S.-i. Naya, et al.
CatalysisCommunications142(2020)106044
As shown by yellow lines, the lattice fringe has an interval of 2.65 Å,
which is in agreement with the d-spacing of SnO2(101). Also, each
particle in the TEM image is an aggregate consisting of sevaral primary
nanocrystals. Fig. 1c shows that the primary nanocrystal size distributes
4000 spectrometer with an integrating sphere were used for recording
the diffuse reflectance UV–Vis-NIR sepectra. The reflectance (R∞) was
recorded with respect to a reference of BaSO4, and the Kubelka-Munk
function [F(R∞)] expressing the relative absorption coefficient was
calculated by the eq. F(R∞) = (1 - R∞)2/2R∞
.
6.3
1.4 nm. Fig. 1d shows X-ray diffraction patterns (XRD) for the
samples. Every sample has peaks at 2θ = 26.88°, 34.08°, 38.10°, and
52.16° assignable to the diffraction from the (110), (101), (200), and
(211) crystal planes of SnO2 with the rutile structure (ICDD No.
01–070-6995), respectively. No signal of antimony oxide phase is ob-
served in the patterns of the Sb-doped samples. The broadening of the
peaks is indicative of small crystalline domain sizes. The crystallite sizes
of SnO2, SnO2: Sb (1), and SnO2: Sb (12) were calculated to be 6.3 nm,
6.2 nm, and 5.9 nm, respectivley, by the Scherrer equation from the
full-width at half maximum of the (110) diffraction peak. The value is
in good agreement with the one determined by the HR-TEM observa-
tion.
Fig. 2a shows diffuse reflectance UV–Vis-NIR absorption spectra of
SnO2: Sb with different doping amount, and the solar spectrum for
comparison. In the spectrum for every sample, the absorption rises
around 440 nm in spite that the band gap of stoichiometric SnO2 is
~3.6 eV [22]. The absorption tail can result from the existence of
oxygen vacancy [23]. Doping of Sb into SnO2 nanocrystals causes the
strong absorption in the red and NIR region. Also, the absorption edge
of the interband transition is slightly blue-shifted by Sb-doping prob-
ably due to the Moss-Burstein effect raising the conduction band (cb)
minimum [24]. X-ray photoelectron (XP) spectra were measured for
SnO2: Sb with different doping level. In the Sn 3d-XP spectra for SnO2:
Sb samples (Fig. S2), two signals due to the emission from the Sn 3d3/2
and Sn 3d5/2 orbitals of SnO2 are observed regardless of xSb at the
binding energies (EB) = 494.7 and 486.6 eV, respectively [25]. In the
Sb 3d-XP spectra (Fig. 2b), the Sb 3d3/2 signal is present at
EB = 539.8 eV, which indicates that the Sb in SnO2 mainly exists as
Sb5+ ion [26,27]. The signal of Sb 3d5/2 at EB ≈ 531 eV is overlapped
with that of O 1 s orbital. Thus, Sb-doping increases the density of free
electrons, and the NIR-absorption in Fig. 2a is induced by the LSPR of
which intensity increases with increasing xSb [1].
2.2. Photocatalytic oxidation of amine
After the suspention of SnO2:Sb or TiO2 (200 mg) in an acetonitrile
solution of benzylamine (100 μM, 200 mL) had been stirred at room
temperature (~ 20 °C), irradiation was started using a 300 W Xe lamp
(λ > 300 nm, HX-500, Wacom). The light intensity integrated from
320 to 400 nm (I320–400) was set to 2.6 mW cm−2. After removing
particles by membrane filter, the amounts of products were quantified
by high performance liquid chromatography (LC-6 CE, SPD-6 A, C-R8A
(Shimadzu)) [measurement conditions: column = Shim-pack CLCODS
(4.6 mm × 150 mm) (Shimadzu); mobile phase acetonitrile; flow
rate = 1.0 mL min−1; λ = 280 nm]. For evaluating the temperature
dependence, the similar reactions were performed under the tempera-
ture controlled conditions from 15 to 42 °C by using double jacket with
circulating water.
2.3. Photoelectrochemical measurement
Mesoporous (mp) nanocrystalline film of SnO2:Sb (12) was formed
on fluorine-doped tin oxide (FTO, Aldrich, TEC 7). SnO2:Sb (12) was
dispersed into a solution of of Triton X-100 (0.1 mL), acetylacetone (1
drop), and polyethylene glycol 20,000 (0.2 g) in H2O (0.4 mL), and
grinded to obtain the uniform paste. By doctor blade, the paste was
coated on FTO with 60 μm thick, and the sample was calcined at 773 K
for 1 h to obtain mp-SnO2:Sb/FTO electrode. Photocurrent response
was examined by the standard three-electrochemical cell with the
structure of mp-SnO2:Sb/FTO (working electrode) │ 0.1 mM benzyla-
mine +0.1 M Bu4N∙ClO4 acetonitrile solution │ Ag/AgCl (reference
electrode) │ glassy carbon (counter electrode). The working electrode
was illuminated by monoclomatic light using LED lamp. The incident
photon-to-current efficiencies (IPCE) were calculated by Eq. 2.
3.2. Photocatalytic amine oxidation
Jph (E)NA hc
IPCE[%] =
× 100
(2)
IFλ
The photocatalytic activity of SnO2: Sb for benzylamine oxidation
was studied. The photocatalytic activity was compared with that of
TiO2 as the representative semiconductor photocatalyst [28]. SnO2: Sb
nanocrystals were dispersed into an acetonitrile solution of benzyla-
mine, and stirred at room temperature in the dark for 0.5 h. Then, light
was irradiated to the aerated suspension without cooling by using Xe
lamp. Whereas no reaction occurred in the dark, photoirradiation of the
solids (SnO2: Sb, SnO2, and TiO2) progresses the oxidation of benzyla-
mine to yield benzaldehyde via the hydrolysis of the intermediate imine
by residual water in the solvent (eq. 3) [29].
where Jph(E) is the photocurrent at an electrode potential of E (dark rest
potential), NA is Avogadro constant, h is Planck constant, and c is speed
of light, I (W cm−2) is light intensity, F is Faraday constant.
2.4. Adsorption property
Adsorption amounts were obtained by exposing SnO2:Sb or TiO2
(100 mg) to an acetonitrile solution of benzaldehyde or benzylamine
(100 μM, 10 mL) at 25 °C or 45 °C for 3 h in the dark. After removing
particles by membrane filter, the concentration of the remaining ben-
zaldehyde or benzylamine was determined by HPLC.
PhCH2NH2 → [PhCH=NH] → PhCH = O
(3)
Fig. 3a shows time courses for benzaldehyde generation under ir-
radiation of light with wavelength (λex) > 300 nm in the presence of
SnO2: Sb, and SnO2 and TiO2 for comparison. TiO2 shows high activity,
but the maximum yield of benzaldehyde is lower than 75%. On the
other hand, in the non-doped SnO2 and SnO2: Sb (1) systems, the yield
at irradiation time = 24 h is only ~60%. Strikingly, SnO2: Sb (12)
exhibits much higher photocatalytic activity to generate benzaldehyde
with a yield of ~90% at irradiation time = 24 h. Further, the yield
continued to increase to reach 98% at irradiation time = 37 h. Irra-
diation of TiO2 generates CO2, which is hardly generated in the SnO2:
Sb (12) system (Fig. S3). Thus, the low yield of benzaldehyde in the
TiO2 system can be attributed to the overoxidation. To check the cat-
alyst stability, 24 h-photocatalytic oxidation reaction of benzylamine
was repeated 5 times using the catalyst particles retrieved from the
suspension after each reaction (Fig. S4). No significant decay of the
3. Results and discussion
3.1. Catalysts characterization
SnO2: Sb particles with Sb-mole fraction (xSb / mol% = {Sb/
(Sn
+
Sb)}
×
g
100)
=
0, 1.0, 11.6 (specific surface
area = 70–80 m2
−1, SN-100P, Ishihara Sangyo), and for comparison,
TiO2 (anatase, SSP-M, Sakai Chemical Industry, mean particle
size = 15 nm, specific surface area = 100 m2g−1) were used as the
photocatalysts. The SnO2 samples doped with xSb = 1.0% and 11.6%
are designated as SnO2: Sb (1) abd SnO2: Sb (12), respectively. Fig. 1a
shows transmission electron microscopy (TEM) image of SnO2: Sb (12).
Almost sphere particles with an apparent size of ~9 nm are observed
(Fig. S1). Fig. 1b shows high resolution (HR)-TEM image of the sample.
2
Naya, Shin-ichi
