High refractive index films of polymer nanocomposites
Lorenz Zimmermann, Martin Weibel, Walter Caseri, and Ulrich W. Suter
Eidgenossische
Technische
Hochschule, [nstitut fur Polymere, ETH-Zentrum, CH-8092 Zurich,
Switzerland
(Received 6 July 1992; accepted 19 March 1993)
Solutions of PbS particles and gelatin were used for the preparation of nanocomposites
by a spin-coating process. This allows for the preparation of nanocomposite films with
controlled thickness, e.g., between 40 nm and 2 /mm for a film containing 45 wt. % PbS.
Surface roughness and film thickness were investigated by surface profilometry and
scanning electron microscopy (SEM). The refractive index at 632.8 nm can be expressed
by a linear function of the volume fraction of PbS in the range of 0 to 55 vol. %
PbS.
In this range, the refractive index increases from 1.5 to 2.5 with increasing PbS
ratio and belongs, therefore, to the highest refractive indices known for polymeric
composite materials.
I. INTRODUCTION
Composites between polymers and inorganic parti-
cles in the micrometer range are often opaque.
1
'
2
Light
scattering, responsible for the opacity, can be suppressed
either by using materials with nearly matching refractive
indices
3
or by decreasing the filler's dimensions to a
range below ca. 50 nm.
1
'
2
'
4
'
5
Therefore, nanocomposites
of polymers and inorganic colloids can act as opti-
cally homogeneous materials with modified optical
properties.
1
'
2
'
4
'
6
'
7
Nanocomposites containing lead sul-
fide, for instance, can act as optical filters,
4
nonlinear
optical,
6
or "ultrahigh" refractive index materials.
7
The refractive index of polymers is usually between
1.3 and 1.7.
8
Examples of higher refractive index are,
e.g., 2.12 for poly(thiophene)
9
and
1.7245
for [CH
2
-
CH = CHCH
2
N
+
(CH
2
CH3)2(CH
2
)
6
N
+
(CH
2
CH
3
)2]« •
n-Hgl4 .
10
Aromatic polyamides can exhibit (isotropic)
refractive indices up to 2.05
11
or 2.07.
12
The refractive index of polymers can be enhanced
by the addition of colloidal particles with high refractive
index,
7
e.g., PbS that exhibits (at least in crystalline
form) a refractive index on the order of 4 in a wide
wavelength range.
13
The highest refractive index for
polymeric materials, i.e., pure polymers or polymer
composites, has been obtained to date, to our best
knowledge, for nanocomposites of lead sulfide and
poly(ethyleneoxide).
7
They exhibit refractive indices of
up to 2.9 at 632.8 nm and of up to 3.0-3.4 in the
whole range of 1000—2500 nm. These nanocomposites
were prepared by adding a hydrogen sulfide solution to
a solution containing the polymer and Pb(n) ions. The
genesis of colloidal PbS particles and the precipitation
of the nanocomposite material take place simultaneously
and immediately after addition of hydrogen sulfide. This
method allows for the preparation of nanocomposites
filled to ca. 50 vol. % PbS; however, attempts to prepare
nanocomposites with a markedly lower PbS-content,
e.g., by variation of the initial Pb
2+
/pctly(ethyleneoxide)
ratio,
failed. Also, experiments aimed at preparing
nanocomposite films were inconclusive because of poor
film quality.
In this report, we describe the preparation of polymer
nanocomposite films with controlled layer thickness and
refractive index, starting from homogeneous solutions
containing colloidal PbS and a polymer.
II.
EXPERIMENTAL
A. Chemical reagents
Gelatin (from pork skin, high gel strength) and lead
acetate were obtained from Fluka (Buchs, Switzerland),
hydrogen sulfide from Pangas (Lucerne, Switzerland).
From gelatin, the following values were obtained by
microanalysis: C,
43.41%,
H, 6.64%, N, 16.25%, and
H
2
O,
5.20%.
B. Preparation of nanocomposite films
Hydrogen sulfide solutions were prepared as fol-
lows:
In a gas wash flask, nitrogen was bubbled for
some minutes through ca. 30 ml water under stirring
to remove some dissolved oxygen. Then, hydrogen sul-
fide gas was bubbled through the gas wash flask for
15—30 min; excess hydrogen sulfide was absorbed in
sodium hydroxide solution. The concentration of such
a hydrogen sulfide solution is 0.04-0.08 M, determined
by back-titration with hydrochloric acid.
7
Higher con-
centrations (up to 0.17 M) were obtained by cooling the
solution during the introduction of the hydrogen sulfide
to ca. 5 °C.
1742
J.
Mater. Res., Vol. 8, No. 7, Jul 1993 © 1993 Materials Research Society
https://doi.org/10.1557/JMR.1993.1742
Downloaded from https:/www.cambridge.org/core. University of Basel Library, on 30 May 2017 at 18:30:58, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms.
L Zimmermann et al.: High refractive index films of polymer nanocomposites
For the preparation of films with low PbS content
(below 40 wt.
%),
gelatin (typically ca. 30 mg) was dis-
solved in water (typically ca. 1 ml) at 50-60 °C, and
lead acetate (the amount corresponding to the desired
final concentration of PbS in the film) was added after
the gelatin was dissolved. For the preparation of films
with high PbS content, gelatin (typically 10-50 mg)
was dissolved in a 0.1 M lead acetate solution (typi-
cally
1.3-2.5
ml) at 50-60 °C. Then, in both cases, the
solution was stirred well and cooled to room tempera-
ture.
Under vigorous stirring hydrogen sulfide solution
was added as fast as possible (ca. 1% stoichiometric
excess with respect to lead, the solution being at room
temperature or at 5 °C, as indicated by the necessary
hydrogen sulfide concentration). For the preparation of
films with more than ca. 40 wt. % PbS, hydrogen sulfide
concentrations of 0.09-0.17 M were used in order to
keep the total volume of the solution low. Uniform stir-
ring is important since otherwise coagulation can occur,
especially for high PbS/gelatin ratios. After complete
mixing the solutions were allowed to stand for 1 h.
During this time, gelation increased the viscosity to
suitable values for the spin-coating process.
Spin coating was performed on a CONVAC 1001 S/
ST 147 using glass substrates, if not otherwise indicated
in the text. About three drops of the solution were spread
over an area of ca. 10 cm
2
with a spatula at a spinning
rate of 50 rpm (the exact amount of liquid is not critical
with respect to the final layer thickness). Thereafter, the
spinning rate was increased to 100-4000 rpm, depend-
ing on the viscosity of the solution and the required layer
thickness. After rotation for 30-90 s, the spinning pro-
cess was interrupted and started again at 150 rpm
for 10 s.
The films then were dried at a pressure of
ca. 100 Torr in a slight and constant stream of air. It
is important that the film is exactly horizontal for films
with high regularity. The stream of air was moistened
in order to prevent fissures in the final film. This drying
procedure was carried out, e.g., during 2 h for a film
thickness of 0.2 /xm and 2 days for a film of 30 /xm.
Films designated in the text as "annealed films" were
subsequently held at 164 °C and 100 Torr for 2 h.
C. Chemical analysis of the nanocomposites
Elemental analysis for carbon and determination of
the water content were carried out by Dieter Manser of
the Microanalytic Laboratory Service of the Institut fur
organische Chemie, ETH-Zurich.
Microanalysis of a film containing 40 wt. % PbS
(calculated on the basis of the known PbS/gelatin ratio
in the solution used for the film preparation) yielded the
following values (in parentheses the calculated values
based on the analysis of the pure gelatin, see A. Chemical
Reagents): C,
26.53%
(26.05%), H, 3.88% (3.98%), N,
9.75%
(9.25%), and H
2
O, 2.45% (3.12%). This shows
in particular that the water/gelatin ratio in the films is
comparable to the ratio in "native" gelatin. Further, it
was found that the annealing procedure does not reduce
the water content significantly.
D. Surface profilometry
Surface profiles, surface roughnesses, and film thick-
nesses were determined by an Alphastep 200 Surface
Profilometer (Tencor Instruments) with a stylus of radius
1.5-2.5
/Am. The stylus force was adjusted to 9 mg, and
a scan time of 40 s was selected. The surface roughness
is calculated as the average deviation in the height
of the profile from the mean level. For film thickness
measurements, an area of the film was selected that
appeared homogeneous to the eye. In this area, a small
part of the film was removed with a spatula, creating an
artificial step at which the film thickness was determined.
E. Refractive index measurements
Refractive index measurements were performed on
a PLASMOS SD 2300 ellipsometer equipped with a
He-Ne laser (A = 632.8 nm) at an angle of incidence
of 70°. Refractive indices were measured at 10 separate
spots,
and 5 measurements were performed at each spot
resulting in a total of 50 measurements per sample. In
the text, the corresponding average value is indicated.
To be sure that the measured refractive index is not
influenced by interference effects, i.e., reflections of a
part of the incident beam on the surface of the substrate
(e.g., glass), films of different thicknesses were investi-
gated. An example follows: Films containing 45 wt. %
PbS were prepared on glass substrates (refractive index
1.423 ± 0.040, 95% confidence level) in a thickness
range between 40 nm and 2 /tm. Between 700 nm and
2 fim thickness, these films show the brown color of
colloidal lead sulfide. Their refractive index does not
depend on thickness (1.807 ± 0.027, 95% confidence
level).
To verify that there is no influence of the sub-
strate, annealed films of 1 and 2
/xvo.
were prepared on
silicon wafers (refractive index 3.8071 ± 0.0024, 95%
confidence level). Their color is also brown, and the
measured refractive indices are 1.818 and
1.802.
On
glass,
refractive indices in the same range are also found
for films with a thickness between 40 and 250 nm. They
also exhibit a brown color. However, between 250 and
700 nm, the films show interference colors different from
brown.
14
'
15
In this thickness region, no consistent indices
could be deduced from the measurements.
F. Electron microscopy
For transmission electron microscopy, a gelatin cap-
sule (diameter and height ca. 5 mm) was filled with
J.
Mater. Res., Vol. 8, No. 7, Jul 1993
1743
https://doi.org/10.1557/JMR.1993.1742
Downloaded from https:/www.cambridge.org/core. University of Basel Library, on 30 May 2017 at 18:30:58, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms.
L. Zimmermann ef a/..
1
High refractive index films of polymer nanocomposites
epon-araldite,
16
turned upside down upon the nanocom-
posite film (on glass), and cured at 60 °C for 24 h.
After cooling to room temperature, the sample was
plunged into liquid nitrogen whereupon the polymer
components detached from the support. The material was
cut in pieces of ca. 4 mm
2
area and ca. 0.5 mm thickness
and embedded again in epon-araldite. After hardening
as described above, thin sections (interference color
silver) were cut on a Reichert Ultracut Microtome with
a diamond knife (Diatome, Biel-Bienne, Switzerland).
Sections were floated on the water as trough liquid. The
electron micrographs were taken on a Philips EM 301
with an acceleration voltage of 80 kV.
The SEM (scanning electron microscopy) samples
were coated with 3 nm platinum-carbon by electron
beam evaporation in a BAF 120 (Balzers) device. Spec-
imens were investigated in a Hitachi
S-900
in-lens field
emission SEM using the secondary electron signal.
III.
RESULTS AND DISCUSSION
A. Preparation and characterization of films
Highly filled PbS-polymer nanocomposites can
be prepared from homogeneous solutions containing
colloidal PbS and polymer (this is favored over exposure
to hydrogen sulfide of a film containing Pb
2+ 7
).
Among several water-soluble polymers tested, i.e.,
poly(vinyl alcohol), poly(acrylamide), poly(acrylic acid),
poly(ethyleneoxide), and gelatin, the most suitable was
found to be gelatin.
The homogeneous polymer-PbS solutions can be
used for the preparation of films. The simplest method
for this purpose is to evaporate the solvent from a stag-
nant "puddle", either at ambient conditions, at elevated
temperature, or at reduced pressure. This allowed for the
preparation of homogeneous films with a PbS content of
up to ca. 65 wt. % PbS. It is also possible to prepare
homogeneous solutions with PbS/gelatin ratios corre-
sponding to higher PbS contents in the film; however,
local phase separation occurs upon solvent evaporation
if the PbS/gelatin ratio is too high. Moreover, the films
manufactured by this simple solvent evaporation are not
uniform (the film thickness decreases with increasing
distance from the center of the film).
Another method for the preparation of polymer
15
'
17
"
21
or polymer-dye
22
films is spin coating. We found that
homogeneous PbS-gelatin films can be created by spin
coating with higher PbS content than is possible with
the other methods investigated. Uniform films could be
obtained when the substrate is initially covered with the
PbS—gelatin solution over the entire area where the film
is finally desired (here typically 10 cm
2
); upon spinning
(100-4000 rpm), the area originally covered with so-
lution yields an even film, while outside of this area
stripes are formed. In this way, it was possible to produce
films with a PbS content of up to 86.4 wt. %. However,
films with a PbS content below ca. 70 wt. % were of
significantly better quality than those containing more
PbS.
The films show the brown color of colloidal PbS.
The viscosity of gelatin or gelatin-PbS solutions can
be controlled by the concentration and the age of the
solution, i.e., the degree of gelation. Good conditions
for spin coating were found for a gelation time of 1 h.
Depending on the thickness and PbS content, a
portion of light passes through the films, and this light
is not scattered as far as can be judged by the eye. It
has been suggested that the refractive index of the two
components must be closely matched, in nanocompos-
ites,
to avoid cloudiness.
1
'
2
Our result on PbS-polymer
nanocomposites agrees with Mahler's conclusion
4
'
5
that
this aspect seems to be of little importance, provided that
the particle size as well as the interparticle distances are
much smaller than the wavelength of light. The absorp-
tion spectrum of the PbS films corresponds to that of
colloidal PbS with a particle diameter of ca. 5-20 nm.
23
Thermogravimetry (TGA) of a film with a PbS
content of 66.6 wt. % shows that the thermal stability
of the composite is comparable with that of "native"
gelatin alone.
For film thickness measurements, a small part in the
interior of the film area was removed mechanically with
a spatula, and the difference between substrate and film
level was measured by surface profilometry.
17
An ex-
ample of such a profile is displayed in Fig. 1. The film
thickness as a function of rotation rate was investigated
for a pure gelatin (20 g/1) and a PbS-gelatin solution
(concentration of gelatin 10 g/1 and of PbS 8.2 g/1,
corresponding to a lead sulfide content in the film of
45 wt.
%).
By varying the rotation rate from 300 to
2000 rpm, the film thickness decreases from ca. 10 /xm
to ca. 200 nm for the pure gelatin solutions (Fig. 2);
varying the rotation rate from 200 to 3000 rpm decreases
the film thickness from ca. 1 /tin to ca. 40 nm when
PbS is present (Fig. 2). It is obvious that the addition of
PbS reduces the film thickness. It has been reported that
I.
L
200nmj[
1 mm
FIG. 1. Step height profile indicating the film thickness for a
spin-coated PbS-gelatin nanocomposite containing 45 wt. % PbS.
1744
J.
Mater. Res., Vol. 8, No. 7, Jul 1993
https://doi.org/10.1557/JMR.1993.1742
Downloaded from https:/www.cambridge.org/core. University of Basel Library, on 30 May 2017 at 18:30:58, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms.
L. Zimmermann et al.: High refractive index films of polymer nanocomposites
2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6
0.6
log (spinning rate/rpm)
FIG. 2. Film thickness of spin-coated films as a function of the
decadic logarithm of the spinning rate (dashed lines are drawn arbi-
trarily). (•): gelatin and (•): PbS-gelatin nanocomposite containing
45 wt.
%
PbS. Dashed lines are arbitrary.
double-logarithmic plots yield a straight line relationship
between film thickness and rotation rate for polymers
17
and polymer-dye films,
22
but other reports disagree.
15
The films described here do not disclose such a linear
dependence (Fig. 2).
The surface roughness of the above films was ana-
lyzed by surface profilometry for a film with 45 wt. %
PbS.
The ratio of surface roughness and film thickness
(relative roughness) gives information on the film qual-
ity. In the following, the "average roughness" is defined
as the average deviation from the mean film level, and
the "peak-to-valley roughness" as half the difference be-
tween the highest and the lowest detected vertical point.
Both types of relative surface roughnesses are displayed
over a distance of 80
/JLVCI
and 5 mm in Figs. 3(a) and
3(b).
By necessity, the relative peak-to-valley roughness
is larger than the relative average roughness. The relative
roughnesses measured over longer distances are larger
than those for shorter distances, indicating that there is
roughness on a characteristic wavelength on the order
of the smaller scanning distance or larger. The relative
average roughness does not exceed 0.08, and the relative
peak-to-valley roughness is below 0.5. Therefore, no
spots were detected on which the substrate is not covered
by film.
No significant difference could be detected by
scanning electron microscopy (SEM) before and after
annealing. Pictures at different magnifications are
displayed in Figs. 4(a)-4(c) (86.4 wt. % PbS). A stereo
image [Fig. 4(b)] discloses a surface roughness, as
evidenced already by surface profilometry. However, no
microphase separation and no objects above ca. 100 nm
are visible. It is not known from these pictures whether
the small, often spherical, particles are single PbS
to
en
cu
c
-C
en
o
0.5 -
0.3 -
Q- 0.2 H
~cu
*" 0.1 -\
0.0
1.5
2.0 2.5 3.0
log (film thickness/nm)
(a)
0.12
1.5 2.0 2.5 3.0 3.5
log (film thickness/nm)
(b)
FIG. 3. Relative peak-to-valley (a) and average surface roughness (b)
as a function of log film thickness (dashed lines are drawn arbitrarily)
of PbS-gelatin nanocomposites containing 45 wt.
%
PbS measured
over a horizontal distance of 5 mm (•) and 80 /am (•).
particles, possibly enveloped by gelatin, or whether these
objects are aggregates of smaller particles.
No evidence for large-scale heterogeneities was
observed by transmission electron microscopy (TEM).
Figure 4(d) shows a micrograph of a nanocomposite with
68 wt. % PbS after heat treatment. The bright stripes
on both sides of the film cross section represent the
embedding material (see Sec. II). Particles with typical
dimensions of 2-5 nm were detected. The particles are
crystalline by electron diffraction, and the diffraction
pattern agrees with a reference of crystalline PbS.
B. Refractive indices
The refractive index at 632.8 nm was measured
before and after annealing of the films (at 164 °C and
J.
Mater. Res., Vol. 8, No. 7, Jul 1993
1745
https://doi.org/10.1557/JMR.1993.1742
Downloaded from https:/www.cambridge.org/core. University of Basel Library, on 30 May 2017 at 18:30:58, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms.
L. Zimmermann et a/.: High refractive index films of polymer nanocomposites
10 ,-m
FIG. 4. Electron micrographs of spin-coated PbS-gelatin films, (a)-(c) SEM micrographs of a film containing 86.4 wt.
%
PbS (the white
elevation in the stereo micrograph (b) is not representative of the film, but it helps in getting the stereo impression when it is fixed) and
(d) TEM micrograph of a film containing 68 wt.
%
PbS.
1746
J.
Mater. Res., Vol. 8, No. 7, Jul 1993
https://doi.org/10.1557/JMR.1993.1742
Downloaded from https:/www.cambridge.org/core. University of Basel Library, on 30 May 2017 at 18:30:58, subject to the Cambridge Core terms of use, available at https:/www.cambridge.org/core/terms.