Electrocatalytic hydrogenation of furfural using non-noble-metal electrocatalysts in alkaline medium

Article abstract of DOI:10.1039/d1gc00579k

The production of bio-oil from agricultural waste is a promising route to improve the agricultural value chain. Herein, furfural (FF), a model bio-oil compound, was subjected to electrocatalytic hydrogenation (ECH) in an alkaline medium to produce economically important furfuryl alcohol (FA) and hydrofuroin (HF). The selectivity of ECH products (FA and HF) on Cu, Pt, and Ni-foam electro-catalysts showed that their generation was dependent upon the availability of Hads, which in turn varied with the choice of electrocatalyst and applied potential. Cu-NPNi/NF was obtained through dealloying Cu from a co-electrodeposited Ni-Cu electrode on a Ni-foam substrate, followed by re-electrodeposition of Cu. A porous, high-surface-area bimetallic Ni-Cu catalyst (Cu-NPNi/NF) on Ni-foam yielded high rates of FA and HF generation from furfural, e.g. 118.7 ± 8 and 176.3 ± 3.4 μmol h-1 cm-2 at -1.45 V vs. Ag/AgCl/sat KCl after 1 h of electrolysis in an alkaline electrolyte. 100% conversion of furfural was observed after 2 h of electrolysis with the same catalyst. The high rate of FA and HF formation was ascribed to enhanced adsorbed FF because of the formation of Cu-nanoplates and bimetallic Ni-Cu. We have provided a rational, high-throughput design for preparing highly active nanoporous electrodes for producing industrially relevant chemicals (furfuryl alcohol and hydrofuroin).

Full text of DOI:10.1039/d1gc00579k

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Electrocatalytic hydrogenation of furfural using
non-noble-metal electrocatalysts in alkaline
medium†
Cite this: Green Chem., 2021, 23,
4201
Ram Ji Dixit,a,b Kaustava Bhattacharyya,c,b Vijay K. Ramani *b and
a,d
Suddhasatwa Basu
*
The production of bio-oil from agricultural waste is a promising route to improve the agricultural value
chain. Herein, furfural (FF), a model bio-oil compound, was subjected to electrocatalytic hydrogenation
(ECH) in an alkaline medium to produce economically important furfuryl alcohol (FA) and hydrofuroin
(HF). The selectivity of ECH products (FA and HF) on Cu, Pt, and Ni-foam electro-catalysts showed that
their generation was dependent upon the availability of Hads, which in turn varied with the choice of elec-
trocatalyst and applied potential. Cu–NPNi/NF was obtained through dealloying Cu from a co-electrode-
posited Ni–Cu electrode on a Ni-foam substrate, followed by re-electrodeposition of Cu. A porous, high-
surface-area bimetallic Ni–Cu catalyst (Cu–NPNi/NF) on Ni-foam yielded high rates of FA and HF gene-
ration from furfural, e.g. 118.7 8 and 176.3 3.4 µmol h−1 cm−2 at −1.45 V vs. Ag/AgCl/sat KCl after 1 h
of electrolysis in an alkaline electrolyte. 100% conversion of furfural was observed after 2 h of electrolysis
with the same catalyst. The high rate of FA and HF formation was ascribed to enhanced adsorbed FF
because of the formation of Cu-nanoplates and bimetallic Ni–Cu. We have provided a rational, high-
throughput design for preparing highly active nanoporous electrodes for producing industrially relevant
chemicals (furfuryl alcohol and hydrofuroin).
Received 14th February 2021,
Accepted 2nd May 2021
DOI: 10.1039/d1gc00579k
rsc.li/greenchem
biomass are generated across the country that could be used
as an alternative source of chemicals and fuels.5,6 The thermo-
1. Introduction
Paddy stubble is lignocellulosic biomass containing 34% cell- chemical treatment of agricultural waste, municipal waste, or
ulose, 23.2% hemicellulose, and 15% lignin, which is 6.7% H, sewage sludge leads to the production of bio-oil,7–9 which can
38.8% C, and 40% O elementally.1 The paddy stubble that is produce fuels and platform chemicals on further processing.10
generated during rice harvesting is usually burnt in China and Bio-oil is a mixture of organics (∼40 wt% O) and water and is
India due to the unavailability of proper disposal technologies. acidic in nature. It is a corrosive, chemically unstable, highly
This stubble has the potential to generate 1500 MW of electri- viscous, and low vapor pressure liquid that has a low heating
city or $930 million per year worth of biofuel in a single state value of 19 MJ kg−1, lower than that of diesel or gasoline (44
(Punjab) alone in India.2–4 The U.S. Department of Energy MJ kg−1).11 The bio-oil can be valorized by subjecting to
reported in 2016 that approximately one million tons of further chemical treatments (e.g. hydrogenation and
hydrodeoxygenation)5,12,13 under high temperature (500–600 K)
and pressure (5–10 MPa) conditions. Bio-oils contain chemicals
aDepartment of Chemical Engineering, Indian Institute of Technology Delhi, Hauz
Khas, New Delhi 110016, India. E-mail: sbasu@iitd.ac.in
bDepartment of Energy, Environmental & Chemical Engineering, Washington
University in St Louis, St Louis 63130, Missouri, USA. E-mail: ramani@wustl.edu
cChemistry Division, Bhabha Atomic Research Centre, Mumbai 400085,
Maharashtra, India
dCSIR-Institute of Minerals and Materials Technology, Bhubaneswar 751013, India
†Electronic supplementary information (ESI) available: Polarization curves for
the HER, cyclic voltammograms, relationship between the current density and
scan rate for Cdl, RFA and RHF at a working electrode potential of −1.45 V, SEM
analysis, % selectivity for all catalysts, XPS survey scan, picture of the Nafion®
membrane, change in the FF concentration in the anode compartment with
respect to time, and the FF ECH scheme. See DOI: 10.1039/d1gc00579k
that can be polymerized at high temperatures, which can cause
plugging and high-pressure drops in the existing petroleum
refinery infrastructure during co-processing with crude oil.13
The thermal valorization approach has low product selectivity,
extreme operating conditions and safety/handling issues,
making it economically unattractive. Electrochemical treatment
(e.g. electrocatalytic hydrogenation (ECH)) is a promising
alternative option for the valorization of bio-oil since it is a low
temperature and pressure process that offers in situ generation
of Hads at the catalyst surface, leading to a greater degree of
tuning of products under mild reaction conditions.14,15
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 4201–4212 | 4201

View Article Online
Paper
Green Chemistry
Herein, we study furfural (FF) as a bio-oil model compound a high HF yield (∼94%) and FEHF (∼93%) in 0.1 M KOH.
since it is a relatively abundant constituent (∼3.5 wt%16) of Though most of the reports show FF-treatment in acidic or
crude bio-oil. The ECH of FF generates furfuryl alcohol (FA), buffered media, handling problems arise due to the polymeriz-
hydrofuroin (HF), 2-methyl furan (MF), n-butanol, 1,5-pentane- ation of FF under acidic conditions.27 One solution is to
diol, and levulinic acid, which are used in the fuel, pharma- perform ECH in an alkaline medium since this (i) enhances
ceutical and polymer industries. Our study is mainly focused the conductivity of bio-oil (pH = 9, conductivity = 21 mS
on FA and HF as both are widely used platform chemicals. FA cm−1 28
and thereby reduces the solution resistance for ECH,
)
is used in the foundry industries and in the energy sector14,17 (ii) lowers the bio-oil polymerization rate, thereby enhancing
while hydrofuroin (market cost >$800 per gram) is a chiral handling capability, and (iii) expands the material space of
compound that can be used in the production of resins, jet electrocatalysts since non-noble transition metal based electro-
fuels and pharmaceutical molecules such as stereoselective catalysts are unstable in acidic medium but can be effective in
sulfoxide.18,19 The separation of FF, FA, and HF can be alkaline medium.29,30 FF also undergoes the Cannizzaro reac-
achieved through distillation or liquid–liquid extraction.20
tion31 in alkaline medium to form FA and furoic acid, and the
The product selectivity (S) and faradaic efficiency (FE) in FF conversion rate is high under highly alkaline conditions.
ECH reactions depend upon the nature of the catalyst, applied We chose 0.5 M NaOH as the supporting electrolyte for ECH so
potential, electrolyte pH, initial concentration of reactants, that FA and furoic acid formation through the Cannizzaro reac-
and temperature. To date, ECH of FF has been reported on tion was minimized.
various electrocatalysts such as Cu, Ni, Pb, Fe, and Pt, over the
The morphology of the catalyst and electrode plays an
entire pH range. The S and FE of products generated by FF important role during ECH. An electrode having high active
ECH are correlated with the hydrogen evolution reaction (HER) site density and surface area, good mechanical stability, and
overpotential, generating FA on low HER overpotential metals properly tuned active site coordination numbers typically will
(Pt and Ni) and pinacol derivatives on high HER overpotential exhibit higher activity.33,34 Mass transfer limitations within the
metals (Pb, Fe, and Al).21 Under mild basic conditions (carbon- electrode pore-structure are a critical factor governing the
ate buffer), the trend in HER overpotential is as follows: Pt < overall rate and selectivity of ECH, making pore size tuning
Ni < Cu < Pb; Cu is the most active element for ECH.22 The and optimization necessary.35
carbon free radical generated by the direct transfer of an elec-
In this work, the ECH of FF was investigated on Cu, Ni-Foam
tron to the CHO moiety of FF is responsible for the higher YFA (NF), and Pt in alkaline medium to analyze their activity
and FEFA on the Cu catalyst. Li’s group23 provided broad towards FA and HF production. A binder-free porous Ni struc-
insights into the mechanism of FF reduction in 0.5 M sulfate ture was developed in situ on a NF substrate by a facile electro-
solution (pH = 3) or 0.5 M sulfuric acid (pH = 0.5) using elec- chemical method. Cyclic voltammetry was performed to obtain
trodes modified with organothiol self-assembled monolayers the reduction and oxidation potentials of Ni and Cu ions, and
to control the diffusion of the reactants on the surface. They these potentials were further employed for subsequent co-depo-
concluded that inner sphere and outer sphere electron transfer sition and etching. The porosity and inter-pore spacing were
reactions occurred for FA and HF generation, respectively. FA optimized to obtain high HER currents. Cu was electrodepos-
(pH = 3, SFA = 39.6%; pH = 0.5, SFA = 11%) and MF (pH = 3, ited on optimized porous Ni to get bimetallic Ni–Cu. The rate of
SMF = 15.4%; pH = 0.5, SMF = 66.8%) were produced by the formation and selectivity towards FA and HF were then obtained
direct interaction of FF with the Cu electrode, while HF for- on the optimized catalysts at various potentials and time inter-
mation did not require the direct interaction with the Cu elec- vals. The products were detected by gas chromatography-mass
trode (pH = 3, SHF = 6.2%; pH = 0.5, SHF = 1.8%). Zhao et al.24 spectrometry and quantified by high-performance liquid chrom-
reported 3 wt% Pt/ACF as an active catalyst for the ECH of FF atography. The FF ECH mechanism was advanced for all three
at pH = 1 (SFA = 99%, FEFA = 78%). An increase in temperature optimized electrodes, NF, porous Ni, and bimetallic Ni–Cu,
up to 50 °C enhanced FEFA up to 85% but a further increase in before and after electrolysis, to determine the change in the
temperature decreased FEFA. La-Doped TiO2 increased the electronic environment and to guide selectivity tuning.
current density during the ECH of FF in DMF due to the for-
mation of the redox couple TiO2/Ti(OH)3 and lattice distortion
(SFA = 88.6%).25 The addition of an organic co-solvent also
2. Experimental
2.1 Catalyst preparation
influenced the yield as an electrolyte of 50% acetonitrile (v/v,
pH = 0) showed a high yield on a Cu-electrode (SFA = 36.4%).26
It was also observed that low pH (∼0) and moderate pH (∼3–4) Cu and Pt plates were pretreated using the following steps
promoted high efficiencies for MF and FA formation, before use: polishing by 600 grit sandpaper, immersing in 1 M
respectively.23,26 HF production on Pb and Cd was analyzed in HCl for 1 h followed by rinsing with DI water and drying. NF
10% KH2PO4 electrolyte at 10 mA cm−2 current density and HF was used as the conductive substrate for all the electrodes pre-
yields were 48% and 43%, respectively.20 Sun’s group19 pared by the electrochemical method due to its high surface
reported HF formation using carbon paper as an electrode in area and 3D hierarchical pore structure.36 NF was pretreated
batch and continuous modes using FF in 0.5 M H2SO4 (pH = by ultra-sonication in acetone for 10 min for degreasing, fol-
0), 0.1 M KPi (pH = 7), and 0.1 M KOH (pH = 13) and obtained lowed by rinsing with DI water and drying. Nano-porous Ni
4202 | Green Chem., 2021, 23, 4201–4212
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
supported on NF was prepared by the template-free selective rates (20, 50, 100, and 200 mV s−1) in 0.5 M NaOH electrolyte
electrochemical etching of Cu from co-deposited Ni–Cu.37 The over the potential range of −0.1 V to −0.5 V. The difference
electrochemical co-deposition was performed in a standard between the anodic and cathodic currents, ΔJ−0.3 V (|Ja − Jc| at
three-electrode undivided electrochemical system of volume −0.3 V), was plotted against the scan rate and the slope of the
300 mL. NF (10 cm2), Pt plate, and Ag/AgCl/sat KCl (Pine curve was related to the Cdl (slope = 2Cdl).38 The electrochemi-
Research) were used as working, counter, and reference electro- cally active surface area (ECSA) was calculated on the basis of
des, respectively, for all the electrodeposition processes. All the the obtained Cdl. All the potentials were iR corrected to com-
electrode potentials were reported against Ag/AgCl/sat KCl. The pensate for the potential drop due to internal resistance.
Pt plate was polished by 600 grit sandpaper followed by rinsing Electrolysis was initially performed in the potentiostatic mode
with DI water before every experiment. The deposition poten- for 1 h to optimize the potential range for FA generation, fol-
tials of Ni2+ and Cu2+ were obtained using cyclic voltammetry lowed by an increase in the electrolysis duration up to 3 h to
(CV) with Ni(II) (0.6 M NiSO4·6H2O with 0.5 M H3BO3), Cu(II) obtain 100% conversion of FF. All the experiments were per-
(0.02 M CuSO4·5H2O with 0.5 M H3BO3), and Ni(II) + Cu(II) (0.6 formed at ambient temperature and pressure.
M NiSO4·6H2O, 0.02 M CuSO4·5H2O with 0.5 M H3BO3) electro-
2.3 Product analysis
lytes. The scan rate and potential range for CV were 50 mV s−1
and 0.8 V to −1.2 V respectively. The co-deposited Ni–Cu on NF FA and HF were generated by FF ECH and FA was also gener-
was prepared by using 0.6 M NiSO4·6H2O, x M (0.01 < x < 0.04) ated by the Cannizzaro reaction.31 Samples collected from the
CuSO4·5H2O, and 0.5 M H3BO3 (pH = 4) as electrolytes at −0.85 product mixture were analyzed by a gas chromatograph-mass
V with the deposition duration of y minutes (10 < y < 40). spectrometer (GC-MS) (Agilent 7890 A) equipped with a HP–
Subsequently, a potential of 0.5 V was applied to the resulting 5MS capillary column (30 m × 0.25 mm, 0.25 µm) and an FID
Ni–Cu-deposited NF electrodes to etch out Cu selectively to yield detector. Extraction of the organic phase was performed by
a porous Ni array on the NF. The samples were designated as x– mixing the collected samples with diethyl ether (1 : 1 v/v) fol-
y-NPNi/NF. The values of x and y were optimized to get the best lowed by vigorous stirring (1 h) and leaving the mixture over-
HER current in 0.5 M NaOH electrolyte. Cu was subsequently night to equilibrate. Subsequently, the organic phase was col-
electrochemically deposited on optimized x–y-NPNi/NF using lected, filtered with a 0.22 µm filter and injected into the
0.02 M CuSO4·5H2O and 0.5 M H3BO3 as electrolytes at −0.85 V GC-MS. The injector was set at 250 °C and 1 µL of injection
for z minutes (5 < z < 30) of electrodeposition. This catalyst was was made in the 10 : 1 split mode. N2 was used as the carrier
designated as 0.02 z Cu–0.02 40 NPNi/NF and the value of z was gas at a flow rate of 1 mL min−1. The MS ion source and quad
optimized on the basis of the rate of formation of FA. The temperatures were 230 °C and 150 °C, respectively. The GC-MS
specific catalysts (including benchmarks) prepared and evalu- program was started by maintaining the oven at 40 °C, halted
ated are tabulated in Table S1.†
for 3 min, followed by heating up to 280 °C with a heating rate
of 10 °C min−1 39
Identification of products was primarily
done by comparing the mass spectra against the National
.
2.2 Electrochemical analysis
All electrochemical analyses were performed using a Gamry Institute of Standards and Technology (NIST) mass spectral
Instruments Version 5.6, 2009 electrochemical workstation library. Thereafter, the standards were run through the GC-MS
equipped with Gamry Echem Analyst software. Linear sweep to reinforce product confirmation.
voltammetry (LSV), cyclic voltammetry (CV), and chronoam-
For NMR analysis, the electrolyte was mixed with ether (1 : 1
perometry were used to analyze the electrochemical behavior v/v) and stirred for 3 h, leaving the mixture for attaining equili-
of the electrocatalysts. Electrochemical analyses were per- brium for 12 h. The aqueous and organic phases were separ-
formed in a H-type electrochemical cell with distinct anode ated from which the organic phase was collected. The ether
and cathode compartments (200 mL each), separated by a present in the organic phase was completely evaporated, and
Nafion® 117 membrane. The compartments were filled with the remaining organics were mixed with 0.5 mL of D2O. 1H
150 mL of electrolyte. The reference electrode and counter and 13C-NMR spectra were obtained using a Bruker Ascend™
electrode used were Ag/AgCl/sat KCl, and Pt plate, respectively. 400 MHz NMR. The presence of FA and HF was confirmed by
The prepared electrodes described earlier (Table S1†) were the NMR spectra (Fig. S11†).19,40
used as the working electrodes. All the obtained currents were
The quantification of chemicals was performed by high-per-
reported in the form of current density (mA cm−2) by normaliz- formance liquid chromatography (HPLC) (Agilent 1100)
ing them with the projected geometrical surface area (10 cm2) equipped with an Agilent C18 column (Zorbax 300 SB-C18;
accounting for both sides of the working electrode. 50 mM FF 4.6 mm × 150 mm, 3.5 µm) and a diode-array detector. The
in 0.5 M NaOH was used as the electrolyte for studying the samples were collected from the product mixture, diluted by
ECH. The NaOH electrolyte was purged with N2 starting mixing with DI water (1 : 1 v/v) and filtered with a 0.22 µm
15 min prior to FF addition, and continued till the end of each filter. 1 µL of the sample was injected under the following con-
analysis. Unless otherwise mentioned, all the LSVs were ditions: column temperature = 22 °C, mobile phase = aceto-
recorded in the potential range of 0 to −1.13 V at a scan rate of nitrile (10% v/v) with water (90% v/v), and flow rate = 1 mL
50 mV s−1. The electrochemical double layer capacitance (Cdl) min−1. The wavelengths were chosen to be 215 nm and
was analyzed by performing CV experiments at various scan 253 nm because of the variable product mixture. The rate of
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 4201–4212 | 4203

View Article Online
Paper
Green Chemistry
formation, selectivity, and conversion were calculated using was calculated after 1 h of ECH for all the catalysts, and A is the
the below equations for all the products.
geometric area of the electrode (10 cm2).
41
The moles of products or reactants in this study were cor-
rected to eliminate the contribution of non-electrochemical 2.
reactions (e.g., Cannizzaro reaction) to ensure only the conver-
sion due to ECH was considered. The moles of organic com-
pounds formed or consumed due to electrolysis (nE) were cal-
culated as follows:
% Selectivityð%SPÞ
¼
moles of FF converted to form product P due to ECH ꢁ 100
:
moles of FF converted due to ECH
41
% Conversion of FF due to ECHð%CÞ
¼
3.
4.
moles of FF converted due to ECH ꢁ 100
:
initial moles of FF
nE ¼ nEþB ꢀ nB
Carbon recoveryð%Þ ¼
ðweight of carbon at t ¼ 0 ꢀ weight of carbon at tÞ
weight of carbon at t ¼ 0
where, nE+B is the moles obtained by taking the sample from
the reactor at time t with the potential applied at t = 0, which
comprises all electrochemical, chemical and physical pro-
cesses, whereas nB (Blank run) corresponds to the samples
taken at identical time t, wherein no electrochemical potential
was applied to the system, thereby comprising the non-electro-
chemical processes.
ꢁ 100:
3. Results and discussion
3.1 Preparation and optimization of electrocatalysts
The NF was modified to develop a nanoporous Ni electrode
using electrochemical co-deposition of Ni and Cu followed by
selective electrochemical etching of Cu. To determine the
potentials for deposition and etching, CV experiments were
performed using Ni(II), Cu(II) and Ni(II) + Cu(II) electrolytes
(Fig. 1). In the case of the Ni(II) electrolyte, a reduction peak at
1. RP was the rate of formation of product P (FA or HF) in
µmol h−1 cm−2 and was calculated using the following equation.
CPE ꢁ V
RP ¼
;
t ꢁ A
−0.54 V indicated the deposition of Ni (ENi2+
= −0.45 V)
/Ni0
where CEP is the concentration of the product (P is FA or HF) gen- (Fig. 1a).42 Two oxidation peaks at 0.06 V and 0.21 V during
erated due to electrolysis, V is the catholyte volume (150 mL), RP the anodic scan (forward scan) are attributed to the dis-
Fig. 1 Cyclic voltammetry for the reduction and oxidation of (a) Ni using 0.6 M NiSO4·6H2O with 0.5 M H3BO3 (Ni(II)), (b) Cu using 0.02 M
CuSO4·5H2O with 0.5 M H3BO3 (Cu(II)), (c) Ni–Cu using 0.6 M NiSO4·6H2O, 0.02 M CuSO4·5H2O with 0.5 M H3BO3 (Ni(II) + Cu(II)). Inset figures are
respective experiments with a small potential window. (d) Chronoamperometry using the Ni(II) + Cu(II) electrolyte for 40 min at −0.85 V followed by
80 min at 0.5 V. Qc is the charge involved during reduction and Qa is the charge involved during oxidation. Substrate: Ni-foam, scan rate: 50 mV s−1
.
4204 | Green Chem., 2021, 23, 4201–4212
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
solution/passivation of α-Ni and β-Ni, respectively.43 In the thesized electrodes. Subsequently, Cu was electrochemically
case of the Cu(II) electrolyte, the deposition of Cu started at deposited onto the NPNi/NF surface to increase the activity for
∼0.08 V (ECu2+
= 0.143 V) (Fig. 1b). In the Ni(II) + Cu(II) elec- the FF ECH, which was the highest for Cu–NPNi/NF (Table S1
/Cu0
trolyte, similar reduction currents and peaks at a potential of and Fig. S3†). This sample was used for further ECH studies.
0.0 V and −0.46 V are attributed to Cu2+ and Ni2+ reduction,
Inspection of the SEM images revealed that optimized
respectively (Fig. 1c). During the anodic scan (forward scan), a NPNi/NF had a rough morphology (Fig. 2a, b, and S4b†). The
broad oxidation peak at 0.58 V was attributed to the combined large pore size range (200–260 nm) and inter-pore spacing for
effect of passivation/dissolution of Ni and Cu (Fig. 1c). The Ni– NPNi/NF facilitated the effective diffusion of reactants and pro-
Cu co-deposition and oxidation potentials were selected as ducts, confirmed via HER current density enhancement
−0.85 V and 0.5 V, respectively, from the above CV analyses. (Fig. S1†).36 Upon electrodepositing Cu on NPNi/NF (Cu–NPNi/
Chronoamperometry was performed using the Ni(II) + Cu(II) NF), SEM images depicted the formation of Cu nanoplates
electrolyte at −0.85 V and 0.5 V with the reaction duration of (Fig. 2c). The XRD of Cu–NPNi/NF shows the characteristic
40 min and 80 min, respectively, to obtain nanoporous Ni. The diffraction features of Ni and Cu metals. The XRD features
chronoamperometric analysis indicated the continuous depo- could be indexed to the (111), (200), and (220) planes of face-
sition of Ni2+ and Cu2+ during reduction followed by metal centered cubic Ni (JCPDS: 87-0712) and face-centered cubic Cu
etching during oxidation (Fig. 1d). The selective etching of Cu (JCPDS: 89-2838) as shown in Fig. 2d. The formation of Cu–Ni
during oxidation was confirmed by energy-dispersive X-ray alloy requires the appearance of distinct features apart from
spectroscopy (EDXS) (Fig. S4d, e and g†), which yielded a pure Cu and Ni metals,44 which was not observed for Cu–
porous Ni electrode. The increase in the oxygen content NPNi/NF. However, a slight positive shift in the XRD features
during oxidation confirmed the passivation of the Ni-surface of Cu planes indicated a small interaction of Cu and Ni planes
(Fig. S4g†).
suggesting the formation of bimetallic Cu–Ni,45 which is also
The surface roughness and pore-size of the electrode were observed by XPS (Fig. S10†). Elemental mapping by EDXS
tuned by modulating the Cu concentration in the electrolyte (Fig. 2f) showed preferential Cu deposition on pore vertices as
solution and the co-electrodeposition time (Table S1 and compared to pore bases of the NPNi/NF substrate due to
Fig. S1†). The HER was performed to show the extent of hydro- diffusion limitations of the Cu ion.46 The XPS survey scan
gen coverage on the roughened surface and its pores under showed the presence of Ni and O in NF, and the presence of
cathodic conditions (0 to −1.13 V) suggesting 0.02–40-NPNi/NF Ni, Cu and O in NPNi/NF and Cu–NPNi/NF (Fig. S7a and b†).
(designated as NPNi/NF), which resulted in the best HER The trend in at% obtained from both XPS and EDXS analyses
activity. A 10-fold increase in the ECSA for NPNi/NF was confirmed the presence of Cu along with Ni in the Cu–NPNi/
observed in comparison with the pristine NF electrode NF catalyst (Table 1). As the reactions occur only on the
(Fig. S2†), illustrating high surface roughness for the syn- surface, XPS measurements were chosen for further analyses.
Fig. 2 SEM images for (a) pristine Ni-foam (NF), (b) NPNi/NF, (c) Cu–NPNi/NF. (d) XRD of Cu–NPNi/NF. Elemental mapping of (e) NPNi/NF and (f)
Cu–NPNi/NF.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 4201–4212 | 4205

View Article Online
Paper
Green Chemistry
Table 1 Elemental concentration of Ni, Cu, and O obtained by EDXS and XPS for NF, NPNi/NF, and Cu–NPNi/NF before and after FF ECH
Elemental concentration from EDXS
(at%)
Elemental concentration from XPS (at%)
Sl no.
Type of sample
Ni
Cu
O
Ni
Cu
O
1
2
3
4
5
6
NF
NPNi/NF
Cu–NPNi/NF
NF, FF
NPNi/NF, FF
Cu–NPNi/NF, FF
98.7
84.4
15.6
—
—
—
0
1.3
36.75
32.11
16.96
18.02
4.41
—
63.25
66.69
74.86
81.98
94.44
97.98
1.8
77.0
—
—
—
13.8
7.4
—
—
—
1.20
8.18
—
1.15
1.15
0.87
3.2 Electrocatalytic hydrogenation of furfural in an alkaline
electrolyte
ECH of FF yielded FA and HF in an alkaline electrolyte. The
formation rate and selectivity of FA (RFA, % SFA) and HF (RHF
% SHF) on NPNi/NF and Cu–NPNi/NF were benchmarked
,
Chronoamperometry analysis was performed to analyze FF against Cu, Pt, and NF by performing chronoamperometry at
ECH by studying 0.5 M NaOH with and without FF (the latter various applied working electrode potentials (Fig. 3a, b and
being the blank experiment). A potential of −1.45 V was Table S3†). Upon analyzing the RFA and RHF for benchmark
applied for 1 h on the following working electrodes: NF, NPNi/ Cu, Pt and NF, we observed that FA formation started at an
NF and Cu–NPNi/NF. In each case, the reduction current applied working electrode potential of −1.15 V for all the three
approached more positive values with time due to the decrease catalysts while HF formation started at −1.15 V for Cu and
in FF concentration owing to ECH; no change in reduction −1.25 for Pt and NF. The binding energy of FF was calculated
current was observed in the blank experiment (Fig. S5c†). The using density functional theory and reported to be −1.86 eV,47
Fig. 3 (a) Rate of formation of FA (RFA), (b) rate of formation of HF (RHF), (c) % conversion of FF at potentials: −1.15 V; −1.25 V; −1.35 V; −1.45 V,
working electrodes: NF, Cu, Pt, NPNi/NF, and Cu–NPNi/NF, electrolyte: 0.5 M NaOH + 50 mM FF, and electrolysis duration: 1 h. (d) Change in con-
centration of FF, FA and HF with respect to time for blank (B, without potential) and with electrolysis (E + B) at −1.45 V using Cu–NPNi/NF. 0.5 M
NaOH + 50 mM FF was used as the electrolyte for all the reaction conditions.
4206 | Green Chem., 2021, 23, 4201–4212
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
−1.59 eV,48 and −0.66 eV (ref. 48), respectively, for Pt (111), Ni Cu content in the Cu-NPNi/NF was 16.96 and 8.18 at% respect-
(111), and Cu (111). This indicated stronger adsorption of FF ively (Table 1). The alloying of a smaller concentration of Cu in
on the surface for Pt and Ni as compared to Cu. Since FF was the Ni bulk decreased the gap between the fermi-energy level
bound more strongly on Pt and NF, the FF ECH reaction and d-band,53 which facilitated enhanced chemisorption of
occurred at a higher potential for these catalysts relative to Cu. species (FFads and Hads in our study) on the catalyst surface.
The C–H coupling required for FA generation did not require For Cu–NPNi/NF, the rate of formation of FA and HF based on
−2
significant additional overpotential, and hence FA was gener- the ECSA was 3.2 µmol h−1 cmECSA
and 4.7 µmol h−1
ated at −1.15 V on all catalysts. Cu also facilitated the specific cmECSA−2, respectively, which were higher than that of NPNi/
adsorption of –O from the –CHO moiety of FF such that the NF indicating both active sites increase and bimetallic Cu–Ni
aromatic furan ring of FF was positioned away from the cata- formation facilitates FF ECH (Table S2†). The RFA (118.7
lyst surface, in contrast to Pt and Ni wherein aromatic furan 8 µmol h−1 cm−2), RHF (176.3 3.4 µmol h−1 cm−2), % SFA
ring adsorption was favored.47,48 The adsorbed complexes on (21.6 1.1%), and % SHF (64.2 3.2%) were higher for Cu–
the surface of Pt and NF were less facile as compared to Cu, NPNi/NF as compared to Cu, NF and NPNi/NF. The perform-
which limited the RFA and RHF for the former as compared to ances reported in previous studies were re-formulated for con-
the latter.48,49 Comparing the rate of formation of FA and HF sistency based on our standard definitions (Section 2.5) and
based on the geometric area and ECSA indicated that Cu had are reported in Table 2. To the best of our knowledge, the Cu–
higher intrinsic activity for FF ECH than Pt and NF NPNi/NF catalyst studied herein showed the highest RFA and
(Table S2†).
RHF in an alkaline electrolyte based on all reports to date using
% SFA and % SHF were noted for the Pt electrocatalyst at non-noble metal catalysts (Table 2). The ECH of hydroxy-
−1.25 V and −1.45 V (Table S3†) and showed a decrease in FA methylfurfural was performed in previous literature studies
selectivity and an increase in HF selectivity at higher poten- using Ag considering it a benchmark catalyst to obtain a sig-
tials. Pt is known to generate H2 from Hads at very high rates at nificant ECH performance,54,55 however, FF ECH has not been
high potentials. This favored faster Hads depletion to H2 com- reported to date using Ag. We electrodeposited Ag on NF (ed-
pared to the rate of transfer of H2O at the electrode surface. Ag/NF) and characterized it as mentioned in the ESI.† FF ECH
Hence, the limited availability of Hads at higher potentials using ed-Ag/NF was compared with the Cu–NPNi/NF under the
favored HF formation.23 The availability of FFads and Hads same operating conditions (applied potential, electrolysis dur-
required for FA production (Scheme S1 in ESI†). FA could be ation, and electrolyte) to provide
a precise comparison
diminished if any of these species (FFads and Hads) became (Table 2). The RFA, % SFA were lower and RHF, % SHF were
limited on the surface. FF conversion (Fig. 3c) for all the elec- higher for Cu–NPNi/NF than those of ed-Ag/NF indicating
trocatalysts showed that FF transport to the electrode surface comparable FF ECH performances while considering both FA
was not the limiting factor. Hads generation was more facile on and HF as the products of interest.
Pt as compared to NF and this led to more RFA, % SFA on Pt.
The ECH and electroreduction (ER) mechanisms were
Similar results were obtained by Sun’s group using carbon obtained for the formation of FA and HF, respectively, in an
paper and Cu electrodes suggesting that HF was formed on acidic supporting electrolyte, which is unclear in an alkaline
carbon paper due to the low Hads concentration, however, HF supporting electrolyte.23,55 ECH involves the direct interaction
19
and FA were formed at Cu due to an enhanced Hads
.
H
can of reactant species (FF and H2O) with the electrode surface for
ads
also be increased by changing the morphology and electronic FF conversion, whereas ER occurs in the electrolyte without a
structure of the catalyst surface;50 therefore, NF was modified specific reactant and surface interaction. To obtain mechanis-
to obtain NPNi/NF. NPNi/NF exhibited an enhanced surface tic insights into FA and HF generation at the Cu–NPNi/NF elec-
area and functionality, which resulted in an enhanced HER trocatalyst in the alkaline supporting electrolyte, we performed
(Hads) (Fig. S1†). The increase in RFA, % SFA and decrease in experiments with distinct initial FF concentrations (10 mM,
RHF, % SHF observed for NPNi/NF as compared to NF was due 50 mM, and 100 mM). The RFA was increased and % SFA was
to the elevated Hads seen in NPNi/NF (Fig. 3, Tables S2 and decreased along with an increase in RHF and % SHF at high
S3†). The comparison of NF with Pt and NPNi/NF confirmed initial FF concentrations (Table 2). An increase in the FF con-
the role of Hads in obtaining more selectivity for FA.
centration limited the Hads concentration at the surface result-
Cu was used as a benchmark due to its high intrinsic ing in the decrease of % SFA. An increase in the FFads concen-
activity for FF conversion in an alkaline medium. The high tration increased RFA and RHF. Therefore, FA formation was
surface area micro-crystalline structure of Cu provided higher dependent upon the FFads and Hads concentrations following
FF conversion than bare Cu foil.51 Cu–NPNi/NF was prepared the ECH mechanism,55 whereas RHF and % SHF could be
by the electro-deposition of Cu nanoplates on the NPNi/NF increased either due to the ECH or ER. Nafion® (5 wt%) was
(Fig. 2c). Cu is known for its FFads property51 and the for- coated on the surface of Cu–NPNi/NF to restrict the direct FF
mation of nanoplates provided enhanced surface-active sites, interaction with the surface. The RFA and RHF were decreased
which contributed to enhanced FFads. The formation of bi- for Nafion®/Cu–NPNi/NF samples compared to Cu–NPNi/NF
metallic Ni–Cu using electrochemical deposition has been samples due to the decrease in the FFads concentration
reported52 and bimetallic formation was also confirmed for (Table 2 and Table S4†) indicating the requirement of strong
the synthesized Cu–NPNi/NF using XPS analysis. The Ni and interaction of FF and H2O with the electrocatalyst surface.23
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 4201–4212 | 4207

View Article Online
Paper
Green Chemistry
Table 2 Comparison of FF conversion along with the rate of formation (R), and selectivity (S) for FA and HF with previously reported literature
studies
Potential/current density
and electrolysis duration
% Furfural
conversion
Cathode
Electrolyte
Furfuryl alcohol
Hydrofuroin
Ref.
Cu–NPNi/NF
0.5 M NaOH
−0.45 V vs. RHEa
RFA = 83.9 2.1c
RHF = 8.9 0.67
75.2 4.0% Our work
(−1.45 V vs. Ag/AgCl/sat KCl)
10 mM FF
0.5 M NaOH
50 mM FF
0.5 M NaOH
100 mM FF
0.5 M NaOH
50 mM FF
0.1 M KOH
10 mM FF
1 M KOH
1 h
SFA = 74.3 3.0d
RFA = 118.7 8c
SFA = 21.6 1.1d
SHF = 15.8 0.9
RHF = 176.3 3.4
SHF = 64.2 3.2
Cu–NPNi/NF
Cu–NPNi/NF
ed-Ag/NF
−0.45 V vs. RHEa
73.2 2.3% Our work
1 h
−0.45 V vs. RHEa
RFA = 135.0 5.6c RHF = 341.1 11.2 66.2 2.8
Our work
Our work
19
1 h
SFA = 13.6 0.6d
SHF = 68.7 3.2
−0.45 V vs. RHEa
RFA = 156.7 9.4c RHF = 125.0 5.8
70.6 3.5
NA
1 h
SFA = 29.6 1.8d
SHF = 47.2 3.2
Carbon paper
Cu3P/CFC
2.1 V b
NA
RHF = 130e
1.4 V b
RFA = 90.75c
SFA = ∼100d
NA
37.5%
66
50 mM FF
10 mA cm−2 b
1 h
Cu/Cu foil-400 nm 0.5 M H2SO4 with
H2O/CH3CN (4/1 v/v)
−0.80 V vs. RHE
RFA = 60.0c
NA
76.0%
16.4%
51
23
100 mM FF
1 h
SFA = 15.79d
RFA = 12.99c
Cu foil
0.5 M Sulphate solution −0.55 V vs. RHE
(pH = 3) with CH3CN
(25% v/v)
RHF = 1.02
50 mM FF
1 h
SFA = 39.6d
SHF = 6.2
a
ERHE = EA°g=AgCl + 0.059 × pH + EAg/AgCl, EA°g=AgCl = 0.197 V vs. SHE, pH=13.69. b Two electrode system. c R = µmol h−1 cm−2
.
d S is given in percen-
tage. e RHF is in µmol h−1
.
These results suggested that both FA and HF followed the ECH after FF ECH, the XRD analysis revealed similar XRD features
mechanism for the conditions reported in our study using Cu– indicating no change in the bulk crystal framework during the
NPNi/NF.
reaction. A slight positive shift in the XRD features of Cu
The change in the cumulative concentration of FF, FA and suggested Cu lattice contraction (0.2%) due to an increase in
HF with respect to time was also analyzed for (i) a blank run the Cu–Ni interaction in the presence of a negative potential
(B) which accounts for all non-electrochemical reactions such (Fig. S12c†). SEM analysis depicted no change in the mor-
as the Cannizzaro reaction,31 FF adsorption on the membrane phology of Cu–NPNi/NF (Fig. S12d†). The XPS analysis rep-
surface, and transport of FF towards the anolyte, and (ii) resented a change in the surface-electronic states due to the
electrochemical reaction in conjunction with the reactions transfer of electrons for FF ECH (Fig. 4 and 5). A small
occurring in the blank sample (E + B) (Fig. 3d). In the blank decrease in RFA, RHF and % conversion of FF was due to a
sample (B), the concentration of FF decreased and the concen- small change in lattice parameters and surface-electronic
tration of FA increased with time, primarily due to the states. The overall analysis of RFA, RHF, XRD, SEM and XPS
Cannizzaro reaction of FF in the presence of a base.32 The indicated the suitability of the catalyst for long-term FF ECH
decrease in the FF concentration with time was also partly due operation.
to FF absorption into Nafion® (Fig. S8†) and the continuous
transport of FF concentration to the anode compartment
(Fig. S9†). The FA and HF concentrations remained constant
3.3 Correlation of % selectivity and rate of formation for
various electrodes
throughout the electrolysis duration upon reaching a certain The % S and R for FA and HF were entirely dependent upon
concentration, confirming that no ECH had occurred (Fig. 3d). the ability of the electrode surface to generate Hads and FFads
.
The carbon balance for sample B after 1 h of mixing of FF and This ability was quite different for NF, NPNi/NF and Cu–NPNi/
NaOH showed 92% of carbon recovery, confirming some FF NF electrodes due to differences in their surface-chemical and
loss via absorption and other means. Similarly, for the E + B electronic states. XPS studies were performed for NF, NPNi/NF
sample, an 85% carbon recovery was obtained after the appli- and Cu–NPNi/NF to understand the electronic and chemical
cation of a working electrode potential of −1.45 V for 1 h state of the various elements in these electrodes before and
(Fig. 3d). The difference in the carbon recovery between the B after the ECH of FF. The Ni-2p spectrum of NF (Fig. 4a) mainly
and E + B samples was attributed to the evaporation of high exhibited a doublet of two peaks 2p3/2 (853 eV) and 2p1/2 (∼873
vapor pressure products (e.g. methyl furan), polymer for- eV), each resulting from the spin orbit coupling which typically
mation, and irreversible adsorption of FF on surfaces.16,50
represented NiO.56,57 Further peak splitting of the 2p3/2 peak
The FF ECH at Cu–NPNi/NF was performed for five con- into two peaks at 853 and ∼855 eV, respectively, occurred due
secutive runs for the electrolysis duration of 1 h for each run at to screening of the electron resulting from hybridisation of the
the applied potential of −1.45 V (Fig. S12a and b†). Before and Ni-3d orbital with the O-2p valence orbital in the NiO lattice.58
4208 | Green Chem., 2021, 23, 4201–4212
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
Fig. 4 XPS spectra of Ni-2p for (a) NF, (b) NF after FF ECH (NF and FF), (c) NPNi/NF, (d) NPNi/NF after FF ECH (NPNi/NF and FF), (e) Cu–NPNi/NF, (f)
Cu–NPNi/NF after FF ECH (Cu–NPNi/NF and FF). FF ECH operating conditions: −1.45 V, 1 h, 50 mM FF + 0.5 M NaOH.
There was another peak at 851.9 eV that was attributed to the SFA (14.4 2.7 µmol h−1 cm−2, 9.0 0.6%) were enhanced and
Ni component in the Ni0 state.58 Therefore, the XPS Ni-2p ana- RHF as well as % SHF (15.7 6.7 µmol h−1 cm−2, 19.4 1%)
lysis of the NF confirmed the formation of NiO along with Ni were decreased when compared to NF at −1.45 V applied
metal (Fig. 4a). NiO is an insulator by nature, however, the potential and 1 h of ECH. The Ni-2p3/2 spectrum of the NPNi/
presence of non-stoichiometry and hydroxides on the surface NF catalyst after the FF ECH (Fig. 4d) showed the complete dis-
of NiO makes it conductive.59,60 The presence of NiO on the appearance of the peak at 853 eV as well as the presence of the
NF surface limited its ability to form Hads and subsequently peak at 855 eV. This suggested the hydroxylation of the Ni by
checked the formation of FA, whereas available FFads com- the solvation of the Ni formed by complete reduction of the
bined with another FFads in the presence of applied potential surface phase during FF ECH. The Ni-2p3/2 spectrum for Cu–
and formed HF. For this reason, there was no FA detected on NPNi/NF before FF ECH (Fig. 4e) was analysed and it was con-
NF, while the RHF and % SHF were 49.1 4.9 µmol h−1 cm−2 cluded that the ratio of the intensities of signature peaks of
and 35.5 1.8% respectively at −1.45 V after 1 h of ECH. In Ni–O at 855.6 eV and 853.7 eV for Cu–NPNi/NF was almost 1 : 1
samples taken after FF ECH (Fig. 4b), the Ni-2p3/2 spectrum at which suggested strong non stoichiometry in the NiO gener-
855.3 eV binding energy for NF had relatively more intensity ated during Cu deposition on NPNi/NF (Fig. 4e).58 The peak
when compared to that of the peak at 853 eV, suggesting the corresponding to Ni0 was obtained at 852.4 eV in Cu–NPNi/NF,
presence of Ni–OH as well as Ni–O on the surface.58 The peak a 0.5 eV higher binding energy as compared Ni0 in NF, con-
corresponding to Ni0 (851.9 eV) disappeared (Fig. 4b) because firming an electronic interaction between the Ni0 and the de-
the Ni metal formed during the reduction process was sub- posited Cu.65 The Ni-2p spectrum peaks for Cu–NPNi/NF after
sequently solvated in the electrolyte medium to form surface FF ECH were not clearly visualized (Fig. 4f) because the pres-
Ni(OH)2. The Ni-2p3/2 spectrum for NPNi/NF before FF ECH ence of Cu enhanced FF adsorption, blocking the surface of
(Fig. 4c) suggested the formation of Ni(OH)2 along with non- the electrode. This confirmed that the presence of FFads was
stoichiometry in NiO.58 The passivation of Ni on NPNi/NF was the reason for enhanced RFA, RHF, % SFA, and % SHF on Cu–
confirmed via the formation of Ni(OH)2 and this phenomenon NPNi/NF electrodes.
was also reported elsewhere.61,62 Ni(OH)2 formation rendered
The XPS O-1s spectrum (Fig. 5) comprised peaks at 528.9
the surface of the NPNi/NF to be very conductive and eV, 531 eV, and 533.4 eV corresponding to the –O of the NiO
enhanced Hads formation.63,64 Therefore the RFA as well as % lattice, surface –OH group, and surface H2O, respectively.56,57
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 4201–4212 | 4209

View Article Online
Paper
Green Chemistry
Fig. 5 XPS spectra of O-1s for (a) NF, (b) NF after FF ECH (NF and FF), (c) NPNi/NF, (d) NPNi/NF after FF ECH (NPNi/NF and FF), (e) Cu–NPNi/NF, (f)
Cu–NPNi/NF after FF ECH (Cu–NPNi/NF and FF). FF ECH operating conditions: −1.45 V, 1 h, 50 mM FF + 0.5 M NaOH.
The ratio of the ∼531.4 peak area for NF, NPNi/NF and Cu– firmed the presence of NiO on the NF surface and the presence
NPNi/NF (Fig. 5a, c and e) was 1 : 1.2 : 0.85, confirming that of Ni(OH)2 on NPNi/NF. Ni(OH)2 is reported to have better
Ni–OH was formed during NPNi/NF electrode preparation and Hads affinity as compared to NiO. The poor Hads property of NF
that a fraction of this Ni–OH was subsequently reduced during resulted in a 1.3-fold higher RHF and no RFA as compared to
Cu deposition to fabricate Cu–NPNi/NF electrodes; these NPNi/NF at −1.45 V after 1 h of ECH. The surface analysis of
results corroborated with the Ni-2p XPS data. The ratio of area NF and NPNi/NF through XPS determined the Hads affinity and
under the ∼529 eV peak for NF and NPNi/NF (Fig. 5a and c) supported the conclusion that FA and HF production was
was 1 : 0.77 confirming less presence of the NiO on the NPNi/ dependent on Hads affinity, similar to the conclusion in an
NF surface due to the reduction of NiO during electrode prepa- earlier section for benchmark catalysts (Cu, Pt, and NF). The
ration. Post FF ECH, the lattice O representing the NiO was XPS for Cu–NPNi/NF confirmed that the electronic interaction
still present only for the NF electrocatalyst (Fig. 5b, d and f) of Cu and Ni enhanced FFads and consequently the rate of for-
which was likely due to the excessive availability of NiO on NF mation of FA and HF on Cu–NPNi/NF.
that did not reduce completely. The disappearance of the peak
at ∼533 eV and the appearance of a peak at 535.5 eV were
attributed to the coverage of the electrode surface by FF and
other organics (FA and HF). The ratio of area under this peak
4. Conclusion
(535.5 eV) for NF, NPNi/NF and Cu–NPNi/NF was 1 : 1.75 : 2.28,
confirming the highest adsorbed oxygen (due to organics) in
Cu–NPNi/NF and further validating the observations shown in
Fig. 4f. The Cu-2p3/2 XPS spectra (Fig. S10†) for Cu–NPNi/NF
and Cu plate before FF ECH were compared and it was
observed that the Cu0 peak (931.8 eV) on Cu–NPNi/NF shifted
towards a higher binding energy of 1.3 eV as compared to the
Cu0 peak on Cu plate (930.5 eV). This increase in binding
energy was attributed to the electronic interaction between Cu
and NPNi/NF along with Cu phase segregation, which signified
the formation of bimetallic Ni–Cu.65 The XPS analysis con-
Electrocatalytic hydrogenation of furfural, a model bio-oil com-
pound, was performed with NF, Cu, Pt, NPNi/NF and Cu–
NPNi/NF electrocatalysts in an alkaline electrolyte. Cu was
identified to be the best catalyst for the generation of furfuryl
alcohol and hydrofuroin while NF was unable to produce fur-
furyl alcohol. The percentages of SFA and RFA depended upon
the catalyst’s property to generate Hads. Hydrofuroin formation
was observed at high potentials on Pt where Hads was preferred
over the HER. Hydrofuroin formation was also observed on NF
where Hads was not generated effectively. Hence, the absence
of Hads and excess Hads resulted in hydrofuroin formation.
4210 | Green Chem., 2021, 23, 4201–4212
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
Porous Ni was generated on NF by electrochemical co-depo-
sition of Ni–Cu followed by selective etching of Cu. NPNi/NF
showed the highest HER current among other prepared porous
electrodes, which showed that it favored high Hads. NPNi/NF
generated higher amounts of furfuryl alcohol and lesser hydro-
furoin because of its intermediate Hads property as compared
to NF and Pt. Furthermore, Cu was deposited on the NPNi/NF
substrate (designated as Cu–NPNi/NF) to get Cu nanoplates
and the resultant bimetallic Ni–Cu generated higher amounts
of furfuryl alcohol and hydrofuroin as compared to Cu, NF
and NPNi/NF. The production of furfuryl alcohol and hydrofur-
oin and % conversion of FF on Cu–NPNi/NF were determined
by varying the electrode potential and electrolysis duration.
XPS studies showed that the presence of NiO on NF, Ni(OH)2
on NPNi/NF, and bimetallic Ni–Cu along with non-stoichio-
metry of NiO on Cu–NPNi/NF played a significant role in the
electrocatalyst’s affinity for furfural electrocatalytic hydrogen-
ation. Cu–NPNi/NF produced the highest RFA and RHF, 118.7
8 and 176.3 3.4 µmol h−1 cm−2 at −1.45 V vs. Ag/AgCl/sat KCl
after 1 h of electrolysis as per the analysis of the relevant alka-
line electrolyte compound that is available in the literature.
report-from-haryana-punjab/articleshow/61609110.cms,
(accessed 16 October 2018).
3 NREL, https://www.nrel.gov/continuum/sustainable_trans-
portation/cellulosic_ethanol.html, (accessed 28 October
2018).
4 V. S. Sikarwar, M. Zhao, P. S. Fennell, N. Shah and
E. J. Anthony, Prog. Energy Combust. Sci., 2017, 61, 189–248.
5 M. B. Griffin, K. Iisa, H. Wang, A. Dutta, K. A. Orton,
R. J. French, D. M. Santosa, N. Wilson, E. Christensen,
C. Nash, K. M. Van Allsburg, F. G. Baddour, D. A. Ruddy,
E. C. D. Tan, H. Cai, C. Mukarakate and J. A. Schaidle,
Energy Environ. Sci., 2018, 11, 2904–2918.
6 G. Afreen, S. Pathak and S. Upadhyayula, Chem. Eng. J.,
2020, 400, 125824.
7 Y. Wang, Z. Zeng, X. Tian, L. Dai, L. Jiang, S. Zhang, Q. Wu,
P. Wena, G. Fua, Y. Liua and R. Ruana, Bioresour. Technol.,
2018, 269, 162–168.
8 Y. Yang, Y. Zhang, E. Omairey, J. Cai, F. Gu and
A. V. Bridgwater, J. Cleaner Prod., 2018, 187, 390–399.
9 R. O. Arazo, D. A. D. Genuino, M. D. G. de Luna and
S. C. Capareda, Sustainable Environ. Res., 2017, 27, 7–14.
This study also provides comprehensive information about the 10 X. Liu, B. Li, G. Han, X. Liu, Z. Cao, D. en Jiang and Y. Sun,
formation of hydrofuroin as a function of operating para- Nat. Commun., 2021, 12, 1–11.
meters such as applied potential and duration of electro- 11 C. Zhao and J. A. Lercher, Angew. Chem., 2012, 124, 6037–
catalytic hydrogenation.
6042.
12 X. Wang and R. Rinaldi, Angew. Chem., Int. Ed., 2013, 52,
11499–11503.
Conflicts of interest
13 M. W. Nolte, A. Saraeian and B. H. Shanks, Green Chem.,
2017, 19, 3654–3664.
14 A. S. May and E. J. Biddinger, ACS Catal., 2020, 10, 3212–
3221.
There are no conflicts of interest to declare.
15 R. Chauhan and V. C. Srivastava, Process Saf. Environ. Prot.,
2021, 147, 245–258.
Acknowledgements
16 R. Trane-Restrup and A. D. Jensen, Appl. Catal., B, 2015,
165, 117–127.
17 R. Mariscal, P. Maireles-Torres, M. Ojeda, I. Sádaba and
M. López Granados, Energy Environ. Sci., 2016, 9, 1144–
1189.
18 K. Okano, Tetrahedron, 2011, 67, 2483–2512.
19 X. Shang, Y. Yang and Y. Sun, Green Chem., 2020, 22, 5395–
5401.
20 W. C. Albert and A. Lowy, Trans. Electrochem. Soc., 1939, 75,
367–375.
The authors acknowledge Indian Institute of Technology,
Delhi; Bioenergy-Awards for Cutting Edge Research (B-ACER)
Fellowship sponsored by Department of Biotechnology and
Indo-US Science and Technology Forum, India; and
Washington University in St Louis, USA for the financial
support at various stages. The authors want to thank Nano
Research Facility, IIT Delhi as well as WUSTL for the use of
HPLC and GC-MS instruments. The authors would like to
appreciate the help from the Institute of Materials Science and
Engineering (IMSE), WUSTL for the use of FE-SEM and XPS.
VR would like to thank the VAJRA adjunct faculty scheme spon-
sored by the Department of Science and Technology, India.
21 P. Nilges and U. Schröder, Energy Environ. Sci., 2013, 6,
2925–2931.
22 L. Liu, H. Liu, W. Huang, Y. He, W. Zhang, C. Wang and
H. Lin, J. Electroanal. Chem., 2017, 804, 248–253.
23 X. H. Chadderdon, D. J. Chadderdon, J. E. Matthiesen,
Y. Qiu, J. M. Carraher, J. P. Tessonnier and W. Li, J. Am.
Chem. Soc., 2017, 139, 14120–14128.
References
1 A. Satlewal, R. Agrawal, S. Bhagia, P. Das and
A. J. Ragauskas, Biofuels, Bioprod. Biorefin., 2018, 12, 83– 24 B. Zhao, M. Chen, Q. Guo and Y. Fu, Electrochim. Acta,
107. 2014, 135, 139–146.
2 How stubble burning in Haryana & Punjab is the biggest 25 F. Wang, M. Xu, L. Wei, Y. Wei, Y. Hu, W. Fang and
culprit for poor air quality in Delhi, https://economictimes. C. G. Zhu, Electrochim. Acta, 2015, 153, 170–174.
indiatimes.com/news/politics-and-nation/how-stubble-burn- 26 S. Jung and E. J. Biddinger, ACS Sustainable Chem. Eng.,
ing-is-the-biggest-culprit-for-poor-air-quality-in-delhi-a-ground-
2016, 4, 6500–6508.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 4201–4212 | 4211

View Article Online
Paper
Green Chemistry
27 Y. Zhu, W. Li, Y. Lu, T. Zhang, H. Jameel, H. M. Chang and 46 J. M. Lee, K. K. Jung and J. S. Ko, J. Electrochem. Soc., 2016,
L. Ma, RSC Adv., 2017, 7, 29916–29924. 163, D407–D413.
28 L. K. E. Park, S. Ren, S. Yiacoumi, X. P. Ye, A. P. Borole and 47 B. Liu, L. Cheng, L. Curtiss and J. Greeley, Surf. Sci., 2014,
C. Tsouris, Energy Fuels, 2016, 30, 2164–2173. 622, 51–59.
29 N. Mahmood, Y. Yao, J. W. Zhang, L. Pan, X. Zhang and 48 N. Shan, M. K. Hanchett and B. Liu, J. Phys. Chem. C, 2017,
J. J. Zou, Adv. Sci., 2018, 5, 1700464. 121, 25768–25777.
30 B. S. De, A. Singh, A. Elias, N. Khare and S. Basu, 49 S. Sitthisa, T. Sooknoi, Y. Ma, P. B. Balbuena and
Sustainable Energy Fuels, 2020, 4, 6234–6244. D. E. Resasco, J. Catal., 2011, 277, 1–13.
31 O. Kikhtyanin, E. Lesnik and D. Kubička, Appl. Catal., A, 50 X. Yang, A.-Y. Lu, Y. Zhu, M. N. Hedhili, S. Min,
2016, 525, 215–225.
K.-W. Huang, Y. Han and L.-J. Li, Nano Energy, 2015, 15,
32 P. Parpot, A. P. Bettencourt, G. Chamoulaud, K. B. Kokoh
and E. M. Belgsir, Electrochim. Acta, 2004, 49, 397–403.
33 C. H. Lam, C. B. Lowe, Z. Li, K. N. Longe, J. T. Rayburn,
634–641.
51 S. Jung, A. N. Karaiskakis and E. J. Biddinger, Catal. Today,
2019, 323, 26–34.
M. A. Caldwell, C. E. Houdek, J. B. Maguire, C. M. Saffron, 52 M. Y. Gao, C. Yang, Q. B. Zhang, Y. W. Yu, Y. X. Hua, Y. Li
D. J. Miller and J. E. Jackson, Green Chem., 2015, 17, 601–
609.
34 B. S. Takale, X. Feng, Y. Lu, M. Bao, T. Jin, T. Minato and
and P. Dong, Electrochim. Acta, 2016, 215, 609–616.
53 S. Ghosh, S. Hariharan and A. K. Tiwari, J. Phys. Chem. C,
2017, 121, 16351–16365.
Y. Yamamoto, J. Am. Chem. Soc., 2016, 138, 10356– 54 J. J. Roylance, T. W. Kim and K. S. Choi, ACS Catal., 2016, 6,
10364. 1840–1847.
35 M. Goepel, H. Kabir, C. Küster, E. Saraçi, P. Zeigermann, 55 X. H. Chadderdon, D. J. Chadderdon, T. Pfennig,
R. Valiullin, C. Chmelik, D. Enke, J. Kärger and R. Gläser,
Catal. Sci. Technol., 2015, 5, 3137–3146.
36 B. You, N. Jiang, M. Sheng, M. W. Bhushan and Y. Sun, ACS
Catal., 2016, 6, 714–721.
37 J.-K. Chang, S.-H. Hsu, I.-W. Sun and W.-T. Tsai, J. Phys.
Chem. C, 2008, 112, 1371–1376.
B. H. Shanks and W. Li, Green Chem., 2019, 21, 6210–6219.
56 M. A. Peck and M. A. Langell, Chem. Mater., 2012, 24,
4483–4490.
57 I. Preda, R. J. O. Mossanek, M. Abbate, L. Alvarez,
J. Méndez, A. Gutiérrez and L. Soriano, Surf. Sci., 2012, 606,
1426–1430.
38 M. A. Lukowski, A. S. Daniel, F. Meng, A. Forticaux, L. Li 58 A. P. Grosvenor, M. C. Biesinger, R. S. C. Smart and
and S. Jin, J. Am. Chem. Soc., 2013, 135, 10274–10277. N. S. McIntyre, Surf. Sci., 2006, 600, 1771–1779.
39 Y. Wei, H. Lei, L. Wang, L. Zhu, X. Zhang, Y. Liu, S. Chen 59 F. J. Morin, Phys. Rev., 1954, 93, 1199–1204.
and B. Ahring, Energy Fuels, 2014, 28, 1207–1212.
40 B. Chatterjee and C. Gunanathan, Org. Lett., 2015, 17,
4794–4797.
41 K. Chen, J.-L. Ling and C.-D. Wu, Angew. Chem., Int. Ed.,
2019, 58, 2–9.
60 R. W. Wright and J. P. Andrews, Proc. Phys. Soc., London,
Sect. A, 1949, 62, 446–455.
61 D. S. Hall, C. Bock and B. R. Macdougall, J. Electrochem.
Soc., 2013, 160, 235–243.
62 J. R. Vilche and A. J. Arvia, Corros. Sci., 1978, 18, 441–463.
42 Z. Jamil, E. Ruiz-Trejo and N. P. Brandon, J. Electrochem. 63 J. Wei, M. Zhou, A. Long, Y. Xue, H. Liao, C. Wei and
Soc., 2017, 164, D210–D217. Z. J. Xu, Nano-Micro Lett., 2018, 10, 1–15.
43 M. Fleischmann and A. Saraby-Reintjes, Electrochim. Acta, 64 X. Yu, J. Zhao, L. Zheng, Y. Tong, M. Zhang, G. Xu, C. Li,
1983, 29, 69–75. J. Ma and G. Shi, ACS Energy Lett., 2017, 3, 237–244.
44 B. Seemala, C. M. Cai, C. E. Wyman and P. Christopher, 65 A. R. Naghash, T. H. Etsell and S. Xu, Chem. Mater., 2006,
ACS Catal., 2017, 7, 4070–4082. 18, 2480–2488.
45 K. K. Patra, P. A. Bharad, V. Jain and C. S. Gopinath, 66 X. Zhang, M. Han, G. Liu, G. Wang, Y. Zhang, H. Zhang
J. Mater. Chem. A, 2019, 7, 3179–3189.
and H. Zhao, Appl. Catal., B, 2019, 244, 899–908.
4212 | Green Chem., 2021, 23, 4201–4212
This journal is © The Royal Society of Chemistry 2021