Preparation and hydrogen storage properties of MgH₂-trimesic acid-TM MOF (TM=Co, Fe) composites

1. Introduction

Undoubtedly, the global increasing demand of energy production and environmental contamination caused by the enormous consumption of fossil fuels are the most significant issues in 21st century [1,2]. As one kind of green and renewable candidates to replace fossil fuels, environmental friendly hydrogen energy system possesses a high energy content per unit mass (142 MJ/kg) with pure water as the only combustion product [3,4]. MgH₂ has been considered as one of the applicable hydrogen storage carriers in virtue of its high hydrogen uptake capacity (7.6 wt%), abundant deposit as well as low cost. However, the enthalpy value (74.7 kJ/mol H₂) of MgH₂ indicates its high thermodynamics stability [5]. The high operating temperature (>623 K) and the sluggish hydrogen de/absorption kinetics of MgH₂ impede the practical application of magnesium-based material for hydrogen storage [6,7]. To enhance the thermodynamics and de/hydriding kinetics of MgH₂, a series of strategies including alloying [[8], [9], [10]], nanoconfinement [[11], [12], [13]] as well as nanoscaling [[14], [15], [16], [17], [18]] have been widely used in Mg based hydrogen storage material. Besides, the addition of catalytic metals e.g. Ni [[19], [20], [21]], Co [22,23], Fe [[22], [23], [24], [25], [26]], Ti [[25], [26], [27]], V [[27], [28], [29]], Cu [30,31], Pd [[30], [31], [32]], Nd [33,34], Y [35], metal oxides [[36], [37], [38]] and metal halides [[39], [40], [41], [42]] etc. into Mg/MgH₂ matrix have been proved to be another effective strategy accelerating the sorption kinetic performances of MgH₂. On the other hand, like other kinds of solid state carriers including carbon nanostructures, pure metals and metallic compounds, metal organic frameworks (MOFs) have been developed as one of candidates for physisorption of hydrogen since the pioneer work of Yaghi et al. in 2003 on the hydrogen absorption in MOFs [43,44]. Numerous researches on the hydrogen storage behaviors of MOFs have been reported in the past decade. For instance, MOFs such as Zn₄O(BDC)₃ (MOF-5) and Zn₄O(BTB)₂ (MOF-177) exhibited uptakes of 10 wt% and 7.0 wt% H₂ at 77 K, respectively [45,46]. Farha et al. [47] reported a MOF named NU-100 with a high Brunauer Emmett Teller (BET) specific surface area of 6143 m²/g and the MOF exhibits a high hydrogen storage capacity (approximately 9 wt%) under 56 bar H₂ pressure at 77 K. With the purpose to further improve the hydrogen physisorption capacities of MOFs at 77 K or room temperature, extensive work on MOFs mainly focus on the functionalization of ligands, manipu-lation of pore size and increment of pore volume and specific surface area [48]. While the catalytic effects of MOFs on the (de)hydriding thermodynamics and kinetics of Mg/MgH₂, have rarely been reported in the literature.

In this work, trimasic acid (TMA) and Co (II), Fe (II) based MOFs (TMA-TM MOFs) with environmentally friendly solvents were firstly synthesized in deionized water (DI water) at 298 K. To explore the catalytic role TMA-TM MOFs played in Mg/MgH₂, the overall hydrogen storage properties and corresponding thermodynamics and kinetics of pure MgH₂ and MgH₂-TMA-TM MOF (MgH₂-TM MOF) (TM = Co, Fe) composites were systematically investigated.

2. Experimental

2.1. Sample preparation

Co(NO₃)₂·6H₂O (Aladdin, 99.99%) and FeSO₄·7H₂O (Aladdin, 99.95%) as metal ion sources, trimasic acid (TMA) (H₃BTC), (Aladdin, 98%) was used as organic linker for preparations of TMA-TM MOFs. Sodium hydroxide (Aladdin, 98%) was used to adjust the pH value of TMA solution. A Co(NO₃)₂ solution was prepared by dissolving 0.009 mol Co(NO₃)₂·6H₂O into 60 mL DI water. A FeSO₄ solution was obtained by the same operation. Meanwhile, 0.0045 × 2 mol TMA was dissolved into 120 mL DI water and the pH of the solution was adjusted to 7 by dropping wise addition of 0.5 M NaOH. The introduction of hydroxyls by adding appropriate NaOH into the TMA solution neutralizes the hydrogen ions derived from H₃BTC and facilitate the complete dissolution of H₃BTC into DI water. Then the prepared TMA solution (120 mL) was equally divided into two beakers and Co²⁺ and Fe²⁺ containing solutions were then added into the TMA solutions, respectively, under a constant stirring rate of 1000 rpm. Minutes later, pink-colored (TMA-Co MOF) and brown-colored (TMA-Fe MOF) precipitates both gradually appeared in their individual beakers. Ten hours of mixing time was given for the complete formations of MOFs. The synthesized precipitates were washed with excess DI water to remove residuary of reactants. Finally, the prepared TMA-TM (TM = Co, Fe) MOFs were dried in vacuum at 333 K for 8 h, followed by another 24 h at 373 K to preliminarily remove the external solvent of MOFs. Then the dried MOFs were preserved in the glove box for further measurements.

The MgH₂-TM MOF (TM = Co, Fe) composites were prepared in a planetary ball mill with the ball-to-powder weight ratio of 20:1. For the preparation of each MgH₂-TM MOF composite, about 1.8 g MgH₂ (produced by Shanghai Mg Power Technology Co. Ltd., purity >98%) [19] and 0.2 g TMA-TM MOF (TM = Co or Fe) powders were added into a sealed steel container (100 mL) under argon atmosphere (1 bar) and ball milled at 150 rpm for 8 h. The rotation speed of 150 rpm for ball milling was selected for the homogenization of TMA-TM MOFs powders with MgH₂ in the MgH₂-TM MOF (TM = Co, Fe) composites without severe destruction of the MOF structures. Same amount (2.0 g) of pure MgH₂ was also ball milled in another container with same parameters for comparison.

2.2. Characterization

The surface areas, average pore volumes and pore sizes of the TMA-TM (TM = Co, Fe) MOFs were examined by N₂ absorption on a 3H-2000PS2 (Beishide Co. Ltd). The structural analysis of the TMA-TM MOFs was performed via Fourier transform infrared spectroscopy (FTIR) by a Nicolet iS5 (Thermo Fisher Scientific Inc.) equipped with a horizontal ATR accessory. The thermal stability of MOFs was evaluated on thermolgravimetry (TG) and differential scanning calorimetry (DSC, Netzsch STA449F3 Jupiter) analysis by heating samples from 293 K to 823 K under 1 bar of argon atmosphere (heating rate =5 K/min). X-ray diffraction (XRD, D/max 2550 V L/PCX) measurements were carried out for the phase identification of the MgH₂-TM MOF (TM = Co, Fe) and the pure MgH₂ powders in various states. A scotch tape was used on the sample holder to prevent the samples from oxygen and H₂O contaminations and the collected data were analyzed using Jade 6.5 software. The morphology and microstructures of the MgH₂-TM MOF (TM = Co, Fe) composites were identified by a transmission electron microscope (TEM, JEM-2100 F) and a scanning electron microscope (SEM, Phenom-XL) equipped with an energy dispersive spectrometer (EDS). The hydriding/dehydriding properties of the samples were evaluated using a Sievert type pressure-composition-temperature (PCT) apparatus. The auto-PCT sorption tests at various temperatures including 598, 623 and 648 K were performed with minimum and maximum hydrogen pressures of 0.01 and 4 MPa, respectively. Before the PCT measurements, the MgH₂-TM MOF (TM = Co, Fe) composites and the as milled pure MgH₂ were all fully activated at 753 K for 1 h by operating a dehydrogenation process in vacuum. Dehydrogenation properties of the samples were examined by DSC under 1 bar argon atmosphere at different heating rates (3, 5 and 10 K/min) from 298 K to 773 K. The sample powders were both hydrogenated at 623 K under 3 MPa hydrogen pressure for 2 h and dehydrogenated at 673 K in vacuum for 2 h in the PCT apparatus for XRD, SEM, TEM as well as DSC measurements.

3. Results and discussion

3.1. MOF characterization

View original graphic| Download| PPT slide

Fig. 1.   Schematic diagram of synthesized TMA-TM MOFs (TM = Co, Fe).

View original graphic| Download| PPT slide

Fig. 2.   SEM images of pure MgH₂ (a), TMA-Co MOF (b) and TMA-Fe MOF (c).

Fig. 3 displays the FTIR spectra of the TMA and TMA-TM (TM = Co, Fe) MOFs obtained in waterless testing environment for structural identification. As shown in Fig. 3(a), the most important functional groups in TMA are carbonyl groups at about 1700 cm^-1, C—O single stretching bonds at about 1454 and 1403 cm^-1. Additionally, The peak at 1274 cm^-1 corresponds to C—O—H stretching bonds. According to Fig. 3(b), the C=O bonds peak split into two peaks at 1616 and 1558 cm^-1 for TMA-Co MOF, demonstrating the interaction of Co ion and carboxylate groups [49]. It is worth noting that the peaks at 1274 and 930 cm^-1 disappear in the FTIR spectra of TMA-Co MOF as compared to that of TMA, which is ascribed to the O–H bonds breaking during synthetic process (see Fig. 1). The FTIR spectra of the TMA-Fe MOF are almost identical to that of TMA-Co MOF, suggesting a similar microstructure of two kinds of MOFs. To estimate the thermostability of the MOFs, the TG and DSC profiles of TMA-Co MOF and TMA-Fe MOF are presented in Fig. 4(a) and (b), respectively. According to the distinct plateaus in TG curves for each of the MOF samples, the decomposition processes both contain two steps at about the temperature ranges of 360-450 and 700-800 K for TMA-Co MOF as well as ones of 400-500 and 750-800 K for TMA-Fe MOF. It is easily to understand that the first mass losses can be due to the loss of solvent (DI water) and the other ones at relatively high temperature range are put down to the degradations of TMA-TM MOFs. It should be pointed out that the residual solvent in the micro-structures of MOFs could not be completely removed until the temperature increases to 360-450 K for TMA-Co MOF as well as ones of 400-500 K for TMA-Fe MOF. Furthermore, two broader endothermic peaks in DSC curve for each of MOF correspond well to the decomposition steps discussed above. Moreover, when heating up to 773 K at the heating rate of 5 K/min, the weight losses of the as prepared TMA-TM MOFs (TM = Co, Fe) are evaluated to be about 53 wt% and 60 wt%, respectively.

View original graphic| Download| PPT slide

Fig. 3.   FT-IR spectra of TMA (a), TMA-Co MOF (b) and TMA-Fe MOF (c) synthesized in DI water.

View original graphic| Download| PPT slide

Fig. 4.   TG and DSC curves of TMA-Co MOF (a) and TMA-Fe MOF (b) at the heating rate of 5 K/min.

3.2. Phase components and microstructure characterizations of the MgH₂-TM MOF (TM = Co, Fe) composites and the pure MgH₂

Fig. 5(a)-(j) displays the XRD patterns of the TMA-TM MOFs (TM = Co, Fe), pure MgH₂ and the MgH₂-TM MOF composites in various states. As shown in Fig. 5(a) and (b), the characteristic peaks for the crystallographic TMA-Co MOF can be found at relatively low diffraction angles (2θ) including 17.62°, 18.81°, 27.26° and 35.58°. They are much similar with those of TMA-Co MOF (17.78°, 18.92°, 27.33° and 35.74°, respectively), suggesting the analogous crystal structures for TMA-TM MOFs (TM = Co, Fe). Apart from a small amount of Mg, almost all diffraction peaks of the as milled MgH₂ powders (Fig. 5(c)) and MgH₂-TM MOF composites (Fig. 5(d) and (e)) can be indexed with pure tetragonal β-MgH₂ MgH₂. Analogously, Same phases can also be observed in pure MgH₂ powder after hydrogenation at 623 K for 2 h (Fig. 5(f)). Moreover, peak from TMA-TM MOFs at 2θ ≈18.8° can also be detected in MgH₂-TM MOF composites after the ball milling, demonstrating the presences of TMA-TM MOFs (TM = Co, Fe) in as milled composites without the destruction of crystal structures. When hydrogenation are performed at 623 K for 2 h, the MgH₂-Co MOF sample is composed of β-MgH₂, Mg₂Co and MgO, while α-Fe phase can be found along with β-MgH₂ and MgO in the hydrogenated MgH₂-Fe MOF composite (see Fig. 5(g) and (h)). The appearance of Mg₂Co phase indicating a reaction of Co and Mg in the MgH₂-Co MOF powder at high temperatures. The reaction can be expressed as follows:

2 Mg + Co = Mg₂Co (1)

View original graphic| Download| PPT slide

Fig. 5.   XRD patterns of TMA-Co MOF (a), TMA-Fe MOF (b), pure MgH₂ and MgH₂-TM MOF (TM = Co, Fe) composites at various states: as milled (c, d, e), re-hydrogenated at 623 K for 2 h (f, g, h), and dehydrogenated at 673 K for 2 h (i, j), respectively.

For the dehydriding processes in Fig. 5(i) and (j), the broad diffraction peaks of two kinds of composites both match for hexagonal Mg as majority phase when the re-hydrogenated MgH₂-TM MOF samples are heated in vacuum at 673 K for 2 h. Meanwhile the Mg₂Co and α-Fe phases can be observed in dehydrogenated MgH₂-Co MOF and MgH₂-Fe MOF composites, respectively. Despite possible oxidation occurred during composites preparations, the presence of oxygen in MgO, cobalt in Mg₂Co and iron in α-Fe phases reveals the degradation of TMA-TM MOFs (TM = Co, Fe) at high temperatures. According to Fig. 4, Fig. 5, it can be deduced that the TMA-TM MOFs in composites are both destabilized by the increment of temperature. The chemical bonds among component in MOFs rupture at the high temperature (753 K) during activation processes and consequently, resulting in the collapses of the structure in TMA-TM MOFs (TM = Co, Fe). The former ball-milling processes increase the contact areas between MgH₂ particles and the MOF additives. The large contact areas make it easy for cobalt and oxygen from TMA-Co MOF to react with surrounding Mg to form Mg₂Co and MgO, respectively. The iron from the decomposition of TMA-Fe MOF does not react with Mg even at high temperatures, thus α-Fe phase can be detected in both re/dehydrogenated MgH₂-Fe MOF composites. It is necessary to point out that there is no trace of Mg₂CoH₅ phase presented in the XRD pattern of hydrogenated MgH₂-Co MOF sample. Indeed, it has been widely reported that high temperature, high hydrogen pressure and long duration are three vital factors dominant the generation of Mg₂CoH₅ phase. For instance, Zolliker et al. [50] reported a sintering technique in a condition of 690-720 K, 4-6 MPa H₂ pressure to prepare Mg₂CoH₅. Selvam et al. [51] obtained Mg₂CoH₅ sample by high-pressure sintering at 723 K under 9 MPa H₂ for more than 7 d. The conditions given in this work (at 623 K under 3 MPa H₂ pressure for 2 h) is unsuitable for the formation of Mg₂CoH₅ during hydrogenation process. Analogously, α-Fe phases can both be detected in rehydrogenated/dehydrogenated MgH₂-Fe MOF samples in Fig. 5(h) and (j), while the characteristic peaks of Mg₂FeH₆ phase are invisible in Fig. 5(h). The similar phenomena have been reported in Mg@Fe composite and Mg-Fe-La ternary composite synthesized through arc plasma method [52,53]. In fact, it has been established that the formation of Mg₂FeH₆ phase needs conditions including high temperature, high H₂ pressure as well as long duration for hydriding [54]. This is the main interpretation for the absence of Mg₂FeH₆ phase in hydrogenated MgH₂-Fe MOF powder under given conditions. The contact areas between the MOF additives and MgH₂ particles increases during the ball-milling process and parts of needle-like TMA-Co MOF and rod-shaped TMA-Fe MOF can penetrate the extremely thin MgO layer covered on MgH₂ particles.

The typical SEM images of as milled, hydrogenated and dehydrogenated MgH₂ samples are given in Fig. 6(A), (H) and (O), respectively. It can be found that the average particle size of the as milled pure MgH₂ remains to be about 10 μm, which means that the ball-milling process at 150 rpm for 8 h could not significantly change the particle size of MgH₂ powder. Besides, the irregular-shaped MgH₂ and Mg with the particle sizes of about 30 μm in Fig. 6(H) and (O) indicating a noticeable agglomeration of MgH₂/Mg particles during (de)/hydriding at high temperature. Similar phenomenon can also be found on the MgH₂-TM MOF (TM = Co, Fe) samples in various states (see Fig. 6(B)-(G)). The particle sizes of the composite powders both increase from about 10 μm (as milled state) to 20-30 μm (re/dehydrogenated states). What’s more, Fig. 6(I)-(K), (L)-(N) and (P)-(U) of EDS maps reveal the distributions of Co, Fe and O, respectively. For the as milled MgH₂-TM MOF (TM = Co, Fe) composites, Co, O and Fe, O are both uniformly distributed on the large MgH₂ particles of the MgH₂-Co MOF and MgH₂-Fe MOF powders, respectively. It can also be clearly observed from Fig. 6(I)-(N) that Co and Fe distributed homogeneously on the MgH₂ (re-hydrogenated) and Mg (dehydrogenated) particles, demonstrating the uniform distributions of Mg₂Co and α-Fe phases on the surface of MgH₂/Mg particles in various states. It is due to the appropriate ball-milling process during the preparations of the composites.

View original graphic| Download| PPT slide

Fig. 6.   SEM micrographs of the as milled (A), hydrogenated (H) and dehydrogenated (O) pure MgH₂ powders, the as milled (B), (E), hydrogenated (C), (F) and dehydrogenated (D), (G) MgH₂-TM MOF (TM = Co, Fe) composites, respectively. The corresponding EDS mappings showing distributions of Co (I-K), Fe (L-N) and O (P-U) (As milled: after ball-milling; Ab: absorption; De: desorption).

TEM observations were conducted in Fig. 7 to further reveal the morphology of the hydrogenated MgH₂-TM MOF composites. Fig. 7(a) and (e) display the typical bright field micrographs of the MgH₂-TM MOF (TM = Co, Fe), respectively. The particle size of the irregular-shaped MgH₂-Co MOF composite is measured to be about 500 nm, which is larger than that of MgH₂-Fe MOF. Some dispersive particles with dark contrast can be observed on the surfaces of the composites. Fig. 7(b) and (f) illustrate the selected area electron diffraction (SAED) patterns of the MgH₂-TM MOF (TM = Co, Fe) samples. All the diffraction rings or points of the MgH₂-Co MOF sample confirm the presences of MgH₂, MgO and Mg₂Co phases. At the same time, the SAED pattern in Fig. 7(f) for the MgH₂-Fe MOF sample can be indexed with (101), (002), (202) and (103) for MgH₂, (110), (211) and (310) for α-Fe and (200) for MgO. The phase identifications for the composites are in accordance with the XRD analyses. The appearance of MgO phase is mainly ascribed to the reaction between Mg and O (provided by the decomposition of TMA-TM MOF) during de/hydriding cycles. With the purpose to reveal the morphology of Mg₂Co and α-Fe phases covered on MgH₂, the diffraction points of Mg₂Co (442) plane and α-Fe (211) plane are taken to draw the dark field TEM micrographs for the MgH₂-Co MOF and MgH₂-Fe MOF, respectively (see Fig. 7(c) and (g)). Both of the Mg₂Co and α-Fe nanoparticles shown in bright contrasts are well dispersed on the surfaces of large-sized MgH₂, which is in good agreement with the EDS mappings of Co and Fe in former SEM observations. It is also found in Fig. 7(c) and (g) that the particle sizes of Mg₂Co particles are in the range of 5-30 nm, which are larger than those of α-Fe (several nanometeres). Moreover, High Resolution Electron Microscope (HRTEM) was performed to investigate the micro-structures of the MgH₂-TM MOF (TM = Co, Fe) composites. Two of the interplanar spacings of the selected zones in Fig. 7(d) can be determined to be about 0.2203 and 0.2189 nm, both of which are nearly identical to that of Mg₂Co (511) plane (0.2199 nm). Similarly, MgH₂ (101) plane and MgH₂ (200) plane with interplanar spacings of 0.2537 nm and 0.2272 nm can also be confirmed in Fig. 7(d). As for the HRTEM of the MgH₂-Fe MOF sample in Fig. 7(h), an atomic layer with the interplanar spacing of 0.2537 nm is identified to be MgH₂ (101) plane (0.2510 nm). Simultaneously, another atomic layer structure is found in the enlarged image presented in the inset Fig. 7(h). The interplanar spacing is measured to be 0.2035 nm, which is in good agreement with that of Fe (110) plane (0.2026 nm).

View original graphic| Download| PPT slide

Fig. 7.   Typical bright field TEM micrographs of the hydrogenated MgH₂-Co MOF (a) and MgH₂-Fe MOF (e) samples, the corresponding SAED patterns (b), (f), the dark field micrographs (c), (g) contributed by Mg₂Co (442) and α-Fe (211) diffraction rings and the HRTEM images (d), (h), respectively.

3.3. Hydrogen storage behaviors of the pure MgH₂ and the MgH₂-TM MOF (TM = Co, Fe) composites

The (de)hydriding thermodynamics and kinetics of the pure MgH₂ and the MgH₂-TM MOF (TM = Co, Fe) powders are estimated by using the PCT technique. The PC isotherms of the as milled MgH₂ and the MgH₂-TM MOF (TM = Co, Fe) composites measured at 598, 623 and 648 K in a H₂ pressure range from 0.1 to 40 bar are presented in Fig. 8(a), (c) and (e), respectively. The parameters of hydrogen storage behaviors for the three samples collected from PCT tests are also listed in Table 2. It can be found in Fig. 8(a) that the single slant plateau associates with the re-dehydrogenation of Mg is presented at each of assigned temperatures. What’s more, there is no distinct hydrogen desorption plateau can be observed in the PC isotherm for pure MgH₂ at each of testing temperatures. In addition, a sharp drop of the hydrogen plateau pressure at the initial stage of dehydrogenation is observed at 598 K. Hence, it is reasonable to believe that the as milled MgH₂ powder still possesses poor hydrogen de/absorption performances during re/dehydrogenation. On the contrary, the single flat plateaus owning to the formation and decomposition of MgH₂ are obtained at 598, 623 and 648 K for the MgH₂-Co MOF and MgH₂-Fe MOF powders in Fig. 8(c) and (e), respectively. Each of the hydrogen desorption profiles for the composites exhibits an evident plateau with no diminution of H₂ pressure, even the temperature decreases to 598 K. It means that the hydrogen (de)absorption kinetic properties of the MgH₂-TM MOF (TM = Co, Fe) composites can be effectively aggrandized with the additions of the TMA-TM MOFs (TM = Co, Fe). According to the data collected in Table 2, the maximum hydrogen uptake amounts at 598, 623 and 648 K for the MgH₂-Co MOF composite are 5.08, 5.14 and 5.19 wt%, respectively, which are a little lower than those of the MgH₂-Fe MOF composite (5.19, 5.30 and 5.37 wt%, respectively). The reversible hydrogen sorption capacities of the composites increase with the rise of temperatures, suggesting the enhanced sorption kinetics at higher temperatures. It is noteworthy that the hydrogen uptake amounts of the MgH₂-TM MOF (TM = Co, Fe) composites at series temperatures are all lower than the theoretical maximum hydrogen storage capacity of pure Mg (7.6 wt%) [4,5], owning to the addition of TMA-TM MOFs (TM = Co, Fe) and the formation of MgO phase. Fig. 8(b), (d) and (f) illustrates the van’t Hoff plots and the corresponding linear fittings (lnP vs. 1000/T) (P donates the hydrogen pressure for ab/desorption plateau and T represents the corresponding temperature) of as milled MgH₂ and the MgH₂-TM MOF samples, respectively. The hydriding and dehydriding enthalpies (ΔH_ab and ΔH_de) of the dehydrogenated MgH₂-Co MOF sample are therefore calculated to be -73.1 ± 3.2 and 78.2 ± 3.4 kJ/mol H₂. As to the MgH₂-Fe MOF composite after dehydriding, the ΔH_ab and ΔH_de are evaluated to be -76.7 ± 2.3 and 77.8 ± 3.4 kJ/mol H₂. They are fairly close to the standard values of Mg (±74.7 kJ/mol H₂) [55] as well as the dehydrogenated MgH₂ (-73.8 ± 3.3 and 76.1 ± 2.5 kJ/mol H₂, respectively) in this work. Herein, the thermodynamic performances of the MgH₂-TM MOF (TM = Co, Fe) remain nearly unchanged with the addition of TMA-TM MOFs.

View original graphic| Download| PPT slide

Fig. 8.   PC isotherms and the van’t Hoff plots of pure MgH₂ (a, b), the MgH₂-Co MOF (c, d) and the MgH₂-Fe MOF (e, f) composites tested at 589, 623 and 648 K.

The absorption profiles of the dehydrogenated MgH₂ and the MgH₂-TM MOF (TM = Co, Fe) composites tested at 448, 473, 498, 523, 548 and 573 K under 3 MPa H₂ are given in Fig. 9(a), (c) and (e), respectively. It can be found that the hydrogen absorption rate of the dehydrogenated MgH₂ evidently declines with the reduction of the temperature. To the opposite, the hydriding rate of the MgH₂-Fe MOF does not sharply decline with the decreasing of the temperature, demonstrating an improved absorption kinetics with respect to that of dehydrogenated pure MgH₂. It has been confirmed that the rate controlling factors of hydrogenation process in metals involve three steps [56]: (1) the hydrogen molecules dissociation on the surface of metal, (2) the penetration of hydrogen atoms from surface into the bulk metal, (3) the diffusion of H in the metal/hydride matrix. To further evaluate the enhanced kinetics of hydrogenation for the MgH₂-TM MOF (TM = Co, Fe) composites, an activation energy (E_a) for hydrogenation is carried out to estimate the overall energy barrier of all steps mentioned above. The reaction mechanism of hydrogen sorption behaviors of hydrogen storage materials can be systematically investigated by using suitable kinetic models [57]. Based on the Johnson-Mehl-Avrami-Kolmogorov (JMAK) model, the hydrogenation data can be systematically calculated and the linear equation of the sorption kinetics is defined as follows:

ln- [ln(1-α)]=ηlnk+ηlnt, (2)

where the value of α represents the fraction of phase transformation from metal (Mg) to its hydride (MgH₂) at time t, k and η donate a parameter of kinetics and the reaction order or Avrami exponent, respectively. With the rate constant k calculated from the equation above, the E_a is evaluated from the Arrhenius equation:

k=Aexp(-Ea/(RT)), (3)

where A donates a coefficient (temperature-independent), R and T represent the gas constant and the absolutely temperature, respectively. As shown in Fig. 9(b), (d) and (f), the E_a values of the dehydrogenated MgH₂-TM MOF (TM = Co, Fe) composites are calculated to be 73.9 ± 2.0 and 66.8 ± 3.3 kJ/mol H₂, respectively, both of which are much lower than that of dehydrogenated pure MgH₂ sample (100.7 ± 5.4 kJ/mol H₂). The notable decrease of the activation energy indicates lower energy barriers of hydrogen absorption for the composites, explaining the superior hydriding kinetics of the dehydrogenated MgH₂-TM MOF samples at relatively low temperatures.

View original graphic| Download| PPT slide

Fig. 9.   Hydrogen absorption profiles and the corresponding lnk-1000/T plots of pure MgH₂ (a, b), the MgH₂-Co MOF (c, d) and the MgH₂-Fe MOF (e, f) composites measured at 448, 473, 498, 523, 548 and 573 K under 3 MPa H₂ pressure for 3 h.

The transition metals (Co, Fe) have been proved to be considerable additives to accelerate the sorption kinetics of MgH₂/Mg [[58], [59], [60], [61], [62]]. The classical Johnson-Mehl-Avrami-Kolmogorov (JMAK) model describes the solid-state phase transitions as three processes including nucleation, growth and impingement. The growth process is considered as interface-controlled or diffusion-controlled mode [63]. For hydrogenation procedure, the transfer of hydrogen atoms to the reaction interface first happens and it involves following steps: (1) physisorption of H₂ on the surface of matrix; (2) chemisorption and dissociation of H₂ on the surface; (3) penetration of hydrogen atoms from surface into the matrix; (4) diffusion of hydrogen atoms to the reaction interface; (5) penetration of hydrogen atoms through the reaction interface. As the probable rate-controlling factors during solid-state phase transitions, the steps above are much slower than that of the chemical reactions [64]. As to the MgH₂-TM MOF (TM = Co, Fe) composites, Co and Fe covered uniformly on Mg play important roles for facilitating the dissociation of hydrogen molecules and consequently, leading to a dramatic decrease of the activation energy (E_a) [65,66]. Furthermore, nano α-Fe distributed on the surface of Mg particles can act as “gateways” to quickly deliver hydrogen throughout the bulk, which can be attributed to the faster hydrogen diffusion rate of bcc transition metals at low temperatures [52,67]. Similarly, Mg₂Co nanoparticles covered on Mg are also critical to facilitate the sorption kinetics of the Mg-Co MOF-H composites by accelerating the hydrogen diffusion into the matrix (MgH₂) [68,69]. In addition, the interfaces between Mg particles and TM (TM = Co, Fe) can be considered as defects, which provide active sites to assist the nucleation for MgH₂ during hydrogenation [70,71]. The beneficial effects expounded above contribute to the outstanding hydriding kinetics of the MgH₂-TM MOF (TM = Co, Fe) composites.

With the purpose to evaluate the dehydrogenation performances of the MgH₂-TM MOF (TM = Co, Fe) samples, the DSC profiles and the ln(β/T_p²) vs. 1000/T_p plots (β donates the heating rate, T_p represents the peak temperature) of pure MgH₂ and the composites powders are presented in Fig. 10. Only one broader endothermic peak can be observed at all heating rates for each sample shown in Fig. 10, which corresponds to the hydrogen release from β-MgH₂ phase. The peak desorption temperatures of the MgH₂-Co MOF composite at the heating rates of 3, 5 and 10 K/min are determined to be 635.7, 648.3 and 661.7 K, respectively, which are lower than those of pure MgH₂ without catalyst. Moreover, compared with other two kinds of samples above, the MgH₂-Fe MOF sample possesses the lowest peak dehydriding temperature (614.6, 624.3 and 640.3 K), respectively. It indicates a higher catalytic efficiency of TMA-Fe MOF in aggrandizing the dehydriding performance of MgH₂ than that of TMA-Co MOF. Additionally, the onset dehydrogenation temperature for the MgH₂-Fe MOF is remarkably declined to 584.4 K (heating rate = 10 K/min), which is 80.6 K lower than that of pure MgH₂ powder (665.0 K), as shown in Fig. 10(a) and (e).

View original graphic| Download| PPT slide

Fig. 10.   DSC profiles and the corresponding ln(β/T_p²)–1000/T_p plots for pure MgH₂ (a, b), the MgH₂-Co MOF (c, d) and the MgH₂-Fe MOF (e, f) composites at different heating rates.

An apparent dehydrogenation activation energy (E_d) is carried out to gain further understanding of the enhanced hydrogen release behaviors of the MgH₂-TM MOF samples. According to the data of DSC measurements at a series of heating rates, the activation energy of dehydrogenation (E_d) can be estimated by the Kissinger’s equation:

ln(β/$T_{ p }^{2}$)=A-E_d/(RT_p), (4)

where β donates the heating rate, T_p, A and R represent the peak temperature, a linear constant and the gas constant, respectively. According to the slope of the plots of ln(β/T_p²) vs. 1000/T_p given in Fig. 10(b), (d) and (f). The E_d of the MgH₂-Fe MOF sample is therefore evaluated to be 142.3 ± 6.5 kJ/mol H₂, much lower than those of MgH₂ (181.4 ± 9.2 kJ/mol H₂) and MgH₂-Co MOF (151.3 ± 9.4 kJ/mol H₂), revealing that the introduction of TMA-Fe MOF reduces the overall desorption energy barrier of MgH₂ more efficiently than that of the TMA-Co MOF.

It has been confirmed that transition metals (Co, Fe) can dramatically reduce the hydrogen desorption activation energy by providing active sites for recombination of hydrogen atoms towards molecular state [72]. During dehydriding process, nano α-Fe and Mg₂Co may play a “pathway” role to speed up the diffusion rate of hydrogen atoms from bulk of matrix to the surface. In addition, nucleation process has been proved to be one of the most crucial factors dominating the hydrogen desorption kinetics during dehydrogenation. For MgH₂-TM MOF(TM = Co, Fe) composites, Mg phase nucleates preferentially at the interfaces of TM (TM = Co, Fe) and MgH₂, thus results in a significant improvement on the desorption kinetic performances of the composites [70,73]. Roy et al. suggested the appropriate introduction of Co or Fe into Mg causes a shift in the Fermi level (E_f), through which the concentration of the hydrogen vacancies improved and consequently, resulting in an enhanced hydrogen diffusion rate during dehydriding process [74]. Most importantly, ab-initio calculations demonstrate that the Mg-H bond can be prominently weakened by the Fe atom between two MgH₂ clusters [75]. Meanwhile, Giusepponi et al. [76] suggested that the higher Fe coordination compared with Mg in the hydride generated a vacancy in the second H-shell. The void can be partially filled by closest Mg atoms to destabilize the crystal structure of hydride, which correspondingly increases the possibility for hydrogen atoms to diffusion toward the interface. Thereby, the existence of nano Mg₂Co in the MgH₂-Co MOF sample and α-Fe in the MgH₂-Fe MOF both exhibit high catalytic effectiveness in accelerating the dehydriding kinetics of MgH₂ [69].

It is necessary to note that, the E_a, E_d as well as peak dehydrogenation temperatures at all heating rates for MgH₂-Fe MOF composite are lower than those of MgH₂-Co MOF, suggesting a more significant enhancement on the hydrogen sorption kinetics for MgH₂ than that of TMA-Co MOF additive. Wang et al. suggested that, the intrinsic activity and the distribution state are crucial factors dominating the catalytic effectiveness of the catalyst particles [77]. Despite the intrinsic catalytic effect in MgH₂-TM MOF composites, Fig. 7(g) illustrates a more homogeneous distribution of α-Fe nanoparticles with smaller particle size than that of Mg₂Co on the surface of MgH₂ in composite powders (see Fig. 7(c)), explaining the better sorption kinetics of MgH₂-Fe MOF composite over that of MgH₂-Co MOF.

4. Conclusions

In the present work, the TMA-TM (TM = Co, Fe) MOFs were successfully synthesized in DI water at room temperature and introduced into MgH₂ to form the MgH₂-TM MOF (TM = Co, Fe) composites via ball milling. The microstructures and hydrogen storage properties of the composites were carefully investigated. The main conclusions are as follows:

(1) XRD analyses reveal the formation of Mg₂Co and α-Fe phases in re/dehydrogenated MgH₂-Co MOF and MgH₂-Fe MOF composites, respectively. SEM and TEM observations show the homogeneous distributions of nano Mg₂Co and α-Fe particles on the surface of large MgH₂ particles in MgH₂-TM MOF (TM = Co, Fe) composites.

(2) The enthalpies of hydrogenation/dehydrogenation for the MgH₂-TM MOF composites are close to the standard values for Mg (± 74.7 kJ/mol H₂), suggesting that the thermodynamics of the MgH₂-TM MOF composites remain unchanged with the addition of TMA-TM (TM = Co, Fe) MOFs.

(3) The E_a values of the dehydrogenated MgH₂-Fe MOF sample are estimated to be 66.8 ± 3.3 kJ/mol H₂, which is much lower than that of pure Mg (100.7 ± 5.4 kJ/mol H₂) and the dehydrogenated MgH₂-Co MOF (73.9 ± 2.0 kJ/mol H₂), indicating the better catalytic effect of Fe MOF over Co MOF on hydrogen absorption of Mg.

(4) The onset dehydriding temperature for MgH₂-Fe MOF at the heating rate of 10 K/min is 80.6 K lower than that of pure MgH₂. The E_d of the MgH₂-Co MOF and MgH₂-Fe MOF composites are 151.3 ± 9.4 and 142.3 ± 6.5 kJ/mol H₂, both of which are much lower than that of pure MgH₂ (181.4 ± 9.2 kJ/mol H₂).

(5) The improved hydrogen storage performances of the MgH₂-TM MOF (TM = Co, Fe) composites over pure MgH₂ can be attributed to the favorable effects of Mg₂Co and α-Fe nanoparticles distributed uniformly on the large-sized Mg/MgH₂ particles during (de)hydriding cycles, respectively.

Acknowledgments

This work was supported financially by the National Natural Science Foundation of China (No. 51771112), the National Key Research & Development Project (No. 2018YFB1502104), the Shanghai Science and Technology Commission (No. 14JC1491600) and the Shanghai Education Commission “Shuguang” Scholar Project (No. 16SG08).