Interfacial and rheological properties of gelatin based solid emulsions prepared with acid or alkali pretreated gelatins

Food Hydrocolloids 43 (2015) 700e707

Contents lists available at ScienceDirect

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

Interfacial and rheological properties of gelatin based solid emulsions prepared with acid or alkali pretreated gelatins Magnus N. Hattrem a, b, *, Silje Molnes a, Ingvild J. Haug c, Kurt I. Draget a a

Department of Biotechnology, Norwegian University of Science and Technology (NTNU), N-7491 Trondheim, Norway Concordix Pharma, N-9008 Tromsø, Norway c Statoil Norge, N-7053 Ranheim, Norway b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 16 May 2014 Accepted 29 July 2014 Available online 13 August 2014

In this study gelled emulsions (10, 20, 40 and 50 wt.% corn oil) and oil-free gels have been prepared using either acid (type A) or alkaline (type B) pre-treated gelatins with varying Bloom values and molecular weights. The gelling kinetics of the gels were monitored by performing small strain oscillatory measurements in combination with a temperature sweep. Increasing setting and melting temperatures as a function of oil content were obtained for the gelled emulsions prepared with type A gelatin. The shear storage moduli for the gels were determined after curing at 20
C for 15 min and compared to estimated values using a simplified version of the van der Poel's formula. A steeper increase in moduli as a function of oil content was obtained for the gelled emulsions prepared with the acid pretreated gelatins, also overshooting the estimated values. It is suggested that the higher moduli and setting and melting temperatures are caused by a hydrogen bond mediated flocculation of the oil droplets. Measurements of interfacial tension between gelatin solutions and corn oil, indicated that type A gelatin samples were more surface active compared to type B gelatin samples. This was attributed to the assumed presence of minor contaminants in the acid pre-treated gelatin samples, which were able to adsorb and reduce the interfacial tension at a faster rate. Finally, gelled emulsions were prepared using a cold water fish gelatin. It was observed that by the introduction of oil droplets, potentially improved physical properties may be obtained for these gels. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Gelatin Emulsion Gel Interface Rheology

1. Introduction Gelatin is a biopolymer obtained from the partial hydrolysis of the parent collagen molecule. A characteristic property of gelatin is its ability to form a thermoreversible gel with a low degree of thermal hysteresis. Gelatin is also amphiphilic due to the presence of both hydrophilic and hydrophobic amino acids (Ward & Courts, 1977). These properties enable gelatin to be an excellent stabiliser of gelled emulsions, as it may both adsorb at oil-water interfaces and form a polymeric network. Gelatin is prepared from collagen by pretreating it with either acid or alkali, in which the final gelatin is referred to as type A or type B gelatin, respectively. The pretreatment leads to hydrolysis of the inter- and intramolecular bonds within and between the collagen molecules. After pretreatment,

* Corresponding author. Department of Biotechnology, Norwegian University of Science and Technology (NTNU), Sem Sælandsvei 6/8, N-7491 Trondheim, Norway. Tel.: þ47 93217710. E-mail addresses: [email protected] (M.N. Hattrem), silje.molnes@uis. no (S. Molnes), [email protected] (I.J. Haug), [email protected] (K.I. Draget). http://dx.doi.org/10.1016/j.foodhyd.2014.07.026 0268-005X/© 2014 Elsevier Ltd. All rights reserved.

gelatin is extracted using hot water or diluted acid, and the soluble fraction after this step is usually referred to as gelatin (Stainsby, 1985). By alkali pretreatment, a conversion of glutamine and asparagine to glutamic and aspartic acid occurs, giving a shift in the isoelectric point (IEP) from 7e9 to ~5. The molecular weight of the final gelatin may also vary, depending on the raw material and preparation procedure. As its precursor collagen, gelatin may interconvert between a helix and random coil. Gelatin is however soluble above the helix-to-coil temperature in water and it will partially regain the triple helical structure below this temperature (Schrieber & Gareis, 2007; Veis, 1964; Ward & Courts, 1977). Gelatin based gelled emulsions are used extensively within food and confectionary, and have also been suggested as a possible delivery vehicle for lipid-based nutraceuticals and pharmaceuticals (Hattrem, Dille, Seternes, & Draget, 2014; Haug et al., 2011; Sato, Moraes, & Cunha, 2014; Thakur et al., 2012). For all of these applications the mechanical properties of the gelled emulsions are of crucial importance. In general, the mechanical properties of gelled emulsions vary strongly depending on the fraction and modulus of the dispersed phase, type and concentration of polymer and

M.N. Hattrem et al. / Food Hydrocolloids 43 (2015) 700e707

interaction between the dispersed droplets and the polymer matrix (Chen & Dickinson, 1999; Chen, Dickinson, Langton, & Hermansson, 2000; van Vliet, 1988). The structural network of gelled emulsions may be formed either by aggregation of dispersed filler, known as a particle gel, and/or by a structural network formed by polymers in the continuous water phase. Gelatin based emulsions would represent the latter, in which the solid like behaviour is mainly determined by the polymer matrix (Dickinson, 2012). The mechanical properties of such gels are known to be highly influenced by the interaction between the droplets and the gel matrix. A structural reinforcement (an increase in the composite modulus) may be obtained if there is a strong interaction between the droplet interfacial layer and the polymer matrix, in which the filler is referred to as being active. This may either occur by the adsorption of the gelling polymer at the droplet interface or by a molecular interaction between the emulsifying agent and the polymer matrix, e.g. electrostatic interaction. Usually, a prerequisite for an increase in composite modulus is that the droplets' modulus (Gfiller), given by the Laplace pressure (equation (1)), is higher than the modulus of the polymer matrix.

Gfiller ¼

2*goilwater droplet radius

(1)

As seen in equation (1), the Gfiller is proportional to the interfacial tension (goilwater) between the oil and water. If there is no/ weak interaction between the interfacial layer and polymer matrix, a reduction in the overall modulus would be expected regardless of the Laplace pressure of the dispersed droplets. The filler of such gels is referred to as being inactive (Chen & Dickinson, 1999; Kim, Renkema, & van Vliet, 2001; van Vliet, 1988). The mechanical properties of gelatin based gelled emulsions have earlier been characterized (Dickinson, Stainsby, & Wilson, 1985; Lorenzo, Checmarev, Zaritzky, & Califano, 2011; Sala, Van Aken, Stuart, & Van De Velde, 2007; Sala, van Vliet, Stuart, van de Velde, & van Aken, 2009). In the study by Dickinson et al. an increase in modulus as a function of oil content was observed. However, at lower gelatin concentrations a minimum in the shear modulus was obtained at a specific fraction of oil. This was suggested to be caused by an adsorption of gelatin at the droplet interface leading to a depletion of gelatin available for gelling in the continuous water phase (Dickinson et al., 1985). Studies by Sala et al. investigated both Young's modulus and stress and strain at fracture for a whey protein stabilised emulsion, in which the droplets were encapsulated in a gelatin matrix (Sala et al., 2007; Sala, van Vliet, Stuart, van Aken, & van de Velde, 2009; Sala, van Vliet, Stuart, van de Velde, et al., 2009). These droplets behaved as active fillers, leading to an increase in Young's modulus with increasing fraction of oil. Additionally, a decrease in both the fracture stress and strain were observed with increasing amount of filler. In a recent study, Lorenzo et al. investigated the influence of the molecular weight of the gelatin on the gelling properties of gelled emulsions. Increased elastic properties were obtained for the gelled emulsions with increasing molecular weights of the gelatin (Lorenzo et al., 2011). Although there have been several studies on the mechanical properties of gelatin based emulsions, the influence of using either acid or alkali pretreated gelatin have not yet been reported. The difference in both molecular weight distribution and IEP between type A and B gelatin may influence their ability to adsorb at and stabilise oil-water interfaces. It is further known that minor contaminants (mucosubstances, proteins and fats) with potential surface active properties may remain in the gelatin samples after preparation. These contaminants are reported to be present in larger amount in type A gelatin samples (Schrieber & Gareis, 2007;

701

Ward & Courts, 1977). As the modulus of the dispersed phase is suggested to be proportional to the interfacial tension (equation (1)), this may further influence the mechanical properties of the gelled emulsions. The scope of the present paper was to investigate if the gelatin type influenced the rheological and interfacial properties of gelled emulsions. Type A or type B gelatin with varying reported bloom values and molecular weight distributions were used to prepare gelled emulsions and oil-free gels. Dynamic interfacial tension was measured for different gelatin solutions, and small strain oscillatory measurements and puncture testing were performed on the resulting gels. 2. Materials and methods 2.1. Materials Acid and alkali pretreated gelatins with reported Bloom values of 160, 200 and 260 g were kindly supplied by Gelita (Eberbach, Germany) and used without further processing. Information about the different gelatins is listed in Table 1. Corn oil and trizma base were purchased from Sigma Aldrich (Seelze, Germany). Acetic acid was provided by VWR international (Darmstadt, Germany). High molecular weight cold water fish gelatin (batch 8004) was supplied by Norland Products (New Jersey, USA). All experiments were performed using deionised water (MQ-water). 2.2. Preparation of gelatin based emulsions and oil-free gels A gelatin solution was prepared by dissolving 25 wt.% gelatin (160A, 160B, 200A, 200B, 260A or 260B gelatin e see Table 1 for information) in MQ-water at 55
C. The present gelatin concentration was chosen in order to give a surplus of gelatin in the bulk phase of the finalised emulsion, reducing potential bulk depletion effects due to gelatin adsorption at the droplet interface. Corn oil (10, 20, 40 or 50 wt.%) was added to the gelatin solution and this sample was further equilibrated at 55
C for 10 min. An oil-in-water emulsion was prepared by mixing the two liquids using a VDI 12 homogeniser (VWR International, Darmstadt, Germany) equipped with a dispersing element (type S12N-12S, VWR International, Darmstadt, Germany) at a mixing speed of 28 000 RPM for 3 min. The homogenization process resulted in a small increase in temperature (~5
C) of the finalized emulsion and the sample was therefore subsequently equilibrated back to 55
C. The oil-free gels were not homogenized, since small strain oscillatory measurements (see Section 2.6) gave insignificant differences in rheological properties between homogenized and non-homogenized samples. Air bubbles were removed by degassing the prepared samples in a vacuum chamber (Nalge company, Rochester, New York, USA) connected to a vacuum pump (Diaphragm Vacuum Pump, Wertheim, Germany). Oil-free gels were prepared by the same

Table 1 Information about the different gelatins used in this study, with reported Bloom strength given by Gelita. Molecular weights for the gelatins have earlier been reported (Hattrem, Molnes, et al., 2014). The sample names listed are used throughout this work. Type

Source

Bloom (g)

Batch number

Sample name

Mna

Mwa

A A A B B B

Pig skin Pig skin Pig skin Limed bovine bone Limed bovine bone Limed bovine bone

160 200 260 160 200 260

629586 630396 628305 630473 626440 630507

160A 200A 260A 160B 200B 260B

36.1 59.4 88.7 58.1 73.0 118.8

90.9 148.2 191.3 161.1 174.6 217.2

a

These values have earlier been reported by (Hattrem, Molnes, et al., 2014).

702

M.N. Hattrem et al. / Food Hydrocolloids 43 (2015) 700e707

procedure as described above, but without the addition of any corn oil. The heated gelled emulsion and oil-free gels were either characterized directly after preparation by small strain oscillatory measurement (see Section 2.6) or casted into steel wells for characterisation of large deformation properties (Section 2.7). Uniform casted gelled emulsions and oil-free gels were prepared using cylindrical steel wells (NTNU Workshop, Norway) with a diameter of 16 mm and height of 19 mm. Prior to casting, the wells were treated with a surfactant solution (0.5 wt.% polysorbate 80 in ethanol) to avoid any adhesion between the wells and the gels. Heated sample was added to the steel wells until a positive meniscus was obtained. A silicone plate was used to cover the sample and pressure was exerted on the silicone by a heavy steel plate mounted at the top. This gave equally shaped gels with specific height and diameter. The samples were stored for 24 h at 22
C prior to the measurement. The influence of pH on the rheological properties was investigated. Gelatin based gelled emulsions (40 wt.% corn oil) and oil-free gels were prepared as described above in combination with either 260A or 260B gelatin, in which a 0.2 M acetic acid buffer (adjusted to pH 4 by the addition of 1.0 M NaOH) or a 0.2 M Trizma base® buffer (adjusted to pH 8 by the addition of 1.0 M HCl) was used as the continuous water phase. 2.3. Measurement of droplet size of the gelled emulsions The droplet size distributions of the gelatin based emulsions were determined by particle size measurements using a Malvern Mastersizer 3000 (Worcestershire, UK) connected to a Hydro MV, wet dispersion unit (Malvern, Worcestershire, UK) and the data was analysed using the respective software (Mastersizer 3000, v1.0.1). The prepared gelled emulsions were dissolved and diluted in MQwater (1:100) at 50
C. The refractive index of water and corn oil was set to 1.33 (solvent) and 1.47 (dispersed phase), respectively. The absorption index of the dispersed droplets was set to 0.01. In order to avoid multiple scattering or low intensity of the scattered light, the dissolved emulsion was added to the dispersion unit (containing ~125 mL water), until an obscuration of approximately 10% was obtained.

2.6. Rheological characterisation by small strain oscillatory measurements Rheological analyses on the prepared gelled emulsions and oilfree gels were performed using a general purpose Rheometer (Stresstech Reologica, Lund, Sweden) with plate-cone geometry (4
, diameter of 40 mm, truncated gap of 150 mm). The rheometer was operated in a strain controlled mode of 0.005 and frequency set to 1 Hz. The chosen strain was confirmed to be within the linear viscoelastic region for all samples. Instrument calibration (zero gap) was performed at 60
C prior to analysis. Approximately 2 g of melted gelatin emulsion was applied and excess sample was removed. In order to avoid evaporation, the applied sample was covered with silicone oil (10 cS fluid, Dow Corning, UK) prior to measurement. The viscoelastic properties of the sample were obtained by using a temperature gradient (2
C/min), with a start and end temperature at 60
C and a holding time of 15 min at 20
C. The gelling and melting temperatures of the samples were estimated as the temperature at which the phase angle corresponded to 45
in the cooling and heating process, respectively. G0 (Pa) was determined as the last measuring point during curing at 20
C. Gelled emulsions and oil-free gels prepared using cold water fish gelatin were characterised by a similar procedure, only changing the curing temperature to 10
C. 2.7. Large deformation properties characterized by a puncture study A puncture test was performed on the gelled emulsion and oilfree gels using a texture analyzer (TA.XT.-Plus Texture Analyzer from Stable Micro Systems, Surrey, UK) equipped with a cylindrical probe (p/2, diameter ¼ 2 mm) operated at a speed of 0.1 mm/s. The gelled emulsions and oil-free gels were prepared as described in Section 2.2, and height and diameter of each sample were manually controlled using a digital calliper. A load cell with a capacity of 5 kg was used and the instrument was calibrated prior to measurement. The measurements were recorded using the corresponding software and force and strain at puncture were determined. The measurements were performed at ambient conditions and 2 independent series were measured with 8 replicates.

2.4. Measurements of dynamic interfacial tension

3. Results and discussion

The dynamic interfacial tension of gelatin solutions/MQ-water and corn oil was measured using a computer controlled Sigma tensiometer 70 (KSV Instruments, Finland) in combination with a Du Noüy ring probe. The probe and sample vessel (40 mm small glass vessel) were cleaned prior to usage by toluene, acetone, water and ethanol and the probe was flame dried using a Bunsen burner. Gelatin solutions were prepared by dissolving gelatin (0.5 wt.% of 160A, 200A, 160B or 200B gelatin) in a water phase containing NaN3 (0.02 w/v %) to prevent microbial spoilage. The probe was immersed in the aqueous solution (15 mL) and corn oil (15 mL) was carefully applied on top of the water phase. The interfacial tension was monitored for a period of 42 h using a probe speed of 0.1 mm/ min at ambient conditions. Two replicates were performed.

In this study, 3 alkaline (type B) and 3 acid (type A) pretreated gelatins were used, with reported bloom values of 160, 200 and 260 g. In total 6 different gelatins were investigated with the sample names (160A, 160B, 200A, 200B, 260A, 260B) earlier given in Table 1.

2.5. Measurements of pH for gelatin solutions The pH of the gelatin solutions prepared in Section 2.4 were measured using a VWR (Radnor, PA, USA) sympHony SB90M5 Benchtop pH Meter connected to a SenTix62 pH electrode (WTW, Weilheim, Germany). Prior to measurements, calibration of the pHmeter was performed in buffered solution of pH 4 and 7. Two replicates were performed at ambient conditions.

3.1. Gelling kinetics The gelling kinetics of the gelled emulsions and oil-free gels were monitored by performing small strain oscillatory measurements in combination with a temperature sweep. The gelling kinetics were monitored by evaluating changes in the phase angle during a cooling and heating process, which may be used for the determination of setting and melting temperatures. This is illustrated in Fig. 1, in which the phase angle of gelled emulsions (40 wt.%) and oil-free gels prepared with either 260A or 260B gelatin is shown. As observed in Fig. 1, the gelled emulsion (40 wt.% corn oil) prepared with the 260A-gelatin, exhibited a gradual transition from a sol to a gel state, with a higher setting and melting temperature (d ¼ 45, 1 Hz) observed compared to the other gels. Similar atypical gelling kinetics was also observed for the gelled emulsions (40 and

M.N. Hattrem et al. / Food Hydrocolloids 43 (2015) 700e707

expected that the proposed droplet interaction was not caused by electrostatic effects. As seen from the gelling kinetics for the 260Aemulsion gels, a decline in the phase angle, i.e. increase in solid like behaviour is observed above the sol-gel transition temperature of the gelatin. This temperature dependent behaviour may indicate a hydrogen-bond mediated interaction. It is known that asparagine is able to form hydrogen bonds with the peptide backbone (Vijayakumar, Qian, & Zhou, 1999). Since only type A gelatin contains such amino acids, it may be suggested that these amino acids contribute to the formation of hydrogen bonds promoting flocculation of the oil droplets. Light microscopy examination of dissolved and diluted emulsion (type A stabilized), confirmed the presence of flocs above the helix-to-coil temperature of the gelatin (data not included). The larger deviation in setting and melting temperatures between the gelled emulsions (40 and 50 wt.% corn oil) and the oilfree gels prepared with type A higher bloom gelatins (Table 2) may be explained by the higher average molecular weight of the gelatin (Table 1), as larger molecular chains present a higher potential for bridging flocculation (Fan, Turro, & Somasundaran, 2000).

90 80 Phase angle (°)

70 60 50 40 30 20 10 0 20

25

40wt.%-Type A

30

35 40 45 Temperature (°C)

40wt.%-Type B

50

55

Oil-free gel - Type A

703

60

Oil-free gel - Type B

Fig. 1. Measured phase angle for gelled emulsions (40 wt.% corn oil) and oil-free gels prepared using acid (type A) or alkali (type B) pretreated gelatin (25 wt.% gelatin of aqueous water phase, 260 Bloom) by performing small strain oscillatory measurements during a cooling-heating process.

50 wt.% corn oil) prepared with the 160A and 200A gelatins (graphs not presented). At lower fractions of oil (10, 20 wt.% corn oil), this effect was however not observed. In order to easier interpret the gelling kinetics, the setting and melting temperatures of the gelled emulsions (10, 20, 40 and 50 wt.% oil) and oil-free gels prepared with different gelatins (160A, 160B, 200A, 200B, 260A and 260B) are shown in Table 2. For type A-gelled emulsions, increased setting and melting temperatures were recorded as a function of oil content (i.e. 40 and 50 wt.%). However, only small differences in the sol/gel transition temperatures were observed for the gelled emulsions prepared with type B gelatins. A plausible explanation would be to attribute the differences in gelling kinetics to a physical interaction between the oil droplets. The main difference between type A and B gelatins is the transformation of glutamine and asparagine into their acid precursors for the alkaline treated samples. This change in the amino acid composition gives a lower IEP as well as a higher charge density for the type B gelatin. To investigate if the difference in gelling kinetics was caused by electrostatic interaction between the droplets, 0.2 M acetic acid (pH 4 e Fig. 2A) and 0.2 M Trizma base (pH 8 e Fig. 2B) buffered gelled emulsions (40 wt.% oil) and oil-free gels were prepared using either 260A or the 260B gelatin. In addition, gelled emulsions (40 wt.%) and oil-free gels prepared without any buffering agents (ambient pH measured in the range of 5.5e6) were used as a comparison (Fig. 2C). As observed in Fig. 2AeC the gelling kinetics of the different gels were independent of the pH in the range of 4e8. Thus, it would be

3.2. Dynamic interfacial tension measurements As previously described, both the interfacial tension (IFT) and the droplet size of the dispersed phase may influence the moduli of the gelled emulsions. The adsorption of gelatin at the oil-water interface is a dynamic process, in which equilibrium is obtained after a certain time. It is further known that larger surface active macromolecules, such as gelatin, usually have a longer equilibra€ bius & tion times compared to small molecular surfactants (Mo Miller, 1998). To evaluate the dynamic properties at the oil-water interface, interfacial tension (Fig. 3) was monitored during 42 h for a corn oil-aqueous solution (MQ-water or 0.5 wt.% of either 160A, 200A, 160B or 200B gelatin) using a Du Noüy ring tensiometer. The measurements were performed at ambient conditions. The 260 bloom gelatins were not measured by this procedure, since 0.5 wt.% is above their critical overlap concentration. As seen in Fig. 3, an equilibrium interfacial tension was not obtained during a period of 42 h for the type B gelatins. It is further observed that the type A gelatins have a much steeper decline in interfacial tension. The measurements were stopped after 5 h due to a limitation in measuring IFT lower than ~4 mN/m using the present procedure. As stated by Schrieber and Gareis, approximately 5% of foreign proteins (non-collagenous) and porcine fat remains as contaminants after acid extraction of gelatin. It is further

Table 2 Setting and melting temperatures of gelled emulsions (10, 20, 40 and 50 wt.%) and oil-free gels prepared with different gelatins (160A, 160B, 200A, 200B, 260A and 260B gelatins) characterized by small strain oscillatory measurements. Five replicates were performed on each gel. Amount of oil (wt.%)

160A

200A

260A

160B

200B

260B

Setting temperature (Tg)

0 10 20 40 50

28.8 28.2 28.3 29.3 39.5

± ± ± ± ±

0.4 0.1 0.2 0.4 0.3

28.6 28.6 29.0 34.5 49.8

± ± ± ± ±

0.2 0.1 0.2 2.7 2.3

31.8 31.9 32.4 46.9 ***

± ± ± ±

0 10 20 40 50

36.1 35.5 35.7 39.5 46.0

± ± ± ± ±

0.4 0.2 0.0 0.3 2.8

35.7 35.9 36.6 45.1 55.3

± ± ± ± ±

0.1 0.2 0.3 2.4 2.3

39.5 39.2 40.5 52.6 ***

± ± ± ±

0.1 0.1 0.2 2.9

29.3 29.5 29.5 29.6 30.3

± ± ± ± ±

0.9 0.1 0.1 0.1 0.2

30.0 29.9 30.2 30.4 31.4

± ± ± ± ±

0.4 0.0 0.1 0.2 0.8

30.7 31.1 31.2 31.5 32.4

± ± ± ± ±

0.1 0.1 0.1 0.3 0.2

0.7 0.1 0.1 0.1 0.1

36.8 36.9 37.1 37.5 38.7

± ± ± ± ±

0.2 0.1 0.1 0.1 0.6

37.6 37.8 37.8 38.5 39.6

± ± ± ± ±

0.3 0.3 0.1 0.3 0.5

Melting temperature (Tm)

*** 260A with 50wt.% oil had a Tg and Tm higher than 60
C.

0.2 0.1 0.3 1.8

36.3 36.8 36.6 36.9 37.5

± ± ± ± ±

M.N. Hattrem et al. / Food Hydrocolloids 43 (2015) 700e707

90 80 70 60 50 40 30 20 10 0

A Phase angle (°)

Phase angle (°)

704

Phase angle (°)

20

30

40 50 Temperature (°C)

60

90 80 70 60 50 40 30 20 10 0

B

20

30

40

50

60

Temperature (°C)

C

90 80 70 60 50 40 30 20 10 0 20

30

40

50

260A-40%

260B-40%

260A-0%

260B-0%

60

Temperature (°C) Fig. 2. Measured phase angle (d(
)) for gelled emulsions (40 wt.% corn oil) and oil-free gels prepared with either a type A or type B gelatin (25 wt.% of the continuous water phase) with a reported Bloom value of 260 by performing small strain oscillatory measurements during a cooling-heating process A) The continuous water phase buffered with an 0.2 M acetic acid buffer (pH 4). B) The continuous water phase buffered with a 0.2 M Trizma base buffer (pH 8) and C) Unbuffered gelled emulsion (control sample). Three measurements were performed on each gel.

stated that for the alkali treated gelatins almost all foreign protein is removed before extraction, giving a high purity of the prepared gelatin (Schrieber & Gareis, 2007). The reported impurities would most likely be surface active, which may explain the observed differences between the alkali and acid pretreated gelatins. The interfacial tension between the gelatin solutions and corn oil may have been influenced by the electrostatic properties of the gelatin molecules. The measurements were performed at ambient conditions with pH of the aqueous solutions measured in the range of 5.5e6. As summarised by Finch and Jobling, studies have indicated a minimum in the interfacial tension at the isoelectric point for the gelatin (Finch & Jobling, 1977). However, by considering that a higher IFT was obtained for the aqueous solutions prepared with type B gelatins (IEP~5) compared to type A-gelatins (IEP ~ 7e9), the difference in IEP does not seem to explain the present results. The molecular weight distributions are also known to influence the

3.3. Measured and estimated shear storage modulus

16 Interfacial tension (mN/m)

adsorption kinetics (Beverung, Radke, & Blanch, 1999; Young, Carroad, & Bell, 1980), in which smaller molecules may diffuse more rapidly to a newly formed interface. This may explain the small difference in IFT between the gelatin solutions prepared with 160B (Mn ~ 58 kda) and 200B (Mn ~ 73 kda) gelatin. Even though type A gelatins are known to contain a small fraction of very low molecular weight peptides (Schrieber & Gareis, 2007), it is not anticipated that this fraction can explain the very large difference in IFT between type A and B gelatins. Similar long equilibration times as for the alkali treated gelatin have earlier been reported for a hydrophobically modified pullulan. The pullulan sample is polydisperse with a high molecular mass similar to the gelatin samples. By performing neutron reflectivity measurements, the long equilibration time was suggested to be caused by a structural rearrangement at the airewater interface (Deme & Lee, 1997). For the present system a similar explanation seems plausible.

14 12

160B

10

160A

8 6

200A

4

200B

2

MQ-Water

0 0

10

20

30

40

Time (hours) Fig. 3. Dynamic interfacial tension measured at ambient conditions for a corn oilaqueous solution (MQ-water or 0.5 wt.% of either 160A, 200A, 160B or 200B gelatin) using a Du Noüy ring tensiometer. Two replicates were performed.

An estimate of the storage modulus for the gelled emulsions and oil-free gels may be calculated using a simplified version of the van der Poel's formula. In order to use this formula, an estimate is needed for the interfacial tension of the corn oil and gelatin solutions. It would be convenient to use the equilibrium interfacial tension for a gelatin solution and corn oil. However, an important consideration is that an emulsion would have an extensively large surface area compared to an unmixed oil-water interface. This increased interfacial area would lower the concentration of contaminants per unit area, reducing its influence on the interfacial tension. Further, since gelatin contains peptides with variation in molecular size and amino-acid composition (i.e. varying affinity for oil-water interface), an unmixed oil and water solution would at best give a minimum value for the interfacial tension of an emulsion. Henceforth, a value of 5 mN/m was decided used as an average

M.N. Hattrem et al. / Food Hydrocolloids 43 (2015) 700e707

705

Table 3 The Sauter mean diameter (D[3,2], mm) for the gelatin emulsions with varying amounts of corn oil (10, 20, 40, 50 wt.%). D[3,2] is represented as the average of five replicates, with each replicate being the average of five measurements. Type gelatin

160A

160B

200A

200B

260A

260B

Oil (wt.%)

D[3;2]

D[3;2]

D[3;2]

D[3;2]

D[3;2]

D[3;2]

10% 20% 40% 50%

0.98 0.96 0.95 0.93

0.91 0.88 0.99 1.00

± ± ± ±

0.06 0.03 0.04 0.17

0.80 0.80 0.78 0.68

interfacial tension of the dispersed droplets for both the alkali and acid pretreated gelatins. The droplet sizes of the gelled emulsions had to be determined in order to calculate the storage moduli. These values are reported in Table 3, as the Sauter mean diameter (D[3,2], mm) for the gelled emulsions (10, 20, 40 and 50 wt.% corn oil) prepared with different types of gelatin (160A, 160B, 200A, 200B, 260A and 260B). As stated earlier the shear storage modulus of filled gels (G0 ) may be estimated using a simplified version of van der Poel's formula (equation (2)) (Poel, 1958; Smith, 1975).

G0 15ð1  vm ÞðM  1Þf 1¼ G0m ð8  10vm ÞM þ 7  5vm  ð8  10vm ÞðM  1Þf

(2)

In equation (2), M equals the ratio between the shear storage modulus of the dispersed droplets and the matrix (Gf/Gm). The shear modulus of the dispersed droplets is estimated to be equal to the Laplace pressure (equation (1)) of the droplets as suggested by van Vliet (van Vliet, 1988). Hence, filler moduli (Gf) were calculated using the measured droplet sizes and estimated interfacial tension (5 mN/m). By assuming an incompressible material, the Poisson's ratio (vm) is estimated to be 0.5. The shear moduli (Gm) for the gelled emulsions (10, 20, 40 and 50 wt.% corn oil) and oil-free gels were obtained by performing small strain oscillatory measurements during a heating-cooling-heating process. By using equation (2), estimated shear moduli were calculated for different amounts of oil (f). The calculated shear moduli are presented using linear fit (Fig. 4) d and compared to the measured (represented by single point symbols) shear moduli for the gelled emulsions and oil-free gels prepared with either type A (Fig. 4A) or type B (Fig. 4B) gelatin. As observed in Fig. 4, an increase in modulus was measured as a function of oil content for gelled emulsions prepared with both type A and type B gelatins. The increase in modulus as a function of oil content is higher for the gelled emulsions prepared with the type A gelatins. Further, only the gelled emulsions prepared with the alkali pretreated gelatins correlated well with the estimates calculated from equation (2). The theoretical assumption behind the van der Poel's theory is that the filler particles are evenly distributed and do not interact (Poel, 1958). Thus, the higher modulus measured for the gelled emulsions prepared with the type A gelatin may be due to the earlier described oil droplet flocculation for these gels. Flocculation can lead to the formation of a structural network, which may promote a reinforcement of the gelled emulsions. A larger deviation from the estimate is observed at higher fractions of the oil. Further, as described in Section 3.1, increased setting and melting temperatures were only observed at higher amounts of dispersed phase (40 and 50 wt.%). These observations suggest that the flocculation mainly contribute to the structural properties at higher packing densities. Chen and Dickinson also reported higher moduli compared to theoretical estimates for a whey protein emulsion gel (Chen & Dickinson, 1999). This was suggested to be caused by flocculation, leading to a higher effective volume of the flocs compared to the individual droplets. This explanation would be reasonable provided a large difference

± ± ± ±

0.02 0.04 0.06 0.11

0.80 0.89 0.85 0.79

± ± ± ±

0.03 0.03 0.10 0.09

0.74 0.69 0.63 0.60

± ± ± ±

0.04 0.01 0.06 0.00

0.77 0.76 0.72 0.69

± ± ± ±

0.02 0.03 0.06 0.00

between the modulus of the filler and matrix (earlier described by the coefficient M). However, for the present system the small difference in modulus between the filler and matrix implies that this may only partially explain the deviation from the estimated moduli. The molecular weight distributions of the gelatins used in this study have earlier been characterized by size exclusion chromatography multi angle light scattering (SEC-MALLS) and the gelling properties as a function of molecular weight have been discussed (Hattrem, Molnes, & Draget, 2014). A correlation between the Bloom values and shear moduli was observed, which was expected since both terms represent an elastic response. The increased shear moduli (Fig. 4) observed for the oil-free gelatin gels with increasing Bloom was suggested to be caused by higher average molecular weights for these gelatins. In the study by Lorenzo et al. increasing elastic properties were obtained for gelled emulsions with increasing molecular weights for alkaline pretreated gelatins (Lorenzo et al., 2011). This is in accordance with the present results,

40000

A

35000 30000 25000 G' (Pa)

0.04 0.01 0.06 0.00

20000 15000 10000 5000 0 0

10 160A-Measured 160A-Es mated

20 30 40 Amount of corn oil (wt.%)

50

200A-Measured 200A-Es mated

260A-Measured 260A-Es mated

60

30000

B

25000 20000 G' (Pa)

± ± ± ±

15000 10000 5000 0 0

10 160B-Measured 160B-Es mated

20 30 40 Amount of corn oil (wt.%) 200B-Measured 200B-Es mated

50

60

260B-Measured 260B-Es mated

Fig. 4. Measured (presented by single point symbols) and estimated (equation (2), linear fit of the estimated values) shear storage moduli (G0 (Pa)) for gelled emulsions (10, 20, 40 or 50 wt.% corn oil) or oil-free gels after 15 min of curing at 20
C by performing small strain oscillatory measurements during a cooling-heating process. The gelled emulsions were prepared with either type A (Fig. 4.A) or type B (Fig. 4.B) gelatin with varying reported Bloom values (160, 200, 260 g), in which the water phase contained 25 wt.% of gelatin. Five measurements were performed on each gel.

706

M.N. Hattrem et al. / Food Hydrocolloids 43 (2015) 700e707

however, as earlier shown, this would only be valid when comparing the gelled emulsions prepared with either the type A or type B gelatin.

90

6000

80 5000 4000

Large deformation properties for the emulsion and oil-free gels were investigated by performing a puncture study. This type of test is widely used for investigating the texture of foods (Bourne, 2002). The tests were performed by compressing a small area of the sample and the force (Fig. 5A) and strain (Fig. 5B) at puncture were determined. As seen in Fig. 5, force and strain at penetration for the gels prepared with 160 and 200 A/B gelatins were almost independent of the fraction oil. However, for the gels prepared with the 260 bloom gelatins (type A and B), a decline in stress and strain was observed with increasing amount of oil. This was most prominent for the gels prepared with the acid pretreated gelatin. A stress concentration would occur at the contact point between the probe and the sample during initial deformation. With increasing deformation, the stress concentration would eventually provide a dissipation of energy by unzipping of physical bonds in the gelatin structural network (van Vliet & Walstra, 1995). The higher Bloom gelatins are expected to have a higher density of polypeptide junction zones compared to the samples with lower Bloom (Joly-Duhamel, Hellio, Ajdari, & Djabourov, 2002). This may explain the larger force at puncture observed for the 260 Bloomgelatin gels. For the gelled emulsions, on the other hand, oil droplets may act as structural defects, leading to stress concentration. In addition, fracture may occur at the oil-water interface (Sala, van Vliet, Stuart, van Aken, et al., 2009). The force required to disrupt oil droplets from the gelatin matrix would probably be of similar magnitude for the different gelatins used in this study. Hence, if initial failure involves unzipping of physical bonds at the gel-oil interphase, it may be suggested that the puncture stress

Force at puncture (N)

6

A

5 4 3 2 1 0 160A

160B 200A 200B 260A Type of gela n (bloom and acid/alkali preatreated) 0%

10%

20%

40%

260B

50%

Strain at puncture

60%

B

50% 40% 30% 20% 10% 0% 160A

160B

200A

200B

260A

260B

Type of gela n (bloom and acid/alkali preatreated) 0%

10%

20%

40%

50%

Fig. 5. Puncture studies performed on gelled emulsions (10, 20, 40 or 50 wt.% corn oil) or oil-free gels prepared with either type A or type B gelatin (25 wt.% of the continuous water phase) with varying reported Bloom (160, 200, 260 g), in which force (5.A) and strain (5.B) at puncture were determined. Two independent replicates were performed on each gel, with 8 measurements for each replicate.

G' (Pa)

3.4. Large deformation properties characterized by puncture studies

60 50

3000 40 2000

30 20

Temperature (°C); δ (°)

70

1000 10 0

0 0

10

20

30

40

50

60

70

Time (min) G'(Pa)-Oil-free gel Temperature δ - Gelled emulsion (40 wt.% corn oil)

G'(Pa) - Gelled emulsion (40 wt.% Corn oil) δ - Oil-free gel

Fig. 6. Measured phase angle (d) and shear storage modulus (Pa) for a gelled emulsion (40 wt.% corn oil) and oil-free gels prepared with a high molecular weight cold water fish gelatin (type A, 25 wt.% gelatin of the aqueous phase) characterized by performing small strain oscillatory measurements during a cooling-heating process. Two replicates were performed for each system.

would converge to a similar value with increasing fractions of oil for all gelatin types. This seems to be valid for the present results since the 260 bloom samples converge towards a similar puncture stress as those obtained for the low Bloom samples at higher oil contents. By considering both the large and small strain deformation studies, the addition of filler particles may be used to manipulate the mechanical properties of gelatin gels. An earlier study by Bot et al., concluded that the only viable way to manipulate the breaking strain permanently for a gelatin gel was the addition of filler particles (Bot, van Amerongen, Groot, Hoekstra, & Agterof, 1996). There has been a large focus on improving the rheological properties of cold water fish gelatin (CWFG) gels, especially emphasising the suboptimal setting temperatures and gel strength. Fish gelatins are known to be mainly produced by acid pretreatment and considering the previous observations, it would be of interest to investigate the rheological properties of a CWFG based emulsion gel. To investigate this further, a gelled emulsion (40 wt.% corn oil) and an oil-free gel were prepared using a high molecular weight CWFG (25 wt.% gelatin of the aqueous phase). The samples were compared using small oscillatory measurements during a cooling-heating process. The shear storage modulus and phase angle for the gels are reported in Fig. 6. As observed in Fig. 6, a decline in the phase angle is observed for the gelled emulsion at temperatures above the helix-to-coil transition temperature of the gelatin. As previously discussed this seems to indicate a hydrogen-bond mediated flocculation. In addition, the presence of oil droplets gives an extensively higher shear storage modulus compared to the oil-free gels. A smaller difference in modulus between the emulsion and oil-free gels were measured for the samples prepared using mammalian gelatins (Fig. 4). This is due to a larger difference between the filler and matrix modulus for the fish gelatin based emulsion (larger value of M), giving rise to a larger increase in shear modulus as a function of oil content. This result indicates that it is possible to tailor the rheological properties of fish gelatin gels by the introduction of oil droplets. 4. Conclusion In this study the rheological and interfacial properties of gelatin based emulsions prepared by either acid or alkali pretreated

M.N. Hattrem et al. / Food Hydrocolloids 43 (2015) 700e707

gelatins have been characterized. Large differences in gelling properties were obtained by using different types of gelatin. At higher fractions of dispersed oil, significant increases in setting and melting temperatures were obtained for the gelled emulsions prepared with the acid pretreated gelatins. In comparison, the sol/ gel transition temperatures for the gelled emulsions prepared with type B-gelatin were independent of the amount of oil. Both modulus and fracture stress for the gelatin based solid emulsions were influenced by the presence of oil droplets. It was observed a larger increase in shear storage modulus for the gelled emulsions prepared with acid pretreated gelatins as a function of oil content. These results suggest that it is possible to tailor gels with specific mechanical properties. From interfacial tension measurements large differences in surface activity was observed between type A and type B gelatins, possible due to a minor fraction of surface active contaminants in type A gelatin. These impurities had a large impact on the dynamic interfacial tension between a gelatin solution and corn oil, which may potentially have a large influence on the properties and preparation procedure of gelatin based emulsions. Finally, a gelled emulsion was prepared using a cold water fish gelatin and it was shown that potentially improved gelling properties may be obtained by the introduction of oil droplets.

Acknowledgement The authors would like to thank Morten J. Dille for very valuable discussions and Sebastian Simon and Bicheng Gao for guidance and discussion regarding the IFT measurements. MNH would also like to thank the Research Council of Norway for financial support, project-nr: 212104/O30.

References Beverung, C. J., Radke, C. J., & Blanch, H. W. (1999). Protein adsorption at the oil/ water interface: characterization of adsorption kinetics by dynamic interfacial tension measurements. Biophysical Chemistry, 81(1), 59e80. Bot, A., van Amerongen, I. A., Groot, R. D., Hoekstra, N. L., & Agterof, W. G. M. (1996). Large deformation rheology of gelatin gels. Polymer Gels and Networks, 4(3), 189e227. Bourne, M. C. (2002). Food texture and viscosity concept and measurement. In Food science and technology international series (2nd ed.). San Diego: Academic Press. Chen, J. S., & Dickinson, E. (1999). Effect of surface character of filler particles on rheology of heat-set whey protein emulsion gels. Colloids and Surfaces B-biointerfaces, 12(3e6), 373e381. Chen, J. S., Dickinson, E., Langton, M., & Hermansson, A. M. (2000). Mechanical properties and microstructure of heat-set whey protein emulsion gels: effect of emulsifiers. Lebensmittel-wissenschaft Und-technologie-food Science and Technology, 33(4), 299e307. Deme, B., & Lee, L. T. (1997). Adsorption of a hydrophobically modified polysaccharide at the air-water interface: Kinetics and structure. Journal of Physical Chemistry B, 101(41), 8250e8258. Dickinson, E. (2012). Emulsion gels: the structuring of soft solids with proteinstabilized oil droplets. Food Hydrocolloids, 28(1), 224e241.

707

Dickinson, E., Stainsby, G., & Wilson, L. (1985). An adsorption effect on the gel strength of dilute gelatin-stabilized oil-in-water emulsions. Colloid and Polymer Science, 263(11), 933e934. Fan, A. X., Turro, N. J., & Somasundaran, P. (2000). A study of dual polymer flocculation. Colloids and Surfaces A-physicochemical and Engineering Aspects, 162(1e3), 141e148. Finch, C. A., & Jobling, A. (1977). The physical properties of gelatin. In A. G. Ward, & A. Courts (Eds.), The science and technology of gelatin. London; New York: Academic Press, (pp. xvi, 564 p.). Hattrem, M. N., Dille, M. J., Seternes, T., & Draget, K. I. (2014). Macro- vs. micromolecular stabilisation of W/O/W-emulsions. Food Hydrocolloids, 37(0), 77e85. Hattrem, M. N., Molnes, S., & Draget, K. I. (2014). Investigation of physico-chemical properties of gelatin matrices in correlation with dissolution studies. In Gums and stabilisers for the food industry 17: The changing face of food manufacture: The role of hydrocolloids (pp. 369e376). The Royal Society of Chemistry. Haug, I. J., Sagmo, L. B., Zeiss, D., Olsen, I. C., Draget, K. I., & Seternes, T. (2011). Bioavailability of EPA and DHA delivered by gelled emulsions and soft gel capsules. European Journal of Lipid Science and Technology, 113(2), 137e145. Joly-Duhamel, C., Hellio, D., Ajdari, A., & Djabourov, M. (2002). All gelatin networks: 2. The master curve for elasticity. Langmuir, 18(19), 7158e7166. Kim, K. H., Renkema, J. M. S., & van Vliet, T. (2001). Rheological properties of soybean protein isolate gels containing emulsion droplets. Food Hydrocolloids, 15(3), 295e302. Lorenzo, G., Checmarev, G., Zaritzky, N., & Califano, A. (2011). Linear viscoelastic assessment of cold gel-like emulsions stabilized with bovine gelatin. LWT-Food Science and Technology, 44(2), 457e464. €bius, D., & Miller, R. (1998). Proteins at liquid interfaces. Amsterdam; New York: Mo Elsevier. Poel, C. (1958). On the rheology of concentrated dispersions. Rheologica Acta, 1(2e3), 198e205. Sala, G., Van Aken, G. A., Stuart, M. A. C., & Van De Velde, F. (2007). Effect of dropletmatrix interactions on large deformation properties of emulsion-filled gels. Journal of Texture Studies, 38(4), 511e535. Sala, G., van Vliet, T., Stuart, M. A. C., van Aken, G. A., & van de Velde, F. (2009). Deformation and fracture of emulsion-filled gels: effect of oil content and deformation speed. Food Hydrocolloids, 23(5), 1381e1393. Sala, G., van Vliet, T., Stuart, M. C., van de Velde, F., & van Aken, G. A. (2009). Deformation and fracture of emulsion-filled gels: effect of gelling agent concentration and oil droplet size. Food Hydrocolloids, 23(7), 1853e1863. Sato, A. C. K., Moraes, K. E. F. P., & Cunha, R. L. (2014). Development of gelled emulsions with improved oxidative and pH stability. Food Hydrocolloids, 34(1), 184e192. Schrieber, R., & Gareis, H. (2007). Gelatine handbook: Theory and industrial practice. Weinheim Chichester: Wiley-VCH, John Wiley distributor. Smith, J. C. (1975). Simplification of van der Poel's formula for the shear modulus of a particulate composite. Journal of Research of the National Bureau of Standards, 79A(2). Stainsby, G. (1985). Gelatin gels. In A. M. Pearson, T. R. Dutson, & A. J. Bailey (Eds.), Advances in meat research (Vol. 4). New York: Van Nostrand Reinhold Company. Thakur, G., Naqvi, M. A., Rousseau, D., Pal, K., Mitra, A., & Basak, A. (2012). Gelatinbased emulsion gels for diffusion-controlled release applications. Journal of Biomaterials Science-polymer Edition, 23(5), 645e661. Veis, A. (1964). The macromolecular chemistry of gelatin. New York: Academic Press. Vijayakumar, M., Qian, H., & Zhou, H. X. (1999). Hydrogen bonds between short polar side chains and peptide backbone: prevalence in proteins and effects on helix-forming propensities. Proteins-structure Function and Genetics, 34(4), 497e507. van Vliet, T. (1988). Rheological properties of filled gels - influence of filler matrix interaction. Colloid and Polymer Science, 266(6), 518e524. van Vliet, T., & Walstra, P. (1995). Large deformation and fracture behaviour of gels. Faraday Discussions, 101, 359e370. Ward, A. G., & Courts, A. (1977). The science and technology of gelatin. London ; New York: Academic Press. Young, M. E., Carroad, P. A., & Bell, R. L. (1980). Estimation of diffusion-coefficients of proteins. Biotechnology and Bioengineering, 22(5), 947e955.