Enhancing electrochemical nitrate reduction toward dinitrogen selectivity on Sn-Pd bimetallic electrodes by surface structure design

Article abstract of DOI:10.1016/j.apcata.2020.117809

Bimetallic palladium (Pd) and tin (Sn) catalysts were electrochemically deposited on stainless steel mesh support by controlling the metal deposition sequence, total electrical charge, and metal composition. Results showed that the preparation procedure a

Full text of DOI:10.1016/j.apcata.2020.117809

AppliedCatalysisA,General606(2020)117809
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Enhancing electrochemical nitrate reduction toward dinitrogen selectivity
on Sn-Pd bimetallic electrodes by surface structure design
Jenn Fang Sua,b, Wei-Fan Kuanc, Ching-Lung Chena,d,e, Chin-Pao Huanga,*
a Department of Civil and Environmental Engineering, University of Delaware, Newark, DE 19716, USA
b Department of Chemical and Materials Engineering, Tamkang University, New Taipei City, 25137, Taiwan
c Advanced Materials Characterization Laboratory, University of Delaware, Newark, DE, 19716, USA
d Department of Safety, Health and Environmental Engineering, Ming Chi University of Technology, New Taipei City, 24301, Taiwan
e Center for Environmental Sustainability and Human Health, Ming Chi University of Technology, New Taipei City, 24301, Taiwan
A R T I C L E I N F O
A B S T R A C T
Keywords:
Bimetallic palladium (Pd) and tin (Sn) catalysts were electrochemically deposited on stainless steel mesh support
by controlling the metal deposition sequence, total electrical charge, and metal composition. Results showed that
the preparation procedure affected the crystal structure of bimetallic Pd-Sn catalysts, which significantly in-
fluenced nitrate removal efficiency and dinitrogen selectivity. Electrode with Sn on the outside surface exhibited
relatively greater nitrate removal rate constant and nitrate conversion. The Sn to Pd molar ratio and the elec-
trical charge applied during electrode preparation also affected the nitrate reduction performance. The SS/
Sn0.2Pd0.8-497 electrode exhibited 88, 89, 79, and 9% of total nitrate removal, dinitrogen selectivity, dinitrogen
yield, and NH4+ selectivity, respectively. Among the three major facets, (214), (131) and (420) of Sn3Pd alloy on
the electrode surface, (420) exhibited the most critical effect on the dinitrogen yield. Crystal structure of cat-
alysts controls the reactivity and selectivity of electrochemical reduction as exemplified by nitrate.
Sn-Pd bimetallic electrode
Nitrate reduction
Dinitrogen selectivity
Surface structure
1. Introduction
nitrogen gas.
As far as drinking water purification is concerned, chemical reduc-
Nitrate is a major groundwater contaminant in the United States
and world. According to US Geological Survey (USGS), about 7% of
2400 private wells show nitrate concentration exceeding the U. S.
Environmental Protection Agency (USEPA) drinking water standard of
10 mg/L. Natural nitrate level in groundwater is typically lower than
10 mg/L. Excessive nitrate concentration is mainly caused by anthro-
pogenic sources, inefficient use of fertilizers, improper disposal of ni-
trate-laden domestic sewage, agricultural or industrial wastewater, and
emissions from combustion engines. High nitrate level in drinking
water may cause methemoglobinemia in infants and gastrointestinal
cancer in adults, therefore there is urgent need of efficient and sus-
tainable technology for the renovation of nitrate-contaminated water is
ever-growing.
tion process [6–10] has numeral advantages over above physical-che-
mical methods, such as fast reaction, easy operation, and generation of
relatively benign nitrogen by-products. Unfortunately, ammonia is an
inevitable byproduct in most chemical reduction processes. Post-treat-
ment processes for ammonium ion removal, such as ion exchange resins
or gas stripping are always needed as to safeguard drinking water
safety, thereby requiring additional treatment costs.
Electrochemical method is a green approach for the water/waste-
water treatment industry [11–16]. It offers a number of advantages
over traditional treatment processes such as environmental compat-
ibility, safety, energy efficiency, selectivity, and versatility.Addition-
ally, electrochemical processes do not generally produce secondary
pollutants, therefore, reduce process significant treatment costs relative
to biodegradation and conventional chemical treatment methods.
A wide variety of cathodic metals and alloys including Pt [17], Pd
[18], Cu [19], Ag [20], Rh [21], Sn [22], Pb [23], Cu-Pt [24], Cu-Sn
[25], Cu-Zn [26], Cu-Pb [20], Cu-Fe [27], Cu-Ni [20,28], Cu-Pd
[29–32], and Sn-Pd [29,33,34] have been studied for electrochemical
nitrate reduction. Recently, many studies have focused on
Common physical-chemical methods for nitrate removal include
adsorption (e.g., ion exchange resins) [1], concentration, e.g., reverse
osmosis [2] and electrodialysis [3]. These methods are reliable, effec-
tive, fast, easy to operate, and appealing to both small and medium-size
water and wastewater treatment plants [4,5]. However, the above
methods do not transform nitrate to benign species, specifically
⁎ Corresponding author.
E-mail address: huang@udel.edu (C.-P. Huang).
https://doi.org/10.1016/j.apcata.2020.117809
Received 26 June 2020; Received in revised form 26 August 2020; Accepted 27 August 2020
Availableonline01September2020
0926-860X/©2020ElsevierB.V.Allrightsreserved.

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
electrochemical denitrification using bimetallic systems as the addition
of a second metal seemed to significantly has improved the nitrate re-
activity, nitrogen selectivity, and electrode stability. Bandarenka and
Koper [35], studied the electrochemical reduction of nitrate using
various monolayer/substrate electrocatalysts and reported that the
presence of a second metal (monolayer) on the electrode surface se-
lectively produced preferential products, such as NO (on Cu/Pt), N2O
+
(on Sn/Pt), NH2OH (on Ge/Pd), or NH4 (on Sn/Pd) [35]. Roué et al.
studied the conversion of nitrate to nitrogen over monometallic Cu
[36,37] and bimetallic Cu0.7Ni0.3 electrode [37], and reported that ni-
trate conversion over the Cu0.7Ni0.3 bimetallic electrode (92%) was
about two times that on the Cu electrode (49%) [36,37]. Xu et al. de-
monstrated that Cu0.8Ni0.2 and Cu0.8Pb0.2 bimetallic electrodes ex-
hibited decreasing ammonia selectivity while improving nitrogen yield
compared to Cu monometallic electrode [38]. Although theoretical and
experimental studies have demonstrated the superiority of bimetallic
catalysts in electrochemical nitrate reduction, a direct 100% selectivity
of nitrate reduction to nitrogen gas remains the ultimate goal and
challenge. Therefore, there were increasing interests on the develop-
ment of new electro-catalytic materials with improved N2 selectivity in
the last few years.
Koper et al. studied the electrochemical activity for nitrate reduc-
tion using density function theory (DFT) calculations incorporated with
a series of laboratory experiments and demonstrated that electro-
chemical nitrate reduction was electrode structure sensitive and that a
perfect monometallic and monocrystalline Pt(100) crystal phase
yielded the highest catalytic activity [39–43]. Most significantly, Koper
et al. reported that active sites consisting of a square or rectangular
arrangement involving four surface metallic atoms on the Pt(100) phase
were essential to most bond-breaking reactions because of higher
bonding energy between bond-breaking precursors and electrode sur-
face atoms [44]. On the contrary, the appearance of Pt(111) phase,
atoms arranged in diamond shapes, would lead to a decrease in cata-
lytic activity because of weak binding energy [35]. Kato et al. compared
the catalytic activity of tin-modified single crystalline electrodes of
palladium, platinum and palladium-platinum alloy with the (111)
surface versus the (100) surface on nitrate reduction and concluded that
the tin-modified (100) surface exhibited greater nitrate reactivity than
that of the (111) surface in acid and neutral media [45].
In short, it has been reported recently reported that the reactivity
and selectivity of nitrate reduction in water was significantly influenced
by the surface roughness and crystallite size of bimetallic electrodes
[34]. The general reaction mechanism was proposed as the following:
NO3 + Sn
Sn[NO3 ]
(1a)
H+ + Pd + e
Pd[H]
(1b)
Sn[NO3−] + 2 Pd[H] → Sn[NO2−] + Pd[H2O]
Sn[NO2−] + Pd[H] → Sn[NO] + Pd[OH−
(1c)
(1d)
(1e)
]
2 Sn[NO] + 2 Pd[H] → Sn[N2O] + Pd[H2O] + Sn+ Pd
Sn[N2O] + 2 Pd[H] → Sn[N2] + Pd[H2O] (1f) → Sn[N2] + Pd[H2O]
(1f)
Sn[N2] → Sn + N2
(1g)
(1h)
(1i)
(1j)
(1k)
(1l)
Sn[NO] + Pd[H] → Pd[NH] + Sn[O]
Pd[NH] + Pd[H] → Pd[NH2] + Pd
Pd[NH2] + Pd[H] → Pd[NH3] + Pd
Pd[NH3] → NH3 + Pd
NH3 + H2 O NH+4 + OH
The major reaction scheme included: (i) adsorption of nitrate onto
2

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 1. Schematic diagram of electrochemical nitrate reduction experiments.
an empty Sn site via its oxygen atoms; (ii) adsorption of hydrogen on
the empty Pd site; (iii) direct surface electrochemical reduction of ni-
trate to nitrite; (iv) a series of surface reactions involving the surface
hydrogen atoms (on the Pd site) and various surface oxy-nitrogen
species (on the Sn site), which eventually led to the formation of ni-
trogen molecules; (v) further reaction between the surface hydrogen
species and surface oxy-nitrogen species generated ammonium ion.
The present study was our continuing effort to further explore the
effects of catalyst preparation process on the electrode surface structure
and how such alloy surface structure affected the reactivity and se-
lectivity of nitrate reduction. Most previous work in bimetallic elec-
trodes studied and discussed the influence of bimetal ratio on the se-
lectivity of the nitrate reduction pathway [46–50]. To the best of our
knowledge, this is among the first study to investigate the surface
structure design and surface atom arrangements of bimetallic alloy
electrodes and its correlation to nitrate reduction performance and se-
lectivity. In the present study, we modified the metal deposition se-
quence, the total electrical charge applied, and the atomic ratio be-
tween the two metals as a means to control the surface property of the
bimetallic electrode. It was hypothesized that adjusting the crystal fa-
cets on electrode surface would influence the bond-breaking and bond-
formation reactions among nitrate ions, electrode surface, and reaction
intermediates, thereby enhancing nitrate reactivity and nitrogen se-
lectivity. Tin-modified palladium, Sn-Pd, catalyst was selected for this
study, in which Sn was intended to be the O-affinity and Pd be the H-
adsorbing metal, respectively, due to its high reactivity compared to
Cu-Pd [33,51]. This work aimed at developing a synthesis procedure for
better design of specific electrochemical electrodes of high efficiency in
transforming nitrate to benign nitrogen gas for water treatment appli-
cations.
methanesulfonic acid were purchased from Sigma-Aldrich, St Luis, MO
USA. Deionized water was treated with Mega-Pure System (Model MP-
290). Platinum wire (Fisher Scientific, 1284987, od: 0.5 mm) was
purchased from Fisher Scientific, Pittsburgh, PA, USA. Stainless steel
mesh (corrosion-resistant 304 stainless steel woven wire cloth,
100 × 100 mesh, 0.0045″ wire diameter) was obtained from McMaster-
Carr Co. Elmhurst, IL USA. For catalytic material preparation, tin (II)
chloride (purity > 98%), and palladium (II) chloride (purity > 98%)
were purchased from Sigma-Aldrich, St Luis, MO USA. All chemicals
were used as received without further purification.
2.2. Electrode preparation
The electrode preparation procedure was as follows. The raw
stainless-steel mesh (SS) was cut into small pieces (ca. 1.5 cm × 8 cm)
and washed with detergent thoroughly. Afterward, the SS mesh was
rinsed with deionized water several times then dried in a dryer at room
temperature for 1 h. The metal ion solutions were prepared by dissol-
ving tin chloride (SnCl2) and palladium chloride (PdCl2) in deionized
water at concentration of 0.1 M and 0.01 M, respectively. All electro-
deposition experiments were carried out at room temperature by using
a two-electrode system, which was connected to a potentiostat (Model
WP705B, Vector-VID). Graphite (ca. 1.5 cm × 8 cm) was the anode and
the surface-cleaned SS was the cathode. Both electrodes were immersed
in the metal ion solution in a 250-mL beaker. Metal deposition process
was performed first at the constant current of 0.3 A @1.8 min for 6
times (total 10.8 min), and then increased current to 0.6 A for 1 min for
additional deposition of both metals, with Sn being deposited ahead of
Pd. For example, a typical SS/Sn-Pd electrode was prepared by im-
mersing the SS in the 0.1 M SnCl2 solution for Sn deposition at 0.3 A @
1.8 min for 6 repeating times (total 10.8 min), and then the current was
increased to 0.6 A to continue deposit Sn for another 5.4 min to com-
plete prepare the SS/Sn precursor electrode. This multi-step electro-
deposition was necessary for better nitrate reduction performance as
detailed previously [34]. The SS/Sn precursor electrode was rinsed with
deionized water followed by immersing in 0.01 M PdCl2 solution for Pd
deposition at 0.3 A for 6 times @ 0.5 min followed by continuous de-
position at 0.6 A for 1 time @1.5 min. Finally, the electrode was washed
2. Materials and methods
2.1. Materials
Sodium nitrate was obtained from Fisher Scientific, Pittsburgh, PA,
USA (ACS certified grade). Perchloric acid was purchased from ACROS,
Fair Lawn, NJ, USA. Sodium hydroxide (purity > 97.0%) and
3

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 3. XRD characterization of (a) SS, (b) SS/Pd0.2Sn0.8-497, (c) SS/Sn0.8Pd0.2
-
497, (d) SS/Sn0.8Pd0.2-166, (e) SS/Sn0.8Pd0.2-332, (f) SS/Sn0.8Pd0.2-1490, (g)
SS/Sn0.5Pd0.5-497, and (h) SS/Sn0.2Pd0.8-497. The standard XRD pattern of Pd
(PDF # 00-046-1043) and Sn (JCPDS # 04-0673) are also presented for com-
parison purpose.
optimized to ensure 100% current efficiency.
To explore the effects of electrode preparation methods on nitrate
reduction, three different factors including the sequence of metal de-
position, the total electrical charge applied, and Sn-Pd ratio were stu-
died in this work. Electrodes of SS/Pd0.2Sn0.8-497 (Sn on the outside
surface) and SS/Sn0.8Pd0.2-497 (Pd on the outside surface) were studied
for the influence of metal deposition sequence on nitrate reduction. The
effects of total charge (Q) for electrode deposition on the nitrate re-
duction were examined by four electrodes prepared with various Q
values including SS/Sn0.8Pd0.2-166, SS/Sn0.8Pd0.2-332, SS/Sn0.8Pd0.2
-
497, and SS/Sn0.8Pd0.2-1490. The effects of Sn-Pd ratio were studied
using SS/Sn0.8Pd0.2-497, SS/Sn0.5Pd0.5-497 and SS/Sn0.2Pd0.8-497.
Prior to each nitrate reduction experiment, the SS/Sn-Pd electrode was
regenerated by electrochemical reduction at constant current of 0.05 A
for 20 min in pure 0.1 M HClO4 solution as to reduce the divalent tin
and palladium ions to their corresponding elemental states, which was
confirmed by analyzing the oxidation state of Sn and Pd with XPS be-
fore and after the pretreatment (data not shown). Furthermore, there
were numeral reports on the electrochemical regeneration of metallic
electrodes [52]. Table 1 summarizes detailed preparation conditions for
all electrodes in this study.
Fig. 2. SEM micrographs of (a) SS/Pd0.2Sn0.8-497, (b) SS/Sn0.8Pd0.2-497, (c)
SS/Sn0.8Pd0.2-166, (d) SS/Sn0.8Pd0.2-332, (e) SS/Sn0.8Pd0.2-1490, (f) SS/
Sn0.5Pd0.5-497, (g) SS/Sn0.2Pd0.8-497, (h) SS/Sn0.2Pd0.8-497 after nitrate re-
duction experiments.
2.3. Characterization of catalysts
Crystalline Sn-Pd alloy was identified by X-ray diffraction (XRD,
Rigaku D-Max B) with Cu Kα radiation at 40 kV beam voltage and
40 mA current, and scan angle from 30° to 70°. The surface element and
crystal structure were characterized by matching XRD pattern of each
sample to known standard patterns. High resolution transmission
electron microscopy (HRTEM, Jeol JEM-2010) was also applied to
support the crystal structure determination. HRTEM was operated at
200 kV, and the magnification of HRTEM was 500 K. A small amount of
sample was put into the sample tube filled with a 95% ethanol solution.
After agitating under an ultrasonic environment for 90 min, one drop of
the dispersed slurry was dipped onto a carbon-coated copper mesh
(300#) (Ted Pella Inc., CA, U.S.A.) and dried at room temperature in
vacuum overnight. Surface morphology and structure of the metal de-
posits were investigated by a scanning electron microscopy (JSM
7400 F) at an accelerating voltage of 3 kV. Cyclic voltammetry (CV) and
linear sweep voltammetry (LSV) were conducted using a three-elec-
trode configuration with an electrochemical workstation (CHI 611).
with deionized water, dried under ambient condition, and stored in a
nitrogen chamber until uses. The mole fraction, xi, of the metal catalyst,
i.e., Pd or Sn, on the electrode was determined by the following
equation:
(Ii × ti)/F
Qi/F
Qi/F
xi =
=
(Ii ti)/F
(2)
where Qi is the total electric charge consumed by the ith species (i.e., Sn
or Pd) in coulombs (C), Ii and ti are the current (A) and deposition time
(sec) applied to the ith species, and F is the Faraday constant (96,485 C/
SS/Sn Pd
eq). All electrodes used in this work were denoted as
-Q,
x
(1 x )
i
i
in which Q was the total charge (C = I x t), I was current (A), and t was
the deposition time (sec). The validity of Pd to Sn ratio was reasonably
assured because the experimental conditions for metal deposition were
4

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 4. (a) Representative HRTEM image of SS/Sn0.2Pd0.8-497 electrode and (b) corresponding selected-area electron diffraction (SAED) patterns.
The counter and reference electrode used were Pt rod and SCE, re-
spectively. LSV tests were measured in 8.0 × 10−3 M nitrate solution
with a scan rate of 10 mV/s from -0.7 to -1.1 V (vs SCE). The electro-
chemical impedance measurements (EIS) were carried out in the fre-
quency range from 100 kHz to 0.1 Hz with oscillation amplitude of
10 mV/s using a Princeton Applied Research potentiostat/galvanostat
(Parstat 2263). The reference electrode was a SCE and all the potential
measurements were based on SEC.
partition keeping the cathodic chamber pH low at 1.5. The pH value
remained relatively unchanged during the reaction due to the presence
of the strong acid in the electrolyzed solution. Nitrate and nitrite were
measured with a Dionex ion chromatograph (IC) system equipped with
a GP50 pump, ED 40 conductance detector, Dionex IonPac AC20
column (4 mm × 250 mm.), and Dionex AS 40 automatic sampler. The
effluent mobile phase was a mixture of deionized water and 50 mM
NaOH with a flow rate of 1 mL/min and injection volume of 25 μL. At
different time intervals, 3 mL of samples were withdrew with-
drawnfrom the electrochemical cell for chemical analysis of related
residual nitrogen species. A Dionex ion chromatograph (IC) system
equipped with a GP40 pump, CD 20 conductance detector, Dionex
IonPac CS16 column (0.5 mm × 250 mm.), and Dionex AS 3500 auto-
matic sampler were used to measure the ammonium concentration. The
effluent mobile phase was a mixture of deionized water and 0.1 M of
methanesulfonic acid, with a flow rate of 1 mL/min and injection vo-
lume of 25 μL. Gaseous products from nitrate reduction reaction were
2.4. Electrochemical nitrate reduction
Nitrate reduction experiments were performed in a divided two-
electrode system, as shown in Fig. 1. Graphite electrode was the anode
and SS/Sn-Pd electrode as the cathode. The working solution was
8.0 × 10−3 M of NaNO3 and the current was constant at -0.04 A. The
volume of the cathode chamber and anodic chamber were 467 and
500 mL, respectively. An H-type reactor with a cationic membrane
5

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 5. Nitrate reduction reaction in different electrode systems. (a) Nitrate concentration versus reaction time; (b) The fitting of kinetic model (dash line) and
experimental data (dot). (c) Nitrite concentration versus reaction time; (d) Ammonium concentration versus reaction time; (e) Nitrogen concentration versus reaction
time. The reported values are averages from three samples with standard deviations indicated by error bars on the data points. Experimental conditions: pH = 1.5;
[NaNO3] = 0.008 M; temperature =25 °C; constant current = -0.04 A; voltage = 5 V to 6 V.
d[NO3 ]t
dt
analyzed with GC-MS (HP 6890 series, Agilent Technologies, Santa
r =
=
k [NO3 ]t
(3)
Clara, USA) equipped with a bonded polystyrene-divinylbenzene based
column (HP-PLOT/Q, 15 m x 0.32 mm id). The mass spectrometric
analysis was undertaken with a mass selective detector 5973 (Agilent
Technologies, Santa Clara, USA). Note that the gaseous nitrous oxide
(NxO) was not detected by GC-MS in all experiments. The concentration
of N2 was determined by the combination of mass balance and GC-MS.
GC-MS was applied mainly to detect the presence of gaseous nitrous
oxides (NxO) produced as end products. As the gaseous nitrous oxide
(NxO) was not detected by GC-MS in the present work, we assumed that
NO2−, NH4+, and N2 were main reduction products and thus gaseous
nitrogen could be quantified by mass-balance.
in which [NO3 ]t was the nitrate concentration (mol/L) at reaction time
t. The fraction of nitrate conversion, XNO3 , was calculated by Eq. 4:
[NO3 ]0 [NO3 ]t
XNO
=
3
[NO3 ]0
(4)
in which [NO3 ]0 was the initial nitrate concentration. The selectivity of
reaction products, i.e., SNO , SNH4+, and SN were calculated by the
2
2
following equations:
[NO2 ]t
[NO3 ]0 [NO3 ]t
SNO
=
2
Nitrate conversion rate constant, k, was obtained by fitting experi-
mental date with the first-order rate equation as shown in Eq. 3 [22,34]:
(5a)
(5b)
[NH4+]t
[NO3 ]0 [NO3 ]t
+
SNH
=
4
6

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Table 2
noise. The result indicated that the orientation of facets (214) and (420)
of the electrode were not paralleled to the sample surface [58,60,61].
The intensity of all XRD patterns were normalized against stainless steel
support, the variance in peak intensity among electrodes suggested that
the electrodeposition procedure influenced the preferred orientation for
the growth of crystallites as well as crystal facet distribution. Fig. 4
gives the HRTEM image and the corresponding selected-area electron
diffraction (SAED) patterns of SS/Sn0.2Pd0.8-497 electrode. Results
showed three strong continuous lattice fringes with lattice spacing of
0.250 nm, 0.202 nm, and 0.138 nm. Note that the spacing calculated
from SAED pattern matched the d-spacing of the (214), (131), and
(420) plane in Sn3Pd phase (i.e. 0.240 nm, 0.204 nm, and 0.145 nm,
respectively, according to ICDD PDF Card No. 00-015-0575). The result
was consistent with XRD characterization, supporting the formation of
Sn3Pd alloy.
Kinetic constants, nitrate removal, selectivity of nitrite, ammonium, and ni-
trogen, nitrogen yield, and charge transfer resistance.
Samples
k (h−1
)
R2
XNO
SNO
SNH
+
SN
Rct (Ω)
N
2
2
3
2
4
SS/Pd0.2Sn0.8-497
SS/Sn0.8Pd0.2-497
SS/Sn0.8Pd0.2-166
SS/Sn0.8Pd0.2-332
0.32
0.24
0.14
0.24
0.96 0.74
0.96 0.65
0.91 0.47
0.98 0.63
0.98 0.56
0.99 0.60
0.92 0.88
0.02
0.03
0.03
0.02
0.03
0.03
0.02
0.30
0.09
0.13
0.18
0.13
0.14
0.09
0.68 0.50 32
0.88 0.58 40
0.84 0.38 72
0.80 0.53 41
0.84 0.48 59
0.83 0.50 48
0.89 0.79 23
SS/Sn0.8Pd0.2-1490 0.20
SS/Sn0.5Pd0.5-497
SS/Sn0.2Pd0.8-497
0.22
0.47
1. k = nitrate removal rate constant; R2 = coefficient of determination; XNO
= nitrate conversion; SNO , SNH+, SN = selectivity of NO2−, NH4+, and N23;
2
2
4
= nitrogen yield; Rct = charge transfer resistance.
2N. 2Example electrode identification: SS/Pd0.2Sn0.8-497 = 20% Pd right on the
SS surface and 80% Sn on the outside surface prepared with total charge of
497C.
3.2. Optimizing electrode preparation procedures
Fig. 5 shows the concentration change of NO3−, NO2−, NH4+, and
N2 as a function of reaction time at constant current of −0.04 A. Results
present in Fig. 5a indicated that nitrate was readily electrochemically
reduced over all SS/Sn-Pd electrodes. Table 2 gives the kinetic constant
of NO3− reduction over seven different electrodes. The kinetic analysis
[NO3 ]0 [NO3 ]t [NO2 ]t [NH4+]t
SN
=
2
[NO3 ]0 [NO3 ]t
(5c)
in which [NO2 ]t and [NH+]t were the concentration of nitrite and am-
monium, respectively, at 4time t.
−
revealed that NO3 reduction follow the first-order-kinetic model
(R2 > 0.91) as shown in Fig. 5b. Electrode with Pd on the outside
surface, i.e., SS/Sn0.2Pd0.8-497 exhibited relatively greater nitrate re-
3. Results and discussion
moval rate constant and nitrate conversion (k =0.47 h−1
,
3.1. Characterization of synthesized bimetallic electrodes
XNO = 0.88) than that with Sn deposited on the outside of the elec-
3
trode, i.e., SS/Pd0.8Sn0.2-497 (k =0.24 h−1
,
XNO = 0.65) under
3
Fig. 2 shows the surface morphology of electrodes characterized by
SEM. Fig. 2a and b are the SEM micrographs of SS/Pd0.2Sn0.8-497 (Sn
on the outer layers) and SS/Sn0.8Pd0.2-497 (Pd on the outer layers),
respectively. It is clear that the electrode surface of SS/Sn0.8Pd0.2-497
composed of multiple nanoparticles, leading to higher roughness
compared to the surface of SS/Pd0.2Sn0.8-497. Fig. 2b-2e represent the
SEM micrographs of four electrodes prepared with different Q values,
exhibited distinct surface architectures among all four electrodes. Al-
though these four electrodes appeared to compose nanoparticles on the
surface, the degree of roughness was clearly different. Considering that
all samples have the same composition and deposition sequence, we
postulate that different Q applied resulted in different electrode surface
properties. SEM micrographs of Fig. 2b, f, g reveal the surface
morphologies of electrodes prepared from different Sn to Pd ratios. It is
noted that the roughness features of SS/Sn0.8Pd0.2-497 (Fig. 2b) and SS/
Sn0.2Pd0.8-497 (Fig. 2g) are much more pronounced than that of the SS/
Sn0.5Pd0.5-497 electrode (Fig. 2f). Fig. 2h shows the SEM micrograph of
SS/Sn0.2Pd0.8-497 after nitrate reduction experiment. The similar fea-
tures between Fig. 2g and h indicate the surface structure was retained,
demonstrating the stability of Sn-Pd bimetallic electrodes prepared in
this study.
otherwise identical conditions, indicating electrodes with Sn-rich outer
surface was beneficial to nitrate conversion.
For electrodes prepared with same Sn to Pd ratio (SS/Sn0.8Pd0.2) but
different Q, it appeared that nitrate conversion increased with in-
creasing Q, reached maximum value at Q of approximately 500 C then
slightly tailed off. The removal rate constant also revealed the same
trend with low k on SS/Sn0.8Pd0.2-166 (k =0.14 h−1) and maintaining
constant k value (k =0.24 h−1) on SS/Sn0.8Pd0.2-332 and SS/Sn0.8Pd0.2
497, except a slight drop of k on SS/Sn0.8Pd0.2-1490 (k =0.20 h−1).
-
For electrodes with the same Q of 497 C, but different Sn to Pd
ratios, namely, SS/Sn0.2Pd0.8-497, SS/Sn0.5Pd0.5-497, and SS/
Sn0.8Pd0.2-497, nitrate conversion decreased from 0.88 at Sn:Pd = 1:4
to 0.60 at Sn:Pd = 1:1 then increased to 0.65 at Sn:Pd = 4:1, while k
(h−1) decreased from 0.47 at Sn:Pd = 1:4 to 0.22 at Sn:Pd = 1:1 then
increased to 0.24 at Sn:Pd = 4:1. The SS/Sn0.2Pd0.8-497 electrode ex-
hibited the highest nitrate conversion among all bimetallic electrodes in
this work, indicating Sn:Pd metal ratio of 1:4 was sufficient for the
reduction of most nitrate ions in the solution. Further increase in Sn
loading might block Pd active sites and reduce the reaction rate and
selectivity [49,50,61].
Fig. 5c shows the variation of nitrite concentration versus reaction
time. In general, nitrite concentration increased with reaction time.
After 5 h, nitrite concentration and selectivity were low, indicating no
nitrite accumulation and fast reaction kinetics [62]. Fig. 5d gives the
concentration of ammonium as a function of reaction time. Results
again showed the importance of the deposition sequence of metals on
ammonium selectivity. Interestingly, ammonium selectivity was in-
hibited when Pd was deposited after Sn, i.e., Pd was on the outer layers
of the electrode structure. All electrodes prepared by depositing Sn first
The surface morphology of electrodes was also characterized by
XRD as shown in Fig. 3. Note that the major diffraction peaks of Pd
metal were located at 2θ = 40, 48, and 68 with (111), (200), and (220)
crystal facets, respectively (ICDD PDF Card No. 00-046-1043) [53–55].
The major diffraction peaks of Sn metal were located at 2θ = 30.7, 32,
and 45 with (200), (101), and (211) crystal facets, respectively (JCPDS
Card No. 04-0673) [55–57]. The surface crystal structure was assigned
by matching XRD pattern of each sample with known standard patterns
[31,58–60]. Interestingly, most samples exhibited X-ray diffraction
peaks at 2θ = 38.1, 44.4, and 64.6, excluding 2θ = 43.5, 50.8 from the
stainless steel (ICDD PDF Card No. 00-033-0397), indicating the for-
mation of Sn3Pd alloy (ICDD PDF Card No. 00-015-0575) with (214),
(131), and (420) crystal facets on the electrode surface, respectively. In
some cases, such as SS/Sn0.8Pd0.2-166, the crystal structure was pre-
sumed to be Sn3Pd alloy based on metal composition because the dif-
fraction peaks at 2θ = 38.1 and 64.6 were indistinguishable from the
+
then Pd (SS/Sn-Pd) exhibited low NH4 production, which might be
resulted from insufficient surface hydrogen on Pd sites and thus
weakening hydrogenation abilities due to the blockage of Pd active sites
by Sn metal. According to Eqs. (1c)–(1f), hydrogenation from Pd[H]
strongly participated in the nitrate reduction pathway toward N2 for-
mation. In the case of SS/Pd0.2Sn0.8-497, Pd[H] might not be enough in
density or availability for the overall reaction as the active sites were
covered or blocked by Sn. Thus, the reduced N-intermediates from
7

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 6. Electrochemical characterization for different electrode systems. (a) Polarization curve from linear sweep voltammetry; (b) Nyquist plot from electrochemical
impedance spectroscopy with equivalent circuit fitting in the inset. Experimental conditions: pH = 1.5; [NaNO3] = 0.008 M; temperature =25 °C; constant current
= -0.04 A; voltage = 5 V to 6 V.
nitrate were not able to pair with the two nearby N-species and form
which again demonstrated the importance of metal deposition se-
quence; Pd on the outer layer of electrode structure displayed higher
nitrogen production. N2 selectivity (SN2) was 0.84, 0.80, 0.88, and 0.84
on SS/Sn0.8Pd0.2-166, SS/Sn0.8Pd0.2-332, SS/Sn0.8Pd0.2-497, and SS/
Sn0.8Pd0.2-1490, respectively.
+
N2, which led to increase in the production of NH4 end products
[50,63].
Fig. 5e shows N2 concentration as a function of reaction time. SS/
Sn0.2Pd0.8-497 exhibited superior N2 selectivity over SS/Pd0.8Sn0.2-497,
8

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Table 3
Comparison of nitrogen selectivity by catalytic electrochemical nitrate reduction.
+
Cathode Material
Support
Cathode Source
Cell Type
Experimental Conditions
NH4 yield (%)
26.4
N2 yield (%)
30
Ref.
Cu
N/A
Commercial
Undivided
pH = 12
[36]
0.1 M NaNO3
0.01 M NaOH
0.5 M NaCl
I =0.6 mA
Al
N/A
N/A
N/A
N/A
N/A
Ni
Commercial
Commercial
Commercial
Commercial
Commercial
Electro-deposition
Divided
Divided
Divided
Divided
Divided
Undivided
−1.8 V (vs. Ag/AgCl)
−2.0 V (vs. Ag/AgCl)
−0.5 V (vs. Ag/AgCl)
−1.6 V (vs. Ag/AgCl)
−2.0 V (vs. Ag/AgCl)
pH = 7
47
64
75
2
38.1
16
[25]
[25]
[25]
[25]
[25]
[31]
Pb
Cu0.6Zn0.4
Sn0.85Cu0.15
Sn0.85Cu0.15
Pd0.4Cu0.6
2.08
7.6
40
49
34.4
41
50 ppm NaNO3
−0.3 V (vs. SCE)
0.1 M NaNO3
1 M NaOH
Pd0.62Cu0.38
Pd0.82Sn0.18
Stainless Steel
Electro-deposition
Electro-deposition
Divided
Divided
10
9
76
80
[32]
[65]
−0.93 V (vs. SCE)
pH = 5
Activated carbon fiber
0.002 M NaNO3
3.2−4.2 V (vs. SCE)
I = -0.04 A
Pd/Sn
Au
Adsorption
Divided
0.01 M NaNO3
0.1 M HClO4
6.8
10
14
21
88
81
[52]
[55]
[34]
Sn-Pd
Red mud
Stainless Steel
Impregnation
Electro-deposition
Undivided
Divided
0.0005 M NaNO3
H2/CO2 gas supply
pH = 1.5
Blended Sn0.8Pd0.2
0.008 M NaNO3
0.1 M HClO4
I = -0.04 A
Sn0.2Pd0.8
Stainless Steel
Electro-deposition
Divided
pH = 1.5
7.9
79
This work
0.008 M NaNO3
0.1 M HClO4
I = -0.04 A
In terms of Sn to Pd ratio, the N2 selectivity (SN2) was decreased
from 0.88 at Sn:Pd = 4:1 (SS/Sn0.8Pd0.2-497) to 0.83 at Sn:Pd = 1:1
(SS/Sn0.5Pd0.5-497) then increased to 0.89 at Sn:Pd = 1:4 (SS/
Sn0.2Pd0.8-497). It was noted that all SS/Sn-Pd electrodes, except SS/
Pd0.2Sn0.8-497, achieved high N2 selectivity (> 0.80), which supported
above argument that the availability and accessibility of Pd[H] during
nitrate reduction were critical to N2 formation process.
was applied to verify the performance of different electrodes in nitrate
reduction reaction (Fig. 6a). The dash line was the LSV curve of SS/
Sn0.2Pd0.8-497 without nitrate ion. Subtracting the current density in
the absence of nitrate from that in the presence of nitrate gave the net
nitrate reduction current [28,65]. Accordingly, the net current density
of nitrate reduction followed the order of SS/Sn0.2Pd0.8-497 > SS/
Pd0.2Sn0.8-497 > SS/Sn0.8Pd0.2-332 > SS/Sn0.8Pd0.2-497 > SS/
Note that the nitrate reduction toward N2 generation can be ex-
Sn0.5Pd0.5-497
SS/Sn0.8Pd0.2-1490 > SS/Sn0.8Pd0.2-166, suggesting
pressed as well in terms of N2 yield ( 2), which is the product of nitrate
that the SS/Sn0.2Pd0.8-497 electrode had superior electrochemical ac-
tivity to all other bimetallic electrodes prepared in this study.
A series of EIS experiments were also conducted on different Sn-Pd
electrodes in nitrate solution as to study the charge transfer kinetics.
The Nyquist plot was obtained as shown in Fig. 6b, and the corre-
sponding equivalent circuit that fitted the electrochemical system was
provided in the inset of Fig. 6b. The simple equivalent circuit consisted
of a solution resistance, Rs, a constant phase element, CPE, and a charge
transfer resistance, Rct. The impedance of CPE can be expressed as Eq. 7
[64].
N
conversion (XNO3 ) and nitrogen selectivity (SN2) using Eq. 6:
= XNO × SN
(6)
N
2
2
3
Table 2 showed the effects of electrode preparation procedure on
the selectivity of several major products, including nitrate (SNO2 ), am-
monium (SNH4+) and nitrogen molecule (SN2), and nitrogen yield ( 2).
In terms of metal deposition sequence, the difference in
betweenNSn
N
2
and Pd was only 8% (0.58 for SS/Sn0.8Pd0.2-497 versus 0.50 for SS/
Pd0.2Sn0.8-497). At Q of 166, i.e., the SS/Sn0.8Pd0.2-166 electrode ex-
hibited the lowest N2 yield (
= 0.38) among all electrodes studied.
N
2
1
ZCPE
=
With respect to metal ratio, nitrogen yield decreased from 0.79 at
Sn:Pd = 1:4 to 0.50 at Sn:Pd = 1:1 then increased to 0.58 at
Sn:Pd = 4:1. Results showed that the electrode prepared at Sn:Pd ratio
of 1:4 (SS/Sn0.2Pd0.8-497) exhibited the largest nitrate removal rate
constant (k =0.47 h−1), nitrate conversion (XNO = 0.88), N2 se-
(CPET)(i × w)n
(7)
where CPET is the pre-factor of CPE, i is the imaginary unit, w is the
angular frequency, and n is the exponent. The exponent n is a constant
value between 0 and 1, and it determines the degree of deviation from
an ideal semicircle in Nyquist plot. In the case of n = 1, CPE is an ideal
capacitor, and CPE describes a pure resistor in the case of n = 0 [64].
The EIS parameters including Rs, CPET, and n are listed in Table S1, and
the values of Rct are provided in Table 2. It could be noticed that the n
value of all electrochemical systems were around 0.64, which explained
the depressed semicircle observed in Fig. 6b and suggested the pseudo-
capacitive behavior of CPE. Rct was calculated from the diameter of
semicircle in Nyquist plot. The smaller Rct value indicated faster ki-
netics in the electron transfer process at the electrode and electrolyte
lectivity (SN
=
0.89), and N2 yield (
=
0.79) than 3all other SS/Sn-Pd
2
N
2
electrodes in this work (Table 2). As all electrodes with Pd deposition
on the outside surface exhibiting similar N2 selectivity due to sufficient
Pd[H] for the reduction reaction. The superior N2 yield on SS/
Sn0.2Pd0.8-497 could be attributed to its outstanding nitrate conversion,
which was resulted from optimal metal ratio as discussed above.
CV and LSV characterized the electrochemical behavior of elec-
trodes prepared in this study as shown in Fig. S1 and Fig. 6a, respec-
tively. Results of CV analysis showed that different preparation para-
meters influenced the cathodic behavior of electrode. Thus, LSV test
interface. In the present study, Rct was in the order of SS/Sn0.2Pd0.8
-
9

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
section, different crystal facet distribution was known to impact surface
active sites responsible for the catalytic reactivity and product se-
lectivity. Therefore, the reactivity and selectivity of electrode SS/
Sn0.2Pd0.8-497 might also influenced by the modified surface structure
of alloy catalyst.
3.3. Effects of surface structure
Fig. 7 shows the relationship between electrode preparation pro-
cedures, i.e., metal deposition sequence (Fig. 7a), charge applied
(Fig. 7b), and Pd content (Fig. 7c), on N2 yield (open read circles) with
respect to crystal facets, namely, Sn3Pd(214), Sn3Pd(131), and Sn3Pd
(420), respectively. Facet peak intensity mostly reflected nitrogen yield.
Generally, nitrogen yield correlated well to the XRD peak intensity of
crystal facets (Fig. 7a, 7b, and 7c). To elucidate the influence of in-
dividual crystal facet on nitrate reduction, the N2 yield was plotted as a
function of peak intensity of the three facets, i.e., Sn3Pd(214), Sn3Pd
(131) and Sn3Pd(420), respectively (Fig. 8). The correlation between N2
yield and peak intensity was evaluated by the Pearson correlation
coefficient according to Eq. 7:
(xi x)(y
y )
i
r =
(xi x)2
(y
y )2
(8)
i
Where r is the linear correlation coefficient, x is the XRD peak intensity
of electrode i, y is the N2 yield of electrode i, and
x
and
y
are the
average values of x and y, respectively. The linear correlation coeffi-
cient, r, measures the strength of a linear relationship between two
variables. The value of r is always between -1 and +1, which means
total negative linear correlation and total positive linear correlation,
respectively. Table 4 shows that the largest r value was 0.77 for the
Sn3Pd(420) facet (peak at 2θ = 64.5) (Fig. 8), which was 10% and 19%
higher than Sn3Pd(131) facet and Sn3Pd(214) facet, respectively, in-
dicating Sn3Pd(420) had stronger impact than the other two facets on
N2 yield.
To evaluate the effects of crystal facets on nitrogen yield, the Sn3Pd
crystal structure, i.e. Cmca space group symmetry, identified from XRD
results was imported to MERCURY software for 3D structure visuali-
zation. Fig. 9 shows the coordination characteristics of Sn3Pd(214),
Sn3Pd(131), and Sn3Pd(420) generated by MERCURY simulation. It can
be seen that the Sn3Pd(214) and Sn3Pd(131) (Fig. 9a, d and b, e, re-
spectively) exhibited undesired diamond-shaped surface atom ar-
rangements [44,66]. On the other hand, Sn3Pd(420) facet (Fig. 9c, f)
hadmainly periodic and rectangular active sites.This particular four-
atom surface arrangement is known to offer optimal electronic and/or
coordination conditions for outstanding electro-catalytic activity as
exemplified by oxygen reduction reaction on Au(100) [67,68], CO2
reduction to ethylene on Cu(100) [69], the oxidation of ammonia on Pt
(100) [70,9,71], and the reduction of nitrate on Pt(100) [43,45].
Based on the above observation, mechanism for nitrate reduction on
Sn3Pd(420) facet was attempted (Fig. 10a) as follows. The main feature
of the proposed reaction mechanism (Fig.10a) are: (1) adsorption of
nitrate ion onto an empty Sn site via oxygen atom; (2) adsorption of
hydrogen on the empty Pd site; (3) bond breaking between N and OH;
(4) bond breaking between N and O; (5) bond formation between ad-
jacent two N atoms; (6) final bond breaking between N and O, leading
to the formation of N2. As shown in Figs. 10b and 10c, the adsorption of
hydrogen on Pd and bond breaking between N and OH were difficult on
Sn3Pd(214) and Sn3Pd(131) facets likely due to steric restrictions. The
result revealed that surface structure design of bimetallic alloy could be
applied to modify the hydrogenation ability of catalyst, enabling a
unique control over nitrate reduction mechanism.
Fig. 7. The relationship among XRD peak intensity, nitrogen yield (open cir-
cles), and (a) electrode deposition sequence, (b) total charge for electrode
preparation, and (c) fraction of Pd in Sn-Pd electrodes.
497 < SS/Pd0.2Sn0.8-497 < SS/Sn0.8Pd0.2-497 < SS/Sn0.8Pd0.2
-
332 < SS/Sn0.5Pd0.5-497 < SS/Sn0.8Pd0.2-1490 < SS/Sn0.8Pd0.2-166.
The EIS result was in agreement with that of LSV analysis and nitrate
conversion data, which confirmed tat SS/Sn0.2Pd0.8-497 was superior
with respect to electrochemical activity toward nitrate reduction reac-
tion. Additionally, the N2 yield on SS/Sn0.2Pd0.8-497 was also relatively
higher than most reported catalytic electrodes in the literature espe-
cially in electrochemical nitrate reduction, while NH4+ yield remained
relatively low at 7.9% (Table 3) [25,31,32,65].
It was further noted that all synthesized electrodes exhibited the
same Sn3Pd XRD pattern (Fig. 3). The major difference among all XRD
spectra was the peak intensity. The variance in XRD peak intensity
suggested that Sn3Pd alloy on the electrode surface was composed of
different crystal facet distribution. As described in the Introduction
In summary, results demonstrated that manipulating electrode
preparation procedures could impact the nitrate conversion and N2
selectivity due to the availability of surface hydrogen on Pd for re-
duction reactions. Additionally, modifying electrode preparation
10

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 8. Relationship between nitrogen yield and crystal facets. Solid line was included for visual clarity.
procedures also changed crystal facet distribution on the electrode
surface, which was found to influence nitrate reduction products.
Different surface atom arrangements of bimetallic alloy on electrodes
and associated several bond-breaking and bond-forming events con-
tributed to nitrate reactivity and selectivity. The protocol highlighted
herein could be extended to other electro-catalytic reactions relevant to
the nitrogen cycle specifically and be employed to synthesize different
bimetallic electrodes with optimal surface-atomic structure and hy-
drogenation ability, for promoting applications in the fields of waste-
water treatment, electrochemical sensors, electrochemical synthesis,
and energy conversion.
Table 4
Values of correlation coefficient for three
crystal facets.
Crystal Facet
r
Sn3Pd(214)
Sn3Pd(131)
Sn3Pd(420)
0.65
0.70
0.77
r: linear correlation coefficient.
11

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 9. MERCURY simulation of Sn3Pd crystal phase with (a) Sn3Pd(214), (b) Sn3Pd(131), and (c) Sn3Pd(420). (d) Perpendicular view to Sn3Pd(214) facet; (e)
Perpendicular view to Sn3Pd(131) facet; (f) Perpendicular view to Sn3Pd(420) facet. Diamond-shape and rectangular atom structures are highlighted by red lines. It
can be seen that undesired diamond-shaped surface atom arrangement appears on the Sn3Pd(214) and Sn3Pd(131) (a, d and b, e). On the other hand, the plane of
Sn3Pd(420) contains mainly periodic and rectangular active sites, which is favorable for the bond-breaking reaction (For interpretation of the references to colour in
this figure legend, the reader is referred to the web version of this article).
4. Conclusion
confirmed tat SS/Sn0.2Pd0.8-497. Results observed in this study, i.e.,
direct nitrate reduction rate, LSV and EIS measurements, clearly in-
dicated that SS/Sn0.2Pd0.8-497 was superior with respect to electro-
chemical activity toward nitrate reduction reaction. Manipulating
electrode synthesis procedure enables control the crystal facet dis-
tribution of bimetallic electrodes, leading to enhanced electrochemical
nitrate reduction. The findings highlighted herein will allow for the
design and synthesis of desirable surface-atomic structure and provide
strategy toward fabricating efficient and cost-effective electrodes for
electrochemical denitrification in water treatment process.
Bimetallic electro-catalyst preparation procedure affected the
crystal structure and hydrogenation ability of the catalysts, which in
turn, governed nitrate reduction reactivity and N2 selectivity. The se-
quence of deposition of the two metals, namely, Pd and Sn, (i.e., de-
position sequence), total electrical charge applied for electrode pre-
paration, and molar ratio of the bimetal together could influence the
availability of surface hydrogen on catalyst as well as the intensity of
the preferred crystal facet orientation, thereby affecting nitrate reduc-
tion performance over the SS/Sn-Pd electrode. Results from simple
analysis of reaction products, SEM, TEM, and XRD revealed that the
crystal facet distribution played a critical role in the nitrate reduction
reaction. Specifically, among the three major facets, Sn3Pd(210), Sn3Pd
(131), and Sn3Pd(420), Sn3Pd(420) exhibited the highest N2 yield,
which was likely due to the favorable arrangement of active sites for
nitrate adsorption and subsequent bond breaking and formation reac-
tions. Electrode prepared by depositing tin then palladium at tin to
palladium atomic ratio of 1:4 and with 497 C applied charge, namely,
the SS/Sn0.2Pd0.8-497 electrode exhibited the best performance in ni-
trate reactivity and dinitrogen selectivity/yield, with 88%, 89% and
79% of total nitrate removal, dinitrogen selectivity and yield, respec-
tively. Results of LSV analysis gave the net nitrate reduction current
density in the order: SS/Sn0.2Pd0.8-497 > SS/Pd0.2Sn0.8-497 > SS/
CRediT authorship contribution statement
Jenn Fang Su: Writing - original draft, Formal analysis, Resources.
Wei-Fan Kuan: Resources. Ching-Lung Chen: Resources. Chin-Pao
Huang: Conceptualization, Writing - review & editing, Project admin-
istration, Supervision, Funding acquisition.
Declaration of Competing Interest
The authors reported no declarations of interest.
Acknowledgement
Sn0.8Pd0.2-332 > SS/Sn0.8Pd0.2-497 > SS/Sn0.5Pd0.5-497
Sn0.8Pd0.2-1490 > SS/Sn0.8Pd0.2-166. EIS analysis showed that Rct
value followed the order of SS/Sn0.2Pd0.8-497 < SS/Pd0.2Sn0.8
SS/
The materials presented in this paper were based upon work par-
tially supported by National Science Foundation, Grant No. 0965984
under the Environmental Engineering Program, Division of Chemical,
Bioengineering, Environmental, and Transport Systems. J. F. Su also
acknowledges the support by the Ministry of Science and Technology,
Taiwan (109-2222-E-032 -001) and 109-2221-E-032-040-MY2.
-
497 < SS/Sn0.8Pd0.2-497 < SS/Sn0.8Pd0.2-332 < SS/Sn0.5Pd0.5
-
497 < SS/Sn0.8Pd0.2-1490 < SS/Sn0.8Pd0.2-166. The EIS result was in
agreement with that of LSV analysis and nitrate conversion data, which
12

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
Fig. 10. Illustration of predicted nitrate reduction mechanism on SS/Sn-Pd electrodes with (a) Sn3Pd(420), (b) Sn3Pd(214), and (c) Sn3Pd(131) planes. The main
feature of the proposed reaction mechanism in (a) are: (1) adsorption of nitrate onto an empty Sn site via their oxygen atom; (2) the adsorption of hydrogen on the
empty Pd site; (3) the bond breaking reaction between N and OH; (4) the bond breaking reaction between N and O; (5) the bond formation reaction between adjacent
two N atoms; (6) the final bond breaking reaction between N and O, leading to the formation of N2. In (b) and (c), the bond breaking reaction between N and OH is
more difficult in Sn3Pd(214) and Sn3Pd(131) likely due to the steric restrictions.
Appendix A. Supplementary data
1016/j.memsci.2009.02.020.
[4] K. Anoop, V. T, J. Environ. Eng. New York (New York) 123 (1997) 371–380,
https://doi.org/10.1061/(ASCE)0733-9372(1997)123:4(371).
[5] G.K. Luk, W.C. Au-Yeung, Adv. Environ. Res. 6 (2002) 441–453, https://doi.org/10.
1016/S1093-0191(01)00072-7.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2020.117809.
[6] C. Della Rocca, V. Belgiorno, S. Meriç, Desalination 204 (2007) 46–62, https://doi.
org/10.1016/j.desal.2006.04.023.
References
[7] X. Fan, X. Guan, J. Ma, H. Ai, J. Environ. Sci. China (China) 21 (2009) 1028–1035,
https://doi.org/10.1016/S1001-0742(08)62378-5.
[8] C.P. Huang, H.W. Wang, P.C. Chiu, Water Res. 32 (1998) 2257–2264, https://doi.
org/10.1016/S0043-1354(97)00464-8.
[1] B.-U. Bae, Y.-H. Jung, W.-W. Han, H.-S. Shin, Water Res. 36 (2002) 3330–3340,
https://doi.org/10.1016/S0043-1354(02)00012-X.
[9] J.M. Rodríguez-Maroto, F. García-Herruzo, A. García-Rubio, C. Gómez-Lahoz,
C. Vereda-Alonso, Chemosphere 74 (2009) 804–809, https://doi.org/10.1016/j.
chemosphere.2008.10.020.
[2] J.J. Schoeman, A. Steyn, Desalination 155 (2003) 15–26, https://doi.org/10.1016/
S0011-9164(03)00235-2.
[3] L.J. Banasiak, A.I. Schäfer, J. Memb. Sci. 334 (2009) 101–109, https://doi.org/10.
[10] K.N. Heck, S. Garcia-Segura, P. Westerhoff, M.S. Wong, Acc. Chem. Res. 52 (2019)
13

J.F. Su, et al.
AppliedCatalysisA,General606(2020)117809
906–915, https://doi.org/10.1021/acs.accounts.8b00642.
[42] D.F. van der Vliet, M.T.M. Koper, Surf. Sci. 604 (2010) 1912–1918, https://doi.org/
[11] D. He, Y. Li, H. Ooka, Y.K. Go, F. Jin, S.H. Kim, R. Nakamura, J. Am. Chem. Soc. 140
(2018) 2012–2015, https://doi.org/10.1021/jacs.7b12774.
10.1016/j.susc.2010.07.027.
[43] I. Katsounaros, M.C. Figueiredo, X. Chen, F. Calle-Vallejo, M.T.M. Koper,
Electrochim. Acta 271 (2018) 77–83, https://doi.org/10.1016/j.electacta.2018.03.
126.
[12] B. Shen, Z. Liu, H. Xu, F. Wang, Fuel 235 (2019) 878–885, https://doi.org/10.
1016/J.FUEL.2018.08.082.
[13] Y.A. Shao, Y.T. Chen, P.Y. Chen, Electrochim. Acta 313 (2019) 488–496, https://
doi.org/10.1016/j.electacta.2019.05.057.
[44] H. Li, F. Calle-Vallejo, M.J. Kolb, Y. Kwon, Y. Li, M.T.M. Koper, J. Am. Chem. Soc.
135 (2013) 14329–14338, https://doi.org/10.1021/ja406655q.
[45] M. Kato, M. Okui, S. Taguchi, I. Yagi, J. Electroanal. Chem. Lausanne (Lausanne)
800 (2017) 46–53, https://doi.org/10.1016/j.jelechem.2017.01.020.
[46] F. Epron, F. Gauthard, C. Pinéda, J. Barbier, J. Catal. 198 (2001) 309–318, https://
doi.org/10.1006/jcat.2000.3138.
[14] S. Garcia-Segura, M. Lanzarini-Lopes, K. Hristovski, P. Westerhoff, Appl. Catal. B
Environ. 236 (2018) 546–568, https://doi.org/10.1016/j.apcatb.2018.05.041.
[15] J. Gao, B. Jiang, C. Ni, Y. Qi, Y. Zhang, N. Oturan, M.A. Oturan, Appl. Catal. B
Environ. 254 (2019) 391–402, https://doi.org/10.1016/j.apcatb.2019.05.016.
[16] N. Aouina, H. Cachet, C. Debiemme-chouvy, T.T.M. Tran, Electrochim. Acta 55
(2010) 7341–7345, https://doi.org/10.1016/j.electacta.2010.07.032.
[17] O.A. Petrii, T.Y. Safonova, J. Electroanal. Chem. Lausanne (Lausanne) 331 (1992)
897–912, https://doi.org/10.1016/0022-0728(92)85013-S.
[47] F. Epron, F. Gauthard, J. Barbier, Appl. Catal. A Gen. 237 (2002) 253–261, https://
doi.org/10.1016/S0926-860X(02)00331-9.
[48] R. Zhang, D. Shuai, K.A. Guy, J.R. Shapley, T.J. Strathmann, C.J. Werth,
ChemCatChem. 5 (2013) 313–321, https://doi.org/10.1002/cctc.201200457.
[49] J. Martínez, A. Ortiz, I. Ortiz, Appl. Catal. B Environ. 207 (2017) 42–59, https://doi.
org/10.1016/j.apcatb.2017.02.016.
[18] G.E. Dima, A.C.A. de Vooys, M.T.M. Koper, J. Electroanal. Chem. Lausanne
(Lausanne) 554–555 (2003) 15–23, https://doi.org/10.1016/S0022-0728(02)
01443-2.
[50] U. Prüsse, K.-D. Vorlop, J. Mol. Catal. A Chem. 173 (2001) 313–328, https://doi.
org/10.1016/S1381-1169(01)00156-X.
[19] K. Bouzek, M. Paidar, A. Sadílková, H. Bergmann, J. Appl. Electrochem. 31 (2001)
1185–1193, https://doi.org/10.1023/A:1012755222981.
[51] K. Shimazu, R. Goto, S. Piao, R. Kayama, K. Nakata, Y. Yoshinaga, J. Electroanal.
Chem. Lausanne (Lausanne) 601 (2007) 161–168, https://doi.org/10.1016/j.
jelechem.2006.11.005.
[20] S. Cattarin, J. Appl. Electrochem. 22 (1992) 1077–1081, https://doi.org/10.1007/
BF01029588.
[21] P.M. Tucker, M.J. Waite, B.E. Hayden, J. Appl. Electrochem. 34 (2004) 781–796,
https://doi.org/10.1023/B:JACH.0000035607.19248.b6.
[52] M. Motahar Hossain, K. Nakata, T. Kawaguchi, K. Shimazu, J. Electroanal. Chem.
Lausanne (Lausanne) 707 (2013) 59–65, https://doi.org/10.1016/j.jelechem.2013.
08.015.
[22] I. Katsounaros, D. Ipsakis, C. Polatides, G. Kyriacou, Electrochim. Acta 52 (2006)
1329–1338, https://doi.org/10.1016/j.electacta.2006.07.034.
[23] J.D. Genders, D. Hartsough, D.T. Hobbs, J. Appl. Electrochem. 26 (1996) 1–9,
https://doi.org/10.1007/BF00248182.
[53] S. Sadjadi, M. Atai, Appl. Clay Sci. 153 (2018) 78–89, https://doi.org/10.1016/j.
clay.2017.12.013.
[54] S. Navaladian, B. Viswanathan, T.K. Varadarajan, R.P. Viswanath, Nanoscale Res.
Lett. 4 (2008) 181, https://doi.org/10.1007/s11671-008-9223-4.
[55] S. Hamid, S. Bae, W. Lee, Chem. Eng. J. 348 (2018) 877–887, https://doi.org/10.
1016/j.cej.2018.05.016.
[24] T.Y. Safonova, O.A. Petrii, J. Electroanal. Chem. Lausanne (Lausanne) 448 (1998)
211–216, https://doi.org/10.1016/S0022-0728(97)00110-1.
[25] C. Polatides, G. Kyriacou, J. Appl. Electrochem. 35 (2005) 421–427, https://doi.
org/10.1007/s10800-004-8349-z.
[56] A. Banotra, N. Padha, J. Korean Inst. Electr. Electron. Mater. Eng. 48 (2019)
304–311, https://doi.org/10.1007/s11664-018-6710-y.
[26] Z. Mácová, K. Bouzek, J. Appl. Electrochem. 35 (2005) 1203–1211, https://doi.org/
10.1007/s10800-005-9028-4.
[57] H. Chen, Q. Ye, X. He, J. Ding, Y. Zhang, J. Han, J. Liu, C. Liao, J. Mei, W. Lau,
Green Chem. 16 (2014) 3841–3845, https://doi.org/10.1039/C4GC00142G.
[58] L. Zhang, A.A.S. Gonçalves, M. Jaroniec, RSC Adv. 10 (2020) 5585–5589, https://
doi.org/10.1039/D0RA00769B.
[27] Y. Zhang, Y. Zhao, Z. Chen, L. Wang, L. Zhou, P. Wu, F. Wang, P. Ou, J.
Electrochem. Soc. 165 (2018) E420–E428, https://doi.org/10.1149/2.0081810jes.
[28] Y. Zhang, Y. Zhao, Z. Chen, L. Wang, P. Wu, F. Wang, Electrochim. Acta 291 (2018)
151–160, https://doi.org/10.1016/j.electacta.2018.08.154.
[59] Y. Yoshinaga, T. Akita, I. Mikami, T. Okuhara, J. Catal. 207 (2002) 37–45, https://
doi.org/10.1006/jcat.2002.3529.
[29] L. Lemaignen, C. Tong, V. Begon, R. Burch, D. Chadwick, Catal. Today 75 (2002)
43–48, https://doi.org/10.1016/S0920-5861(02)00042-1.
[60] Y.J. Shih, Z.L. Wu, Y.H. Huang, C.P. Huang, Chem. Eng. J. 383 (2020) 123157,
https://doi.org/10.1016/j.cej.2019.123157.
[30] A.C.A. de Vooys, R.A. van Santen, J.A.R. van Veen, J. Mol. Catal. A Chem. 154
(2000) 203–215, https://doi.org/10.1016/S1381-1169(99)00375-1.
[31] D. Reyter, D. Bélanger, L. Roué, J. Phys. Chem. C. 113 (2009) 290–297, https://doi.
org/10.1021/jp805484t.
[61] H. Sitepu, B.H. O’Connor, D. Li, J. Appl. Crystallogr. 38 (2005) 158–167, https://
doi.org/10.1107/S0021889804031231.
[62] S. Hamid, M.A. Kumar, W. Lee, Appl. Catal. B Environ. 187 (2016) 37–46, https://
doi.org/10.1016/j.apcatb.2016.01.035.
[32] J.F. Su, I. Ruzybayev, I. Shah, C.P. Huang, Appl. Catal. B Environ. 180 (2016)
199–209, https://doi.org/10.1016/j.apcatb.2015.06.028.
[63] J. Sá, D. Gasparovicova, K. Hayek, E. Halwax, J.A. Anderson, H. Vinek, Catal.
Letters 105 (2005) 209–217, https://doi.org/10.1007/s10562-005-8692-7.
[64] M.R. Shoar Abouzari, F. Berkemeier, G. Schmitz, D. Wilmer, Solid State Ion. 180
(2009) 922–927, https://doi.org/10.1016/j.ssi.2009.04.002.
[65] Y. Wang, J. Qu, R. Wu, P. Lei, Water Res. 40 (2006) 1224–1232, https://doi.org/10.
1016/j.watres.2006.01.017.
[33] K. Shimazu, R. Goto, K. Tada, Chem. Lett. 31 (2002) 204–205, https://doi.org/10.
1246/cl.2002.204.
[34] J.F. Su, W.F. Kuan, H. Liu, C.P. Huang, Appl. Catal. B Environ. 257 (2019) 117909,
https://doi.org/10.1016/j.apcatb.2019.117909.
[35] A.S. Bandarenka, M.T.M. Koper, J. Catal. 308 (2013) 11–24, https://doi.org/10.
1016/j.jcat.2013.05.006.
[66] H. Li, Y. Li, M.T.M. Koper, F. Calle-Vallejo, J. Am. Chem. Soc. 136 (2014)
15694–15701, https://doi.org/10.1021/ja508649p.
[36] D. Reyter, D. Bélanger, L. Roué, Water Res. 44 (2010) 1918–1926, https://doi.org/
10.1016/j.watres.2009.11.037.
[67] R.R. Adić, N.M. Marković, V.B. Vešović, J. Electroanal. Chem. Interfacial
Electrochem. 165 (1984) 105–120, https://doi.org/10.1016/S0022-0728(84)
80090-X.
[37] D. Reyter, D. Bélanger, L. Roué, J. Hazard. Mater. 192 (2011) 507–513, https://doi.
org/10.1016/j.jhazmat.2011.05.054.
[38] L. Li, Y. Yun, Y. Zhang, Y. Huang, Z. Xu, J. Alloys. Compd. 766 (2018) 157–160,
https://doi.org/10.1016/j.jallcom.2018.07.004.
[68] R.R. Adžić, S. Strbac, N. Anastasijević, Mater. Chem. Phys. 22 (1989) 349–375,
https://doi.org/10.1016/0254-0584(89)90005-9.
[39] T. Chen, H. Li, H. Ma, M.T.M. Koper, Langmuir 31 (2015) 3277–3281, https://doi.
org/10.1021/acs.langmuir.5b00283.
[69] Y. Hori, I. Takahashi, O. Koga, N. Hoshi, J. Phys. Chem. B 106 (2002) 15–17,
https://doi.org/10.1021/jp013478d.
́
[40] M. Duca, M.O. Cucarella, P. Rodriguez, M.T.M. Koper, J. Am. Chem. Soc. 132
(2010) 18042–18044, https://doi.org/10.1021/ja1092503.
[70] F.J. Vidal-Iglesias, N. Garcıa-Aráez, V. Montiel, J.M. Feliu, A. Aldaz, Electrochem.
commun. 5 (2003) 22–26, https://doi.org/10.1016/S1388-2481(02)00521-0.
[71] F.J. Vidal-Iglesias, J. Solla-Gullón, V. Montiel, J.M. Feliu, A. Aldaz, J. Phys. Chem. B
109 (2005) 12914–12919, https://doi.org/10.1021/jp051269d.
[41] M. Duca, M.C. Figueiredo, V. Climent, P. Rodriguez, J.M. Feliu, M.T.M. Koper, J.
Am. Chem. Soc. 133 (2011) 10928–10939, https://doi.org/10.1021/ja203234v.
14