panel was able to detect unusual taste in milk from cows consuming the higher level of protected fish oil and disliked the milk from cows given unprotected fish oil. In conclusion, when lactating cows consumed fish oil, milk concentration of long-chain n3 fatty acids increased and mammary de novo synthesis of fatty acids decreased, but milk yield and milk protein content were reduced, and the milk was more susceptible to oxidation and its taste was adversely affected.

Sixteen Holstein cows in mid-lactation were used to determine whether alterations of mammary fatty acid metabolism are responsible for the milk fat depression associated with consumption of fish oil. Cows were given a total mixed ration with no added fish oil (control), unprotected fish oil (3.7% of dry matter), or glutaraldehyde-protected microcapsules of fish oil (1.5% or 3.0% of dry matter) for 4 weeks. Milk samples were taken once a week and a mammary biopsy was taken from a rear quarter at the end of the treatment period. Milk fat content was lower in cows given unprotected fish oil (26.0 g/kg), 1.5% protected fish oil (24.6 g/kg) and 3% protected fish oil (20.4 g/kg) than in cows fed the control diet (36.0 g/kg). This was mainly due to a decrease in the synthesis of short-chain fatty acids. Consumption of protected fish oil decreased the abundance of lipogenic enzymes mRNA in the mammary gland. Acetyl-CoA carboxylase, fatty acid synthase, and stearoyl-CoA desaturase mRNAs for cows given 3% protected fish oil averaged only 30%, 25% and 25% of control values, respectively. Dietary addition of unprotected fish oil slightly decreased mRNA abundance of these enzymes but markedly reduced the amount of lipoprotein lipase mRNA. Milk fat content was significantly correlated with gene expression of acetyl-CoA carboxylase, fatty acid synthase, and stearoyl-CoA desaturase but not lipoprotein lipase. These results suggest that fish oil reduces milk fat percentage by inhibiting gene expression of mammary lipogenic enzymes.

The objective of this study was to determine the effects of stage of lactation (SOL) and time of year on plasmin-derived proteolytic activity in the milk of pasture-fed dairy cows in New Zealand. Four herds of 20 Friesian cows were used, one herd calving in each of January, April, July and October. Cows grazed ryegrass/white clover pasture only, except during June (winter) when all cows received supplementary pasture silage. Milk samples were collected on four occasions during the year (spring, summer, autumn and winter) from each cow in milk, to give a total of three samples per cow (early, mid and late lactation; c. 30, 120 and 220 days after calving, respectively). Milk samples were analysed for plasmin-derived proteolytic activity. There was no effect of either SOL or time of year on plasmin activity and therefore yields of plasmin followed patterns in milk yield (highest in early lactation and in summer). There were effects of both SOL and time of year on plasminogen-derived and total plasmin plus plasminogen-derived activity, both of which were highest in late lactation and in spring. Changes in plasminogen-derived activity and total plasmin plus plasminogen-derived activity due to SOL were not only due to the decrease in milk yield associated with advancing lactation, because enzyme yields were also increased with advancing lactation. Similarly, effects of time of year on plasminogen-derived activity and total plasmin plus plasminogen-derived activity could not be attributed solely to concomitant changes in milk yield, and may be influenced by the variation in the quality and quantity of feed during the year inherent in a pasture-based dairy system. Effects of SOL on proteolytic activity were greater than, and independent of, effects of time of year.

In the context of identifying intrinsic time temperature integrators (TTIs) for evaluating heat processing of milk, the extent to which milk fat content has an effect on alkaline phosphatase (ALP) and lactoperoxidase (Lpo) inactivation and on β-lactoglobulin (β-lg) denaturation kinetics was studied. Inactivation and denaturation kinetics were analysed in whole, semi-skimmed and skimmed milk. In previous experiments (isothermal and non-isothermal heating conditions), heat inactivation of ALP and Lpo and heat denaturation of β-lg were found to follow first order kinetics. This allowed experimental design to be simplified. Data analysis was performed by non-linear regression and results were evaluated by construction of joint confidence regions. The possible effect of milk fat was illustrated by temperature time tolerance (TTT-) diagrams. Although initial ALP activity was lower in skimmed milk compared with semi-skimmed or whole milk, kinetics were comparable and fat content did not seem to substantially affect the ALP test result for pasteurized milk. Unlike ALP, Lpo inactivation and β-lg denaturation kinetics differed significantly in milk with different fat content. Differences between Lpo inactivation kinetics were relatively small and acceptable in the context of quantifying the process impact. Denaturation of β-lg, on the other hand, seemed to be enhanced at higher milk fat content (>72 °C).

The effects of heat treatment and homogenization of whole milk on chemical changes in the milk fat globule membrane (MFGM) were investigated. Heating at 80 °C for 3–18 min caused an incorporation of whey proteins, especially β-lactoglobulin (β-lg), into MFGM, thus increasing the protein content of the membrane and decreasing the lipid. SDS-PAGE showed that membrane glycoproteins, such as PAS-6 and PAS-7, had disappeared or were weakly stained in the gel due to heating of the milk. Heating also decreased free sulphydryl (SH) groups in the MFGM and increased disulphide (SS) groups, suggesting that incorporation of β-lg might be due to association with membrane proteins via disulphide bonds. In contrast, homogenization caused an adsorption of caseins to the MFGM but no binding of whey proteins to the MFGM without heating. Binding of caseins and whey proteins and loss of membrane proteins were not significantly different between milk samples that were homogenized before and after heating. Viscosity of whole milk was increased when milk was treated with both homogenization and heating.

The composition of the volatile fraction of milk from cows was investigated in a survey of milk samples using a headspace sampling technique and gas chromatography coupled to mass spectrometry analysis (GC-MS). Milk samples were collected from 12 farms, selected for similar management, breed and level of production. Farms were also grouped according to the type of forage in the ration: (1) hay; (2) hay and maize silage; (3) hay, maize silage and grass silages. Forty-one compounds in milk were isolated and identified from GC-MS headspace analysis. Quantitatively, the most representative chemical class was ketones (eight compounds, 170 μg/kg), followed by aldehydes (nine compounds, 63 μg/kg), alcohols (eight compounds, 36 μg/kg), and lower amounts of hydrocarbons (six compounds), sulphur compounds (three compounds), esters (four compounds) and terpenes (three compounds). The novel headspace sampling technique, and the consequent reduction of sample pre-treatment, allowed the identification of low-molecular weight volatile compounds, and reduced the risk of producing artefacts during analysis.

Discriminant analysis was used to identify a classification criterion for milk samples, using type of forage in the ration as a grouping variable. Posterior probability error rate indicated that aldehydes provided one of the best discriminant criteria for grouping milks according to ration composition. When all 41 identified volatile compounds were included, discriminant analysis selected nine compounds (acetone, 2,3-butanedione, 2-butanone, ethanol, acetaldehyde, ethylacetate, ethylisovalerate, dimethylsulphone) that did not fail the tolerance test and which correctly classified 100% of the original cases.

Manchego cheese can be manufactured from raw or pasteurized ewes' milk. An automatic purge and trap apparatus, coupled to a GC-MS was used to isolate, identify and compare the relative amounts of the volatile components of raw and pasteurized Manchego cheese during ripening. The majority of volatile compounds were more abundant in raw milk (RM) cheeses than in pasteurized milk (PM) cheeses. Alcohols and esters predominated in the profile of RM Manchego cheese, while methyl-ketones and 2,3-butanedione were quantitatively important in PM cheeses. Branched chain alcohols were much more abundant in RM cheeses. The discriminant analysis separated 100% samples into RM or PM cheeses by using only 16 volatile compounds. Aroma intensity was correlated with esters, branched chain aldehydes and branched chain alcohols in RM cheeses, and with esters, branched chain aldehydes, 2-methyl ketones and 2-alkanols in PM cheeses. Diacetyl was positively correlated with the aroma attribute ‘toasted’ and negatively correlated with aroma quality in PM cheeses.

An automatic purge and trap apparatus, coupled to a GC-MS was used to study the seasonal variability of the volatile fraction of raw milk Manchego cheese. Both season and dairy significantly affected abundance of most volatile compounds. Most aldehydes, methyl ketones, n-alcohols, and secondary alcohols reached significantly higher concentrations in spring cheeses. Branched chain alcohols showed significantly higher concentrations in autumn and winter cheeses, while significantly higher amounts of diketones were found in summer cheeses. Most ethyl esters reached higher concentrations in spring and winter cheeses and lower in autumn cheeses. Lower concentrations of α-pinene were found in spring cheeses, and higher amounts of limonene were observed in winter cheeses. Heptane and octane were significantly more abundant in summer cheeses. No significant seasonal differences were found either for quality or intensity scores.

The structure development of a soft cheese curd model has been studied in relationship to its rheological properties and its biochemical characteristics (pH, amount and partition of minerals, casein proteolysis) at different technical steps including cutting, drawing, three turns and demoulding. Scanning electron microscopy was used to observe structural changes during the drainage of a fat-free soft cheese. The micrographs provided visual evidence of changes in the casein matrix from casein particles aggregated in clusters to uniform strands observed at the demoulding. The initial increase of loss tangent and of the exponent of the power law between G′ and G′′ and frequency (that were maximal at the second turn) was related to the solubilization of micellar calcium phosphate, while intact caseins and large casein fragments accumulated in the curd. After the second turn, the strength, Youngs' and loss moduli of the curd increased greatly. The hydrolysis of αs1-casein into αs1-I-CN f(24–199) may facilitate the rearrangement of casein particles within the curd. The pH-induced solubilization of calcium phosphate continued throughout the manufacture process but was unexpectedly incomplete at the end of the drainage. Combination of electron microscopic observations with dynamic rheological measurements and chemical and biochemical assessments provided increased knowledge about the structure of soft cheese during drainage, an important but poorly understood cheese making stage.

The dairy industry would benefit from rapid and non-destructive determination of moisture content of cheese products. The two components primarily responsible for the low-field magnetic resonance (MR) spin-spin relaxation (T2) signal of cheese products are fat and the water bound to protein. If the moisture component of the signal can be distinguished from the fat component, it should be possible to measure moisture using an MR sensor. Therefore, a key aspect of the development of an MR moisture measurement method is examination of techniques for analysis of T2 relaxation curves. One common method of T2 analysis of complex foods, such as cheese, is to fit multi-term exponential models to the curves. An alternative approach is proposed which uses stretched exponential models. The single-term stretched exponential model has been used for porous rock systems and polymers, but not for foods. The T2 relaxation curves were analysed using both models and the results were compared. The number of unknowns in the three-term exponential and two-term stretched exponential models was reduced by assuming the relaxation curve of the fat component was the same as the relaxation curve of pure fat. In each model, one of the exponential terms described the behaviour of the water in the cheese analogue, while the remaining term or terms described the behaviour of the fat. For each model the T2 relaxation time associated with the water was well correlated with moisture content. Coefficients of determination of the relaxation time versus moisture from each of the two models were nearly identical. The advantages and disadvantages of the two models are discussed.

In the dairy industry re-use and multi-use cleaning-in-place (CIP) systems are operated by circulating chemicals and water without taking the equipment apart. The solutions, which become polluted due to the removal of fouling compounds, are drained periodically when they are considered to be too polluted. This work shows the large variations in composition (pollution, surface tension, etc) of the industrial caustic solutions coming from milk standardization and pasteurization plant CIP throughout their life time (7 days) and from 1 week to another. The work is also intended to show how nanofiltration (1 kg mol−1 molecular weight cut-off) was robust and performed well, with good recovery of caustic solutions, even when faced with large variations of solutions composition: high caustic yield, permeation flux (J) in the range 42–110 l h−1 m−2, average chemical oxygen demand (COD) reduction equal to 0.58 and low surface tension change. Equations have been established for the prediction of J as a function of initial membrane hydraulic resistance (Rm) caustic concentration, volume reduction ratio (VRR) and initial soluble COD. When VRR increased, both J and pollution retention decreased despite the increase in irreversible fouling induced by the increase of soluble pollution concentration in retentate . The higher the initial soluble COD, the sharper the decrease in J vs. VRR. Since irreversible fouling was usually small (0.1–3.4×1013 m−1, that is to say of the same order of magnitude as Rm), the membrane cleaning could be efficiently performed by using single phase sodium hypochlorite alternately with a more expensive acid–base cleaning sequence. The obtained permeate was a clear regenerated cleaning solution with low soluble COD (0.2–3.5 g/l) and surface tension (56–30 mJ m−2) which could be successfully exploited owing to its cleaning potential.