of 102
S
1
Enhanced Ammonia Oxidation Catalysis by a Low
-
Spin Iron Complex Featuring
Cis
Coordination Sites
Michael D. Zott and Jonas C. Peters*
Division of Chemistry and Chemical Engineering, California Institute of Technology
(Caltech), Pasadena,
California
91125, United States
Contents:
S.1
General
p
rocedures
p. S2
S.2 NMR spectra for electronic structure elucidation
p. S
4
S.3 UV
-
vis spectra
p. S
8
S.4 Catalytic controlled potential coulometry experiments
p. S
10
S.5 Electrode rinse test after CPC
p. S
13
S.6 GC
-
MS data for
15
NH
3
experiment
p. S
14
S.7 XPS spectra of BDD plate electrode
p. S
15
S.8 DPV data for E
1
analysis
p. S
18
S.9 Further analysis of speciation related to E
1
p. S
19
S.10 Catalytic rate versus iron and NH
3
concentrations (E
2
)
p. S
22
S.11 Procedure for FOWA
p. S
24
S.12 Further reactivity considered at E
2
p. S
26
S.1
3
Computational methodology
p. S
2
7
S.1
4
DFT spin
-
state ordering
p. S
2
8
S.1
5
DFT tabulated energies
p. S
3
0
S.1
6
DFT structures
p. S
3
2
S
2
S.1 General p
rocedures
General Considerations:
All manipulations were carried out using standard Schlenk or
glovebox
techniques under an N
2
or
Ar atmosphere. Unless otherwise noted, solvents
were deoxygenated and dried by thoroughly sparging with N
2
gas followed by passage
through an activated alumina column in
a
solvent purification system
(
SG Water, USA
LLC
)
. For electrochemical measurements under an Ar atmosphere, solvents were further
degassed and then left under Ar. All solvents were stored over
activated
3 or
4 Å
molecular sieves prior to use.
A
nhydrous ammonia gas was dried by passage through a
calcium oxide drying tube. All reagents were purchased from commercial vendors and
used without further purification unless otherwise stated. Tris(2
-
py
ridylmethyl)amine
(TPA)
,
1
tris(2
-
pyridylmet
hylamine)
iron(II)
triflate
bis
-
acetonitrile
([(TPA)Fe(MeCN)
2
]OTf
2
),
2
6
-
(1,1
-
di(pyridin
-
2
-
yl)ethyl)
-
2,2'
-
bipyridine (bpyPy
2
Me),
3
and
6
-
(1,1
-
di(pyridin
-
2
-
yl)ethyl)
-
2,2'
-
bipyridine
iron(II)
triflate
bis
-
acetonitrile
([(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
)
4
15
NH
4
OTf was prepared from
15
NH
4
Cl (Cambridge
Isotope Laboratories) by anion exchange with silver triflate
or metathesis with triflic acid,
followed by repeated recrystallization from
boiling
acetonitrile.
1
H NMR
chemical
shifts
are reported in ppm relative to tetramethylsilane, using residual solvent resonances as
internal standards.
Electrochemistry:
Voltammetry
experiments were carried out
with
a Biologic
VSP
-
300
or CH
Instruments 600B
potentiostat using a one
-
compartment
three
-
electrode cell, and
coulometry experiments were carried out with a Biologic VSP
-
300 potentiostat using a
one
-
compartment three
-
electrode cell with a septum capped 14/20 joint for headspace
analysis.
For
voltammetr
y
,
a
boron
-
d
oped
d
iamond (BDD
)
working
electrode (3 mm
diameter),
a
Pt wire counter electrode
,
and
a
Ag/AgOTf
reference electrode
(
5 mM AgOTf
and 0.1 M TBAPF
6
in MeCN) were employed. For CPC
, the same reference electrode
was used, but a BDD plate (
geometri
c area:
4
cm
2
) and a Pt mesh were used respectively
as working and counter electrode. All redox potentials in the present work are reported
versus
the
Fc/Fc
+
couple, measured before e
ach
experiment
to be
approximately
+
0.1
2
V
versus our Ag/AgOTf reference
electrode
.
CVs were collected at 10
0 mV·s
1
unless specified otherwise
. E
1/2
values for the
reversible waves were obtained from the half potential between th
e oxidative and
reductive peaks
.
CV
measurement
s
were
performed
apply
ing IR compensation
,
compensating 85% of the resistance measured at one high frequency value (100 kHz).
Potential values for waves that are not fully reversible were obtained as the apparent
standard potential from differential pulse voltammetry measurements
Gas Chromatography:
Gas chromatography was performed in the Environmental
Analysis Center using HP 5890 Series II instruments.
Gas quantification was performed
using
a molecular sieve column attached to a thermal conductivity detector. Argon was
the carrie
r gas. Standard curves were generated by direct injection of hydrogen or
S
3
nitrogen gas. Quantification of background nitrogen was determined using the
background oxygen signal. Isotopic measurements were performed with a separate HP
5890 Series II equipped
with a GasPro column using helium as the carrier gas.
NMR:
NMR spectroscopy was performed using Varian
and Bruker
400 MHz NMR
spectrometer
s equipped with
broadband auto
-
tune probe
s
.
1
H NMR chemical shifts are
reported in ppm relative to tetramethylsilane,
using residual solvent resonances as
internal standards.
UV
-
Vis:
Spectra
were collected using a
Cary 6
0 instrument with Cary WinUV software.
X
-
r
ay Crystallography:
XRD studies were carried out at the Beckman Institute
Crystallography Facility on a Bru
ker
D8 Venture diffractometer (Cu
Kα radiation).
Structures were solved using
direct methods with
SHELXS or SHELXT and refined
against F
2
on all data by full
-
matrix least squares with SHELXL.
5
All of the solutions were
performed in the Olex2 program.
6
The cry
stals were mounted on a glass fiber under
Paratone N oil.
X
-
ray Photoelectron Spectroscopy:
XPS measurements were carried out
in the
Caltech Molecular Materials Resource Center using a Surface Science Instruments M
-
Probe ESCA spectrometer with the sample h
eld at ultra
-
high vacuum (< 2 x 10
9
Torr),
using the Al K α line at 1486.6 eV as a monochromatic X
-
ray source.
XPS measurements
were checked for
surface charging
effects, and the diamond
carbon
(sp3)
1s peak
was
verified to be within ±0.3 eV of
285 eV.
A
full scan from 0
1000 eV was acquired, and
then scans at the binding energies typical for carbon (274
-
294 eV), boron (176
-
196 eV),
nitrogen (389
-
409 eV), oxygen (522
-
542 eV), and iron (690
-
730 eV) were measured.
For
measurements of the BDD plate electrode
after catalysis, the electrode was rinsed with
acetonitrile and air dried prior to data collection.
S
4
S.
2
NMR spectra for electronic structure elucidation
Figure S
1.
1
H NMR spectrum of [(TPA)Fe(
MeCN
)
2
]OTf
2
in CD
3
CN at 25 °C.
Spectrum
also shows 1,3,5
-
trimethoxybenzene used for Evan’s method.
Figure S2.
1,3,5
-
t
rimethoxybenzene signals used for Evan’s method for
[(TPA)Fe(
MeCN
)
2
]OTf
2
in CD
3
CN at 25 °C. [Fe] = 0.011 M, Δf =
3.46
Hz, f = 400.15 MHz,
μ =
0.68
μ
B
.
S
5
Figure S3.
1
H NMR spectrum of [(TPA)Fe(L)
2
]OTf
2
(L = MeCN, NH
3
) formed by mixing
[(TPA)Fe(MeCN)
2
]OTf
2
with 75 equivalents of NH
3
in CD
3
CN at 25 °C.
Spectrum also
shows
1,3,5
-
trimethoxybenzene used for Evan’s method.
Figure S4.
1,3,5
-
t
rimethoxybenzene signals used for Evan’s method for
[(TPA)Fe(L)
2
]OTf
2
(L = MeCN, NH
3
), formed by mixing [(TPA)Fe(MeCN)
2
]OTf
2
with 75
equivalents of NH
3
in
CD
3
CN at 25 °C. [Fe] = 0.011 M, Δf = 200 Hz, f = 400.15 MHz, μ =
5.2μ
B
.
S
6
Figure S
5.
1
H NMR spectrum of [(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
in CD
3
CN at 25 °C.
Spectrum also shows 1,3,5
-
trimethoxybenzene used for Evan’s method.
Figure S
6.
1,3,5
-
t
rimethoxybenzene signals used for Evan’s method for
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
in CD
3
CN at 25 °C. [Fe] = 0.011 M, Δf = 4.62 Hz, f = 400.15
MHz, μ = 0.79μ
B
.
S
7
Figure S
7.
1
H NMR spectrum of [(bpyPy
2
Me)Fe(L)
2
]OTf
2
(L = MeCN, NH
3
) formed by
mixing [(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
with 75 equivalents of NH
3
in CD
3
CN at 25 °C.
Spectrum also shows 1,3,5
-
trimethoxybenzene used for Evan’s method.
Figure S
8.
1,3,5
-
t
rimethoxybenzene signals used for Evan’s method for
[(bpyPy
2
Me)Fe(L)
2
]OTf
2
(L
=
MeCN,
NH
3
),
formed
by
mixing
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
with 75 equivalents of NH
3
in CD
3
CN at 25 °C. [Fe] = 0.011
M, Δf = 11.0 Hz, f = 400.15 MHz, μ = 1.2μ
B
.
S
8
S.
3
UV
-
vis spectra
Figure S
9.
UV
-
vis spectra
of
acetonitrile solution
s
containing 0.12 mM
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
and varying
equivalents
NH
3
in a 1 cm cuvette.
Figure S
10.
UV
-
vis spectra of acetonitrile solution
s
containing 0.12 mM
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
with 250 (black), 400 (blue), 600 (green), or 1000 (red)
equivalents
NH
3
in a 1 cm cuvette. This shows the onset of demetallation around 600
equivalents NH
3
.
S
9
Figure S
11.
UV
-
vis spectra of acetonitrile solution
s
containing 0.12 mM
[(bpy
Py
2
Me)Fe(MeCN)
2
]OTf
2
with 0 (black), 10 (blue), 600 (green), or 1000 (red)
equivalents
NH
3
in a 1 cm cuvette. An arrow highlights an isosbestic point ~500 nm which
is maintained
for 0, 10, and
600 equivalents but has begun to show deviations assigned
to de
metallation at 1000 equivalents.
S
10
S
.4
Catalytic controlled potential coulometry experiments
Procedures for controlled potential coulometry
:
Preparation of the BDD electrode
:
A
10
cm
2
boron
-
doped diamond (BDD) plate
electrode (
IKA
) was
physically attached
to standard electrical wire
coated with chemical
-
resistant insulation
, and the connection was covered with Teflon tape
.
The surface area
of the electrode submerged in solution was
around
4
cm
2
.
BDD has a surface that exists in various s
tates of reduction (H
-
terminated) and oxidation
(O
-
terminated).
7
In order to remove attached nitrogen and iron generated during
controlled potential coulometry experiments
and
to ensure a reliable electrode surface
prior to CPC measurements, the BDD plate
electrode
wa
s oxidatively treated prior to use.
First, the electrode
wa
s soaked in concentrated nitric acid for 5 minutes. Then, a potential
of 3.0 V vs Ag/AgCl
wa
s applied to the BDD electrode in a 0.5 M H
2
SO
4
solution
for 10
minutes. The electrode wa
s then thoroughly rinsed with water prior to use.
Preparation of the platinum counter electrode
: In order to ensure a highly active Pt
surface for
hydrogen evolution
prior to CPC experiments, the Pt mesh counter electrode
was soaked in concentrated hydroc
hloric acid for at least 5 minutes prior to usage.
Preparation of the custom Ag/AgOTf reference electrode
: To ensure reliable potential
measurements, the custom Ag/AgOTf reference electrode was prepared prior to each
CPC experiment and then a CV of ferroc
ene was measured. In a glass tube fitted with a
Vycor
porous glass frit attached by
Teflon heat
shrink tubing
,
an acetonitrile solution
containing 5 mM AgOTf and 0.1 M TBAPF
6
was added
.
A
silver wire was place
d
inside
and the electrode was sealed
.
Preparation of ammonia solutions.
For experiments with natural abundance ammonia,
saturated 2 M solutions
8
in acetonitrile were prepared by bubbling anhydrous ammonia
through acetonitrile in a Schlenk tube under an argon
/ammonia
atmosphere. For
experiments
with labeled
15
NH
3
, ammonia was liberated from
15
NH
4
OTf by addi
tion of 1.1
equivalents of 1,8
-
d
iazabicyclo[5.4.0]undec
-
7
-
ene (DBU) to
15
N labeled
ammonium
triflate dissolved in acetonitrile in a Schlenk tube inside of an argon glovebox. This solution
was then vacuum transferred to
a separate
,
clean Schlenk tube.
CPC:
Inside an argon glovebox, a gas
-
tight electrochemical cell equipped with a 24/40
c
ap containing three tungsten rods for electrical contacts and a
valved
14/20 joint carefully
sealed with a Suba
-
Seal septum was prepar
ed. A BDD plate electrode (A = 4
cm
2
), high
surface area platinum mesh electrode, and custom Ag/AgOTf reference electrode
were
connected to the 24/40 cap. All chemical reagents were then rapidly added to the cell to
prevent evaporation of ammonia
,
and then
,
the cell was sealed with the 24/40 cap. Prior
to each CPC
experiment
, a ZIR and CV were taken.
No
IR compensation was ap
plied
for
CPC measurements
.
The CPC experiment was then conducted for 24 h.
The cell was
then removed from the glovebox for analysis by gas chromatograph
y. For headspace
S
11
analysis, 100 μ
L of the headspace was injected into a
GC
-
TCD for quantification using
a
lockable Hamilton syringe with a 26S gauge needle. For GC
-
MS, only 50
μ
L of the
headspace was injected.
For each experiment, a 10 mL soluti
on containing 0.
0
5 mM
[Fe]
,
20
mM NH
3
(
1.0
mL of
0.2
M solution), and 50 mM NH
4
OTf (8
4
mg) was prepared in acetonitrile.
Reload experiments
:
After a completed CPC experiment, the valved 14/20 joint on the
electrochemical cell was sealed, the septum was removed, and a 14/20 joint
-
to
-
tubing
adapter was connected. This joint was connected to
a double
-
manifold Schlenk line and
put under an argon atmosphere. Under a positive counter
-
flow of argon, the 24/40
electrode cap was removed and replaced with a ground
-
glass stopper. The cell was then
carefully
evaporated to dryness under vacuum and broug
ht into
an argon glovebox. To
this cell containing ammonium triflate electrolyte and used catalyst was added 9 mL
acetonitrile and 1.0 mL of a 0.2 M NH
3
solution to reset the ammonia concentration to the
initial value. Then, the CPC experiment was performed following our standard
procedures.
Table S1
.
Results of catalytic CPC experiments
performed at 0.85 V vs Fc/Fc
+
for 24 h
with 0.05 mM [Fe] and 20 mM
NH
3
(400 equivalents)
.
For entries where reload
experiments were conducted, the entries are listed as x.1 and x.2 for the
initial
and
subsequent reload
experiment, respectively.
LOQ indicates that
the amount of
gas
produced
was below
the limit of quantification.
Entry
Fe Source
Eq. N
2
Charge (C)
FE N
2
(%)
FE H
2
(%)
1
.1
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
89
29.9
86
77
2
.1
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
83
26.5
91
80
3
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
97
31.1
91
93
4*
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
102
36.5
81
84
Avg.
1.1
4
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
93
87
84
1.2
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
55
21.1
77
78
2
.2
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
57
19.7
84
67
Avg. Reload
[(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
56
81
73
3
None (NH
3
only)
LOQ
0.86
LOQ
73
4
.1
FeOTf
2
· 2 MeCN
7.7
8.2
27
74
5
FeOTf
2
· 2 MeCN
7.1
7.5
28
33
Avg. 4
.1
5
FeOTf
2
· 2 MeCN
7.4
28
54
4
.2
FeOTf
2
· 2 MeCN
LOQ
1.2
LOQ
65
* Experiment performed with
15
NH
3
and
15
NH
4
OTf
S
12
NMR spectra showing active catalyst after CPC:
Figure S
12.
1
H NMR spectra of [(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
precatalyst (top) and a
sample extracted after a CPC experiment (bottom)
in CD
3
CN
. The bottom spectrum
shows that active catalyst remains after CPC experiments, as
corroborated by reload
experiments.
S
13
S
.
5
Electrode rinse test after CPC
Figure S
13.
CVs of treated BDD plate electrode and the same electrode after CPC and
rinsing with MeCN
, recorded
in MeCN with 0.05 M NH
4
OTf and 20 mM NH
3
;
P
t counter
and
Ag/AgOTf reference electrodes. The rinsed electrode corresponds to the electrode
after a 2 hour CPC experiment containing the new [(bpyPy
2
Me)Fe(MeCN)
2
]OTf
2
electrocatalyst under the optimized conditions detailed in Table
S1. These
measurements
do not supp
ort the possibility of deposition of catalytically active material onto the
electrode. This is in agreement with our previous investigation of the electrode after CPC
with [(TPA)Fe(MeCN)
2
]OTf
2
.
9
S
14
S.6
GC
-
MS data for
15
NH
3
experiment
Figure S
14.
GC mass spectrum of headspace after CPC experiment using
15
NH
3,
showing
30
N
2
as the major product.
S
15
S
.7
XPS spectra of BDD plate electrode
Figure S
15.
Full XPS spectrum of clean, treated BDD plate electrode and higher
resolution spectra centered on the regions characteristic for Fe 2p and N 1s.
S
16
Figure S
16.
Full XPS spectrum of rinsed BDD plate electrode after 24 h CPC with 0.05
mM [(bpyPy
2
Me)Fe(M
eCN)
2
]OTf
2
and 20 mM NH
3
solution and higher resolution spectra
centered on the regions characteristic for Fe 2p and N 1s.