Integration of Electrocatalysts with Silicon Microcone Arrays
for Minimization of Optical and Overpotential Losses during
Sunlight
-
Driven Hydrogen Evolution
Sisir Yalamanchili
1,3
$
, Paul A. Kempler
2,3
$
,
Kimberly M. Papadantonakis
2
,3
,
Harry A.
Atwater
1,3,4
*
, Nathan S. Lewis
2
-
5
*
1
Division of Engineering and Applied Sciences, California Institute of Technology, Pasadena, CA 91125
2
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA
91125
3
The Joint Center for
Artificial Photosynthesis, California Institute of Technology, Pasadena, CA 91125
4
Kavli Nanoscience Institute, California Institute of Technology, Pasadena, CA 91125
5
Beckman Institute, California Institute of Technology, Pasadena, CA 91125
*Correspondin
g Authors:
haa@caltech.edu
,
nslewis@its.caltech.edu
$
Equal Contributions
Electronic
Supplementary
Material
(ESI)
for
Sustainable
Energy
&
Fuels.
This
journal
is
©
The
Royal
Society
of
Chemistry
2019
Figure S1:
Process flow diagrams
for
two methods of
fabricating
integrated solar
fuels device
s
.
Scalable integration of a wide band gap top cell with a monolithic Si bottom cell could be
achieved via
:
(a)
patterning a
crystalline III
-
V semiconductor
substrate with Au seeds or Cr
masks for vapor
-
liquid
-
solid growth or dry etching, respectively;
(b)
fabricating
a
m
icro
-
or
nanocone array
via vapor
-
liquid
-
solid growth
followed by chemical etching
or dry etching
.
1, 2
(c)
embedding the micro
-
or nanocone array
in
a flexible polymer, and removing i
t
from the
substrate for transfer
;
1
(d)
metallizing the array
with catalysts and ohmic contacts on the top
-
and
bottom
-
facing sides, respectively, for integration with a Si bottom cell.
F
abrication
of
a tandem,
core
-
shell microcone array embedded in an ion conducting membrane could be achieved via: (e)
patt
erning
a Si substrate with Cu
seeds
or Al
2
O
3
masks; (f) fabricating a microcone array as in
(b), and coating with an ohmic contact such as indium
-
tin
-
oxide; (g) depositing a wide band gap
semiconductor via
metal
-
organic chemical vapor deposition
,
3
thermochemical conversion from
solution,
4
or electrochemical growth
,
5
on the surface of the microcone array
,
leading to a core
-
shell architecture
;
(
d
)
infilling the
a
rrays
with
an
ion
exchange membrane
and
peel
ing
from the
substrate.
The substrate can be re
-
patterned
to reduce material
demands
.
Figure S2
:
Schematic diagram of the cell used for
photoelectrochemical hydrogen evolution
testing of devices.
Illumination was directed into the base of the cell onto a bottom facing
photoelectrode.
Fig
ure S3
:
Initial p
erformance
(scan 1)
of
the best
-
performing
n
+
p
-
Si
/Pt
μ
-
cone
photocathode
s
with 8 nm
or
16 nm thick Pt l
oad
ed
on the tips
of the
μ
−
cones
.
The
photocathodes
were tested in
H
2
-
saturated 0.50
M H
2
SO
4
(
aq)
under
100 mW cm
-
2
of
simulated AM1.5
illumination
, with
rapid
stirring of the electrolyte to remove bubbles
from the electrode surface
.
0
0.1
0.2
0.3
0.4
0.5
Potential / V vs. RHE
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
Current Density / mA cm
-2
Champion, 8 nm Pt
Champion, 16 nm Pt
Champion, 4 nm Ti + 16 nm Pt
Figure S
4
:
Stability of n
+
p
-
Si/Pt photocathodes operated in contact with H
2
-
saturated 0.5 M
H
2
SO
4
(aq)
while under
100 mW cm
-
2
of
simulated AM1.5 illumination.
For
n
+
p
-
Si/Pt
μ
−
pyramid
photocathodes with
(a)
4 nm of Pt,
a decrease in fill factor
(red trace)
was observed after the first
potential scan
(blue trace)
. For n
+
p
-
Si/Pt
μ
−
cone
s, devices with (b
) 4 nm of Pt on the
μ
−
cone
tips showed a decrease in fill factor with successiv
e scans
(yellow traces)
, whereas devices
with
(c
) 16 nm of Pt did not show a
n
improvement
in
J
ph
and no decay in fill factor
(blue traces) after
the first scan (purple trace).
Extended
Methods
Stability testing:
Extended stability testing of
n
+
p
-
Si/Ti/Pt and p
-
Si/Co
-
P
μ
-
cone arrays was
performed under nominally identical conditions
as those used
for photoelectrochemical testing,
but with a Pt mesh electrode counter electrode
behind a Nafion membrane (Fuel Cell Store)
. H
2
was bubbled through the electrolyte for the durati
on of the stability tests to maintain a dissolved
concentration of H
2
in equilibrium with 1 atmosphere of H
2
(g)
.
Electrochemical impedance spectroscopy:
Electrochemical
impedance spectra were measured
under
nominally identical conditions
to
those used in
photoelectrochemical testing.
A
sinusoidal
perturbation of amplitude 25 mV was applied to the potential of the working electrode and the
impedance response was recorded over a frequency range from 7 MHz to 20 Hz.
Figure S5:
Stability testing of n
+
p
-
Si/Pt
μ
-
cone array photocathodes in H
2
purged 0.50 M
H
2
SO
4
(aq)
under 1
-
Sun illumination. (a) Chronopotentiometry at
-
10 mA cm
-
2
for 12 h. Cyclic
voltammograms
were periodically recorded
at 50 mV s
-
1
to monitor the
J
-
E
behavior of the
device. (b)
J
-
E
behavior of a representative photocathode before testing (black line), after 12
h
of
chronopotentiometry at
-
10 mA cm
-
2
(red line), and after 10
s
in aqua regia and rinsing with
copious deionized water (red circles). (c)
Nyquist plot for an n
+
p
-
Si/Pt
μ
-
con
e array
photocathode after testing,
recorded
under 1
-
Sun illumination, in 0.5 M H
2
SO
4
(aq),
over a
fre
quency range of 2 MHz to 20 Hz,
at +260 mV (orange) and +340 mV (blue) vs RHE.
Stability testing: Results and Discussion
A comparison of the
J
-
E
behavior of the device before and after 12
h
of chronopotentiometry
(Figure S5
a)
revealed that
the fill factor degraded during
extended operation at
negative
potentials relative to open
-
circuit
(Figure S5
b). Cleaning the device via a 10
s
dip in 3:1 (vol)
hydrochloric acid and nitric acid
,
and rinsing with copious deionized H
2
O
,
restored a
substantial
fraction of the initial fill factor.
The presence of a 4 nm Ti
adhesion layer in n
+
p
-
Si/Ti/Pt
μ
-
cone
array photocathodes did not
affect
the
stability of the
catalyst layer. Furthermore, c
omparisons of
SEMs of devices before and after 12
h
of continuous testing in 0.50 M H
2
SO
4
(aq)
did not reveal
changes to the morphology or coverage of the Ti/Pt layer during extended operation (Figure S
6
).
Thus, the loss in
V
-
10
can be attributed to reduced activity of the thin metal film and not to
delamination or loss of catalyst material.
The frequency
-
dependent impedance response for n
+
p
-
Si/Pt
μ
-
cone array photo
cathodes is
plotted in Figure S
5
c.
Two distinct semicircles w
ere observed when plotting the negative of the
imaginary component of the impedance against the real component of the impedance, a Nyquist
Plot.
Semicircles
that
can be separated by frequency
indicate the presence of
multiple
charge
trapping regions
which
exhibit
time constants differing by at least an order of magnitude
.
6
Such a
response has been reported previously at silicon and hematite photoelectrodes decorated with
catalysts.
6, 7
Multiple equivalent circuits can be fit to this response
. A
n unambiguous assignment
of the individ
ual capacitances and resistances
could be achieved by
a systematic variation of
doping concentration and catalyst
coverages at the devices; this is beyond the focus of the
present study.
In the high
-
frequency limit, the impedance
contribution from capacita
nce
is
minimized
such that
the total impedance
is dominated by
the ohmic resistance of the cell
,
which
was ~10 ohms
.
In the low
-
frequency limit,
the impedance of
all
capacitances is large
.
and the
real component of the
impedance approaches the direct current
polarization resistance of the
cell
.
At all potentials, the polarization resistance was larger than the series resistance.
Figure S
6
:
SEM images at 10 kV and 30 degree sample tilt of n
+
p
-
Si/Ti/Pt
μ
-
cone array
photocathodes (a) before and (b) after 12
h
of chronoamperometric testing in 0.50 M H
2
SO
4
(aq)
under simulated AM1.5 illumination at RHE. Scale bars represent 5
μ
m at 0 degrees tilt.
Fig
ure S
7
:
J
-
E
behavior of the best
-
performing
p
-
Si/CoP μ
-
cone
photocathodes with 400 mC
cm
-
2
Co
-
P deposited photoelectrochemically. The device was cycled from the initial open
-
circuit
potential to
-
0.376 V vs RHE in H
2
-
saturated 0.50 M H
2
SO
4
(aq)
under
100 mW cm
-
2
of
simulated AM1.5
illumination
, with rapid stirring of the electrolyte to remove bubbles
from the
electrode surface
.
Figure S
8
:
Extended photoelectrochemical
testing
of a p
-
Si/CoP μ
-
cone photocathode with 400
mC cm
-
2
Co
-
P. (a
) Current
density vs potential
behavior in H
2
-
saturated 0.50 M H
2
SO
4
(aq)
under
100 mW cm
-
2
of simulated AM1.5 illumination, with rapid stirring of the electrolyte for the 1
st
,
10
th
, and 100
th
cycle at 50 mV s
-
1
, blue, red, and black circles, respectively. The 100
th
scan
occu
rred after 30 min of cycling. (b
)
Comparison of the
J
-
E
behavior
over 24
h
of continuous
H
2
(g)
evolution at RHE under 1
-
Sun illuminatio
n for the device activated in (a
). The i
nitial, 12
h
,
and 24
h
J
-
E
behavior is shown in blue,
red, and black respectively. (c
) Chron
o
amperometry
trace for the device activated in (
a
).
Figure S9:
Scanning
-
electron micrographs of the p
-
Si/CoP μ
-
cone photocathode with 400 mC
cm
-
2
Co
-
P before (A) and after (B) testing in 0.50 M H
2
SO
4
(aq) under 1
-
Sun illumination for 24
h. Scale bars represent 5
μ
m at 0 degrees tilt.
Figure S
10
: Comparison of reflection from bare Si μ
-
pyramids and bare Si μ
-
cones as measured
using a Cary 5000 UV
-
VIS spectrometer with an integrating sphere.
Supporting References
1.
K. T. Fountaine, W.
-
H. Cheng, C. R. Bukowsky and H. A. Atwater,
ACS Photonics
,
2016,
3
, 1826
-
1832.
2.
T. Mårtensson, M. Borgström, W. Seifert, B. Ohlsson and L. Samuelson,
Nanotechnology
, 2003,
14
, 1255.
3.
C. T.
Chen, D. B. Turner
-
Evans, H. Emmer, S. Aloni and H. A. Atwater, 2013 IEEE 39th
Photovoltaic Specialists Conference (PVSC), 2013.
4.
M. R. Shaner, M. T. McDowell, A. Pien, H. A. Atwater and N. S. Lewis,
J. Electrochem.
Soc.
, 2016,
163
, H261
-
H264.
5.
M. R. S
haner, K. T. Fountaine, S. Ardo, R. H. Coridan, H. A. Atwater and N. S. Lewis,
Energy Environ. Sci.
, 2014,
7
, 779
-
790.
6.
D. Klotz, D. A. Grave, H. Dotan and A. Rothschild,
J. Phys. Chem. Lett
, 2018,
9
, 1466
-
1472.
7.
H. Zhang, Q. Ding, D. He, H. Liu, W. Li
u, Z. Li, B. Yang, X. Zhang, L. Lei and S. Jin,
Energy Environ. Sci.
, 2016,
9
, 3113
-
3119.