arXiv:1410.3897v1 [cond-mat.soft] 15 Oct 2014

arX

iv:1

410.

3897

v1 [

cond

-mat

.sof

t] 1

5 O

ct 2

014

Quantifying thermally induced flowability of rennet cheesecurds

Hiroyuki Shimaa, Morimasa Tanimotob

aDepartment of Environmental Sciences, University of Yamanashi, 4-4-37, Takeda, Kofu, Yamanashi 400-8510, JapanbDepartment of Local Produce and Food Sciences, University of Yamanashi, 4-4-37, Takeda, Kofu, Yamanashi 400-8510, Japan

Abstract

Conversion of liquid milk to cheese curds is the first stage incheese manufacture. Changing the rigidity of cheese curds throughheating and pH control is an established method for preparing fresh curds, whereas a similar method to prepare fully coagulatedcurds is largely unknown. This study elucidated the effect of temperature variation on the viscoelastic moduli of fully coagulatedcurds under different pH conditions. The results showed that rennet curds treated at pH 4.8 exhibited drastic changes in theviscoelasticity at 43◦C, above which the degree of fluidity exceeded the degree of rigidity. The viscoelastic moduli exhibitedexponential decay as a function of temperature, which was independent of pH.

Keywords: Casein micelle, Milk protein, Rennet clotting, Viscoelasticity, Dynamic shear modulus

1. Introduction

Why does milk appear white? The answer is because of thepresence of casein, a milk-specific protein comprising manynu-tritionally important amino acids. Casein is one of the best-known phosphorylated proteins, in which the phosphate groupsare bound to many serine residues. The phosphate groups con-fer an overall negative charge to the casein molecule, enablingit to form a complex with Ca ions. The complexes are stabi-lized by assembling each other, resulting in numerous colloidparticles (called casein micelles) that are uniformly suspendedin the milk. In the presence of light, therefore, the casein mi-celles diffusively reflect incident light, because of which milkappears white (i.e., by the Tyndall effect).

The casein in milk can be precipitated and separated byadding an enzyme called rennet to milk. The enzyme prote-olyses the surface of casein micelles, causing them to attracteach other. Consequently, numerous casein micelles aggregate,solidify, and finally precipitate. The rennet-induced aggrega-tion of casein micelles has been utilized in the production ofvarious casein products that serve as functional protein ingredi-ents in diverse applications, including foods, pharmaceuticals,and cosmetics [1, 2, 3, 4]. Against this backdrop, understandingthe essential properties of casein micelles and their aggregatesis important both for the development of optimal dairy productsand for the control of their functionalities [5, 6, 7, 8, 9, 10].

Structural transformation and development of firmness dur-ing milk clotting have been examined by sophisticated mea-surement techniques [11, 12, 13]. The structure and propertiesof casein micelles are highly dependent on the environmentalconditions. In the production of natural cheese, for example,rennet-induced proteolysis and the resultant aggregationof ca-sein micelles lead to cheese curd. The obtained cheese curd,re-inforced by a three-dimensional casein network, exhibits com-plex viscoelasticity in accordance with changes in the temper-ature, pressure, pH, and protein concentration [14, 15]; similar

complex variations have been observed in concentrated milk[16] and mixed food proteins [17]. Furthermore, it is envis-aged that artificially heating and lowering the pH of the curdsafter they have been fully coagulated should cause the caseinnetwork to dissociate and rearrange, thus potentially increas-ing its fluidity. Most previous studies on this issue, neverthe-less, have focused primarily onfresh curds obtainedjust afterthe initiation of casein micelle aggregation. Little attention hasbeen paid to quantitative determination of thermally and/or pH-induced changes in the viscoelasticity ofalready fully coagu-lated curds, though the latter is crucial for the textural controlof processed dairy products.

The present study is designed to investigate the effect of tem-perature variation on the viscoelastic moduli of fully coagulatedrennet curds. Dynamic shear tests are performed to quantita-tively assess the variation of the temperature dependence of themoduli with changes in pH. Molecular interpretations of vis-coelastic behaviours are also presented.

2. Sample Preparation

The milk samples were purchased from dairy farms in thefoothills of Mount Yatsugatake (on the border of Yamanashiand Nagano Prefectures), Japan. Milk clotting was performedby using a device, with a 20 L capacity, that is usually usedfor obtaining natural cheeses (Nichiraku Kikai Co., Ltd.) byfollowing the procedure described below.

The milk samples were first pasteurized by maintaining 18L of raw milk at 65 ◦C for 30 min. This process eliminatesthe bacteria, thus preventing the degradation of milk proteinsat high temperatures. The sample pH was 6.73 immediately af-ter pasteurization. The sample was then cooled to 31◦C, and 18mL of a lactobacillus culture solution was added to it in order toacidify it. To prepare the culture solution, 0.3 g of Direct Vat Set(DVS) lactobacillus starter (CH-N11, Chr. Hansen, Nosawa &

Preprint submitted to International Journal of Food Properties August 16, 2018

Co., Ltd.) containing a blend of several types of concentrated,dried, and frozen lactobacilli was mixed with 300 mL of pas-teurised milk, and the solution was cultured overnight at 21◦C.After adding the culture solution, the sample was maintained at31 ◦C, which is the temperature that produces the greatest lac-tobacillus activity, for 30 min. The sample pH at this stage was6.50.

Rennet was then added to the above sample that slightly acid-ified by lactobacilli. In this experiment, 0.5 g of rennet (CHY-MAX, Chr. Hansen, Nosawa & Co., Ltd.) dissolved in sterilecold water was added to the sample, which was then left tostand at 31◦C for 35 min. After the milk started to coagulate(pH 6.46), the sample was cut into cubes of 12-13 mm to re-move the whey from the curds. Five minutes after cutting, thesample was gently agitated for 15 min to encourage the removalof the whey. As a result, whey corresponding to one-third ofthe original sample weight was eliminated, yielding fresh curdgranules.

To completely eliminate the whey from the sample and ad-just the pH to the target values (4.8-6.0), hot water (80◦C) wasadded to the curd granules. The granules were then gently agi-tated at 0.2◦C/min until a temperature of 38◦C (pH 6.34) wasachieved; eventually, all the whey was removed. In the finalstep, the curds were aged until they reached the target pH. Thetime required to attain curds with the lowest pH was approxi-mately 6 h. After pH adjustment, a series of curds with differentpH values were frozen and stored. Immediately before mea-surement, the curds were defrosted in a refrigerator and thenstirred at 50-60◦C.

3. Dynamic Shear Measurement

A thorough understanding of the viscoelastic properties ofcheese curds is not only a major scientific challenge, but also ofpractical importance since many dairy products and processesrely on the control of these properties under an external load[18, 19, 20]. Small-amplitude oscillatory shear analysis,a non-destructive protocol for determining the viscoelasticityof a ma-terial, is an efficient approach to address this issue [21, 22]. Inthis analysis, oscillatory shear strain is applied to the samplewhile maintaining the strain in the linear viscoelastic region.Two main parameters are determined from this test: one is theelastic (or storage) modulus designated byG′, which is a mea-sure of the elastic energy stored per oscillation cycle. Simplyexpressed, this parameter indicates the degree to which thesam-ple gives a solid-like response to the dynamic load. The otherparameter is the viscous (or loss) modulus designated asG′′,which is a measure of the energy dissipated as heat per cycle,thus indicating the degree to which the sample shows liquid-likebehaviour.

Actual measurements were performed using the Anton PaarMCR 302 rheometer. The samples were thinly sliced and sand-wiched between two flat disk plates with 25 mm radius, facingeach other, separated by a gap of 2 mm. The sample surfacewas coated with silicone oil to prevent evaporation of waterduring measurements. After coating the sample, it was grad-ually cooled from 50◦C to 5 ◦C at a rate of 2◦C/min, during

Figure 1: Temperature dependence of the complex modulus amplitude (|G∗ |)of fully coagulated curds under three different pH conditions. Inset: Enlargedview of the data in the high-temperature region.

which the variations inG′ andG′′ were measured by applyingthe oscillatory shear. For all the measurements, the angular fre-quency of oscillation was set to 1 Hz and the shear strain was0.1%. From theG′ andG′′ data, the amplitude of the complexmodulus|G∗| =

√

(G′)2 + (G′′)2 was also evaluated under eachtemperature and pH condition.

4. Results and Discussion

Figure 1 shows the temperature (T ) dependence of the com-plex modulus (|G∗|) of the fully coagulated curds under threedifferent pH conditions: 4.8, 5.5, and 6.0. At each pH condition,10 samples were analysed and only a slight sample dependencewas evident from the curves of|G∗|. Each data point in Fig. 1represents the mean value of|G∗| for the 10 samples. From thefigure, it is clear that for all the evaluated pH conditions,|G∗| be-low 15 ◦C exhibited a rapid decrease with increasingT , whichchanged to a slow decay till 50◦C. The rapid decrease in|G∗|in the low-T region is attributed to the melting of lipid glob-ules pooled in the sample. In fact, the curds produced hereinwere not delipidated, and thus contained a significant amountof lipid. These lipid globules solidify at low temperaturesandare embedded into the voids of the casein network of the curds.Consequently, the entrained ‘solid’ globules act as inert fillersthat enhance the stiffness of the curds [23]. The effect of suchfillers diminishes with increasingT because of the softening ofthe solid globules. This accounts for the rapid decrease in|G∗|that persists until T reaches the melting point of the lipid (15◦C). It follows from Fig. 1 that even above the melting point,|G∗| continues to decay slowly with increasingT . The slow de-

2

Figure 2: Semi-logarithmic plot of the temperature variance of storage (G′)and loss (G′′) moduli. Respective upward arrows indicate the sol-gel transitionpoint (right) and the shoulder caused by lipid-globule melting (left).

cay indicates thermally excited flow of the molecules that con-stitute the casein network inside the curd.

Figure 2 presents a single-logarithmic plot of the same dataas in Fig. 1, while the storage (G′) and loss (G′′) componentsare plotted separately. The shoulder structures in the curves ofG′(T ) as well asG′′(T ) at 15◦C become clear in these plots, asindicated by an upward arrow on the left. Interestingly, above15 ◦C, most of the curves show a nearly exponential decay withthe functional form,

G′(T ) = C · exp(

−TA

)

(1)

with appropriate constantsC andA. For instance, above 15◦Cthe decay ofG′(T ) at pH 6.0 is represented by the form shownin Eq. (1) withC = 4.0×105 Pa andA = 16.5 ◦C. Above 15◦C,the other curves are fairly well fitted to the exponential form ofEq. (1), wherein the value ofA is insensitive to pH variation.

The most important observation in Fig. 2 is the occurrenceof the sol-gel transition for the data acquired at pH 4.8. Thetransition point is highlighted by the right-hand-side arrow. Ingeneral viscoelastic materials, the structural transition from thesol phase (liquid state) to the gel phase (solid state) is signifiedby the intersection between the two curves ofG′(T ) andG′′(T )[24, 25]. Accordingly, the average critical temperatureTC, atwhich the fully coagulated curds undergo the sol-gel transition,for 10 samples is estimated to be 43◦C. This means that thesample at pH 4.8 exhibits solid-like behaviour belowTC, butassumes viscous liquid behaviour aboveTC. The heat-inducedflowability is believed to result from higher molecular mobilityand reduced cross-linkage that are internal to the casein net-work [26]. These two (and maybe other) physicochemical fac-tors promote the molecular alignment in the direction parallelto the tensile direction, enhancing the flow of the cheese curdsat T > TC. The results also confirmed that the value ofTC isunique and reproducible for an identical sample, though thereis a slight sample-dependent fluctuation of the order of a fewdegrees Celsius.

The structural transition from liquid to solid (and vice versa)demonstrated in Fig. 2 is consistent with the thermal softeningof the casein network in fully coagulated Mozzarella cheesere-ported in the literature [27]. Using the squeezing flow method,it has been reported that Mozzarella cheese shows a decreasedresistance to flow with increasing temperature; the relaxationtime of the Mozzarella cheese was reduced by several times onincreasing the temperature from 30◦C to 60◦C, similar to thepresent observation for|G∗| (see inset of Fig. 1). Compared to aprevious work [27], the present work offers the distinction thatthe characteristic temperatureTC for the transition can be suc-cessfully determined in an objective and quantitative manner byoscillatory shear analysis.

It should be stressed that the physicochemical origin of theexponential decay ofG′ and G′′ is yet to be confirmed. Toour knowledge, however, there is a lacuna in the understand-ing of how casein micelles and other chemical components in-teract with each other in fully coagulated curds and how theseinteractions evolve the viscoelastic properties; this situation isin contrast with the now reasonably well-established structureand composition of individual casein micelles before aggrega-tion [28, 29, 30, 31, 32]. It is thus conjectured that a theoreticaldescription of the exponential decay represented by Eq. (1), ifsuccessfully developed, should shed light on the mechanismofthe interaction of cheese curd constituents; this subject will beaddressed in our future work.

Before closing the article, we would like to point out that acertain class of cheese demonstrates thermally induced solid-ification, opposite to the present observation. This cheeseiscalled paneer [33], a South Asian variety of soft cheese pre-pared by acid coagulation. The ability of paneer to be deepfried (i.e., non-melting property even at high temperatures) is incontrast with the enhanced melting of rennet cheeses. Micro-scopic interpretation of the contrasting thermal responses in theviscoelasticity of paneer (acid-based cheese) relative torennetcheeses has not been established, thus remaining a challenge inthe field of dairy science.

5. Conclusion

The sol-gel phase transition of fully coagulated curds drivenby temperature and pH controls was quantified herein with thehelp of small amplitude oscillatory shear measurements. Forthe curds analysed, the transition occurs at 43◦C when the sam-ples are subjected to pH 4.8. The occurrence of phase transitionin fully coagulated curds implies a novel technique for the ma-nipulation of the cheese texture through pH regulation and heattreatment, which will be complementary to the well-establishedtechnique for preparing fresh curds. The results also show thatthe viscoelastic moduli at different pH conditions universallyfollow an exponential decay as a function of temperature. Thisfinding may provide a platform for better understanding of thestructural relaxation and molecular interaction mechanism ofthe casein network within fully coagulated curds.

3

Acknowledgement

The authors cordially acknowledge the fruitful discussionsheld with Prof. Ryoya Niki and Prof. Katsuyoshi Nishinari.Gratitude is also expressed to Ms. Tazuko Watanabe (AntonPaar Japan K.K.), Mr. Toshiaki Shioya (Unitec Foods Co.,Ltd.), and Mr. Kunio Ueda (Yueisha) for their technical sup-port with measurements and sample preparation. This workwas supported by JSPS KAKENHI Grant Numbers 25390147and 25560035. H.S. gratefully acknowledges financial supportfrom the Kieikai Research Foundation.

References

[1] Korhonen, H.; Pihlanto, A. Bioactive peptides: Production and function-ality. International Dairy Journal 2006, 16, 945-960.

[2] Semo, E.; Kesselman, E.; Danino, D.; Livney, Y.D. Caseinmicelle asa natural nano-capsular vehicle for nutraceuticals. Food Hydrocolloids2007, 21, 936-942.

[3] Livney, Y.D. Milk proteins as vehicles for bioactives. Current Opinion inColloid and Interface Science 2010, 15, 73-83.

[4] Shah, R.; Jana, A.H.; Aparnathi, K.D.; Prajapati, P.S. Process standard-ization for rennet casein based mozzarella cheese analogue. Journal ofFood Science and Technology 2010, 47, 574-578.

[5] Horne, D.S. Scaling behavior of shear moduli during the formation ofrennet milk gels. In Food Macromolecules; E. Dickinson, E.,Lorient, D.,Eds.; Royal Society of Chemistry, Cambridge, UK, 1995; 456-461.

[6] Mellema, M.; Heesakkers, J.W.M.; van Opheusden, J.H.J.; van Vliet, T.Structure and scaling behavior of aging rennet-induced casein gels ex-amined by confocal microscopy and permeametry. Langmuir 2000, 16,6847-6854.

[7] Esteves, C.L.C.; Lucey, J.A.; Pires, E.M.V. Mathematical modelling ofthe formation of rennet-induced gels by plant coagulants and chymosin.Journal of Dairy Research 2001, 68, 499-510.

[8] Udabage, P.; McKinnon, I.; Augustin, M.A. Effects of mineral salts andcalcium chelating agents on the gelation of renneted skim milk. Journalof Dairy Science 2001, 84, 1569-1575.

[9] Srinivasan, M.; Lucey, J.A. Effects of added plasmin on the formation andrheological properties of rennet-induced skim milk gels. Journal of DairyScience 2002, 85,1070-1078.

[10] Mellema, M.; Walstra, P.; van Opheusden, J.H.J.; van Vliet, T. Effects ofstructural rearrangements on the rheology of rennet-induced casein parti-cle gels. Advances in Colloid and Interface Science 2002, 98, 25-50.

[11] Lyndgaard, C.B.; Engelsen, S.B.; van den Berg, F.W. Real-time modelingof milk coagulation using in-line near infrared spectroscopy. Journal ofFood Engineering 2012, 108, 345-352.

[12] Tabayehnejad, N.; Castillo, M.; Payne, F. Comparison of total milk-clotting activity measurement precision using the Berridge clotting timemethod and a proposed optical method. Journal of Food Engineering2012, 108, 549-556.

[13] Djaowe, G.; Bitjoka, L.; Boukar, O.; Libouga, D.G.; Waldogo, B. Mea-surement of the rennet clotting time of milk by digital imagesequences(2D+t) processing. Journal of Food Engineering 2013, 114, 235-241.

[14] Al-Nabulsi, A.; Shaker, R.; Osaili, T.; Clark, S.; Harte, F.; Barbosa-Cnovas, G. Impact of high hydrostatic pressure and heat treatments onmilk gel properties: A comparative rheological study. International Jour-nal of Food Properties 2012, 15, 613-627.

[15] Catarino, I.; Martins, A.P.L.; Duarte, E.; Prudencio, E.S.; de Pinho, M.N.Rennet coagulation of sheep milk processed by ultrafiltration at low con-centration factors. Journal of Food Engineering 2013, 114,249-254.

[16] Morison, K. R.; Phelan, J. P.; Bloore, C. G. Viscosity and non-newtonianbehaviour of concentrated milk and cream. International Journal of FoodProperties 2013, 16, 882-894.

[17] Onwulata, C. I.; Thomas-Gahring, A. E.; Phillips, J. G.Physical proper-ties of mixed dairy food proteins. International Journal ofFood Properties2014, 17, 2241-2262.

[18] Taneya, S.; Izutsu, T.; Kimura, T.; Shioya, T.; Buchheim, W.; Kindstedt,P.; Pechak, D. Structure and rheology of string cheese. FoodStructure1992, 11, 61-71.

[19] Brown, J.A.; Foegeding, E.A.; Daubert, C.R.; Drake, M.A.; Gumpertz,M. Relationships among rheological and sensorial properties of youngcheeses. Journal of Dairy Science 2003, 86, 3054-3067.

[20] Lucey, J.A.; Johnson, M.E.; Horne D.S. Invited review:Perspectives onthe basis of the rheology and texture properties of cheese. Journal of DairyScience 2003, 86, 2725-2743.

[21] Gunasekaran, S.; Ak, M.M. Dynamic oscillatory shear testing of foods -selected applications. Trends in Food Science and Technology 2000, 11,115-127.

[22] Tunick, M.H. Small-strain dynamic rheology of food protein networks.Journal of Agricultural and Food Chemistry 2011, 59, 1481-1486.

[23] van Vliet, T. Rheological properties of filled gels. Influence of filler matrixinteraction. Colloid and Polymer Science 1988, 266, 518-524.

[24] Tung, C.Y.M.; Dynes, P.J. Relationship between viscoelastic propertiesand gelation in thermosetting systems. Journal of Applied Polymer Sci-ence 1982, 27, 569-574.

[25] Ross-Murphy, S.B. Rheological characterisation of gels. Journal of Tex-ture Studies 1995, 26, 391-400.

[26] Mleko, S.; Janas, P.; Pielecki, J. Study on stress relaxation behaviorof rennet cheeses at different temperature. Journal of Food Science andTechnology 2005, 42, 58-60.

[27] Ak, M.M.; Gunasekaran, S. Evaluating rheological properties of moz-zarella cheese by the squeezing flow method. Journal of Texture Studies1995, 26, 695-711.

[28] Fox, P.; Brodkorb, A. The casein micelle: Historical aspects, current con-cepts and significance. International Dairy Journal 2008, 18, 677-684.

[29] Dalgleish, D.G. On the structural models of bovine casein micelles -review and possible improvements. Soft Matter 2011, 7, 2265-2272.

[30] Dalgleish, D.G.; Corredig, M. The structure of the casein micelle of milkand its changes during processing. Annual Review of Food Science Tech-nology 2012, 3, 449-467.

[31] de Kruif, C.G.; Huppertz, T.; Urban, V.S.; Petukhov, A.V. Casein micellesand their internal structure. Advances in Colloid and Interface Science2012, 171-172, 36-52.

[32] Mezzenga, R.; Fischer, P. The self-assembly, aggregation and phase tran-sitions of food protein systems in one, two and three dimensions. Reportson Progress in Physics 2013, 76, 046601.

[33] Khan, S.U.; Pal, M.A. Paneer production: A review. Journal of Food Sci-ence and Technology 2011, 48, 645-660.

4