Introduction

Hydrogen is considered as a promising sustainable energy carrier due to its high energy density and the fact that it can be produced from a variety of renewable sources including biomass and water electrolysis. In addition, hydrogen combustion does not emit green-house gases to the atmosphere which justifies its classification as a clean source of energy. To be able to use hydrogen as an energy source, its safe and efficient storage is a key requirement, but has remained a most challenging issue [1].

Hydrogen storage alloys have been extensively studied for many years as negative electrode materials of Ni–MH batteries [2–6]. Among the various kinds of hydrogen storage materials for Ni–MH batteries, the AB5-type compounds display an average maximum discharge capacity of 300 mAh/g. However, none of the currently commercialized electrode alloys, including AB5 and AB2-types, can meet the demand of power battery owing to the limitation of their properties, such as low discharge capacity for the AB5-type compounds and poor activation capability for the AB2-type Laves phase electrode alloy. Therefore, one of the main challenges in this area is to find new type of electrode materials with higher capacity and longer cycle life.

In the recent years, La–Mg–Ni hydrogen storage alloys, with PuNi3-type structure, have been considered as promising candidates owing to the benefit of lower cost, higher discharge capacity and good electrochemical properties compared to AB5-type alloys [7–11].

Kadir et al. [11, 12] revealed that RMg2Ni9 (R = rare earth, Ca or Y) alloys keep the PuNi3-type rhombohedral structure after hydriding, and that their hydrogen storage capacity, which could reach 1.7–1.8 % (mass fraction), was significantly higher than that of the mischmetal-based AB5-type alloys.

Chen et al. [7] have prepared by induction melting several kinds of R–Mg–Ni based alloys with PuNi3-type structure. Subsequently, they found that the discharge capacity of LaCaMgNi9 alloy could reach 360 mAh/g (1.87 wt%), but the high-rate dischargeability and cyclic stability were poor.

In this work, LaCaMgNi9 compound was prepared by mechanical alloying (MA) to avoid inherent problems of the melting techniques [13]. It is now established that MA is an efficient process to synthesize a wide variety of equilibrium and non-equilibrium structures in solid state. The synthesized alloys exhibit a good compositional and microstructural homogeneity with nanometric sized particles due to the high energy impact during milling. Nanostructuring is one of the possible approaches to improve the hydrogenation properties of these materials [14, 15]. The aim of this work was to study the nanostructuring effect on the electrochemical properties of LaCaMgNi9-type alloy.

Experimental details

Elemental La, Ca, Mg and Ni powders (with at least 99.9 % purity) were mixed with the nominal composition LaCaMgNi9 and charged into carbide vials under controlled atmosphere (purified argon in glove box). MA was performed with a Fritsch Pulverisette P7 planetary ball mill at a rotation speed of 400 rpm with a ball to powder ratio equal to 17:1. The MA duration was varied from 4 to 30 h and the samples were labeled Sd (S for sample and d for duration in hour).

Each powdered sample was sieved to a particle size below 40 µm for X-ray powder diffraction (XRD) analysis. XRD patterns were recorded with a (θ–2θ) Panaltytical X’Pert pro MPD diffractometer with a copper anti-cathode (λ CuKα = 0.15406 nm) in a 2θ range of 10°–100°. The XRD data were analyzed by the Rietveld method [16, 17] using FULLPROF program [18].

Working electrodes were prepared by mixing the alloy powders with black carbon and PTFE in, respectively, 90, 5 and 5 % weight proportions. Electrochemical measurements were performed in a conventional three-electrode open-air cell using a VMP biologic potentiostat–galvanostat. The discharge capacities of the electrodes were determined at room temperature by galvanostatically charge–discharge at, respectively, C/3 and C/6 rates. The Ni(OH)2/NiOOH and Hg/HgO electrodes were used as the counter and the reference electrodes, respectively. Cyclic voltammetry was performed for different scan rates in the −0.5 to −1.1 V potential range. The chronoamperometry was applied after 30 cycles of charge–discharge. This method consists in fully charging the electrode and discharging it at constant potential (−0.6 V versus Hg/HgO).

Experimental results and discussion

Structural properties of the alloys

Starting from elemental La, Ca, Mg and Ni metals, MA leads to the formation of a mixture of nanocrystalline AB3-type phase with PuNi3-type structure (S.G: R-3 m) and LaNi5 phase since 10 h of MA.

Figure 1 gives the XRD patterns of the samples alloyed for different times. In the XRD pattern of the un-milled sample (S0), all the diffraction lines of the starting elements are present. The diffraction peaks of nickel are the most intense while those of the other elements are barely detectable. This fact is mainly related to the low contents of Ca, Mg and La (5.47, 3.32 and 18.9 wt%, respectively) as compared to the Ni one (72.21 wt%).

Fig. 1
figure 1

Evolution of the XRD patterns of samples for increasing MA duration (4–30 h)

Full size image

X-ray diffraction patterns of the mechanically alloyed powders reveal a modification of the microstructure of the powders during MA process. With increasing MA duration, the originally sharp diffraction lines of the powders show a remarkable line broadening and an intensity reduction compared to the S0 sample. This indicates the decrease of the crystallite size and the presence of lattice microstrain within the particles. Furthermore, MA for long time results in the partial amorphization of the alloy. In fact, the increase of MA duration induces more strains and increases the defect concentration in the crystalline structure of elemental powders that leads to the destabilization of part of them into amorphous phase.

As we can see, Ni remains up to 20 h of MA while other elements are not observed after the first hours of milling. Depending on MA duration, the formation of an AB3-type phase with a PuNi3-type structure occurred in coexistence with LaNi5 (S.G: P6/mmm). The LaNi2-type phase appears as an intermediate phase for MA duration ranging from 4 to 8 h and disappears hereafter.

The composition and lattice parameters of the phases were calculated by the Rietveld method and are listed in Table 1. As the MA duration increases, the abundance of the LaCaMgNi9 AB3-type phase increases from about 43–63 wt% whereas that of LaNi5 increases first from 40 to 47 wt% and then decreases to 36 wt%.

Table 1 Phase structural properties and maximum discharge capacities of the samples synthesized by MA at various milling time
Full size table

The SEM micrograph presented in Fig. 2 shows the morphology of the S30 sample. The powder is composed of large agglomerates of more than 5 µm in size. These agglomerates are made of several smaller (0.1–1 µm) deformed particles welded together.

Fig. 2
figure 2

SEM micrography showing the morphology of the S30 sample

Full size image

Electrochemical characterizations

Activation and charge–discharge cycle stability of the alloys

The variation of the measured discharge capacities of the mechanically alloyed samples as a function of the charge–discharge cycle number is shown in Fig. 3. For all samples, the maximum discharge capacity was reached at the first cycle, then slightly decreases for next cycles to remain stable at higher cycle number. This decrease is due to the corrosion of the active phase in the electrolyte during the reaction.

Fig. 3
figure 3

Variation of the discharge capacity versus cycle number of S10–S30 samples

Full size image

As the alloy contains some phases that do not absorb hydrogen, the discharge capacity of the active material (C activematerial, Table 1) was calculated based on the measured discharge capacity of the hole material (Fig. 3) and the contributions of the hydrogen absorbing phases only (AB3-type and LaNi5 phases), as:

$$C_{\text{activematerial}} = C_{\text{measured}} / (x + y )$$
(1)

where C measured is the measured discharge capacity (of the whole material), and x and y are, respectively, the LaNi5 and the AB3-type phase contents.

The effect of the MA duration on the discharge capacities of the active material at the first and the 10th charging–discharging cycle is shown in Fig. 4. This figure shows that the discharge capacity increases up to 20 h of milling to reach about 156 mAh/g. Above 20 h, the discharge capacity decreases with the MA duration to reach values comparable to that obtained at lower milling duration. The storage capacity reduction with the MA duration is attributed to the creation of defects in the material during the synthesis process. During MA, the powder particles which are trapped between the colliding balls are subjected to compressive impact forces. Thus, they are deformed, fractured and cold-welded leading to defects in the system and also an average particle size reduction [19].

Fig. 4
figure 4

Effect of MA duration on the dischage capacities of the active material at the first cycle and the loss of discharge capacity at the 10th charging–discharging cycle

Full size image

As mentioned before, the discharge capacity of LaCaMgNi9 alloy could reach 360 mAh/g [7] while the maximum discharge capacity obtained in this work is 156 mAh/g of active material (146.2 mAh/g of sample S20) after 20 h of MA. In addition, as samples are formed by two hydrogen absorbing phases (LaNi5 and AB3-type phase), it is important to know the contribution of each phase to the total discharge capacity. For a polyphasic alloy, the total measured electrochemical discharge capacity would equal to the sum of the contributions of all phases as:

$$C_{\text{measured}} {\mkern 1mu} {\mkern 1mu} ( {\text{mAh/g of alloy)}} = \sum\limits_{i} {x_{i} } C_{{\phi_{i} }}$$
(2)

where x i and \(C_{{\phi_{i} }}\) are, respectively, the weight content and the electrochemical discharge capacity of the phase ϕ i . In this work, as mentioned above as only LaNi5 and AB3-type phases are the hydrogen absorbing phases, the total measured electrochemical discharge capacity can be expressed as:

$$C_{\text{measured}} \,\, ( {\text{mAh/g of alloy)}} = x \times C_{{{\text{LaNi}}_{ 5} }} + y \times C_{{{\text{AB}}_{ 3} {\text{ - typephase}}}}$$
(3)

where x and y are, respectively, LaNi5 and AB3-type phase weight contents.\(C_{{{\text{LaNi}}_{ 5} }}\) and \(C_{{{\text{AB}}_{ 3} }}\) are, respectively, the discharge capacity of LaNi5 and AB3-type phase.

To determine the maximum discharge capacity of the obtained AB3-type phase, we measured the discharge capacity of LaNi5 synthesized in the same energetic condition that the AB3-type one. The obtained value is 180 mAh/g. The discharge capacity of the AB3-type phase is expressed as below:

$$C_{{{\text{AB}}_{ 3} {\text{ - typephase}}}} = [C_{\text{measured}} - (x \times C_{{{\text{LaNi}}_{ 5} }} ) ] /y$$
(4)

Using expression (4), the maximum discharge capacity of the formed AB3-type phase is about 135 mAh/g. Thus, as the discharge capacity of the LaNi5 is higher than that of AB3-type phase, the decrease of the total measured electrochemical discharge capacity or the electrochemical discharge capacity of the active material beyond 20 h of MA (Fig. 4) is attributed to the decrease of the LaNi5 weight content.

The activation capability is a very important performance for the practical use of Ni-MH battery. It is usually characterized by the number of charge–discharge cycles required to reach the greatest discharge capacity at constant current density. The smaller the number of cycles, the better is the activation performance. The examination of the variation of discharge capacity as function of the cycle number (Fig. 3) reveals that the alloys possess good activation performances, attaining their maximum discharge capacities at the first cycle.

Determination of kinetic parameters of the hydrogen absorption reaction

Cyclic voltametry

Figure 5 shows the cyclic voltammograms for the S30 sample. The working electrode potential was scanned from −1.1 to −0.5 V versus Hg/HgO (1 mol/L KOH solution) with scan rates of 10, 20, 30, 40 and 50 µV/s. The anodic and cathodic peaks reflect the information related to the discharge and charge processes [20]. The observed anodic peak is attributed to the oxidation of the desorbed hydrogen atoms at the surface, while the cathodic peak is attributed to the reduction of hydrogen atoms absorbed by the alloy in the interstitial sites. As we can see in Fig. 5, the anodic peak current increases and its potential shifts towards positive direction with increasing scan rate.

Fig. 5
figure 5

Cyclic voltammograms of S30 sample at potential scan rate: 10, 20, 30, 40 and 50 µV/s

Full size image

Figure 6 gives the variation of the anodic peak potential of the S30 alloy as function of log (v). This relation is linear. In this case, the anodic peak potential can be written as follows:

$$\frac{{{\text{d}}E_{\text{ap}} }}{{{\text{d}}\log (v)}} = \frac{2.3RT}{2\alpha nF}$$
(5)

with R is the gas constant, F is the Faraday constant and T is the temperature.

Fig. 6
figure 6

Variation of the anodic peak potential of cyclic voltammograms of S30 sample as function of log(v)

Full size image

This equation allows the determination of the charge transfer coefficient α (Table 2). This parameter indicates the alloy ability to charge and discharge. As the α value ranges from 0.3 to 0.7, it can be approximated close to 0.5 [21] which means that the charge and discharge are reversible and the system has the same tendency for the charge as for the discharge process.

Table 2 Charge transfer coefficient ‘α’ and hydrogen diffusion coefficient “D” of the S10–S30 samples
Full size table
Chronoamperometry

Figure 7 shows the current responses expressed in log(i) versus time for the synthesized alloys at 298 K. The semi-logarithmic curves of anode current density versus time can be divided into two time regions. In the first one, the current decreases rapidly due to a consumption of hydrogen at the surface, while in the second time region the current decreases slowly in a linear way. In this case, hydrogen is supplied from the bulk alloy proportionally to the concentration gradient of hydrogen and the current is controlled by the diffusion rate of hydrogen atoms [22] with time [23, 24].

Fig. 7
figure 7

Semilogarithmic plots of anodic current–time responses of the S10–S30 samples

Full size image

As all samples are multi-phases, the calculated coefficient of hydrogen diffusion is an average value describing the hydrogen diffusion in the whole sample not in a specific phase. From the slope of the linear region in Fig. 7, it is possible to estimate the average coefficient of hydrogen diffusion value using the following formula, which is valid at a large time [23]:

$$\log (i) = \log \left( {\frac{6FD}{{da^{2} }}} \right)(C_{0} - C_{\text{s}} ) - \left( {\frac{{\pi^{2} D}}{{2.303a^{2} }}} \right)t$$
(6)
$$D = - \frac{{2.303a^{2} }}{{\pi^{2} }}\frac{{{\text{d}}\log (i)}}{{{\text{d}}t}}$$
(7)

where i, D, d, a, C 0, C s and t are, respectively, the diffusion current (A/g), the diffusion coefficient of hydrogen (cm2/s), the density of hydrogen storage, the alloy particle radius (cm), the initial hydrogen concentration in the bulk of alloy (mol/cm), and the hydrogen concentration on the surface of the alloy particle radius, respectively. Assuming that the average particle radius is a = 5 µm (SEM micrograph Fig. 2), D can be calculated according to Eq. (3) and the calculated values are listed in Table 2. As we can notice, the diffusion coefficient of hydrogen D shows the same behavior than the discharge capacity as a function of MA duration. Its value increases up to 20 h of MA and decreases for larger time.

Conclusion

A nanostructured PuNi3-type alloy has been synthesized by mechanical alloying of a stoichiometric mixture of La, Ca, Mg and Ni elements at different milling duration. The electrochemical properties of the milled samples have been investigated by electrochemical methods: chronopotentiometry, cyclic voltammetry and chronoamperometry. The following conclusions can be drawn:

  • All samples exhibit limited discharge capacity with a maximum of about 156 mAh/g after 20 h of MA. Further, the discharge capacity decreases with increasing MA duration. This is attributed to the increase of lattice defects and the decrease of particle size.

  • The values of the charge transfer coefficient range from 0.3 to 0.7 for samples mechanically alloyed between 12 and 25 h which indicate a good reversibility of the electrochemical reaction. For the sample milled for 10 and 30 h, the α values are, respectively, 0.232 and 0.242.

Highlights

  • Synthesis of LaCaMgNi9-type alloy by mechanical alloying.

  • Effect of increasing MA duration on the electrochemical properties of the obtained alloy.

  • ‘α’ values range from 0.3 to 0.7.

  • ‘D’ and the discharge capacity have the same behavior as function of MA duration.