Solar Energy Materials & Solar Cells 157 (2016) 652–659

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells

journal homepage: www.elsevier.com/locate/solmat

An investigation of thermal stability of carbon nanofluids for solar

thermal applications
Sara Mesgari a,n, Robert A. Taylor a,b, Natasha E. Hjerrild b, Felipe Crisostomo a,
Qiyuan Li a, Jason Scott c
a
School of Mechanical and Manufacturing Engineering, University of New South Wales, Sydney, Australia
b
School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, Australia
c
Particles and Catalysis Research Group, School of Chemical Engineering and Industrial Chemistry, University of New South Wales, Sydney, Australia

art ic l e i nf o a b s t r a c t

Article history: Carbon nanofluids are engineered materials with controllable thermal and optical properties. Stable, high
Received 25 January 2016 temperature operation ( 420) of these fluids would enable them to improve upon – and eventually
Received in revised form replace – pure fluids in many important commercial and industrial applications including applications in
17 May 2016
solar thermal collectors. To date, however, much of the nanofluids research focuses on low temperature
Accepted 21 July 2016
( o100 °C) applications and testing. For solar thermal collector applications, carbon nanofluids are
uniquely well-suited due to their high absorptivity over the entire solar spectral range. This study pushes
Keywords: well beyond the 100 °C mark by conducting a range of experiments to identify appropriate base fluids
Solar thermal collectors and functionalization methods to produce stable carbon nanotube (CNT)-based nanofluid dispersions at
Carbon nanotubes
temperatures of up to 250 °C to ensure their suitability for industrial heating applications (typically
Nanofluids
100–250 °C). Different forms of CNTs including, single-walled carbon nanotubes, double-walled carbon
Thermal stability
nanotubes and multi-walled carbon nanotubes were chemically functionalized to obtain stable disper-
sions in water, glycol and Therminol (a synthetic heat transfer oil). The stability of chemically functio-
nalized carbon nanotube dispersions at different temperatures, 20, 80, 100, 150, 200 and 250 °C, was
investigated. The results of broadband UV–VIS–NIR spectroscopy showed no agglomeration in mildly
oxidized multi-walled carbon nanotubes dispersed in Therminol when heated to 250 °C, highlighting this
low-cost composite medium as a potential candidate for use in high temperature nanofluid-based solar
thermal collectors.
& 2016 Elsevier B.V. All rights reserved.

1. Introduction [14–18]. The advantage of the direct absorption of solar thermal

Nanofluids have increasingly attracted attention for use in solar
thermal applications as materials which can be tailored to provide
superior optical and thermos-physical properties [1]. Nanofluids
are desirable for solar thermal application because they can serve
as both the heat transfer fluid and as the solar absorber – effi-
ciently converting sunlight to heat [2–8]. Typical heat transfer
fluids used in conventional solar thermal collectors include water,
molten salts and synthetic heat transfer fluids which are, for the
most part, optically transparent [9,10]. By adding small amounts of
nanoparticles (generally { 0.1% by mass), transparent fluids can
be redesigned to absorb part, or all, of the solar spectrum [11–13].
As compared to conventional collectors, up to a 10% increase in
the efficiency has been reported through the use of nanofluids
n
Corresponding author.
E-mail address: s.hagh@unsw.edu.au (S. Mesgari).
energy is that it avoids intermediate heat transfer steps and
enables the highest temperatures to be generated inside the
working fluid, rather than on the outside surface, which is typical of
conventional surface-based solar thermal collectors [15,19–21].
Apart from taking advantage of the improved heat transfer config-
uration, volumetric absorption using nanofluids can be engineered
to have extremely high solar-weighted absorbance (Fig. 1).
However, the instability of nanofluids at the elevated tem-
peratures required for solar thermal applications and the resulting
clogging of pumps and valves by agglomerated particles have
limited the use of nanofluids to low temperature applications
[16,19]. A number of nanoparticles, including carbon nanoparticles
and metal nanoparticles have been investigated for use in solar
thermal collectors [5,6,16,21–27]. It has been found that dispersing
less than 0.1% volume fraction of nanoparticles can significantly
enhance the solar radiation absorption capability of the base fluids
by up to 7 orders of magnitude [5]. Amongst these particle options,

http://dx.doi.org/10.1016/j.solmat.2016.07.032
0927-0248/& 2016 Elsevier B.V. All rights reserved.
 S. Mesgari et al. / Solar Energy Materials & Solar Cells 157 (2016) 652–659
Fig. 1. A prototype nanofluid-based solar thermal absorber.
carbon-based materials, particularly carbon nanotubes (CNTs),
provide the highest spectral absorptivity (particularly on a per unit
mass basis) over the entire solar range [2,4]. Carbon nanotubes are
present in different forms; single-walled carbon nanotubes
(SWCNTs), double-walled carbon nanotubes (DWCNTs), and multi-
walled carbon nanotubes (MWCNTs). SWCNTs are formed by
wrapping a one-atom-thick layer of graphene into a seamless cy-
linder while DWCNTs and MWCNT are formed by concentrically
wrapping two and multiple layers of graphite, respectively [28].
Although there remain many engineering issues to solve, CNTs
are likely to play a central role in developing nanofluid-based solar
thermal collectors [29–31]. To date, however, applications have
been impeded by the low dispersibility of CNTs in base fluids
[32,33]. Agglomeration of the CNTs in a nanofluid-based solar
thermal collector will result in fouling, clogging, and a consider-
able reduction in the absorbance of incoming solar rays [30,34].
Dispersing CNTs in the base fluid is challenging due to their strong
inter-particle interaction arising from Van der Waals interaction
and their hydrophobic nature [35]. To date, a number of physical
and chemical methods have been developed to achieve stable
dispersions of CNTs [36–49]. Unfortunately, prior research on
dispersion of nanotubes has focused predominantly on dispersion
stability at temperatures that are considerably lower than those
required in typical solar thermal collectors [30,34,50]. While
physical methods, including ultra-sonication and the use of sur-
factants, can effectively disperse CNTs at low temperatures, the
stability of the resulting dispersion has been found to decrease
considerably as temperatures approach (or exceed) 100 °C
[30,34,50,51]. Chemical functionalization has been also used
widely in the literature to produce highly stable solutions at room
temperature [52,53]. During chemical functionalization, functional
groups (such as COOH, OH and COH) are attached to CNTs by
653
treating nanotubes with strong oxidants such as sulfuric acid
(H2SO4) and nitric acid (HNO3) or by mild oxidation using po-
tassium persulfate (K2S2O8) [37,42,50,54–57]. However, the suit-
ability of these methods is unknown for: i) dispersing CNTs in
nonpolar solvents (e.g. glycols and heat transfer oils) and ii)
maintaining a stable dispersion at elevated temperatures. These
issues have not yet been fundamentally or systematically in-
vestigated. Developing stable CNT-based nanofluids for use in high
temperature applications requires identifying the optimal combi-
nation of base fluid, type of CNTs, and dispersion agents (i.e. sur-
factants and/or functional groups) for the desired operating tem-
perature range. To the best of our knowledge, no systematic study
exists which investigates the thermal stability of different types of
CNT solutions at the elevated temperatures. In the present study,
the thermal stability of twenty seven different nanofluids was
probed by varying: a) the type of CNTs (i.e. SWCNTs, DWCNTs and
MWCNTs); b) the dispersion method (i.e. dispersion by surfac-
tants, acid functionalization and base functionalization); and c) the
type of base fluid (i.e. Therminol 55(TH55), propylene glycol (PG)
and water). Thermal stability tests up to 250 °C were conducted to
investigate the suitability of these nanofluids for use in high
temperature applications. The main target temperature for this
study was selected as 250 °C to ensure the suitability of nanofluids
for industrial process heat applications (in the temperature range
of 100–250 °C), a key emerging market for solar energy which is
currently dominated by gas (and sometimes electricity) [58]. Ad-
ditionally, due to importance of optical properties in solar thermal
applications, UV–vis–NIR spectroscopy was used as a sensitive
measurement of their stability, which is directly applicable to their
optical property application in solar thermal collectors.
2. Experimental
2.1. Materials
Multi-walled carbon nanotubes and double-walled carbon na-
notubes were purchased from Sigma Aldrich and used without
further purification. Arc discharge single-walled carbon nanotubes
(AP-SWCNTs) were purchased from Carbon Solutions, Inc. (River-
side, CA, USA). Therminol 55 and propylene glycol (PG) were
purchased from Tru-Blu Oil Australia Pty Ltd. Nitric acid (HNO3),
sulfuric acid (H2SO4), potassium persulfate (KPS, K2S2O8), po-
tassium hydroxide (KOH) and sodium dodecylbenzenesulfonate
(SDBS) were purchased from Sigma Aldrich and used without
further purification.
2.2. Functionalization of CNTs
2.2.1. Chemical functionalization by acid treatment
Acid treatment has been shown to covalently graft oxygenated
functionalities, such as carboxylic groups, onto the surface of CNTs.
The acid functionalization approach used in this study involved
adding 100 mg of each type of CNTs (SWCNTs, DWCNTs and
MWCNTs) to 30 ml of a mixture of H2SO4 and HNO3 in a volu-
metric ratio of 1:3 and stirring for 24 h at 90 °C. The acid treated
CNTs were then collected by vacuum filtration using a polytetra-
fluoroethylene (PTFE) filter paper and washed repeatedly with
water (8–10 times) until a pH of 6 was reached. The solids were
then dried at 90 °C overnight.
2.2.2. Chemical functionalization by KPS
Under an alkaline condition, KPS oxidizes pristine nanotubes by
introducing functional groups such as carboxylic potassium car-
boxylate (–COOK), carbonyl (–C ¼O) and hydroxyl (–C–OH) groups
onto the surface of the CNTs. To achieve mild oxidization by KPS,
654 S. Mesgari et al. / Solar Energy Materials & Solar Cells 157 (2016) 652–659
40 mg of each type of CNTs (SWCNTs, DWCNTs, and MWCNTs)
were added to 50 ml of deionized water and dispersed using bath
sonication for 60 mins. A mass of 0.45 g of KPS was then added to
the flask and the pH of the reaction system was adjusted to 13 by
adding concentrated potassium hydroxide (KOH) solution. The
mixture was stirred for 3 h with a magnetic stirrer at 85 °C. After
cooling to room temperature, the solution was diluted with water
and bath sonicated. After sonication, the solution was centrifuged
at 3000 rpm for 20 min. The supernatant solution was then col-
lected and filtered using a polytetrafluoroethylene (PTFE) filter
paper and washed with water (12–15 times) to neutrality. The
solids were then dried at 80 °C overnight.
2.3. Preparation of functionalised CNT Suspensions
Dispersion of functionalized CNTs involved adding 1.270.1 mg
of chemically functionalized tubes to 15 ml of the base fluid so-
lution (deionized water (DI), PG and TH55). Dispersion was aided
by probe sonication (Misonix ultrasonic liquid processors) at 20 W
in a water-ice bath for 20 min. The nanofluids were prepared with
a CNT concentration of 80 70.1 mg/L, which was identified after
multiple trials as the optimal concentration to achieve a balance
between solar absorption optical depth and stability. In general, an
increase in CNT concentration tends to increase its solar absorp-
tion at the expense of thermal stability and dispersion quality [2],
thus identifying the optimal CNT concentration for a particular
design configuration is crucial.
2.4. Preparation of CNT suspensions with surfactant
The dispersion of CNTs using surfactants was also investigated
as an alternative dispersion technique. SDBS was selected as re-
presentative surfactant due to its reported effectiveness and
common usage as a CNT dispersant in previous studies. Further-
more, SDBS has a relatively high melting point, 4300 °C, making
it a potential candidate for high temperature applications. During
surfactant dispersion, 1.270.1 mg of each type of CNTs (SWCNTs,
DWCNTs, and MWCNTs) was dispersed in 15 ml of the base fluid
solution (DI, PG and TH55) containing 1 wt% of SDBS using probe
sonication at 20 W in a water-ice bath for 20 min.
In the subsequent discussion, acid-treated CNTs will be denoted
as ‘A-CNTs’, KPS-treated CNTs as ‘K-CNTs’, surfactant CNT disper-
sions as ‘S-CNT’ and heated samples as x-CNTs-y-z (with x de-
noting the treatment method, y denoting the base fluid, and z
denoting the temperature of heating (before heating (pristine),
80 °C, 100 °C, 150 °C, 200 °C and 250 °C)).
2.5. Characterization
Thermogravimetric analysis (TGA) was performed on a TA In-
struments TGA Q5000 instrument under air at a flow rate of
200 ml min  1. In this study, 2 mg of the samples were heated
from 20 °C to 800 °C at 10 °C min  1. Raman spectroscopy was
conducted using a Renishaw inVia Raman microscope equipped
with a 514 nm laser. Fourier transform infrared (FTIR) spectro-
scopy was performed using a Perkin-Elmer Spotlight 400 instru-
ment in the transmittance mode across the spectral range
500–4000 cm  1, with a resolution of 1 cm  1. Dynamic Light
Scattering (DLS) (Malvern Nano-ZetaSizer) was used to measure
the particle size distribution of KPS-treated MWCNTs before and
after heating. Ultraviolet-visible-near infrared (UV–vis–NIR) ab-
sorption spectra were measured using a Shimadzu UV-3600
spectrophotometer. The solar weighted absorbance (SWA) was
calculated using the following equation:
1400
∫280 I ( λ)·α ( λ) dλ
SWA= 1400
∫280 I ( λ) dλ
where I is the direct component of the solar spectrum using the
standard solar spectrum ASTM G173 considering only the Direct
(þcircumsolar) component and α is the measured spectral ab-
sorbance of the fluids. The numerator corresponds to the power
per unit of area that can be absorbed by the fluid and the de-
nominator to the total incident power per unit of area reaching the
receiver.
The samples were analyzed by UV–vis–NIR spectrophotometer
before and after heating. The optical absorbance ( 70.00001) was
recorded from 350 nm to 1400 nm with 1 nm intervals. The ab-
sorbance at wavelengths greater than 1400 nm was not recorded
due to the strong absorption bands at such wavelengths.
2.6. Thermal stability tests
The thermal stability of the various dispersions was assessed by
heating at a set temperature for 1 h. In the case of DI-based dis-
persions the temperature was set at 80 °C for DI samples to ac-
count for the boiling point of water and four different tempera-
tures of 100 °C, 150 °C, 200 °C and 250 °C for TH55 and PG sam-
ples. The samples were then cooled to room temperature.
3. Results
The functionalized nanotubes were analyzed using TGA, Ra-
man, and FTIR to evaluate the extent of functionalization through
quantifying the functional groups attached to MWCNTs. As seen in
Fig. 2, all the functionalised CNT samples exhibited a multi-step
decomposition profile between 20 °C and 800 °C. The first stage of
decomposition, between 20 °C and 200 °C, is due to the evapora-
tion of absorbed water; the second stage, between 200 °C and
500 °C, is due to decomposition of the attached functional groups
and; the final decomposition stage, occurring at above 500 °C, is
attributed to nanotube decomposition. The quantity of functional
units was estimated from TGA to be 52 wt% for K-CNTs and 38 wt%
for A-CNTs, indicating a higher level of functionalization for
K-CNTs than A-CNTs which is in line with observations of previous
studies [54,56].
The extent of functionalization was also examined using Raman
spectroscopy. The ratio of the D peak to the G peak (D/G) is related
to the extent of MWCNTs functionalization [59]. As shown in Fig. 3
and Table 1, the D/G ratio is considerably higher for K-CNTs than
100
80
60
40
20
0
100 200 300 400 500 600 700 800
Fig. 2. TGA profiles of acid-treated MWCNTs, KPS-treated MWCNTs and pristine
MWCNTs.
 S. Mesgari et al. / Solar Energy Materials & Solar Cells 157 (2016) 652–659 655

10000 2.0
D
9000
G
8000
1.5
7000 G'
6000

5000
1.0
4000

3000

2000 0.5

1000

0
500 1000 1500 2000 2500 3000 0.0
( )
Fig. 5. The change in solar weighted absorbance of acid-treated (A-CNT), KPS-
Fig. 3. Raman spectra of acid-treated MWCNTs, KPS-treated MWCNTs and pristine treated (K-CNT) and SDBS-assisted dispersed CNTs (S-CNT) in DI after heating at
MWCNTs, collected using a 514 nm laser. 80 °C.

of the MWCNTs changed its IR spectra slightly. For acid treated

Table 1 MWCNTs, the peak at 1633 cm  1 is attributed to the stretching
The ratio of the disorder mode (D peak) and the tangential mode (G
mode of carboxylic groups. In addition, the higher intensity of the
peak) of acid-treated, KPS-treated, and pristine MWCNTs.
OH band for acid-treated versus pristine samples is due to O–H
Sample D/G ratio stretching from the added carboxyl groups [60]. On the other
hand, the OH and C–O–C bands were more pronounced in KPS-
MWCNT-Pristine 0.8231 treated MWCNTs than both the acid-treated MWCNTs and pristine
A-MWCNT 1.0531
K-MWCNT 1.2445
samples due to the introduction of hydroxyl groups onto the sur-
face of tubes from the KPS treatment [56]. The functionalization
characteristic peaks are more intense in K-CNTs than the A-CNTs,
which indicate a higher extent of functionalization for K-CNTs than
A-CNTs.
100 The thermal stability of the nanofluids was evaluated at dif-
ferent temperatures using UV–vis–NIR spectroscopy. Fig. 5 shows
95
the changes in solar weighted absorbance (SWA) of the MWCNTs,
90
DWCNTs and SWCNTs dispersed in DI by SDBS, KPS functionali-
zation and acid functionalization before and after heating at 80 °C
85 for 1 h. An increase in the temperature to 80 °C resulted in a
negligible variation ( o1%) in the SWA of KPS and acid-treated
80 nanotubes dispersed in DI. This confirms the stability of KPS and
acid treatment–induced functional groups on the surface of CNTs
75
1633
1116
at temperatures of up to 80 °C. In addition, as shown in Fig. 5, SDBS
70 was found to be an effective surfactant for achieving a stable
dispersion of CNTs in DI, exhibiting stability up to 80 °C. Further-
65 3446
more, comparing the relative increase in the absorption of differ-
ent nanofluids (Supplementary Figs. S1A, S1B and S1C) indicated
4000 3500 3000 2500 2000 1500 1000 500
that MWCNTs were dispersed more effectively than DWCNTs and
SWCNTs. This may be attributed to the weaker Van der Waals
Fig. 4. FTIR spectra of (a) pristine, (b) A-MWCNTs and (c) K-MWCNTs samples. forces between individual MWCNTs as compared to DWCNTs and
SWCNTs [61].
for the pristine sample and A-CNTs. This confirms that base The temperature induced changes in the solar weighted ab-
treatment leads to an increase in extent of functionalization of sorbance of these dispersions in PG for acid functionalization, KPS
CNTs. The Raman results are also in agreement with the TGA re- functionalization, and SDBS surfactant are compared in Fig. 6. The
sults, which indicated a higher extent of functionalization for change in SWA of K-CNT-PG-150 °C (MWCNTs, DWCNTs, SWCNTs)
compared to that of the K-CNT-PG-100 °C and unheated K-CNT-
K-CNTs than A-CNTs.
PG-Pristine samples was found to be negligible (  0.2%), corro-
Fourier transform infrared (FTIR) spectroscopy was also per-
borating the thermal stability of KPS-functionalized CNTs dis-
formed to investigate the type of functional groups to MWCNTs.
persed in PG. The thermal stability of KPS functionalized CNT
Fig. 4 shows the FTIR spectra of pristine (a), acid treated (b), and
dispersions in PG was also confirmed by the negligible variations
KPS-treated (c) MWCNTs. The vibrational band at 1633 cm  1 is
in their UV–vis–NIR spectra after heating to 150 °C (Supplementary
attributed to the C ¼C bonds of the carbon skeleton of MWCNTs. Fig. S2B). On the contrary, acid functionalised CNTs and SDBS-as-
The peak at 3446 cm  1 is attributed to the O–H stretching of the sisted dispersed CNTs were not as stable with a considerable
alcoholic and phenolic groups as well as the presence of moisture increase (up to 11%) in the SWA being observed at 150 °C as
in CNT samples. The peaks at 1116 cm  1 can be assigned to C–O–C compared to 100 °C and unheated pristine samples. This together
stretching groups. The alcoholic and phenolic groups in pristine with a clear shift in the UV–vis–NIR of samples (Supplementary
MWCNTs are likely due to the purification process by the supplier. Figs. S2A and S2C) indicates that agglomeration of the acid func-
As is shown in Fig. 4, introducing functional groups on the surface tionalized and SDBS-assisted dispersed CNTs begins above 100 °C
656 S. Mesgari et al. / Solar Energy Materials & Solar Cells 157 (2016) 652–659
20
18
16
14
12
10
8 after heating (K-MWCNT-TH55-200 °C).
6
4 TH55-250 °C at wavelengths r700 nm (The peaks at 1200 and
2 1400 nm are due to Therminol 55). This, together with an observed
0
Fig. 6. The change in solar weighted absorbance of acid-treated (A-CNT), KPS-
treated (K-CNT) and SDBS-assisted dispersed CNTs (S-CNT) in PG after heating at
either 100 °C or 150 °C.
20
18
16
14
12
10
8
6 MWCNT dispersions in TH55 up to 250 °C.
4 After the thermal stability tests, KPS-functionalized MWCNT
2
0 centrated solar thermal collector (Supplementary Figs. S5 and S6)
Fig. 7. Change in solar weighted absorbance of acid-treated (A-CNT), KPS-treated
(K-CNT) and SDBS-assisted dispersed CNTs (S-CNT) in Therminol 55 after heating at
either 100 °C or 150 °C.
and becomes significant at 150 °C in PG.
Fig. 7 shows the temperature induced changes in solar
weighted absorbance of CNT dispersions in TH55 achieved using
KPS functionalization, acid functionalization and SDBS surfactant.
The results indicate effective dispersion of acid- and KPS-treated
MWCNTs in TH55 was achieved, with KPS-treated MWCNTs giving
little apparent variations (  0.8%) up to 150 °C in TH55. In contrast,
the results showed a clear change in the SWA of the acid-treated
MWCNTs dispersed in TH55 with an increase in temperature.
Furthermore, as shown in Fig. S3 (Supplementary Fig. S3), the UV–
vis–NIR spectrum of A-MWCNT-TH55-150 °C and A-MWCNT-
TH55-100 °C shifted considerably compared to the unheated na-
nofluids. This implies instability of the acid-treated MWCNTs at
100 °C and 150 °C in TH55. Surprisingly, the same trend with re-
spect to dispersion stability at elevated temperature was not ob-
served for KPS-treated DWCNTs and SWCNTs samples. As shown
in Fig. S3, acid-treated and KPS-treated CNTs did not disperse well
nor did they remain stable in TH55. Better dispersion was achieved
using MWCNTs, which is consistent with the observations of
previous studies. In general, it can be said that better dispersion
was found with MWCNTs than with DWCNTs and SWCNTs, which
can be attributed to the tendency of SWCNTs/DWCNTs to pack into
bundles [62] (see the inset of Fig. S3B). In addition, considerable
agglomeration of the CNTs was observed in SDBS-assisted CNT
dispersions in TH55 at temperatures of 100 °C and 150 °C. The
poor thermal stability of SDBS-assisted CNT dispersion in TH55
was also confirmed through  15% shift in its solar weighted
absorbance after heating (Fig. 7).
Based on the results of these thermal stability tests and particle
size measurements (Supplementary Figs. S4), KPS-treated
MWCNTs appeared to be the best candidate for high temperature
applications, so it was subject to further high temperature thermal
stability tests at temperatures up to 250 °C. As shown in Fig. 8a,
KPS-treated MWCNTs dispersed in TH55 at 200 °C displayed good
stability, since negligible variation in the UV–vis–NIR was observed
On the other hand, minor changes were observed between the
UV–vis–NIR spectra of K-MWCNT-TH55-Pristine and K-MWCNT-
change in the color of the K-MWCNT-TH55-250 °C samples, in-
dicates thermal degradation of TH55 in the presence of oxygen at
250 °C [63]. To confirm the contribution of TH55 thermal de-
gradation to the clear shift in absorbance of K-MWCNT-TH55-
250 °C, a control thermal stability test was performed on pure
TH55. As seen in Fig. 8b, heating TH55 to 200 °C (TH55-200 °C)
and 250 °C (TH55-250 °C) resulted in complete disappearance of
the humps at around 500 nm in the UV–vis–NIR spectra. Also
apparent was a change in the TH55 color (Fig. 8c), indicating that
the degradation of TH55 was the reason for TH55-based dispersion
instability at temperatures higher than 200 °C. Since these fluids
are likely to be used in a closed system (without oxygen), KPS-
treated MWCNTs dispersed in TH55 was purged with nitrogen for
6h to remove dissolved oxygen and then heated to 250 °C. As seen
in Fig. 9a and b, purging with nitrogen led to negligible changes in
both the UV–vis–NIR spectra and the SWA of KPS-functionalized
dispersions in TH55 and pure TH55 (as a control test) were se-
lected for use as volumetric absorber fluids in a prototype con-
to evaluate the nanofluid's performance in a small-scale ‘on-sun’
test. A stagnation test was run to determine the receiver's total
heat loss by exposing it to a known amount of solar radiation
without any heat removal (no fluid flow). Fig. 10 shows the tem-
perature of K-MWCNTs-TH55 and pure Therminol 55 versus time.
As can be seen, K-MWCNTs-TH55 stagnated at a significantly
higher temperature than pure TH55 which suggests that the
K-MWCNTs-TH55 absorb considerably more sunlight than TH55.
Furthermore, the experimental performance of an ITO glass tube
containing nanofluid absorber and Therminol 55 is compared in
Fig. 11. The ITO coating was designed to minimize long wavelength
thermal emission [64]. A multiple regression curve fit was used to
determine efficiency curves for these two absorbers. The effi-
ciencies can be described by Eq. (2) for Therminol 55 and Eq. (3)
for K-MWCNTs-TH55 versus ( T̅ abs  Ta)/Gb[65]. The error in the
evaluated collector efficiency was around 3.9% (including all
sources of errors).
(Tabs
̅ −Ta )
ηth =0. 68 − 3. 4
Gb (2)
(Tabs
̅ −Ta )
ηth =0. 18 − 2. 51
Gb (3)
The optical efficiency of the nanofluid absorber was  3.8 times
higher than pure TH55, indicating that our developed nanofluid is
capable of strong broadband solar absorption.
The results of TGA, Raman, FTIR, DLS, UV–vis–NIR spectroscopy
and the outdoor test presented above confirm the excellent ther-
mal stability of KPS-treated MWCNTs dispersed in DI, PG and TH55
and thus the stability of the KPS-induced functional groups on the
surface of MWCNTs up to 250 °C. The higher thermal stability of
 S. Mesgari et al. / Solar Energy Materials & Solar Cells 157 (2016) 652–659 657

5 5

4 4

3 (c) 3

2 2 6

(b) (b)
1 1
(a)
(a)
(d) (c)
0 0

400 600 800 1000 1200 1400 400 600 800 1000 1200 1400

Fig. 8. (a) UV–vis–NIR spectra of KPS-treated MWCNTs dispersed in Therminol 55; (b) UV–vis–NIR spectra of Therminol 55 base fluid, (c) Photographic images of Therminol
55 base fluid before and after heating to temperatures up to 250 °C.

5
4

4
3

1
1

0 0
400 600 800 1000 1200 1400

Fig. 9. (a) UV–vis–NIR spectra of and (b) changes in solar weighted absorbance for nitrogen-purged KPS-treated MWCNTs dispersed in Therminol 55 before and after heating
at either 200 °C or 250 °C.

KPS-functionalised MWCNTs compared with acid-functionalised
MWCNTs may originate from the greater quantity of COOH func-
tional groups attached to the CNTs through KPS functionalization
as was indicated by the TGA and Raman results. It has been re-
ported that KPS-treated CNTs have a higher potassium carboxylate
(–COOK), carbonyl (–C ¼O) and hydroxyl (–C–OH) functional group
content than acid treated samples [66]. The considerably better
thermal stability of MWCNTs compared to DWCNTs and SWCNTs
may originate from the weaker, by a factor of 6000, Van der Waals
forces between individual MWCNTs [61].
4. Conclusion
The results of a comprehensive experimental study aimed at
developing highly stable nanofluids for use in solar thermal ap-
plications have been presented. The thermal stability of SDBS-
assisted, KPS-functionalised and acid-functionalised SWCNTs,
DWCNTs and MWCNTs dispersed in DI water, propylene glycol,
and Therminol55 heat transfer fluids were investigated at elevated
temperatures of up to 250 °C. A stable nanofluid based on dis-
persion of KPS functionalized MWCNTs in a non-polar solvent
(Therminol 55) was developed to overcome previous low tem-
perature limits of nanofluids and achieve a dispersion which is
stable up to 250 °C. A considerably better thermal stability was
observed for MWCNTs than DWCNTs and SWCNTs based on the
thermal stability tests. Furthermore, the KPS-induced functional
groups on the surface of MWCNTs were found to remain more
stable at higher temperatures of up to 250 °C, rendering KPS
functionalized MWCNT nanofluids suitable for high temperature
applications including solar thermal collectors. The stability of
acid-generated functionalities in PG and TH55 solvents as well as
the stability of SDBS-assisted dispersions were found to decrease
with an increase in temperature from 80 °C to 150 °C. Overall, the
658 S. Mesgari et al. / Solar Energy Materials & Solar Cells 157 (2016) 652–659

180 References

160
[1] P. Phelan, T. Otanicar, R. Taylor, H. Tyagi, Trends and opportunities in direct-
absorption solar thermal collectors, J. Therm. Sci. Eng. Appl 5 (2013) 021003,
140 http://dx.doi.org/10.1115/1.4023930.
[2] N. Hordy, D. Rabilloud, J.-L. Meunier, S. Coulombe, High temperature and long-
120 term stability of carbon nanotube nanofluids for direct absorption solar
thermal collectors, Sol. Energy 105 (2014) 82–90, http://dx.doi.org/10.1016/j.
100 solener.2014.03.013.
[3] R.A. Taylor, P.E. Phelan, T.P. Otanicar, C. a Walker, M. Nguyen, S. Trimble, et al.,
Applicability of nanofluids in high flux solar collectors, J. Renew. Sustain. En-
80 ergy 3 (2011) 023104, http://dx.doi.org/10.1063/1.3571565.
[4] R.A. Taylor, S. Coulombe, T. Otanicar, P. Phelan, A. Gunawan, W. Lv, et al., Small
60 particles, big impacts: A review of the diverse applications of nanofluids, J.
Appl. Phys. 113 (2013) 011301, http://dx.doi.org/10.1063/1.4754271.
40 [5] R.A. Taylor, P.E. Phelan, T.P. Otanicar, R. Adrian, R. Prasher, Nanofluid optical
property characterization: towards efficient direct absorption solar collectors,
Nanoscale Res. Lett. 6 (2011) 225, http://dx.doi.org/10.1186/1556-276X-6-225.
20
[6] J. Liu, Z. Ye, L. Zhang, X. Fang, Z. Zhang, A combined numerical and experi-
0 500 1000 1500 2000 2500
mental study on graphene/ionic liquid nano fluid based direct absorption solar
collector, Sol. Energy Mater. Sol. Cells 136 (2015) 177–186, http://dx.doi.org/
10.1016/j.solmat.2015.01.013.
Fig. 10. Stagnation temperature of K-MWCNT-TH55 nanofluid and Therminol 55 [7] M. Milanese, G. Colangelo, A. Cretì, M. Lomascolo, F. Iacobazzi, A. de Risi,
versus time. Optical absorption measurements of oxide nanoparticles for application as
nanofluid in direct absorption solar power systems – Part II: ZnO, CeO2, Fe2O3
nanoparticles behavior, Sol. Energy Mater. Sol. Cells (2016) 1–6, http://dx.doi.
org/10.1016/j.solmat.2015.12.030.
80 [8] F. Wang, J. Liu, X. Fang, Z. Zhang, Graphite nanoparticles-dispersed paraffin/
water emulsion with enhanced thermal-physical property and photo-thermal
70 performance, Sol. Energy Mater. Sol. Cells 147 (2016) 101–107, http://dx.doi.
org/10.1016/j.solmat.2015.12.013.
[9] T.P. Otanicar, P.E. Phelan, J.S. Golden, Optical properties of liquids for direct
60 absorption solar thermal energy systems, Sol. Energy 83 (2009) 969–977, http:
//dx.doi.org/10.1016/j.solener.2008.12.009.
50 [10] H. Riazi, S. Mesgari, N.A. Ahmed, R.A. Taylor, The effect of nanoparticle mor-
phology on the specific heat of nanosalts, Int. J. Heat Mass Transf. 94 (2016)
40 254–261, http://dx.doi.org/10.1016/j.ijheatmasstransfer.2015.11.064.
[11] F. Crisostomo, R.A. Taylor, A. Mojiri, E.R. Hawkes, D. Surjadi, G. Rosengarten,
Beam splitting system for the development of a concentrating linear fresnel
30
solar hybrid PV/T collector, in: 2013: pp. 1–9.
[12] N.E. Hjerrild, S. Mesgari, F. Crisostomo, J.A. Scott, R. Amal, R.A. Taylor, Hybrid
20 PV/T enhancement using selectively absorbing Ag–SiO 2/carbon nanofluids,
Sol. Energy Mater. Sol. Cells 147 (2016) 281–287, http://dx.doi.org/10.1016/j.
10 solmat.2015.12.010.
[13] F. Crisostomo, R.A. Taylor, D. Surjadi, A. Mojiri, G. Rosengarten, E.R. Hawkes,
0
Spectral splitting strategy and optical model for the development of a con-
0.00 0.05 0.10 0.15 0.20 0.25 centrating hybrid PV/T collector, Appl. Energy 141 (2015) 238–246, http://dx.
doi.org/10.1016/j.apenergy.2014.12.044.
[14] V. Khullar, H. Tyagi, P.E. Patrick, T.P. Otanicar, H. Singh, R. a Taylor, Solar energy
harvesting using nanofluids-based concentrating solar collector, J. Nano-
Fig. 11. Stagnation test efficiency curves for a prototype collector employing technol. Eng. MEd. 3 (2013) 031003, http://dx.doi.org/10.1115/1.4007387.
K-MWCNT-TH55 nanofluid and Therminol 55. [15] Y.L. Hewakuruppu, R.A. Taylor, H. Tyagi, V. Khullar, T. Otanicar, S. Coulombe,
et al., Limits of selectivity of direct volumetric solar absorption, Sol. Energy 114
(2015) 206–216, http://dx.doi.org/10.1016/j.solener.2015.01.043.
[16] T.P. Otanicar, P.E. Phelan, R.S. Prasher, G. Rosengarten, R. a Taylor, Nanofluid-
results of the UV–vis–NIR spectroscopy and solar weighted ab- based direct absorption solar collector, J. Renew. Sustain. Energy 2 (2010)
sorbance presented confirm the important role the functionaliza- 033102, http://dx.doi.org/10.1063/1.3429737.
[17] H. Tyagi, P. Phelan, R. Prasher, Predicted efficiency of a low-temperature na-
tion method plays in achieving carbon-based nanofluids with high nofluid-based direct absorption solar collector, J. Sol. Energy Eng. 131 (2009)
thermal stability. The promising thermal stability of KPS functio- 041004, http://dx.doi.org/10.1115/1.3197562.
nalised MWCNTs nanofluid coupled with its high solar energy [18] T.P. Otanicar, P.E. Phelan, R. a Taylor, H. Tyagi, Spatially varying extinction
coefficient for direct absorption solar thermal collector optimization, J. Sol.
absorption capability makes it an attractive candidate for solar
Energy Eng. 133 (2011) 024501, http://dx.doi.org/10.1115/1.4003679.
thermal applications. [19] V. Khullar, H. Tyagi, N. Hordy, T.P. Otanicar, Y. Hewakuruppu, P. Modi, et al.,
Harvesting solar thermal energy through nanofluid-based volumetric ab-
sorption systems, Int. J. Heat. Mass Transf. 77 (2014) 377–384, http://dx.doi.
org/10.1016/j.ijheatmasstransfer.2014.05.023.
Acknowledgments [20] O. Mahian, A. Kianifar, S. a Kalogirou, I. Pop, S. Wongwises, A review of the
applications of nanofluids in solar energy, Int. J. Heat. Mass Transf. 57 (2013)
582–594, http://dx.doi.org/10.1016/j.ijheatmasstransfer.2012.10.037.
The authors gratefully acknowledge support for this work from
[21] Z. Luo, C. Wang, W. Wei, G. Xiao, M. Ni, Performance improvement of a na-
the Australian Solar Institute and the Australian Renewable Energy nofluid solar collector based on direct absorption collection (DAC) concepts,
Agency (ARENA). Taylor would like to acknowledge the support of Int. J. Heat. Mass Transf. 75 (2014) 262–271, http://dx.doi.org/10.1016/j.
the Australian Research Council (ARC) DECRA Fellowship through ijheatmasstransfer.2014.03.072.
[22] Y. Kameya, K. Hanamura, Enhancement of solar radiation absorption using
(DE160100131). The authors thankfully acknowledge Dr Hewa- nanoparticle suspension, Sol. Energy 85 (2011) 299–307, http://dx.doi.org/
kuruppu's help during outdoor testing. 10.1016/j.solener.2010.11.021.
[23] Z. Meng, D. Wu, L. Wang, H. Zhu, Q. Li, Carbon nanotube glycol nanofluids:
Photo-thermal properties, thermal conductivities and rheological behavior,
Particuology 10 (2012) 614–618, http://dx.doi.org/10.1016/j.partic.2012.04.001.
Appendix A. Supporting information [24] E. Sani, L. Mercatelli, S. Barison, C. Pagura, F. Agresti, L. Colla, et al., Potential of
carbon nanohorn-based suspensions for solar thermal collectors, Sol. Energy
Mater. Sol. Cells 95 (2011) 2994–3000, http://dx.doi.org/10.1016/j.
Supplementary data associated with this article can be found in solmat.2011.06.011.
the online version at http://dx.doi.org/10.1016/j.solmat.2016.07.032. [25] E.P. Bandarra Filho, O.S.H. Mendoza, C.L.L. Beicker, A. Menezes, D. Wen,
 S. Mesgari et al. / Solar Energy Materials & Solar Cells 157 (2016) 652–659
Experimental investigation of a silver nanoparticle-based direct absorption
solar thermal system, Energy Convers. Manag 84 (2014) 261–267, http://dx.
doi.org/10.1016/j.enconman.2014.04.009.
[26] M. Karami, M.A. Akhavan-Behabadi, M. Raisee Dehkordi, S. Delfani, Thermo-
optical properties of copper oxide nanofluids for direct absorption of solar
radiation, Sol. Energy Mater. Sol. Cells 144 (2016) 136–142, http://dx.doi.org/
10.1016/j.solmat.2015.08.018.
[27] Z. Chen, T. Boström, Electrophoretically deposited carbon nanotube spectrally
selective solar absorbers, Sol. Energy Mater. Sol. Cells 144 (2016) 678–683,
http://dx.doi.org/10.1016/j.solmat.2015.10.016.
[28] R.H. Baughman, A. a Zakhidov, W. a de Heer, Carbon nanotubes – the route
toward applications, Science 80 (297) (2002) 787–792, http://dx.doi.org/
10.1126/science.1060928.
[29] N. Hordy, D. Rabilloud, J.-L. Meunier, S. Coulombe, High temperature and long-
term stability of carbon nanotube nanofluids for direct absorption solar
thermal collectors, Sol. Energy 105 (2014) 82–90, http://dx.doi.org/10.1016/j.
solener.2014.03.013.
[30] T.B. Gorji, a a Ranjbar, S.N. Mirzababaei, Optical properties of carboxyl func-
tionalized carbon nanotube aqueous nanofluids as direct solar thermal energy
absorbers, Sol. Energy 119 (2015) 332–342, http://dx.doi.org/10.1016/j.
solener.2015.07.012.
[31] S. Mesgari, S. Coulombe, N. Hordy, R.A. Taylor, Thermal stability of carbon
nanotube-based nanofluids for solar thermal collectors, 19, 2015, pp. 650–653.
[32] L.Y. Yan, W. Li, S. Mesgari, S.S.J. Leong, Y. Chen, L.S. Loo, et al., Use of a chon-
droitin sulfate isomer as an effective and removable dispersant of single-
walled carbon nanotubes, Small 7 (2011) 2758–2768.
[33] Z. Said, R. Saidur, M. a Sabiha, N. a Rahim, M.R. Anisur, Thermophysical
properties of Single Wall Carbon Nanotubes and its effect on exergy efficiency
of a flat plate solar collector, Sol. Energy 115 (2015) 757–769, http://dx.doi.org/
10.1016/j.solener.2015.02.037.
[34] l Karami, M. a Akhavan Bahabadi, S. Delfani, a Ghozatloo, A new application of
carbon nanotubes nanofluid as working fluid of low-temperature direct ab-
sorption solar collector, Sol. Energy Mater. Sol. Cells 121 (2014) 114–118, http:
//dx.doi.org/10.1016/j.solmat.2013.11.004.
[35] R. Bandyopadhyaya, E. Nativ-roth, O. Regev, R. Yerushalmi-rozen, Stabilization
of individual carbon nanotubes in aqueous solutions, Nano (2002) 15–18, http:
//dx.doi.org/10.1021/nl010065f.
[36] L.Y. Yan, W.F. Li, S. Mesgari, S.S.J. Leong, Y. Chen, L.S. Loo, et al., Use of a
chondroitin sulfate isomer as an effective and removable dispersant of single-
walled carbon nanotubes, Small 7 (2011) 2758–2768, http://dx.doi.org/
10.1002/smll.201100708.
[37] Y.P. Sun, K.F. Fu, Y. Lin, W.J. Huang, Functionalized carbon nanotubes: prop-
erties and applications, Acc. Chem. Res. 35 (2002) 1096–1104, http://dx.doi.
org/10.1021/Ar010160v.
[38] R. Scaffaro, A. Maio, S. Agnello, A. Glisenti, Plasma functionalization of mul-
tiwalled carbon nanotubes and their use in the preparation of nylon 6-based
nanohybrids, Plasma Process. Polym. 9 (2012) 503–512, http://dx.doi.org/
10.1002/ppap.201100140.
[39] H. Pan, L. Liu, Z.-X. Guo, L. Dai, F. Zhang, D. Zhu, et al., Carbon nanotubols from
mechanochemical reaction, Nano Lett. 3 (2003) 29–32, http://dx.doi.org/
10.1021/nl025856b.
[40] O.V. Kharissova, B.I. Kharisov, E.G. de Casas Ortiz, Dispersion of carbon na-
notubes in water and non-aqueous solvents, RSC Adv. 3 (2013) 24812, http:
//dx.doi.org/10.1039/c3ra43852j.
[41] C. a Dyke, J.M. Tour, Solvent-free functionalization of carbon nanotubes, J. Am.
Chem. Soc. 125 (2003) 1156–1157, http://dx.doi.org/10.1021/ja0289806.
[42] S. Mesgari, Y.F. Poon, Y. Wang, Y.X. Thong, J. Wang, M.B. Chan-Park, Polymer
removal from electronic grade single-walled carbon nanotubes after gel
electrophoresis, J. Mater. Chem. C 1 (2013) 6813 http://www.scopus.com/in
ward/record.url?eid¼2-s2.0-84885126867&partnerID ¼tZOtx3y1.
[43] L. Vaisman, G. Marom, H.D. Wagner, Dispersions of surface-modified carbon
nanotubes in water-soluble and water-insoluble polymers, Adv. Funct. Mater.
16 (2006) 357–363, http://dx.doi.org/10.1002/adfm.200500142.
[44] J.A. Kim, D.G. Seong, T.J. Kang, J.R. Youn, Effects of surface modification on
rheological and mechanical properties of CNT/epoxy composites, Carbon 44
(2006) 1898–1905, http://dx.doi.org/10.1016/j.carbon.2006.02.026.
[45] N. Mureau, E. Mendoza, S.R.P. Silva, K.F. Hoettges, M.P. Hughes, In situ and real
659
time determination of metallic and semiconducting single-walled carbon
nanotubes in suspension via dielectrophoresis, Appl. Phys. Lett. 88 (2006).
[46] N. Hordy, S. Coulombe, J.-L. Meunier, Plasma functionalization of carbon na-
notubes for the synthesis of stable aqueous nanofluids and poly(vinyl alcohol)
nanocomposites, Plasma Process. Polym. 10 (2013) 110–118, http://dx.doi.org/
10.1002/ppap.201200075.
[47] S. Mesgari, Y.F. Poon, L.Y. Yan, Y. Chen, L.S. Loo, Y.X. Thong, et al., High se-
lectivity cum yield gel electrophoresis separation of single-walled carbon
nanotubes using a chemically selective polymer dispersant, J. Phys. Chem. C.
116 (2012) 10266–10273.
[48] A.K. Sundramoorthy, S. Mesgari, J. Wang, R. Kumar, M.A. Sk, S.H. Yeap, et al.,
Scalable and effective enrichment of semiconducting single-walled carbon
nanotubes by a dual selective naphthalene-based AzO dispersant, J. Am. Chem.
Soc. 135 (2013) 5569–5581, http://dx.doi.org/10.1021/ja312282g.
[49] S. Mesgari, A.K. Sundramoorthy, L.S. Loo, M.B. Chan-Park, Gel electrophoresis
using a selective radical for the separation of single-walled carbon nanotubes,
Faraday Discuss 173 (2014) 351–363, http://dx.doi.org/10.1039/C4FD00092G.
[50] a Ghadimi, R. Saidur, H.S.C. Metselaar, A review of nanofluid stability prop-
erties and characterization in stationary conditions, Int. J. Heat Mass Transf. 54
(2011) 4051–4068, http://dx.doi.org/10.1016/j.ijheatmasstransfer.2011.04.014.
[51] D. Wen, Y. Ding, Effective thermal conductivity of aqueous suspensions of
carbon nanotubes (carbon nanotube nanofluids), J. Thermophys. Heat Transf.
18 (2004) 481–485, http://dx.doi.org/10.2514/1.9934.
[52] B. Zhao, B. Zhao, H. Hu, H. Hu, A. Yu, A. Yu, et al., Synthesis and character-
ization of water soluble single-walled carbon nanotube graft copolymers,
Science 80- (2005) 8197–8203.
[53] I. Kalinina, K. Worsley, C. Lugo, S. Mandal, E. Bekyarova, R.C. Haddon, Synth-
esis, dispersion, and viscosity of poly (ethylene glycol) – functionalized water-
soluble single-walled, Carbon Nanotub. (2011) 1246–1253.
[54] L. Zhang, Q. Ni, Y. Fu, T. Natsuki, One-step preparation of water-soluble single-
walled carbon nanotubes, Appl. Surf. Sci. 255 (2009) 7095–7099.
[55] L. Chen, H. Xie, Surfactant-free nanofluids containing double- and single-
walled carbon nanotubes functionalized by a wet-mechanochemical reaction,
Thermochim. Acta 497 (2010) 67–71, http://dx.doi.org/10.1016/j.
tca.2009.08.009.
[56] L. Chen, H. Xie, Y. Li, W. Yu, Surface chemical modification of multiwalled
carbon nanotubes by a wet-mechanochemical reaction, J. Nanomater. 2008
(2008), http://dx.doi.org/10.1155/2008/783981.
[57] L. Chen, H. Xie, Y. Li, W. Yu, Nanofluids containing carbon nanotubes treated by
mechanochemical reaction, Thermochim. Acta 477 (2008) 21–24, http://dx.
doi.org/10.1016/j.tca.2008.08.001.
[58] A. Ilyes, B. Hassine, A. Helmke, S. Heß, P. Krummenacher, B. Schmitt, et al., IEA
SHC Task 49-Solar Process Heat for Production and Advanced Applications,
2015.
[59] M.S. Strano, C.A. Dyke, M.L. Usrey, C. Kittrell, R.H. Hauge, J.M. Tour, Electronic
structure control of single-walled, Carbon Nanotub. Funct. 301 (2003).
[60] A. Rahimpour, M. Jahanshahi, S. Khalili, A. Mollahosseini, A. Zirepour,
B. Rajaeian, Novel functionalized carbon nanotubes for improving the surface
properties and performance of polyethersulfone (PES) membrane, DES 286
(2012) 99–107, http://dx.doi.org/10.1016/j.desal.2011.10.039.
[61] Y.Y. Huang, E.M. Terentjev, Dispersion of carbon nanotubes: mixing, sonica-
tion, stabilization, and composite properties, Polymers 4 (2012) 275–295,
http://dx.doi.org/10.3390/polym4010275.
[62] Vikas Mittal, Polymer Nanotube Nanocomposites: Synthesis, Properties, and
Applications, 2010.
[63] R. Looser, M. Vivar, V. Everett, Spectral characterisation and long-term per-
formance analysis of various commercial Heat Transfer Fluids (HTF) as Direct-
Absorption Filters for CPV-T beam-splitting applications, Appl. Energy 113
(2014) 1496–1511, http://dx.doi.org/10.1016/j.apenergy.2013.09.001.
[64] R.O.A.T. Taylor, Y.A.H. Ewakuruppu, D.R.E.W.D.E.J. Arnette, Comparison of Se-
lective Transmitters for Solar Thermal Applications, 55, 2016.
[65] J. Beckman, W, Duffie, Solar Engineering of Thermal Process, Wiley- Inter-
science Publication,, 1980.
[66] K. a Wepasnick, B. a Smith, J.L. Bitter, D. Howard Fairbrother, Chemical and
structural characterization of carbon nanotube surfaces, Anal. Bioanal. Chem.
396 (2010) 1003–1014, http://dx.doi.org/10.1007/s00216-009-3332-5.