The reversibility extent of one and two reverse CO2 acidification cycles on the physico-chemical and rennet coagulation properties of milks reconstituted from low- (LH) or medium- (MH) heat skim powder, enriched or not with calcium and pH adjusted or not was investigated. The ionized calcium concentration, buffering properties and average casein micelle size of untreated and CO2-treated milks were evaluated before and after a chilled storage for 2 days. The ionized calcium concentration and buffering properties have been modified by the CO2-treatment, particularly after a second CO2-cycle. These modifications were highly dependent on the initial milk properties and chilled storage. Inversely, the average casein micelle size was not significantly changed. In addition, the rennet-clotting behaviour checked by near infrared spectroscopy (NIR-S) and rheology (SAOR) indicated the main factors responsible for changes in the casein micelles environment and dynamic casein micellar calcium phosphate reorganization, especially after two CO2-cycles. A single CO2-cycle induced a better rennetability for non Ca-enriched milk reconstituted from MH-powder. A second CO2-cycle was particularly efficient to improve Ca-enriched pH-adjusted milks.

The influence of a CO2-acidification cycle on the acid (glucono-δ-lactone, GDL) gelation properties of skim milk with and without added low-methoxyl pectin (LM-pectin) was assessed. Ionic calcium level, zeta potential, particle size, buffering properties and small amplitude oscillatory rheology moduli were monitored. The presence of LM-pectin in milk had an impact on the average size of the casein micelles and a large and dominant influence on its rheological behaviour during GDL acidification. The application of a CO2–pH cycle (pHtarget 4.9) as a milk pretreatment induced during GDL acidification a stabilization of the colloidal system in wide pH range (pH 6.0–5.1) with modifications of the structure of the casein micelles before the onset of gelation. These modifications induced a significant improvement on its acid gelation behaviour. The measurements of Ca2+ level during GDL acidification showed that an important and significant part of the Ca2+ released during the CO2–pH cycling was electrostatistically trapped by pectin molecules in the serum.

Monitoring of rennet-induced coagulation of milk reconstituted from low-heat or medium-heat skim milk powders with or without calcium enrichment (6.25mmolkg−1) was carried out by five analytical methods. The four types of milk samples allowed evaluation of these methods on substrates characterized by very different rennetability. Thirteen parameters describing the coagulation process were extracted from the various measurements for each sample and compared. The visual flocculation and gelation times evaluated by the Berridge test and dynamic small amplitude oscillatory rheometer (DSAOR), respectively, were not significantly different from specific parameters identified from near-infrared (NIR) spectra analysis. DSAOR, piezoelectric rheometer, and NIR spectrometer allowed and evaluation of the beginning of gel firming. It clearly appeared that NIR spectroscopy, an on-line sensor, could best assess rennet-induced coagulation of studied samples.

Response surface methodology was used to investigate the effect of salt supplementation on the micellar composition of reconstituted skim milk subjected to acidification by CO2 pressure to pH 5.8, followed by depressurization under vacuum. Using a Doehlert design, calcium and phosphate were added to skim milk in the range of 0 to 25 mmol/kg and 0 to 16 mmol/kg of milk, respectively, and the pH was adjusted to 6.65 +/- 0.02. After carbonation, the milk sample was depressurized, and the pH returned to its initial value without modification of the ionic strength. Micellar composition was assessed by the concentration of micellar Ca, P, Mg, and protein, and the buffering properties of milk. The second order polynomial models satisfactorily predicted the effect of salt supplementation on the micellar composition (R2adj > 0.75). Added calcium was the most determinant factor, and favored the removal of Ca, P, Mg, and proteins from the soluble phase to the micellar phase when this addition was less than 17.5 mmol/kg of milk. Above this concentration, only the concentration of micellar Ca increased. The buffering response surface showed that the amount of micellar calcium phosphate increased to a maximum upon addition of 17.5 mmol of Ca/kg. By comparison with a control sample (supplemented but untreated skim milk), changes were essentially due to salt supplementation and not to the CO2 treatment. We suggest that Ca formed micellar calcium phosphate when added at a concentration less than 17.5 mmol/kg; whereas above this concentration, Ca bound directly to micellar proteins.

The separation and quantification of the major components of whole casein were performed on an RP-18 column and on a Q Hyper D(TM) column (Beckman Instruments) which is based on hyperdiffusion, a new concept in analytical chromatography. Using reversed phase chromatography, the 4 major caseins (a(s1-), alpha(s2-), beta- and kappa-) were well resolved. With anion-exchange chromatography, as-caseins were eluted as a single peak but gamma-caseins could be distinguished. These methods presented two significant interests: the analyses were carried out in a single run within less than 15 min and a very good quantification and an excellent repeatability were achieved.

The effects of succinylation on the rennet coagulation of milk were investigated to find a relationship between the level of chemical modification and the ability of milk. protein to coagulate. The strengthening rate and the final firmness of the gel decreased as the extent of succinylation increased, which could be related to the microstructural differences caused by milk succinylation. The kinetics of nonprotein N release were not affected by milk succinylation, but the rennet clotting time increased as the extent of chemical modification increased. These results suggest that milk succinylation affected the secondary or aggregation phase of enzymatic coagulation by increasing the electrostatic repulsions between para-casein micelles and, more indirectly, by dissociating casein from micelles and preventing them from participating in network formation of the gel.

Chemical modification of casein micelles by succinylation was investigated in order to assess the role of electrostatic charges in both acid and rennet coagulation, as well as to find a relationship between the chemical modification applied and the ability of milk proteins to coagulate. Succinylation of milk resulted in a delay of gelation and in a decrease in the final firmness of a rennet gel. The resistance of casein micelles to aggregation could be explained by the increase in electrostatic repulsion induced by the introduction of additional negative charges to the proteins. Results showed that the changes in the rheological properties upon rennet coagulation could be related to the microstructural differences observed between the succinylated and the reference milk: so changing the balance of protein charges upon decreasing the pHi increased the coagulation time of acid gels, but might also considerably alter the formation of hydrophobic interactions between succinylated casein micelles during the aggregation phenomena in rennet milk gels.

To evaluate the nature of the main interactions that are involved in the formation of casein gels made by the acidification or rennet coagulation of milk, we investigated the combined effects of ultracentrifugation and specific dissociating agents on protein solubilization. The method used was based on the ability of gelled proteins to resist the dissociating action of solutions of urea, sodium dodecyl sulfate, or EDTA. Results showed that hydrophobic interactions and calcium bonds were the most important forces involved in the rennet milk gel matrix; hydrophobic, hydrogen, and electrostatic interactions were homogeneously distributed in the gel formed from acidified milk.

ABSTRACTReconstituted skim milk with varying concentrations of total solids was coagulated using glucono‐δ‐lactone (GDL). Microscopic, turbidimetric and rheological procedures were used to examine mineral solubilization, buffering capacity, casein dissociation and micellar solvation during gelation. Total solids of the milk affected pH of the onset of gelation attributable to differences in colloïdal calcium phosphate in the casein particles during acidification. Firmness and elasticity of the resulting gel increased with total solids from a more direct contribution of dry matter during the last stage of acid milk gel formation.

During the manufacture of process cheese, biochemical characteristics (casein solubilization, peptization coefficient, and water-holding capacity) were investigated using a combination of microscopic and rheological techniques in order to understand the influence of pH. The contribution of ionic interactions to the stabilization of this structure was also studied.Relationships were observed between pH variation and the characteristics of process cheese that demonstrated the importance of pH control during the manufacturing process. Optimal pH conditions during manufacture ranged from 5.7 to 6.0. Small changes in ionic composition and strength modified the protein interactions substantially and had important repercussions on the final structure and quality of the protein gel that was established during processing of cheese. In addition to ionic interactions, hydrogen and hydrophobic interactions appeared to be important in the structural stabilization of process cheese.

ABSTRACTAn original method for syneresis evaluation was developed, based on image analysis which enabled direct monitoring of curd shrinkage. Image analysis was employed in combination with the customary tracer method to follow the effects of cutting on curd syneresis. The two methods were then utilized to study curd syneresis in the presence of ethanol, a solvent less polar than water. Curd syneresis was reduced in the presence of diluted ethanol which led to the hypothesis that the mechanism which causes rennet curd to synerese probably relates to changes in protein conformation, essentially due to hydrophobic bonding.

Treatment of milk by injection of CO2 and depressurizationwas studied for the effect on milk salts andon the buffering properties of skim milk in order to correlate them to mineral balance. Injection of CO2 bypressure decreased the pH of milk. After depressurization,pH returned to its initial state. No changes were observed in the concentrations of inorganic P, Ca, or Mg in the aqueous phase, determined after ultracentrifugation, but the buffering curves of depressurized milk that had been treated with CO2 were different from those of the original milk. For the initial milk, two buffering peaks at pH 4.95 and 5.40were observed instead of one, showing that the solubilization of CO2 induced the formation of a new salt system. When CO2 pressure increased, the buffering value at pH 4.95 decreased; solubilization occurred for the inorganic calcium phosphate and the Ca directly bound to caseins. The second buffering value at pH 5.40 increased as new salts were formed. Higher pressures shifted the pH of maximal buffering from pH 4.95 to 4.75 and from pH 5.40 to 5.05. The CO2 had a reversible effect on pH and an irreversible effect on inorganic colloidal calcium phosphate, which was changed into other salt forms.

pH-induced changes in casein micelles during direct acidification and bacterial fermentation of reconstituted skim milk at 20 degrees C were monitored by scanning electron microscopy (SEM) in combination with biochemical and rheological methods. For SEM casein micelle observations, an original method of milk sample preparation with porous inorganic membranes was developed. Micrographs suggested that different stages of micellar association were related to pH and that between pH 5.5 and 5.0 casein micelles coalesced. Correlations between microstructural and biochemical changes in casein micelles, and theological behavior of milk or gel, help to explain the different steps leading to the final protein network of the acid milk gel.

A direct cell size measurement technique and an image analysis based sizing method were developed. The former consisted of a manual size measurement of the two-dimensional cell images on a video screen, with automatic data recording. This method was chosen as the reference. The latter, a semiautomatic method took advantage of a commercial computer program designed for image processing and particle morphology analysis. It gave average and median size values which were compatible with the manual method. However, the performance of these time consuming methods is limited. Hence, the laser granulometry technique, intrinsically far more powerful while capable of analysing millions of sample objects in a short time delay, was applied. The comparison revealed that this method gives too low size values, particularly in disagreement with the known dimensions of the bacterial (Zymomonas mobilis) cells. A size correction method was developed to realign the granulometry results of Z mobilis cell samples with those of the direct manual measurement method.

Scanning electron microscopy (SEM) and rheology and permeability were used to evaluate effects of clotting temperature and calcium phosphate on renneted reconstituted milk gels, especially microstructure and permeability. Dynamic measurements with a piezoelectric Viscoprocess determined rheological properties of curd after various times of renneting. An increase in clotting temperature accelerated specific rennet action and gel formation; maximum stiffness was reached faster but its value was weaker. Addition of calcium phosphate increased rate of gel formation. Maximum stiffness was reached faster but its value was not changed. Gel microstructure was not very different but its permeability coefficient decreased.

ABSTRACT Scanning electron microscopy (SEM) and permeability measurements were used to evaluate the effect of coagulation temperature on renneted reconstituted milk gel characteristics (gelation and coagulation), especially with regard to gel microstructure and permability. SEM of milk gel showed casein micelle strands and clusters linked to form a very porous framework with gaps delimited by discontinuous walls. Casein micelles fused together more tightly and strands and clusters became denser and thicker as coagulation temperature increased, resulting in a more open structure with increased permeability.

ABSTRACTChanges in calcium solubilization and rennet reaction rate were investigated after the addition of calcium (6.25 mM) to reconstituted milk. Analysis of soluble and ionic calcium showed that distribution of added calcium was very fast (60 and 25 min, respectively), and that micelles of enriched milk remained more mineralized at pH 6.70 to 5.00. In the presence of sufficient amounts of rennet and hydrogen ions, K‐CSSeh hydrolysis and aggregation rates were increased by addition of calcium. The calcium supplement led to extended maintenance of the micellar structure and to a more easily drained curd. Thus, it could be used advantageously for lactic acid cheese making.

Summary The effect of acidification method (microbiological with or without renneting, HCl addition) on mass transfer, fouling structure and the rheology of the retentate was studied in the ultrafiltration of skim milk coagula using a mineral microfiltration membrane. The increase in fouling with time appeared to determine permeate flow rates, which were higher in biological coagula, and the protein retention rates which were higher in chemical coagula. Fouling was investigated using scanning electron microscopy. The rheological study showed that at the same total solids, biological coagula were more viscous than chemical coagula. The initial coagula (total solids 97 g/kg) all displayed pseudoplastic behaviour at low shear velocities and Newtonian behaviour at high velocities. Ultrafiltration of fat-enriched milk coagulum to a dry weight corresponding to a soft cheese (total solids 334 g/kg; fat in total solids 60%) gave satisfactory permeate flow rates and protein retention rates. Performance was related to the composition of the product, the hydrodynamic parameters used and the resulting fouling. The rheological study showed that the initial coagulum behaved as a pseudoplastic body at low shear rate and for higher velocities as a Newtonian liquid. The concentrated retenate behaved as an ideal viscoplastic body (Bingham body).