1
Supplementary
Information
Cathodic NH
4
+
Leaching of Nitrogen Impurities in CoMo Thin-film Electrodes in
Aqueous
Acidic Solution
Weilai
Yu
a
, Pakpoom Buabthong
b
, Carlos G. Read
a
, Nathan F. Dalleska
c
, Nathan S.
Lewis
a
, Hans-
Joachim Lewerenz
b
, Harry B. Gray
a*
and Katharina
Brinkert
a,d*
a.
Division of Chemistry and Chemical Engineering, California Institute of Technology, 1200 E California Blvd.,
Pasadena,
CA 91125, USA.
b. Division of Engineering and
Applied Science, California Institute of
Technology,
1200 E
California
Blvd.,
Pasadena,
CA 91125, USA.
c. Environmental Analysis Center, Division of Geology and Planetary Science, California
Institute of
Technology,
1200 E California Blvd.,
Pasadena, CA 91125, USA.
d. Department
of Chemistry, University of Warwick, Gibbet Hill Road, Coventry,
CV4
7AL,
UK.
Email: hbgray@caltech.edu
; katharina.brinkert@warwick.ac.uk
Electronic
Supplementary
Material
(ESI)
for
Sustainable
Energy
&
Fuels.
This
journal
is
©
The
Royal
Society
of
Chemistry
2020
2
Experimental
Electrode preparations
n-type
Si
wafers
(111,
As-doped,
resistivity
< 0.005
Ω
· cm,
3”
diameter)
were
obtained
from
Addison
Inc.
To
remove
the
surface
oxide,
the
Si
substrate
was
etched
for
1 min
in
a buffered
HF
etchant
(6:1
volume
ratio
of
40%
NH
4
F(aq)
to
49%
HF(aq)).
All
solutions
were
made
from
ultrapure
water
and
analytical grade chemicals
having
an organic impurity
level
< 50 ppb.
Sputtering
using
a high-vacuum
magnetron-sputtering
system
(AJA
International
Inc.)
was
performed
at
a
constant
Ar
flow
of
20
sccm
and
a working
pressure
of
5 mTorr.
The
Ti
adhesion
layer
was
directly
sputtered
onto
the
freshly
etched
n
+
-Si
wafer
substrate
using
direct-current
(DC)
mode
at 60
W
for
35
min.
The
substrates
were
maintained
at
room
temperature
during
the
sputtering
process.
The
CoMo
thin-film
was
then
co-sputtered
in
radio
frequency
(RF)
mode
by
simultaneously
keeping
the
Co
target
at
150
W
and
the
Mo
target
at 100
W,
while
under
a constant
Ar
flow
of
20
sccm
for
1 h.
FeMo
and
NiMo
were
deposited
using
an
analogous
procedure
in which
the
sputtering
power
for
Fe
and
Mo
targets
was
the
same
as
that
for
Co.
Sputtering
of
the
single-element
Co
and
Mo
thin
films
was
performed
with
the
target
maintained for 1 h at 150 W and
100 W, respectively.
The
wafer
was
cut
into
~1.2
cm
2
pieces
and
ohmic
contacts
to the
electrodes
were
formed
by
rubbing
an
In-Ga
eutectic
onto
the
unpolished
back
sides
of
the
Si
samples.
Conductive
Ag
paste
was
applied
to attach
a
Sn-plated
Cu
wire
to
the
ohmic
contact.
The
wire
was
threaded
into
a glass
tube
and
the
sample
was
encapsulated and sealed to the glass tube using black, chemically
resistant epoxy (Loctite 9460).
Electrochemical nitrogen reduction in aqueous and non-aqueous electrolytes
Electrochemical
experiments
were
performed
using
BioLogic
SP-200
potentiostats
(Biologic,
Grenoble,
France)
controlled
by
standard
EC-Lab
software.
The
experiments
were
performed
in a standard
three-
electrode
potentiostatic
arrangement
in a sealable
electrochemical
cell.
A Pt
wire
(ALS
Co.,
Ltd)
was
used
as
the
counter
electrode
and
Ag/AgCl
(3
M
KCl,
WPI
Europe,
DRIREF-5)
was
used
as
the
reference
3
electrode.
No
Pt deposits
were
observed
by
XPS
on
the
catalyst
(working
electrode)
surface
after
electrochemical
testing.
All
potentials
are
referenced
herein,
unless
otherwise
specified,
relative
to the
potential
of
the
reversible
hydrogen
electrode
(RHE).
The
electrolyte
was
20
mL
of
50
mM
aqueous
or
ethanolic
H
2
SO
4
.
Prior
to
the
measurements,
the
electrolyte
was
purged
for
10
min
with
Ar
(5.0
purity,
flow
rate
165
mL
min
-1
)
under
gentle
stirring
to
remove
O
2
(g)
from
the
electrolyte.
To
saturate
the
electrolyte
with
N
2
prior
to
electrochemical
measurements,
the
electrolyte
was
purged
for
15
min
with
99.999%
pure
N
2
(flow
rate
95
mL
min
-1
).
Before
entering
the
electrochemical
(EC)
cell,
N
2
and
Ar
were
directed through
a gas
bubbler
that contained 50 mM H
2
SO
4
(aq).
Cyclic
voltammetric
and
chronoamperometric
measurements
were
performed
on
electrodes
under
a
constant
N
2
flow
(95
mL
min
-1
).
For
isotope
labelling
experiments,
the
electrolyte
was
purged
with
15
N
2
instead
of
14
N
2
(Sigma-Aldrich,
product
no.
364584-5L)
for
10
min
prior
to
electrochemical
tests.
Before
entering
the
electrochemical
cell,
the
15
N
2
gas
was
directed
through
an
acid-trap
that
contained
5 mL
of
50
mM
H
2
SO
4
(aq).
Electrochemical
tests
were
performed
under
a continuous
15
N
2
purge.
Three
independent
tests
were
carried
out
for
each
condition
using
freshly
prepared
electrodes
to produce
the
error
bars
in
Figure
1 in
the
main
text.
After
sampling
electrolyte
in
the
cell
at
different
stages
of
the
experiment,
the
same
amount
of
fresh
electrolyte
was
added
back
into
the
cell
to keep
the
total
volume
of
electrolyte constant.
Determination of
ammonium, nitrate and nitrite
concentration by IC
Ammonium
concentrations
were
determined
using
a Dionex
(Sunnyvale
CA,
now
Thermo)
ICS
2000.
The
sample
solutions
were
diluted
to
a certain
ratio
and
loaded
into
a 5
μL
sample
loop
by
an
AS40
autosampler.
The
sample
was
then
injected
onto
a CS16
separator
column
(2
x 250
mm)
that
was
protected
by
a CG16
guard
column
(2
x 50
mm).
Isocratic
methanesulfonic
acid
eluent
at
20
mM
and
pumped
at
0.25
mL
min
-1
was
produced
using
an
eluent
generator
cartridge
based
on
methylsulfonic
acid.
Analytes
were
detected
by
suppressed
conductivity
detection
using
a Dionex
CERS-500
2 mm
suppressor
operated
in
eluent
recycle
mode
with
an
applied
current
of
15
mA.
Concentrations
were
calculated
using
4
Chromeleon 6.8 software.
The
concentration
of hydrazine
was
determined
below
the
detection
limit
in the
electrolyte
by
the
Watt
and Chrisp method.
1
14
NH
4
+
/
15
NH
4
+
derivatization with dansyl chloride
and analysis by UPLC-MS
Detection
of
14
NH
4
+
and
15
NH
4
+
as
dansyl
chloride
derivation
products
was
performed
on
an
Acquity
I-
Class
UPLC
coupled
to
a Xevo
G2-S
Quadrupole
Time-of-Flight
(QTof)
mass
spectrometer
(both
from
Waters
Corporation).
2
Dansyl
chloride
derivatization
was
performed
by
preparing
solution
mixtures
in the
following
order
in
2 mL
glass
vials:
0.4
mL
of
~ 290
μM
dansyl
chloride
(Sigma
Aldrich,
≥
99.0%,
for
HPLC)
was
dissolved
in
acetonitrile
(UPLC
grade).
Subsequently,
0.3
mL
of
sodium
carbonate
stock
solution
(2 g/100
mL)
and
0.4
mL
of
collected
electrolyte
in
50
mM
H
2
SO
4
(aq)
were
added
to the
vials.
The
vials
were
vortexed
for
10
s before
loading
onto
the
autosampler
plate
for
analysis.
To
achieve
effective
derivatization
without
heating
the
samples,
the
pH
of
the
sample
solution
was
adjusted
between
9
and
10
using
pH
paper.
The
derivatization
products
(dansyl-ammonia
adducts)
in
the
sample
solution
were
separated
by
reverse
phase
chromatography
with
a binary
water/acetonitrile
solvent
and
a steady
gradient.
The
concentration
was
quantitatively
determined
by
calculating
the
peak
area
at specific
mass
positions
(m/z
= 251.0854
a.u.
for
14
N
and
z = 252.0825
a.u.
for
15
N,
resp.).
Data
analyses
were
performed
using
the
Quanlynx
software
and
each
collected
sample
was
measured
three
times
for
accurate
concentration determination.
Materials characterization
Atomic-force
microscopy
(AFM)
images
were
obtained
on
a Bruker
Dimension
Icon
or
a Mutimode
8
using
Bruker
ScanAsyst-Air
probes
(silicon
tip,
silicon
nitride
cantilever,
spring
constant:
0.4
N/m,
frequency:
50-90
kHz),
operating
in
the
ScanAsyst
mode.
The
scan
sizes
were
1
μm
x 1
μm.
Images
were
analyzed using the
Nanoscope Analysis software (version 1.9).
5
For
transmission-electron
microscopy
(TEM),
samples
of
the
films
were
prepared
using
a focused
Ga-ion
beam
(FIB)
on
a FEI
Nova-600
Nanolab
FIB/FESEM.
Pt
and
C protection
layers
were
deposited
before
exposure
of the
sample
to
the
FIB.
High-resolution
TEM
(HRTEM)
images
were
obtained
using
a Tecnai
Polara (F30) TEM at an accelerating voltage of 300 keV.
Scanning-electron
micrographs
(SEMs)
were
obtained
with
a FEI
Nova
NanoSEM
450,
at an accelerating
voltage of 5.00 kV with a working distance
of 5 mm and an in-lens secondary
electron
detector.
X-ray photoelectron spectroscopy
X-ray
photoelectron
spectroscopy
(XPS)
was
performed
using
a Kratos
Axis
Ultra
system
with
a base
pressure
of 1 x 10
-9
torr
in
the
analysis
chamber.
A monochromatic
Al
Kα
source
was
used
to irradiate
the
sample
with
X-rays
(1486.7eV)
at
450
W.
A
hemispherical
analyzer
oriented
for
detection
along
the
sample
surface
normal
was
used
for
maximum
depth
sensitivity.
High-resolution
spectra
were
acquired
at
a
resolution
of
25
meV
with
a pass
energy
of
10
eV.
The
data
were
analyzed
using
CasaXPS
computer
software.
First,
the
spectra
were
calibrated
by
referencing
the
C 1s
peak
position
to
284.8
eV.
Co
2p,
Mo
3d, N 1s peaks were then fitted
to multiple subspecies each having Gaussian-Lorentz peak shapes.
Peak
assignments
were
based
on
previous
reports.
3–5
The
transmission
function
(T)
of
the
analyzer
was
recorded
automatically
during
the
time
of
measurement.
The
relative
sensitivity
factors
(RSF)
of
each
element
were
provided
by
the
instrument
manufacturer,
and
the
inelastic
mean
free
path
(IMFP)
of
photoelectrons at different
kinetic
energies was calculated using Quases imfptpp2m computer software.
Surface
compositions
of
each
element
were
calculated
by
comparing
the
area
of
the
peaks
for
each
element
normalized by
(RSF*T*IMFP).
XPS
data
were
obtained
ex-situ
i.e.,
after
a short
sample
transfer
through
air,
which
could
potentially
confound
linking
the
surface
composition
and
oxidation
states
found
in
UHV
to
the
ones
present
during
electrocatalysis.
6
Determination of
NO/NO
2
(g) concentration
in
Ar
and
14
N
2
gases
Approximately
15
L of
a gas
sample
was
collected
from
an Ar/N
2
cylinder
(ultra-high
purity,
Airgas)
and
stored
in
a Teflon
bag
before
further
analysis.
A
chemiluminescence
NO/NO
2
/NO
X
Analyzer
(Teledyne
Model
T200)
was
used
to
measure
the
NO
and
NO
2
concentrations
in the
gas
sample
at a 200
cm
3
min
-1
flow
rate.
The
measurement
had
an
uncertainty
of
0.5
ppb.
The
concentration
was
recorded
after
~ 15
min
of equilibration between the gas, sampling line,
and chemiluminescence chamber inside the instrument.
7
Figure
S1.
(a)
Scheme
of the
n
+
-Si-Ti-CoMo
electrode
composition;
(b)
cross-section
TEM
image
of
the
n
+
-Si-Ti-CoMo
electrode
structure
as-prepared;
(c)
HR-TEM
image
of
the
CoMo
surface
after
12
h testing
at
-0.54
V
vs.
RHE
in
50
mM
H
2
SO
4(aq)
and
(d)
magnification
of
(c)
.
The
scale
bars
are:
(b)
100
nm;
(c)
100 nm and
(d)
10 nm.
8
Figure
S2.
Chronoamperometric
(CA)
measurements
of
an
n
+
-Si/Ti/CoMo
electrode
in aqueous
50
mM
H
2
SO
4
(blue)
and
50 mM H
2
SO
4
in ethanol
(green).
9
6
.
7
6
.
8
6
.
9
7
7
.
1
7
.
2
7
.
3
7
.
4
7
.
5
0
0
.
1
0
.
2
0
.
3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
0
5
1
0
1
5
2
0
2
5
3
0
3
5
0
1
2
3
4
5
6
7
8
9
1
0
1
1
1
2
e
l
e
c
t
r
i
c
c
o
n
d
u
c
t
a
n
c
e
/
μ
S
r
e
t
e
n
t
i
o
n
t
i
m
e
/
s
2
0
m
i
n
1
h
1
2
h
P
u
r
g
e
d
e
l
e
c
t
r
o
l
y
t
e
6
.
7
6
.
9
7
.
1
7
.
3
7
.
5
0
0
.
1
0
.
2
0
.
3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
0
0
.
1
0
.
2
0
.
3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
4
.
6
4
.
7
4
.
8
4
.
9
5
5
.
1
5
.
2
5
.
3
5
.
4
e
l
e
c
t
r
i
c
c
o
n
d
u
c
t
a
n
c
e
/
μ
S
r
e
t
e
n
t
i
o
n
t
i
m
e
/
s
K
+
N
H
4
+
Figure
S3.
Ion
chromatogram
of
the
electrolyte
(red)
and
after
20
min
(light
blue),
1 h (cyan)
and
12
h
(dark
blue)
electrocatalysis
of
n
+
-Si/Ti/CoMo
electrode
system
in
aqueous
50
mM
H
2
SO
4
at
-0.54V
vs.
RHE.
The
inset
shows
the
NH
4
+
chromatogram
at
a higher
resolution.
K
+
was
introduced
through
the
junction of
Ag/AgCl (3
M
KCl) reference electrode.
10
Figure
S4.
X-ray
photoelectron
spectra
of the
n
+
-Si/Ti/CoMo
surface
after
1 h of
electrocatalysis
in
aqueous 50 mM H
2
SO
4
electrolyte at -0.54V
vs
RHE.
(a)
Co 2p core levels;
(b)
Mo 3d core levels and
(c)
N 1s core levels. Color coding is given in the legend;
B.E
is the binding energy.
11
Figure
S5.
N 1s
core
level
X-ray
photoelectron
spectra
of
the
CoMo
surface
after
being
held
for
20
min
at
-0.54V
vs.
RHE
in
an aqueous
50
mM
H
2
SO
4
electrolyte
(top)
and
after
incubation
of
the
electrode
afterwards
for
10
min
in
10
mM
KOH
(bottom).
Color
coding
is given
in the
legend;
B.E
is the
binding
energy.
12
Figure
S6.
(a)
Photograph
of our
electrochemical
experimental
setup
for
purging
with
15
N
2
gas
in the
presence
of an
acid-trap
(b)
.
The
15
N
2
gas
was
directed
through
a plastic
tube
containing
a solution
of
50
mM H
2
SO
4
(aq) before
going into the
EC cell.
(a)
(b )
13
Figure
S7.
Comparison
of
averaged
current
density
of
CoMo
electrodes
at
different
potentials
under
constant
15
N
2
purging.
14
Figure
S8.
X-ray
photoelectron
spectra
of
the
n
+
-Si/Ti/CoMo
surface
after
20
min
of
electrocatalysis
in
aqueous
50
mM
H
2
SO
4
electrolyte
at -0.54
V
vs.
RHE
under
14
N
2
purging
(top)
and
15
N
2
purging
(bottom).
(a)
Co
2p core levels;
(b)
Mo 3d core levels and
(c)
N 1s
core levels. Color
coding
is given
in the legend;
B.E. is
the binding energy.
15
Figure
S9.
X-ray
photoelectron
spectra
of
the
n
+
-Si/Ti/CoMo
surface
after
20
min
of
electrocatalysis
in
aqueous
50
mM
H
2
SO
4
electrolyte
at four
different
potentials
vs.
RHE.
(a)
Co
2p
core
levels;
(b)
Mo
3d
core levels and
(c)
N 1s core
levels.
Color coding is given in the legend; B.E. is the binding energy.
16
Figure
S10.
(a)
Comparison
of
cyclic
voltammograms
(CVs)
obtained
at the
CoMo
electrode
tested
in
50
mM
H
2
SO
4
(aq)
with
Ar(g)
purging
and
without
purging.
(b)
Comparison
of
CVs
(CoMo
electrode)
tested
in
50
mM
H
2
SO
4
(aq)
without
purging
in
the
presence
of
different
concentrations
(0,
54,
104,
170
μM)
of
nitrate
ions
(scan
rate:
20
mV
s
-1
).
(c)
Comparison
of
chronoamperometry
(CA)
at
the
CoMo
electrode
held
at -0.54
V vs.
RHE
in
50
mM
H
2
SO
4
(aq)
without
purging
in the
presence
of
different
concentrations
(0, 54,
104, 170
μM)
of nitrate ions.
17
Table
S1
.
Measured
NH
4
+
concentrations
by
IC
obtained
from
various
control
experiments.
The
NH
4
+
-
production
yield
in
Table
1 (main
text)
is
corrected
against
these
background
NH
4
+
values
under
corresponding conditions. The total volume
of electrolyte is 20 mL.
Control Experiment
t [min]
c(NH
4
+
)
[μM]
Total
amount [nmol]
50 mM H
2
SO
4
1.1
22
CoMo at
E
oc
20
1.4
28
CoMo at
E
oc
60
2.6
52
CoMo at
E
oc
720
2.8
56
NiMo at
E
oc
20
1.4
28
FeMo at
E
oc
20
1.7
34
Co at
E
oc
20
1.0
20
Mo at
E
oc
20
2.5
50
CoMo at
E
oc
in 50 mM
HCl
20
1.9
38
CoMo
at
E
oc
in 50 mM HClO
4
20
2.3
46
50 mM
H
2
SO
4
, Ar/N
2
Purge
20
0.8
16
50 mM
H
2
SO
4
, Ar/N
2
Purge
60
0.7
14
50 mM H
2
SO
4
/Ethanol,
Ar/N
2
Purge
20
2.0
40
Pt wire in 50 mM H
2
SO
4
20
0.9
18
Pt wire in 50 mM H
2
SO
4
/Ethanol
20
2.5
50
18
Table
S2.
Summary of
the peak
positions
in the XPS
analysis.
3–5
Species
Binding
energy [eV]
Co
0
778.2
Co
2+
781
Co
3+
782.7
Co-OH
786.5
Mo
2+/3+
228.7-228.8
Mo
3+/4+
230.2
Mo
6+
231.2-232.5
Mo
d+
(2
< d < 4)
229.97
Mo-N
397.2
NH
ad
/NH
2,ad
399.4
NH3
,ad
+
402.1
19
Table
S3.
Comparison
of
electrode
area-normalized
increases
in
14
NH
4
+
ion
concentration
determined
by
the
dansyl
chloride
derivatization/UPLC-MS
method,
after
electrochemical
tests
using
CoMo
electrodes
under
various
potential
and
gas-purging
(
15
N
2
/Ar)
conditions.
The
values
correspond
to
the
data
points
shown in Figure 5 in main text.
Entry
E
[V
vs.
RHE]
Purged gas
a
Increased c(
14
NH
4
+
)
[μM
cm
-2
]
1
-0.29
15
N
2
3.2±0.1
2
-0.54
15
N
2
3.9±0.4
3
-0.54
15
N
2
6.9±0.2
4
-0.79
15
N
2
7.9±0.3
5
-1.29
15
N
2
17.9±0.7
6
-0.29
Ar
3.2±0.2
7
-0.54
Ar
3.7±0.03
8
-0.54
Ar
6.3±0.3
9
-0.54
Ar
7.9±0.2
10
-0.79
Ar
9.6±0.5
11
-1.29
Ar
6.8±0.4
a
All purged gases
were first passed through
a gas bubbler of 50 mM H
2
SO
4
(aq).
20
Table
S4.
Comparison
of
nitrate
(NO
3
-
)
concentration
measured
by
IC
in the
electrolyte
samples
collected
at different
experiment stages for four
independent experiments.
Entry
a
Gas
c(NO
3
-
) before
purge
[μM]
c(NO
3
-
)
after
purge
[μM]
c(NO
3
-
)
after
EC
b
[μM]
14
NH
4
+
yield
[μM]
1
15
N
14.2
12.6
17.1
9.8
2
15
N
10.9
5.3
18.8
3.2
3
Ar
21.5
10.7
18.1
6.3
4
Ar
18.9
4.9
16.4
7
a
All
electrochemical
experiments
were
performed
at -0.54
V
vs.
RHE.
The
electrolyte
volume
was
20
mL.
The
concentrations
of nitrite
ions
were
below
the
detection
limit
for
all electrolyte
samples.
b
EC
represents
electrochemical experiments
The
50
mM
sulfuric
acid
stock
solution
had
low
concentrations
(1.8
μM)
of
nitrate
ions,
which
increased
to
>10
μM
after
transferring
to
the
electrochemical
cell
possibly
due
to
the
additional
contaminants
in the
cell.
After
10-min
of
purging
of
both
15
N
2
and
Ar
gas,
the
nitrate
concentrations
consistently
decreased
in
four
independent
tests.
This
behavior
indicated
that
continuous
purging
of
15
N
2
/Ar
might
have
removed
volatile
HNO
3
species
from
the
solution,
instead
of
introducing
additional
NO
3
-
into
the
electrolyte.
In
comparison,
only
a slight
increase
(6.9
μM)
in
the
concentration
of
nitrate
ions
was
observed
in a small
volume
(5 mL)
of
50
mM
H
2
SO
4
(aq)
electrolyte
that
contained
56.6
μM
of
pre-existing
nitrate
ions
after
purging
with
14
N
2
gas
for
20
min.
Furthermore,
we
used
a chemiluminescence
analyzer
to
measure
the
NO(g)
and
NO
2
(g)
concentrations
in
both
Ar
and
14
N
2
gas
cylinders
(ultra-high
purity,
Airgas).
The
concentrations
of NO(g)
in
both
the
Ar
and
14
N
2
cylinders
was
measured
as
0 ppb,
whereas
~ 1.3
and
3.1
ppb
of
NO
2
(g),
respectively,
were
detected.
As
a reference,
the
concentrations
of
NO(g)
and
NO
2
(g)
in