Effects of xanthan gum rheology on the foaming properties of whey protein concentrate

Accepted Manuscript Effects of xanthan gum rheology on the foaming properties of whey protein concentrate L.P. Martínez-Padilla, J.L. García-Rivera, V. Romero-Arreola, N.B. CasasAlencáster PII: DOI: Reference:

S0260-8774(15)00036-9 http://dx.doi.org/10.1016/j.jfoodeng.2015.01.018 JFOE 8052

To appear in:

Journal of Food Engineering

Received Date: Revised Date: Accepted Date:

31 August 2014 22 November 2014 25 January 2015

Please cite this article as: Martínez-Padilla, L.P., García-Rivera, J.L., Romero-Arreola, V., Casas-Alencáster, N.B., Effects of xanthan gum rheology on the foaming properties of whey protein concentrate, Journal of Food Engineering (2015), doi: http://dx.doi.org/10.1016/j.jfoodeng.2015.01.018

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Effects of xanthan gum rheology on the foaming properties of whey protein concentrate L. P. Martínez-Padilla*, J. L. García-Rivera, V. Romero-Arreola, N. B. Casas-Alencáster Laboratorio de Propiedades Reológicas y Funcionales en Alimentos, Departamento de Ingeniería y Tecnología. Facultad de Estudios Superiores Cuautitlán, Universidad Nacional Autónoma de México. Av. Primero de Mayo s/n, Cuautitlán Izcalli, Edo. de México, 54740, México.

*Corresponding author, Tel: +525556232038; Fax: +525556232026; E-mail address: [email protected]

Abstract The stability of foams with whey protein concentrate (WPC) and xanthan gum (XG) were studied. Flow behavior, density, pH and average particle size of aqueous phases were evaluated (10-25 % WPC, 0.05 or 0.15% XG). Flow properties of the aqueous phases were dominated by XG rheology, where a zero shear viscosity was detected before the classic shear-thinning behavior (Carreau model). In general, an increase of either XG or WPC in the mixtures resulted in an increase in zero shear viscosity, characteristic time and foaming capacity. The foam stability, evaluated by the kinetics of drainage and Ostwald ripening, also increased with WPC or XG concentration, reaching very stable milk foams. The functionality of WPC was improved by the presence of XG, likely as a consequence of biopolymer segregative interactions (thermodynamic incompatibility). Rheology of aqueous phase played a decisive role in WPC-XG foam stability.

Key words: whey protein concentrate, xanthan gum, rheology, Carreau model, foam capacity, foam stability.

Introduction Properties and stability of foams and emulsions depend on many characteristics of the surface-active component in the system and interactions with other foods including food macromolecules like polysaccharides (Dickinson, 1998; Nicurescu et al., 2009). Milk proteins and polysaccharides are known to play an important role in the formation and stabilization of foams and emulsions in food industry. To date, many studies have been conducted with the aim of understanding how these biopolymers adsorb and interact at air1

water and oil-water interfaces in order to control product morphology and stability (Turgeon and Laneuville, 2009).

Previous investigation into whey proteins and other milk proteins have focused on the role of physicochemical properties in foam stability (Marinova et al., 2009), while others concentrated on processes to improve the functionality of whey protein. For example, dynamic heat treatment (Nicurescu et al., 2008, 2009), ultrasound treatment (Jambrak et al., 2014) and the formation of fibrillar dispersions by thermal treatment (Oboroceanu et al., 2014). Recent studies have demonstrated that skim milk powder fortified with a high concentration of WPC improved foaming properties (Martínez-Padilla et al., 2014). Fewer studies related to the effect of food hydrocolloids on whey foam stability exist (LiszkaSkoczylas et al., 2014; Narchi et al., 2009).

It is well known that thickening polysaccharides like xanthan gum (XG) can be used to increase viscosity of the aqueous phase and enhance emulsion stability by retarding the creaming process (Sun et al., 2007). XG is an anionic hetero-polysaccharide with a considerable practical value. Despite its high viscosity at rest (or low shear rates), its “weak gel” properties and high shear-thinning behavior (flows easily) accounts for its ability to provide long-term stability to colloidal systems (Sworn, 2000). This behavior is explained by the rigidity of xanthan chains and by the weakness of the network between macromolecules that involves hydrogen bonds (Narchi et al., 2009).

Microstructure and rheology of protein-XG mixtures differ depending on the XG concentration and the protein type. In WPI-XG mixtures, co-solubility and thermodynamic incompatibility were reported (Bryant and Mc Clements, 2000; Hemar et al., 2001; Li et al., 2006). In aqueous systems, the rheological parameters of WPI-XG mixtures (pH 7, XG 0.20.5% w/w, WPI 14%) are considered synergistic because the parameters of the mixture were larger than those calculated from each independently (Sanchez et al., 1997). Similar results were reported in aqueous mixtures with a lower concentration of WPI (4-10%) and a higher concentration of XG (>0.5%), which were attributed to the water-soluble hybrid denominated complexes or conjugates, these complexes were not coacervates (Benichou et 2

al., 2007). In gels, a dispersion of XG among the protein network is thought to explain an improved elastic modulus of the WPI (pH 6.5-6.0, 0.01-0.06% XG, 16% WPI) (Bertrand and Turgeon, 2007). In both cases, proteins and polysaccharides carried the same negative charge, and the behavior was attributed to the segregative phase separation.

Considering that adsorption at fluid interfaces of whey proteins was also improved by synergistic interactions with XG, resulting in better surface and viscoelastic properties of the film interface (Benichou et al., 2007; Perez et al., 2010), and this mixture was shown to be a good emulsifier in simple oil-in-water emulsion systems (Panaras et al., 2011; Sun et al., 2007; Vázquez-Solorio et al., 2011), the aim of this work is to quantify the effect of XG rheology on the foaming properties of WPC, to propose alternatives for formulating more stable products and to advance our understanding of the functionality of these proteinpolysaccharide blends.

2. Materials and Methods 2.1 Materials Commercial samples of spray-dried WPC (WPC34, 34% protein, 49-52% carbohydrates, 3.8% moisture, pH 6.0-6.7, Origen country USA, Hegart de México, S. A. de C. V., México) and XG (9.94% moisture, Keltrol, CP Kelco, CA, USA) were used. Aqueous samples were prepared by dispersing powders in purified water (E-pura, The Pepsi bottling group Mexico, Mexico).

2.2 Foam preparation Aqueous samples at 10, 15, 20 and 25% WPC alone or with 0.05 and 0.15% XG, were prepared by weight percent on a dry base with a mixer (Heidolph RZR1, Germany, 5931248 rpm, depending on the sample). XG was dissolved first into the mixtures.

To obtain foams, the aqueous samples were whipped using a planetary mixer (Kitchen Aid, K5SS, MI, USA) with a balloon whisk agitator tool (wire whip) for 20 min, at ~170 rpm, incorporating all liquid volume (250 mL) into the foam. Preparation of aqueous samples and foams was conducted at room temperature (24 ± 1oC) maintained by an air conditioner. 3

Foam temperature was not increased during whipping. Each foam was replicated a minimum of three times to obtain average foaming properties. The average and standard deviation were reported. An analysis of variance was performed to compare differences of means between treatment groups (α =0.05).

2.3 Properties of aqueous samples Physical and physicochemical properties of aqueous samples (continuous phase) were measured: relative density (digital density meter, DA-110M, Mettler-Toledo AG Analytical, Switzerland), pH (pH120, Conductronic, Mexico) and particle size distribution (particle size analyzer Cilas 930 L/D, France, based on light diffraction and the Fraunhöfer theory, measured range from 0.2 to 500 µm, in the wet dispersion mode). The samples for particle size analysis were prepared by the dispersion of the aqueous samples in water and treated by ultrasound (55 kHz, <1 min) to breakdown protein agglomerates.

Viscosity was tested using a rheometer for steady shear measurements (MCR301 Physica, Anton Paar, Austria). Flow curves were obtained with a cone and plate, 75 mm diameter and 1o, using a several step program. Each step was divided into 25 points; each reading was taken after 10 s. For WPC samples, an up-down shear rate step program was applied (10-300 s 1 ). For XG and mixtures three steps were applied; first shear stress was varied from 0.001 to 5 Pa, second a constant shear rate was maintained (300 s 1 ) and finally an up-down shear rate step (0.1-300 s 1 ). The later was done to verify shear time independency. Linear (Newtonian fluid) or shear thinning Carreau (Eq.1) model were applied to experimental data. Values of zero shear viscosity ( ), infinite viscosity ( ), characteristic time ( ) and flow behavior index  were calculated using Rheoplus software V3.61 (Anton Paar, Austria).   η η  η  1      /

(1)

In this model,  parameter was neglected (<0.001 Pa s) because of the small values related with  values. Correlation coefficients were higher than 0.99. The temperature was maintained at 25 ± 1 °C.

4

2.4 Properties of foams The same tests were applied as reported previously in a study of skim milk foams fortified with WPC (Martínez-Padilla et al., 2014). Briefly, foaming capacity ( ) was determined immediately after whipping and calculated using Eq. 2, where   is the volume reached by the foam after 20 min of whipping, and   is the initial volume of the aqueous phase (200 mL).       ⁄ 100

(2)

The liquid volume fraction contained in the foam was estimated as a  /  ratio. Foam density was calculated in petri dishes (50 mm), where fresh foam was gently filled to the top. The weight and volume occupied were used to calculate foam density.

Foam stability was measured microscopically from images of the fresh foam. Images were taken using a microscope (10x, Olympus, CX31, Japan) and an Evolution LC camera kit (Media Cybernetics Inc., MD, USA). The diameter of 80 bubbles was measured using the Image-Pro Discovery software (V.4.5.1.29, 2002, Media Cybernetics Inc.); numeric average and standard deviations were calculated. Determination of bubble diameter was repeated 10 and 20 min after foam formation (sitting quiescently after whipping had stopped).

Foam stability was evaluated volumetrically as a function of time. Drainage volume was measured in stability cups (a beaker attached to a 10 mL graduated cylinder). Fresh foam (200 mL) was placed in the cup over a bed of fiberglass. The volume of drained liquid was measured every 5 min or every 30 min, depending on sample stability. The kinetics was fitted to a linear model; in this case, the slope corresponds to the drainage rate. The time ( ) to begin drainage was registered, where a high  value indicated high foam stability and a low  value indicated poor foam stability. Foam stability was also determined optically by measuring the light transmission through the foam using a Turbiscan MA2000 (Formulaction, France) (Mengual et al., 1999). The cylindrical glass tube was filled with fresh foam as described recently. Foam destabilization 5

was analyzed using backscattering profiles as a response to applied pulses (near infrared light source, 850 nm). Foam scans were made every 2 or 30 min. Less stable foams were scanned for 1 h, while more stable foams were scanned for 21 h. Typical backscattering profiles were reported in a previous study; drainage of fluid was identified by a negative peak at the bottom of the glass tube, that showed increasing width or thickness with time; the increase in bubble diameter (Ostwald ripening) is observed as a progressive decrease of backscattered light in the center of the glass tube (Martínez-Padilla et al., 2014).

The kinetics of different destabilization processes was obtained using Turbisoft software (V1.2.1, 1998, Formulaction). For liquid drainage, backscattering absolute thickness was applied and fitted to a kinetic-based sigmoidal model (Eq. 3) (Curve Expert software, V1.34, 1993, unregistered evaluation copy, Microsoft Corp., WA, USA), where three parameters were defined:  , the maximum thickness of the drainage; the exponent ; and  / , the time required to attain half the maximum thickness of the drainage (Raharitsifa et al., 2006).        /



(3)

Normally, drainage is high at the beginning but slows down over time, characterized by a sigmoid curve, where  values are close to 1. When drainage curves became more sigmoidal,  values increased up to 3 or higher. Foam stability is expected to increase as the  decreases;  / is equivalent to half-life. Ostwald ripening kinetics was calculated based on the percent of the backscattering medium values, taken as a function of time, and fitted to a linear model. In this case, the slope corresponds to the rate of the increase of bubble diameter.

3. Results and Discussion 3.1 Properties of aqueous samples Physical properties of aqueous samples of WPC, XG and their mixtures were measured as a quality control and to establish their relationship with foam stability. Relative density of

6

WPC and mixtures ranged from 1.035-1.088, while the density of XG was close to that of water.

The pH of mixtures was close to 6.3. The pH of XG was 6.15 or 6.34, depending on concentration. The pH of WPC diminished from 6.47 to 6.23 when WPC concentration was increased. Values were close to those reported for WPC (González-Tello et al., 2009; Jambrak et al., 2008), but different from the -lactoglubulin isoelectric point (4.2-5.5) (Boland, 2011). Considering that pH between 6 and 7 has little effect overall on bubble diameter of skim milk foam stability (Borcherding et al., 2009), pH was not adjusted in this study.

An example of particle size distributions of mixtures and WPC aqueous samples are shown in Fig. 1a, WPC samples displayed a monomodal distribution with a mode value at approximately 110 µm in size. Samples with or without XG were very similarly sized. The arithmetic mean diameter varying between 99 and 117 µm, and no significant differences were detected (P>0.05). Jambrak et al. (2014) found WPC (protein 60%, 10% dry matter) agglomerates had a median diameter of 104 µm, corresponding to an accumulative frequency of 50%, within the range of this study. Bimodal distributions of particle size have been reported in rehydrated milk protein concentrate powders (1.5% dry matter) after 90 min and 24 h rehydration, where mode 1 was attributed to casein micelles (<1 µm) and mode 2 was attributed to large poorly-dispersible particles, corresponding probably to whey proteins (between ~ 60 and 120 µm) (Crowley et al., 2015). Following ultrasound application, mixtures and WPC aqueous samples showed a break up of protein agglomerates (or large poorly-dispersible particles), and a narrowed and bimodal distribution with particle size values between 0.2 and 25 µm (Fig. 1b). The mode of particles in each population was close to 2.6 and 12 µm, first value corresponding to small aggregates of whey proteins and second one to the biggest ones. Jambrak et al. (2014) reported a median diameter of 0.55 µm for small WPC aggregates (protein 60%, 10% dry matter) when a hard ultrasonic treatment was applied (40 kHz, 15 min). Higher values than those obtained in our study can be attributed to a less intense ultrasound treatment applied 7

in our samples (<1 min). The water-soluble hybrid formation between WPC and XG proposed by Benichou et al. (2007) cannot be evidenced by the particle diameter distribution.

Fig. 2a shows dynamic viscosity values as a function of the shear rate for WPC aqueous samples. Newtonian behavior can be observed at shear rates higher than 50 s 1 , as reported for WPC suspensions (<20% w/w, 67.6% protein) (González-Tello et al., 2009) and characteristic for aqueous systems of proteins at medium or low concentrations. Dynamic viscosity ranged from 0.0019 to 0.0067 Pa s and increased with WPC concentration (Table 1).

Flow behavior of XG aqueous samples obtained at the first step of the shear program is shown in Fig. 2b. These samples exhibited the characteristic non-Newtonian behavior of XG aqueous dissolutions with the presence of a viscosity plateau at small shear rates (0.1 or 1 s 1 ) and shear thinning at high shear rates (Dolz et al., 2007; Martínez-Padilla et al., 2004). The transition from the constant low shear rate viscosity to a power-law behavior was smooth. All mixtures showed curves similar to those of XG aqueous samples and were adjusted to the Carreau model (Eq. 1) (Fig. 2c, Fig. 2d); respective parameters are summarized in Table 1. An increase in XG concentration resulted in greater values of  and viscosity, but with the same overall trend. As expected, in mixtures  increased as WPC concentration increased, with values ranging from 0.027 to 5.3 Pa s. The effects of WPC concentration on  at low and high level of XG were different when comparing to XG alone. For XG 0.05%, with WPC, lower  values were obtained (dilution effect) while at XG 0.15%, higher values were achieved (synergist effect) with WPC 15% or higher concentrations. Lower  in mixtures than respective value of XG (0.05%) aqueous systems could indicate biopolymers co-solubility and that the microstructure of the less entangled system formed by XG was disturbed by WPC aggregates. Higher  in mixtures than respective value of XG (0.15%) imply a biopolymer segregative phase separation, as has been reported for 8

whey protein isolate-XG (Bryant and Mc Clements, 2000; Hemar et al., 2001; Li et al., 2006). In samples with 0.15% XG , the complex structure of the weak network formed by XG was not disturbed even with the high concentration of WPC used. This was considered synergistic because the viscosities of the mixture were larger than those calculated from each biopolymer independently, as reported by Sanchez et al. (1997).

The flow behavior index for mixtures oscillated between 0.69 and 0.87. Comparing this index to that of XG dissolution alone, an increase was observed when WPC was added, independent of concentration, indicating less susceptibility to lose its original structure with shearing. The characteristic time,  corresponds to the reciprocal of the shear rate when the shear thinning behavior began, and increased with an increase in WPC concentration at XG 0.05%, and remained nearly constant at XG 0.15% (19-21 s). This means that at a high  value, the sample starts the shear thinning behavior at a lower shear rate. The critical shear rate obtained from the intersection of the power-law and the constant viscosity region is close to 1 s 1 for XG 0.05%, and to 0.1 s 1 for XG 0.15%, after this, the up and down flow curves overlapped (shear steps 3 and 4).  for XG aqueous phase at 0.05%, was higher when compared to their respective mixtures with WPC, while for XG 0.15%, higher  was detected than in mixtures with WPC.

Rheological parameters of aqueous samples obtained at a low concentration of XG (0.05%), confirmed co-solubility and that the microstructure formed by XG was disturbed by WPC aggregates. As a result, lower  and lower  values, but higher  values were obtained than the corresponding values of XG aqueous dissolution. Inversely, parameters of aqueous samples obtained at high concentrations of XG (0.15%) and WPC (15-25%), confirmed a synergist effect, illustrated by higher values of  and  than corresponding values of XG aqueous dissolution. Closer values of  between mixtures and XG aqueous dissolution showed that the complex structure of the weak network formed by XG was not disturbed even with the high concentration of WPC used.

9

3.2 Foam properties 3.2.1 Foaming capacity The foaming capacity of samples is shown in Table 2. As expected, an increase was observed with increasing WPC. These results agree with those reported by Marinova et al. (2009), who found that foaming capacity increased with increasing WPC or WPI, and can be attributed to the foaming properties of -lactoglobulin, the main protein of the WPC that is able to diffuse and reduce tension at the newly formed fluid interface.

When comparing in both levels of XG, foam capacity was not practically modified by XG presence. Small differences were obtained between the foaming capacities at 0.05 or 0.15% XG, as 600 738% and 550 700%, respectively. These values were higher than those for WPC enriched milk foams (380 534%), prepared with skim milk powder (10%) and WPC (5 15%) (Martínez-Padilla et al., 2014). Similar values between 553.5% and 797.4% was reported for foams prepared with WPI (20% protein) and sugar (16.2%) (Pernell et al., 2002), confirming that an increase in the flow properties due to XG addition did not affected foam formation.

3.2.2 Foam density Average foam density of the studied systems is presented in Table 2. A decrease was observed with increasing WPC. Values varied between 0.128 and 0.149

. These

values were within those reported by Borcherding et al. (2009) for foams of ultrafiltered skim milk (1-6% protein) (0.11-0.22

), some of which were lower than those for

enriched WPC-milk foams (0.14-0.18

) (Martínez-Padilla et al., 2014). However, in

WPC (8%) and XG (2%) foams, a high density (0.686

) has been reported (Liszka-

Skoczylas et al., 2014). The high-density values may be due to the high density and very high viscosity of the aqueous phase developed by XG at 2%.

Samples containing the highest concentration of WPC had the lowest density, and accordingly had the highest foaming capacity. This corresponds to the low liquid volume fraction ( ) contained in the foam (0.154-0.119) (Table 2). Some of these values were 10

higher than those obtained for enriched WPC-milk foams (0.13-0.07) (Martínez-Padilla et al., 2014). Similar

(~0.13-0.14) were reported for foams prepared with whey proteins

(Oboroceanu et al., 2014) and for WPI with sugar (Pernell et al., 2002). However, in WPC (8%) and XG (2%) foams, a very high

(~0.51) was reported (Liszka-Skoczylas et al.,

2014), possibly due to the high concentration of XG used (high density and very high viscosity).

3.2.3 Foam stability Foam stability was determined by three methods: bubble diameter, drainage of liquid using the volumetric method, and light scattering through the foam. Tables 3 and 4 summarize the results of each method for the studied samples. Three methods were used to confirm that stability measures were not altered by high viscosity of aqueous phases. Representative micrographs of fresh foams and 10 or 20 min after foam formation are shown in Fig. 3. Foams obtained showed polydisperse bubble size distributions with spherical and polyhedral shapes.

As expected, an increase in protein concentration decreased the initial average diameter of bubbles from 248 to 165 µm in XG 0.05% foams, and from 220 to 157 µm in XG 0.15%, obtained in fresh foams (Table 3). These results are in agreement with those reported by Borcherding et al. (2009) and Marinova et al. (2009).

As seen for fresh enriched WPC-milk foams (Martínez-Padilla et al., 2014), high standard deviations were observed evidencing a high heterogeneity. When WPC concentration was increased, standard deviations decreased, indicating low heterogeneity. Average bubble diameters were higher when compared to those obtained for fresh enriched WPC-milk foams (116-89 µm) and were within distributions reported for foams prepared with heated WPI dispersions (100-400 µm; 2% WPI, pH 7) (Oboroceanu et al., 2014) or for WPC foams (200 µm, 0.1% WPC, with or without NaCl) (Marinova et al., 2009). The flow properties of WPC-XG aqueous phases influence negatively on the bubble diameter obtained, since higher diameters were observed than those found in enriched WPC-milk foams (similar concentrations of WPC) (Martínez-Padilla et al., 2014). 11

The average bubble diameter obtained after 10 min of foam formation was greater compared with initial average bubble diameter, where values ranged from 305 to 186 µm in XG 0.05% foams, and from 236 to 167 µm in XG 0.15%. Similar behavior was observed after 20 min of foam formation, ranging from 279-181 µm in both XG concentrations. In these conditions, foams containing the lowest concentration of WPC and XG were not stable. In contrast, the smallest bubble diameters after 10 or 20 min of foam formation were obtained in foams containing the highest concentration of WPC and XG. These behaviors can be seen in Fig. 3. The increase in average bubble diameter with time is due to air diffusion through the elastic film and was reduced by increasing WPC and XG.

Drainage of the liquid through the films by gravity or capillary action was determined by quantifying the volume of drained liquid with time in stability cups (volumetric method) (Fig. 4). The initial drainage time was registered and increased with WPC and XG concentration, on the other hand, the drainage rate decreased, indicating in both parameters their ability to improve foam stability (Table 3).

From this method, the more stable foams were also those containing the highest concentration of WPC and XG. It is important to note that the initial drainage time (80-613 min) obtained at XG 0.15% foams is much higher than the half-life obtained for fresh enriched WPC-milk foams (20-30 min) studied previously (Martínez-Padilla et al., 2014). The lowest values of drainage rates were obtained at XG 0.15% (0.05-0.008 mL

),

confirming the increase in foam stability by XG.

In backscattering curves, the drainage front was delimited at the bottom of the cylindrical glass tube (~0-8 mm). The increase in bubble diameter by diffusion of air (Ostwald ripening) was observed in the center of the cylindrical glass tube in the same samples (~865 mm). After delimitation of each zone, the kinetics of different destabilization processes was calculated.

12

Liquid drainage was fit to a sigmoidal model (Eq. 3) and found to be highly sigmoidal in nature (Fig. 5),

values varied between 3.8-4.5 for XG 0.05% and 5.3-10 for XG 0.15%.

The minimum thickness of drainage (6-4 mm) and the maximum

(120-104 min) were

obtained with WPC 25% foams at both XG concentrations (Table 4). These results are in agreement with the increase in the time to begin drainage and the decrease in drainage rate observed using the volumetric method, when concentrations of WPC or XG are increased. increases in both XG levels while increasing WPC concentration.

For the Ostwald ripening process, a curve of the percentage of

backscattering (reference

mode, initial time equal to zero min) medium values as a function of time in the center of the glass tube were plotted (Fig. 6). Initially values decreased rapidly but towards the end decreased slowly. The start of the curve was fitted to a linear model and the slope was calculated (Table 4). A negative slope indicates the rate of decrease of medium values of backscattering and is related with the increase in bubble diameter (Mengual et al., 1999). Ostwald ripening rate decreased with increased WPC concentration at both XG concentrations (Table 4). The lowest rate was obtained at higher levels of WPC and XG.

Foam stability obtained in WPC-XG (0.15%) foams was a consequence of two actions: first the better surface and viscoelastic properties of the film interface developed with two biopolymers, as reported by Perez et al. (2010), and second the synergist viscosity effect obtained in the aqueous phase, as reported by Sanchez et al. (1997), both actions were a result of the segregative phase separation.

Some relationships between aqueous phase properties, physical foam properties and foam stability parameters were obtained by fitting them to a linear model. Pearson correlation coefficients are summarized in Table 5. Protein concentration had a good fit with almost all the properties or parameters. Direct relationships with protein concentrations were found at each XG concentration, for both physical properties of aqueous samples (density and zero shear viscosity), foam capacity and stability times (initial drainage time and

). In the

same manner, a proportional decrease with protein concentrations were obtained for liquid 13

volume fraction, foam density, initial bubble diameter, drainage rate and Ostwald ripening rate.

The liquid volume fraction correlated positively with foam density, initial bubble diameter and drainage rate, but negatively with relative density and zero shear viscosity, foam capacity, initial drainage time and

. Zero shear viscosity had a positive relationship with

relative density, foam capacity, initial drainage time and

. Negative proportionality was

found for liquid volume fraction, foam density, initial bubble diameter, drainage rate and Ostwald ripening rate.

Finally, initial drainage time correlates positively with relative density, zero shear viscosity, foam capacity and

, but negatively with foam density, initial bubble diameter, drainage

rate and Ostwald ripening rate. Correlation coefficients were lower than 0.9 in drainage rate or Ostwald ripening rate, when correlated with liquid volume fraction and zero shear viscosity, or initial drainage time (Table 5). These can be attributed to the complex evaluation of these rates.

4. Conclusions Whey protein concentrates with XG seems to be an ideal biopolymer blend, useful for the development of new food foams in the dairy food industry. Rheology of aqueous phase was the determinant contribution on foam stability. In accordance with consumer preferences, they provide good foaming capacity and stability. Foams with the highest concentration of WPC with both concentrations of XG had the best foaming properties; however, among these, the lowest concentration of XG showed better foaming capacity. Aqueous phase and physical foam properties correlated very well with foam stability parameters, which are also useful to develop new food foams.

Acknowledgments This study was financed by a grant obtained from DGAPA-UNAM (project IN218214). We thank Carolyn Unck for her careful correction of the English language and style.

14

References Benichou, A., Aserin, A., Lutz, R., & Garti, N. (2007). Formation and characterization of amphiphilic conjugates of whey protein isolate (WPI)/xanthan to improve surface activity. Food Hydrocolloids, 21, 379-391. Bertrand, M.E., & Turgeon, S.L. (2007). Improved gelling properties of whey protein isolate by addition of xanthan gum. Food Hydrocolloids, 21, 159-166. Boland, M. (2011). Whey proteins. In R.Y. Yada (Ed.), Proteins in Food Processing (pp. 30-55). Woodhead Publishing Ltd. Cambridge, UK. Borcherding, K., Lorenzen, P.C., & Hoffmann, W. (2009). Effect of protein content, casein-whey protein ratio and pH value on the foaming properties of skimmed milk. International Journal of Dairy Technology, 62, 161-169. Bryant, C.M., & McClements, D.J. (2000). Influence of xanthan gum on physical characteristics of heat-denatured whey protein solutions and gels. Food Hydrocolloids, 14, 383–390. Crowley, S.V., Desautel, B., Gazi, I., Kelly, A.L., Huppertz, T., O'Mahony, J.A. 2015. Rehydration characteristics of milk protein concentrate powders. Journal of Food Engineering, 149, 105-113. Dickinson, E. (1998). Stability and rheological implications of electrostatic milk proteinpolysaccharide interactions. Trends in Food Science & Technology, 9, 347-354. Dolz, M., Hernández, M.J., Delegido, J., Alfaro, M. C., & Muñoz, J. (2007). Influence of xanthan gum and locust bean gum upon flow and thixotropic behaviour of food emulsions containing modified starch. Journal of Food Engineering, 8, 179–186. González-Tello, P., Camacho, F., Guadix, E.M., Luzón, G., & González, P.A. (2009). Density, viscosity and surface tension of whey protein concentrate solutions. Journal of Food Process Engineering, 32, 235–247. Hemar, Y. , Tamehana, M., Munro, P.A., & Singh, H. (2001). Viscosity, microstructure and phase behavior of aqueous mixtures of commercial milk protein products and xanthan gum. Food Hydrocolloids, 15, 565-574. Jambrak, A.R., Mason, T.J., Lelas, V., Herceg, Z., & Herceg, L.J.I. (2008). Effect of ultrasound treatment on solubility and foaming properties of whey protein suspensions. Journal of Food Engineering, 86, 281–28. 15

Jambrak, A.R., Mason, T.J., Lelas, V., Paniwnyk, L., & Herceg, Z. (2014). Effect of ultrasound treatment on particle size and molecular weight of whey proteins. Journal of Food Engineering, 121, 15-23.

Li, J., OuldEleya, M.M., & Gunasekaran, S. (2006). Gelation of whey protein and xanthan mixture: effect of heating rate on rheological properties. Food Hydrocolloids, 20, 678 – 86. Liszka-Skoczylas, M., Ptaszek, A., & Zmudzinski, D., (2014). The effect of hydrocolloids on producing stable foams based on the whey protein concentrate (WPC). Food Hydrocolloids, 29, 1-11. Marinova, K., Basheva, E., Nenova, B., Temelska, M., Mirarefi, A., Campbell, B., & Ivanov, I. (2009). Physico-chemical factors controlling the foamability and foam stability of milk proteins: Sodium caseinate and whey protein concentrates. Food Hydrocolloids, 23, 1864-1876. Martínez-Padilla, L.P., García-Mena, V., Casas-Alencáster, N.B., & Sosa-Herrera, M.G. (2014). Foaming properties of skim milk powder fortified with milk proteins. International Dairy Journal, 36, 21-28. Martínez-Padilla, L.P., López-Araiza, F., & Tecante, A. (2004). Steady and oscillatory shear behavior of fluid gels formed by binary mixtures of xanthan and gellan. Food Hydrocolloids, 18, 471–481. Mengual, O. Meunier, G., Cayré, I. Puech. K. & Snabre, P. (1999). Turbiscan Ma2000: multiple light scattering measurement for concentrated emulsions and suspension instability analysis. Talanta, 50, 445-446. Narchi, I., Vial, C., & Djelveh, G. (2009). Effect of protein-polysaccharide mixtures on the continuous manufacturing of foamed food products. Food Hydrocolloids, 23, 188-201. Nicurescu, I., Loisel, C., Vial, C., Riaublanc, A., Djelveh, G., Cuvelier G., & Legrand J. (2008). Combined effect of dynamic heat treatment and ionic strength on the properties of whey protein foams – Part II. Food Research International, 41, 980–988. Nicurescu, I., Loisel, C., Vial, C., Riaublanc, A., Djelveh, G., Cuvelier G., & Legrand, J. (2009). Effect of dynamic heat treatment on the physical properties of whey protein foams. Food Hydrocolloids, 23, 1209–1219.

16

Oboroceanu, D., Wang, L. Magner, E., & Auty, M.A.E. (2014). Fibrillization of whey proteins improves foaming capacity and foam stability at low protein concentrations. Journal of Food Engineering, 121, 102-111. Panaras, G., Moatsou, G., Yanniotis, S., & Mandala, I. (2011). The influence of functional properties of different whey protein concentrates on the rheological and emulsification capacity of blends with xanthan gum. Carbohydrate Polymers, 86, 433-440. Pernell, C.W., Foegeding, E.A., Luck, P.J., & Davis, J.P. (2002). Properties of whey and egg white protein foams. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 204, 9-21. Perez, A.A., Carrera-Sánchez, C., Rodríguez-Patino, J.M., Rubiolo, A.C., & Santiago, L. G. (2010). Milk whey proteins and xanthan gum interactions in solution at the air-water interface: a rheokinetic study. Colloids and Surfaces B: Biointerfaces, 81, 50-57. Raharitsifa, N., Genovese, B.D., & Ratti, C. (2006). Characterization of apple juice foams for foam-mat drying prepared with egg white protein and methylcellulose. Journal of Food Science, 71, 142-151. Sanchez, C., Schmitt, C. Babak, V.G., & Hardy, J. (1997). Rheology of whey protein isolate-xanthan mixed solutions and gels. Effect of pH and xanthan concentration. Nahrung, 41, 336-343. Sun, C., Gunasekaran, S., & Richards, M.P. (2007). Effect of xanthan gum on physicochemical properties of whey protein isolate stabilized oil-in-water emulsions. Food Hydrocolloids, 21, 555–564. Sworn, G. (2000). Xanthan gum. In G. O. Phillips, & P. A. Williams (Eds.), Handbook of Hydrocolloids (pp. 103-115). Woodhead Publishing Limited. Cambridge, UK. Turgeon, S.L., & Laneuville, S.I. (2009). Protein + polysaccharide coacervates and complexes: From scientific background to their application as functional ingredient in food products. In S. Kasapis, I.T. Norton, & J.B. Ubbink (Eds.), Modern biopolymer science: Bridging the divide between fundamental treatise and industrial application (pp. 327-363). Academic Press, London, UK. Vázquez-Solorio, S.C., Vega-Mendez, D.D., Sosa-Herrera, M.G., & Martínez-Padilla, L.P. (2011). Rheological properties of emulsions, containing milk proteins mixed with xanthan gum. Procedia Food Science, 1, 335-339. 17

Figure Captions Fig. 1. Particle size distribution of aqueous samples (a) and the same samples treated by ultrasound (b). Whey protein concentrate (WPC) and xanthan gum (XG). Fig. 2. Viscosity as a function of shear rate of aqueous samples. Whey protein concentrate (WPC) and xanthan gum (XG). Fig. 3.Microscopic images (objective 4x) of the fresh foams, 10 and 20 min after foam formation. a. WPC10% XG0.05%, b. WPC25% XG0.05%, c. WPC10% XG0.15%, d. WP25% XG0.15%. Whey protein concentrate (WPC) and xanthan gum (XG). Fig. 4. Volume of drained liquid with time using the volumetric method. Whey protein concentrate (WPC) and xanthan gum (XG). Fig. 5. Thickness of drained fluid with time using the optical method. Whey protein concentrate (WPC) and xanthan gum (XG). Fig. 6. ∆Backscattering with time using the optical method. Whey protein concentrate (WPC) and xanthan gum (XG).

18

Table 1. Flow properties of aqueous samples of whey protein concentrate (WPC) and xanthan gum (XG). Newtonian Studied sample

Zero shear viscosity (Pa•s )

Viscosity (Pa•s)

WPC 10% WPC 15% WPC 20% WPC 25% XG 0.05% XG 0.15% WPC 10% - XG 0.05% WPC 15% - XG 0.05% WPC 20% - XG 0.05% WPC 25% - XG 0.05% WPC 10% - XG 0.15% WPC 15% - XG 0.15% WPC 20% - XG 0.15% WPC 25% - XG 0.15%

0.0019a 0.0028b 0.0043c 0.0067d

Carreau model Characteristic Flow behavior time index (s) (-)

(± 0) (± 0) (± 0.0001) (± 0.0002)

0.13a (± 0.01) 4.37a (± 1.5) 0.75a 2.89b (± 0.08) 15.2b (± 0.7) 0.67b c (± 0.5) 0.86c 0.027 (± 0.003) 1.4c (± 0.7) 0.87c 0.037d (± 0.002) 2.1cd (± 0.8) 0.87c 0.055e (± 0.006) 2.9cd f d (± 0.4) 0.84c 0.117 (± 0.002) 3.2 (± 0.23) 2.5g 20.7e (± 2.8) 0.70d h (± 0.24) 3.4 19.2e (± 0.5) 0.69d (± 0.18) 4.1h 21.3e (± 1.1) 0.69d (± 0.29) 5.3i 21.4e (± 1.3) 0.69d a-i Standard deviations are shown in parentheses. Means within a column with common superscripts did not differ significantly (P < 0.05).

19

(± 0.018) (± 0.002) (± 0.019) (± 0.004) (± 0.012) (± 0.009) (± 0.004) (± 0.003) (± 0.001) (± 0.002)

Table 2. Foaming capacity and density of whey protein concentrate (WPC) and xanthan gum (XG) foams.

Studied sample

Foam capacity (%)

WPC 10% WPC 15% WPC 20% WPC 25% WPC 10% WPC 15% WPC 20% WPC 25% -

Volume liquid fraction (calculated) (-)

Foam density (g cm3 )

(± 0) (± 0) 0.137a 600a 0.143 XG 0.05% (± 0) (± 0) 0.136a 625ab 0.138 XG 0.05% b (± 0.003) (± 18) 0.133b 663 0.131 XG 0.05% (± 0.003) (± 18) 0.131c 738c 0.119 XG 0.05% (± 0) (± 0) 0.149d 550d 0.154 XG 0.15% de (± 18) (± 0.004) 588 0.146 0.148d XG 0.15% (± 0.003) (± 18) 0.134e 638e 0.136 XG 0.15% (± 0) (± 0) 0.128f 700f 0.125 XG 0.15% a-f Standard deviations are shown in parentheses. Means within a column with common superscripts did not differ significantly (P < 0.05).

20

(± 0.0003) (± 0.0003) (± 0.0004) (± 0.0006) (± 0.0005) (± 0.0002) (± 0.0008) (± 0.0006)

Table 3. Foam stability. Bubble diameter, initial drainage time and rate (volumetric method) of whey protein concentrate (WPC) and xanthan gum (XG) foams.

Studied sample

Average bubble diameter Twenty Ten minutes minutes Initial* after foam after foam formation (m) formation (m) (m)

Liquid drainage Initial drainage time (min)

Drainage rate (mL min1 )

248a (±95) 305a (±113) 20a (± 0) 0.22a (± 0.0165) b b a a 207 (±57) 242 (±75) 279 (±104) 45 (± 7.1) 0.12b (± 0.0045) 187c (±49) 224b (±62) 257b (±77) 65a (± 7.1) 0.06c (± 0.0059) d c c b 165 (±43) 186 (±60) 209 (±72) 125 (± 21.2) 0.04c (± 0.005) 220e (±79) 236d (±82) 246d (±83) 80c (± 0) 0.05d (± 0.0035) 198f (±51) 224d (±68) 238d (±81) 215d (± 21.2) 0.03e (± 0.003) g e e e 171 (±45) 189 (±62) 203 (±67) 310 (± 14.1) 0.01f (± 0.0003) 157h (±43) 167f (±49) 181f (±52) 613f (± 6.4) 0.008f (± 0.0004) a-h * After 20 min of whipping. Standard deviations are shown in parentheses. Means within a column with common superscripts did not differ significantly (P < 0.05).

WPC 10% - XG 0.05% WPC 15% - XG 0.05% WPC 20% - XG 0.05% WPC 25% - XG 0.05% WPC 10% - XG 0.15% WPC 15% - XG 0.15% WPC 20% - XG 0.15% WPC 25% - XG 0.15%

21

Table 4. Foam stability. Parameters of liquid drainage and Ostwald ripening obtained by backscattering light profiles of whey protein concentrate (WPC) and xanthan gum (XG) foams.

Studied sample

 (mm)

Liquid drainage  (-)

Ostwald ripening rate  / (min)

(%∆Backscattering minି1 )

(± 2.9) 26.6a 0.52a 5.8ab (± 0.4) 3.8a (± 0.4) (± 1.3) 41.0a 0.41ab 7.1a (± 0.5) 4.1a (± 0.8) b a b (± 0.1) 3.9 (± 1.0) (± 4.6) 58.5 0.38ab 5.4 120.7c (± 6.1) 6.0ab (±0.1) 4.5a (± 1.5) 0.36b (± 5.7) 58.2d 0.35c 5.2cd (± 0.4) 5.4b (± 1.4) c b e (± 1) (± 2.9) (± 4.1) 10 79.7 6.5 0.32c (± 2.6) 86.1e 3.9d (± 0.1) 8.8b (± 3) 0.30c (± 3.6) 104f 0.12d 4.0d (± 0.1) 5.3b (± 0.4) a-f Standard deviations are shown in parentheses. Means within a column with common superscripts did not differ significantly (P < 0.05).

WPC 10% - XG 0.05% WPC 15% - XG 0.05% WPC 20% - XG 0.05% WPC 25% - XG 0.05% WPC 10% - XG 0.15% WPC 15% - XG 0.15% WPC 20% - XG 0.15% WPC 25% - XG 0.15%

22

(± 0.04) (± 0.01) (± 0.03) (± 0.05) (± 0.001) (± 0.01) (± 0.05) (± 0.001)

Table 5. Some Pearson correlation coefficients between the aqueous phase and foam properties. Protein concentration

Aqueous phase properties

Physical foam properties

Foam stability (volumetric method) Foam stability (optical method)

Relative density Zero shear viscosity Foam capacity Liquid volume fraction Foam density Initial bubble diameter Initial drainage time Drainage rate

‫ݐ‬1/2

Oswald ripening rate

Liquid volume fraction

Zero shear viscosity

XG concentration (%) 0.15 0.05

0.05

0.15

0.05

0.998

0.999

-0.966

-0.996

0.921

0.993

-0.979

-0.933

0.970

0.994

-0.999

-0.999

-0.980

-0.997

-0.984

-0.956

0.980

-0.984

-0.993

0.965

0.894

Initial drainage time

0.15

0.05

0.15

0.990

0.946

0.958

0.988

0.907

0.988

0.997

0.997

0.984

-0.980

-0.996

-0.995

-0.977

0.965

-0.930

-0.943

-0.956

-0.917

0.938

0.986

-0.858

-0.974

-0.924

-0.931

0.966

-0.995

-0.977

0.988

0.989

-0.958 0.934

-0.957 0.982

0.884 -0.985

0.938 -0.974

-0.775 0.999

-0.917 0.984

-0.861 0.994

-0.852 0.963

-0.922

-0.884

0.838

0.908

-0.729

-0.930

-0.825

-0.973

23

a Density distribution

14

WPC 25%

12

WPC 25% - XG 0.05%

10

WPC 25% - XG 0.15%

8 6 4 2 0 0

100

200

300

400

Diameter (ߤm)

Density distribution

b

WPC 25%

4.5 4 3.5 3 2.5 2 1.5 1 0.5 0

WPC 25% - XG 0.05% WPC 25% - XG 0.15%

0

10

20

30

40

Diameter (ߤm)

Fig. 1. Particle size distribution of aqueous samples (a) and the same samples treated by ultrasound (b). Whey protein concentrate (WPC) and xanthan gum (XG).

24

25

Fig. 2. Viscosity as a function of shear rate of aqueous samples. Whey protein concentrate (WPC) and xanthan gum (XG).

26

0 min

10 min

20 min

a

b

c

d

Fig. 3.Microscopic images (objective 4x) of the fresh foams, 10 and 20 min after foam formation. a. WPC10% XG0.05%, b. WPC25% XG0.05%, c. WPC10% XG0.15%, d. WP25% XG0.15%. Whey protein concentrate (WPC) and xanthan gum (XG).

27

12

Volume (mL)

10 WPC 10% - XG 0.05%

8

WPC 15% - XG 0.05%

6

WPC 20% - XG 0.05%

4

WPC 25% - XG 0.05%

2 0 0

100

200

300

400

Time (min)

14 12

Volume (mL)

WPC 10% - XG 0.15%

10

WPC 15% - XG 0.15%

8

WPC 20% - XG 0.15%

6

WPC 25% - XG 0.15%

4 2 0 0

500

1000

1500

2000

Time (min)

Fig. 4. Volume of drained liquid with time using the volumetric method. Whey protein concentrate (WPC) and xanthan gum (XG).

28

7

Thickness (mm)

6 5

WPC 10% - XG 0.15%

4

WPC 15% - XG 0.15%

3

WPC 20% - XG 0.15% WPC 25% - XG 0.15%

2 1 0 0

200

400

Time (min) 8

Thickness (mm)

7 6

WPC 10% - XG 0.05%

5

WPC 15% - XG 0.05%

4

WPC 20% - XG 0.05%

3

WPC 25% - XG 0.05%

2 1 0 0

200

400

600

Time (min)

Fig. 5. Thickness of drained fluid with time using the optical method. Whey protein concentrate (WPC) and xanthan gum (XG).

29

∆Backscattering (%)

80 70 60

WPC 10% - XG 0.05%

50

WPC 15% - XG 0.05%

40

WPC 20% - XG 0.05%

30

WPC 25% - XG 0.05%

20 10 0 0

200

400

600

90

∆ Backscattering (%)

80 70

WPC 10% - XG 0.15%

60

WPC 15% - XG 0.15%

50

WPC 20% - XG 0.15%

40

WPC 25% - XG 0.15%

30 20 10 0 0

500

1000

1500

Time (min)

Fig. 6. ∆Backscattering with time using the optical method. Whey protein concentrate (WPC) and xanthan gum (XG).

30

Highlights 1. Foaming of whey protein concentrate was improved by the presence of xanthan gum. 2. Flow properties of the aqueous phases were dominated by xanthan gum rheology. 3. Foam stability and capacity were due to biopolymer segregative interactions.

31