Effect of Cold Rolling on the Hydrogen Desorption Behavior of Binary Metal Hydride Powders under Microwave Irradiation

Abstract

In this paper we report that cold rolling could drastically improve hydrogen desorption kinetics under microwave irradiation. Samples of metal hydride powders (TiH₂, ZrH₂, and MgH₂) in as-received conditions and after cold rolling were microwave irradiated in a vacuum using a simple experimental setup. After irradiation, the samples were characterized by X-ray diffraction in other to evaluate the effectiveness of microwave heating. The diffraction patterns indicated that only MgH₂ could be fully decomposed (dehydrided) in the as received state. TiH₂ was only partially decomposed while no decomposition was observed for ZrH₂. However, cold rolling the hydride powders prior to microwave heating led to a significant improvement of hydride decomposition, resulting in the complete dehydriding of TiH₂ and extensive dehydriding of ZrH₂. These results clearly indicated the positive effects of cold rolling on the microwave assisted desorption of the investigated binary hydrides.

1. Introduction

The use of microwave energy in material synthesis and processing has been widely investigated over the years. This approach has been successfully used in the synthesis of micrometric and nanometric particles, microwave sintering, surface modification, etc. [1,2,3,4,5,6]. The most investigated use of microwaves in materials processing and synthesis is the microwave sintering of oxides, carbides, nitrides, silicides, etc. [7].

The microwave heating of metals was considered a minor subject for a long time since the penetration depth of electromagnetic fields into metals is quite small (up to few micrometers), limiting the heating reached in bulk metals. For detailed information on the electromagnetic field interaction with metals during microwave irradiation, the interested reader may refer to the review article by Yoshikawa [8]. However, this scenario changed markedly in 1999 when Roy et al. [9] reported for the first time the microwave heating and sintering of a porous metal compact, extending the applicability of microwave irradiation in materials processing. Subsequently, the interest in the microwave assisted processing of metals, alloys and metal matrix composites became reality [10,11,12,13].

Nowadays, the search for clean energy sources is a major target worldwide and hydrogen appears as a promising energy vector candidate [14,15,16]. Conversely, a number of important technological problems must to be solved in order to allow the large scale utilization of hydrogen in practical energy systems. One of the most challenging issues to be solved is finding safe and reliable technologies for hydrogen storage. Metal hydrides are interesting materials for solid state hydrogen storage, and several hydride systems and different materials processing routes have been extensively investigated [14,15,16,17,18,19,20,21,22,23,24].

Despite of the abovementioned growing interest in the microwave heating of metallic materials, until now only few papers addressing the microwave irradiation in metal hydrides have been published. Luo et al. [25] reported that titanium sintered parts obtained by the microwave irradiation of TiH₂ showed higher densities, finer residual pores and better mechanical properties than those obtained from hydride-dehydrided titanium powder. Li et al. [26] employed microwave sintering to process Mg-La-Ni hydrogen storage alloys from elemental powders and successfully obtained multi-phase alloys. Microwave assisted synthesis of hydrides was also performed by Tapia-Ruiz et al. [3] to produce the Li₄NH hydride by solid state reaction between Li₃N and LiH compounds. This reaction is complex, presenting possible alternative pathways and side reactions with atmosphere or container material contaminations. These authors reported success in producing single phase Li₄NH by microwave synthesis over a much shorter period of time than necessary for conventional heating methods.

Hydrogen desorption promoted by microwave heating was also reported in a few papers. Nakamori et al. [27] investigated the effect of microwave irradiation on some metal hydrides and complex hydrides. These authors reported that some of the investigated transition metal hydrides presented small hydrogen desorption (TiH₂ and VH_0.81), as indicated by shifts in diffraction peaks positions of the hydride phases, whereas others did not indicate any hydrogen release (ZrH₂ and LaH_2.48). Similar behavior was observed for complex hydrides, where LiBH₄ showed partial desorption whereas NaBH₄ and KBH₄ do not show any evidence of desorption by microwave irradiation. The authors concluded that microwave penetration depth and conduction losses are controlling factors on microwave heating and hydrogen desorption for the investigated hydrides. Krishnan [28] also studied the effects of microwave irradiation in NaAlH₄ and TiCl₂ doped NaAlH₄. The authors concluded that, when the metallic phase Al is present in the starting material, microwave exposure leads to the production of reversible Na₃AlH₆ and Al phases. However, when the sample does not contain an Al phase then microwave irradiation produces an amorphization of NaAlH₄. Zhang et al. [29] proposed the use of a Ni-coated honeycomb ceramic monolith (HCM) as susceptor to promote fast heating and hydrogen desorption in MgH₂ and other hydrides. Awad et al. [30] investigated the effect of different types of carbon (carbon fibers, graphite and diamond) on the hydrogen desorption behavior of Mg + C nanocomposites obtained by ball-milling and reported that more effective desorption was obtained by increasing the milling time, the carbon type content and microwave power. Among the investigated composites, the MgH₂ + 10% CF showed the larger hydrogen release, converting about 90% of the initial MgH₂ in Mg.

A possible application of the microwave desorption is for Zircaloy recycling as mentioned by Dupim et al. [31]. As for absorption, to our knowledge absorption under microwave radiation has not been reported yet. The reason may be the necessity of having a sample holder that has to support hydrogen pressure while being permeable to microwaves, which increases the safety issue. In the present investigation only desorption was investigated.

In this work, we investigate the effect of cold rolling on the hydrogen desorption behavior of binary hydride powders during microwave irradiation. This investigation was motivated by the reports concerning the positive effects of cold rolling on the hydrogen desorption kinetics in several hydride systems under conventional heating [31,32,33,34,35,36] and also the lack of investigations focusing the effect of cold rolling on the hydrogen desorption of hydrides during microwave heating.

2. Experimental Section

MgH₂, TiH₂, and ZrH₂ powders having respectively 98%, 99%, and 98.5% purity (metal basis), were used in microwave desorption experiments. Cold rolling was performed by placing the hydride powder between two aluminum plates and rolling it 5 times using a Durston 130 laboratory rolling mill (Durston, High Wycombe, Buckinghamshire, UK). The cold rolling was performed in air.

For microwave irradiation, a domestic microwave oven (Danby model DMW 749SS, Danby, Guelph, ON, Canada) specially adapted for these experiments was used. This microwave oven has a power of 700 W and frequency of 2.5 GHz. A quartz tube with one side dome closed and the other open (200 mm in length and 10 mm diameter) was used as sample holder and a stainless steel cylinder (diameter 0.8 mm) was used as susceptor. The open extremity of this tube was linked to a vacuum pump for dynamic evacuation during the experiments. The level of vacuum was about 10⁻² torr. The mass of the hydride powder was about 200 mg. A schematic representation of the sample holder used for these experiments is shown in Figure 1. The investigated samples (as-received and cold-rolled) were placed into the quartz tube inside an argon filled glove-box. Irradiation experiments were performed with a fixed duration of 30 min.

The morphology of the metal hydrides was observed by scanning electron microscopy (SEM) using a Philips XL30 microscope (FEI, Hillsboro, Oregon, USA). Working distance and accelerating voltage were varied according to the operating mode of the microscope (backscattered or secondary electron). The actual values are indicated on the respective micrographs. The investigated samples were characterized by X-ray diffraction using a Bruker D8 Focus diffractometer (Cu-Kα radiation), Bruker, Madison, WI, USA. Rietveld refinement was performed using the Topas software (Bruker, Madison, WI, USA) [37].

Figure 1. Schematic representation of the sample holder used for microwave irradiation experiments.

Figure 1. Schematic representation of the sample holder used for microwave irradiation experiments.

3. Results and Discussion

3.1. Electron Microscopy

Figure 2 shows the morphologies of hydride powders investigated by scanning electron microscopy images obtained with a secondary electrons detector (SE-SEM). Figure 2a–c correspond to MgH₂, TiH₂, and ZrH₂ as-received powders, respectively. It is clear that all the samples show irregular particle morphology, with relatively widespread particle sizes. Moreover, it is noticeable that ZrH₂ powder has the larger mean particle size, followed by MgH₂ and TiH₂.

The effect of cold rolling on the morphology of the hydride powders can be observed in Figure 2d–f for MgH₂, TiH₂, and ZrH₂ respectively. All the rolled samples presented a morphology of large plate-like agglomerates composed of very fine particles, with particles sizes in the range of sub-microns to few microns. The strong particle size refinement promoted by cold rolling on the hydride powders can be clearly seen. The particle size refinement is a positive feature for microwave heating since the microwave penetration depth in metals is small. Moreover, the decrease in particle size allows the metal powder to reach higher temperatures as demonstrated by Mondal et al. [38] in the case of microwave irradiation of Cu powder.

Figure 2. Morphology of metal hydride powders observed by scanning electron microscopy (SEM). Images (a), (b), and (c) correspond to as-received MgH₂, TiH₂, and ZrH₂, respectively, whereas (d), (e), and (f) correspond to coldrolled MgH₂, TiH₂, and ZrH₂, respectively.

Figure 2. Morphology of metal hydride powders observed by scanning electron microscopy (SEM). Images (a), (b), and (c) correspond to as-received MgH₂, TiH₂, and ZrH₂, respectively, whereas (d), (e), and (f) correspond to coldrolled MgH₂, TiH₂, and ZrH₂, respectively.

3.2. X-ray Diffraction: Cold Rolling Effects

The X-ray diffraction (XRD) patterns of as-received and cold rolled samples of MgH₂, TiH₂, and ZrH₂ are shown in Figure 3, Figure 4 and Figure 5, respectively.

Figure 3. X-ray diffraction patterns of as-received and cold rolled (5 passes) MgH₂ powders. (CR): cold rolled; (AR): as-received.

Figure 3. X-ray diffraction patterns of as-received and cold rolled (5 passes) MgH₂ powders. (CR): cold rolled; (AR): as-received.

Figure 4. X-ray diffraction patterns of as-received and cold rolled (5 passes) TiH₂ powders.

Figure 4. X-ray diffraction patterns of as-received and cold rolled (5 passes) TiH₂ powders.

The XRD patterns of MgH₂ shown in Figure 3 indicate the predominance of the hydride phase with some observable peaks of magnesium. Cold rolling caused broadening of the diffraction peaks which can be ascribed to lattice defects and crystallite size reduction. A quantitative analysis performed using Rietveld refinement indicated the same amount of Mg and MgH₂ in the as-received and cold rolled samples. However, the crystallite size was greatly reduced by cold rolling, going from 114 nm in the as-received sample to 14 nm in the cold rolled sample.

In the case of TiH₂, the X-ray patterns presented in Figure 4 indicate that cold rolling greatly broadened the Braggs peaks. However, contrary to the results found with MgH₂ where the broadening was due to reduction of crystallite size, here the crystallite size remains essentially constant (60 nm) but the microstrain is increased by an order of magnitude from 0.02% in the as-received sample to 0.4% in the cold rolled sample.

Figure 5. X-ray diffraction patterns of as-received and cold rolled (5 passes) ZrH₂ powders.

Figure 5. X-ray diffraction patterns of as-received and cold rolled (5 passes) ZrH₂ powders.

As shown in Figure 5, the X-ray patterns indicate that, in ZrH₂ sample, no other phases are present beside the hydride phase. It should be noticed that the peak broadening induced by cold rolling is not as important as for MgH₂ and TiH₂. In fact, Rietveld analysis indicated that, comparing as-received and cold rolled samples, the crystallite size goes from 29 nm to 20 nm while the microstrain goes from 0.2% to 0.3%.

3.3. X-ray Diffraction: Microwave Irradiation Effects

The X-ray diffraction patterns of as-received and cold rolled samples after microwave irradiation of MgH₂, TiH₂, and ZrH₂ are shown in Figure 6, Figure 7 and Figure 8, respectively. In the case of MgH₂ (Figure 6), both as-received and cold rolled samples were fully dehydrided by microwave irradiation. Nakamori et al. [27] investigated the effect of microwave irradiation in several hydrides and reported the ineffectiveness of that technique for desorption of MgH₂. In their case, the results may be explained by the limited increases of temperature they recorded. However, for ZrH₂ and TiH₂ relatively high temperatures were reached (more than 700 K and 800 K, respectively) and minimal hydrogen desorption was recorded. Therefore, the microwave heating and resulting dehydriding observed in our experiment might be ascribed to the use of a microwave susceptor (metallic cylinder). In our work, the metallic cylinder used as susceptor ensured a close contact with the hydride powder which remained confined between the susceptor and the sample holder wall ensuring an effective heating of the hydride powder. The efficiency on the use of susceptors for microwave irradiation and hydrogen desorption in MgH₂ was demonstrated by Zhang et al. [29] who designed a honeycomb ceramic monolith susceptor for this purpose. Awad et al. [30] prepared MgH₂ composites adding different types of carbon compounds (carbon fiber, graphite, and diamond, which are good microwave absorbers) during ball-milling and observed the effectiveness of this approach in increasing the hydride decomposition during irradiation.

Figure 6. X-ray diffraction patterns of as-received and cold rolled (5 passes) MgH₂ powders after microwave irradiation.

Figure 6. X-ray diffraction patterns of as-received and cold rolled (5 passes) MgH₂ powders after microwave irradiation.

The XRD patterns of TiH₂ samples after microwave irradiation are shown in Figure 7. In both samples (as-received and cold rolled), the hydride decomposition was not complete. The positive effect of cold rolling on the hydrogen desorption is clear from the relative intensities of the diffraction peaks of hydrides and Ti metal phases. From Rietveld refinement we found that, after microwave irradiation, the as-received sample was made of 93 wt.% of titanium hydride (TiH₂ and TiH_1.5) and only 7 wt.% of titanium. In the case of cold rolled sample, after microwave irradiation the proportion of TiH₂ was 46 wt.%, the rest being titanium.

Nakamori et al. [27] highlighted that TiH₂ is suitable for microwave heating (microwave absorber) but showed limited hydrogen desorption. This behavior was ascribed to a limited electromagnetic penetration depth in TiH₂, deduced by these authors as 11 μm for their experimental condition, leading to hydride decomposition only close to the surface. In our work, the hydrogen desorption from the TiH₂ sample was much higher than that reported in Nakamori’s work. These results demonstrated the improvement in hydrogen desorption from TiH₂ promoted by cold rolling.

Figure 7. X-ray diffraction patterns of as-received and cold rolled (5 passes) TiH₂ powders after microwave irradiation.

Figure 7. X-ray diffraction patterns of as-received and cold rolled (5 passes) TiH₂ powders after microwave irradiation.

Cold rolling was most effective in the case of ZrH₂. As shown in Figure 8, the pattern for the as-received sample display only peaks during to the hydride phase which means that microwave heating was totally ineffective. On the contrary, the pattern of the cold rolled sample did not show any ZrH₂ peaks but instead only showed peaks that could be indexed to the Zr metal phase and zirconium oxide. Considering the high thermodynamic stability of ZrH₂, the obtained result is very interesting and may open new possibilities for fast hydride and dehydride Zr and its alloys, which is an interesting topic in several areas, such as the nuclear industry.

As for TiH₂, the absence of dehydrogenation for the as-received sample may be due to by the small electromagnetic field penetration depth in Zr and the high thermodynamic stability of the ZrH₂ phase which prevented the hydride decomposition. By cold rolling, the particle size is reduced and defects are generated which contribute positively to the effectivity of microwave heating. Concerning the positive effect of cold rolling on hydrogen desorption, similar behavior has been observed for the conventional heating of cold rolled hydride powders [31,39,40]. This behavior has been attributed to the decrease in crystallite size and crystal defects generated during rolling which could act as fast diffusion paths and heterogeneous nucleating sites for diffusional phase transformations. Similar behavior may be expected for microwave heating. Particularly concerning ZrH₂, beneficial effect of cold rolling for improving the hydrogen desorption kinetics in conventionally heated ZrH₂ was reported in our previous investigation on the hydriding and dehydriding behavior of Zircaloy chips [32].

Figure 8. X-ray diffraction patterns of as-received and cold rolled (5 passes) ZrH₂ powders after microwave irradiation.

Figure 8. X-ray diffraction patterns of as-received and cold rolled (5 passes) ZrH₂ powders after microwave irradiation.

It should be noticed that after microwave irradiation in a few cases there was formation of oxides while in other cases no oxides were formed. For example, MgH₂ in as-received and cold rolled states presented formation of oxides after microwave irradiation while for TiH₂ no oxides were formed. The explanation may be that when a hydride phase subsisted, it prevented oxidation of the material. As a proof we see that for ZrH₂ the as-received sample after microwave irradiation did not desorbed at all and there were no oxides formed while the cold rolled sample fully desorbed and there is the presence of zirconium oxide.

From these experiments and by comparing with the results of Nakamori et al. [27], it is clear that the use of a susceptor greatly improved the effectiveness of microwave irradiation. A critical measurement that is missing in this investigation is the temperature during the reaction. This lack of recording temperature was due to the interaction between microwave and thermocouple. We are planning to solve this problem in a future investigation. Nevertheless, as MgH₂ decomposed to magnesium powder after microwave irradiation, we know that melting temperature was not reached, which means that in our experiments the temperature reached was probably not so different than in Nakamori’s case.

4. Conclusions

Microwave irradiation in binary metal hydrides was investigated. In the case of MgH₂ the feasibility of microwave heating was observed and its complete dehydriding was made possible by using a susceptor in a simple experimental setup. Despite the low penetration depth of the electromagnetic field in TiH₂, previously reported in the literature, extensive desorption was achieved after cold rolling this hydride powder. As ZrH₂ is much more stable than TiH₂, decomposition of as-received ZrH₂ was even more difficult than TiH₂ under our experimental conditions. On the other hand, full desorption of ZrH₂ was reached after cold rolling this sample. Therefore, for both TiH₂ and ZrH₂ cold rolling of the hydride powders was shown to be a feasible approach for improving the hydrogen desorption. The beneficial effect of cold rolling on the hydride decomposition is attributed to a combination of microstructural and morphological changes promoted by this processing route, which comprises particle size reduction, crystallite size reduction, and increase of residual stress. In the case of particle size reduction, this allows a more effective heating of the particles by approximating the size of the particles to the penetration depth of the electromagnetic field in metallic powders. Moreover, residual stress and crystallite size reduction can increase the number on nucleation sites and accelerate hydrogen diffusivity contributing to effective hydrogen desorption from the hydride powders.