Low-Rate Dynamic Contact Angles on Poly(methyl

methacrylate/ethyl methacrylate, 30/70) and the

Determination of Solid Surface Tensions

D. Y. KWOK,* A. LI, A. W. NEUMANN

Department of Mechanical and Industrial Engineering, University of Toronto, 5 King’s College Road,
Toronto, Ontario M5S 3G8, Canada

Received 19 October 1998; revised 3 February 1999; accepted 8 February 1999

ABSTRACT: Low-rate dynamic contact angles of 12 liquids on a poly(methyl methacry-

late/ethyl methacrylate, 30/70) P(MMA/EMA, 30/70) copolymer were measured by an
automated axisymmetric drop shape analysis-profile (ADSA-P). It was found that five
liquids yield nonconstant contact angles, and/or dissolve the polymer on contact. From
the experimental contact angles of the remaining seven liquids, it is found that the
liquid–vapor surface tension times cosine of the contact angle changes smoothly with
the liquid–vapor surface tension (i.e., g lv cos u depends only on g lv for a given solid
surface or solid surface tension). This contact angle pattern is in harmony with those
from other methacrylate polymer surfaces previously studied.45,50 The solid–vapor
surface tension calculated from the equation-of-state approach for solid–liquid interfa-
cial tensions14 is found to be 35.1 mJ/m2, with a 95% confidence limit of 6 0.3 mJ/m2,
from the experimental contact angles of the seven liquids. © 1999 John Wiley & Sons, Inc.
J Polym Sci B: Polym Phys 37: 2039 –2051, 1999
Keywords: contact angle, dynamic; contact angle, equation of state; contact angle,
drop shape; contact angle, complexity; surface tension, of solid; surface tension, from
contact angle; surface tension, methacrylate polymer

INTRODUCTION tion into columns of particle powder,18 –21 sedi-

Solid surface tensions are important thermody-
namic quantities governing many technological
processes. However, because of the absence of
surface mobility, a solid phase is very different
from a liquid phase; hence, one cannot measure
the surface tension of a solid phase directly as is
the case for a liquid phase. Thus, several indepen-
dent approaches have been used to estimate solid
surface tensions, including direct force measure-
ments,1–9 contact angles,10 –17 capillary penetra-
* Present address: Department of Chemical Engineering,
Massachusetts Institute of Technology, 77 Massachusetts
Ave., Cambridge, Massachusetts 02139
Correspondence to: A. W. Neumann (E-mail: neumann@
mie.utoronto.ca)
Journal of Polymer Science: Part B: Polymer Physics, Vol. 37, 2039 –2051 (1999)
© 1999 John Wiley & Sons, Inc. CCC 0887-6266/99/162039-13
mentation of particles,22–25 solidification fronts of
particles,26,27 gradient theory,28 and Lifshitz the-
ory of van der Waals forces.28 Among these meth-
ods, contact angle measurements are believed to
be the simpliest approach.
At the center of contact angle research is
Young’s equation,
g lvcos u Y 5 g sv 2 g sl (1)
which interrelates the Young contact angle with
the interfacial tensions of the liquid–vapor g lv ,
solid–vapor g sv , and solid–liquid g sl interfaces; u Y
is the Young contact angle (i.e., a contact angle
which can be used in conjunction with Young’s
equation). While there are a number of thermo-
dynamic equilibrium contact angles u e , they are
not necessarily equal to u Y . 29,30
2039
2040 KWOK, LI, AND NEUMANN
Equation (1) implies a single, unique contact
angle; in practice, however, contact angle phe-
nomena are complicated.29,30 For example, the
contact angle made by an advancing liquid ( u a )
and that made by a receding liquid ( u r ) are not
identical; nearly all solid surfaces exhibit contact
angle hysteresis, H (the difference between u a
and u r ):
H 5 ua 2 ur (2)
Contact angle hysteresis can be due to rough-
ness and heterogeneity of a solid surface. If
roughness is the primary cause, then the mea-
sured contact angles are meaningless in terms of
Young’s equation. It is well known that on very
rough surfaces, contact angles are larger than on
chemically identical smooth surfaces21; such an-
gles do not reflect material properties of the sur-
face; rather, they reflect morphological ones. Mod-
els have been employed to gain a better under-
standing of contact angle hysteresis.29 –33
In general, the experimentally observed appar-
ent contact angle, u, may or may not be equal to
the Young contact angle, u Y 29,30: (1) On ideal solid
surfaces, there is no contact angle hysteresis and
the experimentally observed contact angle is
equal to u Y ; (2) On smooth, but chemically heter-
ogeneous solid surfaces, u is not necessarily equal
to the thermodynamic equilibrium angle. Never-
theless, the experimental advancing contact angle,
ua, can be expected to be a good approximation of uY.
This has been illustrated using a model of hetero-
geneous (smooth) vertical strip surfaces.29,30 There-
fore, care must be exercised to ensure that the ex-
perimental apparent contact angle, u, is the advanc-
ing contact angle in order to be inserted into
Young’s equation; (3) On rough solid surfaces, no
such equality between advancing contact angle and
uY exists. Thus, all contact angles on rough surfaces
are meaningless in terms of Young’s equation.
While the receding angle on a heterogeneous and
smooth surface can also be a Young angle, it is
usually found to be nonreproducible often because
of sorption of the liquid into the solid and swelling of
the solid by the liquid.34
The thermodynamic equilibrium angles on
rough and heterogeneous surfaces are the so-
called Wenzel35 and Cassie36,37 angles, respec-
tively. They are not equal to u Y ; furthermore, they
are not experimentally accessible quantities.
Recently, we have shown38 – 40 that measuring
contact angles at very slow motion of the three-
phase contact line allows direct observation of
surface quality. In addition, when such proce-
dures are interpreted by an automated axisym-
metric drop shape analysis—profile (ADSA-P),
complexities such as dissolution of the polymer by
the liquid and slip/stick of the three-phase contact
line, which affect the contact angle interpretation
in terms of surface energetics, can be identi-
fied.40 –50 It was found that many of the readily
accepted assumptions in interpreting contact an-
gles are often violated when contact angles are
measured by conventional goniometer tech-
niques. These assumptions include the applicabil-
ity of Young’s equation and constancy of g lv , g sv ,
and g sl during the experiments. A more complete
discussion of these matters can be found else-
where.51
In previous studies,45,50 low-rate dynamic con-
tact angles of various liquids were reported on
two methacrylate polymers: poly(methyl methac-
rylate) PMMA45 and poly(ethyl methacrylate)
PEMA.50 It was found that g lv cos u changes
smoothly with g lv , on one and the same solid
surface. Thus, one would expect that the result for
a polymer similar to these methacrylate polymers
should behave similarly or should lie somewhere
in between. Thus, it is the aim of this study to
investigate this expectation for a 30 : 70 copoly-
mer of poly(methyl methacrylate) and poly(ethyl
methacrylate), P(MMA/EMA, 30/70).
MATERIALS (SOLID SURFACE AND
LIQUIDS)
Poly(methyl methacrylate/ethyl methacrylate,
30/70) P(MMA/EMA, 30/70) copolymer was pur-
chased from Polysciences (Warrington, PA; cat#
06532) as fine beads. A 2% P(MMA/EMA, 30/70)/
chloroform solution was prepared using chloro-
form (Sigma-Aldrich, 99.91% A.C.S. HPLC
grade) as the solvent. Silicon wafers ^100& (Silicon
Sense, Naschua, NH; thickness: 525 6 50 mi-
crons) were selected as the substrate for the poly-
mer coating. They were obtained as circular discs
of about 10 cm diameter and were cut into rect-
angular shapes of about 2.5 cm 3 5 cm. Each
rectangular wafer surface was then soaked in
chromic acid for at least 24 h, rinsed with doubly-
distilled water, and dried under a heat lamp be-
fore polymer coating. The P(MMA/EMA, 30/70)-
coated surfaces were prepared by a solvent-cast-
ing technique described elsewhere.45– 47
 DYNAMIC CONTACT ANGLES AND SOLID SURFACE TENSIONS 2041

Table I. Supplier, Purity, and Surface Tension of the Liquids Used

Liquid Supplier % Purity Density (g/cm3) g lv (mJ/m2) No. of Drops

Dibenzylamine aldrich 97 1.026 40.80 6 0.06 9

Dimethyl sulfoxide (DMSO) sigma-aldrich 99.9 (HPLC) 1.101 42.68 6 0.03 7
1-Iodonaphthalene aldrich 99 1.740 42.92 6 0.03 10
1-Bromonaphthalene aldrich 98 1.489 44.31 6 0.05 7
1,3-Diiodopropane aldrich 99 2.576 46.51 6 0.13 10
3-Pyridylcarbinol aldrich 98 1.124 47.81 6 0.03 10
1,1,2,2-Tetrabromoethane aldrich 98 2.967 49.29 6 0.05 10
Diiodomethane aldrich 99 3.325 49.98 6 0.02 10
2,29-Thiodiethanol aldrich 991 1.221 53.77 6 0.03 10
Formamide aldrich 99.51 1.134 59.08 6 0.04 10
Glycerol baker analyzed 99.8 1.258 65.02 6 0.04 8
Water LASTa doubly distilled 0.977 72.70 6 0.09 10
a
Laboratory of applied surface thermodynamics.

With respect to the low-rate dynamic contact METHODS AND PROCEDURES

angle measurements by ADSA-P, liquid was sup-
plied to the sessile drop from below the wafer
surface using a motorized syringe device.40,41 In
order to facilitate such an experimental proce-
dure, a hole of about 1 mm diameter was made, by
using a diamond drill bit from Lunzer (New York,
NY; SMS-0.027), in the center of each rectangular
wafer surface before soaking in chromic acid. This
strategy was proposed because of its potential for
avoiding drop vibrations and for measuring true
advancing contact angles without disturbing the
drop profile. In order to avoid leakage between a
stainless steel needle (Chromatographic Special-
ities, Brockville, Ont.; N723 needles pt.# 3,
H91023) and the hole (on the wafer surface), Te-
flon tape was wrapped around the end of the
needle before insertion into the hole. In the liter-
ature, it is customary to first deposit a drop of
liquid on a given solid surface using a syringe or
a Teflon needle; the drop is then made to advance
by supplying more liquid from above using a sy-
ringe or a needle in contact with the drop. Such
experimental procedures cannot be used for
ADSA-P since ADSA determines the contact an-
gles and surface tensions based on a complete and
undisturbed drop profile.
Twelve liquids were chosen in this study. Se-
lection of these liquids was based on the following
criteria: (1) they should include a wide range of
intermolecular forces; (2) they should be nontoxic;
and (3) the liquid surface tension should be higher
than the anticipated solid surface tension.10,14,21
They are listed in Table I, together with the phys-
ical properties and surface tensions (measured at
23.0 6 0.5°C).
ADSA-P is a technique to determine liquid–fluid
interfacial tensions and contact angles from the
shape of axisymmetric menisci (i.e., from sessile
as well as pendant drops).52 Assuming that the
experimental drop is Laplacian and axisymmet-
ric, ADSA-P finds a theoretical profile that best
matches the drop profile extracted from an image
of a real drop, from which the surface tension,
contact angle, drop volume, surface area, and
three-phase contact radius can be computed. The
strategy employed is to fit the shape of an exper-
imental drop to a theoretical drop profile accord-
ing to the Laplace equation of capillarity, using
surface/interfacial tension as an adjustable pa-
rameter. The best fit identifies the correct surface/
interfacial tension from which the contact angle
can be determined by a numerical integration of
the Laplace equation. Details of the methodology
and experimental set-up can be found else-
where.40,41,52–54 A more recent development of the
ADSA-P program is available.55
Sessile drop experiments were performed by
ADSA-P to determine contact angles. The temper-
ature and relative humidity were maintained, re-
spectively, at 23.0 6 0.5°C and at about 40%. It
has been found that, since ADSA assumes an
axisymmetric drop shape, the values of liquid sur-
face tensions measured from sessile drops are
very sensitive to even a very small amount of
surface imperfection, such as roughness and het-
erogeneity, while contact angles are less sensi-
tive. Therefore, the liquid surface tensions used in
this study were independently measured by ap-
plying ADSA-P to a pendant drop, since the axi-
Figure 1. Low-rate dynamic contact angles
on a P(MMA/EMA, 30/70)-coated wafer surface
for (a) water; (b) glycerol; (c) formamide; (d)
2,29-thiodiethanol; (e) 3-pyridylcarbinol; (f) 1,3-
diiodopropane; and (g) 1-iodonaphthalene.
2044 KWOK, LI, AND NEUMANN
symmetry of the drop is enforced by using a cir-
cular capillary. Results of the liquid surface ten-
sion from previous studies are reproduced in
Table I.
In this study, six dynamic contact angle mea-
surements at velocities of the three-phase contact
line in the range from 0.1 to 1.0 mm/min were
performed for each liquid. The choice of this ve-
locity range was based on previous studies38 – 40
which showed that low-rate dynamic contact an-
gles at these velocities are essentially identical to
the static contact angles, for these relatively
smooth surfaces.
RESULTS AND DISCUSSION
Of the 12 liquids used, it was found that only
seven liquids yielded usable contact angles. They
are water, glycerol, formamide, 2,29-thiodietha-
nol, 3-pyridylcarbinol, 1,3-diiodopropane, and
1-iodonaphthalene. The remaining five liquids ei-
ther dissolved the polymer on contact or yielded
nonconstant contact angles during the course of
the experiments.
Figure 1(a)–(g) shows, respectively, typical ex-
perimental results of water, glycerol, formamide,
2,29-thiodiethanol, 3-pyridylcarbinol, 1,3-diiodo-
propane, and 1-iodonaphthalene: all contact an-
gles are essentially constant, as the drop volume
V increases and hence the three-phase contact
radius R. Increasing the drop volume in this man-
ner ensures the measured u to be an advancing
contact angle. These contact angles will be used
for the interpretation in terms of solid surface
tensions.
In Figure 1(b)–(e) and (g), the contact angles of
glycerol, formamide, 2,29-thiodiethanol, 3-pyri-
dylcarbinol, and 1-iodonaphthalene increase ini-
tially at essentially constant R. This is due to the
fact that even carefully putting an initial drop
from above the surface can result in contact an-
gles somewhere between advancing and receding.
Thus, after reaching the proper advancing angles,
the drop front starts to advance with essentially
constant u. It should be noted that the liquid–
vapor surface tension values calculated by ADSA-
P sessile drop are fairly constant, but not as reli-
able as those from pendant drop, as explained
above. The accuracy relies very much on the axi-
symmetry of the drop profile. It can be seen in
Figure 1(f) that the three-phase line of 1,3-diiodo-
propane appears to stick on the surface at the
beginning of the experiment, but not once the
drop has started advancing. This suggests that
the liquid has a tendency to stick on the surface.
Three liquids, 1,1,2,2-tetrabromoethane, 1-bro-
monaphthalene, and dibenzylamine, were found
to dissolve the P(MMA/EMA, 30/70)-coated wafer
on contact, resulting in irregular and flat drops.
The remaining two liquids all show very complex
contact angle behavior, as given in Figure 2(a)
and (b). Figure 2(a) shows the contact angle re-
sults of diiodomethane, a very commonly used
liquid for contact angle measurements.16,17,56 – 61
It can be seen that initially the apparent drop
volume, as perceived by ADSA-P, increases lin-
early, and u increases from 40° to 80° at essen-
tially constant R. Suddenly, the drop front jumps
to a new location as more liquid is supplied into
the sessile drop. The resulting u decreases
sharply from 80° to 40°. As more liquid is supplied
into the sessile drop, the contact angle increases
again. Such slip/stick behavior could be due to
non-inertness of the surface. Phenomenologically,
an energy barrier for the drop front exists, result-
ing in sticking, which causes u to increase at
constant R. However, as more liquid is supplied
into the sessile drop, the drop front possesses
enough energy to overcome the energy barrier,
resulting in slipping, which causes u to decrease
suddenly. It should be noted that as the drop front
jumps from one location to the next, it is unlikely
that the drop is or will remain axisymmetric.
Such a non-axisymmetric drop will obviously not
meet the basic assumptions underlying ADSA-P,
causing possible errors (e.g., in the apparent sur-
face tension and drop volume). This can be seen
from the slight discontinuity of the apparent sur-
face tension and drop volume with time, as the
drop front sticks and slips. Obviously, the ob-
served angles in Figure 2(a) cannot all be the
Young contact angles; since glv, gsv (and gsl) are
constants, u ought to be a constant because of
Young’s equation. In addition, it is difficult to
decide unambiguously at this moment whether
or not Young’s equation is applicable at all.
Therefore, these contact angles should not be
used for the interpretation in terms of surface
energetics.
A different contact angle pattern can be seen in
Figure 2(b) for dimethyl sulfoxide (DMSO) where
g lv and u are not constant. It was observed during
the experiment that DMSO indeed penetrated
into the polymer, causing the polymer film to be
removed from the wafer surface. These angles
cannot be used to determine solid surface tension,
due to physical reaction. Since the observation
 DYNAMIC CONTACT ANGLES AND SOLID SURFACE TENSIONS 2045

Figure 2. Low-rate dynamic contact angles on a P(MMA/EMA, 30/70)-coated wafer

surface for (a) diiodomethane; and (b) dimethyl sulfoxide (DMSO). These angles cannot
be used for the interpretation in terms of surface energetics, see text.

forces us to discard the measurement, it is not
necessary to verify our stipulation of dissolu-
tion of the polymer by the liquid. Even if poly-
mer dissolution would not be the cause of the
observed pattern, the measurement would have
to be discarded. Obviously, these angles should
be disregarded for the following reasons: (1) glv
is different from that of the pure liquids; (2)
we are unsure whether or not gsl and gsv remain
constant and whether Young’s equation is ap-
plicable.
The reproducibility of all solid–liquid sys-
tems is very good. They are summarized in Ta-
ble II for the seven liquids with usable contact
angles, at different rates of advancing and each
on a newly prepared surface. It should be noted
that a total of more than 60 freshly prepared
P(MMA/EMA, 30/70)-coated wafers were pre-
pared and used; more than 4000 images were
acquired and analyzed by ADSA-P. In the spe-
cific case of water in Table II, a final value of u
5 79.91° 6 0.38° was obtained, by averaging the
contact angles from different rates of advancing
(for different experiments). The 95% confidence
limits calculated in this manner (in Table II)
include all possible errors, due to experimental
technique, solid surface preparation, etc. A
summary of the contact angle complexities is
given in Table III, together with the meaningful
results from Table II.
Disregarding the inconclusive contact angle
data in Figure 2, we show in Figure 3 the contact
angle results from Table II, by plotting g lv cos u
vs. g lv , together with the error limits calculated
from the contact angle errors. It can be seen that
all liquids independent of molecular properties
fall on a smooth curve, in agreement with the
patterns obtained in previous studies40 –50,62– 65
for other solid surfaces: the values of g lv cos u
2046 KWOK, LI, AND NEUMANN

change smoothly with g lv , so that we again con-

Table II. Summary of the Advancing Contact Angles (deg.) at Different Rates (mm/min) of Motion of the Three-Phase Contact Line for Liquids Which

37.08 6 0.97a
1-Iodonaphthalene

36.76 6 0.31
37.56 6 0.14
36.32 6 0.14
36.55 6 0.16
38.19 6 0.17
clude that

u
g lv cos u 5 F~ g lv, g sv! (3)
Yielded Constant Contact Angles on a Poly(methyl methacrylate/ethyl methacrylate, 30/70) P(MMA/EMA, 30/70)-Coated Silicon Wafer Surface
and hence, because of Young’s equation

0.748
0.867
0.889
0.951
1.158
Rate g sl 5 g sv 2 F~ g lv, g sv! or g sl 5 f~ g lv, g sv! (4)

44.42 6 1.06a
43.74 6 0.20
44.61 6 0.16
45.60 6 0.35
44.72 6 0.24
43.45 6 0.19
1,3-Diiodopropane

where f is as yet an unknown function. Thus, the

u

surface tension component approaches11,15–17

clash directly with these experimental results:
the surface tension component approaches11,15–17
stipulate that g sl depends not only on g lv and g sv ,
0.446
0.629
0.853
0.867
0.991
Rate

but also on various surface tension components

manifested by the specific intermolecular forces of
47.86 6 0.46a
48.40 6 0.09
47.72 6 0.12
48.07 6 0.15
47.54 6 0.29
47.57 6 0.13
3-Pyridylcarbinol

the liquids and solids. But the above experimen-

tal results allow one to search for a relation in the
u

form of eq. (4). A more complete testing of the

surface tension component approaches is avail-
able.64,66 –70
0.489
0.522
0.590
0.607
0.610
Rate

On phenomenological grounds, an equation of

state approach for solid–liquid interfacial ten-
sions has been formulated14:
55.86 6 0.65a
55.56 6 0.07
55.28 6 0.08
56.63 6 0.07
55.73 6 0.12
56.08 6 0.14
2,29-Thiodiethanol

g sl 5 g lv 1 g sv 2 2 Îg lvg sve 2b~glv2gsv!

u

2
(5)

where b is a constant which was found to be

0.0001247 (m2/mJ)2. Combining this equation
0.491
0.499
0.675
0.700
0.754
Rate

with Young’s equation yields

Î
64.39 6 0.52a
64.69 6 0.07
64.87 6 0.16
64.23 6 0.09
64.34 6 0.04
63.80 6 0.17

g sv 2b~glv2gsv!2
cos u Y 5 21 1 2
Formamide

e (6)
g lv
u

Thus, the solid surface tensions can be deter-

mined from experimental (Young) contact angles
0.448
0.501
0.523
0.845
0.866
Rate

and liquid surface tensions.

Mean u value with the 95% confidence limits.

The applicability of any approach having the

72.52 6 0.38a
72.07 6 0.06
72.37 6 0.08
72.72 6 0.06
72.86 6 0.16
72.56 6 0.09

form of eq. (4) can be tested using the criteria of

the constancy of the calculated g sv values. Rela-
Glycerol

tions of the form of eq. (4) have been in the liter-

ature for a long time. Two examples are An-
tonow’s rule71
0.449
0.456
0.461
0.764
0.807
Rate

g sl 5 u g lv 2 g svu (7)
79.91 6 0.38a
79.82 6 0.15
80.15 6 0.13
80.21 6 0.10
79.45 6 0.05
79.93 6 0.09

and Berthelot’s geometric mean relationship72

u
Water

g sl 5 g lv 1 g sv 2 2 Îg lvg sv (8)
0.240
0.348
0.462
0.491
0.511
Rate

Combining these or similar relationships with

a

Young’s equation yields a relation of the form of

 DYNAMIC CONTACT ANGLES AND SOLID SURFACE TENSIONS
Table III. Summary of Contact Angle Results for a Poly(methyl methacrylate/ethyl methacrylate, 30/70)
P(MMA/EMA, 30/70)-Coated Silicon Wafer Surface
Liquid g lv (mJ/m2)
Dibenzylamine
Dimethyl sulfoxide (DMSO)
1-Iodonaphthalene
1-Bromonaphthalene
1,3-Diiodopropane
3-Pyridylcarbinol
1,1,2,2-Tetrabromoethane
Diiodomethane
2,29-Thiodiethanol
Formamide
Glycerol
Water
a
Part of the polymer was observed to be removed after the experiment.
eq. (3), from which gsv can be calculated. We
show in Table IV the gsv values calculated from
Antonow’s rule,71 Berthelot’s rule,72 and the
equation-of-state approach for solid–liquid in-
terfacial tensions,14 [i.e., eq. (6)], which can be
understood as a modified Berthelot rule.73 It
can be seen that the values of gsv calculated
from Antonow’s rule tend to increase from 39.0
mJ/m2 to 43.0 mJ/m2 as glv increases; the gsv
values calculated from Berthelot’s rule decrease
from 35.0 mJ/m2 to 25.0 mJ/m2 as glv increases.
These variations are more pronounced49,51,66 for
other low-energy polymers (e.g., fluorocarbons).
Figure 3. The values of g lv cos u vs. g lv for the
P(MMA/EMA, 30/70)-coated wafer surface, for the data
in Table II. Since the values of g lv cos u change
smoothly with g lv at constant g sv , g sl can be expressed
as a function of only g lv and g sv because of Young’s
equation.
2047
u (deg.)
40.80 dissolved the polymer on contact
42.68 u 1 and 2 as R 1 (46° 3 42° 3 46°)a
42.92 37.08 6 0.97
44.31 dissolved the polymer on contact
46.51 44.42 6 1.06
47.81 47.86 6 0.46
49.29 dissolved the polymer on contact
49.98 slip/stick (40° 3 80°)
53.77 55.86 6 0.65
59.08 64.39 6 0.52
65.02 72.52 6 0.38
72.70 79.91 6 0.38
The gsv values from the equation-of-state ap-
proach for solid–liquid interfacial tensions are
quite constant (within 6 0.5 mJ/m2), essentially
independent of the liquids used: the gsv value of
P(MMA/EMA, 30/70) was found to be 35.1
mJ/m2 with a 95% confidence limit of 6 0.3
mJ/m2.
It should be noted that the constant b value of
0.0001247 (m2/mJ)2 used in the above calcula-
tions had been obtained only from contact angle
data on three well-prepared solid surfaces63: FC-
721-coated mica, heat-pressed Teflon (FEP), and
poly(ethylene terephthalate) (PET). Alterna-
tively, the g sv value of P(MMA/EMA, 30/70) can
be determined by a two-variable least-squares
analysis,14 by assuming g sv and b in eq. (6) to be
constant. In this procedure, the computer will
search for that pair of g sv and b values which
provides the best fit of eq. (6) to the seven pairs of
experimental ( g lv , u Y ) data points. This leads to a
b value of 0.0001230 (m2/mJ)2 and a g sv value of
35.1 mJ/m2. It is evident that there is good agree-
ment between the g sv values (35.1 6 0.3 mJ/m2
and 35.1 mJ/m2) determined from the two strat-
egies.
Obviously, because of eq. (6), the contact an-
gles for any liquid can be predicted from the gsv
value obtained from a single point of (glv, uY).
This is illustrated in Table V; from the experi-
mental values of glv and u for 1-iodonaphtha-
lene, eq. (6) yields gsv 5 35.2 mJ/m2. With this
value of gsv, the contact angles for the remain-
ing six liquids are calculated from eq. (6) and
listed in Table V. The predicted contact angles
2048 KWOK, LI, AND NEUMANN

Table IV. The Solid–Vapor Surface Tension Values, g sv , of a Poly(methyl methacrylate/ethyl methacrylate, 30/
70) P(MMA/EMA, 30/70)-Coated Silicon Wafer Surface Calculated from Antonow’s Rule, Berthelot’s Rule, and the
Equation-of-State Approach for Solid–Liquid Interfacial Tensions

g sv (mJ/m2)

Liquid g lv (mJ/m2) u (deg.) Antonow’s Rule Berthelot’s Rule Equation-of-State

1-Iodonaphthalene 42.92 37.08 6 0.97 38.6 34.7 35.2

1,3-Diiodopropane 46.51 44.42 6 1.06 39.9 34.2 35.3
3-Pyridylcarbinol 47.81 47.86 6 0.46 39.9 33.4 34.8
2,29-Thiodiethanol 53.77 55.86 6 0.65 42.0 32.8 35.6
Formamide 59.08 64.39 6 0.52 42.3 30.3 35.0
Glycerol 65.02 72.52 6 0.38 42.3 27.5 34.6
Water 72.70 79.91 6 0.38 42.7 25.1 35.5

agree with the experimentally observed ones to
within approximately 6 2° (i.e., the accuracy/
reproducibility normally associated with con-
tact angle measurements). It is apparent that
surface tensions determine the contact angles
completely; the specific intermolecular forces
which give rise to the surface tensions do not
affect the contact angles directly: intermolecu-
lar forces determine the surface tensions; the
surface tensions then determine the contact an-
gle through Young’s equation. The difference
between the predicted and measured angles are
indeed very small and could not even be de-
tected if a conventional contact angle technique
were used.
The above results indeed reconfirm the validity
of the equation of state approach14 to determine
solid surface tensions from contact angles.
The fact that g lv cos u changes smoothly with
g lv for the P(MMA/EMA, 30/70) copolymer should
be confronted with the results obtained in previ-
ous studies45,50 for two other methacrylate poly-
mers: poly(methyl methacrylate) PMMA,45 and
poly(ethyl methacrylate) PEMA.50 We show in
Figure 4 these contact angle results, together
with the results for the P(MMA/EMA, 30/70) poly-
mer. The g sv values of PMMA45 and PEMA50 were
found from the equation-of-state approach to be,
respectively, 38.5 6 0.5 mJ/m2, and 33.6 6 0.5
mJ/m2, so that the value (35.1 6 0.3 mJ/m2) for
P(MMA/EMA, 30/70) is intermediate and closer to
that for PEMA, as expected.
As Figure 4 stands, there might be two pos-
sible concerns: the first one is whether such
data could simply be represented by straight
lines; the next one is whether the curves are
indeed smooth.
With respect to the first point, statistical
tools provide a definitive answer. We have per-
formed a least-squares analysis to fit these ex-
perimental contact angle data separately with
linear and, for simplicity, quadratic equations.
The results are given in Table VI. To illustrate
the generality of such patterns, contact angle
data on other methacrylate polymers46,47 and a
FC-722-coated mica surface40 are also used. It
can be seen in Table VI that, in all cases, the
regression coefficients for quadratic equations
are better than those of the linear equations
and that curvature is indeed present in the

Table V. The Predicted Contact Angles for the Liquids in Table IV, Based Solely on the Contact Angle and
Surface Tension of 1-Iodonaphthalene ( g sv 5 35.2 mJ/m2) Using eq. (6)

Liquid g lv (mJ/m2) Measured u (deg.) Predicted u (deg.)

1,3-Diiodopropane 46.51 44.42 6 1.06 44.6

3-Pyridylcarbinol 47.81 47.86 6 0.46 47.0
2,29-Thiodiethanol 53.77 55.86 6 0.65 56.6
Formamide 59.08 64.39 6 0.52 64.0
Glycerol 65.02 72.52 6 0.38 71.5
Water 72.70 79.91 6 0.38 80.3
 DYNAMIC CONTACT ANGLES AND SOLID SURFACE TENSIONS
Figure 4. The values of g lv cos u vs. g lv for the
PMMA-, P(MMA/EMA, 30/70)- and PEMA-coated wa-
fer surfaces. Again, g lv cos u changes smoothly with g lv
at constant g sv .
plots of these data (i.e., even a simple quadratic
equation represents the data better than a lin-
ear equation); use of eq. (6) would be expected to
give a better fit yet.
With respect to the second point, it does not
appear to us that there is any systematic
change in the contact angles (in Fig. 4). For
example, in the case of formamide with glv
about 59.0 mJ/m2, the data point on the PMMA
curve appears to be slightly higher, while the
point on the P(MMA/EMA, 30/70) is slightly
lower and the one for PEMA slightly higher
again. There is no evidence for any systematic
variation for this and other liquids. Clearly, if
there were a deviation due to specific intermo-
lecular forces, one would, at the most, expect a
monotonous change of the deviation when going
from PMMA to P(MMA/EMA, 30/70), and to
PEMA. To keep matters in perspective, it has to
Table VI. A Summary of the Regression Results Using Linear and Quadratic Equations (see text)
Surface
Poly(n-butyl methacrylate) PnBMA46
Poly(ethyl methacrylate) PEMA50
Poly(methyl methacrylate/n-butyl methacrylate) P(MMA/nBMA)47
Poly(methyl methacrylate/ethyl methacrylate, 30/70)
P(MMA/EMA, 30/70) [this work]
Poly(methyl methacrylate) PMMA45
FC-722-coated mica40
2049
be realized that the curves in Figure 4 would
have to be considered smooth if conventional
goniometric techniques with 6 2° contact angle
accuracy were used. Further, the fact that an
equilibrium spreading pressure of as low as 1
mJ/m2 would easily contribute to such a varia-
tion should not be overlooked; even very minor
swelling of the polymer or creeping of the liquid
could easily introduce a slight deviation of some
points from such curves. Therefore, only after
considering such effects would the need arise
for explanations in terms of direct effects of
intermolecular forces.74
CONCLUSIONS
1. The values of g lv cos u change smoothly
with g lv , in good agreement with those
from other polar and nonpolar solid sur-
faces.40 –50,62– 65
2. The g sv values of poly(methyl methacry-
late/ethyl methacrylate, 30/70) P(MMA/
EMA, 30/70) copolymer calculated from the
equation of state approach14 are quite con-
stant, essentially independent of the liq-
uids used; the average value is g sv 5 35.1
6 0.3 mJ/m2. This reconfirms the sound-
ness of the approach to calculate solid sur-
face tensions from contact angles.
This research was supported by the Natural Science
and Engineering Research Council of Canada (NSERC)
under grants No. A8278 and No. EQP173469, an On-
tario Graduate Scholarship (D.Y.K.), and a NSERC
Postdoctoral Fellowship (D.Y.K.). This article repre-
sents, in part, the Ph.D. thesis of D. Y. Kwok at the
University of Toronto.
R 2 (regression coefficient)
Linear Equation Quadratic Equation
0.968 0.994
0.983 0.999
0.985 0.994
0.985 0.995
0.976 0.990
0.963 0.997
2050 KWOK, LI, AND NEUMANN
REFERENCES AND NOTES
1. Derjaguin, B. J.; Muller, V. M.; Toporov, Y. P. J
Colloid Interface Sci 1980, 73, 293.
2. Johnson, K. L.; Kendall, K.; Roberts, A. D. Proc Roy
Soc (London) 1971, A324, 301.
3. Muller, V. M.; Yushchenko, V. S.; Derjaguin, B. V.
J Colloid Interface Sci 1983, 92, 92.
4. Fogden, A.; White, L. R. J Colloid Interface Sci
1990, 138, 414.
5. Pashley, R. M.; McGuiggan, P. M.; Horn, R. G.;
Ninham, B. W. J Colloid Interface Sci 1988, 126,
569.
6. Christenson, H. K. J Phys Chem 1986, 90, 4.
7. Claesson, P. M.; Blom, C. E.; Horn, P. C.; Ninham,
B. W. J Colloid Interface Sci 1986, 114, 234.
8. Pashley, R. M.; McGuiggan, P. M.; Pashley, R. M.
Colloids Surf 1987, 27, 277.
9. Pashley, R. M.; McGuiggan, P. M.; Ninham, B. W.;
Evans, D. F. Science 1985, 229, 1088.
10. Zisman, W. A. Contact Angle, Wettability and Ad-
hesion, Advances in Chemistry Series; American
Chemical Society: Washington, D.C., 1964; Volume
43.
11. Fowkes, F. M. Ind Eng Chem 1964, 12, 40.
12. Driedger, O.; Neumann, A. W.; Sell, P. J. Kolloid-Z
Z Polym 1965, 201, 52.
13. Neumann, A. W.; Good, R. J.; Hope, C. J.; Sejpal,
M. J Colloid Interface Sci 1974, 49, 291.
14. Spelt, J. K.; Li, D. in A. W. Neumann, J. K. Spelt,
Eds., Applied Surface Thermodynamics; Marcel
Dekker: New York, 1996; pp. 239 –292.
15. Owens, D. K.; Wendt, R. C. J Appl Polym Sci 1969,
13, 1741.
16. van Oss, C. J.; Chaudhury, M. K.; Good, R. J. Chem
Rev 1988, 88, 927.
17. Good, R. J.; van Oss, C. J. in M. Schrader, G. Loeb,
Eds., Modern Approaches to Wettability: Theory
and Applications; Plenum Press: New York, 1992;
pp. 1–27.
18. Bruil, H. G. Colloid Polym Sci 1974, 252, 32.
19. Cheever, G. D. J Coat Technol 1983, 55, 53.
20. Kilau, H. W. Colloids Surf 1983, 26, 217.
21. Grundke, K.; Bogumil, T.; Gietzelt, T.; Jacobasch,
H.-J.; Kwok, D. Y.; Neumann, A. W. Progr Colloid
Polym Sci 1996, 101, 58.
22. Vargha-Butler, E. I.; Zubovits, T. K.; Absolom,
D. R.; Neumann, A. W. J Dispersion Sci Technol
1985, 6(3), 357.
23. Vargha-Butler, E. I.; Moy, E.; Neumann, A. W.
Colloids Surf 1987, 24, 315.
24. Vargha-Butler, E. I.; Zubovits, T. K.; Absolom,
D. R.; Neumann, A. W. Chem Eng Commun 1985,
33, 255.
25. Li, D.; Neumann, A. W. in A. W. Neumann, J. K.
Spelt, Eds., Applied Surface Thermodynamics;
Marcel Dekker: New York, 1996; pp. 509 –556.
26. Omenyi, S. N.; Neumann, A. W. J Appl Phys 1976,
47, 3956.
27. Li, D.; Neumann, A. W. in A. W. Neumann, J. K.
Spelt, Eds., Applied Surface Thermodynamics;
Marcel Dekker: New York, 1996; pp. 557– 628.
28. Moy, E.; Neumann, A. W. in A. W. Neumann, J. K.
Spelt, Eds., Applied Surface Thermodynamics;
Marcel Dekker: New York, 1996; pp. 333–378.
29. Neumann, A. W. Adv Colloid Interface Sci 1974, 4,
105.
30. Li, D.; Neumann, A. W. in A. W. Neumann, J. K.
Spelt, Eds., Applied Surface Thermodynamics;
Marcel Dekker: New York, 1996; pp. 109 –168.
31. Johnson, R. E., Jr.; Dettre, R. H. 1972, 43, 341.
32. Eick, J. D.; Good, R. J.; Neumann, A. W. J Colloid
Interface Sci 1975, 53, 235.
33. Nadkami, G. D.; Garoff, S. Langmuir 1994, 10,
1618.
34. Sedev, R. V.; Petrov, J. G.; Neumann, A. W. J
Colloid Interface Sci 1996, 180, 36.
35. Wenzel, R. N. Ind Eng Chem 1936, 28, 998.
36. Cassie, A. B. D. Disc Faraday Soc 1948, 3, 11.
37. Baxter, S.; Cassie, A. B. D. J Text Inst 1945, 36(T),
67.
38. Kwok, D. Y.; Li, D.; Neumann, A. W. in A. W.
Neumann, J. K. Spelt, Eds., Applied Surface Ther-
modynamics; Marcel Dekker: New York, 1996; pp.
413– 440.
39. Kwok, D. Y.; Budziak, C. J.; Neumann, A. W. J
Colloid Interface Sci 1995, 173, 143.
40. Kwok, D. Y.; Lin, R.; Mui, M.; Neumann, A. W.
Colloids Surf A 1996, 116, 63.
41. Kwok, D. Y.; Gietzelt, T.; Grundke, K.; Jacobasch,
H.-J.; Neumann, A. W. Langmuir 1997, 13, 2880.
42. Kwok, D. Y.; Lam, C. N. C.; Li, A.; Leung, A.;
Neumann, A. W. Langmuir 1998, 14, 2221.
43. Kwok, D. Y.; Lam, C. N. C.; Li, A.; Leung, A.; Wu,
R.; Mok, E.; Neumann, A. W. Colloids Surf A 1998,
142, 219.
44. del Rı́o, O. I.; Kwok, D. Y.; Wu, R.; Alvarez, J. M.;
Neumann, A. W. Colloids Surf A 1998, 143, 197.
45. Kwok, D. Y.; Leung, A.; Lam, C. N. C.; Li, A.; Wu,
R.; Neumann, A. W. J Colloid Interface Sci 1998,
206, 44.
46. Kwok, D. Y.; Leung, A.; Li, A.; Lam, C. N. C.; Wu, R.;
Neumann, A. W. Colloid Polym Sci 1998, 276, 459.
47. Kwok, D. Y.; Lam, C. N. C.; Li, A.; Neumann, A. W.
J Adhes 1998, 68, 229.
48. Kwok, D. Y.; Lam, C. N. C.; Li, A.; Zhu, K.; Wu, R.;
Neumann, A. W. Polym Eng Sci 1998, 38, 1675.
49. Kwok, D. Y.; Neumann, A. W. Prog Colloid Polym
Sci 1998, 109, 170.
50. Kwok, D. Y.; Wu, R.; Li, A.; Neumann, A. W. J
Adhes Sci Technol (submitted).
51. Kwok, D. Y. Contact Angles and Surface Energet-
ics, PhD thesis, University of Toronto, 1998.
52. Rotenberg, Y.; Boruvka, L.; Neumann, A. W. J Col-
loid Interface Sci 1983, 93, 169.
53. Cheng, P.; Li, D.; Boruvka, L.; Rotenberg, Y.; Neu-
mann, A. W. Colloids Surf 1983, 93, 169.
 DYNAMIC CONTACT ANGLES AND SOLID SURFACE TENSIONS
54. Lahooti, S.; del Rı́o, O. I.; Cheng, P.; Neumann,
A. W. in A. W. Neumann, J. K. Spelt, Eds., Applied
Surface Thermodynamics; Marcel Dekker: New
York, 1996; pp. 441–507.
55. del Rı́o, O. I.; Neumann, A. W. J Colloid Interface
Sci 1997, 196, 136.
56. van Oss, C. J.; Ju, L.; Chaudhury, M. K.; Good,
R. J. J Colloid Interface Sci 1989, 128, 313.
57. van Oss, C. J.; Good, R. J. J Macromol Sci-Chem
1989, A26(8), 1183.
58. van Oss, C. J.; Giese, R. F., Jr.; Good, R. J. Lang-
muir 1990, 6, 1711.
59. Chibowski, E.; Kerkeb, M. L.; González-Caballero,
F. J Colloid Interface Sci 1993, 155, 444.
60. Wu, W.; Giese, R. F., Jr.; van Oss, C. J. Langmuir
1995, 11, 379.
61. Lee, L. H. Langmuir 1996, 12, 1681.
62. Li, D.; Neumann, A. W. J Colloid Interface Sci
1992, 148, 190.
63. Li, D.; Xie, M.; Neumann, A. W. Colloid Polym Sci
1993, 271, 573.
2051
64. Kwok, D. Y.; Li, D.; Neumann, A. W. Colloids Surf
A 1994, 89, 181.
65. Kwok, D. Y.; Neumann, A. W. Adv Colloid Interface
Sci (in press).
66. Kwok, D. Y.; Li, D.; Neumann, A. W. Langmuir
1994, 10, 1323.
67. Kwok, D. Y.; Neumann, A. W. Can J Chem Eng
1996, 74, 551.
68. Kwok, D. Y.; Lee, Y.; Neumann, A. W. Langmuir
1998, 14, 2584.
69. Kwok, D. Y. Colloids Surf A (in press).
70. Spelt, J. K.; Moy, E.; Kwok, D. Y.; Neumann, A. W.
in A. W. Neumann, J. K. Spelt, Eds., Applied Sur-
face Thermodynamics; Marcel Dekker: New York,
1996; pp. 293–332.
71. Antonow, G. J Chim Phys 1907, 5, 372.
72. Berthelot, D. Comp Rend 1898, 126, 1703, 1857.
73. Li, D.; Neumann, A. W. J Colloid Interface Sci
1990, 137, 304.
74. van Giessen, A. E.; Bukman, D. J.; Widom, B. J
Colloid Interface Sci 1997, 192, 257.