W. Wu et al. / Journal of Catalysis 361 (2018) 222–229
223
(DMF) three times. Finally, the light brown product denoted as
CN-T was filtered and dried at 60 °C. Similarly, other control sam-
ples of CN-P (melem and BPDA without benzophenone structure
were mixed at the ratio of 1:1) and ME-T (melamine and BTDA
were mixed at the ratio of 1:1) were prepared using the same
method.
2.2. Materials characterization
The morphologies of the samples were characterized by scan-
ning electron microscopy (SEM) (Hitachi S-4800, Japan) and trans-
mission electron microscopy (TEM) (JEM-2100UHR, Japan). XRD
patterns were obtained on a powder X-ray diffractometer at 40
Scheme 1. The modified Jablonski diagram of the photoexcitation energy transfer
process of a photosensitizer (PS) that undergoes efficient intersystem crossing (ISC)
to enable triplet–triplet energy transfer (TTET) to generate singlet oxygen (1O2). The
functionalization of PS with compounds that undergo ISC (spin converters) is one
kV and 15 mA using Cu K
a radiation (X’Pert PRO MPD, Holland).
X-ray photoelectron spectroscopy (XPS) measurements were per-
formed on an ESCALAB 250Xi spectrometer equipped with a prere-
approach to producing 1O2. The smaller singlet-triplet energy gap (
efficient ISC process and 1O2 generation.
DEST) benefits
duction chamber. Solid-state 13C nuclear magnetic resonance (13
C
NMR) was measured on a Bruker Advance III 400 M spectrometer
equipped with a 9.4 T magnet. The UV–Vis spectra were recorded
on a UV–Vis spectrophotometer (UV-2700, Shimadzu, Japan). The
1O2 emission signal of the CN-T photosensitizer was detected in a
fluorescence spectrometer FLS980 (Edinburgh Instruments Ltd.)
with a 450 W Xe lamp and a NIR detector. The sample was excited
at 360 nm. The experimental conditions were excitation and emis-
sion bandwidths of 15 nm, integration time 2 s/step (0.1 s dwell
time ꢀ 20 repeats) and step of 1 nm for the sample. The ns-
Based on this principle, we used precursors of carbon nitride
(melamine and melem) as ligands of heavy atom-free spin convert-
ers (benzophenone) to explore their structure-property relation-
ship for enhancing 1O2 generation. The heavy atom-free spin
converter doped carbon nitride was prepared via the facile copoly-
merization of 3,30,4,40-benzophenonetetracarboxylic dianhydride
(it has the benzophenone structure to function as the spin con-
verter) and a carbon nitride precursor. The structures were con-
firmed by TEM, SEM, solid-state 13C NMR, XPS and XRD. The
modified carbon nitride is shown to greatly enhance 1O2 genera-
tion and selective photooxidation of 1,5-dihydroxynaphthalene
(1,5-DHN). 1,5-DHN is a substance that contaminates water, but
its product (5-hydroxy-1,4-naphthalenedione, Juglone) obtained
after oxidation has hemostatic and antibacterial activity. Further-
more, ESR, steady state/time-resolved luminescence spectroscopy
and DFT calculations were used to explore the structure-property
relationship for enhancing 1O2 generation. This work provides a
basis for broadening the application of carbon nitride in the field
of selective photooxidation due to simple operation, low cost and
high efficiency. More importantly, the heterogeneous photocataly-
sis process enables the product to be easily obtained by simple fil-
tration, which is beneficial for industrial applications.
domain time-resolved fluorescence spectra and the
ls-domain
time-resolved spectra were also obtained on an FLS980 fluores-
cence spectrometer (Edinburgh Instruments Ltd.). Triplet emission
spectra and lifetimes were also characterized with a Horiba Jobin-
Yvon fluorolog spectrometer system.
2.3. Photooxidation 1,5-DHN test
The photochemical reaction was performed at room tempera-
ture under air atmosphere in a round bottom flask (50 mL) with
irradiation by a 35 W xenon lamp (600 W/m2). The MeCN/H2O
(v/v = 5:1) mixed aqueous solution (20 mL) containing 1.0 ꢀ 10ꢁ4
mol/L of 1,5-DHN and photocatalysts (20.0 mg) was irradiated at
k > 385 nm (the light with a wavelength shorter than 385 nm
was blocked by 0.72 M NaNO2 solution). UV–Vis absorption spectra
were used to record at intervals of 10 min. The consumption of 1,5-
DHN was monitored by a decrease in the absorption at 331 nm,
and the concentration of 1,5-DHN was calculated by using its
2. Experimental
molar extinction coefficient (e
= 7664 Mꢁ1 cmꢁ1). Juglone produc-
tion at intervals of 10 min was monitored by an increase in the
2.1. Materials preparation
absorption peak at 419 nm. The concentration of Juglone was also
3,30,4,40-benzophenonetetra-carboxylic dianhydride (BTDA), 3,
3,4,40-biphenyltetracarboxylic dianhydride (BPDA) and melamine
were obtained from Aladdin Industrial Corporation. All the chemi-
cals in our experiment were directly used without further
purification.
calculated by using its molar extinction coefficient (e
= 3567 Mꢁ1
cmꢁ1), and the yield of Juglone was calculated according to the fol-
lowing eqn:
100 ꢀ AiðJugloneÞ
=
eðJugloneÞ
Yield ¼
ꢀ 100%
Melem was synthesized by heating melamine placed in a porce-
lain crucible with a cover at 425 °C for 4 h in a muffle furnace [23].
Similarly, graphitic carbon nitride (g-C3N4) was synthesized by
heating melamine at 550 °C for 4 h in a muffle furnace [24].
Photocatalysts based on carbon nitride were synthesized by
facile thermal condensation of melem and BTDA. In detail, 1 g of
powder mixture of melem and BTDA (the molar ratios of melem
and BTDA were 5:1, 3:1, 1:1, 1:3, 1:5, respectively) was put into
a railboat and heated at 5 °C/min up to 300 °C for 4 h under the
protection of N2. The samples (the molar ratio of melem and BTDA
at 1:1) were calcined under different temperature. The optimum
temperature was chosen at 300 °C (Table S1). In order to avoid
the influence of unreacted BTDA, the resultant block solid was trea-
ted with ultrasound and washed with N,N-dimethylformamide
Cinitialð1;5ꢁDHNÞ
Ai(Juglone) is the absorbance of Juglone in solution; e(Juglone) is the
molar extinction coefficient of Juglone; Cinitial(1,5-DHN) is the initial
concentration of 1,5-DHN.
2.4. Electron spin resonance (ESR) spectroscopy
Electron spin resonance (ESR) spectra were recorded at room
temperature using a JEOL JES FA200 spectrometer at 9.8 GHz, X-
band, with 100 Hz field modulation. Samples were quantitatively
injected into specially made quartz capillaries for ESR analysis in
the dark and illuminated directly in the cavity of the ESR spectrom-
eter. In the whole test process, 2,2,6,6-tetramethylpiperidine