Bio-proton coupled semiconductor/metal-complex hybrid photoelectrocatalytic interface for efficient CO2 reduction

Article abstract of DOI:10.1039/c8gc03066a

Aimed at high-efficiency biomimetic CO₂ photoelectrochemical conversion, a bio-proton coupling metal-complex/semiconductor hybrid photoelectrocatalytic interface (Ru-BNAH/TiO₂/Cu₂O) was constructed by covalently modifying an in situ proton-transfer functionized molecular catalyst (Ru-BNAH) on the surface of a TiO₂/Cu₂O composite semiconductor substrate electrode. Due to the excellent proton coupling of the bio-proton carrier, the light current density in a CO₂ atmosphere of the prepared Ru-BNAH/TiO₂/Cu₂O photoelectrocatalytic interface was twice as high as that without a proton carrier under the same conditions. Simultaneously, based on the excellent photosensitivity of the metal oxide substrate, the photogenerated electrons could rapidly transfer to the molecular catalyst for efficient CO₂ reduction in a water medium. After 8 h irradiation at -0.9 V potential, the Ru-BNAH/TiO₂/Cu₂O photoelectrocatalytic interface produced 409.5 μmol formic acid, which was 2.44 times more than that without a proton transfer carrier. In addition, the in situ UV-visible absorption spectra and in situ Raman spectra indicated that the proton transport carrier supplied protons during CO₂ reduction. Moreover, the generation of HCOO^- in CO₂-saturated D₂O medium confirmed the proton (H) originated from the proton transfer carrier rather than the solvent (D₂O).

Full text of DOI:10.1039/c8gc03066a

Green Chemistry
View Article Online
View Journal
PAPER
Bio-proton coupled semiconductor/metal-
complex hybrid photoelectrocatalytic interface for
Cite this: DOI: 10.1039/c8gc03066a
efficient CO2 reduction†
a,b
Jibo Liu,a,b Chenyan Guo,b Xiaojun Hu*a and Guohua Zhao
*
Aimed at high-efficiency biomimetic CO2 photoelectrochemical conversion, a bio-proton coupling
metal-complex/semiconductor hybrid photoelectrocatalytic interface (Ru-BNAH/TiO2/Cu2O) was con-
structed by covalently modifying an in situ proton-transfer functionized molecular catalyst (Ru-BNAH) on
the surface of a TiO2/Cu2O composite semiconductor substrate electrode. Due to the excellent proton
coupling of the bio-proton carrier, the light current density in a CO2 atmosphere of the prepared
Ru-BNAH/TiO2/Cu2O photoelectrocatalytic interface was twice as high as that without a proton carrier
under the same conditions. Simultaneously, based on the excellent photosensitivity of the metal oxide
substrate, the photogenerated electrons could rapidly transfer to the molecular catalyst for efficient CO2
reduction in a water medium. After 8 h irradiation at −0.9 V potential, the Ru-BNAH/TiO2/Cu2O photo-
electrocatalytic interface produced 409.5 μmol formic acid, which was 2.44 times more than that without
a proton transfer carrier. In addition, the in situ UV-visible absorption spectra and in situ Raman spectra
indicated that the proton transport carrier supplied protons during CO2 reduction. Moreover, the gene-
ration of HCOO− in CO2-saturated D2O medium confirmed the proton (H) originated from the proton
transfer carrier rather than the solvent (D2O).
Received 30th September 2018,
Accepted 30th November 2018
DOI: 10.1039/c8gc03066a
rsc.li/greenchem
activity.10,11 Thanks to the metal ion catalytic site, the mole-
cular catalysts have high selectivity for formic acid or CO in
Introduction
Due to an escalating rate of fossil fuel combustion, the CO2 CO2 reduction, but without the formation of multiple electron
concentration in the atmosphere has reached unprecedented reduction products, such as methanol.12–15 Therefore, CO2
levels and continues to increase, causing concern about reduction oriented by molecular catalysts has attracted deep
climate change and rising sea levels.1,2 The photocatalytic attention in the field of CO2 transformation.16 Although mole-
reduction of CO2 to produce chemicals and fuels (such as cular catalysts used for CO2 reduction have achieved remark-
formic acid, methane and methanol) is regarded as an able results in recent years, the proton coupling process in the
effective way to solve the problem of global warming and CO2 reduction process still restricts the efficiency of the reac-
energy crisis attributed to the inexhaustible supply of solar tion. Though CO2 could convert to various hydrogenated pro-
energy.3–6 The photoelectrocatalytic conversion of CO2 has ducts (such as formic acid, methanol and methane) with the
thus been attracting more and more attention on account as proton coupling process, the simplest product is formic acid,
photoelectrocatalysis can synergize the advantages of photoca- which is generated from activated CO2 with a two-electron
talysis and electrocatalysis.7–9
migration and one-proton coupling process.8 The use of a bio-
Molecular catalysts represent a series of metal–organic com- mimetic redox mediator/redox cofactor can provide useful
plexes equipped with a metal ion catalytic centre that displays reference information for the proton coupling reaction.17
excellent CO2 reduction activity or hydrogen evolution
In the photosynthesis process, NADH, as a derivative of
N-1,4-dihydronicotinamide, plays a vital role in maintaining
the cell differentiation, in the polysaccharide repairing and in
proton transfer of the organism for oxidation/reduction.3,18
The NAD+/NADH couple is a ubiquitous redox mediator/redox
cofactor in chloroplast, distinctive for its proton-coupled elec-
tron transfer (PCET) reactivity based on C–H bond cleavage/
formation reactions with NADH providing C–H-based hydride
equivalents.19–21 To further investigate the proton-transfer
aSchool of Chemical and Environmental Engineering, Shanghai Institute Technology,
100 Haiquan Road, Shanghai, 201418, China
bSchool of Chemical Science and Engineering, Shanghai Key Lab of Chemical
Assessment and Sustainability, Tongji University, 1239 Siping Road,
Shanghai 200092, China. E-mail: g.zhao@tongji.edu.cn
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8gc03066a
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
process, numerous scientists have conducted extensive
research on the application of NADH in bionics.22,23 The
pivotal reactive group of NADH is a 1,4-dihydronicotinamide
derivative. NADH can provide equivalent protons to the oxi-
dation material in the changes of 1,4-dihydronicotinamide
into niacinamide. In order to elucidate the proton-transfer
process of NADH more conveniently and effectively, numerous
NADH substitutes have been widely exploited due to its
complex structure and unstable chemical property. N-Benzyl-
1,4-dihydronicotinamide (BNAH) which is equipped with a 1,4-
dihydronicotinamide structure, is one of the most widely used
substitutes, attributed to its simpler structure and stable per-
formance.24,25 However, when BNAH was employed as a
proton donor, it generally worked well in an organic system
rather than in aqueous solution due to its inferior solubility in
aqueous solution.26 In addition, more than 10 eq. BNAH was
needed to ensure enough protons were supplied for the
reduction to run successfully.27 For improving the efficiency of
the proton transference, it was useful and convenient to
change the proton migration style from intermolecular to
intramolecular. In view of the organic properties of these two
components, covalently bonding BNAH on the molecular cata-
lyst was an optimal method to achieve such an integration. For
many precious-metal centre molecular catalysts, it is necessary
to minimize the amount of required catalyst as much as poss-
ible.28 Heterogenizing the homogeneous catalyst by binding
the molecular catalyst directly onto the surface of the photo-
cathode is a feasible strategy, which not only decreases the
Scheme 1 Construction of the Ru-BNAH/TiO2/Cu2O photoelectro-
catalytic interface.
were employed to explain the probable proton transfer and
CO2 reduction mechanism. An isotopic-labelling experiment
was employed to confirm that BNAH supplied the protons
in situ proton during the CO2 reduction process.
Experimental section
Preparation of the Ru-BNAH/TiO2/Cu2O photoelectrocatalytic
interface
amount of molecular catalyst required but also restrains para- The TiO2/Cu2O nanocomposites were prepared as follows.35,36
sitic light absorption in the homogeneous system caused by With constant stirring, TiO2 (P25, 1 eq.) was added into an
the inherent dark colour of the molecular catalyst.16,29–31
ethanol solution (100 mL) containing copper(II) acetate dihy-
When it comes to the metal oxide substrate electrode, TiO2 drate (200 mg, 1 mmol) and then stirred at 60 °C for 30 min to
is one of the optimal choices thanks to its photostability, obtain a uniform suspension. Subsequently, 100 mL glucose
which protects otherwise unstable photocathode materials aqueous solution (0.2 mol L−1) and 120 mL NaOH aqueous
(such as Cu2O).32,33 The metal oxide mixture constitutes an solution (0.3 mol L−1) were added into the suspension drop-
ideal photoelectrochemical carrier for the covalent binding of wise consecutively. After stirring at such temperature for
molecules by molecular linking groups, such as carboxylates 15 min, the mixture was cooled down to room temperature
or phosphonates. A number of examples of molecular catalysts naturally. The precipitates were collected by filtration, washed
linked to the surfaces of photocathodes have been developed by deionized water and ethanol 3 times and then dried at
due to the attractive prospect of decreasing the catalyst loading 40 °C in vacuum.
while retaining the efficiency.28,34 However, in situ proton-sup-
plying systems for CO2 reduction are rarely utilized.
The freshly prepared TiO2/Cu2O nanoparticles were
immobilized on fluorine-doped tin oxide coated glass (FTO) as
Hence, this manuscript demonstrates a photoelectro-cata- the substrate electrode. Generally, to a suspension of TiO2/
lytic system for formic acid production as a solar fuel by the Cu2O (50 mg) in acetone (50 mL), I2 (10 mg) was added under
covalent attachment of molecular CO2 reduction catalysts continuous stirring. After that, the mixture was ultrasonicated
equipped with a proton carrier on the surface of TiO2-pro- for 5 min to give the electrolyte solution. FTO and Pt plates
tected Cu2O photocathodes by means of carbonic ester linkers were used as the cathode and anode respectively. They were
(as shown in Scheme 1). The photosensitivity of the metal immersed in the solution in parallel at a distance of 1.5 cm,
oxide semiconductor provides electrons for the molecular cata- and 10 V of voltage was added for 10 min using a DC power
lyst, while the BNAH proton carrier supplies the protons supply. The electrode was used directly after drying in vacuum
in situ, and thus CO2 can be effectively reduced on the Ru(II) at 40 °C for 12 h.
catalytic site of the molecular catalyst. Under the photoelectro-
The brief synthesis route of Ru-BNAH is shown in Scheme 2
catalytic reduction, the catalytic system equipped with a BNAH and the specific procedures are listed in the ESI.† The Ru-
proton carrier produced formic acid in twice as high an BNAH/TiO2/Cu2O photocathode interface was prepared as per
amount as that in the absence of the BNAH proton carrier. the following. Ru-BNAH (50 mg) was added to N,N-dimethyl-
In situ UV-visible absorption spectra and Raman spectroscopy formamide (DMF, 50 mL), and then TiO2/Cu2O substrate elec-
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
Typically, 100 mL 0.1 M KCl solution was added as the electro-
lyte solution. The Ru-BNAH/TiO2/Cu2O-loaded FTO working
electrode and Ag/AgCl (filled with saturated KCl) reference
electrode were placed in the cathodic chamber, while a graph-
ite plate counter electrode was placed in the anodic chamber.
A 2 cm−2 quartz window was placed on the cathode chamber
to view the photocatalytic CO2 reduction. Generally, a 0.1 M
KCl aqueous solution (100 mL) was added to the reactor and
then bubbled with high purity CO2 (99.99%) gas at a flow rate
of 30 mL min−1 for 40 min to completely remove the dissolved
oxygen. The potential during the PEC reduction was kept con-
stant at −0.6 V under 100 mW cm−2 irradiation (light source:
PLS-SXE300 xenon lamp, Beijing Perfect Light Co., Ltd, China,
with AM 1.5 filter). Note that the water used for all the solu-
tions preparation was ultrapure with its conductivity reaching
18.2 MΩ cm.
The products in the aqueous phase were determined by
Nash’s reagent method (as described in the ESI†). The gaseous
samples were tested for the potential products from CO2
reduction, such as CO, methane, ethylene and ethane, by
injecting into a Shimadzu GC-2014C gas chromatograph
equipped with a thermal conductivity detector (TCD) and 5 Å
molecular filled column.
Scheme 2 Synthetic procedure for the proton-coupling-type mole-
cular catalyst Ru-BNAH.
trodes were added to the above solution. The mixture was
heated to 95 °C for at least 8 h under the protection of N2 to
achieve saturation adsorption. The electrodes were taken out,
washed with acetonitrile and dried in vacuum at 40 °C for 4 h
to give the Ru-BNAH/TiO2/Cu2O photocathode catalytic
interface.
Chemical and structural characterizations
X-ray photoelectron spectroscopy (XPS, AXIS Ultra HSA, Kratos
Analytical Ltd, UK) was used for the surface element analysis
of the prepared catalyst. All the binding energy values for the
high-resolution spectra were calibrated with the signal of con-
taminated carbon C 1s at 284.8 eV. The crystal structures of
the synthesized TiO2/Cu2O were characterized by X-ray diffrac-
tometry (XRD, Focus D8, Bruker, Germany) with Cu Kα radi-
ation (λ = 0.1542 nm). The 1H NMR spectra were measured
using a Bruker AVANCE 400 spectrometer. A fluorescence
spectrophotometer (Hitachi, F-7000) was employed to charac-
terize the steady-state fluorescence spectra of Ru-BNAH. The
absorption spectra were obtained using an ultraviolet-visible
spectrophotometer (UV-vis, Agilent 8453).
A CHI660C electrochemical workstation (CH Instruments,
Inc., USA) was employed to collect all the PEC performance
data by using a conventional three-electrode system with the
as-prepared electrode as the working electrode, Ag/AgCl (satu-
rated KCl) as the reference electrode and a platinum sheet as
the counter electrode in 0.1 M KCl aqueous solution. All the
potentials were referenced to a normal hydrogen electrode
(NHE) unless stated otherwise. A standard two-compartment
electrolysis cell, which separated the cathode and anode reac-
tions, was used for controlling the potential CO2 reduction in
0.1 M KCl aqueous solution supporting the electrolyte. The
amperometric i–t curve was recorded with an interval of 200 s
for light on/off under the light intensity of 100 mW cm−2 at a
constant potential of −0.6 V (light source: LA-410UV, Hayashi,
Japan).
Results and discussion
Construction of the proton-coupling functional interface
Complexes based on Ru(II), Re(II) and Ir(II) have shown excel-
lent performance as well as robustness when used as homo-
geneous photoelectrocatalysts for the selective reduction of
CO2. Simultaneously, for low-cost p-type light harvesters for
water splitting, TiO2-protected Cu2O photocathodes take up an
important position. Carboxyl linking groups were employed to
enable the BNAH-linked Ru(II) complex (Ru-BNAH) to co-
valently bond on the surface of the TiO2-protected Cu2O
photocathode.
XRD showed that the diffraction peaks of TiO2 and Cu2O in
the TiO2/Cu2O substrate electrode were all represented. As
shown in Fig. S1,† the diffraction peaks appeared at 2θ =
25.26°, 37.77°, 47.98°, 54.01°, 54.01°, 62.70° and 68.86°,
corresponding to the (101), (004), (200), (105), (211), (204) and
(116) planes of typical anatase TiO2. Diffuse reflectance UV-vis
spectra (DRS) were recorded to characterize the optical per-
formance of the TiO2/Cu2O substrate electrode. As shown in
Fig. S2,† the prepared TiO2/Cu2O substrate electrode had a
remarkable optical absorption performance in the UV-visible
region, and the absorption band edge could extend to nearly
650 nm, which indicated the effective harvesting of UV-visible
light. In order to maintain the light absorption and to
decrease the photocorrosion performance of TiO2/Cu2O nano-
particles, TiO2/Cu2O (1 : 1) nanoparticles were electrodeposited
PEC reduction of CO2 and products detection
The PEC reduction of CO2 proceeded under a typical three- on the surface of FTO to prepare the substrate photocathode
electrode-system in a 100 mL homemade double-chamber for CO2 reduction. The band gap value of the prepared TiO2/
reactor, which was divided with a proton-exchange membrane. Cu2O nanoparticles was 2.02 eV, which was attributed to the
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
intrinsic properties of Cu2O, while the values of pure TiO2 and sion wavelength of 400 nm (Fig. 2a), which belonged to the
Cu2O were 3.2 eV and 2.06 eV, respectively. recombination of photogenerated electron–hole in excited
The Ru-BNAH/TiO2/Cu2O catalytic interface was obtained TiO2. By coupling the Cu2O nanoparticles, the fluorescence
by refluxing the TiO2/Cu2O substrate electrode in a solution of intensity clearly decreased more than that of the same weight
Ru-BNAH in ethanol. X-ray photoelectron spectroscopy (XPS) of TiO2, indicating the favourable fluorescence quenching of
measurements were conducted (as shown in Fig. 1) to examine the Cu2O nanoparticles. Furthermore, with Ru-BNAH loading
the chemical composition and chemical status of the constitu- on the surface of TiO2/Cu2O substrate electrode, the fluo-
ent elements in the Ru-BNAH/TiO2/Cu2O catalytic interface. rescence almost vanished, indicating the inhibited recombina-
Referring to Fig. 1a, the XPS survey spectra depicted that C, N, tion of photogenerated electron–hole pairs. The inhibited
O, Ti, Cu and Ru elements were detected in the Ru-BNAH/ recombination of photogenerated electron–hole pairs can
TiO2/Cu2O sample. The binding energy was calibrated against promote CO2 reduction by enabling the electrons to be used
the reference C 1s peak at 284.6 eV. The N 1s spectra could be more effectively. The electron transfer from the TiO2/Cu2O
deconvoluted into three peaks: peaks centred at 398.9 and semiconductor to the Ru-BNAH molecular catalyst was the
401.7 eV derived from the sp2-hybridized nitrogen atoms in main reason for the recombination inhibition of the photo-
CvN–C groups and a peak at around 400.5 eV, attributed to generated electron–hole pairs. Furthermore, Fig. 2b shows the
the tertiary nitrogen N–C3 groups of the ligand. In the fluorescence quenching of the Ru(II) molecular catalyst in the
Cu 2p3/2 core level XPS spectra, the peaks corresponding to presence of BNAH. The non-proton carrier molecular catalyst
the Cu 2p3/2 were observed at 932.6 and 951.2 eV and could (Ru-PLA) showed obvious fluorescence under the 420 nm light
be assigned to Cu2O.37 Further, the existence of the two shake- excitation. However, the fluorescence intensity decreased
up satellite peaks, one in the range 938–948 eV and the other observably upon the addition of the dissociative BNAH. As the
at 961.8 eV in the spectra also reiterated the chemical state of proton carrier (BNAH) bonded on Ru-PLA molecular catalyst
copper in the nanocomposite as a little Cu2+ (CuO) caused by by C–C bond to form an organic whole (Ru-BNAH), the fluo-
the oxidation of Cu2O.38 The major peak at 570.1 eV in the rescence intensity further decreased due to the intramolecular
CuLMM auger spectra further substantiated the presence of rapid electron migration.
copper(I) oxide (As shown in Fig. S3†).39 The peaks at 284.8
and 288.2 eV were assigned to the CvN coordination and
Efficient proton coupling of the Ru-BNAH molecular catalyst
sp2-hybridized carbon in N-containing aromatic rings As is well known, catalytic CO2 reduction has two important
(N–CvN), respectively. Additionally, the binding energy of processes: electron migration and proton coupling. Same as
Ru(II) 3p3/2 at 282.3 eV overlapped with C 1s. Ru(III) was not the electron migration, proton coupling plays an important
detected with the binding energy at 282 eV, which indicated role for the hydrogenate production in catalytic CO2 reduction.
no other phases of the Ru complex were present.
Therefore, a proton carrier equipped with an efficient proton
As a typical inorganic–organic hybrid catalytic interface, the transfer ability could powerfully promote hydrocarbon gene-
electron migration between the components is an important ration during CO2 reduction. As shown in Fig. 2c, the trans-
element to evaluate the catalytic performance. The photo- formation between BNAH and BNA can simulate the proton-
luminescence spectra were introduced to investigate the photo- transfer process. A UV-visible absorption spectrometer was
generated electron migration trend of the Ru-BNAH/TiO2/Cu2O employed to record the proton-transfer capacity of the BNAH.
photoelectrocatalytic interface. Under 325 nm irradiation exci- In the solution of BNA in methanol, an apparent absorption
tation, TiO2 had a strong fluorescence at the maximum emis- peak was observed at 360 nm immediately upon the addition
of excess NaBH4, which was attributed to the changes between
BNA and BNAH (as shown in the scheme of Fig. 2c). Similar
changes should also occur in the transfer with the molecular
catalysts Ru-BNA and Ru-BNAH. As shown in Fig. 2d, an
obvious absorption enhancement of Ru-BNA was recorded by
its UV-visible spectrum accompanied by the addition of
NaBH4, which was due to the proton transfer of Ru-BNAH and
Ru-BNA. To a certain degree, the absorbance located at
350–500 nm of their UV-visible spectra reflect the proton trans-
fer process. The enhanced absorption means the generation of
proton carriers, while the reduced absorption reflects the con-
sumption of proton carriers.
In situ UV-vis spectroscopy was employed to demonstrate
the proton transfer of Ru-BNAH in the catalytic CO2 reduction
process (Fig. 2e). In CO2-saturated 0.1 M KCl aq. solution, the
UV-vis peaks at 275 nm increased significantly accompanied
with sustaining the illumination, which might be attributed to
the intermediate (Ru-COO−) generation based on CO2 injection
Fig. 1 XPS survey scan (a), high-resolution XPS spectra of N 2p (b),
Cu 2p (c), Ru 3d C 1s and C 1s (d).
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
Fig. 2 (a) Photoluminescence (PL) spectra of P25, P25/Cu2O and Ru-BNAH/P25/Cu2O. (b) Photoluminescence spectra of Ru-PLA, Ru-PLA mixed
BNAH and Ru-BNAH. (c) The changes of the UV-visible spectrum of BNAH solution before and after NaBH4 addition. (d) The changes of UV-visible
spectrum of Ru-BNAH solution before and after NaBH4 addition. (e) The in situ UV-visible spectra of Ru-BNAH solution under CO2 atmosphere
upon irradiation with time. (f) Absorbance versus time for selected wavelengths for Ru-BNAH under a CO2 atmosphere upon irradiation.
into the Ru(II) active unit. Meanwhile, a decreased absorption expound the complicated relationship between the CO2- and
tendency also arise at 350–400 nm, attributed to the proton N2-saturated solutions of Ru-PLA and Ru-BNAH clearly, the
contribution of the proton carrier (BNAH). CO2 reduction was difference in values of the current density (Δj ) in the CO2- and
realized following a Ru-COO− intermediate receiving a proton N2-saturated solutions over time are summarized in Fig. 3b.
from BNAH. The evolution of the UV-vis spectra for Ru-BNAH The two catalysts showed evident CO2 reduction because both
under CO2 is shown more clearly in Fig. 2f, in which the absor- of them showed obvious Δj values. Thus, Ru-BNAH possessed
bance at 275 nm (intermediate generation) and 360 nm a much higher Δj value than Ru-PLA at −0.9 V which robustly
(proton transfer) are plotted smoothly versus the irradiation demonstrated the acceleration of CO2 reduction due to the
time. The absorption intensity of both wavelengths was proton carrier.
changed significantly in the first 200 s, which indicates the
The amperometric i–t curves of these electrodes under
rapid intermediate generation and proton migration. However, chopped-light irradiation (Fig. S4a†) also agreed with the
after exceeding 200 s, the absorption intensity flattened out, above results. The i–t curve of the Cu2O/TiO2 electrode showed
which indicates the equilibrium state of the BNAH’s regener- an obvious increase by at least 100 s irradiation duration,
ation and consumption.
under N2 or CO2 atmosphere, which was attributable to the
In order to further investigate the CO2 reduction capacity of slow separation of photogenerated electron–hole pairs within
the designed catalyst, cyclic voltammetry was employed to Cu2O/TiO2. The added value of light current over the dark
evaluate the electrochemical behaviour. The electrochemical current was 0.02 mA cm−2 in the N2-saturated solution, while
responses of Ru-BNAH and Ru-PLA were probed in N2- or CO2- the value was 0.05 mA cm−2 in the CO2-saturated solution.
saturated organic media (1 M H2O in DMF) containing 0.1 M After Ru-BNAH was loaded onto the TiO2/Cu2O electrode, the
tetrabutylammonium hexafluorophosphate (n-Bu4NPF6) as the current increment in the N2-saturated solution between light
supporting electrolyte. In the CO2-saturated solution, the Ru- on and off was rarely 0.025 mA cm−2; whereas when it came to
PLA showed distinctly increasing current density compared the CO2-saturated solution, the value reached 0.09 mA cm−2
.
with that in N2-saturated solution starting at −1.1 V. The The current density increasement indicated that the Ru-BNAH
current changes between the CO2- and N2-saturated solutions loading benefited electron migration, which further caused
indicated the Ru-PLA was equipped with a conspicuous CO2- the enhancement of CO2 reduction. To further investigate the
reduction capacity. However, the current density of Ru-BNAH effect of the BNAH proton carrier bio-enzyme on the photo-
in CO2-saturated solution was much higher than that in electric catalytic reduction of CO2, amperometric i–t curves of
N2-saturated solution. Thus, the current density rapidly Ru-PLA/TiO2/Cu2O electrode which had no BNAH proton
increased beginning at −0.80 V, which also indicated the carrier was introduced. As shown in Fig. S4b,† the light
bonded proton carrier favoured CO2 reduction. To further current of the Ru-BNAH/TiO2/Cu2O electrode was much higher
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 4 (a, b) Product generation and TON growth of formic acid with
different potentials for the Ru-BNAH/TiO2/Cu2O photocathode under
−0.9 V; (c, d) product generation and TON growth of formic acid with
different irradiation times for the BNAH/TiO2/Cu2O photocathode.
g-C3N4 and so on (Table 1).10,29,40–45 The high efficiency was
mainly attributed to the effective proton transfer between the
BNAH proton carrier and Ru(II) molecular catalyst through
covalent bond connection. Generally, the BNAH proton carrier
was used in homogeneous reactions, which require a large
number of BNAH to fortify the chance of a collision. However,
the BNAH covalently bonded on a molecular catalyst can
change the intermolecular proton to transfer it into an intra-
molecular one. Additionally, formic acid production reached
80 μmol h−1 in the initial 4 h, but the value then decreased to
22.4 μmol h−1 in the final 4 h. The probable reason for this
was the higher concentration of CO2 and the preeminent cata-
lytic ability of the catalyst in the original 4 h.
Fig. 3 (a) Cyclic voltammetry of Ru-PLA and Ru-BNAH in N2- or CO2-
saturated DMF/1.0 M H2O with 0.1 M (n-Bu4N)(PF6) as the supporting
electrolyte; glassy carbon working electrode (area 0.07056 cm2),
AgNO3/Ag reference electrode (filled with 10 mM AgNO3 in MeCN, 0.46
V vs. NHE, calibrated with the ferrocenium/ferrocene couple), room
temperature; (b) variation of j (CO2) − j (N2) of Ru-BNAH and Ru-PLA
with the applied potential.
Furthermore, the substrate electrode fulfils the following
functions: Cu2O serves as photoabsorber and TiO2 nano-
particles serve as a protective layer to avoid the photocorrosion
of Cu2O. Simultaneously, TiO2 serves as a scaffold to support
the molecular catalyst.28 The formic acid production and TON
were tested following the ratios of Cu2O. As shown in Fig. S6,†
the formic acid yield of 1 : 1 TiO2/Cu2O was higher than those
than that of the Ru-PLA/TiO2/Cu2O electrode in the CO2-satu-
rated solution which clearly indicated the efficiency electron
migration by the bonding.
Excellent photoelectrochemical CO2 reduction activity
Formic acid was the only liquid product detected in the final of other substrates, which reflected the synergistic effect
liquid phase after 8 h PEC CO2 reduction. Due to the variety of between the repressed photocorrosion and favourable light
products in CO2 reduction, trace CO and ethylene were harvesting of Cu2O.
detected in the final gas phase (Fig. S5†). The presence of the
As a pivotal aspect of the photoelectrocatalytic reaction, the
photosensitizer (Cu2O) responded to the generation of ethyl- applied potential plays an important role in the synergetic
ene and the molecular catalyst caused the CO generation. As photoelectrocatalytic CO2 conversion. Therefore, to investigate
shown in Fig. 4, formic acid production increased persistently the impact of the potential for CO2 reduction, formic acid
under the irradiation of simulated solar at −0.9 V potential (vs. generation following the changed potential was analyzed, as
NHE). After 8 h PEC reduction, the formic acid production shown in Fig. 4c. Without the potential applied on the photo-
reached 409.5 μmol with a TON of 321. The catalytic efficiency cathode, the pure photocatalytic CO2 reduction process gave
achieved was significantly comparable with many similar 57.2 μmol formic acid production. Simultaneously, the formic
molecular catalysts in the literature, such as Ru(H4P2O6-C2H4- acid amount increased constantly following the raised applied
bpy)(CO)2Cl2/g-C3N4, Ru(dcbpy)/N-Ta2O5, Ru(II)–Re(I) catalyst potential. However, only 139.5 μmol formic acid was detected
with the presence of BNAH, RuRu′ binuclear catalyst/Au/ at −0.5 V, which indicated that formic acid generation
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
increased slowly at a cathode potential of less than −0.5 V. As
the cathode potential exceeded −0.6 V, formic acid generation
increased significantly alongside the cathode potential rising.
For instance, 409.5 μmol formic acid was detected when a
−0.9 V cathode potential was applied, according to the result
shown in Fig. 4b (the current density grew rapidly when the
cathode potential exceeded −0.65 V in the CO2-saturated solu-
tion). The applied potential was mainly used to separate the
photogenerated electron–hole pairs when the cathode poten-
tial was less than −0.6 V (photocatalytic CO2 reduction domi-
nant process). When the cathode potential exceeded −0.7 V,
the electrocatalytic process not only functioned as an assistant
for the photocatalytic process, but also catalyzed CO2
reduction directly (synergetic photoelectrocatalytic process).
To further investigate the enhancement of CO2 reduction
caused by the BNAH proton carrier, the catalytic efficiency of
the non BNAH proton carrier modified photocathode (Ru-PLA/
TiO2/Cu2O) was assessed (the structure of Ru-PLA is shown in
Scheme S1†). After irradiation for 8 h at −0.9 V, the Ru-PLA/
TiO2/Cu2O photocathode produced 161.7 μmol formic acid,
which was far less than the amount generated by the
Ru-BNAH/TiO2/Cu2O photocathode under the same con-
ditions. When BNAH (0.1 M) was added to the Ru-PLA/TiO2/
Cu2O system, the generated formic acid increased to
231.0 μmol, which was also much less than the Ru-BNAH/
TiO2/Cu2O
photocathode
produced.
This
intuitively
indicated the acceleration of CO2 reduction caused by covalent
bonding.
Probable mechanism for Ru-BNAH catalytic CO2 reduction
In the CO2-saturated solution, in situ Raman spectroscopy was
employed to detect the probable mechanism for Ru-BNAH-
oriented CO2 reduction. The working electrode was prepared
by uniformly loading a Ru-BNAH coating on the surface of the
polished Ag electrode, using Pt wire electrode as the opposite
electrode, and Ag/AgCl as the reference electrode. As shown in
Fig. 5a, from −0.1 V to −1.4 V, two constantly enhanced peaks
at 430 and 570 wavenumbers appeared following the continu-
ous increased potential. The changes might be attributed to
the stretching vibration of Ru(II)–O bond in the Ru-COO− inter-
mediate, which was generated from activated CO2 inserting in
to the d orbital of the Ru(II) centre because this region belongs
to the vibration of metal–ligand (M–L) (CO2 inserting process,
Fig. 5 (a) In situ Roman spectra um in the region 200–600 cm−1 of Ru-
BNAH under CO2 atmosphere following the potential changing in
different wavenumbers; (b) in situ Roman spectra in the region
900–1600 cm−1 of Ru-BNAH under CO2 atmosphere following the
potential changing.
Table 1 Products and TON compared to results presented in the literature
Catalyst
Product
Production
TON
Ref.
Ru-BNAH/TiO2/Cu2O
HCOOH
HCOOH
HCOOH
HCOOH
HCOOH
HCOOH
HCOOH
CO
409.5 μmol
19.3 μmol
1.2 μmol
321
200/20 h
103
89
141
125
17
This work
Ru(H4P2O6-C2H4-bpy)(CO)2Cl2/g-C3N4
RuRu′-binuclear catalyst/Ta2O5
Ru(dcbpy)/N-Ta2O5
g-C3N4/[Ru(bpy)2(CO)2Cl2]
RuRu′ binuclear catalyst/Au/g-C3N4
InP/Ru(deb)CO2
40
41
42
29
43
44
10
45
3.5 μmol h−1
8.8 μmol h−1
1.7 μmol g−1
42 μmol
Ru(II)–Re(I) catalyst + BNAH
NiO-RuRe(II) binuclear catalyst
0.15 μmol
7.9 μmol
207
32
CO
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
Scheme 3 The proposed pathway for the CO2 reduction. The polypyri-
dine ligand of Ru-BNAH was replaced by L to aid showing the CO2
reduction pathway clearly.
as shown in Scheme 3).46,47 It’s worth noting that from −0.1 V
to −0.8 V, the Raman signal intensity changed rarely. From
−0.8 V to −1.4 V, the Raman signal at 430 and 570 wavenum-
bers were enhanced significantly, indicating a demarcation
point of the catalyst for CO2 electrocatalytic reduction begin-
ning at −0.8 V. The region from −0.1 V to −0.8 V was predomi-
nantly related to a photocatalytic process, while the region
beginning from −0.8 V was a photoelectrocatalytic synergetic
process.
Simultaneously, significant Raman peaks were observed at
the wavenumbers of 1130, 1220, and 1560 cm−1, which
increased in line with the increment in potential. The
enhanced peaks may be attributed to the generation of C–O
single bonds and cleaving of CvO double bonds followed with
electron migration (CO2 activation process or electron
migration process).48,49 Furthermore, the weakening Raman
peak at the wavenumber of 1350 cm−1 is attributed to the
proton donor process of the BNAH proton donor (proton coup-
ling process in Scheme 3). Following the proton migration, the
activated CO2 was converted into formic acid, before finally
realizing the catalytic CO2 reduction (product generation
process).
Fig. 6 (a) Headspace GC/MS analyses of pure formic acid and liquid
phase in D2O solution after photoelectrocatalytic CO2 conversion,
respectively. (b) MS peak intensities of pure formic acid and liquid phase
in D2O solution after photoelectrocatalytic CO2 conversion, respectively;
(c) 1H NMR spectrum of the CO2-saturated D2O solution after photo-
electrocatalytic CO2 conversion.
In order to further investigate the proton transfer and
proton source during the Ru-BNAH/TiO2/Cu2O photocathode
catalytic CO2 reduction, the isotope-labelling method was
employed. First, the GC-MS spectrum of unlabelled formic
acid aqueous solution was prepared as the standard. In the nificant red peak appeared at 1.88 min, which was attributed
headspace GC-MS spectrogram, the retention time of standard to the formic acid produced from photoelectrocatalytic CO2
formic acid was 1.88 min with the 46 m/z+ value corresponding reduction. By comparing the GC-MS spectrum of standard
to the rod-like signal peak (as shown in Fig. 6). Then, the CO2 formic acid and the reaction solution, the 47 m/z+ in reaction
photoelectrocatalytic reduction proceeded in CO2-saturated solution was attributed to single deuterium formic acid
D2O solution with Ru-BANH/TiO2/Cu2O as the photocathode. (HCOOD). If the protons of formic acid were derived from the
The headspace GC-MS spectrum of the resulting mixture after solvent directly, the detected MS value of formic acid should
20 min irradiation was recorded. As shown in Fig. 6a, the sig- be 48(DCOOD). Thus, the detection of formic acid with the 47
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
m/z+ indicated a proton (H) replacing a deuterium proton (D)
of deuterium formic acid (DCOOD) during CO2 reduction.
Acknowledgements
Further analysis showed that Ru-BNAH and the solvent were This work was financially supported by the National Natural
the proton sources for CO2 hydrogenation. Therefore, Ru- Science Foundation of China (NSFC, no. 21876128, 21537003,
BNAH was the only remaining compound that could supply H 21477085) and Scientific Research Foundation of Shanghai
atoms for CO2 reduction in the D2O system. Furthermore, by Institute of Technology (39120K186028-YJ2018-7).
structural analysis, BNAH was the only active proton source for
CO2 reduction in Ru-BNAH. Based on the above information,
the generation of mono-deuterium formic acid further proved
that the necessary protons for CO2 photoelectrocatalytic
reduction originated from the BNAH proton carrier.
Notes and references
For the purpose of further investigation of the generation of
mono-deuterium formic acid, the 1H NMR method was used
to detect the production after 8 h irradiation in the CO2-satu-
rated D2O solution. As shown in Fig. 6c, there was a significant
1 R. Francke, B. Schille and M. Roemelt, Chem. Rev., 2018,
118, 4631–4701.
2 S. B. Wang, W. S. Yao, J. L. Lin and X. C. Wang, Angew.
Chem., Int. Ed., 2014, 53, 1034–1038.
3 S. K. Kuk, R. K. Singh, D. H. Nam, R. Singh, J. K. Lee and
C. B. Park, Angew. Chem., Int. Ed., 2017, 56, 3827–3832.
4 S. Sorcar, J. Thompson, Y. Hwang, Y. H. Park, T. Majima,
C. A. Grimes, J. R. Durrant and S. In, Energy Environ. Sci.,
2018, 11, 3183–3193.
5 Y. J. Zhang, J. M. Liu, Y. Zhang and Y. P. Bi, Nano Energy,
2018, 51, 504–512.
6 H. H. Ou, X. R. Chen, L. H. Lin, Y. X. Fang and X. C. Wang,
Angew. Chem., Int. Ed., 2018, 57, 8729–8733.
7 Z. F. Jiang, H. L. Sun, T. Q. Wang, B. Wang, W. Wei,
H. M. Li, S. Q. Yuan, T. C. An, H. J. Zhao, J. G. Yu and
P. K. Wong, Energy Environ. Sci., 2018, 11, 2382–2389.
8 Z. Cao, J. S. Derrick, J. Xu, R. Gao, M. Gong, E. M. Nichols,
P. T. Smith, X. W. Liu, X. D. Wen, C. Coperet and
C. J. Chang, Angew. Chem., Int. Ed., 2018, 57, 4981–4985.
9 R. Long, Y. Li, Y. Liu, S. M. Chen, X. S. Zheng, C. Gao,
C. H. He, N. S. Chen, Z. M. Qi, L. Song, J. Jiang, J. F. Zhu
and Y. J. Xiong, J. Am. Chem. Soc., 2017, 139, 4486–4492.
1
aldehyde H atom chemical shift at 8.23 ppm in the H NMR
spectrum, which corresponded to the characteristic H atom of
HCOO−. If the protons in the formic acid originated from the
solvent directly, only deuterium formic acid DCOO− would be
generated because D2O could only supply the D atom, thus no
aldehyde H would be detected. Therefore, H atoms can only be
obtained from BNAH proton carrier according to the structural
analysis. The result also proved that the BNAH proton carrier
transferred protons to CO2 during photoelectrocatalytic CO2
reduction indirectly.
Conclusion
To sum up, the biomimetic CO2 catalytic reduction interface
(Ru-BNAH/TiO2/Cu2O) equipped with a proton transfer carrier
was constructed by bridging a proton-coupling functionized
molecular catalyst (Ru-BNAH) on the surface of a TiO2/Cu2O 10 Y. Tamaki and O. Ishitani, ACS Catal., 2017, 7, 3394–3409.
substrate electrode. The bionic BNAH proton transfer carrier, 11 A. Taheri and L. A. Berben, Chem. Commun., 2016, 52,
molecular catalyst and photocathode were constructed as a
unified interface, thus achieving efficient photoelectrocatalytic 12 X. Wang, V. Goudy, G. Genesio, J. Maynadie, D. Meyer and
CO2 conversion by in situ proton coupling and photogenerated M. Fontecave, Chem. Commun., 2017, 53, 5040–5043.
electron migration. Under the synergetic photoelectrocatalytic 13 H. Rao, J. Bonin and M. Robert, Chem. Commun., 2017, 53,
function, the formic acid production was much higher than 2830–2833.
that without the proton carrier, following a selectivity of more 14 W. W. Kramer and C. C. L. Mccrory, Chem. Sci., 2016, 7,
than 98%. Furthermore, in situ Raman spectroscopy, in situ 2506–2515.
UV-visible spectroscopy and the isotope experiments in CO2- 15 X. X. Qiao, Q. Q. Li, R. N. Schaugaard, B. W. Noffke,
1768–1777.
saturated D2O solution confirmed the efficient proton-transfer
process during photoelectrocatalytic CO2 reduction. This work
Y. J. Liu, D. P. Li, L. Liu, K. Raghavachari and L. S. Li, J. Am.
Chem. Soc., 2017, 139, 3934–3937.
represented a new demonstration of using covalent linking to 16 A. Nakada, K. Koike, K. Maeda and O. Ishitani, Green
immobilize a proton carrier to generate a semiconductor/ Chem., 2016, 18, 139–143.
metal-complex hybrid photoelectrocatalytic interface and 17 N. Song, M. T. Zhang, R. A. Binstead, Z. Fang and
showed the path towards further research into improving the
system stability and efficiency.
T. J. Meyer, Inorg. Chem., 2014, 53, 4100–4105.
18 B. Siritanaratkul, C. F. Megarity, T. G. Roberts,
T. O. M. Samuels, M. Winkler, J. H. Warner, T. Happe and
F. A. Armstrong, Chem. Sci., 2017, 8, 4579–4586.
19 Y. C. Liu, X. Z. Li, C. Yang and Q. X. Guo, Bioorg. Chem.,
2001, 29, 14–18.
Conflicts of interest
20 S. Shipovskov, A. Bonamore, A. Boffi and E. E. Ferapontova,
Chem. Commun., 2015, 51, 16096–16098.
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
21 J. Patel, K. Majee, E. Ahmad, K. Tanaka and S. K. Padhi, 35 S. Fuldner, R. Mild, H. I. Siegmund, J. A. Schroeder,
̣
Dalton Trans., 2015, 44, 920–923.
22 M. Sahin and E. Ayranci, Electrochim. Acta, 2015, 166, 261– 36 D. Wang, X. Y. Pan, G. T. Wang and Z. G. Yi, RSC Adv.,
270. 2015, 5, 22038–22043.
23 D. Ghosh, T. Fukushima, K. Kobayashi, S. Sen, S. Kitagawa, 37 Z. L. Zhao, X. Wang, J. Q. Si, C. T. Yue, C. G. Xia and
M. Gruberc and B. Konig, Green Chem., 2010, 12, 400–406.
T. Katod and K. Tanaka, Dalton Trans., 2017, 46, 4373–
4381.
24 M. M. Elguindy and E. Nakamaru-Ogiso, J. Biol. Chem.,
2015, 290, 20815–20826.
25 H. Shuo, J. B. Zhang, X. Y. Gao and H. J. Ying, J. Mater.
Chem. B, 2015, 3, 6626–6633.
26 X. Q. Zhu, J. Y. Zhang and J. P. Cheng, J. Org. Chem., 2006,
71, 7007–7015.
F. W. Li, Green Chem., 2018, 20, 832–837.
38 L. X. Yang, Z. Y. Li, H. M. Jiang, W. J. Jiang, R. K. Su,
S. L. Luo and Y. Luo, Appl. Catal., B, 2016, 183, 75–85.
39 I. Platzman, R. Brener, H. Haick and R. Tannenbaum,
J. Phys. Chem. C, 2008, 112, 1101–1108.
40 K. Maeda, K. Sekizawa and O. Ishitani, Chem. Commun.,
2013, 49, 10127–10129.
41 A. Nakada, T. Nakashima, K. Sekizawa, K. Maeda and
O. Ishitani, Chem. Sci., 2016, 7, 4364–4371.
42 S. Sato, T. Morikawa, S. Saeki, T. Kajino and T. Motohiro,
Angew. Chem., Int. Ed., 2010, 49, 5101–5105.
27 H. J. Xu, X. Wan, Y. Y. Shen, S. Xu and Y. S. Feng, Org. Lett.,
2012, 14, 1210–1213.
28 M. Schreier, J. S. Luo, P. Gao, T. Moehl, M. T. Mayer and
M. Grätzel, J. Am. Chem. Soc., 2016, 138, 1938– 43 R. Kuriki, H. Matsunaga, T. Nakashima, K. Wada,
1946.
A. Yamakata, O. Ishitani and K. Maeda, J. Am. Chem. Soc.,
2016, 138, 5159–5170.
44 S. Sato, T. Arai, T. Morikawa, K. Uemura, T. M. Suzuki,
H. Tanaka and T. Kajino, J. Am. Chem. Soc., 2011, 133,
15240–15243.
45 G. Sahara, R. Abe, M. Higashi, T. Morikawa, K. Maeda,
Y. Uedaa and O. Ishitani, Chem. Commun., 2015, 51, 10722–
10725.
29 R. Kuriki, K. Sekizawa, O. Ishitani and K. Maeda, Angew.
Chem., Int. Ed., 2015, 54, 2406–2409.
30 C. Xie, J. L. Song, H. R. Wu, Y. Hu, H. Z. Liu, Y. D. Yang,
Z. R. Zhang, B. F. Chen and B. X. Han, Green Chem., 2018,
20, 4655–4661.
31 X. F. Huang, Q. Shen, J. B. Liu, N. J. Yang and G. H. Zhao,
Energy Environ. Sci., 2016, 9, 3161–3171.
32 E. Kecsenovity, B. Endrődi, Z. Pápa, K. Herńadi, 46 S. Kim and Y. Seo, Appl. Energy, 2015, 154, 987–994.
K. Rajeshwar and C. Jańaky, J. Mater. Chem. A, 2016, 4, 47 Y. L. Deng, A. D. Handoko, Y. H. Du, S. B. Xi and B. S. Yeo,
3139–3147.
33 Y. Yang, S. Ajmal, X. Z. Zheng and L. W. Zhang, Sustainable 48 S. Ponnurangam, C. M. Yun and I. V. Chernyshova,
Energy Fuels, 2018, 2, 510–537. ChemElectroChem, 2016, 3, 74–82.
34 J. A. Hamilton, T. Pugh, A. L. Johnson, A. J. Kingsley and 49 M. C. Figueiredo, I. Ledezma-Yanez and M. T. M. Koper,
S. P. Richards, Inorg. Chem., 2016, 55, 7141–7151. ACS Catal., 2016, 6, 2382–2392.
ACS Catal., 2016, 6, 2473–2481.
Green Chem.
This journal is © The Royal Society of Chemistry 2018