lable at ScienceDirect

Food Hydrocolloids 35 (2014) 484e493

Contents lists avai

Food Hydrocolloids

journal homepage: www.elsevier .com/locate/ foodhyd

Rheology, texture and microstructure of gelatin gels with and withoutmilk proteins

Zhihua Pang a,*, Hilton Deeth a, Peter Sopade b, Ranjan Sharma c, Nidhi Bansal a,*a School of Agriculture and Food Sciences, The University of Queensland, St. Lucia, Qld. 4072, AustraliabCentre for Nutrition and Food Sciences, Queensland Alliance for Agriculture and Food Innovation, University of Queensland, St. Lucia, Qld. 4072, AustraliacDairy Innovation Australia Limited, Werribee, Vic. 3030, Australia

a r t i c l e i n f o

Article history:Received 9 May 2013Accepted 8 July 2013

Keywords:GelatinGelationMilk proteinsRheologyTextureMicrostructure

* Corresponding authors.E-mail address: n.bansal@uq.edu.au (N. Bansal).

0268-005X/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.foodhyd.2013.07.007

a b s t r a c t

The effects of gelatin concentration, pH and addition of milk proteins on the physical and microstructuralproperties of type B gelatin gels were studied by small deformation rheology, texture analysis andscanning electron microscopy. Whey protein isolate (WPI), milk protein concentrate (MPC) and skimmilk powder (SMP) were used as sources of milk proteins. The elasticity of gelatin gels was significantlyaffected by the concentration of gelatin. Higher gelatin concentrations led to a stronger gel, and highergelling and melting temperatures. However, all the gelatin gels at concentrations from 1.0 to 5.0% meltedbelow human body temperature. Rheological properties of gelatin gels were independent of pH in therange pH 4.6e8.0. At pH 3.0 gelation of gelatin was significantly inhibited. Addition of SMP and MPCsignificantly enhanced the rheological properties of gelatin gels, while addition of WPI had a negativeeffect on them. However, the effect of addition of milk proteins was dependent on the gelatin concen-tration. Textural results showed that addition of all milk powders increased the hardness of gelatin gelsat high gelatin concentration (5.0%). The fracturability of the gels was greatly influenced by pH. Additionof milk proteins and high gelatin concentration (5.0%) both caused loss of gel fracturability. Micro-structural results showed that gelatin concentration and pH had a marked influence on the gel structure,and the addition of MPC and SMP changed the structure of the gelatin gels; a structure similar to puregelatin gel was observed after addition of WPI.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Gelatin is an animal protein produced from collagen (Boran,Mulvaney, & Regenstein, 2010). It has high flexibility of the poly-peptide chains and a non-random occurrence of imino acids (i.e.,proline or hydroxyproline) in its sequence which is unusual amongthe gel-forming agents (Karim & Bhat, 2009). The intermolecularcontacts in gelatin gels are hydrogen bonds, which make the gelsthermally reversible. Specifically, a gelatin gel melts below humanbody temperature, which gives it the well-known “melt-in-mouth”property (Djabourov, 1988). These unique properties make gelatinan important and widely used biopolymer in the food industry.However the properties of gelatin gels are affected by factors suchas pH and concentration. Gelation of a gelatin solution and subse-quent changes in the gel network arise through the partial return ofdisordered gelatin molecules (coil) to the collagen-like structure(polyproline II helix) (Djabourov, Lechaire, & Gaill, 1993). It has

All rights reserved.

been reported that at low gelatin concentration, three regions ofthe helix may be derived from one chain to give an intramolecularcollagen-like structure which makes no contribution to the gelnetwork. At higher gelatin concentrations, the three regions of thehelix can come from two or three different chains, so that usefuljunction zones that induce gelation can be formed (Djabourov,1988). Although gelatin provides stable gels over a very widerange of pH values, pH should still be considered in gelatin gelation.pH can greatly affect the viscosity of gelatin solutions, which isminimum at its isoionic point (IP) because of the maximum mo-lecular folding at that pH (Petrie and Becker, 1970). It was reportedthat aggregation of gelatin type A (IP 9.0) increased and the gelatingel turned from transparent to opaque as the pH was increasedfrom 5.4 to 7.5 (Walkenstrom & Hermansson, 1997). Gelation ofboth gelatin type A and B is inhibited greatly and the gel strength islow outside the pH range 4.0e10. This is attributed to strongelectrostatic forces that inhibit the ability of chains to form junctionzones (Bello, Vinograd, & Bello, 1962).

Milk proteinegelatin mixtures are widely used in food products,as they play an essential role in the texture and stability of many

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493 485

food systems. Gelatin is also used widely for modifying the textureand shelf-life of dairy-based foams, gels, dispersions and emulsions(Hemar, Liu, Meunier, & Woonton, 2010; Koh, Merino, & Dickinson,2002). The interaction between milk proteins and food hydrocol-loids has been reviewed extensively (Dickinson, 1998; Lal,O’Connor, & Eyres, 2006; Syrbe, Bauer, & Klostermeyer, 1998). Hy-drocolloids can be classified as non-ionic and anionic, which de-termines the behaviour of proteinehydrocolloid solutions (Syrbeet al., 1998). Gelatin is expected to interact with milk proteins atpH values where the two polymers carry opposite charges. GelatinA (IP 9.0) interacts with the oppositely charged micellar casein atpH 6.7, while gelatin B (IP 5.0) does not (Lefebvre & Antonov, 2001).However, in the gelatin Aewhey proteins system, no interactionhas been observed at pH 4.6 or 5.4 (Walkenstrom & Hermansson,1994). Therefore, the milk protein type plays an important role inmixed system. In most previous studies of milk proteinegelatinsystem, milk proteins have been denatured either by heating oracidification, which leads to gelation of the milk proteins (Fiszman& Salvador, 1999; Koh et al., 2002; Walkenstrom & Hermansson,1996).

The aim of this study was to investigate the effects of pH, con-centration and addition of milk proteins on the gelling behavioursof type B gelatin. Small deformation rheology, texture analysis andmicroscopy were used to investigate the properties of the gels. Inthis study, gelatin was the only gelling agent in the systems used toinvestigate the effect of milk proteins on gelation properties ofgelatin.

2. Materials and methods

2.1. Materials

The gelatin used in this study was supplied by Gelita (Beau-desert, Australia). It was a light coloured edible bovine skin (type B)gelatin powder with bloom 200 and mesh 20, which is a com-mercial product commonly used in the food industry. The milkpowders, whey protein isolate (WPI), milk protein concentrate(MPC) and skim milk powder (SMP) were obtained from MurrayGoulburn Co-Operative Ltd (Melbourne, Australia). The proteincontents of WPI, MPC and SMP were 90.2, 85.0 and 33.3%, respec-tively (information provided by supplier).

2.2. Methods

2.2.1. Sample preparationSolutions with three concentrations (1.0, 2.5 and 5.0%, w/v) of

gelatin were prepared by allowing the gelatin to swell in distilledwater overnight (about 15 h) followed by heating at 45 �C for30 min to dissolve it. Then 1 M NaOH or HCl was used to adjust thepH to 3.0, 4.6, 5.3, 6.6 or 8.0. In mixed gels, the milk protein con-centration used was 4.5% (w/w), which was obtained by adding theappropriate amounts of WPI, SMP or MPC. Milk powders andgelatin were dissolved together in distilled water overnight fol-lowed by heating at 45 �C for 30 min. The pH was then adjusted to6.6 and 8.0 for gels containing MPC and SMP, since MPC and SMPeasily form aggregates at pH < 5.3, and to 3.0 to 8.0 for WPI-containing gels.

2.2.2. Small deformation rheological measurementDynamic oscillatory measurements were performed on a stress-

controlled rheometer (Model AR-G2, TA Instruments, Elstree, UK).Test samples were poured at about 45 �C onto the bottom plate ofthe rheometer, and a cone (4 cm, diameter; 2� angle) and plategeometry was used. A strain sweep test revealed that 0.5% strain at1 Hz frequency was within the linear viscoelastic region (LVR) for

the samples. The measurements were carried out in a three-stageprocess (Salvador & Fiszman, 1998):

a. Cooling: equilibration at 40 �C and a temperature sweep to10 �C at a cooling rate of 1 �C/min to promote gelatin gelformation.

b. Annealing: a time sweep at 10 �C for 2.5 h to observe thematuration of the gelling samples.

c. Heating: a temperature sweep from 10 to 40 �C at a heating rateof 1 �C/min to observe melting of gelatin gels.

The gelling (TG) and melting (TM) temperatures were calculatedwhen there were appreciable increases and decreases, respectively,in complex viscosity (h*), and two values were obtained for eachtemperature to calculate the average gelling and melting temper-atures. The complex viscosity, h* was defined as in Eq. (1):

h* ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiG02 þ G002=u

q(1)

where, G0 ¼ storagemodulus, G00 ¼ loss modulus and u¼ frequency.Following the procedures in Sopade, Halley, and Junming (2004):

i. The cross-over temperature was obtained when G00 equals G0

or loss tangent, which is the ratio of G00 to G0, equals to 1.ii. Temperature of maximum or minimum change in complex

viscosity per unit change in temperature. This was defined asthe point of inflection. It was obtained by differentiating thecomplex viscosity with respect to temperature (first deriva-tive, dh*/dT) and finding the temperature when the derivativewas zero (¼0).

All rheological measurements were performed in duplicate andthe samples were randomised for the analysis.

2.2.3. Texture analysisTexture measurements were performed using a TAeXT2 Texture

Analyser (Godalming, Surrey, UK). Samples were transferred to anincubator at 10 �C after the pH was adjusted, and kept for 2.5 hbeforemeasurement. All measurements were carried out at 10 �C intriplicate. The probe used was cylindrical with a flat base of12.7 mm diameter, operating at a speed of 1 mm/s. The sampleheight was 30 mm in a cylindrical container of about 40 mm. Theprobe penetrated the gel during a total displacement of 10 mm.Two parameters were obtained from the forceetime curves: (a)fracturability (N/mm), defined as the force at the first significantbreak in the curve; (b) firmness (N/mm), defined as the initial slopeof the penetration curve within the first 2 s (Fiszman & Salvador,1999).

2.2.4. MicrostructureGels were formed in the same way as for texture analysis. Gels

were cut into small pieces (w1 mm3) and fixed with 2.5% (v/v)glutaraldehyde in 0.1 M phosphate buffer (pH 6.8), dehydrated inethanol with a serial concentration of 50, 70, 90 and 100% (v/v) anddried with a CO2 critical point dryer (Tousimis Automatic, Rockville,USA) prior to mounting on aluminium stubs and sputter-coatedwith a Baltek platinum coater. The microstructure of the gels wasexamined using a scanning electron microscope (JEOL 6610, Tokyo,Japan) at an acceleration voltage of 10 kV.

2.2.5. Statistical analysisMinitab ver. 16 software (Minitab Inc., USA) was used for anal-

ysis of variance (ANOVA), test of significance (p < 0.05).

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493486

3. Results and discussion

3.1. Rheology results

Fig. 1 shows representative rheological results for a 2.5% gelatingel at pH 6.6 through the three stages (cooling, annealing andheating). G0 increased rapidly and surpassed G00 during the coolingstage, which indicated that the system gelled (Fig. 1A). When thetemperature was held at 10 �C for 2.5 h during annealing, both G0

and G00 showed an increase (Fig. 1B), with G0 increasing more thanG00. After around 40 min, the slope of the curve decreased andgelation became slow and stable, but gelation was not complete inthe 2.5 h. Similar observations were made by Joly-Duhamel, Hellio,Ajdari, and Djabourov (2002). From optical rotation measurements,these authors found that the helix concentration of gelatinincreased with the annealing time and did not reach 100% withinthe “normal time scale” of observation. During the heating stage, G0

decreased quickly and became lower than G00 (Fig. 1C), whichindicated melting of the gel. Similar trends for G0 and G00 wereobserved for samples at different concentrations and pH, and onaddition of the milk proteins. However, samples with 1 and 2.5%gelatin at pH 3.0 did not gel during the cooling stage.

3.1.1. Gelling temperatureAs described above, the gelling temperatures (TG) of the samples

were measured in the cooling stage, and Table 1 summarises theANOVA output on the dependence of the gelling temperature ontype of proteins, concentration and pH. Only the main effects(concentration and protein) were significant, and the trends areshown in Fig. 2A and B.

At any pH, the higher the concentration of gelatin, the higher theTG of the gels (Fig. 2A). The 5.0% gelatin solution with or withoutmilk proteins gelled at temperatures in the range 20e22 �C and the2.5% solution in the range 15e18 �C (except at pH 3.0, explained inthe next paragraph); samples with 1.0% gelatin did not gel duringthe cooling stage. Similar results were reported by Michon,Cuvelier, and Launay (1993). In their study, TG of the gelatin gel

Fig. 1. Changes in G0 (.) and G00 (�) of pure gelatin gels at concentration 2.5%, pH 6.6. Cooling

ranged from 26.4 �C at 1.1% to 32.6 �C at 20% concentration. Joly-Duhamel et al. (2002) also reported that TG increased with con-centration for both mammalian and fish gelatin. The gelling tem-peratures of gelatin gels can also be different at the sameconcentration because of different thermal histories. The gellingpoint in a polymerizing system is considered to be the point atwhich a three-dimensional network, infinite in extent, first appears(Eldridge & Ferry, 1954). For a gelatin gel, a threshold level of pol-yproline II helixes is needed to form the three-dimensionalnetwork, which could be achieved within a relatively short timeand high temperature at high gelatin concentration.

pH from 4.6 to 8.0 had no significant effect on the TG of gelatingels with or without milk proteins at the concentrations studied(data not shown). The gelling properties of all samples weresubstantially affected by pH 3.0. At pH 3.0, the TG of samples with5.0% gelatin was much lower than that at other pH, and sampleswith 1.0 or 2.5% gelatin did not gel at all during cooling stage.Thus, it could be concluded that gelation of gelatin was inhibitedat pH 3.0. This is attributable to protonation of amino acids ofgelatin at low pH, which prevents formation of hydrogen bonds.Hydrogen bonds are very important in forming the gelatin gelframework (Bello et al., 1962). In a circular dichroism study of a0.2% gelatin solution, Wustneck, Wetzel, Buder, and Hermel(1988) observed that when the temperature of the gelatin solu-tion was decreased from 25 to 10 �C, an increase in peak area fortriple helical structures occurred at pH 4.9, 7.0 and 10, but not atpH 2.0. They attributed this to partial destruction of the gelatinmolecules at very low pH.

In the pH range studied, addition of MPC and SMP significantlyincreased the TG of gelatin gels at all gelatin concentrations both atpH 6.6 and 8.0, while addition of WPI did not affect TG significantly(Fig. 2B). These results indicated that interactions occurred be-tween gelatin and MPC/SMP, but not between gelatin and WPI.From these observations it can be inferred that, under the experi-mental conditions of this study, gelatin was able to interact withcaseins but not with whey proteins. The reasons for these differ-ences are discussed in Section 3.1.3.

step from 40 to 10 �C (A); annealing step at 10 �C (B); heating step from 10 to 40 �C (C).

Table 1Summary of ANOVA for the gelatin and gelatineprotein mixed gels.a

Parameterb Concentration pH Protein Concentration � pH Concentration � protein pH � protein Concentration � pH � protein

Firmness *** *** *** *** ** NS NS[***] [NS] [**] [NS] [*] [***] [NS]

TG *** NS NS NS NS NS NS[***] [NS] [***] [NS] [NS] [NS] [NS]

TM *** ** * NS NS ** **[***] [***] [***] [NS] [**] [***] [*]

G00 *** *** *** *** *** * NS

[***] [*] [***] [NS] [***] [**] [*]GN

0 *** * *** ** *** NS NS[***] [NS] [***] [NS] [***] [NS] [NS]

K *** NS *** ** *** NS NS[***] [***] [***] NS [***] [***] [***]

a Deductions are for pure gelatin and gelatinewhey protein mixtures or for pure gelatin and gelatinemilk proteins mixtures (in brackets [ ]); P < 0.05, *; p < 0.01, **;p < 0.0001, ***.

b TG ¼ gelling temperature, TM ¼ melting temperature, G00 ¼ initial G0 of time sweep, GN

0 ¼ G0 at infinite time in time sweep, K ¼ rate of gelation during annealing.

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493 487

3.1.2. Melting temperatureThe melting temperatures (TM) of the samples were obtained

from the heating stage, but 1.0% gelatin did not gel at pH 3.0, and nomelting temperature was obtained. TM of all samples was signifi-cantly affected by gelatin concentration (Table 1), and the higherthe concentration, the higher was TM (Fig. 3A and B). From pH 4.6 to8.0, gels with 5.0, 2.5 and 1.0% gelatinmelted at temperatures in therange 29e33 �C, 27e32 �C and 25e27 �C, respectively. These resultsare in agreement with previous studies (Bello et al., 1962; Eldridgeand Ferry, 1954; Michon et al., 1993; Salvador & Fiszman, 1998).A higher concentration of gelatin leads to shorter distances be-tween gelatin coils, hence stronger and more junction zones areformed, and a higher temperature is needed to destroy the struc-ture (Haug, Draget, & Smidsrod, 2004). It should be noted that thegelatin gels still melted below body temperature in all cases, whichensures their melt-in-mouth property. TM at pH 3.0 was substan-tially lower (25e27 �C) than at other pH for 2.5 and 5.0% gelatin gelsirrespective of the milk protein addition.

Low TM values at low pH were also observed by Choi andRegenstein (2000), who investigated gelatins from pig and fish;the TM of all gelatin gels decreased markedly at pH less than 4.0.Bello et al. (1962) reported that the melting point of a 5% gelatintype B gel was independent of pH between pH 5.0 and 11 butdecreased between pH 4.0 and 3.0. The effect of pH on TM wasmainly observed on gels with SMP (Fig. 3B), which showedsignificantly higher TM at pH 8.0 than at pH 6.6 at all gelatinconcentrations. This could be because of the minerals in the SMP.

Fig. 2. The main effect of concentration (A) and addition

pH has a great effect on the charge state of minerals, which canchange the electrolyte content and affect the gelation of gelatin(Bello et al., 1962; Boedtker & Doty, 1954). Although MPC alsocontains some minerals, the amount is less than that in SMPbecause of their removal during manufacture (Hemar, Hall,Munro, & Singh, 2002). It was also found that gels containingSMP showed significantly higher melting temperatures at allconcentrations and pH studied than other gels (Fig. 3B). Resultscontradictory to this study have been reported by Salvadorand Fiszman (1998) who found that the melting temperatureof a gelatin gel was not affected by the addition of milk com-ponents. However a different type of gelatin was used in thatstudy.

3.1.3. Annealing stageIn order to fully understand the gelation process during the

annealing stage, the rheograms were described using a modifiedfirst-order kinetic model (Eq. (2)) as it was thought that the time-sweep outputs described a first order kinetic. The storagemodulus was focussed on because G00 was effectively zero for thegels at all the conditions studied. The modified first-order kineticmodel has been extensively used to describe time-course starchdigestion, and Sorba and Sopade (2013) used it, written with vis-cosity parameters, to describe time-course viscosity changes instarch during digestion. Moreover, these authors opined that themodel is identical to some creep models, even though the param-eters might indicate different concepts.

of proteins (B) on gelling temperature of gelatin gels.

Fig. 3. Effect of the protein � concentration � pH interactions for melting temperature of gelatin gel with or without WPI (A) and gelatin gels with or without three milk proteins (B).

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493488

Gt0 ¼ G0

0 þ GN�00½1� expð�K tÞ� (2)

GN0 ¼ G0

0 þ GN�00 (3)

where G00 ¼ G0 at time t¼ 0, GN

0 ¼ G0 at infinite time (t/N) andK ¼ rate of gelation during annealing. These results are shown inFig. 3. Within the time range (250 min) studied, regression analysisrevealed that the first-order kinetic model gave significant corre-lation coefficients (P < 0.05) greater than 0; thus suggesting thatthis model is suitable for describing the experimental data of thisstudy. Since data of gels with 1.0% gelatin at pH 3.0 were notavailable, pH 3.0 was excluded from statistical analysis and dis-cussed separately.

From Fig. 4A and B, it could be seen that the initial G0 ðG00Þ of all

gels at the beginning of the annealing step, was higher for highergelatin concentration, suggesting that higher gelatin concentra-tions resulted in stronger gels at the end of cooling stage. This isbecause of a greater number of junction zones formed in the gels athigher concentrations of gelatin. Pure and protein-containinggelatin samples at 1.0% gelatin showed very low G0

0 (around zero)at all pH, indicating that the system had not gelled at the beginningof the annealing stage. G0

0 was independent of pH in the range 4.6e8.0 for all samples (Table 1).

In the pH range 4.6e8.0, at both 2.5 and 5.0% gelatin concen-trations, gels with WPI showed significantly lower G0

0 than puregelatin gels, while MPC- and SMP-containing gels showed signifi-cantly higher G0

0 (Fig. 4A and B). The enhancement of gelation bycasein has been studied in i-carrageenan gels, where gels madewith reconstitutedmilk and permeatewere compared. It was foundthat the presence of casein micelles increased G0 during the coolingand annealing stages and increased the gelation temperature; thiswas attributed to the additional bridging of carrageenan chains bycasein micelles (Langendorff, Cuvelier, Launay, & Parker, 1997).

The final G0 ðGN0Þ values of the gels in the annealing step are

shown in Fig. 4C and D. GN0 increased with gelatin concentration

and pH from 4.6 to 8.0 did not show a significant effect on GN0 of

the gels (Table 1). Only pH 3.0 dramatically lowered G0 of samples,which was only in the range of 200e300 Pa for 2.5% gels and 1300e1500 Pa for 5.0% gels. The effect of pH on the G0 did not seem to berelated to the IP of gelatin (gelatin used in this study was type B, theisoionic point (IP) of which is around 5.0). The same conclusionwasobtained from the study of Yoshimura et al. (2000) in which pigand shark gelatins with similar isoionic points (IP around 9.0)were used. Different effects of pH on the G0 of these gelatin gelswere observed. The G0 of pig gelatin gels showed no changes over apH range of 4.0e10, while the G0 of shark gelatin gels showed

significantly lower values at pH 4.0 and 8.0. The authors suggestedthat in the viscoelasticity of the gelatin gel, the intramoleculardistribution of charge is more important than the total or averagecharge of the whole molecule. Other authors reported that theviscoelastic properties of gelatin gels were greatly improved byhigher amounts of triple helical structure, instead of the aggregatesof gelatin molecules (Boedtker & Doty, 1954; Sarabia, Gomez-Guillen, & Montero, 2000). Large aggregates could be formedaround the IP (Boedtker & Doty, 1954; Lin et al., 2002), while nosignificant difference of G0 of gelatin gels was observed among thepH close to IP (4.6 and 5.3) and far away from IP (8.0) in our study,except pH 3.0.

Addition of WPI significantly decreased the value of GN0 at

gelatin concentrations 2.5 and 5.0%, but not at 1.0%. MPC and SMPcaused a significant increase in GN

0 at a gelatin concentration of5.0%, but not at 2.5 and 1.0% (Fig. 4C and D), which indicated thegelatin concentration dependence of the milk protein effect. Thusthe effect of milk proteins on gelatin gels was more significant athigher gelatin concentrations, which is in agreement with a studyon caseinmicelles/k-carrageenan interactions. It was found that thechance for these two polymers to interact increased with increasedconcentration of k-carrageenan. At very low carrageenan concen-trations, there were fewer carrageenan chains so that the junctionzones connected by casein micelles may not be statistically signif-icant (Ji, Corredig, & Goff, 2008; Langendorff et al., 1997). Alsoincreasing the gelling agent concentration could lead to an increasein the amount of milk protein aggregates (Martin, Goff, Smith, &Dalgleish, 2006), which would affect the rheological properties ofthe gels.

Increase in G0 value of gelatin gel by SMP has been reportedpreviously (Salvador & Fiszman, 1998). It was attributed to stabili-zation of the network by SMP through changes in hydrogenbonding, which is the basis of the formation of the gelatin network.The disturbance of the gelatin network by addedWPC has also beenreported (Walkenstrom & Hermansson, 1996). Considering thedifferences among these three milk protein preparations, it isapparent that the casein component may be the cause of the higherG0 of the mixed gels than that of pure gelatin gels. However theseresults are different from those reported previously. Based on sizedistribution and turbidimetric titration results, Lefebvre andAntonov (2001) reported that no interaction occurred betweengelatin type B and caseins at pH 6.7, at gelatin: casein ratios from0.03 to 13. A rheological study by Koh et al. (2002) indicated thataddition of gelatin B had little influence on G0 of sodium caseinategels at pH 4.6. They reported that in the pH range 4.5e5.0, bothcasein and gelatin B carry a very low net charge, so they would nothave a strong interaction. Our study of gelatin/MPC and gelatin/

Fig. 4. Effect of protein � concentration � pH interaction during annealing stage on G00 (A), GN

0 (C) and K (E) for gelatin gel with or without WPI, G00 (B), GN

0 (D) and K (F) forgelatin gels with or without three milk proteins.

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493 489

SMP gels was conducted at pH 6.6 and 8.0 (both gelatin and themilk proteins carried some negative charge at those pH) which isdifferent from the study of Koh et al. (2002). Moreover it has beensuggested that a “positive patch” exists between residues 97 and112 of k-casein, which is on the surface of casein micelles, and thatthis positive patch even exists at pH above PI (Snoeren, 1975).Therefore, it is possible that interactions occurred between gelatinand the positive patch on k-casein under the experimental condi-tions of this study.

Similar to those obtained in this study, effects of addition of milkproteins have been reported on gelation of k-carrageenan gels. Itwas reported that k-carrageenan gel with SMP showed a lower losstangent with increase of pH from 6.0 to 8.0, indicating higherelasticity. However, no explanations were given for this result (Xu,Stanley, Goff, Davidson, & Lemaguer, 1992). Hemar et al. (2002)reported that addition of SMP and MPC showed a greater effecton the viscoelastic properties of k-carrageenan gels than addition ofWPI, and the results obtained for MPC- and SMP-containing gelswere similar. They attributed this effect to the phase separationcaused by addition of MPC and SMP. The casein particle networkformed due to depletion flocculation contributed to the increase inviscosity and elasticity of the k-carrageenan gel, while for WPI-containing gels, no significant phase separation occurred that

would cause improvement in the rheological properties of the gels.Those authors reported in another study that the depletion floc-culation mechanism depended on the biopolymer size; particles inSMP and MPC solutions have an average diameter of 0.2 mm, whichismainly due to caseinmicelles, while particles inWPI solutions areof nanometer size (Hemar, Tamehana, Munro, & Singh, 2001). Thishypothesis is also reasonable for explaining the results of our studyon gelatin. Phase separation of different mixtures could be seen inour microscopy study (Fig. 6).

Fig. 4E and F show the results of K0 during the annealing stage,which indicated the gelation rate. The K0 of all the gels increasedwith increasing concentration of gelatin at all pH values. At pH 3.0 asignificant decrease in K0 was observed for pure gelatin gels at 2.5and 5.0% gelatin concentrations, but no significant difference couldbe seen for WPI-containing gels. Interestingly, it was observed thatthe K0 of WPI-containing gels was higher than for pure gelatin gels,except at 1.0% gelatin concentration (Fig. 3E). However, WPI-containing gels showed the lowest values for other parameters.A significant increase in K0 due to addition of MPC and SMP couldonly be seen at 1.0% gelatin concentration, and not at higher con-centrations like the other parameters. The gelation rate not onlydetermines the final gel strength, but also is crucial to the structureof the gels. High gelation rates tend to lead to a coarse network

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493490

(Cavallieri & Da Cunha, 2008). Further study needs to be done tounderstand the mechanism of rate of increase of K0.

3.2. Texture analysis

In the texture analysis, only two gelatin concentrations (2.5 and5.0%) were studied. Since a 1.0% gelatin gel was too soft, no pene-tration force peak could be obtained. Fig. 5A and B show repre-sentative penetrometry profiles of the samples. Gel firmness wascalculated as initial slope from the penetrometry curves and theresults are shown in Fig. 5C and D.

Fig. 5A and B show how the penetration curve varied amongdifferent gels. All gels showed a rapid increase in the force over ashort time as the probemoved into the samples, although the initialslope was different for different gels. The gel firmness was muchhigher at 5.0% concentration than at 2.5% for all gels (Fig. 5C and D),similar to the concentration effect on rheological results. Theseresults are also in agreement with previous studies (Ferry, 1948;Fiszman & Salvador, 1999; Salvador & Fiszman, 1998).

The gel firmness at pH 3.0 was significantly lower than that atother pH (p < 0.05) for all gels (Fig. 5C), which corresponded to therheological results. The low initial slope indicated that the gels atpH 3.0 deformed easily and tended to flow more than break. Theeffect of pH on the firmness of gelatin gels is probably due tochanges in the electrostatic interactions in the system (Fiszman &Salvador, 1999). The firmness of a high concentration (27%)gelatin gel was studied in the pH range 2.0e12 by Cumper andAlexander (1952). They found that at the extreme pH, the num-ber of basic or acidic amino acid residues available for bond for-mation decreased rapidly and consequently a sharp decrease infirmness occurred. This could explain the low firmness of thegelatin gel at pH 3.0 in our study. Our results were also consistentwith the study of Choi and Regenstein (2000), in which different

Fig. 5. Representative penetrometry profiles for gels and the effect of protein � concentratioprotein on the texture profile of gelatin gels at pH 6.6 (.5% pure gelatin gel, d 2.5% pure gepure gelatin gels (. pH 3.0, d pH 4.6, －－ 5.3, －$$－ pH 8.0) (B); firmness of pure gelatinthree milk proteins (D).

kinds of gelatin were studied. They observed a marked decrease ingel strength of gelatin gels below pH 4.0.

Gel firmness was independent of pH in the range 4.6e8.0 for allgels except gels containing SMP. Gels with SMP showed signifi-cantly higher firmness at pH 8.0 than at pH 6.6 at both 2.5 and 5.0%gelatin concentration. This could be due to the minerals in the SMPas discussed before for gel melting temperature. Addition of milkproteins significantly increased the gel firmness at 5.0% gelatinconcentration, while at 2.5% the values of the mixed gels were notsignificantly different from those of the pure gelatin gels. Theseresults again indicated the dependence on the gelatin concentra-tion of the effect of the milk proteins on gelatin gels.

In addition to gel firmness, another important characteristic thatcould be observed from penetrometry curves was the breakingpoint, which is a measure of the fracturability of the gel. As can beseen from Fig. 5B, only some of the pure gelatin gels at 2.5% gelatinconcentration showed a breaking point. At pH 5.3 and 6.6, the puregelatin gels were not easily deformed and showed a clear breakingpoint during penetration (between 8 and 9 s) (Fig. 5B). No breakingpoint was observed for pure gelatin gels at pH 3.0 and the pene-tration force kept increasing until the compression ended at 10 s,indicating that the gel had no fracturability. The force recorded atthe end of compression (10 s) is not indicative of any physicalcharacteristic (Fiszman & Salvador, 1999) and hence was notrecorded. At pH 4.6, the profile of the pure gelatin gel showed ashoulder at about 8.5 s, which indicated it had an initial resistanceto penetration; however this was considered as a questionablebreaking point because the penetration force did not show anapparent decrease after this point. The small inflection also indi-cated a structural changewhich was not strong enough to break thegel and needs to be confirmed by a microscopy study. At pH 8.0, noapparent breaking point was observed either. The fracturability ofgels could be related to IP. Molecular aggregation could be caused

n � pH interaction on gel firmness. Effect of gelatin concentration and addition of milklatin gel, －－ 2.5% gelatin gel with MPC) (A); effect of pH on the texture profile of 2.5%gel and gelatin gel with WPI (C) and firmness of pure gelatin gel and gelatin gels with

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493 491

by the strong attraction of oppositely charged groups on gelatinchains around the IP (Boedtker & Doty,1954), leading to fragile gels.However a gel at pH 4.6, which is also very close to the IP of gelatintype B, did not show apparent fracturability. The difference be-tween the gels formed at pH 4.6 and at pH 5.3 and 6.6 could also beseen in microscopy results (Fig. 6).

Additionally, no apparent breaking point was observed for puregelatin gels at 5.0% concentration at all pH conditions studied. Athigh concentrations, gelatin gels have higher elasticity, whichmakes them harder to break (Hansen, Blennow, Pedersen,Norgaard, & Engelsen, 2008). The absence of a breaking point in5.0% gelatin gels indicated that the changes in electrostatic in-teractions due to pH only influenced the aggregation of the mobilechains in weak gels like the 2.5% gelatin gels. The results were inagreement with a study by Salvador and Fiszman, (1998), althoughdifferent texture profiles were obtained because of a differentgelatin type used and different experimental conditions. Theyfound that at 1.5% gelatin concentration, the breaking force ofgelatin gels decreased as the pH moved further away from its IP,

Fig. 6. Scanning electronic micrographs of gels: 2.5% pure gelatin gels at pH 6.6 (A); 1% pureWPI (G), MPC (H) and SMP (I) at pH 6.6. Scale bars in the images are 1 mm.

while no significant differences were found at 5.0% concentration(Fiszman & Salvador, 1999). It can also be observed that all the gelscontaining milk protein showed no obvious breaking point, sug-gesting that the presence of milk proteins changes the texture ofgelatin gel in such a way that prevents fracturability. This could bebecause milk proteins filled in the pores of gelatin gel, therebymaking the gel less fracturable.

3.3. Microstructure

The microstructure of gels with high gelatin concentration (2.5and 5.0%) was too dense and no clear gelatin strands could beseen. Hence, the effect of pH and addition of milk protein on themicrostructure of gelatin gel was studied at 1.0% gelatin concen-tration. The structural changes in pure gelatin gels were studied inthe pH range 3.0e8.0. The effect of addition of milk proteins onthe microstructure of gelatin gels was studied at pH 6.6. High milksolids content tended to cover the gelatin structure, so in thisstudy only 2% milk solids were added. The microstructure of

gelatin gel at pH 3.0 (B), pH 4.6 (C), pH 5.3 (D), pH 6.6 (E) and pH 8.0 (F); 1% gelatin with

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493492

gelatin gels with and without milk proteins were examined bySEM (Fig. 6).

As can be seen in the micrographs, the microstructure of gelatingels differed greatly with concentration. A 2.5% gelatin gel at pH 6.6formed a very dense structure with small voids (Fig. 6A). No strandscould be seen in the gel structure while for all the gels with 1.0%gelatin, the structurewasmuch looser and the strands could be seenclearly (Fig. 6BeF). Some particles were observed inpure gelatin gels(Fig. 6A), which could have been an impurity in the gelatin product,such as unhydrolyzed collagen. These results were in agreementwith the rheology and texture study, which showed that strongerand firmer gels were formed at higher gelatin concentrations.

The microstructure of pure gelatin gels was also significantlyaffected by pH (Fig. 6BeF). The microstructure of 1.0% gelatin gelformed at pH 3.0 was much looser with larger pores than those athigher pH (Fig. 6B). The microstructure at higher pH was muchdenser, suggesting that the gels were more organized at these pHvalues. Pure gelatin gels at pH 4.6 appeared to form more strandsthan those at pH 3.0 (Fig. 6C). From pH 5.3, the network becamemore three dimensional, and no significant differences wereobserved between themicrostructures of pure gelatin gels at pH 5.3and 6.6 (Fig. 6D and E). The microstructure of the pure gelatin gel atpH 8.0 was dense with some large pores, but still individual strandscould be observed (Fig. 6F). These results could further explain theresults obtained during the textural measurements. At pH 3.0, thegel was too soft to be fractured. pH 4.6 seemed to be the transitionpH for the gels to change from flat to “three-dimensional”, whichcould have induced a questionable breaking point in the penetr-ometry curve and a relatively low melting temperature. At pH 5.3and 6.6, the gels formed well-organised structures, which couldhave led to enough gel strength to have a clear breaking point. AtpH 8.0, the gel structurewas very well organised and it could not bebroken within the penetration distance used in this study.

The gelatin gel with WPI at pH 6.6 had a similar microstructureas the pure gelatin gel (Fig. 6G). The strands’ organization did notchange, and the fine and homogeneous gel structure could still beobserved. The only difference was that the gel appeared denser andmore protein aggregates were seen in the WPI-containing gel,which could be attributed to the presence of protein aggregates inWPI (Hemar et al., 2001). No apparent phase separation between thegelatin and whey proteinwas observed. This was in agreement witha previous study byWalkenstrom and Hermansson (1994), inwhichlight microscopy was used to study the microstructure of WPI/gelatin mixed gels. In the gelatin gel with MPC, no individual caseinmicelles could be seen; they appeared as large aggregates linked bygelatin strands (Fig. 6H). The gel structure had changed significantlyfrom that of the pure gelatin gel. The regularity of the pure gelatingel seemed disturbed. It was not as homogenous as the pure gelatingel and large voids were observed. It seems there were two separatephases, containing dense milk protein aggregates and gelatinstrands, which could be caused by depletion flocculation of milkproteins in the aqueous phase (Hemar et al., 2002). These micro-structure results suggest interactions between the gelatin and milkproteins, which is in agreement with the observation of therheology study. Similar results have been obtained by confocal mi-croscopy of MPC and k-carrageenan gels (Hemar et al., 2002). In thegelatin gel with SMP (Fig. 6I), a compact and interconnectednetwork was formed and more branched structures were observedthan in the pure gelatin gel. Phase separation could again beobserved, although the milk proteins filled the voids and formedclusters and chains connected by gelatin strands. Different from thegel with MPC, the SMP-containing gel showed smaller protein ag-gregates and smaller voids, which was in agreement with theconfocal microscopy study of Hemar et al. 2002. The SMP-containing gel also appeared to be more branched, which could be

attributed to the ions in SMP affecting the structure of the gelatin geland the interaction between gelatin and casein by influencing thehydrogen bonding (Salvador & Fiszman,1998). SMP/carrageenan gelwas also studied by Martin et al. (2006), who found that the SMPaggregates were linked by carrageenan. It was attributed to attrac-tive electrostatic interactions between carrageenan and casein.Similar electrostatic interactions could be responsible for thechanges in the microstructure of gelatin gels containing MPC andSMP in this study. The interaction could be between the “positivepatch” on k-casein and gelatin, as discussed before. There are noreports in the literature indicating such positive patches on wheyprotein molecules. Hence, no interactions between gelatin and WPIare expected, since both carry a small negative charge at pH 6.6.

4. Conclusions

This study provides an insight into the effect of concentration,pH and addition of milk proteins on the gelation behaviour ofgelatin. According to our results, mixed gels of gelatin and milkproteins with different properties could be obtained by manipu-lating these three factors. Stronger and firmer gels could be formedwith high gelatin concentrations and addition of casein, whilewhey protein has different effects on the rheology and texture ofgelatin gels. Gelatin behaviour is constant in pH range from 4.6 to8.0 in terms of rheology and texture, which makes it suitable for awide range of applications. However the difference of fracturabilitycaused by pH should be noted. The differences caused by additionof different milk proteins on gelatin gel can be understood in twoways. Firstly, k-casein has a “positive patch” on the surface of itsmolecule, which can interact with negatively charged gelatin at pH6.6 and 8.0, while whey proteins do not have such a “positivepatch”. Secondly, casein micelles in MPC and SMP can more easilycause depletion flocculation, which induces phase separation, thanthe much smaller particles present in WPI. This phase separationmay lead to higher elasticity of the gels. A further study with pu-rified milk proteins, such as k-casein, b-casein, a-lactalbumin andb-lactoglobulin, could be undertaken to better understand themechanism of interaction between gelatin and milk proteins. SEMseems to be a useful technique to explain the physical properties ofgels at the microstructure level. In this study, modelling theannealing stage was very useful to obtain a comprehensive un-derstanding of the gel properties. The parameters modelled in thisstudy revealed some useful information which was not apparentfrom just measuring the traditional rheological parameters such asG0. The dependence of gelatin concentration on the effect of milkproteins was only observed from GN

0 and the slightly weaker gelformed at pH 4.6 could only be seen from parameter G0

0. Furtherstudy needs to be done to elucidate the mechanism of increase inrate of G0 during the annealing stage.

Acknowledgements

The authors acknowledge the financial support to Dairy Inno-vation Australia. We also acknowledge the facilities, and the sci-entific and technical assistance, of the Australian Microscopy &Microanalysis Research Facility at the Centre for Microscopy andMicroanalysis, The University of Queensland.

References

Bello, J., Vinograd, J. R., & Bello, H. R. (1962). Mechanism of gelation of gelatin e

influence of ph, concentration, time and dilute electrolyte on gelation of gelatinand modified gelatins. Biochimica et Biophysica Acta, 57(2), 214e221.

Boedtker, H., & Doty, P. (1954). A study of gelatin molecules, aggregates and gels.Journal of Physical Chemistry, 58(11), 968e983.

Z. Pang et al. / Food Hydrocolloids 35 (2014) 484e493 493

Boran, G., Mulvaney, S. J., & Regenstein, J. M. (2010). Rheological properties ofgelatin from silver carp skin compared to commercially available gelatins fromdifferent sources. Journal of Food Science, 75(8), E565eE571.

Cavallieri, A. L. F., & Da Cunha, R. L. (2008). The effects of acidification rate, pH andageing time on the acidic cold set gelation of whey proteins. Food Hydrocolloids,22(3), 439e448.

Choi, S. S., & Regenstein, J. M. (2000). Physicochemical and sensory characteristics offish gelatin. Journal of Food Science, 65(2), 194e199.

Cumper, C. W. N., & Alexander, A. E. (1952). The viscosity and Rigidity of gelatin inconcentrated aqueous systems. 1. Viscosity. Australian Journal of ScientificResearch Series A e Physical Sciences, 5(1), 146e152.

Dickinson, E. (1998). Stability and rheological implications of electrostatic milkprotein-polysaccharide interactions. Trends in Food Science & Technology, 9(10),347e354.

Djabourov, M. (1988). Architecture of gelatin gels. Contemporary Physics, 29(3),273e297.

Djabourov, M., Lechaire, J. P., & Gaill, F. (1993). Structure and rheology of gelatin andcollagen gels. Biorheology, 30(3e4), 191e205.

Eldridge, J. E., & Ferry, J. D. (1954). Studies of the cross-linking process in gelatingels. 3. Dependence of melting point on concentration and molecular weight.Journal of Physical Chemistry, 58(11), 992e995.

Ferry, J. D. (1948). Mechanical properties of substances of high molecular weight. 4.Rigidities of gelatin gels e dependence on concentration, temperature andmolecular weight. Journal of the American Chemical Society, 70(6), 2244e2249.

Fiszman, S. M., & Salvador, A. (1999). Effect of gelatine on the texture of yoghurt andof acid-heat-induced milk gels. Zeitschrift Fur Lebensmittel-Untersuchung Und-Forschung a e Food Research and Technology, 208(2), 100e105.

Hansen, M. R., Blennow, A., Pedersen, S., Norgaard, L., & Engelsen, S. B. (2008). Geltexture and chain structure of amylomaltase-modified starches compared togelatin. Food Hydrocolloids, 22(8), 1551e1566.

Haug, I. J., Draget, K. I., & Smidsrod, A. (2004). Physical and rheological properties offish gelatin compared to mammalian gelatin. Food Hydrocolloids, 18(2), 203e213.

Hemar, Y., Hall, C. E., Munro, P. A., & Singh, H. (2002). Small and large deformationrheology and microstructure of kappa-carrageenan gels containing commercialmilk protein products. International Dairy Journal, 12(4), 371e381.

Hemar, Y., Liu, L. H., Meunier, N., & Woonton, B. W. (2010). The effect of high hy-drostatic pressure on the flow behaviour of skim milk-gelatin mixtures. Inno-vative Food Science & Emerging Technologies, 11(3), 432e440.

Hemar, Y., Tamehana, M., Munro, P. A., & Singh, H. (2001). Viscosity, microstructureand phase behavior of aqueous mixtures of commercial milk protein productsand xanthan gum. Food Hydrocolloids, 15(4e6), 565e574.

Ji, S., Corredig, M., & Goff, H. D. (2008). Aggregation of casein micelles and kappa-carrageenan in reconstituted skim milk. Food Hydrocolloids, 22(1), 56e64.

Joly-Duhamel, C., Hellio, D., Ajdari, A., & Djabourov, M. (2002). All gelatin networks:2. The master curve for elasticity. Langmuir, 18(19), 7158e7166.

Karim, A. A., & Bhat, R. (2009). Fish gelatin: properties, challenges, and prospects asan alternative to mammalian gelatins. Food Hydrocolloids, 23(3), 563e576.

Koh, M. W. W., Merino, L. M., & Dickinson, E. (2002). Rheology of acid-inducedsodium caseinate gels containing added gelatin. Food Hydrocolloids, 16(6),619e623.

Lal, S. N. D., O’Connor, C. J., & Eyres, L. (2006). Application of emulsifiers/stabilizersin dairy products of high rheology. Advances in Colloid and Interface Science, 123,433e437.

Langendorff, V., Cuvelier, G., Launay, B., & Parker, A. (1997). Gelation and floccula-tion of casein micelle/carrageenan mixtures. Food Hydrocolloids, 11(1), 35e40.

Lefebvre, J., & Antonov, Y. (2001). Stability against aggregation of bovine caseinmicelles in the presence of acidic and alkaline gelatin. Colloid and PolymerScience, 279(5), 434e441.

Lin, W., Yan, L. F., Mu, C. D., Li, W., Zhang, M. R., & Zhu, Q. S. (2002). Effect of pH ongelatin self-association investigated by laser light scattering and atomic forcemicroscopy. Polymer International, 51(3), 233e238.

Martin, A. H., Goff, H. D., Smith, A., & Dalgleish, D. G. (2006). Immobilization ofcasein micelles for probing their structure and interactions with poly-saccharides using scanning electron microscopy (HEM). Food Hydrocolloids,20(6), 817e824.

Michon, C., Cuvelier, G., & Launay, B. (1993). Concentration-dependence of thecritical viscoelastic properties of gelatin at the gel point. Rheologica Acta, 32(1),94e103.

Petrie, S. E. B., & Becker, R. (1970). Thermal behavior of aqueous gelatin solutions. InR. S. Porter, & J. F. Johnson (Eds.), Analytical Calorimetry (pp. 225e238). NewYork: Plenum Press.

Salvador, A., & Fiszman, S. M. (1998). Textural characteristics and dynamic oscilla-tory rheology of maturation of milk gelatin gels with low acidity. Journal ofDairy Science, 81(6), 1525e1531.

Sarabia, A. I., Gomez-Guillen, M. C., & Montero, P. (2000). The effect of addedsalts on the viscoelastic properties of fish skin gelatin. Food Chemistry, 70(1),71e76.

Snoeren, T. H. M., Payens, T. A. J., Jeunink, J., & Both, P. (1975). Electrostatic inter-action between k-carrageenan and k-casein. Milchwissenschaft-Milk ScienceInternational, 30, 3.

Sopade, P. A., Halley, P. J., & Junming, L. L. (2004). Gelatinisation of starch in mix-tures of sugars. II. Application of differential scanning calorimetry. CarbohydratePolymers, 58(3), 311e321.

Sorba, A., & Sopade, P. A. (2013). Changes in rapid visco-analysis (RVA) viscosityreveal starch digestion behaviours. Starch/Stärke, 65, 437e442.

Syrbe, A., Bauer, W. J., & Klostermeyer, N. (1998). Polymer science concepts in dairysystems e an overview of milk protein and food hydrocolloid interaction. In-ternational Dairy Journal, 8(3), 179e193.

Walkenstrom, P., & Hermansson, A. M. (1994). Mixed gels of fine-stranded andparticulate networks of gelatin and whey proteins. Food Hydrocolloids, 8(6),589e607.

Walkenstrom, P., & Hermansson, A. M. (1996). Fine-stranded mixed gels of wheyproteins and gelatin. Food Hydrocolloids, 10(1), 51e62.

Walkenstrom, P., & Hermansson, A. M. (1997). Mixed gels of gelatin and wheyproteins, formed by combining temperature and high pressure. Food Hydro-colloids, 11(4), 457e470.

Wustneck, R., Wetzel, R., Buder, E., & Hermel, H. (1988). The Modification of thetriple helical structure of gelatin in aqueous-solution. 1. The influence of anionicsurfactants, Ph-value, and temperature. Colloid and Polymer Science, 266(11),1061e1067.

Xu, S. Y., Stanley, D. W., Goff, H. D., Davidson, V. J., & Lemaguer, M. (1992).Hydrocolloid milk gel formation and properties. Journal of Food Science, 57(1),96e102.

Yoshimura, K., Terashima, M., Hozan, D., Ebato, T., Nomura, Y., Ishii, Y., et al. (2000).Physical properties of shark gelatin compared with pig gelatin. Journal ofAgricultural and Food Chemistry, 48(6), 2023e2027.