STM imaging (29, 32, 38). We performed STM
spectroscopic mapping in a 2.5-nm area around
one graphitic N dopant (STM junction parameters are V = 0.8 V, I = 1 nA). Shown in Fig. 4B
are a subset of these maps, acquired at bias voltages (applied to the sample relative to the tip)
from –0.78 to 0.78 V. The maps did not show
much contrast at high positive bias, but the local
DOS around the N atom was strongly suppressed
at energies below the EF. The local DOS recovered its background value within a few lattice
constants of the dopant atom. We plotted the
radial distribution of the dI/dV intensity (Fig. 4C)
from the maps in Fig. 4B as a function of distance
from the N atom, normalizing the background
value of the dI/dV to unity for each energy. The
intensity of the spectral weight changes caused by
the N dopant were energy-dependent, but the decay lengths were ~7 Å for all energies (Fig. 4D).
This result indicates that the electronic perturbation caused by a nitrogen dopant is localized near
the dopant atom, which is confirmed in large-area
dI/dV maps, and seen in the calculated charge
distribution [see SOM II (e) and fig. S5] and
simulated STM image in Fig. 2B (26).
References and Notes
1. N. W. Ashcroft, N. D. Mermin, Solid State Physics
(Holt, New York, 1976).
2. A. H. Castro Neto, F. Guinea, N. M. R. Peres,
K. S. Novoselov, A. K. Geim, Rev. Mod. Phys. 81,
109 (2009).
3. X. R. Wang et al., Science 324, 768 (2009).
4. I. Gierz, C. Riedl, U. Starke, C. R. Ast, K. Kern, Nano Lett.
8, 4603 (2008).
5. D. C. Wei et al., Nano Lett. 9, 1752 (2009).
6. Y.-C. Lin, C.-Y. Lin, P.-W. Chiu, Appl. Phys. Lett. 96,
133110 (2010).
7. L. S. Panchokarla et al., Adv. Mater. 21, 4726 (2009).
8. Z. Z. Sun et al., Nature 468, 549 (2010).
9. D. Deng et al., Chem. Mater. 23, 1188 (2011).
10. L. T. Qu, Y. Liu, J. B. Baek, L. M. Dai, ACS Nano 4, 1321
(2010).
11. Y. Wang, Y. Y. Shao, D. W. Matson, J. H. Li, Y. H. Lin,
ACS Nano 4, 1790 (2010).
12. J. C. Meyer et al., Nat. Mater. 10, 209 (2011).
13. X. Li et al., Science 324, 1312 (2009); 10.1126/
science.1171245.
14. A. C. Ferrari et al., Phys. Rev. Lett. 97, 187401 (2006).
15. M. S. Dresselhaus, A. Jorio, A. G. Souza Filho, R. Saito,
Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 368, 5355
(2010).
16. A. Jorio et al., J. Phys. Condens. Matter 22, 334204
(2010).
17. M. M. Lucchese et al., Carbon 48, 1592 (2010).
18. J. Yan, Y. B. Zhang, P. Kim, A. Pinczuk, Phys. Rev. Lett.
98, 166802 (2007).
19. A. Das et al., Nat. Nanotechnol. 3, 210 (2008).
20. I. Shimoyama, G. Wu, T. Sekiguchi, Y. Baba, Phys. Rev. B
62, R6053 (2000).
21. N. Hellgren et al., Thin Solid Films 471, 19 (2005).
22. N. Hellgren et al., Appl. Phys. Lett. 79, 4348 (2001).
23. X. Li et al., J. Am. Chem. Soc. 131, 15939 (2009).
24. S. Point, T. Minea, B. Bouchet-Fabre, A. Granier,
G. Turban, Diamond Relat. Mater. 14, 891 (2005).
25. M. Endo, T. Hayashi, S. H. Hong, T. Enoki,
M. S. Dresselhaus, J. Appl. Phys. 90, 5670 (2001).
26. B. Zheng, P. Hermet, L. Henrard, ACS Nano 4,
4165 (2010).
27. J. Tersoff, D. R. Hamann, Phys. Rev. B 31, 805 (1985).
28. P. Giannozzi et al., J. Phys. Condens. Matter 21,
395502 (2009).
29. G. M. Rutter et al., Science 317, 219 (2007).
30. E. Cockayne et al., Phys. Rev. B 83, 195425 (2011).
31. Y. B. Zhang et al., Nat. Phys. 4, 627 (2008).
32. Y. B. Zhang, V. W. Brar, C. Girit, A. Zettl, M. F. Crommie,
Nat. Phys. 6, 74 (2010).
33. G. H. Li, A. Luican, E. Y. Andrei, Phys. Rev. Lett. 102,
176804 (2009).
34. A. Lherbier, X. Blase, Y. M. Niquet, F. Triozon, S. Roche,
Phys. Rev. Lett. 101, 036808 (2008).
35. G. Giovannetti et al., Phys. Rev. Lett. 101, 026803
(2008).
36. P. A. Khomyakov et al., Phys. Rev. B 79, 195425
(2009).
37. K. S. Kim et al., Nature 457, 706 (2009).
Guided Growth of Millimeter-Long
Horizontal Nanowires with
Controlled Orientations
David Tsivion,1 Mark Schvartzman,1 Ronit Popovitz-Biro,2 Palle von Huth,2 Ernesto Joselevich1*
The large-scale assembly of nanowires with controlled orientation on surfaces remains one
challenge preventing their integration into practical devices. We report the vapor-liquid-solid
growth of aligned, millimeter-long, horizontal GaN nanowires with controlled crystallographic
orientations on different planes of sapphire. The growth directions, crystallographic orientation,
and faceting of the nanowires vary with each surface orientation, as determined by their
epitaxial relationship with the substrate, as well as by a graphoepitaxial effect that guides their
growth along surface steps and grooves. Despite their interaction with the surface, these
horizontally grown nanowires display few structural defects, exhibiting optical and electronic
properties comparable to those of vertically grown nanowires. This paves the way to highly
controlled nanowire structures with potential applications not available by other means.
S
emiconductors with controlled structures are
at the core of the most advanced technologies in our everyday life, but defects induced
by stress during growth often degrade the electronic and optical properties. For example, singlecrystal gallium nitride was first produced in 1969
www.sciencemag.org
SCIENCE
VOL 333
38. A. Deshpande, W. Bao, F. Miao, C. N. Lau, B. J. LeRoy,
Phys. Rev. B 79, 205411 (2009).
Acknowledgments: This material is based on work supported
as part of the Center for Re-Defining Photovoltaic
Efficiency Through Molecule Scale Control, an Energy
Frontier Research Center funded by the U.S. Department
of Energy (DOE), Office of Science, Office of Basic Energy
Sciences under award no. DE_SC0001085. Support was
also provided by the Air Force Office of Scientific Research
under grant no. FA9550-11-1-0010 (A.N.P); by the
DOE under grants DE-FG02-88ER13937 (G.W.F) and
DE-FG02-07ER15842 (T.H.) for research carried out in part
at the Center for Functional Nanomaterials, Brookhaven
National Laboratory, contract no. DE-AC02-98CH10886
(M.S.H.) and at the National Synchrotron Light Source,
contract no. DE-AC02-98CH10886; by the Office of Naval
Research under Graphene Multidisciplinary University
Research Initiative (A.P. and P.K.); by the Defense Advanced
Research Projects Agency Carbon Electronics for RF
Applications program (P.K.); by the NSF under grant no.
CHE-0641523 (A.P.); by the New York State Office of Science,
Technology and Academic Research and by the Priority
Research Centers Program (2011-0018395) through the
National Research Foundation of Korea funded by the
Ministry of Education, Science and Technology (K.S.K.).
Equipment and material support was provided by the NSF
under grant CHE-07-01483 (G.W.F.). Portions of this research
were carried out at the Stanford Synchrotron Radiation
Lightsource (SSRL), a Directorate of SLAC National Accelerator
Laboratory and an Office of Science User Facility operated for
the DOE Office of Science by Stanford University. We thank C.
Jaye and D. Fischer for assistance in using National
Synchrotron Light Source beamline U7A, H. Ogasawara for
assistance at SSRL beamline 13-2, and C. Marianetti and D.
Prezzi for useful discussions. The authors declare no
competing financial interests. Requests for materials should
be addressed
to A.N.P.
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6045/999/DC1
Materials and Methods
SOM Text
Figs. S1 to S7
23 May 2011; accepted 29 June 2011
10.1126/science.1208759
by epitaxial growth on sapphire (1), but it took
three decades of research to reduce its defect concentration before it became the basis of the
blue light-emitting diodes (LEDs) and violet lasers
that enabled outdoor TV screens and the Bluray disk. The vapor-liquid-solid (VLS) growth
method, first described in 1964 (2), has attracted
overwhelming renewed attention during the past
decade (3), as a means of producing stress-free
single-crystal semiconductor nanowires with unparalleled electronic and optical properties suitable
for ultraminiaturized electronics (4), optoelectronics
(5), renewable energy (6), and chemical and biological sensing (7). Yet the large-scale assembly
of horizontal nanowires with controlled orientation on surfaces remains a challenge toward their
integration into practical devices. To this end, several assembly processes have been devised, including the use of liquid flows (8), electric fields (9),
Downloaded from www.sciencemag.org on August 20, 2011
REPORTS
1
Department of Materials and Interfaces, Weizmann Institute of
Science, Rehovot 76100, Israel. 2Chemical Research Support,
Weizmann Institute of Science, Rehovot 76100, Israel.
*To whom correspondence should be addressed. E-mail:
ernesto.joselevich@weizmann.ac.il
19 AUGUST 2011
1003
REPORTS
Downloaded from www.sciencemag.org on August 20, 2011
Fig. 1. Guided VLS growth of horizontal nanowires:
Concept and realization. (A) Schematic view of
guided VLS growth (right) versus conventional VLS
growth (left). (B) Three postulated modes of guided
VLS growth (schematic cross-sectional views): (a)
epitaxial growth along specific lattice directions, (b)
graphoepitaxial growth along L-shaped nanosteps of
an annealed miscut substrate, and (c) graphoepitaxial
growth along V-shaped nanogrooves of an annealed
unstable low-index substrate. (C) Experimental realization of (B) for GaN nanowires on different planes
of sapphire (cross-sectional TEM images): (a) C
(0001), (b) annealed miscut C (0001) by 2° toward
[ 1100], and (c) annealed M (1010). (D) Ultralong
(>1 mm), unidirectional GaN nanowires grown on R
(1102) sapphire. (E) Perfectly aligned GaN nanowires grown on annealed M-plane sapphire. Detail:
magnification of the dense nanowires (highlighted
in yellow) along the sapphire nanofacets. (Note that
nanowires in SEM images may appear broadened as
a result of static charging.) (F) Atomic force microscope (AFM) image of unidirectional GaN nanowires
grown on nonannealed M-plane sapphire.
Fig. 2. Graphoepitaxial effect on the guided VLS
growth of horizontal nanowires. (A) Graphoepitaxial effect on C-plane sapphire (schematic). (B)
Corresponding scanning electron microscope (SEM)
images showing how the guided growth of GaN
nanowires changes from six directions on wellcut C plane (left side) to bidirectional growth, by
the introduction of nanosteps when the substrate
is miscut toward [1100] (top right) and [1120]
(bottom right). (C) Graphoepitaxial effect on M-plane
sapphire (schematic). (D) Corresponding SEM images
showing how the guided growth changes from unidirectional (left) to bidirectional at T90°, by the introduction of nanogrooves when the substrate is
annealed (right).
1004
19 AUGUST 2011
VOL 333
SCIENCE
www.sciencemag.org
Langmuir-Blodgett compression (10), and mechanical shear (11).
Although these postgrowth assembly methods
yield relatively well-aligned arrays on a variety of
surfaces on a wafer-scale size, their alignment is
subject to thermal and dynamic fluctuations. The
assembled nanowires are usually not much longer
than 10 mm, and there is no control over their
crystallographic orientation. VLS growth enables
the production of nanowires with controlled polar
or nonpolar crystallographic orientations that af-
fect their optical and electronic properties (12), as
well as nanowire p-n and p-i-n heterojunctions with
modulated doping (13) and composition (14),
suitable for electronics, optoelectronics, and photovoltaics (15). But when these polar nanowires and
heterojunctions are collected and dispersed to be
assembled, their directionalities become irreversibly scrambled.
Guided growth of horizontally aligned nanowires
(Fig. 1A, right) could overcome the limitations of
postgrowth assembly by combining synthesis and
Fig. 3. Different morphologies of guided GaN nanowires (cross-sectional TEM images, white-line squares
indicating the zoom areas to the right): (A) Perfectly matched epitaxial nanowires, exemplified by a GaN
nanowire on R-plane sapphire. The nanowire and substrate fringes display a perfect 3:2 ratio with no misfit
dislocations (sample slightly tilted to highlight vertical fringes). (B) Mismatched epitaxial nanowires,
exemplified by a GaN nanowire on M-plane sapphire. The 16% lattice mismatch is released by the formation of misfit dislocations (indicated by white arrows) consistent with the 6:7 ratio between the nanowire
and substrate lattice parameters in this orientation. (C) Graphoepitaxial nanowires along L-shaped nanosteps,
exemplified by a GaN nanowire on annealed C-plane sapphire miscut 2° toward [1100]. (D) Graphoepitaxial
nanowires along V-shaped nanogrooves, exemplified by a GaN nanowire on annealed M-plane sapphire.
The atomic models on the right are consistent with fringes and TEM simulations.
www.sciencemag.org
SCIENCE
VOL 333
assembly in a single step. The guided growth of
carbon nanotubes, demonstrated by our group
along atomic steps (16) and nanosteps (17) of
miscut C-plane sapphire, is now widely used
on a variety of planes of sapphire and quartz (18).
VLS growth of horizontally aligned semiconductor nanowires has been reported for a limited
number of materials and substrates, including ZnO
nanowires on A-plane sapphire (19) and GaAs
nanowires on GaAs(100) (20). In both cases, the
alignment was induced by a single epitaxial relationship, and hence the possibility of controllably varying the crystallographic orientation of the
nanowires was not demonstrated; nor did previous
work exploit the strong guiding effect of surface
steps and grooves, known as “graphoepitaxy” (21).
Overall, the guided VLS growth was not sustainable enough to produce horizontally aligned
nanowires much longer than 10 mm, whereas guided
growth of carbon nanotubes can yield aligned
arrays longer than 1 mm.
We report the guided VLS growth of aligned,
millimeter-long, horizontal GaN nanowires with
controlled crystallographic orientations on several different planes of sapphire, including both
atomically flat and spontaneously faceted surfaces. We focused our study on GaN because,
besides its great technological importance, this
material is known to possess a variety of epitaxial
relationships with different planes of sapphire
(22). This gives us an opportunity to investigate
the effects of epitaxy and graphoepitaxy, acting
in cooperation or in competition, to determine
the orientation, morphology, atomic structure,
and physical properties of horizontally grown
nanowires.
We postulate three general modes of guided
VLS growth of horizontal nanowires, depending
on the substrate morphology, as depicted in Fig.
1B: (a) epitaxial growth along specific lattice directions, driven by minimization of interfacial energy and strain; (b) graphoepitaxial growth along
L-shaped nanosteps spontaneously formed upon
annealing a miscut substrate, driven by maximization of the interface area between the substrate
and the nanowire or the catalyst; and (c) graphoepitaxial growth along V-shaped nanogrooves
spontaneously formed upon annealing an unstable
low-index surface, driven similarly to (b). Sapphire
(a-Al2O3) is a trigonal crystal of the R 3c space
group. Its equilibrium Wulff shape is characterized by facets C{0001}, Rf1102g, Sf1011g,
Pf1123g, Af1120g, and Mf1010g (too unstable
to be present), in order of increasing surface energy
(23). Thus, when sapphire substrates with unstable
orientations, such as M plane or miscut C plane,
are annealed at a high temperature, they spontaneously become periodically faceted, as shown in
Fig. 1B, (b) and (c), respectively (24).
We grew GaN nanowires on eight different
planes of sapphire: C, A, R, M, annealed M, and
annealed miscut C planes tilted by 2° toward
½1120, ½1100, and ½1100. The GaN nanowires
were grown by chemical vapor deposition (CVD)
from Ga2O3 powder in a flow of N2, H2, and
19 AUGUST 2011
Downloaded from www.sciencemag.org on August 20, 2011
REPORTS
1005
Fig. 4. Optical and electronic properties of guided GaN nanowires. (A)
Typical room-temperature, single-nanowire photoluminescence spectra of (a)
horizontally grown GaN nanowires on M-plane sapphire, (b) horizontally
grown GaN nanowires on annealed C-plane miscut sapphire toward [1120],
and (c) vertically grown GaN nanowires dispersed on A-plane sapphire. Excitation wavelength lex = 325 nm. The sharp peaks at 329 to 333 nm are from
Raman scattering. (B) Electrical characterization of guided GaN nanowires
in a single-nanowire FET configuration at room temperature. Graphs of
NH3 at 950°C and 400 mbar (25). The catalyst
consisted of Ni nanoparticles patterned as islands or stripes by photolithography. After CVD,
the bulk of the catalyst stripes was covered with
entangled forests of vertically grown GaN nanowires, whereas the stripe edges led to a large
number of horizontal nanowires extending onto
the clean sapphire surface. Sonication for a few
seconds in isopropanol removed the nanowire
forests, leaving only the horizontal nanowires
in place.
Growth from catalyst stripes on atomically
flat R-plane sapphire yielded horizontally aligned
GaN nanowires longer than 1 mm (Fig. 1D), although the alignment was not perfect, whereas on
annealed M-plane sapphire the nanowires were
perfectly aligned with no observable deviations
(Fig. 1E). The better alignment of the latter is attributed to the graphoepitaxial effect. The nanowires
have a typical width of 5 to 50 nm, and their
arrays can be as dense as 20 nanowires/mm.
The graphoepitaxial effect is exemplified on
annealed miscut C-plane and annealed M-plane
sapphire: On well-cut C-plane sapphire (Fig. 2, A
and B, left), GaN nanowires formed random
triangular patterns by growing along the six isomorphic ⟨1120⟩ directions defined by the threefold symmetry of the C plane and the threef1120g
glide planes [contrary to (26), which reports growth
along ⟨1010⟩]. However, when the C plane was
miscut by 2° toward ½1100, the GaN nanowires
chose to grow only along two directions T½1120,
forming parallel arrays (Fig. 2, A and B, right).
Here, the VLS growth was clearly guided by the
L-shaped nanosteps, as postulated in Fig. 1B, (b).
Furthermore, when the C plane was miscut by 2°
toward ½1120, the GaN nanowires were coerced
to grow along the T½1100 directions of the steps,
at 30° from the growth directions preferred by the
well-cut C plane (Fig. 2B, bottom right). Hence,
1006
source-drain current (Isd) versus source-drain voltage (Vsd) are displayed for
different gate voltages (Vg). Upper left inset: AFM image of a typical FET
device, showing the nanowire and four electrodes before deposition of the
dielectric and top gate. Lower right inset: Isd-Vg curves for different Vsd bias
voltages. The charge carrier density can be extracted from the linear part of
the graph (25). The electrical measurements were preformed on nanowires
horizontally grown on R-plane and annealed M-plane sapphire, yielding
similar results.
graphoepitaxy overrules epitaxy. Analogously, on
nonannealed M-plane sapphire, GaN nanowires
grew unidirectionally along½0001 (Fig. 2, C and D,
left, and Fig. 1F). But when the same M-plane sapphire was annealed before CVD, the GaN nanowires
switched their growth direction by T90°, growing
bidirectionally along the T ½1120 directions of
the spontaneously faceted V-shaped nanogrooves
(Fig. 2, C and D, right, and Fig. 1E), as postulated
in Fig. 1B, (c).
We characterized the crystallographic orientation of the GaN nanowires and their epitaxial/
graphoepitaxial relationship with the sapphire substrate by cutting thin (50 to 100 nm) slices across
the nanowires by means of a focused-ion beam
(FIB) and observing them under a high-resolution
transmission electron microscope (TEM). Lowmagnification images (Fig. 1C) confirmed the
postulated modes of guided growth (Fig. 1B), and
most nanowires displayed well-defined facets.
Higher-magnification images displayed clear
fringes on both the GaN and Al2O3, which can
be identified with known crystallographic data
and fit to atomic models (Fig. 3). All the GaN
nanowires had a hexagonal wurtzite structure,
but their morphology and orientation varied with
the substrate orientation. The images in Fig. 3, A
to D, exemplify guided nanowires with four different morphologies: a perfectly matched epitaxial nanowire, a mismatched epitaxial nanowire,
a graphoepitaxial nanowire along an L-shaped
nanostep, and a graphoepitaxial nanowire along a
V-shaped nanogroove. These four cases are exemplified, respectively, on the R, M, annealed miscut
C toward ½1100, and annealed M planes of sapphire. The growth directions, crystallographic orientations, epitaxial relationships, and faceting
of these and other nanowires and substrates
are described in fig. S1 and tables S1 and S2. For
instance, on R-plane sapphire (Fig. 3A), the
19 AUGUST 2011
VOL 333
SCIENCE
GaN nanowires usually grew in epitaxy with the
ð1102ÞAl2O3 plane, along the ½1101Al2O3 direction, and with a nonpolar ½1010GaN orientation.
These nanowires have isosceles triangular cross
sections with their (0001)GaN face downward in
epitaxy with the substrate, and ð1212ÞGaN and
ð11212ÞGaN facets exposed. Here the mismatch is
almost negligible (0.49%), and no misfit dislocations are observed. In contrast, on M-plane
sapphire (Fig. 3B) there can be a large mismatch
(16%) across the nanowire axis, where the
nanowire-substrate interface displays misfit dislocations every six fringes of the sapphire. This is
consistent with the 6:7 ratio between the lattice
parameters of the GaN and Al2O3 across the
nanowire in this orientation. Here, the nanowire
axis has a semipolar ½1123GaN orientation. On
annealed miscut C plane toward ½1100 (Fig. 3C),
the nanowires grew, as mentioned (Fig. 2C, top
right), along the T½1120Al2O3 directions, namely
a subset of the same directions preferred on the
well-cut C plane. Surprisingly, though, the crystallographic orientation of the GaN nanowires on
the miscut C plane switched from the nonpolar
½11010GaN axis preferred on the well-cut C plane
to an unusual semipolar ½1453GaN axis. Thus,
graphoepitaxy not only overrules the preferred epitaxial relationship, but can also induce nanowires
to grow with unusual epitaxial relationships and
crystallographic orientations. On annealed M-plane
sapphire (Fig. 3D), the nanowires grow in epitaxial
relationship with either the S- or R-plane facets
of the V-shaped nanogrooves, with the same
½1453
1
GaN axial orientation as on miscut C plane.
We performed similar analyses for more than
85 nanowires on eight different substrate orientations (figs. S1 to S16 and table S1). In general,
the nanowires grew in well-defined and reproducible directions with a single crystallographic
orientation, although in some cases there were
www.sciencemag.org
Downloaded from www.sciencemag.org on August 20, 2011
REPORTS
REPORTS
the optical properties by single-nanowire photoluminescence (Fig. 4A). The photoluminescence
spectra were typical of GaN, although with a
slight blue shift (lmax = 354 nm), possibly a result
of quantum confinement or compressive strain
(28), and did not show any signs of the typical
“yellow tail” that is usually observed in defective
GaN because of optical transitions involving
in-gap states (22). Moreover, the photoluminescence spectra from horizontally grown nanowires
were very similar to those from nanowires vertically grown in the same conditions and dispersed
onto the same substrate with an even higher
intensity.
Another property that may be affected by
defects is the charge carrier mobility. We characterized the electronic properties of the horizontal epitaxial and graphoepitaxial GaN nanowires
by building single-nanowire field-effect transistors (Fig. 4B) and thin-film transistors (fig. S17)
with top gates. The nanowires displayed n-type
behavior (Fig. 4B) with mobility m = 60 to 250
cm2 V–1 s–1 and electron density ne = 1 × 1018 to
2 × 1019 cm−3. These values are comparable with
those reported for nonhorizontally grown GaN
nanowires (29, 30).
Overall, both the optical and electronic properties of horizontally grown GaN nanowires were
surprisingly similar or even superior to those of
nonhorizontally grown ones. This indicates that
the interaction with the substrate during growth
does not induce defects that degrade the optical
and electronic properties of the semiconductor, as
opposed to epitaxial growth of 2D films. This
important difference can be attributed to the fact
that in a continuous film, stress accumulates in
two dimensions, whereas in a nanowire, stress
accumulates only in the macroscopic axial direction and can be effectively released by contraction, expansion, or localized misfit dislocations
in the transverse direction. Thus, guided VLS
growth enjoys the beneficial orientation control of epitaxial growth without suffering the
stress-related problems of traditional epitaxial
growth in 2D layers.
The mechanism of guided VLS growth remains to be understood. A recent theoretical
model predicts horizontal VLS growth of nanowires, on the basis of catalyst droplet statics and
the formation and growth of nanowire facets
(31). Although this model appears to be highly relevant to our results, as a continuum model, it does not refer to the epitaxial relationship
between the nanowire material and the substrate [for further discussion, see (25) and figs.
S18 to S20].
We believe that the guided VLS growth of
nanowires, demonstrated here for GaN on sapphire, could be extended to a large variety of semiconductors and substrates, producing ordered
arrays of nanowires and heterojunctions with
coherently modulated composition and doping.
This will enable the production of semiconductors with highly controlled structures and unique
properties, suitable for a wide range of applica-
www.sciencemag.org
SCIENCE
VOL 333
tions, including nanoscale high-power circuits,
LEDs, lasers, photovoltaic cells, photodetectors,
and radio-frequency, photonic, and nonlinear
optical devices.
References and Notes
1. H. P. Maruska, J. J. Tietjen, Appl. Phys. Lett. 15,
327 (1969).
2. R. S. Wagner, W. C. Ellis, Appl. Phys. Lett. 4, 89
(1964).
3. P. D. Yang, R. X. Yan, M. Fardy, Nano Lett. 10,
1529 (2010).
4. W. Lu, C. M. Lieber, Nat. Mater. 6, 841 (2007).
5. R. X. Yan, D. Gargas, P. D. Yang, Nat. Photonics 3,
569 (2009).
6. A. I. Hochbaum, P. D. Yang, Chem. Rev. 110, 527
(2010).
7. Y. Cui, Q. Q. Wei, H. K. Park, C. M. Lieber, Science
293, 1289 (2001).
8. Y. Huang, X. F. Duan, Q. Q. Wei, C. M. Lieber,
Science 291, 630 (2001).
9. P. A. Smith et al., Appl. Phys. Lett. 77, 1399
(2000).
10. S. Jin et al., Nano Lett. 4, 915 (2004).
11. Z. Y. Fan et al., Nano Lett. 8, 20 (2008).
12. T. Kuykendall et al., Nat. Mater. 3, 524 (2004).
13. C. Yang, Z. H. Zhong, C. M. Lieber, Science 310,
1304 (2005).
14. M. S. Gudiksen, L. J. Lauhon, J. Wang, D. C. Smith,
C. M. Lieber, Nature 415, 617 (2002).
15. T. J. Kempa et al., Nano Lett. 8, 3456 (2008).
16. A. Ismach, L. Segev, E. Wachtel, E. Joselevich,
Angew. Chem. Int. Ed. 43, 6140 (2004).
17. A. Ismach, D. Kantorovich, E. Joselevich, J. Am.
Chem. Soc. 127, 11554 (2005).
18. S. J. Kang et al., Nat. Nanotechnol. 2, 230 (2007).
19. B. Nikoobakht, C. A. Michaels, S. J. Stranick,
M. D. Vaudin, Appl. Phys. Lett. 85, 3244 (2004).
20. S. A. Fortuna, J. G. Wen, I. S. Chun, X. L. Li, Nano Lett. 8,
4421 (2008).
21. M. W. Geis, D. C. Flanders, H. I. Smith, Appl. Phys. Lett.
35, 71 (1979).
22. H. Morkoç, Handbook of Nitride Semiconductors and
Devices (Wiley-VCH, Weinheim, Germany, 2008), vol. 1.
23. J. H. Choi et al., J. Am. Ceram. Soc. 80, 62 (1997).
24. R. Gabai, A. Ismach, E. Joselevich, Adv. Mater. 19,
1325 (2007).
25. See supporting material on Science Online.
26. Z. Wu, M. G. Hahm, Y. J. Jung, L. Menon, J. Mater. Chem.
19, 463 (2009).
27. P. Waltereit et al., Nature 406, 865 (2000).
28. J. B. Schlager et al., J. Appl. Phys. 103, 124309
(2008).
29. Y. Huang, X. F. Duan, Y. Cui, C. M. Lieber, Nano Lett.
2, 101 (2002).
30. T. Kuykendall et al., Nano Lett. 3, 1063 (2003).
31. K. W. Schwarz, J. Tersoff, Nano Lett. 11, 316
(2011).
Acknowledgments: We thank L. Zeiri for assistance in
PL measurements; M. Bar-Sadan and L. Houben for
assistance with TEM simulations; and W. D. Kaplan,
M. Baram, H. Meltzman, R. Tenne, and L. Kronik for
helpful discussions and critical comments. Supported by
the Israel Science Foundation, Kimmel Center for
Nanoscale Science, Moskowitz Center for Nano and
Bio-Nano Imaging, and the Djanogly, Alhadeff, and
Perlman foundations.
Downloaded from www.sciencemag.org on August 20, 2011
secondary epitaxial relationships leading to nanowires with other crystallographic orientations. In
these cases, the axial orientation was usually similar
for all the nanowires grown on each substrate,
whereas the rotational orientation of the nanowires
around their axis varied from one nanowire to
another (table S1). This can be attributed to the
fact that stress can be more easily relieved across
the nanowire than along its axis. The rotational
variability was especially pronounced in graphoepitaxial nanowires along V-shaped nanogrooves.
This may be related to the fact that, along these
faceted substrates, the nanowires must accommodate incompatible epitaxial relationships with two
different facets (fig. S13). We note that the epitaxial nanowires usually prefer nonpolar axial
orientations (½1010GaN on C-, A-, and R-plane
sapphire, and ½1120GaN on M-plane sapphire),
whereas the graphoepitaxial nanowires tend to
adopt unusual semipolar orientations such as
1½1453GaN and ½1102GaN. Elementary mapping
of Ga, N, Al, and O by electron energy loss spectroscopy (EELS) revealed sharp GaN-Al2O3 interfaces without observable interdiffusion (fig. S4).
Our results show a relationship between the
nanowire growth directions and the symmetry of
the chosen substrate plane, as shown in fig. S1.
Flat substrates with D3, C2, or D1 morphological
symmetries led to epitaxial nanowire growth along
six directions, two directions, and one single direction, respectively. When the substrate had D1
symmetry and the nanowires grew parallel to the
only symmetry plane, as on R- and M-plane sapphire, there was no symmetry operation that could
exchange between the forward and backward directions of growth. This asymmetry caused the
nanowires to grow in one single direction. Such
unidirectional growth has clear advantages over
bidirectional growth both for polarity control and
for integration purposes, because one can pattern
the catalyst and have all the nanowires extending from it in the same direction (Fig. 2D, left).
The photophysical properties of semiconductor
nanowires and thin films were found to be affected
by their polarity; this is because radiative lifetimes and quantum efficiencies are affected by
changes in electron-hole overlap due to internal
fields (27). Epitaxial VLS growth on different
substrates has enabled the selective production of
polar and nonpolar nanowires as vertical arrays
(12), but no available postgrowth method based
on dispersions can horizontally align them with
coherent polarity, whereas guided growth could
yield coherently oriented nanowires with similar
polarity. This polarity control may have important advantages in photonic, optoelectronic, and
radio-frequency applications, as well as in nonlinear optics (22).
Once the guided growth of semiconductor
nanowires has been demonstrated, the next question is to what extent the horizontal epitaxial or
graphoepitaxial growth may lead to stress-induced
defects, which can affect the optical and electronic
properties of the nanowires. We used a confocal
microscope with a He-Cd laser to characterize
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6045/1003/DC1
Materials and Methods
SOM Text
Tables S1 and S2
Figs. S1 to S20
References (32–37)
16 May 2011; accepted 11 July 2011
10.1126/science.1208455
19 AUGUST 2011
1007
www.sciencemag.org/cgi/content/full/333/6045/1003/DC1
Supporting Online Material for
Guided Growth of Millimeter-Long Horizontal Nanowires with Controlled
Orientations
David Tsivion, Mark Schvartzman, Ronit Popovitz-Biro, Palle von Huth, Ernesto Joselevich*
*To whom correspondence should be addressed. E-mail: ernesto.joselevich@weizmann.ac.il
Published 19 August 2011, Science 333, 1003 (2011)
DOI: 10.1126/science.1208455
This PDF file includes:
Materials and Methods
SOM Text
Tables S1 and S2
Figs. S1 to S20
References (32–37)
3
1.
Materials and Methods
1.1
Substrate preparation
Sapphire (α-Al2O3) wafers with eight different orientations were used. Singular
(i.e., well-cut) sapphire wafers, including C-plane (0001), -plane ( 101 0 ), -plane
(1120) and R-plane ( 1 1 02), were purchased from Roditi International. Miscut C-plane
(0001) sapphire wafers with different miscut directions were custom-produced as
previously described (16,32). The manufacturer (Gavish Ltd., Omer, Israel), produced the
wafers as directed, and the exact miscut tilt and azimuth angles were determined by our
own developed procedure of asymmetric double exposure back-reflection XRD (17).
Miscut C-plane and M-plane sapphire substrates were annealed at 1100 ºC and 1400 ºC,
respectively, in air, using a high-temperature muffle furnace (Carbolite RHF 16/8). Prior
to use, the substrates were sonicated for 10 minutes in acetone, then rinsed in acetone,
isopropyl alcohol (IPA), and distilled H2O, and blow-dried in N2.
1.2
Catalyst deposition and patterning
The Ni catalyst was deposited by electron-beam evaporation of a thin (5-20Å)
metal layer. First, the areas for catalyst deposition were defined using standard
photolithography with negative or positive photoresists. After pattern development and
surface cleaning by oxygen plasma (March Plasmod GCM 200, 2 min, 1 sccm of O2, 100
W) thin films (5-20 Å) of Ni was deposited by electron-beam evaporation. These thin Ni
films undergo dewetting upon heating, thus generating the nanoparticles that serve as
catalyst for the VLS growth of nanowires (Fig. S18-20). We also performed experiments
using Ni, Fe and Au catalysts from colloidal suspensions and salts, obtaining similar
4
results. However, all the results presented in this report were performed with evaporated
Ni.
1.3
Nanowire synthesis
Nanowires were grown in a home-built hot-wall CVD with fast heating capability.
The gallium source was Ga2O3 powder placed inside the oven. The nitrogen source is dry
NH3 (99.999%, Maxima). H2, Ar or N2 are used as carrier gases (all gases 99.999%,
Gordon Gas). In a typical experiment, a sample is placed on a fused-silica carrier plate
and inserted to a 25 mm diameter quartz tube. The tube is inserted into a split oven and
purged of oxygen by 5 cycles of pumping to 10 mbar and purging with N2 or Ar at
elevated temperature. Once the tube is purged, dry NH3 (2-30 sccm) and H2 gas (30-120
sccm) are streamed into the tube, and pressure is maintained between 50 torr to
atmospheric pressure. Once the desired temperature (930-1000 Cᵒ) is achieved, the oven
is moved over the Ga2O3 powder, and turned off at the end of the growth time (10 min to
8 hours).
1.4
Microscopic characterization
The grown nanowires were characterized using field-emission SEM (Supra 55VP
FEG LEO Zeiss) at low working voltages of 1-5kV. Atomic force microscopy (AFM,
Veeco, Multimode Nanoscope IV). Images were acquired in air tapping mode using 70
kHz (FESP1) and 300 kHz (RTESP7) silicon tips (Nanoprobes).Thin lamellae for TEM
characterization were made using a FEI Helios DualBeam microscope, and inspected
using a FEI Tecnai F30-UT field emission TEM, equipped with a parallel EELS (Gatan
imaging filter) operating at 300 kV. TEM digital images were recorded using a Gatan
Ultrascan1000 CCD camera. TEM images were analyzed to determine crystallographic
5
orientation and epitaxial relationships using Fourier transform (FFT) from selected areas
in the nanowire cross-sections. Indexing of the FFT peaks was done according to
crystallographic tables for bulk GaN, as shown in Fig. S11. Atomic models were verified
with TEM simulations, using EMS software (33) for nanowires grown on C, A, and Mplane sapphire.
1.5
Optical characterization
Photoluminescence spectra from single GaN nanowires were acquired with a Jobin-
Yvon LabRam HR 800 micro-Raman system, equipped with a liquid-N2-cooled detector.
For the excitation, we used a He-Cd laser at 325 nm. The laser power on the sample was
5 mW. The measurements were taken with a 2400 g mm-1 grating for the UV (spectral
resolution 4 cm-1) and a confocal microscope with a 100 m aperture and a 10 m laser
spot. Care was taken to measure spectra from isolated nanowires. Photoluminescence
spectra were measured up to 730 nm for (i) horizontally grown nanowires on C(0001),
R( 1 1 02 ), M( 10 1 0 ), annealed M( 10 1 0 ) and miscut C(0001) toward [10 1 0] , (ii)
nanowires vertically grown as forests under the same conditions as (i) and dispersed on
A( 1120 ) sapphire. Spectra from more than 20 single nanowires were measured, yielding
quite similar results. No yellow photoluminescence was observed in any of the
nanowires. Occasionally, a few nanowires, both horizontally and vertically grown,
displayed a small shoulder around 420 nm. No other peaks were observed at longer
wavelengths up to 730 nm.
6
3.6
Electrical characterization
GaN nanowire field effect transistors (FETs) were fabricated by defining source
and drain electrodes using electron-beam lithography (JEOL 6400 SEM equipped with
JC Nabity Nanopattern Generator System), followed by electron-beam evaporation of Ti
(20 nm) and Au (30 nm) (34), and liftoff in acetone. Rapid thermal annealing (500oC,
3min) was done to ensure low-resistance ohmic characteristics of the fabricated contacts.
The typical electrodes were of 2 m width, separated by 2-5m from each other. For
gate dielectric, 400 nm thick SiO2 was deposited by plasma enhanced chemical vapor
deposition (PECVD, Nextral 890). A top-gate electrode was patterned by electron-beam
lithography, electron-beam evaporation (Cr 5 nm/Au 50 nm) and liftoff.
Two-terminal electrical characterizations were done by applying source-drain DC
bias, and measuring I-V curves as a function of gate voltage. To characterize the possible
influence of the contact resistance on our measurements, two additional external
electrodes were added to allow 4 point resistance measurement, giving the value of 183
k vs. 197 k obtained by direct two point measurement. The slight difference between
the two values should be attributed to the contact resistance, which was found to be two
orders of magnitude lower than that of the characterized nanowires, and can then be
neglected.
The charge carrier mobility () was extracted from the transconductance gm, which
is defined as the slope of Isd-Vg curve in the linear region and is given by equation (S2),
where Vg is the gate voltage, Idc is the mobility, L is the nanowire channel length, and C is
the capacitance (35). The capacitance is calculated using quasi-circular cross-section
approximation given by equation (S3), where h is the dielectric thickness and a is the
7
width of the horizontal GaN nanowire (36). The carrier concentration ne was extracted
from the threshold voltage Vth using equation (S4), where r is the nanowire radius and the
total charge Q = CVth .
Overall, about 20 nanowire transistors with channel lengths in the range of 3-5
m were fabricated and characterized, giving the mobility values ranging from 60 cm/V.s
to 250 cm/V.s, and the carrier density values ranging from 1x1018 cm-3 to 2x1019 cm-3.
gm 
C
dI dc
dV g
 (
Vsd  const .
C
)Vsd
L2
2 0 L
ln(4h / a)
Q  CV th  n e e  r 2 L
2.
(S2)
(S3)
(S4)
Further discussion on the guided growth mechanism
As mentioned in the main text, the mechanism of guided growth is not yet fully
understood. Ref (31) presents a model that could be relevant to our results. However, as a
continuum model, it does not refer to the epitaxial relationship between the nanowire
material and the substrate. To explain and predict the orientation control in our results, a
more detailed model, beyond the scope of this Report, will be required. To this end, a few
important observations need to be taken into account. First, the epitaxial relationships of
most of our nanowires are consistent with those reported in literature for thin films
produced by different methods (22). This suggests a significant degree of thermodynamic
control. On the other hand, the most common facets observed in our nanowires are not
necessarily the most thermodynamically stable ones. For instance, the {1122 } facets,
which are quite common in our results (Fig. S1 and Table S1), are not part of the
equilibrium Wulff shape of GaN, but are known as fast-growing planes in the kinetic
8
Wulff plot (37). This indicates a significant degree of kinetic control. Moreover, in terms
of symmetry, the unidirectional growth of nanowires with nonpolar orientations on the R
and M planes of sapphire cannot be explained by the thermodynamics of the nanowiresubstrate interaction alone, so it is clear that kinetics and the catalysts play a major role in
the mechanism of guided VLS growth. This also suggests that by changing the growth
conditions, the orientation control could be further improved.
The mechanism of termination of nanowire growth is also yet to be understood. We
may speculate that growth termination can be caused by pinning or poisoning of the
catalysts by defects or impurities on the substrate. Since we pattern the catalyst in stripes
and the nanowires grow onto the clean sapphire areas (Figs. S18-S20), a nanowire can
grow very long before the catalyst gets pinned or poisoned. The guided growth of the
nanowire should exert on the catalyst a large and highly anisotropic force, which can
overcome pinning forces, thus allowing the nanowire to keep growing for very large
distances exceeding 1 mm. The force exerted on the catalyst by graphoepitaxial growth
on a faceted surface is probably more highly anisotropic than that exerted by epitaxial
growth on an atomically flat surface. This could explain why graphoepitaxial nanowires
are generally better aligned than epitaxial nanowires.
9
3. Complete Nanowire Characterization
10
Fig. S1. Growth directions, crystallographic orientations and faceting of horizontally
grown GaN nanowires on different sapphire substrates. The growth directions are
schematically represented in relation with the morphological symmetry of the substrate.
The SEM images show the observed growth direction from patterned catalyst islands.
The cross-sectional TEM images show the most common epitaxial relationship,
crystallographic orientation, faceting and size of the GaN nanowires for each sapphire
plane. The model is based on these observations. Comprehensive data are shown in Table
S1 and Figs. S2-S16.
11
Table S1. Comprehensive crystallographic data for horizontally grown GaN nanowires.
Substrate # of Orientation wires Epitaxial relationship
Horizontal Axial
Transversal
GaN║Al2O3 GaN║Al2O3 GaN║Al2O3 Nanowire facets
C 0001 9
0001 ║ 0001
1010 ║ 1120
1210 ║ 1100 1212 , 0001 , 1212
A 1120 7
0001 ║ 1120
1010 ║ 1010
1210 ║ 0001 1212 , 0001 , 1212
10
1216 ║ 1102
1010 ║ 1101
1212 ║ 1120 1212 , 0001 , 1212
0001 ║ 1102 1010 ║ 1101
1210 ║ 1120
1212 , 0001 , 1212
2
3210 ║ 1102
1453 ║ 1101
1013 ║ 1120 1012 , 0113
13
1103 ║ 1010
1120 ║ 0001
1101 ║ 1210 0001 , 1101
3*
1100 ║ 1010
1120 ║ 0001
0001 ║ 1210 0001 , 1101
4
3210 ║ 1010
1453 ║ 0001
1013 ║ 1210 1012 , 0113
3
1122 ║ 1010
1123 ║ 0001
1100 ║ 1210 0111 , 1122 , 1011
9†
XXXX ‡║ 1010
1453 ║ 1210
XXXX ‡║ 0001 1
1010 ║ 1010
0001 ║ 1210
1210 ║ 0001 0110 , 1010 , 1100
1
1212 ║ 1010
1010 ║ 1210
1215 ║ 0001 0001 , 0001 , 1212
AnnealedC 4§
0001 miscut
toward 1100 1
1101 ║ 0001
1453 ║ 1120
1012 ║ 1100 0001 ║ 0001
1120 ║ 1120
1010 ║ 1100 1102 , 0001 , 1102
1101 ║ 0001
1453 ║ 1120
1012 ║ 1100 1101 ║ 0001
1102 ║ 1100
1120 ║ 1120 3122 , 1322 , 1101
1214 ║ 0001
2241 ║ 1100
1120 ║ 1010 1122 , 3122
0111 ║ 0001
1123 ║ 1100
3121 ║ 1120
1010 , UUUU
R 1102 M 1010 AnnealedM
1010 8
*
AnnealedC
0001 miscut 4§
toward 1100
3
AnnealedC
0001 miscut 2
toward 1120 1
1012 , 1101
1012 , 1101
1012 , 1101
Nanowire axial orientations are highlighted in yellow.
*
These nanowires have the same crystallographic axis orientation as the previous case, but are rotated with respect
to their axis.
†
These nanowires have all the same crystallographic axis orientation, but are rotated with different rotations with
respect to their axis.
‡
XXXX indicates variable directions or planes.
§
Some of these nanowires have the same crystallographic axis orientation, but are rotated with different rotations
with respect to their axis.
ǁ
UUUU Indicates undefined planes.
ǁ
12
Table S2. Mismatch between GaN and Al2O3 along and across the nanowires
Longitudinal
Epitaxy
ao (Å)
GaN║Al2O3
GaN : Al2O3
0001 ║ 0001 2.762 : 2.380
Transversal
ao (Å)
Ratio
Mismatch
GaN : Al2O3
GaN : Al2O3
1:1
-0.16
Ratio
Mismatch
GaN : Al2O3
1.595 : 4.123
1:1
0.60322
7.975 : 8.246
5:2
0.033
2.762 : 4.123
1:1
0.33
1.595 : 2.165
1:1
0.263
8.286 : 8.246
3:2
-0.0048
7.975 : 8.660
5:4
0.079
2.762 : 5.189
1:1
0.461
1.595 : 2.380
1:1
0.330
5.524 : 5.189
2:1
-0.077
4.785 : 4.760
3:2
-0.0049
1100 ║ 1010 1.595 : 2.165
6.380 : 6.495
1:1
0.263
2.594 : 2.380
1:1
-0.0897
4:3
0.018
0001 ║ 1120 0001 ║ 1102 ao : bulk lattice parameter
The mismatch is calculated using equation (S1), where a0 is the lattice spacing for the
substrate and the nanowire in the relevant direction. The mismatch is calculated for 1:1
ratio, and also for other ratios, where such a ratio improves the matching. The table
shows no clear trend favoring a lower mismatch along than across the nanowire.
Therefore, we are not able to claim that strain minimization along the nanowire is the
major factor determining the growth direction. It is thus likely that other factors, such as
the kinetics of nanowire growth could also play a role in determining the growth
directions.
mismatch
ao ( Al2 O3 )  ao (GaN)
ao ( Al2O3 )
(S1)
13
Fig. S2. Cross-sectional TEM of GaN nanowires on C(0001) sapphire. All the nanowires
have the same orientation and the same epitaxial relationship with the substrate, as
summarized in Table S1.
14
Fig. S3. Side-view TEM of a GaN nanowire on C(0001) sapphire. The epitaxial
relationships are in agreement with the cross-sectional TEM images (Fig. S1).
15
Fig. S4. EELS elemental map of GaN on C-plane sapphire (left: Al and Ga, right: N and
O). No interdiffusion between nanowire and substrate is observed.
16
Fig. S5. Cross-sectional TEM of GaN nanowires on A (1120) sapphire. All the
nanowires have the same orientation and the same epitaxial relation with the substrate, as
summarized in Table S1.
17
Fig. S6. Cross-sectional TEM of GaN nanowires on R (1 1 02) sapphire. The vast
majority of nanowires grow along their [ 10 1 0 ]GaN direction. In about half of these
18
nanowires, the (0001) GaN plane is parallel to the substrate, as the nanowire shown in Fig.
2A. The other wires have similar axis orientation, but are rotated by 30° with respect to
their axis, such that their ( 1216 )GaN plane is parallel to the substrate. The facets are
always 1212 , 1212 and 000 1 ). The nanowire growth axis is usually [ 10 1 0 ]GaN,
but sometime it can slightly deviate toward the [ 220 1 ] direction. This small deviation is
within the range of sample alignment at the TEM. An additional nanowire, growing along
the [ 1453]GaN orientation is also shown, but this orientation was only seen twice, on nonflat regions of the substrate. Orientations and epitaxial relationships are summarized in
Table S1.
19
Fig. S7. Cross-sectional TEM of GaN nanowires on M 10 1 0 sapphire having a
1120
GaN axis,
and a ( 1 1 03 )GaN║ 10 1 0
Al2O3interface,
as summarized in Table S1.
20
Fig. S8 Cross-sectional TEM of GaN nanowires on M 10 1 0 sapphire, having a
1120
GaN axis,
and a ( 1 1 00 )GaN║ 10 1 0
Al2O3interface,
as summarized in Table S1.
Fig. S9. Cross-sectional TEM of GaN nanowires on M(1010) sapphire having a 1453 .
axis, as summarized in Table S1.
21
Fig. S10. Cross-sectional TEM of GaN nanowires on M 1010 sapphire having a 1123 axis, as the nanowire shown in Fig 2B, and summarized in Table S1.
Fig. S11. Explanation of the variation in the rotation position of GaN nanowires on Mplane sapphire, based on symmetry of the substrate material. The nanowire growth on M
plane (1010) sapphire can be divided into 3 families, according to the growth axis: (i)
[1120] (by far, the most common), (ii) [112 3] , and (iii) 1453 . However, due to the
symmetry of the substrate in the nanowire growth direction, the nanowires that grow
along the [1120] GaN can be found in 3 orientations, rotated by 60° around the nanowire
22
axis. This is demonstrated in this figure, where the GaN (0001) planes can match the
Al2O3 1120) planes in three ways: (A) The GaN (0001) planes match the sapphire
(1120) planes. (B) The GaN (0001) planes match the sapphire (1210) planes. (C) The
GaN (0001) planes are aligned with the sapphire ( 2110) planes.
23
Fig. S12. Cross-sectional TEM of GaN nanowires on annealed M 1010 sapphire. Three
such types of nanowires, labeled A, B and C can be identified, as summarized in table S1.
The nanowires with the most common orientation, labeled A, share a common grow axis.
24
However, these nanowires can be rotated in different angles with respect to their growth
axis, so they have different epitaxial relationships with the substrate. This could be due to
the competition between the interfaces of the GaN with the R(1102) and S(1101) planes
of the sapphire simultaneously, where each plane has its own preferred orientation.
25
Fig. S13. Analysis of cross-sectional TEM images and identification of crystallographic
nanowire orientation, exemplified for two different GaN nanowires grown on annealed
M-plane sapphire. (A) TEM image. (B) Fourier transform (FFT) from a selected area in
26
the nanowire cross-section. Indexing of the FFT peaks is done according to
crystallographic tables for bulk GaN. The nanowire axis corresponds to the zone axis,
namely the cross product of the first linearly independent indices, (1012) × (1101
1453 . (C) Simulated electron diffraction pattern for the latter crystallographic
orientation, including d-spacings and  = 87.7º, for comparison with the FFT (B). Here,
the R(1012) planes of GaN nanowire are aligned with the R(1102) planes of the
sapphire substrate. The second nanowire (D-F) has the same axial
orientationalong 1453 , but is rotated around its axis with respect to that of the first
nanowire (A-C), so that the S(1101) planes of the GaN are roughly aligned with the
C(0001) planes of the sapphire. It is interesting to note that on the non-annealed M-plane
sapphire the GaN usually grow along their own 1120 direction, which is also an
intersection between R and S planes of GaN, although different from those whose
intersection defines the 1453 direction.
27
Fig. S14. Cross-sectional TEM of GaN nanowires on annealed vicinal miscut C(0001)
sapphire 2ᵒ toward [1 1 00] . Two types of such nanowires, labeled A and B can be
identified, as summarized in table S1.
28
Fig. S15. Cross-sectional of GaN nanowires on annealed miscut C(0001) sapphire 2ᵒ
toward [1 100] . All wires have the same orientation and the same epitaxial relationships
with the substrate, as summarized in table S1.
29
Fig. S16. Cross-sectional TEM of GaN nanowires on annealed miscut C(0001) sapphire
miscut 2ᵒ toward [11 2 0] . Three types of nanowires, labeled A, B and C, can be
identified, as summarized in table S1.
30
Fig. S17. Demonstration of a thin-film transistor (TFT) built on an array of perfectly
aligned horizontal GaN nanowires. (A) Schematic of the device. (B) SEM image (before
31
deposition of the top gate). (C) I-V characteristics of the device for different gate
voltages. The device comprises ca. 40 nanowires spread across a width of 40 m, with
common source and drain electrodes separated by 5 m, and a top gate over a thin (400
nm) SiO2 film. The transconductance at Vsd=1.0 V is gm = 4.5x10-7 Ω-1, corresponding to
an electron mobility  = 246 cm2/Vs per nanowire, in consistency with our singlenanowire FET measurements.
32
Fig. S18. SEM images showing a catalyst nanoparticle (top) at the end of each epitaxial
GaN nanowire horizontally grown on R-plane sapphire from Ni islands (bottom).
33
Fig. S19. SEM images showing a catalyst nanoparticle (bottom) at the end of each
graphoepitaxial GaN nanowire horizontally grown on annealed M-plane sapphire from Ni
islands (top). The substrate nanogrooves are visible in the bottom images.
34
Fig. S20. SEM images showing catalyst nanoparticles at the ends of many
graphoepitaxial GaN nanowires horizontally grown on annealed M-plane sapphire from
Ni islands (bottom) after a short period of time, prior to removal of the vertically grown
nanowires. Each of the 17 nanowires visible in the bottom image shows a catalyst
nanoparticle at its end. The substrate nanogrooves are visible in the bottom images.
35
4.
References
32.
33.
34.
35.
A. Ismach, E. Joselevich, Nano Lett. 6, 1706 (2006).
P.A. Stadelmann, Ultramicroscopy 21, 131 (1987).
J. S. Hwang et al., Appl. Phys. Lett. 85, 1636 (2004).
R. Martel, T. Schmidt, H. R. Shea, T. Hertel, P. Avouris, Appl. Phys. Lett. 73,
2447 (1998).
Y. Cui, X. Duan, J. Hu, C. M. Lieber, J. Phys.Chem. B 104, 5213 (2000).
V. Jindal, F. Shahedipour-Sandvik, J. Appl. Phys. 106, (2009).
36.
37.