Journal of the American Chemical Society
Article
−
1
new ν(N−H) stretching at 3072 cm together with a new
EXPERIMENTAL SECTION
■
−
1
σ(N−H) bending at 1697 cm , while the ν(NN) stretching
Catalyst Preparation. BiOCl-OVs was prepared by a solvother-
−
1
35
at 2103 cm increases significantly. This indicates that, as
mal method. Bi(NO ) ·5H O (1.45 g, 3.0 mmol) and KCl (0.22 g,
3
3
2
II
shown in Figure 4a → b → c, N is adsorbed on the two Bi
sites in an end-on/end-on bridging mode
donation from the latter to the respective N atoms, producing
a NN species (b). Fast donation of H atoms from the
adjacent surface −OH to the species creates a diazene
intermediate (c).
3.0 mmol) were stirred in ethylene glycol (16 mL) for 30 min, and the
mixture was left in an autoclave at 433 K for 12 h. The resulting
powder was recovered by centrifugation, washed with water and
ethanol, and dried at 353 K for 24 h. BiOCl with perfect
stoichiometry was prepared in water in a manner similar to that
of BiOCl-OVs.
Photoreaction. The catalyst (0.2 g) and solution (0.1 L) were
added to a glass tube (internal diameter, 45 mm; capacity, 0.2 L).
Water, 550 mM KCl solution, or seawater was used for reactions.
Seawater was prepared by dissolving 33.4 g of red sea salt in 1 L of
2
59,60
by electron
3
4
Figure 5c shows the change in the DRIFTS chart of the N2-
adsorbed BiOCl-OVs under photoirradiation. Broad absorp-
tion bands appear at 1000−1600 cm− and are attributed to
1
+
4
−
the symmetric and antisymmetric deformations of NH and
water to form a solution containing ∼550 mM of Cl , and the
solution pH was adjusted by adding HClO
2
0,36
−1
44
NH3.
A sharp band formed at 2333 cm is assigned to N2
4
.
The reaction solution
6
1
was stirred using a magnetic stirrer and placed under N flow (0.3 L
chemisorbed on the Brønsted acid site, which may be formed
2
−
1
48
−
+
min ), and the tube was photoirradiated by a Xe lamp. The
solution temperature was ∼303 K during photoirradiation. For the
SCC efficiency determination, the reactions were performed using a
by oxidation of interlayer Cl by the VB h (eqs 6 and 9).
These results suggest that the N adsorbed on the surface OVs
is reduced to NH by the CB e , while the VB h oxidize
interlayer Cl . As shown in Figure 4, the diazene intermediate
c) can undergo stepwise reduction and produce NH and O
via oxidation of interlayer Cl , followed by photodecomposi-
tion of HClO. Photoexcitation of BiOCl-OVs with the
intermediate (c) produces e and h pairs. The CB e reduces
the two Bi sites, regenerating them to Bi (d). The VB h
oxidizes interlayer Cl , and the photodecomposition of the
HClO formed produces O and regenerates Cl . The two Bi
sites promote NN reduction and produce a hydrazine
intermediate (e). Further reduction of the strongly adsorbed
intermediate may give amine intermediates (f) without
formation of hydrazine in solution, finally producing NH
with the Bi regeneration (a). The photocatalytic sequence
around the surface OVs involving photodecomposition of
2
−
+
3
solar simulator, whose irradiance was adjusted to the AM1.5 spectrum
−
56
(
1 sun).2 The efficiency was calculated using the following
4
(
3
2
equation:
−
SCC efficiency (%)
−
+
−
−1
[
ΔG for NH formation (J mol )] × [NH formed (mol)]
III
II
+
3
3
=
−
[total input energy (W)] × [reaction time (s)]
−
II
× 100
(13)
2
The total input energy was 0.314 W based on the free energy for NH
3
−
1
generation (339 kJ mol ), the irradiance at 300−2500 nm (1000 W
−2 56
−4
2
m ), and the irradiation area (3.14 × 10 m ). Photographs of the
3
II
+
catalyst was recovered by centrifugation. The NH4 amount in the
HClO facilitates efficient NH production with water as the
electron donor.
resulting water or KCl solutions was analyzed using an ion
chromatograph equipped with a conductivity detector. A 3.5 mM
3
(
COOH) solution containing 2.0 mM 18-crown-6 ether was used as
2
62
the eluent to suppress the interference by the K cation. The NH4+
+
CONCLUSION
■
amount in seawater was determined by UV−visible absorption
26
spectroscopy after the indophenol assay.
We found that ultraviolet light irradiation of the BiOCl-OVs
−
Analysis. Electrochemical measurements were carried out using
catalyst in Cl -containing solution (e.g., seawater) at acidic pH
1
.0 M Na SO solution or 550 mM KCl solution with a Pt wire
2
4
efficiently promotes N fixation with water under ambient
2
(counter) and a Ag/AgCl (reference) electrode. The solution pH was
conditions. The surface OVs act as the sites of N reduction by
2
adjusted with H SO , and N was bubbled through the solution for 10
−
+
−
2
4
2
the CB e . The VB h oxidize the interlayer Cl of the catalyst
self-oxidation). Subsequent photodecomposition of the HClO
formed facilitates O2 evolution. The Cl in solution
min prior to measurements. The working electrode was prepared with
2
6
3
(
a glassy carbon substrate (2 cm ). All potential values are expressed
with respect to the RHE for ease of comparison of the data obtained
−
−
64
compensates for the removed interlayer Cl and suppresses
at different pH values using the following equations:
catalyst deactivation. These reactions facilitate efficient and
E(vs RHE) = E(vs Ag/AgCl) + EAg/AgCl(ref) + 0.0591V × pH
stable N fixation with water as the electron donor. The SCC
2
efficiency of N2 fixation in seawater is 0.05%, which is
comparable to the average efficiency of natural photosynthesis
in plants. This new system has several advantages: (i) naturally
abundant seawater and sunlight are available, (ii) noble-metal-
free photocatalyst made by facile solvothermal method can be
(
EAg/AgCl(ref) = 0.1976V vs NHE at 25°C)
(14)
The DRIFTS charts were obtained using a FT-IR system with an in
situ diffuse reflectance cell. The catalyst (20 mg) was added to the
cell, which was then evacuated (10 Pa) at 303 K for 6 h. Water vapor
65
used, and (iii) storable and transportable NH solution can
3
directly be obtained. For practical application of the present
process, there is several challenging issues such as generation of
pure N gas from air containing a large amount of O , which
(42 μmol) and N (42 μmol) were injected into the cell at 100 K, and
measurements were started in the dark. The spectra were then
monitored under photoirradiation by a Xe lamp at λ > 300 nm and
2
2
2
1
00 K. The ESR spectra were recorded on the Bruker EMX-10/12
suppresses N reduction, and separation of NH from seawater
2
3
66
spectrometer. The catalyst (20 mg) was added to a quartz tube,
containing a large amount of electrolytes. Nevertheless, the
which was evacuated at 298 K for 6 h, and analyzed at 77 K. N (20
2
new concept for N fixation based on a cheap oxychloride
2
Torr) was added to the tube at 298 K and left for 6 h, and the tube
was analyzed at 77 K. The instruments and procedures used for the
DR UV−visible spectroscopy, XRD, XPS, and SEM measurements are
catalyst and seawater presented herein may contribute to the
design of new artificial photosynthesis systems for clean solar
fuel production.
67
described in our previous work.
G
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX