Journal of Food Engineering 78 (2007) 1410–1426

www.elsevier.com/locate/jfoodeng

Foaming properties of egg white proteins affected

by heat or high pressure treatment
Iesel Van der Plancken, Ann Van Loey, Marc E. Hendrickx *

Center for Food and Microbial Technology, Faculty of Bioscience Engineering, Katholieke Universiteit Leuven, Kasteelpark, Arenberg 22,
B-3001 Heverlee, Belgium

Received 15 September 2005; accepted 7 January 2006

Available online 28 February 2006

Abstract

This paper deals with the effect of heat (50–85 C) and high pressure treatment (400–700 MPa at 10–60 C) on the foaming properties
of egg white solutions (10% v/v or 9.64 mg protein/mL). These physical treatments (20 min) were performed at two pH levels: pH 7.6
corresponding to the pH of fresh egg white and pH 8.8, corresponding to that of older egg white. Both heat and pressure treatment
affected the foaming properties of egg white proteins. While foams from untreated egg white solutions were crispy and subject to collapse
of the foam column after a long standing period, foams from heat and pressure-treated egg white solutions were moist and creamy, show-
ing smaller bubble size and little or no sensitivity to foam collapse. The effect of physical treatments was strongly dependent on the pH
during treatment. The most voluminous foams were obtained at pH 8.8, while the most stable, dense foams were obtained at pH 7.6, for
both heat and pressure treatment. Based on previous data on the effect of heat and pressure treatment on the physicochemical properties
of egg white proteins, the relationship between the processing-induced changes in these physicochemical properties and the foaming
properties of egg white was investigated. Treatments resulting in a high level of protein unfolding, yet accompanied by a certain degree
of residual protein solubility (primarily at pH 8.8), resulted in egg white solutions with improved foaming ability. A high level of unfold-
ing combined with extensive protein solubility loss (primarily at pH 7.6) was associated with increased foam stability and density. Foams
with high volume and average stability and density were obtained by pressure treatment at pH 8.8 (above 500 MPa). The processing-
induced changes in the foaming properties could not be attributed to the changes in a single physicochemical property. The foaming
ability was in part determined by the sulfhydryl content and protein flexibility. Improved protein–protein interactions (solubility and
exposed SH groups) contributed to increased foam stability of treated egg white solutions. Other properties, not measured in this study,
probably also contribute to the foaming properties of processed egg white solutions, especially after pressure treatment.
 2006 Elsevier Ltd. All rights reserved.

Keywords: Egg white; Foam; Heat-treatment; High hydrostatic pressure; Physicochemical properties

1. Introduction study focuses on the foaming properties of egg white pro-

teins. The excellent foaming capacity of the egg white pro-
It is almost impossible to ignore the significance of eggs tein and the stability of the resulting foams (even when
in the world of food. Chicken egg is a multi-purpose ingre- subjected to heating) are applied in food industry in the
dient, meeting the requirements of a variety of food formu- preparation of, among others, meringues and cakes.
lations. It combines not only excellent foaming, gelling and For most applications in food industry, pasteurization
emulsifying properties, but also a high nutritional quality of of egg white is a prerequisite, because of possible contam-
its proteins (Hatta, Hagi, & Hirano, 1997; Mine, 1995). This ination with pathogen Salmonella sp. (Cunningham, 1986).
It is known that heat-treatment can induce protein dena-
turation, which can, depending on the severity of the pro-
*
Corresponding author. Tel.: +32 16 32 15 72; fax: +32 16 32 19 60. cess, result in changes in the functional properties of the
E-mail address: [email protected] (M.E. Hendrickx). treated proteins. The correlation between the structural

0260-8774/$ - see front matter  2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jfoodeng.2006.01.013
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1411

properties of egg white proteins and their functional prop- aggregates that are only partly stabilized by disulfide bonds
erties has been the subject of many studies, with an empha- (resulting from SH–SS exchange mechanisms), but more
sis on the contribution of the major egg white protein, importantly by hydrophobic interactions. The larger size
ovalbumin (Donovan, Mapes, Davis, & Garibaldi, 1975; of the aggregates formed at elevated temperature at ambi-
Kato, Fujimoto, Matsudomi, & Kobayashi, 1986; Mine, ent pressure might explain why the resulting suspensions
Noutomi, & Haga, 1990; Van der Plancken, Delattre, are more turbid (Van der Plancken et al., 2005b).
Indrawati, Van Loey, & Hendrickx, 2004; Van der Planc- Pressure-treated egg white proteins are more flexible, as
ken, Van Remoortere, Indrawati, Van Loey, & Hendrickx, evidenced by the higher susceptibility to enzymatic hydro-
2003; Van der Plancken, Van Loey, & Hendrickx, 2005b, lysis compared to the untreated protein (Van der Plancken
2006). In these studies much attention was paid to the et al., 2004). An antagonistic effect between pressure and
changes in structural properties of ovalbumin or the other temperature is observed, since at elevated pressure the
egg white proteins induced by heating, leading to changes changes in physicochemical properties are faster at lower
in the functional properties. Heating of egg white solutions temperature, while at atmospheric pressure this occurs at
in the temperature range of 50–85 C results in significant elevated temperature (Suzuki, 1960; Van der Plancken
unfolding of the proteins, as evidenced by an exposure of et al., 2004, in press). These pressure-induced changes in
hydrophobic groups and sulfhydryl (SH) groups, priorly protein structure and physicochemical properties will affect
buried in the protein core, and a higher sensitivity towards the foaming properties of the pressurized egg white solu-
proteases. The decrease in denaturation enthalpy indicates tions. As the mechanism of protein denaturation appeared
loss of secondary protein structure (Ma & Harwalkar, to be different for heat and pressure treatment, it can be
1991; Privalov & Khechinashvili, 1974). Depending on expected that this difference is reflected in the foaming
the pH during heat-treatment, these changes result in mas- properties of the treated egg white proteins. Thus, similar
sive loss of protein solubility and the formation of turbid to heat-treatment, it will affect the functional properties
protein suspensions after prolonged heating at elevated of the treated egg white. Although several studies have
temperature, due to hydrophobic interactions and the for- been conducted on the pressure-induced gelation of egg
mation of disulfide (SS) bonds through SH/SS exchange white (Bridgman, 1914; Doi, Shimizu, Oe, & Kitabatake,
reactions and SH oxidation (Van der Plancken et al., 1991; Hayashi, Kawamura, Nakasa, & Okinaka, 1989;
2005b). These changes in physicochemical properties due Richwin, Raasch, Teichgraber, & Knorr, 1992), few studies
to heat-treatment are expected to result in changes in the are available on the effect of pressure on the foaming prop-
foaming characteristics of the treated egg white solutions. erties of treated egg white. The existing literature is chiefly
High pressure processing is being more and more sug- focused on the lower pressure range, up to 400 MPa
gested as an interesting alternative for heat-treatment as (Richwin et al., 1992; Strohalm et al., 2000).
this non-thermal processing technique often results in a Therefore, the objective of this work was to compare the
better balance between food safety on the one hand and foaming properties of egg white solutions treated either by
food quality on the other. High hydrostatic pressure can, heat or by high pressure and to relate possible differences in
however, also induce protein denaturation, depending on these properties to differences in the denaturation mecha-
the protein concentration, pressure, temperature and pH nisms of the two preservation techniques. As the pH of
(Balny & Masson, 1993; Barbarosa-Cánovas, Pathakam- egg white increases during storage up to 9.5 (Gu, Matsuda,
ury, Palou, & Swanson, 1997). Similar to heat-treatment, & Nakamura, 1986), the processing-induced changes in all
pressure treatment of egg white proteins above 450 MPa parameters were studied both at the pH of fresh egg white
results in a loss of secondary structure (Hayakawa, Linko, (pH 7.6) and at pH 8.8, corresponding to the pH of older
& Linko, 1996). These pressure-induced structural changes egg white.
in egg white proteins can also be demonstrated by the In our previous work (Van der Plancken et al., 2004,
exposure of previously buried hydrophobic and SH groups 2005a, 2005b; Van der Plancken, Van Loey, & Hendrickx,
and increased flexibility (measured as susceptibility to enzy- 2006, in press), the effect of the heat and pressure treat-
matic hydrolysis) (Iametti et al., 1998, 1999; Van der Planc- ments on the physicochemical properties of egg white solu-
ken et al., 2004, 2005b; Van der Plancken, Van Loey, & tions was discussed. Based on these data, the relationship
Hendrickx, 2005a, Van der Plancken, Van Loey, & Hend- between the processing-induced changes in these physico-
rickx, in press). However, the increase in surface hydropho- chemical properties and the foaming properties of egg
bicity is less pronounced at elevated pressure, compared to white was investigated.
elevated temperature at ambient pressure. On the other
hand, the decrease of exposed SH groups due to SH oxida- 2. Materials and methods
tion is enhanced by pressure. Both observations might
explain why pressure treatment induced protein solubility 2.1. Materials
loss to a lesser extent compared to heat-treatment. Due
to high pressure, smaller aggregates stabilized by disulfide Fresh eggs (68) were obtained from the Zootechnologi-
bonds (arising from SH oxidation) are formed, while cal Centre (K.U. Leuven, Belgium). The egg white (approx-
heat-treatment induces the formation of large, insoluble imately 2.6 L) was separated from the egg yolk and the
1412 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

chalazae were removed. The albumen was gently mixed 2.3. Determination of foaming properties
and stored at 40 C, without any conversion to S-ovalbu-
min, as demonstrated by differential scanning calorimetry Heat-treated or pressurized egg white solutions (10%
measurement (data not shown) (Donovan & Mapes, v/v) were diluted to 8% v/v with demineralized water to
1976). The protein content of the egg white was determined prepare foams. A volume of 50 mL (20 C) foaming solu-
to be 9.64 ± 0.78% (w/v). A 10-fold dilution of the egg tion was placed in a glass beaker of 400 mL (diameter =
white (10% v/v or 9.64 g protein/L) was prepared in 7.7 cm) and whipped for 4 min with a rotating anchor
0.2 M TrisHCl at pH 7.6 and 8.8. Under these conditions, (45 mm diameter) at 1300 rpm and ambient temperature,
no precipitation was observed, but at pH 8.8, a hazy solu- using a laboratory stirrer with microprocessor controlled
tion was obtained. constant speed (OST basic, IKA, Germany). In Fig. 1, a
schematic presentation is shown of the experimental set-
2.2. Physical treatments up. The foam produced and any remaining liquid were
gently transferred into a graduated glass cylinder (3.5 cm
Thermal treatment of 10 mL egg white solution (10% diameter, 31 cm height, graded up to 0.250 L) using a small
v/v) was performed in 14 mL polypropylene tubes with size plastic spatula and plastic, broad neck funnel, within
screw cap (15 mm in diameter, Greiner Bio-One, Fricken- 5 min. Any air pockets were removed by holding up the
hausen, Germany). The tubes were heated for 20 min base of the cylinder with one hand and the top of the cyl-
in a temperature-controlled water bath (Thermo Haake, inder with the other hand. With the cylinder held upright,
Karlsruhe, Germany) set at a constant temperature, rang- two quick downward shakes were given (Patel, Stripp, &
ing from 50 to 85 C. After heating, samples were taken Fry, 1988).
out of the water bath and immediately transferred to an At regular time intervals, close-up photographs of the
ice bath to stop further denaturation. Under the conditions foam column were taken using the super macro modus of
applied, no gel was obtained. However, a turbid suspension a high-end digital camera (Canon Powershot Pro 1) from
of protein aggregates was observed. a distance of 5 cm. Arbitrary sections (25 mm2) of these
High pressure experiments were performed in a photographs were analyzed using imaging analysis software
laboratory scale multivessel high-pressure equipment (analySIS pro 3.1 and analySIS 5, Soft imaging System,
(HPIU-10.000, serial no. 95/1994, Resato, Roden, The Bensheim). The number and cross section area (mm2) of
Netherlands), consisting of six individual vessels (vol- the air bubbles surrounded by protein film were determined.
ume = 40 mL) surrounded by a thermostatic mantel, con- The foam-forming ability (FA) was defined as the foam
nected to a cryostat. The protein solutions were filled in volume (L) 5 min after the end of whipping. The liquid in
14 mL polypropylene tubes with flexible silicone stopper foam (LF) was expressed as the volume of liquid retained
(15 mm in diameter, Greiner Bio-One, Frickenhausen, in a fixed volume of foam, 5 min after whipping had
Germany), packed in double polyethylene plastic packs,
to avoid any direct contact with the pressure medium,
and vacuum sealed (down to 30 mbar, Multivac A300/16,
Wolferteschwenden, Germany). Next, the tubes were
enclosed in the pressure vessels already equilibrated at
the desired temperature. Introduction of the samples in
the pressure equipment took exactly 3 min, after which
pressure was built up slowly (100 MPa/min) to minimize
adiabatic heating. After attaining the desired pressure, all
individual vessels were isolated and the central circuit
was decompressed. The vessels were decompressed after
20 min of high pressure treatment. In these experiments,
a pressure range of 400–700 MPa and temperature range
of 10–60 C was used. Exactly 1 min after pressure release,
the samples were taken out of the pressure equipment and
cooled in ice water to stop any further heat-denaturation.
Under the conditions applied, no gel was obtained. How-
ever, a turbid suspension of protein aggregates was
observed.
Heated or pressurized solutions from different tubes
were mixed to render the necessary volume for determina-
tion of foaming properties. Foaming characteristics were
assessed after 24 h storage at 4 C. In this way, only the Fig. 1. Schematic presentation of the experimental set-up of the whipping
irreversible changes in egg white proteins were taken into device (dimensions in mm). Fifty milliliters of foaming solution (8% v/v)
account. was whipped in a glass beaker of 400 ml with a rotating anchor.
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1413

stopped. Drainage of liquid was recorded during 1 h. The 3. Results and discussion
foam stability (FS) was defined as the percentage of liquid
still present in the foam after 1 h compared to the situation 3.1. Foaming properties of untreated egg white solutions
at 5 min after whipping. Samples were determined in
triplicate. Egg white solutions (8% v/v) were whipped using the
experimental set-up as described above. Increasing both
2.4. Determination of physicochemical properties whipping time and rotation speed of the anchor resulted
in the formation of higher volume foams with lower density
The procedures for and the results of the determination (data not shown). Foam making conditions of 4 min at
of the physicochemical properties of treated egg white solu- 1300 rpm and ambient temperature were chosen.
tions were described in previous studies (Van der Plancken A significant difference (a = 0.05) was observed in foam
et al., 2004, 2006). Residual denaturation enthalpy was volume of foams prepared from untreated egg white solu-
determined using differential scanning calorimetry (DSC). tions at pH 7.6 and pH 8.8 (0.1837 ± 0.0033 and 0.2170 ±
Solubility and turbidity were determined spectrophotomet- 0.0054 L, respectively). Furthermore, it was observed that
rically. Samples (1:10 dilutions of the 10% v/v samples in foams prepared at the lower pH contained more liquid
distilled water) were centrifuged for 15 min at 19,900g and (0.103 ± 0.007 and 0.064 ± 0.006 L liquid/L foam at pH
4 C. Protein content of the supernatant was determined 7.6 and pH 8.8, respectively). Once formed, foams of
using Sigma Procedure No. TRPO-562 at 562 nm. Solubil- untreated egg white solutions were only marginally stable,
ity was expressed as the percentage of protein remaining in as the liquid retained in the foam drained as a function of
the supernatant as compared to the untreated sample. The standing time. From Fig. 2, it is clear that drainage occurs
transmittance of a 1/20 dilution of the (un)treated egg white in two stages. Initially, liquid flows freely as a result of
solution (10% v/v) in demineralized water was measured at gravitational force, however, flow is counteracted as func-
650 nm. A turbidity of 0% corresponds to a totally clear tion of the interface tension gradient. In a later stage, the
solution (transmittance in comparison to the blank is zero). slower liquid drainage is due to the collapse of foam bub-
(Aromatic) surface hydrophobicity (S0) was determined bles (German, O’Neill, & Kinsella, 1985). The stability of
using the fluorescent probe ANS (anilino-naphthalenesulf- the pH 8.8 foams was significantly lower than at pH 7.6,
onic acid) as described by Alizadeh-Pasdar and Li-Chan as at the former pH only 26.0 ± 0.8% of the liquid initially
(2000). The total, buried and exposed sulfhydryl (SH) con- present in the foam was retained after 1 h, while at pH 7.6,
tent was determined spectrophotometrically using Ellman’s this was 28.8 ± 1.0%. The foam volume could not be deter-
reagent (5 0 ,5-dithiobis (2-nitrobenzoic acid) or DTNB) mined as a function of time, as the foams of untreated sam-
(Kalab, 1970). The susceptibility to enzymatic hydrolysis ples collapsed internally. Consequently, foam stability had
(DH10), a measure for protein’s flexibility, was determined to be determined based on the amount of liquid retained in
using a pH-stat method (Adler-Nissen, 1986; Pedersen & the foam after a fixed delay time.
Eggum, 1981). Hammershoj, Prins, and Qvist (1999) studied the surface
and foaming properties of diluted (dried) egg white solu-
2.5. Data analysis tions (0.01% protein w/v) at different pH values. They used
two methods to prepare the foams: 25 s shaking and 60 s
The foaming properties of egg white solutions treated
under different conditions of temperature and pressure
were analyzed by Tukey’s Studentized Range Test using 50
the ANOVA procedure of the SAS Software (SAS Insti-
tute, 1999). Significant differences between means were
40
decided based on a significance level (a) of 0.05.
The data on the physicochemical and foaming proper-
ties of treated egg white solutions were subjected to multi-
Lin foam (ml)

30
ple regression analysis using least squares methodology
and the REG procedure of the SAS software (SAS Insti-
tute, 1999). Stepwise multiple regression analysis was per- 20
formed on the individual foaming properties as
dependent variables and the physicochemical properties 10
as independent variables. Logarithmic transformations of
the variables and interaction terms between two variables
were also taken into account. The best combination of 0
independent variables was found by eliminating variables 0 15 30 45 60
of no significance to simultaneously improve the standard t (min)
error of the estimate (S), the adjusted r2 ðr2adj Þ and the Fig. 2. Liquid drainage from foams of untreated egg white solutions at pH
sum of squares of the residuals. 7.6 () and pH 8.8 ().
1414 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

stirring with a fan-shaped whisk at 2500 rpm. The latter finally breaking of the interfacial film. Therefore, it was
resembled better the method used in our work. For that concluded that besides liquid drainage, due to gravitational
method, a somewhat lower overrun was observed at pH force, bubble coalescence was the main process leading to
7.0 compared to 9.2, while for the overrun of foams pre- foam instability.
pared using the shaking method, the opposite was
observed. For both methods, the foam stability (based on 3.2. Effect of heat-treatment on foaming properties of egg
the residual foam volume), was the lowest at pH 9.2. white proteins
Brittle foams were formed upon whipping of untreated
egg white solutions. After foam transfer and 5 min after Temperature- and pH-dependent changes could be
the end of whipping, bubble cross section area ranged from observed in the foaming properties of heat-treated egg
0.05 mm2 to 0.55 mm2 with an average of 0.077 ± white solutions, as shown in Fig. 3. Increasing temperature
0.010 mm2 at pH 7.6. At the higher pH, bubbles were resulted in a decrease of the volume of the foams formed,
somewhat larger, ranging from 0.011 to 0.70 mm2 with until a pH-dependent minimum was reached (Fig. 3A).
an average of 0.090 ± 0.009 mm2. Bubble-size distribution Further increasing the temperature, again improved the
measurements through a glass wall can be erroneous, as a foaming ability, at pH 8.8 even to a higher level than of
relatively higher number of larger bubbles is observed foams made out of untreated egg white solutions. At pH
(Bisperink, Ronteltap, & Prins, 1992). Furthermore, as 8.8, this decrease in FA coincided with an increase in tur-
the boundaries of these air bubbles were selected manually bidity of heat-treated egg white solutions and the first
(because of the low contrast between air bubble and sur- phase in protein solubility loss, shown in an earlier study
rounding medium, although this boundary was clearly dis- (Van der Plancken et al., 2006). With increasing treatment
tinguishable by eye), and only a small section of a single temperature, foams with lower foam volume were formed,
sample was examined, these values are approximative and until the temperature at which maximum turbidity occurs,
therefore merely informational. Keeping in mind the deter- was reached. From this temperature on, the FA of heat-
mination conditions used, the difference in average bubble treated egg white solutions increased again until the second
cross section area between the two pH values cannot be phase of protein solubility loss commenced.
regarded significant. At the lower pH, however, FA initially did not decrease
In course of time, egg white solution drained from the when the turbidity started to increase (above 55 C). Only
foam, resulting in drying-up of the foam and distortion when the temperature of the second phase in the turbidity
of the bubbles. While the bubbles were more or less spher- increase was reached (above 65 C), FA of heat-treated egg
ical at first, they became more polyhedrical after time. The white solutions decreased until a minimum was reached at
amount of bubbles decreased from 12.2 to 3 bubbles/mm2 75 C, the temperature above which no further changes in
at pH 7.6 and from 9.8 to 4 bubbles/mm2 at pH 8.8, within turbidity could be observed. Above this temperature,
45 min after whipping. A clear shift to larger bubbles and a heat-treated egg white solutions showed a further loss in
broad range of bubble cross section areas was observed as protein solubility and improved foam volume (Van der
after a delay of 45 min, the average bubbles cross section Plancken et al., 2006).
area was increased to 0.323 mm2, with a maximum of Thus, extensive protein solubility loss at pH 7.6 (maxi-
4.47 mm2 at pH 7.6 and 0.231 mm2, with a maximum of mal at 85 C) did not impair the ability of the egg white
3.13 mm2 at pH 8.8. It has to be stressed that at the glass proteins to form voluminous foam (83% of the untreated
interface, bubbles can become distorted and disproportion- foam volume). This would indicate that the loss of soluble
ation and coalescence rates may be affected. Therefore, protein able to unfold readily at the interface was counter-
these values are not an adequate representation of the over- balanced by the other structural changes induced by heat-
all changes in bubble size distribution due to aging of the treatment. At pH 8.8, changes in protein structure were not
foam. The foam column collapsed 45 min after whipping, accompanied by extensive protein insolubilization (Van der
leaving a dry, fragile matrix of egg white proteins. Plancken et al., 2006). This can explain the improved foam-
These changes in bubble cross section area and number ing ability of egg white proteins treated at temperatures
can be accounted for by two possible mechanisms. Firstly, above 70 C. The proteins can readily unfold at the inter-
coalescence of two bubbles due to the rupture of the inter- face by heat-induced increased flexibility (Van der Planc-
facial film results in larger, though less numerous bubbles. ken et al., 2004), without being hindered by large
Secondly, disproportionation due to internal gas pressure aggregate sizes. Hagolle, Relkin, Popineau, and Bertrand
differences between large and small bubbles, results in the (2000) also observed an increase in foaming ability when
shrinking and even disappearing of small bubbles in favour ovalbumin and lysozyme solutions (pH 7.0) were pre-
of the larger bubbles. Distinguishing between these two heated. For ovalbumin, the solution had to be heated
phenomena is often difficult (Bisperink et al., 1992). Under above 70 C to obtain improved foaming power. Lysozyme
the present conditions of bubble cross section area determi- showed no foaming ability, unless it was heated above
nation, no significant shrinking of bubbles, typical for bub- 80 C. The heat-induced conformational changes in this
ble disproportionation, could be observed. The drainage of rigid protein were indicated as the cause for this
liquid due to gravitational force leads to thinning and phenomenon.
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1415

175 200

150 175

150
Foaming ability (%)

Liquid in foam (%)

125
125
100
100
a a
75 b
75
a c a c a a a c
c a a b a
50 b d a d d a d d a
d b d 50 b b b b b
e f c c b b b
25 d d 25

0 0
25 50 55 60 65 70 75 80 85 25 50 55 60 65 70 75 80 85
(A) T (°C) (B) T (°C)

250

200
Foam stability (%)

150
a a
b
100 b
b a b b
c c c c
d
d d e d
50 e e

0
25 50 55 60 65 70 75 80 85
(C) T (°C)

Fig. 3. Effect of 20 min heat-treatment of egg white solutions (10% v/v) at pH 7.6 (grey bars) and pH 8.8 (white bars) on foaming ability (A), foam density
(B), and foam stability (C) as compared to the untreated egg white solution of each individual pH (%). Error bars represent the standard deviation of
triplicate measurement. For each individual pH, means with the same letter are not significantly different (a = 0.05).

At the neutral pH, maximum foam density was obtained approach of adjacent films. Heat-treatment at high temper-
in the same temperature range wherein a minimum in foam atures (above 55 C) seemed to enhance these protein–
volume occurred (Fig. 3B). Egg white solutions treated at protein interactions at the interface, possibly by interactions
pH 7.6 and 75 C, formed low volume, highly dense foams, between heat-exposed hydrophobic groups or sulfhydryl
that were however relatively stable (Fig. 3C). At pH 8.8, groups (Van der Plancken et al., 2005b, 2006).
however, the FD was only mildly affected by heat- The higher density of foams prepared from egg white
treatment. A heat-induced change in foam volume was thus solutions treated at temperatures above 65 C, resulted in
paralleled by a proportional change in liquid retained in a moist and creamy appearance, contrasting with the crispy
the foam. An increase in foam density was also observed and dry appearance of the foams prepared from untreated
for pre-heated ovalbumin solutions (pH 7) above 70 C. egg white solutions. These foams were sticky and resem-
The foams of heated lysozyme showed also increased initial bled clotted cream. No changes in color (white) of the
density when the temperature of the heat-treatment was foams could be observed with the naked eye.
increased (Hagolle et al., 2000). Foams prepared from heat-treated samples showed
For both pH values, foam stability was improved by a more and smaller bubbles compared to those from
20 min heat-treatment at temperatures above 55 C untreated egg white solutions. For example, egg white solu-
(Fig. 3C). Maximal FS was observed for treatments result- tions treated for 20 min at 75 C and pH 7.6 had an initial
ing in minimal FA. The FS is dependent on the ability of average bubble cross section area of 0.043 mm2. Foams
the proteins at the interface to form a cohesive network treated under the same conditions but at pH 8.8 showed
by both covalent and non-covalent interactions. In stable similar cross section areas. The higher foam density
foams, this film can resist physical perturbations and the observed after heat-treatment at pH 7.6, thus indicates that
1416 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

at the lower pH, the amount of interlamellar liquid was in earlier studies (Van der Plancken et al., 2004, 2005b,
higher, as no significant difference in bubble size was 2006) and the properties of the foams prepared from
observed for the different pH levels. This difference in foam heat-treated egg white solutions were examined. When cor-
structure compared to the untreated egg white solutions relating the foaming properties of heat-treated egg white
was also observed for ovalbumin and lysozyme foams solutions with the observed changes in their physicochem-
obtained by sparging of heat-treated protein solutions ical properties, it is important to keep in mind that during
(Hagolle et al., 2000). foam formation, the proteins undergo additional denatur-
The collapse of the foam column observed for foams ation at the interface (Clarkson, Cui, & Darton, 1999).
prepared from untreated egg white solutions was reduced For instance, it has been shown that conformational
or even prevented by heat-treatment. Foams prepared from changes in ovalbumin at the air–water interface lead to
egg white solutions treated at 50 or 55 C, showed some the formation of new intermolecular disulfide bonds due
collapse of the foam column, but for egg white solutions to oxidation of exposed sulfhydryl groups (Kitabatake &
treated at temperatures above 60 C this was not observed Doi, 1987). Heat-induced changes in the physicochemical
upon aging. For these treatments, the bubble size did properties of the egg white proteins can facilitate or coun-
increase as a function of time, but not as pronounced as teract these changes, and thus enhance or diminish their
for foams from untreated egg white solutions (Fig. 4). foaming properties. Nevertheless, the physicochemical
For instance, egg white heat-treated at 75 C for 20 min properties of the treated egg white solutions do not neces-
had an average cross section area of 0.090 mm2 at pH 7.6 sarily correspond to those of the proteins at the interface.
and 0.18 mm2 at pH 8.8 1 h after whipping. At pH 7.6, Pair wise correlation analysis of all foaming properties
the heat-induced protein–protein interactions seem to be of the heat-treated egg white solutions showed that, how-
involved in limiting the drainage and subsequently coales- ever significant, no strong linear correlation existed
cence of bubbles, resulting in a very stable structure, with between the individual properties, especially at pH 8.8
smaller bubbles. (Table 1). Pair wise correlation analysis between the indi-
The relationship between heat-induced changes in the vidual foaming properties on the one hand and the corre-
physicochemical properties of egg white solutions reported sponding physicochemical properties of these egg white

Fig. 4. Appearance of foams prepared from untreated egg white solutions (A) and egg white solutions heat-treated at 65 C (B) and 75 C (C), at pH 7.6
(1) or pH 8.8 (2), 60 min after whipping. The width of the photograph corresponds to 9 mm.
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1417

Table 1
Pearson correlation coefficients among foaming properties of egg white solutions (8% v/v) heat-treated at a concentration of 9.64 g protein/L, and their
physicochemical properties, presented in earlier studies (Van der Plancken et al., 2003, 2005b, 2006)
Property pH 7.6 pH 8.8
FA LF FS FA LF FS
a a
FA – 0.816 0.883 – 0.402 0.252
LF 0.816a – 0.901a 0.402 – 0.090
FS 0.883a 0.901a – 0.252 0.090 –
Enthalpy 0.742a 0.955a 0.943a 0.641 0.593 0.526
Solubility 0.747a 0.956a 0.938a 0.464 0.567 0.668
Turbidity 0.775a 0.898a 0.970a 0.008 0.171 0.898a
A650 0.717a 0.942a 0.925a 0.011 0.178 0.888a
S0 0.716a 0.917a 0.920a 0.449 0.525 0.663
Total SH 0.790a 0.953a 0.932a 0.647a 0.616 0.497
Exposed SH 0.713a 0.961a 0.895a 0.719a 0.503 0.469
Buried SH 0.733a 0.963a 0.907a 0.694 0.557 0.485
DH10 0.778a 0.945a 0.964a 0.620 0.537 0.570
Correlation coefficients higher than 0.95 are highlighted.
FA foaming ability (L) of 8% v/v egg white solution.
LF liquid in foam (L/L) of 8% v/v egg white solution.
FS foam stability (%) of 8% v/v egg white solution.
A650 absorbance of 1/20 dilution of the treated egg white solution (10% v/v).
S0 surface hydrophobicity (a.u.).
SH sulfhydryl content (% of total SH content of untreated egg white solution).
DH10 degree of hydrolysis after 10 min of enzymatic hydrolysis with trypsin and a-chymotrypsin.
a
Significant at P < 0.0001.

solutions on the other hand revealed that although some different mechanisms, depending on the pH. Therefore, it is
significant correlation existed (and this primarily at pH not surprising that at the different pH values, different
7.6), the heat-induced changes in none of the foaming physicochemical properties are determinant for the foam-
properties could be accounted for by the heat-induced ing ability of heat-treated egg white solutions.
changes in one single physicochemical property measured At both pH values, the sulfhydryl groups seemed to be
(Table 1). This could be expected as both protein-stabilized involved in the mechanism of foam formation. This was
foam formation and heat-induced protein denaturation are also observed for foams prepared from ovalbumin. During
complex phenomena involving several different types of foam formation, new intermolecular disulfide bonds are
protein interactions. formed due to oxidation of exposed sulfhydryl groups
Therefore, stepwise multiple regression analysis was per- (Kitabatake & Doi, 1987). These covalent bonds contribute
formed on the individual foaming properties as dependent to the formation of a protein film around the bubble. Heat-
variables and the physicochemical properties (transformed induced changes in these groups thus affect the foaming
if necessary) as independent variables. The optimal combi- ability of the egg white proteins.
nation of independent variables was selected based on the The foaming ability of a protein strongly depends on its
lowest standard error of the estimate (S), the highest flexibility at the interface. Susceptibility to enzymatic
adjusted r2 ðr2adj Þ and the lowest sum of squares of the resid- hydrolysis was demonstrated to be a measure for protein
uals. Only models consisting of maximum three physico- flexibility (Kato, Komatsu, Fujimoto, & Kobayashi,
chemical properties and maximum four parameters 1985). At pH 7.6, an increase in DH10 had a positive effect
(including intercept) were considered. The denaturation on the foaming ability of the heat-treated egg white pro-
enthalpy was not considered as the determination of this teins (Table 2). This was also observed for ovalbumin
property was performed on egg white solutions treated at and lysozyme solutions by Kato et al. (1986). Although
a much higher concentration (75% v/v) than the other the exposure of hydrophobic groups at the interface
properties. reduces the surface tension, no positive effect of S0 was
For each individual pH, a specific set of physicochemical observed. At pH 8.8, even a negative contribution of the
properties could be found to predict accurately the foaming heat-induced exposure of hydrophobic groups to the foam-
ability of heat-treated egg white solutions (more than 95% ing ability was observed, when the other physicochemical
of the variation in the property was explained by the pre- properties were also taken into account (Table 5).
diction equation), as shown in Table 2. Several factors con- No significant relationship could be found between
tribute to the formation of protein-stabilized foam: among foaming ability and surface hydrophobicity of different
others, protein flexibility, hydrophobic–hydrophilic ratio untreated proteins (Townsend & Nakai, 1983). This obser-
and electrostatic forces. Furthermore, heat-induced egg vation was ascribed to the extensive exposure of hydropho-
white protein denaturation was shown to proceed through bic groups upon unfolding a protein at the interface. As a
1418 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

Table 2
Multiple regression models for prediction of the foaming properties of egg white solutions (8% v/v) heat-treated at a concentration of 9.64 g protein/L,
based on their physicochemical properties, presented in earlier studies (Van der Plancken et al., 2003, 2005b, 2006)
Property pH Predictor variable Parameter estimate t-Value P>t S r2adj
a b
FA (mL) 7.6 Intercept 3263.6 ± 201.4 16.2 <0.001 8.71 0.962
Exposed SH 17.52 ± 1.15b 15.3 <0.001
ln(buried SH) 665.1 ± 43.6b 15.3 <0.001
ln(DH10) 28.54 ± 9.19b 3.1 0.0052

8.8a Intercept 7402.7 ± 638.5b 11.6 <0.001 6.30 0.968

Total SH 34.75 ± 2.49b 14.0 <0.001
ln(total SH) 2450.3 ± 193.8b 12.6 <0.001
ln(S0) 34.58 ± 1.67b 20.7 <0.001

LF (L/L) 7.6a Intercept 1.359 ± 0.163b 8.35 <0.001 0.0062 0.984

Exposed SH 0.0331 ± 0.0039b 8.33 <0.001
Total SH · exposed SH 280.6 106 ± 40.0 106b 7.01 <0.001
ln(buried SH) 0.3305 ± 0.0353b 9.36 <0.001

8.8 No model established

FS (%) 7.6a Intercept 169.88 ± 16.99b 10.00 <0.001 1.916 0.983

ln(solubility) 43.40 ± 3.64b 11.92 <0.001
Turbidity · DH10 0.3145 ± 0.0184b 17.10 <0.001

8.8a Intercept 60.22 ± 1.09b 55.38 <0.001 1.293 0.977

Exposed SH 0.121 ± 0.018b 6.84 <0.001
ln(A650 nm) 8.077 ± 0.299b 26.95 <0.001
FA foaming ability (L) of 8% v/v egg white solution.
LF liquid in foam (L/L) of 8% v/v egg white solution.
FS foam stability (%) of 8% v/v egg white solution.
A650 absorbance of 1/20 dilution of the treated egg white solution (10% v/v).
S0 surface hydrophobicity (a.u.).
SH sulfhydryl content (% of total SH content of untreated egg white solution).
DH10 degree of hydrolysis after 10 min of enzymatic hydrolysis with trypsin and a-chymotrypsin.
a
Significant at P < 0.001.
b
Standard error.

result, the S0 measured in solution may be quite different to the heat-induced changes in the physicochemical proper-
the actual exposed hydrophobicity of a protein unfolded at ties. The foam density of egg white solutions heat-treated
the interface. In that study, a strong contribution of total at pH 7.6 could be adequately predicted based on their sulf-
hydrophobicity, viscosity and dispersibility to the foaming hydryl content (Table 2).
capacity of untreated proteins was demonstrated. Surface Although proteins need to be sufficiently flexible to
hydrophobicity of heat-treated ovalbumin or lysozyme unfold at the interface, they also need to retain a consider-
was not related to their foaming ability. However, a strong able secondary and even tertiary structure at the interface
correlation with emulsifying activity, another surface activ- in order to form a stable film around the foam air bubbles
ity, was demonstrated (Kato et al., 1986). (German et al., 1985). Therefore, it can be assumed that
Significant correlations existed between the physico- physicochemical properties facilitating the formation of a
chemical properties, especially at pH 7.6 (data not shown). protein–protein network will contribute to the foam stabil-
Therefore, the contribution of the other structural proper- ity. In Table 2, the best prediction equations for the foam
ties to foaming properties of heat-treated egg white solu- stability of heat-treated egg white proteins based on their
tions cannot be excluded based on the models obtained. physicochemical properties, are listed.
Furthermore, other physicochemical properties that were Although a relatively strong correlation was observed
not considered in this study (e.g. charge density and aggre- between the foam stability and the sulfhydryl content of
gate size) can also contribute to the foaming ability of heat- egg white solutions heat-treated at pH 7.6, the former did
treated egg white proteins. not form part of the optimal model predicting the foam
No model meeting the prerequisites could be found to stability at this pH. At pH 8.8, the amount of exposed
predict the foam density of egg white solutions treated at SH groups contributed negatively to the foam stability
pH 8.8. This could be expected as little significant change (Table 2). The heat-induced exposure of the buried SH
(a = 0.05) in this property occurred at the alkaline pH. groups of ovalbumin can facilitate the formation of disul-
Probably some other physicochemical properties are fide bonds observed at the interface. However, the forma-
responsible for the constant density by counterbalancing tion of these disulfide bonds proved not to be essential
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1419

for a stable foam, as the addition of DTNB to ovalbumin 175

prior to foam formation does not impair the formation of a
stable foam (Kitabatake & Doi, 1987). 150
For each individual pH, the optimal model predicting a a a a a
a a a
the foam stability consisted of a term corresponding to 125 b b

Foaming ability (%)

a a a
the degree of aggregation of the egg white proteins (either c b b
turbidity or solubility). Strong protein–protein interactions 100 c
d c
are necessary to form a stable foam. Increased aggregation c c c
had a positive influence on the foam stability. 75 d

3.3. Effect of pressure treatment on foaming properties of egg 50

white proteins
25
Similar to heat-treatment, pressurization of egg white
solutions resulted in irreversible changes in the foaming 0
10 25 40 60
properties of the treated solutions. These changes were (A) T (°C)
either beneficial or detrimental, depending on the pH and
the temperature–pressure combination applied (Figs. 5–7). 175
The effect of pressure on the foaming ability of pressure- a
treated egg white solutions was strongly dependent on the a a
150 a
pH of the egg white solutions. At pH 8.8, a beneficial effect a ab a a b b a
b
could be observed at all levels of temperature studied 125 b b c
(Fig. 5B). With increasing pressure, foams of higher vol- Foaming ability (%) c c
dc d
ume could be obtained. At the highest pressures, the 100 b
b
improvement of the FA was even more pronounced than
obtainable by heat-treatment at this pH. The effect of tem- 75
perature was not clear below 60 C at any pressure. For the
pressure levels studied, no significant lower pressure was 50
needed to change the FA at lower pressure, as would be
expected based on the antagonistic effect between pressure 25
and temperature observed for the physicochemical proper-
ties of pressure-treated egg white proteins. At 60 C how- 0
ever, pressure had a significantly stronger, enhancing 10 25 40 60
(B) T (°C)
effect on the FA of egg white solutions.
At pH 7.6, the effect of pressure on the foaming ability Fig. 5. Effect of 20 min pressure treatment of egg white solutions (10%
was less straightforward than at the higher pH (Fig. 5A). v/v) at pH 7.6 (A) or pH 8.8 (B) and at 0.1 ( ), 400 ( ), 500 ( ), 600 ( ), and
Below 60 C, an initial increase in FA due to pressure treat- 700 MPa ( ) on foaming ability as compared to the untreated egg white
ment at pressures below 500 MPa, was followed by a solution of each individual pH (%). Error bars represent the standard
deviation of triplicate measurement. For each individual pH and
decrease, resulting in egg white solutions with lower foam- temperature, means with the same letter are not significantly different
ing capacity compared to the untreated proteins. This det- (a = 0.05).
rimental effect at elevated pressure was stronger at lower
temperature. At 60 C, no significant effect of pressure on
the FA of the treated egg white solutions could be observed In the study by Strohalm et al. (2000), however, no signif-
(a = 0.05). icant change in foam volume could be observed for egg
Except for 60 C, pressure treatment resulting in exces- white pressurized at 25 C and pressures up to 400 MPa.
sive loss of protein solubility (600–700 MPa at pH 7.6) This difference can be due to the fact that our study was
reduced the foaming ability of the treated egg white solu- conducted on diluted egg white solutions, while in the other
tions. At pH 8.8 little insolubilization of protein occurred studies both pressure treatment and foam formation where
due to pressure treatment. The combination of high resid- performed with pure egg white. In the first study, foams
ual protein solubility and pressure-induced changes in the were prepared by stirring, while in the second the pressur-
protein structure can explain the improved foaming ability ized egg white was whipped.
of egg white proteins pressure-treated at this pH. As was observed for heat-treatment, high pressure treat-
Richwin et al. (1992) observed that pressure treatment ment exerted little effect on the density of the foams pre-
up to 400 MPa results in decreasing foam expansion of pared from egg white solutions pressure-treated at pH 8.8
the pressure-treated egg white, almost independent of tem- (Fig. 6B). Little or no effect of temperature during pressure
perature (20–50 C), while in our study the FA was treatment was observed either. Since the volume of the
improved by pressure treatment at 400 MPa below 60 C. foams formed was indeed affected by pressure treatment
1420 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

200 250

175
a
150 a a 200 a
Liquid in foam(%)

b a
a a

Foam stability (%)

b b b b a a
125 c b b
c c c c d c c b a
150 b a b
d c c c c b
100 c c c c c
bdb
b d d d
d c
75
100
50

25 50
0
10 25 40 60
(A) T (°C) 0
10 25 40 60
(A) T (°C)
200

175 250

150
Liquid in foam (%)

200
125 a a a
a a a a a a Foam stability (%)
100 a a a a a a b b b b b
a a a a
150
a
75 a b b
a a a
b b b a a a a a a c
50 100
a c c

25

0 50
10 25 40 60
(B) T (°C)

Fig. 6. Effect of 20 min pressure treatment of egg white solutions (10% 0

10 25 40 60
v/v) at pH 7.6 (A) or pH 8.8 (B) and at 0.1 ( ), 400 ( ), 500 ( ), 600 ( ), and (B) T (°C)
700 MPa ( ) on foam density as compared to the untreated egg white
solution of each individual pH (%). Error bars represent the standard Fig. 7. Effect of 20 min pressure treatment of egg white solutions (10%
deviation of triplicate measurement. For each individual pH and v/v) at pH 7.6 (A) or pH 8.8 (B) and at 0.1 ( ), 400 ( ), 500 ( ), 600 ( ), and
temperature, means with the same letter are not significantly different 700 MPa ( ) on foam stability as compared to the untreated egg white
(a = 0.05). solution of each individual pH (%). Error bars represent the standard
deviation of triplicate measurement. For each individual pH and
temperature, means with the same letter are not significantly different
(a = 0.05).
(Fig. 5B), at pH 8.8 a pressure-induced change in foam vol-
ume was paralleled by a proportional change in liquid
retained in the foam. (Fig. 7), as was the case for heat-treated egg white solutions
At the lower pH, pressure-treated egg white solutions (Fig. 3C). The positive effect on this foaming property of
(above 500 MPa) gave rise to denser foams (Fig. 6A). As the non-thermal treatment was however less pronounced.
for heat-treated egg white solutions, a higher density coin- The low volume, dense foams produced from egg white
cided with a lower foam volume (Fig. 5A). Under the con- solutions pressure-treated above 500 MPa (10–40 C) were
ditions applied, the foams prepared from pressure-treated significantly more stable than those prepared from egg
egg white solutions were less dense compared to those from white solutions with pressure-induced improved foaming
heat-treated solutions (Fig. 3B). Iametti et al. (1999) ability. Probably the formation of aggregates in the former
showed that pressure treatment (5 min at 600 MPa and solutions did not allow the incorporation of large volumes
25 C) enhanced the foam density of the treated egg white; of air, but the strong intermolecular interactions stabilized
the addition of 10% sucrose prior to pressure treatment the small bubbles formed to a higher extent.
enhanced this effect, while the presence of 10% NaCl Literature data present contradictory observations
resulted in less dense foams. regarding the stability of foams prepared from egg white
At both pH values, it could be observed that pressure pressurized up to 400 MPa. Richwin et al. (1992) observed
treatment had no detrimental effect on the stability of the that increasing pressure results in decreasing foam volume
foams prepared from the resulting egg white solutions of the pressure-treated egg white, almost independent of
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1421

temperature (30–50 C). In the higher temperature regions Some significant correlation existed between the individ-
(40–50 C), foam stability decreased most likely due to ual foaming properties on the one hand and the corre-
lower soluble protein content. When the foam stability sponding physicochemical properties of these egg white
was defined based on the amount of liquid drained from solutions on the other hand in the lower temperature range
the foam (like in our study), it was observed that increasing (and this primarily at pH 7.6) (Table 3). At 60 C however,
pressure up to 400 MPa resulted in more stable foams little linear correlation was observed between foaming
(Strohalm et al., 2000). properties and physicochemical properties (Table 4). Pres-
Foams prepared from pressure-treated egg white solu- sure treatment at 60 C indeed had a different effect on
tions had a moist and creamy appearance, similar to foams foaming properties than pressure treatment in the lower
from egg white solutions heat-treated above 65 C. Severe temperature range. For instance, at 60 C, pressure treat-
pressure treatments (700 MPa) led to dense foams that ment resulting in low protein solubility did not impair
were sticky and resembling clotted cream. During the whip- the foaming ability of the treated egg white proteins, while
ping of egg white solutions processed under these condi- in the lower temperature range a low residual protein solu-
tions, splashing of the solution occurred and the resulting bility of the pressurized egg white solutions was accompa-
foam adhered to the glass wall and the rotating anchor. nied by poor foaming ability. These opposite relationships
The collapse of the foam column observed for foams for different temperature ranges were also observed
prepared from untreated egg white solutions was prevented between the other physicochemical properties and foaming
by pressure treatment (400–700 MPa). No coarsening was properties. This clearly indicates that pressure-induced
observed upon aging of foams from pressure-treated egg denaturation at elevated temperatures occurs through a
white solutions. Foams prepared from pressure-treated different mechanism compared to moderate temperatures.
egg white solutions with low foaming ability (600– None of the foaming property changes were accounted
700 MPa at 10–40 C and pH 7.6) showed small bubble for by pressure-induced changes in a single physicochemi-
size. For instance, foams of egg white solutions pressurized cal property. Therefore, stepwise multiple regression anal-
for 20 min at 700 MPa and 10 C had an average bubble ysis was performed on the individual foaming properties
cross section area of 0.035 mm2. This small bubble size as dependent variables and the physicochemical properties
was more or less maintained during standing of the foam (transformed if necessary) as independent variables. Again,
(Fig. 8). This is not surprising as these foams showed high this analysis was performed separately for the two temper-
stability. Under conditions of improved foaming ability, a ature ranges.
bubble size comparable to that of untreated egg white An acceptable model (80% of the variation in the prop-
foams was observed at pH 7.6, while at pH 8.8 smaller bub- erty was explained by the prediction equation) was found
ble sizes were observed compared to untreated foams, both to describe the increase in foaming ability at moderate
at low and high temperature. For instance after 20 min pressure (400 and 500 MPa) and the decrease in this prop-
treatment at 700 MPa and 10 C, the average initial bubble erty at high pressure (600 and 700 MPa) at pH 7.6, based
cross section area was 0.064 mm2. on the susceptibility to enzymatic hydrolysis, surface
At pH 8.8, the bubble size after 60 min standing was hydrophobicity and buried SH content of the pressure-
similar to those of foams of heat-treated egg white solu- treated egg white solutions (Table 5). At pH 8.8, the
tions, both for egg white solutions pressurized at low tem- changes in this property could be described based on the
peratures and at 60 C. Foams prepared from egg white egg white protein’s susceptibility to enzymatic hydrolysis
solutions pressurized at 60 C at the lower pH showed bub- and surface hydrophobicity alone (Table 5). The lower
ble sizes after 60 min of standing comparable to those of r2adj in comparison to the models describing this property
untreated egg white solutions (Fig. 8). for heat-treated egg white solutions can be due to the lim-
The relationship between the pressure-induced changes ited amount of data points or to the contribution of
in the physicochemical properties of egg white solutions unmeasured physicochemical properties to the FA of pres-
reported in earlier studies (Van der Plancken et al., 2004, sure-treated egg white solutions. Contrary to what was
2005a, in press) and the properties of the foams prepared observed for foams from heat-treated egg white solutions,
from these pressure-treated egg white solutions were exam- surface hydrophobicity had a positive effect on the FA of
ined. It was observed that the pressure-induced changes at egg white proteins pressure-treated in the temperature
60 C in both physicochemical properties and foaming range of 10–40 C.
properties strongly deviated from those of egg white solu- No model meeting the prerequisites could be found to
tions pressurized at 10–40 C (Van der Plancken et al., in describe the changes observed in foaming ability of egg white
press). Therefore, pair wise correlation analysis of all foam- solutions pressure-treated at 60 C based on their physico-
ing properties of the pressure-treated egg white solutions chemical properties. Under the restrictions applied, no
was performed for the two temperature domains sepa- higher r2adj than 0.460 could be obtained. This can be due
rately. No strong, significant linear correlation existed to the insignificant changes (a = 0.05) in this property. Fur-
between the individual properties for neither of the two thermore, other properties than the ones measured can be
temperature domains, as was also observed for heat-treated responsible for the foaming activity at this temperature. As
egg white solutions (Tables 3 and 4). shown in Table 5, the foaming ability of proteins pressurized
1422 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

Fig. 8. Appearance of foams prepared from egg white solutions pressure-treated at 10 C and 500 MPa (A) or 700 MPa (B) and at 60 C and 500 MPa (C)
or 700 MPa (D), at pH 7.6 (1) or pH 8.8 (2), 60 min after whipping. The width of the photograph corresponds to 9 mm.

at 60 C and pH 8.8 could be accurately described based on sible to construct a predictive model for this property based
their solubility and surface hydrophobicity. on the physicochemical properties measured (Table 5).
Similar to heat-treated egg white solutions, the insignif- Probably other physicochemical properties are responsi-
icant changes in the foam density at pH 8.8, made it impos- ble for this invariable density at all temperature levels.
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1423

Table 3
Pearson correlation coefficients among foaming properties of egg white solutions (8% v/v) pressure-treated at 10–40 C at a concentration of
9.64 g protein/L, and their physicochemical properties, presented in earlier studies (Van der Plancken et al., 2004, 2005a, in press)
Property pH 7.6 pH 8.8
FA FD FS FA FD FS
a a a
FA – 0.832 0.742 – 0.585 0.032
FD 0.832a – 0.876a 0.585a – 0.603a
FS 0.742a 0.876a – 0.032 0.603a –
Enthalpy 0.689a 0.868a 0.906a 0.860a 0.518b 0.045
Solubility 0.752a 0.911a 0.936a 0.423 0.510 0.648
Turbidity 0.564a 0.637a 0.773a 0.681a 0.296 0.029
A at 650 nm 0.593a 0.642a 0.785a 0.680a 0.298 0.028
S0 0.674a 0.856a 0.904a 0.915a 0.557b 0.041
Total SH 0.577a 0.815a 0.868a 0.837a 0.519b 0.039
Exposed SH 0.708a 0.848a 0.805a 0.756a 0.289 0.288
Buried SH 0.726a 0.928a 0.926a 0.829a 0.438 0.097
DH10 0.771a 0.920a 0.925a 0.837a 0.515b 0.109
FA foaming ability (L) of 8% v/v egg white solution.
LF liquid in foam (L/L) of 8% v/v egg white solution.
FS foam stability (%) of 8% v/v egg white solution.
A650 absorbance of 1/20 dilution of the treated egg white solution (10% v/v).
S0 surface hydrophobicity (a.u.).
SH sulfhydryl content (% of total SH content of untreated egg white solution).
DH10 degree of hydrolysis after 10 min of enzymatic hydrolysis with trypsin and a-chymotrypsin.
a
Significant at P < 0.0001.
b
Significant at P < 0.001.

Table 4
Pearson correlation coefficients among foaming properties of egg white solutions (8% v/v) pressure-treated at 60 C at a concentration of 9.64 g protein/L,
and their physicochemical properties, presented in earlier studies (Van der Plancken et al., 2004, 2005a, in press)
Property pH 7.6 pH 8.8
FA FD FS FA FD FS
FA – 0.581 0.405 – 0.046 0.847
FD 0.581 – 0.669 0.046 – 0.337
FS 0.405 0.669b – 0.847a 0.337 –
Enthalpy 0.778 0.828b 0.407 0.796b 0.205 0.757
Solubility 0.686 0.746 0.328 0.423 0.715 0.043
Turbidity 0.756 0.651 0.266 0.526 0.811b 0.067
A at 650 nm 0.800b 0.762 0.415 0.523 0.812b 0.065
S0 0.796b 0.754 0.388 0.773b 0.129 0.699
Total SH 0.628 0.709 0.290 0.506 0.033 0.349
Exposed SH 0.706 0.660 0.445 0.636 0.158 0.546
Buried SH 0.710 0.584 0.218 0.522 0.047 0.371
DH10 0.805b 0.842b 0.440 0.745 0.165 0.787
FA foaming ability (L) of 8% v/v egg white solution.
LF liquid in foam (L/L) of 8% v/v egg white solution.
FS foam stability (%) of 8% v/v egg white solution.
A650 absorbance of 1/20 dilution of the treated egg white solution (10% v/v).
S0 surface hydrophobicity (a.u.).
SH sulfhydryl content (% of total SH content of untreated egg white solution).
DH10 degree of hydrolysis after 10 min of enzymatic hydrolysis with trypsin and a-chymotrypsin.
a
Significant at P < 0.0001.
b
Significant at P < 0.001.

The surface hydrophobicity seemed to contribute to the ture range. With increasing degree of protein precipitation,
density of foams of heat-treated egg white solutions. the foam stability improved. However, other physicochem-
Only for egg white solutions pressurized at pH 7.6 in the ical properties can attribute to this property as less than
temperature range of 10–40 C, a model could be found to 90% of the variation in the FS was explained by the predic-
predict the foam stability (Table 5). This is not surprising, tion equation. This again shows that protein–protein inter-
as pressure treatment did not affect significantly the foam actions are important to form stable foams. The current
stability at the other level of pH, or in the higher tempera- data are however insufficient in number to fully grasp the
1424 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

Table 5
Multiple regression models for prediction of the foaming properties of egg white solutions (8% v/v) pressure-treated at a concentration of 9.64 g protein/L,
based on their physicochemical properties, presented in earlier studies (Van der Plancken et al., 2004, 2005a, in press)
Property T-Range pH Predictor variable Parameter estimate t-Value P>t S r2adj
a b
FA (%) 10–40 C 7.6 Intercept 71.1 ± 16.71 4.25 0.0001 9.95 0.802
DH10 15.9 ± 1.7b 9.25 <0.0001
ln(S0) 15.7 ± 2.4b 6.65 0.0023
ln(buried SH) 8.5 ± 2.6b 3.27 <0.0001

8.8a Intercept 61.8 ± 4.7b 13.11 <0.0001 3.63 0.905

DH10 1.139 ± 0.296b 3.85 0.0004
ln(S0) 6.908 ± 0.720b 9.60 <0.0001
60 C 7.6 No model established

8.8a Intercept 1479 ± 161b 9.18 <0.0001 5.81 0.972

S0 0.0465 ± 0.00465b 10.00 <0.0001
ln(S0) 176.5 ± 14.5b 12.21 <0.0001
ln(solubility) 580.2 ± 52.7b 11.01 <0.0001

LF (%) 10–40 C 7.6a Intercept 153.5 ± 9.3b 16.45 <0.0001 5.55 0.934
DH10 7.57 ± 0.96b 7.91 <0.0001
ln(S0) 6.80 ± 1.3b 5.17 <0.0001
ln(buried SH) 5.32 ± 1.45b 3.67 0.0007

8.8a No model established

60 C 7.6a Intercept 260.5 ± 25.5 10.22 <0.0001 3.75 0.923

S0 0.0095 ± 0.0010 9.47 <0.0001
ln(S0) 25.0 ± 3.7 6.76 <0.0001
8.8 No model established

FS (%) 10–40 C 7.6a Intercept 371.3 ± 13.5b 27.41 <0.0001 9.24 0.888
ln(solubility) 57.6 ± 3.2b 18.02 <0.0001
8.8 No model established

60 C 7.6 No model established

8.8 No model established
FA foaming ability (L) of 8% v/v egg white solution.
LF liquid in foam (L/L) of 8% v/v egg white solution.
FS foam stability (%) of 8% v/v egg white solution.
A650 absorbance of 1/20 dilution of the treated egg white solution (10% v/v).
S0 surface hydrophobicity (a.u.).
SH sulfhydryl content (% of total SH content of untreated egg white solution).
DH10 degree of hydrolysis after 10 min of enzymatic hydrolysis with trypsin and a-chymotrypsin.
a
Significant at P < 0.001.
b
Standard error.

effect of pressure-induced denaturation on the foaming foams were obtained at pH 7.6, for both treatments. Treat-
properties of egg white solutions. ments resulting in a high level of protein unfolding however
combined with a certain degree of residual protein solubil-
4. Conclusions ity (primarily at pH 8.8) resulted in egg white solutions
with improved foaming ability. A high level of unfolding
Both heat and pressure treatment affected the foaming combined with extensive protein solubility loss (primarily
properties of egg white proteins. While foams from at pH 7.6) was associated with increased foam stability
untreated egg white solutions were crispy and subject to and density.
collapse of the foam column after a long standing period, A pH-dependent minimum could be observed in the
foams from heat and pressure-treated egg white solutions foaming ability of heat-treated egg white proteins coincid-
were moist and creamy, showing smaller bubble size and ing with maximal foam stability. The heat-induced changes
lower sensitivity to bubble coalescence as evidenced by in the foaming properties could not be attributed to the
the reduced tendency of collapse of these foams. changes in one single physicochemical property. The foam-
The effect of physical treatments was strongly dependent ing ability of heat-treated egg white solutions was in part
on the pH during treatment. The most voluminous foams determined by the sulfhydryl content of ovalbumin, the
were obtained at pH 8.8, while the most stable, dense major egg white protein. Increased protein flexibility
 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426 1425

(measured as the susceptibility to enzymatic hydrolysis) Balny, C., & Masson, P. (1993). Effects of high pressure on proteins. Food
due to heating also contributed to the volume of foams pre- Reviews International, 9, 611–628.
Barbarosa-Cánovas, G. V., Pathakamury, U. R., Palou, E., & Swanson,
pared from heat-treated egg white solutions. Surface B. G. (1997). High hydrostatic pressure food processing. In G. V.
hydrophobicity showed to have no or a negative effect on Barbarosa-Cánovas, U. R. Pathakamury, E. Palou, & B. G. Swanson
the foam volume. This was also observed in literature (Eds.), Non thermal preservation of foods (pp. 9–52). New York, USA:
(Kato et al., 1986; Townsend & Nakai, 1983). The foaming Marcel Dekker Inc.
ability of pressure-treated egg white solutions was Bisperink, C. G. J., Ronteltap, A. D., & Prins, A. (1992). Bubble-size
distribution in foams. Advances in Colloid and Interface Science, 38,
improved at all levels of temperature and pressure at pH 13–32.
8.8, while at pH 7.6 in the lower temperature range a neg- Bridgman, P. W. (1914). The coagulation of albumen by pressure.
ative effect of pressure was observed for pressure treat- Proceedings of the American Academy of Arts and Sciences, 19,
ments resulting in pronounced protein denaturation. This 511–512.
difference in foaming ability of egg white solutions pres- Clarkson, J. R., Cui, Z. F., & Darton, R. C. (1999). Protein denaturation
in foam. 1. Mechanism study. Journal of Colloid and Interface Science,
sure-treated at different pH values can be attributed to 215, 323–332.
the marked difference in aggregation behaviour under pres- Cunningham, F. E. (1986). Egg-product pasteurization. In W. J. Stadel-
sure. Furthermore, an effect of pH during foam formation man & O. J. Cotterill (Eds.), Egg science and egg technology
(for instance electrostatic interactions) cannot be excluded. (pp. 243–272). Connecticut: AVI Publishing Wesport.
This observation does not necessarily imply that it is advi- Doi, E., Shimizu, A., Oe, H., & Kitabatake, N. (1991). Melting of heat-
induced ovalbumin gel by pressure. Food Hydrocolloids, 5, 409–425.
sory to use older eggs before physical treatment. Firstly, Donovan, J. W., & Mapes, C. J. (1976). A differential scanning
aging of eggs can reduce the microbial quality of eggs calorimetric study of conversion of ovalbumin to S-ovalbumin in
before pasteurization, compelling more severe treatments. eggs. Journal of the Science of Food and Agriculture, 27, 197–204.
Secondly, this study was performed on egg white in a Donovan, J. W., Mapes, C. J., Davis, J. G., & Garibaldi, J. A. (1975). A
Tris–HCl buffer system (pKa = 8.1). A contribution of the differential scanning calorimetric study of the stability of egg white to
heat denaturation. Journal of the Science of Food and Agriculture, 26,
buffer to the different aggregation behaviour at different 73–83.
levels of pH, through differences in shielding of negatively German, J. B., O’Neill, T. E., & Kinsella, J. E. (1985). Film forming and
charged egg white proteins, cannot be excluded. foaming behavior of food proteins. Journal of the American Oil
While little or no effect of physical treatment was observed Chemists Society, 62, 1358–1366.
on the foam density at pH 8.8, a clear increase in density of Gu, J., Matsuda, T., & Nakamura, R. (1986). Antigenicity of ovomucoid
remaining in boiled shell eggs. Journal of Food Science, 51, 1448–1450.
foams prepared from egg white solutions heat- or pressure- Hagolle, N., Relkin, P., Popineau, Y., & Bertrand, D. (2000). Study of the
treated at pH 7.6 under conditions of pronounced protein stability of egg white protein-based foams: Effect of heating protein
aggregation (above 65 C at ambient pressure and above solution. Journal of the Science of Food and Agriculture, 80, 1245–1252.
500 MPa at 10–40 C, respectively) was observed. The high Hammershoj, M., Prins, A., & Qvist, K. B. (1999). Influence of pH on
density of foams prepared from egg white solutions treated surface properties of aqueous egg albumen solutions in relations to
foaming behaviour. Journal of the Science of Food and Agriculture, 79,
under these conditions was also shown by a small bubble 859–868.
size. The stability of dense foams was in general high. Pro- Hatta, H., Hagi, T., & Hirano, K. (1997). Chemical and physicochemical
tein–protein interactions (solubility, SH groups) showed to properties of hen eggs and their application in foods. In T. Yamamoto,
be determinant for this property. L. R. Juneja, H. Hatta, & M. Kim (Eds.), Hen eggs. Their basic and
The best foams (high volume and average density) were applied science (pp. 13–24). Boca Raton: CRC Press.
Hayakawa, I., Linko, Y. Y., & Linko, P. (1996). Mechanism of high
obtained by pressure treatment at pH 8.8. For practical pressure denaturation of proteins. Lebensmittel-Wissenschaft und
applications in meringues and angel cakes, the heat-stabil- Technologie, 29, 756–762.
ity of the foams prepared from treated egg white should be Hayashi, R., Kawamura, Y., Nakasa, T., & Okinaka, O. (1989).
investigated. Also, microbiological studies are required to Application of high pressure to food processing: Pressurization of
explore the effect of both heat and high pressure treatment egg white and yolk, and properties of gels formed. Agricultural and
Biological Chemistry, 53, 2935–2939.
on Salmonella, so that processes can be designed that are Iametti, S., Donnizzelli, E., Pittia, P., Rovere, P. P., Squarcina, N., &
both safe and result in excellent foaming properties. Bonomi, F. (1999). Characterization of high-pressure-treated egg
albumen. Journal of Agricultural and Food Chemistry, 47, 3611–3616.
Acknowledgement Iametti, S., Donnizzelli, E., Vecchio, G., Rovere, P. P., Gola, S., &
Bonomi, F. (1998). Macroscopic and structural consequences of high-
pressure treatment of ovalbumin solutions. Journal of Agricultural and
This research has been supported by KU Leuven and Food Chemistry, 46, 3521–3527.
the Fund for Scientific Research Flanders (FWO). Kalab, M. (1970). Factors affecting the Ellman determination of sulfhy-
dryl groups in skimmilk powder and gels. Journal of Dairy Science, 53,
References 711–718.
Kato, A., Fujimoto, K., Matsudomi, N., & Kobayashi, K. (1986). Protein
Adler-Nissen, J. (1986). Enzymatic hydrolysis of food proteins (1st ed.). flexibility and functional properties of heat-denatured ovalbumin and
New York: Elsevier Applied Science Publishers (p. 427). lysozyme. Agricultural and Biological Chemistry, 50, 417–420.
Alizadeh-Pasdar, N., & Li-Chan, E. C. Y. (2000). Comparison of protein Kato, A., Komatsu, K., Fujimoto, K., & Kobayashi, K. (1985).
surface hydrophobicity measured at various pH values using three Relationship between surface functional properties and flexibility of
different fluorescent probes. Journal of Agricultural and Food Chem- proteins detected by the protease susceptibility. Journal of Agricultural
istry, 48, 328–334. and Food Chemistry, 33, 931–934.
1426 I. Van der Plancken et al. / Journal of Food Engineering 78 (2007) 1410–1426

Kitabatake, N., & Doi, E. (1987). Conformational change of hen egg Suzuki, K. (1960). Studies on the kinetics of protein denaturation under
ovalbumin during foam formation detected by 5,5 0 -dithiobis-(2-nitro- high pressure. Review of Physical Chemistry of Japan, 29, 91–97.
benzoic acid). Journal of Agricultural and Food Chemistry, 35, 953–957. Townsend, A.-A., & Nakai, S. (1983). Relationships between hydropho-
Ma, C., & Harwalkar, V. R. (1991). Thermal analysis of food proteins. bicity and foaming characteristics of food proteins. Journal of Food
Advances in Food and Nutrition Research, 35, 317–366. Science, 48, 588–594.
Mine, Y. (1995). Recent advances in the understanding of egg white Van der Plancken, I., Delattre, M., Indrawati, Van Loey, A., &
protein functionality. Trends in Food Science and Technology, 6, Hendrickx, M. E. G. (2004). Kinetic study on the changes in the
225–232. susceptibility of egg white proteins to enzymatic hydrolysis induced by
Mine, Y., Noutomi, T., & Haga, N. (1990). Thermally induced changes in heat and high hydrostatic pressure pretreatment. Journal of Agricul-
egg white proteins. Journal of Agricultural and Food Chemistry, 38, tural and Food Chemistry, 52, 5621–5626.
2122–2125. Van der Plancken, I., Van Loey, A., & Hendrickx, M. (2005a). Combined
Patel, P. D., Stripp, A. M., & Fry, J. C. (1988). Whipping test for the effect of high pressure and temperature on selected properties of egg
determination of foaming capacity of protein: A collaborative study. white proteins. Innovative Food Science and Emerging Technologies, 6,
International Journal of Food Science and Technology, 23, 57–63. 11–20.
Pedersen, B., & Eggum, B. O. (1981). Prediction of protein digestibility by Van der Plancken, I., Van Loey, A., & Hendrickx, M. (2005b). Changes in
in vitro procedures based on two multi-enzyme systems. Zeitschrift für sulfhydryl content of egg white proteins due to heat and pressure
Tierphysiologie, Tiernährung und Futtermittelkunde, 45, 190–200. treatment. Journal of Agricultural and Food Chemistry, 53, 5726–
Privalov, P. L., & Khechinashvili, N. N. (1974). A thermodynamic 5733.
approach to the problem of stabilization of globular protein structure: Van der Plancken, I., Van Loey, A., & Hendrickx, M. (2006). Effect of
A calorimetric study. Journal of Molecular Biology, 86, 665–684. heat treatment on the physico-chemical properties of egg white
Richwin, A., Raasch, A., Teichgraber, P., & Knorr, D. (1992). Effects of proteins: A kinetic study. Journal of Food Engineering, 75, 316–326.
combined temperature and high pressure treatment on the function- Van der Plancken, I., Van Loey, A., Hendrickx, M. (in press). Kinetic
ality of egg white proteins. European Food Science, 43, 27–31. study on the combined effect of high pressure and temperature on the
SAS Institute (1999). SAS user’s guide, Statistics. Cary, USA: SAS physico-chemical properties of egg white proteins. Journal of Food
Institute Inc. Engineering, doi:10.1016/j.jfoodeng.2005.09.018.
Strohalm, J., Novotna, P., Houska, M., Kyhos, K., Vavreinova, S., Van der Plancken, I., Van Remoortere, M., Indrawati, Van Loey, A., &
Gabrovska, D., et al. (2000). Influence of high pressure treatment on Hendrickx, M. E. (2003). Heat-induced changes in the susceptibility of
rheological and other properties of egg white. High Pressure Research, egg white proteins to enzymatic hydrolysis: A kinetic study. Journal of
19, 137–143. Agricultural and Food Chemistry, 51, 3819–3823.