Enhancements in Hydrogen Storage Properties of Magnesium Hydride Supported by Carbon Fiber: Effect of C–H Interactions

Abstract

Carbon-based materials with excellent catalytic activity provide new ideas for the development of magnesium-based hydrogen storage. C-H bonding interactions may play a key role in performance improvement. In this work, we comprehensively compare the magnesium-carbon cloth composites (CC) prepared by method of dry ball milling and wet impregnation. The results were that the hydrogen release activation energy (E_a) of MgH₂@CC composites prepared by wet immersion method was 175.1 ± 19.5 kJ·mol⁻¹, which was lower than that of pure MgH₂ (E_a = 213.9 ± 6.4 kJ·mol⁻¹), and the activation energy of MgH₂-CC composites prepared by ball milling method was 137.3 ± 8.7 kJ·mol⁻¹, which provided better results. The kinetic enhancement should be attributed to C-H interactions. The presence of carbon carriers and electron transfer to reduce the activation energy of Mg-H bond fracture. These results will provide further insights into the promotion of hydrogen ab-/desorption from metal hydrides.

1. Introduction

Energy is a potential driver of human production and social activity, and the depletion of fossil fuels and global climate change have prompted the search for clean energy carriers [1,2,3]. As a clean, efficient and abundant secondary energy source, hydrogen is considered one of the ideal energy carriers for storing and transferring energy in the future [4,5,6,7,8]. However, gaseous and liquid hydrogen storage have limitations in terms of safety and storage density for various applications [9,10,11,12]. Solid-state hydrogen storage, especially in the form of metal hydrides, offers the most compact and safe technology for storing hydrogen, and among the various solid-state hydrogen storage materials, MgH₂ is valuable for research due to its high hydrogen storage capacity (7.6%) and abundant magnesium metal resources (2.3%) [1,13,14,15,16,17,18,19]. However, the high thermodynamic stability and slow reaction kinetics of MgH₂ make it difficult to meet the requirements of practical applications [20,21,22,23].

Nanosizing [13,24,25,26,27] has been shown to be one of the effective ways to improve the thermodynamics/kinetics of the MgH₂ reaction, and chemical reduction and hydrogenation [28,29,30,31] have attracted interest due to their low cost and portable operating conditions [28,32]. Liu et al. [33] used carbon aerogel (CA) as a precursor, wet impregnated with dibutyl magnesium solution and hydrogenated to prepare Mg-CA composites. The average particle size of the Mg nanoparticles was 19.3 nm, and the enthalpies of hydrogenation and dehydrogenation were −65.1 and 68.8 kJ·mol⁻¹. However, the hydrogen uptake kinetics of the Mg–CA composites were slower than those of the LaMg-containing ones prepared by the arc plasma method, and the kinetics need to be further improved.

In recent decades, a variety of novel carbon isomers have been synthesized, such as nanotubes [26,34], nanofibers [28,35], graphene [36,37], fullerenes [38,39], Mxenes [40,41], and carbon nanospheres [42,43]. Due to carbon’s light weight, large specific surface area, thermal stability, load-bearing capacity, and carbon atom substitutability, the carbon composites studied those all contribute to enhance the kinetics of hydrogen uptake and/or desorption [44]. Carbon-based materials have been investigated as catalytic materials with the promise of further enhancing the MgH₂ kinetics. Andrey Lakhnik et al. [45] prepared MgH₂/C composites of elemental magnesium with graphite powder using high-energy ball milling method with particle sizes predominantly in the range of 10–30 μm. The materials have a minimum activation energy for hydrogen absorption of 65 kJ·mol⁻¹ and a hydrogen capacity of ~4.3 wt%, which gives them an excellent kinetic performance. It was also found that the activation energy of the absorption process monotonically decreases with the increase of the graphite powder surface area during sample synthesis. Carbon composites also have excellent cycling performance, as they affect hydrogen storage kinetics primarily by preventing particle agglomeration and improving thermal conductivity [46]. R.M. Carson et al. [47] found that Mg-5 wt% C retains a capacity of 5.5 wt% H₂ after 1000 cycles and is suitable for most commercial applications.

The surface of carbon fiber has a high degree of surface defects and abundant growth sites, which can be used as a good growth carrier. Meanwhile, carbon fibers can improve cycling performance and inhibit MgH₂ agglomeration due to their hydrogen storage absorption/desorption cycling stability [8,48,49,50]. Herein, carbon fiber materials were selected to be combined with Mg-based hydrogen storage, and two different preparation processes, wet impregnation and dry ball milling, were designed in order to investigate the effect of the preparation processes on the hydrogen storage properties of magnesium-carbon fiber composites. It was found that the two preparation processes had a significant effect on the enhanced hydrogen storage kinetic properties of MgH₂. The introduction of carbon fiber carriers caused the carbon-hydrogen bond to weaken the interaction between hydrogen atoms and magnesium atoms, thus promoting the migration and release of hydrogen atoms.

2. Results and Discussion

2.1. Structural Features and Microstructures Induced by Different Processes

2.1.1. Structural Features of MgH₂@CC and MgH₂-CC Composites

We chose carbon fiber materials combined with magnesium-based hydrogen storage and designed two different preparation processes, wet impregnation and mechanical milling, as shown in Figure 1, with details as discussed in experimental sections. First, in order to investigate the formation process of MgH₂ on the carbon cloth, XRD tests were performed on the samples. The crystal structure analysis of MgH₂@CC and MgH₂-CC composites by XRD is shown in Figure 2a. The MgH₂@CC composite shows the presence of two broadened diffusion diffraction peaks near 25.2° and 43.1° that correspond to the characteristic diffraction peaks for carbon. The impregnated and vacuum-treated samples showed characteristic peaks of MgBu₂ near 12.4°, 19.2°, and 21.1°, respectively, indicating that the heptane solution had been completely removed and MgBu₂ was successfully loaded on the surface of the carbon cloth. The disappearance of the MgBu₂ phase and the appearance of the MgH₂ phase on the surface of the carbon cloth after the hydrogenolysis reaction imply that the hydrogenolysis reaction is complete. At the same time, no other heterogeneous phases appeared, suggesting that the dibutyl magnesium was completely transformed into MgH₂.

In addition, the MgH₂-CC composites prepared by ball milling showed characteristic peaks belonging to MgH₂ at 27.8°, 35.6°, and 39.7°, but no more obvious carbon peaks were detected, which may be attributed to the fact that the carbon fibers were uniformly dispersed in the MgH₂ matrix after ball milling. In order to demonstrate the successful loading of carbon fibers with MgH₂ in the MgH₂-CC composites, Raman tests were carried out on the MgH₂-CC composites. As shown in Figure 2b, typical Raman peaks belonging to the D-band and G-band of the graphite phase, respectively, appeared at wave numbers of 1350 cm⁻¹ and 1590 cm⁻¹. D-band represents the sp3 defects in the carbon and the G-band represents the E2g vibration of the carbon due to the sp2 hybridization. The results also confirmed that the carbon fibers were successfully loaded with MgH₂ in the MgH₂-CC composites. The samples were then analyzed for particle size. The Scheller formula can be used to calculate the crystal size from the XRD data:

$$\mathrm{D}={\displaystyle \frac{\mathrm{K}\lambda }{\beta \cos \theta }}$$

(1)

where D is the grain size perpendicular to the observed grain surface; K is a constant, usually taking the value of 0.9; λ is the wavelength of X-rays, β is the radian of the half-height width of the sample’s diffraction peak, and θ is the diffraction angle. According to the calculation results, the average grain size of MgH₂@CC is 35.75 nm, the average grain size of MgH₂-CC is 20.01 nm, and the average grain size of commercial MgH₂ is 181.56 nm, which are both improved.

2.1.2. Distinct Morphologies of MgH₂@CC and MgH₂-CC Composites

The microstructure of MgH₂@CC composites obtained by hydrogenolysis of organic magnesium was observed by scanning electron microscopy and is shown in Figure 3a–c. It can be clearly observed that the MgH₂ particles are uniformly distributed on the surface of the carbon fibers, and the high-density defects on the surface of the carbon cloth provide the growth sites for MgH₂, which grows along the direction of the rod-shaped carbon fibers. It was also observed that some of the MgH₂ on the surface of the carbon fibers was deposited due to physical changes during the impregnation process, but the MgH₂ particles on the surface still showed a diffuse distribution. The microstructure of MgH₂-CC composites obtained by ball milling is shown in Figure 3d–f. It can be clearly seen in the image that the MgH₂ particles and carbon fibers are diffusely distributed on the surface of the conductive adhesive, and the particles do not show an obvious aggregation phenomenon. Compared with MgH₂@CC composites, the particles are more uniformly dispersed, and it is easier to form a uniform carbon fiber loading system by the ball milling method, while the dispersed carbon after ball milling can inhibit the agglomeration of MgH₂ particles.

2.2. Improved Hydrogen Storage Performance

2.2.1. Kinetic Properties of MgH₂@CC and MgH₂-CC Composites

In order to investigate the kinetic properties of MgH₂@CC composites obtained after hydrolysis of organic magnesium, the composites were heated to 500 °C, with a heating rate of 3 °C/min, 5 °C/min, 8 °C/min, and 10 °C/min, and their DSC curves were tested. As a comparison, the DSC curves of commercially pure MgH₂ and BM-MgH₂ were tested under the same conditions. For the dehydrogenation, the exothermic decomposition of MgH₂→Mg is accompanied by the exothermic process during the temperature rise stage, and the corresponding exothermic peaks will appear in the DSC curve. As shown in Figure 4, the peak exothermic temperatures of MgH₂-CC, MgH₂@CC, BM-MgH₂, and pure MgH₂ are 332 °C, 358 °C, 371 °C, and 449 °C, respectively. The peak hydrogen release temperatures of MgH₂-CC and MgH₂@CC composites were reduced by 117 °C and 91 °C, respectively, compared with that of pure MgH₂, indicating that the introduction of carbon fiber carriers can improve the kinetic properties of MgH₂ and reduce the peak hydrogen release temperature significantly.

To further investigate the hydrogen release properties of MgH₂@CC and MgH₂-CC composites, Figure 5a–h shows the DSC curves of Pure MgH₂, BM-MgH₂, MgH₂@CC, and MgH₂-CC. It can be observed that both MgH₂@CC and MgH₂-CC show a significant decrease in the peak heat uptake temperature at different heating rates. The exothermic activation energy is also an important measure of the kinetic performance, and in general, the exothermic activation energy can be calculated based on the peak temperature of heat absorption and the rate of warming in combination with the Kissinger equation:

$$\mathrm{l}\mathrm{n}\left({\displaystyle \frac{\mathsf{\beta }}{T_P^2}}\right)=\mathrm{A}-{\displaystyle \frac{E_{\mathrm{a}}}{\mathrm{R}{T}_P}}$$

(2)

where β is the rate of temperature increase, T_p is the peak temperature of heat absorption, and A is a linear constant. The peak heat absorption temperatures of MgH₂@CC were 348.4 °C, 358.6 °C, 364.5 °C, and 372.1 °C at 3 °C/min, 5 °C/min, 8 °C/min, and 10 °C/min, respectively, and the scatter plots of ln(β/T_p²) versus 1000/T_p were made subsequently, as shown in Figure 5b. Based on the slope of the straight line in the fitting result, the hydrogen release activation energy of MgH₂@CC can be obtained as ~175.1 kJ·mol⁻¹, which is 38.8 kJ·mol⁻¹ lower than that of commercially pure MgH₂ (213.9 kJ·mol⁻¹), suggesting that the MgH₂@CC composites obtained by hydrolysis of organomagnesium improve their hydrogen release properties under the synergistic effect of nanoeffects and the carbon fiber carrier. Meanwhile, the hydrogen release properties of MgH₂-CC composites and BM-MgH₂ were investigated, and Figure 5e–h shows the DSC curves of MgH₂-CC composites and BM-MgH₂ and the straight line of ln(β/T_p²) fitted to 1000/T_P. It can be observed that the exothermic peak temperatures of MgH₂-CC composites are 321.3 °C, 332.8 °C, 341.5 °C, and 346.6 °C, respectively, and the activation energy of hydrogen release is ~137.3 kJ·mol⁻¹, which is also significantly improved compared with that of the ball-milled MgH₂ samples (~176.4 kJ·mol⁻¹).

2.2.2. Thermodynamic Properties of MgH₂@CC and MgH₂-CC Composites

Thermodynamic tests were based on the equilibrium pressure of the desorption platform in PCI at different temperatures. The enthalpy (ΔH) and entropy (ΔS) of deuterium desorption were determined according to the van’t Hoff equation:

$$\mathrm{l}\mathrm{n}\left({\displaystyle \frac{P_{\mathrm{e}}}{P_0}}\right)=\left({\displaystyle \frac{∆H}{\mathrm{R}T}}\right)-\left({\displaystyle \frac{∆S}{\mathrm{R}}}\right)$$

(3)

where P_e is the equilibrium desorption pressure, P₀ is the standard atmospheric pressure, R is the gas constant, and T is the temperature in Kelvin.

To investigate the thermodynamic properties of MgH₂@CC and MgH₂-CC composites, as shown in Figure 6a–d, P-C-T tests were performed at 275 °C, 300 °C, 325 °C, and 350 °C, and the hydrogen absorption and release plateau pressures were obtained for the corresponding temperatures. The plateau pressures (P_eq) at different temperatures are shown in

Table 1. Hydrogen absorption and desorption plateau of MgH₂@CC and MgH₂-CC composite under different temperature.

Temperature (°C)	Absorption Plateau (MPa)		Desorption Plateau (MPa)
	MgH2@CC	MgH2-CC	MgH2@CC	MgH2-CC
350	0.853	0.823	0.526
325	0.472	0.458	0.310	0.303
300	0.262	0.237	0.130	0.138
275	0.127	0.121	0.081	0.079

, and each plateau pressure is positively correlated with the temperature; ∆H and ∆S were calculated according to the Van’t Hoff Equation (3). Based on the slope and intercept of the straight line in the fitting results, the enthalpy changes of hydrogen release (∆H_des) and entropy change of hydrogen release (∆S_des) of MgH₂@CC composites are 73.5 kJ·mol⁻¹ and 140.5 J·mol⁻¹·K⁻¹, respectively, and the enthalpy change of hydrogen absorption (∆H_abs) and entropy change of hydrogen absorption (∆S_abs) were −71.7 kJ·mol⁻¹ and −132.2 J·mol⁻¹·K⁻¹, respectively. The enthalpy changes of hydrogen release ∆H_des and entropy change of hydrogen release ∆S_des of MgH₂-CC composites were 73.0 kJ·mol⁻¹ and 140.5 J·mol⁻¹·K⁻¹, respectively, and the enthalpy change of hydrogen absorption ∆H_abs and entropy change of hydrogen absorption ∆S_abs were −72.8 kJ·mol⁻¹ and −136.2 J·mol⁻¹·K⁻¹, respectively. Compared with the commercially pure MgH₂ (∆H_des = 76 kJ/mol⁻¹ and ∆S_des = 134.1 J·mol⁻¹·K⁻¹), the thermodynamic properties have not been improved, indicating that reducing the size of the MgH₂ particle size and preventing MgH₂ particle aggregation is more likely to be reflected in the improvement of kinetic properties.

2.3. Mechanism Understanding

In order to investigate the reason for the enhanced hydrogen release properties of MgH₂@CC composites, Fourier Transform Infrared Spectroscopy (FTIR) was conducted to analyze the material; the results are shown in Figure 7a. Among them, the peaks in the 1150–1500 cm⁻¹ region correspond to the Mg-H stretching bands, the peaks in the 759–1000 cm⁻¹ region correspond to the Mg-H bending bands, and the peaks in the 1450–1550 cm⁻¹ region correspond to the O-H bonding, which means that the material undergoes partial oxidation. The peaks in the 2850–2960 cm⁻¹ region correspond to the C-H bonding, which means that the carbon fiber matrix interacts with MgH₂ at the hydrogen release stage; this implies that the carbon fiber matrix interacts with MgH₂ during the hydrogen release phase [51,52,53]. When the hydrogen absorption reaction occurs, the carbon in magnesium-carbon composites adsorbs hydrogen from H₂ atmosphere or MgH₂ to form sp² C-H and sp³ C-H bands [54]. When the hydrogen release reaction occurs, the carbon becomes a carrier for H atom transfer due to the C-H bond [55,56], the high temperature energy will drive the C-H bond to break, and the broken carbon bond will act as a carrier for H transfer and reduce the activation energy of the Mg-H bond, which will promote MgH₂ to overcome the hydrogen release energy barrier, similar to the photocatalytic Ni/TiO₂ composites, where the electrons are excited by the visible light and transferred from Ni to Ti, which drives the photocatalytic hydrogen production [57]. The activation energy is reduced, which improves the hydrogen storage performance of MgH₂@CC composites. In contrast, in MgH₂-CC composites, Mg has more binding sites with C. The C-H bonds and high-energy ball mills synergize with each other to provide more reaction sites and energy for hydrogen uptake/desorption, and thus produce better catalytic effects. The schematic diagram of the desorption enhancement mechanism of magnesium-carbon composites is shown in Figure 7b.

3. Experimental

3.1. Material Preparation

Carbon fiber cloth (CDD, 200 × 200 mm, Yongfu Electrochemical, Nanajing, China) was purchased and further processed. Dibutyl magnesium (1.0 mol/L in heptane, Aladdin, solution), MgH₂ (>95% from Mg Power Technology Co., Ltd., Shanghai, China), and anhydrous ethanol (C₂H₅OH, AR, Anhui Tiandi High Purity, Anqing, Anhui) were purchased and directly used as received without further purification. First, as shown in Figure 1a, the carbon fiber cloth was pretreated. The carbon fiber cloth was rinsed with ethanol, dried in an oven, and then placed in a stainless-steel autoclave and heated at 300 °C under 2 MPa hydrogen atmosphere for 3 h to remove hydrophilic groups and other impurities. In a sealed glove box filled with argon gas, weigh 0.43 g of pretreated carbon fiber cloth into a beaker and draw up a total of 3 mL of dibutyl magnesium solution using a syringe with a volume capacity of 5 mL. The absorbed solution was added to the surface of the carbon fiber cloth with 0.5 mL drops at a time, and after the carbon fiber cloth was completely impregnated, another drop was added until 3 mL was added. Subsequently, the beaker was placed in a vacuum chamber and vacuum dried for 12 h to remove the heptane solvent to obtain MgBu₂@CC. Finally, the sample was placed in a stainless-steel sample chamber and kept at 5 MPa hydrogen pressure and 200 °C for 2 h. The final sample obtained was named as MgH₂@CC composite. Subsequently, the sample was vacuum-dried in a vacuum oven at 80 °C for 6 h to obtain the carbon fiber-loaded precursor composite. The equation of the reaction was as follows:

$${\left({\mathrm{C}}_4{\mathrm{H}}_9\right)}_2\mathrm{Mg}+{2\mathrm{H}}_2\to \mathrm{Mg}{\mathrm{H}}_2+2{\mathrm{C}}_4{\mathrm{H}}_9$$

(4)

The loading of MgH₂ in MgH₂@CC can be calculated to be ~24.2 wt% from the before and after mass difference.

MgH₂-CC composites were prepared as shown in Figure 1b. In an argon glove box, pure MgH₂ and carbon fiber cloth were weighed according to the mass percentage of 75:25 and put into a stainless-steel ball milling jar with a ball to material ratio of 40:1. The stainless-steel balls were DECO-304-B, with diameters of 5 mm, 6 mm, 8 mm, and 10 mm, respectively. The ball milling time was 10 h. The ball milling speed was 400 rpm, and the interval between each 30-min revolution of the ball mill was 5 min. Ball milled MgH₂ (BM-MgH₂) samples were prepared in the same way described above.

3.2. Material Characterization

The phase composition of the composites was analysed by using an X-ray diffractometer (XRD, Mini Flex 600, Rigaku, Japan) with Cu Kα radiation (λ = 0.154056 nm) at 40 kV and 15 mA. The 2θ angle ranged from 10° to 90°with increments of 0.02°; samples were placed in a hermetically sealed XRD sample stage and sealed with polymer tape. The morphologies and microstructures of the samples were examined by using scanning electron microscopy (SEM, MIRA3 XMU, Tescan). We analysed the elemental distribution in the local region of the samples by EDS (Xplore30, OXFORD) attached to SEM, with magnification rates ranging from 200 to 5000. For SEM samples, ethanol or organic magnesium was used as the dispersant, and a small amount of the solution was taken and dropped on the conductive adhesive after sonication. SEM tests were performed at magnifications of 5 K, 10 K, and 20 K, with a 5 kv energy. Thermogravimetric analysis (TG, DTG-60H, SHIMADZU) was performed in the range of 30~700 °C with a heating rate of 10 °C/min to verify the loaded composition, Weigh about 5 mg of the sample in the glove box, place it in an aluminium crucible with a lid to seal the aluminium crucible and then take it out and put it into the DTG-60H instrument for testing. Maintain the argon gas environment at 0.2 Mpa during the testing process with a flow rate of 300 mL/min, and determine the peak value by selecting the beginning and end intervals of the melting peak in the DSC test. Raman spectra (HORIBA LabRAM HR Evolution spectrometer) were measured and processed with a laser wavelength of 532 nm. Fourier Transform Infrared Spectroscopy (FTIR) was conducted to analyse the material with a detection wavelength range of 500~3000 cm⁻¹. The samples were mixed with KBr and then pelletized in the argon-filled glove box.

The thermodynamic properties were evaluated by measuring the pressure−composition isotherms (PCI) on an automatic Sievert-type apparatus. Prior to the PCI measurements, we weighed 0.1–0.5 g of sample and heated it at 300 °C under vacuum conditions and activated it through two hydrogen sorption cycles consisting of a 0.5 h absorption under 3.0 MPa of H₂ gas and a 0.5 h desorption under vacuum. All the tests were performed under initial pressures of 0.0001 MPa for desorption and 5 MPa hydrogen pressure for absorption.

4. Conclusions

In summary, two carbon fiber-loaded MgH₂ hydrogen storage composites were successfully prepared by hydrogenation of organic magnesium and ball milling. Both the obtained MgH₂@CC and MgH₂-CC composites show significantly improved kinetics, with the MgH₂-CC composite being better in terms of preparation cost and kinetics. Our experimental results reveal that C-H bonding interactions in the two composite hydrogen storage materials are responsible for promoted hydrogen absorption/desorption from hydrides. These results provide further insights into promoting the hydrogen ab-/desorption of metal hydrides. Unfortunately, the thermodynamics were not improved, and we will work on the improvement of thermodynamic properties in subsequent work.