Functionalization of single-walled carbon nanotubes with well-defined polystyrene by "click" coupling

Article abstract of DOI:10.1021/ja054958b

Covalent functionalization of alkyne-decorated SWNTs with well-defined, azide-terminated polystyrene polymers was accomplished by the Cu(I)-catalyzed [3 + 2] Huisgen cycloaddition. This reaction was found to be extremely efficient in producing organosolub

Full text of DOI:10.1021/ja054958b

Published on Web 09/23/2005
Functionalization of Single-Walled Carbon Nanotubes with
Well-Defined Polystyrene by “Click” Coupling
†
†
‡
,†,‡
Huaming Li, Fuyong Cheng, Andy M. Duft, and Alex Adronov*
Contribution from the Department of Chemistry and Brockhouse Institute for Materials Research
(BIMR), McMaster UniVersity, Hamilton, Ontario L8S 4M1, Canada
Received July 23, 2005; E-mail: adronov@mcmaster.ca
Abstract: Covalent functionalization of alkyne-decorated SWNTs with well-defined, azide-terminated
polystyrene polymers was accomplished by the Cu(I)-catalyzed [3 + 2] Huisgen cycloaddition. This reaction
was found to be extremely efficient in producing organosoluble polymer-nanotube conjugates, even at
relatively low reaction temperatures (60 °C) and short reaction times (24 h). The reaction was found to be
most effective when a CuI catalyst was employed in the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene
as an additive. IR spectroscopy was utilized to follow the introduction and consumption of alkyne groups
2
on the SWNTs, and Raman spectroscopy evidenced the conversion of a high proportion of sp carbons to
3
sp hybridization during alkyne introduction. Thermogravimetric analysis indicated that the polymer-
functionalized SWNTs consisted of 45% polymer, amounting to approximately one polymer chain for every
200-700 carbons of the nanotubes, depending on polymer molecular weight. Transmission electron
microscopy and atomic force microscopy were utilized to image polymer-functionalized SWNTs, showing
relatively uniform polymer coatings present on the surface of individual, debundled nanotubes.
1
5
Introduction
approaches to the functionalization of SWNTs, including
1
6,17
covalent sidewall coupling reactions,
and noncovalent
have been developed to overcome
the solubility limitations. Among these approaches, covalent
sidewall modification with polymeric structures has shown
promise in improving the solubility of nanotube-polymer
conjugates, even with a relatively low degree of functionaliza-
Due to the unique structural and electronic properties of
single-walled carbon nanotubes (SWNTs), they have been
investigated for numerous potential applications that include
molecular electronics,2 sensors, field-emission devices, and
components in high-performance composites. Although some
of these applications can be realized through in-situ growth
1
8-22
1
exohedral interactions,
,3
4-6
7
8
9
,10
2
3,24
tion.
Furthermore, the versatility of polymer chemistry
or solid-phase deposition of carbon nanotubes at the site of
action,10-12
allows for control over the final properties of the nanotube-
polymer conjugate, which are dictated by the chemical and
physical characteristics of the grafted polymer. For example,
the attachment of functional polymers with controlled archi-
tectures may provide opportunities for the preparation of hybrid
materials capable of environmental responsiveness and, poten-
tially, supramolecular organization. Current methods for prepa-
ration of covalent polymer-SWNT conjugates include “grafting
from” methods using initiator-functionalized SWNTs,
well as “grafting to” methods involving radical coupling2
many others will require solution-phase processing
and manipulation to achieve appropriate assemblies, orientations,
and homogeneous dispersions of carbon nanotubes within host
1
3
materials. Solution-phase processing, however, poses great
difficulty considering the general insolubility of these structures
in most organic and aqueous solvents.14 Recently, several
†
Department of Chemistry.
Brockhouse Institute for Materials Research.
‡
25-28
as
(
(
(
(
(
1) Iijima, S.; Ichihashi, T. Nature 1993, 363, 603-605.
2) Avouris, P. Acc. Chem. Res. 2002, 35, 1026-1034.
3) Collins, P. G.; Avouris, P. Sci. Am. 2000, 283, 62-69.
4) Dai, H. J. Acc. Chem. Res. 2002, 35, 1035-1044.
5) Kong, J.; Franklin, N. R.; Zhou, C. W.; Chapline, M. G.; Peng, S.; Cho,
K. J.; Dai, H. J. Science 2000, 287, 622-625.
9-32
(15) Hirsch, A. Angew. Chem., Int. Ed. 2002, 41, 1853-1859.
(16) Niyogi, S.; Hamon, M. A.; Hu, H.; Zhao, B.; Bhowmik, P.; Sen, R.; Itkis,
M. E.; Haddon, R. C. Acc. Chem. Res. 2002, 35, 1105-1113.
(17) Banerjee, S.; Hemraj-Benny, T.; Wong, S. S. AdV. Mater. 2005, 17, 17-
29.
(
6) Kong, J.; Chapline, M. G.; Dai, H. J. AdV. Mater. 2001, 13, 1384-1386.
7) Choi, W. B.; Chung, D. S.; Kang, J. H.; Kim, H. Y.; Jin, Y. W.; Han, I.
T.; Lee, Y. H.; Jung, J. E.; Lee, N. S.; Park, G. S.; Kim, J. M. Appl. Phys.
Lett. 1999, 75, 3129-3131.
(
(18) Nakashima, N.; Tomonari, Y.; Murakami, H. Chem. Lett. 2002, 638-639.
(19) Chen, R. J.; Zhang, Y.; Wang, D.; Dai, H. J. Am. Chem. Soc. 2001, 123,
3838-3839.
(
8) Ajayan, P. M. Chem. ReV. 1999, 99, 1787-1799.
(
9) Wei, B. Q.; Vajtai, R.; Jung, Y.; Ward, J.; Zhang, R.; Ramanath, G.; Ajayan,
P. M. Nature 2002, 416, 495-496.
(20) Chen, R. J.; Bangsaruntip, S.; Drouvalakis, K. A.; Kam, N. W. S.; Shim,
M.; Li, Y. M.; Kim, W.; Utz, P. J.; Dai, H. J. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 4984-4989.
(
(
10) Dai, H. J. Surf. Sci. 2002, 500, 218-241.
11) Rueckes, T.; Kim, K.; Joselevich, E.; Tseng, G. Y.; Cheung, C. L.; Lieber,
C. M. Science 2000, 289, 94-97.
(21) Liu, L.; Wang, T. X.; Li, J. X.; Guo, Z. X.; Dai, L. M.; Zhang, D. Q.; Zhu,
D. B. Chem. Phys. Lett. 2003, 367, 747-752.
(
(
(
12) Dai, L. M.; Patil, A.; Gong, X. Y.; Guo, Z. X.; Liu, L. Q.; Liu, Y.; Zhu,
(22) Gomez, F. J.; Chen, R. J.; Wang, D. W.; Waymouth, R. M.; Dai, H. J.
Chem. Commun. 2003, 190-191.
D. B. ChemPhysChem 2003, 4, 1150-1169.
13) Huang, Y.; Duan, X. F.; Wei, Q. Q.; Lieber, C. M. Science 2001, 291,
(23) Sun, Y. P.; Fu, K. F.; Lin, Y.; Huang, W. J. Acc. Chem. Res. 2002, 35,
1096-1104.
6
30-633.
14) Tasis, D.; Tagmatarchis, N.; Georgakilas, V.; Prato, M. Chem.-Eur. J. 2003,
, 4001-4008.
(24) Hill, D. E.; Lin, Y.; Rao, A. M.; Allard, L. F.; Sun, Y. P. Macromolecules
2002, 35, 9466-9471.
9
1
4518 J. AM. CHEM. SOC. 2005, 127, 14518-14524
9
10.1021/ja054958b CCC: $30.25 © 2005 American Chemical Society

Functionalization of Single-Walled Carbon Nanotubes
A R T I C L E S
46
or reaction with surface-bound carboxylic acid groups on
of dendrimers,39,44,45 dendronized polymers, side-chain func-
tional linear polymers,
oxidized SWNTs.3
3-36
Although the “grafting from” approach
47,48
and cross-linked materials.49 It was
promises high graft densities, attachment of initiator groups to
SWNTs and control over polymer molecular weight and
architecture can be difficult to achieve. Additionally, the widely
used controlled radical polymerizations that have been reported
to occur from nanotube surfaces may be affected by radical
coupling to carbon nanotubes, potentially causing side reactions
leading to decreased control over the polymerization process.
Conversely, the “grafting to” approach, which involves polymer
preparation prior to grafting, allows full control over polymer
length and architecture, but typically results in low graft density
due to steric hindrance from attached polymer chains and
relatively low reactivity of functional groups on the surface of
SWNTs. In addition, most of the reported nanotube grafting
reactions require relatively harsh conditions, typically involving
high temperatures and long reaction times. Such conditions may
be incompatible with many of the functional molecules that are
desirable for grafting onto SWNTs.
therefore postulated that the use of click chemistry would be
an ideal modular methodology for the introduction of a wide
variety of molecules onto the surface of SWNTs. Here, we report
the application of the Huisgen cycloaddition to the functional-
ization of SWNTs with polystyrene. To achieve a high degree
of functionalization, we chose to introduce alkyne groups on
the nanotube surface using the Pschorr-type arylation previously
5
0
reported by Tour and co-workers, which has been shown to
modify a significant proportion of carbons within the nanotube
5
1-53
sidewall.
Subsequent introduction of polystyrene was
achieved by first installing an azide functionality at the polymer
chain end. The Cu(I)-catalyzed formation of 1,2,3-triazoles by
coupling azide-terminated polymer and alkyne-functionalized
SWNTs was found to occur in an efficient manner under a
variety of favorable conditions. This resulted in relatively high
nanotube graft densities, full control over polymer molecular
weight, and good solubility in organic solvents.
One potential method for achieving high graft densities while
maintaining control over the polymer structure involves the
application of a modular approach where SWNTs bearing a
controllable number of highly reactive surface species are
coupled to separately prepared, end-functionalized polymers
using an efficient (preferably quantitative) coupling protocol.
For such an approach to be successful, the coupling reaction
would have to be relatively mild and highly selective so as to
allow incorporation of various functional groups on the polymer
Results and Discussion
In these studies, pristine SWNTs prepared by the HiPco
process were used without further treatment. p-Aminophenyl
propargyl ether (1) was prepared by etherification of p-
nitrophenol with propargyl bromide, followed by reduction to
the corresponding aniline derivative. The p-aminophenyl
propargyl ether (1) was subsequently reacted with SWNTs using
54
50
a solvent-free diazotization and coupling procedure to produce
37
without risk of side reactions. Recently, “click chemistry” has
attracted significant attention in polymer and materials science.
alkyne-functionalized SWNTs 2 (Scheme 1). This chemistry
allows for a high degree of functionalization and can be
performed on relatively large-scale. In a typical experiment, 160
mg of SWNTs was reacted with p-aminophenyl propargyl ether
(4 equiv/mol of carbon) in the presence of isoamyl nitrite (5
equiv/mol of carbon).
38
Click reactions are attractive for carrying out polymerizations
as well as for modification of macromolecules39-41 due to their
modular nature, selectivity, and high yields. This allows for
diverse arrays of building blocks to be prefabricated with
appropriate functional groups that allow their linkage in a single
convergent step. In particular, the copper(I)-catalyzed [3 + 2]
Huisgen cycloaddition4 has been utilized for the preparation
A series of well-defined polystyrene (PS) polymers was
prepared by atom transfer radical polymerization (ATRP) of
styrene using ethyl 2-bromoisobutyrate (EBiB) as an initiator
and a CuBr/2,2′-bipyridine (BPy) complex as the catalyst. All
polymerizations proceeded with good control, as evidenced by
low product polydispersity (Tables 1 and 2). In addition, the
ATRP reactions were stopped at approximately 60% conversion
to ensure that bromine end-groups were retained. Azide terminal
groups were installed on these polymers by end-group substitu-
2,43
(
25) Qin, S. H.; Oin, D. Q.; Ford, W. T.; Resasco, D. E.; Herrera, J. E. J. Am.
Chem. Soc. 2004, 126, 170-176.
(
26) Kong, H.; Gao, C.; Yan, D. Y. J. Am. Chem. Soc. 2004, 126, 412-413.
27) Yao, Z.; Braidy, N.; Botton, G. A.; Adronov, A. J. Am. Chem. Soc. 2003,
(
1
25, 16015-16024.
(
(
28) Liu, Y. Q.; Adronov, A. Macromolecules 2004, 37, 4755-4760.
29) Lou, X. D.; Detrembleur, C.; Sciannamea, V.; Pagnoulle, C.; Jerome, R.
Polymer 2004, 45, 6097-6102.
3
8
(
(
(
(
(
(
30) Qin, S. H.; Qin, D. Q.; Ford, W. T.; Herrera, J. E.; Resasco, D. E.; Bachilo,
S. M.; Weisman, R. B. Macromolecules 2004, 37, 3965-3967.
31) Qin, S. H.; Qin, D. Q.; Ford, W. T.; Resasco, D. E.; Herrera, J. E.
Macromolecules 2004, 37, 752-757.
tion with NaN3 in DMF, as depicted in Scheme 1. Complete
transformation of bromo-terminated PS to azide-terminated PS
1
was monitored using H NMR spectroscopy, which indicated a
32) Liu, Y. Q.; Yao, Z. L.; Adronov, A. Macromolecules 2005, 38, 1172-
1
179.
33) Riggs, J. E.; Guo, Z. X.; Carroll, D. L.; Sun, Y. P. J. Am. Chem. Soc.
000, 122, 5879-5880.
34) Sano, M.; Kamino, A.; Okamura, J.; Shinkai, S. Langmuir 2001, 17, 5125-
128.
(44) Wu, P.; Feldman, A. K.; Nugent, A. K.; Hawker, C. J.; Scheel, A.; Voit,
B.; Pyun, J.; Frechet, J. M. J.; Sharpless, K. B.; Fokin, V. V. Angew. Chem.,
Int. Ed. 2004, 43, 3928-3932.
2
5
(45) Joralemon, M. J.; O’Reilly, R. K.; Matson, J. B.; Nugent, A. K.; Hawker,
C. J.; Wooley, K. L. Macromolecules 2005, 38, 5436-5443.
(46) Helms, B.; Mynar, J. L.; Hawker, C. J.; Frechet, J. M. J. J. Am. Chem.
Soc. 2004, 126, 15020-15021.
35) Lin, Y.; Zhou, B.; Fernando, K. A. S.; Liu, P.; Allard, L. F.; Sun, Y. P.
Macromolecules 2003, 36, 7199-7204.
(
(
36) Fernando, K. A. S.; Lin, Y.; Sun, Y. P. Langmuir 2004, 20, 4777-4778.
37) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2001,
(47) Parrish, B.; Breitenkamp, R. B.; Emrick, T. J. Am. Chem. Soc. 2005, 127,
7404-7410.
4
0, 2004-2021.
(38) Tsarevsky, N. V.; Sumerlin, B. S.; Matyjaszewski, K. Macromolecules 2005,
(48) Mantovani, G.; Ladmiral, V.; Tao, L.; Haddleton, D. M. Chem. Commun.
2005, 2089-2091.
3
8, 3558-3561.
(
39) Malkoch, M.; Schleicher, K.; Drockenmuller, E.; Hawker, C. J.; Russell,
T. P.; Wu, P.; Fokin, V. V. Macromolecules 2005, 38, 3663-3678.
40) Opsteen, J. A.; van Hest, J. C. M. Chem. Commun. 2005, 57-59.
41) Lutz, J. F.; Borner, H. G.; Weichenhan, K. Macromol. Rapid Commun.
(49) Diaz, D. D.; Punna, S.; Holzer, P.; McPherson, A. K.; Sharpless, K. B.;
Fokin, V. V.; Finn, M. G. J. Polym. Sci., Part A: Polym. Chem. 2004, 42,
4392-4403.
(
(
(50) Dyke, C. A.; Tour, J. M. J. Am. Chem. Soc. 2003, 125, 1156-1157.
(51) Bahr, J. L.; Tour, J. M. Chem. Mater. 2001, 13, 3823-3824.
(52) Bahr, J. L.; Yang, J. P.; Kosynkin, D. V.; Bronikowski, M. J.; Smalley, R.
E.; Tour, J. M. J. Am. Chem. Soc. 2001, 123, 6536-6542.
(53) Dyke, C. A.; Tour, J. M. Chem.-Eur. J. 2004, 10, 813-817.
(54) Agag, T.; Takeichi, T. Macromolecules 2001, 34, 7257-7263.
2
005, 26, 514-518.
(
42) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2002, 41, 2596.
(43) Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002, 67, 3057-
3
064.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 41, 2005 14519

A R T I C L E S
Scheme 1 a
Li et al.
a
(i) Isoamyl nitrite, 60 °C; (ii) EBiB, CuBr/BPy, DMF, 110 °C; (iii) NaN3, DMF, room temperature; (iv) Cu(I), DMF.
Table 1. Results of the Polymerizations and Subsequent Coupling
Reactions in DMF with [(PPh3)3CuBr] as Catalyst
a
M
n,GPC
/
temp/
time/
h
solubility /
mg L-
1
1
run
polymer
kg mol-
PDI
°C
a
b
c
d
e
f
PS-N3 [3a]
PS-N3 [3b]
PS-N3 [3c]
PS-N3 [3b]
PS-N3 [3b]
PS-N3 [3b]
PS-N3 [3b]
2.01
4.83
8.62
4.83
4.83
4.83
4.83
1.10
1.09
1.13
1.09
1.09
1.09
1.09
100
100
100
80
60
100
100
65
65
65
65
65
48
24
112
232
87
136
24
180
83
g
a
The solubility of full-length polystyrene-functionalized SWNTs in THF
(see Supporting Information).
Table 2. Results of the Polymerizations and Subsequent Coupling
Reactions in DMF with CuI/1,8-Diazabicyclo[5.4.0]undec-7-ene
Figure 1. Photograph of three separate SWNT samples in THF. (A) Pristine
SWNTs; (B) alkyne-functionalized SWNTs 2; (C) polymer-functionalized
SWNTs 4.
(DBU) as Catalyst
a
M
n,GPC
/
temp/
time/
h
solubility /
mg L-
run
polymer
kg mol-
1
PDI
°C
1
the PS-SWNT conjugates. It is the polymer, therefore, that
dictates the solubility properties of the overall material. Figure
a
b
c
d
e
PS-N3 [3a]
PS-N3 [3b]
PS-N3 [3c]
PS-N3 [3b]
PS-N3 [3b]
2.01
4.83
8.62
4.83
4.83
1.10
1.09
1.13
1.09
1.09
60
60
60
90
20
24
24
24
24
48
97
233
172
226
88
1
shows three vials containing equal volumes of THF and equal
masses of pristine, unmodified SWNTs (vial A), alkyne-
functionalized SWNTs 2 (vial B), and the PS-SWNT conjugate
4 (vial C), where the PS number average molecular weight (Mn)
is ca. 4800 g/mol. Clearly, the umodified and alkyne-function-
alized SWNTs are both completely insoluble in THF (and all
other common solvents), although the latter material exhibits a
higher density than the former. The PS-SWNT conjugate
contained in vial C forms a clear, dark-brown solution that
exhibits no discernible particulate materials and remains stable
for a period of at least 3 weeks. This solubility is remarkable
considering that the dissolved SWNTs are, on average, greater
than 1 µm in length (vide infra).
Raman spectroscopy was not only utilized to verify the
structural integrity of the modified SWNT materials, but also
to gather information regarding the degree of nanotube func-
tionalization. Spectrum A in Figure 2 corresponds to pristine
SWNTs, which exhibit the characteristic radial breathing (ωr
a
The solubility of full-length polystyrene-functionalized SWNTs in THF
(
see Supporting Information).
clear shift of the methylene protons adjacent to the end-groups
from 4.4 to 3.9 ppm.
The alkyne-functionalized SWNTs (2) and azide-modified PS
3) were coupled via [3 + 2] Huisgen cycloaddition between
the alkyne and azide end groups using either CuI or the
organosoluble Cu(I) species, [(PPh3)3CuBr],44 as catalyst. All
(
reactions were performed in DMF, yielding polymer-function-
alized SWNTs (Scheme 1) as summarized in Table 1. In this
study, excess PS was used in an attempt to drive the click
reactions to high conversion, and this excess polymer was easily
removed after each reaction by ultra-filtration and prolonged
washing with THF. In addition, trace amounts of copper salts
in the products were removed by washing with an aqueous
ammonium hydroxide solution.
The first and simplest qualitative test to determine whether
the “click” coupling procedure was successful involved checking
for solubility of the product in common organic solvents. It was
found that the PS-functionalized SWNTs indeed exhibited high
solubility in THF, CH2Cl2, CHCl3, and other solvents known
to effectively solvate PS polymers. Conversely, any solvent in
which PS does not dissolve was also incapable of dissolving
-
1
-1
52
≈ 250 cm ) and tangential (ωt ≈ 1590 cm ) modes. In
-
1
addition, a weak disorder band (ω ≈ 1290 cm ) can be
observed, indicating the presence of a small number of sp3
hybridized carbons within the nanotube framework. Upon
reaction with aniline 1, a significant change in the disorder band
was observed, in good agreement with previous studies.5 Its
intensity increased dramatically relative to both the radial
breathing and the tangential modes, indicating that a large
0,52
14520 J. AM. CHEM. SOC.
9
VOL. 127, NO. 41, 2005

Functionalization of Single-Walled Carbon Nanotubes
A R T I C L E S
4
(spectrum C), having a polymer molecular weight of 4800
g/mol. Clearly, the unmodified SWNTs are devoid of any
identifiable functional groups, as expected (the band at ∼3500
-
1
cm corresponds to trace amounts of water present in the KBr
used for preparation of the sample pellet). However, small, but
-
1
clearly discernible, signals at 2120 and 3280 cm are present
in spectrum B, corresponding to the C-C and C-H stretching
frequencies of the appended terminal alkyne functionalities,
respectively. Additionally, signals corresponding to the C-C
and C-H stretches of the aromatic ring that serves as a linker
between the nanotubes and the alkyne functionality in compound
-
1
2
can be seen at 1660 and 2920 cm , respectively. Upon
cycloaddition, the IR spectrum of the product (Figure 4i, curve
5
5
C) bears the characteristic signals for polystyrene, indicating
that polymer had been grafted. Although the alkyne signal at
Figure 2. Raman spectra of (A) pristine SWNTs, (B) alkyne-functionalized
SWNTs (2), and (C) polystyrene-functionalized SWNTs (4).
-
1
2120 cm is weak, magnification of the three spectra in the
-
1
region between 2000 and 2250 cm indicates that it is not
present prior to reaction with p-aminophenyl propargyl ether
or after the Huisgen cycloaddition (Figure 4ii). The disappear-
ance of the alkyne stretch after the click coupling indicates that
most of the alkynes must have been consumed during this
reaction, although the low intensity of this IR absorption makes
quantitation of conversion difficult.
The successful implementation of the click reaction on the
SWNT sidewall prompted us to investigate the effect of reaction
time, temperature, catalyst type, and polymer molecular weight
on the solubility of the resulting SWNT-polymer conjugates.
Reaction conditions were systematically varied (Tables 1 and
2
), and their effect was evaluated by spectrophotometrically
estimating the relative solubility of the PS-SWNT conjugates
in THF. All samples were typically sonicated for 3 min and
then centrifuged at 5000 rpm for 20 min, followed by standing
undisturbed overnight prior to UV/vis absorption measurements,
which were performed on the supernatant of the resulting
sample. A previously determined specific extinction coefficient
Figure 3. Thermogravimetric analysis data for polystryrene-functionalized
SWNTs (4) having a polymer Mn ) 4800 g/mol, acquired under argon.
The temperature profile involved a 1 h hold at 150 °C followed by a ramp
-
1
of 5 °C min to 500 °C, and a 2.5 h hold at 500 °C.
2
3
-1
-1
number of sp hybridized carbons have been converted to sp
of 0.0103 L mg cm was used to estimate the nanotube
3
2
hybridization. The intensity of the tangential mode relative to
the radial breathing mode also increased, again in agreement
with previous observations.52 Upon reaction with azide-func-
tionalized polystyrene, the intensity of the disorder band relative
to the radial breathing and tangential modes remained un-
changed, as would be expected because the click reaction does
not alter the hybridization of carbon atoms within the nanotube
framework. To gain a more quantitative picture of the extent
of nanotube functionalization, thermogravimetric analysis (TGA)
was performed on the reaction product (Figure 3). A mass loss
of approximately 45%, due to polymer decomposition, was
observed when the product was heated to 500 °C under Ar for
concentrations. We first determined the effect of polymer
molecular weight, reaction time, and temperature using the
44
organosoluble Cu(I) species, [(PPh ) CuBr], as the catalyst
3 3
(all cycloaddition reactions were performed in DMF). Compar-
ing entries a, b, and c of Table 1 provides an indication that
polymer molecular weight has an impact on the solubility of
the conjugate materials. Interestingly, at a molecular weight of
4
800 g/mol, corresponding to approximately 50 styrene repeat
units, the polymer imparts the highest solubility at 233 mg/L.
When the molecular weight is decreased to ∼2000 g/mol, the
solubility drops by a factor of 2. This is likely due to a decreased
polymer-solvent interaction that diminishes the solubilizing
strength of the polymer at lower molecular weight. Alternatively,
when the molecular weight is incleased to ∼8600 g/mol, the
conjugate solubility also dropped. This result can be explained
by the decreased reactivity of the chain-end azide functionality
that becomes less accessible as the polymer adopts a more
random coil structure at this higher molecular weight. Therefore,
as the polymer molecular weight is increased, the polymer graft
density decreases, as also indicated by TGA. Clearly, a balance
between end-group accessibility and polymer solubilizing
2.5 h. Considering the polymer molecular weight of 4800 g/mol,
this mass loss corresponds to functionalization of approximately
one in 500 carbon atoms on the SWNT sidewalls. The TGA
experiment, when carried out using nanotubes grafted with 2000
and 8600 g/mol polymers, resulted in similar mass losses,
corresponding to a functionalization of one in 200 and one in
7
00 nanotube carbons, respectively.
IR spectroscopy provided information about the structures
appended to the surface of the SWNTs, which is not available
from the Raman data. Figure 4i depicts the IR spectra of
unmodified SWNTs (spectrum A), the alkyne-functionalized
SWNTs 2 (spectrum B), and the polymer-functionalized SWNTs
(
55) Silverstein, R. M.; Webster, F. X. Spectrometric Identification of Organic
Compounds, 6th ed.; John Wiley & Sons: New York, 1998.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 41, 2005 14521

A R T I C L E S
Li et al.
Figure 4. (i) IR spectra of (A) pristine SWNTs, (B) alkyne-functionalized SWNTs, and (C) polystyrene-functionalized SWNTs. (ii) Expanded region
-
1
-1
between 2000 and 2250 cm of the same three samples, highlighting the appearance and disappearance of the alkyne stretch at 2120 cm
.
strength must be reached to maximize solubility of the conju-
gates.
Comparing entries b, d, and e (Table 1) provides information
on the effect of reaction temperature on solubility of the final
product. Using this catalyst system, it appears that higher
temperatures improve the efficiency of the cycloaddition reac-
tion. The effect of time can also be gleaned from entries b, f,
and g, indicating that longer reaction times improve solubility,
but this improvement diminishes beyond reaction times of 48
h.
In an attempt to decrease reaction time and temperature, while
maintaining a high conjugate solubility, the catalyst system was
changed to CuI with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)
as an additive. This combination was previously reported by
Gin and co-workers as an effective combination for the Huisgen
cycloaddition.56 Using this catalyst system, it was possible to
achieve solubilities comparable to the ones with [(PPh3)3CuBr],
but at a reaction temperature of 60 °C after only 24 h (Table 2,
Figure 5. TEM image of polystyrene-functionalized SWNTs (inset is a
entry b). Longer reaction times under these conditions did not
lead to improved solubility. Again, it was found that the highest
solubility was achieved with the 4800 g/mol polymer (compare
entries a, b, and c of Table 2). Increasing the reaction
temperature to 90 °C resulted in no improvement of solubility
magnified version of the area within the white square).
or recycling homogeneous catalysts, where nanotube bound
reagents or catalysts are kept in homogeneous solution while
stirring, but precipitate when stirring is stopped at the end of
the reaction.
(entry d), while decreasing to 20 °C lowered the solubility
significantly, even when reaction time was doubled. It should
be noted that using the CuI/DBU system resulted in greater
solubility at a temperature of 20 °C than what was achieved
with the [(Ph3P)3CuBr] at 60 °C (compare entry e from Tables
Recently, Ford and co-workers reported the direct cycload-
dition of azide-terminated PS to unmodified SWNTs via nitrene
formation upon thermolysis of the terminal azide group. This
31
chemistry was carried out in 1,2-dichlorobenzene at 130 °C
under a nitrogen atmosphere, with a reaction time of 60 h. As
a control experiment, we attempted coupling our azide-
terminated PS to as-received, unmodified SWNTs. We found
that stirring a mixture of polymer, SWNTs, and either of the
Cu(I) catalysts in DMF at 100 °C for 90 h led to reaction
products that exhibited no solubility. We can therefore rule out
any direct nitrene coupling through the thermolysis of the azide-
terminated PS under our reaction conditions.
High-resolution transmission electron microscopy (TEM)
analysis of the PS-SWNT conjugates was performed by placing
a single drop of the THF solution onto a holey carbon-coated
copper grid. Figure 5 depicts representative features showing
interconnected fiberlike structures composed largely of amor-
phous carbon in which the characteristic striations indicative
of individual, embedded SWNTs are observed. The inset of
1
and 2). In all of these experiments, deoxygenation of the
reagent mixture at the beginning of each reaction was critical
to maintaining an active Cu(I) catalyst throughout the cycload-
dition. All mixtures were therefore evacuated and refilled with
N2 at least three times to remove oxygen before commencing
the reaction.
Interestingly, the SWNTs that were functionalized with 8600
g/mol PS tended to precipitate from solution over 24 h.
However, the precipitated material (in the form of a fine black
powder) could be redispersed to form a completely homoge-
neous solution by simply shaking the mixture for a few seconds,
without the need for sonication (see Supporting Information).
Such materials may have applications in solid-phase synthesis
(
56) Bodine, K. D.; Gin, D. Y.; Gin, M. S. J. Am. Chem. Soc. 2004, 126, 1638-
1
639.
14522 J. AM. CHEM. SOC.
9
VOL. 127, NO. 41, 2005

Functionalization of Single-Walled Carbon Nanotubes
A R T I C L E S
Figure 6. AFM images of pristine SWNTs (A-C) and polystyrene-functionalized SWNTs (D-F), with height profiles of different sections of images C
and F given on the right. Total scales of images vary from 15 µm (A and D) to 5 µm (B and E) and to 1 µm (C and F).
Figure 5 shows a particularly well-resolved feature in which
two or three individual polymer-coated SWNTs can be seen.
The presence of such individual SWNTs is indicative of
debundling of SWNTs as a result of polymer functionalization.
To complement the TEM data, atomic force microscopy
(AFM) was also performed. Figure 6 consists of two sets of
AFM images that represent the pristine SWNTs (images A-C)
and the 4800 g/mol PS-functionalized SWNTs 4 (images D-F).
In each set, the magnification increases such that the image scale
totals 15, 5, and 1 µm on going from left to right (magnified
sections of images A, B, D, and E are indicated by a white
box). The height profiles for the three marked sections of each
2
1
µm image are given on the right. The two samples were
prepared by spin-coating THF suspensions onto freshly cleaved
mica at 2500 rpm. Due to the lack of solubility of the unmodified
SWNTs, this compound required several minutes of sonication
immediately prior to the spin-coating procedure. Conversely,
the polymer-functionalized material was used after centrifugation
and standing (vide supra) and required no sonication as it
remained in the form of a stable solution for a period of at least
Figure 7. AFM surface plot of a PS-SWNT strand passing through a
separate multi-nanotube rope.
3
weeks. Comparing the lowest magnification images, it can
aggregate into thicker “ropes” exhibiting height profiles of ca.
be observed that the unmodified SWNTs result in very few
individually dispersed features (image A) that were relatively
difficult to find, with most of the sample forming large
aggregates that were not imageable without damaging the AFM
tip. However, the polymer-functionalized material leads to
numerous easily measurable nanotube-based features in every
1
5 nm. It is likely that the thicker nanotube ropes consist of
individual polymer-functionalized SWNTs that have become
associated with one another during the spin-coating process,
rather than bundles of SWNTs that are polymer-functionalized
only on their exterior. The polymer functionalization process
must therefore cause alkyne-functionalized SWNT bundles to
exfoliate, resulting in the formation of individual polymer-coated
nanotubes in solution. This is supported by the observation of
numerous features corresponding to ropes splitting into several
smaller polymer-coated strands as well as regions where one
strand passes between two strands of a separate rope (Figure
7). The observed high solubility of these structures, as well as
their long-term stability in solution (even during high-speed
centrifugation), are also consistent with individual nanotubes,
rather than larger bundles, having been grafted with polymer.
1
5 × 15 µm spot that was imaged. Comparing images C and
F, a clear difference due to polymer functionalization can be
observed. The polymer-functionalized material exhibits nanotube
strands that are thicker than the original SWNT bundles.
Although most of the material in image C resides in bundles
that are 4-7 nm in height, some individual tubes with heights
around 1 nm are observable. Upon functionalization, the isolated
soluble material exhibits individual nanotube strands that have
a relatively uniform polymer coating, with a minimum height
of approximately 4 nm. In this case, multiple strands still
J. AM. CHEM. SOC.
9
VOL. 127, NO. 41, 2005 14523

A R T I C L E S
Li et al.
Conclusion
Acknowledgment. We would like to thank Fred Pearson for
help with TEM measurements and Frank Gibbs for TGA data.
Financial support for this work was provided by the Natural
Science and Engineering Research Council of Canada (NSERC)
Strategic Grants program, the Canada Foundation for Innovation
(CFI), the Ontario Innovation Trust (OIT), and the Materials
and Manufacturing Ontario Emerging Materials Knowledge
Fund (EMK). H.L. would also like to gratefully acknowledge
the China Scholarship Council for financial support in the form
of a fellowship for international study.
We have demonstrated a highly efficient, modular approach
to the functionalization of SWNTs with polystyrene using the
Cu(I)-catalyzed [3 + 2] Huisgen cycloaddition. The separate
preparation of alkyne-functionalized SWNTs and azide-
terminated polymers allowed for characterization of both
materials prior to the convergent grafting step. A number of
reaction parameters were investigated, allowing the identification
of relatively mild coupling conditions, using CuI/DBU as the
catalyst system. TGA of the PS-SWNT conjugates indicated a
grafting density of 1 polymer chain for every 200-700 nanotube
carbons, resulting in a material that consisted of approximately
Supporting Information Available: Full experimental details
for all compounds, as well as a video clip showing dissolution
of polymer-functionalized SWNTs upon shaking. This material
is available free of charge via the Internet at http://pubs.acs.org.
4
5% polymer. These materials exhibited high solubility in
organic solvents such as THF, CHCl3, and CH2Cl2. TEM and
AFM analysis clearly showed the presence of polymer-coated
SWNTs in solutions that remained stable for at least 3 weeks.
Currently, we are extending this approach to the functionaliza-
tion of SWNTs with other polymeric and functional materials.
JA054958B
14524 J. AM. CHEM. SOC.
9
VOL. 127, NO. 41, 2005