On direct-writing methods for electrically contacting GaAs and Ge nanowire devices

Article abstract of DOI:10.1063/1.3441404

The electronic transport and gating characteristics in GaAs and Ge nanowires (NWs) are altered significantly following either indirect or direct exposure to a focused Ga⁺ ion beam (FIB), such as that used to produce Pt electrical contacts to NWs. While these results challenge the assumptions made in some previously reported work relating to the electronic properties of semiconductor NWs using FIB-assisted production of contacts and/or their leads, local electron beam induced deposition is shown to be a reliable and facile route for producing robust electrical contacts to individual vapor phase-grown NWs in a manner that enables study of their actual carrier transport properties.

Full text of DOI:10.1063/1.3441404

On direct-writing methods for electrically contacting GaAs and Ge nanowire devices
Guannan Chen, Eric M. Gallo, Joan Burger, Bahram Nabet, Adriano Cola, Paola Prete, Nico Lovergine, and
,
Jonathan E. Spanier
Citation: Appl. Phys. Lett. 96, 223107 (2010); doi: 10.1063/1.3441404
View online: http://dx.doi.org/10.1063/1.3441404
View Table of Contents: http://aip.scitation.org/toc/apl/96/22
Published by the American Institute of Physics

APPLIED PHYSICS LETTERS 96, 223107 ͑2010͒
On direct-writing methods for electrically contacting GaAs and Ge
nanowire devices
Guannan Chen,1 Eric M. Gallo,2 Joan Burger,1 Bahram Nabet,2 Adriano Cola,3
Paola Prete,3 Nico Lovergine,4 and Jonathan E. Spanier1,a͒
1Department of Materials Science and Engineering, Drexel University, Philadelphia,
Pennsylvania 19104, USA
2Department of Electrical and Computer Engineering, Drexel University, Philadelphia,
Pennsylvania 19104, USA
3Istituto per la Microelettronica e Microsistemi (IMM), Consiglio Nazionale delle Richerche (CNR), Lecce
73100, Italy
4Department of Innovation Engineering, Universitá del Salento, Lecce 73100, Italy
͑Received 5 April 2010; accepted 9 May 2010; published online 2 June 2010͒
The electronic transport and gating characteristics in GaAs and Ge nanowires ͑NWs͒ are altered
significantly following either indirect or direct exposure to a focused Ga+ ion beam ͑FIB͒, such as
that used to produce Pt electrical contacts to NWs. While these results challenge the assumptions
made in some previously reported work relating to the electronic properties of semiconductor NWs
using FIB-assisted production of contacts and/or their leads, local electron beam induced deposition
is shown to be a reliable and facile route for producing robust electrical contacts to individual vapor
phase-grown NWs in a manner that enables study of their actual carrier transport properties. © 2010
American Institute of Physics. ͓doi:10.1063/1.3441404͔
Semiconductor nanowires ͑NWs͒ continue to attract sig-
nificant interest due to their potential as elements in high-
performance electronic and optoelectronic devices.1–3 Rapid
evaluation of their electronic transport properties is essential
to quantifying dopant composition, to achieving growth pro-
cess control, and to the rapid prototyping of latest devices. A
single-step method capable of electrically interfacing NWs
with contacts that are stable under thermal and voltage cy-
cling, and with precise control of their placement, can aid
significantly in developing reliable NW-based devices. In re-
cent years, local deposition of metal contacts on Si,4 Ge,5
SiGe,6 GaAs,7 GaN,8 ZnO,9 and chalcogenide10 NWs using a
focused Ga+ ion beam ͑FIB͒ has emerged as an attractive
alternative to multi-step processes involving electron-beam
or photolithographic patterning of a resist followed by physi-
cal vapor-deposited metal and lift-off. Despite numerous re-
ports of application of FIB to contacting individual NWs and
their subsequent analyses, remarkably little is known about
the effects of FIB exposures,11 both direct and proximally
indirect, on the electronic transport properties of semicon-
ductor NWs. Here, we quantify the effects of focused Ga+
beam exposures on sets of GaAs and Ge NWs that have been
contacted using several different methods as follows: elec-
tron beam lithography ͑EBL͒, electron beam induced depo-
sition ͑EBID͒ of Pt, and ion beam induced deposition ͑IBID͒
of Pt. FIB treatments, including exposures introduced at
some distance from each NW, are seen to modify signifi-
cantly the response of the NW to electrostatic gating.
grown and not intentionally doped GaAs NWs exhibiting
Ohmic contact behavior obtained without the use of a
FIB. Here, GaAs NWs were grown using colloidal Au-
nanoparticle-catalyzed metal-organic vapor phase epitaxy on
͑111͒B-terminated GaAs substrates using ͑CH3͒3Ga and
͑C4H9͒AsH2 and at a growth temperature of 400 °C.16 Ge
NWs were grown using Au film-catalyzed chemical vapor
deposition with GeH4 in H2 gas. The growth procedure in-
volved a 10 min H2 preannealing at 330 °C, followed by
flowing the precursor and carrier gases at 300 °C for 10 min.
The lengths of GaAs and Ge NWs vary in the 4–6 ␮m
interval; diameters of the GaAs and Ge NWs were between
70–100 nm and 40–90 nm, respectively. Following sonica-
tion in isopropyl alcohol to remove the NWs from their
growth substrates, NWs in solution were then transferred
onto p-type Si͑100͒ substrates, each possessing a 200-nm-
thick thermally-grown SiO2 gate oxide. Ti/Au bilayers ͑10
nm/150 nm͒ were previously deposited by electron-beam
evaporation, and patterned into contact pads using EBL.
Contacts to GaAs and to Ge NWs were each produced
using the following methods: ͑i͒ EBID of Pt; ͑ii͒ IBID of Pt,
and ͑iii͒ EBL patterning of metal deposited via electron beam
evaporation. While the EBID process is nondestructive, the
composition and electrical properties of EBID-produced Pt
are known to evolve with thickness,17 resulting in a change
from thermally activated to metallic conduction. Here, a
500-nm-thick, 250-nm-wide EBID-produced Pt lead ͑5 kV
and 56 pA͒ exhibited a resistivity of ͑8.3ϫ10−3 ⍀ cm͒ at
both 80 and 400 K ͑not shown͒. EBID was then used to
deposit 500-nm-thick Pt contacts onto 10 GaAs NWs and
onto 5 Ge NWs, and to connect the leads to Ti/Au contact
pads by employing ͑CH3͒Pt͑CpCH3͒. For comparison, IBID
of Pt ͑30 kV and 30 pA͒ was also carried out on a separate
set of GaAs and Ge NWs. Separately, EBL patterning of
metal film deposited via evaporation ͑electron-beam deposi-
tion of Pd/Ge/Pd trilayers, 10 nm/40 nm/10 nm͒ was used to
GaAs and Ge NW materials, grown without intentional
doping, were chosen for this study to provide two examples
of the impact of ion beam processing in the contacting of
NWs. Ge NWs were chosen since their transport properties
are well-characterized.12–15 In contrast, there have been no
reports published on the transport properties of vapor-phase
a͒
Electronic mail: spanier@drexel.edu.
0003-6951/2010/96͑22͒/223107/3/$30.00
96, 223107-1
© 2010 American Institute of Physics

223107-2
Chen et al.
Appl. Phys. Lett. 96, 223107 ͑2010͒
FIG. 2. I-V traces before ͑solid͒ and after ͑dashed͒ ion beam irradiation of
EBID-contacted ͑a͒ GaAs and ͑b͒ Ge NWs.
FIG. 1. ͑Color online͒ Measured source-drain I-V traces on ͑a͒ an EBL-
contacted GaAs NW for −20 VϽVg Ͻ20 V ͑2.5 V increment͒ and ͑b͒ an
EBID-contacted Ge NW ͑−10 VϽVg Ͻ10 V, 3 V increment͒. The inset in
͑a͒ is an SEM top-view ͑scale=1 ␮m͒ of the EBID contacts on the GaAs
NW.
and 2͑b͒ correspond to the I-V traces for the same EBID-
contacted GaAs and Ge NWs following ion beam exposure
as described above. In all NWs studied the current measured
following ion beam exposure was seen to increase by two to
three orders of magnitude, and the light response ͑in the
GaAs NW͒ and variation with gating ͑in both GaAs and Ge
NWs͒ could no longer be discerned.
contact other subsets of GaAs NWs. An optimization of the
film thickness, composition and a rapid thermal anneal
͑400 °C, 30 s͒ was performed to obtain Ohmic contact to the
GaAs NWs. In addition, identical sets of both IBID and
EBID contacts were also prepared, without NWs, to validate
that the measured currents were not due to a possible leakage
current path.
The measured transport in the Pt-contacted NWs can be
described in terms of an MSM structure.18 The current in-
crease by two to three orders of magnitude after ion-beam
irradiation on NWs is consistent with a relatively high-dose
ion beam implantation. We calculated the 30 kV Ga+ ion
stopping range by a transport range of ions in matter ͑TRIM͒
simulation.19 The results showed that the ion traveling dis-
tance peaked at 18.3 nm and 18.2 nm in GaAs and Ge, re-
spectively, and an appreciable concentration of ions extends
50 nm under the surface. Considering that the NWs have a
diameter range between 40 and 100 nm, the implantation can
be expected to reach well within each NW. As observed in
bulk GaAs, the Ga+ beam implantation can be expected to
generate electrically active defects, creating acceptor states
and resulting in an increase in carrier concentration in p-type
GaAs as a function of Ga+ dose.20 Thus, the increase in cur-
rent indicates that the acceptor states, through implantation
in the p-type NWs, causing the Fermi level to move closer to
the valence band. Changes in both NW conductivity and
metal-semiconductor contact properties21 can be responsible
for the observed substantial increase in conduction. The lack
of light response in postprocessed NWs is also expected as
the number of optically generated carriers is negligible com-
pared to dark conditions. Further, high doping of the NWs
can be expected to reduce the gating effect, consistent with
the data.
The electronic transport properties of the EBL and
EBID-contacted GaAs and Ge NW devices were collected
under vacuum ͑Ͻ10−5 torr͒, at selected temperatures be-
tween 80 and 400 K, and under a range of combinations of
source-drain and gate biases enabled by a Ti–Au film on the
back of each substrate. Response to optical illumination was
facilitated by an unfocused 150 W halogen lamp placed 5 cm
from the NWs. Three-terminal measurements were carried
out on EBL-contacted GaAs and Ge NWs using substrate
gating ͑Ϯ10, Ϯ20, and Ϯ50 V͒ and source-drain bias
͑−5 VϽVsdϽ+5 V͒ in order to ascertain the doping type
of the as-grown NWs, not previously reported in GaAs NWs
grown from the vapor phase without intentional doping. Rep-
resentative I-V data plotted at selected gate voltages for the
as-EBL-contacted GaAs and Ge NWs ͓Figs. 1͑a͒ and 1͑b͒,
respectively͔ demonstrate gating in each NW type, with
p-type character. Representative I-V traces from EBID-
contacted GaAs NWs exhibit a light/dark ratio of ϳ1.5–5
͑not shown͒. After the initial transport measurements the NW
devices were transferred immediately into a FIB ͑FEI Strata
235͒ for the ion-beam exposures. The electron beam and ion
beam ͑30 kV and 10–15 pA͒ were well-aligned in order to
precisely locate the ion beam. The final setting enabled the
incident ion-beam irradiation to be confined within the NW
area, resulting in a typical ion beam dose on the order of
2.2ϫ1015 cm−2 on each NW. An identical series of electrical
transport measurements were performed on the EBL- and
EBID-contacted GaAs and Ge NWs following their expo-
sures to the ion beam.
GaAs and Ge NWs, contacted using Pt via IBID, were
reported to have exhibited
a
linear current-voltage
response,7,11 indicative of Ohmic contacts. The Pt work func-
tion and electron affinity of GaAs suggest that deposited Pt,
even with annealing, cannot account for the observed re-
sponses. A focused 30 keV ion-beam has a dropletlike-
shaped interaction volume ͑ϳ6.5ϫ104 nm3͒ that is much
larger than that for a typical NW, and the implanted Ga+ are
expected to have a lateral Gaussian distribution Ͼ500 nm
from the beam spot.20 Besides Ga+ ion implantation, the
electronic transport in long ͑Ͼ1 ␮m͒ NWs is also influ-
enced by IBID lateral proximity effects. To demonstrate the
spatial proximity extent of ion beam exposure, the NW de-
vices were also placed in the scanning electron microscope
͑SEM͒-FIB while the ion beam ͑30 kV/30 pA͒ was incident
on the substrate for IBID of a Pt patch over an area of
Shown in Figs. 2͑a͒ and 2͑b͒ are representative I-V
traces of an EBID-contacted Ge NW and an EBID-contacted
GaAs NW prior to ͑solid line͒ the ion beam exposure, re-
spectively. These data were similar to those collected from
the other EBID-contacted GaAs and Ge NWs, where the
currents in each were in the picoampere and nanoampere
ranges, respectively. For Ge NWs of this diameter range, this
I-V characteristic and current at 3 V are consistent with pre-
vious reports on Ge NWs grown without intentional doping
and contacted using Ti/W.12 The dashed lines in Figs. 2͑a͒

223107-3
Chen et al.
Appl. Phys. Lett. 96, 223107 ͑2010͒
The authors acknowledge the Central Research Facility
at Drexel University and the assistance of Craig L. Johnson
and Ed Gasgall in using the SEM-FIB in this work. This
work was supported by the NSF under Grant Nos. DMR-
0907381, DMR-0722845, and ECCS-0702716, by the ARO
under Grant No. W911NF-08-1-0067, and with additional
support from Air Products Corporation.
1A. Javey, S. Nam, R. S. Friedman, H. Yan, and C. M. Lieber, Nano Lett.
7, 773 ͑2007͒.
2Y. Gu, E. S. Kwak, J. L. Lensch, J. E. Allen, T. W. Odom, and L. J.
Lauhon, Appl. Phys. Lett. 87, 043111 ͑2005͒.
FIG. 3. I-V traces for ͑a͒ GaAs and ͑b͒ Ge NWs collected before ͑solid͒ and
after ͑dashed͒ an IBID Pt patch was deposited 10 ␮m from each NW. Inset:
SEM image of the IBID Pt patch in relation to the GaAs NW. ͑scale: 5 ␮m͒.
3B. Z. Tian, X. L. Zheng, Y. Fang, N. Yu, G. Yu, J. Huang, T. J. Kempa,
and C. M. Lieber, Nature ͑London͒ 449, 885 ͑2007͒.
4A. I. Hochbaum, R. Chen, R. D. Delgado, W. Liang, E. C. Garnett, M.
Najarian, A. Majumdar, and P. D. Yang, Nature ͑London͒ 451, 163
͑2008͒.
1.5 ␮m2/dose of 3ϫ1017 cm−2 at 10 ␮m from the NW ͓in-
set of Fig. 3͑a͔͒. The currents were seen to increase by one to
two orders of magnitude, even when the ion beam exposure
was limited to some distance from the NW device ͓Figs. 3͑a͒
and 3͑b͔͒.22 This effect may be attributed to the surface depo-
sition of recoiled Ga+ ions that readily oxidize under the
relatively low vacuum environment of the FIB, suggesting
that connecting EBID contacts to probe and/or bonding con-
tact pads with IBID lines would also change the NW trans-
port properties.
In our comparison experiments, the transport measure-
ments on IBID contacted GaAs NW devices produced with
30 pA/30 kV exhibited a linear I-V response and a resistance
of 3.75ϫ107 ⍀ ͑80 nA at 3 V͒. However, the transport ex-
periments from EBL-contacted GaAs NWs taken from the
same batch also exhibited Ohmic contacts but showed that
these GaAs NWs in fact have a resistivity four orders of
magnitude higher. This enormous difference, taken with
other reports on the implantation of Ga+ and amorphization11
demonstrate that IBID of Pt significantly modifies the NW
transport properties.
Low-current 30 kV FIB irradiation on or near semicon-
ductor NWs is seen to alter their carrier transport. Orders of
magnitude increases in current are accompanied by dimin-
ished photocurrent and gating responses. For direct ion beam
exposure, these results can be ascribed primarily to Ga+ im-
plantation in the NWs, resulting in the additional doping of
GaAs NWs which can also alter contact properties. For indi-
rect exposure the results can be attributed to the recoiling of
Ga+ ions that readily oxide under the relatively low vacuum
environment of the FIB. Thus, EBID of Pt, rather than IBID,
produces contacts to semiconductor NWs in a manner that
permits study of their as-grown transport properties.
5T. Hanrath and B. A. Korgel, J. Phys. Chem. B 109, 5518 ͑2005͒.
6C. Qi, Y. Rangineni, G. Goncher, R. Solkani, K. Langworthy, and J. Jor-
dan, J. Nanosci. Nanotechnol. 8, 457 ͑2008͒.
7S. Thunich, L. Prechtel, D. Spirkoska, G. Abstreiter, A. Fontcuberta i
Morral, and A. W. Holleitner, Appl. Phys. Lett. 95, 083111 ͑2009͒.
8C. Y. Nam, J. Y. Kim, and J. E. Fischer, Appl. Phys. Lett. 86, 193112
͑2005͒.
9Y. Ma, Z. Zhang, L. Lu, A. Jin, and C. Gu, Nanotechnology 16, 746
͑2005͒.
10Y. Jung, C. Yang, S. Lee, and R. Agarwal, Nano Lett. 9, 2103 ͑2009͒.
11T. Hanrath and B. A. Korgel, Proceedings of the Institution of Mechanical
Engineers, Part N: Journal of Nanoengineering and Nanosystems ͑Profes-
sional Engineering Publishing, Westminster, London, 2004͒, Vol. 218, p.
25.
12D. Wang, Q. Wang, A. Javey, R. Tu, H. Dai, H. Kim, R. B. Wehrspohn, P.
C. McIntyre, T. Krishnamohan, and K. C. Saraswat, Appl. Phys. Lett. 83,
2432 ͑2003͒.
13S. Zhang, E. R. Hemesath, D. E. Perea, E. Wijaya, J. L. Lensch-Falk, and
L. J. Lauhon, Nano Lett. 9, 3268 ͑2009͒.
14W. Lu, J. Xiang, B. P. Timko, Y. Wu, and C. M. Lieber, Proc. Natl. Acad.
Sci. U.S.A. 102, 10046 ͑2005͒.
15B. Yu, X. H. Sun, G. A. Calebotta, G. R. Dholakia, and M. Meyyappan, J.
Cluster Sci. 17, 579 ͑2006͒.
16P. Paiano, P. Prete, E. Speiser, N. Lovergine, W. Richter, L. Tapfer, and A.
M. Mancini, J. Cryst. Growth 298, 620 ͑2007͒.
17J. M. De Teresa, R. Cordoba, A. Fernandez-Pacheco, O. Montero, P. Stri-
chovanec, and M. R. Ibarra, J. Nanomater. 2009, 936863͑2009͒.
18S. M. Sze, D. J. Coleman, and A. Loya, Solid-State Electron. 14, 1209
͑1971͒.
19J. F. Ziegler, http://www.srim.org.
20S. J. Brown, P. D. Rose, G. A. C. Jones, E. H. Linfield, and D. A. Ritchie,
Appl. Phys. Lett. 74, 576 ͑1999͒.
21S. M. Sze and K. K. Ng, Physics of Semiconductor Devices, 3rd ed.
͑Wiley, New York, 2007͒.
22Y. K. Park, T. Nagai, M. Takai, C. Lehrer, R. Frey, and H. Ryssel, Pro-
ceedings of the International Conference on Ion Implantation Technology,
͑IEEE, New York, 1998͒, Vol. 2, pp. 1137.