Supramolecular functionalization of single-walled carbon nanotubes with triply fused porphyrin dimers: A study of structure-property relationships

Article abstract of DOI:10.1021/cm2004648

A triply fused porphyrin dimer bearing long alkyl chains for enhanced solubility was prepared and investigated for its ability to supramolecularly functionalize single-walled carbon nanotubes. It was found that this porphyrin dimer indeed binds strongly to the nanotube surface, allowing the removal of excess unbound porphyrin from solution without diminishing nanotube solubility. UV-vis spectroscopy indicated a bathochromic shift of the porphyrin Soret band upon binding to the nanotube surface, while Raman spectroscopy indicated that functionalization with the porphyrin dimer does not lead to any defects being formed on the nanotube wall. Transmission electron microscopy revealed the presence of individual nanotubes and small nanotube bundles that were heavily coated with porphyrin dimers. Comparison to analogous fused dimers bearing tert-butyl groups for solubility clearly demonstrated that long alkyl chains are necessary for prolonged solution stability of the nanotube complexes. While a previously investigated tert-butyl derivative was found to bind to the nanotube surface, the resulting complex precipitated out of solution within minutes. It was also found that the position of bulky substituents on the porphyrin dimer had a dramatic effect on whether it was able to interact with the nanotube surface.

Full text of DOI:10.1021/cm2004648

ARTICLE
pubs.acs.org/cm
Supramolecular Functionalization of Single-Walled Carbon
Nanotubes with Triply Fused Porphyrin Dimers: A Study of
StructureꢀProperty Relationships
Fuyong Cheng,† Jason Zhu,‡ and Alex Adronov*,‡
†Steacie Institute for Molecular Sciences, National Research Council Canada, Ottawa, Ontario, Canada
‡Department of Chemistry and the Brockhouse Institute for Materials Research, McMaster University, Hamilton, Ontario, Canada
ABSTRACT: A triply fused porphyrin dimer bearing long alkyl chains for enhanced solubility was
prepared and investigated for its ability to supramolecularly functionalize single-walled carbon nanotubes.
It was found that this porphyrin dimer indeed binds strongly to the nanotube surface, allowing the removal
of excess unbound porphyrin from solution without diminishing nanotube solubility. UVꢀvis spectros-
copy indicated a bathochromic shift of the porphyrin Soret band upon binding to the nanotube surface,
while Raman spectroscopy indicated that functionalization with the porphyrin dimer does not lead to any
defects being formed on the nanotube wall. Transmission electron microscopy revealed the presence of
individual nanotubes and small nanotube bundles that were heavily coated with porphyrin dimers.
Comparison to analogous fused dimers bearing tert-butyl groups for solubility clearly demonstrated that
long alkyl chains are necessary for prolonged solution stability of the nanotube complexes. While a
previously investigated tert-butyl derivative was found to bind to the nanotube surface, the resulting
complex precipitated out of solution within minutes. It was also found that the position of bulky
substituents on the porphyrin dimer had a dramatic effect on whether it was able to interact with the nanotube surface.
KEYWORDS: triply fused porphyrin dimer, porphyrin synthesis, single-walled carbon nanotube, supramolecular functionalization,
structureꢀproperty relationships, carbon nanotube solubility
’ INTRODUCTION
In these cases, although the resulting materials could be sus-
pended in solvent by sonication in the presence of excess
porphyrin, the complexes were not stable upon removal of
the excess unbound porphyrin, indicating that the interaction
strength was relatively weak and that an equilibrium between
bound and free porphyrin was established. From the perspective
of increasing the interaction strength, it is reasonable to expect
that multivalent binding, where macromolecular structures bear
multiple porphyrin units as side-chains or as polymer backbone
components, would be beneficial. To this end, nanotube com-
plexes with PMMA19 or polypeptide20 polymers bearing por-
phyrins as side chains have been prepared and were found to be
relatively soluble in polar organic solvents, such as DMF.
Conjugated porphyrin polymers containing butadiyne bridges
linked directly to the porphyrin meso carbons are fully con-
jugated, rigid, and strongly electron-donating structures.21ꢀ23
These properties make conjugated porphyrin polymers highly
complementary to the conjugated, rigid, but electron-accepting
SWNTs. It was therefore feasible that the interaction between
SWNTs and conjugated porphyrin polymers would be much
stronger than the analogous interaction with monomeric por-
phyrins. Recently, we have found that such polymers indeed form
strong interactions with the nanotube surface in THF.24,25
Unfortunately, structural flexibility within the polymer backbone
Porphyrins have been extensively investigated in practically all
disciplines of science, including chemistry, physics, biology, and
medicine.1 Such immense interest stems from their ability to
serve as ligands for a variety of metals,1 to catalyze a number of
reactions,2 and to absorb and convert light into other forms of
energy.3 Recently, porphyrins have been increasingly investi-
gated as components of nanoscale functional assemblies. In parti-
cular, the covalent and supramolecular chemistry of porphyrins
with carbon-based materials, such as fullerenes, has received sig-
nificant attention.4 In addition, functionalization of carbon nano-
tubes with porphyrins has been increasingly investigated since
interaction of porphyrins with the extended π-electron network of
carbon nanotubes leads to complexes with interesting optoelec-
tronic properties.5ꢀ7 The preparation of porphyrinꢀnanotube
conjugates can be achieved through covalent functionalization,
such as esterification with oxidized (carboxylic acid-bearing) nano-
tubes,8 1,3-dipolar cycloaddition,9 diazonium chemistry,10 and
Suzuki coupling to a prefunctionalized nanotube surface.11 How-
ever, all of these covalent methods introduce defect sites on the
carbon nanotube sidewall and greatly diminish the stability and
conductivity properties of the resulting materials.12,13
Conversely, noncovalent functionalization of carbon nano-
tubes with porphyrins is particularly attractive as it enables the
introduction of various porphyrin structures without disturbing
the π-system of the nanotube framework.14 In this approach,
significant attention has been given to the use of monomeric
porphyrin molecules as supramolecular nanotube adducts.15ꢀ18
Received: February 14, 2011
Revised:
May 24, 2011
Published: June 14, 2011
r
2011 American Chemical Society
3188
dx.doi.org/10.1021/cm2004648 Chem. Mater. 2011, 23, 3188–3194
|

Chemistry of Materials
ARTICLE
’ RESULTS AND DISCUSSION
Three different triply fused porphyrin dimers, 1, 2, and 3, were
investigated as part of this study (structures shown in Figure 1).
These three structures differ in the substituent types that are
located at the two different meso positions, labeled A and B
(Figure 1). Compound 1 exhibits only bulky 3,5-di-tert-butyl-
phenyl substituents, while compound 2 is functionalized with 3,
5-di-tert-butylphenyl at positions A and the less sterically deman-
ding 4-cyanophenyl substituents at positions B. Both of these
structures weresynthesized according to literatureprocedures.29,30
Dimer 3 is functionalized with 3,5-di-n-tridecylphenyl substitu-
ents at positions A for enhanced solubility and 4-(1-oxyethyl)-
phenyl as sterically nondemanding substituents in the B posi-
tions. The synthetic approach for the preparation of porphyrin
dimer 3 is shown in Scheme 1. Briefly, treatment of commercially
available 5-bromo-m-xylene with NBS resulted in 1-bromo-3,
5-bis(bromomethyl) benzene, which was reacted with P(OMe)3
to give diphosphonate 5. Subsequent treatment of 5 with dodecyl
aldehyde via the HornerꢀWadsworthꢀEmmons reaction, fol-
lowed by hydrogenation in the presence of palladium/carbon
catalyst, afforded 1-bromo-3,5-di-n-tridecylbenzene (7). Phenyl
bromide 7 was then reacted with BuLi and DMF to produce the
long alkyl chain substituted benzaldehyde 8 (24% yield over five
steps). Condensation of aldyhede 8 with dipyrromethane and
metalation with Zn(II) acetate afforded Zn-porphyrin 10. Por-
phyrin 10 was then treated with one equivalent of NBS, yielding
porphyrin bromide 11, which was reacted with the phenylboro-
nic ester 12 by Pd-catalyzed Suzuki cross-coupling to give the
porphyrin derivative 13. Porphyrin monomer 13 was then con-
verted into the triply fused dimer 3 in one step using the oxidative
ring closure mediated by DDQ and Sc(OTf)3.30
Figure 1. General structure of the triply fused porphyrin dimer super-
imposed on the structure of a carbon nanotube (top, not to scale) and
the chemical structures of the substituents on the three porphyrin dimers
(1, 2, and 3) used in this study (bottom).
Our recent work has shown that a solution of THF containing
5% TFA is a good solvent medium to form and solubilize the Zn-
porphyrin polymer and triply fused porphyrin oligomer com-
plexes with SWNTs.24,29 In the present work, this solvent system
was again used for the investigation of supramolecular interac-
tions between the three porphyrin dimers and SWNTs. In a
typical experiment, a HiPco SWNT sample (1.0 mg) was added
to separate solutions of porphyrin dimers 1, 2, and 3 in the THF/
TFA solution (0.4 mg/mL), and the mixture was sonicated for
1 h. The resulting nanotube suspension was filtered through a
200 nm pore-diameter Teflon membrane and washed repeatedly
with the 5% TFA in THF solution until all the excess porphyrin
dimer was removed, as indicated by the complete loss of color in
the filtrate. The remaining residue was then resuspended in 5 mL
of the 5% TFA/THF solution by sonicating for 5 min. When this
procedure was carried out with dimer 1, sonication of the residue
did not result in any observable SWNT solubility, indicating that
the washing procedure likely removed all the porphyrin dimers,
and no significant interaction between 1 and the SWNTs exists.
In contrast, when the same procedure was carried out with a
solution of dimer 2, sonication of the washed nanotube resi-
due resulted in the initial formation of a homogeneous solution.
However, as observed previously,29 upon standing for several
hours, the SWNTs were found to completely precipitate, leaving
a clear, colorless supernatant. Conversely, when the nanotubes
were added to a solution of dimer 3, and the above procedure was
again repeated, a very stable, dark solution was obtained after
sonication of the washed residue for 5 min. This solution re-
mained stable upon standing indefinitely, and no sedimentation
was observed even after centrifugation for 20 min at 2576g.
limited the binding strength to nanotubes and allowed equilibra-
tion between bound and unbound polymers.24
To increase the nanotube interaction strength, we recently
turned to triply fused Zn(II)-porphyrin oligomers, in which each
repeat unit is absolutely restricted to a fully coplanar orientation
with its neighbor.26ꢀ28 This coplanarity produces “molecular
tapes” that exhibit enhanced conjugation and greater electron-
donating ability relative to those of all other porphyrin oligomers
and polymers. These triply fused oligomers were therefore
deemed ideal candidates for strong supramolecular interactions
with SWNTs. In our recent work, we investigated the nanotube
binding strength of tert-butyl-functionalized triply fused porphyr-
in dimer and trimer molecules,29 relative to the analogous mono-
meric control compound, and found that the trimer alone formed
extremely strong and nearly irreversible supramolecular interac-
tions with SWNTs, leading to stable solutions.29 Interestingly, in
the case of the dimer, we found that a highly concentrated
homogeneous solution could be formed upon sonication, but the
complexes precipitated on standing for several hours. We there-
fore hypothesized that either the binding strength between the
dimer and the nanotubes is too low to impart long-term solution
stability or the tert-butyl groups in this structure do not provide
enough solubility to keep the nanotubeꢀporphyrin complexes in
solution. In order to answer this question, we have undertaken
the synthesis and investigation of a triply fused porphyrin dimer
that is functionalized with n-tridecyl chains for enhanced solubi-
lity. Here, we compare the nanotube binding and solubilizing
ability of this new porphyrin dimer to that of two previously
reported analogous structures (Figure 1).
3189
dx.doi.org/10.1021/cm2004648 |Chem. Mater. 2011, 23, 3188–3194

Chemistry of Materials
ARTICLE
Scheme 1. Synthesis of Porphyrin Dimer 3a
a Conditions: (a) NBS, CCl4, 30%; (b) P(OMe)3, 100%; (c) C11H23CHO, NaH, THF, 85%; (d) Pd/C, H2, THF, 100%; (e) BuLi, THF, DMF, 97%; (f)
dipyrromethane, TFA, DDQ, CH2Cl2, 4%; (g) Zn(OAc)2, CHCl3/MeOH, 95%; (h) NBS, CH2Cl2/pyridine; (i) 12, Pd(PPh3)4, Cs2CO3, toluene, 73%
(over two steps); (j) DDQ, Sc(OTf)3, toluene, 60%.
These observations can be explained by considering the posi-
tion and structure of the substituents on the porphyrin dimer. To
maximize π-stacking interactions, it is expected that the porphyr-
in dimer would preferentially orient its long axis along the length
of the SWNT structure (Figure 1). In this configuration, large
solubility-enhancing substituents bound to the porphyrin meso-
carbons at the two ends (positions B) can actually prevent the
close interaction of the dimer with the nanotube surface that is
necessary for the π-stacking interaction to be significant. Con-
versely, large substituents that are present at the sides of the
porphyrin dimer (positions A) will not hinder π-stacking and will
only serve to impart solubility to the overall structure. It is
therefore not surprising that dimer 1, which bears the bulky 3,5-
di-tertbutylphenyl groups at both ends, does not exhibit any
Figure 2. UVꢀvis spectra of porphyrin dimer 3 (black) and the 3-
SWNT complex (red).
significant interaction with the SWNT wall and cannot impart
even temporary solubility to the 1-SWNT complex. The struc-
ture of 2, which only differs from that of dimer 1 by the linear and
relatively nonsterically encumbered p-cyanophenyl groups at its
two ends, leads to a temporary solubilization of the 2-SWNT
complex. This is understandable because the two end groups do
not hinder the π-stacking interactions between the porphyrin
dimer and the SWNT surface, allowing the formation of a stable
complex. However, in this structure, the flanking 3,5-di-tert-
butylphenyl groups that were installed to impart solubility are
only temporarily effective in this role. Upon sonication, nanotube
solubility is observed, but the 2-SWNT complexes precipitate
over time as the t-butyl solubilizing groups are too rigid and not
effective enough in permanently stabilizing the complexes in
solution. However, the structure of dimer 3 exhibits both types of
necessary substituents for maximal nanotube interactions and
solubility enhancement. The relatively small and unencumbered
4-acetylphenyl substituents present at the two ends do not hinder
close π-stacking interactions with SWNTs, while the multiple
flexible tridecyl chains significantly enhance the solubility of the
3-SWNT complex, to the point that the complexes remain stable
in solution indefinitely. It should be noted that a previously
described monomeric porphyrin unit with long alkyl side-
chains,24 analogous to those discussed above, was found to not
impart any solubility to SWNTs. Thus, a balance between the
extended π-conjugation of the porphyrin oligomers and the size
of solubilizing side chains must be achieved in order to impart a
high degree of solubility to the porphyrinꢀSWNT complexes. As
the oligomer gets shorter, the side chain must become longer,
and vice versa.
The supramolecular interaction of dimer 3 with SWNTs was
investigated by UVꢀvis absorption spectroscopy. Figure 2 de-
picts the UVꢀvis absorption spectra of dimer 3 and the 3-SWNT
complex in acidified THF. Upon complexation of the dimer to
3190
dx.doi.org/10.1021/cm2004648 |Chem. Mater. 2011, 23, 3188–3194

Chemistry of Materials
ARTICLE
Figure 3. Representative TEM images of the 3-SWNT complex deposited from solution. Arrows indicate several positions of dimer-functionalized
nanotube sections (black arrows) and sections of bare nanotubes (white arrows). Most of the nanotube area is covered by the porphyrin dimer. Scale bars
correspond to 10 nm.
Figure 4. Raman spectra of pristine SWNTs (A) and the 3-SWNT complex. The inset shows a close-up of the spectral region between 100 and
350 cmꢀ1, corresponding to the radial breathing mode (RBM) region.
SWNTs, a bathochromic shift relative to the dimer alone was
observed. The longer wavelength component of the split Soret
band was observed to shift from 576 to 603 nm. The shift in the
absorption spectra upon oligoporphyrin complexation to
SWNTs is a clear manifestation of electronic interactions be-
tween the porphyrins and the nanotubes, possibly signifying a
porphyrin-to-nanotube electron transfer.
Further characterization of the 3-SWNT complex was accom-
plished by transmission electron microscopy (TEM). A drop of
the 3-SWNT solution, prepared as described above, was placed
on a holey carbon coated copper TEM grid and was allowed to
air-dry prior to loading into the microscope. Figure 3 shows four
representative images of the sample, indicating the presence of
long fiber-like structures that are mostly coated along their length
3191
dx.doi.org/10.1021/cm2004648 |Chem. Mater. 2011, 23, 3188–3194

Chemistry of Materials
ARTICLE
with organic material. These structures clearly resemble exfo-
liated carbon nanotubes and small nanotube bundles that have
been heavily coated with the porphyrin dimer 3 (indicated by
black arrows). A few areas where individual uncoated nanotubes
can be seen are indicated by white arrows (Figure 3).
instrument. FTIR was performed on a BIO-RAD FTS-40 instrument.
UVꢀvis spectra were measured using a Cary 50 UVꢀvisible spectro-
photometer. High resolution ESI-MS measurements were done on
the Micromass Ultima Global instrument (quadrupole time-of-flight),
and MALDIꢀMS was performed on the Waters/Micromass MALDI
Micro instrument (R-cyano-4-hydroxycinnamic acid as matrix). Raman
spectroscopy was performed on a Renishaw InVia Raman spectrometer
equipped with a 25 mW argon ion laser (514 nm), a 300 mW Renishaw
785 nm laser, and 1800 L/mmand 1200 L/mm gratings for the two lasers,
respectively. The Raman system also is equipped with a Leica microscope
having 5ꢁ, 20ꢁ, and 50ꢁ objectives as well as a USB camera for sample
viewing. The 785 nm laser was operated at 5% intensity to avoid damage
to the sample. Ultrasonication was done in a Branson Ultrasonics B1510
bath sonicator. Filtration was done through a 200 nm-pore Teflon mem-
brane (Millipore).
Synthesis of 1-Bromo-3,5-bis(bromomethyl) Benzene (4).
5-Bromo-m-xylene (34.94 g, 0.19 mol), N-bromo-succinimide (70.57 g,
0.40 mol), and benzoyl peroxide (0.5 g, 2.1 mmol) were added to dry
CCl4 (300 mL) in a 500 mL round-bottom flask, equipped with a stir bar
and a reflux condenser, under Ar. The reaction was then heated to reflux
in the dark and stirred for 16 h. The mixture was filtered through a fritted
funnel, and the filtrate was evaporated in vacuo. Recrystallization from
the minimum amount of hexanes afforded the desired product 4 as a
white solid (19.43 g, 30%). 1H NMR (CDCl3, 200 MHz): δ 4.40
(s, 4 H), 7.33 (s, 1 H), 7.46 (s, 2 H). 13C NMR (CDCl3, 50 MHz): δ
31.44, 122.67, 128.23, 131.94, 140.27. IR (KBr, cmꢀ1): 1580 (m), 1445
(m), 1262 (m), 1212 (s), 878 (m), 864 (m), 818 (w), 691 (s), 602 (s),
586 (m), 539 (s), 484 (w). HRMS (EI) m/z calcd for C8H7Br3 (Mþ)
339.8098; found, 339.8085.
Additional characterization of the 3-SWNT complexes was
accomplished by Raman spectroscopy with excitation at 785 nm.
The characteristic absorptions at ∼1590 cmꢀ1, ∼1300 cmꢀ1
,
and 150ꢀ300 cmꢀ1, corresponding to the graphitic (G) band,
the disorder (D) band, and the radial breathing modes (RBMs)
of SWNTs, respectively, are clearly visible (Figure 4, curve B). A
comparison to pristine, unfunctionalized SWNTs (Figure 4,
curve A) revealed only a few differences. First, the line shape of
the G band of 3-SWNT shows a slightly decreased Gꢀ band
(∼1560 cmꢀ1) relative to the pristine sample of SWNTs, indi-
cating a lower concentration of metallic nanotubes.31 Similarly,
the D band has a slightly lower intensity in 3-SWNT, indicating
that functionalization with porphyrin dimer 3 does not lead to
any defects being formed on the nanotube wall, as expected for
supramolecular functionalization. Finally, close examination of
the RBM signals of 3-SWNT indicates that signals at ∼260 and
∼220 cmꢀ1 are increased in intensity relative to the other signals
of pristine SWNTs, while the small signal at 150 cmꢀ1, present in
the pristine nanotubes, is undetectable in 3-SWNT (Figure 4,
inset). These signals could be interpreted to indicate a slight
preference of the dimer for specific nanotube diameters,32ꢀ34 but
it is difficult to make conclusions with these polydisperse,
surface-functionalized SWNT samples. Surface functionalization
can shift peak positions, and can change the nanotube types that
are in resonance with the excitation wavelength.
Synthesis of Tetramethyl [(5-Bromo-1,3-phenylene)bis-
(methylene)]bis-phosphonate (5). Compound 4 (18.27 g, 53.29
mmol) was added to trimethyl phosphite (200 mL) in a 500 mL round-
bottom flask, equipped with a stir bar and a reflux condenser, under inert
Ar atmosphere. The reaction was heated to reflux for 16 h, then the
mixture was cooled, and thephosphite was removed in vacuo. The desired
’ CONCLUSIONS
This work demonstrates that triply fused porphyrin dimers can
exhibit strong interactions with the nanotube surface, leading to
prolonged solution stability, as long as they are appropriately func-
tionalized. To impart the observed solubility, large solubilizing
groups (long alkyl chains) must be installed along the long axis of
the dimer structure. However, sterically bulky substituents at the
two ends (along the short axes) of the triply fused dimer hinder the
close interaction of the porphyrin with the nanotube surface and
should be avoided. In the present case, tridecyl chains along the
long axis and methyl phenyl ketone groups along the short axis
successfully provided the desired interaction strength and nano-
tube solubility. It was found that excess porphyrin dimer could be
removed by ultrafiltration and excessive washing with THF,
resulting in a residue that retained its solubility. TEM analysis
revealed the presence of individual nanotubes and small nanotube
bundles that were heavily coated with porphyrin molecules, while
UVꢀvis spectroscopy indicated that the porphyrin Soret band
exhibits a bathochromic shift upon nanotube complexation. De-
tailed Raman analysis indicated that complexation of nanotubes by
the porphyrin strucures does not lead to any structural changes to
the nanotubes.
1
product 5 was obtained as a white solid (21.61 g, 100%). H NMR
(CDCl3, 200 MHz): δ 3.07 (d, 4 H, J = 22.0 Hz), 3.68 (d, 12 H, J = 10.8
Hz), 7.15 (s, 1 H), 7.32 (s, 2 H). 13C NMR (CDCl3, 50 MHz): δ 30.93,
33.68, 52.82, 52.96, 122.53, 129.74, 131.25, 133.62. IR (KBr, cmꢀ1):
2952 (w), 2916 (w), 1569 (w), 1444 (w), 1240 (s), 1224 (m), 1050 (s),
1032 (s), 863 (m), 835 (w), 805 (w), 724 (w), 688 (m), 556 (w),
537 (w). HRMS (EI) m/z calcd for C12H19BrO6P2 (Mþ) 399.9840;
found, 399.9823.
Synthesis of 1-Bromo-3,5-di-1-tridecenyl-benzene (6). Do-
decyl aldehyde (39.48 g, 0.21 mol) and NaH (40 g, 1.67 mol) were
added to THF (250 mL) in a 500 mL three-neck round-bottom flask,
equipped with a stir bar, an addition funnel, and a reflux condenser,
under inert Ar atmosphere. Diphosphonate 5 (21.50 g, 53.57 mmol) was
dissolved in THF (150 mL), transferred to the funnel, and added
dropwise to the reaction vessel. After heating at reflux for 16 h, water
(50 mL) was added dropwise to the mixture until effervescence sub-
sided. THF was removed in vacuo. Organic compounds were extracted
into hexanes (100 mL) and subsequently washed with water (3 ꢁ
100 mL). Hexane was removed in vacuo. The desired product 6 was
obtained as a white solid (23.64 g, 85%). 1H NMR (CDCl3, 200 MHz):
δ 0.88 (t, 6 H, J = 6.5 Hz), 1.27 (m, 36 H), 2.20 (m, 4 H), 6.26ꢀ6.24 (m,
4 H), 7.17 (s, 1 H), 7.30 (s, 2 H). 13C NMR (CDCl3, 50 MHz): δ 14.12,
22.68, 29.22, 29.64, 31.91, 33.00, 122.45, 122.83, 126.86, 128.40, 132.81,
139.98. IR (KBr, cmꢀ1): 2924 (s), 2853 (s), 1706 (w), 1464 (m),
963 (w). HRMS (EI) m/z calcd for C32H53Br (Mþ) 516.3331; found,
516.3322.
’ EXPERIMENTAL SECTION
General. Single-walled carbon nanotubes (SWNTs) were pur-
chased from Carbon Nanotechnologies, Inc. (Houston, TX). All other
reagents were purchased from commercial suppliers and used as
received. TEM analysis was performed using JEOL 2010F operating
at 200 keV. NMR was performed on a Bruker 200, 500, or 600 MHz
Synthesis of 1-Bromo-3,5-di-n-tridecyl-benzene (7). Com-
pound 6 (23.64 g, 45.65 mmol) and palladium/carbon catalyst (0.5 g)
were added to THF (200 mL) in a 500 mL round-bottom flask,
3192
dx.doi.org/10.1021/cm2004648 |Chem. Mater. 2011, 23, 3188–3194

Chemistry of Materials
ARTICLE
equipped with a stir bar, under inert Ar atmosphere. The mixture was
stirred vigorously at room temperature (22 °C) for 20 h. The mixture
was then filtered through Celite and then silica, washing with THF (3 ꢁ
20 mL). THF was removed in vacuo. The target product 7 was obtained
as a white solid (23.62 g, 100%). 1H NMR (CDCl3, 200 MHz): δ 0.88 (t,
6 H, J = 6.1 Hz), 1.25 (m, 40 H), 1.57 (m, 4 H), 2.52 (t, 2 H, J = 7.7 Hz),
6.89 (s, 1 H), 7.12 (s, 2 H). 13C NMR (CDCl3, 50 MHz): δ 14.13, 22.68,
29.66, 31.32, 31.59, 31.93, 35.63, 127.34, 128.59, 144.97. IR (KBr, cmꢀ1):
2924 (s), 2853 (s), 1600 (w), 1569 (m), 1464 (m). HRMS (EI) m/z
calcd for C32H57Br (Mþ) 520.3644; found, 520.3626.
127.13, 127.92, 128.75, 131.15, 131.38, 132.94, 141.06, 141.29, 145.08,
147.32. IR (KBr, cmꢀ1): 2923 (s), 2852 (s), 1596 (m), 1464 (m), 1392
(w), 1059 (m), 994 (m), 848 (m), 782 (m), 726 (w). UVꢀvis (CHCl3,
λmax, nm): 436, 627. MALDI-TOF-MS m/z calcd for C84H124N4Zn
(Mþ) 1252.9117; found, 1252.90.
Synthesis of [10,20-Bis(3,5-di-n-tridecylphenyl)-5-bromo-
porphyrinato(2ꢀ)-kN21,kN22,kN23,kN24]zinc(II) (11). Zn-por-
phyrin 10 (250 mg, 0.20 mmol) and pyridine (0.3 mL) were added to
CH2Cl2 (30 mL) in a 100 mL round-bottom flask, equipped with a stir
bar. The solution was cooled to 0 °C and NBS (39 mg, 0.22 mmol) was
added under inert Ar atmosphere. After 10 min, the reaction was
quenched with acetone (4 mL), and the solvent was removed in vacuo.
The residue was washed with methanol (3 ꢁ 50 mL) and dried in vacuo.
A mixture of unbrominated, monobrominated, and dibrominated por-
phyrin was obtained as a purple solid (290 mg) and used directly in the
Suzuki coupling reaction without product isolation.
Synthesis of 3,5-Di-n-tridecyl-benzaldehyde (8). Compound
7 (23.00 g, 44 mmol) was added to THF (200 mL) in a 500 mL round-
bottom flask, equipped with a stir bar, under inert Ar atmosphere
and cooled to ꢀ60 °C. While stirring, BuLi (60 mL, 2.5 M in hexane,
150 mmol) was added dropwise. After 30 min, DMF (20 mL) was added
dropwise. The mixture was warmed to room temperature. After 2 h,
5 vol % H2SO4 (50 mL) was added. CH2Cl2 (500 mL) was added to the
mixture, which was then washed with brine. CH2Cl2 was then removed in
vacuo. The product was purified by column chromatography on silica gel
using hexanes as the initial eluent and then switching to CH2Cl2. The
solvent was then removed in vacuo. The target product 8 was obtained as
Synthesis of 2-(4-Acetylphenyl)-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (12). Pinacol (10.81 g, 92 mmol) and 4-acetylphe-
nylboronic acid (5 g, 31 mmol) were added to ethyl ether (150 mL) in a
250 mL round-bottom flask, equipped with a stir bar. The reaction was
stirred for 16 h at room temperature. The mixture was then washed with
water (5 ꢁ 100 mL) to remove excess pinacol, and ether solvent was
removed in vacuo. The target product 12 was obtained as a pale yellow
solid (6.72 g, 90%). 1H NMR (CD2Cl2, 200 MHz): δ 1.35 (s, 12 H),
2.61 (s, 3 H), 7.87 (d, 2 H, J = 8.3 Hz), 7.93 (d, 2 H, J = 8.1 Hz). 13C
NMR (CDCl3, 50 MHz): δ 24.78, 26.65, 84.10, 127.17, 134.83, 138.90,
198.27. IR (KBr, cmꢀ1): 2980 (m), 1684 (s), 1555 (w), 1507 (m), 1397
(s), 1358 (s), 1265 (s), 1214 (w), 1142 (m), 1094 (m), 1015 (w), 961
(w), 857 (m), 827 (w), 731 (w), 655 (m), 670 (m), 599 (w). HRMS
(EI) m/z calcd for C14H19BO3 (Mþ) 246.1427; found, 246.1433.
Synthesis of [10,20-Bis(3,5-di-n-tridecylphenyl)-5-(4-acetyl-
phenyl)porphyrinato(2ꢀ)-kN21,kN22,kN23,kN24]zinc(II) (13).
The bromo-porphyrin mixture 11 (290 g), borate 12 (160 mg, 0.651
mmol), and CsCO3 (0.7 g, 2.17 mmol) were added to toluene (60 mL)
in a 100 mL round-bottom flask, equipped with a stir bar, and deaerated
under inert Ar atmosphere by sonication for 40 min. The solution was
then heated to reflux and stirred for 16 h. TLC revealed three distinct
products. The solution was then filtered through a Celite plug and
washed with toluene (3 ꢁ 15 mL). The solvent was removed from the
filtrate in vacuo. The products were purified by column chromatography
on silica gel. Hexanes/CH2Cl2 (5:1) was used to elute the unfunctio-
nalized porphyrin, hexanes/CH2Cl2 (2:1) eluted the monofunctiona-
lized porphyrin, and pure CH2Cl2 eluted the difunctionalized porphyrin.
The combined two-step process of porphyrin bromination and Suzuki
coupling afforded the monofunctionalized porphyrin 13 as a red solid
(200 mg, 73%) and difunctionalized porphyrin 14 as a purple solid (70
mg, 23%). Data for 13: 1H NMR (CDCl3, 200 MHz) δ 0.83 (t, 12 H, J =
6.3 Hz), 1.21 (m, 40 H), 1.84 (m, 8 H), 2.85 (s, 3 H), 2.85 (t, 8 H, J = 7.5
Hz), 7.40 (s, 2 H), 7.87 (s, 4 H), 8.31 (s, 4 H), 8.89 (d, 2 H, J = 4.6 Hz),
9.03 (d, 2 H, J = 4.8 Hz) 9.13 (d, 2 H, J = 4.8 Hz), 9.38 (d, 2 H, J = 4.4),
10.24 (s, 1 H). 13C NMR (CDCl3, 50 MHz): δ 14.13, 22.70, 26.52,
29.67, 31.93, 36.09, 105.77, 119.44, 121.37, 126.34, 127.72, 131.23,
131.52, 132.31, 132.68, 134.58, 135.69, 140.87, 142.25, 148.39, 148.93,
149.59, 150.25, 198.29. IR (KBr, cmꢀ1): 2923 (s), 2852 (s), 1688 (m),
1597 (w), 1465 (m), 1263 (w), 995 (m), 791 (w), 717 (w). UVꢀvis
(CHCl3, λmax, nm): 444, 642. MALDI-TOF-MS m/z calcd for
C92H130N4OZn (Mþ) 1370.9536; found, 1370.93.
1
a white solid (19.50 g, 94%). H NMR (CD2Cl2, 200 MHz): δ 0.87
(t, 6 H, J = 7.7 Hz), 1.27 (m, 40 H), 1.61 (m, 4 H), 2.65 (t, 2 H, J = 7.7
Hz), 7.27 (s, 1 H), 7.49 (s, 2 H), 9.94 (s, 1 H). 13C NMR (CDCl3, 50
MHz): δ 22.68, 29.25, 29.35, 29.45, 29.65, 31.31, 31.91, 35.61, 127.12,
134.99, 136.62, 143.80, 192.81. IR (KBr, cmꢀ1): 2918 (s), 2850 (s), 1697
(m), 1595 (w), 1465 (m), 1142 (m), 721 (w). HRMS (EI) m/z calcd for
C33H58O (Mþ) 470.4488; found, 470.4491.
Synthesis of 5,15-Bis(3,5-di-n-tridecylphenyl)-porphyrin
(9). Aldehyde 8 (1.94 g, 4.13 mmol) and dipyrromethane (0.6 g, 4.11
mmol) were added to deaerated CH2Cl2 (1 L) in a 2 L round-bottom
flask, equipped with a stir bar, under inert Ar atmosphere. TFA (0.3 mL)
was added, and the reaction was stirred in the dark for 16 h at room
temperature. DDQ (3.1 g, 13.65 mmol) and triethylamine (1.0 mL)
were then added, and the solution was stirred for a further 2 h. CH2Cl2
was removed in vacuo. The product was first roughly isolated by column
chromatography on silica gel using a mixture of hexanes/CH2Cl2 (1:1)
as the eluent, then finely purified using a second round of column
chromatography on silica gel using a mixture of hexanes/CH2Cl2 (5:1)
as the eluent. The solvent was then removed in vacuo. The desired
1
product 9 was obtained as a purple solid (200 mg, 4%). H NMR
(CDCl3, 200 MHz): δ ꢀ3.08 (s, 2 H), 0.86 (t, 12 H, J = 6.4 Hz), 1.27 (m,
40 H), 1.90 (m, 8 H), 2.92 (t, 8 H, J = 7.6 Hz), 7.45 (s, 2 H), 7.93 (s, 4 H),
9.14 (d, 4 H, J = 4.7 Hz), 9.39 (d, 4 H, J = 4.7 Hz), 10.30 (s, 2 H). 13C
NMR (CDCl3, 50 MHz): δ 14.16, 22.73, 29.70, 31.85, 31.96, 36.09,
105.08, 119.77, 127.91, 131.17, 131.40, 132.92, 141.04, 141.30, 145.08,
147.29. IR (KBr, cmꢀ1): 2922 (s), 2851 (s), 1632 (m), 1590 (w), 1435
(m), 1162 (m), 731 (w). UVꢀvis (CHCl3, λmax, nm): 436, 574, 625,
674. HRMS (ESI) m/z calcd for C84H127N4 (Mþ) 1192.0061; found,
1191.9993.
Synthesis of [5,15-Bis(3,5-di-n-tridecylphenyl)-porphyri-
nato(2ꢀ)-kN21,kN22,kN23,kN24] Zinc(II) (10). A solution of Zn-
(OAc)2 2H2O (1.16 g, 5.3 mmol) in methanol (20 mL) was added to a
3
solution of porphyrin 9 (520 mg, 0.44 mmol) in CHCl3 (40 mL) in a
100 mL round-bottom flask, equipped with a stir bar. The solution was
stirred in the dark at room temperature overnight. The organic layer was
washed with water (3 ꢁ 30 mL) and dried over Na2SO4. The product
was purified by column chromatography on silica gel using a mixture
of hexanes/CH2Cl2 (5:1) as the eluent. The target product 10 was ob-
tained as a dark pink solid (550 mg, 100%). 1H NMR (CDCl3, 200 MHz):
δ 0.86 (t, 12 H, J = 6.3 Hz), 1.25 (m, 40 H), 1.88 (m, 8 H), 2.87 (t, 8 H,
J = 7.5 Hz), 7.43 (s, 2 H), 7.90 (s, 4 H), 9.17 (d, 4 H, J = 4.7 Hz), 9.44 (d,
4 H, J = 4.7 Hz), 10.30 (s, 2 H). 13C NMR (CDCl3, 50 MHz): δ 14.18,
22.73, 29.70, 30.33, 31.84, 31.96, 34.21, 36.10, 105.07, 119.76, 125.56,
Data for [5,15-Bis(4-acetylphenyl)-10,20-bis(3,5-di-n-tridecyl-
phenyl)porphyrinato(2ꢀ)-kN21,kN22,kN23,kN24]zinc(II) (14).
1H NMR (CDCl3, 200 MHz): δ 0.86 (t, 12 H, J = 6.7 Hz), 1.23 (m,
40 H), 1.87 (m, 8 H), 2.75 (s, 6 H), 2.89 (t, 8 H, J = 7.2 Hz), 7.43 (s, 2 H),
7.90 (s, 4 H), 8.24 (d, 4 H, J = 8.1 Hz), 8.34 (d, 4 H, J = 8.0 Hz) 8.91 (d,
4 H, J = 4.7 Hz), 9.05 (d, 4 H, J = 4.7). 13C NMR (CDCl3, 50 MHz):
δ 14.10, 22.67, 26.76, 29.34, 29.67, 31.76, 31.89, 36.65, 122.32, 126.49,
3193
dx.doi.org/10.1021/cm2004648 |Chem. Mater. 2011, 23, 3188–3194

Chemistry of Materials
ARTICLE
127.85, 131.45, 132.54, 134.58, 136.03, 140.90, 142.24, 148.08, 149.55,
150.53, 198.24. IR (KBr, cmꢀ1): 2923 (s), 2852 (s), 1687 (m), 1659
(m), 1598 (s), 1465 (w), 1265 (m), 1070 (w), 997 (m) 795 (w), 717
(m). UVꢀvis (CHCl3, λmax, nm): 452, 667. MALDI-TOF-MS m/z
calcd for C100H136N4O2Zn (Mþ) 1488.9955; found, 1488.93.
(11) Cheng, F. Y.; Adronov, A. Chem. Mater. 2006, 18, 5389–5391.
(12) Mickelson, E. T.; Huffman, C. B.; Rinzler, A. G.; Smalley, R. E.;
Hauge, R. H.; Margrave, J. L. Chem. Phys. Lett. 1998, 296, 188–194.
(13) Sinnott, S. B. J. Nanosci. Nanotechnol. 2002, 2, 113–123.
(14) Britz, D.; Khlobystov, A. Chem. Soc. Rev. 2006, 35, 637–659.
(15) Murakami, H.; Nomura, T.; Nakashima, N. Chem. Phys. Lett.
2003, 378, 481–485.
(16) Li, H. P.; Zhou, B.; Lin, Y.; Gu, L. R.; Wang, W.; Fernando,
K. A. S.; Kumar, S.; Allard, L. F.; Sun, Y. P. J. Am. Chem. Soc. 2004, 126,
1014–1015.
(17) Chen, J. Y.; Collier, C. P. J. Phys. Chem. B 2005, 109,
7605–7609.
(18) Hasobe, T.; Fukuzumi, S.; Kamat, P. V. J. Am. Chem. Soc. 2005,
127, 11884–11885.
(19) Guldi, D. M.; Taieb, H.; Rahman, G. M. A.; Tagmatarchis, N.;
Synthesis of {μ-[10,100-Bis(4-acetylphenyl)-5,50,15,150-
tetrakis(3,5-di-n-tridecylphenyl)-18,180:20,200-dicyclo-2,20-
210
220
230
bipo0rphyrinato(4ꢀ)-kN21,kN22,kN23,kN24:kN ,kN ,kN
,
kN24 ]} Dizinc(Ii) (3). The monofunctionalized porphyrin (50 mg,
0.036 mmol), Sc(OTf) (90 mg, 0.182 mmol), and DDQ (41 mg, 0.182
mmol) were added to dry toluene (60 mL) in a 100 mL round-bottom
flask, equipped with a stir bar, under inert Ar atmosphere. The solution
was heated to 80 °C and stirred for 3 h. After cooling to room
temperature, the mixture was passed through a short alumina column,
and washed with CH2Cl2 and then with CH2Cl2/THF (5:1). The
CH2Cl2/THF solvent was removed in vacuo. The solid was dissolved in
hexanes, and silica gel column chromatography was used to isolate the
product. Side products were eluted with hexanes/CH2Cl2 (3:1), and the
dimer was eluted using pure CH2Cl2. The desired product dimer 3 was
Prato, M. Adv. Mater. 2005, 17, 871–875.
(20) Saito, K.; Troiani, V.; Qiu, H. J.; Solladie, N.; Sakata, T.; Mori,
H.; Ohama, M.; Fukuzumi, S. J. Phys. Chem. C 2007, 111, 1194–1199.
(21) Anderson, H. L.; Martin, S. J.; Bradley, D. D. C. Angew. Chem.,
Int. Ed. 1994, 33, 655–657.
(22) Screen, T. E. O.; Lawton, K. B.; Wilson, G. S.; Dolney, N.;
Ispasoiu, R.; Goodson, T.; Martin, S. J.; Bradley, D. D. C.; Anderson,
H. L. J. Mater. Chem. 2001, 11, 312–320.
(23) Screen, T. E. O.; Thorne, J. R. G.; Denning, R. G.; Bucknall,
D. G.; Anderson, H. L. J. Mater. Chem. 2003, 13, 2796–2808.
(24) Cheng, F. Y.; Adronov, A. Chem.—Eur. J. 2006, 12, 5053–5059.
(25) Cheng, F. Y.; Adronov, A. J. Porphyrins Phthalocyanines 2007,
11, 198–204.
(26) Tsuda, A.; Furuta, H.; Osuka, A. Angew. Chem., Int. Ed. 2000,
39, 2549–2552.
(27) Tsuda, A.; Osuka, A. Science 2001, 293, 79–82.
(28) Tsuda, A.; Osuka, A. Adv. Mater. 2002, 14, 75–79.
(29) Cheng, F. Y.; Zhang, S.; Adronov, A.; Echegoyen, L.; Diederich,
F. Chem.—Eur. J. 2006, 12, 6062–6070.
obtained as a black solid (30 mg, 60%). H NMR (CDCl3/1% d5-
1
pyridine, 200 MHz): δ 0.85 (t, 24 H, J = 6.7 Hz), 1.24 (m, 80 H), 1.72
(m, 16 H), 2.71 (t, 16 H, J = 7.2 Hz), 2.74 (s, 6 H), 7.02 (s, 4 H), 7.18 (s,
4 H), 7.34 (s, 8 H), 7.47 (d, 4 H, J = 4.5 Hz), 7.52 (d, 4 H, J = 4.5 Hz) 7.84
(d, 4 H, J = 8.1 Hz), 8.11 (d, 4 H, J = 8.1 Hz). 13C NMR (CDCl3/1% d5-
pyridine, 50 MHz): δ 14.05, 22.64, 29.33, 29.47, 29.69, 31.54, 31.89,
35.91, 117.87, 125.35, 126.88, 127.30, 129.88, 130.79, 132.91, 133.03,
136.01, 147.23, 152.03, 153.21, 153.53, 154.54, 197.96. IR (KBr, cmꢀ1):
2923 (s), 2852 (s), 1687 (m), 1599 (m), 1465 (m), 1264 (m), 1142 (m),
997 (w), 721 (w). UVꢀvis (THF, λmax, nm): 422, 576. MALDI-TOF-
MS m/z calcd for C184H254N8O2Zn2 (Mþ) 2735.8603; found, 2735.78.
’ AUTHOR INFORMATION
(30) Kamo, M.; Tsuda, A.; Nakamura, Y.; Aratani, N.; Furukawa, K.;
Kato, T.; Osuka, A. Org. Lett. 2003, 5, 2079–2082.
(31) Holzinger, M.; Abraha, J.; Whelan, P.; Graupner, R.; Ley, L.;
Hennrich, F.; Kappes, M.; Hirsch, A. J. Am. Chem. Soc. 2003, 125,
8566–8580.
Corresponding Author
*Department of Chemistry, McMaster University, 1280 Main St.,
W. Hamilton, ON L8S 4M1. Tel: (905) 525-9140 x23514. Fax:
(905) 521-2773. E-mail: adronov@mcmaster.ca.
(32) Wang, J. S.; Wai, C. M.; Shimizu, K.; Cheng, F.; Boeckl, J. J.;
Maruyama, B.; Brown, G. J. Phys. Chem. C 2007, 111, 13007–13012.
(33) Chiang, I. W.; Brinson, B. E.; Huang, A. Y.; Willis, P. A.;
Bronikowski, M. J.; Margrave, J. L.; Smalley, R. E.; Hauge, R. H. J. Phys.
Chem. B 2001, 105, 8297–8301.
’ ACKNOWLEDGMENT
Financial support for this work was provided by the Natural
Science and Engineering Research Council (NSERC) of Canada,
the Canada Foundation for Innovation (CFI), and the Ontario
Innovation Trust (OIT). Wealso thank Professor Gianluigi Botton
and the Canadian Centre for Electron Microscopy (CCEM) for
assistance with TEM characterization of samples.
(34) Chiang, I. W.; Brinson, B. E.; Smalley, R. E.; Margrave, J. L.;
Hauge, R. H. J. Phys. Chem. B 2001, 105, 1157–1161.
’ REFERENCES
(1) Kadish, K., Smith, K. M., Guilard, R., Eds. The Porphyrin Hand-
book; Academic Press: New York, 1999.
(2) Drain, C. M.; Varotto, A.; Radivojevic, I. Chem. Rev. 2009, 109,
1630–1658.
(3) Hasobe, T. Phys. Chem. Chem. Phys. 2010, 12, 44–57.
(4) Boyd, P. D. W.; Reed, C. A. Acc. Chem. Res. 2005, 38, 235–242.
(5) Guldi, D. M.; Rahman, G. M. A.; Zerbetto, F.; Prato, M. Acc.
Chem. Res. 2005, 38, 871–878.
(6) Fukuzumi, S.; Kojima, T. J. Mater. Chem. 2008, 18, 1427–1439.
(7) Chitta, R.; D’Souza, F. J. Mater. Chem. 2008, 18, 1440–1471.
(8) Li, H. P.; Martin, R. B.; Harruff, B. A.; Carino, R. A.; Allard, L. F.;
Sun, Y. P. Adv. Mater. 2004, 16, 896–900.
(9) Campidelli, S.; Sooambar, C.; Diz, E. L.; Ehli, C.; Guldi, D. M.;
Prato, M. J. Am. Chem. Soc. 2006, 128, 12544–12552.
(10) Liu, Z. B.; Tian, J. G.; Guo, Z.; Ren, D. M.; Du, T.; Zheng, J. Y.;
Chen, Y. S. Adv. Mater. 2008, 20, 511–515.
3194
dx.doi.org/10.1021/cm2004648 |Chem. Mater. 2011, 23, 3188–3194