Bending of nanowire-flexible substrate assemblies integrated via direct synthesis methods

Article abstract of DOI:10.1002/pssa.201026722

Nanowires are successfully synthesized directly onto two types of flexible substrates, Kapton and silicone. Reduced temperature catalyst-assisted chemical vapor deposition techniques enable the direct integration. The mechanical properties of these assemblies are evaluated using bending tests where tensile or compressive strains are applied to the samples. Substrate-dependent nanowire properties, thin film formation, and failure of the integrated assemblies are observed. Image: Failure of a germanium nanowire (GeNW) array synthesized on a flexible substrate once subjected to compressive strains.

Full text of DOI:10.1002/pssa.201026722

Phys. Status Solidi A 208, No. 10, 2443–2449 (2011) / DOI 10.1002/pssa.201026722
a
p s s
www.pss-a.com
applications and materials science
Bending of nanowire–flexible
substrate assemblies integrated via
direct synthesis methods
Justin Rodgers, Julia Edel, Johan Rivera, and Ongi Englander*
FAMU-FSU College of Engineering, Florida State University, 2525 Pottsdamer St, Tallahassee, FL 32310, USA
Received 28 November 2010, revised 18 May 2011, accepted 20 May 2011
Published online 17 June 2011
Keywords bending tests, direct integration, flexible substrates, germanium nanowires
* Corresponding author: e-mail englander@eng.fsu.edu, Phone: þ1 850 410 6624, Fax: þ1 850 410 6337
Nanowires are successfully synthesized directly onto two types
of flexible substrates, Kapton and silicone. Reduced temper-
ature catalyst-assisted chemical vapor deposition techniques
enable the direct integration. The mechanical properties of
these assemblies are evaluated using bending tests where tensile
or compressive strains are applied to the samples. Substrate-
dependent nanowire properties, thin film formation, and failure
of the integrated assemblies are observed.
Image: Failure of a germanium nanowire (GeNW) array
synthesized on a flexible substrate once subjected to compres-
sive strains.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1 Introduction As interest in flexible electronics and thin flexible substrates is of great interest as such approaches
microfluidics has grown over the past decade, polymer and eliminate the need to transfer, pick and place nanostructures
plastic substrates have become commonplace in standard yet provide basic building blocks for nanoscale applications
microfabrication processes [1–3]. Such substrates offer and devices. We can envision successful integration
many desirable properties including flexibility, bendability, processes serving as the basis for nanowire-based electrodes
biocompatibility, transparency, and reduced weight. The for flexible fuel cells, the direct integration of nanowire arrays
integration of one-dimensional (1D) nanostructures with within microfluidic channels for enhanced flow control or as
flexible substrates to yield flexible and bendable composite thebasisformanipulatingtheresponseofnovelmetamaterials.
assemblies and devices has been widely demonstrated
The direct integration of germanium nanowires
[4–11]. In these examples, individual 1D nanostructures, (GeNWs) with polyimide substrates using reduced tempera-
1D nanostructure arrays, and 1D nanostructure-based films ture vapor–liquid–solid (VLS) techniques has been recently
have been drop cast, transferred by pressing, stamping, or reported [12, 13]. However, the mechanical properties and
rolling techniques onto a wide range of flexible substrates. failure mechanisms of VLS synthesized nanowires–flexible
The direct integration of bottom-up synthesis mechanisms substrate assemblies have yet to be evaluated. The integrity
for 1D nanostructure synthesis with a wide range of of nanowire-substrate assembly is important in validating the
substrates is often limited by the high temperature require- direct integration of bottom-up processes with flexible
ments of the catalyst-assisted chemical vapor deposition substrates.
mechanisms for 1D nanostructure synthesis. Integrated The properties of thin films and thin ribbons on stiff and
processes which enable nanowire synthesis directly onto flexible substrates have been studied in depth [14–21].
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
2444
J. Rodgers et al.: Bending of nanowire–flexible substrate assemblies
Characterization techniques such as bend tests, indentation, bottom-up reaction’s catalyst. The silicon substrates are not
MEMS-based testing, and bulge tests are commonly used to etched to remove the native oxide as we seek to maintain
evaluate mechanical properties [22–25]. Stress evolution identical sample preparation steps for all substrates.
and failure modes are often of interest and may be caused by
GeNW synthesis takes place over a range of process
thermal mismatches, processing stresses, residual stresses, parameters. A temperature of 285 8C and a pressure of
and surface indentation and scratching [14, 17]. With stiff 80 Torr using 40 sccm GeH4 and 100 sccm H2 is most often
films-flexible substrates combinations, bending causes the used, however multiple parameter combinations in the 270–
substrate to reversibly deform causing the stiff film to crack 285 8C temperature range and 7.5–100 Torr pressure range
or debond [16, 17, 26]. Significant efforts have been placed have successfully yielded nanowires. We also demonstrate
on characterizing electronic devices (thin film transistors and the integration of SiNWs with Kapton substrates. This
solar cells) fabricated onto flexible substrates and subjected synthesis process takes place at 400 8C and 40 Torr using
to bending tests [8, 10, 17, 22, 27, 28]. Findings suggest that 100 sccm SiH4 and 100 sccm H2. For SiNW synthesis, we
device performance can be limited by the applied strain but note that VLS reactions at 440–460 8C yield more desirable
operation of bent devices is possible and in fact, electron results however these temperatures are no longer compatible
mobilities may be enhanced under particular loading with the Kapton substrate. Following a 30-min long
conditions [17]. Further, these studies suggest that different synthesis process, the furnace is cooled and the samples
devices exhibit different failure modes [27]. Novel removed.
mechanics as well as the possibilities offered by flexible
substrate-based devices are truly exemplified in these on the synthesis results, for each synthesis process, we
studies. simultaneously place a silicon substrate, a Kapton substrate
To eliminate potential effects of process variations
Here, we report on the direct synthesis of GeNWs onto and a silicone substrate within the synthesis chamber (note,
Kapton and silicone substrates. We also demonstrate that silicone is not used for SiNW synthesis). The substrates are
silicon nanowire (SiNW) synthesis directly onto Kapton placed in close proximity to each other (ꢀ0.5 cm between
substrates is possible. We exclusively utilize bottom-up samples) and thus see identical flow conditions. In all cases,
synthesis techniques such that no post synthesis nanowire the substrates are kept flat during the synthesis process.
transfer or manipulation is needed. Next, we characterize the
processing results. More specifically, we evaluate fabrica-
tion results as a function of substrate, perform bending tests
to assess nanowire-substrate adhesion and delamination,
and note the role of thin film formation on the resulting
assemblies.
2.2 Characterization Mechanical properties of the
composite assemblies are examined using bending tests
where a bending moment is applied to the sample. The
samplesare subjected to either tensile or compressive strains.
Bending experiments are performed by bending the samples
so that the nanowire–substrate interface is strained. Samples
are bent to a predetermined radius of curvature (R) as
illustrated in Fig. 1 and then allowed to return to the initial
position (R ¼ 1). The bending experiments are performed at
discrete radii of curvature: 3, 1.5, 1 mm and a hard bend
(where the sample folded in half). In all cases, bending takes
place along the long axis (1 cm) of the sample. Tensile strain
on the assembly is realized by bending to the sample such
that nanowires-on-substrate configuration yields a concave
down configuration (Fig. 1). Similarly, compressive strain on
2 Experimental methods
2.1 Fabrication Nanowire synthesis follow standard
bottom-up processes and takes place in a low pressure
chemical vapor deposition (LPCVD) chamber. We select the
nanowire synthesis conditions based on robust processes
previously established in our lab using standard silicon
substrates. Kapton E and silicone are utilized as the flexible
substrates in this work and silicon (100) substrates serve as
the control substrate. The Kapton substrates are 50 mm-thick,
the silicone substrates are 1 mm-thick, and the silicon
substrates are 400 mm-thick. According to material specifi-
cations, Kapton E can sustain temperatures up to 400 8C
and silicone can sustain temperatures up to 315 8C. The
substrate selection is motivated by thermal requirements
of the nanowire synthesis process. Since to large extent
the nanowire synthesis process does not actively involve the
substrate (with the exception of the role of pristine single
crystalline substrates in controlling the growth direction) a
wide range of substrates may be accommodated by the
bottom-up synthesis process [29, 30].
The Kapton and silicone samples are cut to size and a
sample size of 1 cm by 0.5 cm is used in all experiments. All
substrates are first cleaned in acetone and de-ionized (DI)
water and dried with a nitrogen gun prior to catalyst
Figure 1 Experimentalsetupusedtoapplybendingstrainsontothe
samples. The bending experiments are performed at discrete radii of
application. Thin sputtered gold films are utilized as the curvature.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A 208, No. 10 (2011)
2445
Figure 2 Synthesis results of SiNWs at 400 8C on a Kapton sub-
strate.
the nanowires-on-substrate configuration yields a concave
up configuration. The samples are imaged following each
bending test using a scanning electron microscope.
3 Results and discussion
3.1 Fabrication results GeNW and SiNW are suc-
cessfully grown on stiff and flexible substrates under
identical synthesis conditions for each material. The flexible
substrates consist of Kapton and silicone substrates but the
SiNW synthesis is limited to the Kapton substrates due to
temperature incompatibilities with the silicone substrate.
The SiNWs are of uniform diameter throughout their length
(Fig. 2) and we note no observable differences between
nanowires synthesized on silicon and on Kapton substrate.
The GeNWs exhibit a characteristic taper [Fig. 3(a, b)]
[29, 31] on silicon and Kapton substrates while no tapering is
observed on the silicone substrate [Fig 3(c)]. With identical
gold thin films serving as the reaction catalyst, dense GeNW
arrays are obtained, while the SiNW arrays are fairly sparse
revealing the substrate. No nanowire alignment or order is
noted regardless of substrate.
Figure 3 Synthesis of GeNWs at 285 8C. (a) GeNWs on a silicon
substrate, (b) GeNWs on a Kapton substrate, and (c) GeNWs on a
silicone substrate. Differences in nanowire geometry (tapered vs.
cylindrical) are observed. Insets present optical images of the
samples.
Images of GeNWs on the various substrates are seen in
Fig. 3 and measurements of nanowire growth rates, tip measurement, is consistent with nanowire diameters on the
diameter, length, and nanowire density are presented in silicon and Kapton substrates. This nanowire array, however,
Table 1. There are no observable differences in nanowire is much denser than arrays synthesized on the other
properties (diameter, geometry, nanowire density, and substrates. Additionally, a reduced growth rate is observed
growth rate) for nanowires synthesized on the standard on the silicone substrates. Ongoing studies are designed to
silicon substrate and those synthesized onto the Kapton investigate the differences in growth properties. The silicone
substrate. The nanowires show a typical taper (thick base and surface is very rough and we have also observed changes to
narrow tip) on both substrates and as previously reported the surface morphology of the silicone substrate during or
[13, 29, 31, 32]. The tapering of GeNWs [Fig. 3(a, b)] is following the synthesis process. Specifically, large grain-like
speculated to occur as a result of uncatalyzed germanium sections appear on the surface. We suspect that the silicone
deposition onto the growing nanowire. As a result, the base
of the nanowire becomes significantly larger than the tip
resulting in a cone-shaped nanostructure. The average tip
Table 1 GeNW synthesis results.
diameter and nanowire density (number of nanowires/mm2)
GeNWs on GeNWs on GeNWs on
silicon
are almost identical for nanowires synthesized on silicon and
Kapton substrates. In both cases, growth rates of ꢀ0.1 mm/
min are recorded. Figure 3(c) shows GeNWs on silicone
substrates. In this case, we note somewhat different nanowire
growth characteristics. Specifically, nanowires with no
significant taper, and hence a uniform diameter from base
to tip, are observed. The average nanowire diameter is
Kapton
silicone
growth rate (mm/min) 0.11 ꢁ 0.02 0.1 ꢁ 0.04
0.05 ꢁ 0.01
58 ꢁ 12
1.36 ꢁ 0.23
10
tip diameter (nm)
lengtha (mm)
nanowire density
(no. of NW/mm2)
66 ꢁ 12
3.4 ꢁ 0.68
5
72 ꢁ 15
3.08 ꢁ 1.1
6
ꢀ58 nm, and considering the standard deviation in aFollowing a 30-min long synthesis process.
www.pss-a.com
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
2446
J. Rodgers et al.: Bending of nanowire–flexible substrate assemblies
chemistry, which is largely unknown, as well as surface
properties, could contribute to these observations.
Additionally, the reduced growth rate on silicone substrates
could be related to the absence of observable tapering.
Optically, GeNWs on silicon and Kapton substrates appear
black while GeNWs on the silicone substrates appear dark
brown (Fig. 3 insets). We considered the effect of nanowire
length (by reducing the nanowire synthesis time) and
substrate material on optical response but have found that
neither impacts the optical response. Instead, we have
observed that tapered nanowire arrays appear black while
untapered arrays appear brown.
3.2 Nanowire–substrate interface and thin film
formation Next, we investigate the nanowire–substrate
interface. To better assess interfacial properties, Focused ion
beam (FIB) capabilities are used to uncover the nanowire–
substrate interface. Figure 4 illustrates the nanowire–
substrate interface for the GeNW-Kapton assembly. The
formation of a thin film, while not consistent with the
bottom-up synthesis process, has been observed previously
in conjunction with the process [13, 32]. We suspect that the
thin film formation is due to uncatalyzed germanium
deposition onto the side walls of the growing GeNWs which
is prevalent at the synthesis temperature. Further, uncata-
lyzed germanium deposition not only extends the diameter of
the nanowires but also fills in regions between the densely
growing nanowires forming a germanium thin film. Thus, the
formation of the film is consistent with the observed GeNW
tapering. In order to avoid FIB related effects and ensure that
the apparent continuity of the film is not a result of this
processing, or more specifically, the melting of the
nanowires during the material removal process we reduce
the ion beam energy to 50 pA. We note that the thin film–
Kapton interface is smooth, continuous and free of voids and
unbonded regions. The Kapton substrate appears to accom-
modate the deposited film and nanowires well. This thin film
formation mechanism does leave behind voids, or incom-
plete filling, with increasing distance from the interface.
Figure 5 GeNWs on a silicone substrate with a thin film-like
structure protruding from within the dense nanowire array.
Eventually, individual nanowires are observed. The film is
ꢀ3–4 mm thick but voids in the film are visible within less
than 1 mm from the interface. The presence of the voids could
contribute to crack formation upon bending as discussed in
subsequent sections.
On the silicone samples, thin film formation appears
along the surface as seen in Fig. 5. While this behavior is not
well understood, we suspect that due to the large difference
in the coefficient of thermal expansion of silicone
(ꢀ100 ꢂ 10ꢃ6/8C) and germanium (5.8 ꢂ 10ꢃ6/8C), the thin
film-nanowire array experiences large compressive thermal
strains (ꢀ10 times larger than thermal strains experienced by
the Kapton-GeNW samples) which drive the contraction of
the sample upon cooling thus pushing the film away from the
surface. The film appears as isolated vertical regions where
the nanowire growth is oriented horizontally away from the
apparent vertical thin film formation. On these occasions
nanowire growth is limited to one edge of the film which is
consistent with a structure which is attached to the surface
during at least some portion of the synthesis process. Based
on the observed solid regions (Fig. 5), the film thickness is
estimated to be 0.5–1 mm thick. Unlike the thin film
formation on the Kapton substrates, this film formation does
not appear to be continuous throughout the surface of the
sample. Overall, this analysis points to differences in thin
film formation among the two substrates. The substrate along
with the resulting nanowire geometry (tapered vs. uniform
diameter) could have an effect on the thin film formation
mechanism and the growth rate.
3.3 Bending test results We have determined that
standardized adhesion test methods are not compatible with
the samples at hand. Standardized adhesion tests typically
rely on an interaction between a known adhesive and a test
surface prior to pull and delamination tests. The point contact
which would be formed between such an adhesive and
nanowire tips greatly reduces the contact area and thus the
effectiveness of such test. Therefore, tensile and compres-
sive bending tests are used to analyze failure mechanisms
and evaluate allowable strains for each of the composite
Figure 4 AcrosssectionoftheKapton–GeNWinterfacegenerated
using FIB capabilities. The Kapton substrate (bottom), the Kapton–
thin film interface, thegermanium thin filmas wellasthe emergence
of individual nanowires (inset) are seen.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A 208, No. 10 (2011)
2447
assemblies. We have studied multiple samples prepared
under identical synthesis conditions. On the Kapton
substrates distinct characteristics are observed consistent
with thin film failure modes. Under tensile strains, crack
behavior showcases channeling characteristics consistent
with tensile loads in thin films as illustrated in Fig. 6(a).
Generally, a crack begins at a flaw in the thin film and
propagates across the film thickness. Once it reaches the
surface it elongates laterally and spans the length of sample
[14]. While flaws have not been identified within the as-
synthesized nanowire arrays, voids noted within the thin film
(Fig. 4) could contribute to the observed cracking behavior.
Additionally, the proposed thin film formation mechanism is
not expected to yield high quality films. We note that cracks
originate along the axis of the bend, and additional, parallel
cracks are formed away from the central crack extending to
the edge of the sample (perpendicular to the bend axis). For
the samples studied, tension (or channeling) cracks measure
50–200 nm in width where for each crack, the crack width is
rather uniform (within 10 nm) throughout the length of the
crack. In compression, a debonding behavior is observed.
Debonding is most commonly observed under compressive
loading in films with poor adhesion to the surface. Cracks
propagate due to unbonded areas that can lead to buckling of
the material away from the substrate [Fig. 6(c)]. The
presence of parallel cracks, about the bend axis which
extend to the edge of the sample (in the direction
perpendicular to the bend axis), is observed in compression
as well. In tension, a higher crack density is noted and the
cracks are in close proximity to each other with an average
distance of 5 mm between cracks as illustrated in Fig. 6(a, b).
Compressive strains lead to debonding cracks with an
average spacing of ꢀ24 mm. While cracks extend to the edge
of the sample in the direction perpendicular to the bend axis,
the extent of the cracking along to bend axis is a function of
the applied strain. The cracking region extends over 100 mm
for all samples. The extent of the cracking region in
compression is significantly wider (four to nine times) than
that in tension and we note wide variations among the
samples in this case. A closer examination of the debonding
cracks reveals what appears to be the thin film formation
between the nanowires and the Kapton substrate as seen in
Fig. 6(d). Debonding cracks measure between 800 nm and
4 mm in width. When compared to channeling cracks,
debonding cracks are much wider and lack uniformity in
width as the crack width varies by as much as 3 mm along its
length. As a result, compressive strains appear to yield a
more significant perturbation to the nanowire array and the
sample as a whole. As seen in Fig. 6(c, d), we note the
displacement of the nanowires and the nanowire-thin film
combination away from the substrate all along the length of
each crack.
Figure 6 Channeling and debonding behavior of Kapton–GeNW
assemblies.(a)Parallelchannelsformastensilestrainisappliedand(b)
acloseupview of two channelsillustratingclean cracks(which appear
asdarklines)alongthesurface.(c)Parallelcracksformascompressive
strain is applied. A debonding behavior is noted as material appears to
be removedfrom the surface. (d) A close up viewof a debonding crack
illustratingnanowiresremovedfromthesurfaceaswellasthepresence
of a film and the Kapton substrate (darkest regions).
The response of the GeNWs-silicone substrate assem-
blies to strains via bending showcases different character- cracks are seen in Fig. 7. In compression, on the other hand,
istics. Overall, the damage to these samples is rather minimal the samples display no signs of damage or debonding; even
and highly localized. In tension, either no cracks or up to two when subjected to a hard bend, the surface appears to be
cracks appear with the onset of bending. Such channeling undamaged and undisturbed. We note that the nature of these
www.pss-a.com
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

a
p s s
2448
J. Rodgers et al.: Bending of nanowire–flexible substrate assemblies
where d is the thickness of the substrate and R is the radius of
curvature (applied during the bending test):
e ¼ d=2R:
(1)
This relation may be used assuming the thin film is much
thinner than its substrate.
Suo and coworkers define various regimes for thin films
on flexible substrates which are subjected to bending strains
[17, 27]. Specifically, a safe region (no evidence of failure), a
transition region and a failure region are all defined as a
function of percent strain. We use this approach to map our
findings. This analysis includes an observation of four
samples in each group, subjected to a sequence of bending
tests at R ¼ 3, 1.5, 1 mm and a hard bend as well as individual
bending tests at each radius. The samples are imaged
immediately following each bend test. In this manner we can
assess the impact of ‘‘pre-existing conditions’’ on sub-
sequent cracking behavior. With the exception of tensile
bending tests on silicone samples, we note consistent
behavior within each group of samples and no changes in
the onset of failure due to prior bending. The Kapton samples
exhibit a safe region in both tension and compression up to a
radius of curvature of 1.5 mm or 1.7% strain. Within the safe
region there is no evidence of cracking or damage of the
Figure 7 Silicone substrate–GeNWs sample subjected to tensile
strain. (a) Two channeling cracks (within dotted boxes) form upon
the application of tensile strain. The cracks are distinguishable
despite of the sample’s rough morphology. (b) A close up view of
a channeling crack.
samples and in particular the surface properties of the nanowire–thin film–substrate structure. Beyond this safe
silicone (Fig. 7) make the analysis of cracking behavior more region the samples behave differently depending on the type
complex that those of the Kapton based assemblies. We of strain. A transition region is observed for samples placed
suspect that the nature and final location and orientation of in compression within which small cracks appear and a low
the thin film could affect the samples’ response to bending degree of debonding is observed. This period is followed by
strains. More specifically, the absence of a continuous thin pronounced debonding at a radius of curvature of 1 mm or
film could prevent traditional failure modes observed on 2.5% strain. We suspect that once the debonding region
Kapton. Table 2 summarizes the observed results for three becomes sufficiently large, film buckling and then fracture
sample sets (in each group) evaluated at the largest applied takes place [17]. Thewidth of the debonding cracks increases
strain values (hard bend). Intermediate strain effects are with increasing bending strains. At low strains (transition
addressed in subsequent discussion.
region), the cracks are initiated but hardly observable. The
cracks become wider and displace much material from the
surface with increasing strain. Samples subjected to tensile
strains exhibit channeling at radii of curvature greater than
1.5 mm or 1.7% strain. The areal extent of the channels and
the frequency of the channels increase with increasing strain
however, a transition zone prior to failure is not observed.
In comparison, the silicone samples exhibit a different
response to the bending tests. Keeping the experimental radii
of curvature consistent for the silicone samples and noting
that the silicone substrate is 20 times thicker than the Kapton
substrate, the silicone samples are subjected to strains
20 times larger. However, the lack of observable cracking,
places silicone samples, subjected to compressive strains,
within the safe region at all applied strains tested. In tension,
we observe two distinct behaviors: either no failure or an
immediate failure with the onset of strain with no transition
or safe region. Ongoing studies are designed to better
understand these observations but the current analysis
suggests that the silicone-based assemblies are more tolerant
of bending strains.
3.4 Quantifying strain Film strain (e) as a function
of radius of curvature may be determined using Eq. (1) [26],
Table 2 Summary of cracking behavior under applied strains.
Kapton
substrate
silicone
substrate
tensile strain
distance between cracks (mm)
extent of channeling region
(parallel to bend axis) (mm)
number of cracks
5 ꢁ 2a
none-2500b
none-2500b
145 ꢁ 30a
18–36b
0–2b
compressive strain
distance between cracks (mm)
extent of debonding region
(parallel to bend axis) (mm)
number of cracks
24 ꢁ 19a
none
none
780 ꢁ 290a
18–40b
0
aAverage value plus/minus standard deviation (measurement accuracy
ꢁ0.2 mm); brange of observed values.
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.pss-a.com

Original
Paper
Phys. Status Solidi A 208, No. 10 (2011)
2449
4 Conclusions We demonstrate the bottom-up syn-
thesis of GeNWs directly onto two types of flexible
substrates, Kapton and silicone, yielding integrated assem-
blies via a single step process. We also observe that it is
possible to synthesize SiNWs directly onto Kapton sub-
strates, however, the process renders the Kapton brittle and
the assembly fractures easily. This approach is unique as it
eliminates complex transfer protocols which are often
required to move nanostructures from a stiff donor substrate
onto a flexible substrate.
[7] T. Y. Tsai, C. Y. Lee, N. H. Tai, and W. H. Tuan, Appl. Phys.
Lett. 95, 013107 (2009).
[8] Z. Fan, H. Razavi, J.-W. Do, A. Moriwaki, O. Ergen, Y.-L.
Chueh, P. W. Leu, J. C. Ho, T. Takahashi, L. A. Reichertz,
S. Neale, K. Yu, M. Wu, J. W. Ager, and A. Javey, Nature
Mater. 8, 648 (2009).
[9] H. Ko, Z. X. Zhang, J. C. Ho, K. Takei, R. Kapadia, Y. L.
Chueh, W. Z. Cao, B. A. Cruden, and A. Javey, Small 6, 22
(2010).
[10] K. Heo, J. W. Park, J. E. Yang, J. Koh, J. H. Kwon, Y. M.
Jhon, M. Kim, M. H. Jo, and S. Hong, Nanotechnology 21, 6
(2010).
We find that the resulting nanowire properties, the
fabricated assemblies’ response to applied strains and thin [11] Y. Qi, N. T. Jafferis, K. Lyons, C. M. Lee, H. Ahmad, and
M. C. McAlpine, Nano Lett. 10, 524 (2010).
[12] K. Kang, D. A. Kim, H. S. Lee, C. J. Kim, J. E. Yang, and
M. H. Jo, Adv. Mater. 20, 4684 (2008).
[13] Y. L. Chueh, Z. Y. Fan, K. Takei, H. Ko, R. Kapadia, A. A.
Rathore, N. Miller, K. Yu, M. Wu, E. E. Haller, and A. Javey,
Nano Lett. 10, 520 (2010).
film formation attributes are dependent on the type of flexible
substrate used. Nanowire properties are evaluated in terms of
nanowire geometry, density, and growth rate and show a
notable difference when synthesized on Kapton and silicone
substrates under identical synthesis conditions. The mech-
anical properties of the GeNW–flexible substrate assemblies
are evaluated using bending tests and differences in the
assemblies’ response are observed. The Kapton–GeNW
samples illustrate a resistance to failure at up to 2.5% strain in
compression while the silicone-GeNW samples show
resistance to failure at all levels of applied compressive
strain. In tension, the Kapton–GeNW samples are able to
withstand 1.7% strain before cracking is observed. The
silicone–GeNW samples, on the other hand, may show crack
formation when subjected to the smallest amount of tensile
strain. With both substrates, the formation of germanium thin
films is observed; however, the nature of these films varies
with substrate and is suspected to affect bending results.
This study contributes to developing a better understand
of the interaction between nanoscale components and
compliant substrates. The robustness of the resulting
assemblies, their potential failure modes and the overall
feasibility of the integration process are all of great
importance. Ultimately such studies could provide general
and application-specific substrate selection guidelines.
Ongoing investigations are designed to better characterize
and understand the role of the substrate and its surface
properties on the resulting nanowires, the formation of the
thin film, and the fabricated assemblies.
[14] K. H. J. Buschow, R. W. Cahn, M. C. Flemings, B. Ilschner,
`
E. J. Kramer, S. Mahajan, and P. Veyssiere (eds.), Encyclo-
pedia of Materials: Science and Technology (Elsevier
Science, Amsterdam, 2001), p. 3290.
[15] L. B. Freund and S. Suresh, Thin Film Materials: Stress,
Defect Formation, and Surface Evolution (Cambridge
University Press, New York, 2003).
[16] T. Li, Z. Huang, Z. Suo, S. P. Lacour, and S. Wagner, Appl.
Phys. Lett. 85, 3435 (2004).
[17] G. P. Crawford (ed.), Flexible Flat Panel Displays (Wiley,
Chichester, 2005), p. 263.
[18] W. W. Gerberich and M. J. Cordill, Rep. Prog. Phys. 69, 2157
(2006).
[19] D. Passeri, A. Bettucci, and M. Rossi, Anal. Bioanal. Chem.
396, 2769 (2010).
[20] S.-I. Park, J.-H. Ahn, X. Feng, S. Wang, Y. Huang, and
J. A. Rogers, Adv. Funct. Mater. 18, 2673 (2008).
[21] J. L. Ni, X. F. Zhu, Z. L. Pei, J. Gong, C. Sun, and P. Zhang,
J. Phys. D 42, 175404 (2009).
[22] H. Gleskova, S. Wagner, and Z. Suo, Appl. Phys. Lett. 75,
3011 (1999).
[23] B. E. Alaca, J. C. Selby, M. T. A. Saif, and H. Sehitoglu, Rev.
Sci. Instrum. 73, 2963 (2002).
[24] T. Sekitani, Y. Kato, S. Iba, H. Shinaoka, T. Someya,
T. Sakurai, and S. Takagi, Appl. Phys. Lett. 86,
(2005).
[25] X. Chen, B. L. Kirsch, R. Senter, S. H. Tolbert, and V. Gupta,
Mech. Mater. 41, 839 (2009).
[26] J. Lewis, Mater. Today 9, 38 (2006).
[27] H. Gleskova, I.-C. Cheng, S. Wagner, J. C. Sturm, and Z. Suo,
Sol. Energy 80, 687 (2006).
Acknowledgements Thisworkwascompletedwithsupport
from Florida State University’s start up funds. The authors thank
Dr. F. Kametani for assistance with the FIB.
[28] M. Lee, J. Koo, E. A. Chung, D. Y. Jeong, Y. S. Koo, and
S. Kim, Nanotechnology 20, 6 (2009).
References
[1] S. Roy, J. Phys. D, Appl. Phys. 40, R413 (2007).
[2] J. Song, H. Jiang, Y. Huang, and J. A. Rogers, J. Vac. Sci.
Technol. A 27, 1107 (2009).
[3] A. C. Siegel, S. T. Phillips, M. D. Dickey, N. Lu, Z. Suo, and
G. M. Whitesides, Adv. Funct. Mater. 20, 28 (2010).
[4] B. Weintraub, Y. L. Deng, and Z. L. Wang, J. Phys. Chem. C
111, 10162 (2007).
[5] M. C. McAlpine, H. Ahmad, D. Wang, and J. R. Heath,
Nature Mater. 6, 379 (2007).
[6] S. Tawfick, K. O’Brien, and A. J. Hart, Small 5, 2467 (2009).
[29] H. Adhikari, A. F. Marshall, C. E. D. Chidsey, and P. C.
McIntyre, Nano Lett. 6, 318 (2006).
[30] H. Jagannathan, M. Deal, Y. Nishi, J. Woodruff,
C. Chidsey, and P. C. McIntyre, J. Appl. Phys. 100,
024318 (2006).
[31] D. E. Perea, E. R. Hernesath, E. J. Schwalbach, J. L. Lensch-
Falk, P. W. Voorhees, and L. J. Lauhon, Nature Nanotechnol.
4, 315 (2009).
[32] C.-B. Jin, J.-E. Yang, and M.-H. Jo, Appl. Phys. Lett. 88,
193105 (2006).
www.pss-a.com
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim