1

The Measurement of surface energy of polymer by means

of contact angles of liquids on solid surfaces

A short overview of frequently used methods

by

Finn Knut Hansen

Department of Chemistry
University of Oslo
Definitions

Surface tension and surface energy

L F

dx

Surface tension form force:

The force, F, involved in stretching a film is: F = L = surface tension (constant)
This means: = F/L i.e. force/unit length Units: N/m or mN/m (= dyn/cm in c.g.s units)
Surface energy from work:
The work, dW, involved in increasing the surface by a length dx is: dW = dG = L dx = dA
2
This means: = dG/dA i.e. free energy/unit area Units: J/m = N/m

Surface tension and surface energy are interchangeable definitions with the same units

Work of adhesion and work of cohesion

Work of adhesion The work of adhesion between 2 (incompatible)
substances is:
1 a
W = W12 = 1 + 2 - 12
12 a
1 W12 2 or: 12 = 1 + 2 - W
2

Work of cohesion The work of cohesion of a single substance is:

1 c
W = W11 = 1 + 1 - 0 = 2 1
12 c c a
W11 I.e. = 0.5 (W + W ) - W

1 1 2
Youngs equation
1 When the liquid does not spread, a drop has a
contact angle on the surface. The balance
between the forces on the surface gives:
Gas(air) Subst.1
Youngs Equation: 2 = 12 + 1 cos
2 NB: Only valid in dry wetting
12
Subst.2

In
In cases with wet
these cases, wetting,lowered
2 becomes the surface pressure
by the surfaceofvapor
the liquid vapor
pressure .on the solid is substantial.
So that: 2 = + 12 + 1 cos

Expressed by the work of adhesion we can write:

a
W = 1 + 2 - 12 = 1 + 1 cos = 1(1+cos ) This is the Young - Dupree
equation

Critical surface tension - Zisman plot

1 c
Zisman et al. (1950) found an empirical
connection between cos and 1: 1.0-
0.8-
Polymer: PTFE
If we measure
liquids on the the contact
same angle
surface, and of many
plot 0.6-
cos

cos against 1 , we get

be extrapolated to cos =1. a curve that can
0.4-
The extrapolated value is called the
critical surface tension of the solid 0.2-
surface.
0-
Note: This is
the surface not necessarily
tension thesame
of the solid, as
2.
-0.2-

-0.4-

0 10 20 30 40 50 60 70 80
1 (mN/m)

There are 2 problems: 1. The line is not really straight (it is more hyperbolic)
2. c is not the same as 2 (only if 12= 0 when = 0)

In DROPimage the Zisman Plot tool performs an ordinary linear Zisman plot.
The interaction parameter and the work of adhesion
2
The work of adhesion has been expressed by Good and Girifalco (1960) by the geometric
mean of the surface tensions in the same way as the Hamaker constant:
a 1/
W = 2 ( 1
2 ) where is the Interaction parameter, 0.5 < < 1.15
2

is a function of the molar volumes of substance 1 and 2:

=
41r 2r A12 where r1, r2 = molecular radii

A + A
( r1 + r2 )2 ( 11 22 )1/ 2
and A is the sum of London constants (or corresponding) for all types of intermolecular
forces (dispersive, polar, acid-bas, etc.)
a
If we insert for W in the Young-Dupree equation, we get:
2
a
W = 2 ( )1/ 2 = (1 + cos ) = (1 + cos )=
or
1 2 1 2 1 2
4

has been calculated theoretically, but the results have often been misleading.

It is possible to calculate empirical values for by using values of measured

by liquid/liquid interactions in systems of similar polarity.

3
In a recent publication,
expression, Kwokofand
i.e. and 2equation Neumann
state (K&N)
for . Their have argued
expression is: for using an analytical
= exp ( )
1 2

It is easily seen that if 1 = 2 then = 1. The magnitude of is therefore crucial in giving a

universally correct work of adhesion, if such an expression is possible. K&N have determined
this experimentally from an extensive amount
2 2of measurements of low energy polymer
surfaces.
it They found = 0.0001247 (m /mJ) to give the best all-over results, although
varied some.

This method has been implemented in DROPimages Surface Energy (One Liquid) tool.
 Fowkes theory

Fowkes theory is based on 2 fundamental assumptions: Additivity and the geometric mean

d p h i ab
1. Surface forces (energies) are additive: = + + + +

+ ... where d = dispersion force

p = polar force
h = hydrogen bonding force
i = induction force (Debye)
ab = acid/base force
etc.

2. Geometric
d
mean is used for the work of adhesion for each type of force (energy):
p pd d 1/ 2
p
(
W 12 = 2 1 1/ 2
2
, W 12 = 2 1 2
) ) (
, W h12 = 2 1h h2 1/ 2 , etc.
) (
12 12 1 d 2 p 2
The general expression for W is: W = (1 + cos ) = W
1 +W
1 + ...
d d p p
And for 12 thus: 12 = 1 + 2 2 1 ( 1/ 2
2 1 2 ( 1/ 2
...

2 ) )
By using a liquid that only interacts with the surface by dispersion forces, we can write:
1 d
(1 = 1 ) 2d (1 + cos
W12 = 1 (1 + cos ) = 2
d
1/ 2 and ) )=
2
= 1
4
2 (

Extended Fowkes theory

The combination of additivity and geometric mean has been used by many:
1/ 2 1/ 2 1/ 2
(
W12 = 2 d1 d 2 )
+ 2 p1
p
2 ( ) (
+ h1 h )
(Kitazaki and Hate, 1972)4
2 2

( )
1/ 2
d d

( )
1/ 2
W12 = 2 1 + 2 n1 n (n = all non-dispersive componenets)
2 2

By doing measurements with m number of liquids on the same surface, we can calculate m
different components of the surface energy, if the corresponding components of the liquids
are known.

By using 2 liquids, A and B, we can write

 1/ 2 1/ 2
W12A = ( 1 + cos ) (
= 2 1Ad d ) p
+ 2 1A p
2 ( )
1A A 2
1/ 2 1/ 2
W12B = ( 1 + cos ) (
= 2 1Bd d ) p
+ 2 1B p
2 ( )
1B B 2

The 2 equations can be linearized to give:

1/ 2
( 1Ad ) ( 1A ) 1/ 2
1/ 2
p 1 cos A
1/ 2 = +

1A
(2 )
d
+
1A
( 2)
p
2

1/ 2
( 1Bd ) ( 1B ) 1/ 2
1/ 2
p 1 cos B
1/ 2 = +

1B
( 2 ) + 1B ( )
d p
2
2

d 1/2 p 1/2
These equation set is solved for (2 ) and (2 ) .

Usually, one polar (water) and one unpolar (methylene iodide) liquid are used. This is the so-
called two-liquid method. In the solution, care must be taken to check if a square root is
negative. This indicates errors in the measurements. This procedure is used in DROPimages
5
Surface Energy (Two Liquids) tool (Geometric Mean) .

The question of geometric or harmonic mean

Fowkes assumption of the geometric mean is based on Berthelots hypothesis, again based
on Londons theory of dispersive forces.

The London (Lennard-Jones) attraction constants between like and dislike substances are
d = 3 1 2 d 3 2 d =3 2
A h A = h A h
, 12 2 1 2 , 11 1 1 22 2 2
1 + 2 4 4

where h = Plancks constant, = frequency of vibration, = polarizability

If is eliminated, we obtain
1/ 2 d 1/ 2 d d 1/ 2
d = 2(1 2 ) d ) , and if = , then = (A d A )
A (A A
A
12 11 1 2 12 11 22
1 + 2
22

After Girifalco and Good, the works of adhesion are expressed by

2 2
n n A 1 1 n A 1 1 n A
1 1
W = 1 2 12 , W = 1 11 , W = 2 22
12 2 11 2 22 2
6d12 2 m 4 6d11 2 m 4 6d 22 2 m 4
where n = molecular density, d = equilibrium distance between the phases, m = the repulsion
constant (Lennard-Jones). By eliminating n, and using the expressions for Aij above, we
obtain:
d d d d d 1/ 2
W = 1 2 (W W ) (There are only dispersion energies here)
12 2 11 22
d12

12 11 22 12d d d 1/ 2
If then d = (d d )
1/2
we obtain: W = W22 ) and therefore d
= 2( d d )1/ 2
(W 11W 12 1 2

These are therefore the assumption involved in using the geometric mean.

It is, however, possible that some of these assumptions are less acceptable. It is possible to
6
derive another mean, based on slightly different assumptions :

If we eliminate instead of from the equations above, we obtain

d 2A11d A 22 d
A12 =
A11d (1 / 2 ) + d (1 / 2 )
A 22

d d
d 2A 11A 22
If now 1 = 2 then the expression for A12 is: A12 = , i.e. a harmonic mean
A11d + A 22d

By using the same equations for Wij as above, we now obtain

d d d d d
2W11 W22d 11 22 1 2 2
d n1 n 2 d12
W12 =
W11d (d11 / n1 )2 + W22 d (d 22 / n 2 )2

1/ 2
If now again d12 (d11d 22 ) and in addition n1 / d11 n 2 / d 22 , the work of adhesion
becomes

d 2 W11d W22d d d
W = , and therefore W d = 411 22
12 12
W11d + W22 d 1d + 22d
1

This is the harmonic mean for the work of adhesion. Wu has claimed that the harmonic mean
6
is better suited for low energy surfaces, such as polymers .

On the next page are shown two figures of the work of adhesion plotted as a function of the
dispersive and polar surface energy components. The plot using the harmonic mean is seen to
give generally lower values for the work of adhesion than the geometric mean.
 Harmonic me G e ome tric me an
an

90 90

80 80

70

Work of adhesion (mN /m)

70
Work of adhesion (mN /m)

60 60

50 50

40 40

30 30

20 20

10 10

0 35 0 35

2
2

6
6

10
5
10

14
5
14

18
18

22
22

26
26

30
30

Measuring surface energies with 2 liquids by the harmonic mean

In the same way as with using the geometric mean, the harmonic mean may be used to
calculate the dispersive and polar components of the surface energy by measuring the angles
of two liquids, A and B.

The two equations are then:

d d p p
4 4
W12A = 1A (1 + cos A ) = 1A 2 + 1A 2
1A d + d 1 p + 2p
2 A

d d p p
4 4 1B 2
W12B = 1B (1 + cos B ) = 1B 2 +
1B d + 2d 1 p + 2p
B

Here, the harmonic mean is used also for the polar components. The justification of this
assumption is not necessarily solid, but is necessary here to obtain an analytical solution. This
set of equations may be converted to two second degree equations as given below.
a
d
( )2 + b d d d d
+ c =0
a
p
( )2 + b
2 p
2
p2
2
p p
+ c =0

Where the constants are

d p
a = AB B A A A BB a = A B CA A AC B
d p
b = C BB A CA BB DA AB + DB AA and b = B BC A B AC B DA AB + DB AA
d p
c = D BC A DA CB c = D BB A DA BB
 A i = 1i (3 cos ) / 4
B =
p d
(1 + cos ) / 4

i 1i 1i 1i i
and for i = A and B, respectively
C =
d
(1 + cos ) / 4
p
i 1i 1i 1i i

D = ( (1 + cos ) / 4
d p

i 1i 1i 1i i

The roots of the second degree equations are found in the usual way, taking care to obtain the
correct, positive, roots. This procedure is used in DROPimages Surface Energy (Two
Liquids) tool (Harmonic Mean).

Combinations of means

Different combinations of geometric and harmonic means have also been proposed, for
6
instance the following combination :

p p
d d 1/ 2 4
W12 = 2( 1 2 ) + p 1 p2
1 + 2

Wu proposed that this equation should be used for high energy surfaces, such as mercury,
glass, oxides and graphite, while the harmonic mean also should be uses for the dispersive
components in the case of polymer surfaces.

Acid-base interactions

Also, even if these types of relationships fit for the dispersion forces, and possibly also for
some polar (Debye and Keesom forces), they are not so good for acid/base and hydrogen
7-9
bonding. For this, Van Oss and coworkers have proposed other combination of surface
energies.

The argument is that often, the polar (Keesom and Debye) forces are weak, and can be
included in the dispersive contribution. The combined contribution is denoted by LW
Lifschitz-van der Waals. In addition, there is a short-range interaction (SR), that is caused by
acid-base interactions (hydrogen bonding is a type of acid-base).
We can write: = LW + AB

LW
( )
1/ 2
W can be expressed by the geometric mean (after Fowkes): W LW
= 2 1 LW 2 LW
12

AB
However, W can not be expressed in this way, as the basic components of the surface only
interact with the acid components of the liquid, and vice versa.
( ) ( )
1/ 2 1/ 2
Therefore, Van Oss and coworkers write: W 12 AB = 2 1 + 2
+ 2 1 2 +
 + -
where i is the acidic part and i the basic part. Some substances that have only acidic or
only basic properties are classified as monopolar, while substances with both types of
properties are bipolar.

Calculation

By measuring the contact angle of three different liquids, A, B, and C, with known LW, acidic
and basic components, the corresponding surface energies of the solid can be calculated.
1/ 2 + 1/ 2 + 1/ 2
W12A = ( 1 + cos ) (
= 2 1ALW LW
) + 2 1A ( + 2 1A 2 ) ( )
1A A 2 2
1/ 2 + 1/ 2 + 1/ 2
W12B = ( 1 + cos ) (
= 2 1BLW LW
) + 2 1B( + 2 1B 2 ) ( )
1B B 2 2
1/ 2 + 1/ 2 + 1/ 2
W12C = ( 1 + cos ) (
= 2 1CLW LW
) + 2 1C( + 2 1C 2 ) ( )
1C C 2 2

These equations are solved in the same way as the 2-equaions set above. We may express the
set in matrix form:
Coefficient vector: a = [(
2
LW 1/
) ( )
2
1/
( )
+ 1/ 2
]
2 2 2

( 1ALW )1/ 2 ( 1A )
+ 1/ 2
( 1A )
1/ 2

1A 1A 1A
( LW )1/ 2 (
+ 1/ 2
) ( 1B)
1/ 2
The x-matix: = 1B 1B

1B 1 1B
B
LW 1/ 2 + 1/ 2 1/ 2
( 1C ) ( 1C ) ( 1C )
1C 1 1C
C

(1 + cos(A )) / 2
The y-vector: = (1 + cos(
B )) / 2

(1 + cos(C )) / 2

-1
The solution is then obtained from the solution: a = by the usual matrix inversion and
multiplication. This procedure is used by DROPimages Acid-Base tool.

Using more than 2 liquids in Fowkes method

In the modified Fokes theory only two liquids are used to calculate the dispersive and polar
components of the surface energy. If more than two liquids are used, all combination of liquid
pairs may be used to calculate the components, and then the average of these may be taken.
10
However, there is also another method that has been proposed in the literature by using a
regression line.
The extended Fowkes equation
1 (1 + cos ) = 1 2
d d
)1/ + 2 (1 p 2 p )1/ 2
2 ( 2

is modified by dividing by 2 1 ( d )1/ 2 and we obtain

1/ 2
1 1/ 2 1 p
( ( )
1/ 2
(1 + cos ) = 2d + 2p
) 2 1/ 2 d
( d1) 1
1/ 2
p
If we plot the left side of this equation as a function of 1 d we get a straight line with the

1

( p) ( d)
slope equal to 2 1/ 2 and intercept on the y-axis equal to 2 1/ 2 . With only two liquids the
result is the same as with the Surface Energy (Two Liquids) method, using the geometric
mean. But the advantage with the regression is that more liquids may be used, and that by this
method it also possible to do a realistic evaluation of the errors involved. This procedure is
implemented in the DROPimage Surface Energy (Multi Liquids) tool.

One should be careful not to accept negative slopes in this regression because the square
always comes out positive. A negative slope is clearly caused by errors in the contact angle
data or the surface energies of the liquids.

Measurements of high energy surfaces

As most liquids are spread on a high energy surface, the contact angle cannot be measured.
11
However, Shultz (1977) has developed a method where these angles may be measured by
submerging the surface in one liquid and using a second liquid to measure the contact angle.
Usually, hydrocarbons like n-hexane, n-octane, n-decane, and n-hexadecane are used as the
submerging liquids and water as the contact angle liquid.

We can show this by the following derivation (using the geometric mean)
d 1/ 2
For the hydrocarbon: SH = S + H S H )
2(

d d 1/ 2
For the water drop: SW = S + W S W ) ESW
2(

where S denotes solid, H hydrocarbon, and W water. The parameter ESW is the excess
interaction energy, i.e. the polar part. By subtracting the second from the first equation, we
obtain

SH SW = H W + d )1/ 2[( Wd 1/ 2 ( H )1/ 2 ] + ESW

2(S )
 From Youngs equation SH SW = WH cos , we obtain

WH cos = H W + d )1/ 2[( Wd 1/ 2 ( H )1/ 2 ] + ESW , or

2(S )

W H + WH cos = d )1/ 2[( Wd 1/ 2 ( H )1/ 2 ] + ESW

2(S )

( H ) we get a straight line with the slope

d 1/ 2
By plotting the left side against ( W )
1/ 2

d 1/ 2

(mN/m)
2(S and intercept ESW.
)
Using the geometric mean, ESW is W - H + WH cos() 110
expressed according to the ESW n-hexane
extended Fowkes theory as 105 n-octane
p p 1/ 2 S

ESW = 2(S W ) 100

Slope = 2( d)1/2
.

d
Knowing the polar part of the n-dodecane = 123 mN/m
S
surface tension of water, the solids 95 S p = 43 mN/m
polar component can be easily n-hexadecane
calculated and thus the total 90
-1 -0.5 0 0.5 1
surface energy of the solid. The H

figure to the left shows an example ( )

d 1/2
(
)
1/2 (mN/m)
1/2

of this type of plot. W

The described method is applied in the Solid-Liquid-Liquid (SLL) Surface Energy tool in
DROPimage.

References

1. W.A. Zisman, ACS Adv. Chem. Ser. 43, 1 (1964).

2. L.A. Girifalco, and R.J. Good, J. Phys. Chem. 61, 904 (1957).
3. D.Y. Kwok, and A.W. Neumann, Adv. Colloid Interface Sci., 81 (1999), 167-249.
4. Y. Kitazaki, and T. Hate, J. Adhesion 4, 123 (1972).
5. D.K. Owens and R.C. Wendt, J. Appl. Polym. Sci. 13, 1741 (1969).
6. S. Wu, J. Polym. Sci C 34, 19 (1971).
7. C.J. van Oss, R.J. Good and M.K. Chaudhury; Adv. Colloid Interface Sci. 28, 35
(1987).
8. C.J. van Oss, R.J. Good and M.K. Chaudhury, J.; Chromatography 191, 53 (1987).
9. C.J. van Oss, R.J. Good and M.K. Chaudhury, J.; Langmuir 4, 884 (1988).
10. W. Rabel, Farbe + Lack, 77(10), 997-1006 (1971).
11. J. Schultz, K. Tsutsumi and J.-B. Donnet, J.Colloid Interface Sci 59, 272 and 277
(1977).