Properties of activated MgH2 + mischmetal nanostructured composite produced by ball-milling

Properties of activated MgH2 + mischmetal nanostructured composite produced by ball-milling MgH + mischmetal nanostructured composite was synthesized from MgH plus 6 and 10 wt% of mischmetal by ball- 2 2 milling at various times. XRD studies revealed that cerium hydride was produced during the milling in all samples. Sievert test results indicated that the samples containing 6 wt% of mischmetal showed a higher desorption compared with the ones containing 10 wt% of mischmetal. The high amount of cerium hydride in the samples may be the reason, while hydrogen desorption properties decreased by adding more catalyst. Furthermore, BET results showed that the addition of the catalyst to the samples resulted in agglomerate formation in shorter milling times. The agglomerate formation increased with adding more amounts of mischmetal, thus decreasing the hydrogen desorption properties of the composite. The best results were obtained from the 30 h-milled sample containing 6 wt% of catalyst. The on-set desorption temperature of this sample was 100 °C lower than that of as-received MgH . Keywords Magnesium hydride · Milling · Catalyst · Mischmetal · Hydrogen storage Introduction disadvantages such as the need for low temperature in liquid storages or hazards and inefficiencies in the pressurized gas Supplying energy through fossil fuels creates challenges method. such as global warming and climate change due to the Extensive research on magnesium and its alloys for use as emission of greenhouse gases, pollution of urban areas, and hydrogen storage has been carried out due to its high storage reduction of oil resources [1]. Because of its abundance and capacity and low price [3]. Magnesium-based hydrides also lack of contamination, hydrogen is among the best options have good application properties such as thermal resistance, for replacing fossil fuels. Finding a safe method for hydro- reversibility, and recycling capability. Hence, in recent years, gen storage is one of the challenges of using hydrogen as more attention has been paid to research on magnesium and a new fuel. Hydrogen can be stored as a pressurized gas, its alloys. MgH has a high hydrogen storage capacity of liquid gas, and as solid-state hydrogen in materials such as 7.7 wt% with low production costs due to its abundance and magnesium hydride (MgH ) [2]. The storage of hydrogen recyclability [4]. in the form of reversible metal hydrides is the main alterna- The hydrogen desorption temperature of MgH is 327 °C tive for low-capacity methods such as storage under pressure under the hydrogen pressure of 1  bar. However, various or in liquid form at low temperatures. Hydrides have the temperatures of more than 400 °C have also been reported potential to store hydrogen in low-volume reservoirs without depending on the activation process of MgH . Although this temperature is lower than that of other hydrides, it is extremely high for practical applications [5]. The main dis- advantages of MgH , as a hydrogen storage medium, are * Shahram Raygan its high hydrogen desorption temperature, slow kinetics of shraygan@ut.ac.ir hydrogen absorption, and high tendency to react with oxy- School of Metallurgy and Materials Engineering, College gen [6]. These problems can be solved by methods such as of Engineering, University of Tehran, Tehran, Iran mechanical milling and catalyst addition. Department of Material Engineering, Hamedan University of Technology, Hamadan, Iran Vol.:(0123456789) 1 3 15 Page 2 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 One of the most common methods for improving the composite containing titanium demonstrated the fastest sorption of hydrogen from magnesium hydrides is ball-mill- absorption kinetics. The added elements helped combine ing. Mechanical milling increases the free surface, creates hydrogen atoms in a catalytic reaction. The catalytic activ- micro/nanostructures, enhances grain boundaries, decreases ity of the added elements was influenced by the high num- grain size, creates a porous surface structure with high active ber of structural defects, low compound stability, chemical sites for the adsorption and desorption of hydrogen, creates composition of the intermediate metal ions in the composi- micro-strains, and forms defects on the surface or inside tion, and high prevalence of hydrogen ion–metal ions [14]. the material. The created defects encourage the diffusion of High-energy ball-milling also plays a role in the synthesis hydrogen into the material by decreasing the diffusion acti- of hydrides and catalysts composites, creating a good inter- vation energy. Milling also changes the microstructure and face between hydride and catalyst and promoting the proper size of magnesium/magnesium hydride crystals and grains, distribution of the catalyst [15]. exerting a great inu fl ence on the sorption of hydrogen. These The enthalpy of hydrogenation reaction can be reduced by factors improve kinetics and surface activation to absorb or alloying magnesium with some elements. Thus, alloying is a desorb hydrogen, decrease the activation energy, reduce the traditional and efficient solution for improving the thermo- temperature of desorption, and improve kinetics and hydro- dynamics of hydrogen absorption and desorption from mag- gen diffusion in the material. As already mentioned, milling nesium. As an example, Ni Mg can react with hydrogen and reduces the size of grains and crystallites. These two factors form Mg NiH . The hydrogenation enthalpy of Mg NiH 2 4 2 4 have a great influence on the kinetics of hydrogen absorp- (− 64.5 kJ/mol) is less than the enthalpy of MgH forma- tion/desorption. A smaller crystallite size is proportional to tion (− 75 kJ/mol) [10]. Morinaga et al. [16] found that the more grain boundaries. Due to the low density of atoms in binding energy between hydrogen and nickel is stronger than grain boundaries, the diffusion of hydrogen is usually faster that of magnesium in Mg NiH . This nickel–hydrogen bond 2 4 in grain boundaries than inside grains. In addition, grain in Mg NiH is weaker than that of magnesium and hydrogen 2 4 boundaries are suitable nucleation sites for the formation or in pure MgH , resulting in the low formation enthalpy of decomposition of the hydride phase. Reducing the crystallite Mg NiH . Magnesium and copper can create a Cu Mg alloy. 2 4 2 size of the powder material increases the surface area and During the absorption of hydrogen, Cu Mg is decomposed decreases the length of the hydrogen diu ff sion path. The area into MgH and MgCu and the equilibrium temperature of 2 2 is proportional to the rate of surface reaction with hydrogen, hydrogen desorption under 1 bar H pressure decreases to while the length of the hydrogen diffusion path is propor - 240 °C. However, this reaction is not reversible [12]. tional to the kinetics of hydride formation or decomposi- In this research, the nanostructured composite of tion [7–10]. Wagemans et al. [11] investigated the hydrogen MgH –mischmetal was prepared by ball-milling to inves- desorption properties of MgH using quantum mechanical tigate the simultaneous effect of milling and addition of calculations and found that MgH clusters with nanoscale mischmetal as a catalyst. Findings of various studies have dimensions had a significantly lower (63 kJ/mol) desorp- revealed that some elements have a good catalytic effect on tion enthalpy than MgH . Thus, hydrogen desorption occurs the process of hydrogen absorption and desorption from at lower temperatures for nano-sized magnesium clusters. MgH . Nevertheless, few studies have been conducted on The desorption temperature of magnesium clusters (with the the effect of rare earth elements on the hydrogen desorption approximate size of 0.9 nm) is calculated only at 200 °C. It properties of MgH . Therefore, a combination of cerium and is reported that ball-milling also increases the rate of hydro- lanthanum transition metals (mischmetal) as the catalyst was gen absorption and desorption from MgH by removing the employed in this study. oxide layers on the surface [12]. The use of catalysts is one way to improve surface kinet- Materials and research process ics. Therefore, different catalysts such as metal oxides, intermetallic compounds, and carbon materials have been In this study, MgH (Alfa Aesar < 140  μm, purity > 95%) used to improve the properties of hydrides [3–7]. The desir- and small slices cut from a bulk of mischmetal containing able catalytic effect of metals and their oxides has also been 65.37 wt% of cerium, 34.37 wt% of lanthanum, 0.1 wt% of demonstrated on hydrogen desorption/absorption in mag- neodymium, and 0.1 wt% of praseodymium were utilized nesium-based hydrides. Liang et al. [13] mixed MgH with as raw materials. MgH was ball-milled for 10, 30, and 40 h 2 2 titanium, vanadium, manganese, iron, and nickel through to examine the effect of ball-milling on the properties of the mechanical milling. All the added elements reduced the used powder. To prepare the nanostructured composite, 3 g activation energy of hydrogen desorption from MgH . The of MgH powder with the addition of 6 wt% of mischmetal 2 2 MgH –vanadium composite showed the fastest kinetics was ball-milled for 10, 20, 30, and 40 h. Moreover, to study of desorption and its activation energy (62.3 kJ/mol) was the effect of the amount of catalyst on the hydride properties much less than that of MgH (120 kJ/mol). Meanwhile, the of the nanostructured composites, 10 wt% of mischmetal 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 3 of 11 15 was added to 3 g of MgH powder and then ball-milled for field-emission scanning electron microscope were utilized. 20 and 30 h. Particle size was calculated using the MIP image analysis Ball-milling was performed in a high-energy planetary software in the form of a circle diameter equal to the parti- 1/2 ball mill (PM2400 Asia Rakhsh Co.) using a hardened chro- cle size ECD = (4A/π) . In this formula, A represents the mium-nickel steel (150 mL) vial and hardened steel balls magnitude of the surface of the particle. Particle size was (5, 10, and 20 mm in diameter) with ball to powder weight determined by several images and by specifying 300–700 ratio of 20 and rotation speed of 250 rpm under argon atmos- particles in each measurement. To measure the free surface phere. It is reported that, in ball-milling, different ball sizes of the samples (BET test), the BELSORP-mini II device are sometimes mixed to randomize the motion of balls and with liquid nitrogen was used. Before BET experiments, the increase the efficiency of milling [ 17]. After the completion samples were heated up to 180 °C and maintained for 4 h of the ball-milling process, the vial was opened in a glove to remove all possible moisture. The main parameters were box under high-purity argon gas to prevent the oxidation of 195.65 °C for 55 min and 10 points BET measurement. To samples. measure the properties of hydrogen desorption, a home- To recognize and analyze the phases, X-ray diffraction made Sievert apparatus was employed. Before the experi- (XRD) (Philips X’Pert Pro diffractometer) with CuK radia- ments, all the samples were subjected to activation using tion, at the wavelength of λ = 0.1541874 nm, in the range of a Sievert machine under the hydrogen pressure of 4 Mpa 2θ = 20–100°, and step size of 0.02° was used. The X’Pert at 180° C for 4 h. A sample of 300 mg was placed in the HighScore Plus v2.2b software of PANalytical Company Sievert apparatus and washed with argon gas. The samples was employed to identify the phases. Mean crystallite size were then heated up to 500 °C with the rate of 10 °C/min. and the lattice micro-strain of the particles were measured Finally, the results were reported as the curve of the amount using the Williamson–Hall method [18]: of hydrogen desorbed versus temperature. All Sievert tests were performed at least twice to ensure reproducibility. cos  = K∕ + 2 sin , (1) sample where β is the full width at half maximum (FWHM) of sample the milled powder, θ is the position of peak maximum, K is Results and discussion the Scherrer constant (about 0.9), λ is the beam wavelength, δ is the crystallite size, and ε is the lattice micro-strain intro- Phase identifying duced by ball-milling. For instrumental correction, a Gauss- ian relationship was used [19]: Figures 1 and 2 depict the XRD patterns of mischmetal and MgH , respectively. In Fig.  2, low-intensity magnesium 2 2 =  −  , (2) sample peaks are observed, which were predictable according to experimental instrumental the > 95 wt% purity of primary MgH . where β is the measured FWHM of the annealed experimental Figure  3 illustrates the XRD patterns of as-received nickel powders. MgH and ball-milled MgH . According to these pat- 2 2 To explore the morphology of the samples, a Tescan- terns, increasing the milling time increased the width and MV 2300 scanning electron microscope and Zeiss-Sigma reduced the intensity of the peaks. Moreover, some of the Fig. 1 XRD pattern of the mis- chmetal used in this study 1 3 15 Page 4 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 Fig. 2 XRD pattern of the magnesium hydride used in this study Fig. 3 XRD patterns of a as-received MgH and MgH 2 2 milled for b 10, c 30, and d 40 h weak peaks disappeared by increasing the milling time due Table 1 Mean crystallite size and lattice strain of the ball-milled to the reduction of crystallite size as well as the increase MgH in lattice strain and internal energy during milling [20]. In Time of ball-milling (h) Mean crystallite size (nm) Lattice the ball-milled samples, a weak peak was observed from strain magnesium oxide. The formation of this compound may (%) be due to the presence of Mg in the initial MgH and ten- 0 87 – dency towards oxidation during ball-milling. It is well- 10 58 0.72 known that ball-milled powders are very active. One rea- 30 35 0.95 son for the oxidation of active Mg during ball-milling may 40 32 1.07 be the presence of oxygen as an impurity in the flowing argon gas especially when the majority of argon in the gas container is used and the amount of gas in the container is powder to 32 nm for the 40 h-milled M gH , and lattice less than 1/4 of container volume. Another reason can be strain increased to 1.07% due to the 40 h milling. the leakage of glove box and entrance of oxygen into it. Oxidation can also be performed during the movement of Morphology of ball‑milled MgH samples from the laboratory towards the XRD machine. Table 1 shows the variation of crystallite size and lattice The SEM images of as-received MgH and ball-milled strain of MgH with ball-milling. It can be observed that MgH (10, 30, and 40 h) are depicted in Fig. 4. Based on crystallite size was reduced from 87 nm for the unmilled Fig. 4b, after 10 h of milling, MgH particles became spheri- cal and their mean size reached 338 nm. The presence of 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 5 of 11 15 Fig. 4 Scanning electron microscopy (SEM) images of a as-received MgH and MgH 2 2 ball-milled for b 10, c 30, and d 40 h small particles along with the coarse particles or agglom- erate of particles in the sample milled for 10 h shows the non-uniform distribution of particle size in this sample. With an increase in ball-milling time from 10 to 30 h, the parti- cles were slightly smaller and the mean size of the particles reached 306 nm. An increase in the ball-milling time up to 30 h also led to a more uniform distribution of the par- ticle size. The particles were agglomerated in the sample milled for 40 h and the mean size of the particles increased to 372 nm. After milling for a certain period, a state of equi- librium was created and fine particles became larger, while coarse particles became smaller due to the balance between the cold-welding rate (which tended to increase the mean particle size) and the rate of fracture (that led to reduce the mean particle size). The mean particle size eventually reached to the equilibrium size [21]. It can be assumed that no reduction in the mean particle size would be observed at longer milling times. Fig. 5 Hydrogen desorption behavior of as-received MgH and ball- milled MgH for 10, 30, and 40 h Eec ff t of ball‑milling on the hydrogen desorption properties of  MgH temperature can be the surface oxidation and the need for The results of hydrogen desorption studies on as-received more activation. However, due to the uniformity of activa- and ball-milled MgH (10, 30, 40 h) are illustrated in Fig. 5. tion conditions for all samples, the results are comparable The on-set desorption temperature of MgH decreased from 2 with each other. The difference between the on-set desorp- 480  °C for as-received MgH to 400  °C for 40  h-milled tion temperature of as-received MgH and 40 h ball-milled MgH in the condition of these experiments. The on-set des- 2 MgH (400 and 480 °C) is the increased specific surface area orption temperature of as-received MgH has been reported and lattice strain, reduced size of crystallites, and formation to be above 300 °C [1, 5]. The reason for the difference of structural defects in the material, which facilitate hydro- in temperature measured in this study with the reported gen desorption. The positive role of milling in reducing 1 3 15 Page 6 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 the hydrogen desorption temperature has been confirmed. Nevertheless, with increasing the milling time from 30 to 40 h, the desorption curve slightly shifted to the right (higher temperature), perhaps due to the agglomeration of particles and reduction of their free surface area. Therefore, to further decrease the temperature of hydrogen desorption, the cata- lyst was added to MgH prior to the milling process. Eec ff t of catalyst addition on  MgH Phase identifying of  MgH : 6 wt% of mischmetal The XRD patterns of MgH ball-milled for 10 h as well as Fig. 7 The XRD pattern of 30  min ball-milled MgH -6  wt% of mis- chmetal the MgH containing 6 wt% of mischmetal milled for 10, 20, 30, and 40 h are shown in Fig.  6. By comparing the XRD patterns of MgH milled for 10 h and MgH contain- 2 2 ing 6 wt% of mischmetal milled for 10 h, a broadening and which was ball-milled for 30 min. It was observed that the intensity of the cerium hydride peak was low, indicating the decrease in the peak intensities of the catalyst-containing sample can be observed. This is due to the decrease in crys- formation of a low amount of this compound at short milling times. However, as the milling time increased, the intensity tallite size and increase in lattice strain through the addition of catalyst (mischmetal). Catalyst particles can act as hard of the cerium hydride peak increased in the samples. It can be concluded that an increment in the milling time led to a balls during milling and contribute to structural defects [22]. Peaks of CeH were observed in the catalyst-containing higher amount of hydride in the samples. Cerium hydride 2.73 may be formed on the surface of cerium. It has been reported samples. Considering the lower Gibbs free energy of forma- tion for cerium hydride (− 204.8 kJ/mol H ) in comparison that this surface layer controls the reaction rate of the for- mation of cerium hydride [24]. Creating new surfaces and with MgH (− 75 kJ/mol H ), it seems that cerium reacted 2 2 with MgH during high-energy ball-milling according to fracture of cerium hydride layers on the surface of cerium due to mechanical milling facilitates the formation of cerium Eq. 3 and formed cerium hydride [23]. hydrides. The formation of hydrides during the milling pro- Ce + MgH → CeH + Mg ΔG =−129.8 kJ/mol. 2 2 (3) cess has also been reported by Shang et al. [23]. This layer Due to the non-equilibrium nature of the milling process, also affects the amount of hydrogen desorption from the the formation of non-stoichiometric hydrides of cerium such powder mixture. as CeH was also expected. Figure 7 shows the diffrac- 2.73 tion pattern of a sample containing 6 wt% of mischmetal Particle morphologies Figure 8 shows the SEM images of MgH -6 wt% of mis- chmetal ball-milled for 10, 20, 30, and 40 h. It has been reported in Sect. 2.2 that the mean particle size of the 10 h ball-milled MgH is 338 nm, while the particle size of the 10 h ball-milled MgH -6 wt% of mischmetal is 287 nm. By increasing the milling time of MgH -6 wt% of mischmetal from 10 to 20 h, the particle size increased and the mean particle size reached 381 nm. In the 30 h-milled sample, the mean particle size was 360 nm. At shorter milling times, large and small particles were created together. However, with increasing the milling time and approaching the equi- librium level, small particles stuck together and large parti- cles broke into small ones [21]. The increased particle size in the 20 h-milled sample can be attributed to the agglom- eration of fine particles and reduction of particle size in the 30 h-milled sample can be due to the fracture of large par- Fig. 6 The XRD pattern of a 10  h ball-milled MgH and ticles. By increasing this time to 40 h, the particles stuck MgH -6wt % of mischmetal ball-milled for b 10 h, c 20 h, d 30 h, and e 40 h together and agglomerated with the mean size of 417 nm. By 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 7 of 11 15 Fig. 8 The SEM images of MgH -6 wt% of mischmetal ball-milled for a 10, b 20, c 30, and d 40 h comparing these results with those of the samples without a was observed. The increase in the milling time up to 30 h catalyst, it can be seen that smaller particles were achieved led to the fracture of large particles, resulting in an increase in short ball-milling times by adding the catalyst. This can in free surface relative to the 20 h-milled sample. During the be attributed to the role of the catalyst in the application ball-milling process, the removing, mixing and shrinkage of of more work on particles, resulting in cold-welding and pores and forming new porosities occurred simultaneously particle-bonding [21]. by agglomeration of new particles [21]. According to the results of the BET test, it seems that the agglomeration and Determination of specific surface area (BET) removal of porosities resulted in a reduced free surface and mean pore volume in the 40 h ball-milled MgH -6 wt% of Free surface changes, mean porosity diameter, and mean mischmetal sample. pore volume for the MgH samples milled for 10 h and the MgH samples with 6 wt% of catalyst milled for 10, 20, Hydrogen desorption properties 30, and 40 h are shown in Fig. 9. According to these dia- grams, the sample with the catalyst that is milled for 10 h Figure 10 illustrates the results of the hydrogen desorption of has the highest free surface with the smallest mean diameter MgH, MgH milled for 10 h, and MgH with 6 wt% of cata- 2 2 2 of porosity and the highest pore volume. By comparing the lyst milled for 10, 20, 30, and 40 h. By comparing the MgH MgH sample milled for 10 h with the MgH sample with samples milled for 10 h with MgH with the catalyst milled 2 2 2 the catalyst milled for 10 h, it can be observed that the addi- for 10 h, it was observed that the addition of catalyst reduced tion of the catalyst increased the free surface and mean pore the temperature of hydrogen desorption and decreased the volume, but decreased the pore size. It can be concluded amount of desorbed hydrogen. The decrease in the on-set that the addition of the catalyst led to a smaller particle size, desorption temperature of hydrogen can be due to the posi- which is also confirmed by SEM images. By increasing the tive effect of the catalyst on the hydrogen desorption pro- milling time to 20 h for the catalyst containing MgH , the cess by increasing the free surface and accumulated energy free surface area and mean pore volume decreased and the in the milled powder and reducing the activation energy. mean diameter of the pores increased. According to the The reduction in the amount of desorbed hydrogen can be results of SEM images, it can be concluded that the closing due to the formation of the cerium hydride phase during the and mixing of the porosities occurred simultaneously with milling process. The cerium hydride phase formed during the particle size increase, and hence the free surface loss ball-milling was stable at high temperatures [23] and, thus, 1 3 15 Page 8 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 the formation of this phase reduced the amount of desorbed hydrogen from the milled MgH -catalyst mixture. Increasing the milling time to 20 and 30 h also reduced the tempera- ture of hydrogen desorption. By comparing the results of MgH -6 wt% of mischmetal milled for 10, 20, and 30 h and considering the results of the BET test, it can be observed that the on-set desorption temperature of the sample milled for 10 h is more than that of the samples ball-milled for 20 and 30 h despite the lower free surface in the 20 and 30 h-milled mixtures. This can be attributed to the increase in the internal energy, formation of defects in the structure, and reduction in the diffusion path of hydrogen atoms. The on-set desorption temperature of hydrogen in the 40 h ball- milled sample increased compared to the 30 h ball-milled sample. This is because of the increase in particle size and free surface loss in the 40 h ball-milled sample due to the cold-welding phenomenon. Eec ff t of increasing the amount of the catalyst To examine the effect of the amount of catalyst on the hydro- gen desorption properties of MgH , the MgH + 10 wt% of 2 2 mischmetal powder mixture was prepared by ball-milling for 20 and 30 h. Figure 11 shows the X-ray diffraction patterns of the samples containing 6 and 10 wt% of catalyst which were milled for 20 and 30 h. Table 1 also shows the crystallite size and lattice strain of MgH -10 wt% of mischmetal ball-milled for 20 and 30 h and MgH -6wt % of mischmetal ball-milled for 20 and 30 h. Fig. 9 Free surface, mean porosity diameter, and mean pore volume Similar to the samples containing 6 wt% of catalyst, cerium for the 10 h ball-milled MgH and MgH -6 wt% of mischmetal ball- 2 2 milled for 10, 20, 30, and 40 h hydride peaks were observed in the samples containing 10 wt% of catalyst. By adding more catalyst in the same milling time, the peaks of the XRD pattern became wider. Fig. 10 Hydrogen desorption behavior of as-received MgH , 10 h ball-milled MgH , and MgH -6 wt% of mischmetal ball-milled for 10, 20, 30, and 40 h 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 9 of 11 15 time. This increase in the mean particle size also showed the effect of catalyst addition on the achievement of equilibrium between cold-welding and fracture at lower times of ball- milling. Subsequently, cold-welding became the predomi- nant process and agglomeration occurred. Thus, particle size was higher in MgH -10 wt% of mischmetal due to the further agglomeration. Agglomeration can also be seen in Fig. 12b for the 30 h-milled powder. The hydrogen desorption behavior of MgH -6 wt% of mischmetal and MgH -10 wt% of mischmetal ball-milled for 20 and 30 h, and also of the 30 h-milled MgH can be observed in Fig. 13. The latter sample helped understand the effect of catalyst addition. It was observed that an increase in the amount of catalyst led to a reduction in the amount of hydrogen desorption because of the further formation of cerium hydride with increase in the catalyst. Increasing the Fig. 11 XRD patterns of MgH -6 wt % of mischmetal ball-milled for amount of catalyst from 6 to 10 wt% also resulted in a 15 °C a 20, c 30, and MgH -10 wt% of mischmetal ball-milled for b 20 and increment in the on-set hydrogen desorption temperature of c 30 h the sample milled for 20 h and a 20 °C increase in the on- set hydrogen desorption temperature of the sample milled Table 2 Mean crystallite size and lattice strain of the ball-milled for 30 h. Shang et al. [23] explained the rise in hydrogen MgH -10 wt% and MgH -6 wt% of mischmetal after 20 and 30 h of 2 2 desorption temperature as a result of the formation of a sta- ball-milling ble layer of cerium hydride on particle surfaces, stating that Sample Time of Mean crys- Lattice the creation of this surface layer would be problematic for ball milling tallite size strain the diffusion of hydrogen from inside the particles outward. (h) (nm) (%) This seems to be true for high amounts of cerium hydride. In addition, it is possible to consider the increase of particle MgH -6 wt% of mischmetal 20 33 1.02 MgH -10 wt% of mischmetal 20 33 1.08 size and reduction of free surface as a reason for increasing the temperature of hydrogen desorption. MgH -6 wt% of mischmetal 30 33 1.19 MgH -10 wt% of mischmetal 30 30 1.65 Conclusion Table 2 confirms the role of adding catalyst in creating lat- tice strain in crystallites in comparison with MgH (Fig. 3). The results of this research can be summarized as follows: It is clear from Fig. 10 that adding to the amount of catalyst at a constant milling time led to more lattice strain in the 1. The XRD patterns of as-received MgH and ball- samples.milled MgH show that increasing the milling time led According to the SEM images of MgH -10 wt% of mis- to an increase in the width and reduction in the inten- chmetal milled for 20 and 30 h in Fig. 12, the mean parti- sity of peaks. It is shown that crystallite size reduced cle size of the 20 h-milled sample increased in comparison from 87 nm for the unmilled powder to 32 nm for the with the MgH -6 wt% of mischmetal milled for the same Fig. 12 SEM images of MgH -10 wt% of mischmetal ball-milled for a 20 and b 30 h 1 3 15 Page 10 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 Fig. 13 Hydrogen desorption behavior of MgH -6 wt% of mischmetal and MgH -10 wt% of mischmetal milled for 20 and 30 h and also of the 30 h-milled MgH Open Access This article is distributed under the terms of the Crea- 40 h-milled MgH and the lattice strain increased to tive Commons Attribution 4.0 International License (http://creat iveco 1.07% due to 40 h of milling. mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- 2. Increasing the milling time up to 40 h resulted in an tion, and reproduction in any medium, provided you give appropriate improvement in the hydrogen desorption properties of credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. MgH and 80 °C reduction in hydrogen desorption tem- perature 3. By adding the mischmetal, cerium reacted with MgH References during ball-milling and formed cerium hydride. It seems that the amount of this hydride increased with an incre- 1. Varin, R.A., Czujko, T., Wronski, Z.S.: Nanomaterials for solid ment of mischmetal from 6 to 10 wt%. This hydride was state hydrogen storage. Springer Science & Business Media, also stable at high temperatures, resulting in the harder New York (2009) desorption of hydrogen in large quantities. 2. Ogden, J.M.: Developing an infrastructure for hydrogen vehi- cles: a Southern California case study. Int. J. Hydrogen Energy 4. Comparing the XRD results of 10 h-milled MgH -6 wt% 24, 709–730 (1999) of mischmetal to 10 h-milled MgH revealed that crys- 3. Grochala, W., Edwards, P.P.: Thermal decomposition of the non- tallite size decreased and the lattice strain increased by interstitial hydrides for the storage and production of hydrogen. the addition of the catalyst (mischmetal). Chem. Rev. 104, 1283–1316 (2004) 4. Alefeld, G., Volkl, J.: Topics in Applied Physics. Hydrogen In 5. Mischmetal exhibited a suitable catalytic effect on the Metals. I.-Basic Properties, vol. 28. Springer, New York (1978) hydrogen desorption properties of MgH and, in the best 5. Walker, G.: Solid-State Hydrogen Storage: Materials and Chem- case, the reduction in hydrogen desorption temperature istry. Elsevier, Amsterdam (2008) of the sample containing 6 wt% of mischmetal milled 6. Zaluska, A., Zaluski, L., Ström-Olsen, J.: Nanocrystalline mag- nesium for hydrogen storage. J. Alloy Comd. 288, 217–225 for 30 h was 100 °C.Therefore, the powder containing (1999) 6 wt% of mischmetal milled for 30 h is the most suitable 7. Guoxian, L., Erde, W., Shoushi, F.: Hydrogen absorption and sample from the desorption point of view. desorption characteristics of mechanically milled Mg 35 wt% 6. Increasing the amount of mischmetal from 6 to 10 wt% FeTi1. 2 powders. J. alloy compd. 223, 111–114 (1995) led to an increment in the hydrogen desorption tempera- ture of the milled MgH –mischmetal mixture. 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 11 of 11 15 8. Wang, P., Wang, A., Zhang, H., Ding, B., Hu, Z.: Hydriding 17. Takacs, L.: Self-sustaining reactions induced by ball milling. Prog. properties of a mechanically milled Mg–50 wt% ZrFe 1.4 Cr Mater Sci. 47, 355–414 (2002) 0.6 composite. J. alloy compd. 297, 240–245 (2000) 18. Williamson, G., Hall, W.: X-ray line broadening from filed alu- 9. Li, F., Jiang, L., Du, J., Wang, S., Liu, X., Zhan, F.: Synthesis minium and wolfram. Acta Metall. Mater. 1, 22–31 (1953) and hydrogenation properties of Mg–La–Ni–H system by reactive 19. Hosseini-Gourajoubi, F., Pourabdoli, M., Uner, D., Raygan, S.: mechanical alloying. Int. J. hydrogen Energy 31, 581–585 (2006) Effect of process control agents on synthesizing nano-structured 10. Barkhordarian, G., Klassen, T., Bormann, R.: Kinetic investi- 2  Mg–9Ni–Y catalyst by mechanical milling and its catalytic gation of the effect of milling time on the hydrogen sorption effect on desorption capacity of MgH . Adv. Powder Technol. reaction of magnesium catalyzed with different Nb O contents. 26, 448–453 (2015) 2 5 J. Alloy. Compd. 407, 249–255 (2006) 20. Raygan, S., Pourabdoli, M., Abdizadeh, H., Medraj, M.: Synthe- 11. Wagemans, R.W.P., van Lenthe, J.H., de Jongh, P.E., Van Dil- sizing nanostructured Ni75Mg16.66Y8.34 (at%) powder by solid len, A.J., de Jong, K.P.: Hydrogen storage in magnesium clusters: state reaction and mechanical milling. Mater. Manuf. Process. 27, quantum chemical study. J. Am. Chem. Soc. 127, 16675–16680 1300–1305 (2012) (2005) 21. Suryanarayana C.: Mechanical Alloying and Milling. CRC Press 12. Zhu, M., Lu, Y., Ouyang, L., Wang, H.: Thermodynamic tun- (2004) ing of Mg-based hydrogen storage alloys: a review. Materials. 6, 22. Sakintuna, B., Lamari-Darkrim, F., Hirscher, M.: Metal hydride 4654–4674 (2013) materials for solid hydrogen storage: a review. Int. J. Hydrogen 13. Liang, G., Huot, J., Boily, S., Van Neste, A., Schulz, R.: Catalytic Energy 32, 1121–1140 (2007) effect of transition metals on hydrogen sorption in nanocrystalline 23. Shang, C., Guo, Z.: Structural and desorption characterisations of ball milled MgH –Tm (Tm = Ti, V, Mn, Fe and Ni) systems. J. milled (MgH + Y, Ce) powder mixtures for hydrogen storage. Int. 2 2 Alloy. Compd. 292, 247–252 (1999) J. Hydrogen Energy 32, 2920–2925 (2007) 14. Barkhordarian, G., Klassen, T., Bormann, R.: Catalytic mecha- 24. Sarussi, D., Jacob, I., Bloch, J., Shamir, N., Mintz, M.: The kinet- nism of transition-metal compounds on Mg hydrogen sorption ics and mechanism of cerium hydride formation. J. alloy compd. reaction. J. Phys. Chem. B 110, 11020–11024 (2006) 191, 91–99 (1993) 15. Dornheim, M., Doppiu, S., Barkhordarian, G., Boesenberg, U., Klassen, T., Gutfleisch, O., Bormann, R.: Hydrogen storage in Publisher’s Note Springer Nature remains neutral with regard to magnesium-based hydrides and hydride composites. Scripta jurisdictional claims in published maps and institutional affiliations. Mater. 56, 841–846 (2007) 16. Morinaga, M., Yukawa, H.: Nature of chemical bond and phase stability of hydrogen storage compounds. Mater. Sci. Eng. A 329, 268–275 (2002) 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Materials for Renewable and Sustainable Energy Springer Journals

Properties of activated MgH2 + mischmetal nanostructured composite produced by ball-milling

Free
11 pages

Loading next page...
 
/lp/springer_journal/properties-of-activated-mgh2-mischmetal-nanostructured-composite-qEi3g6b9jj
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Materials Science; Materials Science, general; Renewable and Green Energy; Renewable and Green Energy
ISSN
2194-1459
eISSN
2194-1467
D.O.I.
10.1007/s40243-018-0122-z
Publisher site
See Article on Publisher Site

Abstract

MgH + mischmetal nanostructured composite was synthesized from MgH plus 6 and 10 wt% of mischmetal by ball- 2 2 milling at various times. XRD studies revealed that cerium hydride was produced during the milling in all samples. Sievert test results indicated that the samples containing 6 wt% of mischmetal showed a higher desorption compared with the ones containing 10 wt% of mischmetal. The high amount of cerium hydride in the samples may be the reason, while hydrogen desorption properties decreased by adding more catalyst. Furthermore, BET results showed that the addition of the catalyst to the samples resulted in agglomerate formation in shorter milling times. The agglomerate formation increased with adding more amounts of mischmetal, thus decreasing the hydrogen desorption properties of the composite. The best results were obtained from the 30 h-milled sample containing 6 wt% of catalyst. The on-set desorption temperature of this sample was 100 °C lower than that of as-received MgH . Keywords Magnesium hydride · Milling · Catalyst · Mischmetal · Hydrogen storage Introduction disadvantages such as the need for low temperature in liquid storages or hazards and inefficiencies in the pressurized gas Supplying energy through fossil fuels creates challenges method. such as global warming and climate change due to the Extensive research on magnesium and its alloys for use as emission of greenhouse gases, pollution of urban areas, and hydrogen storage has been carried out due to its high storage reduction of oil resources [1]. Because of its abundance and capacity and low price [3]. Magnesium-based hydrides also lack of contamination, hydrogen is among the best options have good application properties such as thermal resistance, for replacing fossil fuels. Finding a safe method for hydro- reversibility, and recycling capability. Hence, in recent years, gen storage is one of the challenges of using hydrogen as more attention has been paid to research on magnesium and a new fuel. Hydrogen can be stored as a pressurized gas, its alloys. MgH has a high hydrogen storage capacity of liquid gas, and as solid-state hydrogen in materials such as 7.7 wt% with low production costs due to its abundance and magnesium hydride (MgH ) [2]. The storage of hydrogen recyclability [4]. in the form of reversible metal hydrides is the main alterna- The hydrogen desorption temperature of MgH is 327 °C tive for low-capacity methods such as storage under pressure under the hydrogen pressure of 1  bar. However, various or in liquid form at low temperatures. Hydrides have the temperatures of more than 400 °C have also been reported potential to store hydrogen in low-volume reservoirs without depending on the activation process of MgH . Although this temperature is lower than that of other hydrides, it is extremely high for practical applications [5]. The main dis- advantages of MgH , as a hydrogen storage medium, are * Shahram Raygan its high hydrogen desorption temperature, slow kinetics of shraygan@ut.ac.ir hydrogen absorption, and high tendency to react with oxy- School of Metallurgy and Materials Engineering, College gen [6]. These problems can be solved by methods such as of Engineering, University of Tehran, Tehran, Iran mechanical milling and catalyst addition. Department of Material Engineering, Hamedan University of Technology, Hamadan, Iran Vol.:(0123456789) 1 3 15 Page 2 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 One of the most common methods for improving the composite containing titanium demonstrated the fastest sorption of hydrogen from magnesium hydrides is ball-mill- absorption kinetics. The added elements helped combine ing. Mechanical milling increases the free surface, creates hydrogen atoms in a catalytic reaction. The catalytic activ- micro/nanostructures, enhances grain boundaries, decreases ity of the added elements was influenced by the high num- grain size, creates a porous surface structure with high active ber of structural defects, low compound stability, chemical sites for the adsorption and desorption of hydrogen, creates composition of the intermediate metal ions in the composi- micro-strains, and forms defects on the surface or inside tion, and high prevalence of hydrogen ion–metal ions [14]. the material. The created defects encourage the diffusion of High-energy ball-milling also plays a role in the synthesis hydrogen into the material by decreasing the diffusion acti- of hydrides and catalysts composites, creating a good inter- vation energy. Milling also changes the microstructure and face between hydride and catalyst and promoting the proper size of magnesium/magnesium hydride crystals and grains, distribution of the catalyst [15]. exerting a great inu fl ence on the sorption of hydrogen. These The enthalpy of hydrogenation reaction can be reduced by factors improve kinetics and surface activation to absorb or alloying magnesium with some elements. Thus, alloying is a desorb hydrogen, decrease the activation energy, reduce the traditional and efficient solution for improving the thermo- temperature of desorption, and improve kinetics and hydro- dynamics of hydrogen absorption and desorption from mag- gen diffusion in the material. As already mentioned, milling nesium. As an example, Ni Mg can react with hydrogen and reduces the size of grains and crystallites. These two factors form Mg NiH . The hydrogenation enthalpy of Mg NiH 2 4 2 4 have a great influence on the kinetics of hydrogen absorp- (− 64.5 kJ/mol) is less than the enthalpy of MgH forma- tion/desorption. A smaller crystallite size is proportional to tion (− 75 kJ/mol) [10]. Morinaga et al. [16] found that the more grain boundaries. Due to the low density of atoms in binding energy between hydrogen and nickel is stronger than grain boundaries, the diffusion of hydrogen is usually faster that of magnesium in Mg NiH . This nickel–hydrogen bond 2 4 in grain boundaries than inside grains. In addition, grain in Mg NiH is weaker than that of magnesium and hydrogen 2 4 boundaries are suitable nucleation sites for the formation or in pure MgH , resulting in the low formation enthalpy of decomposition of the hydride phase. Reducing the crystallite Mg NiH . Magnesium and copper can create a Cu Mg alloy. 2 4 2 size of the powder material increases the surface area and During the absorption of hydrogen, Cu Mg is decomposed decreases the length of the hydrogen diu ff sion path. The area into MgH and MgCu and the equilibrium temperature of 2 2 is proportional to the rate of surface reaction with hydrogen, hydrogen desorption under 1 bar H pressure decreases to while the length of the hydrogen diffusion path is propor - 240 °C. However, this reaction is not reversible [12]. tional to the kinetics of hydride formation or decomposi- In this research, the nanostructured composite of tion [7–10]. Wagemans et al. [11] investigated the hydrogen MgH –mischmetal was prepared by ball-milling to inves- desorption properties of MgH using quantum mechanical tigate the simultaneous effect of milling and addition of calculations and found that MgH clusters with nanoscale mischmetal as a catalyst. Findings of various studies have dimensions had a significantly lower (63 kJ/mol) desorp- revealed that some elements have a good catalytic effect on tion enthalpy than MgH . Thus, hydrogen desorption occurs the process of hydrogen absorption and desorption from at lower temperatures for nano-sized magnesium clusters. MgH . Nevertheless, few studies have been conducted on The desorption temperature of magnesium clusters (with the the effect of rare earth elements on the hydrogen desorption approximate size of 0.9 nm) is calculated only at 200 °C. It properties of MgH . Therefore, a combination of cerium and is reported that ball-milling also increases the rate of hydro- lanthanum transition metals (mischmetal) as the catalyst was gen absorption and desorption from MgH by removing the employed in this study. oxide layers on the surface [12]. The use of catalysts is one way to improve surface kinet- Materials and research process ics. Therefore, different catalysts such as metal oxides, intermetallic compounds, and carbon materials have been In this study, MgH (Alfa Aesar < 140  μm, purity > 95%) used to improve the properties of hydrides [3–7]. The desir- and small slices cut from a bulk of mischmetal containing able catalytic effect of metals and their oxides has also been 65.37 wt% of cerium, 34.37 wt% of lanthanum, 0.1 wt% of demonstrated on hydrogen desorption/absorption in mag- neodymium, and 0.1 wt% of praseodymium were utilized nesium-based hydrides. Liang et al. [13] mixed MgH with as raw materials. MgH was ball-milled for 10, 30, and 40 h 2 2 titanium, vanadium, manganese, iron, and nickel through to examine the effect of ball-milling on the properties of the mechanical milling. All the added elements reduced the used powder. To prepare the nanostructured composite, 3 g activation energy of hydrogen desorption from MgH . The of MgH powder with the addition of 6 wt% of mischmetal 2 2 MgH –vanadium composite showed the fastest kinetics was ball-milled for 10, 20, 30, and 40 h. Moreover, to study of desorption and its activation energy (62.3 kJ/mol) was the effect of the amount of catalyst on the hydride properties much less than that of MgH (120 kJ/mol). Meanwhile, the of the nanostructured composites, 10 wt% of mischmetal 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 3 of 11 15 was added to 3 g of MgH powder and then ball-milled for field-emission scanning electron microscope were utilized. 20 and 30 h. Particle size was calculated using the MIP image analysis Ball-milling was performed in a high-energy planetary software in the form of a circle diameter equal to the parti- 1/2 ball mill (PM2400 Asia Rakhsh Co.) using a hardened chro- cle size ECD = (4A/π) . In this formula, A represents the mium-nickel steel (150 mL) vial and hardened steel balls magnitude of the surface of the particle. Particle size was (5, 10, and 20 mm in diameter) with ball to powder weight determined by several images and by specifying 300–700 ratio of 20 and rotation speed of 250 rpm under argon atmos- particles in each measurement. To measure the free surface phere. It is reported that, in ball-milling, different ball sizes of the samples (BET test), the BELSORP-mini II device are sometimes mixed to randomize the motion of balls and with liquid nitrogen was used. Before BET experiments, the increase the efficiency of milling [ 17]. After the completion samples were heated up to 180 °C and maintained for 4 h of the ball-milling process, the vial was opened in a glove to remove all possible moisture. The main parameters were box under high-purity argon gas to prevent the oxidation of 195.65 °C for 55 min and 10 points BET measurement. To samples. measure the properties of hydrogen desorption, a home- To recognize and analyze the phases, X-ray diffraction made Sievert apparatus was employed. Before the experi- (XRD) (Philips X’Pert Pro diffractometer) with CuK radia- ments, all the samples were subjected to activation using tion, at the wavelength of λ = 0.1541874 nm, in the range of a Sievert machine under the hydrogen pressure of 4 Mpa 2θ = 20–100°, and step size of 0.02° was used. The X’Pert at 180° C for 4 h. A sample of 300 mg was placed in the HighScore Plus v2.2b software of PANalytical Company Sievert apparatus and washed with argon gas. The samples was employed to identify the phases. Mean crystallite size were then heated up to 500 °C with the rate of 10 °C/min. and the lattice micro-strain of the particles were measured Finally, the results were reported as the curve of the amount using the Williamson–Hall method [18]: of hydrogen desorbed versus temperature. All Sievert tests were performed at least twice to ensure reproducibility. cos  = K∕ + 2 sin , (1) sample where β is the full width at half maximum (FWHM) of sample the milled powder, θ is the position of peak maximum, K is Results and discussion the Scherrer constant (about 0.9), λ is the beam wavelength, δ is the crystallite size, and ε is the lattice micro-strain intro- Phase identifying duced by ball-milling. For instrumental correction, a Gauss- ian relationship was used [19]: Figures 1 and 2 depict the XRD patterns of mischmetal and MgH , respectively. In Fig.  2, low-intensity magnesium 2 2 =  −  , (2) sample peaks are observed, which were predictable according to experimental instrumental the > 95 wt% purity of primary MgH . where β is the measured FWHM of the annealed experimental Figure  3 illustrates the XRD patterns of as-received nickel powders. MgH and ball-milled MgH . According to these pat- 2 2 To explore the morphology of the samples, a Tescan- terns, increasing the milling time increased the width and MV 2300 scanning electron microscope and Zeiss-Sigma reduced the intensity of the peaks. Moreover, some of the Fig. 1 XRD pattern of the mis- chmetal used in this study 1 3 15 Page 4 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 Fig. 2 XRD pattern of the magnesium hydride used in this study Fig. 3 XRD patterns of a as-received MgH and MgH 2 2 milled for b 10, c 30, and d 40 h weak peaks disappeared by increasing the milling time due Table 1 Mean crystallite size and lattice strain of the ball-milled to the reduction of crystallite size as well as the increase MgH in lattice strain and internal energy during milling [20]. In Time of ball-milling (h) Mean crystallite size (nm) Lattice the ball-milled samples, a weak peak was observed from strain magnesium oxide. The formation of this compound may (%) be due to the presence of Mg in the initial MgH and ten- 0 87 – dency towards oxidation during ball-milling. It is well- 10 58 0.72 known that ball-milled powders are very active. One rea- 30 35 0.95 son for the oxidation of active Mg during ball-milling may 40 32 1.07 be the presence of oxygen as an impurity in the flowing argon gas especially when the majority of argon in the gas container is used and the amount of gas in the container is powder to 32 nm for the 40 h-milled M gH , and lattice less than 1/4 of container volume. Another reason can be strain increased to 1.07% due to the 40 h milling. the leakage of glove box and entrance of oxygen into it. Oxidation can also be performed during the movement of Morphology of ball‑milled MgH samples from the laboratory towards the XRD machine. Table 1 shows the variation of crystallite size and lattice The SEM images of as-received MgH and ball-milled strain of MgH with ball-milling. It can be observed that MgH (10, 30, and 40 h) are depicted in Fig. 4. Based on crystallite size was reduced from 87 nm for the unmilled Fig. 4b, after 10 h of milling, MgH particles became spheri- cal and their mean size reached 338 nm. The presence of 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 5 of 11 15 Fig. 4 Scanning electron microscopy (SEM) images of a as-received MgH and MgH 2 2 ball-milled for b 10, c 30, and d 40 h small particles along with the coarse particles or agglom- erate of particles in the sample milled for 10 h shows the non-uniform distribution of particle size in this sample. With an increase in ball-milling time from 10 to 30 h, the parti- cles were slightly smaller and the mean size of the particles reached 306 nm. An increase in the ball-milling time up to 30 h also led to a more uniform distribution of the par- ticle size. The particles were agglomerated in the sample milled for 40 h and the mean size of the particles increased to 372 nm. After milling for a certain period, a state of equi- librium was created and fine particles became larger, while coarse particles became smaller due to the balance between the cold-welding rate (which tended to increase the mean particle size) and the rate of fracture (that led to reduce the mean particle size). The mean particle size eventually reached to the equilibrium size [21]. It can be assumed that no reduction in the mean particle size would be observed at longer milling times. Fig. 5 Hydrogen desorption behavior of as-received MgH and ball- milled MgH for 10, 30, and 40 h Eec ff t of ball‑milling on the hydrogen desorption properties of  MgH temperature can be the surface oxidation and the need for The results of hydrogen desorption studies on as-received more activation. However, due to the uniformity of activa- and ball-milled MgH (10, 30, 40 h) are illustrated in Fig. 5. tion conditions for all samples, the results are comparable The on-set desorption temperature of MgH decreased from 2 with each other. The difference between the on-set desorp- 480  °C for as-received MgH to 400  °C for 40  h-milled tion temperature of as-received MgH and 40 h ball-milled MgH in the condition of these experiments. The on-set des- 2 MgH (400 and 480 °C) is the increased specific surface area orption temperature of as-received MgH has been reported and lattice strain, reduced size of crystallites, and formation to be above 300 °C [1, 5]. The reason for the difference of structural defects in the material, which facilitate hydro- in temperature measured in this study with the reported gen desorption. The positive role of milling in reducing 1 3 15 Page 6 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 the hydrogen desorption temperature has been confirmed. Nevertheless, with increasing the milling time from 30 to 40 h, the desorption curve slightly shifted to the right (higher temperature), perhaps due to the agglomeration of particles and reduction of their free surface area. Therefore, to further decrease the temperature of hydrogen desorption, the cata- lyst was added to MgH prior to the milling process. Eec ff t of catalyst addition on  MgH Phase identifying of  MgH : 6 wt% of mischmetal The XRD patterns of MgH ball-milled for 10 h as well as Fig. 7 The XRD pattern of 30  min ball-milled MgH -6  wt% of mis- chmetal the MgH containing 6 wt% of mischmetal milled for 10, 20, 30, and 40 h are shown in Fig.  6. By comparing the XRD patterns of MgH milled for 10 h and MgH contain- 2 2 ing 6 wt% of mischmetal milled for 10 h, a broadening and which was ball-milled for 30 min. It was observed that the intensity of the cerium hydride peak was low, indicating the decrease in the peak intensities of the catalyst-containing sample can be observed. This is due to the decrease in crys- formation of a low amount of this compound at short milling times. However, as the milling time increased, the intensity tallite size and increase in lattice strain through the addition of catalyst (mischmetal). Catalyst particles can act as hard of the cerium hydride peak increased in the samples. It can be concluded that an increment in the milling time led to a balls during milling and contribute to structural defects [22]. Peaks of CeH were observed in the catalyst-containing higher amount of hydride in the samples. Cerium hydride 2.73 may be formed on the surface of cerium. It has been reported samples. Considering the lower Gibbs free energy of forma- tion for cerium hydride (− 204.8 kJ/mol H ) in comparison that this surface layer controls the reaction rate of the for- mation of cerium hydride [24]. Creating new surfaces and with MgH (− 75 kJ/mol H ), it seems that cerium reacted 2 2 with MgH during high-energy ball-milling according to fracture of cerium hydride layers on the surface of cerium due to mechanical milling facilitates the formation of cerium Eq. 3 and formed cerium hydride [23]. hydrides. The formation of hydrides during the milling pro- Ce + MgH → CeH + Mg ΔG =−129.8 kJ/mol. 2 2 (3) cess has also been reported by Shang et al. [23]. This layer Due to the non-equilibrium nature of the milling process, also affects the amount of hydrogen desorption from the the formation of non-stoichiometric hydrides of cerium such powder mixture. as CeH was also expected. Figure 7 shows the diffrac- 2.73 tion pattern of a sample containing 6 wt% of mischmetal Particle morphologies Figure 8 shows the SEM images of MgH -6 wt% of mis- chmetal ball-milled for 10, 20, 30, and 40 h. It has been reported in Sect. 2.2 that the mean particle size of the 10 h ball-milled MgH is 338 nm, while the particle size of the 10 h ball-milled MgH -6 wt% of mischmetal is 287 nm. By increasing the milling time of MgH -6 wt% of mischmetal from 10 to 20 h, the particle size increased and the mean particle size reached 381 nm. In the 30 h-milled sample, the mean particle size was 360 nm. At shorter milling times, large and small particles were created together. However, with increasing the milling time and approaching the equi- librium level, small particles stuck together and large parti- cles broke into small ones [21]. The increased particle size in the 20 h-milled sample can be attributed to the agglom- eration of fine particles and reduction of particle size in the 30 h-milled sample can be due to the fracture of large par- Fig. 6 The XRD pattern of a 10  h ball-milled MgH and ticles. By increasing this time to 40 h, the particles stuck MgH -6wt % of mischmetal ball-milled for b 10 h, c 20 h, d 30 h, and e 40 h together and agglomerated with the mean size of 417 nm. By 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 7 of 11 15 Fig. 8 The SEM images of MgH -6 wt% of mischmetal ball-milled for a 10, b 20, c 30, and d 40 h comparing these results with those of the samples without a was observed. The increase in the milling time up to 30 h catalyst, it can be seen that smaller particles were achieved led to the fracture of large particles, resulting in an increase in short ball-milling times by adding the catalyst. This can in free surface relative to the 20 h-milled sample. During the be attributed to the role of the catalyst in the application ball-milling process, the removing, mixing and shrinkage of of more work on particles, resulting in cold-welding and pores and forming new porosities occurred simultaneously particle-bonding [21]. by agglomeration of new particles [21]. According to the results of the BET test, it seems that the agglomeration and Determination of specific surface area (BET) removal of porosities resulted in a reduced free surface and mean pore volume in the 40 h ball-milled MgH -6 wt% of Free surface changes, mean porosity diameter, and mean mischmetal sample. pore volume for the MgH samples milled for 10 h and the MgH samples with 6 wt% of catalyst milled for 10, 20, Hydrogen desorption properties 30, and 40 h are shown in Fig. 9. According to these dia- grams, the sample with the catalyst that is milled for 10 h Figure 10 illustrates the results of the hydrogen desorption of has the highest free surface with the smallest mean diameter MgH, MgH milled for 10 h, and MgH with 6 wt% of cata- 2 2 2 of porosity and the highest pore volume. By comparing the lyst milled for 10, 20, 30, and 40 h. By comparing the MgH MgH sample milled for 10 h with the MgH sample with samples milled for 10 h with MgH with the catalyst milled 2 2 2 the catalyst milled for 10 h, it can be observed that the addi- for 10 h, it was observed that the addition of catalyst reduced tion of the catalyst increased the free surface and mean pore the temperature of hydrogen desorption and decreased the volume, but decreased the pore size. It can be concluded amount of desorbed hydrogen. The decrease in the on-set that the addition of the catalyst led to a smaller particle size, desorption temperature of hydrogen can be due to the posi- which is also confirmed by SEM images. By increasing the tive effect of the catalyst on the hydrogen desorption pro- milling time to 20 h for the catalyst containing MgH , the cess by increasing the free surface and accumulated energy free surface area and mean pore volume decreased and the in the milled powder and reducing the activation energy. mean diameter of the pores increased. According to the The reduction in the amount of desorbed hydrogen can be results of SEM images, it can be concluded that the closing due to the formation of the cerium hydride phase during the and mixing of the porosities occurred simultaneously with milling process. The cerium hydride phase formed during the particle size increase, and hence the free surface loss ball-milling was stable at high temperatures [23] and, thus, 1 3 15 Page 8 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 the formation of this phase reduced the amount of desorbed hydrogen from the milled MgH -catalyst mixture. Increasing the milling time to 20 and 30 h also reduced the tempera- ture of hydrogen desorption. By comparing the results of MgH -6 wt% of mischmetal milled for 10, 20, and 30 h and considering the results of the BET test, it can be observed that the on-set desorption temperature of the sample milled for 10 h is more than that of the samples ball-milled for 20 and 30 h despite the lower free surface in the 20 and 30 h-milled mixtures. This can be attributed to the increase in the internal energy, formation of defects in the structure, and reduction in the diffusion path of hydrogen atoms. The on-set desorption temperature of hydrogen in the 40 h ball- milled sample increased compared to the 30 h ball-milled sample. This is because of the increase in particle size and free surface loss in the 40 h ball-milled sample due to the cold-welding phenomenon. Eec ff t of increasing the amount of the catalyst To examine the effect of the amount of catalyst on the hydro- gen desorption properties of MgH , the MgH + 10 wt% of 2 2 mischmetal powder mixture was prepared by ball-milling for 20 and 30 h. Figure 11 shows the X-ray diffraction patterns of the samples containing 6 and 10 wt% of catalyst which were milled for 20 and 30 h. Table 1 also shows the crystallite size and lattice strain of MgH -10 wt% of mischmetal ball-milled for 20 and 30 h and MgH -6wt % of mischmetal ball-milled for 20 and 30 h. Fig. 9 Free surface, mean porosity diameter, and mean pore volume Similar to the samples containing 6 wt% of catalyst, cerium for the 10 h ball-milled MgH and MgH -6 wt% of mischmetal ball- 2 2 milled for 10, 20, 30, and 40 h hydride peaks were observed in the samples containing 10 wt% of catalyst. By adding more catalyst in the same milling time, the peaks of the XRD pattern became wider. Fig. 10 Hydrogen desorption behavior of as-received MgH , 10 h ball-milled MgH , and MgH -6 wt% of mischmetal ball-milled for 10, 20, 30, and 40 h 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 9 of 11 15 time. This increase in the mean particle size also showed the effect of catalyst addition on the achievement of equilibrium between cold-welding and fracture at lower times of ball- milling. Subsequently, cold-welding became the predomi- nant process and agglomeration occurred. Thus, particle size was higher in MgH -10 wt% of mischmetal due to the further agglomeration. Agglomeration can also be seen in Fig. 12b for the 30 h-milled powder. The hydrogen desorption behavior of MgH -6 wt% of mischmetal and MgH -10 wt% of mischmetal ball-milled for 20 and 30 h, and also of the 30 h-milled MgH can be observed in Fig. 13. The latter sample helped understand the effect of catalyst addition. It was observed that an increase in the amount of catalyst led to a reduction in the amount of hydrogen desorption because of the further formation of cerium hydride with increase in the catalyst. Increasing the Fig. 11 XRD patterns of MgH -6 wt % of mischmetal ball-milled for amount of catalyst from 6 to 10 wt% also resulted in a 15 °C a 20, c 30, and MgH -10 wt% of mischmetal ball-milled for b 20 and increment in the on-set hydrogen desorption temperature of c 30 h the sample milled for 20 h and a 20 °C increase in the on- set hydrogen desorption temperature of the sample milled Table 2 Mean crystallite size and lattice strain of the ball-milled for 30 h. Shang et al. [23] explained the rise in hydrogen MgH -10 wt% and MgH -6 wt% of mischmetal after 20 and 30 h of 2 2 desorption temperature as a result of the formation of a sta- ball-milling ble layer of cerium hydride on particle surfaces, stating that Sample Time of Mean crys- Lattice the creation of this surface layer would be problematic for ball milling tallite size strain the diffusion of hydrogen from inside the particles outward. (h) (nm) (%) This seems to be true for high amounts of cerium hydride. In addition, it is possible to consider the increase of particle MgH -6 wt% of mischmetal 20 33 1.02 MgH -10 wt% of mischmetal 20 33 1.08 size and reduction of free surface as a reason for increasing the temperature of hydrogen desorption. MgH -6 wt% of mischmetal 30 33 1.19 MgH -10 wt% of mischmetal 30 30 1.65 Conclusion Table 2 confirms the role of adding catalyst in creating lat- tice strain in crystallites in comparison with MgH (Fig. 3). The results of this research can be summarized as follows: It is clear from Fig. 10 that adding to the amount of catalyst at a constant milling time led to more lattice strain in the 1. The XRD patterns of as-received MgH and ball- samples.milled MgH show that increasing the milling time led According to the SEM images of MgH -10 wt% of mis- to an increase in the width and reduction in the inten- chmetal milled for 20 and 30 h in Fig. 12, the mean parti- sity of peaks. It is shown that crystallite size reduced cle size of the 20 h-milled sample increased in comparison from 87 nm for the unmilled powder to 32 nm for the with the MgH -6 wt% of mischmetal milled for the same Fig. 12 SEM images of MgH -10 wt% of mischmetal ball-milled for a 20 and b 30 h 1 3 15 Page 10 of 11 Materials for Renewable and Sustainable Energy (2018) 7:15 Fig. 13 Hydrogen desorption behavior of MgH -6 wt% of mischmetal and MgH -10 wt% of mischmetal milled for 20 and 30 h and also of the 30 h-milled MgH Open Access This article is distributed under the terms of the Crea- 40 h-milled MgH and the lattice strain increased to tive Commons Attribution 4.0 International License (http://creat iveco 1.07% due to 40 h of milling. mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- 2. Increasing the milling time up to 40 h resulted in an tion, and reproduction in any medium, provided you give appropriate improvement in the hydrogen desorption properties of credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. MgH and 80 °C reduction in hydrogen desorption tem- perature 3. By adding the mischmetal, cerium reacted with MgH References during ball-milling and formed cerium hydride. It seems that the amount of this hydride increased with an incre- 1. Varin, R.A., Czujko, T., Wronski, Z.S.: Nanomaterials for solid ment of mischmetal from 6 to 10 wt%. This hydride was state hydrogen storage. Springer Science & Business Media, also stable at high temperatures, resulting in the harder New York (2009) desorption of hydrogen in large quantities. 2. Ogden, J.M.: Developing an infrastructure for hydrogen vehi- cles: a Southern California case study. Int. J. Hydrogen Energy 4. Comparing the XRD results of 10 h-milled MgH -6 wt% 24, 709–730 (1999) of mischmetal to 10 h-milled MgH revealed that crys- 3. Grochala, W., Edwards, P.P.: Thermal decomposition of the non- tallite size decreased and the lattice strain increased by interstitial hydrides for the storage and production of hydrogen. the addition of the catalyst (mischmetal). Chem. Rev. 104, 1283–1316 (2004) 4. Alefeld, G., Volkl, J.: Topics in Applied Physics. Hydrogen In 5. Mischmetal exhibited a suitable catalytic effect on the Metals. I.-Basic Properties, vol. 28. Springer, New York (1978) hydrogen desorption properties of MgH and, in the best 5. Walker, G.: Solid-State Hydrogen Storage: Materials and Chem- case, the reduction in hydrogen desorption temperature istry. Elsevier, Amsterdam (2008) of the sample containing 6 wt% of mischmetal milled 6. Zaluska, A., Zaluski, L., Ström-Olsen, J.: Nanocrystalline mag- nesium for hydrogen storage. J. Alloy Comd. 288, 217–225 for 30 h was 100 °C.Therefore, the powder containing (1999) 6 wt% of mischmetal milled for 30 h is the most suitable 7. Guoxian, L., Erde, W., Shoushi, F.: Hydrogen absorption and sample from the desorption point of view. desorption characteristics of mechanically milled Mg 35 wt% 6. Increasing the amount of mischmetal from 6 to 10 wt% FeTi1. 2 powders. J. alloy compd. 223, 111–114 (1995) led to an increment in the hydrogen desorption tempera- ture of the milled MgH –mischmetal mixture. 1 3 Materials for Renewable and Sustainable Energy (2018) 7:15 Page 11 of 11 15 8. Wang, P., Wang, A., Zhang, H., Ding, B., Hu, Z.: Hydriding 17. Takacs, L.: Self-sustaining reactions induced by ball milling. Prog. properties of a mechanically milled Mg–50 wt% ZrFe 1.4 Cr Mater Sci. 47, 355–414 (2002) 0.6 composite. J. alloy compd. 297, 240–245 (2000) 18. Williamson, G., Hall, W.: X-ray line broadening from filed alu- 9. Li, F., Jiang, L., Du, J., Wang, S., Liu, X., Zhan, F.: Synthesis minium and wolfram. Acta Metall. Mater. 1, 22–31 (1953) and hydrogenation properties of Mg–La–Ni–H system by reactive 19. Hosseini-Gourajoubi, F., Pourabdoli, M., Uner, D., Raygan, S.: mechanical alloying. Int. J. hydrogen Energy 31, 581–585 (2006) Effect of process control agents on synthesizing nano-structured 10. Barkhordarian, G., Klassen, T., Bormann, R.: Kinetic investi- 2  Mg–9Ni–Y catalyst by mechanical milling and its catalytic gation of the effect of milling time on the hydrogen sorption effect on desorption capacity of MgH . Adv. Powder Technol. reaction of magnesium catalyzed with different Nb O contents. 26, 448–453 (2015) 2 5 J. Alloy. Compd. 407, 249–255 (2006) 20. Raygan, S., Pourabdoli, M., Abdizadeh, H., Medraj, M.: Synthe- 11. Wagemans, R.W.P., van Lenthe, J.H., de Jongh, P.E., Van Dil- sizing nanostructured Ni75Mg16.66Y8.34 (at%) powder by solid len, A.J., de Jong, K.P.: Hydrogen storage in magnesium clusters: state reaction and mechanical milling. Mater. Manuf. Process. 27, quantum chemical study. J. Am. Chem. Soc. 127, 16675–16680 1300–1305 (2012) (2005) 21. Suryanarayana C.: Mechanical Alloying and Milling. CRC Press 12. Zhu, M., Lu, Y., Ouyang, L., Wang, H.: Thermodynamic tun- (2004) ing of Mg-based hydrogen storage alloys: a review. Materials. 6, 22. Sakintuna, B., Lamari-Darkrim, F., Hirscher, M.: Metal hydride 4654–4674 (2013) materials for solid hydrogen storage: a review. Int. J. Hydrogen 13. Liang, G., Huot, J., Boily, S., Van Neste, A., Schulz, R.: Catalytic Energy 32, 1121–1140 (2007) effect of transition metals on hydrogen sorption in nanocrystalline 23. Shang, C., Guo, Z.: Structural and desorption characterisations of ball milled MgH –Tm (Tm = Ti, V, Mn, Fe and Ni) systems. J. milled (MgH + Y, Ce) powder mixtures for hydrogen storage. Int. 2 2 Alloy. Compd. 292, 247–252 (1999) J. Hydrogen Energy 32, 2920–2925 (2007) 14. Barkhordarian, G., Klassen, T., Bormann, R.: Catalytic mecha- 24. Sarussi, D., Jacob, I., Bloch, J., Shamir, N., Mintz, M.: The kinet- nism of transition-metal compounds on Mg hydrogen sorption ics and mechanism of cerium hydride formation. J. alloy compd. reaction. J. Phys. Chem. B 110, 11020–11024 (2006) 191, 91–99 (1993) 15. Dornheim, M., Doppiu, S., Barkhordarian, G., Boesenberg, U., Klassen, T., Gutfleisch, O., Bormann, R.: Hydrogen storage in Publisher’s Note Springer Nature remains neutral with regard to magnesium-based hydrides and hydride composites. Scripta jurisdictional claims in published maps and institutional affiliations. Mater. 56, 841–846 (2007) 16. Morinaga, M., Yukawa, H.: Nature of chemical bond and phase stability of hydrogen storage compounds. Mater. Sci. Eng. A 329, 268–275 (2002) 1 3

Journal

Materials for Renewable and Sustainable EnergySpringer Journals

Published: May 29, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off