LETTER

pubs.acs.org/NanoLett

In Situ Nanomechanics of GaN Nanowires

Jian Yu Huang,*, He Zheng,, S. X. Mao, Qiming Li, and George T. Wang

Sandia National Laboratories, Albuquerque, New Mexico 87185, United States

Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, Pennsylvania 15261, United States

Department of Physics, Center for Electron Microscopy and Key Laboratory of Acoustic and Photonic Materials and Devices, Wuhan
University, Wuhan 430072, China

bS Supporting Information
ABSTRACT: The deformation, fracture mechanisms, and the
fracture strength of individual GaN nanowires were measured in
real time using a transmission electron microscope scanning
probe microscope (TEM-SPM) platform. Surface mediated
plasticity, such as dislocation nucleation from a free surface
and plastic deformation between the SPM probe (the punch)
and the nanowire contact surface were observed in situ.
Although local plasticity was observed frequently, global plasticity was not observed, indicating the overall brittle nature of
this material. Dislocation nucleation and propagation is a
precursor before the fracture event, but the fracture surface shows brittle characteristic. The fracture surface is not straight but
kinked at (10-10) or (10-11) planes. Dislocations are generated at a stress near the fracture strength of the nanowire, which ranges
from 0.21 to 1.76 GPa. The results assess the mechanical properties of GaN nanowires and may provide important insight into the
design of GaN nanowire devices for electronic and optoelectronic applications.
KEYWORDS: GaN nanowire, nanomechanics, dislocation, plasticity, fracture, in-situ electron microscopy

allium nitride (GaN) is a technologically important wide

band gap (Eg = 3.39 eV) semiconductor used in optoelectronic and high frequency and high power electronic
applications.1 12 GaN thin lms heteroepitaxially grown on
sapphire or other substrates exhibit a high density of threading
dislocations, which degrades the eciency and the lifetime of the
GaN-based devices.13 19 In this context, GaN nanowires have
the advantage of low or even no dislocations, which makes them
attractive candidate for GaN-based applications. The electrical,
optoelectrical, and thermal properties of GaN nanowires have
been studied extensively, but relatively little is known on their
mechanical properties.20 Such studies are important because
GaN nanowires may be subjected to mechanical or electrical
stress during device processing or operation, and such stresses
may impact the properties of the nanowires for example via
dislocation generation. In this work, the deformation, fracture
mechanisms, and the fracture strength of individual GaN nanowires were measured in real time using a transmission electron
microscope scanning probe microscope (TEM-SPM) platform.
The GaN nanowires were grown by Ni-catalyzed metal
organic chemical vapor deposition (MOCVD) method on a
sapphire substrate wafer,10,21,22 and the as-grown nanowires were
triangular in cross-section (Figure 1) with stacking faults in the
basal planes similar to that reported in the literature.23,24 The
nanowire diameter varied from 100 nm to about 500 nm, and its
growth direction was generally [1-210] (Figure 1). A tiny piece
was cut o the main wafer and glued to an Al rod with diameter of
about 280 m (Supporting Information Figure S1). Individual
r 2011 American Chemical Society

nanowires were then manipulated to approach either a nanoindentor or a at-ended scanning tunneling microscopy probe
(STM) for in situ compression experiments. In the former, the
force-displacement could be directly measured and recorded by a
computer. The compression was displacement controlled with
the speed varying from 0.1 to 6.8 nm/s.
Figure 2a d and Supporting Information movie M1 show
that dislocations were nucleated from a free surface (Figure 2a,b),
and then propagated along a prismatic plane of the nanowire,
leading to a displacement of the upper segment with respect to
the lower segment of the nanowire (Figure 2c); eventually the
nanowire broke from the location where slippage had occurred
(Figure 2d). Dislocation nucleation from a free surface has been
observed in metallic nanomaterials,25,26 but it has not been
reported in semiconductor nanomaterials to our knowledge.
The result suggests that surface dislocation nucleation may be
a general phenomena to many nanostructured materials, regardless of their metallic, semiconducting, or ceramic nature.
Figure 2e i (Supporting Information movie M2) is another set
of data showing the slip event before the fracture of the nanowire.
There was an abrupt diameter change near the base of the
nanowire, causing a surface step as pointed out by an arrowhead
in Figure 2e. Slip or shear initiated from the step (Figure 2e),
propagated along a prismatic slip plane (Figure 2f), and
Received: January 1, 2011
Revised:
February 13, 2011
Published: March 18, 2011
1618

dx.doi.org/10.1021/nl200002x | Nano Lett. 2011, 11, 16181622

Nano Letters

LETTER

Figure 1. The cross-section of the GaN nanowire is triangular. (a) An atomic structural model of the cross-sectional view of the nanowire. The nanowire
with a growth direction of [1-210] is enclosed by the (000-2), (-1011), and (10-11) planes. (b) A TEM image showing the triangular cross-section of the
nanowire. The enclosed angles between dierent planes are consistent with the model shown in (a). (c) A plan-view of the triangular nanowire. (d) A
general electron diraction pattern from a GaN nanowire. Streaks along the (0002) diraction series indicated stacking faults in the nanowires.

Figure 2. (see also Supporting Information movies M1 and M2) Two

sets of sequential TEM images showing fracture initiated from a slip or a
shear nucleated from the free surface of the nanowire (pointed out by
arrows). (a,b) and (e,f) Dislocation nucleation from free surfaces. (c,f)
Slip of the dislocations. (d,h,i) Fracture of the nanowire. The punch was
advanced at 0.5 and 0.7 nm/s in movies M1 and M2, respectively.

eventually the nanowire fractured along the same slip plane

(Figure 2g i). The fracture surface is smooth. This shows again

that fracture was initiated from a slip or a shear which was

nucleated from a free surface.
Dislocation nucleation and propagation right before fracture
were captured in situ, as shown in Figure 3a d and Supporting
Information movies M3 and M4. The nanowire may have fractured
from the middle in a previous compression experiment, but the
broken segment was still attached to the nanowire, forming a kink.
As the nanowire was compressed, dislocations were nucleated
about 50 nm away from the punch and nanowire contact
(pointed out by an arrow in Figure 3b), crossed the nanowire
along a prismatic plane (pointed out by an arrow in Figure 3c), and
then escaped to the other free surface. Similar dislocation emission
accompanying the nanowire fracture was also observed in other
nanowires (Supporting Information Figure S2). This indicated
that dislocation activity was a precursor of fracture. In many
cases, residual dislocations were observed near the fracture
surface (Figure 3e), and the dislocations were mostly 1/3 [1120] type (Figure 3g) with the slip plane being (1-100)
(Figure 3f,g).
Signicant local plastic deformation was observed right before
fracture in some nanowires. Figure 3h,i and Supporting Information movies M5 to M9 show the fracture process of GaN
nanowires under compression. Note the plastic deformation
induced by piling up of dislocations occurred in the contact
surfaces between the nanowires and the punches (Figure 3h and
Supporting Information movies M5 to M7). Plastic deformation
was not obvious in the nanowires shown in Supporting Information movies M8 and M9. Interestingly, the fracture location is
about 140 nm away from the contact surface (Figure 3h,i), and
the fracture surface is very sharp from low-magnication TEM
images (Figure 3i) with an inclined angle of 64 with respect to
the vertical direction. Similar fracture surfaces were observed in
1619

dx.doi.org/10.1021/nl200002x |Nano Lett. 2011, 11, 16181622

Nano Letters

Figure 3. Dislocation slip before fracture. (a d) (see also Supporting

Information movies M3, M4) Sequential TEM images showing that
dislocations were emitted underneath the punch, crossed the nanowire,
and escaped to the other surface right before fracture. (e g) Residual
dislocations below the fracture surface. The Burgers vector was determined to be 1/3 [11-20] (see the Burgers circuit in (g)) and the slip plane
was (1-100). The directions indicated in (g) are projections. (h j) (see
also Supporting Information movies M5 to M7) Signicant plastic
deformation occurred underneath the punch before fracture. A lowmagnication TEM image showed that the fracture plane is inclined 64
with respect to the vertical direction (i), while an HRTEM image showed
that the fracture plane is kinked at (10-10) and (10-11) planes (j).

many nanowires (Supporting Information Figure S3). Dislocations were also observed near the fracture area (Figure 3i,
arrows). High-resolution transmission electron microscopy
(HRTEM) (Figure 3j) shows that the atomic scale fracture
surface is not at but kinked at (10-10) and (10-11) planes,
both of which are the slip planes in a hexagonal system.
Figure 4a c (Supporting Information movies M10, M11) and
Figure 4g,h (Supporting Information movie M12) are higher
magnication images showing the local plasticity in GaN nanowires under compression. The nanowire is a single crystal with a
triangular cross-section and a diameter of 135 nm, and dislocations are present near the top surface (Figure 4a), possibly
introduced unintentionally during the alignment of the probe
with the nanowire. As the punch pushed on the nanowire, the
nanowire collapsed near the punch and nanowire contact surface,
forming a mushroom-shaped heavily plastically deformed zone
(Figure 4b,c). Detailed structural investigation indicated that the
lattice in the deformed zone was heavily distorted with small
nanocrystals broken o the nanowire (Figure 4d), and a high
density of dislocations were observed in the heavily deformed
zone (Figure 4e). The Burgers vector of the dislocation was
determined to be 1/3[-2110] and the slip plane is (01-1-1).
These results show that there is signicant surface plasticity
in the GaN nanowires, and that the plastic deformation is

LETTER

Figure 4. (see also Supporting Information movies M10, M11) Local

plastic deformation of GaN nanowires. The plastic deformation
occurred in the punch and nanowire contact area. (a c) and (g,h)
are two sets of time-lapsed TEM images showing local plasticity in GaN
nanowires. (a) The initial nanowire. (b) The nanowire under compression by a at-end punch on the right. The punch was advanced at
0.9 nm/s. (c) After the punch was released, the nanowire was deformed
permanently to a mushroom shape. (d) Heavy lattice distortion in the
deformed zone with a small nanocrystal was broken o the nanowire in
the heavily deformed zone. (e) A number of dislocations exist in the
plastically deformed zone. The dislocation Burgers vector is 1/3[-2110]
and the slip plane is (01-1-1). The lattice directions marked in (e) are
projections. (f) An HRTEM image of the pristine nanowires, and the
inset is a diraction pattern from the same nanowire. (g,h) (see also
Supporting Information movie M12) HRTEM images showing plastic
deformation on the nanowire and punch contact surface. The punch
was advanced at 0.1 nm/s. Note the heavy lattice distortion after
punching in (b).

accommodated by a dislocation nucleation and pile-up, leading

to the formation of a local plastic deformation zone.
Surface plasticity was also observed in other GaN nanowires.
Figure 4g,h and Supporting Information movie M12 are
HRTEM images showing surface plasticity. The lattice on the
nanowire surface was heavily distorted after punching, forming a
small plastically deformed zone near the contact area.
Despite the plastic behavior observed in the TEM, the forcedisplacement curves show a linear behavior before fracture,
indicating little global plasticity in the nanowire (Figure 5a). This
1620

dx.doi.org/10.1021/nl200002x |Nano Lett. 2011, 11, 16181622

Nano Letters

LETTER

Figure 5. Force-displacement plot (a) and fracture strength (b) of GaN nanowires with dierent diameters.

suggest that a small plastic event may not be reected in the forcedisplacement curve, which was also observed in metallic pillars.27
We found that the fracture strength of GaN nanowires
increases with a decreasing diameter (Figure 5b), and the reason
for this size eect may be related to the strain gradient induced by
bending during the compression. Because the aspect ratio in our
nanowire is usually larger than 5, bending or buckling usually
occurred in the compression experiments, which introduced a
strain gradient in the nanowire.
The local plasticity observed in GaN nanowire is dierent
from the global plasticity observed in metallic pillars in which
dislocations quickly run out of the surface of the pillars before
new dislocations are nucleated, leading to a dislocation starvation
state in the pillar.28 33 In GaN nanowires, the dislocations were
nucleated from the punch and nanowire contact surface, and
then piled in the contact area until a local plastic zone was
formed. The local plasticity was caused by surface roughness of
the nanowire and the punch. Similar local plasticity may also exist
in metallic materials.
It is interesting to compare our results with recent compression experiments of Si and GaAs nanopillars.34,35 In the case of Si
pillars, in uniaxial compression tests pillars having a diameter
exceeding a critical value develop cracks, whereas smaller pillars
show ductility comparable to that of metals.34 The critical
diameter is between 310 and 400 nm. This kind of brittle to
ductile transition under a critical diameter was not observed in
GaN nanowires. In the case of GaAs nanopillars, ductile plasticity
was observed in pillars with a diameter of 1 m.35 The diameter
of all the GaN investigated in our studies is less than 500 nm, yet
no global plasticity was observed in these nanowires. It is noted
that our nanowires are in their pristine state before the in situ
compression experiments, while the Si and GaAs pillars were
fabricated by focused ion beam (FIB), and it is well-known that
FIB can cause signicant damage to semiconductor materials. It
is possible that the FIB damage may contribute to the large
plasticity observed in the Si and GaAs pillars. However, this needs
further experimental conrmation. We do nd that the fracture
strength in GaN nanowires is similar to that observed in GaAs
pillars, implying that the strength of GaAs pillars was not aected
by FIB process.
The dislocation slip and lattice plane shearing-initiated fracture
agrees well with the recent simulation results in Si nanowires.36,37

On the basis of refs 36 and 37, a cleavage fracture is initiated

from a crack, while a shear fracture is initiated from a dislocation
slip. In our experiments, many of the fracture events were
initiated from a shear or a dislocation slip (Figures 2 and 3),
and the fracture surface exhibits brittle characteristic, e.g., it was
kinked on the (10-10) and (10-11) planes (Figure 3j). It is
unlikely that this kind of fracture surface was a result of
dislocation slip, because it requires cross-slip of dislocations in
dierent slip planes, which requires very high stress. All the
dislocations observed in our experiments slipped in a single slip
plane, and the stress to initiate dislocation slip is close to that
required to break the nanowire, that is, the fracture stress.
Our results indicate that dislocations can nucleate and propagate before fracture even though the nanowire failed in a brittle
manner. This again agrees with refs 36 and 37, which shows that
dislocation may be nucleated and propagates across the Si
nanowire before fracture occurs.
In summary, we observed signicant local plastic deformation
in GaN nanowires near the contact surface and also from the free
surface. Dislocation nucleation from the free surface leading to
the fracture of nanowires was directly observed. Furthermore, we
found that dislocation emission is the precursor of the fracture
event, even though the fracture may show brittle characteristic.
Local plasticity but no global plasticity was observed, indicating
the overall brittle nature of the nanowires. Very high stress, close
to the fracture strength of GaN nanowires which varies from 0.21
to 1.76 GPa, is required to activate dislocations. The results may
provide important guidance in rational design, fabrication, and
use of GaN nanowire based optoelectronic and electronic
devices.

ASSOCIATED CONTENT

bS

Supporting Information. Additional gures and movies

with descriptions. This material is available free of charge via the
Internet at http://pubs.acs.org.

AUTHOR INFORMATION
Corresponding Author

*E-mail: jhuang@sandia.gov.
1621

dx.doi.org/10.1021/nl200002x |Nano Lett. 2011, 11, 16181622

Nano Letters

ACKNOWLEDGMENT
We acknowledge support from Sandias Solid State Lighting
Science Energy Frontier Research Center, funded by DOE BES,
and the NNSAs Laboratory Directed Research and Development program. A part of this work was performed at the Center
for Integrated Nanotechnologies, a U.S. DOE, BES user facility.
Sandia National Laboratories is a multiprogram laboratory
managed and operated by Sandia Corporation, a wholly owned
subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energys National Nuclear Security Administration
under contract DE-AC04-94AL85000. H.Z. thanks Chinese
Scholarship Council for nancial support. S.X.M. acknowledges
NSF CMMI 08 010934 through University of Pittsburgh and
Sandia National Laboratories support.
REFERENCES
(1) Cheng, G. S.; Zhang, L. D.; Zhu, Y.; Fei, G. T.; Li, L.; Mo, C. M.;
Mao, Y. Q. Appl. Phys. Lett. 1999, 75 (16), 24552457.
(2) Duan, X. F.; Lieber, C. M. J. Am. Chem. Soc. 2000, 122 (1),
188189.
(3) Goldberger, J.; He, R. R.; Zhang, Y. F.; Lee, S. W.; Yan, H. Q.;
Choi, H. J.; Yang, P. D. Nature 2003, 422 (6932), 599602.
(4) Hersee, S. D.; Fairchild, M.; Rishinaramangalam, A. K.; Ferdous,
M. S.; Zhang, L.; Varangis, P. M.; Swartzentruber, B. S.; Talin, A. A.
Electron. Lett. 2009, 45 (1), 75U24.
(5) Hersee, S. D.; Sun, X. Y.; Wang, X. Nano Lett. 2006, 6 (8),
18081811.
(6) Huang, Y.; Duan, X. F.; Cui, Y.; Lieber, C. M. Nano Lett. 2002, 2
(2), 101104.
(7) Johnson, J. C.; Choi, H. J.; Knutsen, K. P.; Schaller, R. D.; Yang,
P. D.; Saykally, R. J. Nat. Mater. 2002, 1 (2), 106110.
(8) Kuykendall, T.; Pauzauskie, P. J.; Zhang, Y. F.; Goldberger, J.;
Sirbuly, D.; Denlinger, J.; Yang, P. D. Nat. Mater. 2004, 3 (8), 524528.
(9) Law, M.; Goldberger, J.; Yang, P. D. Annu. Rev. Mater. Res. 2004,
34, 83122.
(10) Talin, A. A.; Wang, G. T.; Lai, E.; Anderson, R. J. Appl. Phys.
Lett. 2008, No. 093105.
(11) Westover, T.; Jones, R.; Huang, J. Y.; Wang, G.; Lai, E.; Talin,
A. A. Nano Lett. 2009, 9 (1), 257263.
(12) Zhong, Z. H.; Qian, F.; Wang, D. L.; Lieber, C. M. Nano Lett.
2003, 3 (3), 343346.
(13) Heying, B.; Wu, X. H.; Keller, S.; Li, Y.; Kapolnek, D.; Keller,
B. P.; DenBaars, S. P.; Speck, J. S. Appl. Phys. Lett. 1996, 68 (5), 643645.
(14) Morkoc, H.; Strite, S.; Gao, G. B.; Lin, M. E.; Sverdlov, B.;
Burns, M. J. Appl. Phys. 1994, 76 (3), 13631398.
(15) Nakamura, S.; Mukai, T.; Senoh, M. Appl. Phys. Lett. 1994, 64
(13), 16871689.
(16) Nakamura, S.; Senoh, M.; Nagahama, S.; Iwasa, N.; Yamada, T.;
Matsushita, T.; Kiyoku, H.; Sugimoto, Y. Jpn. J. Appl. Phys., Part 2 1996,
35 (1B), L74L76.
(17) Strite, S.; Morkoc, H. J. Vac. Sci. Technol., B 1992, 10 (4),
12371266.
(18) Tsao, J. Y. IEEE Circuits Devices 2004, 20 (3), 2837.
(19) Wu, X. H.; Brown, L. M.; Kapolnek, D.; Keller, S.; Keller, B.;
DenBaars, S. P.; Speck, J. S. J. Appl. Phys. 1996, 80 (6), 32283237.
(20) Wong, E. W.; Sheehan, P. E.; Lieber, C. M. Science 1997, 277
(5334), 19711975.
(21) Arslan, I.; Talin, A. A.; Wang, G. T. J. Phys. Chem. C 2008, 112
(30), 1109311097.
(22) Wang, G. T.; Talin, A. A.; Werder, D. J.; Creighton, J. R.; Lai, E.;
Anderson, R. J.; Arslan, I. Nanotechnology 2006, 17 (23), 57735780.
(23) Nam, C. Y.; Jaroenapibal, P.; Tham, D.; Luzzi, D. E.; Evoy, S.;
Fischer, J. E. Nano Lett. 2006, 6 (2), 153158.
(24) Tham, D.; Nam, C. Y.; Fischer, J. E. Adv. Funct. Mater. 2006, 16
(9), 11971202.

LETTER

(25) Lancon, F.; Ye, J.; Caliste, D.; Radetic, T.; Minor, A. M.;
Dahmen, U. Nano Lett. 2010, 10 (2), 695700.
(26) Zheng, H.; Cao, A.; Weinberger, C. R.; Huang, J. Y.; Du, K.;
Wang, J. B.; Ma, Y. Y.; Xia, Y. N.; Mao, S. X. Nat. Commun. 2010, 1, 18.
(27) Minor, A. M.; Asif, S. A. S.; Shan, Z. W.; Stach, E. A.;
Cyrankowski, E.; Wyrobek, T. J.; Warren, O. L. Nat. Mater. 2006, 5
(9), 697702.
(28) Greer, J. R.; Nix, W. D. Phys. Rev. B 2006, 73 (24), No. 245410.
(29) Shan, Z. W.; Mishra, R. K.; Asif, S. A. S.; Warren, O. L.; Minor,
A. M. Nat. Mater. 2008, 7 (2), 115119.
(30) Uchic, M. D.; Dimiduk, D. M.; Florando, J. N.; Nix, W. D.
Science 2004, 305 (5686), 986989.
(31) Volkert, C. A.; Lilleodden, E. T. Philos. Mag. 2006, 86 (33 35),
55675579.
(32) Zhu, T.; Li, J. Prog. Mater. Sci. 2010, 55 (7), 710757.
(33) Greer, J. R.; Hosson, J. Prog. Mater. Sci. 2011in press.
(34) Michler, J.; Leifer, K.; Ballarini, R.; O uml stlund, F.; Hale,
L. M.; Gerberich, W. W.; Yuye, T.; Rzepiejewska-Malyska, K. Adv. Funct.
Mater. 2009, 243944.
(35) Michler, J.; Wasmer, K.; Meier, S.; Ostlund, F.; Leifer, K. Appl.
Phys. Lett. 2007, 90 (4), No. 431231.
(36) Kang, K.; Cai, W. Philos. Mag. 2007, 87, 21692189.
(37) Kang, K. W.; Cai, W. Int. J. Plast. 2010, 26 (9), 13871401.

1622

dx.doi.org/10.1021/nl200002x |Nano Lett. 2011, 11, 16181622