Electroreduction of a CoII coordination complex producing a metal-organic film with high performance toward electrocatalytic hydrogen evolution

Article abstract of DOI:10.1039/c8ta07340f

This paper describes the synthesis and structural characterization of a novel, cheap and simple Co^II complex (Co^II(L)₂Cl₂) based on the 1,3,5-trisubstituted-pyrazoline ligand along with the electrochemical production of metal-organic electroactive films derived from this new complex. These systems were applied as electrocatalysts for hydrogen production (ACN/ACA or ACN/TFA medium) where both materials presented high performance toward hydrogen evolution. Compared to the Co^II complex, the electroactive films exhibited significant electroactivity toward hydrogen evolution, presenting a remarkable TOF for H₂ production (312:900 s^-1, corrected by Faraday efficiency) in the presence of TFA. In addition, the generated metal-organic film showed high stability toward the electrocatalytic hydrogen production, supporting at least 1000 cycles at 20 mV s^-1 in the large potential range investigated, as well as good performance and stability in the presence of 0.5 M H₂SO₄. Relevant insights into the mechanistic details and the role played by the Co^II complex and the films during the catalytic hydrogen production are also discussed in light of the structural features and electrochemical experiments.

Full text of DOI:10.1039/c8ta07340f

View Article Online
View Journal
Journal of
Materials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: L. S. Bezerra, P. P.
Rosa, G. V. Fortunato, L. Pizzuti, G. A. Casagrande and G. Maia, J. Mater. Chem. A, 2018, DOI:
10.1039/C8TA07340F.
This is an Accepted Manuscript, which has been through the
Volume
4 Number 1 7 January 2016 Pages 1–330
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry A
Accepted Manuscripts are published online shortly after
Materials for energy and sustainability
www.rsc.org/MaterialsA
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2050-7488
PAPER
Kun Chang, Zhaorong Chang et al.
Bubble-template-assisted synthesis of hollow fullerene-like
MoS2 nanocages as a lithium ion battery anode material
rsc.li/materials-a

Page 1 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
Electroreduction of CoII Coordination Complex
Producing MetalꢀOrganic Film with High
Performance Toward Electrocatalytic Hydrogen
Evolution
Leticia S. Bezerraa, Persiely P. Rosab, Guilherme V. Fortunatoa, Lucas Pizzutib, Gleison A.
Casagrandeb*, Gilberto Maiaa*
a) Instituto de Química, Universidade Federal de Mato Grosso do Sul; Av. Senador Filinto
Muller, 1555; Campo Grande, MS 79074ꢀ460, Brazil.
b) Grupo de Pesquisa em Síntese e Caracterização Molecular do Mato Grosso do Sul, Instituto
de Química, Universidade Federal de Mato Grosso de Mato Grosso do Sul (laboratório 2),
Av. Senador Filinto Muller 1555, Campo Grande, MS 79074ꢀ460, Brazil.
KEYWORDS: hydrogen evolution, cobalt catalysts, electrocatalysis, metal-organic films,
pyrazoline ligands
ABSTRACT: This paper describes the synthesis and structural characterization of a novel,
cheap and simple CoII complex (CoII(L)2Cl2) based on 1,3,5ꢀtrisubstitutedꢀpyrazoline ligand
along with the electrochemical production of metalꢀorganic electroactive films derived from this
new complex. These systems were applied as electrocatalysts for hydrogen production
(ACN/ACA or ACN/TFA medium) where both materials presented high performance toward
hydrogen evolution. Compared to the CoII complex, the electroactive films exhibited significant
electroactivity toward hydrogen evolution, presenting a remarkable TOF for H2 production (312
900 sꢀ1, corrected by Faraday efficiency) in the presence of TFA. In addition, the generated
metalꢀorganic film showed high stability toward the electrocatalytic hydrogen production,
supporting at least 1000 cycles at 20 mV s–1 under the large potential range investigated, as well
1

Journal of Materials Chemistry A
Page 2 of 32
View Article Online
DOI: 10.1039/C8TA07340F
as good performance and stability in presence of 0.5 M H2SO4. Relevant insights into the
mechanistic details and the role played by the CoII complex and the films during the catalytic
hydrogen production are also discussed in light of the structural features and electrochemical
experiments.
1.
INTRODUCTION
The production of hydrogen from protons reduction has attracted an increasingly growing
attention owing to the renewable and carbonꢀfree energy characteristic of the process
(interconversion of electricity and chemical energy),1 which is considered as a simple chemical
reaction for the storage of renewable energy.2
While it is an indisputable fact that platinum (Pt) is one of the most effective catalysts for H2
production, its scarcity (high cost),3 however, tends to hinder the use of catalysts based on
precious metals in large scale.2 The last few years have witnessed the development of catalysts
using firstꢀrow transition (or complexes based on earthꢀabundant metals)4 instead of rare noble
metals for the reduction of protons to hydrogen.5,6 These catalysts are normally bioinspired
mononuclear complexes,7ꢀ8 which may undergo degradation under protic and reducing
conditions,9 being deposited as heterogeneous material on the electrode surface9,10 but yet
maintaining their activity in the hydrogen production.9
Some examples of Ni complexes11ꢀ14 comply the function of electrocatalysts for hydrogen
production as well as the electrode adsorbed films based on Ni metalorganic species produced
from Ni(II) complexes degraded electrochemically9,15 or from electrodeposition on glassy carbon
(GC) surface from [Bu4N][Ni(bdt)2] [tetrabutylammonium bis(1,2ꢀbenzenedithiolate)nickel] in
the presence of acid.16 Tri,17ꢀ20 bi,21ꢀ25 and monovalent26,27 Co complexes, as well as di,28ꢀ30 and
tricobalt31 complexes have been reported to act as electrocatalysts for hydrogen production.
The electrodeposition of Co nanoparticles on the electrode surface based on the reduction of
[Co (dpg)3(BF)2]BF4 (dpg = diphenylglyoxime as ligand; a CoIII complex) in acetonitrile solution
containing acid yields the active nanoparticles for H2 production.32 Ghachtouli et al.33 reported
the results of their study on cobalt (II) ꢀ trisglyoximato clathrochelate complex. Their findings
detected the deposition of active particles on the surface of GC electrode for H2 evolution in the
presence of HClO4. Difluoroborylꢀbisꢀglyoximatoꢀdiphenyl CoII complex (Co(dpgBF2)2) acted as
a preꢀcatalyst since it was electrochemically altered in acetonitrile in the presence of strong acid
2

Page 3 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
(HClO4). This electrochemical alteration contributed towards the formation of Coꢀbased
nanoparticles on the surface of the carbon electrode which is found to be prone to H2 production
through the reduction of protons.34
Ladderꢀtype crystalline coordination polymers (CPs) have been produced via complexation of
transition metal ions (= Fe, Co, and Ni) with selective bis(chelate) ligands containing sulfur and
nitrogen;35 singleꢀlayer metalꢀorganic frameworks (2D MOFs) constituted of MS2N2 complexes
along with MN4 and MS4 moieties incorporated into the hexagonal networks through metal
dithiolene–diamine coordination (THATꢀM) (M=Co and Ni) and twoꢀdimensional
supramolecular polymer singleꢀlayer sheet composed of triphenyleneꢀfused nickel bis(dithiolene)
complexes (THTNi 2DSP) were produced by using the Langmuir–Blodgett (LB) method
followed by horizontal transference onto a GC surface.36,37 Diimine–dioxime cobalt catalyst or
Ph
[Ni(P2 N2PhCH2CH2COOPht)2](BF4)2 catalyst were attached covalently to the surface of carbon
nanotube electrode decorated with
a
polyꢀ4ꢀ(2ꢀaminoethyl)phenylene or
a
4ꢀ(2ꢀ
aminoethyl)phenyl layer from the electroreduction of the corresponding aryldiazonium salt.38,39
A cobalt dithiolene polymer was immobilized onto a GC electrode by immersing the GC face
into the reaction mixture to produce a polymer, generating a metal−organic surface.40,41 A pyrene
appended bipyridine ligand (P) was used as linker between metal complex and graphitic carbon
electrode surfaces, immobilizing the metal complex via noncovalent interactions.42 These films
aforementioned presented electrocatalytic activity toward hydrogen evolution reaction (HER) in
H2SO4 (0.5 M or pH 1.3).
High catalytic activity toward HER such as small Tafel slope, low onset overpotential and
high stability were obtained by Li and coꢀworkers during their studies on the PANI nanodots
43
decorated CoP hybrid nanowires supported on carbon fibers in 0.5 M H2SO4 along with Co
nanotubes decorated with TiO2 nanodots44 and Cu nanodotsꢀdecorated Ni3S2 nanotubes,45 both in
1 M KOH solution and supported on carbon fibers.
Regarding the use of complexes based on pendant amines ligands, which can act as proton
relay to cobalt metallic sites in the second coordination sphere of metallic complexes, has
essentially been understood as a key step for the electrocatalytic hydrogen evolution when these
systems are in acid solutions. This step has been related to the enhancement of the
electrocatalytic performance by accelerating proton movement during the catalysis, prompting an
increase in the rate and a decrease in the overpotential catalysis. Most of these experiments were
3

Journal of Materials Chemistry A
Page 4 of 32
View Article Online
DOI: 10.1039/C8TA07340F
made using cobalt coordination compounds based on sophisticated mixed amine/phosphine
polydentate ligands (1,3,5,7ꢀtetraphenylꢀ1,5ꢀdiazaꢀ3,7ꢀdiphosphacyclooctane, 1,5ꢀdiphenylꢀ3,7ꢀ
diꢀtertꢀbutyl1,5ꢀdiazaꢀ3,7ꢀdiphosphacyclooctane and P–N–P bidentate ligands).21ꢀ24 Motivated by
these studies, we conducted an investigation regarding the use of a simple and cheap (only U$
0.22 per gram) 1,3,5ꢀtrisubstitudedꢀpyrazoline ligand (1ꢀthiocarbamoylꢀ5ꢀ(4ꢀmethoxyphenyl)ꢀ3ꢀ
phenylꢀ4,5ꢀdihydroꢀ1Hꢀpyrazole) believing that this system is capable of acting as a medium of
accepting and delivering proton to the CoII active metallic site in a new coordination compound
during the catalytic production of hydrogen. The estimated cost of the synthesized CoII complex
is U$ 0.40 per gram. Figure 1A explains the manner in which this new CoII complex can be
assessed in light of its electroactive performance during the catalysis.
Figure 1. Molecular structure of the CoII complex (CoII(L)2Cl2) used as electrocatalyst for
hydrogen production. Emphasis is laid on the catalytic region of the structure (A). Molecular
structure of the synthesized CoII complex (CoII(L)2Cl2). Thermal ellipsoids are drawn at 50%
probability level (B). Coordination sphere of the CoII(L)2Cl2 complex with emphasis on the
intramolecular interactions present in the structure (C).
The synthesized complex possesses a CoII active metallic site and two available amino (NH2)
groups in the second coordination sphere, which can function as an acceptor and deliverer of
protons to the metallic atom during the electrocatalytic hydrogen production constituted by the
4

Page 5 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
area referred as the catalytic region. In this sense, our research group started studying the
preparation, characterization and application of a new, cheap and simple CoII complex based on
the thiocarbamoylꢀpyrazoline ligand as a catalyst and preꢀcatalyst for the production of very
active electrocatalytic metalꢀorganic film for hydrogen production reaction.
2. EXPERIMENTAL SECTION
2.1. Materials and Methods. The chemical substances employed in the experiments included
the following: CoCl2.6H2O, ferrocene (FcCp2), trifluoroacetic acid (TFA), sodium
trifluoroacetato, thiosemicarbazide, solvents (all from SigmaꢀAldrich), acetic acid (ACA)
(Vetec), tetrabutylammonium perchlorate and acetonitrile (ACN) UVꢀHPLC (both from Fluka
Analytical), and sulfuric acid (Merck). A threeꢀelectrode glass cell with a working electrode
(Pine Research Instrumentation) consisting of a Teflonꢀembedded GC disk rotating assembly
(0.196 cm2 geometric area) was employed for the voltammetric measurements. A Teflonꢀ
embedded GC disk (0.196 cm2 geometric area) and Ag wire were used as counter electrode and
pseudoꢀreference electrode respectively.
The procedures involving the synthesis of the new CoII complex and the ligand46ꢀ48 are
presented in the Supporting Information (Scheme S1, Figures S1 and S2).
2.2. Electrode Preparation. The working and counterꢀelectrode GC disk electrodes were
polished with alumina slurries (1 and 0.05 ꢁm) and were afterwards cleaned by sonication in
MilliꢀQ water (Millipore), acetone, and MilliꢀQ water (Millipore), for 5 min in each solvent.
Subsequently, the working and counterꢀelectrode GC disk electrodes were dried by flowing N2
upon the electrodes and immersed in ACN aiming at the complete removal of water. The pseudoꢀ
reference Ag wire electrode was cleaned by immersion in concentrated HCl (∼37%). For the
films produced by electrodeposition on the GC surface, the modified electrodes were immersed
in ACN prior to being transferred to 0.1 M TBAClO4/ACN, and were in turn saturated with N2
(5.0 purity from White Martins).
2.3. Preparation of the Films from CoII(L)2Cl2 Complex. The films were produced by
0/+
electrodeposition onto GC surface (working electrode) fixing the potential at −2.1 V vs FcCp2
during 1800 s followed by 20 cycles at 20 mV s–1 from −0.8 to −2.1 V vs FcCp2 in a 0.1 M
0/+
TBAClO4/ACN solution containing 0.5 mM CoII(L)2Cl2, named Film 1. The Film 2 was obtained
by using similar procedure to produce Film 1, with the solution containing in addition 10 mM
5

Journal of Materials Chemistry A
Page 6 of 32
View Article Online
DOI: 10.1039/C8TA07340F
ACA. The Film 3 was obtained by electrodeposition onto GC surface (working electrode) fixing
0/+
the potential at −2.1 V vs FcCp2 during 1800 s followed by 20 cycles at 20 mV s–1 from −0.5
to −2.1 V vs FcCp20/+ in a 0.1 M TBAClO4/ACN solution containing 0.5 mM CoII(L)2Cl2 and 10
mM TFA.
2.4. Sample Preparation for TEM, IR, and XPS. Film 2 was taken from the GC surface and
diluted in ACN. The diluted solution was dripped onto ultrathin carbon films supported on lacey
carbon film on 400ꢀmesh copper grids (Ted Pella) and used for the TEM experiments.
The IR spectra were obtained from pellets produced by grinding KBr together with the
samples (CoII(L)2Cl2, Film 1, and Film 2), using a mortar until a fine and homogeneous powder
was attained. After milling, the material was placed in a mold and pressed for about 5 minutes,
using hydraulic press, to produce the pellets. Prior to reading the sample in the
spectrophotometer, we had to perform background reading obtained with KBr pellet without the
presence of the samples.
The XꢀRay photoelectron spectroscopy (XPS) spectra were obtained from an electrodeposited
Film 2 onto a graphite plate (1 × 1 × 0.2 cm).
2.5. Apparatuses and Measurements. The electrochemical techniques employed (cyclic
voltammetry (CV) and hydrodynamic cyclic voltammetry (HCV)) were conducted on an AFP2
WaveDriver 20 bipotentiostatꢀgalvanostat (Pine Research Instrumentation) coupled to an
AFMSRCE modulatedꢀspeed rotator. An IRꢀFTIRꢀPerkin Elmer Frontier spectrophotometer was
used at spectral window ranging from 400 to 4000 cmꢀ1 so as to obtain the IR spectra. TG
characterizations of CoII(L)2Cl2 and Film 2 were performed using a TGAꢀ50 thermogravimetric
analyzer (Shimadzu) with 5 mg samples under a Synthetic Air 5.0 FID gas flow (50 mL min−1) at
temperatures ranging from ambient to 900 °C and a heating rate of 10 °C min−1 using alumina
crucibles. Inductive coupled plasma optical emission spectrometry (ICP OES) determination of
Co and S contents at Film 2 was performed on a Thermo Fisher Scientific iCAP 6300 Duo
equipment (operating conditions: 1250 W RF power, 0.45 L min−1 sample flow, 12 L min−1
plasma gas flow rate, 5 s integration time, 20 s stabilization time, 30 psi nebulization pressure,
axial view mode, λ = 238.892 nm to Co and 180.731 nm to S). (Detailed information on the Co
and S ICP OES determination can be found in the Supporting Information). Film 2 nanostructure
was characterized by TEM, which was performed on a JEOL JEM2100 LaB6 apparatus operating
1
at 200 kV. NMR spectra were recorded on a Bruker DPX 300 spectrometer (300.13 MHz for H
6

Page 7 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
and 75.48 MHz for 13C) at 300 K. The reactions applied for the synthesis of the CoII complex and
ligand (see Supporting Information for the synthesis procedures) were carried out with a microtip
probe connected to a 500 W Sonics Vibraꢀcell ultrasonic processor operating at 20 KHz at 25%
of the maximum power output. Elemental analysis (CHN) was performed using a Perkin–Elmer
2400 analyzer. The refinement of the crystal structure of the complex was conducted using a
SHELX package.49 All nonꢀhydrogen atoms were refined with anisotropic displacement
parameters, and the hydrogen atoms were included at their theoretical ideal positions. More
detailed information on the structure determinations, bond lengths and bond angles are given in
Tables S1, S2, and S3 in the Supporting Information. XPS analysis was performed on a
spectrometer system (ESCA+, ScientaꢀOmicron) equipped with a hemispherical analyser
(EA125) and a monochromated Al Kα (Xm1000, 1486.7 eV) radiation source. The chamber was
evacuated to 2 × 10−9 Pa during the analysis, and the energy steps were 50 and 20 eV for the
survey and highꢀresolution spectra respectively. Spectra were deconvoluted using a Voigtꢀtype
function with Gaussian (70%) and Lorentzian (30%) combinations. The UV measurements were
made in a double beam Hitachi Uꢀ3000 Spectrophotometer using quartz cuvette with 1 cm of
optical pathway and capacity of 3 cm3.
3. RESULTS AND DISCUSSION
3.1. Characterization of the CoII(L)2Cl2 and Metal-Organic Films Derived from the CoII
Complex.
3.1.1. X-Ray Studies. The CoII(L)2Cl2 complex was initially characterized by the single
crystal Xꢀray diffraction technique. Figure 1B shows the crystalline and molecular structure of
the new CoII complex.
The CoII(L)2Cl2 can be described as a fourꢀcoordinate CoII compound. The complex has been
presented as a neutral compound where the cobalt atom has Co+2 charge while the chlorine atoms
present Clꢀ1 charges. The sulfur atom from the ligand is bonded covalently to the Co+2 atom
throughout a lone electrons pair. The 1,3,5ꢀtrisubstituted pyrazoline ligand is bonded to the
metallic atom through two Co–S(1) and Co–S(2) bonds of 2.333(2) and 2.327(2) Å, respectively.
The coordination sphere of CoII atom is completed by two Co–Cl(1) and Co–Cl(2) bonds of
2.242(2) and 2.251(3) Å, respectively. Figure 1C shows the coordination sphere environment of
the synthesized CoII complex.
7

Journal of Materials Chemistry A
Page 8 of 32
View Article Online
DOI: 10.1039/C8TA07340F
Based on the models proposed by Addyson and Houser50,51 for the prediction of coordination
geometry in fourꢀcoordinate systems, we applied the τ4 parameter for a better description of the
coordination geometry around the CoII atom. Using the proposed equation (τ4 = 360° ꢀ
(α+β)/141°) and considering the bond angles reported in Table S2 (Supporting Information), the
calculated value for the CoII atom is τ4 = 0.919, indicating a slightly distorted tetrahedral
coordination geometry for the cobalt atom in the complex. A detailed investigation of the
intramolecular secondary bonds present in the structure revealed that the complex exhibits two
weak intramolecular hydrogen bondings, namely (N(3)⋅⋅⋅Cl(1) of 3.255(8) and N(4)⋅⋅⋅Cl(2) of
3.305(7) Å), possessing H⋅⋅⋅Cl interactions estimated to be (N(4)ꢀH(4b)⋅⋅⋅Cl(2) of 2.48 and N(3)ꢀ
H(3b)⋅⋅⋅Cl(1) of 2.44 Å) between Cl and H atoms. These values are in line with similar
compounds reported in the literature.52,53 The NH2 amino groups fill the second coordination
sphere of the CoII atom through two secondary bonds including (Co⋅⋅⋅N(3) and Co⋅⋅⋅N(4))
ranging from 3.441(7) to 3.464(6) Å respectively. These secondary bonds Co⋅⋅⋅N are weaker than
normal covalent bonds, on the other hand they should not be discarded inasmuch as they are
occupying the second coordination sphere of the CoII atom; it is worth mentioning that they also
are within of the Van der Waals radii sum (3.75 Å) for these elementsthe maximum value
considered for an interaction between two atoms.54
3.1.2. Thermogravimetric and ICP OES Analyses, TEM, HRTEM, Electron Diffraction
Pattern, HAADF-STEM, and EDX Mapping Images. TG analysis (Figure S3, Supporting
Information) showed that CoII(L)2Cl2 began burning at 275 °C with nearly 48% mass loss at 315
°C. Beyond the aforementioned temperature, a new burn is noted up to 507 °C (with additional
16% mass loss), and an additional burn up to 550 °C (with additional 28% mass loss), without
perceptible additional mass loss up to 900 °C (remaining mass: 8.0% attributed to cobalt
oxides).55 Film 2 lost water (10% mass loss) at 265 °C and began burning at 265 °C with nearly
30.5% mass loss at 306 °C. Beyond that temperature, a new burn was observed up to 453 °C
(with additional 16.5% mass loss), without appreciable mass loss up to 900 °C (remaining mass:
38% which is also attributed to cobalt oxides). We noted that the CoII(L)2Cl2 complex and Film 2
exhibited similar thermogravimetric behavior (except for the fact that the film lost water whereas
the CoII complex did not; this is in consonance with the absence of water in CoII complex Xꢀray
structure). The similarity occurs until approximately 507° C. Beyond this temperature, only the
8

Page 9 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
CoII complex presented mass loss. Taking into account the difference between the remaining
cobalt oxides residue (38% for Film 2 and 8% for the CoII complex), it is clearly reasonable to
infer that Film 2 is approximately 30% richer in cobalt (or organically poorer) compared to
CoII(L)2Cl2. The underlying implication is that some organic fragments were lost during the
electrochemical production of the film. To definitively confirm the Co amount present in the
Film 2, ICP OES analysis was performed aiming the comparison with TG results. The Co
amount (% mass/mass) determined by ICP OES presented values of 38.54 ± 0.43. Additionally
the amount of S was also quantified by ICP OES as being 3.38 ± 0.03 (% mass/mass). The
amount of Co determined by ICP OES (Tables S4, S5, and S6, Supporting Information) is in
agreement to that obtained by TG measurement, revealing the Co enrichment (or organic
impoverishment) during the electrochemical film formation. This line of thinking is also
supported by the elemental analysis of the Film 2 (C 12.81, H 2.79, and N 1.10%) as compared
to that of the CoII complex (C 54.13, H 4.55, and N 11.07%) in which the organic content is
found to be higher in the latter than in the former. Figure 2 shows the representative TEM,
HRTEM, electron diffraction pattern, HAADFꢀSTEM, and EDX mapping images for the Film 2.
9

Journal of Materials Chemistry A
Page 10 of 32
View Article Online
DOI: 10.1039/C8TA07340F
Figure 2. Representative TEM (A,A’), electron diffraction pattern from the material viewed in
A’ (B), HRTEM (C), HAADFꢀSTEM and EDX mapping (D) images for the Film 2 generated in
the presence of 10 mM ACA.
The TEM images (Figure 2A,A’) of Film 2 show relatively thin, foamꢀlike or cloudlike
structure. It is discarded a possible metallic diffraction pattern (Co metal nanoparticles) due to
the absence of electron diffraction pattern (Figure 2B) at the Film 2; even in the HRTEM images
(Figure 2C) which revealed small dark spots (5ꢀ10 nm) exhibiting fringes features (Moiré
fringes,56 represented in the red ring in Figure 2C). The occurrence of a variation in the lattice
distances (0.22 to 0.33 nm, Figure 2C) should be related to the Moiré fringes characteristic.
Finally the HAADFꢀSTEM and EDX mapping images (Figure 2D) for the Film 2, are showing
high content of Co surrounded by less amount of S and negligible presence of Cl (Table S7,
Supporting Information). The Co enrichment in the Film 2 agrees to the ICP OES and TG
measurements. The sulfur content also agrees to the ICP OES measurements.
10

Page 11 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
3.1.3. IR Spectroscopy Studies. Aiming at establishing a correlation between the organic
fragments which could be present in the produced films structures and in the CoII complex, we
investigated the structural similarities displayed by these materials by using IR spectroscopy.
Figure 3 shows the IR spectra of the CoII(L)2Cl2 and the electrochemically generated films from
the CoII complex.
Figure 3. Infrared spectra of the CoII(L)2Cl2 complex and films generated from the CoII complex.
Film 1 was generated from the CoII complex in the absence of acid medium while Film 2 was
generated in the presence of 10 mM ACA.
The CoII(L)2Cl2 IR spectrum shows the main vibrational modes which belong to the ligand
described as follows: 3401ꢀ3269 assigned to the ν(N–H) stretching vibrations, 3120 ν(C–Har.),
2944 ν(CH2), 1526ꢀ1446 ν(C=C + C=Nring), 1346 ν(C=S), 1245 ν(H3C–O), 1175 ν(C–C), 1035
ν(C–H) along with 770 and 580 cmꢀ1 (δ(C–H) and δ(C–C)) bending modes.47,57,58 The IR spectra
of the films (1 and 2) are found to be largely comparable to each other, indicating that the
methodology of preparation (in the presence or absence of acid) does not, in essence, affect the
material composition at least when it comes to the vibrational/structural characteristics. The
metalꢀorganic nature of the films can be observed by means of the infrared results in view of the
fact that the main vibrational modes of the organic fragments present in the CoII complex were
also found in the IR spectra of the films (this is in line with the metalꢀorganic nature of Film 2
11

Journal of Materials Chemistry A
Page 12 of 32
View Article Online
DOI: 10.1039/C8TA07340F
which has also been proved from the C, H, N elemental analysis, TG experiments, ICP OES and
EDX results). The IR spectra of the films presented broad bands at ca. 3250 up to 3600 cmꢀ1
which were assigned to the ν(O–H) stretching related to the presence of water molecules (also
found in the TG experiments). The sharp bands seen at ca. 3549, 3618 (Film 1) and 3620 cmꢀ1
(Film 2) suggest the presence of ν(N–H) stretchings that are possibly from amino (NH2) groups
which might be present in the film structures. The absence of ν(C–Har.) stretching that usually
occurs at ca. 3080 up to 3200 cmꢀ1 along with the backbone ν(C=C) stretching that usually occur
as sharp and intense bands at ca. 1600 up to 1450 cmꢀ1 and the disappearance of ν(H3C–O)
stretching at ca. 1245 cmꢀ1 (present in the complex and absent in the films) might be an
experimental evidence of the loss of aromatic rings in the film structures. On the other hand, the
presence of ν(CH2) stretchings at ca. 2967 ꢀ 2928, ν(C=S) at ca. 1306, ν(C–C) at ca. 1175, ν(C–
H) at ca. 1028 and the observed (δ(C–H) and δ(C–C)) bending modes at ca 764 and 525 cmꢀ1,
respectively, are experimental evidence of the metalꢀorganic nature of the films. A plausible
interpretation that could be inferred from these results is that some organic fragments and active
metallic site related to the electrochemical properties of the complex might also be present in the
structure of the produced films. It is worth pointing out that the films were found to present
remarkable electrocatalytic performance and higher electrochemical stability compared to the
CoII complex (this will be further explained in the electrochemical studies section of this paper).
3.1.4. XPS Analysis. Figure S4A (Supporting Information) depicts the XPS spectrum of Film
2 where Co 2s, Co 2p, O 1s, N 1s, C 1s, S 2s, S 2p, Co 3s, Co 3p, O 2s, and Co 3d peaks are
distinguishable. Positions, atomic and mass percentages (Table S8, Supporting Information)
were obtained from the XPS spectrum of Figure S4A.
The Film 2 surface analysis from the XPS measurement (Figure S4A, Supporting
Information) shows mass percentages of Co (32%), C (30.7%), O (31%), S (5.8%) and N
(0.26%) (Table S8, Supporting Information), which are in agreement to the ICP OES, TG, and
EDX results, in terms of enrichment of Co, as well as in terms of presence of S in the Film 2.
The deconvoluted highꢀresolution Co 2p XPS spectrum (Figure 4A) contains two peaks at
780.3 and 795.9 eV with their respective satellites shakeup at 784.4 and 801.7 eV corresponding
to the 2p3/2 and 2p1/2 levels respectively, in a 2:1 ratio. A spinꢀorbital splitting of 15.6 eV
between the peaks and the presence of satellites peaks indicate that CoII species36 are present in
12

Page 13 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
the Film 2. This is in agreement to the absence of electron diffraction pattern (Figure 2B),
reinforcing the occurrence of metalꢀorganic film in counterpoint to Co metal nanoparticles.
The deconvoluted S 2p XPS spectrum (Figure 4B) has been correlated to the threeꢀsulfur
chemical environment. The small peak at 162.0 eV was assigned to the S–Co group36 followed
by S=C group at 166.9 eV and S oxidized group35,37 at 168 eV (see Table S9, Supporting
Information, for % content). The presence of S oxidized should be related to the partial oxidation
of thiol groups when the Film 2 was exposed to the air.37 The presence of S in the Film 2 was
also evidenced by IR spectroscopy (Figure 3) which has been presented ν(C=S) stretchings at ca.
1306 cmꢀ1, as well as by the EDX mapping image (Figure 2D) and EDX results (Table S7,
Supporting Information).
Figure 4. Highꢀresolution Co 2p (A), S 2p (B), and N 1s (C) XPS spectra of Film 2.
The main component of the deconvoluted N 1s XPS spectrum (Figure 4C) can be attributed to
the N–H group (401.2 eV),35,36 followed by N–C group (398.5 eV),35,36 Nox group (408.1 eV),
and N–N group (402.8 eV) (see Table S9, Supporting Information, for % content). The
occurrence of N oxidized is related to the partial oxidation of N groups when the Film 2 is
exposed to the air. The presence of N–H groups was also observed in the IR measurements (ν(Nꢀ
H) stretchings at ca. 3620 cmꢀ1, Figure 3).
The deconvoluted C 1s XPS spectrum (Figure S4B, Supporting Information) should be related
to the treeꢀcarbon chemical environment. The main peak can be attributed to the C–C and C=C at
283.8 eV,59 followed by C–S, C–O groups at 284.7 eV,59 and C–N group at 287.5 eV. The
deconvoluted components in the O 1s XPS spectrum (Figure S4C, Supporting Information)
should be assigned to the O–C group (531.0 eV) and O2− group (530.2 eV). The highꢀresolution
13

Journal of Materials Chemistry A
Page 14 of 32
View Article Online
DOI: 10.1039/C8TA07340F
Cl 2p XPS spectrum (Figure S4D, Supporting Information) shows two peaks at 198.1 and 207.8
eV, corresponding to the 2p3/2 and 2p1/2 levels respectively, in a 1.45:1 ratio. The Co enrichment
as well as the presence of S and Cl were also determined by EDX mapping images (Figure 2D)
and EDX results (Table S7).
3.1.5. Electrochemical Characterization of the CoII(L)2Cl2 Complex, Films, and
Proposed Mechanism for the Electrocatalytic Hydrogen Production. Figure S5A (Supporting
Information) shows the CV response of the redox couple of FcCp2 (FcCp20/+) with ꢀEp/2 (ꢀEp =
Eap − Ecp) value of 0.3 V vs Ag wire. When the CoII(L)2Cl2 is added to the solution, the ꢀEp/2 is
displaced to more positive potential and a cathodic peak is visible at −1.44 V vs Ag wire
followed by a shoulder at more negative potential associated with the reduction of CoII(L)2Cl2.
The anodic peak at −0.04 V vs Ag wire is attributed to the oxidation of the products
(CoII(L)2Cl2 combined with FcCp2) formed during reduction of CoII(L)2Cl2 at more negative
potential than −1.12 V vs Ag wire, reflecting the ample irreversibility of this reduction. This
irreversibility along with the production of a blocking film was observed from the 4th cycle
onwards (Figure S5B, Supporting Information). We noted a decline of less than half of the
0/+
currents for the redox couple of FcCp2 and a complete change in the cathodic response for
CoII(L)2Cl2 as well as the complete disappearance of faradaic events in the 6th cycle. Considering
that the CV responses were affected in the simultaneous presence of CoII(L)2Cl2 and FcCp2, we
decided to reference the electrochemical potential for the redox couple (FcCp20/+) as a separate
system. Figure S6 (Supporting Information) is selfꢀexplanatory; it shows how the
electrochemical potential was referenced in this work. The small reduction peak at −0.95 V vs
0/+
FcCp2 (Figure S6B, Supporting Information) is attributed to the residual oxygen which is
0/+
irremovable from the solution, while the reduction peak at −1.88 V vs FcCp2 (Figure S6B,
Supporting Information) is related to the irreversible reduction of CoII(L)2Cl2. No visible
oxidation peaks are observed in the absence of FcCp2 (Figure S6B, Supporting Information). The
0/+
electrochemical behavior of FcCp2 does not undergo any changes after several cycles (not
shown).
As can be noted in Figure 5, no considerable faradaic current is seen for the bare GC and the
bare GC in the presence of CoCl2 or ligand (L). By contrast, for the CoII(L)2Cl2 complex, one
can observe a first reduction peak at −1.70 V and a discrete second peak at −1.85 V. These two
14

Page 15 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
peaks are related to the reduction of CoII to CoI and a possible electron transfer through the
metallic site to the organic moiety prompting the disruption of bonds in the ligand (loss of
organic fragments). Here, one should not rule out the production of organometallic species of
cobalt in lower oxidation states.9,13 This electrochemistry behavior is likely to be responsible for
the irreversibility of the CoII(L)2Cl2 electroreduction.
Figure 5. Cyclic voltammograms for bare GC and bare GC in the presence of 0.5 mM CoCl2, 0.5
mM ligand, 0.5 mM CoII(L)2Cl2, 0.5 mM Co(II)(L)2Cl2 + 0.5 mM ACA, and 0.5 mM CoII(L)2Cl2
+ 0.5 mM TFA in N2ꢀsaturated 0.1 M TBAClO4/ACN. Potential scan rate: 20 mV s–1. Potential
scan starts at −0.80 V (negative direction).
When CoII(L)2Cl2 is in the presence of acetic acid (ACA) in similar concentration, one clearly
notices the occurrence of two reduction peaks, which are catalytically defined (increased
currents) at −1.69 and −1.96 V respectively. Furthermore, the first reduction peak (−1.69 V)
exhibited more than twoꢀfold increase in (2.4) current value in relation to (−1.70 V) that of the
CoII(L)2Cl2 complex. The second reduction peak appears negatively displaced (approximately
100 mV) in relation to that of the CoII(L)2Cl2 complex, presenting an increased peak current
compared to the current of the first reduction peak.
When CoII(L)2Cl2 is in the presence of trifluoroacetic acid (TFA) at similar concentration, one
notices an anticipation of both reduction peaks for more positive potentials at −1.35 and −1.71 V
respectively. In addition, the first reduction peak (at −1.35 V) in the presence of TFA exhibits
15

Journal of Materials Chemistry A
Page 16 of 32
View Article Online
DOI: 10.1039/C8TA07340F
approximately twoꢀfold increase in (1.9) current value of the first reduction peak (at −1.70 V) for
the sole CoII(L)2Cl2. This result characterizes catalysis in terms of the potential and current
values.
Additional information was obtained following the hydrodynamic cyclic voltammetry
responses (Figure S7, Supporting Information) for the bare GC in the presence of CoII(L)2Cl2,
CoII(L)2Cl2 + ACA and CoII(L)2Cl2 + TFA. For CoII(L)2Cl2, a wave can be seen with increased
current at around −1.77 V, this coincides with the potentials related to the two peaks observed in
Figure 5. Following the addition of ACA into the solution containing CoII(L)2Cl2, a shoulder
(Figure S7, Supporting Information) appeared at around −1.64 V exhibiting a fourꢀfold increase
in the current value compared to that of CoII(L)2Cl2 in the absence of ACA, considering negative
potential scan direction. This behavior reinforces the catalytic effect of the CoII(L)2Cl2 on the
production of H2. The electroreduction of 0.5 mM of CoII(L)2Cl2 involves one electron, meaning
oneꢀfold increase in current value. The current continues to increase to more negative potentials.
When the scan is reversed (toward positive potential direction), the shoulder appears at more
positive potential (−1.56 V). Upon the addition of 0.5 mM of ACA, the electrode surface is
found to be “in need” of another electron to reduce one H+ and two electrons for the reduction of
both the complex and another H+ (one electron each). This results in a pair of electrons
equivalent in current to produce H2 and another pair of electrons equivalent in current to reduce
CoII(L)2Cl2) (see proposed mechanism for H2 production in Figure 6, which will be discussed
further in more detail).
However, upon the addition of TFA (Figure S7, Supporting Information), the catalytic effect
with respect to H2 production, in terms of potential, is found to be sufficiently pronounced with a
wave exhibiting an increased current at around −1.42 V. The wave at −1.42 V approximately
coincides with the potential related to the first reduction peak current depicted in Figure 5 and
with a visible peak coincident with the second reduction peak, also observable in Figure 5, in the
negative potential direction. In the reversed scan (positive potential direction), the currents that
appear on the wave are slightly smaller (without a peak) and are seen to be closer to the wave
currents of CoII(L)2Cl2 in the presence of ACA, indicating at least 4 electrons involved in the
reduction (see reasoning in the previous paragraph).
16

Page 17 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
Figure 6. Proposed mechanism for CoII(L)2Cl2 reduction coupled with hydrogen production.
For both acids, the increase in concentration up to 10 mM (Figure S8, Supporting
Information) results in the observance of a peak at −1.75 and −1.51 V, respectively, for ACA and
TFA at a concentration of 10 mM of both acids, with a more evident preꢀshoulder for TFA at
−1.32 V. The catalytic current (cathodic peak current) dependence is that of a second
order4,7,8,21,24,28,30 with respect to the ACA and TFA concentrations (see inset plots in Figure S8,
17

Journal of Materials Chemistry A
Page 18 of 32
View Article Online
DOI: 10.1039/C8TA07340F
Supporting Information), based on equation (1)3,13 (linear behavior of Icat vs. ACA or TFA
concentrations intercepting the origin, meaning Icat being proportional to ACA or TFA
concentration). A first–order dependence21 is observed relative to CoII(L)2Cl2 concentration (see
Figure S9, Supporting Informaiton):
ꢀꢁꢂꢃ = nFA[cat] ꢅ(ꢆ[ꢇꢈ]ꢉ)
(1)
ꢄ
where Icat is the catalytic current, n is the number of electrons involved in the catalytic reaction, F
is the Faraday constant, A is the electrode area, [cat] is the catalyst concentration, D is the
diffusion coefficient, k is the rate constant, and x being the order of the reaction related to the
acid.3,13
When CoII(L)2Cl2 (0.5 mM) is in the presence of 10 mM of ACA or TFA (see Figure S10,
Supporting Information), the current values are found to increase by nearly ten times in
comparison to the current results shown in Figure S7 (Supporting Information). The
electrochemical behavior shown in Figure S10 (Supporting Information) presents resemblance to
that shown in Figure S7 (Supporting Information), the exception being the current values.
Moreover, in the positive potential scan direction, the catalytic effect of TFA was revealed
through the onset potential value occurring at −0.88 V in comparison with the lower value of
−1.12 V shown in Figure S7. The onset potential for bare GC in the presence of 10 mM TFA
occurred at −1.32 V and the currents are at least 2.2 times lower in comparison to the CoII(L)2Cl2
(0.5 mM) in the presence of 10 mM of ACA or TFA (Figure S10, Supporting Information).
Figure S11 (Supporting Information) presents the cyclic voltammetry responses for
CoII(L)2Cl2 (0.5 mM) in the presence of increased amounts of ACA and TFA (larger than 10
mM). For the increase in ACA concentration (see Figure S11A, Supporting Information), one
will notice the presence of a main peak which has previously been observed in Figure S8A
(Supporting Information). This main peak becomes invisible when it draws close to the
concentration of 200 mM, with Icat being linear with ACA concentration up to the value close to
the aforementioned concentration (200 mM) and intercepting the origin (see inset of Figure
S11A, Supporting Information). In the case of the increase in TFA concentration (Figure S11B,
Supporting Information), one can clearly observe the development of a preꢀshoulder at around
−1.30 V as a peak; this peak is followed by the increase in TFA concentration (linear behavior is
18

Page 19 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
noted up to concentration value close to 240 mM, see inset of Figure S11B, Supporting
Information). The second peak does not appear after 200 mM of TFA concentration at the
potential range. The linear behavior of Icat vs. ACA and TFA concentrations which passes
through the origin reinforces the secondꢀorder dependence reaction4,7,8,21,24,28,30 (see inset plots in
Figure S11, Supporting Information).
Following the concentration of 200 mM ACA and 240 mM TFA (Figure S11, Supporting
Information) the reaction becomes pseudoꢀzeroꢀorder in relation to ACA and TFA
concentrations,3 assuming that two electrons (n = 2) are required for the production of H2. This
pseudoꢀzeroꢀorder situation to which few changes were observed in the Icat (few changes were
seen in Icat with an increase in acidic concentration during the electrochemical experiment)
allowed us to calculate the catalytic rate constant (kobs, eq. 2), or yet the turnover frequency
(TOF) eq. 3, simplified to eq. 41,3,4,7,8,11ꢀ13,21,25,60ꢀ64
.
ꢆꢊꢋꢌ = ꢆ[ꢇꢈ]ꢉ
(2)
(3)
ꢍ
ꢎꢏꢐ
ꢒ
ꢘꢙꢚ
ꢛꢜꢝ
ꢗ
ꢞꢟ
=
ꢍ
ꢓ.ꢔꢔꢕꢖ
ꢑ
ꢧ
ꢆꢊꢋꢌ = 1.94ꢠꢡꢢ ∙ ꢣ ꢤꢍꢎꢍꢏꢐꢦ
(4)
ꢥ
where Ip is the peak current at the potential of the CoII/I couple in the absence of acid, R is the gas
constant, T is the temperature in K, and ν being the potential scan rate at V s−1.
Based on the appropriate values taken from Figure 5 for CoII(L)2Cl2 Ip and CoII(L)2Cl2 + ACA
(or CoII(L)2Cl2 + TFA) Icat from Figures S11 and S12 (CV curves with concentration of 200 mM
ACA and 240 mM TFA, see Supporting Information), we were able to find the maximum TOF
value for H2 production as being 1900 and 4237.4 s−1, respectively, which are higher values
when compared with several values reported in the literature (Table S10, Supporting
Information). The value obtained is 2.2 times more catalytic for CoII(L)2Cl2 in the presence of
TFA compared to ACA.
The Ecat (catalytic potential) was determined using the CV curves with concentrations of 200
mM ACA and 240 mM TFA in the presence of 0.5 mM CoII(L)2Cl2 (Figure S12, Supporting
Information), which are related to Icat obtained in acid concentrations where the first current
plateau is seen on the catalytic waves (insets of Figure S11, Supporting Information).
The EACA0 and ETFA0 were calculated using eq. 5 below:65,66
19

Journal of Materials Chemistry A
Page 20 of 32
View Article Online
DOI: 10.1039/C8TA07340F
ꢨꢩꢓꢪꢩ ꢊꢫ ꢙꢞꢩ = ꢨꢬꢓ /ꢬ − ꢯꢧ.ꢖꢓꢖꢘꢙꢰ ∙ ꢱꢲꢳ(ꢴꢵꢴ ꢶꢷ ꢸꢹꢴ)
(5)
ꢭ
ꢮ
ꢞ
where ꢨꢬꢓ /ꢬ value was assumed to be −0.028 V (vs. FeCp2+/0).65,66 Here, the pKa values in ACN
ꢭ
ꢮ
were 12.65 for TFA and 23.51 for ACA,66 which yielded the values of −0.776 V51 for ꢨꢙꢓꢞꢩ and
+/0
−1.419 V51 (vs. FeCp2 ) for ꢨꢩꢓꢪꢩ
.
The overpotential (eq. 6)12,21,22 is thus obtained (see Supporting Information for comments
about usefulness of eq. 5 as well as the determination of overpotential (eq. 6)):
ꢺꢣꢻꢷꢱꢶꢼꢻꢽꢼꢾꢿꣀ = ꢨꢩꢓꢪꢩ ꢊꢫ ꢙꢞꢩ − ꢨꣁꢳꣂ/ꢧ
(6)
The overpotential values were found to be 230 and 600 mV for CoII(L)2Cl2 in the presence of
ACA and TFA, respectively (see Figure S12, Supporting Information), which are in general
lower values than those reported in the literature (Table S10, Supporting Information).
The information described above can be attributed to the initial reduction of CoII(L)2Cl2
(involving one irreversible electron transfer) to CoI(L)2Cl2, followed by an extremely fast
protonation of CoI(L)2Cl2, involving proton migration which is in turn followed by the fast
regeneration of NH3 species (this protonation gives rise to HCoIII(L)2Cl2, depicted in Figure 6).
+
Subsequently, a second electron transfer combined with proton delivery occurs followed by the
+
fast regeneration of NH3 species (Figure 6) which leads to hydrogen production, regenerating
CoII catalyst.23ꢀ25 The proposed mechanism for the aforementioned catalytic cycle is EiCiCiEi
(Figure 6). One should note, however, that neither ECEC nor EECC can be ruled out here.30
As mentioned previously, the electrochemical irreversibility behavior of CoII(L)2Cl2 produces
a blocking film (Figure S5B, Supporting Information) when a continuous cyclic voltammetry
experiment is carried out (Figure S13, Supporting Information). It is worth noting that the small
0/+
reduction peak at −1.02 V vs FcCp2 is attributed to some residual oxygen which has not been
0/+
removed from the solution while the reduction peak at −1.92 V vs FcCp2 is related to the
irreversible reduction of CoII(L)2Cl2 observed in the first cycle. The irreversible reduction of
CoII(L)2Cl2 completely disappeared after the 5th cycle (Figure S13, Supporting Information) and
the global currents totally decreased at the 10th cycle (Figure S13, Supporting Information).
The GC surface modified with the metalꢀorganic film produced from CoII(L)2Cl2 reduction
(GC/Film 1) was placed in contact with increased concentrations of TFA (see Figures 7 and S14,
Supporting Information).
20

Page 21 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
Figure 7. (A) Cyclic voltammograms for bare GC and GC modified with Film 1 (GC/Film 1) in
the presence of different concentrations of TFA. Inset Figure 7A: Cyclic voltammogram for
GC/Film 1 electrode in N2ꢀsaturated 0.1 M TBAClO4/ACN. (B) Cathodic catalytic peak currents
for GC/Film 1 electrode and (C) normalized cathodic catalytic peak currents per cathodic current
0/+
at −1.7 V vs FcCp2 for GC/Film 1 electrode in N2ꢀsaturated 0.1 M TBAClO4/CAN (inset
Figure 7A) as a function of TFA concentration. All experiments were carried out in N2ꢀsaturated
0.1 M TBAClO4/ACN. Potential scan rate: 20 mV s–1. Potential scans start at −0.50 V in
negative direction.
When the GC/Film 1 electrode is placed in contact with increased concentration of TFA up to
10 mM (Figure S14, Supporting Information), it exhibits a peak at −1.70 V with an evident preꢀ
shoulder at −1.42 V, which appeared at more negative potentials in comparison with 0.5 mM
CoII(L)2Cl2 in similar solution (Figure S8B, Supporting Information). This is, at first thought,
suggestive (that GC/Film 1 is less catalytic compared to the complex CoII(L)2Cl2 in similar
solution) of the presence of a less catalytic surface for H2 production. The catalytic current
dependence (cathodic peak current) is second order4,7,8,21,24,28,30 relative to different
concentrations of TFA (see inset plot in Figure S14, Supporting Information), according to
equation (1),3,13 (linear behavior of Icat vs. TFA concentration which passes through the origin,
meaning Icat proportional to the TFA concentration), as observed for 0.5 mM CoII(L)2Cl2 in
similar solution.
Figure 7A shows the CV responses for GC/Film 1 electrode in the presence of increased
amounts of TFA (greater than 10 mM). Clearly, this indicates the development of a preꢀshoulder
at around −1.46 V as a peak, which is followed by an increase in TFA concentration (linear
21

Journal of Materials Chemistry A
Page 22 of 32
View Article Online
DOI: 10.1039/C8TA07340F
behavior up to nearly 240 mM, Figure 7B, intercepting the origin). For purposes of comparison
between the currents for bare GC and GC/Film 1 electrode let us consider the concentration of
240 mM for TFA. One will notice a positive overpotential of around 300 mV for the catalytic
effect in H2 production for GC/Film 1 electrode (Figure 7A). The second peak does not appear
after the addition of 100 mM TFA at the potential range investigated. The linear behavior of Icat
vs. TFA concentrations intercepting the origin reinforces the second order reaction4,7,8,21,24,28,30
(Icat being proportional to TFA concentration, Figure 7B). These results suggest that the
CoII(L)2Cl2 complex and GC/Film 1 electrode display a comparable performance when it comes
to electrocatalytic hydrogen production in the presence of TFA. Hence, the elements that
compose the active catalytic region in both structures should be common.
After reaching the concentration of 240 mM TFA (Figure 7B), the reaction becomes pseudoꢀ
zeroꢀorder in relation to TFA concentration,3 as observed previously to 0.5 mM CoII(L)2Cl2 in
similar solution. The assumption that two electrons (n = 2) are needed to produce H2, a situation
in which the Icat changes very little (scanty changes are noted in Icat given an increase in the
acidic concentration during the electrochemical experiment) enabled us to calculate the catalytic
rate constant (kobs, eq. 2), or yet the turnover frequency (TOF) eq. 3, simplified to eq. 4.1,3,4,7,8,11ꢀ
13,21,25,60ꢀ64
Through the use of the appropriate values from the inset of Figure 7A for GC/Film 1
0/+
electrode I at −1.7 V vs FcCp2 and GC/Film 1 electrode + TFA Icat from Figures 8AꢀC (CV
curve with concentration of 240 mM TFA, Figure 7A), we were able to find the maximum TOF
value for H2 production as being 1.49 × 106 s−1. This value is 351.6 times more catalytic for
GC/Film 1 electrode in the presence of TFA compared to 0.5 mM CoII(L)2Cl2 in similar solution.
It is worth pointing out that the maximum TOF value obtained here (1.49 × 106 s−1 ) is 8.8 times
better than that reported in Ref. 1, which is considered one of the best TOF values reported in the
literature. This, in effect, reveals the remarkable nature of the TOF value obtained for our
GC/Film 1 electrode.
The overpotential was found to be 470 mV for the GC/Film 1 electrode in the presence of
TFA (Figure S12, Supporting Information); this value is 130 mV lower compared to the
overpotential of CoII(L)2Cl2 in the presence of TFA, (Figure S12, Supporting Information),
coincident with the anticipated value of the onset potential (130 mV) for H2 catalytic production.
This overpotential value obtained for the GC/Film 1 electrode is in considerable agreement with
the range of overpotential values reported in the literature (Table S10, Supporting Information).
22

Page 23 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
The improvement observed in TOF and in the overpotential for the GC/Film 1 electrode in
comparison with CoII(L)2Cl2, both in the presence of TFA, can be attributed to the easier charge
transfer from the GC surface to the catalytic region of the metalꢀorganic film, which contains
common elements as Co, S, N, C and Cl in the electroactive site of both structures, in the light of
the IR, TG, ICP OES, XPS, HAADFꢀSTEM, EDX mapping and elemental analysis results.
For a better understanding of the production of the film which was electrochemically
generated on the GC surface, some studies were conducted when this material is produced in a
chosen acid concentration of 10 mM. Initially, the Figure S15 (Supporting Information) shows
the CVs up to the 20th cycle during the deposition of Film 1, Film 2, and Film 3, and GC for
comparison purposes.
Here, one will notice the presence of extremely small currents for the production of Film 1
(inset of Figure 7A and Figure S13), which are greater than those of the bare GC (Figure S15,
Supporting Information)). However, for the GC in the presence of 0.5 mM CoII(L)2Cl2 and 10
mM ACA or 10 mM TFA, the currents are found to be very high, presenting a peak current at
0
−1.6 V and −1.28 V vs FcCp2 for Films 2 and 3, respectively. The currents appear at more
positive potential than that depicted in Figure S6. The peak current reduction observed in Figure
S15 takes place at a potential range comparable to that of the current shoulder of the first
voltammetric cycle for the 0.5 mM of CoII(L)2Cl2 in the presence of 10 mM TFA or ACA (see
Figure S8, Supporting Information). Thus, when the films were produced from 0.5 mM of
CoII(L)2Cl2 in the presence of 10 mM ACA or TFA, one will notice that the system became more
catalytic compared to the first cycle apart from being more catalytic as well than the first cycle of
the GC/Film 1 modified electrode (Figure S14, Supporting Information).
When the films are developed using CoII(L)2Cl2 in the presence of ACA or TFA, we observe
that the continuous catalytic H2 production at the 20th cycle occurs simultaneously with
CoII(L)2Cl2 reduction (high currents are observed, see Figure S15). Conversely, when the films
are produced from CoII(L)2Cl2 in the absence of acid, the faradaic currents are negligible from
the 6th cycle onwards due to the production of a blocking film on the GC surface (Figure S13,
Supporting Information). During the production of Film 3, an irreversible oxidation (positive
currents between −0.69 up to −0.50 V) is observed. This oxidation is likely to be related to the
products from the irreversible reduction of the CoII(L)2Cl2 complex (that escapes to the solution),
and/or the oxidation of Film 3 or perhaps the oxidation of the produced H2.
23

Journal of Materials Chemistry A
Page 24 of 32
View Article Online
DOI: 10.1039/C8TA07340F
Figure 8 shows the CVs for bare GC, GC/Film 1, GC/Film 2, and GC/Film 3 electrodes in the
presence of 10 mM ACA or TFA.
Figure 8. Cyclic voltammograms for bare GC, GC/Film 1, GC/Film 2, and GC/Film 3 electrodes
in the presence of 10 mM ACA or TFA. All experiments were performed in N2ꢀsaturated 0.1 M
TBAClO4/ACN. Potential scan rate: 20 mV s–1. Potential scans begin at −0.80 or −0.50 V in
negative direction.
It is clear that all the modified electrodes with the metalꢀorganic film catalyze the H2
production in a more effective manner than the bare GC in the presence of 10 mM TFA, even
when compared with the films in the presence of 10 mM ACA. The films produced in the
presence of the respective acids (Figure 8) are relatively more catalytic (presented more positive
potentials for the two peaks/shoulders reductions, at least 50 mV more positive) than those
produced in the absence of acids (Figure 8 and Figure S15, Supporting Information). Films 2 and
3 are less catalytic (with regard to the potential) than the 0.5 mM CoII(L)2Cl2 in the presence of
10 mM of ACA or TFA, since the peaks/shoulders for the GC/Film 2 and GC/Film 3 electrodes
are at more negative potentials (negatively shifted at least 80 mV) (Figure 8) in comparison with
the peaks/shoulders of 0.5 mM CoII(L)2Cl2) in the presence of 10 mM of ACA or TFA (Figure
S8). In this sense, the overall catalytic behavior exhibited by the films and the CoII(L)2Cl2)
complex in acid medium is comparable to each other (with regard to the potential and
voltammetric profile).
24

Page 25 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
For the purpose of studying the electrochemical stability of the films playing the role of
electrocalysts in H2 production, Figure 9 shows the CV responses presented by the metalꢀorganic
films during the catalytic H2 evolution over time.
Figure 9. Cyclic voltammograms for GC/Film 1 and GC/Film 2 electrodes in the presence of 10
mM TFA (A) and ACA (B), respectively. All experiments were performed in N2ꢀsaturated 0.1 M
TBAClO4/ACN. Potential scan rate: 20 mV s–1. Potential scans begin at −0.80 or −0.50 V in
negative direction.
It can be seen from Figure 9, that the onset potential became relatively more positive while
the current peak underwent a decrease after 1000 cycles at 20 mV s–1 for the Film 1 in presence
of TFA (Figure 9A). This current peak decrease is attributed to the concentration lowering of
TFA (near to zero) as confirmed by UV analysis of the solution after 1000 cycles at 20 mV s–1
(see Figure S16, Supporting Information). The result of onset potential shows that the catalytic
activity for H2 production has been enhanced over time. The persistent high catalytic
performance of the Film 1 was also demonstrated during the experiment performed and depicted
in Figure 7 where the film went through several TFA additions but yet maintained
electrocatalysis for the production of H2.
For providing experimental evidence on the H2 evolution, chronoamperometric measurements
were performed (Figure S17, Supporting Information) which revealed negligible currents for
bare GC and bare GC in presence of CoII(L)2Cl2 or bare CG in presence of sodium
25

Journal of Materials Chemistry A
Page 26 of 32
View Article Online
DOI: 10.1039/C8TA07340F
trifluoroacetate (240 mM) (Figure S18AꢀB, Supporting Information, H2 bubbles were not
observed on the GC surface). However, proceeding with the addition of TFA 240 mM (Figure
S17, Supporting Information) the currents are enough decreased for bare GC (charge equal 13 C
meaning 6.72 × 10−5 H2 mol from Faraday’s law (nmol = Q (charge in C)/z+ . F (z+ = moles of
electrons involved in the reaction)) presenting H2 bubbles totally visible (Figure S18CꢀD,
Supporting Information). These bubbles are responsible for the variation on the currents, which
increase when the bubbles number is enhanced and decrease when the bubbles are removed from
the electrode surface throughout the unsettling the cell. For the GC modified with Film 1, the
currents were also decreased (Figure S17, Supporting Information, Q = 17.7 C meaning 8.13 ×
10−5 H2 mol also presenting H2 bubbles totally visible (Figure S18EꢀF, Supporting Information)).
As well as for the GC in presence of CoII(L)2Cl2, the currents were also decreased (Figure S17,
Supporting Information, Q = 21.4 C meaning 11.07 × 10−5 H2 mol and presenting H2 bubbles
totally visible (Figure S18GꢀH, Supporting Information)).
The Faraday efficiency towards H2 evolution was underestimated by considering the volume
(and/or the halfꢀvolume) of the visible bubbles H2 sphere’s which evolve during the experiments
Figure S18, (see Supporting Information for more details about these calculations), which
resulted 4.3, 5.6, and 3.7 % for bare GC, GC modified with Film 1, and GC in presence of
CoII(L)2Cl2 respectively. These underestimated Faraday efficiency for H2 evolution should be
related to the fact that we are considering in the calculations only the visible bubbles on the
electrode surface, naturally it is expected that innumerable and invisible bubbles or nanobubbles
should be escaping from the electrode surface to the solution before agglomerating to produce
visible bubbles. However, considering the underestimated values obtained for the Faraday
efficiency it is possible to identify that the H2 production for GC modified with Film 1 and also
for the GC in presence of CoII(L)2Cl2 is more effective in comparison with bare GC.
Importantly, the Faraday efficiency for H2 evolution contribute to confirm the found TOF
values once chronoamperometric currents that appeared for the sodium trifluoroacetate
experiments are negligible, meaning that the observed currents are not related to the reduction of
trifluoroacetate but related to the H2 evolution from TFA.
Based on the calculated charges of the chronoamperometric results (Figure S17, Supporting
Information) it is possible to attribute for GC modified with Film 1 and GC in presence of
CoII(L)2Cl2 electrodes, both in presence of TFA 240 mM, an improved Faraday efficiency for H2
26

Page 27 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
evolution of 21 and 65% respectively in comparison with bare GC in presence of TFA 240 mM.
Applying these improved percentages to the previously determined TOFs (1.49 × 106 and 4237.4
sꢀ1, respectively) it is possible to correct these TOFs values to 312 900 and 2754.3 sꢀ1
respectively, which for the GC electrode modified with Film 1 results in a TOF of 1.84 times
higher than the best TOF reported in Ref. [1].
In addition, the GC modified with Film 2 presented best performance for HER in comparison
to the bare GC, both in presence of 0.5 M H2SO4 (Figure S20, Supporting Information), showing
an overpotential of 606 mV vs RHE at 10 mA cm−2geo in 0.5 M H2SO4 (Figure S20B, Supporting
Information), value comparable to those shown in Table S10, Supporting Information. For the
graphite (2 cm2 in geometric area) modified with Film 2 (Figure S20C, Supporting Information)
operating at overpotential of 606 mV vs RHE and after 450 cycles between 0 and −1.0 V vs RHE
(50 hours of potential polarization), the density current was increased in 1.9 times in relation to
before the 450 cycles, as well as to a chronoamperometric experiment at −700 mV vs RHE
during 50 hours (inset of Figure S20D, Supporting Information), the density current was
increased over time (in average 8.3 mA cm−2 in the end of each chronoamperometric stop, inset
of Figure S20D, Supporting Information) and also in 1.23 times in relation to before the 50 hours
of chronoamperometric experiment (Figure S20D, Supporting Information) in an overpotential
of 700 mV vs RHE, evidencing clearly the strong stability of Film 2 even in a harsh medium and
electrochemical condition.
4. CONCLUSION
A novel, cheap (U$ 0.40 per gram) and simple CoII(L)2Cl2 complex was synthesized and
applied as catalyst and preꢀcatalyst in the production of metalꢀorganic electroactive films. Both
the complex and the films exhibited high performance toward H2 evolution. The CoII(L)2Cl2
complex and metalꢀorganic films derived from the CoII complex presented comparable catalytic
behavior (with respect to the potential and voltammetric profile). This behavior is thought to be
related to the presence of common elements as Co, S, N, C and Cl in the electroactive site of
both structures. We believe that the elements which compose the catalytic region of CoII(L)2Cl2
(Co active metallic site and two available amino (NH2) groups in the second coordination sphere
working as an acceptor and deliverer of protons to the metallic atom during the electrocatalytic
27

Journal of Materials Chemistry A
Page 28 of 32
View Article Online
DOI: 10.1039/C8TA07340F
hydrogen production) is also likely to be present in the metalꢀorganic films structure given that
the results obtained from IR, TG, ICP OES, XPS, HAADFꢀSTEM, EDX mapping and Elemental
analysis along with those of the electrochemical experiments are in line with this hypothesis. To
the best of our knowledge, the metalꢀorganic films produced in this work exhibited the best TOF
value reported in the literature to date relative to H2 production. The TOF value obtained in this
investigation (312 900 sꢀ1, corrected by Faraday efficiency) is 1.84 times higher than the TOF
value reported previously in the literature. The overpotential value obtained (470 mV) for the
GC/Film 1 electrode in the presence of TFA during hydrogen production is 130 mV lower than
that of CoII(L)2Cl2 in the presence of TFA and 240 mV higher than that of CoII(L)2Cl2 in the
presence of ACA. The electroreduction mechanism for the CoII(L)2Cl2 in the presence of acid
involves an irreversible electron transfer followed by an extremely fast protonation of the active
region which gives rise to HCoIII(L)2Cl2 intermediate. This intermediate is generated from proton
+
migration followed by the fast regeneration of NH3 species from the protonation of the protonꢀ
accepting NH2 groupements. Furthermore, the second electron transfer combined with the proton
+
delivery and followed by the fast regeneration of NH3 species leads to the production of
hydrogen and the regeneration of the catalytic region of CoII(L)2Cl2. The persistent high catalytic
performance of the films was demonstrated during the H2 production. The GC and graphite
modified with Film 2 have presented better performance toward HER in comparison to the bare
GC and graphite, both in the presence of 0.5 M H2SO4.
ASSOCIATED CONTENT
Supporting Information
Electronic Supplementary Information (EIS) available: The Supporting Information contains
crystallographic data (CCDC 1581490), figures, images, equations, and tables addressing
supplementary results, and references (PDF).
AUTHOR INFORMATION
Corresponding Authors
*Eꢀmail for G.A.C.: gleison.casagrande@ufms.br
*Eꢀmail for G.M.: gilberto.maia@ufms.br
ORCID
Leticia S. Bezerra: 0000ꢀ0003ꢀ2077ꢀ2123
28

Page 29 of 32
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C8TA07340F
Guilherme V. Fortunato: 0000ꢀ0002ꢀ0768ꢀ4156
Lucas Pizzuti: 0000ꢀ0002ꢀ3950ꢀ0972
Gleison A. Casagrande: 0000ꢀ0002ꢀ1667ꢀ913x
Gilberto Maia: 0000ꢀ0003ꢀ2449ꢀ0887
Notes
The authors declare no competing financial interest.
ACKNOWLEDGEMENTS
The authors G.M. and G.A.C wish to express their gratitude and indebtedness to LMEꢀIQSCꢀ
USPꢀSão Carlos for providing the TEM facilities, to XꢀRay Photoelectron Spectroscopy
Laboratory from IFSCꢀUSPꢀSão Carlos for providing the XPS facilities, Xꢀray singleꢀcrystal
collection data (V.M.D), and to CNPq (grants 301403/2011ꢀ2, 473991/2012ꢀ8, 405695/2013ꢀ6,
303759/2014ꢀ3, and 442268/2014ꢀ9) and FundectꢀMS (grants 23/200.583/2012,
23/200.735/2012, 23/200.246/2014, and 59/300.184/2016) for their financial support. L.S.B,
P.P.R, and G.V.F are grateful to CNPQ and CAPES for the fellowships granted during the course
of this research.
REFERENCES
(1) R. M. Bullock, A. M. Appel and M. L. Helm, Chem. Commun., 2014, 50, 3125–3143.
(2) T. N. Huan, R. T. Jane, A. Benayad, L. Guetaz, P. D. Tran and V. Artero, Energy Environ.
Sci., 2016, 9, 940–947.
(3) S. Wiese, U. J. Kilgore, M.ꢀH. Ho, S. Raugei, D. L. DuBois, R. M. Bullock and M. L. Helm,
ACS Catal., 2013, 3, 2527–2535.
(4) V. S. Thoi, Y. Sun, J. R. Long and C. J. Chang, Chem. Soc. Rev., 2013, 42, 2388–2400.
(5) A. Das, Z. Han, W. W. Brennessel, P. L. Holland and R. Eisenberg, ACS Catal., 2015, 5,
1397–1406.
(6) S. Cherdo, S. El Ghachtouli, M. Sircoglou, F. Brisset, M. Oriob and A. Aukauloo, Chem.
Commun., 2014, 50, 13514–13516.
(7) L. Gan, T. L. Groy, P. Tarakeshwar, S. K. S. Mazinani, J. Shearer, V. Mujica and A. K.
Jones, J. Am. Chem. Soc., 2015, 137, 1109–1115.
(8) M. L. Helm, M. P. Stewart, R. M. Bullock, M. R. DuBois and D. L. DuBois, Science, 2011,
333, 863–866.
(9) B. D. Mccarthy, C. L. Donley and J. L. Dempsey, Chem. Sci., 2015, 6, 2827–2834.
(10) S. Roy, M. Bacchi, G. Berggren and V. Artero, ChemSusChem, 2015, 8, 3632–3638.
(11) U. J. Kilgore, J. A. S. Roberts, D. H. Pool, A. M. Appel, M. P. Stewart, M. R. DuBois, W.
G. Dougherty, W. S. Kassel, R. M. Bullock and D. L. DuBois, J. Am. Chem. Soc., 2011, 133,
5861–5872.
(12) H. J. S. Brown, S. Wiese, J. A. S. Roberts, R. M. Bullock and M. L. Helm, ACS Catal.,
29