Structurally defined graphene nanoribbons with high lateral extension

Article abstract of DOI:10.1021/ja307697j

Oxidative cyclodehydrogenation of laterally extended polyphenylene precursor allowed bottom-up synthesis of structurally defined graphene nanoribbons (GNRs) with unprecedented width. The efficiency of the cyclodehydrogenation was validated by means of MALDI-TOF MS, FT-IR, Raman, and UV-vis absorption spectroscopies as well as investigation of a representative model system. The produced GNRs demonstrated broad absorption extended to near-infrared region with the optical band gap of as low as 1.12 eV.

Full text of DOI:10.1021/ja307697j

Communication
pubs.acs.org/JACS
Structurally Defined Graphene Nanoribbons with High Lateral
Extension
Matthias Georg Schwab,†,§ Akimitsu Narita,† Yenny Hernandez,†,∥ Tatyana Balandina,‡ Kunal S. Mali,‡
Steven De Feyter,‡ Xinliang Feng,† and Klaus Mullen*,†
̈
†Max Planck Institute for Polymer Research, Ackermannweg 10, D-55128 Mainz, Germany
‡Division of Molecular and Nanomaterials, Department of Chemistry, and INPAC-Institute of Nanoscale Physics and Chemistry,
Katholieke Universiteit Leuven, Celestijnenlaan, 200 F, B 3001 Leuven, Belgium
S
* Supporting Information
polyphenylene precursors. For comparison, a representative
model polycyclic aromatic hydrocarbon (PAH, C78,16,17 with
ABSTRACT: Oxidative cyclodehydrogenation of laterally
extended polyphenylene precursor allowed bottom-up
synthesis of structurally defined graphene nanoribbons
(GNRs) with unprecedented width. The efficiency of the
cyclodehydrogenation was validated by means of MALDI-
TOF MS, FT-IR, Raman, and UV−vis absorption
spectroscopies as well as investigation of a representative
model system. The produced GNRs demonstrated broad
absorption extended to near-infrared region with the
optical band gap of as low as 1.12 eV.
78 carbon atoms in the aromatic core) of GNRs is also
reported, for which a new efficient synthesis is validated by
means of matrix-assisted laser desorption/ionization time-of-
flight (MALDI-TOF) mass spectrometry (MS), 1H NMR
spectroscopy, and scanning tunneling microscope (STM).
First, para-terphenyl-based oligophenylene precursor 4a,
which corresponds to target PAH C78,16,17 was designed as a
model compound to examine its suitability for efficient
cyclodehydrogenation (Scheme 1). Synthesis of 4a was initiated
with Sonogashira-Hagihara cross-coupling of 2,2″-diiodoter-
phenyl 1a with trimethylsilyl acetylene followed by depro-
tection to yield 2,2″-diethynylterphenyl 2a. Two-fold Diels−
Alder cycloaddition between 2a and functionalized tetraphe-
nylcyclopentadienone 318 under microwave conditions effi-
raphene nanoribbons (GNRs), characterized by a large
G
aspect ratio with lateral quantum confinement, are
predicted to possess a band gap1−5 in contrast to graphene
which itself is a zero band gap semimetal.6 It has been
theoretically and experimentally demonstrated that both width
and edge structure of GNRs strongly govern the electronic
characteristics of graphene materials.1,7,8 Their appealing
electronic properties can only be accessed, however, if new
methods for reliable and reproducible build-up of well-defined
GNRs with precise lateral and longitudinal dimensions are
developed. Top-down approaches such as lithographic “cutting”
of graphene,1,2 longitudinal “unzipping”3 or “etching”4 of
carbon nanotubes, and the surfactant-assisted extraction from
graphite dispersions5 cannot reach the required chemical
precision. In our search for synthetic pathways toward GNRs,
we have worked out a bottom-up strategy that relies on the
intramolecular oxidative cyclodehydrogenation9,10 of tailored
polyphenylene precursors.11−15 Hence, a number of different
GNR geometries have been realized, ranging from fully
linear11,14,15 to kinked.12,13,15 The versatility of this method
was successfully demonstrated both for solution-11−14 and
surface-based15 protocols, resulting in GNRs of atomic
accuracy. This stands in sharp contrast to the aforementioned
top-down methods.1−5
a
Scheme 1. Synthetic Route to PAH C78 and GNR 6
a
Geometric dimensions of GNR 6 were derived from Merck
Molecular Force Field 94 (MMFF94) calculations. Reagents and
conditions: (a) trimethylsilyl acetylene, Pd(PPh3)2Cl2/CuI, NEt3, rt;
(b) K2CO3, THF/MeOH, rt, 2a: 51% (two steps), 2b: 58% (two
steps); (c) xylene, 160 °C, μW, 300 W, 4a: 81%, 4b: 85%; (d) FeCl3,
CH2Cl2/CH3NO2, rt, 93%; (e) MoCl5, CH2Cl2, rt, 85%; (f) PIFA/
BF3, CH2Cl2, −60 °C, then −10 °C, 81%; (g) bis(cycloocta-(1,5)-
diene)nickel(0) cycloocta-(1,5)-diene, 2,2′-bipyridine, toluene/DMF,
80 °C, 81%; (h) FeCl3, CH2Cl2/CH3NO2, rt, 92%.
GNRs so far fabricated by bottom-up approaches are limited
to those with lateral dimensions smaller than 1 nm, which show
absorption only up to 670 nm and calculated band gap larger
than 1.6 eV.11−15 For future applications in electronic devices, it
is essential to synthesize GNRs with tailored band gaps. In this
work, we present a solution synthesis and characterization of
unprecedentedly broad, low band gap, and structurally defined
GNRs, which can be derived from laterally extended
Received: August 3, 2012
Published: October 19, 2012
© 2012 American Chemical Society
18169
dx.doi.org/10.1021/ja307697j | J. Am. Chem. Soc. 2012, 134, 18169−18172

Journal of the American Chemical Society
Communication
ciently produced 4a in 81% yield. Subsequent experiments
confirmed that 4a can be readily transformed into C78 by using
either the established FeCl3 mediated cyclodehydrogena-
tion9,10,16,17 or less common oxidative procedures involving
MoCl510,19 or PIFA/BF3.10,18 In all three cases, characterization
of the products by a combination of MALDI-TOF MS and H
NMR spectroscopy proved the formation of C78 as an
chains. The aromatic core showed three parallel stripes running
along the long axis, which correlated with electron densities in
the highest occupied molecular orbitals.17 The dimensions of
the core extracted from the image (length = 2.3 0.1 nm and
width 1.4 0.1 nm) were in good agreement with the values
obtained by molecular modeling (Figure 2b). The remaining
interspace between four C78 molecules was filled with TCB
guest molecules that appeared as small bright spots. These
observations clearly support the defect-free synthesis of C78
(see Supporting Information for more details). It should be
mentioned here that, despite the successful synthesis of C78 in
previous reports,16,17 the present protocol shows the obvious
advantage regarding the cyclodehydrogenation efficiency of
precursors that can avoid the rearrangement under the
experimental conditions. More importantly, precursor 4 can
be considered as a repeating unit of polyphenylene precursor 5
which renders one to produce laterally extended GNR 6 with a
width of 1.54−1.98 nm (Scheme 1).
1
exclusive product (Figures 1 and S1). No partially fused
So far, all chemical routes toward the synthesis of
polyphenylene precursors made use of A2B2-type polyconden-
sation reactions such as Suzuki-Miyaura cross-coupling11,12 and
Diels−Alder reaction.13,14 Because of the intrinsic sensitivity of
these protocols to stoichiometry,21 an AA-type Yamamoto
polycondensation system which can circumvent this drawback
is believed to be more efficient and can yield high molecular-
weight precursors. Therefore, monomer 4b substituted with
reactive halogen groups was synthesized, allowing an AA-type
Yamamoto polymerization to yield kinked polyphenylene
precursor 5 (Scheme 1).21 MALDI-TOF MS characterization
of 5 indicated the presence of a regular pattern with molecular
weight up to 35 000−40 000 g mol−1 (Figure 3a). On this basis,
Figure 1. (a) MALDI-TOF MS spectrum of C78; inset: the isotopic
distribution is in agreement with the simulated result. (b) H NMR
spectrum of C78 in 1,2-dichlorobenzene-d4 at 170 °C.
1
intermediates or chlorinated products could be detected by
MALDI-TOF MS, and comparison of the isotropic distribution
with simulation proved the complete cyclodehydrogenation
(Figure 1a). Generally, PAHs suffer from strong aggregation in
1
solution, which results in broadening of peaks in H NMR
spectroscopy.20 Remarkably, however, a well resolved 1H NMR
spectrum could be recorded for C78 at 170 °C by using high-
boiling-point solvent, 1,2-dichlorobenzene-d4 (Figure 1b).
Because of its extended π-system, the signals of the protons
on the aromatic backbone of C78 appeared strongly downfield
shifted in a range between δ = 8.0 and 9.5 ppm. With the help
of peak integration and Nuclear Overhauser enhancement
spectroscopy (NOESY) measurements, a number of aliphatic
and aromatic signals could be assigned as indicated in Figure
1b, which further confirmed the chemical identity of the
product. To our best knowledge, this is the largest PAH that
1
Figure 3. (a) Linear-mode MALDI-TOF MS spectra of polyphenylene
precursor 5 (before separation). (b) Molecular weight distribution of
5-I and 5-II (SEC, eluent: THF, PS standard).
allows for structural characterization by H NMR.
Furthermore, STM rendered the visualization of C78 at the
solid−liquid interface of Au(111)/1,2,4-trichlorobenzene
(TCB) (Figure 2a). The alkyl chains were surface-crystallized
and split into a set of four lateral and two kinked longitudinal
the number of repeating units in 5 was 21−24, which
corresponded to approximately 30 nm of the resultant GNRs.
This is most likely not the maximal value, however, due to the
limitation of MALDI-TOF MS for the analysis of high-
molecular-weight species with a broad molecular weight
distribution.12,22 Further investigation of 5 by size exclusion
chromatography (SEC) revealed its relatively high polydisper-
sity index (PDI) of 2.2. Therefore, the crude polymer 5 was
separated into two fractions by utilizing preparative SEC. The
fraction of larger molecular weight 5-I showed weight-average
molecular weight (Mw) of 52 000 g mol−1 with PDI of 1.2,
whereas the other fraction of smaller molecular weight 5-II
showed Mw of 7200 g mol−1 with PDI of 1.1 (Figure 3b). These
results based on SEC are approximately relative values
according to the polystyrene (PS) standard calibration,23 but
they highlight the superiority of the AA-type Yamamoto
approach over the A2B2-type polymerization for the preparation
of high molecular weight precursors.11−13
Figure 2. (a) STM image of C78 physisorbed at the Au(111)/TCB
interface; blue circles indicate entrapped guest molecules; imaging
conditions: It = 0.1 nA, Vbias = 464 mV. (b) A tentative molecular
model for the self-assembly of C78 on Au (111).
18170
dx.doi.org/10.1021/ja307697j | J. Am. Chem. Soc. 2012, 134, 18169−18172

Journal of the American Chemical Society
Communication
Next, intramolecular cyclodehydrogenation of precursors 5,
5-I, and 5-II was performed using FeCl3 as oxidant in a mixture
of dichloromethane and nitromethane, yielding GNRs 6, 6-I,
and 6-II, respectively. MALDI-TOF MS analysis of GNR 6-II
suggested a similar number of repeating units with respect to
precursor 5-II (Figure S11). However, GNRs 6 and 6-I were
precluded from MALDI-TOF MS characterization possibly
owing to the strong aggregation of high molecular weight
GNRs.19,22
Fourier transform infrared (FTIR) spectroscopy analysis of
precursor 5 and GNR 6 showed the disappearance of the band
at 4050 cm−1 which originates from the rotation of free phenyl
rings (Figure 4).13,24 Further, the signal triad at 3025, 3056, and
Figure 5. (a) Raman spectrum of GNR 6-I measured at 488 nm
(powder, laser power: 1 mW). (b) Normalized UV−vis absorption
spectra of GNR 6-I (red, in NMP), GNR 6-II (orange, in NMP), and
C78 (blue, in THF); inset: a picture of solutions of GNR 6-I and 6-II
in NMP.
Thus, UV−vis absorption spectra of GNR 6-I and 6-II were
recorded in NMP and compared with that of C78 (17 μM in
THF) as shown in Figure 5b. For C78, the β-band was
observed at 431 nm and the p- and α-band were located at 514
and 580 nm, respectively. The relatively high intensity of the α-
band could be attributed to the low symmetry of C78.18,29 The
optical band gap of C78 could not be determined from the
onset of the p-band due to the broadened absorption bands,29
but was estimated to be larger than 1.92 eV from the absorption
edge of 647 nm. It is remarkable to note that the longest GNR
6-I showed absorption peaks at 690 and 960 nm with broad
absorption extended to the near-infrared (NIR) region. The
absorption edges of GNR 6-I and 6-II were derived from the
spectra to be 1109 and 812 nm, corresponding to the optical
band gap of 1.12 and 1.53 eV, respectively. Thereby, this result
undoubtedly demonstrated the extraordinarily low band gap of
the laterally extended GNRs compared to previously bottom-up
synthesized GNRs.11−15 Furthermore, the optical band gap of
1.12 eV is in a good agreement with the calculated band gap of
1.08 eV, proving that GNR 6-I is sufficiently elongated, namely
more than 15 repeating units, or approximately 20 nm, to
obtain the lowest band gap achievable with its lateral
structure.30
In summary, structurally defined and laterally extended
GNRs were synthesized via a bottom-up chemical approach.
Characterizations by MALDI-TOF MS, FTIR, Raman, and
UV−vis absorption spectroscopies as well as investigation of a
model system validated the efficiency of the cyclodehydroge-
nation. For the first time, it was possible to access GNRs with
band gap as low as 1.12 eV, reaching a broad absorption up to
the NIR region. These GNRs may hold the potential in a
number of opoelectronic devices such as solar cell, optical
switching, and infrared imaging.
Figure 4. Representative FTIR spectral regions before (4a, 5) and
after (C78, GNR 6) oxidative cyclodehydrogenation.
3083 cm−1 which is typical for aromatic C−H stretching
vibrations was diminished, and the fingerprint bands at 698 and
755 cm−1 originating from monosubstituted benzene rings were
attenuated.24,25 These observations were in line with the
analogous conversion of model compound 4a into C78, and
indicated highly efficient “graphitization” of precursor 5 into
GNR 6. Raman spectrum of a powder sample of GNR 6-I was
measured at 488 nm with laser power of 1 mW, and showed
first-order D band (disorder band) and G band (graphite band)
at 1316 and 1595 cm−1, respectively, consistent with literature
values for GNRs (Figure 5a).3,4,15,26 The relatively high
intensity of the D band could be explained by the contribution
of the edges as defects.4,12,15,26 Moreover, well-resolved double-
resonant signals were also observed at 2632, 2910, and 3194
cm−1, which could be assigned to 2D, D+G, and 2G bands,
respectively.26 It should be mentioned that the G* band
(∼2450 cm−1) typical for graphene was apparently too small to
be observed, because a powder (multilayer) sample was used
for the measurement in this work.26d
ASSOCIATED CONTENT
* Supporting Information
Experimental details, NMR, MALDI-TOF MS, UV−vis
absorption, fluorescence, and DSC spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
The GNR samples were virtually insoluble in conventional
organic solvents such as dichloromethane, chloroform, and
toluene. However, use of N-methylpyrrolidone (NMP), which
is known as a good solvent for the exfoliation of graphene27 as
well as for the debundling of carbon nanotubes,28 enabled the
exfoliation of GNRs with the assistance of mild sonication to
obtain stable dispersions (∼60 μg/mL) (Figure 5b, inset).
AUTHOR INFORMATION
Corresponding Author
muellen@mpip-mainz.mpg.de
■
Present Addresses
§BASF SE, Carl-Bosch-Strasse 38, 67056 Ludwigshafen,
Germany
18171
dx.doi.org/10.1021/ja307697j | J. Am. Chem. Soc. 2012, 134, 18169−18172

Journal of the American Chemical Society
Communication
∥Physics Department, Universidad de los Andes, Carrera 1 18A-
(21) (a) Carothers, W. H. Trans. Faraday Soc. 1936, 32, 39.
(b) Odian, G. G. Principles of Polymerization; J. Wiley & Sons:
Hoboken, NJ, 2004.
́
10, Bloque Ip. Bogota, Colombia
Notes
(22) Martin, K.; Spickermann, J.; Rader, H. J.; Mullen, K. Rapid
̈
̈
The authors declare no competing financial interest.
Commun. Mass Spectrom. 1996, 10, 1471.
(23) The PS standard was used for the SEC analysis instead of the
poly(para-phenylene) standard, considering the kinked and nonrigid
structure of polyphenylene precursor 5.
ACKNOWLEDGMENTS
■
(24) Centrone, A.; Brambilla, L.; Renouard, T.; Gherghel, L.; Mathis,
We are grateful to the financial support from ERC grant on
NANOGRAPH, EU Project SUPERIOR (PITN-GA-2009-
238177) and GENIUS, DFG Priority Program SPP 1459,
ESF Project GOSPEL (Ref Nr: 09-EuroGRAPHENE-FP-001),
the Max Plank Society through the program ENERCHEM,
DFG Priority Program SPP 1355, DFG MU 334/32-1.
C.; Mullen, K.; Zerbi, G. Carbon 2005, 43, 1593.
̈
(25) Shifrina, Z. B.; Averina, M. S.; Rusanov, A. L.; Wagner, M.;
Mullen, K. Macromolecules 2000, 33, 3525.
̈
(26) (a) Malard, L. M.; Pimenta, M. A.; Dresselhaus, G.; Dresselhaus,
M. S. Phys. Rep 2009, 473, 51. (b) Ryu, S.; Maultzsch, J.; Han, M. Y.;
Kim, P.; Brus, L. E. ACS Nano 2011, 5, 4123. (c) Bischoff, D.;
Guttinger; Droscher, S.; Ihn, T.; Ensslin, K.; Stampfer, C. J. Appl. Phys.
̈
̈
2011, 109, 073710. (d) Yoon, D.; Moon, H.; Cheong, H.; Choi, J. S.;
Choi, J. A.; Park, B. H. J. Korean Phys. Soc. 2009, 55, 1299.
(27) Hernandez, Y.; Nicolosi, V.; Lotya, M.; Blighe, F. M.; Sun, Z.;
De, S.; McGovern, I. T.; Holland, B.; Byrne, M.; Gun’Ko, Y. K.;
Boland, J. J.; Niraj, P.; Duesberg, G.; Krishnamurthy, S.; Goodhue, R.;
Hutchison, J.; Scardaci, V.; Ferrari, A. C.; Coleman, J. N. Nat.
Nanotechnol 2008, 3, 563.
(28) Bergin, S. D.; Nicolosi, V.; Streich, P. V.; Giordani, S.; Sun, Z.;
Windle, A. H.; Ryan, P.; Niraj, N. P. P.; Wang, Z.-T. T.; Carpenter, L.;
Blau, W. J.; Boland, J. J.; Hamilton, J. P.; Coleman, J. N. Adv. Mater.
2008, 20, 1876.
REFERENCES
■
̈
(1) Han, M. Y.; Ozyilmaz, B.; Zhang, Y.; Kim, P. Phys. Rev. Lett. 2007,
98, 206805.
̈
(2) (a) Ozyilmaz, B.; Jarillo-Herrero, P.; Efetov, D.; Kim, P. Appl.
Phys. Lett. 2007, 91, 192107. (b) Jia, X.; Hofmann, M.; Meunier, V.;
Sumpter, B. G.; Campos-Delgado, J.; Romo-Herrera, J. M.; Son, H.;
Hsieh, Y. P.; Reina, A.; Kong, J. Science 2009, 323, 1701.
(3) (a) Kosynkin, D. V.; Higginbotham, A. L.; Sinitskii, A.; Lomeda,
J. R.; Dimiev, A.; Price, B. K.; Tour, J. M. Nature 2009, 458, 872.
(b) Shimizu, T.; Haruyama, J.; Marcano, D. C.; Kosinkin, D. V.; Tour,
J. M.; Hirose, K.; Suenaga, K. Nat. Nanotechnol. 2011, 6, 45.
(4) Jiao, L.; Zhang, L.; Wang, X.; Diankov, G.; Dai, H. Nature 2009,
458, 877.
(29) Rieger, R.; Mullen, K. J. Phys. Org. Chem. 2010, 23, 315.
̈
(30) Osella, S.; Narita, A.; Schwab, M. G.; Hernandez, Y.; Feng, X.;
Mullen, K.; Beljonne, D. ACS Nano 2012, 6, 5539.
̈
(5) Li, X.; Wang, X.; Zhang, L.; Lee, S.; Dai, H. Science 2008, 319,
1229.
(6) (a) Wallace, P. R. Phys. Rev. Lett. 1947, 71, 622. (b) Castro Neto,
A. H.; Guinea, F.; Peres, N. M. R.; Novoselov, K. S.; Geim, A. K. Rev.
Mod. Phys. 2009, 81, 109.
(7) (a) Schwierz, F. Nat. Nanotechnol. 2010, 5, 487. (b) Chen, Z.;
Lin, Y. M.; Rooks, M. J.; Avouris, P. Physica E 2007, 40, 228.
(8) (a) Barone, V.; Hod, O.; Scuseria, G. E. Nano Lett. 2006, 6, 2748.
(b) Yang, L.; Park, C.-H.; Son, Y.-W.; Cohen, M. L.; Louie, S. G. Phys.
Rev. Lett. 2007, 99, 186801. (c) Radovic, L. R.; Bockrath, B. J. Am.
Chem. Soc. 2005, 127, 5917.
(9) Scholl, R.; Seer, C. Liebigs Ann. Chem. 1912, 394, 111.
(10) Rempala, P.; Kroulík, J.; King, B. T. J. Am. Chem. Soc. 2004, 126,
15002.
(11) Yang, X.; Dou, X.; Rouhanipour, A.; Zhi, L.; Rader, H. J.;
̈
Mullen, K. J. Am. Chem. Soc. 2008, 130, 4216.
̈
(12) Dossel, L.; Gherghel, L.; Feng, X.; Mullen, K. Angew. Chem., Int.
̈
̈
Ed. 2011, 50, 2540.
(13) Wu, J.; Gherghel, L.; Watson, M. D.; Li, J.; Wang, Z.; Simpson,
C. D.; Kolb, U.; Mullen, K. Macromolecules 2003, 36, 7082.
̈
(14) Fogel, Y.; Zhi, L.; Rouhanipour, A.; Andrienko, D.; Rader, H. J.;
̈
̈
Mullen, K. Macromolecules 2009, 42, 6878.
(15) Cai, J.; Ruffieux, P.; Jaafar, R.; Bieri, M.; Braun, T.; Blankenburg,
S.; Muoth, M.; Seitsonen, A. P.; Saleh, M.; Feng, X.; Mullen, K.; Fasel,
̈
R. Nature 2010, 466, 470.
(16) (a) Muller, M.; Iyer, V. S.; Kubel, C.; Enkelmann, V.; Mullen, K.
̈
̈
̈
Angew. Chem., Int. Ed. 1997, 36, 1607. (b) Morgenroth, F.; Kubel, C.;
̈
Muller, M.; Wiesler, U. M.; Berresheim, A. J.; Wagner, M.; Mullen, K.
Carbon 1998, 36, 833.
̈
̈
(17) Bohme, T.; Simpson, C.; Mullen, K.; Rabe, J. Chem.Eur. J.
̈
̈
2007, 13, 7349.
(18) Takada, T.; Arisawa, M.; Gyoten, M.; Hamada, R.; Tohma, H.;
Kita, Y. J. Org. Chem. 1998, 63, 7698.
(19) Kramer, B.; Frohlich, R.; Waldvogel, S. Eur. J. Org. Chem. 2003,
̈
3549.
(20) (a) Wasserfallen, D.; Kastler, M.; Pisula, W.; Hofer, W. A.;
Fogel, Y.; Wang, Z.; Mullen, K. J. Am. Chem. Soc. 2006, 128, 1334.
̈
(b) Kastler, M.; Pisula, W.; Wasserfallen, D.; Pakula, T.; Mullen, K. J.
̈
Am. Chem. Soc. 2005, 127, 4286.
18172
dx.doi.org/10.1021/ja307697j | J. Am. Chem. Soc. 2012, 134, 18169−18172