1
Supporting Inf ormation
Appendix A: Metallic Catalyst Removal Procedure
The
Si
microwires were grown on (111)
-
oriented
Si
substrates
by a chemical
-
vapor
-
deposition,
vapor
–
liquid
–
solid (VLS) growth process
.
13,20,48
For most samples,
the Cu VLS catalys
t was
r
emoved from the top and sidewalls of the microwires by chemical etching
.
After growth a
slow
cool down procedure
was performed,
diffusing the metallic catalyst
from the Si
. The etch
procedure for removing the catalyst and for etching off the SiO
2
that resu
lt
ed
from the catalyst
removal
wa
s:
10 s, 10% aq. HF
RCA2 etch: 20 min, DI water:HCl:H
2
O
2
(7:1:1, v/v/v)
10 s, 10% aq. HF
RCA2 etch: 20 min, DI water:HCl:H
2
O
2
(7:1:1, v/v/v)
10 s, 10% aq. HF
1 min
,
3
0 wt.% aq. KOH
10 s, 10% aq. HF
After each step, the micr
owires were rinsed thoroughly with DI water and were dried under a
stream of N
2
(g)
. An
RCA2 etch
was used to remove the metallic catalyst. The KOH(aq.)
removed
Cu impurities at the Si surface
. The
diluted
HF(aq.) was used to remove the na
tive
oxide as well
as any oxides that formed
during the catalyst removal process.
Electronic Supplementary Material (ESI) for Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2012
2
Appendix B: Microwire Methylation Procedure
F
ollow
ing
the methylation procedure that has been
reported for planar Si,
27
the
microwires were
first etched in 10% aq.
HF
to
remove the native oxi
de.
Samples were then placed
into a N
2
(g)
-
purged flushbox.
Using
a saturated solution of PCl
5
in chlorobenzene
,
t
he
micro
wires
we
re
chlorinated at 90 °C for 45 min. The
micro
wires were
then
rinsed and
placed in a 1
.0
M methyl
Grignard
(CH
3
MgBr)
solution
fo
r between 1 h and 1 day at 60 °C.
Appendix
C
: Preparation of Conducting Polymer Films
PEDOT:PSS was purchased from
Clevios
as an
aqueous
solution. To prepare 12 wt.%
PEDOT:PSS:Nafion, 1250 mL of Nafion solution
(from
Sigma
-
Aldrich
, 10 wt.% dispersion in
wa
ter) and
750 mL PEDOT:PSS w
ere
mixed
for 2
-
5 min on a vortex mixer. T
he solution was
then spin
coated
on the target substrate
at 2000 rpm for 20 s.
The films were allowed to dry for
6
–
8 h and then annealed for 1 h at 100
–
150
ºC
.
Appendix
D
: Formation of Mi
crowire / Polymer Junctions
After removal of the metallic catalyst (Figure
D.1
) and native oxide (appendix A), a corner of the
Si substrate
was scraped
using a razor blade,
to
remov
e
a small portion of the microwires
which
were then suspended in isopropano
l or acetonitrile.
Single
-
microwire measurements were
performed by deposition of this sol
ution (~10 μL) onto an insulating substrate (e.g. glass). Direct
contacts to the individual microwires were formed by use of tungsten probes in
a
probe station.
Electronic Supplementary Material (ESI) for Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2012
3
Figure
D
.1:
Schematic diagram o
f
a
microwire array. The average diameter of the microwires i
s 1.5 μm
and the array pitch size is approximately 7 μm. the length of the microwires is
9
0
±15
μm. The
VLS
grown catalyst
metallic caps are shown in yellow.
The W probes were etched using KOH(aq.), to remove tungsten native oxide and to improve the
qualit
y of the contacts.
An o
hmic contact to the conductive polymer was formed by sputteri
ng
Au directly on the polymer
.
The
I
-
V
profile for the Au contacts to the conducting polymer is
shown in Figure D.2.
Figure
D
.2:
Two
-
electrode measurements
of the c
urren
t
-
voltage response
between
two 32 nm thick
Au
pads and
intervening
conducting polymer films coated on
a
glass substrate.
Electronic Supplementary Material (ESI) for Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2012
4
Appendix E:
Use of
the Resistance vs. Tungsten Probe Spacing
to
Estimate the Doping Concentration of
Methyl
-
Terminated Si
Microwires
F
igure
E.1
depicts the resistance vs
.
probe separation data obtained
for
methyl
-
terminated
p
-
type
and n
-
type Si microwires.
E
ach data point
is
the
average of seven independent measurements in
which the probe
was completely disconnected from the microwire be
fore the contact
for the next
measurement was made. The resistance per unit length was constant
for all of the measurements,
having values of 0.50 kΩ
.
μm
-
1
for
the p
-
type and 0.18 kΩ
.
μm
-
1
for the n
-
type Si microwires. The
contact resistance was calculated b
y performing a linear fit to
the resistance versus probe
separation data, in conjunction with
the evaluation of the intercept of
the
plot (Figure
E.1
).
Figure
E
.1:
Contact resistance measurements for
methyl
-
terminated (CH
3
-
terminated)
p
-
and n
-
type Si
m
icrowires with
∼
1.5 μm diameter. Seven independent measurements
were performed for each selected
point across the length of the microwire.
These
measurements
were repeat
ed
with other microwires of
different doping type, different doping concentration, and originating
fro
m different fabrication batches.
Observation of the expected trends confirmed the validity of the measurement.
Electronic Supplementary Material (ESI) for Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2012
5
The calculated
contact resistance values
(i.e. y intercepts)
were reasonably similar for both
doping types of Si microwires
. Although the
data
analysis confirmed
that the contact resistance
was a negligible contribution to the
total measured resistance, the contact resistance was
still
subtracted
prior to calculating the resistivity of the
microwires and, in turn, prior
to estimation of
the dopin
g concentration of the microwires.
T
he
range of microwire
doping concentration
s was
calculated to be
10
17
–
10
18
cm
-
3
,
and n
o
detectable
variation in
the resistance measurements was
observed between different regions
of the
microwires or between different mi
crowires,
suggest
ing that
the p
-
and n
-
type doping concentration was uniform over the
length scales
considered
14,47
.
Appendix F: Comparative XPS Analysis of H
-
Terminated and
CH
3
-
Terminated Microwires vs. Time
Figure F.1 shows the initial XPS analysis resu
lts on H
-
terminated and CH
3
-
terminated n
-
Si
microwire samples. To collect these data, a large amount of microwires (760 μg) were scraped
from the growth substrate and the placed onto a Au
-
coated substrate. H
-
terminated Si microwires
were freshly etched and
then mounted in the XPS chamber (day 1). Methyl
-
terminated
microwires were also collected from a recently methylated batch of Si microwires. Both samples
were kept under lab conditions for 1 month, and then subjected to the XPS analysis again
. The
relativ
e amount of oxidized Si was determined from the ratio of the area fitted to the ~103 eV
(oxidized Si) and the ~99 eV (neutral Si) peaks.
Although the H
-
terminated wires were etched
before the measurements, the time interval between the etch and mounting th
e samples into the
XPS chamber was long enough for the native oxide to form. The oxide ratio for the case of H
-
terminated microwires remained roughly constant over the course of a month indicating the
presence of this saturated native oxide layer.
The unchanged ratio for
H
-
terminated samples
Electronic Supplementary Material (ESI) for Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2012
6
(
~24
%)
following a one
-
month period contrasts greatly with the changed ratio recorded for the
methyl
-
terminated samples (
from ~5% initially to ~
1
1
%
after one month
)
. This clearly indicated
that methylation impro
ved oxidative stability
.
T
he oxide observed for methyl
-
terminated Si samples was attributed to imperfect surface
methylation and/or to oxide formation at the backside of the microwires, where the wires had
been mechanically removed from the growth substrat
e.
Electronic Supplementary Material (ESI) for Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2012
7
(a)
(b)
Figure F.1:
XPS analyses for H
-
terminated and CH3
-
terminated n
-
Si microwire samples in the Si 2p
region; (a) at day 1 and (b) after one month. A higher rate of oxide growth was observed for H
-
terminated
samples compared to CH
3
-
term
inated samples. The oxide observed on methyl
-
terminated samples was
attributed to the backside of the microwires, where the samples had been removed from the growth
substrate.
Electronic Supplementary Material (ESI) for Energy & Environmental Science
This journal is © The Royal Society of Chemistry 2012