Comprehensively Probing the Contribution of Site Activity and Population of Active Sites toward Heterogeneous Electrocatalysis

Article abstract of DOI:10.1002/cctc.201902316

Due to the enormous demand for clean energy generation and for efficient energy storage, developing green and storable energy has become an urgent task. This subject, thus, stimulates studies on the improvement of new electrocatalysts. Specifically, through improving the population of active sites and/or the activity of individual sites, the performance of the catalysts can be significantly improved. However, clarifying the contribution of these two factors is still ambiguous and must be done with great care. To allow researchers in related fields to present their findings objectively, having a clear insight into the electrocatalytic activity is indispensable. In this study, a series of cobalt-based pre-electrocatalysts are taken as an example to examine their corresponding oxygen evolution reaction (OER) behaviors. Wherein, the three important metrics – electrochemical active surface area (ECSA)/roughness factor (RF), Tafel slope, and turnover frequency (TOF) – are used to profile the population of active sites of electrocatalysts and the activity of related individual sites. The results indicate the correction of OER behavior by ECSA/RF manifests the more accurate identification of OER activity. Meanwhile, the timing to obtain RF/ECSA significantly influence such an evaluation. Altogether, a fair benchmarking is proposed for the comparison between the activities of electrocatalysts.

Full text of DOI:10.1002/cctc.201902316

Accepted Article
Title: Comprehensively probing the contribution of site activity and
population of active site toward heterogeneous electrocatalysis
Authors: Hsiao-Chien Chen, Tai-Lung Chen, Sheng-Chih Lin, Chia-
Shuo Hsu, Ting‐Shan Chan, Mei-Yi Liao, and Hao Ming Chen
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201902316
Link to VoR: http://dx.doi.org/10.1002/cctc.201902316
A Journal of
www.chemcatchem.org

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
Comprehensively probing the contribution of site activity and population of
active site toward heterogeneous electrocatalysis
Hsiao-Chien Chen,[a, b] Tai-Lung Chen,[a] Sheng-Chih Lin,[a] Chia-Shuo Hsu,[a] Ting-Shan Chan,*[c] Mei-
Yi Liao,*[d] and Hao Ming Chen*[a,c]
[a]
Dr. H. C. Chen, T. L. Chen, S. C. Lin, C. S. Hsu, Prof. Hao Ming Chen
Department of Chemistry
National Taiwan University
Taipei 10617 (Taiwan)
E-mail: haomingchen@ntu.edu.tw
Dr. H. C. Chen
Department of Photonics
National Cheng Kung University
Tainan 70101 (Taiwan)
Dr. Ting-Shan Chan
National Synchrotron Radiation Center
Hsinchu 30076 (Taiwan)
E-mail: chan.ts@nsrrc.org.tw
Prof. Mei-Yi Liao
[b]
[c]
[d]
Department of Applied Chemistry
National Pingtung University
Pingtung 90003 (Taiwan)
E-mail: myliao@mail.nptu.edu.tw
Supporting information for this article is given via a link at the end of the document.
highest energy density for energy storage.[5] However,
bottlenecks of the overall water splitting are the sluggish half
reaction – the oxygen evolution reaction (OER), which consists of
multi-step electron transfer paths and the high activation energy
barrier for the O–O bond formation. Such sluggish kinetics,
therefore, hinder the widespread implementation of water
electrolysis.[6] Even though IrO2 and RuO2 exhibit the excellent
activity, their intrinsic poor stability and high prices prevent the
rutile-type oxides from large-scale application.[7,8] Consequently,
close attention has been devoted to investigating earth-abundant
transition metal-based electrocatalysts.[3,9.10] In connection with
advanced projects for significantly increasing the efficiency of
water electrolysis, two noticeable research domains are adopted.
The first one is to create high surface areas of electrocatalysts for
improving the number of exposed active sites, while the second
one is to design a new electrocatalyst for enhancing intrinsic
activity (Table 1).[11-30] Loading or in-situ generation of
Abstract: Due to the enormous demand for clean energy generation
and for efficient energy storage, developing green and storable
energy has become an urgent task. This subject, thus, stimulates
studies on the improvement of new electrocatalysts. Specifically,
through improving the population of active site and/or the activity of
individual sites, the performance of the catalysts can be significantly
improved. However, clarifying the contribution of these two factors is
still ambiguous and must be done with great care. To allow
researchers in related fields to present their findings objectively,
having a clear insight into the electrocatalytic activity is indispensable.
In this study, a series of cobalt-based pre-electrocatalysts are taken
as an example to examine their corresponding oxygen evolution
reaction (OER) behaviors. Wherein, the three important metrics ‒
electrochemical active surface area (ECSA)/roughness factor (RF),
Tafel slope, and turnover frequency (TOF) ‒ are used to profile the
population of active sites of electrocatalysts and the activity of related
individual sites. The results indicate the correction of OER behavior
by ECSA/RF manifests the more accurate identification of OER
activity. Meanwhile, the timing to obtain RF/ECSA significantly
influence such an evaluation. Altogether, a fair benchmarking is
proposed for the comparison between the activities of electrocatalysts.
electrocatalysts
on
two-dimensional/three-dimensional
frameworks, such as graphene derivatives or metal-organic
frameworks,[18,31] bears large specific area for improving the
exposure of active sites. Controlling nanostructured shape, array
and size of electrocatalysts is another strategy to increase the
population of active sites.[23,32] On the other hand, doping
elements, tuning ratio of transition metal complexes, replacing
substrate or anchoring atoms to modulate local coordination
environments have been a frequent approach for promoting the
activity of electrocatalysts.[11-29,33]
Regrettably, the assessment of contribution of the amount of
reacting catalyst or catalytic ability to OER behavior is ambiguous.
For example, in most literature, the evaluation of OER activity has
been finalized immediately after obtaining overpotential at 10
mA/cm-2 (η10). However, the better OER behavior surely exhibits
the lower overpotential but does not necessarily represent the
higher OER activity, because OER behavior is also dominated by
the population of exposed active sites. That is to say, the
evaluation of OER activity without consideration of involved
population of active species is biased. Although researchers in
Introduction
In response to the growing energy demand but the finite supply,
i.e., fossil fuels, the energy-related issues have been focused on
exploring renewable and sustainable sources.[1-4] Among the
myriad alternatives, research in the field of electrochemical
heterogeneous catalysis is in majority with more than ten
thousand published articles per year.
For example, electrocatalytic water splitting has been
received great attention because the resulting hydrogen can
produce near-zero emissions of greenhouse gas and owns the
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
Table 1. Relevant important parameters for OER electrocatalysts in 1 M KOH.
Electrocatalyst
MnGa4
Substrate
NF
Tafel slope (mV/dec)
η10 (mV)
293
TOF (s-1)
Ref.
11
98
NiCoP
NF
70.5
83.6
59.5
58
0.00692@305 mV
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
NP-NiFe2O4/SWNTs
Fe5Co4Ni20Se36Bx
Co3Ni1-LDH
W doped Ni(OH)2
NiFe0.1Cr0.1
S|NiNx-C
GC
247
279.8
350
237
200
280
283
228
237
289
230
288
106
183
255
330
240.4
340
GC
PCPs
GC
33
0.74@250 mV
0.04@250 mV
10.9@300 mV
Cu NA
EG
28
45
Ni–Fe PBA
Ni3S2/MnS
Au/NiFe LDH
CoP/CN@MoS2
Cu-Co3O4 NAs
S:Co2P
GC
54
NF
46
Ti mesh
RDE
NF
36
0.11@280 mV
0.127@350 mV
69
86
NF
71
CoFeP
CC
53.1
36
LaNiFe
GC
CoSeP
CC
87
CoSx/NGF
Co1Ni2Fe1
GC
62
0.100@300 mV
0.081@350mV
GC
38.8
56
Co LDH
GC
NF: nickel foam; CC: carbon cloth; PCPs: PPy-Coated Cotton Pads; Cu NA: Cu nanoarray; EG: exfoliated graphene
surface areas, such as physical surface areas by Brunauer-
Emmett-Teller (BET) theory or electrochemically active surface
area (ECSA) by cyclic voltammograms.[34] Enjoying the high
popularity, ECSA can be derived by extracting the double-layer
capacitance (Cdl) through the measurement of cyclic
voltammetry (CV) or electrochemical impedance spectroscopy
(EIS) in non-Faradaic potential region before applying the
following equation:
related fields attempt to clarify this ambiguousness and to
propose a benchmarking protocol to define the population of
active sites and the activity of individual sites of electrocatalysts,
a criterion to implement a fair comparison of electrocatalytic
activity has not yet been proposed.[34-36] Herein, a series of cobalt
derivate-based electrocatalysts for OER is taken as an example
to probe the correlation within electrocatalytic behavior,
electrocatalytic activity and population of active site. Firstly, Co-
bearing catalysts electrodeposited on various supports are
performed to clarify the influence of the population of active sites
in OER behavior. Further, the comparison between the activities
of Co-based electrocatalysts according to the electrochemical
active surface area (ECSA)/roughness factor (RF) corrected
OER behavior are performed. Additionally, the OER activity is
also evaluated by Tafel slope and turnover frequency (TOF).
ECSA = Cdl/Cs
(1)
where Cs is the specific capacitance of electrocatalysts.
However, some conflicting opinions emerge on the issue about
the employment of EIS to determine Cdl. McCrory et al.
suggested this measure was acceptable because Cdl only
decreased 15 % in comparison of outcome from CV.[34] On the
other hand, some investigations revealed that the employment
of EIS should take account of the potential-dependent
conductivity, in which the obtained values of Cdl from insulating
(lower oxidation potential) and conducting (higher oxidation
potential) electrocatalysts were from the contribution of bare
working electrodes and electrocatalysts, respectively.[37,38] To
exclude the effect of bare electrode, the method to obtain Cdl is
to perform CVs at Faradaic silence potential range with various
scan rates. Additionally, roughness factor (RF) – which is the
ratio of real surface area to geometric surface area – can be
further calculated by dividing the ECSA with geometric area of
electrode. By plotting the current (the difference of forward and
reverse current at middle of potential window) against scan rates
(Figure 1b and S2), the obtained slope was Cdl (Figure 1c).
Furthermore, the value of ECSA could be calculated through
dividing Cdl by specific capacitance of 40 μF cm−2 for the
transition-metal oxide in 1 M alkaline solution.[34] After correcting
the initial OER behavior of electrodeposited Co on different
supports by ECSA/RF, the results of OER current density were
very similar (Figure 1d), indicating the similar numbers of active
sites among as-prepared Co catalysts. Furthermore, this result
evidently demonstrated the cobalt metal-based nickel foam
electrode with the better OER behavior was attributed to the
more population of active sites but not the increase of site activity.
Hence, any comparison regarding to activity of electrocatalyst
Results and Discussion
Given their large surface areas, rough support materials – such
as carbon cloth or nickel foam – have proved to enhance the
activity of electrocatalysts toward the OER. Figure 1a showed
the OER behavior of electrodeposited Co on different surface
roughness of support electrodes. In which, the electrodeposited
Co was characterized by X-ray diffraction (XRD) pattern, and the
diffraction peaks at 2θ = 41.6, 44.6, 47.5 and 76.0° could be
indexed to (100), (002), (101), and (110) of hexagonal-phase
cobalt (JCPDS: 05-0727) (Figure S1). Although the respective
currents of Co metal have been normalized by the geometric
area of individual electrode, the OER behavior of Co metal on
the nickel foam outcompetes those on carbon cloth, glass carbon
and fluorine doped tin oxide (FTO), respectively. This result is
predictable because the difference in surface area of the
substrate creates difference in population of reactive sites. It had
been pointed out that the evaluation of OER activity should take
into account the surface areas of corresponding catalysts.[36]
Several methodologies have been proposed to estimate the
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
Figure 1. (a) LSV curves of Co on FTO, GC, nickel foam and carbon cloth electrode. (b) CVs in the potential range of faradaic silence with various scan rates for
Co on FTO. (c) Measured capacitive currents of Co on different substrates at various scan rates. (d) Corrected RF for LSV curves of Co on FTO, GC, nickel foam
and carbon cloth.
should not be concluded until the correction by ECSA.
were chosen for the following investigation. The peak positions
of CoO, c-CoSe2 and CoFeO, illustrated in Figure S7, were in
good agreement with the standard JCPDS card no. 74-2392, 09-
0234 and 22-1086, respectively, indicating that those pre-
catalysts were prepared successfully. Also, the experimental
conditions were strictly controlled, namely the same loadings of
Co cations which was quantified by inductively coupled plasma
mass spectrometry (ICP-MS) and the same rotating speed of
1,600 rpm using rotating disk electrodes (RDE). Figure 3a
showed the linear sweep voltammetry (LSV) polarization curves
in 1 M Fe free KOH for OER while each of specimens reached
steady-state after scanning by several cycles. The
electrocatalysts of CoFeO, CoO, CoOOH and c-CoSe2
individually showed an η10 of 330, 470, 400 and 430 mV,
indicating that the ranking of OER behavior was in order of
CoFeO, CoOOH, c-CoSe2 and CoO. Figures 1d and 2c have
been demonstrated the correction of OER behavior by RF could
exclude influences of the population of active sites. Numerous
studies have also pointed out and supported this
correction.[34,35,40] Editorial of American Chemical Society has
announced the fairer comparison of electrocatalytic activity
should be corrected by ECSA firstly.[40] However, in this case of
testing various electrocatalysts, the timing to execute the
measurement of ECSA, which had not been clarified, becomes
extremely important. Especially, it has been confirmed that most
of electrocatalysts underwent a structural self-reconstruction or
phase transformation during OER and even happened
immediately when immersed into alkaline electrolyte.[25,27] For
Additionally, electrocatalyst loadings also influenced the
related OER behavior. We synthesized cobalt oxyhydroixde
(CoOOH) confirmed by XRD for the sake of such an evaluation
(Figure S3).[39] Figures 2a and2b shows the loading dependence
of OER behavior of CoOOH on GC, short for glass carbon,
electrodes, and the corresponding individual RFs were estimated
by aforementioned measurements and displayed in Figures S4-
S6. At lower CoOOH loadings, the OER behavior was enhanced
obviously with the increase of CoOOH loadings because nearly
all of the active sites were accessible for the OER. However, an
excess loading of electrocatalyst did not expectedly further
improve the OER behavior, indicating a saturation effect on the
number of surface active sites. The correction of Figure 2a by the
population of active sites clearly illustrated these OER behavior
as well as site activity were almost the same (Figures 2b and 2c).
Meanwhile, the reason of adopting corrected overpotential at
current density of 0.1 mA/cm2 would be discuss latter.
So far, studies related to OER for the most part had been
focused on developing new electrocatalysts. However, the lack
of benchmarking misled the conclusion that good OER behavior
was attributed to highly active electrocatalysts. The above-
mentioned results have been demonstrated that the same
electrocatalyst exhibited different OER behavior as the loading
or support electrodes were changed (Figures 1 and 2). To
comprehensively dissect the enhancement factor of different
electrocatalysts toward the OER, cobalt-based electrocatalysts
including CoO, cubic CoSe2 (c-CoSe2), CoOOH and CoFeO
Figure 2. (a) LSV curves of various amount of CoOOH on GC electrode. (b) OER performance of overpotential at η10 without RF correction and at η0.1 with RF
correction to various amount of CoOOH. (c) Corrected LSV curves of various amount of CoOOH on GC electrode by RF.
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
Figure 3. (a) LSV curves of CoFeO, CoO, CoOOH and c-CoSe2 on GC electrodes. (b) RF of CoFeO, CoO, CoOOH and c-CoSe2 on GC electrodes before and after
OER. (c) Correction of RF for LSV curves of CoFeO, CoO, CoOOH and c-CoSe2 on GC electrodes.
instance, in-situ Raman spectroscopy has probed the phase
transformation of cobalt metal into thermodynamic stable Co3O4
while cobalt metal was immersed in 0.1 KOH. At the onset
potential of the OER, the features of Co3O4 gradually attenuated
along with the occurrence of second phase transformation to
instance, image of scanning electron microscope in the literature
showed that the electrocatalyst, NiFe2O4, did not completely
cover the surface of hybrid material, single-walled carbon
nanotube which was inactive toward OER.[13] Additionally, in the
case of c-CoSe2, the restriction of phase transformation causes
mixing phases which were real active CoOOH and slightly
active/inactive c-CoSe2 toward OER. The correction of LSV
curve of c-CoSe2 by ECSA without excluding the residual c-
CoSe2 would also overestimate the population of active sites,
CoOOH, and underestimate the real activity provided by CoOOH.
Altogether, complicated is to dissect various possible factors that
influenced the measurements of the surface area of exposed real
active species. As yet, most of the comparisons of OER activity
depended on the values of η10 which were obtained by the
normalization of geometrical surface area of the electrode. This
normalization is based on assuming RF value of 1 which is under
the situation of smooth film-typed catalyst on flat electrode.
However, the real situation is not the case. Figures 1 and 3
indicated the values of RF was at least 10 to 100 times greater
than 1 when high specific surface area of nano-sized catalyst or
three-dimensional bare electrode was used. It should be noticed
that the current indicator parameter for the comparison of OER
behavior was commonly used η10 which was corrected by
geometric area with assuming RF was 1. Furthermore, the OER
behavior would be decreased 10 to 100 times while the
contribution of RF was taken into account, meaning that the
value of OER current should be higher than 100 mA to 1 A before
calculating η10. However, most prepared electrocatalytic
electrodes in laboratory can not achieve to such high threshold
of OER current. The simple method is to decrease 100 times of
the adopted current density from 10 mA/cm2 (η10-geometric area) to
CoOOH.[41]
Additionally,
operando
X-ray
absorption
spectroscopy showed the CoSx was spontaneously transforming
into Co(OH)2 in phosphate buffered solution.[42] It is worth noting
that the phase transformation also influences the morphology of
electrocatalyst. The comparisons of SEM images of pre-
electrocatalysts and post-electrocatalysts were clearly indicated
that the surface morphologies were changed in the duration of
OER (Figure S8). These phenomena further influenced the
evolutions of surface area and RF (Figure 3b). For this reason, it
is incorrect to evaluate the RF of electrocatalyst in an initial state
rather than in a final steady-state after performing the OER. It
could be observed that the ranking of OER behavior to the as-
prepared electrocatalysts was changed after normalizing their
individual RFs which were obtained after reaching stable LSV.
Meanwhile, the pre-catalysts of CoO and CoOOH exhibited the
same OER behavior. For CoO toward OER, the less stable CoO
eventually transformed into real reactive species of CoOOH,[43]
even though the formation of Co(OH)2 occurred through
hydration process in alkaline solutions.[44] Since the transformed
CoOOH from CoO dominated the OER behavior, it could be
concluded that the pre-catalyst CoO exhibited the same OER
activity as the prepared CoOOH. In contrast, the
thermodynamically stable c-CoSe2 suppressed the major part of
phase transformation and therefore was unfavorable to the
OER.[27] By excluding the contribution of surface areas, c-CoSe2
exhibited lower OER activity comparing to the CoOOH and CoO.
The incorporation of elements into Co-based oxides to decrease
surface adsorption energy of intermediates or to increase
roughness factor has been disclosed for efficiently improving
OER behavior.[12,25,27,29] Among these elements, Fe significantly
enhances the electrocatalytic activity to OER. Our pervious study
has reported the introduction of Fe could occupy an octahedral
site as well as confine Co to an active tetrahedral site toward
OER.[45,46] Also, Fe-doped cobalt phosphides accelerated the
transformation of CoOOH. According to the literature, the
incorporation of Fe into CoOOH intensified the intrinsic activity
toward the OER by ~100 times over to pure CoOOH.[47] The
obtained LSV curves were accordant with the previous reports
that CoFeO showed the highest OER activity followed by
CoOOH/CoO and c-CoSe2 after correcting by RF (Figure 3c). It
is worth noting that the evaluation of electrocatalytic activity
through the normalization of RF or ECSA is based on the
assumption that the real reactive species covers the surface of
support or hybrid material completely. However, we need to point
out that although being inactive, the support or hybrid material
could contribute its double-layer capacitance to the
corresponding ECSA, resulting in an overestimation of ECSA
and therefore an underestimation of catalytic activity. For
Figure 4. Correction of overpotential at the current density of 0.1 mA/cm2 to
Co-based electrocatalysts by RF.
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
0.1 mA/cm2 (η0.1-RF/ECSA). Figure 4 showed the corrected η0.1 of
electrodeposited Co on different substrates and different pre-
catalysts. The pre-catalysts of Co, CoO and CoOOH exhibited
nearly the same activity since they all converted to active site,
CoOOH. This result highly suggested that the OER activity could
be accurately distinguished based on the overpotential at current
density of 0.1 mA/cm2 (η0.1-RF/ECSA) after the correction of
RF/ECSA, which could be used as benchmarking for evaluating
and differentiating the contribution of site activity and population
of active site toward OER behavior.
Furthermore, Tafel plot which is obtained by plotting the
logarithm of the current density as a function of the overpotential
is often adopted to discern the mechanism and reaction kinetics
of OER. Figure 5a showed the Tafel plots of CoFeO, CoOOH,
CoO and c-CoSe2 with Tafel slope of 58, 74, 73 and 79 mV/dec,
respectively. The lowest Tafel slope of CoFeO manifested its fast
kinetics for the OER. This result was also in agreement with
literature that the incorporation of Fe could reduce the Tafel
slope of Co3O4.[48] Nevertheless, the catalytic mechanism was
not changed while the present Fe in the lattice.[45] Ideally, Tafel
slope could be an indicator for evaluating catalyst activity; a fast
charge transfer across the electrocatalyst interface in the
duration of OER leads to a lower activation overpotential, having
a smaller Tafel slope. However, resistances from electrocatalyst,
electrocatalyst-support electrode interface, and electrolyte also
influence the obtained Tafel slopes, suggesting Tafel slopes did
not truly reflect the intrinsic activity of electrocatalysts. This
phenomenon was observed in the comparison of numerous
studies which revealed the ranking of Tafel slopes did not
correspond to those of η10 (Figure 5b).[11,12-30]
Figure 6. TOF values of CoFeO and CoOOH at overpotential of 0.37 V in 1m
KOH electrolyte for various amount of loading Co.
the current density at a certain overpotential, A is geometrical
surface area of the electrode, F is he Faraday constant (96485
C/mol), and m is the electrocatalyst loadings. Generally,
electrocatalysts with high values of TOF toward OER suggested
the fast OER kinetics and electron transfer. The calculation of
TOF was assumed ideally that all the active sites were exposed
on the surface and could contribute electrocatalytic current
density. Once again, it still presents some imprecise
measurements, especially to the estimation of the electrocatalyst
loadings. As shown in Figure 6, the values of TOF at the
overpotential of 0.37 V decreased with the increase in the
amount of CoOOH, indicating that the derived values of current
density and the amount of CoOOH were positively correlated but
not proportional. It should be noticed that the same
electrocatalyst should perform the same activity if the use of TOF
as a metric parameter for the comparison of electrocatalytic
activity. Also, quantifying the concentration of surface active sites
by ICP-MS was based on the premise that entire loading
electrocatalysts were involved in electrocatalytic reaction.
However, this ideal assumption to most catalysts was farfetched
because this quantification includes the inner part of
electrocatalyst itself which in fact did not react. Take Co3O4 as
an example. It has been demonstrated that Co3O4 was size-
independent OER activity based on the correction of ECSA,[32]
rather than the smaller particle size owning the higher activity
based on estimated TOF by using ICP-MS. Therefore, using
Another important metric for examining the intrinsic activity of
electrocatalyst is turnover frequency (TOF) which is defined as
the rate of species produced in mol per surface metal atom per
second. The value of TOF toward OER could be calculated
according to the equation of TOF = (j × A)/(4 × F × m), where j is
TOF as
a criterion for comparison of activity among
electrocatalysts is doubtful before developing the accurate
quantification of concentration of surface active sites.
Conclusion
Facing
explosive
developments
in
heterogeneous
electrocatalysts, scientists should take a clear insight into the
activity parameters to avoid being misled by tricks in terms of
ignoring population of activity site. Therefore,
a
fair
benchmarking, involving the effect of ECSA or RF to
electrocatalytic current, is proposed for the comparison of
electrocatalyst activity. Meanwhile, the timing to measurements
of ECSA or RF should be after electrocatalysis, considering the
phase transformation of as-prepared electrocatalysts. By ECSA
or RF correction, more accurately evaluated is that whether the
Figure 5. (a) Tafel slopes of CoFeO, CoO, CoOOH and c-CoSe2 derived from
Figure 3a. (b) Comparison of Tafel slopes and η10 from different
electrocatalysts toward OER.
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
prepared catalysts modified electrode as the working electrode. LSV was
recorded within potential window ranged from 0 to 0.8 V (vs SCE) with
the scanning rate of 1 mV/s for several times until the LSV curve did not
change. The double-layer capacitance (Cdl) is recorded cyclic
voltammograms (CV) at different scan rates within Faradaic silence
potential range.
overriding electrocatalytic current is attributed to the population
of actives sites or to the activity of individual sites. We also
suggested that the contribution of site activity and population of
active site toward OER behavior could be fully profiled through
the estimation of overpotential at current density of 0.1 mA/cm2
(η0.1-RF or η0.1- ECSA) after the correction of RF/ECSA. This
proposed benchmarking not only provided a greatly significant
assessment in the OER activity but also precluded some
deliberately misleading means in evaluating electrocatalytic
activity.
Acknowledgements
H.-C.C. and T.-L.C. contributed equally to this work. The authors
acknowledge support from the Ministry of Science and
Technology, Taiwan (Contract Nos. MOST 107-2628-M-002-
015-RSP, 108-2113-M-153 -001 and 108-2113-M-213-006).
Experimental Section
Chemicals: Cobalt(II) nitrate hexahydrate (99%, Co(NO3)3·6H2O),
cobalt(II) chloride hexahydrate (98%, CoCl2·6H2O), lithium perchlorate
(≥99%, LiClO4) and urea (99.5%) were purchased from ACROS.
Hydrogen peroxide (35 wt%, H2O2) and iron(II) chloride (98%) were
purchased from Alfa Aesar. Potassium hydroxide (KOH) was purchased
from Fisher Scientific. Selenium powder (≥99.5%) was purchased from
Sigma-Aldrich.
Keywords: electrocatalyst • benchmarking • ECSA • OER •
activity
[1]
[2]
P. Boldrin, N. P. Brandon, Nat. Catal. 2019, 12, 571-577.
S. C. Lin, C. S. Hsu, S. Y. Chiu, T. Y. Liao, H. M. Chen, J. Am. Chem.
Soc. 2017, 139, 2224-2233.
Electrodeposition of Co: The electrodeposition of Co on GC, nickel
foam, FTO and carbon cloth electrodes were similar to that reported in
the literature.[49] The mixture solution of 0.05 M of CoCl2·6H2O and 0.95
M LiClO4 was prepared as an electrolytic solution for electrodeposition of
Co on bare electrode (GC, nickel foam, FTO and carbon cloth) under the
potentiostatic conditions of -0.81 V vs. saturated calomel electrode (SCE)
for 120s under stirring rate of 200 rpm. In which, the three-electrode cell
contained Pt wire as the counter electrode, SCE as the reference
electrode and various bare electrode as the working electrode. After
electrodeposition, the samples were quickly washed with deionized water
and was dried in vacuum oven.
[3]
[4]
N. T. Suen, S. F. Hung, Q. Quan, N. Zhang, Y. J. Xu, H. M. Chen, Chem.
Soc. Rev. 2017, 46, 337-365.
Y. Y. Hsu, N. T. Suen, C. C. Chang, S. F. Hung, C. L. Chen, T. S. Chan,
C. L. Dong, C. C. Chan, S. Y. Chen, H. M. Chen, ACS Appl. Mater.
Interfaces 2015, 7, 22558-22569.
[5]
R. Subbaraman, D. Tripkovic, D. Strmcnik, K. C. Chang, M. Uchimura,
A. P. Paulikas, V. Stamenkovic, N. M. Markovic, Science 2011, 334,
1256-1260.
[6]
[7]
V. I. Birss, A. Damjanovic, P. G. Hudson, J. Electrochem. Soc. 1986,
133, 1621-1625.
R. Kötz, H. J. Lewerenz, S. Stucki, J. Electrochem. Soc. 1983, 130, 825-
829.
Preparation of CoOOH: The preparation of CoOOH was according to
previous literature.[39] In which, 1.25 g of Co(NO3)2·6H2O and 2.1 g of urea
were dissolved in 10 mL deionized water under stirring. The clear solution
was dried at 80 oC to obtain powder before transferring to a preheated
furnace at 400 oC for 10 min. The blue foam was ground to a powder by
agate mortar. Then, 200 mL of H2O2 (30 wt%) was slowly added to the
powder for transforming Co-based complex to CoOOH. The precipitate
was collected and was repeated the above process for seven times for
complete transformation. Finally, the CoOOH was obtained after drying
at 80 oC.
[8]
[9]
R. Kötz, H. Neff, S. Stucki, J. Electrochem. Soc. 1984, 131, 72-77.
C. S. Hsu, N. T. Suen, Y. Y. Hsu, H. Y. Lin, C. W. Tung, Y. F. Liao, T.
S. Chan, H. S. Sheu, S. Y. Chen, H. M. Chen, Phys. Chem. Chem. Phys.
2017, 19, 8681-8693.
[10] J. Qi, W. Zhang, R. Cao, ChemCatChem 2018, 10, 1206-1220.
[11] P. W. Menezes, C. Walter, J. N. Hausmann, R. Beltrμn-Suito, C.
Schlesiger, S. Praetz, V. Y. Verchenko, A. V. Shevelkov, M. Driess.
Angew. Chem. Int. Ed. 2019, 58, 16569-16574.
[12] B. Cao, Y. Cheng, M. Hu, P. Jing, Z. Ma, B. Liu, R. Gao, J. Zhang. Adv.
Funct. Mater. 2019, 29, 1906316.
[13] M. Fei, H. Shi, J. Zhao, N. Kang, W. He, H. Li, F. Yang, ChemCatChem
2018, 10, 5174-5181.
[14] Y. Zuo, D. Rao, S. Ma, T. Li, Y. H. Tsang, S. Kment, Y. Chai. ACS Nano
2019, 13, 11469-11476.
Preparation of c-CoSe2: The c-CoSe2 was synthesized according to our
previous report.[27] Firstly, cobalt hydroxide carbonate hydrate was
prepared by a mixture solution (10 mL) containing 0.625 g of urea and
0.437 g of Co(NO3)3·6H2O at 90 oC after 6 h. The collected precipitate
was washed with deionized water and dried under vacuum. Furthermore,
[15] W. Wang, Y. Lu, M. Zhao, R. Luo, Y. Yang, T. Peng, H. Yan, X. Liu, Y.
Luo, ACS Nano 2019, 13, 12206-12218.
[16] J. Yan, L. Kong, Y. Ji, J. White, Y. Li, J. Zhang, P. An, S. Liu, S.T. Lee,
Tianyi Ma. Nat. Comm. 2019, 10, 2149.
0.06
g of selenium powder and the precipitate, cobalt hydroxide
carbonate hydrate, were placed separately in a glass tube before being
heated in a tube furnace at 400 oC for 3 h under a 5% H2/Ar atmosphere
with the heating rate of 4 °C min. After cooling to room temperature, the
c-CoSe2 was obtained and was re-dispersed into deionized water.
[17] L. Fan, P. Zhang, B. Zhang, Q. Daniel, B. J. J. Timmer, F. Zhang, L.
Sun, ACS Energy Lett. 2018, 3, 2865-2874.
[18] Y. Hou, M. Qiu, M. G. Kim, P. Liu, G. Nam, T. Zhang, X. Zhuang, B.
Yang, J. Cho, M. Chen, C. Yuan, L. Lei, X. Feng, Nat. Comm. 2019, 10,
1329.
Preparation of CoFeO: CoFeO was prepared by chemical bath
deposition and postcalcination which had been reported in the previous
literature.[45] Firstly, the precipitate was obtained from a mixture solution
(10 mL) containing 0.625 g of urea, 0.437 g of Co(NO3)3·6H2O and FeCl2
(0.152 g) at 90 oC after 3 h. The collected precipitate was washed with
deionized water and dried under vacuum. Furthermore, the obtained
precipitate was placed in a glass tube before being heated in a tube
furnace at 300 oC with the heating rate of 2 °C min for 2 h. After cooling
to room temperature, the CoFeO was obtained and was re-dispersed into
deionized water.
[19] Z. Y. Yu, Y. Duan, J. D. Liu, Y. Chen, X. K. Liu, W. Liu, T. Ma, Y. Li, X.
S. Zheng, T. Yao, M. R. Gao, J. F. Zhu, B. J. Ye, S. H. Yu. Nat. Comm.
2019, 10, 2799.
[20]
Y. Zhang, J. Fu, H. Zhao, R. Jiang, F. Tian, R. Zhang, Appl. Catal., B
2019, 257, 117899.
[21] J. Zhang, J. Liu, L. Xi, Y. Yu, N. Chen, S. Sun, W. Wang, K. M. Lange,
B. Zhang, J. Am. Chem. Soc. 2018, 140, 3876-3879.
[22] J. G. Li, K. Xie, H. Sun, Z. Li, X. Ao, Z. Chen, K. K. Ostrikov, C. Wang,
W. Zhang, ACS Appl. Mater. Interfaces 2019, 11, 36649-36657.
[23] Y. Tian, L. Cao, P. Qin. ChemCatChem 2019, 11, 4420-4426.
[24] M. A. R. Anjum, M. D. Bhatt, M. H. Lee, J. S. Lee, Chem. Mater. 2018,
30, 8861-8870.
[25] S. F. Hung, Y. Zhu, G. Q. Tzeng, H. C. Chen, C. S. Hsu, Y. F. Liao, H.
Ishii, N. Hiraoka, H. M. Chen, ACS Energy Lett. 2019, 4, 2813-2820.
[26] G. Chen, Y. Zhu, H. M. Chen, Z. Hu, S.F. Hung, N. Ma, J. Dai, H.J. Lin,
C.T. Chen, W. Zhou, Z. Shao, Adv. Mater. 2019, 31, 1900883.
[27] Y. Zhu, H. C. Chen, C. S. Hsu, T. S. Lin, C. J. Chang, S. C. Chang, L.
D. Tsai, H. M. Chen, ACS Energy Lett. 2019, 4, 987-994.
Preparation of CoO: The method of preparing CoO was according to the
literature.[50] In which, Co3O4 powders were placed in quartz tube furnace
with flowing Ar under ambient pressure. The tube was heated from 25 oC
to 800 oC in 30 mins and from 800 oC to 1000 oC in 2 hrs. The temperature
was kept at 1000 oC for 10 hrs before cooling to room temperature.
Electrochemical measurements: Electrochemical behavior was
assessed on a Bio-Logic SP-300 workstation at RT. All tests of LSV and
ECSA were carried out in 1 M KOH electrolyte by three-electrode cell with
Pt wire as the counter electrode, SCE as the reference electrode and as-
[28] K. Fan, H. Zou, Y. Lu, H. Chen, F. Li, J. Liu, L. Sun, L. Tong, M. F.
Toney, M. Sui, Jiaguo Yu, ACS Nano 2018, 12, 12369-12379.
[29] C. Dong, L. Han, C. Zhang, Z. Zhang, ACS Sustainable Chem. Eng.
2018, 6, 16096-16104.
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
[30] X. Xu, Z. Zhong, X. Yan, L. Kang, J. Yao, J. Mater. Chem. A 2018, 6,
5999-6006.
[31] X. Wang, H. Xiao, A. Li, Z. Li, S. Liu, Q. Zhang, Y. Gong, L. Zheng, Y.
Zhu, C. Chen, D. Wang, Q. Peng, L. Gu, X. Han, J. Li, Y. Li, J. Am. Chem.
Soc. 2018, 140, 15336-15341.
[32] S. Sun, H. Li, Z. J. Xu, Joule 2018, 2, 1019-1027.
[33] L. Bai, C. S. Hsu, D. T. L. Alexander, H. M. Chen, X. Hu, J. Am. Chem.
Soc. 2019, 141, 14190-14199.
[34] C. C. L. McCrory, S. Jung, J. C. Peters, T. F. Jaramillo, J. Am. Chem.
Soc. 2013, 135, 16977-16987.
[35] S. Jung, C. C. L. McCrory, I. M. Ferrer, J. C. Peters, T. F. Jaramillo, J.
Mater. Chem. A 2016, 4, 3068-3076.
[36] C. Wei, S. Sun, D. Mandler, X. Wang, S. Z. Qiao, Z. J. Xu, Chem. Soc.
Rev. 2019, 48, 2518-2534.
[37] P. Acharya, J. N. Burrow, M. Abolhassani, L. F. Greenlee, ECS Trans.
2018, 85, 81-89.
[38] A. S. Batchellor, S. W. Boettcher, ACS Catal. 2015, 5, 6680-6689.
[39] W. Wen, D. Liang, J. P. Cheng, J. M. Wu, RSC Adv. 2016, 6, 70947-
70951.
[40] D. Voiry, M. Chhowalla, Y. Gogotsi, N. A. Kotov, Y. Li, R. M. Penner, R.
E. Schaak, P. S. Weiss, ACS Nano 2018, 12, 10, 9635-9638.
[41] B. S. Yeo, A. T. Bell, J. Am. Chem. Soc. 2011, 133, 5587-5593.
[42] N. Kornienko, J. Resasco, N. Becknell, C. M. Jiang, Y. S. Liu, K. Nie, X.
Sun, J. Guo, S. R. Leone, P. Yang, J. Am. Chem. Soc. 2015, 137, 7448-
7455.
[43] C. W. Tung, Y. Y. Hsu, Y. P. Shen, Y. Zheng, T. S. Chan, H. S. Sheu,
Y. C. Cheng, H. M. Chen, Nat. Commun. 2015, 6, 8106.
[44] X. Ma, K. Li, X. Zhang, B. Wei, H. Yang, L. Liu, M. Zhang, X. Zhang, Y.
Chen, J. Mater. Chem. A 2019, 7, 14904-14915.
[45] S. F. Hung, Y. Y. Hsu, C. J. Chang, C. S. Hsu, N. T. Suen, T. S. Chan,
H. M. Chen, Adv. Energy Mater. 2018, 8, 1701686.
[46] S. F. Hung, Y. T. Chan, C. C. Chang, M. K. Tsai, Y. F. Liao, N. Hiraoka,
C. S. Hsu, H. M. Chen, J. Am. Chem. Soc. 2018, 140, 17263-17270.
[47] M. S. Burke, M. G. Kas, L. Trotochaud, A. M. Smith, S. W. Boettcher, J.
Am. Chem. Soc. 2015, 137, 10, 3638-3648.
[48] C. Xiao, X. Lu, C. Zhao, Chem. Commun. 2014, 50, 10122-10125.
[49] C. Q. Cui, S. P. Jiang, A. C. C. Tseung, J. Electrochem. Soc. 1990, 137,
3418-3423.
[50] L. Liao, Q. Zhang, Z. Su, Z. Zhao, Y. Wang, Y. Li, X. Lu, D. Wei, G.
Feng, Q. Yu, X. Cai, J. Zhao, Z. Ren, H. Fang, F. Robles-Hernandez, S.
Baldelli, J. Bao, Nat. Nanotechnol. 2014, 9, 69-73.
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902316
ChemCatChem
FULL PAPER
Entry for the Table of Contents
Contribution factor to electrocatalysis: Taking population of
active site into account toward heterogeneous electrocatalysis is
indispensable before further investigating site activity. A series
of cobalt-based pre-electrocatalysts are taken as an example to
probe their individual correlation within electrocatalytic behavior,
electrocatalytic activity and population of active site in order to
propose
electrocatalytic activity. This proposed benchmarking precluded
some deliberately misleading means in evaluating
electrocatalytic activity.
a criterion to implement a fair comparison of
8
This article is protected by copyright. All rights reserved.