Friction Anisotropy−Driven Domain Imaging on Exfoliated Monolayer

Graphene
Jin Sik Choi et al.
Science 333, 607 (2011);
DOI: 10.1126/science.1207110

This copy is for your personal, non-commercial use only.

If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.

Downloaded from www.sciencemag.org on January 31, 2014

Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.

The following resources related to this article are available online at

www.sciencemag.org (this information is current as of January 31, 2014 ):

Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/333/6042/607.full.html
Supporting Online Material can be found at:
http://www.sciencemag.org/content/suppl/2011/06/30/science.1207110.DC1.html
A list of selected additional articles on the Science Web sites related to this article can be
found at:
http://www.sciencemag.org/content/333/6042/607.full.html#related
This article cites 22 articles, 5 of which can be accessed free:
http://www.sciencemag.org/content/333/6042/607.full.html#ref-list-1
This article has been cited by 1 articles hosted by HighWire Press; see:
http://www.sciencemag.org/content/333/6042/607.full.html#related-urls
This article appears in the following subject collections:
Materials Science
http://www.sciencemag.org/cgi/collection/mat_sci

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2011 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.
Suswapna Patnaik,9 Rosa Quan,6 Gopalakrishna Ramaswamy,9#
Paul Shinn,6 Geetha M. Swamilingiah,9 Stacy Wu,6 Joseph R.
Ecker6,7† (chair).
Interactome data acquisition group: Matija Dreze1,2,5 (project
leader), Danielle Byrdsong,1,2 Amélie Dricot,1,2 Melissa Duarte,1,2
Fana Gebreab,1,2 Bryan J. Gutierrez,1,2 Andrew MacWilliams,1,2
Dario Monachello,12** M. Shahid Mukhtar,11†† Matthew M.
Poulin,1,2 Patrick Reichert,1,2 Viviana Romero,1,2 Stanley Tam,1,2
Selma Waaijers,1,2‡‡ Evan M. Weiner,1,2 Marc Vidal1,2† (co-
chair), David E. Hill1,2† (co-chair), Pascal Braun1,2† (chair).
wNAPPA interactome validation group: Mary Galli6 (project
leader), Anne-Ruxandra Carvunis,1,2,3 Michael E. Cusick,1,2 Matija
Dreze,1,2,5 Viviana Romero,1,2 Frederick P. Roth,1,8‡ Murat
Tasan,8 Junshi Yazaki,7 Pascal Braun1,2† (co-chair), Joseph R.
Ecker6,7† (chair).
Bioinformatics and analysis group: Anne-Ruxandra
Carvunis1,2,3 (project leader), Yong-Yeol Ahn,1,10 Albert-László
Barabási,1,10 Benoit Charloteaux,1,2,4 Huaming Chen,6 Michael E.
Cusick,1,2 Jeffery L. Dangl,11 Matija Dreze,1,2,5 Joseph R. Ecker,6,7†
Changyu Fan,1,2 Lantian Gai,6 Mary Galli,6 Gourab Ghoshal,1,10
Tong Hao,1,2 David E. Hill,1,2† Claire Lurin,12 Tijana Milenkovic,13
Jonathan Moore,14 M. Shahid Mukhtar,11†† Samuel J.
Pevzner,1,2,15,16 Natasa Przulj,17 Sabrina Rabello,1,10 Edward A.
Rietman,1,2§§ Thomas Rolland,1,2 Frederick P. Roth,1,8‡ Balaji
Santhanam,1,2 Robert J. Schmitz,7 William Spooner,18,19 Joshua
Stein,18 Murat Tasan,8 Jean Vandenhaute,5 Doreen Ware,18,20
Pascal Braun1,2† (co-chair), Marc Vidal1,2† (chair).
Writing group: Pascal Braun1,2† (chair), Anne-Ruxandra
Carvunis,1,2,3 Benoit Charloteaux,1,2,4 Matija Dreze,1,2,5 Mary
Galli,6 Marc Vidal1,2† (co-chair).
1 References
Center for Cancer Systems Biology (CCSB) and Department
of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA
02215, USA. 2Department of Genetics, Harvard Medical School,
Boston, MA 02115, USA. 3Computational and Mathematical
Biology Group, Techniques de l’Ingénierie Médicale et de
la Complexité–Informatique, Mathématiques et Applica-
tions de Grenoble, CNRS UMR5525 and Université de
Grenoble, Faculté de Médecine, 38706 La Tronche Cedex,
France. 4Unit of Animal Genomics, GIGA-R and Faculty of
Veterinary Medicine, University of Liège, 4000 Liège, Wallonia-
Brussels Federation, Belgium. 5Unité de Recherche en Biologie
Moléculaire, Facultés Universitaires Notre-Dame de la Paix,
5000 Namur, Wallonia-Brussels Federation, Belgium. 6Genomic
Analysis Laboratory, Salk Institute for Biological Studies, La
Jolla, CA 92037, USA. 7Plant Biology Laboratory, Salk Insti-
tute for Biological Studies, La Jolla, CA 92037, USA. 8Depart-
ment of Biological Chemistry and Molecular Pharmacology, 9 #Present address: Thermo Fisher Scientific, BioSciences Di-
Harvard Medical School, Boston, MA 02115, USA. Life
10
Technologies, Carlsbad, CA 92008, USA. Center for
Complex Network Research (CCNR), Department of Physics,
Northeastern University, Boston, MA 02115, USA. 11Depart-
ment of Biology, University of North 12
Carolina at Chapel Hill,
Chapel Hill, NC 27599, USA. Unité de Recherche en
Génomique Végétale, Unités Mixtes de Recherche Institut
Nationale de la Recherche Agronomique/Université Evry-Val
d’Essonne ERL Centre National de la Recherche Scientifique
91057, Evry Cedex, France. 13Department of Computer Science
and Engineering, University of Notre Dame, Notre Dame, IN
46556, USA. 14Warwick Systems Biology Centre, Coventry
House, University of Warwick, Coventry, CV4 7AL, UK.
15
Biomedical Engineering Department, Boston University,
Boston, MA 02215, USA. 16Boston University School of
Medicine, Boston, MA 02118, USA. 17Department of Comput-
ing, Imperial College London, London SW7 2AZ, UK. 18Cold
Spring Harbor Laboratory, Cold Spring Harbor, NY 11724, USA.
19
Eagle Genomics Ltd., Babraham Research Campus, Cam-
bridge, CB4 1JD, UK. 20United States Department of Ag-
riculture, Agricultural Research Service, Robert W. Holley
Center for Agriculture and Health, Cornell University, Ithaca,
NY 14853, USA.
‡Present address: Donnelly Centre for Cellular and Biomolec-
ular Research, University of Toronto, Toronto, Ontario M5S3E1,
Canada and Samuel Lunenfeld Research Institute, Mt. Sinai
Hospital, Toronto, Ontario M5G1X5, Canada.
§Present address: Foley & Lardner LLP, 3579 Valley Centre
Drive, Suite 300, San Diego, CA 92130, USA.
||Deceased.
¶Present address: Pacific Biosciences, 940 Hamilton Drive,
Menlo Park, CA 94025, USA.
vision, Bangalore-560011, India.
**Present address: Centre de Génétique Moléculaire du C.N.R.S.,
1 Avenue de la Terrasse, 91190 Gif-sur-Yvette, France.
††Present address: Department of Biology, University of Ala-
bama at Birmingham, Birmingham, AL 35294, USA.
‡‡Present address: University of Utrecht, 3508 TC Utrecht,
The Netherlands.
§§Present address: Center of Cancer Systems Biology,
St. Elizabeth's Medical Center, Tufts University School of
Medicine, Boston, MA 02135, USA.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6042/601/DC1
SOM Text
Figs. S1 to S47
Tables S1 to S12
7 February 2011; accepted 10 June 2011
10.1126/science.1203877

REPORTS
Atomic force microscopy (AFM) can be used
Friction Anisotropy–Driven to study the mechanical properties of surfaces be-
cause it provides local information about hard-

Domain Imaging on Exfoliated ness, deformation, slipperiness, and chemical

state. Friction force microscopy (FFM) has been
used to investigate elastic deformation, atomic
Monolayer Graphene structure, dislocation, and defects (8, 9). Recent
studies show that friction depends on the num-
Jin Sik Choi,1 Jin-Soo Kim,1 Ik-Su Byun,1 Duk Hyun Lee,1 Mi Jung Lee,1 Bae Ho Park,1* ber of graphene layers as well as the nature of the
Changgu Lee,2 Duhee Yoon,3 Hyeonsik Cheong,3 Ki Ho Lee,4 Young-Woo Son,4 graphene-substrate bond (10–12). Puckering in-
Jeong Young Park,5* Miquel Salmeron6 duced by AFM tip scanning has been proposed
as the origin of the thickness effects on friction
Graphene produced by exfoliation has not been able to provide an ideal graphene with (10). Here, we show the existence of domains on
performance comparable to that predicted by theory, and structural and/or electronic defects exfoliated monolayers of graphene deposited on
have been proposed as one cause of reduced performance. We report the observation of domains silicon oxide that are distinguished by their dif-
on exfoliated monolayer graphene that differ by their friction characteristics, as measured by ferent friction characteristics when an AFM tip
friction force microscopy. Angle-dependent scanning revealed friction anisotropy with a periodicity slides over them. These domains cannot be ob-
of 180° on each friction domain. The friction anisotropy decreased as the applied load served in AFM topographic images, optical mi-
increased. We propose that the domains arise from ripple distortions that give rise to anisotropic croscopy, or micro-Raman spectroscopy.
friction in each domain as a result of the anisotropic puckering of the graphene.
1
he mechanical exfoliation method to trans- kles or ripples (3–5), and microscopic corruga-

T
fer a monolayer of graphene to a substrate
is thought to be a facile method to obtain
a single crystalline graphene (1). Mechanical
exfoliation, however, may induce strain on the
graphene layer during deposition on a substrate
and can create wrinkled films and other defects,
because the interaction with the substrate might
introduce uneven compressive and tensile stresses
that are nonuniformly distributed across the film.
Structural defects such as atomic defects (2), wrin-
tion (6) have already been reported on supported
graphene. These defects tend to lower the elec-
trical performance of graphene devices because
they break translational or rotational symmetry.
In addition, the boundaries of microscale domains
also break the symmetry, as reported for graphene
grown by chemical vapor deposition (7). How-
ever, no experimental observations of microscale
domains on mechanically exfoliated monolayer
graphene have been reported to date.
Division of Quantum Phases and Devices, Department of
Physics, Konkuk University, Seoul 143-701, Korea. 2Depart-
ment of Mechanical Engineering, Sungkyunkwan University,
Suwon 440-746, Korea. 3Department of Physics, Sogang Uni-
versity, Seoul 121-742, Korea. 4Korea Institute for Advanced
Study, Seoul 130-722, Korea. 5Graduate School of Energy, En-
vironment, Water, and Sustainability, NanoCentury KI, Korea
Advanced Institute of Science and Technology, Daejeon 305-701,
Korea. 6Materials Science Division, Lawrence Berkeley National
Laboratory, Berkeley, CA 94720, USA.
*To whom correspondence should be addressed. E-mail:
baehpark@konkuk.ac.kr (B.H.P.); jeongypark@kaist.ac.kr
( J.Y.P.)

www.sciencemag.org SCIENCE VOL 333 29 JULY 2011 607

REPORTS
The black dashed lines in the AFM topog-
raphy and optical microscopy images in Fig. 1, A
and B, delineate a single graphene layer. To con-
firm the number of graphene layers, we performed
micro-Raman spectroscopy at the six points indi-
cated in Fig. 1B (13). The 2D band (~2686 cm−1)
has a single Lorentzian 2D line shape, and there
is no D peak, the presence of which is associ-
ated with disordered carbon atoms or defects
(Fig. 1C). Moreover, there are no appreciable
variations in the Raman spectra, which indicates
that the graphene layer does not have charge dop-
ing variations or high-strain domains. The expanded
AFM topographic image in Fig. 1D, correspond-
ing to the red dashed square in Fig. 1A, shows the
topographic uniformity of the graphene layer.
Although the topographic image is completely
uniform, a striking result was obtained in simul-
taneously acquired friction force images, which
reveal the existence of several domain structures
in the same layer with different friction proper-
ties (Fig. 1E). Figure 1F shows the FFM line pro-
files in forward and backward scans along the
black dashed line shown in Fig. 1E. The difference
between the friction signals in the two scan direc-
tions is proportional to the friction force. Such
frictional domains in single layers were frequently
observed in other graphene layers that we mea-
sured, both in air and in high-vacuum environ-
ments (~10−4 torr).

A B C

Raman intensity (arb. units)

G band 2D
2D band
Monolayer
graphene
1550 1600

2650 2700 2750

G
SiO2
G*

5 µm 1200 1600 2000 2400 2800

Raman shift (cm-1)
D E F

Friction signal (arb. units)

Forward scan
SiO2

SiO2 I III II

Monolayer Backward scan

graphene
0 1 2 3 4 5
1 µm
Distance (µm)
Fig. 1. Frictional domains on a monolayer graphene flake. (A) AFM to- friction force images, respectively, of the red dashed square in (A).
pography, (B) optical microscopy image, and (C) Raman spectra at the The red arrow indicates the forward scanning direction. (F) Friction
six points marked in (B). The two inset graphs show detailed shapes of force traces from forward and backward scans along the black dashed
G-band and 2D-band peaks. (D and E) Expanded AFM topographic and line in (E).

A B 0.4
0° 16° 27° 46° F2
I
0.3
Friction (a.u.)

II
III
1 µm
0.2

60° 73° 85° 94°

0.1 F1
experiments : I II III
fittings : I II III
0.0
0 60 120 180
Rotation Angle (θ)
126° 141°
Fig. 2. (A) Friction force images showing the changing friction
contrast as the sample is rotated counterclockwise from 0° to
184° relative to the horizontal scan direction (red dashed arrow).
Roman numerals indicate the three friction domains identified in
165° 184° Fig. 1E. (B) Normalized friction force versus rotation angle for each
domain, showing 180° periodicity. The lines show that the varia-
tions in friction can be fitted by a simple sine modulus function.

608 29 JULY 2011 VOL 333 SCIENCE www.sciencemag.org


Another surprising result is that the friction
force in each domain showed a peculiar 180°
periodic dependence on the angle between the
scan direction and sample orientation (Fig. 2).
The images in Fig. 2A were obtained by incre-
mentally rotating the sample while keeping the
scan direction along the horizontal axis (red dashed
arrow) fixed. The brighter areas correspond to
greater friction. During these measurements, the
loading force (0 nN), scan area (5 mm by 5 mm),
and scanning speed (0.5 Hz) were kept fixed. The
different friction measured in each domain is not
caused by any interaction with the silicon ox-
ide substrate, which has been shown to be two-
dimensionally isotropic (14). Nor can it be due
to different local chemical properties, because the
tip-surface adhesion was found to be the same in
each domain within the error bar. The values for
A 4
0.8
work of adhesion measured on domains I, II, and
III were 7.3 T 0.2, 7.4 T 0.2, and 7.6 T 2.3 mJ/m2,
respectively, as derived from the Derjaguin,
Muller, and Toporov (DMT) model with a tip
radius of 100 nm. These are all smaller than the
work of adhesion measured on the bare silicon
oxide surface (9.2 T 0.8 mJ/m2).
Figure 2B shows normalized friction data
versus sample rotation angle (q) for each do-
main. Normalization was performed by subtract-
ing the forward and backward friction signals
on each domain, followed by dividing this val-
ue by the corresponding value on the surround-
ing SiO2 area, which is independent of the
scanning direction.
The 180° angular periodicity of the friction
cannot be the result of the hexagonal periodicity
of graphene, which would give rise to a period of
B 2.0
Friction ( I) / friction (III)
REPORTS
60°. Instead, the observed periodic friction in each
domain can be fit with a simple sinusoidal curve
of the type Fi (q) = F1 + (F2 − F1) × |sin(q – qi)|,
where the phase term qi for each domain is de-
termined from fitting the experimental data in
Fig. 2B. F1 and F2 are the friction force minimum
and maximum values, respectively. From this fit,
we obtain a friction anisotropy ratio F2/F1 of 2.15 T
0.08, which is the same for each domain, indicat-
ing a common origin of the friction anisotropy.
We also investigated the dependence of fric-
tion on applied load. As shown in Fig. 3A, the
friction force on the SiO2 substrate shows a good
fit to the DMT model, which is applicable to
hard bulk materials (15). The friction value at a
loading force of –2 nN confirms that the tip-
surface adhesion force on SiO2 and monolayer
graphene is between –1 nN and –2 nN. Figure
3B shows the effect of applied load on the ratio
of the friction in I (maximum) and III (minimum)
domains for a fixed scan direction. The ratio de-
creases gradually as the load increases until it fi-

SiO 2
nally converges to 1 when the applied load reaches
Friction (a.u.)

0.4 I DMT
II fitting
5 nN or higher. Because domain II is oriented in a
0.0
III
1.5 diagonal direction with respect to the ripple line
2 -2 0 2 4 6
and exhibits an intermediate friction value, the ef-
fect of applied load on the ratio of the friction in
I
II
the I (maximum) and II (medium) domains is not
III 1.0
as prominent.
0 Friction anisotropy has been previously ob-
-2 0 2 4 6 -2 0 2 4 6 served on various systems and rationalized in terms
Applied load (nN) Applied load (nN) of lattice commensurability arguments (16), dif-
Fig. 3. (A) Friction force dependence on applied load for each domain and for the bare SiO2 substrate. ferences in phonon excitation probability along
(B) Effect of applied load on the friction ratio between domains I and III for a fixed scan direction. different crystallographic directions (17), and an-
isotropic deformation in films of organic molecules
(18, 19). Our results, however, show frictional
domains with an anisotropy period of 180°, un-
related to the 60° angle between principal crystal-
lographic directions of graphene. Having excluded
the substrate influence as well as electronic and
chemical effects, we are led to assign the friction
differences in the domains to out-of-plane elastic
puckering, as proposed by Lee et al. (10), result-
ing from differently oriented ripple structures in
the graphene layer in each domain (Fig. 4).
We propose that the ripple structure is the re-
sult of inhomogeneous interactions of the graphene
with the SiO2 substrate, which causes stress-induced
deformations in the form of ripples along the stiff
directions of the graphene lattice, presumably the
armchair or zigzag directions (20, 21). This ex-
plains the 60° oriented frictional domains shown
in Fig. 2B. Ripple structures were also reported by
Morozov et al. and are related to the impossibility
of simultaneously attaching graphene to the sub-
strate over the entire surface during mechanical
exfoliation (22).
Rippled graphene under unidirectional strain
Fig. 4. Schematic diagrams illustrating the model used to explain the existence of friction domains and will have anisotropic bending stiffness depending
the 180° period friction anisotropy. The model is based on the existence of graphene ripples originated on the direction. If the tip scans along the ripple
by the inhomogeneous pinning of the flake to the SiO2 substrate. The model assumes that the ripple line, bending (or puckering) of the graphene is
line follows directions of high stiffness in the graphene (e.g., the armchair or zigzag directions). (A) prohibited because deformation resistance is high-
Illustration of three ripple domains. (B) Definition of the relative angle (Q) between scan direction and er along the direction perpendicular to the strain.
the ripple lines. (C) Illustration of the puckering effect when the tip deforms the graphene layer by the In contrast, for perpendicular scanning, bending
lateral force exerted along the scan direction. The puckering effect depends on the applied load. is enhanced, leading to an increased contact area

www.sciencemag.org SCIENCE VOL 333 29 JULY 2011 609

REPORTS
between the tip and graphene. Puckering deforma-
tion due to the tip pushing the ripple crest forward
along the scanning direction explains the 180°
anisotropy. Therefore, if Q is the relative angle
between the ripple line and the scan direction, we
can then expect minima and maxima separated
by 90° with an angular dependence following a
|sin Q| relation, because the in-plane force com-
ponent along the direction perpendicular to the
ripple line of the AFM tip varies in this way.
A high loading force induces a large contact
area (100 to 1000 nm2), which can decrease the
relative contribution of ripple deformations to
the friction and thus the dependence on Q. This
is because puckering takes place at the exposed
graphene in contact proximity to the tip (Fig. 4C).
Although the contact area increases at the high
load, the contribution of the ripple deformation
around the tip in relation to the contact area de-
creases, and thus the bending of graphene becomes
less dependent on the scan direction. We can
expect, therefore, more isotropic puckering and
friction at the high load. High-resolution friction
images showed the existence of lattice distortions
that could be associated with the puckering (fig.
S1). These distortions were not observed on bulk
graphite images obtained under identical condi-
tions. Because bulk graphite can be considered
an extremely thick graphene, such distortion as-
sociated with puckering seems to decrease as the
graphene thickness increases.
As a further test of this model, we performed
FFM measurements on a graphene sheet acci-
dentally deposited over a particle in the sub-
strate. The images revealed mechanical wrinkles
and friction domains induced by stress in the
proximity of the particle (figs. S2 and S3). Ripples
created by the radial stress around the particle
are manifested in the formation of domains ex-
hibiting friction anisotropy with similar 180° pe-
riodicity (Fig. 2). The 60° shift between ripple
lines of adjacent domains supports our assump-
tion that the ripple lines are related to the crys-
tallographic direction of the graphene. The friction
contrast in the domains also became weaker af-
ter thermal annealing at 200°C and disappeared
completely after annealing at 400°C (fig. S4).
The initial anisotropy was a characteristic of the
graphene flake previous to equilibration. Because
two-dimensional structured graphene has a neg-
ative thermal expansion coefficient, heating and
cooling processes can restore the graphene to
equilibrium, changing a stress-induced ripple (4).
Our results indicate that friction mapping on
the graphene layer constitutes a powerful tool to
study the ripple structure formed on graphene
during mechanical exfoliation processes. Anoth-
er interesting outcome of our research is that the
control of friction and anisotropic mechanical
properties, such as the one presented here with a
large anisotropy ratio of 215%, could be ex-
ploited in solid lubrication of micro- or nano-
electromechanical systems.
References and Notes
1. K. S. Novoselov et al., Science 306, 666 (2004).
2. A. Hashimoto, K. Suenaga, A. Gloter, K. Urita, S. Iijima,
Nature 430, 870 (2004).
3. A. Fasolino, J. H. Los, M. I. Katsnelson, Nat. Mater. 6, 858
(2007).
4. W. Bao et al., Nat. Nanotechnol. 4, 562 (2009).
5. K. Xu, P. Cao, J. R. Heath, Nano Lett. 9, 4446
(2009).
6. C. H. Lui, L. Liu, K. F. Mak, G. W. Flynn, T. F. Heinz,
Nature 462, 339 (2009).
7. P. Y. Huang et al., Nature 469, 389 (2011).
8. J. Y. Park, P. A. Thiel, J. Phys. Condens. Matter 20,
314012 (2008).
9. C. Lee et al., Phys. Status Solidi B 246, 2562 (2009).
10. C. Lee et al., Science 328, 76 (2010).
11. H. Lee, N. Lee, Y. Seo, J. Eom, S. W. Lee, Nanotechnology
20, 325701 (2009).
12. T. Filleter et al., Phys. Rev. Lett. 102, 086102 (2009).
13. A. C. Ferrari et al., Phys. Rev. Lett. 97, 187401 (2006).
14. J. Y. Park, D. F. Ogletree, P. A. Thiel, M. Salmeron,
Science 313, 186 (2006).
15. D. S. Grierson, E. E. Flater, R. W. Carpick, J. Adhes. Sci.
Technol. 19, 291 (2005).
16. M. Dienwiebel et al., Phys. Rev. Lett. 92, 126101 (2004).
17. J. Y. Park et al., Science 309, 1354 (2005).
18. M. Liley et al., Science 280, 273 (1998).
19. R. W. Carpick, D. Y. Sasaki, A. R. Burns, Tribol. Lett. 7, 79
(1999).
20. K. Min, N. R. Aluru, Appl. Phys. Lett. 98, 013113 (2011).
21. A. Sakhaee-Pour, Solid State Commun. 149, 91 (2009).
22. S. V. Morozov et al., Phys. Rev. Lett. 97, 016801
(2006).
Acknowledgments: Supported by the National Research
Foundation of Korea funded by the Ministry of Education,
Science, and Technology (National Research Laboratory
program grant 2008-0060004; World Class University
program grants R31-2008-000-10057-0 and R31-2008-
000-10055-0; Basic Science Research program grants
KRF-2008-314-C00111, KRF-2010-0005390, 2010-0015035,
2011-0014209, and 2011-0017605; and Quantum
Metamaterials Research Center grant R11-2008-053-03002-0).
M.B.S. was supported by the Office of Basic Energy Sciences,
Division of Materials Sciences and Engineering, U.S.
Department of Energy, under contract DE-AC02-05CH11231.
J.S.C. was supported by a Hi Seoul Science/Humanities
Fellowship from the Seoul Scholarship Foundation.
Supporting Online Material
www.sciencemag.org/cgi/content/full/science.1207110/DC1
Materials and Methods
Figs. S1 to S4
References
18 April 2011; accepted 16 June 2011
Published online 30 June 2011;
10.1126/science.1207110

Synthesis and Characterization of a (CR2) and nitrenes (NR), have been spectro-
scopically characterized in solid inert gas matri-
ces at temperatures of a few K (10, 11) but to
Neutral Tricoordinate Organoboron date have eluded preparative isolation. Nonethe-
less, Braunschweig et al. (12) have shown that
Isoelectronic with Amines borylenes can be incorporated into the ligand
sphere of stable, isolable transition metal com-
plexes (13).
Rei Kinjo,1 Bruno Donnadieu,1 Mehmet Ali Celik,2 Gernot Frenking,2 Guy Bertrand1*
In recent years, stable singlet carbenes such
Amines and boranes are the archetypical Lewis bases and acids, respectively. The former can as N-heterocyclic carbenes (NHCs) (14, 15) and
readily undergo one-electron oxidation to give radical cations, whereas the latter are easily cyclic (alkyl)(amino)carbenes (CAACs) (16) have
reduced to afford radical anions. Here, we report the synthesis of a neutral tricoordinate proven as powerful as transition metal centers
boron derivative, which acts as a Lewis base and undergoes one-electron oxidation into the for stabilizing highly reactive main group element
corresponding radical cation. These compounds can be regarded as the parent borylene (H-B:) species (17, 18). In the boron series, Robinson
and borinylium (H-B+.), respectively, stabilized by two cyclic (alkyl)(amino)carbenes. Ab initio and co-workers (19, 20) have reported that re-
calculations show that the highest occupied molecular orbital of the borane as well as the singly duction of the (NHC)BBr3 adduct A produced
occupied molecular orbital of the radical cation are essentially a pair and a single electron, the isolable stable neutral diborene B, which can
respectively, in the p(p) orbital of boron. be regarded as a dimer of the parent borylene-

1
he chemistry of boron is dominated by numerous clusters involving hypervalent boron UCR-CNRS Joint Research Chemistry Laboratory (UMI 2957),

T compounds in which the element adopts

the +3 oxidation state and acts as a potent
electron pair acceptor, or Lewis acid. To com-
pensate its intrinsic electron deficiency, boron
also often participates in multicenter bonds, and
centers are known (1). At the opposite extreme,
it is only recently that low-valent boron deriv-
atives have been thoroughly explored (2–9).
Among these species, borylenes (BR), the sub-
valent boron(I) derivatives analogous to carbenes
Department of Chemistry, University of California Riverside
(UCR), Riverside, CA 92521–0403, USA. 2Fachbereich Chemie,
Philipps-Universitat Marburg, Hans-Meerwein-Strasse, 35032
Marburg, Germany.
*To whom correspondence should be addressed. E-mail:
guy.bertrand@ucr.edu

610 29 JULY 2011 VOL 333 SCIENCE www.sciencemag.org