Supporting Information
S
-
1
Supporting Information for
Nanoporous Gold as a
Highly Selective and Active Carbon Dioxide
Reduction Catalyst
Alex J. Welch
1,2
, Joseph S. DuChene
1,2
, Giulia Tagliabue
1,2
, Artur Davoyan
1,2
,
Wen
-
Hui Cheng
1,2
, and Harry A. Atwater
1,2
*
1
Department of Applied
Physics and Material S
cience, California Institute of
Technology, Pasadena, CA 91125, USA.
2
Joint Center for Artificial Photosynthesis, California Institute of Technology,
Pasadena, CA 91125, USA.
Corresponding Author
* E
-
mail:
haa@caltech.edu
;
Supporting Information
S
-
2
Experimental Methods:
Materials.
Potassium
car
bonate (
99.995%, Sigma Aldrich
), Nitric Acid (
18
M EMD Millipore Corporation
),
Si [
p
-
type, 0
-
10
cm, (100) orientation, 620
±
25
m thick, University
Wafers
]
,
Au foil (
99.9975%,
0
.1 mm thick, Alfa Aesar
), Pt foil
(
99.99% ,
0
.05
mm thick, Alfa Aesar
) and Copper
(II)
sulfate (
≥99%, Sigma Aldrich
)
were used without modification unless otherwise noted.
The materials for electron
beam
deposition we
re ordered from Plasma
terials.
The
Au
target
was 99.99% pure
with 3
-
6
mm random size pieces, Ag wa
s
99.99% pure with 3
-
6
mm random size
pieces, and Ti was 99.995% pure in
0
.25” diameter pellets.
All water used for
experiments was deionized and filtered thr
ough a 0.22
m Millipak Express 40,
serial number 0826.
Fabrication of nanoporous
Au
(np
-
Au)
films
.
An
AMOD dual
electron
beam
deposition
system (System 02520, Angstrom Engineering)
was used to
fabricate all samples. First
,
Si substrates were cleaned by sonicating sequentially
in acetone, isopropanol, and
de
ionized water for five minutes each. The samples
were then stored in deionized water until they were dried with N
2
prior
to be
ing
placed in the electron beam deposition
chamber
. First
,
2
nm of Ti was deposited
at a rate of 1 Å/s
,
then 50
nm of Au was deposited at a rate of 2 Å/s.
Next,
Au and
Ag were co
-
deposited at a rate of 2Å/s and 6Å/s
,
respectively
,
to create a 25% Au
and 75% Ag alloy
. Over the course of the depos
ition the partial pressure of the
chamber would rise from ~10
-
7
torr to ~10
-
6
torr and the temperature would rise
from 20
C to 60
C
for a 1μm thick sample
. If the samples were being etched at
room temperature they were then placed in a beaker of 70 wt.%
HNO
3
for 10
min.
After this
time had elapsed,
they were rinsed with deionized water 10 times before
being dried with N
2
.
Scanning Electron Microscopy (SEM)
characterization
.
The fabricated
samples were
imaged in a Nova200 Nanolab Dualbeam
FIB/
SEM with
an
acceleration voltage of 15
keV and spot size of 3. To image a cross section
of the
film,
samples were cracked with
a
diamond scribe and placed at
a
90
orientation
relative
to
the
electron beam.
Helium Focused Ion Beam (HFIB)
characterization
.
The fabricated
samples were imaged in a
Zeiss
HFIB
with an acceleration voltage of
30
k
V
, beam
current between 1
-
3
pA
, working distance of 8
mm, and a scan dwell time of 6.5
s.
X
-
ray d
iffraction (XRD
)
characterization
.
X
RD spectra were taken using
an X’PERT
-
PRO MRD Serial # DY3178 made by PANalytic.
The scan went from
30
to 120
. The voltage was 45
kV, current 40
mA and the beam attenua
tor was
Ni 0.125
mm automatic.
Cu
Underpotential Deposition
(Cu UPD)
of Au films
.
The surface area
of the np
-
Au was determined using Cu UPD.
A CuSO
4
solution (
0.1
M
) in 0.5
M
H
2
SO
4
was prepared and used as the deposition bath for all Cu UPD experiments.
T
he solution was bubbled with N
2
(Research grade from Airgas)
for 30 min to
remove
dissolved O
2
from the solution
prior to starting the experiment
. The working
electrode was a
planar Au film or a
np
-
Au film
of various thickness
es
with a Pt
mesh counter electrode and a Ag/AgCl reference electrode. The potential
of the
Supporting Information
S
-
3
working electrode
wa
s swept from 450 mV to 50 mV vs. Ag/AgCl at a scan rate of
5 mV s
-
1
.
A total of three to five electrodes were
measured
at each thickness
ranging from 100 nm to
1.6
μm and the electrochemical surface area enhancement
was obtained by taking the average surf
ace area
of the np
-
Au film
relative to that
of a planar Au film
of known geometrical surface area
.
Electrocatalytic CO
2
r
eduction
reaction
(CO
2
R
R
)
experiments
.
A two
-
compartment
electrochemical
cell made of polyether ether ketone (PEEK) was
used to perform
CO
2
R
R
experiments.
A
volume
of
2
mL
of 50 mM K
2
CO
3
was
added to each compartment,
which
were separated by an anion exchange
membrane (AGC, Selemion AMV).
The np
-
Au film served as the working
electrode
with a
Ag/AgCl leakless reference electrode
and a Pt foil counter electrode.
The Pt
foil was soaked in 10 wt.% HNO
3
for 1 h
and then flame annealed to remove
contaminants before each experiment. The flame annealing process entails
holding a flam
e to the foil until it glows red then rinsing the foil in water and drying.
This process is repeated twice. The same procedure was also applied to the Au
foil before testing.
CO
2
saturated 50
mM K
2
CO
3
(
pH
6.8
)
was prepared by bubbling
CO
2
(Research grade
from Airgas) into the electrolyte for 30 min prior to
experiments.
Each electrolyte compartment was bubbled with
CO
2
at a rate of 5
SCCM through a fine glass dispersion frit to maximize the speed of delivery of CO
2
into solution. The outflowing gas was se
nt
through a
flow meter to check
that
the
flow
of CO
2
in and out of the cell was the same
,
ensuring that it was
thoroughly
sealed
against gas leaks
.
T
he outflowing gas was sent through a vapor trap to
remove all water from the air before i
t was fed into a
(
SRI
-
8610
) gas
chromato
graph
.
All experiments were performed at room temperature
with a
Biologic
VSP
-
300
potentiostat
.
All potentials were con
verted to
the reversible hydrogen
electrode (
RHE
)
scale
using
t
he equation
:
E
vs. RHE
=
E
vs. Ag/AgCl
+ 0.197
V
+
0.059
V pH
-
1
×
solution
pH
. Before each expe
riment
,
potentiostatic
electrochemical impedance spectroscopy (
PEIS
)
was
performed
to determine the
solution
resistance of the cell,
which was
typically between 30
–
60
. The applied
electrochemical
potentia
l
was
then
compensated by 85%
using iR compensation
of the potentiostat
.
T
he
electrochemical
cell was dismounted and rinsed multiple
times
after each experiment and
then stored in 10
wt.
% HNO
3
. Before using the
cell for the next experiment
,
it
was sonicated for 10 min
in water
at least 4 times.
Analysis of chemical p
roduct
s
.
To analyze the
chemical
products
,
the
electrode was held at the desired potential for at least 2
h
allowing
for the
completion of
eight gas chromatography measurem
ents.
The gas chromatograph
(SRI
-
8610) used a Hayesep D column and a Mo
l
sieve 5A column with N
2
as the
carrier gas. The gaseous products were detected using a thermal
conductivity
detector
(TCD)
for CO detection
and
a
flame ionization detector (FID) equi
pped
with a methanizer
for H
2
detection
. Quantitative
analysis
of
gaseous
products was
based on calibration with several gas standards over many orders of magnitude in
concentration
.
To measure liquid products
,
the electrolyte on the ano
de and
cathode wer
e sampled at
the end of the run and tested
with
high performance
liquid c
hromato
graphy (
HPLC).
However, no liquid products were ever observed.
Between different potential experiments all of the electrolyte was removed and
Supporting Information
S
-
4
then the cell was rinsed three
times with water before new electrolyte was add and
bubbled with CO
2
.
Supporting Information
S
-
5
Figure
S1
. XRD spectra of (a)
RT
np
-
Au (~800 nm thick)
on a glass substrate, (b)
planar Au
film
on glass
,
and (c) Au foil after flame an
nealing. (d) Zoom
-
in of (200)
peak where planar Au and np
-
Au have been increased by 20x.
From the XRD it is
evident that the planar Au film is highly oriented in the (111) orientation. The data
also shows that the full width half max of the np
-
Au is much larger than that of the
Au foil
,
indicating that the np
-
Au has smaller grains.
Supporting Information
S
-
6
Figure
S2
. Bright field t
ransmission electron microscopy (TEM)
images
of
RT
np
-
Au film
.
The red arrows
denote
grain boundaries
.
A
ll scale bars
represent
5 nm.
Supporting Information
S
-
7
Figure
S3
. Transmission electron microscopy (TEM) analysis of RT np
-
Au films.
(a) Bright
-
field TEM of a particular region of the np
-
Au film along with the
corresponding selected area electron diffraction (SAED) pattern
below it
.
The dark
-
field TEM images numbered 1
-
7 correspond to the spot
s numbered in the SAED
pattern. (b
)
Bright
-
f
ield TEM image of np
-
Au film
along with dark
-
field TEM images
numbered 1
-
3
. All scale bars in all TEM images represent 20 nm.
Supporting Information
S
-
8
Figure
S4
. Transmission electron microscopy (TEM) analysis of RT np
-
Au films.
(a
-
c
) Bright
-
field TEM
image
of a particular region of the np
-
Au film along with
the
corresponding dark
-
fi
eld TEM image obtained from the same
region of the np
-
Au
film. All scale bars in all TEM images represent 20 nm.
(d) Size distribution
histogram of grain sizes obtained from analy
sis of dark
-
field TEM images.
Supporting Information
S
-
9
Figure
S5
.
(
a) Cyclic voltammetry of
LT np
-
Au
(grey curve) and
RT np
-
Au
(dark
blue curve) in 0.5 M H
2
SO
4
obtained at a scan rate of 50
mV s
-
1
. From these
data
we can determine that the LT sample
has ~3
x
greater
electrochemical
surface area
as compared to the RT sample.
(b) shows a histogram of pore widths measured
on these two samples at three different locations on the sample.
Representative
SEM cross
-
section images of
(c)
RT np
-
Au
and (d)
LT np
-
Au
film
s.
The sc
ale bars
on both images correspond to 100 nm.
These images
correspond to the actual
electrodes used
for electrochemical tests involving different electrolyte
concentrations.
Supporting Information
S
-
10
Figure S6
.
(a) SEM cross
-
section image of
~150
nm
-
thick
RT
np
-
Au film.
The
scale bar represents 100 nm.
(b) Faradai
c efficiency as a function of applied
potential
(
E
)
for 150 nm
-
thick
RT
np
-
Au film.
(c) Partial current density for CO
(
J
CO
)
from
a 1
50 nm
-
thick
and
~800
nm
-
thick RT
np
-
Au sample
.
C
onsidering
the
4x
smaller surface area
of the thinner film, t
he
relative
J
CO
between the two films is
unexpected.
We
hypothesize
that this is due to mass transport limitations.
Supporting Information
S
-
11
Figure S
7
.
Extended electrochemical stability data for
a
RT np
-
Au
film
(~800 nm
thick)
. The Faradaic efficiency for CO (filled points) and H
2
(open points) was
measured every 15 min via gas chromatography over the course of
110
h at an
applied potential of
E
=
–
0.5 V
RHE
with iR compensation
.
Supporting Information
S
-
12
Fig
ure S8
.
(a,b)
SEM images
of a ~800 nm thick RT np
-
Au film (a)
be
fore and
(b)
after testing for 110
h at −
0.5
V vs. RHE
.
(c,d
)
SEM images
of
a planar Au film
(c)
before
and (d) after testing for 24 h at
−
0.5 V vs RHE
.
(e,f)
SEM images of
a Au
foil (e) before and (f) after testing for 24 h at
−
0.5 V vs RHE.
There is no visible
difference between any of the planar samples before and after
testing.
In the np
-
Au sample there is some
minor
coarsening of the ligaments
, but no significant
changes to t
he film morphology are observed
.
Supporting Information
S
-
13
Figure S9
.
XRD spectra of Au films before and after 24
h
of testing for (a)
~800
nm thick
RT
np
-
Au film, (b) planar Au film, and (c) Au foil. The peak at 68
in the
RT np
-
Au film and the planar Au film is due to the
Si substrate.
In
Fig. S1
, the XRD
patterns were collected from films supported on a glass substrate to avoid the peak
from the Si substrate. Negligible differences
were observed between Au
peaks
obtained from on Si
vs. glass
substrate
s
.