Synthesis and characterization of macroporous Ni, Co and Ni–Co electrocatalytic deposits for hydrogen evolution reaction in alkaline media

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Synthesis and characterization of macroporous Ni, Co and NieCo electrocatalytic deposits for hydrogen evolution reaction in alkaline media C. Gonza´lez-Buch, I. Herraiz-Cardona, E. Ortega, J. Garcı´a-Anto´n, V. Pe´rez-Herranz* Ingenierı´a Electroquı´mica y Corrosio´n (IEC), Departamento de Ingenierı´a Quı´mica y Nuclear, Universitat Polite`cnica de Vale`ncia, Camino de Vera, s/n. 46022 Valencia, Spain

article info

abstract

Article history:

In this work, macroporous Ni, Co and NieCo electrodes have been developed by co-

Received 28 December 2012

deposition at high current density on stainless steel (AISI 304) substrates. The obtained

Received in revised form

materials were characterized both morphologically and chemically by confocal laser

7 May 2013

scanning microscopy, and SEM coupled with EDX analysis. The activity for hydrogen

Accepted 5 June 2013

evolution reaction (HER) on the obtained layers was assessed by using pseudo-steady-state

Available online 11 July 2013

polarization curves and electrochemical impedance spectroscopy (EIS) in alkaline solution (30 wt.% KOH). The electrochemical results show that HER on these electrodes takes place

Keywords:

by the VolmereHeyrovsky mechanism. The synthesized coatings present higher catalytic

Porous electrodes

activity for HER than commercial smooth Ni electrode. As the Co content increases in the

Nickelecobalt alloys

electrodeposition bath the obtained structures show lower surface roughness factors. Ni

HER

eCo deposit with a Co content of 43 at.% manifests the highest intrinsic activity for HER as

Roughness factor

a consequence of the synergetic combination of Ni and Co. Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

Current global energy system is mainly based on fossil fuels: oil, natural gas and coal [1]. In fact, 81% of the primary world energy production comes from these fuels, originating serious problems, which will grow with time [1]. The most important drawbacks of this system are: (1) environmental problems derived from the greenhouse effect gasses emission; and (2) fossil fuels reserves are finite. Hydrogen is one of the most promising candidates as an energy carrier, and it can be positioned as the main alternative to fossil fuels to reduce CO2 emissions. It is a versatile, efficient, and clean fuel [1e3]. Most of the methods used for H2

production are based on fossil fuels, due to their easy usage in present designed machines and low costs [4e6]. Nevertheless, it is not consistent with the policies on the way to a green energy system. On the other hand, H2 generation from alkaline water electrolysis has a lot of possibilities of coupling renewable energy sources, minimizing pollution emissions [7]. However, this technique is not widely used in global H2 generation due to its low energy efficiency (high operating costs) [6,8]. This problem may be solved by the development of inexpensive electrode materials with low HER overpotentials, i.e. with good electrocatalytic activity. The increase in the electrode catalytic activity can be carried out by enlarging its real surface area and/or its intrinsic activity [9]. In this way, Ni,

* Corresponding author. Tel.: þ34 96 3877632; fax: þ34 96 3877639. E-mail address: [email protected] (V. Pe´rez-Herranz). 0360-3199/$ e see front matter Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2013.06.016

10158

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

and Ni-based compounds and alloys are the most important and studied electrode materials, which respond to their high catalytic activity and stability at low cost [10e12]. Different techniques have been used to enlarge the electrode real surface area, being one of them the electrodeposition at large current densities [13e16]. Alternatively, Ni intrinsic activity has been improved by means of the combination of Ni with metals such as: NiCo [17e19], NiFe [20e22], NiMo [20,23,24], NiW [20,25,26], NiLa [27,28]. The objective of the present research is the development of macroporous Ni and NieCo coatings deposited on stainless steel (AISI 304) at high current density from Ni chloride baths modified with different cobalt amounts. The influence of Co incorporation in the Ni porous matrix was tested by means of SEM, confocal scanning microscopy and EDX. The activity towards hydrogen evolution reaction (HER) was assessed by pseudo-steady-state polarization curves and electrochemical impedance spectroscopy (EIS) in alkaline media.

Table 1 e Synthesis operating conditions, bath composition and nomenclature of employed the developed macroporous Ni, Co and NiCo electrodes. Ni electrode Nomenclature Ni

Experimental

2.1.

Preparation of electrodes

The electrocatalytic layers were deposited onto AISI 304 stainless steel disc electrodes (0.5 cm2 geometric area). These substrate materials were set up with the pre-treatment process described in our previous works [15e17,19], as an initial step. A porous nickel electrode (Ni electrode) was galvanostatically synthesized at a current density of 1 A cm2, in a bath composed of NiCl2$6(H2O) and NH4Cl (Ni electrode). Nickelecobalt electrodes were obtained under the same operating conditions by adding different amounts of CoCl2$6(H2O), diluted in the minimum distilled water volume (ca. 1 mL), to the electrodeposition bath. The cobalt salt was not initially added to the electrodeposition bath in order to avoid the Co occlusion down to a Ni layer, as a consequence of the anomalous co-deposition of NiCo alloys [29]. Both the synthesis operating conditions and the electrode nomenclature used in the following are collected in Table 1. Note that the parameter tCo in Table 1 is the time during the electrodeposition process at which Co is added to bath. Cobalt electrode was fabricated in two steps, firstly a nickel layer was electrodeposited onto AISI 304 stainless steel with the same electrodeposition bath that the Ni electrode during 45 min and then a Co layer is obtained in a bath composed of CoCl2$6H2O and NH4Cl. Chemical grade reagents used for electrolyte preparation with distilled water were not subjected to a supplementary purification. Ni, Co and/or NieCo electrodeposition at high current densities takes place simultaneously to the gas bubbling, i.e. gas bubbles act as a dynamic template [13,15,16], following the schema shown in Fig. 1. Electrodepositions were accomplished in a thermostated one-compartment cell with the substrate surface to be coated in horizontal “face-up” position (see Fig. 2), allowing the free departure of the generated gas bubbles [16]. A three-electrode configuration was employed for the electrodeposition process. A large-area graphite electrode of high purity was used as a counter-electrode, and an AgeAgCl (3 M KCl electrolyte)

g L1

NiCl2$6(H2O) NH4Cl

48 170

NiCo electrodes Nomenclature

Base bath composition

g L1

tCo (min)

NiCl2$6(H2O) NH4Cl CoCl2$6(H2O) NiCl2$6(H2O) NH4Cl CoCl2$6(H2O) NiCl2$6(H2O) NH4Cl CoCl2$6(H2O) NiCl2$6(H2O) NH4Cl CoCl2$6(H2O) NiCl2$6(H2O) NH4Cl CoCl2$6(H2O)

48 170 10 48 170 10 48 170 10 48 170 15 48 170 40

55

NiCo1

NiCo2

NiCo3

NiCo4

2.

Base bath composition

NiCo5

45

15

15

15

Co electrode Nomenclature Step Base bath composition g L1 t (min) Co

1 2

NiCl2$6(H2O) NH4Cl CoCl2$6(H2O) NH4Cl

48 170 48 170

45 15

Operating conditions Temperature ( C) Current density (A cm2) Time (min) pH

25 1 60 4.5

electrode was used as reference. The experiments were carried out by means of an AUTOLAB PGSTAT302N potentiostat/ galvanostat. The surface morphologies and compositions of the obtained electrocatalytic coatings were studied by means of an OLIMPUS LEXT OLS3100-USS confocal laser scanning microscope, and a JEOL JSM-3600 scanning electron microscope coupled with an Energy-Dispersive X-Ray (EDX) analysis.

2.2.

Electrochemical measurements

HER on the synthesized electrocatalysts was accomplished by pseudo-steady-state polarization curves and electrochemical impedance spectroscopy (EIS) in oxygen free 30 wt.% KOH solutions, obtained by bubbling N2 (15 min) before the tests. Potentiodynamic polarization curves were recorded at a scan rate of 1 mV s1 from 1.60 V vs Ag/AgCl (1.40 V vs SHE) up to the equilibrium potential, at 30, 40, 50, 60, 70 and 80  C. The working electrode was held at 1.60 V vs Ag/AgCl (1.40 V vs SHE) in the same solution before the experiments, for the time needed to set up reproducible representations.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

10159

Fig. 1 e Simplified description of the formation process of Ni and NiCo electrodes at high current density.

After the polarization curves the EIS measurements were recorded. These AC experiments were accomplished at 30, 50, and 80  C, at different cathodic overpotentials. The frequency range used for the measurements was: 10 kHze3 mHz. Ten frequencies per decade were scanned using a sinusoidal signal of 10 mV peak-to-peak. The complex nonlinear least square (CNLS) fitting of the impedance data was carried out with the Zview 3.0 software package. The electrochemical measurements were carried out in the electrochemical cell P200002526 [30]. In this system, the

developed electrode was used as the working electrode, placing the electrode/electrolyte interface on a vertical plane, in order to allow the free evolution of the produced hydrogen bubbles when necessary. The counter-electrode was a largearea Ni foam (INCOFOAM), and the reference electrode was the AgeAgCl electrode. All these tests were obtained by using an AUTOLAB PGSTAT302N potentiostat/galvanostat. Fig. 2 shows a schema of the experimental setup and the electrochemical cell used in both the electrodeposition process and the electrochemical characterization.

Fig. 2 e Schema of the experimental setup and the electrochemical cells used.

10160

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

3.

Results and discussion

3.1.

Morphology and composition of the electrodes

Figs. 3 and 4 show SEM and confocal laser micrographs of the developed coatings. The superficial morphology of electrodes is characterized by a continuous matrix provided by circular macropores distributed in the total surface, as a consequence of the electrodeposition strategy used, briefly described in the Experimental section (see Fig. 1) and in Refs. [13,15,16]. According to Fig. 3(a), pure Ni electrodeposit microstructure is characterized by closely packed dendrites. The addition of 10 g L1 of CoCl2$6(H2O) to the electrodeposition bath does not considerably modify the obtained macro and microstructure, independently of tCo, as it is shown in the SEM images of Figs. 3 and 4(a). Fig. 4(a)e(c) shows the effect of Co concentration modifications added at the same tCo on both

macro and microstructures of the deposits. 3D confocal laser images show that the increase in the bath salt concentration causes homogeneity loses in the global macrostructure derived from a decrease in the macropore number. This phenomenon can be attributed to a change in the microstructure, from dendrites to globules (cauliflower structure) that grow even inside the macropores, blocking them as the Co content increases. From Fig. 4(d) it is evident that a macropore morphology is also obtained in the case of the pure porous Co electrode. Nevertheless, its microstructure is characterized by not agglomerated dendrites, providing a more open microstructure, comparing to the rest of developed materials. Table 2 shows the chemical composition analysis of the developed electrodes obtained by means of energy-dispersive X-ray spectroscopy. With respect to the electrodes obtained with the same Co content in the deposition bath and different tCo (NiCo1, NiCo2 and NiCo3) it can be concluded that NiCo2

Fig. 3 e SEM micrographs of the Ni (a) and NiCo2 (b) developed electrodes; magnifications: 30 (a.1 and b.1), 300 (a.2 and b.2), 3000 (b.3) and 5000 (a.3).

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

10161

Fig. 4 e 3D confocal laser (1) and SEM (2) micrographs of NiCo3 (a), NiCo4 (b), NiCo5 (c) and Co (d) developed coatings. Magnifications: 100 (3D confocal laser images) and 300 (SEM images).

10162

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

Table 2 e Composition of the investigated electrocatalytic coatings in at.%. Catalyst Ni NiCo1 NiCo2 NiCo3 NiCo4 NiCo5 Co

Ni

Co

100.0 70.7 57.0 71.2 42.5 12.1 e

e 29.3 43.0 28.8 57.5 87.9 100.0

electrode is the Co-richest material (43 at.%). This is because Co is in the bath the time necessary to carry out the complete Co electrodeposition, according to the Faraday’s Law (15 min). On the other hand, NiCo1 and NiCo3 electrodes manifest lower Co percentages, which can be explained according to tCo. In the case of NiCo1 electrode (29.3 at.%), tCo is very high and the Co salt is not in the electrodeposition bath the time necessary to accomplish the total Co electrodeposition. With respect to NiCo3 electrode (28.8 at.%) the Co salt is added in the initial step of the process (tCo ¼ 15 min), as NieCo codeposition is anomalous, and Co is preferentially deposited [30], Co can be occluded down to a Ni layer, and it is not completely detected in the EDX analysis. From Table 2 it is also clear that the higher Co salt amount in the electrodeposition bath the higher Co percentage in the deposit, for a same tCo.

3.2.

h ¼ a þ blog j

(1)

where h (V) represents the overpotential responsible of the current density j (A cm2), b (V decade1) is the Tafel slope, and a (V) is the intercept, which is defined by Eq. (2). Exchange current density j0 (A cm2) can be determined by the equation (2) too: a ¼ ð2:3RTÞ=ðane FÞ  log j0

(2)

The charge-transfer coefficient, a, can be obtained from the Tafel slope by using Eq. (3): b ¼ ð2:3RTÞ=ðane FÞ

(3)

where ne represents the exchanged electrons, and R (¼8.314 J mol1 K1) and F (¼96,485 C mol1) are the gas and the Faraday constants, respectively. The mechanism of HER in alkaline solution involves the formation of an adsorbed hydrogen intermediate, MHads (Volmer reaction, Eq. (4)), followed by an electrochemical (Heyrovsky reaction, Eq. (5)) and/or a chemical hydrogen desorption step (Tafel reaction, Eq. (6)): H2 O þ M þ e /MHads þ OH

(4)

H2 O þ MHads þ e /H2 þ M þ OH

(5)

MHads þ MHads /H2 þ 2M

(6)

Polarization measurements

Fig. 5 shows the Tafel plots recorded on the synthesized electrocatalysts at 30  C in 30 wt.% KOH solution. Overpotentials were obtained from the pseudo-steady-state polarization curves by correcting the potential for the ohmic drop and the reversible HER potential at the operating conditions. From Tafel polarization data it is clear that the developed coatings yield very high HER catalytic activity. NiCo2 electrode manifests the highest apparent catalytic activity for HER of all the tested cathodes. On the other hand, the Corichest alloy (NiCo5, 87.9 at.% in Co) is the worst electrocatalyst, which can be attributed to the combination of a high Co content with a lower number of macropores evidenced in 0.00 30 ºC -0.05 -0.10

η/V

the micrographic study. The linear polarization representations plotted for all the investigated coatings (Fig. 5) show a classical Tafelian behaviour, i.e. HER is completely controlled by the reaction kinetics, and can be expressed by means of the Tafel equation [15,16,28,31]:

-0.15 Ni NiCo1 NiCo2 NiCo3 NiCo4 NiCo5 Co

-0.20 -0.25 -0.30 -0.35 -5

-4

-3

-2

-1

-2

log |j| / A cm

Fig. 5 e Linear Tafel polarization curves recorded on the investigated electrocatalytic coatings in 30 wt.% KOH solution at 30  C.

0

where M is a free site on the metal surface. Tafel slope plays an important function in estimating the mechanism when it is assessed from the rate-determining step (rds) of a multi-step reaction [32e34]. If it is assumed HER via the VolmereHeyrovsky mechanism [10,20] according to Eqs. (4) and (5), the Tafel slope at 30  C should be ca. 120 mV dec1 when Volmer step is rate determining (rds) (140 mV dec1 at 80  C) or 40 mV dec1 at 30  C for Heyrovsky as rds (47 mV dec1 at 80  C) [10,20]. However, from the literature it has been proved the impossibility of distinguish the rds when b is ca. 120 mV dec1, because this Tafel slope value is also obtained when the surface coverage by adsorbed hydrogen, q, tends to 1, being Heyrovsky the rds [10]. When b is 30 mV dec1 at 30  C (35 mV dec1 at 80  C) Tafel reaction (6) is the rds. Table 3 collects the kinetic parameter values derived from the polarization curves linear fitting. According to the literature of HER on transition metals [9] and the kinetics parameters shown in Table 3, b ranging from 90 to 140 mV dec1, at 30 and 80  C, respectively, and a close to 0.5 for all the coatings, HER takes place by means of the same VolmereHeyrovsky mechanism [10]. The pure Co electrode manifests a different behaviour, i.e. two different Tafel regions (or slopes), reaching slopes higher than 120 mV dec1 at low cathodic overpotentials, indicating the existence of Co oxides on the electrode surface [20,35]. As it was expected, the exchange current densities, j0, of the studied cathodes increase with temperature. NiCo2

10163

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

Temperature ( C)

Catalyst

|η100| / mV

Ni b (mV dec1) j0 (mA cm2) a h100 NiCo1 b (mV dec1) j0 (mA cm2) a h100 NiCo2 b (mV dec1) j0 (mA cm2) a h100 NiCo3 b (mV dec1) j0 (mA cm2) a h100 NiCo4 b (mV dec1) j0 (mA cm2) a h100 NiCo5 b (mV dec1) j0 (mA cm2) a h100 Co b (mV dec1) j0 (mA cm2) a h100

500 450 400 350 300 250 200 150 100 50 0

30

40

50

60

70

80

91.3 0.33 0.66 224

88.9 0.41 0.70 213

91.4 0.68 0.70 197

97.2 0.86 0.68 200

105.8 1.46 0.64 194

114.2 2.34 0.61 185

91.0 0.10 0.66 272

89.8 0.18 0.69 247

93.6 0.33 0.68 230

96.1 0.49 0.69 219

97.4 0.92 0.70 199

107.7 1.21 0.65 208

117.7 3.91 0.51 166

120.9 4.89 0.51 159

125.2 6.86 0.51 146

127.0 6.95 0.52 147

130.4 8.73 0.52 138

135.6 10.52 0.52 133

86.9 0.28 0.69 225

91.9 0.39 0.68 221

98.2 0.67 0.65 214

104.7 1.05 0.63 207

111.7 1.87 0.61 193

122.2 3.08 0.57 185

129.9 0.86 0.46 267

113.5 0.38 0.55 274

114.7 0.54 0.56 258

121.9 0.62 0.54 263

126.1 1.02 0.54 242

124.6 1.41 0.56 225

111.1 0.05 0.54 369

136.9 0.11 0.45 401

155.9 0.24 0.41 403

185.9 0.84 0.36 379

200.3 1.10 0.34 385

218.4 1.42 0.32 396

113.7 0.25 0.53 224.5

109.7 0.41 0.57 221.7

122.1 0.78 0.52 217.7

135.9 1.57 0.49 215.4

155.5 3.54 0.44 227.3

168.2 7.02 0.42 218.0

-2

-2 log j0 / A cm

Table 3 e Kinetic parameters of the HER obtained from the polarization curves recorded in 30 wt.% KOH solution at different temperatures.

-3

-4

-5 2.8

2.9

3.0

3.1 1000 T-1 / K

3.2

3.3

Fig. 7 e Arrhenius representation for the investigated electrocatalytic coatings in 30 wt.% KOH solution.

electrode, which shows the best catalytic activity, also presents the highest j0 value, one order of magnitude higher than that calculated for the pure Ni electrode, as a consequence of the possible synergism effect of NiCo alloying in this composition range. In addition, it must be remarked that the pure Co electrode also manifests a high j0 value, which, in this case, can be attributed to its more open surface morphology with respect to the rest electrode materials. Fig. 6 shows the overpotential values at a fixed current density of 100 mA cm2, h100, in 30 wt.% KOH solution at different temperatures. The higher the operating temperature, the lower the jh100j, as a consequence of an intrinsic catalytic activity improvement. From Fig. 6, it can be confirmed the main conclusions addressed from the j0 study, i.e. NiCo2 cathode is the best overall electrocatalysts. NiCo1, NiCo3 and NiCo4 electrodes manifest jh100j values very close to that reported for both pure Ni and Co porous electrodes, which is maintained in the whole range of temperatures. Apparent activation energy (Ea) obtained at equilibrium potential (h ¼ 0), with the exchange current densities, is a widely used parameter in electrocatalysis to compare the catalytic activity of an electrode in a given electrolyte [36]. The lower the activation energy, the lower the energy requirements for hydrogen production. The Ea is related to the kinetic coefficient by means of the Arrhenius equation (7): k ¼ AeEa =RT

(7)

Table 4 e Activation energy values, Ea (kJ molL1), of the investigated electrocatalytic coatings in 30 wt.% KOH solution. Ni

20

NiCo1

30

NiCo2

40

NiCo3

NiCo4

50 60 Temperature / ºC

70

NiCo5

80

Co

Catalyst

Linear regression

90

Fig. 6 e Comparison of the electrocatalytic activity of the investigated electrocatalytic coatings in terms of the overpotential needed for a fixed hydrogen production rate determined by the current density of L100 mA cmL2, h100, in 30 wt.% KOH at different temperatures.

Ni NiCo1 NiCo2 NiCo3 NiCo4 NiCo5 Co

log j0 log j0 log j0 log j0 log j0 log j0 log j0

¼ 1:991$103 $T1 þ 2:972; R2 ¼ 0:991 ¼ 2:363$103 $T1 þ 3:808; R2 ¼ 0:995 ¼ 0:893$103 $ T1 þ 0:550; R2 ¼ 0:972 ¼ 2:275$103 $ T1 þ 3:891; R2 ¼ 0:990 ¼ 1:546$103 $ T1 þ 1:502; R2 ¼ 0:968 ¼ 3:364$103 $ T1 þ 6:816; R2 ¼ 0:963 ¼ 3:156$103 $T1 þ 6:726; R2 ¼ 0:987

Ea (kJ mol1) 38.1 45.2 17.1 43.6 29.6 64.4 60.4

10164

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

Fig. 8 e Impedance data obtained in 30 wt.% KOH solution at 30  C for a. Ni catalyst; at 50  C for b. NiCo2 catalyst, c. NiCo5 catalyst and d. Co catalyst. 1. Nyquist representation, 2. Bode representation of the phase angle as a function frequency. Symbols are the experimental points and solid lines are modelled data.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

Due to the fact that the current density is proportionally related to the kinetic coefficient, the activation energy can be obtained by the following equation: log j ¼ A0 

Ea 1 $ 2:303$R T

(8)

where A0 is a constant in the operating conditions. Fig. 7 shows the Arrhenius representation obtained for the developed electrocatalysts. The linear regression equations and the obtained apparent activation energy values are collected in Table 4. From Fig. 7 it is clear that NiCo2 electrocatalyst manifests the lower slope in the Arrhenius representation and, therefore, this material exhibits the lowest apparent activation energy value. As a consequence, it can be concluded that NiCo2 electrode presents the highest intrinsic and apparent catalytic activities. On the other hand, NiCo5 and Co electrodes are characterized by the highest Ea values, which can be attributed to the considerably lower activity of Co with respect to Ni.

3.3. Electrochemical impedance spectroscopy measurements Fig. 8 shows examples of EIS spectra recorded on the Ni, NiCo2, NiCo5 and Co electrocatalysts. The complex plane representation of the impedance data obtained on the Ni and NiCo2 coatings (Fig. 8(a.1 and b.1)) reveals the presence of a straight line close to 45 at high frequencies followed by two strongly overlapped semicircles (that is two-time constants). The same shape of the impedance curves was observed for NiCo1, NiCo3 and NiCo4 electrodes. The behaviour reported at high frequencies, independent of both temperature and overpotential, indicates a cylindrical pore geometry of finite length pores, as it was observed from the SEM study, with no mass transfer effects [37]. From Fig. 8(a.1 and b.1) it is shown that, as the cathodic overpotential increases, the diameter of the two semicircles observed in the complex plane plot decreases for all the studied coatings. The increase in temperature reported the same behaviour. This phenomenon indicates that both semicircles (i.e. both time constants) are associated to the HER kinetics [38]. In all the cases, at the highest cathodic overpotentials applied, where the vigorous gas bubbling causes too much interference, only one deformed semicircle appears, as evidenced by other authors [39e41]. Contrarily, the impedance spectra of the NiCo5 electrode present two clearly differentiated semicircles in the complex plane plot (Fig. 8(c.1)), i.e. two maximums in the phase angle Bode representation (Fig. 8(c.2)), being the high frequency semicircle diameter practically constant with the overpotential (see Fig. 8(c.1)). This behaviour at high frequencies has been reported by other authors, and is attributed to pear[12,42] or diamond-shaped pore geometries [42]. Therefore, the high frequency impedance response of NiCo5 electrode is not dominated by the cylindrical pore geometry, due to the decrease of this kind of pores as a consequence of the change in the microstructure evidenced in the morphologic study. For pure Co electrode (Fig. 8(d)) only one semicircle is observed in the EIS complex plane plot, which is related to its more open superficial morphology that inhibits the pore contribution on the impedance spectra.

10165

The one-time constant (1T) electric equivalent circuit (EEC) model was used to fit the impedance experimental data characterized by only one capacity time constant (i.e. one semicircle). This is the classical Randles EEC in which a constant phase angle element (CPE) replaces the double layer capacitance (Cdl), see Fig. 9(a) [12]. CPE can be defined by the equation:
n 1 ZCPE ¼ Q$ðiuÞ

(9)

where Q is the CPE constant, u is the angular frequency (in rad s1), i2 ¼ 1 is the imaginary number, and n is the CPE exponent. The studied cathodes characterized in the impedance complex plane plot by two semicircles dependents on both overpotential and temperature were modelled by using the two-time constant parallel (2TP) EEC model (see Fig. 9(b)), originally proposed by Armstrong and Henderson [43]. As discussed in Refs. [15e17,19], for this EEC, the high frequency (HF) time constant, s1 (CPE1, R1), is associated to HER kinetics; whereas the low frequency (LF) time constant, s2 (Cp, R2), can be related to the formation of the intermediate MHads. In the case of the NiCo5 catalyst, the two-time constant serial (2TS) EEC model (Fig. 9(c)) has been used to model the experimental impedance response. This EEC model, proposed by Chen and Lasia [12], reflects the response of a system composed of two semicircles, being the HF one related to the electrode porosity, and the LF semicircle linked to HER kinetics. Fig. 8 shows that the EEC used properly model the alternating current response of the investigated materials, manifesting an excellent agreement between the experimental (symbols) and CNLS fitting (lines) data. Table 5 reports the EEC parameters values obtained from the experimental impedance data fitting on Ni, NiCo2 and NiCo5 electrocatalysts at

Fig. 9 e EEC models used to explain the EIS response of the HER on investigated electrocatalytic coatings: a. one-time constant (1T), b. two-time constant parallel model (2TP), and c. two-time constant serial model (2TS).

10166

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

different temperatures and overpotentials. The equation proposed by Brug et al. [44] was employed to average double layer capacitances, Ci, calculations: i1=ni h   1 ð1ni Þ Ci ¼ Q i = R1 S þ Ri

(10)

According to the circuit values collected in Table 5, the twotime constants, s1 (CPE1, R1) and s2 (Cp, R2) for Ni and NiCo2 electrocatalysts, diminish with the temperature and the cathodic overpotential. This fact indicates that both semicircles are connected to HER kinetics [38]. The parameters obtained from the HF semicircle, which define s1 (C1 and R1), decrease with the cathodic overpotential. This is the typical response of charge-transfer processes (i.e. HER kinetics). Attending to the LF semicircle, Cp and R2 values manifest a response which is typically connected to adsorption processes on the electrode surface, i.e. an inverse relationship between both parameters, decreasing R2 with an increase in the cathodic overpotential [15e17,20,38].

With respect to the NiCo5 electrocatalyst, from Table 5 it is clear that HF time constant, s1, does not significantly change with the cathodic overpotential, which is usually reported for HF studies on porous electrodes [45]. For the LF semicircle, both C2 and R2 decrease with the overpotential and, therefore, s2 decreases. This behaviour is consistent with the chargetransfer phenomenon. The real surface area of the fabricated cathodes can also be estimated through the EIS results. Thus, if it is subtracted the surface area effect, it is possible to differentiate the HER intrinsic catalytic activity of the material. The real electrochemically active electrode area, in terms of surface roughness factor (Rf), may be determined by means of the quotient between the Cdl of porous and smooth electrodes [46]. Cdl value depends on the metal composition by means of Eq. (11) [45]: Cdl ¼

X

qMi $CdlMi

(11)

i

Table 5 e EEC parameters obtained by fitting EIS experimental spectra recorded at various overpotentials and temperatures in 30 wt.% KOH solution on the investigated electrocatalytic coatings. Temperature/ C

Catalyst 30 Ni h/mV c2 RS/U cm2 R1/U cm2 R2/U cm2 Q1/mU1 cm2 sn n1 C1/mF cm2 Cp/F cm2 s1/s s2/s NiCo2 h/mV c2 RS/U cm2 R1/U cm2 R2/U cm2 Q1/mU1 cm2 sn n1 C1/mF cm2 Cp/F cm2 s1/s s2/s NiCo5 h/mV c2 RS/U cm2 RP/U cm2 Rct/U cm2 Q1/mU1 cm2 sn n1 C1/mF cm2 Q2/mU1 cm2 sn n2 C2/mF cm2 s1/s s2/s

80

0 2.32$104 0.69 78.1 13.1 40 0.97 36 0.49 2.8$100 6.4$100

53 6.05$105 0.68 42.9 2.8 30 0.98 27 0.84 1.11$100 2.4$100

101 1.82$105 0.69 26.8 0.4 18 0.98 16 1.97 2.2$101 7.9$101

146 8.89$105 0.72 3.3 e 15 0.98 12 e 4.0$102 e

0 1.91$104 0.38 16.5 1.3 32 0.96 25 1.27 4.1$101 1.6$100

41 6.49$104 0.42 10.6 2.2 27 0.96 23 2.17 2.4$101 2.0$100

88 1.40$104 0.42 5.0 3.5 19 0.97 15 6.2 7.5$102 2.1$100

132 1.46$104 0.42 1.5 0.1 16 0.97 12 8.7 1.8$102 1.1$101

0 1.16$103 0.69 5.95 0.22 68 0.93 65 0.18 3.9$101 4.0$102

44 7.31$103 0.69 3.47 0.06 50 0.94 50 0.19 1.7$101 1.0$102

92 2.91$104 0.67 1.79 e 37 0.94 35 e 6.0$102 e

119 1.38$103 0.64 0.81 e 30 0.95 29 e 2.0$102 e

0 1.63$103 0.36 2.70 0.07 49 0.90 43 0.09 1.2$101 1.3$102

52 5.17$103 0.35 2.83 0.01 37 0.90 32 2.06 9.2$102 7.4$103

99 1.13$104 0.39 1.97 e 22 0.97 22 e 6.7$102 e

109 3.54$104 0.38 0.96 e 20 0.96 20 e 1.9$102 e

0 4.04$104 0.75 4.41 432 6.34 0.83 2.01 28.2 0.86 20.6 8.9$103 8.9$100

40 1.86$104 0.75 4.19 340 5.46 0.84 1.85 21.8 0.89 16.3 7.7$103 5.5$100

90 2.73$104 0.76 3.74 180 4.51 0.85 1.66 18.4 0.87 12.8 6.2$103 2.3$100

139 1.90$104 0.74 4.13 66 4.04 0.85 1.42 14.5 0.87 9.7 5.9$103 6.5$101

0 1.83$104 0.47 3.43 20.4 7.69 0.78 6.31 54.7 0.82 37.9 2.2$102 7.7$101

69 3.23$104 0.45 2.91 17.9 4.28 0.83 3.65 23.9 0.82 13.2 1.0$102 4.4$101

93 1.96$104 0.45 3.38 13.2 3.74 0.83 3.19 19.9 0.86 12.5 1.1$102 1.6$101

119 2.58$104 0.45 3.77 9.75 3.28 0.83 2.79 17.9 0.88 11.9 1.1$102 1.2$101

10167

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

Table 6 e Surface roughness factors, Rf, determined from the EIS study on the investigated electrocatalytic coatings at 30 wt.% KOH solution. Temperature/ C

Catalyst 30 h/mV Rf h/mV Rf h/mV Rf h/mV Rf h/mV Rf h/mV Rf h/mV Rf

Ni NiCo1 NiCo2 NiCo3 NiCo4 NiCo5 Co

0 1790 0 1083 0 947 0 749 0 402 0 173 0 2153

53 1338 61 982 44 705 53 562 61 268 40 150 40 1417

50 101 789 110 740 92 511 101 373 94 245 90 124 89 960

146 624 134 614 119 417 146 318 112 176 139 80 138 790

0 1551 0 888 0 821 0 688 0 338 0 170 0 1849

where qMi is the surface percentage occupied by metal Mi, whose double layer capacitance is CdlMi . In this case, due to the fact that the atomic radiuses of nickel and cobalt have similar values, the surface percentage occupied by each metal can be approximated to the atomic percentage obtained by EDX. Values of 20 mF cm2 [11,19] and 135 mF cm2 [47] were considered for the Cdl of smooth polycrystalline nickel and cobalt surfaces in alkaline media, respectively. Table 6 collects the Rf values determined for the investigated cathodes at all the tested conditions. Rf values decrease with the increase in the cathodic potential, indicating that H2 gas bubbles block a fraction of the electrode active surface area, avoiding the electrolyte access [48,49]. Macroporous pure Co and Ni electrodeposits present the highest Rf values, ca. 103, in the same magnitude order than those reported in literature for electrodeposited Raney NieZn [50], pressed powder Raney NieZn [12], and thermal arc sprayed porous Ni cathodes [46]. The lower Rf values were reported for the Corichest alloy (NiCo5), which is in agreement with the micrographic study.

57 1040 49 684 61 601 67 525 71 207 35 135 75 1146

30 ºC -0.05

η/V

-0.10

Ni NiCo1 NiCo2 NiCo3 NiCo4 NiCo5 Co

-0.25 -0.30 -8

-7

-6

-5 -1 -2 log (j Rf ) / A cm

-4

-3

127 620 138 445 115 344 115 352 134 144 134 94 123 661

0 1241 0 618 0 620 0 621 0 280 0 186 0 1534

41 1135 69 458 52 469 51 507 65 212 69 108 72 1152

88 752 92 392 99 315 98 336 88 195 93 94 94 737

132 605 114 346 109 283 140 258 130 152 119 82 116 631

Conclusions

Ni, Co and NieCo macroporous cathodes, synthesized by galvanic deposition at high current densities, were characterized morphologically and electrochemically for hydrogen evolution reaction (HER) in 30 wt.% KOH solution, manifesting important catalytic properties for commercial application. HER on these catalytic coatings was studied by means of direct and alternating current techniques. Main results of this research allowed us to enhance that:

0.00

-0.20

81 733 96 516 106 431 81 421 91 178 85 120 99 787

Fig. 10 shows the linear Tafel polarization curves corrected considering the surface roughness factor. This plot is a very useful tool to directly evaluate the intrinsic catalytic activity of the investigated electrocatalysts, by subtracting the surface area effect. According to Fig. 10, it can be concluded that NiCo2 electrode, with an intermediate Rf value, manifests the highest apparent and intrinsic catalytic activities (the lowest Ea and jh100j values) as a result of the synergism between the properties of Ni and of Co in this composition range (43.0 wt.% in Co) and the synthesis conditions. On the other hand, the electrodes with a high Co content (NiCo5 and pure Co electrodes) and NiCo1 electrode show the lowest intrinsic catalytic activity. This effect is related to its particularly high superficial Co content, even in the case of NiCo1 electrode, which is connected to the fact that Co is added in the last minutes of the electrodeposition process, when the nickel bath is exhausted, and it can be generated a very thin Co-rich superficial layer.

4.

-0.15

80

-2

Fig. 10 e Linear Tafel polarization curves recorded on the investigated electrocatalytic coatings in 30 wt.% KOH solution at 30  C, corrected considering the surface roughness factor, Rf.

1. The developed electrodes are characterized by cylindrical macropores. An excess in the Co content during the electrodeposition process leads to a decrease in the roughness factor of the obtained catalysts, as a consequence of a change in the microstructure (from dendrites to globules); which is to the detriment of the electrode apparent catalytic activity. Contrarily, the pure porous Co electrode manifests a more open microstructure, provided with not agglomerated dendrites.

10168

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

2. All that the developed coatings yield very high HER catalytic activity, which was mainly attributed to its very high surface area. From the study of pseudo-steady-state polarization curves and EIS, the electrocatalysts with Co contents around 40e60 at.% present the highest intrinsic catalytic activities, indicating that, in this composition range, it exists a synergism between the HER properties of Ni and of Co. 3. HER on the investigated cathodes takes place by the VolmereHeyrovsky mechanism.

Acknowledgements The authors acknowledge the support of Generalitat Valenciana (PROMETEO/2010/023) and Universidad Polite´cnica de Valencia (PAID-06-10-2227). I. Herraiz-Cardona is grateful to Fundacio´n Iberdrola for the financial support.

[16]

[17]

[18]

[19]

[20]

[21]

references [22] [1] Barbir F. Transition to renewable energy systems with hydrogen as an energy carrier. Energy 2009;34:308e12. [2] Veziroglu TN, Barbir F. Hydrogen e the wonder fuel. Int J Hydrogen Energy 1992;17:391e404. [3] Bockris JO, Veziroglu TN. A solar-hydrogen economy for USA. Int J Hydrogen Energy 1983;8:323e40. [4] Trommer D, Noembrini F, Fasciana A, Rodriguez D, Morales A, Romero M, et al. Hydrogen production by steamgasification of petroleum coke using concentrated solar power e I. Thermodynamic and kinetic analyses. Int J Hydrogen Energy 2005;30:605e18. [5] Momirlan M, Veziroglu TN. Current status of hydrogen energy. Renew Sustain Energy Rev 2002;6:141e79. [6] Mazloomi K, Gomes C. Hydrogen as an energy carrier: prospects and challenges. Renew Sustain Energy Rev 2012;16:3024e33. [7] Zeng K, Zhang D. Recent progress in alkaline water electrolysis for hydrogen production and applications. Prog Energy Combust Sci 2010;36:307e26. [8] Ganley JC. High temperature and pressure alkaline electrolysis. Int J Hydrogen Energy 2009;34:3604e11. [9] Lasia A. Hydrogen evolution. In: Vielstich W, Lamm A, Gasteiger HA, editors. Handbook of fuel cell technology. John Wiley and Sons Ltd; 2003. p. 416e40. [10] Lasia A, Rami A. Kinetics of hydrogen evolution on nickel electrodes. J Electroanal Chem 1990;294:123e41. [11] Rami A, Lasia A. Kinetics of hydrogen evolution on NieAl alloy electrodes. J Appl Electrochem 1992;22:376e82. [12] Chen LL, Lasia A. Study of the kinetics of hydrogen evolution reaction on nickelezinc powder electrodes. J Electrochem Soc 1992;139:3214e9. [13] Marozzi CA, Chialvo AC. Development of electrode morphologies of interest in electrocatalysis. Part 1: electrodeposited porous nickel electrodes. Electrochim Acta 2000;45:2111e20. [14] Marozzi CA, Chialvo AC. Development of electrode morphologies of interest in electrocatalysis. Part 2: hydrogen evolution reaction on macroporous nickel electrodes. Electrochim Acta 2001;46:861e6. [15] Herraiz-Cardona I, Ortega E, Garcı´a-Anto´n J, Pe´rezHerranz V. Assessment of the roughness factor effect and the intrinsic catalytic activity for hydrogen evolution

[23]

[24]

[25]

[26]

[27]

[28]

[29] [30] [31] [32]

[33] [34]

reaction on Ni-based electrodeposits. Int J Hydrogen Energy 2011;36:9428e38. Herraiz-Cardona I, Ortega E, Va´zquez-Go´mez L, Pe´rezHerranz V. Double-template fabrication of threedimensional porous nickel electrodes for hydrogen evolution reaction. Int J Hydrogen Energy 2012;37:2147e56. Herraiz-Cardona I, Ortega E, Va´zquez-Go´mez L, Pe´rezHerranz V. Electrochemical characterization of a NiCo/Zn cathode for hydrogen generation. Int J Hydrogen Energy 2011;36:11578e87. Lupi C, Dell’Era A, Pasquali M. Nickelecobalt electrodeposited alloys for hydrogen evolution in alkaline media. Int J Hydrogen Energy 2009;34:2101e6. Herraiz-Cardona I, Ortega E, Pe´rez-Herranz V. Impedance study of hydrogen evolution on Ni/Zn and NieCo/Zn stainless steel based electrodeposits. Electrochim Acta 2011;56:1308e15. Navarro-Flores E, Chong Z, Omanovic S. Characterization of Ni, NiMo, NiW and NiFe electroactive coatings as electrocatalysts for hydrogen evolution in an acidic medium. J Mol Catal A Chem 2005;226:179e97. Solmaz R, Kardas‚ G. Electrochemical deposition and characterization of NiFe coatings as electrocatalytic materials for alkaline water electrolysis. Electrochim Acta 2009;54:3726e34. Giz MJ, Benito SC, Gonza´lez ER. NiFeZn codeposited as a cathode material for the production of hydrogen by water. Int J Hydrogen Energy 2000;25:621e6. Krstajic NV, Jovic VD, Gajic-Krstajic L, Jovic BM, Antozzi AL, Martelli GN. Electrodeposition of NieMo alloy coatings and their characterization as cathodes for hydrogen evolution in sodium hydroxide solution. Int J Hydrogen Energy 2008;33:3676e87. Tasic GS, Maslovara SP, Zugic DL, Maksic AD, Kaninski MPM. Characterization of the NieMo catalyst formed in situ during hydrogen generation from alkaline water electrolysis. Int J Hydrogen Energy 2011;36:11588e95. Reshkov R, Arnaudova M, Avdeev G, Zielonka A, Jannakoudakis P, Jannakoudakis A, et al. NiW/TiOx composite layers as cathode material for hydrogen evolution reaction. Int J Hydrogen Energy 2009;34:2095e100. Oliver-Tolentino MA, Arce-Estrada EM, Corte´s-Escobedo CA, Boları´n-Miro AM, Sa´nchez-De Jesu´s F, Gonza´lez-Huerta R, et al. Electrochemical behavior of NixW1x materials as catalyst for hydrogen evolution reaction in alkaline media. J Alloys Compd 2012;536:S245e9. Han Q, Chen J, Liu K, Li X, Wei X. The heat-treatment effect of amorphous NieS(La) electrode on the hydrogen evolution reaction in an alkaline media. Int J Hydrogen Energy 2004;29:597e603. Domı´nguez-Crespo MA, Torres-Huerta AM, BrachettiSibajab B, Flores-Vela A. Electrochemical performance of NieRE (RE ¼ Rare earth) as electrode material for hydrogen evolution reaction in alkaline medium. Int J Hydrogen Energy 2011;36:135e51. Brenner A. Electrodeposition of alloys: principles and practice, vol. 2. Academic Press Inc.; 1963. Garcı´a-Anto´n J, Igual-Mun˜oz A, Guin˜o´n JL, Pe´rez-Herranz V. 2000. P200002526. Southampton Electrochemistry Group. Instrumental methods in electrochemistry. Wiley; 1985. Chen LL, Lasia A. Hydrogen evolution reaction on nickelemolybdenum powder electrodes. J Electrochem Soc 1992;139:3458e64. Bockris JO, Reddy AKN. Modern electrochemistry. 2nd ed. Kluwer/Plenum Press; 2000. Rosalbino F, Delsante S, Borzone G, Angelini E. Electrocatalytic behaviour of CoeNieR (R ¼ Rare earth metal)

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 0 1 5 7 e1 0 1 6 9

[35]

[36] [37]

[38]

[39]

[40]

[41]

[42]

crystalline alloys as electrode materials for hydrogen evolution reaction in alkaline medium. Int J Hydrogen Energy 2008;33:6696e703. Dominquez-Crespo MA, Ramirez-Meneses E, MontielPalma V, Huerta AMT, Rosales HD. Synthesis and electrochemical characterization of stabilized nickel nanoparticles. Int J Hydrogen Energy 2009;34:1664e76. Trasatti S. Electrocatalysis: understanding the success of DSA. Electrochim Acta 2000;45:2377e85. De Levie R. Electrochemical responses of porous and rough electrodes. In: Delahay P, Tobias CW, editors. Advances in electrochemistry and electrochemical engineering, vol. 6. Interscience; 1967. p. 329e97. Birry L, Lasia A. Studies of the hydrogen evolution reaction on Raney-nickelemolybdenum electrodes. J Appl Electrochem 2004;34:735e49. Choquette Y, Brossard L, Lasia A, Menard H. Investigation of hydrogen evolution on Raney-nickel composite-coated electrodes. Electrochim Acta 1991;35:1251e6. Choquette Y, Brossard L, Lasia A, Menard H. Study of the kinetics of hydrogen evolution reaction on Raney nickel composite-coated electrode by AC impedance technique. J Electrochem Soc 1990;137:1726e30. Okido M, Depo JK, Capuano GA. The mechanism of hydrogen evolution reaction on a modified Raney-nickel compositecoated electrode by AC impedance. J Electrochem Soc 1993;140:127e33. Keyser H, Beccu KD, Gutjahr MA. Estimation of the pore structure of porous electrodes by impedance measurements. Electrochim Acta;197(21):539e43.

10169

[43] Losiewicz B, Budniok A, Rowinski E, Lagiewka E, Lasia A. The structure, morphology and electrochemical impedance study of the hydrogen evolution reaction on the modified nickel electrodes. Int J Hydrogen Energy 2004;29:145e57. [44] Brug GJ, Vandeneeden ALG, Sluytersrehbach M, Sluyters JH. The analysis of electrode impedances complicated by the presence of a constant phase element. J Electroanal Chem 1984;176:275e95. [45] Va´zquez-Go´mez L, Cattarin S, Guerriero P, Musiani M. Influence of deposition current density on the composition and properties of electrodeposited Ni þ RuO2 and Ni þ IrO2 composites. J Electroanal Chem 2009;634:42e8. [46] Kellenberger A, Vaszilcsin N, Brandl W, Duteanu N. Kinetics of hydrogen evolution reaction on skeleton nickel and nickeletitanium electrodes obtained by thermal arc spraying technique. Int J Hydrogen Energy 2007;32:3258e65. [47] Jukic A, Piljac J, Metiko s-Hukovic M. Electrocatalytic behavior of the Co33Zr67 metallic glass for hydrogen evolution. J Mol Catal A Chem 2001;166:293e302. [48] Rausch S, Wendt H. Morphology and utilization of smooth hydrogen-evolving Raney nickel cathode coatings and porous sintered-nickel cathodes. J Electrochem Soc 1996;143:2852e62. [49] Pierozynski B, Smoczynski L. Kinetics of hydrogen evolution reaction at nickel-coated carbon fiber materials in 0.5 M H2SO4 and 0.1 M NaOH solutions. J Electrochem Soc 2009;156:B1045e50. [50] Chen L, Lasia A. Study of the kinetics of hydrogen evolution reaction on nickelezinc alloy electrodes. J Electrochem Soc 1991;138:3321.