Selective and reversible noncovalent functionalization of single-walled carbon nanotubes by a pH-responsive vinylogous tetrathiafulvalene-fluorene copolymer

Article abstract of DOI:10.1021/ja409918n

A vinylogous tetrathiafulvalene (TTFV) monomer was prepared and copolymerized with fluorene to give a conformationally switchable conjugated copolymer. This copolymer was shown to undergo a conformational change upon protonation with trifluoroacetic acid

Full text of DOI:10.1021/ja409918n

Article
pubs.acs.org/JACS
Selective and Reversible Noncovalent Functionalization of Single-
Walled Carbon Nanotubes by a pH-Responsive Vinylogous
Tetrathiafulvalene−Fluorene Copolymer
Shuai Liang,† Yuming Zhao,‡ and Alex Adronov*,†
†Department of Chemistry, McMaster University, Hamilton, Ontario, Canada L8S 4M1
‡Department of Chemistry, Memorial University of Newfoundland, St. John’s, Newfoundland, Canada A1B 3X7
S
* Supporting Information
ABSTRACT: A vinylogous tetrathiafulvalene (TTFV) monomer was prepared
and copolymerized with fluorene to give a conformationally switchable
conjugated copolymer. This copolymer was shown to undergo a conformational
change upon protonation with trifluoroacetic acid (TFA). When mixed with
single-walled carbon nanotubes (SWNTs), this TTFV−fluorene copolymer
exhibited strong interactions with the SWNT surface, leading to stable,
concentrated nanotube dispersions in toluene. Photoluminescence excitation
mapping indicated that the copolymer selectively disperses low-diameter
SWNTs, as would be expected from its ability to form a tightly coiled conformation on the nanotube surface. Addition of TFA to
the copolymer−SWNT dispersion resulted in a rapid conformational change and desorption of the polymer from the SWNT
surface, resulting in precipitation of pure SWNTs that were completely free of polymer. Importantly, the nanotubes isolated after
dispersion and release by the TTFV−fluorene copolymer were more pure than the original SWNTs that were initially dispersed.
on-current and on/off ratio, relative to those of devices
prepared with pristine SWNTs that are not coated with
polymers.35 Thus, strategies for achieving strong and selective
interactions between conjugated polymers and SWNTs to
produce concentrated dispersions of specific nanotube
chiralities, combined with reversible complexation such that
the polymer can be desorbed from the nanotube surface upon
application of a specific stimulus, are required.36
Recently, a class of novel vinylogous tetrathiafulvalene−
phenylacetylene copolymers was shown to disperse SWNTs in
organic solvents.37,38 These copolymers exhibited reversible
conformational switching behavior upon oxidation/reduction of
the tetrathiafulvalene vinylogue (TTFV) units, which resulted
in a redox-mediated release of SWNTs. However, the low
molecular weights and poor solubility of the copolymers
resulted in low SWNT concentrations dispersed in solution,
and the overall copolymers did not show appreciable selectivity
for specific SWNT chiralities. To address these issues, we
hypothesized that introduction of highly π-conjugated aromatic
units in the TTFV polymer backbone would assist in addressing
these shortcomings. In previous studies, fluorene-based
polymers have been found to show excellent efficiency and
selectivity in SWNT dispersion.26 Here, we report the
preparation of a TTFV−fluorene copolymer that exhibits
strong interactions with SWNTs and imparts selectivity as a
result of incorporation of the fluorene monomer unit. This
novel copolymer is also capable of reversible SWNT binding
INTRODUCTION
■
Functionalization of single-walled carbon nanotubes (SWNTs)
enables their homogeneous dispersion in various solvents and
bulk materials.1−10 Of the recently reported functionalization
strategies, supramolecular approaches have the advantage of
producing concentrated dispersions without deteriorating the
conjugated sidewall structure of SWNTs.11−13 This enables
retention of their unique structural, electronic, thermal, optical,
and mechanical properties and improves their applicability
within devices.14−17 Conjugated polymers have been exten-
sively used as supramolecular adducts and dispersion agents, as
they form multiple π-stacking interactions with the nanotube
surface, leading to strong interactions.18−24 In this respect, it
has also been shown that rationally designed conjugated
polymers, such as poly(9,9-dialkylfluorene)s, are capable of
selectively interacting with specific, mostly semiconducting,
nanotube species.25−30 These observations have spawned a
significant effort in determining design rules for conjugated
polymers such that their selectivity toward specific nanotube
types is optimized, with the vision of isolating pure fractions of
specific SWNTs.15,31−33
Although selective interactions between conjugated polymers
and SWNTs have allowed isolation of specific SWNT
chiralities, the purified SWNTs still retain the conjugated
polymer on their surface.23,26,34 It has been shown that removal
of the surface-bound polymer via solvent washing is difficult, as
the interaction with the SWNT surface is very strong and
prevents polymer desorption.24 Recent studies have also
demonstrated that application of polymer−SWNT complexes
within field effect transistor (FET) devices results in decreased
Received: September 25, 2013
Published: December 26, 2013
© 2013 American Chemical Society
970
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of TTFV−Fluorene Copolymer 7
Figure 1. 1H NMR (600 MHz, CDCl3) spectra for (A) the copolymer 7 and (B) TTFV precursor 5. The inset shows a comparison of the magnified
aromatic regions for the copolymer 7 (top) and the TTFV monomer 5 (bottom).
and was shown to release SWNTs upon protonation of the
TTFV unit with trifluoroacetic acid (TFA).
nonpolar organic solvents such as toluene, THF, and
chloroform. Gel permeation chromatographic (GPC) analysis
of copolymer 7 revealed a number-average molar mass (Mn) of
19 000 g/mol and a polydispersity index (PDI) of 1.8, relative
to polystyrene standards. Characterization of copolymer 7 by
1H NMR spectroscopy confirmed the 1:1 ratio of the TTFV
and fluorene units within the structure. As expected, the
protons of the phenyl rings of TTFV shift from 7.42 and 7.35
ppm in the monomer (Figure 1B) to 7.48 and 7.41 ppm in the
copolymer (Figure 1A).
The electronic absorption properties of copolymer 7, the
fluorene precursor 6, and the TTFV precursor 5 were
investigated by UV−vis absorption spectroscopy (Figure 2).
From these data, it can be seen that the copolymer’s absorption
maximum (λmax = 407 nm) is red-shifted significantly from that
of either monomer (λmax = 314 and 377 nm for fluorene and
TTFV-m monomer 5, respectively), indicating that delocaliza-
tion of electrons has occurred within the copolymer as a result
RESULTS AND DISCUSSION
■
Synthesis of the TTFV−fluorene copolymer was carried out
according to Scheme 1. Compounds 1 and 2 were prepared
following literature procedures.39 Phosphite-mediated coupling
of 1 and 2 afforded the phenyl dithiafulvene 3, which was
subsequently subjected to an iodine-promoted oxidative
dimerization followed by reductive workup with Na2S2O3 to
produce the neutral phenylacetylene-substituted TTFV com-
pound 4. Removal of the trimethylsilyl (TMS) groups of 4
under basic conditions gave the free terminal bisalkyne product
5, which was subsequently subjected to a Sonogashira cross-
coupling polymerization with 1 equiv of 2,7-diiodofluorene (6),
catalyzed by Pd(PPh3)4 and CuI with diisopropylamine (DIPA)
as base. The resulting TTFV−fluorene copolymer 7 was
isolated as a yellow solid, which was reasonably soluble in
971
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977

Journal of the American Chemical Society
Article
characteristic simultaneous two-electron transfer process
occurring on the diphenyl-TTFV unit that assumes a pseudo-
cis conformation in the neutral state.39 The actual molecular
geometry of TTFV in the neutral state has been unambiguously
elucidated by X-ray single-crystal structural analysis,40−43 which
reveals a twisted pseudo-cis molecular structure. The CV data
for copolymer 7 exhibit similar features for the reduction and
oxidation of the TTFV-m unit, though the peaks are broadened
and peak maxima are shifted closer together, located at +0.50
and +0.68 V, as a result of the increased conjugation within the
copolymer. An additional redox couple at +0.81 and +0.98 V is
also observed for the copolymer, which possibly arises from a
minor trans conformer of the TTFV-m unit within the
copolymer structure.44 These CV data are consistent with the
copolymer structure of 7.
Treatment of copolymer 7 with trifluoroacetic acid TFA
resulted in the formation of TTFV mono- and dications, which
exhibit distinct electronic absorption features. Like other known
TTFV derivatives, the protonation of copolymer 7 occurs via a
two-step process, where the formation of TTFV monocation
and dication takes place sequentially.38 This is evidenced by
UV−vis titration analysis (Figure 4) where a solution of
copolymer 7 was titrated with TFA up to 2250 equiv. The UV−
vis titration data clearly show spectroscopic changes occurring
Figure 2. UV−vis absorption spectra for the fluorine precursor 6, the
TTFV-m monomer 5, and the copolymer 7. Spectra were measured in
CH2Cl2 at room temperature.
of π-conjugation between the constituent units. The electro-
chemical properties of the TTFV-m monomer 5 and copolymer
7 were investigated by cyclic voltammetry (CV) in CH2Cl2 at
room temperature. The measured cyclic voltammograms are
shown in Figure 3. The CV profile for TTFV-m 5 shows a
reversible redox wave pair at +0.48 and +0.72 V, attributed to a
Figure 4. UV−vis spectral changes monitoring the titration of
copolymer 7 with TFA (top, from 1 to 300 equiv of TFA; bottom,
from 500 to 2250 equiv of TFA). All titration experiments were done
in CH2Cl2 at room temperature.
Figure 3. Cyclic voltammograms of TTFV-m monomer 5 and
copolymer 7.
972
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977

Journal of the American Chemical Society
Article
in two stages. In the first stage (up to 300 equiv of TFA), the
intensity of the absorption band at 405 nm, which is attributed
to the π−π* transition of neutral TTFV, continuously
decreases with increasing acidity. Simultaneously, two new
longer wavelength absorption bands at 643 and 948 nm emerge
and grow steadily. These two bands are consistent with the
known UV−vis absorption features of TTF radical cations,45
and hence are assigned to the monoprotonated [TTFV·H]+. In
addition, a shoulder at 710 nm is observable in this process,
which is attributed to partial formation of the dication [TTFV·
2H]2+. As the acidification continues, the formation of the
dication leads to further spectral changes, where the two bands
at 643 and 948 nm (attributed to [TTFV·H]+) diminish and
the band at 710 nm (attributed to [TTFV·2H]2+) grows
significantly (Figure 4). Since the final diprotonated [TTFV·
2H]2+ units have been known to undergo a conformational
switch to the trans conformation to minimize charge
repulsions,46 the entire conformation of protonated polymer
7 is predicted to take a linear structure. This acidity-responsive
conformational change serves as the driving force for effective
release of SWNTs from the corresponding polymer−SWNT
assemblies (see below).
Dispersion of SWNTs with the TTFV−Fluorene
Copolymer. The ability of copolymer 7 to form supra-
molecular complexes with SWNTs and enable their homoge-
neous dispersion in organic solvents was investigated using
previously published protocols.37 Briefly, 3 mg of raw SWNT
powder (CoMoCat or HiPco) was added to a polymer solution
containing 50 mg of polymer dissolved in 10 mL of toluene.
The polymer−SWNT mixtures were sonicated for 30 min in a
bath sonicator that was chilled with ice, followed by 60 min of
centrifugation at 8346g. The supernatant was carefully removed
from the centrifuge tube to obtain a stable dispersion of the 7−
SWNT complexes. This dispersion was extremely dark and
stable with no observable flocculation over several months. The
formation of these stable, concentrated dispersions is consistent
with the theoretical predictions that the copolymer can adopt a
suitable helical conformation to tightly wrap individual SWNTs
(see above). Analysis by UV−vis−NIR spectroscopy revealed
the presence of SWNT absorption bands in the wavelength
range of 500−1500 nm, consistent with the presence of a high
concentration of exfoliated SWNTs (Figure 6).47 In the
Molecular Modeling Studies. Molecular structural
properties of copolymer 7 as well as the complex of 7 with a
(10,0) SWNT were investigated by molecular mechanics
calculations using the SYBYL force field implemented with
the Spartan’10 software package (the alkyl chains on TTFV
were replaced with H atoms for clarity and computational
simplification). The calculated conformations indicate that the
free polymer has a relatively flexible backbone and tends to
form an irregular folded conformation, in which the TTFV
units take a pseudo-cis geometry similar to those of diphenyl-
TTFV derivatives determined by X-ray crystallography (Figure
5A,B).46 However, upon interacting with SWNTs, the polymer
Figure 6. UV−vis−NIR absorption spectra for copolymer 7−HiPco
dispersion (blue trace) and copolymer 7−CoMoCAT dispersion (red
trace) in toluene.
absorption spectrum of the CoMoCAT SWNT dispersion, a
prominent absorption peak at 999 nm was observed, which is
assigned to SWNTs with a chiral index of (6,5).48 Lower
intensity absorptions corresponding to the (7,5) and (8,3)
SWNTs were also observed.48 In addition, the lack of
background absorbance across the entire spectrum indicates
that these samples do not contain significant amounts of
amorphous carbon and SWNT aggregates. This suggests that
complexation with copolymer 7 enables purification and
debundling of commercial SWNT samples. The analogous
absorption spectrum of the HiPco SWNT dispersion exhibits
well-resolved peaks in the S11 (approximately 830−1600 nm)
and S22 (approximately 600−830 nm) ranges, with very little
signal corresponding to M11 transitions (450−650 nm)
observable. However, overlap with the copolymer absorption
in the M11 region makes it difficult to conclusively determine
the extent to which metallic SWNTs are present in this
dispersion.
Figure 5. Conformation of a model hexamer of TTFV−fluorene
copolymer: (A) side view, (B) front view. Two projections of the
conformation of a model hexamer wrapping around a (10,0) SWNT:
(C) side view, (D) front view.
adopts a more tightly packed helical conformation that exhibits
a 1.5 nm diameter hollow inner cavity that is well suited for
wrapping an exfoliated (10,0) SWNT (ca. 1.1 nm in diameter)
via π−π interactions (Figure 5C,D). It is worth noting that all
the fluorene units show direct face-to-face contact with the
surface of the SWNT, suggesting that the fluorene−SWNT π-
stacking contributes the most significant driving force for the
polymer to wrap around an individual SWNT.
Photoluminescence excitation (PLE) mapping is a powerful
tool for characterizing different chiralities in bulk SWNT
samples. The PLE maps for both CoMoCAT and HiPco
973
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977

Journal of the American Chemical Society
Article
SWNTs dispersed by copolymer 7 in toluene are shown in
Figure 7. On the basis of the high emission intensity observed
Figure 8. Graphene sheet maps showing normalized PL intensities for
copolymer 7−HiPco complexes. The darker fill shading represents
higher PL intensity.
highest intensity signals from the 7−HiPco dispersion are also
listed in Table 1, and although most of these SWNTs have
Table 1. Table of Chiral Index vs Diameter and Chiral Angle
chiral index diameter
chiral
angle (θ)
chiral index diameter
chiral
angle (θ)
(m,n)
(nm)
(m,n)
(nm)
(8,7)
(9,5)
(8,6)
(9,4)
1.032
0.976
0.966
0.916
27.80
20.63
25.28
17.48
(7,6)
(10,2)
(7,5)
(6,5)
0.895
0.884
0.829
0.757
27.46
8.95
24.50
27.00
chiral angles above 24°, the presence of several species with
lower chiral angles indicates that there is little chiral angle
selectivity with this polymer. It is also interesting to compare
the PLE map of 7−HiPco to what is obtained when the same
HiPco SWNTs are dispersed with a nonselective surfactant,
such as sodium dodecylbenzenesulfonate (SDBS) (see the
Supporting Information, Figure S1). Most notably, the (7,5)
species that is prominent in the dispersion with the copolymer
is a relatively minor component in the SDBS dispersion,
indicating that copolymer 7 is indeed selective for specific
SWNT types.
Tapping mode atomic force microscopy (AFM) was carried
out on copolymer−SWNT complexes that were spin-cast onto
freshly cleaved mica substrates. To observe individual SWNTs
and eliminate the presence of free polymer, copolymer−SWNT
samples were filtered through a Teflon membrane (0.2 μm pore
diameter) and washed extensively until the filtrate contained no
detectable free polymer. The resulting residue was dispersed in
toluene, as described above, and used for sample preparation.
Figure S2 (Supporting Information) shows the AFM image for
the 7−CoMoCAT complex, spin-cast from a 100-fold-diluted
dispersion in toluene. From the image, it is clear that individual
polymer-wrapped nanotubes are present. Height analysis of the
AFM images indicates that the majority of nanotube features
are approximately 2.5−3.2 nm high, which is consistent with
single nanotube structures wrapped with a monolayer of
polymer. Unfortunately, even after the extensive washing
procedure, some features that resemble free polymer on the
mica surface are still observable by AFM.
Figure 7. PLE maps of CoMoCAT (A) and HiPco (B) SWNTs
dispersed in toluene with copolymer 7.
for the SWNTs, it can be concluded that the copolymer
efficiently debundled both CoMoCAT and HiPco bundles into
individual, polymer-wrapped tubes. Although the intensity of
signals in the PLE maps does not provide a quantitative
measure of the SWNT concentration because the quantum
efficiency of a given nanotube chirality may depend on its
environment, it is still a useful indicator of a given polymer’s
selectivity for specific SWNTs when considered together with
absorption data. On the basis of the PLE data in Figure 7, the
7−CoMoCAT dispersion predominantly contains the (6,5)
nanotube species, which is consistent with the UV−vis−NIR
absorption data, while the 7−HiPco dispersion predominantly
contains the (7,5) SWNT species, although a number of other
nanotube chiralities, including (9,4), (8,6), (6,5), (8,7), (9,5),
(7,6), and (10,2), are also present (Figure 8). These major
components in the 7−HiPco dispersion are all low-diameter
SWNTs (1.03 nm or lower), indicating that a certain degree of
diameter selectivity is achieved with copolymer 7. Considering
the tight helical conformation that it is expected to adopt, it is
not surprising that selective interactions based on nanotube
diameter would be possible. The chiral angles of the eight
Release of SWNTs from Copolymer. As shown in Figure
4, the addition of acid to a solution of copolymer 7 leads to a
significant change in its absorption properties and a dramatic
conformational switch resulting from cis−trans isomerization of
the TTFV units. Similarly, addition of two drops of TFA to a
dispersion of the 7−CoMoCAT SWNT complex at room
temperature, followed by agitation of the vial to mix the
974
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977

Journal of the American Chemical Society
Article
contents, resulted in an immediate color change of the
polymer−SWNT dispersion from dark brown to green,
accompanied by a simultaneous precipitation of the dispersed
SWNTs (Figure 9 and Figure S3 of the Supporting
Figure 9. Photograph of (A) pure copolymer 7 solution in toluene,
(B) a dispersion of copolymer 7−CoMoCAT SWNTs in toluene, and
(C) the dispersion after addition of TFA, showing precipitated
nanotubes (the sample was diluted 4-fold to clearly see the
precipitates).
Information). Clearly, the conformational change of the
TTFV units leads to a polymer conformation that is not
conducive to interaction with the SWNT surface. The released
SWNTs were easily removed from the polymer solution via
either filtration or centrifugation. The isolated acidified polymer
solution was then treated with aqueous NaHCO3 to neutralize
the polymer and recover the original, orange solution of
1
copolymer 7. H NMR analysis of the recovered polymer
confirmed that there were no changes in its molecular structure
as a result of nanotube complexation, release, and neutralization
steps (Figure S4, Supporting Information). This recovered
polymer could again be used to redisperse the precipitated
SWNTs, demonstrating that the dissolution−precipitation
process is reversible (Figure S5, Supporting Information).
Characterization of the released and isolated SWNTs was
initially conducted using Raman spectroscopy. In particular, the
radial breathing mode (RBM) signals49 of SWNTs are strongly
dependent on the nanotube diameter as well as any surface-
adsorbed chemical species.50−52 Raman data using 785 nm
excitation were acquired for the CoMoCat SWNT starting
material, the 7−CoMoCat complexes, and the CoMoCat
SWNTs recovered after release from the copolymer (Figure
10). The corresponding Raman data for the 7−HiPco
complexes are provided in the Supporting Information (Figure
S6). For comparison, we also attempted to measure the Raman
spectrum of the pure polymer at the same excitation
wavelength, but only observed background signals from the
silicon substrate (Figure S7, Supporting Information). In the
RBM region, the CoMoCat starting material exhibits three
prominent signals at 230, 264, and 301 cm−1 (Figure 10B). The
signal at 264 cm−1 is often referred to as a “bundling peak”, as it
arises when nanotubes are aggregated in bundles.53−55 Upon
complexation with copolymer 7, the bundling peak at 264 cm−1
dramatically decreases in intensity as a result of the exfoliation
that occurs when the copolymer interacts with the SWNTs.
More interestingly, the signal at 230 cm−1 undergoes a red shift
by 3 cm−1, indicating that a π-stacking interaction has occurred
between the adsorbed polymer and the nanotube surface.
Furthermore, the shoulder at 225 cm−1 increased in intensity
relative to the peak at 230 cm−1, possibly signifying enrichment
in this SWNT species. The spectrum corresponding to the
released SWNTs after polymer isomerization shows the
Figure 10. Raman spectra (A) of (i) raw CoMoCAT SWNTs, (ii) a
copolymer 7−CoMoCAT SWNT complex, and (iii) released
CoMoCAT SWNTs. The expanded RBM region from the same
data is given in (B). Samples were prepared by drop-casting
dispersions onto silicon wafers followed by air-drying overnight and
were excited at 785 nm.
reappearance of the bundling peak and a blue shift of the
peak at 233 cm−1 back to 230 cm−1, where it was in the original
starting material. The shoulder peak at 225 cm−1 retained its
relative intensity with respect to the 230 cm−1 signal, indicating
that the enrichment of this nanotube species persisted in the
nanotube sample recovered upon release. The signal at 301
cm−1 was found to mirror the one at 264 cm−1, with a
significant decrease in relative intensity upon interaction with
copolymer 7 and then a recovery upon release of the nanotubes
(Figure 10B). Similar data were observed with the 7−HiPco
complexes using 785 nm excitation (Figure S6, Supporting
Information).
The thermal stability of copolymer 7, the polymer−SWNT
supramolecular complexes, and released SWNTs was charac-
terized by thermogravimetric analysis (TGA), carried out under
an Ar atmosphere. These data are shown in Figure 11 and
indicate that the major mass loss for the copolymer occurs at
234 °C, amounting to 47%. This mass loss corresponds to the
loss of side chains (−SC8H17) within the polymer. For the
polymer−SWNT complex, the mass loss amounts to 30%,
which indicates that the SWNT composition within the
polymer−SWNT complexes was approximately 45%. The
thermogram for the SWNTs that were released from the
975
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977

Journal of the American Chemical Society
Article
CONCLUSIONS
■
The synthesis of a novel conformationally switchable
conjugated copolymer consisting of alternating TTFV and
fluorene units was achieved. Absorption spectroscopy indicated
that the polymer undergoes a two-step protonation process
with TFA, accompanied by dramatic conformational changes in
the polymer structure. This polymer exhibits strong interactions
with the surface of SWNTs, allowing their homogeneous
dispersion in a variety of organic solvents, including toluene,
THF, and chloroform. Furthermore, the interaction of the
copolymer with SWNTs was found to be size-selective, with a
preference for low-diameter semiconducting SWNTs. Upon
addition of TFA to the copolymer−SWNT dispersion, a rapid
color change and simultaneous precipitation of the nanotube
material were observed. From TGA measurements, it was
found that the precipitated nanotubes were free of polymer,
indicating that the polymer underwent the expected conforma-
tional change and desorbed from the nanotube surface upon
protonation. Furthermore, on the basis of TGA data, the
nanotube material isolated after the dispersion/release
procedure was of higher purity than the SWNT starting
material. It should also be noted that the polymer could be
easily recovered and reused. This work demonstrates the
feasibility of selectively dispersing a specific subpopulation of
SWNTs from an impure mixture and their rapid, facile release
and isolation in pure form upon application of a simple release
trigger (acidification). Our work has significant implications for
future nanotube enrichment/purification procedures, in terms
of both nanotube chirality and overall separation from non-
nanotube impurities.
Figure 11. TGA profiles showing mass loss upon heating for
copolymer 7, 7−SWNT complexes, raw SWNT, and released
SWNT samples under Ar.
copolymer shows no appreciable mass loss up to 800 °C. This
experiment definitively proves that the released SWNTs do not
have any polymer on their surface and supports the hypothesis
that the copolymer is able to completely desorb from the
nanotube surface upon application of a stimulus, namely, TFA.
Interestingly, comparison of the TGA data for the original, as-
received CoMoCat sample with those for the released
nanotubes after dispersion with copolymer 7 shows some
difference. The original sample exhibits a reproducible mass
loss with an onset temperature of around 600 °C. Considering
that SWNTs should be stable to beyond 800 °C, this mass loss
is attributed to impurities present in the as-received commercial
sample. Several TGA measurements on this commercial batch
of CoMoCAT SWNTs revealed identical data. Surprisingly,
when TGA is carried out on the nanotubes released from
copolymer 7, as described above, the thermogram exhibits a
lower mass loss upon heating to 800 °C. This indicates that
dispersion with the copolymer is selective for SWNTs and
results in exclusion of any amorphous carbon impurities and
metal catalyst particles that were present in the original sample,
thus leading to enrichment of the SWNT sample. It is evident
that the purity of the nanotubes released from the copolymer is
improved relative to that of this commercial sample.
ASSOCIATED CONTENT
* Supporting Information
■
S
Full experimental details, PLE map for HiPco SWNTs
dispersed with SDBS, AFM data, nanotube dispersion images,
NMR data for the recovered polymer, and Raman data for the
copolymer and copolymer-dispersed HiPco SWNTs. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
adronov@mcmaster.ca
■
Notes
The authors declare no competing financial interest.
Since SWNTs can rapidly and cleanly be released from the
copolymer solution triggered by addition of TFA, the solubility
of SWNTs in the polymer−SWNT dispersion in toluene can be
easily determined. Upon addition of TFA, the resulting mixture
was subjected to ultracentrifugation, and the SWNTs under-
went membrane filtration and extensive washing with toluene.
The released SWNTs were slowly stirred in aqueous NaHCO3
solution to neutralize any protonation of SWNTs that may have
happened during the release experiment.56 The resulting
isolated neutralized SWNTs were dried to constant mass.
The recovered amounts of CoMoCat and HiPco SWNTs were
1.1 and 2.0 mg, respectively. This indicates that the
concentrations in the original solutions (10 mL volume) were
approximately 0.1 and 0.2 mg/mL for CoMoCat and HiPco
SWNTs, respectively, when the concentration of copolymer
was 5 mg/mL.
ACKNOWLEDGMENTS
■
This work was supported by the National Science and
Engineering Research Council of Canada (NSERC), the
Canada Foundation for Innovation (CFI), and McMaster
University. S.L. gratefully acknowledges Nicole Rice and Patigul
Imin for providing the PLE map of raw HiPco SWNTs
dispersed by SDBS and Guang Chen for electrochemical
experiments.
REFERENCES
■
(1) Hirsch, A. Angew. Chem., Int. Ed. 2002, 41, 1853−1859.
(2) Hirsch, A.; Vostrowsky, O. Top. Curr. Chem. 2005, 245, 193−237.
(3) Bahr, J. L.; Tour, J. M. J. Mater. Chem. 2002, 12, 1952−1958.
(4) Rana, S.; Cho, J. W. Nanoscale 2010, 2, 2550−2556.
(5) Hu, C.-Y.; Xu, Y.-J.; Duo, S.-W.; Zhang, R.-F.; Li, M.-S. J. Chin.
Chem. Soc. 2009, 56, 234−239.
976
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977

Journal of the American Chemical Society
Article
(6) Grossiord, N.; Loos, J.; Regev, O.; Koning, C. E. Chem. Mater.
2006, 18, 1089−1099.
(39) Chen, G.; Mahmud, I.; Dawe, L. N.; Zhao, Y. Org. Lett. 2010,
12, 704−707.
(40) Yamashita, Y.; Tomura, M.; Badruz Zaman, M. Chem. Commun.
1998, 1657−1658.
(7) Dyke, C.; Tour, J. J. Phys. Chem. A 2004, 108, 11151−11159.
(8) Tasis, D.; Tagmatarchis, N.; Bianco, A.; Prato, M. Chem. Rev.
2006, 106, 1105−1136.
(41) Yamashita, Y.; Tomura, M.; Tanaka, S.; Imaeda, K. Synth. Met.
1999, 102, 1730−1731.
(9) Schnorr, J. M.; Swager, T. M. Chem. Mater. 2011, 23, 646−657.
(10) Singh, P.; Campidelli, S.; Giordani, S.; Bonifazi, D.; Bianco, A.;
Prato, M. Chem. Soc. Rev. 2009, 38, 2214−2230.
(11) Britz, D.; Khlobystov, A. Chem. Soc. Rev. 2006, 35, 637−659.
(12) Liu, P. Eur. Polym. J. 2005, 41, 2693−2703.
(13) Zhao, Y.-L.; Stoddart, J. F. Acc. Chem. Res. 2009, 42, 1161−
1171.
(42) Hascoat, P.; Lorcy, D.; Robert, A.; Carlier, R.; Tallec, A.;
Boubekeur, K.; Batail, P. J. Org. Chem. 1997, 62, 6086−6089.
(43) Guerro, M.; Carlier, R.; Boubekeur, K.; Lorcy, D.; Hapiot, P. J.
Am. Chem. Soc. 2003, 125, 3159−3167.
(44) Bouzan, S.; Chen, G.; Mulla, K.; Dawe, L. N.; Zhao, Y. Org.
Biomol. Chem. 2012, 10, 7673−7676.
(45) Kirketerp, M.-B. S. M.; Leal, L. A. E. L.; Varsano, D. D.; Rubio,
(14) Guo, Z.; Feng, Y.; Zhu, D.; He, S.; Liu, H.; Shi, X.; Sun, J.; Qu,
A. A.; Jørgensen, T. J. D. T.; Kilsa, K. K.; Nielsen, M. B. M.; Nielsen, S.
̊
M. Adv. Funct. Mater. 2013, 23, 5010−5018.
B. S. Chem. Commun. 2011, 47, 6900−6902.
(15) Park, S.; Lee, H. W.; Wang, H.; Selvarasah, S.; Dokmeci, M. R.;
Park, Y. J.; Cha, S. N.; Kim, J. M.; Bao, Z. ACS Nano 2012, 6, 2487−
2496.
(46) Zhao, Y.; Chen, G.; Mulla, K.; Mahmud, I.; Liang, S.; Dongare,
P.; Thompson, D. W.; Dawe, L. N.; Bouzan, S. Pure Appl. Chem. 2012,
84, 1005−1025.
(16) Pang, X.; Imin, P.; Zhitomirsky, I.; Adronov, A. J. Mater. Chem.
(47) O’Connell, M.; Bachilo, S.; Huffman, C.; Moore, V.; Strano, M.;
Haroz, E.; Rialon, K.; Boul, P.; Noon, W.; Kittrell, C.; Ma, J.; Hauge,
R.; Weisman, R.; Smalley, R. Science 2002, 297, 593−596.
(48) Bachilo, S.; Strano, M.; Kittrell, C.; Hauge, R.; Smalley, R.;
Weisman, R. Science 2002, 298, 2361−2366.
2012, 22, 9147−9154.
(17) Pang, X.; Imin, P.; Zhitomirsky, I.; Adronov, A. Macromolecules
2010, 43, 10376−10381.
(18) Dalton, A. B.; Stephan, C.; Coleman, J. N.; McCarthy, B.;
Ajayan, P. M.; Lefrant, S.; Bernier, P.; Blau, W. J.; Byrne, H. J. J. Phys.
Chem. B 2000, 104, 10012−10016.
(49) Rao, A.; Richter, E.; Bandow, S.; Chase, B.; Eklund, P.; Williams,
K.; Fang, S.; Subbaswamy, K.; Menon, M.; Thess, A.; Smalley, R.;
Dresselhaus, G.; Dresselhaus, M. Science 1997, 275, 187−191.
(50) Dresselhaus, M. S.; Eklund, P. C. Adv. Phys. 2000, 49, 705−814.
(51) Dresselhaus, M. S.; Dresselhaus, G.; Jorio, A.; Souza Filho, A.
G.; Pimenta, M. A.; Saito, R. Acc. Chem. Res. 2002, 35, 1070−1078.
(52) Fantini, C.; Jorio, A.; Souza, M.; Strano, M. S.; Dresselhaus, M.
S.; Pimenta, M. A. Phys. Rev. Lett. 2004, 93, 1−4.
(19) Star, A.; Stoddart, J. F.; Steuerman, D.; Diehl, M.; Boukai, A.;
Wong, E. W.; Yang, X.; Chung, S.-W.; Choi, H.; Heath, J. R. Angew.
Chem., Int. Ed. 2001, 113, 1771−1775.
(20) Star, A.; Stoddart, J. Macromolecules 2002, 35, 7516−7520.
(21) Star, A.; Liu, Y.; Grant, K.; Ridvan, L.; Stoddart, J. F.;
Steuerman, D. W.; Diehl, M. R.; Boukai, A.; Heath, J. R.
Macromolecules 2003, 36, 553−560.
(22) Cheng, F.; Adronov, A. Chem.Eur. J. 2006, 12, 5053−5059.
(23) Cheng, F.; Imin, P.; Maunders, C.; Botton, G.; Adronov, A.
Macromolecules 2008, 41, 2304−2308.
(24) Imin, P.; Imit, M.; Adronov, A. Macromolecules 2012, 45, 5045−
5050.
(53) O’Connell, M.; Sivaram, S.; Doorn, S. Phys. Rev. B 2004, 69,
235415.
(54) Luo, Z.; Doorn, S. K.; Li, R.; Papadimitrakopoulos, F. Phys.
Status Solidi B 2006, 243, 3155−3160.
(55) Jiang, C.; Saha, A.; Xiang, C.; Young, C. C.; Tour, J. M.;
Pasquali, M.; Martí, A. A. ACS Nano 2013, 7, 4503−4510.
(56) Engtrakul, C.; Davis, M. F.; Gennett, T.; Dillon, A. C.; Jones, K.
M.; Heben, M. J. J. Am. Chem. Soc. 2005, 127, 17548−17555.
(25) Chen, F.; Wang, B.; Chen, Y.; Li, L. Nano Lett. 2007, 7, 3013−
3017.
(26) Nish, A.; Hwang, J.; Doig, J.; Nicholas, R. Nat. Nanotechnol.
2007, 2, 640−646.
(27) Hwang, J.-Y.; Nish, A.; Doig, J.; Douven, S.; Chen, C.-W.; Chen,
L.-C.; Nicholas, R. J. J. Am. Chem. Soc. 2008, 130, 3543−3553.
(28) Sturzl, N.; Hennrich, F.; Lebedkin, S.; Kappes, M. M. J. Phys.
̈
Chem. C 2009, 113, 14628−14632.
(29) Gao, J.; Kwak, M.; Wildeman, J.; Herrmann, A.; Loi, M. Carbon
2010, 49, 333−338.
(30) Ozawa, H.; Ide, N.; Fujigaya, T.; Niidome, Y.; Nakashima, N.
Chem. Lett. 2011, 40, 239−241.
(31) Lemasson, F. A.; Strunk, T.; Gerstel, P.; Hennrich, F.; Lebedkin,
S.; Barner-Kowollik, C.; Wenzel, W.; Kappes, M. M.; Mayor, M. J. Am.
Chem. Soc. 2011, 133, 652−655.
(32) Lemasson, F.; Berton, N.; Tittmann, J.; Hennrich, F.; Kappes,
M. M.; Mayor, M. Macromolecules 2012, 45, 713−722.
(33) Lee, H. W.; Yoon, Y.; Park, S.; Oh, J. H.; Hong, S.; Liyanage, L.
S.; Wang, H.; Morishita, S.; Patil, N.; Park, Y. J.; Park, J. J.; Spakowitz,
A.; Galli, G.; Gygi, F.; Wong, P. H. S.; Tok, J. B.-H.; Kim, J. M.; Bao, Z.
Nat. Commun. 2011, 2, 541−548.
(34) Imin, P.; Imit, M.; Adronov, A. Macromolecules 2011, 44, 9138−
9145.
(35) Wang, W. Z.; Li, W. F.; Pan, X. Y.; Li, C. M.; Li, L.-J.; Mu, Y. G.;
Rogers, J. A.; Chan-Park, M. B. Adv. Funct. Mater. 2011, 21, 1643−
1651.
(36) Lemasson, F.; Tittmann, J.; Hennrich, F.; Sturzl, N.; Malik, S.;
̈
Kappes, M. M.; Mayor, M. Chem. Commun. 2011, 47, 7428.
(37) Liang, S.; Chen, G.; Peddle, J.; Zhao, Y. Chem. Commun. 2012,
48, 3100−3102.
(38) Liang, S.; Chen, G.; Zhao, Y. J. Mater. Chem. C 2013, 1, 5477−
5490.
977
dx.doi.org/10.1021/ja409918n | J. Am. Chem. Soc. 2014, 136, 970−977