A catalytic mechanism investigation of TiF₃ on hydriding/dehydriding properties of Mg₈₅Cu₅Ni₁₀ alloy

Abstract

In this research, Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys were synthetized via ball milling and the catalytic mechanism of TiF₃ on hydrogenation and dehydrogenation of Mg₈₅Cu₅Ni₁₀ was studied. The microstructure, hydriding/dehydriding kinetics and thermodynamics of the alloys were discussed in detail. The TiF₃ catalyzed alloys have faster hydriding/dehydriding kinetics and lower thermodynamic stability. After hydrogen absorption and desorption, TiF₃ decomposes into TiH₂ and MgF₂. TiF₃, TiH₂ and MgF₂ promote to forming crystal defects, dislocations, grain boundaries and nanocrystals which are advantageous to speeding up the rate of hydrogen absorption and desorption. The dehydrogenation activation energy E_a(de) and dehydrogenation enthalpy ΔH(de) are reduced to 81.462 from 116.767 kJ mol⁻¹ and 72.456 from 93.372 kJ mol⁻¹ respectively by 6 wt% TiF₃. An appropriate amount of TiF₃ can improve the hydriding/dehydriding kinetics and thermodynamics of Mg₈₅Cu₅Ni₁₀.

---

Introduction

Energy is an important material basis for the survival and development of human society.^1,2 The rapid development of modern industry relies on the use of a large number of fossil fuels. Basically, our life and work are inseparable from fossil fuels. Nevertheless, all fossil fuels are non-renewable. In addition, environmental pollution and extreme climate change caused by fossil fuels make human beings face severe challenges. At present, all countries around the world are considering carbon emission reduction by reducing the use of fossil fuels and developing and using renewable energy. Therefore, the development of clean and efficient new energy has become an important issue for all countries, which is also an important guarantee to solve the energy crisis and human survival.^3–9 So far, there are many new energies that can replace traditional fossil fuels, including solar energy, wind energy, tidal energy, geothermal energy, nuclear energy, and hydrogen energy. Considering the factors of manufacture, transportation, safety, reliability, and environmental friendliness, hydrogen energy is one of the best choices. Since the 1990s, Japan, the United States, and some European countries have carried out hydrogen research and development plans. Toyota launched the second generation Mirai hydrogen fuel cell vehicle in 2019, marking that the fuel cell electric vehicle (FCEV) technology had entered a new stage. Nevertheless, there are still some important technical problems for the hydrogen energy system to fully implement and replace fossil fuels.^10–17 The first is cheap and mass-produced hydrogen production technology. The second is hydrogen storage and transportation. The third is to convert hydrogen into electricity and heat energy safely and efficiently in the application process. Among them, high-density, low-cost, safe, and high efficiency hydrogen storage technology are the key links of large-scale use of hydrogen energy.^18–20

The advantages of the metal hydride hydrogen storage method are high hydrogen storage capacity, no need for high pressure and heat insulation container and good safety compared with the traditional gas and liquid hydrogen storage methods.^21,22 Among many metal hydride hydrogen storage materials, magnesium has a high reversible hydrogen storage capacity (7.6 wt%) which attracts many scholars' interest.^23–26 Besides, magnesium has a low price, abundant reserves, excellent heat resistance, good recyclability, and can form solid solution with other elements in equilibrium state. Unfortunately, MgH₂ has high thermal stability and needs high temperature to release hydrogen, and the rate of hydrogenation/dehydrogenation evolution is relatively slow, which limits its wide application. Many scholars have improved the poor dynamic and thermodynamic properties of magnesium based alloys by alloying, nanocrystallization and amorphization.²⁷ Generally, transition metal elements or rare earth elements are used to alloy with magnesium to ameliorate the hydrogen storage performance. Yuan et al.²⁸ synthetized Sm₅Mg₄₁ and found that its initial dehydriding temperature was 270 °C indicating that alloying with Sm reduced the thermal stability of hydrides. Some studies have shown that nanocrystallization can effectively ameliorate the kinetics of hydriding and dehydriding of magnesium. This is because nanoparticles and nanocrystals have many interfaces. The channel effect and surface effect introduced by nanocrystallization shorten the diffusion distance of hydrogen in the particles, reduce the time of hydrogenation reaction, and accelerate the rate of hydrogen absorption and desorption. Liu et al.²⁹ prepared Mg–6.9 at% Zn ultrafine particles via hydrogen plasma-metal reaction method. They demonstrated that the Mg–Zn nanoparticles could absorb 5.0 wt% H₂ at 300 °C in 20 min with 56.3 kJ mol⁻¹ hydrogen absorption activation energy. Furthermore, amorphization by mechanical ball milling and vacuum rapid cooling technology also could improve hydrogen storage performance of magnesium. Zhang et al.³⁰ synthesized (Mg₂₄Ni₁₀Cu₂)_100−xNd_x (x = 0, 5, 10, 15, 20) alloy through rapidly quenched under vacuum and found the alloy with amorphous structure had good electrochemical performance. Moreover, it is easy to add all kinds of catalysts during ball milling to prepare composite materials. As a catalyst, TiF₃ is often used to improve the hydrogen storage properties of magnesium based alloys. N. S. Mustafa and M. Ismail³¹ investigated the effect of TiF₃ on the performance of 2NaAlH₄–Ca(BH₄)₂ hydrogen storage properties and they found that the addition of TiF₃ reduced the onset decomposition temperature and the ab/desorption kinetic was also improved as compared to the undoped composite. N. N. Sulaiman³² studied the H₂ storage properties of MgH₂/Cd–TiF₃ for the first time. The sorption properties of MgH₂/Cd were improved after adding with TiF₃. E_A for hydrogen desorption of doped composite was lowered to 107.0 kJ mol⁻¹. In addition, our study has proved that TiF₃ significantly improved the kinetics of hydrogen absorption and desorption of Mg₈₅Zn₅Ni₁₀ alloy, and reduced the dehydrogenation activation energy to 75.514 kJ mol⁻¹.¹² Therefore, we choose TiF₃ as catalyst to study the catalytic effect on the hydrogen storage properties of Mg–Cu–Ni alloy.

In this investigation, we prepared Mg–Cu–Ni alloy with TiF₃ catalyst to study the hydrogen storage properties and catalytic mechanism. The hydriding/dehydriding kinetics and thermodynamics were investigated in detail. The microstructure and characterization were tested and observed by XRD, SEM and HRTEM. The dehydrogenation activation energy, entropy and enthalpy were also fitted and calculated.

Experimental

First of all, the as-cast Mg₈₅Cu₅Ni₁₀ alloy was smelted. We used a vacuum induction furnace to prepare the as-cast Mg₈₅Cu₅Ni₁₀ alloy under 0.04 MPa He atmosphere as the protective gas. As a result of the high vapor pressure of Mg, it is very easy to burn and volatilize during smelting. According to smelting experience, more 10 wt% Mg was mixed in the smelting process in order to guarantee the designed composition of the as-cast Mg₈₅Cu₅Ni₁₀ alloy. Subsequently, the as-cast Mg₈₅Cu₅Ni₁₀ alloy was crushed into powders (particle size < 75 μm) mechanically. Finally, the Mg₈₅Cu₅Ni₁₀ powders and different contents of TiF₃ (produced by Thermo Fisher Technology (China) Co., Ltd) were mixed to ball mill by a planetary-type mill (Made by Instrument Factory of Nanjing University) under Ar atmosphere. The contents of TiF₃ were 0, 2, 4, 6 and 8 wt%. The ball to powder was 40 : 1. The milling speed was 350 rpm. In consideration of excess heating caused by milling, the mill process was rested for 1 h after every 1 h working. The total milling time was 5 h. All the block metal materials in this experiment were provided by China Iron & Steel Research Institute Group.

Hydrogen absorption/desorption kinetics of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys were measured by a Sieverts-type apparatus (made by General Research Institute for Nonferrous Metals). At the beginning of tests, all the samples were absorbed and desorbed hydrogen for several cycles to activate completely. Hydrogenation was tested at a pressure of 3 MPa from 100 to 360 °C, and dehydrogenation was tested at a pressure of 1 × 10⁻⁴ MPa from 240 to 360 °C. The purities of He, Ar and H₂ gases were 99.999% supplied by China Iron & Steel Research Institute Group.

Phase structures, morphologies and crystalline states of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys were detected and observed by X-ray diffraction (XRD) (D/max/2400) using CuKα radiation with scanning rate of 2° min⁻¹, scanning electron microscopy (SEM) (QUANTA 400), scanning electron microscopy (SEM) (QUANTA 400) and high resolution transmission electron microscope (HRTEM) (JEM-2100F).

Results and discussion

Microstructural characteristics

X-ray diffraction (XRD) was used for clarifying the phase composition and hydrogenation and dehydrogenation reaction mechanism of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys. Fig. 1 describes the XRD patterns of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 4, 8) alloys after ball milling, hydrogenation and dehydrogenation. Because the patterns of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys are basically similar, in order to facilitate displaying and reading, Fig. 1 only shows the XRD patterns of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 4, 8) alloys. As shown in Fig. 1(a), after ball milling, catalyst TiF₃ did not decompose. Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ consists of Mg, Mg₂Ni and Mg₂Cu. Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys are composed of Mg, Mg₂Ni, Mg₂Cu and TiF₃. The little broadened diffraction peaks show that ball milling refines the alloy particles. According to Fig. 1(b), the Mg, Mg₂Ni and Mg₂Cu in Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ become into MgH₂, Mg₂NiH₄ and MgCu₂ respectively after hydriding. However, the TiF₃ decomposes into TiH₂ and MgF₂ in the catalyzed Mg₈₅Cu₅Ni₁₀ alloys after hydrogenation with the formation of MgH₂, Mg₂NiH₄ and MgCu₂. From Fig. 1(c), MgH₂, Mg₂NiH₄ and MgCu₂ change into Mg, Mg₂Ni and Mg₂Cu respectively after dehydriding, meanwhile, the TiH₂ and MgF₂ still exist in the alloys. It indicates that the thermodynamic properties of TiH₂ and MgF₂ are stable and do not decompose under the condition of dehydrogenation evolution in this experiment. On the basis of analysis above, the possible hydriding/dehydriding mechanism of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys are:

Mg + H₂ ↔ MgH₂

Mg₂Ni + 2H₂ ↔ Mg₂NiH₄

2Mg₂Cu + 3H₂ ↔ MgCu₂ + 3MgH₂

2TiF₃ + 3Mg + 2H₂ → 2TiH₂ + 3MgF₂.

Fig. 1 XRD patterns of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 4, 8) alloys: (a) before hydrogen absorption, (b) after hydrogen absorption, (c) after hydrogen desorption.

High resolution transmission electron microscope (HRTEM) was selected to further study the phase transition and microstructure of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys before and after hydriding and dehydriding. Fig. 2 shows the TEM images of Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ and Mg₈₅Cu₅Ni₁₀–4 wt% TiF₃ after hydrogenation and dehydrogenation. On the basis of Fig. 2(a) and (b), Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ consists of MgH₂, Mg₂NiH₄ and MgCu₂ after hydriding. After dehydriding, it contains Mg, Mg₂Ni and Mg₂Cu, which is consistent with the results of XRD analysis in the previous section. Instead, Mg₈₅Cu₅Ni₁₀–4 wt% TiF₃ is composed of MgH₂, Mg₂NiH₄, MgCu₂, TiH₂ and MgF₂ after hydrogenation indicating that the catalyst TiF₃ decomposes into TiH₂ and MgF₂ during hydriding process from Fig. 2(c). Meanwhile, according to Fig. 2(d), Mg₈₅Cu₅Ni₁₀–4 wt% TiF₃ contains Mg, Mg₂Ni, Mg₂Cu, TiH₂ and MgF₂ after dehydrogenation suggesting that TiH₂ and MgF₂ still exist in the alloy. The TEM image further proves that for Mg₈₅Cu₅Ni₁₀ alloy, the catalyst TiF₃ decomposes and forms stable TiH₂ and MgF₂ during the hydrogen absorption and desorption. Furthermore, we can observe the existence of crystal defects, dislocations, grain boundaries, nanocrystals and amorphous illustrating that after ball milling for 5 hours with catalyst TiF₃, the microstructure of Mg₈₅Cu₅Ni₁₀ alloy changes obviously after hydriding and dehydriding cycles. TiF₃ promotes the formation of nanocrystals on the surface of Mg₈₅Cu₅Ni₁₀ alloy and introduces high density grain boundaries.

Fig. 2 HRTEM micrographs of Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ and Mg₈₅Cu₅Ni₁₀–4 wt% TiF₃ after hydrogen absorption (a) and (c), after hydrogen desorption (b) and (d).

Fig. 3 shows the SEM images of Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ and Mg₈₅Cu₅Ni₁₀–4 wt% TiF₃ after ball milling, after hydrogen absorption and after hydrogen desorption. From Fig. 3, the alloys particles present irregular spheres. After ball milling, the particle diameter is about 30 μm. However, Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ particles agglomerate, and Mg₈₅Cu₅Ni₁₀–4 wt% TiF₃ particles are pulverized after hydriding and dehydriding. It indicates that TiF₃ is helpful to refine alloy particles during hydriding and dehydriding cycles which is beneficial to increase the contact area of hydrogen absorption and desorption reaction. In addition, after the cycles of hydrogen absorption and desorption, some cracks appear on the surface of the particles. Our previous study has confirmed that these cracks contribute to accelerating the rate of hydrogen absorption and desorption.^12,27 In addition, it is worth noting that TiF₃ decomposes into TiH₂ and MgF₂ during hydriding and dehydriding process of Mg₈₅Cu₅Ni₁₀, while in our previously published work,¹² TiF₃ does not decompose in Mg₈₅Zn₅Ni₁₀. It may be due to fact that MgZn₂ is a stable phase in Mg₈₅Zn₅Ni₁₀ alloy while there is a transformation between Mg₂Cu and MgCu₂ in Mg₈₅Cu₅Ni₁₀ alloy during hydrogenation and dehydrogenation. The stable MgZn₂ and transformation between Mg₂Cu and MgCu₂ may explain the decomposition of TiF₃ in Mg–Cu–Ni alloy. There is a small amount of excess Mg appearing in the Mg₂Cu and MgCu₂ transformation process, making Ti and F reacting with Mg and H to form MgF₂ and TiH₂.

Fig. 3 SEM images of Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ and Mg₈₅Cu₅Ni₁₀–4 wt% TiF₃ after ball milling (a) and (d), after hydrogen absorption (b) and (e), after hydrogen desorption (c) and (f).

Kinetic properties of hydrogenation and dehydrogenation

In order to study the effect of TiF₃ on the kinetics of hydrogen absorption and desorption of Mg₈₅Cu₅Ni₁₀, the kinetics of hydriding and dehydriding of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys was measured at different temperatures after fully activating. Fig. 4(a)–(e) describe the hydrogenation curves of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys from 100 to 360 °C. All five alloys have very fast hydrogen absorption rates at high temperatures, and all of them can basically reach the saturation state in a short time on the basis of Fig. 4. However, with the decrease of hydrogen absorption temperature, the kinetic properties of the five alloys are gradually different. According to Fig. 4(a), Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ has a rapid hydrogen absorption rate when the temperature is higher than 320 °C. When the temperature is at the range of 300–200 °C, the hydrogen absorption rate and capacity of Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ decrease along with temperature decreasing. The hydrogen absorption speed and capacity of Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ decrease obviously when the temperature is lower 200 °C. However, the TiF₃ catalyzed alloys have faster hydrogen absorption rates when the temperature is lower than 300 °C. The results show that TiF₃ improves the hydrogen absorption kinetics of Mg₈₅Cu₅Ni₁₀ below 300 °C. The most intuitive embodiment is that the hydrogen absorption curves of the four TiF₃ catalyzed alloys at the range of 300–200 °C are closer to those above 300 °C. Fig. 4(f) shows the time required to absorb 3 wt% hydrogen of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys at 100 °C. Obviously, the times required for the four alloys containing TiF₃ to absorb 3 wt% hydrogen are significantly shorter than that for Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ at 100 °C. It illustrates that TiF₃ significantly improves and increases the hydrogen absorption rate of Mg₈₅Cu₅Ni₁₀. However, due to the decomposition of TiF₃ and the formation of TiH₂ and MgF₂, the saturated hydriding capacities of the alloys also decrease.

Fig. 4 Hydrogen absorption kinetic curves of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys at different temperatures: (a) x = 0; (b) x = 2; (c) x = 4; (d) x = 6; (e) x = 8 and the time required to absorb 3 wt% hydrogen at 100 °C (f).

Fig. 5 shows the dehydrogenation curves of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys from 240 to 360 °C. As Fig. 5 describing, temperature affects the dehydrogenation performance of the alloys very much. Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys can complete all hydrogen release in 10 min with fast rate of dehydriding over 320 °C. When the dehydrogenation temperature is 360 °C, the time required for complete dehydrogenation is 246 s, 144 s, 216 s, 150 s and 192 s respectively (Fig. 5(f)). The dehydrogenation time increases with the decrease of dehydrogenation temperature. Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ can completely release hydrogen in 25 min, while the TiF₃ catalyzed alloys can completely dehydrogenate in 20 min at 300 °C. Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ only can release some hydrogen with more than 1 h, while the TiF₃ catalyzed alloys can completely dehydrogenate in 50 min at 280 °C. When the dehydrogenation temperature is below 280 °C, all of the five alloys show poor dehydrogenation performance, and it takes a long time for partial dehydrogenation. The above results and analysis show that the dehydrogenation performance of Mg₈₅Cu₅Ni₁₀ is obviously improved after TiF₃ catalyzing.

Fig. 5 Hydrogen desorption kinetic curves of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys at different temperatures: (a) x = 0; (b) x = 2; (c) x = 4; (d) x = 6; (e) x = 8 and the time required for complete dehydrogenation at 360 °C (f).

Hydrogenation and dehydrogenation cyclic stability

Fig. 6 and 7 express the hydrogenation and dehydrogenation cyclic stability curves of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys at 360 °C. As shown in the images, there is nearly no change of hydrogen absorption and desorption capacities for Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys after 20 cycles. It takes a long time for the five alloys to absorb hydrogen for the first time due to the long process of activating alloys. However, the first dehydrogenation time of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys is shorter than that of Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ indicating that TiF₃ shortens the first dehydrogenation. After the second cycle, the hydrogenation and dehydrogenation rates are obviously accelerated. With the increase of cycle times, the hydrogenation and dehydrogenation rates are gradually improved with good kinetics. The hydrogen absorbing and desorbing cyclic curves are almost identical. It illustrates that the decomposition of TiF₃ does not affect the cyclic stability of the alloys. But it reduces the reversible hydrogen storage capacity. Because the decomposing of TiF₃ and forming of TiH₂/MgF₂ can not react with hydrogen in this condition. Therefore, Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys can not reach the theoretical saturated hydrogen storage capacities. The saturated hydrogen storage capacity decreases with the increase of TiF₃ content. On the basis of Fig. 6 and 7, Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys still have good hydrogenation and dehydrogenation cyclic stability.

Fig. 6 Hydrogenation cycling curves of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys at 360 °C: (a) x = 0; (b) x = 2; (c) x = 4; (d) x = 6; (e) x = 8.

Fig. 7 Dehydrogenation cycling curves of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 2, 4, 6, 8) alloys at 360 °C: (a) x = 0; (b) x = 2; (c) x = 4; (d) x = 6; (e) x = 8.

Dehydrogenation activation energy

For further studying the catalytic effect of TiF₃ on the dehydrogenation performance of Mg₈₅Cu₅Ni₁₀, we calculated the dehydrogenation activation energy of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys. The dehydrogenation activation energy is a key parameter to describe the dehydriding properties. The kinetic performance of dehydriding evolution is mainly determined by the energy barrier required by MgH₂ changing to Mg by releasing H₂. The dehydrogenation activation energy is the total energy needed to overcome. In general, the kinetic curve of dehydrogenation is fitted by JMA model, and the dehydrogenation activation energy E_a(de) can be calculated by Arrhenius method. Johnson–Mehl–Avrami (JMA) model is expressed by following equation:^33–35

ln[−ln(1 − α)] = η ln k + η ln t (1)

in this equation, α means the reaction fraction at time t, η represents the Avrami exponent reaction order and k indicates an effective kinetic parameter. Fig. 8 describes JMA curves of ln[−ln(1 − α)] vs. ln t at different temperatures. The fitting line of JMA curve is almost linear, which illustrates that the dehydriding follows a three-dimensional growth process of instantaneous nucleation followed by interface control. The values of η and η ln k express the slope and intercept of JMA curve, from which the value of rate constant k can be determined. As a result, the dehydrogenation activation energy (E_a) can be calculated based on Arrhenius equation:^36,37

k = A exp[−E_a/RT] (2)

in this equation, A is a temperature independent coefficient, T represents the absolute temperature and R means universal gas constant. Fig. 8 also shows the Arrhenius plots and dehydrogenation activation energy E_a(de) of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys. The dehydrogenation activation energy E_a(de) of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys are 116.767, 95.749, 95.531, 81.462 and 89.971 kJ mol⁻¹ respectively listed in Table 1. As a comparison, the dehydrogenation activation energy of MgH₂ is 160 kJ mol⁻¹. In our previous study,²⁷ the dehydrogenation activation energy of as-cast Mg₈₅Cu₅Ni₁₀ was 119.142 kJ mol⁻¹, and the E_a(de) of Mg₈₅Cu₅Ni₁₀–x wt% CeO₂ (x = 0, 4, 8) alloys were 116.767, 84.824, and 81.211 kJ mol⁻¹ respectively. According to the results above, TiF₃ can decrease the dehydrogenation activation energy of Mg₈₅Cu₅Ni₁₀ obviously. But there is a point worth noting, the dehydrogenation activation energy E_a(de) of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys decreases first and then increases with the increase of TiF₃ content. This may be due to the increase of TiF₃ content, which leads to the increase of TiH₂ and MgF₂ contents, which cover the surface of the alloys, thus affecting the dehydriding evolution performance. In conclusion, TiF₃ can significantly reduce the dehydrogenation activation energy E_a(de) and ameliorate the kinetic performance of Mg₈₅Cu₅Ni₁₀.

Fig. 8 JMA graphs of Mg₈₅Cu₅Ni₁₀ + x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys different temperatures: (a) x = 0; (b) x = 2; (c) x = 4; (d) x = 6; (e) x = 8 and E_a dehydrogenation (f).

Table 1 Hydrogen desorption activation energy (E_a(de)), enthalpy change (ΔH) and entropy change (ΔS) of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys and as-cast Mg₈₅Cu₅Ni₁₀

Mg85Cu5Ni10–x wt% TiF3	Ea(de) (kJ mol^−1)	ΔHab (kJ mol^−1)	ΔSab (J mol^−1 K^−1)	ΔHde (kJ mol^−1)	ΔSde (J mol^−1 K^−1)
As-cast Mg85Cu5Ni10 (ref. 27)	119.142	−98.287	−166.292	98.892	165.755
x = 0	116.767	−97.254	−155.682	93.372	151.331
x = 2	95.749	−79.420	−135.903	80.114	134.065
x = 4	95.531	−74.908	−126.432	77.227	129.059
x = 6	81.462	−69.336	−118.058	72.465	121.317
x = 8	89.971	−70.926	−121.503	77.911	131.419

---

Hydrogen storage thermodynamics

For investigating the hydrogen storage thermodynamics of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys, the PCT curves were measured at 360, 340 and 320 °C shown in Fig. 9. All the five alloys almost absorb no hydrogen before reaching the platform pressure due to the front of each PCT curve is almost linear rise. When it reaches the plateau pressure with pressure increasing, the alloy begins to absorb a lot of hydrogen, and the first hydrogen absorption platform is formed. This platform belongs to Mg/MgH₂ hydriding. With the continuous increase of hydrogen pressure, the amount of hydrogen absorption also increases slowly, and then a second hydrogen absorption platform appears which belongs to Mg₂Ni/Mg₂NiH₄ hydriding. After that, the alloy reaches the saturated state with hydrogen pressure increasing. The dehydriding evolution is the opposite with the decrease of hydrogen pressure. Here we only discuss the pressure of the first platform pressure belonging to Mg/MgH₂. According to the pressure value of platform pressure at different temperatures, the enthalpy change (ΔH) and entropy change (ΔS) can be calculated by the van't Hoff equation:³⁸

ln[P(H₂)/P₀] = ΔH/(RT) − ΔS/R (3)

in this equation, P(H₂) is the equilibrium hydrogen gas pressure, P₀ means the standard atmospheric pressure, T represents the sample temperature and R expresses the gas constant, respectively. The van't Hoff plots, ΔH and ΔS are also shown in the inserted graphs in Fig. 9. On the basis of Fig. 9(f), the ΔH(de) of Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys is 93.372, 80.114, 77.227, 72.456 and 77.911 kJ mol⁻¹ respectively shown in Table 1. For comparison, the ΔH(de) of as-cast Mg₈₅Cu₅Ni₁₀ was 98.892 kJ mol⁻¹ and the ΔH(de) of Mg₈₅Cu₅Ni₁₀–x wt% CeO₂ (x = 0, 4, 8) alloys were 92.441, 79.548, and 80.018 kJ mol⁻¹ respectively.²⁷ The ΔH(de) values of TiF₃ catalyzed alloys are lower than that of as-cast Mg₈₅Cu₅Ni₁₀ alloy and Mg₈₅Cu₅Ni₁₀–0 wt% TiF₃ illustrating that thermodynamic stability of the alloys decrease. Although TiF₃ is decomposed into TiH₂ and MgF₂ during the hydrogen absorption and desorption cycles, some defects, grain boundaries and nano interfaces are also added. Because the bond energy of Mg–H bond at the grain boundary is relatively low, the thermodynamic properties of the alloys are improved.

Fig. 9 PCT curves and corresponding van't Hoff plots of Mg₈₅Cu₅Ni₁₀ + x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys different temperatures: (a) x = 0; (b) x = 2; (c) x = 4; (d) x = 6; (e) x = 8 and enthalpy change (ΔH) of dehydrogenation (f).

Conclusions

Mg₈₅Cu₅Ni₁₀–x wt% TiF₃ (x = 0, 2, 4, 6, 8) alloys were prepared via ball milling. The compounds contained Mg, Mg₂Ni, Mg₂Cu and TiF₃. The alloys catalyzed by TiF₃ have faster hydriding/dehydriding kinetics and lower thermodynamic stability, even though TiF₃ decomposes into TiH₂ and MgF₂ during hydriding and dehydriding. TiF₃, TiH₂ and MgF₂ promote to forming crystal defects, dislocations, grain boundaries and nanocrystals which have a positive effect on accelerating the rate of hydrogen absorption and desorption. The dehydrogenation activation energy E_a(de) and dehydrogenation enthalpy ΔH(de) are reduced to 81.462 from 116.767 kJ mol⁻¹ and 72.456 from 93.372 kJ mol⁻¹ respectively by 6 wt% TiF₃. An appropriate amount of TiF₃ can improve the kinetics and thermodynamics of hydrogen absorption and desorption of Mg₈₅Cu₅Ni₁₀.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is financially supported by the National Natural Science Foundations of China (51901105 and 51871125), Natural Science Foundation of Inner Mongolia, China (No. 2019BS05005), and Inner Mongolia University of Science and Technology Innovation Fund (2019QDL-B11).

References

1. K. Manickam, P. Mistry, G. Walker, D. Grant, C. E. Buckley, T. D. Humphries, M. Paskevicius, T. Jensen, R. Albert, K. Peinecke and M. Felderhoff, Int. J. Hydrogen Energy, 2019, 44, 7738–7745.
2. J. J. Liu, H. H. Cheng, S. M. Han, H. F. Liu and J. Huot, Energy, 2020, 192, 116617.
3. C. Milanese, T. R. Jensen, B. C. Hauback, C. Pistidda, M. Dornheim, H. Yang, L. Lombardo, A. Zuettel, Y. Filinchuk, P. Ngene, P. E. de Jongh, C. E. Buckley, E. M. Dematteis and M. Baricco, Int. J. Hydrogen Energy, 2019, 44, 7760–7774.
4. S. Yasuoka, J. Ishida, K. Kishida and H. Inui, J. Power Sources, 2017, 346, 56–62.
5. Y. H. Zhang, M. Ji, Z. M. Yuan, W. G. Bu, Y. Qi and S. H. Guo, RSC Adv., 2017, 7, 37689–37698.
6. X. Y. Chen, R. R. Chen, X. Ding, H. Z. Fang, X. Z. Li, H. S. Ding, Y. Q. Su, J. J. Guo and H. Z. Fu, Energy, 2019, 166, 587–597.
7. N. A. A. Majid, N. Maeda and M. Notomi, Int. J. Hydrogen Energy, 2019, 44, 29189–29195.
8. B. Galey, A. Auroux, S. S. Etienne, M. Grellier, S. Dhaher and G. Postole, Sustainable Energy Fuels, 2018, 2, 2335–2344.
9. F. A. H. Yap, N. N. Sulaiman and M. Ismail, Int. J. Hydrogen Energy, 2019, 44, 30583–30590.
10. V. A. Yartys, M. Baricco, J. B. von Colbe, D. Blanchard and C. Zlotea, Int. J. Hydrogen Energy, 2019, 44, 7809–7859.
11. B. X. Dong, H. Tian, Y. C. Wu, F. Y. Bu, W. L. Liu and Y. L. Teng, Int. J. Hydrogen Energy, 2016, 41, 14507–14518.
12. Y. Yin, B. Li, Z. M. Yuan, Y. Qi and Y. H. Zhang, RSC Adv., 2018, 8, 34525–34535.
13. İ. Muz and M. Atiş, J. Alloys Compd., 2016, 667, 275–281.
14. Q. Luo, J. D. Li, B. Li, B. Liu, H. Y. Shao and Q. Li, J. Magnesium Alloys, 2019, 7, 58–71.
15. N. Lobo, A. Takasaki, K. Mineo, A. Klimkowicz and K. Goc, Int. J. Hydrogen Energy, 2019, 44, 29179–29188.
16. K. Kitabayashi, K. Edalati, H. W. Li, E. Akiba and Z. J. Horita, Adv. Energy Mater., 2020, 22, 1900027.
17. S. C. Gao, X. H. Wang, H. Z. Liu, T. He, Y. Y. Wang, S. Q. Li and M. Yan, J. Power Sources, 2019, 438, 227006.
18. J. Li, B. Li, H. Shao, W. Li and H. Lin, Catalyst, 2018, 8, 89.
19. H. Zhang, X. Zheng, X. Tian, Y. Liu and X. Li, Prog. Nat. Sci.: Mater. Int., 2017, 27, 50–57.
20. K. Alsabawi, A. Mac, E. Gray and C. J. Webb, Int. J. Hydrogen Energy, 2019, 44, 2976–2980.
21. M. M. Hu, X. B. Xie, M. Chen, C. X. Zhu and T. Liu, J. Alloys Compd., 2020, 817, 152813.
22. M. Ismail, Y. F. A. Halim, N. N. Sulaiman and M. H. I. Ishak, J. Alloys Compd., 2016, 678, 297–303.
23. Z. M. Yuan, B. W. Zhang, Y. H. Zhang, S. H. Guo, X. P. Dong and D. L. Zhao, J. Mater. Sci. Technol., 2018, 34, 1851–1858.
24. Z. M. Yuan, W. Zhang, P. L. Zhang, Y. H. Zhang, W. G. Bu, S. H. Guo and D. L. Zhao, RSC Adv., 2017, 7, 56365–56374.
25. Y. H. Zhang, W. Zhang, J. L. Gao, Z. M. Yuan, W. G. Bu and Y. Qi, Acta Metall. Sin., 2017, 11, 10–18.
26. Y. H. Zhang, M. Ji, Z. M. Yuan, J. L. Gao, Y. Qi, X. P. Dong and S. H. Guo, J. Rare Earths, 2018, 36, 409–417.
27. Y. Yin, B. Li, Z. M. Yuan, Y. Qi and Y. H. Zhang, J. Rare Earths, 2020, 38, 983–993.
28. Z. M. Yuan, T. Yang, W. G. Bu, H. W. Shang, Y. Qi and Y. H. Zhang, Int. J. Hydrogen Energy, 2016, 41, 5994–6003.
29. T. Liu, T. W. Zhang, X. Z. Zhang and X. G. Li, Int. J. Hydrogen Energy, 2011, 36, 3515–3520.
30. Y. H. Zhang, T. Yang, W. G. Bu, Y. Cai, G. F. Zhang and D. L. Zhao, Trans. Nonferrous Met. Soc. China, 2013, 23, 3668–3676.
31. N. S. Mustafa and M. Ismail, Int. J. Hydrogen Energy, 2019, 44, 21979–21987.
32. N. N. Sulaiman and M. Ismail, Int. J. Hydrogen Energy, 2019, 44, 30574–30582.
33. M. Pourabdoli, S. Raygan, H. Abdizadeh and D. Uner, Int. J. Hydrogen Energy, 2013, 38, 11910–11919.
34. G. Barkhordarian, T. Klassen and R. Bormann, J. Alloys Compd., 2006, 407, 249–255.
35. L. J. Wei, H. Gu, Y. F. Zhu and L. Q. Li, Int. J. Hydrogen Energy, 2012, 37, 17146–17152.
36. T. Liu, Y. R. Cao, C. G. Qin, W. S. Chou and X. G. Li, J. Power Sources, 2014, 246, 277–282.
37. T. Kimura, H. Miyaoka, T. Ichikawa and Y. Kojima, Int. J. Hydrogen Energy, 2013, 38, 13728–13733.
38. H. Falahati and D. P. J. Barz, J. Hydrogen Energy, 2013, 38, 8838–8851.

---