Asymmetric zinc porphyrin derivatives bearing three pseudo-pyrimidine: Meso -position substituents and their photosensitization for H2evolution

Article abstract of DOI:10.1039/d0nj02056g

Novel asymmetric zinc porphyrin derivatives (ZnPy-5 and ZnPy-6) with meso-positions bearing one benzoic acid and three pseudo-pyrimidines with two N atoms located at different positions were synthesized and utilized as sensitizers for Pt-loaded g-C3N4 (PCN). Compared to the analogue (ZnPy-1 bearing one benzoic acid and three phenyl meso-position substituents), ZnPy-5 and ZnPy-6 exhibit significantly enhanced photosensitization on PCN under visible light (λ ≥ 420 nm) illumination. In particular, ZnPy-5/PCN and ZnPy-6/PCN exhibit H2 evolution activities of 418 and 585 μmol h-1, corresponding to turnover numbers (TON) of 8845 and 12?381 h-1, respectively. Both of these values are much better that (316 μmol h-1) of ZnPy-1/PCN, which has a TON of 6687 h-1. In addition, ZnPy-5/PCN and ZnPy-6/PCN show apparent quantum yields of 32.6% and 33.1% at 420 nm monochromatic light, and these are much higher than that (10.6%) for ZnPy-1/PCN. Compared with ZnPy-5, the two N atoms of the pseudo-pyrimidines in ZnPy-6 are further away from the porphyrin macrocycle, which can more effectively combine with the sacrificial reagent and g-C3N4, thus promoting dye regeneration and the photoexcited charge transfer for delivering better photocatalytic performance. The present results demonstrate that the number and positions of the N atoms in the peripheral substituents of the porphyrin derivatives have a great influence on the photosensitization, and that the fine-tuning of molecular structures is crucial for improving the H2 evolution activity of the dye-sensitized semiconductors.

Full text of DOI:10.1039/d0nj02056g

NJC
View Article Online
View Journal
PAPER
Asymmetric zinc porphyrin derivatives bearing
three pseudo-pyrimidine meso-position
substituents and their photosensitization
Cite this:DOI: 10.1039/d0nj02056g
for H evolution†
2
ab
b
b
b
b
Peng Zeng,‡ Ya Zheng,‡ Shengtao Chen, Haoran Liu, Renjie Li * and
b
Tianyou Peng
*
Novel asymmetric zinc porphyrin derivatives (ZnPy-5 and ZnPy-6) with meso-positions bearing one
benzoic acid and three pseudo-pyrimidines with two N atoms located at different positions were
3 4
synthesized and utilized as sensitizers for Pt-loaded g-C N (PCN). Compared to the analogue (ZnPy-1
bearing one benzoic acid and three phenyl meso-position substituents), ZnPy-5 and ZnPy-6 exhibit
significantly enhanced photosensitization on PCN under visible light (l Z 420 nm) illumination.
ꢀ
1
In particular, ZnPy-5/PCN and ZnPy-6/PCN exhibit H
2
evolution activities of 418 and 585 mmol h ,
ꢀ1
corresponding to turnover numbers (TON) of 8845 and 12 381 h , respectively. Both of these values are
ꢀ1
ꢀ1
much better that (316 mmol h ) of ZnPy-1/PCN, which has a TON of 6687 h . In addition, ZnPy-5/PCN
and ZnPy-6/PCN show apparent quantum yields of 32.6% and 33.1% at 420 nm monochromatic light,
and these are much higher than that (10.6%) for ZnPy-1/PCN. Compared with ZnPy-5, the two N atoms
of the pseudo-pyrimidines in ZnPy-6 are further away from the porphyrin macrocycle, which can more
effectively combine with the sacrificial reagent and g-C
photoexcited charge transfer for delivering better photocatalytic performance. The present results
demonstrate that the number and positions of the atoms in the peripheral substituents of
the porphyrin derivatives have a great influence on the photosensitization, and that the fine-tuning
3 4
N , thus promoting dye regeneration and the
Received 23rd April 2020,
Accepted 29th May 2020
N
DOI: 10.1039/d0nj02056g
of molecular structures is crucial for improving the H
semiconductors.
2
evolution activity of the dye-sensitized
rsc.li/njc
TiO with a wide bandgap can only absorb UV photons that
account for merely 4–5% of sunlight. Therefore, great efforts
2
Introduction
1
,6
As an ideal energy carrier, hydrogen (H
2
) will be the main have been made to explore new photocatalysts, enabling the
energy source in the foreseeable future since it can be produced efficient utilization of visible light since it represents B50% of
1
,6
through water-splitting using a variety of energy sources such sunlight (AM1.5).
as fossils, nuclear energy and solar energy. Among the various Among the many strategies used to harvest visible light,
hydrogen production techniques, photocatalytic H evolution photosensitization is an effective way to extend the spectral
2
6
,13–19
from water-splitting using sunlight has attracted considerable response range of the semiconductors with a wide bandgap.
1
–5
attention due to it being cost-effective and eco-friendly. Up to As a light-harvesting center, the sensitizer must have a series of
now, inorganic semiconductors such as metal oxides, metal basic requirements, including a broad spectral response range, an
sulfides and metal nitrides have been the most extensively used appropriate redox potential for efficient electron injection and dye
6
–12
6,13–17
photocatalysts.
Among them, the most widely investigated regeneration, long-term irradiation stability and low cost.
Currently, a number of organic compounds and metal complexes
1
3
14
a
such as Eosin Y (EY), poly(3-hexylthio-phene) (P3HT), and
School of Food and Pharmaceutical Engineering, Zhaoqing University,
Zhaoqing 526061, P. R. China
15,16
metal phthalocyanines (MPc)
have been used as a sensitizer
b
College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072,
P. R. China. E-mail: lirj@whu.edu.cn, typeng@whu.edu.cn
for semiconductors with a wide bandgap to achieve a visible-light-
responsive ability for H2 evolution. Metal porphyrin and its
derivatives have also attracted great interest as sensitizers
in mimicking the natural photosynthesis due to their strong
photo-absorption ability in the visible region (with a strong
1
†
Electronic supplementary information (ESI) available: MALDI-TOF and H MNR
spectra, cyclic voltammograms, photoluminescence and UV-vis absorption spectra.
See DOI: 10.1039/d0nj02056g
‡
These authors contributed to this work equally.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020
New J. Chem.

View Article Online
Paper
NJC
B-band around B420 nm and moderate Q-bands in the range sample, while ZnPy-5 and ZnPy-6 contain one benzoic acid and
1
7,18
of 500–650 nm) and thermal stability.
For example, an iron three pseudo-pyrimidines, the two N atoms of which are located
porphyrin derivative ((FeTPP) O) was found to be an efficient at different positions, and the two N atoms of the pseudo-
2
sensitizer of g-C N for H evolution due to its p–p interactions pyrimidines in ZnPy-6 are further away from the porphyrin
3
4
1
2
8
and Fe–N bonding.
macrocycle compared with those in ZnPy-5. The experimental
Recently, asymmetric zinc porphyrins (ZnPy) bearing one results show that both ZnPy-5 and ZnPy-6 deliver much better
benzoic acid and three pseudo-pyridine peripheral substituents photosensitization on PCN than ZnPy-1 under visible light
were utilized by our group to sensitize Pt-loaded g-C
3
N
4
(l Z 420 nm) illumination, and that ZnPy-6/PCN delivers
2
0–22
for visible-light-driven H
2
evolution.
It was found that the higher H evolution activity than ZnPy-5/PCN. The above differ-
2
different steric positions of the N atom in the pseudo-pyridines ences in the photosensitization of these ZnPy derivatives on
resulted in different interactions between the porphyrin and PCN and the effect of the number and positions of the N atoms
ascorbic acid (AA) as the sacrificial reagent, and this strongly in the peripheral substituents of the porphyrin on the photo-
influenced the photocatalytic activity. Among them, the ZnPy activity were studied in detail, in order to investigate more
derivative with the lowest steric hindrance of the N atom in the efficient porphyrin-based sensitizers and their regeneration
pseudo-pyridines delivered the best photosensitization. When processes in artificial photosynthesis.
the three pseudo-pyridines were substituted with phenyls,
however, the corresponding ZnPy derivative showed signifi-
2
0,21
cantly decreased photosensitization.
One possible reason
Results and discussion
Spectroscopic analyses
for this is that the two lone-pair electrons and basicity of the N
atom of the pyridine increase the electron cloud density of the
ZnPy molecules and therefore its interaction with AA and The obtained ZnPy derivatives (ZnPy-5 and ZnPy-6) were identified
1
3 4
g-C N . This can promote the charge transfer and dye regenera- using spectroscopic characterization techniques ( H nuclear
tion, and then promote the photoactivity of the dye-sensitized magnetic resonance (NMR) spectroscopy; matrix assisted laser
system. Similarly, a pyrimidine with two N atoms that was desorption/ionization time-of-flight mass spectrometry (MALDI-
introduced into a dye (INPA) linked to a carboxylic acid facili- TOF-MS); ultraviolet visible (UV-vis), Fourier-transform infrared
tated the interaction between the dye and TiO
2
, and therefore (FTIR) and fluorescence spectroscopies) and elemental analysis
enhanced the photovoltaic performance of the dye-sensitized (see Fig. S1–S8 and the Experimental section, ESI†). UV-Vis
2
3
solar cells (DSSCs). Density functional theory (DFT) calcula- absorption spectra (Fig. 1a) show that the metal-free porphyrin
tions also demonstrated that when a pyrimidine group is derivative (H Py-5) in DMF solution has five obvious absorption
2
introduced in a p-type sensitizer of DSSCs, it can result in bands with maximum absorption wavelengths (lmax
) at
an increased charge transfer ability compared to with the B416 nm (B-band), 511 nm (Q-band), 546 nm (Q-band),
2
4
introduction of a thiophene group. These results imply that 587 nm (Q-band) and 642 nm (Q-band). H
pyrimidines (with two N atoms) as peripheral groups, when similar absorption spectrum but with a bathochromic shift of
introduced in ZnPy derivatives, might more effectively increase B2 nm relative to the corresponding absorption bands of H Py-5.
After coordination with Zn , the resulting ZnPy-5 and ZnPy-6
action with AA, and thus would be a benefit to dye regeneration exhibit very similar absorption spectra (Fig. 1b) with one strong
and photo-sensitization on semiconductors. B-band in the range of 350–450 nm, due to the a (p) to e *(p)
Herein, novel asymmetric zinc porphyrin derivatives (ZnPy-5 electronic transitions and two less intensive Q-bands in the range
and ZnPy-6) with meso-positions bearing one benzoic acid of 450–670 nm that arise from the a2u(p) to e *(p) electronic
In particular, ZnPy-5 and ZnPy-6 show respective
synthesized, and they were then utilized as sensitizers for B-bands at a lmax of B425 and B427 nm (Fig. 1b), with both of
Pt-loaded g-C (hereafter denoted as PCN) for visible-light- them redshifted by B9 nm compared with those of H Py-5 and
2
Py-6 displays a
2
2+
the electron density of the porphyrin macrocycle and its inter-
1u
g
g
20–22
and three pseudo-pyrimidines (Scheme 1) were designed and transitions.
3
N
4
2
driven H evolution. In Scheme 1, ZnPy-1, with one benzoic acid H Py-6 (Fig. 1a), respectively.
2
2
and three phenyl meso-position substituents, is the contrast
Similar to the metal-free porphyrins (H Py-5 and H Py-6),
2 2
the absorption spectrum of ZnPy-6 also shows a bathochromic
shift of B2 nm with respect to that of ZnPy-5, and this can be
ascribed to the different positions of the two N atoms in the
pseudo-pyrimidine substituents (Scheme 1). Compared with
2 2
ZnPy-5 (or H Py-5), ZnPy-6 (or H Py-6) has the two N atoms
in the peripheral pseudo-pyrimidines further away from the
porphyrin macrocycle, and this could weaken the electron-
withdrawing ability of the pseudo-pyrimidines and increase
the electron cloud density of the porphyrin macrocycle, thus
enhancing the photo-absorption ability, as shown in Fig. 1.
The FTIR spectra (Fig. 2) indicate that the porphyrin macro-
cycles have their characteristic IR absorption bands in the range
Scheme 1 Molecular structures of the ZnPy derivatives.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
2 2
Fig. 2 FTIR spectra of the metal-free porphyrins (H Py-5 and H Py-6) and
zinc porphyrins (ZnPy-5 and ZnPy-6).
2
Fig. 1 UV-vis absorption spectra of (a) the metal-free porphyrins (H Py-5
and H
2
Py-6) and (b) the zinc porphyrins (ZnPy-5 and ZnPy-6) in DMF
solution.
Fig. 3 UV-vis diffusion reflectance absorption spectra (DRS) of PCN and
its dye-sensitized products (ZnPy-5/PCN and ZnPy-6/PCN).
ꢀ
1
of 700–799 cm , originating from the out-of-plane bending vibra-
tion of QC–H in the pyrrole rings, and the characteristic IR peak
ꢀ1
at 793 cm can be assigned to the p-disubstituted benzene of derivative in DMF solution (Fig. 1a). The slight redshifts
the porphyrin derivatives. In addition, the broad IR peak near (B8 nm) and enhanced peak intensities of those B-band and
ꢀ
1
3
426 cm can be attributed to the stretching vibration of O–H in Q-band absorptions may be caused by the p–p stacking inter-
20
the –COOH group and the H-bond between the –COOH groups.
The metal-free porphyrin derivatives (H Py-5 and H Py-6) exhibit IR their similar p-conjugated and planar structures.
peaks at 952 and 965 cm , which can be ascribed to the N–H Furthermore, the slight redshifts can be due to the chemical
vibration in the pyrrole rings (Fig. 2a), but the peaks disappear after interaction between the –COOH of the ZnPy derivatives and the
the coordination with Zn for ZnPy-5 and ZnPy-6. In addition, terminal –NH of g-C
3 4
action between the ZnPy molecules and g-C N in terms of
1
5,25–27
2
2
ꢀ
1
2+
9,21
3
N
4
via a condensation reaction.
With
ZnPy-5 and ZnPy-6 display their featured stretching vibration band the same dye-loading content, ZnPy-6/PCN delivers slightly
ꢀ
1
of N–Zn at 995 cm . In particular, the pseudo-pyrimidines in the higher peak intensities than ZnPy-5/PCN. The possible reason
porphyrin derivatives create the IR peaks that are mainly located at for this is that the two N atoms of the pseudo-pyrimidines in
ꢀ
1
B1566 and B1408 cm , and these positions are slightly different ZnPy-6 are further away from the macrocycle, and this results in
due to the different positions of the two N atoms in the pseudo- the electron-withdrawing ability not being as strong as that of
pyrimidines of the ZnPy derivatives.
ZnPy-5, thus the macrocycle of ZnPy-6 has more concentrated
UV-vis diffuse reflectance absorption spectra (DRS) of the electron density than the macrocycle of ZnPy-5. This then
ZnPy derivative (ZnPy-5 and ZnPy-6) sensitized Pt-loaded g-C
3
N
4
causes ZnPy-6 to be more easily adsorbed and aggregated on
(PCN) complexes are depicted in Fig. 3. As seen in the spectra, PCN, enhancing the photo-absorption ability.
the pristine PCN exhibits an absorption edge at B450 nm in
addition to an absorption tail that is probably due to the n–p*
Energy band structure analyses and theoretical calculations
2
0–22
electronic transitions.
ZnPy-5/PCN and ZnPy-6/PCN display Fig. 4 depicts the normalized UV-vis absorption and fluores-
the absorption features of the combination of PCN with cence (FL) spectra of ZnPy-5 and ZnPy-6 at the same concen-
the ZnPy derivatives, but the adsorption bands in the range tration in DMF solution. As seen from the spectra, both ZnPy-5
of 420–650 nm are much broader than their respective ZnPy and ZnPy-6 have a maximum emission wavelength of 610 nm at
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020
New J. Chem.

View Article Online
Paper
NJC
Table 1 Spectroscopic and electrochemical data of the obtained ZnPy
derivatives
a
Dye
lmax/nm
log e
Eox/V (vs. NHE)
E0–0/eV
E* /V
ZnPy-5
ZnPy-6b
ZnPy-1
425
427
426
5.87
5.90
5.77
0.90
0.89
0.91
2.06
2.05
2.05
ꢀ1.16
ꢀ1.16
ꢀ1.14
a
b
Calculated with formula, E* = Eox ꢀ E0–0
.
Data from ref. 20.
Fig. 5 Experimental values of the molecular orbital energy levels of the
ZnPy derivatives and the VB/CB levels of g-C
3 4
N .
Fig. 4 Normalized UV-vis absorption and fluorescence emission spectra
of ZnPy-5 and ZnPy-6 in DMF solution with lex = 426 nm.
DFT calculations were carried out on the ZnPy derivatives at
the B3LYP/6-31G(d) level to explore their energy level distribu-
tion and electron transfer. The calculated frontier orbital
energy levels (Fig. 6) indicate that the HOMOꢀ2 electrons of
the ZnPy derivatives are mainly concentrated around the metal
center, and the HOMOꢀ1 and HOMO electrons are delocalized
lex = 426 nm. It has been reported that the optical bandgap
(E0–0) of porphyrins can be estimated from the intersection
3
4
over the porphyrin marcocycles. On the other hand, the
LUMO+1 and LUMO electrons of the p* orbitals are delocalized
across the porphyrin macrocycle and move towards the
carboxyl group, whereas the LUMO+2 electrons are mainly
localized over the benzene ring and the carboxyl group. These
frontier molecular orbitals of ZnPy-5 and ZnPy-6 demonstrate
that the photoexcited electrons can transfer from the metal
point wavelength (lint.) of the normalized absorption and
emission spectra (Fig. 4) using the formula,
E
0–0
=
2
7,28
(1240/lint.).
ZnPy-5 and ZnPy-6 exhibit intersection point
wavelengths at 601 and 604 nm, corresponding to E0–0 values of
2.06 and 2.05 eV (Table 1), respectively.
From the cyclic voltammograms (CV in Fig. S9, ESI†) of
the ZnPy derivatives in DMF solution, corresponding electro-
chemical data can be estimated by a tangent method and the
3
4
center of porphyrin towards the carboxyl group, and this
would facilitate the directional injection of electrons into
2
9
data are summarized in Table 1, in which Eox (V vs. NHE) is
calculated from the Leeuw formula [Eox (V vs. Ag/AgNO
2
0,25–27
3 4 2
chemically bonded g-C N for H evolution.
3
) ꢀ E1/2
+
+
+
30–32
(
Fc /Fc vs. Ag/Ag ) + E (Fc /Fc vs. NHE)].
The highest
1
/2
occupied molecular orbital (HOMO) of the ZnPy derivative
2
0–23
corresponds to the Eox value,
and the lowest unoccupied
molecular orbital (LUMO) can be estimated using the equation,
E* = Eox ꢀ E0–0 (Table 1). Based on these results, the corres-
ponding experimental LUMO/HOMO levels of the ZnPy deriva-
tives can be plotted as shown in Fig. 5, in which the energy
band structure, including the conduction band (CB) and
valence band (VB), of g-C N is also depicted. It can be seen
3
4
that both ZnPy-5 and ZnPy-6 have a LUMO energy level (E*) of
ꢀ
1.16 V, which is slightly more negative than the CB (ꢀ1.12 V)
33
of g-C
3
N
4
.
This indicates that the photogenerated electrons of
these excited state ZnPy derivatives can readily be injected into
Fig. 6 Frontier molecular orbitals of ZnPy-1, ZnPy-5 and ZnPy-6, fully
the CB of g-C
3
N
4
due to thermodynamics.
optimized at the B3LYP/6-31G level.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Photocatalytic performance analyses
spectra (Fig. S10a and b, ESI†). It can be seen that both ZnPy-5
and ZnPy-6 in DMF solution exhibit a strong emission band at
lex = 426 nm, implying fast charge recombination of their excited
states. After addition into PCN, the emission intensities of both
ZnPy-5 and ZnPy-6 were significantly reduced, indicating that the
charge recombination rate of the porphyrin derivatives can
The photoreaction conditions can influence the photoactivity of
4
,6,35
a dye-sensitized semiconductor.
Considering the similarity
and our
we adopted a similar photoreaction
condition, as follows: 15 mg of 1.5 wt% Pt/g-C (PCN),
between the present dye-sensitized Pt-loaded g-C
3 4
N
2
0–22
previous systems,
3 4
N
3
6–38
clearly be slowed down by the addition of PCN.
The time-
sensitized with the ZnPy derivatives dispersed in 10 mL of
ascorbic acid (AA) aqueous solution (50 mM). Rotary evapora-
tion was conducted to load the ZnPy derivative in order to
ensure the same dye-loading content (6.30 mmol g ), and this
was confirmed by a desorption experiment.
resolved fluorescence spectra (TRFS) indicate that the fluores-
cence lifetimes (t) of the ZnPy-5 and ZnPy-6 solutions can be
decreased by adding PCN (Fig. S10c and d, ESI†), and this also
confirms that a charge transfer process exists from the two ZnPy
ꢀ1
3
9
dyes to the PCN, since g-C N does not absorb 475 nm light.
3
4
2
Fig. 7a depicts the photocatalytic H evolution activity of the
This issue is also supported by the transient photocurrent curves
of a dye-sensitized PCN suspension irradiated with l Z 420 nm
light (Fig. S11, ESI†). Compared to pristine PCN, both ZnPy-5/
PCN and ZnPy-6/PCN exhibit clearly enhanced photocurrent
values, suggesting that the photoexcited charge carriers of
the porphyrin derivatives can be efficiently transferred and
separated. The above results confirm that the fast charge separa-
tion is the main reason for the clear photosensitization for H2
evolution of the ZnPy derivatives on PCN (Fig. 7a).
obtained PCN and its dye-sensitized products under visible
light (l Z 420 nm) illumination. As seen from the figure, the
ZnPy derivatives that are loaded onto PCN show clear photo-
sensitization for H evolution compared with the pristine PCN
2
ꢀ1
that only delivers a H2 evolution activity of 112 mmol h
.
As mentioned above, the asymmetric structures of the ZnPy
derivatives can promote the movement of intramolecular direc-
ted electrons from the metal center towards the carboxyl group,
and the more negative LUMO levels than the CB of g-C
3 4
N
Moreover, ZnPy-5 and ZnPy-6 on PCN exhibit a much better
,20,21 photosensitization performance than the contrast sample
(
Fig. 5) are beneficial for the excited state electrons to transfer
9
from porphyrin to g-C N connected via the carboxyl groups.
3
4
(
ZnPy-1) bearing three peripheral phenyls (Scheme 1). For
instance, ZnPy-5/PCN and ZnPy-6/PCN exhibit H evolution
activities of 418 and 585 mmol h (Fig. 7a), corresponding to
This conjecture can be corroborated by the photoluminescence (PL)
2
ꢀ1
ꢀ1
turnover numbers (TONs) of 8845 and 12 381 h , respectively.
ꢀ
1
Both of these values are much better than that (316 mmol h
ꢀ
1
with a TON of 6687 h ) of ZnPy-1/PCN. This implies that the
pyrimidines that replace the phenyls in ZnPy-1 are responsible
for more efficient photosensitization of ZnPy-5 and ZnPy-6, and
2
0–22
this is similar to our previous results.
Since ascorbic acid
(AA) as an acidic sacrificial reagent can provide electrons to
regenerate the oxidized dye molecules, the ZnPy derivatives
with three basic pyridine substituents can form an electron
transfer channel through proton coupling with AA to promote
dye regeneration, and thus deliver better H evolution activity
2
2
0,21
than ZnPy-1 bearing three phenyls.
It is possible that the
two N atoms of the three pseudo-pyrimidines in ZnPy-5 and
ZnPy-6 greatly benefit the proton coupling process with AA
molecules, and this then contributes to more efficient dye
regeneration compared to ZnPy-1 without pseudo-pyrimidines.
Therefore, the substitution of pseudo-pyrimidines for phenyls in
ZnPy-1 and the effective electron transfer from the dye to g-C N4
3
should be the main reasons for the better photoactivity of ZnPy-
5
/PCN and ZnPy-6/PCN compared with ZnPy-1/PCN (Fig. 7a).
ꢀ1
In addition, the H
2
evolution activity (585 mmol h ) of ZnPy-
ꢀ1
6/PCN is much better than that (418 mmol h ) of ZnPy-5/PCN
under the same photoreaction conditions. Since ZnPy-5 and
ZnPy-6 have very similar absorption spectra, molar extinction
coefficients, photoelectrochemical properties and energy band
2
0
structures (Table 1 and Fig. 5), implying their similar driving
force for the photoexcited electrons of the dye injecting into the
PCN, the difference in H evolution performance between the
2
two dye-sensitized products can be mainly attributed to
the differences in the peripheral substituents. ZnPy-6 has its
Fig. 7 (a) Comparison of H
sensitized products (ZnPy-5/PCN and ZnPy-6/PCN). (b) Time course of
evolution over ZnPy-5/PCN and ZnPy-6/PCN. Conditions: 15 mg photo-
2
evolution rates over the PCN and its dye-
H
2
ꢀ1
catalyst, 10 mL of AA solution (50 mM), 1.5 wt% Pt-loading, 6.30 mmol g
dye adsorbed and l Z 420 nm light irradiation.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020
New J. Chem.

View Article Online
Paper
NJC
2
0
two N atoms in the pseudo-pyrimidine groups further away pseudo-pyridines (ZnPy-2 and ZnPy-3). Therefore, it can be
from the porphyrin macrocycle than the ZnPy-5 does. This concluded that increasing the number of N atoms in the
means that ZnPy-6 has a lower steric hindrance than ZnPy-5 peripheral groups can improve the photocatalytic H evolution
2
when the two N atoms in the pseudo-pyrimidines combine with activity to a certain extent.
AA molecules through proton coupling, hence promoting the
Fig. 7b displays the photostability of dye-sensitized PCN for
has visible-light-responsive H evolution under the above photo-
rich surface features such as basicity, electron-rich and reaction conditions. As seen from the figure, both ZnPy-5/PCN
4
0
regeneration of the oxidized dye molecules. Since g-C
3
N
4
2
4
1–44
H-bonds,
porphyrin macrocycle in ZnPy-6 are more easily combined with and fairly good stability for a total of 15 h in the three
g-C via H-bonds than those in ZnPy-5, hence promoting the consecutive runs with fresh sacrificial reagent solution periodi-
photogenerated electron transfer from porphyrin to the semi- cally replaced during each run. Under l Z 420 nm light
conductor for the photocatalytic reaction. illumination, the average H evolution activity over ZnPy-5/
The above conjecture can be validated by the changing PCN is 402 mmol h in the first run of 5 h, and this subse-
2
the N atoms that are further away from the and ZnPy-6/PCN show sustained enhancement in H evolution
3 4
N
2
ꢀ1
ꢀ
1
trends of PL, TRFS spectra and transient photocurrent curves. quently declines to 381 and 369 mmol h in the second and
For example, ZnPy-6 in DMF solution presents more obvious third runs, whilst the activity retention for the third run is
fluorescence quenching than ZnPy-5 after the addition of PCN B92% compared to the first run. Similarly, the average H
2
ꢀ1
(
Fig. S10a and b, ESI†). The fluorescence lifetime (t) decrease evolution activity over ZnPy-6/PCN is 565 mmol h in the first
ꢀ1
of ZnPy-6 with the addition of PCN is 51% (from 3.09 ns to run, and this decreases to 558 and 537 mmol h in the second
1
3
.51 ns), and this is more obvious than the decrease (45% from and third runs, whilst the activity retention ratio for the third
.12 ns to 1.73 ns) of ZnPy-5 when PCN is added (Fig. S10c and run is B95% compared to the first run. These results demon-
d, ESI†), and this also indicates that the electron transfer strate that both ZnPy-5 and ZnPy-6 on PCN have excellent
between ZnPy-6 and PCN is faster than that between ZnPy-5 photosensitization and relatively good stability during the
and PCN. In addition, ZnPy-6/PCN shows a more obvious long-term photoreaction process.
higher photocurrent than ZnPy-5/PCN (Fig. S11, ESI†). The
ZnPy derivatives of dye-sensitized PCN before and after
above results demonstrate that the electron transfer between irradiation for 1 h were desorbed using KOH DMF/water
ZnPy-6 and PCN is more efficient than that between ZnPy-5 and (v/v = 1 : 1) solution (0.1 M) under ultrasonication, and the
PCN, and this is consistent with the above-mentioned changing collected desorbed solutions were determined using UV-vis
trends of photoactivity (Fig. 7a). Therefore, it can be concluded absorption spectra. As shown in Fig. S12 (ESI†), the ZnPy
that ZnPy derivatives, with their N atoms further away from the derivatives after the photocatalytic experiments show absorp-
porphyrin macrocycle, are not only beneficial for dye regenera- tion properties and peak positions that are very similar to those
tion, but they also promote charge separation and transfer for without irradiation, indicating the relatively good stability of
efficient H
2
evolution.
the present ZnPy dye-sensitized system. In addition, no absorp-
2
0
According to our previous literature, ZnPy-2 and ZnPy-3 tion signal of the ZnPy derivatives could be detected from
with pseudo-pyridines as peripheral substituents have N posi- the filtrates of the dye-sensitized product suspensions after
tions similar to those in ZnPy-5 and ZnPy-6 bearing three irradiation, indicating that there is no dye present in the
pseudo-pyrimidines with two N atoms, respectively. The corres- aqueous solution and the dye molecules can be anchored
ponding ZnPy-2/PCN and ZnPy-3/PCN delivered H evolution strongly onto g-C N .
activities of 316 and 451 mmol h
2
1,22
Moreover, the dye solutions desorbed
2
3 4
ꢀ
1 20
,
respectively. It is apparent from the dye-sensitized products before and after irradiation
that the present ZnPy-5 and ZnPy-6 on PCN have more effective exhibit very similar UV-vis absorption spectra (Fig. S12, ESI†),
photosensitization than ZnPy-2 and ZnPy-3 under the same implying that unimpaired skeletons of the ZnPy derivatives
photoreaction conditions (Fig. 7a). This indicates that during the photoreaction process are present. The above fairly
porphyrin-bearing pseudo-pyrimidines with two N atoms exhibit good stability of sensitized PCN could be due to efficient dye
2
0–22
better H evolution activity than the derivatives bearing pseudo- regeneration in the presence of AA as a sacrificial reagent.
2
pyridines with one N atom. There are two possible reasons for
On the basis of the H evolution activities of the photocata-
2
this issue: (1) the increase in N atoms in the pseudo-pyrimidines lyst illuminated with various monochromatic light wavelengths
of ZnPy-5 and ZnPy-6 can promote a more efficient excited state (l = 420, 450, 500, 550, and 600 nm) acquired using their
charge transfer than that of ZnPy-2 and ZnPy-3; (2) the additional respective narrow band-pass filters, the calculated wavelength-
N atoms in the pseudo-pyrimidines of ZnPy-5 and ZnPy-6 dependent AQY curves are shown in Fig. 8. The AQY values of
can more effectively interact with the sacrificial reagent (AA) ZnPy-5/PCN and ZnPy-6/PCN are functions of incident mono-
through proton coupling to promote electron transfer and dye chromatic light wavelength, and their changing trends are
2
0–22
regeneration, compared to ZnPy-2 and ZnPy-3.
Namely, the similar to the respective DRS spectra. For instance, ZnPy-5/
basic pseudo-pyrimidine substituents in ZnPy-5 and ZnPy-6 are PCN delivers AQY values of 32.6%, 10.8%, 0.3%, 3.1% and 1.8%
beneficial for the proton coupling process with acidic AA when illuminated at 420, 450, 500, 550 and 600 nm mono-
molecules, and this then causes more efficient hole consump- chromatic light, respectively. Similarly, ZnPy-6/PCN delivers
tion and dye regeneration, and thus better photosensitization AQY values of 33.1%, 11.3%, 0.3%, 3.0% and 2.0% when
than the porphyrin derivatives bearing phenyls (e.g. ZnPy-1) or illuminated at 420, 450, 500, 550 and 600 nm monochromatic
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Scheme 2 Typical synthetic route for 2-pyrimidinecarbaldehyde.
with methanol several times, filtered and dried by rotary
evaporation to obtain the milky-white product (2-pyrimidine-
1
carboxylic acid). Yield: 5.50 g, 90%. H NMR (CDCl
3
, 400 MHz):
d = 11.0 ppm (1H), 9.27 ppm (2H), 8.19 ppm (1H).
2) 2-Pyrimidinecarboxylic acid (5.5 g, 44 mmol) was added
(
to 30 mL of methanol and stirred for 10 min, and then 3 mL of
concentrated H SO solution was dropped into the mixed
2
4
solution, and this was then refluxed at 85 1C for 2 h. After
cooling to room temperature, the solvent was removed by rotary
evaporation, and then 20 mL of distilled water was added. After
3
extraction with CHCl , the organic phase was dried by rotary
evaporation. The crude product was purified by silica gel
column chromatography with CH Cl and CH OH (v/v =
2
2
3
1
0 : 1) as the eluent to obtain a white solid product (methyl 2-
1
pyrimidinecarboxylate). Yield: 4.40 g, 68%. H NMR (DMSO-d ,
6
4
00 MHz): d = 9.38 ppm (2H), 7.94 ppm (1H), 4.49 ppm (3H).
3) Methyl 2-pyrimidinecarboxylate (2.0 g, 14.49 mmol) was
added to 32 mL of absolute alcohol under stirring, and then a
mixture of NaBH (0.68 g, 20 mmol) and absolute alcohol
32 mL) was dropped into the mixed solution at 0 1C. After
(
4
(
Fig. 8 Comparison of DRS spectra and AQY curves of (a) ZnPy-5/PCN
and (b) ZnPy-6/PCN, irradiated by various monochromatic light
wavelengths.
reacting for 5 h at room temperature, the solvent was removed
by rotary evaporation, and then 6 mL of saturated K CO3
2
solution was added. After extraction with ethyl acetate, the
organic phase was washed and dried with anhydrous Na SO .
2 4
light, respectively. It is worth noting that ZnPy-6/PCN exhibits The liquid product was collected by rotary evaporation and was
the best AQY value of 33.1% when illumined at 420 nm mono- then purified by column chromatography to obtain a light-
chromatic light, and this is slightly higher than that of ZnPy-5/ yellow liquid (2-pyrimidinemethanol). Yield: 1.593 g, 84%.
1
PCN (32.6%), and much higher than those for ZnPy-1/PCN
H NMR (DMSO-d , 400 MHz): d = 8.79 ppm (2H), 8.79 ppm
6
(
10.6%), ZnPy-2/PCN (11.5%) and ZnPy-3/PCN (30.1%), as (2H), 7.41 ppm (1H), 4.95 ppm (2H).
2
0
reported in our previous literature.
(4) SeO2 (0.378 g, 3.40 mmol) was added to 15 mL of
The above results demonstrate that ZnPy-6 is the most 1,4-dioxane, and this was then heated at 60 1C for 30 min,
efficient sensitizer of g-C N among the synthesized ZnPy before a mixture of 2-pyrimidinemethanol (0.50 g, 4.54 mmol)
3
4
derivatives, and that the number and positions of the N atoms and 1,4-dioxane (10 mL) was dropped into the mixed solution.
in the peripheral substituents of the porphyrin molecules can During the reaction, the solution changed from cream yellow to
significantly influence the photoactivity of the corresponding reddish brown. After the dropping process, the solution was
dye-sensitized systems, thus opening up a new path for further refluxed at 110 1C for 4 h, and the reaction process was
improvement of the solar-to-hydrogen energy conversion effi- monitored by TLC. After cooling to room temperature, 15 mL
ciency of dye-sensitized semiconductors through porphyrin of saturated NaCl solution was added. After extraction with
molecular designing.
CH Cl, the organic phase was filtered and collected by rotary
3
evaporation, and it was then dried to obtain a yellowish oily
product (2-pyrimidine carbaldehyde). Yield: 0.45 g, 91%.
H NMR (DMSO-d , 400 MHz): d = 9.98 ppm (1H), 9.10 ppm
6
Experimental section
ZnPy derivative syntheses
1
(2H), 7.78 ppm (1H).
2-Pyrimidinecarbaldehyde. This was prepared through a
5-(4-Carboxyphenyl)-10,15,20-tri(2-pyrimidine)porphyrin zinc
four-step procedure, as shown in Scheme 2.
(ZnPy-5). This was prepared through a three-step procedure, as
1) 2-Cyanopyrimidine (5.0 g, 47.6 mmol) was dropped into shown in Scheme 3.
(
NaOH (5.7 g, 142.5 mmol) solution, and then stirred at 65 1C for
(1) Desiccated CH Cl (500 mL) was added into a flask with
2 2
4
2
h. After cooling, the pH value of the mixture was adjusted to three necks that was vacuumed and filled with N in advance,
B5.0 using diluted HCl solution, and then the solvent was and then 2-pyrimidinecarbaldehyde (0.450 g, 4.16 mmol),
removed by rotary evaporation. The crude product was washed methyl 4-formylbenzoate (0.341 g, 2.08 mmol) and re-steamed
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020
New J. Chem.

View Article Online
Paper
NJC
(3) Zinc acetate (0.22 mg, 0.5 mmol) was added to the H
2
Py-5
(30 mg, 0.05 mmol) solution of dry DMF (30 mL) under a N
2
atmosphere, and this was then refluxed at 100 1C for 10 h. After
cooling, the solvent was removed by rotary evaporation.
The product was washed with 7 mL water for three times and
dried overnight under vacuum to obtain the final product
[5-(4-carboxyphenyl)-10,15,20-tri(2-pyrimidine)porphyrin zinc
(
5
ZnPy-5)]. Yield: 32 mg, 89%. UV-vis lmax (DMF)/nm: 421, 553,
1
6
94. H NMR (DMSO-d , 400 MHz): d = 9.36 ppm (6H, H-Py),
8.82–8.87 ppm (8H, b-H), 8.21–8.52 ppm (4H, H-PhCOOH), 7.97
ppm (3H, H-Py). TOF-MS (m/z): calc. for C H N O Zn
3
9
22 10 2
+
[
M + 1] , 726.12; found, 727.25. Elemental analysis calc. for
C H N O ZnꢁCH OH: C 58.28, H 3.34, N 16.58; found: C
3
9
22 10
2
3
59.02, H 3.88, N 17.20.
5-(4-Carboxyphenyl)-10,15,20-tri(3-pyrimidine)porphyrin
zinc (ZnPy-6). This was prepared through a two-step procedure,
as shown in Scheme 4.
Scheme 3 Typical synthetic route for ZnPy-5.
(
1) 5-Pyrimidinecarbaldehyde (0.648 g, 6.0 mmol) and
-carboxybenzaldehyde (0.302 g, 2.0 mmol) were added to
pyrrole (0.388 mL, 5.54 mmol) were added and stirred for 80 mL of propanoic acid, and this was then refluxed at
min. Afterwards, BF (0.25 g) ether solution was added to 130 1C for 1 h. Afterwards, a mixture of re-steamed pyrrole
the above mixture in the dark, the reaction was kept at room (0.56 mL, 8.0 mmol) and propanoic acid (10 mL) were dropped
temperature for h, and then 2,3-dicyano-5,6-dichloro- into the reaction solution, and this was further refluxed at
benzoquinone (DDQ, 0.74 g) was added to the reaction solution 135 1C for 8 h. The solvent was removed by vacuum distillation
and stirred at room temperature overnight. After that, 1.0 mL of and the crude product was washed using CH Cl to remove the
4
5
3
1
2
2
triethylamine was added to quench the reaction. The resulting tarry impurity. The collected organic phase was dried by rotary
crude solid was dried by rotary evaporation and purified by evaporation and purified by silica gel column chromatography
silica gel column chromatography using CH Cl and CH OH using CH Cl and CH OH (v/v = 10 : 1) as the eluent to obtain a
2
2
3
2
2
3
(v/v = 25 : 1) as the eluent to obtain the product [5-(methyl purple product [5-(4-carboxyphenyl)-10,15,20-tri(3-pyrimidine)-
4
2
2
-benzoate)-10,15,20-tris(2-pyrimidine)porphyrin]. Yield: 0.15 g, porphyrin (H Py-6)]. Yield: 243 mg, 17%. UV-vis lmax (DMF)/nm:
1
1
0.0%. UV-vis lmax (DMF)/nm: 416, 513, 547, 589, 647. H NMR 418, 514, 548, 588, 644. H NMR (DMSO-d
, 400 MHz): d = 9.41 ppm (6H, H-Py), 8.94–8.99 ppm d = 13.32 ppm (H, H-COOH), 9.70 ppm (3H, H-Py), 9.67 ppm
8H, b-H), 8.43 ppm (4H, H-PhCOOMe), 8.03 ppm (3H, H-Py), (6H, H-Py), 8.96–9.04 ppm (8H, b-H), 8.36–8.43 ppm (4H,
.06 ppm (3H, H-COOMe), ꢀ3.00 ppm (2H). TOF-MS (m/z): calc. H-PhCOOH), ꢀ2.98 ppm (2H). TOF-MS (m/z): calc. for
6
, 300 MHz):
(DMSO-d
6
(
4
+
+
for C H N O [M + 1] , 679.15; found, 679.02. Elemental C H N O [M + 1] , 664.41; found, 665.31. Elemental analysis
4
0
26 10
2
39 24 10 2
analysis calc. for C H N O ꢁCH OHꢁCH Cl : C 70.79, H 3.86, calc. for C H N O ꢁCH OHꢁCH Cl : C 63.00, H 3.87, N 17.92;
4
0
26 10
2
3
2
2
39 24 10
2
3
2
2
N 20.64; found: C 71.03, H 3.90, N 20.34.
2) 5-(Methyl 4-benzoate)-10,15,20-tris(2-pyrimidine) porphyrin
0.15 g, 0.2 mmol) was dissolved in 15 mL of THF solvent. The
resulting solution was dropped into 10 mL of LiOH solution
1.0 M) and then refluxed at 70 1C for 10 h. The reaction process
found: C 63.53, H 4.01, N 18.00.
(
(
(
was monitored by TLC. Afterwards, the pH of the reaction
solution was adjusted to 5.00 using diluted HCl solution, and
then extracted three times with ethyl acetate. The collected
2 4
organic phase was dried with anhydrous Na SO and then
separated by silica gel column chromatography. The resulting
product was re-crystallized using methanol and anhydrous
diethyl ether to obtain the final product [5-(4-carboxyphenyl)-
1
4
0,15,20-tri(2-pyrimidine) porphyrin (H
7%. UV-vis lmax(DMF)/nm: 416, 511, 546, 587, 642. H NMR
2
Py-5)]. Yield: 60 mg,
1
(
DMSO-d , 400 MHz): d = 9.41 ppm (6H, H-Py), 8.91–8.96 ppm
6
(8H, b-H), 8.27–8.41 ppm (4H, H-PhCOOH), 8.01 ppm (3H,
H-Py), ꢀ2.96 ppm (2H). TOF-MS (m/z): calc. for C39
24 10 2
H N O
+
[
M + 1] , 664.21; found, 665.23. Elemental analysis calc. for
C
39
H
24
N
10
O
2
ꢁCH
3
OH: C 68.96, H 4.05, N 20.10; found: C 68.74,
H 3.95, N 20.34.
Scheme 4 Typical synthetic route for ZnPy-6.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
(2) Zinc acetate (0.236 g, 1.08 mmol) was added to H
2
Py-6 acetonitrile). The measurements were performed in [NBu
4
]-
(
60 mg, 0.09 mmol) in dry DMF (20 mL) under a N
2
atmosphere, [ClO ] solution (0.1 M) of dried DMF under a N atmosphere
4
2
ꢀ
1
and this was refluxed at 100 1C for 10 h. After being cooled to at room temperature with a scan rate of 100 mV s . The results
room temperature and with the addition of 25 mL of distilled were corrected for junction potentials using a normal hydrogen
water, the resulting purple-green solid was filtered and washed electrode (NHE), which was referenced internally to the ferro-
repeatedly with water, and then vacuum-dried to obtain cenium/ferrocene (Fc /Fc) couple [E1/2(Fc /Fc) = 0.74 V vs.
the final product [5-(4-carboxyphenyl)-10,15,20-tri(3-pyrimidine)- NHE].
+
+
1
5
porphyrin zinc (ZnPy-6)]. Yield: 42 mg, 64%. UV-vis lmax (DMF)/
1
Materials preparation
6
nm: 427, 558, 599. H NMR (DMSO-d , 300 MHz): d = 9.66 ppm
1
2
(
3H, H-Py), 9.61 ppm (6H, H-Py), 8.87–8.91 ppm (8H, b-H), g-C N was prepared as follows: urea (5.0 g) was placed in a
3
4
8
.28–8.42 ppm (4H, H-PhCOOH). TOF-MS (m/z): calc. for covered crucible and heated at 580 1C for 3 h at a heating rate of
+
ꢀ1
C H N O Zn [M + 1] , 726.13; found, 727.25. Elemental analysis 5 1C min . After cooling to room temperature, the product was
39 22 10 2
calc. for C H N O Zn: C 64.34, H 3.05, N 19.24; found: C 65.02, purified using HNO solution (0.1 M) and distilled water, and it
39
22 10
2
3
H 3.35, N 19.58.
was then dried at 70 1C overnight to obtain the yellow-colored
g-C
g-C
3
N
N
4
(denoted as CN). A Pt co-catalyst was deposited onto
, as follows: g-C N (0.60 g) was dispersed in a water
3 4
Chemical and material characterization
21
3
4
Dichloromethane (DCM), N,N-dimethylformamide (DMF) and (150 mL) and methanol (35 mL) mixed solution containing
pyrrole were distilled from CaH under a N atmosphere, and H PtCl (0.077 M, 0.60 mL), and then the suspension was
propionic acid was distilled from Na SO . Column chromato- irradiated for 3 h using a 500 W Hg-lamp under stirring. The
graphy was performed on a silica gel column (Merck, Kieselgel product was collected by centrifugation, washed with water and
0, 70–230 mesh). Thin layer chromatography (TLC) was carried dried at 70 1C overnight to obtain 1.5 wt% Pt-loaded g-C N4
2
2
2
6
2
4
6
3
out on a glass plate (20 ꢂ 20 cm) using silica gel (GF-254). (denoted as PCN).
2
0
All reagents and solvents were of pure quality and were used as
Dye-sensitized PCN was prepared as follows: the above
1.5 wt% Pt-loaded g-C N (PCN, 0.1 g) was added to a ZnPy
received, unless otherwise stated.
3
4
1
H NMR spectra were recorded using a Bruker Ascend 400 derivative DMF solution. After stirring for 6 h, rotary evapora-
as a solvent and tetramethyl- tion was used to collect the product to ensure the same amount
spectrophotometer with DMSO-d
6
ꢀ
1
silane as the reference. MALDI-TOF-MS was performed using a of dye-loading (6.3 mmol g ).
Bruker BIFLEX III ultrahigh resolution Fourier-transform ion
Photocatalytic experiments
The photoreaction for H evolution was performed in a home-
made outer-irradiation-type photoreactor (Pyrex glass) con-
nected to a closed gas-circulation system. The amount of
H evolution was measured using gas chromatography (GC,
2
SP6890, TCD detector, 5 Å molecular sieve columns and a Ar
carrier). In a typical process, the photocatalyst was dispersed in
cyclotron resonance (FTICR) mass spectrometer with a-cyano-4-
hydroxycinnamic acid as a matrix. UV-vis absorption spectra
were recorded using a TU-1810 spectrophotometer. A K2 ISIS
spectrometer was used to determine the fluorescence spectra
and a Biorad FTS-165 spectrometer was used to record the FTIR
spectra using a KBr pellet. A MPC-3100 spectrophotometer was
used to record the UV-vis diffuse reflectance spectra (DRS) with
2
4
BaSO as a reference, and a F-4600 fluorescence spectrophoto-
1
0 mL of water containing ascorbic acid (AA, 50 mM) as an
electron donor, and this was treated in an ultrasonic bath for
0 min before illumination. After thoroughly removing the air,
the suspension was irradiated from the top with visible light
meter was used to determine the photo-luminescence (PL)
spectra. Elemental analyses were carried out using a Vario EL
III Elementar. Time-resolved photo-luminescence spectra
2
(TRPS) were recorded on a FL SP920 fluorescence lifetime
(
(
l Z 420 nm), which was obtained using a cutoff filter
l Z 420 nm) placed on a 300 W Xe-lamp (PLS-SXE300, Beijing
spectrophotometer (Edinburgh Instruments, UK) at excitation
and detection wavelengths of 475 and 610 nm, respectively. The
fluorescence lifetime (t) was calculated according to an expo-
Trusttech Co. Ltd, China).
The apparent quantum yield (AQY) was calculated based on
4
5
nential fitting equation:
the amount of H evolution per hour obtained under the same
2
Fit = A + B ꢂ eꢀt/t + B
1
ꢂ eꢀt/t + B
2
ꢂ eꢀt/t
3
photoreaction conditions but with monochromatic light illu-
1
1
(1)
mination derived from a band pass filter (e.g. l = 420 ꢃ 10 nm)
2
0–23
1 2 3
where A, B , B and B are constants and are obtained after
fitting the decay curve. The statistic values of the lifetimes were
obtained from the fluorescence behavior of 10 photons after
excitation by computer software.
using eqn (2):
5
2 ꢂ Number of evolved H molecules
2
AQY ð%Þ ¼
ꢂ 100%
Number of incident photons
2
ꢂ CNA
Electrochemical measurements
¼
ꢂ 100%
Ptl=hc
Cyclic voltammetry (CV) measurements were conducted using a
CHI600C voltammetric analyzer in a three-electrode setup,
(
2)
ꢀ
1
2 A
which included a glassy carbon electrode, a Ag wire counter where C is the H production amount per hour (mmol h ); N
+
23
ꢀ1
electrode and a Ag/Ag reference electrode (0.01 M AgNO
3
in dry is the Avogadro constant (6.02 ꢂ 10 mol ); t is the light
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020
New J. Chem.

View Article Online
Paper
NJC
3 4
irradiation time (1 h); l is the monochromatic light wavelength faster electron transfers between ZnPy-6 and g-C N than between
ꢀ
34
(
nm); h is the Planck constant (6.626 ꢂ 10
J s); c is the ZnPy-5 and g-C
vacuum light velocity (3 ꢂ 10 m s ); P is the irradiation flux optimize the porphyrin molecular structure for more efficient
W) on the irradiated area, and is equal to the product of the combination of the semiconductor and sacrificial reagent, and
3 4
N . The present results provide new insight to
8
ꢀ1
(
2
irradiation area (A = 11.3 cm ) and the incident monochromatic demonstrate that the N atom number and positions in the
light intensity (p, mW cm ) that was detected by a calibrated Si peripheral substituents of porphyrins have great influence on
ꢀ
2
photodiode (SRC-1000-TC-QZ-N, Oriel).
2
photosensitization for H evolution.
The turnover number (TON) is usually defined by the
number of reacted molecules and the number of active sites.
Conflicts of interest
In the present work, we assumed that H production takes
2
place when two electrons injected from the excited dye mole-
cules react with two protons. One dye molecule can produce
only one photogenerated electron that can be injected into the
There are no conflicts to declare.
CB of g-C
with two H ions and produce one H
can be calculated based the number of evolved H
per hour and the active site using eqn (3):
3
N
4
, therefore two dye molecules are needed to react Acknowledgements
+
2
molecule, and so the TON
molecules
This work was supported by the National Natural Science
Foundation of China (21975190, 21871215, 21573166 and
2
2
7
2
1271146), the Science & Technology Planning Project of
2
ꢂ Number of evolved H2 moleculesperhour
Shenzhen Municipality (JCYJ20180302153921190), the Natural
Science Foundation of Jiangsu Province (BK20151247) and the
Funds for Creative Research Groups of Hubei Province
TON ¼
(3)
Number of dyemoleculesadsorbed
where the evolved H molecules per hour is the H production
rate (mmol h ), and the number of dye molecules is calculated
from the dye-loading amount (6.3 mmol g ) and the photo-
2
2
ꢀ1
(2014CFA007), China.
ꢀ1
catalyst dosage (0.015 g).
Notes and references
1
T. Hisatomi, J. Kubota and K. Domen, Chem. Soc. Rev., 2014,
3, 7520–7535.
Conclusions
4
Novel asymmetric zinc porphyrin (ZnPy) derivatives (ZnPy-5
and ZnPy-6) bearing one benzoic acid and three pseudo-
pyrimidines as meso-position substituents were synthesized
2 C. F. Fu, X. J. Wu and J. L. Yang, Adv. Mater., 2018,
30, 1802106.
3 Y. Ma, X. L. Wang, Y. S. Jia, X. B. Chen, H. X. Han and C. Li,
Chem. Rev., 2014, 114, 9987–10043.
3 4
and utilized to sensitize Pt-loaded g-C N (PCN). Among the
ZnPy derivatives, benzoic acid acts as an electron-withdrawing/
anchoring group and the pseudo-pyrimidines act as electron-
donating groups, whereby the photoexcited electrons can be
transferred from the metal center of porphyrin towards the
carboxyl group and injected into g-C N for H evolution. The
4 X. B. Chen, S. H. Shen, L. J. Guo and S. S. Mao, Chem. Rev.,
2010, 110, 6503–6570.
5 Q. Guo, C. Y. Zhou, Z. B. Ma and X. M. Yang, Adv. Mater.,
2019, 50, 1901997.
6 X. H. Zhang, T. Y. Peng and S. S. Song, J. Mater. Chem. A,
2016, 4, 2365–2402.
7 Y. Yang, L. C. Yin, Y. Gong, P. Niu, J. Q. Wang, L. Gu,
X. Chen, G. Liu, L. Wang and H. M. Cheng, Adv. Mater.,
2018, 30, 1704479.
3
4
2
pseudo-pyrimidines that replace the phenyls (ZnPy-1) or
pseudo-pyridines (ZnPy-2 and ZnPy-3) show significantly
improved photosensitization on PCN, even though they have
very similar photo-absorption abilities and energy band struc-
tures. Moreover, the different positions of the two N atoms in
the pseudo-pyrimidines resulted in different interactions of the
ZnPy derivatives with the sacrificial reagent (AA) and g-C N ,
8 Q. Q. Lang, Y. H. Chen, T. L. Huang, L. L. Yang, S. X. Zhong,
L. J. Wu, J. R. Chen and S. Bai, Appl. Catal., B, 2018, 220,
182–190.
3
4
and this then strongly influences the activity. For example,
ZnPy-6 with the lowest steric hindrance of N atoms in its
9 S. S. Song, Z. C. Liang, W. L. Fu and T. Y. Peng, ACS Appl.
Mater. Interfaces, 2017, 9, 17013–17023.
3 4
pseudo-pyrimidines is easy to combine with g-C N and AA, 10 S. S. Song, J. M. Wang, T. Y. Peng, W. L. Fu and L. Zan, Appl.
and thus ZnPy-6/PCN exhibits a H
2
evolution production activity of
Catal., B, 2018, 228, 39–46.
ꢀ1
ꢀ1
585 mmol h with an extremely high TON of 12 381 h , and these 11 W. L. Fu, J. M. Wang, S. Y. Zhou, R. J. Li and T. Y. Peng, ACS
values are better than those of ZnPy-5/PCN, which has a H
2
Appl. Nano Mater., 2018, 1, 2923–2933.
ꢀ1
ꢀ1
evolution activity of 418 mmol h
and TON of 8845 h
.
12 Y. Zhang, L. L. Wu, X. Y. Zhao, Y. N. Zhao, H. Q. Tan,
X. Zhao, Y. Y. Ma, Z. Zhao, S. Y. Song, Y. H. Wang and
Y. G. Li, Adv. Energy Mater., 2018, 8, 1801139.
In particular, ZnPy-6/PCN also delivers an impressively high
apparent quantum yield of 33.1% at 420 nm monochromatic
light, and this is greater than that (32.6%) of ZnPy-5 with a higher 13 S. X. Min and G. X. Lu, J. Phys. Chem. C, 2012, 116,
steric hindrance. The above differences in photosensitization 19644–19652.
between ZnPy-5 and ZnPy-6 on PCN can mainly be ascribed to 14 H. J. Yan and Y. Huang, Chem. Commun., 2011, 47,
the pseudo-pyrimidines with different N atom positions, leading to 4168–4170.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
1
1
1
1
1
5 L. J. Yu, X. H. Zhang, C. S. Zhuang, L. Lin, R. J. Li and 31 Y. Ma, P. Zhang, P. H. Zhu, X. Y. Zhang, Y. N. Gao, D. D. Qi,
T. Y. Peng, Phys. Chem. Chem. Phys., 2014, 16, 4106–4114.
6 Q. W. Liu, J. M. Wang, D. Liu, R. J. Li and T. Y. Peng, J. Power
Sources, 2018, 396, 57–63.
7 K. Li, L. Lin, T. Y. Peng, Y. Y. Guo, R. J. Li and J. Zhang,
Chem. Commun., 2015, 51, 12443–12446.
Y. Z. Bian, N. Kobayashi and J. Z. Jiang, J. Mater. Chem.,
2011, 21, 6515–6524.
32 J. J. He, G. Benk o¨ , F. Korodi, T. Pol ´ı vka, R. Lomoth, B. Sun,
L. C. Åkermark, A. Hagfeldt and V. Sundstr o¨ m, J. Am. Chem.
Soc., 2002, 124, 4922–4932.
8 D. H. Wang, J. N. Pan, H. H. Li, J. J. Liu, Y. B. Wang, 33 S. W. Cao, J. X. Low, J. G. Yu and M. Jaroniec, Adv. Mater.,
L. T. Kang and J. N. Yao, J. Mater. Chem. A, 2016, 4, 290–296. 2015, 27, 2150–2176.
9 O. Bettucci, T. Skaltsas, M. Calamante, A. Dess `ı , M. Bartolini, 34 W. Sinha, M. G. Sommer, N. Deibel, F. Ehret, B. Sarkar and
A. Sinicropi, J. Filippi, G. Reginato, A. Mordini, P. Fornasiero
and L. Zani, ACS Appl. Energy Mater., 2019, 2, 5600–5612.
0 Y. Zheng, J. M. Wang, J. Zhang, T. Y. Peng and R. J. Li,
Dalton Trans., 2017, 46, 8219–8228.
1 J. M. Wang, Y. Zheng, T. Y. Peng, J. Zhang and R. J. Li, ACS
Sustainable Chem. Eng., 2017, 5, 7549–7556.
2 J. M. Wang, D. Liu, Q. W. Liu, T. Y. Peng, R. J. Li and
S. Y. Zhou, Appl. Surf. Sci., 2019, 464, 255–261.
3 Z. F. Wu, X. Li, H. Agren, J. L. Hua and H. Tian, ACS Appl.
Mater. Interfaces, 2015, 7, 26355–26359.
4 F. Y. Zhang, P. Yu, W. Shen, M. Li and R. X. He, Comput.
Theor. Chem., 2016, 1095, 118–124.
5 D. M. Chen, K. W. Wang, W. Z. Hong, R. L. Zong, W. Q. Yao 40 R. Amorati, A. Baschieri, G. Morroni, R. Gambino and
and Y. F. Zhu, Appl. Catal., B, 2015, 166–167, 366–372. L. Valgimigli, Chem. – Eur. J., 2016, 22, 7924–7934.
6 V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, 41 J. J. Zhu, P. Xiao, H. L. Li and S. A. C. Carabineiro, ACS Appl.
S. Kar, Chem. – Eur. J., 2014, 20, 15920–15932.
35 K. Maeda, M. Eguchi, W. J. Youngblood and T. E. Mallouk,
Chem. Mater., 2008, 20, 6770–6778.
36 Y. S. Xu and W. D. Zhang, ChemCatChem, 2013, 5,
2343–2351.
37 B. Chai, T. Y. Peng, X. H. Zhang, J. Mao, K. Li and
X. G. Zhang, Dalton Trans., 2013, 42, 3402–3409.
38 M. S. Zhu, Z. Li, B. Xiao, Y. T. Lu, Y. K. Du, P. Yang and
X. M. Wang, ACS Appl. Mater. Interfaces, 2013, 5, 1732–1740.
39 U. Bhattacharjee, L. Men, B. A. Rosales, S. R. Alvarado,
J. Vela and J. W. Petrich, J. Phys. Chem. C, 2017, 121,
676–683.
2
2
2
2
2
2
2
N. Kim, K. C. Kemp, P. Hobza, R. Zboril and K. S. Kim,
Chem. Rev., 2012, 112, 6156–6214.
7 Y. Y. Guo, S. S. Song, Y. Zheng., R. J. Li and T. Y. Peng,
Dalton Trans., 2016, 45, 14071–14079.
Mater. Interfaces, 2014, 6, 16449–16465.
42 Q. Li, S. C. Wang, Z. X. Sun, Q. J. Tang, Y. Q. Liu, L. Z.
Wang, H. Q. Wang and Z. B. Wu, Nano Res., 2019, 12,
2749–2759.
2
2
2
3
8 K. Pei, Y. Wu, A. Islam, Q. Zhang, L. Han, H. Tian and 43 J. W. Fu, J. G. Yu, C. J. Jiang and B. Cheng, Adv. Energy
W. Zhu, ACS Appl. Mater. Interfaces, 2013, 5, 4986–4995. Mater., 2018, 8, 1701503.
9 B. Liang, C. Y. Jiang, Z. Chen, X. J. Zhang, H. H. Shi and 44 Y. B. Jiang, Z. Z. Sun, C. Tang, Y. X. Zhou, L. Zeng and
Y. Cao, J. Mater. Chem., 2006, 16, 1281–1286. L. M. Huang, Appl. Catal., B, 2019, 240, 30–38.
0 D. M. de Leeuw, M. M. J. Simenon, A. R. Brown and 45 X. H. Zhang, B. S. Peng, S. Zhang and T. Y. Peng, ACS
R. E. F. Einerh, Synth. Met., 1997, 87, 53–59. Sustainable Chem. Eng., 2015, 3, 1501–1509.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020
New J. Chem.