Optical Properties of Single-Wall Carbon Nanotubes Films

Abstract

The paper describes a set of simple experiments performed to develop an optical model to describe Si/SiO(2) substrates coated with two transparent films of carbon nanotubes. The final goal is to use such optical model to investigate the interaction of proteins with carbon nanotubes. Experiments were performed to assess light reflection as a function of the wavelength or angle of incidence using two substrates (same material, different amounts) composed of oxidized carbon nanotubes. The experimental results indicate that the selected carbon nanotubes layers are anisotropic and significantly different from each other. Experiments performed by spectroscopic ellipsometry (as a function of the wavelength and incident angle) enabled the development of an Effective Medium Approximation model consisting in a two-fraction phase (arc-evaporated carbon and void space). Furthermore, the model enabled calculating the amount of protein adsorbed on the surface of the carbon nanotubes film.

Full text

Optical properties of single-wall carbon nanotube films deposited on Si/SiO
2
wafers
Hariyadi Soetedjo
1
, Maria F. Mora, Carlos D. Garcia ⁎
Department of Chemistry, The University of Texas at San Antonio, One UTSA Circle, San Antonio, TX 78249, USA
abstractarticle info
Article history:
Received 28 May 2009
Received in revised form 3 February 2010
Accepted 12 February 2010
Available online 19 February 2010
Keywords:
Carbon nanotubes
Spectroscopic ellipsometry
Optical model
Protein adsorption
The paper describes a set of simple experiments performed to develop an optical model to describe Si/SiO
2
substrates coated with two transparent films of carbon nanotubes. The final goal is to use such optical model
to investigate the interaction of proteins with carbon nanotubes. Experiments were performed to assess light
reflection as a function of the wavelength or angle of incidence using two substrates (same material,
different amounts) composed of oxidized carbon nanotubes. The experimental results indicate that the
selected carbon nanotubes layers are anisotropic and significantly different from each other. Experiments
performed by spectroscopic ellipsometry (as a function of the wavelength and incident angle) enabled the
development of an effective medium approximation model consisting in a two-fraction phase (arc-
evaporated carbon and void space). Furthermore, the model enabled calculating the amount of protein
adsorbed on the surface of the carbon nanotube film.
Published by Elsevier B.V.
1. Introduction
Carbon nanotubes (CNT) are conductive, well-ordered, all-carbon
hollow nanomaterials with a high aspect ratio [1]. Just like other
nanostructured materials [2–5], CNT have shown unique properties
(thermal, mechanical, electrical, biological, etc.) not found in other
allotropic forms of carbon, such as graphite or diamond. Most of these
extraordinary properties are linked to higher surface area, surface
roughness, surface defects, and altered electron distributions [6]. For
these reasons, CNT have contributed to the development of basic and
applied studies in the fields of physics, chemistry, and material
sciences. In particular, single-wall carbon nanotubes offer a well-
defined atomic structure, high length to diameter ratio, and chemical
stability. Depending on the atomic structure, single-wall CNT exhibit
specific electronic, chemical, and physical properties [7]. Among
biomedical applications [8–11], single-wall CNT are particularly useful
as substrates for the preparation of biosensors because they are
biocompatible [12,13] and stable over a large range of conditions,
demonstrate catalytic activities towards many reactions [14–16], and
provide significant increases in surface area [14]. Due to their small
size and conductivity they can also be regarded as the smallest
possible electrodes, with diameters as small as one nanometer [17].
When using CNT as substrates forbiosensorsitiscrucialto
understand the interaction between the recognition element (typi-
cally a protein) and the CNT because they can affect the protein
conformation [18–22], cellular adhesion [23], and biocompatibility
[18,24,25]. To understand the roots of these interactions, a detailed
characterization of the CNT substrates must be performed.
Several techniques have been used to perform a morphological
and structural characterization of CNT. Among others, scanning
tunneling microscopy, transmission electron microscopy, X-ray
photoelectron spectroscopy, neutron diffraction, X-ray diffraction,
infrared spectroscopy, and Raman spectroscopy are the most
commonly reported characterization techniques [26].Themain
problem associated with these techniques is their limitation to
perform in-situ measurements of adsorption of biological molecules
under physiological conditions. An alternative technique to perform
in-situ measurements is ellipsometry [21].
Ellipsometry is an optical technique that measures changes in the
reflectance and phase difference between the parallel (R
P
) and
perpendicular (R
S
) components of a polarized light beam upon
reflection from a surface. Using Eq. (1),
tan ΨðÞeiΔ=RP
RS
ð1Þ
the intensity ratio of R
P
and R
S
can be related to the amplitude ratio
(tan Ψ)and the phase difference (Δ) between the two components
[27]. Because ellipsometry measures the ratio of two values (R
P
and
R
S
) originated from the same signal, the data collected are highly
accurate and reproducible. The changes in polarization measured by
ellipsometry are extremely sensitive to film thickness (down to the
monolayer level), optical constants, and film microstructure (such as
surface roughness, index grading, and intermixing). Most impor-
tantly, ellipsometry (and specifically the setup described in this
manuscript) enables investigating the adsorption of proteins in real-
time, on a variety of substrates, and using aqueous environments.
Thin Solid Films 518 (2010) 3954–3959
⁎Corresponding author. Tel.: +1 210 458 5774; fax: +1 210 458 7428.
E-mail address: carlos.garcia@utsa.edu (C.D. Garcia).
1
Permanent address: CIRNOV, University of Ahmad Dahlan, Jalan Cendana 9a,
Semaki, Yogyakarta, 55166, Indonesia.
0040-6090/$ –see front matter. Published by Elsevier B.V.
doi:10.1016/j.tsf.2010.02.037
Contents lists available at ScienceDirect
Thin Solid Films
journal homepage: www.elsevier.com/locate/tsf

However, because the measurement is affected by the optical
constants, thickness, and film microstructure, the interpretation of
ellipsometric data requires an optical model. This is probably the
biggest limitation of the technique, which will enable formulating
conclusions that are only as good as the selected optical model.
Considering the potential advantages of using ellipsometry to
study the interaction of biomolecules with single-wall CNT, the
objective of this paper was to develop a model able to describe the
optical properties of transparent layers of single-wall CNT. To develop
the model, the optical properties of two different substrates were
investigated as a function of the wavelength and angle of incidence
using reflection and ellipsometric experiments. Finally, the model was
used to gain insight into the interaction kinetics between a selected
protein and carbon nanotubes [21].
2. Materials and measurements
2.1. Substrates
Single-wall CNT-coated silica (Si/SiO
2
/CNT) surfaces were used for
the present studies. As described in earlier studies [20,21,28], Si/SiO
2
/
CNT surfaces were prepared by Eikos Inc. (Franklin, MA), using b100N
silicon wafers (Sumco, Phoenix, AZ) as substrates. According to the
provider, a layer of CNT was deposited on the wafers using arc-
produced single-wall CNT having about a 1.3 nm diameter. The raw
material formed in the arc reactor was purified, to remove metal
catalyst and non-tubular forms of carbon, by a process of acid reflux,
followed by washing and centrifugation. This process is known to
introduce carboxylic acid groups on the surface of the CNT, and
therefore the results discussed in this paper should not be broadly
applied to CNT films prepared by other methods. Once purified, the
nanotubes were dispersed in water and alcohol to form an ink. This
dispersion was then spray-coated onto the Si/SiO
2
wafer heated to
65 °C while monitoring deposition rate. The coating formed is
essentially a layer of pure single-wall CNT and contains no residual
organic additives or polymeric constituents. For the experiments
described in this paper, two types of CNT films were prepared in the
same way, but containing different amounts of single-wall CNT. These
two films will be denoted as Si/SiO
2
/CNT
LOW
(for the film containing
the lowest amount of CNT) and Si/SiO
2
/CNT
HIGH
(for the film
containing the highest amount of CNT). Control experiments were
also performed using the same Si/SiO
2
wafers without the CNT layer.
2.2. Spectroscopic ellipsometry
As describ ed elsewhere [21], thesubstrate characterizationas well as
the dynamic adsorption experiments were performed at room
temperature using a variable angle spectroscopic ellipsometer (VASE,
J.A. Woollam Co; Lincoln, NE). Initially, different substrates were
analyzed using light reflection for p-polarization and s-polarization by
varying the angle of incidence from 60° to 85°, in 0.5° increments.
Ellipsometry data (amplitude ratio (Ψ) and phase difference (Δ)asa
function of wavelength (λ) or time) were obtained and modeled using
the WVASE® software package (J.A. Woollam Co; Lincoln, NE). The
difference between the experimental and model-generated data was
assessed by the mean square error (MSE) defined by Eq. (2),
MSE =1
2N−M∑
N
i=1
Ψmod
i−Ψexp
i
σexp
Ψ;i
!
2
+Δmod
i−Δexp
i
σexp
Δ;i
!
2
"#
=1
2N−MX2
ð2Þ
where Nis the number of Ψand Δpairs used, Mis the number of
parameters varied in the regression analysis, and σis the standard
deviation of the experimental data points. Although smaller MSE
values indicate better fittings, MSEb15 are typically considered
acceptable [28,29].
In all cases, the sample under investigation was mounted on a
micrometer-position-controlled translation stage with the gradient
direction perpendicular to the plane of incidence. Substrates were
initially characterized in air, varying the incident angle between 60° and
70° (with respect to the substrate), and the wavelength between
250 nm and 900 nm. The structure of the adsorbed films of protein was
evaluated in aqueous media using a commercial cell (J.A. Woollam Co;
Lincoln, NE). More details of the cell can be found elsewhere [21].
2.3. Reagents and solutions
All aqueous solutions were prepared using 18 MΩcm water (NANO-
pure Diamond, Barnstead; Dubuque, Iowa) and analytical reagent grade
chemicals. Sodium pyrophosphate (Na
4
P
2
O
7
), sodium chloride, and
sodium hydroxide were purchased from Fisher Scientific(FairLawn,
NJ). The pH of the solutions was adjusted using either 1 M NaOH or 1 M
HCl (Fisher Scientific; Fair Lawn, NJ) and measured using a glass electrode
and a digital pH meter (Orion 420A+, Thermo; Waltham, MA). D-amino
acid oxidase (Sigma; Saint Louis, MO) from porcine kidney was purchased
as a lyophilized powder (≥1.5 U mg
−1
)andkeptat−4°C until used.
Stock solutions of protein (0.1 mg mL
−1
) were prepared by dissolving a
known amount of protein in pyrophosphate buffer. Other solutions of
proteins were prepared by diluting the corresponding amount of stock in
pyrophosphate buffer. In all cases, the concentration of the stock solution
was confirmed by spectrophotometry (Genesys 10uv, Thermo Elec.
Corp.).
3. Results and discussion
3.1. Optical characterization
Despite the enormous technological relevance, there are only a few
reports describing the optical properties of thin CNT films [7,30].The
most relevant problems associated with thesemeasurements stemfrom
the fact that CNT films are typically composed of a mixture of metallic
and semiconducting scrambled carbon nanotubes. As a consequence,
CNT films are neither flat nor smooth. Recently, Wu et al. described the
optical properties of self-standing CNT films (240 nm thick) in the IR
region of the spectrum [31]. Barnes et al. reported optical and electrical
properties of single-wall CNT films deposited on glass and quartz
substrates [28] and noted a large absorption peak at around 4.4 eV
(∼282 nm). This peak was attributed to the π-plasmon resonance of the
metallic single-wall CNT. However, they pointed out that although
absorption spectra can be determined simply by subtracting the
transmission and reflection spectra from unity, it is not known if
neglecting the reflection data is an acceptable approximation for CNT
films. In order to explain the optical reflectance of thick films of single-
wall CNT and taking advantage of the reflective nature of the Si/SiO
2
substrates, reflection measurements were performed at p-polarization
(parallel to the plane of incidence) using 70° and 80° as the angles of
incidence (in air). The first angle (70°, Fig. 1) was selected because it is
the incident angle in the liquid cell used for the protein adsorption
experiments and it is close to the Brewster angle of most substrates.
Based on preliminary experiments, the second angle (80°, Fig. 2)was
selected to highlight the differences between the two CNT films under
investigation. In both cases, the spectra corresponding to the three
selected substrates showed two reflection peaks at approximately
265 nm and 365 nm, corresponding to energy values of 4.69 and
3.41 eV, respectively. However, when 80° was used, a third reflection
peak was obtained at 334 nm in the Si/SiO
2
/CNT
HIGH
substrate. These
electronic transitions have been associated with π-electrons present in
the CNT network that (in the case of nanoscale objects) could be
collectively excited originating a surface plasmon [32]. To identify the
magnitude and position of these transitions is relevant because they can
3955H. Soetedjo et al. / Thin Solid Films 518 (2010) 3954–3959

be related to specific properties of single-wall CNT such as anisotropy
[28,33],structure[26,34], size distribution [35], chirality [36],ormethod
of synthesis. Specifically, it has been reported that a dominant feature is
present at approximately 4.5 eV when the CNT are randomly oriented,
but shifts to approximately 5.25 eV if the film is vertically aligned on the
substrate [37]. Although the results shown in Figs. 1 and 2 indicate that
both peaks could be attributed not only to the CNT layer but also to the
SiO
2
substrate, the position of these features are in agreement with the
presence of a film of randomly distributed CNT with the expected
anisotropic behavior [28]. A comparison of Rs (as a function of the
wavelength, data not shown) with respect to Rp (Figs. 1 and 2), along
with the relatively sharp reflection peaks observed at both angles,
suggested that the propagation of the electric field (normal to the
substrate plane) couples more strongly than the electric field in the
substrate plane, again supporting the anisotropic behavior.
The two single-wall CNT films (Si/SiO
2
/CNT
LOW
and Si/SiO
2
/CNT
HIGH
)
and the Si/SiO
2
substrate were also investigated as a function of the angle
of incidence (in the 60° to 85° range) at a wavelength of 365 nm. This
wavelength was selected based on the absorption peak observed in Figs. 1
and 2. As shown in Fig. 3,thereflectivity curves generally possess a
parabolic dependence with respect to the incident angle. It is also evident
that although the two studied CNT films exhibit different optical
properties, the film Si/SiO
2
/CNT
HIGH
showed the most significant
difference with respect to the Si/SiO
2
substrate. According to these
results, pseudo-Brewster angles of 81°, 80°, and 71° were observed for the
Si/SiO
2
,Si/SiO
2
/CNT
LOW
, and Si/SiO
2
/CNT
HIGH
substrates, respectively. The
relevance of these results is threefold. First, considering that shifts in
pseudo-Brewster angles as small as 1° indicate the presence of a layer
with significantly different properties, the shifts observed in Fig. 3 (up to
10°) confirmed the need to characterize these films individually. Second,
the results demonstrate that even simple experiments performed with
the ellipsometer could provide complementary information regarding the
substrates that other techniques (such as atomic force microscopy or
scanning electron microscopy) cannot provide. Third, because the
ellipsometric parameters are most sensitive to changes in the vicinity of
the Brewster angle, experiments performed in commercial liquid cells
may require additional considerations to overcome the limitations
imposed by the fixed angle of incidence. Although shifts in the pseudo-
Brewster angle have been linked to different phenomena [38–40],our
results are in agreement with previous reports stating that angular shifts
can be correlated to increased layer phase thickness [41].
The two single-wall CNT films (Si/SiO
2
/CNT
LOW
and Si/SiO
2
/CNT
HIGH
)
and the Si/SiO
2
substrate were also investigated using spectroscopic
ellipsometry (SE). Figs. 4 and 5 show the dependence of psi (Ψ)anddelta
(Δ)asafunctionofthewavelength(λ) of the incident light beam using air
as the ambient for Si/SiO
2
/CNT
LOW
and Si/SiO
2
/CNT
HIGH
, respectively.
Although both substrates were investigated using 65°, 70°, and 75° as
Fig. 1. Reflection at p-polarization for substrates of Si/SiO
2
, Si/SiO
2
/CNT
LOW
, and Si/SiO
2
/
CNT
HIGH
as a function of the wavelength at an angle of incidence of 70°.
Fig. 2. Reflection at p-polarization for substrates of Si/SiO
2
, Si/SiO
2
/CNT
LOW
, and Si/SiO
2
/
CNT
HIGH
as a function of the wavelength at an angle of incidence of 80°.
Fig. 3. Reflection at p-polarization for substrates of Si/SiO
2
, Si/SiO
2
/CNT
LOW
, and Si/SiO
2
/
CNT
HIGH
as a function of the angle of incidence, collected using 365 nm as the incident
wavelength.
Fig. 4. Ψ–Δcurves for substrates of Si/SiO
2
/CNT
LOW
as a function of the wavelength
collected at an angle of incidence of 70° and using air as the ambient (MSE =5.51).
3956 H. Soetedjo et al. / Thin Solid Films 518 (2010) 3954–3959

angles of incidence, only the results obtained at 70° are shown in the
figures. In agreement with the data previously discussed, both CNT-
coated samples and the Si/SiO
2
substrate showed two distinctive features
at 265 nm and 362 nm. By comparing Figs. 4 and 5,itisalsoevidentthat
both samples (although made with the same starting material) displayed
significantly different optical properties. The most challenging aspect of
these results was to develop an optical model that could describe both
substrates.
3.2. Optical model
Carbon films have been represented by a wide variety of optical
models including Tauc–Laurenz [42–44], effective medium approxima-
tion (EMA) [45], and general oscillators [28].However,thevalidityof
these models is heavily influenced by the particular composition and
microstructure of the substrate. Consequently, different models were
evaluated in order to describe the selected substrates in terms of the
refractive index (n), extinction coefficient (k), and thickness (d). Based on
the difference between the experimental and model-generated data, the
mean square error (MSE) was used to select the most appropriate optical
model. According to our results, the best fit was obtained when the
substrates were modeled by three uniform uniaxial layers, with optical
axes parallel to the substrate surface. The dielectric functions of the
substrates were described using a layer of Si (bulk; d=1 mm),a layerof
SiO
2
(d=3 nm), and a two-media Bruggeman EMA model consisting of a
fraction (f) of arc-evaporated carbon [46,47] and a fraction of void space
(1-f). The thickness of the EMA layer ranged from 15 to 40 nm depending
on the substrate used. As shown in Figs. 4 and 5, the same optical model
could be used to accurately describe the properties of both CNT films (Si/
SiO
2
/CNT
LOW
and Si/SiO
2
/CNT
HIGH
) with MSE values of 5.03 and 7.54,
respectively. The main difference is that the fraction of void space in Si/
SiO
2
/CNT
LOW
was found to be 98%, while the fraction of void space in the
Si/SiO
2
/CNT
HIGH
film was found to be 30%. This optical model is in good
agreement with previous reports [20] but differs slightly from other
models used to describe CNT films prepared under different condi-
tions [28,31,33,48]. This singularity was expected, as nanotubes prepared
(or purified) by different methods may have different electrical and
optical properties due to different excitonic transitions between the van
Hove singularities [49–53].
It is worth noting that the model was also able to describesubstrates
with the same fraction of void space but ranging in thicknesses from
15 nm up to 40 nm (data not shown). The calculated thickness values as
well as the structure of the film used in the optical model were verified
by performing atomic force microscopy and scanning electron micros-
copy on the selected substrates (data not shown). SEM pictures of the
substrates have been previously reported in literature [21]. Although it
is not possible to precisely quantify the empty space from the SEM and
AFM pictures, it is clear that the proposed model is consistent with the
physical characteristics of the films. It is worth mentioning that
ellipsometry (light spot ∼7mm
2
) was not sensitive to slight variations
(spot-to-spot and sample-to-sample) observed by microscopy.
As previously stated, the dielectric functions of the substrates were
describedusing a layer of Si, a layer of SiO
2
, and a two-media Bruggeman
effective medium approximation (EMA) model consisting of arc-
evaporated carbon and void space. The corresponding values of nand
Fig. 5. Ψ–Δcurves for substrates of Si/SiO
2
/CNT
HIGH
as a function of the wavelength
collected at an angle of incidence of 70° and using air as the ambient (MSE = 6.97).
Fig. 6. Optical constants (nand k) for the materials used in the manuscript. A) arc-evaporated
carbon, B) the CNT layer of Si/SiO
2
/CNT
HIGH
, and C) the CNT layer of Si/SiO
2
/CNT
LOW
.
3957H. Soetedjo et al. / Thin Solid Films 518 (2010) 3954–3959

kvalues for the three materials used in the manuscript (arc-evaporated
carbon, the CNT layer of Si/SiO
2
/CNT
HIGH
, and the CNT layer of Si/SiO
2
/
CNT
LOW
) are provided in Fig. 6A, B, and C, respectively. As it can be
observed, the dependence of nand kin the three substrates is very similar,
but decreasing in magnitude as the amount of CNT in the films decreases.
3.3. Adsorption of proteins to CNT
As previously stated, thegoal of this paper was to develop an optical
model that enables studying both the formation kinetics as well as the
structure of proteins adsorbed to the single-wall CNT layers by
spectroscopic ellipsometry [21]. Spectroscopic ellipsometry experi-
ments were performed by placing the substrates in the liquid cell,
covering them with buffer, and measuring the ellipsometric angles(as a
function of wavelength) using water as the ambient medium.
Althoughthe proposed optical model was ableto describe the optical
properties of the Si/SiO
2
/CNT
LOW
(98% void) substrate (MSE = 5.0)
when immersed in aqueous solutions (and using water as ambient), a
significant disagreement was obtained between the experimental and
the model-generated data for the Si/SiO
2
/CNT
HIGH
(30% void) substrate
(MSEN20). Preliminary results (to be reported) suggest that the
substrates containing a higher load of single-wall CNT (Si/SiO
2
/CNT
HIGH
,
30% void) may significantly swell (up to 3 times) when submerged in
buffer (pH= 5.6) for ∼10 h. For this reason, further experiments related
to the interaction of proteins to carbon nanotubes were performed using
the Si/SiO
2
/CNT
LOW
(98% void) substrates.
For the adsorption experiments, the adsorbed proteins were
modeled with an additional non-absorbing layer (denoted as
“protein”in Fig. 7), whose optical constants were described using a
Cauchy parameterization model, according to Eq. (3),
nλðÞ=A+B
λ2+C
λ4ð3Þ
where A,B,andCare computer-calculated fitting parameters and λis
the wavelen gth of the incident beam. Fig. 8 shows the dependence of the
ellipsometric angles (Ψand Δ) as a function of the wavelength (in the
250 to 850 nm range) for a layer of protein (D-amino acid oxidase,
DAAO) adsorbed to a Si/SiO
2
/CNT
LOW
(98% void) substrate. The
thickness of the protein layer was calculated to be 5.2± 0.1 nm. As
can be observed, a very good agreement (MSE=3.49) was obtained
between the experimental data and the data generated by the proposed
optical model. As reported elsewhere [21], these results suggest the
formation of a monolayer of DAAO on the CNT surface. Considering that
DAAO exhibits an elongated ellipsoidal framework with approximate
dimensions of 11 nm (length)× 4 nm (width) [54], the thickness values
obtained under the selected experimental conditions indicate that
DAAO adopted a tip-to-surface alignment (slightly tilted) with respect
to the CNT surface. The amount of adsorbed protein (Γ,expressedin
mg m
−2
) can be determined using Eq. (4),
Γ=dn−n0
ðÞ
dn =dcðÞ ð4Þ
where dis the thickness value of the adsorbed layer, nis the refractive
index of the protein, and n
0
is the refractive index of the ambient (H
2
O)
[55]. In accordance with previous reports [56–59],therefractiveindex
increment for the molecules in the layer (dn/ dc) was assumed to be
0.187 mL g
−1
. Under these conditions, the amount of adsorbed DAAO,
calculated from Fig. 7 was 3.8± 0.2 mg m
−2
.
Table 1 summarizes the most important results related to the
thickness of each layer in the studied films. Two other aspects are also
worth noting. First, the reflection peak observed at ∼275 nm (at Ψ
axis in Fig. 8) was attributed to an absorption band of the adsorbed
DAAO molecules. This absorption peak was verified by measuring a
spectral curve of DAAO in solution under the same experimental
conditions (data not shown). Second, the use of alternative optical
models, such as those considering the possibility of DAAO penetrating
in the CNT layer, or those including surface roughness did not yield
any significant improvement in the MSE value.
4. Conclusions
The paper describes a set of simple experiments performed to develop
an optical model to describe Si/SiO
2
substrates coated with two different
transparent films of single-wall CNT. Although a considerable contribu-
tion from the Si/SiO
2
substrate overlapped with the expected π-plasmon
resonance of the CNT, the obtained results (p-reflection as a function of
Fig. 7. Optical model proposed to describe the formation of a layer of DAAO on the Si/
SiO
2
/CNT
LOW
substrate. Adsorption conditions: 5 mM pyrophosphate buffer pH =5.75
and [DAAO] =0.01 mg/mL; flow rate: 1.8 mL/min.
Fig. 8. Relationship of Ψand Δas function of wavelength obtained from the spectroscopic
ellipsometer for DAAO layer (5.2 ±0.1 nm) adsorbed on the Si/SiO
2
/CNT
LOW
(98% void)
substrate (MSE= 3.49).
Table 1
Summary of the composition, thickness of each layer (Si, SiO
2
, CNT, and DAAO), ambient, and mean square error (MSE) calculated for the selected substrates described in the
corresponding figures.
Substrate Void (%) Si SiO
2
(nm) CNT (nm) DAAO (nm) MSE Ambient Fig.
Si/SiO
2
/CNT
LOW
98 Bulk 3.0±0.1 22±2 –5.51 Air 4
Si/SiO
2
/CNT
HIGH
30 Bulk 3.0±0.1 21.5±0.1 –6.97 Air 5
Si/SiO
2
/CNT
LOW
98 Bulk 3.0±0.1 24±2 –2.77 Buffer 7
Si/SiO
2
/CNT
LOW
98 Bulk 3.0±0.1 24±2 5.2±0.1 3.49 Buffer 7
3958 H. Soetedjo et al. / Thin Solid Films 518 (2010) 3954–3959

the incident wavelength or angle) suggest that the single-wall CNT layers
are anisotropic and significantly different. Experiments performed by
spectroscopic ellipsometry (as a function of the wavelength and incident
angle) enabled the development of an EMA model consisting of a
combination of arc-evaporated carbon and void space. Although one of
the films (Si/SiO
2
/CNT
HIGH
, 30% void) significantly changed when
introduced in water (possibly swelling), the other film (Si/SiO
2
/CNT
LOW
,
98% void) enabled studying the interaction of a protein (DAAO) with
single-wall CNT. It is expected that further developments in the area of
ellipsometry will enable the systematic investigation of the interactions of
proteins with nanophased materials.
Acknowledgments
Financial support for this project was provided in part by The
University of Texas at San Antonio, and the National Institute of
General Medical Sciences (NIGMS)/National Institutes of Health
(1SC3GM081085). Authors would also like to thank David Olmos/
Dr. Miguel Jose-Yacaman (UTSA) and Dr. M. Miller (SwRI) for collecting
the SEM and AFM images, respectively.
References
[1] S. Iijima, Nature 354 (1991) 56.
[2] T.J. Webster, C. Ergun, R.H. Doremus, R.W. Siegel, R. Bizios, J. Biomed. Mater. Res.
51 (2000) 475.
[3] F.-Y. Cheng, S.P.-H. Wang,C.-H. Su, T.-L. Tsai, P.-C.Wu, D.-B. Shieh, J.-H. Chen,P.C.-H.
Hsieh, C.-S. Yeh, Biomaterials 29 (2008) 2104.
[4] Y.-C. Chung, I.H. Chen, C.-J. Chen, Biomaterials 29 (2008) 1807.
[5] C. Deng, J. Chen, X. Chen, C. Xiao, L. Nie, S. Yao, Biosens. Bioelectron. 23 (2008)
1272.
[6] H. Liu, T.J. Webster, Biomaterials 28 (2007) 354.
[7] V.N. Popov, Mater. Sci. Eng. R 43 (2004) 61.
[8] V.N. Khabashesku, J.L. Margrave, E.V. Barrera, Diamond Relat. Mater. 14 (2005)
859.
[9] W. Yang, P. Thordarson, J.J. Gooding, S.P. Ringer, F. Braet, Nanotechnology (2007)
412001.
[10] B.S. Harrison, A. Atala, Biomaterials 28 (2007) 344.
[11] M. Foldvari, M. Bagonluri, Nanomed. Nanotechnol. Biol. Med. 4 (2008) 173.
[12] Y. Liu, X. Qu, H. Guo, H. Chen, B. Liu, S. Dong, Biosens. Bioelectron. 21 (2006) 2195.
[13] D. Khang, S.Y. Kim, P. Liu-Snyder, G.T.R. Palmore, S.M. Durbin, T.J. Webster,
Biomaterials 28 (2007) 4756.
[14] M. Zhang, W. Gorski, Anal. Chem. 77 (2005) 3960.
[15] J. Weber, A. Kumar, A. Kumar, S. Bhansali, Sens. Actuators B 117 (2006) 308.
[16] H. Qi, C. Zhang, X. Li, Sens. Actuators B 114 (2006) 364.
[17] J.J. Gooding, Electrochim. Acta 50 (2005) 3049.
[18] D. Pantarotto, C.D. Partidos, R. Graff, J. Hoebeke, J.P. Briand, M. Prato, A. Bianco,
J. Am. Chem. Soc. 125 (2003) 6160.
[19] D. Pantarotto, J.P. Briand, M. Prato, A. Bianco, Chem. Commun. (2004) 16.
[20] L.E. Valenti, P.A. Fiorito, C.D. Garcia, C.E. Giacomelli, J. Colloid Interface Sci. 307
(2007) 349.
[21] M.F. Mora, C.E. Giacomelli, C.D. Garcia, Anal. Chem. 81 (2009) 1016.
[22] J.-W. Shen, T. Wu, Q. Wang, Y. Kang, Biomaterials 29 (2008) 3847.
[23] S. Garibaldi, C. Brunelli, V. Bavastrello, G. Ghigliotti, C. Nicolini, Nanotechnology 17
(2006) 391.
[24] D. Pantarotto, C.D. Partidos, J. Hoebeke, F. Brown, E. Kramer, J.P. Briand, S. Muller,
M. Prato, A. Bianco, Chem. Biol. 10 (2003) 961.
[25] A.A. Shvedova, E.R. Kisin, R. Mercer, A.R. Murray, V.J. Johnson, A.I. Potapovich, Y.Y.
Tyurina, O. Gorelik, S. Arepalli, D. Schwegler-Berry, A.F. Hubbs, J. Antonini, D.E.
Evans, B.-K. Ku, D. Ramsey, A. Maynard, V.E. Kagan, V. Castranova, P. Baron, Am. J.
Physiol. Lung Cell. Mol. Physiol. (2005) 00084.2005 2005.
[26] T. Belin, F. Epron, Mater. Sci. Eng. B 119 (2005) 105.
[27] H. Fujiwara, Spectroscopic Ellipsometry. Principles and Applications, J. Wiley &
Sons, West Sussex, England, 2007.
[28] T.M. Barnes, J. van de Lagemaat, D. Levi, G. Rumbles, T.J. Coutts, C.L. Weeks, D.A.
Britz, I. Levitsky, J. Peltola, P. Glatkowski, Phys. Rev. B 75 (2007) 23541001.
[29] R.P. Shrestha, D. Yang, E.A. Irene, Thin Solid Films 500 (2006) 252.
[30] J. Zhao, X. Chen, J.R.H. Xie, Anal. Chim. Acta 568 (2006) 161.
[31] Z. Wu, Z. Chen, X. Du, J.M. Logan,J. Sippel, M. Nikolou, K. Kamaras, J.R. Reynolds, D.B.
Tanner, A.F. Hebard, A.G. Rinzler, Science 305 (2004) 1273.
[32] H.I. Elim, W. Ji, G.H. Ma, K.Y. Lim, C.H. Sow, C.H.A. Huan, Appl. Phys. Lett. 85 (2004)
1799.
[33] G. Fanchini, S. Miller, B.B. Parekh, M. Chhowalla, Nano Lett. 8 (2008) 2176.
[34] R.B. Weisman, S.M. Bachilo, D. Tsyboulski, Appl. Phys. A 78 (2004) 1111.
[35] X. Liu, T. Pichler, M. Knupfer, M.S. Golden, J. Fink, H. Kataura, Y. Achiba, Phys. Rev. B
66 (2002) 045411.
[36] S. Tasaki, K. Maekawa, T. Yamabe, Phys. Rev. B 57 (1998) 9301.
[37] Y. Murakami, E. Einarsson, T. Edamura, S. Maruyama, Carbon 43 (2005) 2664.
[38] K. Spaeth, G. Kraus, G. Gauglitz, Fresenius' J. Anal. Chem. 357 (1997) 292.
[39] U. Stange, M.G. Hutchins, N. Groome, L. Tarassenko, Biomaterials 9 (1988) 58.
[40] T. Kawanishi, Opt. Commun. 186 (2000) 251.
[41] Z. Fu, M.M. Santore, Colloids Surf. A 135 (1998) 63.
[42] S. Lousinian, S. Kassavetis, S. Logothetidis, Diamond Relat. Mater. 16 (2007) 1868.
[43] S. Lousinian, S. Logothetidis, Microelectron. Eng. 84 (2007) 479.
[44] S. Lousinian, S. Logothetidis, A. Laskarakis, M. Gioti, Biomol. Eng. 24 (2007) 107.
[45] K. Matsumoto, H. Maeda, Y. Kawaguchi, K. Takahashi, M. Aoyama, T. Yamaguchi, K.
Postava, Thin Solid Films 455–456 (2004) 339.
[46] E.T. Arakawa, S.M. Dolfini, J.C. Ashley, M.W. Williams, Phys. Rev. B 31 (1985) 8097.
[47] E.D. Palik, Handbook of Optical Constants of Solids, Academic Press, New York, NY,
1997.
[48] H.I. Elim, W. Ji, G.H. Ma, K.Y. Lim, C.H. Sow, C.H.A. Huan, Appl. Phys. Lett. 85 (2004)
1799.
[49] M.Y. Sfeir, F. Wang, L. Huang, C.-C. Chuang, J. Hone, S.P. O'Brien, T.F. Heinz, L.E.
Brus, Science 306 (2004) 1540.
[50] F. Wang, G. Dukovic, L.E. Brus, T.F. Heinz, Science 308 (2005) 838.
[51] F. Wang, D.J. Cho, B. Kessler, J. Deslippe, P.J. Schuck, S.G. Louie, A. Zettl, T.F. Heinz,
Y.R. Shen, Phys. Rev. Lett. 99 (2007) 227401.
[52] S. Kilina, S. Tretiak, S.K. Doorn, Z. Luo, F. Papadimitrakopoulos, A. Piryatinski, A.
Saxena, A.R. Bishop, Proc. Natl. Acad. Sci. 105 (2008) 6797.
[53] E. Lioudakis, C. Kanari, A. Othonos, I. Alexandrou, Diamond Relat. Mater. 17 (2008)
1600.
[54] H. Mizutani, I. Miyahara, K. Hirotsu, Y. Nishina, K. Shiga, C. Setoyama, R. Miura,
J. Biochem. 120 (1996) 14.
[55] J.A. De Feijter, J. Benjamins, F.A. Veer, Biopolymers 17 (1978) 1759.
[56] B. Lassen, M. Malmsten, J. Colloid Interface Sci. 179 (1996) 470.
[57] R. Kurrat, J.E. Prenosil, J.J. Ramsden, J. Colloid Interface Sci. 185 (1997) 1.
[58] C.E. Giacomell i, M.J. Esplandiu, P.I. Ortiz, M.J. Avena, C.P. De Pauli, J. Colloid
Interface Sci. 218 (1999) 404.
[59] M. Vinnichenko, R. Gago, N. Huang, Y.X. Leng, H. Sun, U. Kreissig, M.P. Kulish, M.F.
Maitz, Thin Solid Films 455–456 (2004) 530.
3959H. Soetedjo et al. / Thin Solid Films 518 (2010) 3954–3959