Journal of Food Engineering 92 (2009) 63–69

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier .com/locate / j foodeng

Physical characteristics during storage of soy yogurt made from ultra-highpressure homogenized soymilk

V. Ferragut *, N.S. Cruz, A. Trujillo, B. Guamis, M. CapellasCentre Especial de Recerca Planta de Tecnologia dels Aliments (CERPTA), XaRTA, XiT, Departament de Ciència Animal i dels Aliments, Facultat de Veterinària,Universitat Autònoma de Barcelona, 08193-Bellaterra, Spain

a r t i c l e i n f o

Article history:Received 31 May 2008Received in revised form 14 October 2008Accepted 22 October 2008Available online 5 November 2008

Keywords:SoymilkUltra-high pressure homogenization (UHPH)SyneresisViscoelasticityFirmnessColourMicrostructure

0260-8774/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.jfoodeng.2008.10.026

* Corresponding author.E-mail address: victoria.ferragut@uab.es (V. Ferrag

a b s t r a c t

The effects of ultra-high pressure homogenization (UHPH) on soymilk for producing soy yogurts and theirevolution during cold storage for 28 days (analysis were performed on days 1, 7, 14, 21 and 28) wereinvestigated. Soy yogurts were prepared from UHPH-treated soymilk preheated at 50 �C at 200 and300 MPa. Moreover a combined treatment at 300 MPa with a retention time of 15 s (at the temperaturereached after the high pressure valve) was investigated. Soymilk treated at 95 �C for 15 min (HT) wasused as control. This study included the evaluation of viscoelasticity by dynamic oscillation, puncturetest, microstructure by laser confocal microscopy, water holding capacity (WHC) determination and col-our evaluation. Results showed that soy yogurts from UHPH-treated soymilk presented higher values ofmechanical parameters related to firmness and G*, and better WHC. In addition, soy yogurts maintainedthese positive characteristics during cold storage. However, in UHPH soy yogurts colour parametersshowed some differences compared to control which may indicate changes in chemical compositionsin addition to the colloidal characteristics of these products.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Soymilk can be used as an inexpensive source of proteins com-pared to meat, and offers an interesting alternative when fer-mented. Studies about the incorporation of lactic acid bacteriainto soymilk attracted interest by consumers, resulting in the in-creased production of fermented soymilk during the last few years.Soymilk fermentation is produced in a similar way to milk fermen-tation by adding Streptoccocus thermophilus and Lactobacillus del-bruekii subsp. bulgaricus to the liquid food. Sugar fermentationcauses the pH to drop, modifying the surface charge of proteinswhich aggregate to build a network which entraps the liquid phaseand fat droplets. The demand for alternatives to yogurt producedfrom cow’s milk is growing due to some problems of milk proteinallergenicity, vegetarian alternatives, introduction of soy derivatesin the occidental diets, etc. Several studies (Chumchuere andRobinson, 1999; Donkor et al., 2007; Wang et al., 2003) reportedthat lactic acid bacteria fermentation provided an improved vola-tile profile to soymilk. In order to provide desirable body and tex-ture, soy yogurt is often fortified with different origin dry matter,especially milk proteins (Buono et al., 1990; Karleskind et al.,1991; Yadav et al., 1994). These studies used a traditional technol-ogy to obtain soymilk which was inoculated with lactic acid bacte-

ll rights reserved.

ut).

ria, but treatment conditions were quite extreme (121 �C for15 min, 95 �C for 15 min, 80 �C for 20 min) leading to importantchanges of soy components.

Ultra-high pressure homogenization (UHPH) is a technologywhich has demonstrated its potential benefit in the food industryas an alternative to conventional technologies, such as heat treat-ments. UHPH is based on conventional homogenization (40–50 MPa) but uses pressures from 100 to 350 MPa (Floury et al.,2004a). When liquid food passes through the high pressure valveit generates an increment of flow speed and a loss of pressure,bringing about cavitation, chisel effect, turbulence, and collisionof dispersed particles of liquid foods, such as fat droplets (Flouryet al., 2000, 2004b) and micro organisms, which may be destroyed(Diels et al., 2005; Hayes et al., 2005; Pereda et al., 2007). Severalauthors (Desrumaux and Marcand, 2002; Hayes and Kelly, 2003;Sandra and Dalgleish, 2005; Zamora et al., 2007) have studiedUHPH application to milk; some of them focusing on protein dena-turation, which is produced to a lesser extent than in heat treat-ments, and also focusing on fat globule size reduction, which isresponsible for the high stability of liquid foods. However, UHPHtreatment of soymilk has only been studied recently (Cruz et al.,2007). Results so far have shown, that this technology reduced soy-milk microbial load, lead to a certain degree of protein denatur-ation, and produced a highly physically stable product, not onlybecause of the reduction of fat globules size but also due to the so-lid particles in suspension.

64 V. Ferragut et al. / Journal of Food Engineering 92 (2009) 63–69

Studies on milk coagulation and yogurt stability produced fromUHPH (Serra et al., 2007; Serra et al., 2008) and more recently, onsoymilk coagulation (Cruz et al., 2008) have demonstrated the ade-quacy of this technology for this purpose, since gel characteristicsof the fermented soy products obtained are improved when com-pared to those obtained from conventionally treated milk or soy-milk. Considering these preliminary results (Cruz et al., 2008),the objective of this investigation was to study the evolution of rel-evant physical characteristics of soy yogurt related to quality stan-dards and their stability during cold storage using UHPHtechnology. Moreover, UHPH treatments applied were single orin combination with heat retention time (i.e. temperature reachedin the pressure valve was maintained during little time to enhancethe heat effect on the sample) to know whether the combination ofpressure and temperature improved the quality characteristics ofsoy yogurts.

2. Materials and methods

2.1. Soymilk elaboration and treatments

Soymilk used in this study was provided by Liquats Vegetals,S.A. (Girona, Spain). Hydrated (1:6 water:soybean w/w) and de-hulled soybean was milled in a colloidal mill (E. Bachiller B.S.A,Parets del Valles, Spain) at 80 �C followed by centrifugation to sep-arate the liquid phase. The soymilk obtained was then treated byconventional pasteurization at 95 �C for 15 min (HT). UHPH treat-ments were performed at 200 and 300 MPa at 50 �C inlet temper-ature and immediately cooled down. Other UHPH treatment at300 MPa, 50 �C inlet temperature was made with 15 s of retentiontime, maintaining the temperature reached at the high pressurevalve (�108 �C). Temperatures reached at the UHPH pressure valveat 200 and 300 MPa were �95 �C and �108 �C respectively. Esti-mated time for liquid passing through the valve was �0.07 s.

UHPH treatments were performed with a high pressure homog-enizer (model FPG11300, Stansted Fluid Power Ltd., Essex, UK).This device comprises a high pressure ceramic valve which is ableto support 350 MPa and a second pneumatic valve, located afterthe first one, able to support up to 50 MPa. Inlet and outlet temper-atures of soymilk were controlled by two tubular heat exchangers(Garvía, Barcelona, Spain) located before the machine entrance andafter the second homogenization valve, respectively. Inlet temper-ature (Ti), temperature after the first homogenization valve (T1), aswell as outlet temperature (To) reached after the pass through theheat exchanger were monitored in all productions.

2.2. Soymilk fermentation

Frozen commercial cultures mixture (DVS YF-3331, Chr. Han-sen, Horsholm, Denmark) of S. thermophilus and L. delbrueckii subsp.bulgaricus stored at �70 �C were thawed and diluted at 0.1% (v/v)in sterile peptone water just before inoculation, and culture wasinoculated at 0.03% (v/v) in samples heated at 45 �C in a water bathand incubated at the same temperature for 5 h. Multiple samples ofsoy yogurt were prepared for each day of analysis. Samples werestored at 4 �C until analysis at 1, 7, 14, 21 and 28 days of storage.

2.3. Water holding capacity (WHC)

Inoculated soymilk samples (30 g) were incubated in polypro-pylene centrifuge tubes (radius 32 mm, height 115 mm) until pH4.6 was reached and stored at 4 �C. To estimate the water holdingcapacity of soy yogurts, stored samples were centrifuged (Sigma4K15, Postfach, Germany) at 480g for 10 min at 20 �C, and the ex-pelled whey was weighed. WHC was expressed as the weight of

drained whey per gram of fermented sample. This measurementof water expulsion induced by centrifugation provides informationabout the water retention capacity in forced conditions.

On the other hand, the whey expelled spontaneously, that pro-vides information of the water retention capacity of the undis-turbed structure of the gel (100 g in plastic containers) wasweighed and expressed as the weight of drained whey per gramfermented sample.

2.4. Texture analysis

For textural analysis, soy yogurts were prepared in plastic con-tainers of 55 mm diameter. A Stable Micro System Texture Ana-lyzer (Model TA.TX2, Surrey, UK) equipped with a 25 mmcylindrical probe was used to perform a puncture test, operatingat a constant speed of 2.0 mm/s until a sample depth of 20 mm.The force-distance curves were performed at 4 �C, and fracturabil-ity and firmness were calculated from the texturometer software.

2.5. Viscoelasticity determination

Oscillatory testing of soy yogurt samples was performed at 4 �Cusing 20 mm parallel serrated plates, and a gap setting of 2.5 mm,in a rheometer ThermoHaake RS1 (Thermo- Haake GmbH, Kar-lsruhe, Germany). Gels were prepared in 50-ml syringes (1 syringeper day of storage), and cylindrical slices were obtained by pushingup the plunger of the syringe and cutting with a blade. The sliceswere transferred into the rheometer and, after the measuring sys-tem reached the testing position, sample edges were cut and left torelax for 5 min. To determine the viscoelastic properties of yogurt,frequency sweep tests were performed from 0.1 to 1 Hz at a max-imum strain of 0.01. Measurements were carried out within thelinear viscoelastic range, previously determined by a stress sweeptest from 0.001 to 1 Pa. Dynamic moduli (G0, G00 and G*) and tandwere calculated with the Rheowin software (ThermoHaake GmbH).G0 (the storage modulus) is related to the molecular events of elas-tic nature; G00 (the loss modulus) related to the viscous character ofmaterial; G* (complex modulus, corresponding to (G02+G002)1/2) andtand, which correspond to G”/G’ (values closer to 0 are more solid-like).

2.6. Colour evaluation

The colour of soy yogurts was measured with a Hunter Lab col-orimeter (MiniScan XETM, Hunter Associates Laboratory Inc., Res-ton, Virginia, USA) using the D65 illuminant with an angle ofobservation of 10�. Soy yogurt samples were tempered to 20 �C be-fore analysis, and thoroughly homogenized with a spatula beforeplacing into the measuring cup. A 10 mm-high ring was insertedinto the sample cup. The cup was then filled with the sample,and a white ceramic disk was then pushed down through the sam-ple until it rested on the top of the disk. The cup with the samplewas then placed on the instrument port for measurement and cov-ered with a black cup to prevent external light reaching thesample.

Colour L*, a*, b* coordinates were obtained, and difference be-tween each treatment and UHT and BP soymilks were calculatedas DE ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiDL2 þ Da2 þ Db2

p:

2.7. Microstructure

Confocal laser scanning microscopy observations were per-formed in fluorescence mode in a Leica TCS4D (Heidelberg, Ger-many). The images were obtained following the methoddescribed by Cruz et al. (2008) with an oil-immersion 60x lens ata wavelength of 488 nm. Inoculated soymilk samples (5 ml) were

V. Ferragut et al. / Journal of Food Engineering 92 (2009) 63–69 65

mixed with 3–4 drops of fluorescein isothiocyanate (FITC) and 4–5drops of Nile red in 0.5% of ethanol solution. Some drops werepoured on a concave glass slide, covered with a cover slip andsealed. Samples were incubated at 45 �C during 5 h and stored at4 �C for 24 h.

2.8. Statistical analysis

ANOVA was carried out to determine significant differences be-tween samples at the 5% level of probability, using the SPSS� Sys-tem for WINTM version 12.0.1 (SPSS Inc, Chicago, IL). The Tukey testwas used for comparison of data. Data were obtained from threeindependent experiments and each soymilk sample was analyzedin triplicate.

3. Results and discussion

3.1. Mechanical and microstructural characteristics

Microstructure of soy yogurts obtained by laser confocalmicroscopy revealed the interaction degree between fat and pro-tein and protein–protein fractions in the gel structure develop-ment. Images in Fig. 1 showed the protein fraction stained,which gave more information about the gel matrix. In fact, in HTsoy yogurt images the fat globules could be easily appreciated,even without using the fat stained. These soy yogurts revealedthe high number and size of fat globules which migrated to thegel surface during the fermentation process. However, in soy yo-gurts from UHPH-treated soymilk, fat globules were not easily vis-ible since they remained hidden inside the protein network. On theother hand, although protein network density in UHPH soy yogurtsis homogeneous, a more compact structure may be observed inthose obtained from 300 MPa than in 200 MPa soy yogurts. The300 MPa, 15 s soy yogurts presented a similar aspect but with aprotein matrix structure slightly more open, showing black holeswhere water was entrapped. In all cases, at the end of the storageperiod, a settlement of the structure was observed.

Frequency sweep tests were used to determine the viscoelasticcharacteristics of soy yogurts. G0 values were always higher thanG00, showing the predominating solid character of soy yogurts. Tandvalues were similar for all treatments applied to produce soy yo-gurts. During cold storage this parameter showed a small decrease,which was only significantly different (P > 0.05) on the last day ofstorage compared to the first day in 200 and 300 MPa soy yogurts.Mean values of tand for all treatments applied to soymilk rangedfrom 0.224 ± 0.005, on the first day, to 0.215 ± 0.007 at the end ofthe storage period when differences were more marked, indicatinga slight development of the gel structure.

The development of rigidity was estimated from G* values whichrepresent the total resistance of a material against the appliedstrain. G* values (Table 1) presented significant differences(P < 0.05) between all treatments applied to soymilk. HT soy yogurtsexhibited the lowest rigidity value while those obtained from UHPHtreatment of soymilk were higher, and this parameter increased aspressure application increased. The highest rigidity was shown bysoy yogurt treated at 300 MPa, 15 s. The evolution of G* during stor-age for 200 and 300 MPa did not show significant differences(P > 0.05) of gel rigidity at the different periods analysed, althougha slight increase was observed from day 1 to day 7. HT soy yogurtsand those obtained from 300 MPa, 15 s UHPH treatment showedthe lowest and highest G* values, respectively, and their evolutionduring storage showed more fluctuation in this parameter, indicat-ing less stabilization of the structure at the storage conditions. Aprevious study of the coagulation characteristics of UHPH-treatedsoymilk compared with that thermal-treated was performed by

Cruz et al. (2008). It was concluded that the different behaviour ob-served between UHPH and heat treated soy yogurts were due to themechanisms implied in the gel structure stabilization at refrigera-tion. In general, heat treatment favoured hydrophobic interactionsin the gel formation while UHPH treatments led to a gel whichwas stabilized by a predominance of hydrogen bonds, thus improv-ing gel characteristics in refrigerated conditions.

Firmness is defined as the force necessary to attain a givendeformation (Van Vliet et al., 1991). The lowest firmness values(Table 2) were shown by HT soy yogurts. The UHPH soy yogurtswere significantly (P < 0.05) firmer and this parameter increasedwith pressure application. No further firmness was developed inyogurts obtained from 300 MPa, 15 s treatment. In general, theevolution of firmness in soy yogurts during storage showed an in-crease 7 or 14 days after soy yogurt production and a further sta-bilisation in their values.

As mentioned above, the highest G* value was presented by soyyogurts obtained from 300 MPa, 15 s followed by 300 MPa and thatfollowed by 200 MPa. The lowest G* value was shown by HT soyyogurts. Mechanical characteristics determined by the puncturetest (gel firmness) had a similar tendency except in the case ofsoy yogurt obtained from 300 MPa, 15 s, in which no longer in-crease of firmness was caused by the additional effect of tempera-ture (retention time) to the UHPH treatment applied. Thisdifference in the mechanical parameters of G* and firmness canbe attributed to the different magnitude of force applied in themechanical evaluation. G* is obtained from a test in which struc-ture of the network is not broken, thus its sensitivity for detectingthe contribution of different interactions is higher than in thepuncture test, in which macroscopic failure of the structure isexperienced. The effect of UHPH treatment on physical propertiesof milk (Hayes and Kelly, 2003; Pereda et al., 2007) and soymilk(Cruz et al., 2007) has demonstrated that the main changes ob-served in these products involve fat globules size reduction. More-over, fat globules are well protected by proteins, and a certaindegree of protein denaturation is caused depending on the magni-tude of pressure applied (Cruz et al., 2007). These colloidal struc-tures are responsible for the network characteristics of gelswhich led to a protein network with the fat globules entrappedand homogeneously distributed. These characteristics produced ahigh development of rigidity and firmness. On the other hand,the highest rigidity developed by 300 MPa, 15 s soy yogurts couldbe attributed to an additional effect of temperature with the UHPHtreatment, which caused protein denaturation contributing to en-hance hydrophobic interactions. Thus, stabilisation of 300 MPa,15 s was due to a higher contribution of hydrophobic interactionthan in single UHPH treatments but also by a high contributionof hydrogen bonds, which are characteristics of the refrigeratedsoy yogurts produced by UHPH-treated soymilk (Cruz et al.,2008). In this respect, after the main gel network is formed duringthe fermentation process, the final stabilisation of gel structure isreached in the first hours after soy yogurts are submitted to lowtemperature. These conditions led to the fat solidification and theestablishment of interactions between particles and molecules,especially those which are favoured at low temperatures, i.e.hydrogen bonds. However, hydrophobic interactions are favouredat higher temperatures during the soy yogurt fermentation(45 �C). This is why soy yogurt produced from heat treatment(95 �C, 15 min) presented the lowest G* values and firmness char-acteristics since, once the gel has been developed, the contributionto stabilisation of soy yogurt structure in cold conditions is mainlydue to hydrogen bonds which in heat treated soymilk are less plen-tiful than in UHPH-treated soymilk. Microstructure of soy yogurtson day 1 (Fig. 1) is in accordance to mechanical characteristics de-scribed. A protein–fat mixed and homogeneous network wasdeveloped in UHPH soy yogurts, which is associated to the

Fig. 1. Confocal scanning light microscopy showing the structure of soy yogurt elaborated from soymilks treated: (a, b) 95 �C 15 min; (c, d) 200 MPa 50 �C; (e, f) 300 MPa50 �C and (g, h) 300 MPa 50 �C, 15 s.

66 V. Ferragut et al. / Journal of Food Engineering 92 (2009) 63–69

mechanical behaviour previously described. However, in soy yo-gurts from heat treated soymilk, in addition to the reasons men-

tioned before; fat globules could act as disruptors of the networkhomogeneity, acting against strong mechanical characteristics.

Table 1Mean values (±SD) of G* (Pa) during cold storage of soy yogurt elaborated from UHPH and heat treated soymilk.

Treatments Storage (days)

1 7 14 21 28

95 �C 15 min 1527 ± 084dz 1676 ± 096dy 1763 ± 068dxy 1792 ± 092dx 1786 ± 093dx

200 MPa 50 �C 2003 ± 031cx 2120 ± 185cx 2095 ± 100cx 2067 ± 176cx 2148 ± 146cx

300 MPa 50 �C 2251 ± 125bx 2293 ± 134bx 2286 ± 157bx 2385 ± 236bx 2363 ± 078bx

300 MPa 50 �C, 15 s 2814 ± 087ax 2542 ± 078az 2671 ± 048ay 2627 ± 082ayz 2891 ± 129ax

a–d Different superscript in the same column are significantly different (p < 0.05). x–z Different superscript in the same row are significantly different (p < 0.05).

Table 2Mean (±SD) values of firmness (N) during cold storage of soy yogurt elaborated from UHPH and heat treated soymilk.

Treatments Storage (days)

1 7 14 21 28

95 �C 15 min 1.33 ± 0.24cy 1.53 ± 0.30cx 1.60 ± 0.33cx 1.67 ± 0.38cx 1.55 ± 0.35cx

200 MPa 50 �C 1.66 ± 0.20by 1.97 ± 0.34bx 1.95 ± 0.35bx 2.00 ± 0.37bx 2.05 ± 0.26bx

300 MPa 50 �C 2.11 ± 0.23ayz 2.04 ± 0.38bz 2.27 ± 0.37axy 2.40 ± 0.35ax 2.19 ± 0.44abxyz

300 MPa 50 �C, 15 s 2.10 ± 0.23ay 2.40 ± 0.21ax 2.29 ± 0.22ax 2.31 ± 0.24ax 2.35 ± 0.38ax

a–c Different superscript in the same column are significantly different (p < 0.05). x–z Different superscript in the same row are significantly different (p < 0.05).

V. Ferragut et al. / Journal of Food Engineering 92 (2009) 63–69 67

3.2. Water holding capacity

Water holding capacity (WHC) was estimated by two differentways: by spontaneous syneresis (Table 3) and by centrifugation(Fig. 2). In the first case, no significant differences (P > 0.05) wereobserved in the spontaneous expulsion of water phase comparedto the treatment applied to soymilk and, in general, statistical dif-ferences did not reveal changes during the storage period for anyindependent treatment applied. Only soy yogurts obtained at300 MPa, 15 s showed a slight increase in the spontaneous synere-sis on day 28.

Expulsion of aqueous phase induced by centrifugation, how-ever, showed different results in contrast to spontaneous syneresis.Expulsed water was significantly lower (P < 0.05) in soy yogurtsobtained by UHPH than in those treated by conventional heattreatment and, this effect was more marked when increasing pres-sure application. However, treatment at 300 MPa, 15 s did not pro-duce further water retention when compared with soy yogurtsobtained from only 300 MPa. The evolution during storage of alltreatments applied led to stable soy yogurts in terms of WHC as,in general, no significant differences (P > 0.05) were observed dur-ing the storage period for all treatments applied, except for HT soyyogurts and those obtained from 200 MPa, in which the WHC ofsoy yogurts on day 28 were significantly higher (P < 0.05) thanthose of day 1 of storage.

All microstructure aspects mentioned before are in close relationto the water holding capacity, which is a relevant quality parameterof food gels since syneresis is an undesirable defect in this kind ofproducts. As mentioned before, spontaneous syneresis was low,even when compared to syneresis reported by Kovalenko and Brig-gs (2002) for soy yogurts obtained from soymilk treated at 95 �C,15 min, which was in the range of 4 to 15%. Results indicated that,

Table 3Mean (±SD) values of syneresis espontaneous (%) during cold storage of soy yogurt elabor

Treatments Storage (days)

1 7

95 �C 15 min 1.42 ± 0.67az 1.79 ± 0.20ay

200 MPa 50 �C 1.64 ± 0.52ay 1.88 ± 0.46axy

300 MPa 50 �C 1.34 ± 0.65az 1.69 ± 0.41ay

300 MPa 50 �C, 15 s 1.75 ± 0.40az 1.74 ± 0.37az

a–b Different superscript in the same column are significantly different (p < 0.05). w–z Di

although low, the main water expulsion of water from the gel net-work is produced during the first 24 h when gel structure is stabi-lizing in cold conditions, and a slight settlement of gel isproduced during the cold storage, as can also be seen in micropho-tographs at 28 days storage period. This implies the expulsion ofaqueous phase due to a small gel retraction caused by interactionestablishment of particles and molecules during storage.

The expulsion of aqueous phase induced by centrifugationshowed the improved WHC of UHPH compared to heat treatment.This parameter, in general, was similar throughout storage and dif-ferences were only found between day 1 and day 28, except in soyyogurt from 300 MPa, 15 s, which presented a high stabilized val-ued of WHC throughout the storage period. It can also be observedthat increasing pressure led to an increase of WHC and, that300 MPa, 15 s was the best treatment for this purpose.

Soy protein gels with a higher number of physicochemicalbonds (or more developed structure) have a better ability to entrapwater into the network (Kovalenko and Briggs, 2002). In UHPH soyyogurts, in addition to the protein–protein interaction producedduring gel formation, the small fat globules which were very wellprotected by protein could act as particles which, in terms of gelbuilding, behaved as proteins, increasing the surface area of colloi-dal structures to the gel development (Serra et al., 2007; Cruz et al.,2007) and hence, entrapping water.

3.3. Colour evaluation

The evolution of colour coordinates L*, a*, b* during storage arereported in Table 4. L* (lightness) values of HT soy yogurts weresignificantly lower (P < 0.05) than those from UHPH on day oneof storage. However, UHPH treatments did not show any differ-ences between them in this parameter. The evolution of L* during

ated from UHPH and heat treated soymilk.

14 21 28

2.01 ± 0.40axy 1.92 ± 0.28bxy 2.12 ± 0.32bx

2.09 ± 0.30ax 2.10 ± 0.48abx 2.15 ± 0.33bx

1.80 ± 0.50axy 1.96 ± 0.58bwx 2.13 ± 0.36bw

2.04 ± 0.23ayz 2.35 ± 0.48ayx 2.49 ± 0.22ax

fferent superscript in the same row are significantly different (p < 0.05).

cx

cxy

bx

ax

cxbcx

axy

bxy

dxcy

axy

bxy bxbxbxy

axy

bxbxby

ay

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

95ºC 15 min 200 MPa 50ºC 300 MPa 50ºC 300 MPa 50ºC. 15 s

Exp

elle

d w

ater

(%

)

1 7 14 21 28

Treatment

Storage (days)

Fig. 2. Expelled water values (%) during cold storage of soy yogurt elaborated from UHPH and heat treated soymilk. Letters a–d compares differences between treatments,w–z compare differences of each treatment during storage. Different superscript letters indicate significant difference, (p < 0.05).

Table 4Mean (±SD) values of L*, a*, b* colour coordinates and colour difference (DE) of soy yogurt elaborated from UHPH and heat treated soymilk.

Coordinates Storage (days) Treatments

95 �C 15 min 200 MPa 50 �C 300 MPa 50 �C 300 MPa 50 �C, 15 s

L* 1 85.45 ± 0.37bz 86.41 ± 0.10ax 85.99 ± 0.78axy 86.10 ± 0.23ax

7 86.28 ± 0.16a 85.73 ± 0.33by 85.40 ± 0.52cx 85.77 ± 0.40bxy

14 86.59 ± 0.18ax 84.95 ± 0.50bz 85.47 ± 0.40bx 85.17 ± 0.80byz

21 86.50 ± 0.09axy 84.89 ± 0.39bz 85.30 ± 0.37bx 85.23 ± 1.22byz

28 86.24 ± 0.05ay 85.17 ± 0.06bz 84.61 ± 0.27by 84.95 ± 1.13bz

a* 1 0.14 ± 0.01cz 0.43 ± 0.07az 0.25 ± 0.03bz 0.31 ± 0.02bz

7 0.33 ± 0.03dxy 1.19 ± 0.02by 0.81 ± 0.04cy 1.37 ± 0.02ay

14 0.34 ± 0.02cxy 1.44 ± 0.03aw 1.39 ± 0.03bw 1.37 ± 0.01by

21 0.31 ± 0.01dy 1.30 ± 0.02bx 1.16 ± 0.17cx 1.47 ± 0.03ax

28 0.36 ± 0.03bx 1.37 ± 0.01ax 1.38 ± 0.02aw 1.38 ± 0.01ay

b* 1 14.25 ± 0.04ax 11.83 ± 0.46bx 11.34 ± 0.16cw 11.25 ± 0.10cx

7 13.60 ± 0.27az 8.54 ± 0.56dy 9.12 ± 0.07cx 10.89 ± 0.03bxy

14 14.19 ± 0.51axy 8.28 ± 0.27dy 8.98 ± 0.47cx 10.52 ± 0.16by

21 13.93 ± 0.56axy 7.33 ± 0.05dz 8.47 ± 0.20cy 10.56 ± 0.07by

28 13.94 ± 0.63axy 7.41 ± 0.05cz 7.03 ± 0.27cz 9.34 ± 0.47bz

DE 1 – 2.62 ± 0.47bz 2.95 ± 0.16abz 3.07 ± 0.05az

7 – 5.17 ± 0.57ay 4.61 ± 0.15by 2.97 ± 0.07cz

14 – 6.24 ± 0.38ax 5.44 ± 0.52bx 4.13 ± 0.08cy

21 – 6.87 ± 0.14ax 5.67 ± 0.16bx 3.93 ± 0.22cy

28 – 6.70 ± 0.0 7.17 ± 0.23aw 4.97 ± 0.66bx

a–d Different superscript in the same rows are significantly different (p < 0.05). w–z Different superscript in the same columns are significantly different (p < 0.05).

68 V. Ferragut et al. / Journal of Food Engineering 92 (2009) 63–69

storage varied arbitrarily from period to period, showing in somecases differences between the first day and day 28. However, inall cases differences observed were approximate of 1 unit of L*

value, which is probably undetected by consumers. Compared withUHPH-treated soymilk (Cruz et al., 2007), in which lightness de-creased with pressure application due to the particles dispersion,soy yogurt solid state made that light reflection on the gel surfacecaused higher values of lightness. Moreover, the fact that UHPHsoy yogurt presented higher L* values than HT soy yogurts is prob-ably due to the higher homogeneity of the gel obtained by UHPHwhich led to a more compact surface.

The a* (red–green axis) values were positive in all cases imply-ing that they were in the red colour space. However, UHPH and

heat treated soymilk (Cruz et al., 2007) presented negative contri-butions of a* (green) to the colour of soymilk. Although the soy yo-gurts appeared white to the human eye, the colorimeter detectedthe red shade contribution. In all cases, a* contribution to the col-our of UHPH soy yogurts were higher than that showed by heattreatment, and during storage this parameter showed an increasein all cases, specially from day 1 to day 7 and from the latter today 14.

The b* (yellow–blue axis) values were also positive in all caseswith the yellow contribution to match the soy yogurt colour. Thehighest significant value of this coordinates corresponded to thesoy yogurt from heat treatment (P < 0.05) which also showedthe most stable values during storage. The b* values of UHPH soy

V. Ferragut et al. / Journal of Food Engineering 92 (2009) 63–69 69

yogurts decreased significantly during storage, showing the biggestdifference between day 1 and day 7.

Colour difference for each treatment between day 1 and day 28gave an idea of stability of overall colour changes during storage.The most stable soy yogurt in terms of colour difference was thatobtained from heat treated soymilk (DE = 0.88 ± 0.11), followedby 300 MPa, 15 s (DE = 2.44 ± 0.59), and with similar values, 200and 300 MPa soy yogurts (DE = 4.67 ± 0.33). On the other hand,since HT soy yogurt is the standard product to consumers, we cal-culated the colour difference of each UHPH treatment against HTsoy yogurt (Table 4). As storage advanced, colour differences in-creased, mainly due to the different b* values showed by UHPHsoy yogurts compared to the HT soy yogurts. The lowest colour dif-ference was shown by 300 MPa, 15 s soy yogurts. These resultsmay be due to chemical changes produced in UHPH compared toheat treatment, since the combination of UHPH with a small reten-tion time at high temperature (i.e. around 108 �C) influenced col-our characteristics, leading to colour parameters more similar tothe conventional heat treatment. This is an aspect which is beingstudied, especially changes in oxidation products which could beresponsible for the colour changes observed in UHPH soy yogurts.Once the complete study is finished we will be able to select thebest UHPH conditions for soy yogurt elaboration. At the same time,sensory analysis will be performed to find out the acceptability ofUHPH soy yogurt. At the moment, some preliminary sensory trialsperformed by the Food Technology Department personnel have ledus to study the application of UHPH for soy yogurt elaboration.

4. Conclusion

In the present study, all of the UHPH treatment of soymilk pro-duced set-type soy yogurts with improved mechanical characteris-tics and water holding capacity when compared with yogurtsmade from soymilk using conventional heat treatment (95 �C,15 min). These results are in accordance with the microstructureobservations by laser confocal microscopy, which showed a morehomogeneous and compact network structure of the UHPH soy yo-gurts. Moreover, evolution of soy yogurt during cold storage didnot produce important changes in any treatment, which ensuredthe maintenance of typical physical characteristics of this productduring at least 28 days.

However, some colour differences between the conventionalheat treatment with each of the UHPH treatments applied to soy-milk were detected. This could indicate chemical differences ofsoymilk due to the treatment, and it is being studied to evaluatetheir origin and possible impact of the UHPH technology on finalcomposition of the product. In any case, UHPH treatment in com-bination with 15 s at high temperature (300 MPa, 15 s) has shownin this study to be the best condition for obtaining soy yogurt interms of mechanical properties, water holding capacity and colourcharacteristics. This assumption should be contrasted by sensoryanalysis.

Acknowledgements

This study was supported by the Ministerio de Educación yCiencia (AGL2003-03494), the Commission of the European Com-munities Union (CRAFT Project 512626), and a predoctoral fellow-ship from Programa de Mejoramiento a Profesorado (PROMEP) bySecretaria de Educación Pública (SEP) and Universidad Autónoma

Del Estado de Hidalgo of México. The confocal microscopy was per-formed by Servei de Microscòpia of the Universitat Autònoma deBarcelona.

References

Buono, M.A., Setser, C., Erickson, L.E., Fung, D.Y.C., 1990. Soymilk yogurt: sensoryevaluation and chemical measurement. Journal of Food Science 55 (2), 528–531.

Chumchuere, S., Robinson, R.K., 1999. Selection of starter cultures for thefermentation of soya milk. Food Microbiology 16 (2), 129–137.

Cruz, N., Capellas, M., Hernández, M., Trujillo, A.J., Guamis, B., Ferragut, V., 2007.Ultra-high pressure homogenization of soymilk: microbiological,physicochemical and microstructural characteristics. Food ResearchInternational 40 (6), 725–732.

Cruz, N.S., Capellas, M., Jaramillo, D.P., Trujillo, A.J., Guamis, B., Ferragut, V., 2008.Soymilk treated by ultra high pressure homogenization: acid coagulationproperties and characteristics of a soy yogurt product. Food Hydrocolloids.doi:10.1016/j.foodhyd.2008.03.010.

Desrumaux, A., Marcand, J., 2002. Formation of sunflower oil emulsions stabilizedby whey proteins with high pressure homogenization (up to 350 MPa): effect ofpressure on emulsion characteristics. International Journal of Food Science andTechnology 37 (3), 263–269.

Diels, A.J., Callewaert, L., Wuytack, E., Masschalck, B., Michiels, C., 2005. Inactivationof Escherichia coli by high pressure homogenization is influenced by fluidviscosity but not by water activity and product composition. InternationalJournal of Food Microbiology 101 (3), 281–291.

Donkor, O., Henriksson, A., Vasiljevic, T., Shah, N.P., 2007. a-Galactosidase andproteolytic activities of selected probiotic and dairy cultures in fermentedsoymilk. Food Chemistry 104 (1), 10–20.

Floury, J., Desrumaux, A., Lardières, J., 2000. Effect of high pressure homogenizationon droplet size distributions and rheological properties of model oil-in-wateremulsions. Innovative Food Science and Emerging Technologies 1 (2), 127–134.

Floury, J., Bellettre, J., Legrand, J., Desrumaux, A., 2004a. Analysis of a new type ofhigh pressure homogeniser. A study of the flow pattern. Chemical EngineeringScience 59 (4), 843–853.

Floury, J., Legrand, J., Desrumaux, A., 2004b. Analysis of a new type of high pressurehomogeniser. Part B. study of droplet break-up and recoalescence phenomena.Chemical Engineering Science 59 (6), 1285–1294.

Hayes, M.G., Fox, P.F., Kelly, A.L., 2005. Potential applications of high pressurehomogenization in processing of liquid milk. Journal of Dairy Research 72 (1),25–33.

Hayes, M.G., Kelly, A.L., 2003. High pressure homogenization of raw whole bovinemilk (a) effects on fat globule size and other properties. Journal of DairyResearch 70 (3), 297–305.

Karleskind, D., Laye, I., Halpin, E., Morr, C.V., 1991. Improving acid production insoy-based yogurt by adding cheese whey proteins and mineral salts. Journal ofFood Science 56 (4), 999–1001.

Kovalenko, I.V., Briggs, J.L., 2002. Textural characterization of soy-based yogurt bythe vane mehod. Journal of Texture Studies 33 (2), 105–118.

Pereda, J., Ferragut, V., Quevedo, J.M., Guamis, B., Trujillo, A.J., 2007. Effects of ultra-high pressure homogenization on microbial and physicochemical shelf life ofmilk. Journal of Dairy Science 90 (3), 1081–1093.

Sandra, S., Dalgleish, D.G., 2005. Effects of ultra-high pressure homogenization andheating on structural properties of casein micelles in reconstituted skim milkpowder. International Dairy Journal 15 (11), 1095–1104.

Serra, M., Trujillo, A.J., Quevedo, J.M., Guamis, B., Ferragut, V., 2007. Acid coagulationproperties and suitability for yogurt production of cows’ milk treated by highpressure homogenization. International Dairy Journal 17 (7), 782–790.

Serra, M., Trujillo, A.J., Guamis, B., Ferragut, V., 2008. Evaluation of physicalproperties during storage of set and stirred yogurts made from ultra-highpressure homogenization-treated milk. Food Hydrocolloids. doi:10.1016/j.foodhyd.2007.11.015.

Van Vliet, T., Van Dijk, H.J.M., Zoon, P., Walstra, P., 1991. Relation between syneresisand rheological properties of particle gels. Colloid and Polymer Science 269 (6),620–627.

Wang, Y., Yu, R., Yang, H., Chou, C., 2003. Sugar and acid contents in soymilkfermented with lactic acid bacteria alone or simultaneously with bifidobacteria.Food Microbiology 20 (3), 333–338.

Yadav, V.B., Jha, Y.K., Garg, S.K., Mital, B.K., 1994. Effect of soymilk supplementationand additives on sensory characteristic and biochemical-changes of yogurtduring storage. Australian Journal of Dairy Technology 49 (1), 34–38.

Zamora, A., Ferragut, V., Jaramillo, P.D., Guamis, B., Trujillo, A.J., 2007. Effects ofultra-high pressure homogenization on the cheese-making properties of milk.Journal of Dairy Science 90 (1), 13–23.