of 8
1
Supporting Information
Appendix A: Metallic Catalyst Removal Procedure
The silicon microwires were received as arrays grown on <111> silicon substra
tes.
Different samples were grown using both Au and Cu catalysts. The microwires
were p-type
doped with boron (using BCl
3
) as the dopant. All the samples had the metallic catalyst on
the end of each microwire and also have some minor amounts deposited on the sides. A
slow cool down procedure has been done after growth so that the metallic catalyst
diffuses
out readily from the Si. The etch procedure for removing the catalyst and for etc
hing off the
SiO
2
that results from the catalyst removal is as follow:
10 s, 10% aq. HF
30 min. 30 wt.% aq. FeCl
3
10 s, 10% aq. HF
1 min 20 wt.% aq. KOH
10 s, 10% aq. HF
After each step, the microwires were rinsed thoroughly with DI water and dri
ed under a
stream of N
2
. FeCl
3
was used to remove the metallic catalyst. The KOH was used to
remove any leftover FeCl
3
. Finally, buffered HF was used to remove the native oxide and
any oxide formed during the catalyst removal process. XPS analysis confir
med the catalyst
removal from the microwires after etching process (Figure A.1).
2
Figure A.1:
XPS measurements (a) before and (b) after the etch process confirm met
allic
catalyst removal from copper-based microwires.
3
Appendix B: Conductive Polymer Film Preparation
B.1 Polymer Solutions
Two different conductive polymer solutions were prepared for the measurements. Ea
ch
procedure includes the polymer solution preparation and spin coating the solution on the
target substrates. The solution was coated at 2000 rpm for 20 seconds. The film prepara
tion
for each polymers followed by a rinsing process to remove the residual PSS or PMA
.
Microwires were aligned before the rinsing process in all of the cases
.
B.1.1 PEDOT:PMA
PEDOT:PMA solution was prepared using acetonitrile (CH
3
CN) as the solvent. The
procedure is as follows:
PMA solution: CH
3
CN (1 mL) + PMA (1.09 g)
EDOT solution: CH
3
CN (1 mL) + EDOT (42.6
μ
L)
These solutions were mixed and spin coated on the target substrates to form the final
PEDOT:PMA membranes. For the rinsing process, the films were placed in
dichloromethane (CH
2
Cl
2
) solution with a small amount of acetonitrile (3~4 mL) for ~ 30
min.
B.1.2 PEDOT:PSS:Nafion
PEDOT:PSS was purchased from Sigma Aldrich as a solution. In order to prepar
e 12 wt.%
PEDOT:PSS:Nafion, 1250 mL of Nafion solution was mixed with 750 mL PEDOT:PSS
and this solution was coated on the glass substrate. For the rinsing process, the fil
ms were
placed in acetonitrile for ~ 30 min. following the rinsing process, the PEDOT:PSS:
Nafion
films were baked under vacuum at 100ºC for one hour.
4
Appendix C: Microwire / Polymer Junction Formation
Following removal of the metallic catalyst (Figure C.1), and native oxide re
moval process
(appendix A) a solution of microwires was prepared for coating on the target substr
ates by
scraping a corner of the substrate using a razor blade and removing a smaller por
tion of the
microwires. This is a desirable approach for single microwire measureme
nts as the final
solution was more diluted and, when coated on the substrate resulted in completely
separated microwires. Single microwire measurements were performe
d by deposition of
this solution (~10
μ
L) onto an insulator substrate (e.g. glass). Direct contacts to the
individual microwires were formed using tungsten probes and InGa in the probe stat
ion.
Figure C.1:
Schematic diagram on the microwire arrays as received. The average di
ameter
of the microwires is 1.5 μ m and the array pitch size is approximately 7 μ m. the aver
age
length of the microwires is 100 μ m. The metallic caps are shown in yellow.
5
Tungsten probes are also etched using KOH before the measurements to remove the
tungsten native oxide and improve the quality of the contacts. Using tungsten probes
provides the ability to mechanically manipulate the microwires and make conta
cts to the
individual wires as demonstrated in Figure C.2.
Figure C.2
: Mechanical manipulation of silicon microwires using tungsten probes.
The probes were placed on both ends of the microwire and the current passing through the
microwire was measured for a range of applied voltage using an
Agilent 4155c
semiconductor parameter analyzer.
Conductive polymer solutions, required for the microwire-polymer junction investiga
tions,
were prepared according to the established procedures (Appendix B). Ohmic conta
ct to the
conductive polymer was formed by sputtering gold directly on the polymer. The mic
rowire
solution (~10
μ
L) was deposited directly on the glass substrate after removing the paraf
fin
tape. The next step was to align microwires at the border between the microwire
and
25 μm
25 μm
25 μm
6
conductive polymer in order to make electrical contact between the two elements. T
he
schematic diagram of these measurements is shown in Figure C.3. Three important
resistances in the system have been labeled in Figure C.3(c) as R
polymer
, R
contact
and R
wire
which were going to be characterized.
Figure C.3:
Schematic diagrams for (a) single microwire measurements and (b)
microwire/polymer junction characterization. The diameter of tungsten (W)
probes is
7
approximately 1 μ m. (c) an optical micrograph of a microwire aligned at t
he polymer/glass
border with important resistances.
Appendix D: Quantifying the Applied Mechanical Force on the Single Si
licon
Microwires
The required pressure to induce a phase transition in silicon is ~112,000 kg.cm
-2
which is
equal to ~11 mN.μ m
-2
. The actual required pressure inducing such a transition in silicon
microwires might be smaller considering the fact that the reported pressur
e in [19] was
exerted on silicon indirectly through an interfacial layer (aluminum). Consider
ing the
weight of the probes (~ 0.2 g) and the contact area close to 1 μ m
2
, based on the probe
diameter, there is approximately 2 mN.μ m
-2
applied to the microwire from the probes
having no additional force applied to the probe.
However, to accurately determine the pressure applied by the probe to the Si
microwire
during the measurements, the force of the probe holder setup was measured using a
balance. Placing the probe holder on a labjack adjacent to a tared balance the probe was
lowered until the balance recorded a weight. The probe was then retracted until
the balance
returned to zero, this was taken as just touching the balance. From this point the probe was
lowered a number of turns until a constant weight was achieved. After 3/4 of a turn t
he
weight stabilized at approximately 3.80 g. This is equivalent to 37.3 mN and given the
probe’s radius the applied pressure from this setup is 12 mN.μ m
-2
which is enough to pass
the required transition threshold.
We were able to observe the phase transition in the contacts by backing off the press
ure on
the microwires. Figure D.1(a) shows a measured I-V profile for a highly dope
d microwire
with one probe fixed at one end of the wire with ~ 37 mN of force applied from the probe,
and the second probe loosely connected to the other end with no additional force applied.
Figure D.1(b) shows the change in the I-V profile as ~37 mN force was appl
ied to the
8
second probe. This was a reversible procedure which repeated by increasing or decr
easing
the force on the probes.
Figure D.1:
Local phase transition as a result of pressure at the contact area can cause
a
change in the contact behavior. A highly doped 100 μ m long microwire was investigate
d
with the first probe fixed at one end of the wire (with ~37 mN of applied force) while
the
second probe (a) touched the other end with almost no applied force to the probe (b)
touched the other end with the same amount of force applied.