Abstract
A series of AB5-type alloys with different particle and grain size were characterized electrochemically as materials for negative electrodes in Ni-MH batteries. Besides, the effect of alloy composition on the discharge capacity and cycle life was studied. Ball milling under hydrogen atmosphere was applied to reduce significantly the grain size of the alloys and to avoid the initial activation at the electrochemical charge/discharge cycling. The alloy morphology and microstructure after different mechanical treatments were characterized by XRD and SEM. The best effect in respect of discharge capacity and cycle life was achieved for the materials milled for 5 h under hydrogen atmosphere.
1. Introduction
Over decades, materials for negative electrodes in Ni-MH batteries are one of the most investigated topics in material science, the main reason being the growing need for new sources of energy and the few alternatives of the fossil fuels [1–3].
Among all types of hydrogen storage alloys for Ni-MH battery application, AB5 type are of great importance, being even commercially implemented in rechargeable batteries for portable power tools and hybrid electrical vehicles (HEVs) due to their comparatively high energy density and low level of environmental toxicity [4–6]. The main drawback and limiting factor for the commercial application of LaNi5 alloy is its low theoretical electrochemical capacity of 372 mAh/g and high cost [7]. A large number of investigation during the last decade have shown that substitution of La by Mm (Ce-rich or La-rich mish metal) and partial substitution of Ni by Co, Mn, Al, Cr, etc. lead to significant improvement of the electrochemical characteristics of LaNi5 and reduce the price of the final product [8–10]. The addition of Co, Mn, Al, and other transition metals modifies the crystal structure of the LaNi5 alloy thus increasing the thermodynamic stability of the hydride phase and the alloy corrosion resistance, proving a long cycle life. The AB5 alloys used in commercial Ni-MH batteries have typical composition of MmNi3.55Co0.75Mn0.4Al0.3 [11].
Another approach for improvement of the electrode characteristics is mechanical milling or alloying for the preparation of nanocrystalline AB5 alloys and AB5-TM (Transitional Metal) composites. Mechanical treatment is being used to obtain materials with smaller crystallite sizes and higher surface area thus improving the hydrating/dehydrating kinetics and hydrogen sorption capacities [12–14]. The microstructure of the alloys and composites synthesized using mechanical milling strongly depends on the conditions used—milling time and velocity, gas atmosphere, etc. However, there is no data if such treatment affects the electrochemical characteristics of the AB5-TM composites and how it depends on the milling conditions.
In the present work, we have compared the electrochemical behavior of three AB5-type materials—ball-milled MmNi5, ball-milled MmNi4.1Co0.6Al0.3, and MmNi5-Ni composite obtained mechanically as well. The influence of both microstructure and composition on the electrochemical properties is studied. Ni is selected as additive to MmNi5 because it is known to have a positive effect on the performance of the electrodes—forms unstable hydrides and has high electrocatalytic activity [4]. It is interesting to check if these positive characteristics can be observed electrochemically for materials synthesized mechanically because they are usually characterized through gas phase p-c absorption/desorption curves. The addition of other elements like Co and Al is also another way to improve the performance of the MmNi5 alloy. In this case, the reduction of the grain size, i.e., the effect of grinding, is interesting to be analyzed as well. While milling hydrides instead of ductile metals is often used as the method for more pronounced particle size reduction of H-storage materials [15, 16], in this study, we apply milling of metals under hydrogen atmosphere, an approach which relies on the partial hydrogenation of the material, thus facilitating its decrepitation during milling.
2. Materials and Methods
Two separate samples of as-received MmNi5 alloy (Sigma-Aldrich Product no. 685976) are ball milled for 5 and 20 hours under hydrogen atmosphere in planetary-type mill Fritsch Pulverisette 6 with a rotation speed of 300 rpm and ball-to-powder mass ratio of 10 : 1. Stainless vial and milling balls were used. Hydrogen is used mainly to promote the grinding and ease the process of reducing the particle size. Milling time of 5 and 20 hours is selected for all other samples to study the effect of particle and grain size reduction on the electrochemical behavior of the materials as negative electrodes in Ni-MH batteries. MmNi5-Ni composite is obtained by adding 2 wt.% pure Ni powder (99.5%) to MmNi5 using the same milling conditions. MmNi4.1Co0.6Al0.3 was synthesized by high-frequency induction melting method.
The structure of all samples is determined using XRD (Co radiation). Morphology and particle size are studied using scanning electron microscope JEOL 5510.
The electrochemical behavior of the alloys during charge/discharge cycling is studied using three-electrode cell. The working electrode is prepared using 70 mg of the bulk or synthesized materials, 100 mg polytetrafluoroethylene (PTFE)/carbon black (Teflonized carbon), and heptane. The mixture is pressed at about 150 atm to form the electrode and left on air to dry. NiOOH/Ni(OH)2 is used as counter electrode and Ag/AgCl as reference. Each electrode is charged for 3 hours at 5 mA and discharged to 500 mV in 6 M KOH water solution.
3. Results and Discussion
X-ray powder diffraction revealing the microstructure of all studied samples is shown in Figure 1. It is clear that after milling for 5 hours MmNi5 has sufficient reduction of the crystallite size from 88 nm for bulk MmNi5 to 34 nm. Further milling for 20 hours leads to lower crystallite size of 10 nm and the sample loses partly its crystallinity (Figure 1). SEM images however show that the average particle size in both cases is nearly the same—about 2-3 μm. The particle size distribution being wide with both smaller separate particles and agglomerates consisting of much smaller particles is presented in Figures 2(b) and 2(c). Milling of MmNi5 with 2 wt.% Ni leads to the formation of a composite—the most intense peak of Ni at 52° being present at the pattern. Obviously, applying these milling conditions, nickel does not incorporate in the crystal lattice of MmNi5 since there is no shift of the diffraction peaks. The broadening of the peaks in this case is smaller than in the sample milled for 5 hours without the addition of Ni and the crystallite size according to Scherrer’s equation is 42 nm. It is interesting that only 2 wt.% Ni leads to a pattern with more intensive and better expressed crystal peaks from the sample milled for 5 hours without Ni (Figure 1). The SEM image of the composite (Figure 2(d)) reveals a wide particle size distribution with an average particle size of 12 μm. Bulk defects like cracks and fractures are not so well manifested; the particles retained the morphology of the bulk MmNi5 but the particle size is reduced. All these effects result from the addition of 2 wt.% Ni. Srivastava and Panwar [13] have observed the same effect when adding nickel to MmNi5 and milling for 3 hours. They explained it with respect to the hardness of the transition element added to MmNi5—less elastic metals preserve the morphology of the sample to some extent during mechanical processing.
(a)
(b)
(c)
(d)
(e)
(f)
Diffraction peaks in the pattern of bulk MmNi4.1Co0.6Al0.3 are shifted towards smaller angles in comparison to bulk MmNi5 (Figure 3) proving that the added Co and Mn substitute Ni atoms in the crystal lattice, thus changing only the cell size, but keeping the CaCu5 structure type—the same as for MmNi5. The morphology of both alloys is similar (Figures 2(a) and 2(e))—particles with wide size distribution from 50 to 500 μm. Mechanical grinding of MmNi4.1Co0.6Al0.3 for 5 hours leads to the significant lowering of the crystallite size from 90 nm for bulk material to 29 nm. Comparison of ball milled for 5 hours MmNi5 and MmNi4.1Co0.6Al0.3 shows similarity in the structure of both alloys—they are both nanocrystalline with similar crystallite size (Figure 3). There is no significant difference in the morphology as well—MmNi4.1Co0.6Al0.3 milled for 5 hours has the same average particle size of about 2 μm, but a little bit wider distribution in size (Figures 2(e) and 2(f)).
All of the ball-milled materials have nanocrystalline structure and reduced particle size and the influence of both microstructure and composition on their electrochemical properties is of high interest. The electrochemical discharge capacity of MmNi5 alloys before and after milling is compared in Figure 4. The maximum discharge capacity of 78 mAh/g is measured for bulk MmNi5 which drops in the second cycle to 36 mAh/g and becomes even lower with cycling reaching about 20 mAh/g. These results are close to those published by Jurczyk et al. [17]. Here, it has to be mentioned that the working electrode structure in our study was not optimized to maximize the electrode capacity but rather to achieve good reproducibility of the electrochemical results and to be able to compare them quantitatively (Figure 5). One possible explanation for the higher discharge capacity obtained in this case than that of the milled samples can be the hydrogen atmosphere during grinding, which can cause formation of difficult to decompose cerium hydride and block some of the active centers on the surface of the particles of the milled MmNi5 (mish metal used in the study is Ce-rich). Milled MmNi5 for 5 and 20 hours has nearly the same initial discharge capacity of 35 and 37 mAh/g correspondingly, lower than that of the bulk alloy. The microstructure developed after mechanical grinding for 5 hours has the most positive effect on the discharge capacity. Despite the low values at the second and third cycles, there is a tendency for increase with cycling and the value for the last tenth cycle is 36 mAh/g. Milling for 20 hours leads to partial amorphization of the alloy but the discharge capacities vary with a minimum of 20 mAh/g at the third cycle to a maximum of 35 mAh/g at the sixth cycle, followed by a decay and a final capacity of 26 mAh/g. This is lower than the final capacity of the sample milled for 5 hours, the main reason being partial agglomeration of the particles after such long grinding. For all materials, there is a clear tendency for a drop of the capacity in the second cycle which probably is due to corrosion processes on the surface of the electrode and difficult access to the particles of active material inside.
The lowest discharge capacities are obtained for the composite with 2 wt.% Ni—the lowest initial capacity of 27 mAh/g followed by additional decrease in the second cycle and constant value of 16 mAh/g (Figure 4). In this case, mechanical grinding, which is the activation process for MmNi5 alloys [18], is suppressed due to nickel presence and the larger particle and crystallite sizes cause difficulties in the electrochemical hydrogenation.
The electrochemical discharge curves for bulk MmNi4.1Co0.6Al0.3 and milled for 5 hours are shown in Figure 6. They are both characterized by higher capacities than MmNi5 samples. Co and Al are known for their positive effect on the electrochemical behavior of the metal hydride electrodes, namely, decrease in the corrosion rate, improvement in the cycle life, and increase in the amount of hydrogen stored [4]. These positive effects are true in our study as well. Bulk MmNi4.1Co0.6Al0.3 has typical electrochemical behavior with the lowest initial discharge capacity of 35 mAh/g which becomes higher in the next cycles reaching a plateau of 45 mAh/g. Milling for five hours improves the microstructure and therefore the electrochemical performance of the electrode—initial discharge capacity of 45 mAh/g followed by a maximum of 75 mAh/g at the fourth cycle and a final one of 65 mAh/g.
4. Conclusions
Three nanocrystalline AB5-type materials were obtained by mechanical grinding—MmNi5, MmNi4.1Co0.6Al0.3, and MmNi5-2 wt.% Ni composite. The effective reduction of particle and crystallite sizes down to 2 μm and 34 nm, respectively, is observed after milling for 5 hours under hydrogen atmosphere. Grinding for 20 hours further reduces the size of the crystals to 10 nm without altering the average particle size due to their agglomeration. Adding 2 wt.% Ni to MmNi5 during milling preserves the microstructure of the sample and the composite MmNi5-2 wt.% Ni showed the least size reduction. The electrochemical performance of the materials showed that the discharge capacity strongly depends on particle and grain sizes and partial substitution of Ni by Co and Al. The highest discharge capacity was obtained for MmNi4.1Co0.6Al0.3 after milling for 5 hours.
Data Availability
All data are provided in Results and Discussion. Any additional information about the data used to support the findings of this study are available from the corresponding author upon request.
Disclosure
The research was performed as part of the employment of the authors (Sofia University “St. Kliment Ohridski”).
Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this paper.
Acknowledgments
This work was supported by the European Union Horizon 2020 Project Materials Networking under Grant REP-692146-1 and European Regional Development Fund within the Operational Programme “Science and Education for Smart Growth 2014-2020” under the Project CoE “National Center of Mechatronics and Clean Technologies” (BG05M2OP001-1.001-0008-С01).