doi: 10.1111/1471-0307.12050
ORIGINAL RESEARCH
Microstructure of industrially manufactured goat cheese Queso de Murcia al Vino during synaeresis S I L V I A R O V I R A , * V I C T O R G A R C I A , J O S E L A E N C I N A and M A R I A EZ B E L E N L OP Department of Food Science and Technology, Veterinary Faculty, University of Murcia, Espinardo, E-30071, Murcia, Spain
The microstructural parameters of an industrially manufactured goat cheese curd (pore number, area and perimeter, strand thickness and porosity) were analysed by scanning electron microscopy and image analysis during synaeresis. The water-holding capacity, whey fat, pH and moisture content were also determined to establish any relationship with the curd microstructure. The quantification of the different microstructural parameters made it possible to assign pitching and stirring as important processing steps because these steps impart different features to the curd microstructure. Higher pore number was related to reduced pore area, perimeter and strand thickness, but higher porosity and moisture. Keywords Curd, Microstructure, Image analysis, Synaeresis, Physicochemical.
INTRODUCTION
*Author for correspondence. E- mail:
[email protected] © 2013 Society of Dairy Technology
Curd structure evolves during cheese manufacture, while the changes that occur influence the final product. Rearrangements in the protein network are closely correlated with the different stages of cheesemaking, especially the coagulation process, cutting time, setting and whey drainage. The curd is subjected to different treatments to promote the expulsion of the whey from the protein network and to concentrate the casein and fat to a degree that is characteristic of the cheese variety. When the gel is cut or pressed, the matrix contracts and expels the whey contained in the matrix pores. This process, known as synaeresis, enables the cheesemaker to control the moisture content of the cheese variety. The magnitude of cheese moisture variations has a direct influence on plant profitability (Lacroix et al. 2000). Hence, synaeresis is considered one of the most important processes to monitor in order to control any variability in cheese moisture. A higher moisture content is related to faster maturation, a higher yield but lower stability. The ideal moisture content cannot be considered the same for all cheese varieties and, as discussed by Fox and Cogan (2004), the factors that promote and regulate synaeresis in a cheese are specific to each variety. Vol 66 International Journal of Dairy Technology
Synaeresis has been extensively studied and references can be found concerning the factors that influence it (Akkerman et al.1993; Dejmek and Walstra 2004). Mathematical models of cheese curd synaeresis have been developed based on the theory of mechanics of porous media (Tijskens and Baerdemaeker 2004). Synaeresis has also been monitored online during cheesemaking by near infrared light (NIR) sensor (Fagan et al. 2007a,b, 2008; Mateo et al. 2009a,b). Monitoring is based on the changes that take place in the gel, which are characterised by different spectra, making it possible to predict cheesemaking indices and other parameters such as cutting time, moisture and whey fat (Fagan et al. 2007a, 2008; Mateo et al. 2009b). However, the relationship between synaeresis and the development of the curd microstructure during manufacture is not well understood. Using scanning electron microscopy, Boutrou et al. (2002) observed that a more compact network develops with time, and the same was mentioned for cheese storage by Tunick et al. (1997) using transmission electron microscopy (TEM). Nevertheless, these studies did not determine which steps enhance compaction or how the distribution of aggregates changes after each of the setting processes involved in the industrial manufacture of cheese. It was 1
Vol 66
therefore thought necessary to study in greater depth the relationship between synaeresis, the development of curd microstructure and the influence of the setting processes during a typical industrial cheese manufacturing process. Researchers have highlighted the importance of the mechanical shears involved in the different manufacturing steps for the physical properties of cheese, the total volume expelled from the protein matrix and, therefore, for cheese yield. Some authors have focused on the influence of the time and speed of cutting, stirring and pressing (Patel et al. 1972) and the influence of the heating speed on the firmness and physicochemical properties of different types of cheese (Noronha and O’ Riordan 2008). It is known that the mechanical processes involved in the curd-setting process (cutting, heating, stirring and washing) influence curd microstructural changes (Dejmek and Walstra 2004). However, further study is necessary to understand which setting process exerts the greatest influence during cheesemaking, how the curd microstructure develops and how any changes are related to physicochemical parameters, such as water-holding capacity, moisture and pH. Akkerman et al. (1993) stated that the scaling down of the curd-making process leads to curd grains that have little resemblance to those produced in the factory. Therefore, to understand and thereby to improve the knowledge in the manufacturing process of goat cheese, it is essential to apply a quantitative method to describe the changes that take place in particle distribution and in the bonds between aggregates, and how the whey is retained in the protein matrix in curd made under industry conditions. There is insufficient scientific information concerning goat milk and cheese products in relation to synaeresis; this, together with the increasing commercial potential and particular composition of goat milk, justifies focussing this study on the manufacturing process of goat cheese. This will improve the knowledge on its chemical and protein behaviour during manufacture, as is already available for other type of milks. The first aim of this research was to characterise the evolution of microstructural (pore number, area and perimeter, strand thickness and porosity) and physicochemical (whey fat, moisture, pH and water-holding capacity) parameters during synaeresis at short intervals in an industrial goat cheese manufacturing process. The second aim was to observe any significant differences in these parameters according to the time when a given mechanical step is applied in order to ascertain whether these manufacturing steps influence the product microstructure. Finally, microstructural and physicochemical parameters were correlated to determine whether pore number, area and perimeter, strand thickness and porosity, moisture, water-holding capacity and whey fat are influenced by each other. 2
MATERIAL AND METHODS
Samples Curd samples were obtained during the industrial manufacture of Queso de Murcia al Vino, a cheese made with pasteurised Murciano-Granadina goat milk coagulated with calf rennet (chymosin 85%). This cheese is a washed-curd, noncooked, semihard pressed cheese with a cylindrical shape, which is ripened for 45 days. The rind is smooth and bathed in red wine. The samples were taken at 11 different times of the goat cheese production process for each of 13 production days using sterile 250-mL plastic containers (Table 1). The samples were fixed at manufacture as described by Rovira et al. (2011). For each curd sample (200 g), four pieces were processed and two scanning electron microscopy images per piece were analysed. A total of 104 values per sampling time were considered for the study (Table 1). Scanning Electron Microscopy The fixed samples were treated as described by Rovira et al. (2011) and observed at 15 kV (Jeol T-300 SEM; Sollentuna, Sweden). The internal surface of the samples was observed at 65009 magnification. Two representative areas of each piece of the sample were randomly selected and scanned in a horizontal direction at a distance of 9 mm. At least two micrographs were taken and captured for each piece (Inca Oxford image capture system, High Wycombe, UK). Image analysis To compute the microstructural parameters, micrographs were analysed by means of the image analysis software MIP v4.5 (Consulting Image Digital, Barcelona, Spain) in a blind manner. The microstructural parameters of the industrially manufactured goat cheese curd (pore number, area and perimeter, strand thickness and porosity) were defined as described in previous studies (Rovira et al. 2011). Micrographs were transformed, normalised and calibrated as described by Rovira et al. (2011) to guarantee the correct and standard contrast for all the images. For the variables porosity, pore number, area and perimeter, the grey levels were segmented to identify which pixels belonged to pores and which to the protein matrix regions as was described by Rovira et al. (2011). Porosity in the binary image was calculated as described by Rovira et al. (2011). Interactive measurements were used to calculate strand thickness as described by the same authors. Physicochemical analysis The samples were collected at the factory and fixed for scanning electron microscopy analysis and then stored immediately at 4 °C. The physicochemical analyses were © 2013 Society of Dairy Technology
After 10-min moulding 178.8 11.8 Just before moulding 154.6 6.5
Just after moulding 165.2 9.1
Just after pressing 275.5 18.9
Before salting 293.6 35.8
conducted immediately on their reception (1 h) in the University of Murcia Food Technology laboratories. The pH determinations were made using a Crisonâ pH meter (micropH 2001; Barcelona, Spain) connected to a Crisonâ glass combined electrode (1952–2002) previously calibrated at room temperature. Two values were considered per sample, a total of 26 values per production time. For moisture content, curd samples (3 0.01 g) were dried to constant weight according to ISO (1997). Whey fat was analysed according to ISO (1975) using a Gerber Van Gulik butyrometer. To assess the water-holding capacity, curd samples (30 0.1 mg) were centrifuged at 4500 g for 10 min and the exudate was measured and expressed as described by Pandey et al. (2000). Two values were considered per sample, a total of 26 values per production time.
Statistical analysis Statistical treatment of the data was performed using SPSS v15.0 (2006, SPSS Iberica S.L.U. Madrid, Spain). Descriptive analyses, one-way ANOVA, Tukey’s post hoc test and Pearson’s correlation test were used with a significance level of P < 0.05, and the correlations were previously calculated for each of the sampling times, only the parameters which showed correlations in all the sampling times were indicated in Table 2. RESULTS
38.5 1.5 34.5 0.6 Production time (min) (mean SD)
© 2013 Society of Dairy Technology
SD, standard deviation.
2º cut 1 cut Step
Before washing 67.6 3.0
After washing 89.0 2.7
End of heating 97.9 3.0
Second pitching 132.6 2.7
t8.2 t7 t4 t3 t2 t1 Sampling
Table 1 Steps and production times
er
t5
t6
t8.1
t9.1
t9.2
Vol 66
Figure 1 represents the evolution during synaeresis of the microstructural parameters (pore number, area and perimeter, strand thickness and porosity) and of the physicochemical parameters (whey fat, moisture, pH and water-holding capacity). Figure 1(a) details the evolution of porosity, number of pores, moisture content and water-holding capacity. Figure 1(b) displays the development of pH, whey fat content, and the other microstructural parameters analysed, both vs different times of the industrial goat cheese manufacturing process, which are shown on the X-axis. The curd was cut for 10 s, 34.5 0.6 min after rennet addition (t1) (Table 1). Then, another cut lasting 10 min was made (t2). During cutting (t1 t2), whey fat, moisture, porosity, pore area and perimeter decreased but strand thickness and the number of pores increased. No pronounced changes were observed for any of the parameters between t2 and t3, except for the number of pores, which fell sharply. The moisture content also decreased during this period (t2 t3), fell sharply after the heating step (t5) and particularly sharply (4%) after moulding (between t7 and t8.1). The pH remained constant until the end of heating (t5), when it began to decrease slowly. This gradual decrease continued until pressing (between t8.2 and t9.1), when the fall in pH became more accentuated. The whey fat content showed a behaviour similar to the moisture content, although the greatest decrease occurred 3
Vol 66
Table 2 Differences in microstructural and physicochemical parameters throughout cheesemaking steps (post hoc test of Tukey’s) Pore number (absolute value) dfa SLb Processing step First cut Second cut Washing Heating Pitching Pitching Moulding Pitching Pressing Pitching
Periodsc t1 vs t2 t2 vs t3 t3 vs t4 t4 vs t5 t5 vs t6 t6 vs t7 t7 vs t8.1 t8.1 vs t8.2 t8.2 vs t9.1 t9.1 vs t9.2
Pore area (lm2)
525 ***
525 ***
24.85ns 23.66ns 10.43ns 5.63ns 47.57*** 6.37ns 18.70ns 3.167ns 16.22ns 9.45ns
Pore perimeter (lm) 525 ***
0.09*** 0.00ns 0.01ns 0.04ns 0.01ns 0.03ns 0.01ns 0.03ns 0.02ns 0.00ns
0.10ns 0.07ns 0.12ns 0.12ns 0.07ns 0.10ns 0.06ns 0.05ns 0,10ns 0.03ns
Strand thickness (lm) 525 *** 0.03ns 0.42ns 0.02ns 0.02ns 0.07*** 0.03ns 0.14*** 0.03ns 0.01ns 0.03ns
Porosity (%)
Moisture (%, w/w)
pH
104 ***
286 ***
286 ***
3.18ns 2.92ns 3.63ns 2.26ns 1.88ns 3.11ns 6.45ns 0.89ns 1.60ns 0.25ns
6.24*** 3.14*** 4.03*** 2.88* 4.25*** 3.86*** 2.38ns 0.95ns 0.43ns 1.46ns
0.00ns 0.01ns 0.00ns 0.02* 0.04*** 0.08*** 0.06*** 0.12*** 0.52*** 0.11***
d.f, Degrees of freedom: Which corresponds in the case of the first four parameters in the number of strands and pores (and its respective area and perimeter) identified in each of the 104 values per each time considered. b SL, ANOVA signification level (***P < 0.001, *P < 0.05, ns: not significant). c t1-t9.2, times expressed in minutes are detailed in Figure 1. a
(a) 350 325 300 275 250 225 200 175 150 125 100 75 50 25 0
t1
t2
t3
t4
t5
t6
t7
t8.1
t8.2
t9.1
t9.2
Elaboration time ranges porosity(%)
Number of pores
Moisture(%)
WHC (%)*
(b) 7 6.5 6 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0
t1
t2
t3
t4
t5
t6
t7
t8.1
t8.2
t9.1
t9.2
Elaboration time ranges Whey fat (g/100g)
pH (units)
Perimeter of pores (μm)
Strand thickness (μm) Area of pores (μm2)
Figure 1 (a, b) Evolution of the microstructural and physicochemical parameters during cheese manufacture.
4
during the first four periods (between t1 and t4) (65%), after which it remained constant. Pore number decreased during synaeresis by 30%, while the greatest changes in this parameter were observed between t5 and t6. Pore area and perimeter also fell during synaeresis, the former by 35% and the latter by 19%. However, the parameters which most varied during synaeresis were porosity and strand thickness, perhaps the two parameters most related to visible differences in curd microstructure (Figure 2). Porosity decreased (62%) steadily while the pore number, area and perimeter decreased irregularly with shallow peaks. To provide a visual characterisation, the representative scanning electron microscopy images are illustrated in Figure 2. Some pores and strands are identified in each image to provide more information related to differences between the different cheesemaking steps. To study the effects of the different processing steps on the curd microstructure and the changes involved, a contrast Tukey’s test was performed (Table 2). The effect between t1 and t2 corresponds to the first cut. As can be observed, this processing step significantly affected porosity and pore area but did not affect strand thickness, moisture, pore perimeter and number. The second cut (between t2 and t3) that lasted 10 min, only significantly affected porosity and the same applied to the washing step (between t3 and t4); in other words, the second cut and the washing process had no effect on moisture or pore number, area and perimeter, although all these parameters fell between these processing steps, as can be observed in Figure 1. In the heating process (between t4 and t5), porosity significantly changed © 2013 Society of Dairy Technology
Vol 66
Pores
Pores
Pores
(t1)
8 μm
Strands
Strands
Strands
(t2)
8 μm
Strands
Strands
Strands
Pores
Pores
(t4)
8 μm
(t3)
8 μm
Pores
(t5)
8 μm
(t6)
8 μm
Strands
Strands
Strands
Pores
Pores
Pores
(t7)
8 μm
(t8.1)
8 μm
Strands
(t8.2)
Strands
Pores
8 μm
8 μm
Pores
(t9.1)
8 μm
(t9.2)
Figure 2 Scanning electron microscopy images of industrial goat curd at 65009 magnification and examples of strands and pores identified: (t1) at the first cut; (t2) at the second cut; (t3) before washing; (t4) after washing; (t5) at the end of heating; (t6) at the second pitching; (t7) just before moulding; (t8.1) just after moulding; (t8.2) after 10-min moulding; (t9.1) just after pressing; (t9.2) before salting.
(P < 0.05) but no differences were observed for the other microstructural parameters, despite the 5 °C increase in temperature recorded during this processing step. The moulding process (t8.1) significantly affected strand thickness of the curd (P < 0.001). During the pressing step, the microstructure was not affected as no significant changes were observed in porosity, pore number, area or perimeter. Table 3 shows the results of the Pearson’s correlation test carried out to ascertain which relations between the parameters measured might provide information about the parameters which can be used to predict other parameters which that may be of interest to expand the knowledge related to the microstructure and their interaction in the rearrangement © 2013 Society of Dairy Technology
of the protein matrix. As can be observed, there was a moderate negative association between pore number and pore area, pore number and perimeter, and pore number and strand thickness, with a significance level of P < 0.001. Strand thickness and pore area and perimeter decreased as pore number increased. Higher area values were related to higher perimeter values and porosity. There was no correlation between strand thickness and the area and perimeter of the pores, although the strand thickness was negatively correlated with the number of pores, porosity and all the physicochemical parameters. As regards porosity, the results point to a moderate positive interaction with pore perimeter, area and strand thickness and a slightly positive interaction 5
Vol 66
Table 3 Relationship between microstructural and physicochemical parameters Pore number (absolute value) Pore area (lm2) Pore perimeter (lm) Strand thickness (lm) Porosity (%) Moisture (%) WHC (%) Whey Fat (%) pH
0.50*** 0.54*** 0.44*** 0.14*** 0.29*** 0.14ns 0.09ns 0.31***
Area (lm2) 0.89*** 0.03ns 0.48*** 0.33*** 0.59*** 0.28*** 0.08*
Pore perimeter (lm)
0.01ns 0.46*** 0.34*** 0.48*** 0.22*** 0.07ns
Strand thickness (lm)
0.49*** 0.77*** 0.67*** 0.48*** 0.68***
Porosity (%)
Moisture (%)
WHC (%)
Whey Fat (%)
0.77*** 0.74*** 0.63*** 0.42***
0.92*** 0.85*** 0.65***
0.78*** 0.73***
0.46***
*P < 0.05; ***P < 0.001; ns, not significant.
with pore number. Furthermore, higher porosity values are related to higher moisture content, water-holding capacity, whey fat content and pH. Moisture was positively but slightly associated with pore number, area and perimeter and strongly associated with strand thickness and porosity. Moreover, a high water-holding capacity was significantly related to higher values of porosity, moisture, and pore area and perimeter. As regards whey fat content, higher values were strongly related to higher porosity, pore area and perimeter, moisture and water-holding capacity but lower strand thickness. This correlation illustrated the evolution of these parameters during synaeresis, taking into account the differences in the fat content of the raw material, which will be related to differences in the fat content expelled with the whey. No correlation was observed between whey fat content and pore number. The values of all the parameters analysed, except strand thickness, decrease with time, which reflects the behaviour of the microstructural and physicochemical properties of the cheese during synaeresis. As shown in Table 3, higher pH values are associated with higher values of moisture, waterholding capacity, whey fat content, porosity and pore area and perimeter and lower values of strand thickness. DISCUSSION The increase in the number of pores observed between t1 and t2 may be related to the increase in permeability. Dejmek and Walstra (2004) suggested that deformation of the curd, such as that involved in the process of cutting, causes local rupture of the network and that this rupture may be related to an increase in permeability. This increase in the number of pores was assumed to be associated with lower moisture and porosity – hence, the decrease in pore area and perimeter observed in Figure 1. The slopes observed of moisture content in this figure are consistent with the observations of Dejmek and Walstra (2004), who mention that the steepest slope for the decrease in water content is observed after moulding (t8.1). 6
Fox and Cogan (2004) described how removing the whey from the curd and continuous stirring promotes synaeresis, which suggests a decrease in the moisture content, as was seen in our case (Figure 1). However, these authors suggested that washing the curd was related to an increase in the moisture content, which is not evident from Figure 1, where the moisture content is seen to decrease by 5% (from t4 to t5, washing-heating process). Also, Casiraghi et al. (1987) reported that adding water after part of the whey has been removed enhances synaeresis and leads to a nonsignificant fall in the water content. The fact that the curd moisture content in our study decreased (5%) and did not increase during the washing process is presumably related to the 5 ° increase in temperature (from 33 to 38 °C), which was recorded during curd washing in this study. In our study, the decrease in the water content due to moulding was more pronounced than the decrease corresponding to the pressing step. As can be observed in Figure 1, the decrease in the moisture content during synaeresis is associated with decreased waterholding capacity, pH and whey fat. Dejmek and Walstra (2004) mentioned that a decrease in pH is associated with enhanced synaeresis and so less water remains in the curd network. The decrease in the water-holding capacity during synaeresis (24%) (Figure 1) was lower than expected and mainly occurred between t3 and t7 during washing, heating and stirring. However, the decrease was lower than the total decrease in the moisture content (45%) and porosity (62%), both of which are supposedly related to the capacity of the curd to retain water. The manufacturing process, therefore, had a higher effect on the moisture content and porosity than on the water-holding capacity during this time. The pH gradually decrease from heating (t5) to the pressing step, in agreement with Dejmek and Walstra (2004), who reported that the pH falls as cheesemaking progresses due to acid production, a well-known fact that enhances synaeresis. The fact that the strand thickness increased by 62% during synaeresis (although it remained constant during the last four periods analysed) confirms the quantitative results of © 2013 Society of Dairy Technology
Vol 66
Dejmek and Walstra (2004), namely that the grains carry on fusing until the pores are no longer interconnected, which takes place after the moulding process. From the images shown in Figure 2, differences in the curd matrix can be observed between the different processing steps considered, which are more pronounced between micrographs (t1, t2, t3, t4) and (t5), (t7) and (t8.1), (t8.1) and (t8.2), and finally (t9.1) and (t9.2). The results obtained in the Tukey’s test between t3 and t4 are in agreement with Casiraghi et al. (1987), who reported that the washing step has no significant effect on the moisture content. As shown in Table 2, the most pronounced changes in pore number, strand thickness, porosity occurred during the continuous pitching and stirring of the curd (between t5 and t7), although the second pitching did not significantly influence pore number. Hence, pitching and stirring impart different features to the curd microstructure and so can be considered as important processing steps in cheesemaking, influencing the final properties of the cheese curd. Curd compaction was enhanced by moulding (t8.1) and not by pressing (t9.1), as was described by Dejmek and Walstra (2004), who observed significant differences in strand thickness in moulding but not in pressing. The fact that moulding was the process that produced the greatest differences in strand thickness is important. Fusion of the grains is greatly enhanced by curd deformation and such deformation is the result of moulding rather than of pressing (Dejmek and Walstra 2004). The changes in pH which accompany moulding became more pronounced during pitching (between t8.1 and t8.2) and pressing (t9.1), as described by Fox and Cogan (2004). The pH was the parameter most influenced through cheesemaking, mainly due to the production of lactic acid by the starter bacteria. The parameters analysed during synaeresis have been a good way of observing the variations in pH, which are closely related to the water expelled from the protein matrix and so with the moisture content in which the time elapsed plays an important role. The interactions observed in Table 3 confirm that the association between the pore area, pore perimeter and microstructure porosity is higher than the corresponding association between porosity and the number of pores in the protein matrix. This is of great importance because controlling the processing steps that most influence the pore area and perimeter (i.e. cutting and pitching of the curd) will enable the desired porosity to be obtained in the final cheese. Furthermore, the interactions observed between porosity and moisture, water-holding capacity, pH and whey fat content suggest that the microstructure is strongly related to physicochemical parameters which, in turn, are strongly related to cheese yield and so to profitability. The fact that all the parameters are intercorrelated and demonstrate a high degree of correlation with porosity © 2013 Society of Dairy Technology
(Table 3) is of great importance because porosity plays an important role in texture perception (Langton et al. 1996). It can be assumed that the variables studied in this work can be regarded as a good tool for relating curd microstructure during synaeresis with final product features. CONCLUSIONS Porosity plays an important role in controlling the physicochemical features of cheese particularly moisture, pH and whey fat content, which are closely related to product quality and, hence, profitability. This relationship confirms the potential of the microstructure to predict other parameters such as moisture and water-holding capacity. The continuous pitching and stirring of the curd between heating and moulding are the processing steps that most affect the microstructure because pore number, strand thickness, porosity and pH significantly changed during this time. Pitching and stirring impart different features to the curd microstructure and so can be considered as important processing steps for obtaining the desired properties in the cheese curd. ACKNOWLEDGEMENTS This work was financially supported by the Spanish Ministry of Education with a research fellowship. The samples were generously provided by Central Quesera Montesinos S.L. (Murcia. Spain). The assistance of the services of Microscopy and Image Analysis of the University of Murcia is also acknowledged. REFERENCES Akkerman J C, Lewis R O and Walstra P (1993) Fusion of curd grains. Netherlands Milk Dairy Journal 47 137–144. Boutrou R, Famelart M H, Gaucheron F, Le Graet Y, Gassi J Y, Piot M and Leonil J (2002) Structure development in a soft cheese curd model during manufacture in relation to its biochemical characteristics. Journal Dairy Research 69 605–618. Casiraghi F M, Peri C and Piazza L (1987) Effect of calcium equilibria on the rate of syneresis and on the firmness of curds obtained from milk UF retentates. Milchwissenschaft 42 232–235. Dejmek P and Walstra P (2004) The syneresis of rennet-coagulated curd. In Cheese: Chemistry, Physics and Microbiology, vol 1, 3rd edn. pp. 71–101. Fox P F, ed. London: Chapman and Hall. Fagan C C, Castillo M, Payne F A, O’Donnell C P and O’Collaghan D J (2007a) Effect of cutting time, temperature and calcium on curd moisture, whey fat losses and curd yield by response surface methodology. Journal Dairy Science 90 4499–4512. Fagan C C, Leedy M, Castillo M, Payne F A, O’Donnell C P and O’Collaghan D J (2007b) Development of a light scatter sensor technology for on-line monitoring of milk coagulation and whey separation. Journal Food Engineering 83 61–67. Fagan C C, Castillo M, O’Donnell C P, O’Callaghan D J and Payne F A (2008) On-line prediction of cheese making indices using backscatter of near infrared light. International Dairy Journal 18 120–128.
7
Vol 66
Fox P F and Cogan T M (2004) Factors that affect the Quality of Cheese. In Cheese: Chemistry, Physics and Microbiology, vol. 1, 3rd edn, pp. 583–608. Fox P F, ed. London: Chapman and Hall. ISO (1975) Determination of fat Content. Van Gulik Method. ISO standard 3432/33. Geneva: International Organization for Standardization. ISO (1997) Determination of Moisture Content. ISO standard 1442. Geneva: International Organization for Standardization. Lacroix S, Jimenez-Marquez S A and Emmons D B (2000) Practical Guide for Control of Cheese Yield: Factors Affecting the Accuracy of Plant Cheese Yield Estimates, pp. 99–113. Brussels, Belgium: International Dairy Federation. Langton M, Astr€om A and Hermansson A M (1996) Texture as a reflection of microstructure. Food Quality Preference 7 185–191. Mateo M J, O’Collaghan D J, Everard C D, Castillo M, Payne F A and O’Donnell C P (2009a) Validation of a curd-syneresis sensor over a range of milk composition and process parameters. Journal Dairy Science 92 5386–5395. Mateo M J, O’Collaghan D J, Everard C D, Fagan C C, Castillo M, Payne F A and O’Donnell C P (2009b) Influence of curd cutting
8
programme and stirring speed on the prediction of syneresis indices in cheese-making using NIR light backscatter. LWT-Food Science Technology 42 950–955. Noronha N and O’ Riordan E D (2008) Influence of processing parameters on the texture and microstructure of imitation cheese. European Food Research Technology 226 385–393. Pandey P K, Ramaswamy H S and St-Gelais D (2000) Water Holding capacity and gel strength of rennet curd as affected by high-pressure treatment of milk. Food Research International 33 655–663. Patel M C, Lund D B and Olson N F (1972) Factors affecting syneresis of renneted milk gels. Journal Dairy Science 55 913–918. Rovira S, L opez M B, Ferrandini E and Laencina J (2011) Microstructure quantification by scanning electron microscopy and image analysis of goat cheese curd. Journal Dairy Science 94 1091–1097. Tijskens E and Baerdemaeker J (2004) Mathematical modelling of syneresis of cheese curd. Mathematics and Computers in Simulation 65 165–175. Tunick M H, Cooke P H, Malin E L, Smith P W and Holsinger V H (1997) Reorganization of casein submicelles in Mozzarella cheese during storage. International Dairy Journal 7 149–155.
© 2013 Society of Dairy Technology
