ADVANCESIN

COLLOID AND
INTERFACE
Advancesin Colloid and Interface Science
ELSEVIER 70(1997) 125-169
SCIENCE

Hamaker constants of inorganic materials

Lennart Bergstr6m
Institute for Surface Chemistry, P.O. Box 5607, S-114 86 Stockholm, Sweden

Abstract

Calculations of H a m a k e r constants using Lifshitz t h e o r y require the a v a i l a b i l i t y of

a c c u r a t e dielectric data, especially in the ultraviolet spectral region, a n d the use of a
convenient a n d a p p r o p r i a t e m a t h e m a t i c a l representation. In this review, a m u l t i p l e
oscillator model - - t h e so-called N i n h a m - P a r s e g i a n ( N - P ) r e p r e s e n t a t i o n - - h a s been
used a n d spectral p a r a m e t e r s for 31 different inorganic m a t e r i a l s (including diamond)
h a v e been g e n e r a t e d from critically e v a l u a t e d optical d a t a or collected from the litera-
ture. F o r m o s t m a t e r i a l s , a two-oscillator model (one UV a n d one IR t e r m ) was used b u t
more d e t a i l e d r e p r e s e n t a t i o n s were included when available. The spectral p a r a m e t e r s
p r e s e n t e d h e r e can be combined with previous data, m a i n l y focused on hydrocarbon a n d
organic systems, to yield a n extensive spectral d a t a b a s e for both solids a n d liquids
e n a b l i n g Lifshitz calculations of H a m a k e r constants for m a n y m a t e r i a l s combinations.
N o n - r e t a r d e d H a m a k e r constants for s y m m e t r i c m a t e r i a l combinations across vac-
u u m (Alvl) a n d w a t e r (Alwl) have been calculated for the different m a t e r i a l s ; these
calculations were p e r f o r m e d using t h e full Lifshitz theory. A s y m m e t r i c combinations,
A lv3 a n d A lw3, a g a i n s t four commonly used m a t e r i a l s in atomic force microscopy studies:
silica, a m o r p h o u s silicon nitride, s a p p h i r e , a n d muscovite mica, have also been covered.
The use of a new dielectric r e p r e s e n t a t i o n for w a t e r r e s u l t e d in significantly lower values
of A l w l c o m p a r e d to previous calculations. Analytical a p p r o x i m a t i o n s to the full Lifshitz
t h e o r y were e v a l u a t e d a n d found to give s u r p r i s i n g l y accurate r e s u l t s (the T a b o r - W i n -
t e r t o n a p p r o x i m a t i o n ) for A l v l when t h e IR contribution is of m i n o r importance. An
a t t e m p t to m a k e t h e TW a p p r o x i m a t i o n more general by e s t a b l i s h i n g some scaling
r e l a t i o n s h i p b e t w e e n no and couv was m e t with little success; only t h e UV spectral
p a r a m e t e r s of the covalent oxides, sulphides and n i t r i d e s m a y be fitted to a simple power
law relation.
The Lifshitz calculations in this s t u d y were compared with an a l t e r n a t i v e m e t h o d
w h e r e a more d e t a i l e d dielectric r e p r e s e n t a t i o n in the visible-ultraviolet spectral r a n g e
w a s o b t a i n e d t h r o u g h K r a m e r s - K r o n i g ( K - K ) t r a n s f o r m a t i o n of reflectivity d a t a over
a broad frequency range. Despite the difference in dielectric information, the two

0001-8686/97/$32.00 © 1997 - - Elsevier Science B.V. All rights reserved.

PII: S0001-8686(97)00003-1
126 L. BergstrSm /Adv. Colloid Interface Sci. 70 (1997) 125-169

m e t h o d s g e n e r a l l y yield n o n - r e t a r d e d H a m a k e r constants which do not differ signifi-

cantly. This is not t r u e for all m a t e r i a l s , e.g. water, w h e r e a more d e t a i l e d r e p r e s e n t a t i o n
using e i t h e r an N - P r e p r e s e n t a t i o n with several oscillators or the K - K r e p r e s e n t a t i o n
m u s t be used. It was shown t h a t the omission of the static and low frequency contribution
in the l a t t e r m e t h o d m a y r e s u l t in a significant u n d e r e s t i m a t i o n of the value for Alwl
w h e n the dispersive contribution becomes v e r y small.

Keywords: H a m a k e r constant, v a n der W a a l s interaction, inorganic m a t e r i a l s , i n t e r p a r -

ticle forces, A F M

Contents

I. Introduction .................................... 126

II. Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
IIA. Dielectric response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
IIB. Calculation of n o n - r e t a r d e d H a m a k e r constants . . . . . . . . . . . . 134
III. Dielectric r e p r e s e n t a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
IIIA. Dielectric d a t a of m a t e r i a l s . . . . . . . . . . . . . . . . . . . . . . . . 137
IIIB. D e t e r m i n a t i o n of spectral p a r a m e t e r s . . . . . . . . . . . . . . . . . . 139
IV. S p e c t r a l p a r a m e t e r s of m a t e r i a l s . . . . . . . . . . . . . . . . . . . . . . . . 142
IVA. A m o r p h o u s m a t e r i a l s . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
IVB. Cubic h a l i d e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
IVC. Cubic covalent m a t e r i a l s . . . . . . . . . . . . . . . . . . . . . . . . . 145
IVD. Hexagonal, t e t r a g o n a l and trigonal m a t e r i a l s . . . . . . . . . . . . . . 147
IVE. Birefringent m a t e r i a l s . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
V. C a l c u l a t e d H a m a k e r constants . . . . . . . . . . . . . . . . . . . . . . . . . 150
VA. I m p o r t a n c e of the IR r a n g e . . . . . . . . . . . . . . . . . . . . . . . . 150
VB. Full Lifshitz calculations a n d analytical a p p r o x i m a t i o n s . . . . . . . . 154
VC. H a m a k e r constants for use in A F M studies . . . . . . . . . . . . . . . 155
VD. S p e c t r a l p a r a m e t e r scalings . . . . . . . . . . . . . . . . . . . . . . . . 159
VE. C o m p a r i s o n w i t h a K r a m e r s - K r o n i g dielectric r e p r e s e n t a t i o n . . . . . 162
VI. S u m m a r y and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

I. I n t r o d u c t i o n

The surface forces between two bodies in close proximity play an

important role in many scientific and technical areas, e.g. adhesion,
wetting, adsorption, colloidal phase equilibrium, flocculation rates, rheologi-
c a l p r o p e r t i e s a n d o t h e r c o l l o i d a n d i n t e r f a c e p h e n o m e n a [1,2]. I n p h e n o m -
ena involving inorganic materials, examples include the important role of
interparticle forces in determining the rheological behaviour of the ceramic
suspensions in the powder preparation and consolidation step [2-4], and
 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169 127

the suggestion t h a t a surface force balance controls the thickness of t h i n

i n t e r g r a n u l a r glass films in polycrystalline ceramics [5].
A m o n g the m a n y contributions to the interaction between surfaces,
e.g. double layer, structural, steric, depletion, h y d r a t i o n and hydropho-
bic forces, t h e r e is one type of interaction which is always present, the
van der Waals interaction. This ubiquitous interaction m a y be of v a r y i n g
importance depending on the system. The H a m a k e r constant represents
a conventional and convenient way of assessing the m a g n i t u d e of this
interaction. Recent work has shown t h a t m a n y inorganic systems are
characterised by relatively high H a m a k e r constants [6,7], hence m a k i n g
this interaction of significant importance. It is obvious t h a t accurate
estimations of the H a m a k e r constants is necessary for a quantitative
u n d e r s t a n d i n g of the effect of interparticle forces on various phenomena.
The van der Waals force has an electrodynamic origin as it arises
from the interactions between atomic or molecular oscillating or rotating
electrical dipoles within the i n t e r a c t i n g media. There are t h r e e types of
interactions w h i c h contribute to the van der Waals force:
(i) The Keesom force which is the interaction between two p e r m a n e n t
dipoles.
(ii) The Debye force which is the interaction b e t w e e n one p e r m a n e n t
dipole a n d one induced dipole.
(iii) The London or dispersion force which is the interaction between
two induced dipoles.
B e t w e e n molecules, each of these contributions has an interaction
free e n e r g y w h i c h varies w i t h the inverse sixth power of the distance at
small and intermediate distances of separation (from 1-2 .~ to several nm).
H a m a k e r [8] calculated the distance dependence of the free e n e r g y
of macroscopic bodies by performing a pair-wise s u m m a t i o n over all the
atoms in the bodies. For example, the van der Waals interaction free
energy, VvdW, per u n i t area, between semi-infinite parallel plates at
separation, L, can be expressed as
A
Vvd w - 12xL 2 (1)

w h e r e A is the H a m a k e r constant. As seen in Eq. (1), t h e r e is a direct

proportionality b e t w e e n the m a g n i t u d e of the van der Waals interaction
and the H a m a k e r constant. The H a m a k e r constant is a m a t e r i a l s
c o n s t a n t t h a t depends on the properties of the two m a t e r i a l s and the
i n t e r v e n i n g media. The distance d e p e n d e n c e of the van der Waals
e n e r g y depends essentially on the geometry of the two i n t e r a c t i n g bodies
128 L. BergstrSm /Adv. Colloid Interface Sci. 70 (1997) 125-169

being proportional to L -2 for parallel plates and scale to L -1 for two

spherical particles at short separation distances where retardation can
be ignored.
In the original treatment, also called the microscopic approach, the
H a m a k e r constant was calculated from the polarizabilities and number
densities of the atoms in the two interacting bodies [8]. Lifshitz pre-
sented an alternative, more rigorous approach where each body is
treated as a continuum with certain dielectric properties [9]. This
approach automatically incorporates many-body effects which is ne-
glected in the microscopic approach. In the Lifshitz treatment, the van
der Waals interaction is the result of fluctuations in the electromagnetic
field between two macroscopic bodies, modified by the separating media,
where the interaction can be referred to the standing waves which only
occur at certain frequencies. Hence, the van der Waals interaction, and
thus the associated H a m a k e r constants, can be estimated from a knowl-
edge of the frequency dependent dielectric properties of the interacting
materials together with the intervening medium, and the geometry of
the bodies. The accuracy of the estimated H a m a k e r constants are
directly related to the precision and accuracy of the dielectric spectra
and the mathematical representation of this data.
The main focus of this review is to provide a collection of dielectric
representations for a range of inorganic materials presented in a con-
densed but appropriate manner, suitable for calculations of accurate
H a m a k e r constants. The dielectric representation follows the form
presented by Hough and White [10] in their collection of spectral
parameters and H a m a k e r constants for a number of materials, focusing
mainly on liquids and organic solids. Since their work was published,
collections of accurate dielectric data for many inorganic materials have
become easily available [11,12] which makes the determination of the
necessary spectral parameters an easy task. The two handbooks of
Optical Constants of Solids I and II [11,12] will constitute the main
source of dielectric spectral information but previously published data
will also be included.
In the next, theoretical section, the basis of the dielectric response
representation and an outline of the Lifshitz theory for van der Waals
forces together with a description of simplifying assumptions made to
facilitate calculations will be described. It should be noted that no
theoretical advances are being presented here; the theoretical section
completely relies on previously published material and merely forms a
basis for the estimations of spectral parameters and calculations of
 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169 129

H a m a k e r constants. This is followed by a presentation of dielectric data

and discussion how the spectral parameters in the ultraviolet (UV) and
the infrared (IR) frequency range can be determined. The spectral
parameters for 31 different inorganic materials (including diamond) are
either taken directly from the literature or estimated from data on the
static dielectric constant, ~(0), the frequency dependence of the refrac-
tive index, n(o~), in the UV frequency range, and the frequency of the
strongest IR-absorption band.
Non-retarded Hamaker constants for the inorganic materials across
vacuum and water was calculated using the full Lifshitz theory. In
addition, Hamaker constants for systems important in atomic force
microscopy (AFM) studies have also been calculated. A recently publish-
ed improved spectral parametric representation of water [13] has been
used for these calculations which resulted in some significant changes
compared to previous estimates. The full Lifshitz calculations have been
compared with various analytical simplifications and a surprisingly
good agreement was found for the Hamaker constants with vacuum as
the intervening medium. The importance of the IR range for the calcu-
lation of H a m a k e r constants was evaluated. In the final section, the
H a m a k e r constants obtained in this study were compared with calcula-
tions based on an alternative dielectric representation using K r a m e r s -
Kronig relations.

II. Theory

IIA. Dielectric response

The dielectric properties of a material are commonly represented by

the frequency dependent dielectric response function, e(co)
c(¢o) = c'(co) + ic"(co) (2)
which is a complex function where e'(co) is the real part, and e"(co) the
imaginary part. This function is intimately related to the transmission
(c'(co)) and absorption (e"(o~)) of the applied electromagnetic radiation.
In frequency regions where e"(¢o) is zero, the material is transparent and
the real part of the dielectric response is directly related to the refractive
index, n(co) through

c(co) = c'(¢9) = n2(~o) (3)

130 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169

For a static applied field (co = 0), non-conductor m a t e r i a l s are charac-

terised by e"(co) ~- 0, h e n c e ~'(0) = a(0) which is the static dielectric
constant.
As a completely equivalent a l t e r n a t i v e to the dielectric response
function, e(o~), can the complex refractive i n d e x , N(c0), be used
N(o~) = n(co) - i k ( o ) (4)
w h i c h m i g h t be more familiar to some readers. The real part, n(o2),
r e p r e s e n t s the refractive index and k(o3) is the extinction coefficient.
R e m e m b e r i n g the transmission and absorption d e p e n d e n c e of the com-
plex dielectric function, the m a t h e m a t i c a l connection between a(co) and
N(o~) can be expressed as:
~(o~) = N2(co) (5)
N2(co) = (n(co) - ik(¢o))2 (6)
e'(co) = n2(o)) - k2(¢o) (7)
e"(¢o) = 2n(co)k(¢o) (8)
Dielectric d a t a can be obtained with a n u m b e r of methods including
capacitance bridge, optical reflectance and ellipsometric techniques,
and electron loss spectroscopy. Most of these techniques yield only one
of the components of the complex dielectric response function w h i c h
often necessitate the t r a n s f o r m a t i o n from the real to the i m a g i n a r y
c o m p o n e n t or v i c e v e r s a . Such a t r a n s f o r m a t i o n can be performed using
the K r a m e r s - K r o n i g (K-K) relations, one of t h e m defined as

xe"(o))
a'(co) = 1 + 2 i ~ _ - ~ 2 dx (9)
0

w h e r e x r e p r e s e n t s real values of the frequency. The K - K relation in Eq.

(9) shows t h a t if e"(o~) (i.e. the absorption spectrum) is k n o w n in the
entire frequency range, 0 < co < ~, t h e n e'(o~) can be fully defined. Similar
K - K relations can be defined for other dielectric properties such as the
f r e q u e n c y d e p e n d e n t i n t e r b a n d transition s t r e n g t h [7].
N i n h a m a n d P a r s e g i a n [14,15] showed t h a t by constructing an imagi-
n a r y dielectric response function, e(i~m), a m u c h simpler dielectric
r e p r e s e n t a t i o n could be used for the purpose of calculating H a m a k e r
constants. H o u g h and White [10] have given a t h o r o u g h analysis of this
representation. The reason for using this i m a g i n a r y r e p r e s e n t a t i o n is
 L. Bergstr~m /Adv. Colloid Interface Sci. 70 (1997) 125-169 131

completely m a t h e m a t i c a l , the i m a g i n a r y p a r t of the frequency has no

physical significance t h a t refers to the m a t e r i a l s behaviour. The com-
plex i m a g i n a r y function decreases monotonically from ~0 w h e n ~ = 0,
a n d goes to 1 as ~ approaches infinity. The final form of the Lifshitz
expression for calculation of H a m a k e r constants (see next section) shows
t h a t i m a g i n a r y dielectric function only has to be e v a l u a t e d at discrete,
equally spaced i m a g i n a r y frequencies, i~m, w i t h
~m = rn( 4~2kB T/h ) (10)
w h e r e h is Planck's constant, k B represents Boltzmann's constant, m is
an integer; m = 0,1,2,3,4.., and T is the t e m p e r a t u r e in Kelvin. The
i m a g i n a r y dielectric response function, e(i~m), can be obtained from
g'(co) using a n o t h e r K r a m e r s - K r o n i g relation [10]:
oo

_2f
~(i~) = 1 + ~ ~ x2 + ~2 dx (11)
0

The K - K relation in Eq. (10) has been used in several studies to

construct the n e c e s s a r y a(i~m) function [7,16,17] for estimations of
H a m a k e r constants. However, since the K - K relation is formally correct
only w h e n ~"(co) is k n o w n in the entire frequency range, 0 < 0`) < ~, this
p r o c e d u r e is prone to n u m e r i c a l error since approximations for parts of
the absorption s p e c t r u m invariably has to be used due to lack of spectral
information.
With limited dielectric data available, the function s(i~m) is usually
represented by a model based on a damped oscillator. The so-called
N i n h a m - P a r s e g i a n (N-P) representation of ~(i~) can be given as [14,15,
18,191
N
B Cj (12)
~(i~m) = 1 + 1 + ~----~+ y=" 1 + (~ /coj)2 + gj~ /0.)2
j=l
w h e r e the first t e r m after u n i t y represents the contribution from the
orientation of p e r m a n e n t dipoles (present in polar liquids like water),
a n d t h e final s u m m a t i o n relates to the absorption peaks in the infrared
and ultraviolet range. B and gj are related to the oscillator s t r e n g t h in
the microwave r a n g e and d a m p i n g coefficient of the oscillator, respec-
tively, and Cj is r e p r e s e n t e d by

cj- 2 fj (13)
132 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169

where fj is the oscillator strength and coj is the relaxation frequency (in
the unit rad/s) of the absorption band. The N-P representation simplifies
the construction of the required function e(i~m) since the frequencies and
relative strengths of the peaks in the absorption spectrum can be used.
With a well-characterised absorption spectrum over a relatively wide
frequency range, many spectral parameters can be defined. Water is such
a material and spectral parameters in the ultraviolet, infrared as well as
the microwave range have been determined [13,20]. Roth and Lenhoff [13]
have recently published an improved UV representation of water based
on previously published reflectivity data [21]. Figure 1 shows that the
real and imaginary dielectric data in the important frequency range can
be closely fitted by using the summation in Eq. (12) and six [IV terms
including band widths. The full spectral representation of water includ-
ing one microwave term and five infrared terms from Parsegian [20] and
the six UV terms from Roth and Lenhoff [13] are collected in Table 1.
The accuracy of Hamaker constants calculated from Lifshitz theory
relies on the quality of the experimental dielectric data and the accuracy
of the spectral representation. With dielectric data over a wide fre-
quency range, the two alternative methods: K-K transformation using
Eq. (11), or the N - P representation, Eq. (12), should yield a similar
result. If the dielectric data is limited, both approaches may result in
inaccurate representations. The K-K transformation is sensitive to the

Table 1
Spectral parameters* for water**

eli V O~uvx 1016 guv× 1015 CIR C01RX 1014 gIR × 1013
(rad/s) (tad/s) (rad/s) (rad/s)

0.0484 1.25 0.957 1.46 0.314 2.29

0.0387 1.52 1.28 0.737 1.05 5.78
0.0923 1.73 3.11 0.152 1.40 4.22
0.344 2.07 5.92 0.0136 3.06 3.81
0.360 2.70 11.1 0.0751 6.46 8.54
0.0383 3.83 8.11

B = 76.8 a n d 1/TMW = 1.08 x 1011 rad/s.

* C m a n d Cuv are the absorption strengths, and com and a)uv r e p r e s e n t the characteristic
absorption frequencies in the IR and UV range, respectively. The d a m p i n g t e r m s are
denoted by gia a n d guy.
** UV r e p r e s e n t a t i o n from Roth and Lenhoff [13], and IR a n d microwave r e p r e s e n t a t i o n
from P a r s e g i a n [20].
 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169 133

(a)

... Roth and L e n h o f f [13]

2.5 - - P a r s e g l a n [14]

- - Data [21._._~_~]m

1.5
LL

0.5

0 l i i i I
0 5 10 15 20 25 30
Frequency (eV)

(b)
1.8
.,. Roth and L e n h o f f [13]
1.6 /',~ . - P a r s e g l a n [14]

1.4 ~" ~ uuta L~:lj

1.2

1
%
0.8

0.6

0.4

0.2

0 ! I i i i
5 10 15 20 25 30
Frequency (eV)

Fig. 1. Comparison of experimental dielectric data (solid line) for water of the real (a)
and imaginary (b) component [21], and two different mathematical representations
(broken lines) from Parsegian [14] and Roth and Lenhoff [13]. Each representation uses
six oscillators including band widths. From Roth and Lenhoff [13].
134 L. Bergstr6rn/Adv. ColloidInterface Sci. 70 (1997) 125-169

approximations outside the available spectral data, and the N - P repre-

sentation, using a limited n u m b e r of oscillators, m a y be of poor quality
for m a t e r i a l s w i t h complicated spectral features.
However, t h e r e is some evidence t h a t m a n y m a t e r i a l s w i t h "simpler"
spectral f e a t u r e s can be reasonably well described by less complicated
N - P r e p r e s e n t a t i o n s [10]. At room t e m p e r a t u r e , ~m are sampled at
i n t e g r a l multiples of 2.4 × 1014 rad/s w h i c h corresponds to one static
term, only a few t e r m s in the infrared and m a n y in the visible-ultravio-
let region. Hence, since the dielectric response in the visible-UV r a n g e
completely d o m i n a t e s the v a n der Waals interaction for most systems,
it is of outmost importance to have access to accurate spectral d a t a in
this frequency range. In this study, E(i~m) for most of the inorganic
m a t e r i a l s will be r e p r e s e n t e d by one UV and one IR relaxation

CIR CUV
e(i~m) = 1 + + (14)
l+(~/com) 2 l+(~/O)uv) 2

w h e r e each m a t e r i a l is characterised by four parameters; CIR and Cvv

are t h e absorption s t r e n g t h s in the IR and UV range, respectively, a n d
0)IR a n d a)vv r e p r e s e n t the characteristic absorption frequencies in the
IR a n d UV range, respectively. The d a m p i n g t e r m s are omitted. In some
cases w h e n a more detailed r e p r e s e n t a t i o n has already been presented,
this d a t a will be included. The limitations and accuracy of this dielectric
r e p r e s e n t a t i o n will be discussed in more detail later.

IIB. Calculation of non-retarded Hamaker constants

The interaction free e n e r g y per u n i t area, E132, b e t w e e n two half-

spaces of m a t e r i a l 1 and m a t e r i a l 2 interacting over m e d i a 3, can in the
n o n - r e t a r d e d approximation (small separation distance, L) be w r i t t e n
as [9,10,18]

kBT
E132- ~ 2" I ydy ln(1 - A13A23 e -2yL) (15)
m--0 0
w i t h the differences in dielectric response, Akl, defined as

ak(i~m ) -- ez(i{m )
(16)
Akl- ek(i~m ) + e/(i~ m)
 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169 135

where ek(i~m) and el(i~m) are the dielectric response function of material
k and l, respectively, evaluated at the imaginary frequency, i~ n accord-
ing to Eq. (10). The prime on the first summation in Eq. (15) indicates
that the m = 0 term (the static contribution) is given half weight. By a
change of variables (x = 2yL), it is possible to define El32 according to
Eq. (1),

A132
El3 2 = - 12EL 2 (17)

thus identifyingthe non-retarded Hamaker constantA132 which is given

as

3kBT
A132 - 2 ~' f dxx ln(1 - A13A23e -x) (18)
m=O 0

This integral can be performed analytically to yield

3kBT (A13A23)s
A132- 2 ~'~ s3 (19)
rn=0 s = l

A detailed account of how this double summation equation (Eq. (19)) is

derived directly from the original expression given by Lifshitz [9] can be
found in previous work [18]. The frequency sum can be replaced by an
integral to yield

3h (A13A23)s
A132: f at E s3 (20)
0 s=l

which is correct only in the limit, T --->0, where the distance between
the sampled frequencies (Eq. (10)) tend to zero [9,18]. This transforma-
tion is equivalent to omitting the static term (m = 0, ~ = 0) in the
frequency summation (Eq. (19)). The temperature independent disper-
sive part of the non-retarded Hamaker constant, represented by the
summation m > 1, however, is not affected by the transformation. Hence,
when the static contribution to the non-retarded Hamaker constant is
negligible, the error of using Eq. (20) will be minor. However, when
water or another material characterised by a high static dielectric
constant is one part of the system, the static contribution may be
significant thus restricting the use of the integral equation (Eq. (20)).
136 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169

T h i s will be i l l u s t r a t e d in m o r e d e p t h later. Also, at e l e v a t e d t e m p e r a -

t u r e s , t h e static c o n t r i b u t i o n to t h e H a m a k e r c o n s t a n t m a y be large.
Simplified a p p r o x i m a t i o n s for less a c c u r a t e e s t i m a t i o n s of non-re-
t a r d e d H a m a k e r c o n s t a n t s h a v e been derived for a n u m b e r of cases by
only r e t a i n i n g a single UV r e l a x a t i o n to r e p r e s e n t t h e dielectric r e s p o n s e
of each m a t e r i a l . T h e s e simplifications h a v e b e e n m o t i v a t e d by t h e
pivotal role of t h e u l t r a v i o l e t r e l a x a t i o n s on t h e m a g n i t u d e of t h e
d i s p e r s i v e interactions. W i t h s u c h a simple dielectric r e p r e s e n t a t i o n ,
e a c h m a t e r i a l is fully described by t h r e e p a r a m e t e r s , e(0), t h e charac-
teristic f r e q u e n c y in t h e ultraviolet, ¢otr¢, a n d t h e low-frequency limit of
t h e refractive i n d e x in t h e v i s i b l e - U V r a n g e , n o (since Cuv = n2s - 1).
F o r e x a m p l e , t h e n o n - r e t a r d e d H a m a k e r c o n s t a n t for two identical
m a t e r i a l s 1, i n t e r a c t i n g across a m e d i u m 3 w a s calculated by u s i n g Eq.
(19), n e g l e c t i n g all s u m m a t i o n t e r m s s > 1, c o n v e r t i n g t h e s u m m a t i o n
over n to an i n t e g r a l for m > 1, a n d finally s e t t i n g (01 -- 033 ~ 03. This
r e s u l t s in [1]

3 k T tel(0) _ e3(0)~2 3he0 (n 2 -- n2) 2

A131 = T el(0 ) + ~ !) + 3 2 ~ / 2 n21 + n32)3/2 (21)

w h e r e t h e first t e r m on t h e r i g h t h a n d side r e p r e s e n t s t h e static

c o n t r i b u t i o n a n d t h e second t e r m r e p r e s e n t s t h e dispersive c o n t r i b u t i o n
a n d n I a n d n 3 is t h e refractive index in t h e visible of t h e m a t e r i a l a n d
m e d i u m , respectively. If t h e m e d i u m is v a c u u m a n d t h e static contribu-
tion is ignored, Eq. (21) r e d u c e s to t h e so called T a b o r - W i n t e r t o n (TW)
r e p r e s e n t a t i o n [22]. A m o r e complicated expression for t h e dispersive
c o n t r i b u t i o n is o b t a i n e d if t h e i n t e r a c t i n g m a t e r i a l a n d t h e m e d i u m are
allowed to h a v e different a b s o r p t i o n frequencies 031 ¢ (03 [23].

Adisp,131 = 3h ~ (X2 ~ + 2XAe a 1/2 + Ae2(3 + 211)) (22)

1 2 8 / ~ 7'4 [ ( y _ h/~-~-i-)1/2 + (Y + a ] - ~ _ lil/213

where
= 0.5(nl2 + n32)

AE = n 2 - n 2

°)1 (03
X = - - ( n 2 - i ) - - - (n~°- 1)
(03- (01
 L. BergstriJm/Adv. Colloid Interface Sci. 70 (1997) 125-169 137

Y = 4_~1/----~ (n 2 + 1) + - - (n 2 + 1)
COl

It m a y be shown t h a t Eq. (22) will be identical to the dispersive p a r t of

Eq. (21) if °31 and (03 are replaced with co. These approximations of the
n o n - r e t a r d e d H a m a k e r constant and others r e p r e s e n t i n g dissimilar
m a t e r i a l s across a dielectric and interactions b e t w e e n coated m a t e r i a l s
have been presented in a condensed form by Prieve and Russel [19]. In this
study, the appropriateness of the different approximations for inorganic
materials will be evaluated and the importance of the IR-range and the
static contribution to the m a g n i t u d e of the H a m a k e r constant will be
evaluated. For the full Lifshitz calculations using Eq. (19), a s u m m a t i o n
over s = 1 to 4 and the first 3000 t e r m s of m w h e r e employed.

III. Dielectric r e p r e s e n t a t i o n

IIIA. Dielectric data of materials

The dielectric response can be directly referred to the electronic and

vibrational states of a material. In the infrared region, lattice vibrations
d o m i n a t e while in the ultraviolet-visible frequency range, electronic or
i n t e r b a n d transitions d o m i n a t e for a non-conducting solid. T h e r e is a
w e a l t h of d a t a on the frequency dependence of the optical properties of
solid inorganic m a t e r i a l s available in the literature. M a n y inorganic
m a t e r i a l s h a v e been critically s u m m a r i s e d in the two H a n d b o o k s of
Optical C o n s t a n t s of Solids [11,12] which constitute an excellent source
of d a t a for the p r e s e n t t a s k of elucidating spectral p a r a m e t e r s for
inorganic materials.
Some recently published optical d a t a on ceramic m a t e r i a l s [6] illus-
t r a t e the typical features of the optical characteristics in the i m p o r t a n t
UV-visible range. Figure 2 shows a plot of the refractive index, n(c0),
a n d extinction coefficient, k(co) (ZnO only), as a function of photon e n e r g y
for the polycrystalline materials: 6H-SiC, tetragonal ZrO2, a-A1203, a n d
ZnO based on spectroscopic ellipsometry m e a s u r e m e n t s . Three of these
materials: SiC, ZrO 2 and A1203, display the n o r m a l dispersive b e h a v i o u r
of a t r a n s p a r e n t material; the refractive index increases w i t h an in-
crease in photon e n e r g y (frequency). This is related to a low photon
absorption w i t h the extinction coefficient, k(co), close to zero and always
below 0.1 (not shown). ZnO on the other hand, shows a distinct p e a k in
the refractive index and a s h a r p increase in the extinction coefficient at
 138 L. Bergstr6mlAdv. Colloid Interface Sci. 70 (1997) 125-169

(a) 3.5

J
3.0
C

.E
2.5
"5

O
iv-
2,0 ¸
AI203

i i ,
1.5
2 3 4
P h o t o n e n e r g y (eV)

(b) 2.8 1.0

ZnO

2.6! 0.8

j
e,-

.E
,>
~'2.4

L_. 0.6 "~

0
0

2.2 0.4.
o
c
x
2.0 0.2 I,U

1.8 ' 0.0

2 3 4 5
P h o t o n e n e r g y (eV)

Fig. 2. Optical characteristics versus photon energy from ellipsometry measurements

for (a) 6H-SiC, tetragonal, partially stabilized ZrO2, and a-A1203; and (b) ZnO. Taken
from Ref. [6].

a critical p h o t o n e n e r g y . A t p h o t o n e n e r g i e s b e l o w 3.2 eV, t h e e x t i n c t i o n

coefficient is close to zero (k < 0.05). H o w e v e r , a b o v e 3.3 eV, k s u d d e n l y
i n c r e a s e s to k > 0.5. H e n c e t h e m a t e r i a l c e a s e s to b e t r a n s p a r e n t in t h e
h i g h p h o t o n e n e r g y range. This t r a n s i t i o n is a s s o c i a t e d w i t h a n electronic
e x c i t a t i o n since Z n O is a s e m i c o n d u c t o r w i t h a b a n d g a p a t 3.3 eV [24].
 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169 139

For the simplified spectral r e p r e s e n t a t i o n used in this study, optical

d a t a over a relatively n a r r o w frequency range (as shown in Fig. 2)
suffices for the elucidation of the UV term. However, if a more detailed
spectral r e p r e s e n t a t i o n of the ultraviolet spectral region is to be con-
structed, dielectric d a t a over a wider frequency r a n g e is needed, as
i l l u s t r a t e d for w a t e r in Fig. 1.
Anisotropic m a t e r i a l s can display different dielectric responses in
different crystal orientations. This can be r e p r e s e n t e d by a dielectric
response function in tensorial form w h e r e the n u m b e r of i n d e p e n d e n t
components depends on the crystal symmetry. Cubic systems only h a v e
one component, i.e. display no anisometry, while tetragonal, hexagonal
and trigonal systems have two components, r e s u l t i n g in optical birefrin-
gence. More complicated crystal forms such as orthorhombic, monoclinic
a n d triclinic systems display a h i g h e r n u m b e r of components. In a
birefringent material, the following n o m e n c l a t u r e is often used:
no, n± = o r d i n a r y ray, E p e r p e n d i c u l a r to the optical axis
neo, n//= e x t r a o r d i n a r y ray, E parallel to the optical axis
w h e r e E denotes the electric field. The s e p a r a t e components can only be
d e t e r m i n e d for single crystals while polycrystalline m a t e r i a l s yields an
a v e r a g e due to the r a n d o m orientation of the grains.

IIIB. Determination of spectral parameters

In the simplified r e p r e s e n t a t i o n of the ~(i~m) function used in this
study, the ultraviolet frequency region is r e p r e s e n t e d by only one
oscillator w h i c h can be w r i t t e n as
Cuv
~(i~m) = 1 + (23)
1 + (~ / tour) 2
valid at frequencies below the band gap and m u c h above the i n f r a r e d
absorption band. For m a t e r i a l s t h a t do not absorb in the visible and UV
range, the dielectric s p e c t r u m is simplified according to Eq. (3). H o u g h
and W h i t e [10] utilised a simple graphic m e t h o d of obtaining Ouv and
Cuv from the frequency d e p e n d e n c e of the refractive index, n(c0); the
so-called "Cauchy equation"

n2(co) - 1 = (n2(co) - 1) °)2 + Cuv (24)

which should result in a linear plot of [(n2(o~) - 1)] versus [(n2(co) - 1) (o2].
140 L. B e r g s t r f m / A d v . Colloid Interface Sci. 70 (1997) 1 2 5 - 1 6 9

"C

~c I I I

• o. t":, ~.. o. ~. ~. ~. m. °q. ~ O. ~.. ~. m. ~. e,!. ~ ~. ~

g~

ea~

0 0
° ~ ~ ~ .~-~- F

~ ~.~
 L. Bergstr6m /Adv. Colloid Interlace Sci. 70 (1997) 125-169 141

~0

~D ~D

~ ~ ~ o ~ o ~ ~
a3

¢9

¢9

~9

¢9

° ~° ~°
~ o~°
~
~ o o ~ . o~~
~

~9 ¢9
4.z
O 0
142 L. BergstrSm /Adv, Colloid Interface Sci. 70 (1997) 125-169

Figures 3-6 show "Cauchy" plots based on optical data from the
literature [11,12] for a number of inorganic materials. The data was
selected from the frequency range where the materials display negli-
gible extinction coefficients (below the band gap) and sufficiently above
the IR absorption range. More details of the spectral data is given below
for each of the materials. The slope and intercept of the straight lines
were determined by least square regression, which yields Cu r and tour.
The UV-visible optical data for all the materials in this review can be
well fitted to this type of single oscillator model. Materials where the
optical data was too sparse or the Cauchy fit too poor, have not been
included here. The values of the spectral parameters are collected in
Table 2.
A simplified infrared representation was used since the infrared and
microwave contribution to the ~(i~m) function usually is of minor impor-
tance. This is due to the sparse sampling which is related to the large
spacing of the sampling frequencies, ~m is about 2.5 × 1014 rad/s at room
t e m p e r a t u r e (see Eq. (10)). The spectral constant CIR is given by

CIR -- E(O) -- C U V - 1 (25)

and the characteristic absorption frequency, C0iR,will be defined as the

frequency of the major absorption peak in the IR spectrum. The values
of static dielectric constants for the different materials and the IR data
were obtained from two major sources [12,25]. More details of the
infrared data are given below for each material.

IV. Spectral p a r a m e t e r s of materials

The values of both the IR and UV spectral parameters for the 31

inorganic materials (including diamond) are summarised in Table 2.
The dielectric data or representation of each material is briefly described
below. All the spectral data represents room-temperature values.

IVA. Amorphous materials

Two amorphous materials were included in Table 2: silicon nitride

(Si3N 4) and silica (SiO2). Senden and Drummond [26] have published a
thorough analysis of the spectral parameters of these materials in both
the UV-visible and infrared spectral region, included in Table 2.
 L. Bergstri~m/Adv. Colloid Interface Sci. 70 (1997) 125-169 143

IVB. Cubic halides

All of the spectral parameters for the halides have been obtained from
Cauchy plots (UV-visible spectral range), and infrared absorption peaks
(IR spectral range) from the literature. Figure 3 shows that all of the
cubic halides display a similar behaviour, well described by the single
oscillator expression in the UV-visible range.

1. CaF2
Bezuidenhout [27] has compiled the data for calcium fluoride which
is a highly ionic compound having a cubic fluorite crystal structure. The
data of Malitson [28] on the refractive index values in the ultraviolet-
visible frequency range was used in the Cauchy plot (Fig. 3). The
infrared absorption data from Kaiser et al. [29] was used to determine
the frequency of the main absorption peak.

r , , ' J = ! I i , , , , i i , ,

A
I
I " b
c~
r-

i .I I I I I I t I J i ~ ! I i i f i I

0 1 10 32 2 10 32

(o2(n2(oj)-1) [(rad/s) 2]
Fig. 3. Cauchy plots for the cubic ionic halide materials; (a) CsI, (b) KBr, (c) NaC], (d)
KC1, (e) CaF2, (f) LiF, (g) NaF.
144 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169

2. CsI
Cesium iodide which has a cubic CsC1 structure does also display a
very wide transmission band making it a common window material in
infrared spectroscopy. Eldridge [30] has compiled the dielectric data and
the ultraviolet-visible data used in the Cauchy plot (Fig. 3) were based
on the tabulated values by Li [31]. The very low frequency of the infrared
absorption peak was determined by Beairsto and Eldridge [32] using
transmission measurements.

3. K B r
The spectral data of potassium bromide having a cubic rock salt
crystal structure has been compiled by Palik [33]. The collection and
analysis of refractive index data by Li [34] was used for the Cauchy plot
in Fig. 3. Infrared data from Johnson and Bell [35] was used for
absorption peak determination.

4. K C l
Palik [36] has also collected the spectral data of potassium chloride
having a cubic rock salt crystal structure. The ultraviolet-visible data
used in the Cauchy plot (Fig. 3) was based on refractive index data
tabulated by Li [31]. The infrared absorption peak was identified with
the transverse phonon frequency, also from Li [31].

5. L i F
Lithium fluoride also display the cubic rock salt crystal structure.
The dielectric data has been compiled by Palik and H u n t e r [37], and the
Cauchy plot in Fig. 3 is based on tabulated refractive index data by Li
[38]. The identification of the main absorption peak in the infrared was
based on the data by Kachare et al. [39].

6. N a C l
The refractive index data of sodium chloride used in the Cauchy plot
in Fig. 3 was taken from Li [31] again, compiled by Eldridge and Palik
[40]. The infrared data are from Eldridge and Staal [41].

7. N a F
The dielectric data of sodium fluoride, having a cubic rock salt crystal
structure, was compiled by Ohlidal and Navratil [42]. The refractive
index data in the ultraviolet-visible range was obtained from Harting
[43], while the infrared data originates from Pai et al. [44].
 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169 145

IVC. Cubic covalent materials

The spectral parameters of two of the cubic, covalent materials,

magnesium oxide (MgO) having a periclase structure and magnesium
aluminate (MgA1204) having a spinel structure, were collected from a
previous study using literature optical data [6].

1. C (diamond type IIa)

Although diamond (cubic carbon) cannot be classified as an inorganic
solid, it was included in this review due to the general interest in this
material. Diamond is classified into four types depending on its optical
and electrical properties. Edwards and Philipp [45] have compiled
spectral data for the natural types Ia and IIa and they concluded that
the difference of the refractive index in the visible range is very small
(less than 1%) between these types. In the Cauchy plot (Fig. 4), the

i ! = = i : i i i i , i | ! j = = : i i = r i i i i = : =

|St ......... i .........

171 ~
I S

16

I- 7
3 o 5 1030 1 SiC
CM
I=
6

S rTiO a ~ ~
C (diamond II)

I I I I I I I I I ~ I I I I I I 1 l I I I I I I I I I I I

1 10 32 2 10 32 3 10 32

0)2(n2(~o)-1) [(radls) 2]
Fig. 4. Cauchy plots for the cubic materials: p-SiC, C (diamond type II), SrTiO3, ZnS,
and PbS (insert).
146 L. BergstrOm/Adv. Colloid Interface Sci. 70 (1997) 125-169

sparse refractive index data from Peter [46] was used. The infrared
absorption is of minor importance since the absorption, and thus the CIR
value, is so low (Table 2). However, the value of the phonon frequency
[47] was used for the characteristic frequency in the infrared.

2. PbS
Lead sulphide is a semiconductor having a cubic rock salt crystal
structure. Guizzetti and Borghesi [48] have compiled the dielectric data,
generally focusing on materials with low free carrier concentrations.
Since lead sulphide displays significant absorption in the ultraviolet-vis-
ible frequency range, the Cauchy plot in Fig. 4 rests on some sparse data
in the near-infrared, originating from Zemel et al. [49]. Fortunately, the
infrared absorption band occurs at very low frequencies [50], leaving a
sufficiently broad transparent frequency region for Cauchy treatment.

3. p-SiC
The spectroscopic ellipsometry data on cubic p-silicon carbide re-
ported by Alterovitz and Woollam [51] was used in the Cauchy plot (Fig.
4). The spectral data of this material, which is of the zincblende struc-
ture, agree very well with reported values of n o for hexagonal 6H-silicon
carbide [52]. The infrared absorption data are from Spitzer et al. [53].

4. SrTi03
Dielectric data on undoped strontium titanate single crystals of the
cubic perovskite structure have been collected by Gervais [54]. The n(¢o)
data plotted in the Cauchy plot (Fig. 4) were taken from reflectivity
measurements by Cardona [55]. The infrared absorption data are from
Spitzer et al. [56].

5. Z n S
The natural crystal modifications of zinc sulphide include the cubic,
zincblende, and the hexagonal, wurtsite type. Zinc sulphide minerals
usually contain substantial amounts of impurities with iron being the
principal one. These impurities affect the colour of the minerals with the
purest minerals having a light yellow body colour [57]. Palik and Ad-
damiano [57] have collected the dielectric data of both of the zinc sulphide
forms, focusing on synthetic materials or pure minerals. The calculated
refractive index values presented by Pikhtin and Yaskov [58] in the
ultraviolet-visible frequency range for the cubic modification were used
for the Cauchy plot in Fig. 4. Infrared data was taken from Deutsch [59].
 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169 147

IVD. Tetragonal, hexagonal and trigonal materials

In this section, the spectral parameters of tetragonal, hexagonal or

trigonal materials with negligible or undetermined birefringence are
reported. The spectral parameters of several polycrystalline ceramic
materials were recently reported by BergstrSm et al. [6]. They used
spectroscopic ellipsometry to determine n(¢o) in the ultraviolet-visible
spectral range and calculated the oscillator parameters from Cauchy
plots (UV range). The spectral constant CIR was determined from the
static dielectric constant using Eq. (25). Their spectral parameters for
polycrystalline hexagonal 6H-SiC and ~-Si3N4, and tetragonal, partially
stabilised ZRO2(3% Y) are included in Table 2 with some slight modifi-
cations. The number of valid figures has been increased and a slightly
lower value of e(0) for polycrystalline, hexagonal {~-A1203- - which is the
average value of~(0) for the ordinary and extraordinary ray - - w a s used
in this study. The spectral parameters of hexagonal yttrium oxide (Y203)
were also collected from the study by BergstrSm et al. [6] using optical
data from the literature.
Crystalline quartz (SiO 2) of the trigonal crystal structure displays a
small birefringence, as analysed by Hough and White [10]. Only the
average values are reported in Table 2. Senden and Drummond [26]
have performed a thorough analysis of the dielectric data of muscovite mica
which is a potassium aluminosilicate (KA]2Si3AIO10(OH)2)belonging to the
monoclinic crystal class. They concluded that the spectral parameters in
different crystallographic directions affect the calculated by no more
than +1%; hence, only the average values are reported in Table 2.
1. CdS
The semiconducting compound cadmium sulphide (CdS) is normally
of the hexagonal close-packed (wurtzite) crystal structure but a cubic form
may occur. Here, only the hexagonal form will be treated, summarised by
Ward [60]. Data for both the ordinary and extraordinary rays are
available but since the difference in dielectric response is negligible,
data for polycrystalline CdS [61] in the ultraviolet-visible range was
used for the Cauchy plot in Fig. 5. The infrared data was from Manabe
et al. [62].
2. MgF2
Magnesium fluoride (MgF 2) is an ionic, tetragonal material with
rutile crystal structure. The dielectric data of this slightly birefringent
material has been compiled by Cotter et al. [63]. Refractive index data
148 L. BergstrSm/Adu. Colloid Interface Sci. 70 (1997) 125-169

' ' ' ' i . . . . I . . . . I " ' ' ' I '

CdS

ZnS

Zl'~O i # l l g l l l i l l l l l l l U i l ii ill_l! i-I I!!-! I-!

I

£
e-

MgF 2
0 O----e 0 0

0 = i i i I = = ~ = I I I I I [ I ~ ~ ~ I i i t 1

0 1 103t 2 103~ 310 ~t 41031 51031

(o2(n2(o~)-1) [(rad/s) 2]
Fig. 5. Cauchy plots for tetragonal and hexagonal materials with negligible birefrin-
gence; CdS, MgFe, ZnO, ZnS.

for the ordinary ray, presented in a fitted form by Dodge [64] was used
for the Cauchy plot in Fig. 5. The infrared data was from Barker's
compilation of IR modes [65].
3. Z n O
The optical data in the ultraviolet-visual range for polycrystalline,
hexagonal zinc oxide (ZnO) was taken from the Fig. 2, previously
reported by BergstrSm et al. [6]. Only the data in the transparent region,
thus sufficiently below the band gap energy, was used for the Cauchy
plot in Fig. 5. The infrared data was from Farmer [66].
4. Z n S
The optical data in the ultraviolet-visual spectral range for the
hexagonal modification of zinc sulphide was taken from Bienewski and
Czyzak [67]. The material only displays a minor birefringence in this
transparent spectral region so only the ordinary ray data was used for
the Cauchy plot (Fig. 5). The infrared data was from Manabe et al. [62].
Comparison of the spectral parameters for cubic and hexagonal zinc
 L. Bergstr6m /Adv. Colloid Interface Sci. 70 (1997) 125-169 149

sulphide show relatively small differences; hence, it is not critical to

know the exact composition when estimating the Hamaker constant for
a phase mixture, e.g. the commercially available IRTRAN 2 material.

IVE. Birefringent materials

Previous analysis of the dielectric data in different crystallographic
orientations has shown that the trigonal material calcite (CaCO 3) and the
tetragonal material rutile (TiO 2) display significant birefringence. The
data for calcite from Hough and White [10] and the spectral parameters
for rutile, recently reported by Buscall [68], are shown in Table 2.
1. BaTi03
The Cauchy plot for tetragonal BaTiO 3 having a perovskite crystal
structure, is shown in Fig. 6. Spectral data have been collected by Wong

2,2 BaTiO 3
. . . . . . . ' . . . . Be5
EO
2 O

1,8 . . . . . . . I ........ A
I 0 4 1031 8 10 31
5

4
0 5 1 0 31 1 1 0 32

0)2(n2(~)-1) [(rad/s) 2]
Fig. 6. Cauchy plots for the birefringent materials; tetragonal BaTiO3, and inserted;
hexagonal BeO. The ordinary (O) and extraordinary (EO) rays represent the refractive
indices perpendicular and parallel to the optical axis, respectively. The averaged
representation (A) can be obtained by n a = (1/3)neo + (2/3)n o.
150 L. Bergstrbm/Adv. Colloid Interface Sci. 70 (1997) 125-169

et al. [69] and the optical properties of melt-grown single crystal BaTiO 3
[70] w a s used in the C a u c h y plot. It can be seen t h a t the optical
anisotropy for this m a t e r i a l is quite significant; the curves for the
o r d i n a r y a n d e x t r a o r d i n a r y rays are well s e p a r a t e d and t h e r e is a 4-6%
difference in obtained spectral p a r a m e t e r s for the two crystallographic
orientations. The average r e p r e s e n t a t i o n for the ultraviolet-visible
spectral r a n g e was obtained by adding one third of the c-axis (extraor-
dinary) contribution and two thirds of the a-axis (ordinary) contribution
following r a t i o n a l e above. The infrared and static spectral p a r a m e t e r s
will be discussed in detail later.
2. BeO
The relatively sparse optical d a t a for hexagonal beryllium oxide
(BeO) h a v e been collected by E d w a r d s and White [71]. The d a t a from
N e w k i r k et al. [72] was used for the C a u c h y plot (insert in Fig. 6). The
optical a n i s o m e t r y is quite significant also for this material, in particu-
lar r e g a r d i n g the (Ouv w h e r e a 20% difference b e t w e e n the two crystal-
lographic orientations was found. The infrared d a t a was from Loh [73].

V. C a l c u l a t e d H a m a k e r c o n s t a n t s

VA. Importance of the IR range.

The i m p o r t a n c e of the IR r a n g e for the calculation of H a m a k e r
c o n s t a n t s was tested for two materials: BaTiO 3 which is c h a r a c t e r i s e d
by a v e r y h i g h e(0), and t h u s a high Cm, and SiO 2 which h a s a low e(0).
Full Lifshitz calculations, using Eq. (19), were performed across t h r e e
different media: v a c u u m , w a t e r and n-dodecane; investigating the effect
of the i n f r a r e d contribution. The calculated values collected in Table 3
show t h a t the IR contribution to the H a m a k e r constant is significant for
BaTiO 3 in all media; for SiO 2 the IR contribution is negligible except in
t h e case of n-dodecane w h e r e the Cuv and O~v of the m e d i u m and the
m a t e r i a l are very similar resulting in a very low UV contribution to the
n o n - r e t a r d e d H a m a k e r constant. Hence, for a m a t e r i a l characterised by
a relatively low value of CIR , like most ceramic materials, the IR
contribution to the H a m a k e r constant is very small, typically only a few
percent, except in the case w h e n the i n t e r v e n i n g m e d i u m has a similar
dielectric response function in the UV r a n g e as the material.
In the case of a m a t e r i a l characterised by a high Ciz and sufficiently
h i g h ¢OiR, e.g. BaTiO 3, the IR contribution is significant irrespective of
 L. Bergstr6m/Adv. Colloid Interface Sci, 70 (1997) 125-169 151

Table 3

N o n - r e t a r d e d H a m a k e r constants for t e t r a g o n a l BaTiO 3 and silica i n t e r a c t i n g across

v a c u u m , w a t e r and n-dodecane at room t e m p e r a t u r e (298 K)

Material Medium

Vacuum Water n-dodecane

B aT i O 3
com (1014 rad/s) H a m a k e r constant (10 .2o J)
- 14.1 4.45 3.77
0.7 17.6 7.91 7.44
1.0 19.3 9.49 8.99

SiOe
~om (10 TM rad/s) H a m a k e r constant (10 .2o J)
- 6.35 0.48 0.097
See Table 2 (3 IR 6.50 0.46 0.14
contributions)

t h e dielectric properties of the m e d i u m . With w a t e r or n-dodecane as

t h e i n t e r v e n i n g media, t h e IR contribution a t t r i b u t e s for m o r e t h a n 50%
of t h e total v a l u e of the n o n - r e t a r d e d H a m a k e r constant, t h u s m a k i n g
an a c c u r a t e dielectric r e p r e s e n t a t i o n of the i n f r a r e d r a n g e i m p o r t a n t .
In t h e case of BaTiO 3, Wemple et al. [70] r e p o r t e d room t e m p e r a t u r e
dielectric c o n s t a n t s m e a s u r e d at 100 kHz of: Ea = 3600, ec = 150. The
a v e r a g e static dielectric c o n s t a n t was e s t i m a t e d as eaver = 2400. Busca
et al. [74] h a v e studied the IR characteristics of a t e t r a g o n a l BaTiO 3
powder. T h e y found t h a t t h e IR s p e c t r u m h a s two d o m i n a n t complex
absorptions: one in the region 600-500 cm -1 a n d the o t h e r in the region
450-350 cm -1. T h e s e absorptions correspond to (~IR at 1.0 × 10 -14 a n d
0.7 × 10 -14 rad/s, respectively. With t h e u n c e r t a i n t y in the IR absorption
frequency, t h e e r r o r in t h e H a m a k e r c o n s t a n t becomes relatively large.
A r e a s o n a b l e e s t i m a t e of t h e n o n - r e t a r d e d H a m a k e r c o n s t a n t for
t e t r a g o n a l BaTiO 3 in v a c u u m is 18 + 1.5 × 10 -20 J, across w a t e r Alw I =
8 _+ 1.5 × 10 -20 J, a n d across n-dodecane A = 7.5 + 1.5 × 10 -20 J. Hence,
for such a m a t e r i a l w i t h a significant IR contribution, the e r r o r in the
H a m a k e r c o n s t a n t calculation becomes relatively large (10-20%) using
t h e p r e s e n t r e p r e s e n t a t i o n . The a c c u r a c y of the calculations m a y be
i m p r o v e d using a m o r e detailed IR r e p r e s e n t a t i o n .
152 L. BergstrOm/Adv. Colloid Interface Sci. 70 (1997) 125-169

v$

¢g

~D

¢9

~9

z
 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169 153

~3

~9

0
~9
~9
~9

~D

~9
154 L. BergstrSm /Adv. Colloid Interface Sci, 70 (1997) 125-169

VB. Full Lifshitz calculations and analytical approximations

N o n - r e t a r d e d H a m a k e r c o n s t a n t s w e r e calculated for all t h e s y m m e t -

ric s y s t e m s , m a t e r i a l 1 i n t e r a c t i n g across a m e d i u m 3, u s i n g full Lifshitz
c a l c u l a t i o n s , Eq. (19), a c r o s s v a c u u m a n d w a t e r . N o n - r e t a r d e d
H a m a k e r c o n s t a n t s w e r e also e s t i m a t e d u s i n g two different a p p r o x i m a -
tions, Eq. (21) for t h e H a m a k e r c o n s t a n t in v a c u u m ( T a b o r - W i n t e r t o n
e q u a t i o n i n c l u d i n g t h e static c o n t r i b u t i o n (TWS)), a n d Eq. (22) t o g e t h e r
w i t h t h e static t e r m in Eq. (21) ( H o r n - I s r a e l a c h v i l i (HIS) e q u a t i o n ) for
t h e H a m a k e r c o n s t a n t across water. T h e calculated v a l u e s are collected
in Table 4. T h e n o n - r e t a r d e d H a m a k e r c o n s t a n t s in v a c u u m , Alvl, v a r y
b e t w e e n 296 zJ (1 zJ = 10 -21 J) for d i a m o n d down to 40.5 zJ for N a F .
Most of t h e h a l i d e s h a v e relatively low n o n - r e t a r d e d H a m a k e r c o n s t a n t s
while t h e m o r e covalently b o n d e d oxides, carbides a n d n i t r i d e s h a v e
significantly h i g h e r values. T h e TWS a p p r o x i m a t i o n gave a s u r p r i s i n g l y
good c o r r e s p o n d e n c e to t h e full Lifshitz calculations. It s h o u l d be ob-
s e r v e d t h a t for t h e s e calculations, t h e a p p r o p r i a t e O>~v a n d n o for e a c h
m a t e r i a l (from Table 2) w e r e used. T h e a v e r a g e error for t h e 31 i n o r g a n i c
m a t e r i a l s w e r e only =3%, except for BaTiO 3 w h e r e t h e TWS a p p r o x i m a -
tion p r e d i c t s a v a l u e 23% too low. T h e m a i n difference b e t w e e n t h e TWS
a p p r o x i m a t i o n a n d t h e full Lifshitz calculation is t h a t t h e TWS approxi-
m a t i o n only considers t h e c o n t r i b u t i o n from t h e UV-vis region, h e n c e
i g n o r i n g t h e IR c o n t r i b u t i o n to e(i~), a n d t h a t t h e double s u m m a t i o n in
Eq. (19) is a p p r o x i m a t e d w i t h an analytical expression. T h e good agree-
m e n t b e t w e e n t h e full Lifshitz a n d t h e TWS a p p r o x i m a t i o n s u p p o r t s t h e
p r e v i o u s s t a t e m e n t t h a t t h e IR c o n t r i b u t i o n to t h e n o n - r e t a r d e d
H a m a k e r c o n s t a n t is negligible for m o s t ceramic m a t e r i a l s in v a c u u m
or air. This is a n e n c o u r a g i n g r e s u l t since it w a s difficult to obtain
d e t a i l e d i n f o r m a t i o n of t h e IR absorption b e h a v i o u r for s o m e of t h e
m a t e r i a l s . A s m a l l e r error in O~IR a n d C m s h o u l d n o t r e s u l t in a n y
s i g n i f i c a n t c h a n g e in t h e e s t i m a t e d v a l u e of A i r 1.
W i t h w a t e r as t h e i n t e r v e n i n g m e d i u m , t h e r e is a m u c h l a r g e r s p r e a d
in t h e n o n - r e t a r d e d H a m a k e r c o n s t a n t s , Alwl; from 3.1 zJ for N a F to
138 zJ for d i a m o n d . All t h e s e calculation w e r e p e r f o r m e d u s i n g t h e n e w
s p e c t r a l r e p r e s e n t a t i o n of w a t e r [13], see Table 1. In general, t h e
n o n - r e t a r d e d H a m a k e r c o n s t a n t s p r e s e n t e d h e r e are significantly lower
t h a n e a r l i e r e s t i m a t e s u s i n g t h e p r e v i o u s , less exact, U V r e p r e -
s e n t a t i o n s of w a t e r . For example, t h e n o n - r e t a r d e d H a m a k e r c o n s t a n t s
across w a t e r H o u g h a n d W h i t e [10] calculated for silica (8.49 zJ), calcite
(22.3 zJ), c a l c i u m fluoride (10.4 zJ) a n d s a p p h i r e (53.2 zJ) are u p to two
 L. BergstrSm /Adv. Colloid Interface Sci. 70 (1997) 125-169 155

times larger t h a n the values calculated in this study. Similarly, the

n o n - r e t a r d e d H a m a k e r constants across w a t e r p r e s e n t e d by BergstrSm
et al. [6] for nine different inorganic m a t e r i a l s are significantly larger
t h a n the p r e s e n t values. One m a t e r i a l w h e r e the m a g n i t u d e of the vdW
interaction is of special importance is muscovite mica; an extensively
used m a t e r i a l in direct surface force m e a s u r e m e n t s [1,22,75]. Com-
m o n l y used values for Alwl(mica) = 20 + 2 zJ [1,26] are almost 50%
h i g h e r compared to the p r e s e n t value OfAlwl(mica) = 13.4 zJ. A lowering
of the H a m a k e r c o n s t a n t for this system m a y have i m p o r t a n t implica-
tions r e g a r d i n g the i n t e r p r e t a t i o n of the obtained force curves.
T h e r e is a relatively poor correspondence between the full Lifshitz
calculations and the HIS approximation across water. The HIS approxi-
mation, w h i c h utilises only one value for COuvand Cuv for the m a t e r i a l
and the m e d i u m , respectively, appears to overestimate the H a m a k e r
c o n s t a n t across w a t e r for almost all the m a t e r i a l s while the TWS
a p p r o x i m a t i o n u n d e r e s t i m a t e s the n o n - r e t a r d e d H a m a k e r constant in
v a c u u m . Note t h a t the TWS approximation can not be used for calcula-
tions w i t h an i n t e r v e n i n g m e d i u m since the i n h e r e n t a s s u m p t i o n of a
c o m m o n ekyv for the m a t e r i a l and the m e d i u m n e v e r holds. D e p e n d i n g
on the spectral p a r a m e t e r s of the materials, the HIS approximation
o v e r e s t i m a t e s the n o n - r e t a r d e d H a m a k e r constant w i t h almost a factor
of two. While the u n d e r e s t i m a t i o n in the TWS case can be explained by
t h e absence of the IR r a n g e contribution, the overestimation using the
HIS approximation across w a t e r is probably due to the poor spectral
r e p r e s e n t a t i o n of w a t e r using only one UV oscillation. For m e d i a with
simpler spectral features, e.g. dodecane, a very good correspondence
b e t w e e n the full Lifshitz calculations and the HIS approximation is
obtained (not shown).

VC. H a m a k e r constants for use in A F M studies

The use of atomic force microscopy (AFM) to probe the distance

d e p e n d e n t forces between various m a t e r i a l s s e p a r a t e d by air, v a c u u m
or liquids h a v e recently gained m u c h interest. Initially, the forces
b e t w e e n the cantilever tip - - typically m a d e of silicon nitride - - and a
substrate, commonly mica, was monitored and m a n i p u l a t e d [75-79].
This w o r k has been extended by m e a s u r i n g the force curves between
silicon nitride tips and s u b s t r a t e in liquids at different conditions
[26,80,81], a n d a t t a c h i n g colloidal spheres of different m a t e r i a l s to the
AFM cantilevers [82-85]. These developments have made AFM a versatile
156 L. Bergstriirn /Adv. Colloid Interface Sci. 70 (1997) 125-169

tool for studying the complex interactions between surfaces of different

materials.
To assist the interpretation of AFM force measurements, the non-re-
tarded Hamaker constants of materials interacting across vacuum (air)
and water against four materials: silica, amorphous silicon nitride,
s a p p h i r e (a-A1203), a n d m u s c o v i t e m i c a h a v e b e e n c a l c u l a t e d ( T a b l e 5).
These materials are commonly used either as cantilever tips (silicon
n i t r i d e ) , c o l l o i d a l p r o b e s ( s i l i c a ) or i n o r g a n i c s u b s t r a t e s ( s a p p h i r e , m i c a )
in AFM studies. In general, the magnitude of the Hamaker constants
follows the expected trends with a large value for high refractive index

Table 5
Non-retarded H a m a k e r constants in vacuum and across water (vacuum/water) of
inorganic materials interacting against four materials at room temperature

H a m a k e r constant (10 -2oJ)

Material Silica Silicon nitride Alumina Mica

BaTiO 3 10.1/0.62 16.5/4.84 15.2/3.55 12.4/1.98

BeO 9.67/0.95 15.4/3.87 14.8/3.50 1L9/2.06
C (diamond) 13.7/1.71 22.0/7.94 21.117.05 17.0/4.03
CaCO 3 8.07/0.69 12.9/2.53 12.3/2.17 9.94/1.35
CaF 2 6.70/0.45 10.6/1.17 10.3/1.10 8.26/0.73
CdS 8.03/0.72 13.1/3.12 12.0/2.15 9.86/1.43
KC1 5.94/0.37 9.53/0.73 9.00/0.51 7.3110.46
MgA1204 9.05/0.85 14.5/3.39 13.8/2.97 11.2/1.79
MgF 2 6.15/0,36 9.74/0.66 9.42/0.69 7.57/0.50
MgO 8.84/0.81 14.2/3.26 13.5/2.79 10.9/1.69
Mica 8.01/0.69 12.8/2.45 12.2/2.15 9.86/1.34
NaC1 6.45/0.44 10.3/1.17 9.77/0.88 7.93/0.66
PbS 5.37/-0.08 8.88/0.64 7.90/-0.20 6.57/-0.03
6H-SiC 12.6/1.52 20.3/7.22 19.2/6.05 15.5/3.54
~-Si~N 4 10.8/1.17 17.3/5.13 16.5/4.43 13.3/2.61
SiO 2 (quartz) 7.59/0.63 12.112.07 11.6/1.83 9.35/1.16
SrTiO~ 9.44/0.57 15.4/4.02 14.2/2.98 11.6/1.69
TiO 2 9.46/0.69 15.4/4.26 14.2/3.11 11.6/1.83
Y20:~ 9.24/0.89 14.9/3.80 14.0/3.11 11.4/1.89
ZnO 7.38/0.58 12.0/2.30 11.111.58 9.06/1.08
ZnS (cubic) 9.69/1.02 15.7/4.56 14.6/3.55 11.9/2.19
3Y-ZrO 2 11.4/1.25 18.4/5.89 17.4/4.95 14.112.89
 L. Bergstr6m /Adv. Colloid Interface Sci. 70 (1997) 125-169 157

m a t e r i a l like silicon nitride and a lower value for a low refractive index
m a t e r i a l like silica. In vacuum, the n o n - r e t a r d e d H a m a k e r constants
span is relatively narrow: from 53.7 zJ (PbS against silica) to 220 zJ
( d i a m o n d a g a i n s t silicon nitride). With w a t e r as the i n t e r v e n i n g me-
dium, the m a g n i t u d e of the v a n der Waals attraction is of course r e d u c e d
and also b r o a d e n e d d e p e n d i n g on the material: from 3.7 zJ (KC1 a g a i n s t
silica) to 79.4 zJ (diamond against silicon nitride).
In fact, negative n o n - r e t a r d e d H a m a k e r constants, indicating a re-
pulsive v a n der Waals interaction, w e r e calculated for several m a t e r i a l
combinations involving PbS across water. A negative H a m a k e r c o n s t a n t
r e p r e s e n t s a destructive electrodynamic interaction between the mate-
rials and it is only recently t h a t direct m e a s u r e m e n t s of repulsive van
der Waals forces w e r e d e m o n s t r a t e d [86,87]. In general, repulsive van
der W a a l s forces are predicted for systems w h e r e the m a g n i t u d e of the
dielectric response of the m e d i u m is i n t e r m e d i a t e to the m a g n i t u d e of
t h e dielectric response of the two different i n t e r a c t i n g materials. For
PbS, h a v i n g the spectral characteristics of a very high e(0), e x t r e m e l y
h i g h n o (or Cur), and a very low muv, the calculations against t h e four
m a t e r i a l s across w a t e r (Table 5) display several i n t e r e s t i n g features.
The static contribution (m = 0) is negative (repulsive) for all these
m a t e r i a l combinations since ~(0) of the m e d i u m (water) is i n t e r m e d i a t e
to ~(0) for PbS and the materials. Since the static contribution u s u a l l y
is of m i n o r importance, it is of interest to analyse the dispersive contri-
bution in some detail. However, it should be noted t h a t the spectral
p a r a m e t e r s in the UV-visible spectral r a n g e for PbS relies on relatively
sparse data, t h u s m a k i n g the spectral r e p r e s e n t a t i o n less accurate. As
i l l u s t r a t e d in Fig. 7 for two of the m a t e r i a l s combinations: PbS against
silica, a n d PbS a g a i n s t silicon nitride, the dispersive contribution
changes sign from positive to negative above a critical frequency. Hence,
in the i n f r a r e d and visible p a r t of the s p e c t r u m (m < 25), the interaction
is c o n s t r u c t i v e while t h e contribution from the ultraviolet region
(m > 25) is destructive. In the system silica against PbS, this balance is
v e r y delicate r e s u l t i n g in a v e r y small, positive dispersive contribution
(0.2 zJ) to the n o n - r e t a r d e d H a m a k e r constant w h e n the contributions
from all the frequencies h a v e been s u m m a t e d . Since t h e static contri-
bution (-1.0 zJ) d o m i n a t e s the total van der Waals interaction for this
system, the calculations suggests t h a t screening this interaction, e.g. by
addition of salt [2J, would result in a system with a negligible vdW
attraction.
158 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169

(a) 400 I ' ' ' ' I

i i , i I L i i , 5000

Silica against PbS

A~w2=-0.8 zJ
300 4000
c
E O

\
,~ 200 3000
iI \
Q.
\ c
0
c-
o
O
'~ 100 2000
\V-
,I1 oN

L_

c-
\\ _

o 0 1000 E
O
O

-100 , , , I , , , i 0

0 100 200 300 400

m

(b) 30 ' ' " i . . . . I . . . . , . . . . 400

SiaN 4 against PbS o~"

Alw2= 6.4 zJ "-'c
E 20 / \\ 300 .o
.r.
\\ c
g lO -" 20o

C
0
j
1 0 0 "~
o E
o
o
. . . . I , , , I . . . . I . . . . 0
-10

0 100 200 300 400

Fig. 7. Contributions from i n d i v i d u a l t e r m s in Eq. (19) of the dispersive p a r t (m > 0) to

the n o n - r e t a r d e d H a m a k e r constants, Alw2 for PbS a g a i n s t (a) silica, and (b) silicon
nitride, across water. The static contribution was - 1 . 0 zJ, a n d - 0 . 9 zJ for (a) a n d (b),
respectively.
 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169 159

VD. Spectral parameter scalings

With the absence of detailed spectroscopic data, we are limited to t h e

simplified expressions for calculating n o n - r e t a r d e d H a m a k e r constants.
The TWS a n d HIS approximations are examples of such simplifications.
However, even for t h e simple TWS approximation of the n o n - r e t a r d e d
H a m a k e r c o n s t a n t in v a c u u m , we h a v e to know t h r e e p a r a m e t e r s : okrv,
n 0, a n d e(0). Usually, values of the refractive index in the visible a n d the
static dielectric c o n s t a n t can be found for most m a t e r i a l s b u t O)uv is
u s u a l l y not t a b u l a t e d a n d h a s to be e x t r a c t e d from d a t a of t h e f r e q u e n c y
d e p e n d e n c e of t h e refractive index which m a y be h a r d to find. W i t h o u t
such data, t h e a s s u m p t i o n originating from Tabor and W i n t e r t o n [22]
of a common value of ¢Ouv = 2 x 10 i6 rad/s for all materials is frequently
used.
F i g u r e 8 illustrates how a common value ofa)cv = 2 x 1016 rad/s results
in a s y s t e m a t i c o v e r e s t i m a t i o n of t h e n o n - r e t a r d e d H a m a k e r c o n s t a n t s
for m a n y of t h e inorganic materials. For simplicity, the original T a b o r -
W i n t e r t o n (TW) a p p r o x i m a t i o n was used, t h u s not including t h e static
contribution w h i c h is negligible for vdW interactions across v a c u u m .
The deviation of t h e Lifshitz value for A i r i from this simplified TW

5 0 ' ' ' ' I ' ' ' ' I ' ' ~ ~ I ~ / ' I ' ' ' j ' ' ' ' I ' ' ' ' I ' '

~"
40

30
/
/.c
/, oSiC
'; 20
¢'.- ,' . s,o
_/~ • CdS
10 ~' o • Z n O • PbS
11 Csl

0 ,Ji,, I , ~ , ~I, , , ill j j j I, ,, , , , , , I , , , , I , ,

1 2 3 4 5
no
Fig. 8. Non-retarded Hamaker constants for inorganic materials across vacuum (Airl)
versus the limiting refractive index, no, from Table 2. Solid line denotes the TW
approximation with ¢Ovv=2x101Grad/s.
160 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169

a p p r o x i m a t i o n i n c r e a s e s w i t h i n c r e a s i n g refractive index. This agrees

w i t h previous s t u d i e s [7,88] w h i c h concluded t h a t t h e T W a p p r o x i m a -
tion r e s u l t s in poor c o r r e s p o n d e n c e to full Lifshitz calculations for h i g h
refractive index m a t e r i a l s . However, this does n o t reflect a n y significant
i n h e r e n t errors in t h e TWS a p p r o x i m a t i o n s ; it m e r e l y r e l a t e s to a poor
choice of ¢Ouv. If t h e a p p r o p r i a t e v a l u e s of t h e spectral p a r a m e t e r s , in
p a r t i c u l a r Ouv, are used, t h e r e s u l t s in Table 4 clearly show t h a t t h e
TWS a p p r o x i m a t i o n gives quite a c c u r a t e r e s u l t s w h e n c o m p a r e d to full
Lifshitz calculations for m o s t m a t e r i a l s .
T h e r e l a t i o n b e t w e e n t h e refractive index, no, a n d ¢Ovv is s h o w n in
Fig. 9. All t h e m a t e r i a l s display a d e c r e a s i n g mvv w i t h an i n c r e a s i n g
refractive i n d e x w h i c h can explain t h e i n c r e a s i n g d e v i a t i o n in t h e v a l u e
ofAlv 1 b e t w e e n t h e full Lifshitz calculations a n d t h e TW a p p r o x i m a t i o n
u s i n g a c o n s t a n t v a l u e of O)uv. An a t t e m p t w a s m a d e to e s t a b l i s h a
g e n e r a l scaling relation for all t h e s e m a t e r i a l s . D u e to t h e large s c a t t e r
of t h e d a t a in Fig. 9a, however, it w a s n o t possible to describe all t h e
d i f f e r e n t m a t e r i a l s w i t h s u c h a scaling relation. If we s o m e w h a t arbi-
t r a r i l y limit t h e d a t a to covalent oxides, s u l p h i d e s a n d nitrides, t h u s
e x c l u d i n g t h e ionic halides, a n d SiC a n d d i a m o n d , a s i m p l e scaling
r e l a t i o n at h i g h e r refractive indices, following t h e power law

8.8 × 1016
O)uv - n2.S (26)

m a y be e s t a b l i s h e d , a l t h o u g h a significant scatter of t h e d a t a still

r e m a i n s (Fig. 9b). F i g u r e 9b indicates t h a t for covalent m a t e r i a l s w i t h
a low refractive i n d e x (n o < 1.7), O)cv a p p e a r s to be i n d e p e n d e n t o f n 0. A
s i m p l e scaling b e t w e e n t h e refractive index in t h e visible a n d t h e
n o n - r e t a r d e d H a m a k e r c o n s t a n t h a s been i l l u s t r a t e d previously by
Buscall [68] for m a t e r i a l s such as polymers, h y d r o c a r b o n s , q u a r t z a n d
calcite. T h e success of this scaling can be referred to t h e n e a r l y identical
Ouv of t h e s e low refractive index m a t e r i a l s . Buscall also concluded t h a t
s u c h a correlation is n o t applicable w h e n Ovv deviates s u b s t a n t i a l l y
f r o m this c o m m o n value, as observed w i t h rutile, a h i g h refractive index

Opposite: Fig. 9. Relation between the characteristic frequency in the UV range, ¢Otrv,
and the refractive index in the visible, no: (a) all the inorganic materials, (b) only the
covalent materials thus excluding the halides. The high refractive index materials,
diamond and silicon carbide have also been excluded in (b) where the full line represents
the power law fit according to Eq. (26).
 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169 161

(a)
3 1016 . . . . I . . . . I . . . . I ' ' '

• Fluorides
• BeO

2 1016

'0 • C
% •
im

>
Halides• • SiC
1 1016
ZnO • 11~
• • TiO 2

• PbS
0 , , , , I . . . . I , , , I . . . .

1 2 3 5
n
0

(b) 1 017 ' ' ' ' 1 ' ' ' ' 1 ' ' ' ' 1 ' '

A ,%

"1o
I,.
1 016

I I I I I m l l l l l | l l l l J i
1 0 is
2 3 4

no
162 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169

material (n = 2.46). In this work, however, no attempt to implement the

scaling in Eq. (26) in the TWS approximation was made due to the large
scatter of the data.

VE. Comparison with a Kramers-Kronig dielectric representation

French and co-workers [7,88] have recently published several papers

where the non-retarded H a m a k e r constants of ceramic materials have
been calculated using a slightly different scheme applying the integral
form of the H a m a k e r constant expression (Eq. (20)). The imaginary
dielectric representation, ~(i~), of the materials were obtained through
K r a m e r s - K r o n i g transformations of experimentally determined inter-
band transition strengths over a relatively wide photon energy range
(0.46 to 44 eV) from optical reflectivity spectra. Non-retarded H a m a k e r
constants for different combinations were calculated using a commer-
cially available collection of the necessary programs and a spectral data
base of m a n y materials. As previously stated, the accuracy of H a m a k e r
constants calculated from Lifshitz theory relies on the quality of the
experimental dielectric data and the accuracy of the spectral repre-
sentation. Since French and co-workers [7,88] use dielectric data over a
wide frequency range, their method has the potential of obtaining a very
detailed spectral representation in the important frequency range. In
principle, the N - P representation has the same potential if a sufficient
n u m b e r of relaxations is introduced. Of course, accurate dielectric
representations can only be obtained from accurate dielectric data.
Ackler et al. [88] recently presented data on non-retarded H a m a k e r
constants calculated by the method described above, using the integral
equation and Kramers-Kronig relations (IKK method), and compared
their values with previously published values where the method in this
study, the summation equation (Eq. (19)) with an N - P dielectric repre-
sentation (SNP method) was used. Also, other simplified models were
used for 5 different materials: muscovite mica, A1203, SiO2, Si3N4, and
rutile TiO2, separated by vacuum and water. In the IKK calculations
[88], the dielectric representation for the inorganic materials was ob-
tained from VUV reflectivity data as described above while the a(i~)
function for water was determined from literature data [12] and K - K
transformed to be incorporated in the data base.
Here, their values for the non-retarded H a m a k e r constants using the
IKK method are compared with values using the SNP method (Table 6)
in an attempt to assess the accuracy of representing the materials with
 L. Bergstr6m /Adv. Colloid Interface Sci. 70 (1997) 125-169 163

Table 6
C o m p a r i s o n of calculated n o n - r e t a r d e d H a m a k e r c o n s t a n t s using two different meth-
ods: the integral Kramers-Kronig (IKK) method and the summation Ninham-
Parsegian (SNP) method

Material H a m a k e r constant (10 -20 J)

Vacuum Water

I K K [881 SNP I K K [88] SNP SNP-UV*

Mica 6.96 9.86 0.29 1.34 1.08

A1203 14.5 15.2 2.75 3.67 3.48
SiO 2 6.60 6.50 0.16 0.46 0.17
Si3N 4 17.4 18.0 4.6 5.47 5.22

TiO 2 r u t i l e 17.3 15.3 6.0 5.35 5.02

* S N P - U V calculations were performed using Eq. (19) including only the UV contribu-
tion of the ~(i~m) function of both the m a t e r i a l s a n d water.

t h e simplified, two oscillator, N - P r e p r e s e n t a t i o n . Observe t h a t t h e

v a l u e s for t h e n o n - r e t a r d e d H a m a k e r c o n s t a n t s u s i n g t h e S N P m e t h o d
differs in this s t u d y a n d in Ackler et al. [32] since a n e w w a t e r repre-
s e n t a t i o n is u s e d here. W i t h v a c u u m as t h e i n t e r v e n i n g m e d i a , t h e r e is
a v e r y good c o r r e s p o n d e n c e of t h e calculated v a l u e s ofAlv I for t h r e e of
t h e m a t e r i a l s : a l u m i n a , silica, a n d silicon nitride, w i t h an a v e r a g e
difference of only 3%. For mica, t h e I K K m e t h o d yields a significantly
lower v a l u e while t h e IKK m e t h o d yields a 12% h i g h e r v a l u e of t h e
H a m a k e r c o n s t a n t for rutile. T h e difference in t h e c a l c u l a t e d v a l u e s
b e t w e e n t h e two m e t h o d s is i n c r e a s e d significantly for t h e n o n - r e t a r d e d
H a m a k e r c o n s t a n t s across water.
T h e r e a r e two m a i n differences b e t w e e n t h e I K K a n d t h e S N P
m e t h o d s . In t h e I K K m e t h o d , t h e i n t e g r a l e q u a t i o n for t h e n o n - r e t a r d e d
H a m a k e r c o n s t a n t is u s e d (Eq. (20)) w h i c h ignores t h e static contribu-
tion. T h e S N P m e t h o d uses t h e double s u m m a t i o n e q u a t i o n w h i c h
i n c l u d e s this c o n t r i b u t i o n , r e f e r r i n g to t h e first s u m m a t i o n (m = 0) in
Eq. (19). T h e I K K m e t h o d typically h a s an i m a g i n a r y dielectric repre-
s e n t a t i o n over a significantly b r o a d e r f r e q u e n c y r a n g e , in p a r t i c u l a r t h e
h i g h f r e q u e n c y r a n g e , c o m p a r e d to t h e simplified N - P r e p r e s e n t a t i o n
u s i n g only one UV a n d possibly a few IR t e r m s (Eq. (14)). T h e I K K
164 L. BergstrSm /Adv. Colloid Interface Sci. 70 (1997) 125-169

method, however, also ignores the IR contribution at p r e s e n t [7]. It is of

i n t e r e s t to e v a l u a t e the effect and m a g n i t u d e of these two differences
b e t w e e n the I K K and SNP method. Unfortunately, such a comparison
is complicated by the fact t h a t dielectric d a t a of different origin was used
for most materials. Hence, the difference in the calculated values of the
n o n - r e t a r d e d H a m a k e r constants can be caused by either the difference
in m a t h e m a t i c a l t r e a t m e n t (integral and summation), the accuracy of
the original dielectric d a t a (frequency d e p e n d e n c e of the refractive index
or i n t e r b a n d transition strengths) and using an appropriate i m a g i n a r y
dielectric representation.
I g n o r i n g the static and low frequency contribution of the a(i~) repre-
s e n t a t i o n will only have a m a r g i n a l effect on the n o n - r e t a r d e d H a m a k e r
c o n s t a n t for m a t e r i a l s i n t e r a c t i n g across v a c u u m since the dispersive
p a r t completely d o m i n a t e s the value of Air 1. The static contribution is
u s u a l l y of m i n o r importance in all systems since its m a x i m u m v a l u e is
0.75 k T (4.1 zJ at room t e m p e r a t u r e ) but it can h a v e a significant
influence at high t e m p e r a t u r e s or w h e n the dispersive p a r t of the
H a m a k e r c o n s t a n t is very small. The significance of the IR contribution
was discussed previously and no major influence is expected except for
m a t e r i a l s w i t h very high values of the static dielectric constant. Hence,
since no major differences are expected using the integral or the sum-
m a t i o n equations (Eqs. (20) and (19), respectively) for low a(0) m a t e r i a l s
in v a c u u m , the n e a r l y identical values of A i r 1 for A1203, SiO 2 and Si3N 4
suggests t h a t the dielectric r e p r e s e n t a t i o n s are similar in the two
studies. This is supported by the close correspondence of the index of
refraction (An < 0.05) for these m a t e r i a l s in Ackler et al. [32] and in this
work. For rutile, the h i g h e r A l v 1 using the I K K m e t h o d m a y be explained
by a difference in the original dielectric data, Ackler et al. [32] reporting
n = 2.6 while this study uses n = 2.46 based on d a t a from Buscall [68].
For mica, a similar explanation m a y apply, or the significantly lower
Alv 1 v a l u e using the I K K m e t h o d m a y be related to the fine details of
t h e a(i~) r e p r e s e n t a t i o n over a wider frequency range.
T u r n i n g to the n o n - r e t a r d e d H a m a k e r constants across w a t e r , t h e r e
is a significant difference in Alw 1 b e t w e e n the two m e t h o d s for all
materials. The I K K m e t h o d always yields lower values compared to the
S N P method. In water, the dispersive contribution becomes very small
for some materials, t h u s m a k i n g the static and low frequency contribu-
tion i m p o r t a n t . For silica, the result in Table 6 shows t h a t the difference
b e t w e e n t h e two methods can be fully related to the omission of this
contribution in the I K K method. For all the other materials, except mica,
 L. Bergstr6m /Adv. Colloid Interface Sci. 70 (1997) 125-169 165

the difference between the IKK and SNP-UV (omitting static and IR
contribution) was less than 30%. The large difference in Alw I for mica,
almost a factor of four, must be related to major differences in the
imaginary dielectric representation, either due to major differences in
the original dielectric data or to the capability of the IKK method to yield
a detailed e(i~) representation over a wide frequency range. In summary,
a comparison of these two methods suggests that the simplified N - P
representation over a limited frequency range in general yields non-re-
tarded H a m a k e r constants which do not differ significantly from calcu-
lations using a more detailed representation over a broader frequency
range. This is not true for all materials, e.g. water, where a more detailed
representation using either an N - P representation with several oscilla-
tors or the K - K representation. It is possible that mica also is such a
material.

V/. S u m m a r y a n d c o n c l u s i o n s

Calculation of non-retarded H a m a k e r constants from Lifshitz theory

relies on the accuracy of the dielectric spectral representation and the
quality of the experimental dielectric data of the materials and media.
In this review, spectral parameters for 31 different inorganic materials
(including diamond) have been generated from critically evaluated
optical data [11,12] or collected from different sources. The spectral
parameters: the oscillator strengths and characteristic frequencies in
the UV and IR range; Cur, tour, and Cm, ~ a , respectively, and the static
dielectric constant, a(0), for the different inorganic materials generated
here can be combined with previous data, mainly focused on hydrocar-
bon and organic systems [10], to yield an extensive spectral data base
for both solids and liquids.
Non-retarded H a m a k e r constants across vacuum and water have
been calculated using Lifshitz theory for the different inorganic mate-
rials (including diamond). All the symmetric combinations (material 1
against material 1) have been treated and the asymmetric combinations
with most of the materials against silica, silicon nitride, alumina and
mica. These calculations were performed using the full Lifshitz theory
utilising the N i n h a m - P a r s e g i a n representation of the imaginary dielec-
tric response function for the materials and medium. A recently reported
improved spectral parametric representation of water [13] was used for
these calculations which resulted in some significant changes in the
value for Alvd, e.g. for mica, compared to previous studies.
166 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169

T h e full Lifshitz calculations were c o m p a r e d w i t h v a r i o u s analytical

simplifications a n d a n a l t e r n a t i v e Lifshitz m e t h o d . W h e n t h e appropri-
a t e v a l u e s of o~y¢ a n d n o w e r e used, a s u r p r i s i n g l y good a g r e e m e n t w a s
f o u n d - - w i t h v a c u u m as t h e i n t e r v e n i n g m e d i u m - - b e t w e e n t h e full
t r e a t m e n t a n d t h e T a b o r - W i n t e r t o n (TWS) approximation. The corre-
spondence between t h e full Lifshitz t r e a t m e n t a n d the TWS approximation
s u g g e s t e d t h a t t h e IR frequency range of t h e dielectric r e p r e s e n t a t i o n is
of m i n o r i m p o r t a n c e for m o s t m a t e r i a l s . This w a s c o r r o b o r a t e d by a
s e p a r a t e e v a l u a t i o n s h o w i n g t h a t t h e IR c o n t r i b u t i o n to t h e non-re-
t a r d e d H a m a k e r c o n s t a n t is significant only for m a t e r i a l s , like BaTiO3,
w h i c h are c h a r a c t e r i s e d by a v e r y h i g h static dielectric c o n s t a n t .
T h e calculations in this s t u d y utilising t h e N i n h a m - P a r s e g i a n rep-
r e s e n t a t i o n of t h e dielectric d a t a (SNP m e t h o d ) w e r e c o m p a r e d w i t h a n
a l t e r n a t i v e m e t h o d i g n o r i n g t h e static a n d low f r e q u e n c y c o n t r i b u t i o n
a n d u s i n g t h e K r a m e r s - K r o n i g t r a n s f o r m a t i o n for t h e v i s i b l e - U V r a n g e
(IKK method). In general, t h e S N P m e t h o d u s i n g a simplified N - P
r e p r e s e n t a t i o n yields n o n - r e t a r d e d H a m a k e r c o n s t a n t s w h i c h do n o t
differ significantly from calculations u s i n g t h e I K K m e t h o d c o n t a i n i n g
a m o r e d e t a i l e d dielectric r e p r e s e n t a t i o n over a b r o a d e r f r e q u e n c y
r a n g e . However, this is n o t t r u e for all m a t e r i a l s , e.g. w a t e r , w h e r e a
m o r e d e t a i l e d r e p r e s e n t a t i o n u s i n g e i t h e r an N - P r e p r e s e n t a t i o n w i t h
s e v e r a l r e l a x a t i o n s or t h e K - K r e p r e s e n t a t i o n m u s t be used. It is
possible t h a t mica is also s u c h a m a t e r i a l w h i c h implies t h a t t h e m o r e
d e t a i l e d dielectric r e p r e s e n t a t i o n of t h e IKK m e t h o d s h o u l d yield a m o r e
a c c u r a t e v a l u e of t h e n o n - r e t a r d e d H a m a k e r c o n s t a n t for t h i s m a t e r i a l .
It s h o u l d be n o t e d t h a t t h e v a l u e s for Alw I calculated for mica u s i n g t h e
S N P (13.4 zJ) a n d t h e I K K m e t h o d (2.9 zJ) are significantly lower t h a n
e x p e r i m e n t a l l y d e t e r m i n e d v a l u e s u s i n g t h e surface force a p p a r a t u s (22
zJ). U s i n g silica as a n example, it w a s s h o w n t h a t t h e omission of t h e
static a n d low f r e q u e n c y c o n t r i b u t i o n in t h e I K K m e t h o d m a y r e s u l t in
a significant u n d e r e s t i m a t i o n of t h e v a l u e for Alw I w h e n t h e dispersive
c o n t r i b u t i o n becomes v e r y small.

Acknowledgements

T h e a u t h o r g r a t e f u l l y a c k n o w l e d g e s t h e S w e d i s h N a t i o n a l B o a r d for
I n d u s t r i a l a n d T e c h n i c a l D e v e l o p m e n t (NUTEK) for financial support.
A n d e r s M e u r k is t h a n k e d for s t i m u l a t i n g discussions.
 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169 167

References

[1] J. Israelachvili, Intermolecular and Surface Forces, 2nd edn. Academic Press,
London, 1992.
[2] W.B. Russel, D.A. Saville and W.R. Schowalter, Colloidal Dispersions. Cambridge
University Press, Cambridge, 1989.
[3] L. BergstrSm, in: R.J. Pugh and L. BergstrSm (eds.), Surface and Colloid Chemis-
try in Advanced Ceramics Processing. Marcel Dekker, New York, 1994, pp. 193.
[4] A. Bleier, J. Phys. Chem., 87 (1983) 3493.
[5] D.R. Clarke, J. Am. Ceram. Soc., 70 (1987) 15.
[6] L. BergstrSm, A. Meurk, H. Arwin, and D.J. Rowcliffe, J. Am. Ceram. Soc., 79
(1996) 339.
[7] R.H. French, R.M. Cannon, L. K. DeNoyer, and Y.-C. Chiang, Solid State Ionics,
75 (1995) 13.
[8] H.C. Hamaker, Physica, 4 (1937) 1058.
[9] E.M. Lifshitz, Sov. Phys. JETP, 2 (1956) 73.
[10] D.B. Hough and L.R. White, Adv. Colloid Interface Sci., 14 (1980) 3.
[11] E.D. Palik, Handbook of Optical Constants of Solids. Academic Press, Orlando,
FL, 1985.
[12] E.D. Palik, Handbook of Optical Constants of Solids II. Academic Press, New York,
1991.
[13] C.M. Roth and A.M. Lenhoff, J. Colloid Interface Sci., 179 (1996) 637.
[14] V.A. Parsegian and B.W. Ninham, Nature, 224 (1969) 1197.
[15] B.W. Ninham and V.A. Parsegian, Biophys J., 10 (1970) 646.
[16] H. Krupp, Adv. Colloid Interface Sci., 1 (1967) 111.
[17] H. Krupp, W. Schnabel and J. Walter, J. Colloid Interface Sci., 39 (1972) 421.
[18] J. Mahanty and B.W. Ninham, Dispersion Forces. Academic Press, London, 1976.
[19] D.C. Prieve and W.B. Russel, J. Colloid Interface Sci., 125 (1988) 1.
[20] V.A. Parsegian, in: H. van Olphen and K.J. Mysels (eds.), Physical Chemistry:
Enriching Topics from Colloid and Surface Science. Theorex, La Jolla, CA, 1975.
[21] J.M. Heller Jr., R.N. Hamm, R.D. Birkhoffand L.R. Painter, J. Chem. Phys., 60
(1974) 3483.
[22] D. Tabor and R.H.S. Winterton, Proc. R. Soc. London A, 312 (1969) 435.
[23] R.G. Horn and J.N. Israelachvili, J. Chem. Phys., 75 (1981) 1400.
[24] T. Shiosaki, in: A.S. Bhalla, E.M. Vogel and K.M. Nair (eds.), Ceramics Transac-
tions, Vol. 14, Electro-Optics and Nonlinear Optic Materials. American Ceramic
Society, Westerville, OH, 1990, pp. 197.
[25] K.F. Young and H.P.R. Friedrikse, J. Phys. Chem., Ref. Data, 2 (1973) 313.
[26] T.J. Senden and C.J. Drummond, Colloids Surfaces A: Physicochem. Eng. Aspects,
94 (1995) 29.
[27] D.F. Bezuidenhout, in Ref. [12], p. 815.
[28] I.H. Malitson, Appl. Opt., 2 (1963) 1103.
[29] W. Kaiser, W.G. Spitzer, R.H. Kaiser and L.E. Horwarth, Phys. Rev., 127 (1962)
1950.
[30] J.E. Eldridge, in Ref. [12], p. 853.
168 L. BergstrSm/Adv. Colloid Interface Sci. 70 (1997) 125-169

[31] H.H. Li, J. Phys. Chem. Refer. Data, 5 (1976) 329.

[32] J.A.B. Beairsto and J.E. Eldridge, Can. J. Phys., 51 (1973) 2550.
[33] E.D. Palik, in Ref. [12], p. 989.
[34] H.H. Li, J. Phys. Chem. Refer. Data, 5 (1976) 345.
[35] K.W. Johnson and E.E. Bell, Phys. Rev., 187 (1969) 1044.
[36] E.D. Palik, in Ref. [11], p. 703.
[37] E.D. Palik and W.R. Hunter, in Ref. [11], p. 675.
[38] H.H. Li, J. Phys. Chem. Ref. Data, 9 (1980) 561.
[39] A. Kachare, G. Anderman and L.R. Brantley, J. Phys. Chem. Solids, 33 (1972) 467.
[40] J.E. Eldridge and E.D. Palik, in Ref. [11], p. 775.
[41] J.E. Eldridge and P.R. Staal, Phys. Rev. B, 16 (1977) 4608.
[42] I. Ohlidal and K. Navratil, in Ref. [12], p. 1021.
[43] H. Harting, Stzungsber. Deutsch. Akad. Wiss. Ber., 4 (1948) 1.
[44] K.F. Pai, T.J. Parker, N.E. Tornberg, R.P. Lowndes and W.G. Chambers, Infrared
Physics, 18 (1978) 199.
[45] D.F. Edwards and H.R. Philipp, in Ref. [11], p. 665.
[46] F. Peter, Z. Phys., 15, (1923) 358.
[47] D.F. Edwards and E. Ochoa, J. Opt. Soc. Am., 71 (1981) 607.
[48] G. Guizzetti and A. Borghesi, in Ref. [11], p. 525.
[49] J.N. Zemel, J.D. Jensen and R.B. Schoolar, Phys. Rev. A, 140 (1965) 330.
[50] R. Geick, Phys. Lett., 10 (1964) 51.
[51] S.A. Alterovitz and J.A. Woollam, in Ref. [12], p. 705.
[52] W.J. Choyke and L. Patrick, J. Opt. Soc. Am., 58 (1968) 377.
[53] (a) W.G. Spitzer, D.A. Kleinman and C.J. Frosch, Phys. Rev., 113 (1959) 133.
(b) W.G. Spitzer, D.A. Kleinman, C.J. Frosch and D.J. Walsh, in: J.R. O'Connor
and J. Smiltens (eds.), Silicon Carbide, A High Temperature Semiconductor.
Pergamon, New York, 1960, pp. 347.
[54] F. Gervais, in Ref. [12], p. 1035.
[55] M. Cardona, Phys. Rev., 140 (1965) A651.
[56] W.G. Spitzer, R.C. Miller, D.A. Kleinman and L.E. Howarth, Phys. Rev., 126 (1962)
1710.
[57] E.D. Palik and A. Addamiano, in Ref. [11], p. 597.
[58] A.N. Pikhtin and A.D. Yaskov, Sov. Phys. Semicond., 12 (1978) 622.
[59] T. Deutsch, Proc. Int. Conf. Phys. Semicond., 6th Exeter, The Institute of Physics
and the Physical Society, London, 1962, pp. 505.
[60] L. Ward, in Ref. [12], p. 579.
[61] A.B. Francis and A.J. Carlsson, J. Opt. Soc. A., 50 (1960) 118.
[62] A. Manabe, A. Mitsuishi and H. Yoshinga, Jpn. J. Appl. Phys., 6 (1967) 593.
[63] T.M. Cotter, M.E. Thomas and W.J. Troph, in Ref. [12], p. 899.
[64] M.J. Dodge, Appl. Opt., 23 (1984) 1980.
[65] A.S. Barker, Phys. Rev., 136 (1964) A1290.
[66] V.C. Farmer, in: V.C. Farmer (ed.), The Infrared Spectra of Minerals. Mineralogi-
cal Society of London, UK, 1974, pp. 183.
[67] T.M. Bienewski and S.J. Czyzak, J. Opt. Soc. Am., 53 (1963) 496.
[68] R. Buscall, Colloids Surf. A, 75 (1993) 269.
[69] C. Wong, Y.Y. Teng, J. Ashok and P.L.H. Varaprasad, in Ref. [12], p. 789.
 L. Bergstr6m/Adv. Colloid Interface Sci. 70 (1997) 125-169 169

[70] S.H. Wemple, M. Didomenico, Jr. and I. Camlibel, J. Phys. Chem. Solids, 29 (1968)
1797.
[71] D.F. Edwards and R.H. White, in Ref. [12], p. 805.
[72] H.W. Newkirk, D. K. Smith, and J. S. Kahn, Am. Mineral., 51 (1966) 141.
[73] E. Loh, Phys. Rev., 166 (1968) 673.
[74] G. Busca, G. Ramis, J.M. Gallardo Amores, V.S. Escribano and P. Piaggio, J.
Chem. Soc. Faraday Trans., 90 (1994) 3181.
[75] P.M. Claesson, Forces between surfaces immersed in aqueous solutions. Thesis,
Royal Institute of Technology, Stockholm, Sweden, 1986.
[76] A.L. Weisenhorn, P.K. Hansma, T.R. Albrecht and C.F. Quate, Appl. Phys. Lett.,
54 (1989) 2651.
[77] U. Hartmann, Adv. Mater., 2 (1990) 594.
[78] J.L. Hutter and J. Bechhoefer, J. Vac. Sci. Technol. B, 12 (1994) 2251.
[79] J.L. Hutter and J. Bechhoefer, J. Appl. Phys., 73 (1993) 4123.
[80] X.-Y. Lin, F. Creuzet and H. Arribart, J. Phys. Chem., 97 (1993) 7272.
[81] T.J. Senden, C.J. Drummond and P. Kekicheff, Langmuir, 10 (1994) 358.
[82] W.A. Ducker, T.J. Senden and R.M. Pashley, Nature, 353 (1991) 239.
[83] I. Larson, C.J. Drummond, D.Y.C. Chan and F. Grieser, J. Am. Chem. Soc., 115
(1993) 11885.
[84] S. Biggs and P. Mulvaney, J. Chem. Phys., 100 (1994) 8501.
[85] S. Biggs and T.W. Healy, J. Chem. Soc. Faraday Trans., 90 (1994) 3415.
[86] A. Milling, P. Mulvaney and I. Larson, J. Colloid Interface Sci., 180 (1996) 460.
[87] A. Meurk, P. Luckham and L. BergstrSm, Langmuir, submitted.
[88] H.D. Ackler, R.H. French and Y.-M. Chiang, J. Colloid Interface Sci., 179 (1996)
460.