Jung Han. Innovations in Food Packaging
Подождите немного. Документ загружается.
Surface
chemistry
of
food,
packaging and biopolymer materials
47
iLVi
Vapor
(\/)
^
P\e
Liquid
(L) X
^
1
Solid
(S)
Figure 4.1
The
wetting
of
a
solid surface
by a
liquid drop.
liquid-vapor interfacial energy
(yiv)^
the solid-vapor interfacial energy
(ysv)^
and the
solid-liquid interfacial energy
(ysi)
(Figure 4.1) (Han and Krochta, 1999). The wet-
ting of a solid surface
is
governed by
the Young
equation, which establishes the equi-
librium relationship between
the
contact angle
6 and the
three aforementioned
interfacial tensions (Ornebro et ah, 2000):
JLV
XCOSO = ysv - JSL (4-1)
According
to
this equation,
a
droplet with high surface tension resting
on a low-
energy solid tends
to
adopt
a
spherical shape due
to
the establishment
of
a
high con-
tact angle with
the
surface. Conversely, when
the
solid surface energy exceeds
the
liquid surface tension,
the
droplet tends
to
adopt
a
flatter shape due
to
the establish-
ment
of
a
low
contact angle with
the
surface. Thus, wetting means that
the
contact
angle between
a
liquid
and a
solid
is so
nearly zero that the liquid spreads over
the
solid
easily.
Similarly, non-wetting means that the angle is greater than
90°,
and there-
fore the liquid will ball up and run off the surface easily (Adamson and Gast, 1997).
Surface energy and spreading coefficient
Another valuable thermodynamic parameter
to
describe wetting
and
other surface
properties
is
the surface free energy. The condition
for
wetting
or
spreading
of
a
liq-
uid on
a
solid surface
is
decided by the spreading coefficient, Sys'.
SL/S =
ysv - JLV - JSL (4.2)
This equation means that
a
liquid will spread out on the surface
of
a
solid iiSys
> 0,
but will form
a
droplet with
a
finite
contact angle if Sys
<
0.
In
other
words,
wetting
depends
on
the surface tension
of
the
liquid and
on
the free energy
of
the
interfaces
(Ornebro
et al,
2000).
For
good adherence
to
take place,
the
interfacial energy
between the two materials must be small. As
it
applies
to
food packaging,
the
liquid
usually denoted is water. Thus,
to
prevent wetting from occurring, biopolymer pack-
aging materials should be designed to have
a
high interfacial tension with respect
to
water (Lawton, 1995).
48 Innovations in Food Packaging
As mentioned previously, there are two separate and additive attractive forces oper-
ating across the soHd-liquid interface; the dispersion Lifshitz-van der Waals forces,
and the
Lewis acid-base forces. These
are
notated
as
y^^and y^^, respectively. The y^^
forces are mainly due to dipoles induced between adjacent
molecules,
whereas the 7^^
forces are generated by polar interactions between electron-acceptor (Lewis acid: y"^)
and electron-donor (Lewis base: y~) species (Kaya et a/., 2000; Bialopiotrowicz,
2003).
Surface free energy is then expressed as:
where y^^
is
the Lifshitz-van der Waals component, and y^^ is actually a geometric
mean between
y"^
and y~:
jAB = 2(y+ X y-)i/2 (4.4)
Taking into account equations (4.3) and (4.4), van Oss (1994) derived the following
equation describing the work of adhesion of
a
liquid to a solid:
^SL
—
7L >^ (1 +
COS
0) = ys + JL ~ JSL
= 2{yf
X
yi^f^ + 2(yJ
X
ylf'^ + 2{ys X ylf^ (4.5)
In this equation,
WSL
represents the work of adhesion of a liquid to a solid (that
is,
the
work necessary to separate a unit area of
the
interface SL into two liquid-vapor and
solid-vapor interfaces); y^-^and y^^are the Lifshitz-van der Waals components of
solid S and liquid L surface free energies, respectively; yj and y^ are the electron-
acceptor
and
electron-donor parameters of
the
Lewis acid-base component of the surface
free energy of a solid; and ji and yl are the electron-acceptor and electron-donor
parameters of
the
Lewis acid-base component of
the
surface free energy of
a
liquid.
According to Lawton (1995), for good adherence to take place, the interfacial energy
between the two materials must be small. This model has been usefiil to compute bac-
terial adhesion to surfaces involving acid-base and electrostatic interactions (Michalski
et
al.,
1998).
The computation of the above parameters for the unknown solid through
equation (4.5) can be made by a three-liquid method, namely two polar and one apolar
liquid substances of known surface tension parameters. Thus, the polar components
can
be
evaluated from the settings of the equation using the two polar
liquids,
whereas
y^^is evaluated from the apolar liquid for which the polar components are zero (Kaya
et ai, 2000; Bialopiotrowicz, 2003).
Zisman equation and critical surface tension
Zisman and coworkers found that cos 6 is usually a monotonic function of ji for a
homologous series of liquids:
cos^
=
a-byL=l-
KJL - 7c) (4.6)
where jc
is
the critical surface tension and is
j8
a constant (Zisman,
1973).
The critical
surface tension can be obtained from the Fox-Zisman plot (cos
6
vs. ji) by extra-
polation
to cos 6
= \ with various liquids of different ji
values.
According to McGuire
and Kirtley (1988), jc is solely a function of the surface properties of a solid, and
Surface
chemistry
of
food,
packaging
and
biopolymer
materials
49
related to its true surface energy. Thus y^ becomes a characteristic property, and has
provided
a
useful means of summarizing wetting behavior and allowing predictions of
interpolative nature. Liquids with surface tension < y^ will spread on the surface,
completely wetting the
solid.
Nevertheless, it
is
important
to
remember that yc
is
not y^;
the latter is most probably larger than the former, especially when considering polar
materials such as starch (Toussaint and Luner,
1993;
Lawton, 1995).
McGuire's theory and equation
In 1972, Fowkes emphasized that work of adhesion
{W^j)
may also be divided into
dispersive energy
{d)
and polar energy (p) components:
WsL
= WIL +
W^sL
(4.7)
With respect to solid-liquid contact, a further identification of
WSL
was generated as:
WsL
=WIL+ Wl^ = 2(yi X y|)i/2 + 2(yf
X
y§y^^ (4.8)
From equation (4.5),
^SL = 7z X (1 + COS0) - 2(yi X ylY^^ (4.9)
Contact angle data recorded for a series of diagnostic liquids with known values of yf
on a surface of known y| should therefore lead to the evaluation of
WSL-
Since y^ of
each diagnostic liquid is known, plots of W^i vs y ^ can be constructed for any mate-
rial targeted for food contact (McGuire and Kirtley, 1988). Experimentally derived
plots led to the finding that the relationship between
W§i
and y^ was linear for all the
testing materials, which strongly suggested that the relationship was independent
from the diagnostic liquids. McGuire and Kirtley (1989) arranged this fact into math-
ematical terms as:
W^L
= kyl + b (4.10)
where the slope k (dimensionless) and intercept b (mJ/m^) are specific to each solid
surface.
The value of the polar contribution to the work of adhesion J^f^ depends upon the
polar character of both the solid and the liquid that are in contact, and provides either
a measure of solid surface hydrophilicity (Yang et a/., 1991) or a relative index of
surface hydrophobicity/hydrophilicity (McGuire and Sproull, 1990).
Neumann's equation
The solid surface tension, ysv, can be calculated by combining
the
Young
equation:
y^KCOS0 = ysv-
JSL
(4.11)
and the Neumann's equation of
state
for interfacial tensions:
ysL
=
JLV
+ ysv -
2(yLv
X
ysv)'^'
X e-^(yLv-ysvf (4.12)
50 Innovations in Food Packaging
where y^^j/is the hquid surface tension,
JSL
the soUd
Hquid interfacial tension, and
j8
the
Neumann's constant
(j8
= 0.000115
mVmJ).
By ehminating y^^froni these equations,
we get:
COS0 = -1 +
2{ysvlyLvf'^
X
e-^^^Lv-ysvf
(4.13)
Therefore, solid surface tension jsv
can
be computed by equation (4.13) based on the
experimentally measured contact angle and
the
known liquid surface tension (Matuna
and Balatinecz, 1998).
Harmonic mean method
Another way to estimate surface energy of
solids
is derived from liquid-solid contact
angle measurements, as follows:
(1 +
cos
eyyL = ^y'f
X
7^/(7^ + rD + 4rf
X
o^/(yf
+ 4) (4.14)
In this equation, there are two unknowns:
75-^
and 7^. When the contact angle is
measured for two different liquids on the solid given that the liquid's yf ^ and y£are
known, the following relationships are produced:
(1 + cos0,)/yLi = ^iT
X
yWHyW + TD + 4rf x yli{yl, + -i) (4.i5)
(1 + cos02)/yi2 = Ayf
X
yWlilW + TD + 4^?
X
yl^l{yl^ + T|) (4.16)
where
y^and
yi^^are the surface energy for the two probe liquids. In fact, these two
equations, known as the harmonic mean method, are able simultaneously to obtain
solid surface energy and its polar and apolar components (Lawton, 1995). There are
two general criteria for selecting probe liquids for the harmonic mean method: the
probe liquids should
have
very different polarities (water and methylene iodide are the
two liquids most often used), and the liquids should be inert with respect to the solid
(i.e.
the liquid should not react, swell or dissolve the solid being tested) (Lawton,
1995).
Germain's method
An alternative procedure to evaluate liquid-to-solid adhesion phenomena is the
Germain's method, based on the following equation:
WsL
=Wi+W^,
= l{yi X yi^l^ + 2(yf
X
ylf^ + 2(yi X
y>^sf^
(4.17)
where y^ is the surface tension component due to hydrogen bonds, and
yP
the polar
Lifshitz-van der Waals component. This model has been used to study the adhesion
of ink to polymer surfaces (Michalski et al, 1998).
Techniques for measuring the contact angle
Measuring the contact angle is the most common method to estimate surface
hydrophobicity (McGuire and Kirtley, 1988; Jones et al., 1999; Boonaert and
Surface
chemistry of
food,
packaging and biopolymer materials 51
Rouxhet, 2000). There are two basic techniques for measuring contact angles of non-
porous solids: goniometry and tensiometry.
Goniometry
Analysis of the shape of
a
drop of test liquid placed on a solid is the basis for gonio-
metry, also referred to as the sessile drop technique. Clean, dry samples of polymer
films are mounted on a glass slide using double-sided sticky tape. Samples should
have no surface irregularities (such as seams, lettering or visible scratches). A
5-|UL1
drop of distilled water
is
placed
on the
sample using a glass
syringe.
The tangent
to
the
point of contact at the solid-liquid-vapor interface is then measured using a contact
angle goniometer to give the advancing contact angle 6 (Jones et al., 1999). Basic
elements of a goniometer are a light source, sample stage, and lens, and an image
capture device. The contact angle can also be roughly estimated using the following
formula:
eC) = larctg
(2
H/W) (4.18)
where H
is
the height of the drop and ^is the width of the solid-liquid interface. To
use this method, water drops
(5-|IJL1)
are deposited on the solid surface of interest and
photographs are taken within 5 seconds. Negatives are then projected on screen to
allow estimation of the droplet dimensions (Sinde and Carballo, 2000).
Goniometry can be used in many situations where substrates can provide regular
curvature.
A very small quantity of liquid
is
needed. Sessile drop techniques and asso-
ciated equipment are currently being used for quality control at food packaging man-
ufacturing plants (e.g. polymeric coatings applied to paper). Tests can be done very
easily if solid substrates have a relatively flat portion for testing and can fit on the
stage of
the
instrument (McGuire and
Yang,
1991). However, there are some limita-
tions.
Measurements have been performed under different experimental conditions
and with different procedures in different laboratories, which makes comparison of
results very difficult - if not impossible (McGuire and Yang, 1991). It has been found
that contact angles decrease with decreasing drop diameter for certain solid-liquid
systems (Good and Koo, 1979), and increases with increasing drop diameter in a
range below a limiting value identified as the critical drop volume
(CDV).
CDV varies
among materials, and needs to be known to allow measurement standardization
(McGuire and Yang, 1991). In addition, contact angles of most test liquids with
hydrophilic surfaces tend
to
be rather small, which increases the possibility of error in
experimental measurements (Noda, 1993).
Recently, more advanced equipment has been developed to test contact angles. A
digital microscope is connected to a personal computer. Polymer films are glued onto
a well-leveled platform, and the microscope is positioned horizontally for capturing
the side-view image (Figure 4.2). At the lens position of 10, the side-view image is
acquired and converted into a binomial edge-enhancing picture using conventional
photo-editing software. Contact angles of the digital images are measured by hand if
the sample size is small, or using line-slope calculation software if sample sizes are
large (Han and Krochta, 1999).
52 Innovations in Food Packaging
Figure 4.2 Different contact angles of various liquids on polypropylene film surface (30 mj/m^ of surface energy). A: DMSO
(44.0 mj/m^ of surface tension); B: ethylene glycol (48 mj/m^); C: formamide (58.2 mj/m^); D: glycerol (63.4 mj/m^); and E: water
(72.8mJ/m2).
Tensiometry
The tensiometric method for the determination of contact angles measures the forces
present when a sample solid is brought into contact with a test liquid. If the forces of
interaction, geometry of the solid, and surface tension of the liquid are known, the
contact angle may be calculated. The sample of the solid to be tested is hung on the
balance and
tared.
The liquid, of known surface tension, is then raised
to
make contact
with the solid. When the solid is in contact with the liquid, the change in forces is
detected and the height is registered as zero depth of immersion. As the solid is
pushed into the liquid, the forces on the balance are recorded. An alternative and prob-
ably now more widely used procedure is to raise the liquid level gradually until it just
touches the hanging solid suspended from the balance. The change of weight is then
noted. A general equation is:
y^^cos0 = APF/p (4.19)
where AfF is the change in weight (that is, the force exerted by the solid when it is
brought into contact with the liquid),/? the geometrical perimeter of the solid, and jiy
is the liquid surface tension. When the solid is advanced into the liquid, this contact
angle is the advancing contact angle; otherwise, it is the receding contact angle. At
any point on the immersion graph, all points along the perimeter of
the
solid at that
depth contribute to the recorded force measurement. Thus, the force used to calculate
contact angles at any given depth of immersion is already an averaged value. This
technique is also very useful in studying hysteresis. Variations of contact angles, both
advancing and receding, for the entire length of
the
sample tested, are visualized on
the same graph (Adamson and Gast, 1997).
'•tH\
ed research
Some commonly used food packaging synthetic polymers show rather low surface
energies (Table
4.1).
Improving these synthetic polymers surface energies is better for
package bonding, printing, and coating with other films. Good adhesion in the seal
areas is also highly desirable at polymer-polymer
or
polymer-metal interfaces in food
Surface chemistry of
food,
packaging and biopolymer materials 53
Table 4.1 Sutface energies for some polymers (mj/m^)
PE PET PVP TPX PMMA PVCH PS
32.0 ± 1.6 38.0 ± 2 50.0 ± 2.0 21.5 ± 0.1 40.0 ± 0.2 29.0 ± 1.0 30.0 ± 1.0
PVCH,
poly (vinylcyclohexane); PE, polyethylene; PET, poly (ethylene terephthalate); TPX, poly
(4-methyl-1-pentene); PS, polystyrene; PVP, poly (2-vinylpyridine); PMMA, poly (methyl methacrylate).
From Tirrell (1996).
packages, particularly those composed of laminated materials. Without adequate
adhesion, food may become contaminated with food-borne organisms or extraneous
materials (Ozdemir et
al.,
1999).
Some technologies that have been developed to improve the surface energy of food
packaging polymers include flame, corona, and plasma treatments, the latter being
considered the most effective. Plasma treatments can drastically increase bond
strengths (adhesion) of commonly used packaging polymers. For example, the bond
strength of LDPE and PP increased 20-fold and
7-fold,
respectively, compared to
untreated films (Ozdemir et ai, 1999). Plasma treatments have also been reported as
effective in improving adhesion between a polymer and a metal substrate in the
absence of an adhesive (Sapieha et al., 1993;
O'Kell
et al., 1995). Figure 4.3 shows
the peel strength of air- and nitrogen-plasma-treated aluminum/PE/aluminum lami-
nates prepared by direct melting and pressing of PE films onto the aluminum sub-
strates without any adhesives. Peel strength values of air-plasma-treated PE samples
were considerably higher than those of nitrogen-plasma-treated counterparts. A
70-
to
75-fold increase in peel strength improvement was achieved at the PE/aluminum
interface of air-plasma-treated samples, compared
to
only
3-
to
6-fold
increases follow-
ing exposure to nitrogen treatment. The improvement in peel strength was due to the
incorporation of 02-containing
fimctional
groups into the PE surface during the plasma
treatment. In addition, the peel strength improvement between polymer-polymer and/or
polymer-metal interfaces without the use of adhesives following exposure to a plasma
treatment can significantly limit the use of volatile organic solvents. Eliminating these
volatile substances
fi*om
adhesive formations will reduce damage to the environment
and limit health risks
fi-om
these hazardous solvents (Ozdemir et al, 1999).
Surface properties of conventional plastic films can be modified by biopolymer
coating. Whey protein has been coated onto plastic films to increase the surface
energy of the plastic films to be hydrophilic (Hong et ai, 2004).
With respect to the adhesion of food materials to synthetic package surfaces, most
studies in
the
literature are empirical, without taking into account
the
existing theoretical
adhesion models (Michalski et
al.,
1997). Oil raises most concerns, because adhesion
of fat or oil on packages increases recycling costs and enhances interactions that may
alter the food product, leading to a poor product appearance. Michalski et al (1998)
studied the phenomenon using virgin olive oil, refined first-draft sunflower oil, soy-
bean
oil,
and pure white vaseline oil to test
the
correlation between model-predicted and
experimental adhesion. Low-density polyethylene (LDPE), polyethylene terephthalate
54 Innovations in Food Packaging
90
80-1
70
J
Z
60-j
D) 50H
c
0
"S 40
"55
I
30
20
10H
0
•
Air plasma
•
Nitrogen plasma
/
0 10 20 30 40 50
Treatment time (sec)
60 70
80
Figure
4,3
Peel strength
of
air-plasma- and nitrogen-plasma-treated polyethylene/aluminum laminates
(modified from
O'Kell
etal.,
1995).
(PET),
stainless steel AISI 304, and glass were chosen to be solid package surfaces.
Results showed that experimental adhesion correlated well with the McGuire's and
Germain's methods. These two approaches seemed to be efficient in predicting bulk
adhesion in oils, particularly the latter one.
Unlike research on synthetic films, most research on biopolymer films has focused
on water barrier properties, because low water barrier ability
is the
biggest short-
coming of biopolymer films and seriously limits their application as packaging mate-
rials.
Biopolymer films are expected to be moderate
to
excellent O2 barriers due
to
their tightly packed, ordered hydrogen-bonded network structure (Fang and Hanna,
2000;
Diaz-Sobac and Beristain, 2001), but because of their inherent hydrophilicity,
their water vapor permeability (WVP) is high. Thus, lipids or other hydrophobic sub-
stances such as resins, waxes, fatty acids or even some non-soluble proteins need to
be added
to
retard moisture transfer, essentially related to the low mobility
of
fatty
acid chains (Fairley et
ai,
1995;
Guilbert
et
al,
1996; Callegarin et
al.,
1997; Sherwin
et
ai,
1998). Sebti et
al.
(2002) used hydroxy-propyl-methyl cellulose (HPMC) com-
bined with stearic acid to make packaging
films.
They found that the incorporation of
stearic acid increased the contact angle
fi-om
49° to 82°, and decreased WVP by 40%.
Viroben et
al.
(2000) tested the hydrophilicity
of
films
made from pea-protein iso-
lates,
adding
1,3-propanediol
(PRD) and
1,2-propanediol
(PRG) as plasticizers. The
results are presented in
Table
4.2, which also includes previously obtained data by the
same researchers (Guegen et al., 1998)
for
comparison purposes. The contact angle
was much higher for PRG than for PRD, and even higher than those for other plasti-
cizers.
This was primarily attributed
to
the low surface energy
of
PRG (38mJ/m^)
compared
to
that of PRD (49
mJ/m^).
However, PRG-added
films
also showed a much
higher
WVP.
In
order to make pea-protein films more hydrophobic, monoglycerides
were incorporated into the film-forming dispersion, but the resulting contact angles
Surface chemistry of
food,
packaging and biopolymer materials 55
Table 4.2 Effect of plasticizer type on barrier properties of pea-protein films
Plasticizer type
Ethylene glycol*
1,2-propanediol(PRG)
1,3-propanediol
(PRD)
Glycerol*
Diethylene glycol*
Triethylene glycol*
Tetraethylene glycol*
No.
of
C atoms
2
3
3
3
4
6
8
Film plasticizer
content
9.2
23.8
25.0
33.9
35.4
38.3
40.4
(%)
Contact
of water
21
67
32
40
13
21
40
angle
n
Water vapor permeability
(XlO^V'"^sPa)
19
37
47
29
21
28
28
*Data from Gueguen
etal.
(1998) included for comparison with PRG and PRD (from Viroben
etal.,
2000).
Table 4.3 Effect of added monoglycerides on contact angle of pea-protein films
Fatty acid
Oleic
Linoleic
Ricinoleic
Myristic
Heptanoic
Undecenoic
Contact angle (**)
21
19
18
29
15
19
From Viroben
etal.
(2000).
were rather low and did not improve hydrophobicity (Table 4.3). The only advantage
of using these monoglycerides was that they allowed for the preparation of very soft
films, which did not dry after exposure to air at ambient temperature for as long as
30 days, significantly improving their aging behavior.
Starch has received considerable attention because of its totally biodegradable,
inexpensive
nature.
Law1:on (1995) tested the hydrophilicity of
films
and ribbons made
fi*om three kinds of com starch: high amylose, waxy, and
normal.
Eight different probe
liquids with known surface energy
7^^
and ^f were used
to
test contact angles and sur-
face energy It was found that contact angles decreased as surface energy decreased.
Critical surface energy ranged from 38 to 43mJ/m^, and there were no differences
related
to
the type of starch or processing
conditions.
The surface energy of starch films
and ribbons ranged from 35.7 to
41.2
mJ/m^
- that is, less than the corresponding 7^
but similar to polyethylene. The author considered that because of the interaction
between the polar probe liquid and starch, not enough time was available for the polar
probe liquid to get into equilibrium with the starch films or ribbons when testing the
contact angle. This led to quick measurements of
the
contact angle, and large values
resulted - comparable to those of non-polar solids. To predict which coating will
adhere to starch, the author emphasized that unbiased values for surface energy of
starch and its polar and apolar components need to be obtained through more reliable
methods than contact angle measurements.
56 Innovations in Food Packaging
—^^—
-
<)-
--<>•-"
PEA
PCL
PLA
40 60 80
Polyester content (wt. %)
Figure
4.4
Effect
of
polyester content (PEA, PLA, and PCL, % w/w)
on the
hydrophilic character
of
plasticized wheat starch films (from Martin
etal.,
2001; reprinted with permission ofWiley-VCH).
One strategy
to
overcome starch hydrophilic weakness
is to
associate
it
with
a
moisture-resistant polymer. Martin
et
al. (2001) reported water contact angle mea-
surements performed
on
starch-polyester blends
as
well
as on
pure polyesters
and
plasticized starch (Figure 4.4). Starch films, plasticized with glycerol (glycerol/starch
ratio
=
0.54), were blended with polyesteramide (PEA), polylactic acid (PLA), and
polycaprolactone (PCL). All blends showed a rapid decrease of the hydrophilic char-
acter as the amount of polyester in the mix increased from
0
to
10%,
and kept decreas-
ing until the value of pure polyester was reached. These researchers also performed
water immersion tests of the starch-polyester multilayers to check their ability
to
resist
water penetration. They showed good moisture resistance
by
standing the water
for
days without delamination of product swell.
Another strategy to overcome the hydrophilicity of biopolymer films is to incorpo-
rate lipids into the film matrix. Compared to laminated films, emulsified biopolymer
films are less efficient water vapor barriers, but require only
a
single step in process-
ing, exhibit good mechanical and adhesive properties, and can be made at room tem-
perature (Kamper
and
Fennema, 1984; Quezada-Gallo
et
al., 2000). Peroval
et al.
(2002) used arabinoxylans with four different kinds
of
lipids (palmitic acid,
Ci6;
stearic acid, Cig; hydrogenated palm oil, OK35; and triolein) to make edible emulsi-
fied films. The experimental results are presented in Table 4.4.
Arabynoxylan-only films were characterized by contact angles
of
about 70°,
and
did not differ significantly from those with addition of
Ci6
or
Cig.
The best hydropho-
bic surface, showing contact angles >90°,
was
obtained
by
adding OK35. These
values were similar to those regularly observed with LPDE. Nevertheless, angles
of
arabinoxylan-OK35 films decrease
10
times faster versus time (data not shown), sug-
gesting
a
rapid change
in
surface properties. This study showed that higher water