Hydrogen storage and hydrogen generation properties of CaMg2-based alloys

Accepted Manuscript Hydrogen storage and hydrogen generation properties of CaMg2-based alloys Miaolian Ma, Ruoming Duan, Liuzhang Ouyang, Xiaoke Zhu, Zhiling Chen, Chenghong Peng, Min Zhu PII:

S0925-8388(16)32709-8

DOI:

10.1016/j.jallcom.2016.08.307

Reference:

JALCOM 38809

To appear in:

Journal of Alloys and Compounds

Received Date: 5 July 2016 Revised Date:

28 August 2016

Accepted Date: 29 August 2016

Please cite this article as: M. Ma, R. Duan, L. Ouyang, X. Zhu, Z. Chen, C. Peng, M. Zhu, Hydrogen storage and hydrogen generation properties of CaMg2-based alloys, Journal of Alloys and Compounds (2016), doi: 10.1016/j.jallcom.2016.08.307. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Hydrogen storage and hydrogen generation properties of CaMg2-based alloys Miaolian Maa, b, Ruoming Duana, b, Liuzhang Ouyanga, b, c*, Xiaoke Zhua, Zhiling Chena, Chenghong Penga, Min Zhua, b School of Materials Science and Engineering and Key Laboratory of Advanced Energy Storage Materials of

RI PT

a

Guangdong Province, South China University of Technology, Guangzhou, 510641, PR China b

China-Australia Joint Laboratory for Energy & Environmental Materials, South China University of Technology, Guangzhou, 510641, PR China c

Key Laboratory of Fuel Cell Technology of Guangdong Province, South China University of Technology,

SC

Guangzhou 510641, PR China

Abstract:

M AN U

The hydrogen storage and hydrogen generation of CaMg2 and CaMg1.9Ni0.1 alloys were investigated by X-ray diffraction (XRD) and pressure-composition-isotherm (PCI) measurements. The results confirmed that the CaMg2 alloy cannot absorb hydrogen at room temperature. While the addition of Ni to the CaMg2-based alloys resulted in room-temperature hydrogen absorption

TE D

without an activation process, and a maximum hydrogen-absorption capacity of 5.65 wt.%. The hydrolysis performance of hydrogenated CaMg2 and hydrogenated CaMg1.9Ni0.1 (abbreviated as H-CaMg2 and H-CaMg1.9Ni0.1 hereafter) was also evaluated. The hydrolysis reaction of H-CaMg2

EP

occurred rapidly in water and resulted in a hydrogen yield of 800 mL/g. Furthermore, the hydrolysis properties of H-CaMg1.9Ni0.1 were significantly enhanced with the addition of Ni, as

AC C

evidenced by a hydrogen yield of 1053 mL/g in 12 min, this yield is 94.7% of the theoretical hydrogen yield of H-CaMg1.9Ni0.1.

Keywords: Hydrogen storage; CaMg2-based alloys; Hydrogen absorption; Kinetics; Hydrolysis.

* Corresponding author: Liuzhang Ouyang, E-mail: [email protected], Tel.: 86-20-87114253, Fax: 86-20-87112762.

ACCEPTED MANUSCRIPT

1. Introduction Hydrogen is a promising renewable energy for replacing increasingly depleted fossil fuels

RI PT

due to its high energy density and clean emissions [1, 2]. As a hydrogen storage materials, magnesium-based materials are well suited for use as energy carrier media owing to their high

theoretical hydrogen-storage capacities (7.6 wt.% for MgH2), natural abundance, and low cost [3].

SC

However, the practical applications of these materials are blocked by the kinetic and

thermodynamic properties. In the last few decades, several methods, such as ball-milling [4],

M AN U

addition of catalysts [5], surface modification [6] or element substitution [7], and a combination thereof, have been used to develop Mg-based alloys. Development of a new type of Mg-based alloys with relatively high hydrogen capacity at moderate temperatures and hydrogen pressure is therefore essential.

TE D

Recent studies have reported that magnesium alloys can react with hydrogen at room temperature. For example, Mg-rare-earth-based alloys (such as Mg3Pr, Mg3La, and Mg3Mm with reversible hydrogen capacities of 2.2 wt.%, 2.89 wt.%, and 2.91 wt.%) absorb hydrogen at room

EP

temperature [8-10]; Mg-based pseudo-binary alloys (Mg0.67Ca0.33)Ni2 can reversibly absorb and desorb 1.4 mass% hydrogen at room-temperature [11]. However, rare-earth metals are extremely

AC C

expensive and their hydrogen capacities are inadequate for commercial application. Therefore, the development of alloys with satisfactory hydrogen capacity under moderate reaction conditions is crucial.

Chiotti et al. [12] found that CaMg2 reacts with hydrogen and then transforms to CaH2 and Mg at 450 °C. The CaMg2 alloy is expected to absorb more than 6.3 wt.% of hydrogen, at an H/M ratio of two. It is well known that the transition metal Ni has catalytic effect of hydrogen on the Ni

ACCEPTED MANUSCRIPT

surface during the hydriding process of alloys. Researchers have reported some excellent hydrogen storage properties of CaMg2-based alloys added with Ni. For example, Lupu et al. [13]

RI PT

reported the CaMg1.8Ni0.5 alloy can absorb 5.7 wt.% of hydrogen at 338 °C with fast hydriding process. Terashita et al. [14] found that the (Ca0.8La0.2)Mg2.2Ni0.1 alloy can absorb 5.1 wt.%

hydrogen at room temperature. To obtain a higher hydrogen capacity of Ca-Mg-Ni alloys and

SC

reduce costs, we designed CaMg2-based alloys by adding a small amount of Ni.

In this work, the CaMg2-based alloys were prepared by induction melting and the

M AN U

hydrogen-storage behavior of the resulting CaMg1.9Ni0.1 alloy was described by evaluating the phase-structure transformation during hydrogenation and dehydrogenation, de/hydriding kinetics, apparent activation energy, and thermodynamic parameters (enthalpy and entropy) of the de/hydriding reaction. The hydrolysis properties of CaMg2, CaMg1.9Ni0.1, and their hydrides,

TE D

which constitute promising candidates for either hydrogen-storage or hydrolysis-induced hydrogen-generating materials, are also considered in this study.

EP

2. Experiment

2.1. Sample preparation

AC C

Raw materials are high purity Mg (≥99.9%), Ni (≥99.9%), and Ca (≥95%). The

CaMg2 alloy was prepared via the sealed tube method: placing an argon-filled pure iron pipe in a resistance furnace maintained at 900 °C for 20 h. The CaMg1.9Ni0.1 alloy was

prepared via induction melting in an alumina crucible under the protection of pure argon atmosphere. This melting process was repeated at least three times. Approximately 6–8 wt.% of extra Mg and Ca was calculated for the loss of evaporation. The resulting Mg and Ni

ACCEPTED MANUSCRIPT

ingots were pulverized by saws and the oxide layer was polished with fine-grit sandpaper. The as-melted ingots were broken into small particles and filtered through a 300-mesh

RI PT

sieve. This filtering and subsequent storage of the samples were performed in an argon-filled glove box equipped with a recirculation system, in order to prevent sample oxidation and/or hydroxide formation. The CaMg2 and CaMg 1.9Ni0.1 alloys were

SC

hydrogenated by using an Advanced Materials Corporation (AMC) gas reaction controller. 2.2. Hydrolysis experiment

M AN U

Hydrogen-generation measurements were performed using in-house-developed equipment [15]. During these measurements, 0.1–0.2 g of CaMg2, CaMg1.9Ni0.1, and/or their hydrides reacted with pure water in a 50-mL Pyrex glass reactor that has three openings: one for water addition, one for inserting the thermometer, and one for hydrogen

TE D

exhausting. Hydrolysis began with contact between the sample and the water. The generated hydrogen was passed through a Tygon tube and then through a Monteggia washing bottle filled with room-temperature water, in order to condense the water vapor.

EP

The amount of hydrogen generated was determined by extracting water into a beaker,

AC C

which was then placed on an electronic scale. The weight changes, over time, were recorded and the quantity of hydrogen was thereby determined. The hydrogen-generation rate and yield can be calculated from the reaction time and hydrogen volume that was measured and analyzed by the computer. For the sake of reproducibility, each experiment was repeated at least twice. 2.3. Sample characterization

ACCEPTED MANUSCRIPT

The chemical composition of the alloys was determined by energy dispersive x-ray spectroscopy (EDX) accessory attached to a Philips XL-30 FEG scanning electron

RI PT

microscope (SEM). Moreover, the phase structure of alloys was determined via X-ray diffraction (XRD; Philips X’Pert MPD) using Cu-Kα radiation (λ=0.154056 nm). Scans were performed over a 2θ range of 20–80°, at an accelerating voltage, current, and

SC

scanning rate of 40 kV, 15 mA, and 0.02°/s, respectively. The samples were wrapped with

liquid paraffin to prevent oxidation during XRD measurements. In addition, the hydrogen

M AN U

absorption–desorption properties of the CaMg2 and CaMg1.9Ni0.1 alloys were evaluated at different temperatures. The measurement temperature of each sample was controlled, to an accuracy of ±2 K, via induction heating in the reaction cell. The △H and △S were determined from the van’t Hoff plot. For these calculations, the plateau pressure was taken

PCI curves.

TE D

as the hydrogen pressure at the midpoint of the desorption pressure plateau of the measured

EP

3. Results and discussion

3.1. Phase transition of CaMg2 and CaMg1.9Ni0.1 during hydrogenation and dehydrogenation

AC C

The Ca-Mg-Ni alloy, with an EDX-determined chemical composition of 62.3 at%

Mg, 33.9 at% Ca, and 3.8 at% Ni, is referred to as CaMg1.9Ni0.1 alloy. The XRD patterns of

the as-melted, hydrogenated, and dehydrogenated CaMg2 and CaMg1.9Ni0.1 alloys are shown in Fig. 1 and Fig. 2, respectively. The pattern shown in Fig. 1(a) indicates that induction melting of pure Ca and Mg metals yields a CaMg2 compound consisting of C14 Laves phases and described by the space group, P63/mmc. Based on the ICDD file, the

ACCEPTED MANUSCRIPT

peaks indexed to the CaMg2 compound are attributed to a close-packed hexagonal structure (hcp) phase with lattice constants of a=b=0.6225 nm and c=1.028 nm. Fig. 1(b) shows the

RI PT

XRD pattern of the hydrogenated CaMg2 alloy (abbreviated as H-CaMg2) at 350℃. The peaks in this pattern indicate that the CaMg2 alloy reacts with hydrogen during the hydriding process, thereby forming MgH2 and Ca4Mg3H14 phases, and a small amount of

SC

Mg. The Mg peaks in the pattern of the dehydrogenated alloy (Fig. 1(c)), are sharper than

their counterparts in the pattern of H-CaMg2 (Fig. 1(b)). In addition, peaks corresponding

M AN U

to MgH2 and Ca4Mg3H14 are absent, but new peaks corresponding to CaH2, are observed. The CaH2 phase in H-CaMg2 arises from the dehydrogenation reaction of Ca4Mg 3H14, which proceeds as follows:

Ca Mg  H  → CaH + Mg + H ↑

(1)

TE D

Owing to the splitting catalytic effect of hydrogen on the Ni surface, the addition of Ni may lead to accelerated hydrogenation of the CaMg2 alloy. CaMg1.9Ni0.1 alloy can absorb hydrogen at room temperature and a hydrogen pressure of 3.8 MPa. Therefore, the

EP

addition of Ni may accelerate dehydrogenation of the CaMg2-base alloy, even under mild

AC C

conditions. Fig. 2(a) shows the XRD pattern obtained from the CaMg 1.9Ni0.1 alloy prepared via induction melting. The diffraction peaks can all be indexed as CaMg2 and MgNi2. The

strongest diffraction peaks of CaMg2 are left-shifted by 0.26°, indicative of increases in the lattice constant and volume. In other words, the Ni catalyst and lattice expansion facilitate hydrogenation of the CaMg1.9Ni0.1 alloy, even at room temperature. Peaks corresponding to the CaMg2 and MgNi2 phases occur in the XRD patterns (Fig. 2(b)) of the hydrogenated CaMg1.9Ni0.1 alloy. In addition, several sharp peaks, which appear inconsistent with the

ACCEPTED MANUSCRIPT

ICDD file, are attributed to Ca5Mg9H28, based on the XRD, PCI and EDX measurements. The Ca5Mg 9H28 phase, whose hydrogen absorption/desorption properties differ from those

RI PT

of the H-CaMg2 system, is a new Ca-Mg hydride that merits further study. Furthermore, except for the occurrence of peaks corresponding to the MgNi2 phase, the pattern of the dehydrogenated CaMg1.9Ni0.1 alloy (Fig. 2(c)) is identical to that of the dehydrogenated

SC

H-CaMg2 alloy (Fig. 1(c)). The phase transformation of the CaMg1.9Ni0.1 alloy during

TE D

M AN U

hydrogenation and dehydrogenation will be discussed further in the following section.

AC C

EP

Fig. 1 XRD patterns of CaMg 2 alloys at different stages: (a) as-melted, (b) hydrogenation and (c) dehydrogenation

Fig. 2 XRD patterns of CaMg1.9Ni0.1 alloys at different stages: (a) as-melted, (b) hydrogenation and (c) dehydrogenation 3.2. Dehydriding thermodynamic properties of CaMg2 and CaMg1.9Ni0.1 alloys

ACCEPTED MANUSCRIPT

The thermodynamic parameters associated with the dehydriding reaction of H-CaMg1.9Ni0.1 were determined from pressure-composition isotherm (PCI) curves. Fig.

RI PT

3(a) shows the PCI curves obtained at 573, 593, 613, and 633 K during hydrogen desorption of this alloy. The maximum hydrogen desorption capacity (2.57 wt.%) can be obtained at 633 K. Moreover, two plateaus occur in each PCI desorption curve, indicating

SC

that the dehydrogenation process of H-CaMg1.9Ni0.1 can be divided into two steps. The high flat plateau occurs at high pressure and is composed of a few data points, indicating

M AN U

that the desorption process occurs more easily in the initial stage than in subsequent stages. The dehydrogenation mechanism of H-CaMg 1.9Ni0.1 was further investigated via quasi-in situ XRD analysis.

The dehydrogenation reaction was stopped when the first point (denoted as I in Fig.

TE D

3(a)) appeared on plateau 1 of the PCI curve. As shown in Fig. 4(a), peaks corresponding to MgCaH3.72, MgH2, and Mg phases appear in the pattern during the first step of dehydrogenation. Therefore, the initial plateau region corresponds to the decomposition of

EP

Ca5Mg9H28, which yields H2, Mg, MgH2, and MgCaH3.72. Plateau 2 forms during the dehydrogenation reaction; the XRD pattern of the resulting product (denoted as point II in

AC C

Fig. 3(a)) is shown in Fig. 4(b). The relative strength of the peaks corresponding to the Mg and CaH2 phases increased, whereas the MgH2 phase nearly disappeared. Therefore, the second plateau region corresponds to the dehydrogenation of MgH2 and MgCaH3.72, which

subsequently undergo complete transformation to Mg and CaH2 phases. The dehydriding pressure of CaMg1.9Ni0.1 is higher than that of pure Mg at the same temperature, and is indicative of unstable dehydriding thermodynamics.

SC

RI PT

ACCEPTED MANUSCRIPT

TE D

M AN U

Fig. 3 (a) PCI curves for the hydrogen desorption and (b) Van’t Hoff plots for the two plateaus of H-CaMg1.9Ni0.1 composite.

Fig. 4 XRD patterns of the dehydrogenation products: (a) plateau 1 and (b) plateau 2.

EP

The van’t Hoff plots (Fig. 3(b)) of the H-CaMg1.9Ni0.1 sample were obtained from the aforementioned PCI curves. Furthermore, the enthalpy and entropy of the dehydriding

AC C

process were determined from the Van’t Hoff equation (Eq. (2)), which is given as follows:  = −

∆ 

+

∆ 

(2)

where Kθ is the standard equilibrium constant, Kθ= PH2 during dehydriding, and K = 1/PH2

during hydriding. The midpoint of the plateau was used as the equilibrium pressure for the calculation. Enthalpy and entropy values of 94.80 kJ/mol H2 and 160.2 J/(K· mol) H2, respectively, were obtained for the dehydrogenation plateau of Ca5Mg9H28. Corresponding

ACCEPTED MANUSCRIPT

values of 97.34 kJ/mol H2 and 168.2 J/(K· mol) H2 were obtained for the dehydrogenation reaction of MgCaH3.72 and MgH2.

RI PT

3.3. Hydriding properties of CaMg2 and CaMg1.9Ni0.1 alloys Fig. 5(a) shows the hydriding-kinetic curves for the CaMg2 alloy at 573 K and

CaMg1.9Ni0.1 alloy at different temperatures. A capacity of 5.10 wt.% was realized for the

SC

hydrogenation of CaMg1.9Ni0.1 alloy at room temperature. The maximum hydrogen

capacity of 5.65 wt.% in this alloy were attained at 353 K. However, the hydriding kinetics

M AN U

of the CaMg 2 alloy were sluggish, as shown in Fig. 5, even after 16 h, the hydrogenation process was incomplete.

The hydriding-kinetic curves were analyzed by using the Johnson–Mehl–Avrami– Kolmogorov (JMAK) model [16, 17]:

(3)

TE D

α =  −  [−"η ]

where α(t), k, and η are the reaction fraction at time t, rate constant, and Avrami exponent, respectively. The JMAK equation is widely used to describe the time-dependent kinetic

EP

behavior of isothermal solid-state reactions [18, 19]. In most cases, Eq. (2) is rewritten as: (4)

AC C

[−  − α] = η  + η $

A hydrogen absorption fraction ranging from 0.2–0.5 was used to fit the kinetic

curves. By plotting ln[-ln(1-α(t))] versus lnt, the Avrami exponent η and the rate constant $ can be obtained from the slope and intercept, respectively, of the fitted line as shown in

Fig. 5(b). The η values obtained (i.e., 0.92–1.25) were close to unity, indicating that the hydriding reaction of the sample follows a diffusion-controlled mechanism [20]. In

ACCEPTED MANUSCRIPT

addition, the apparent activation energy (Ea) of hydrogenation was determined by substituting the values of $ into the Arrhenius equation (Eq. (5)): '(

$ = %  & *+) ,

RI PT

(5)

where A, R, and T are the pre-exponential factor, gas constant (R=8.314 J/(mol· K)), and

absolute temperature, respectively. The hydrogen absorption of CaMg1.9Ni0.1 exhibits

SC

significant linearity when ln k is plotted as a function of 1/T. Moreover, the hydrogenation

AC C

EP

TE D

and Mg 2Ni (∼50 kJ/mol) [23-25].

M AN U

Ea (41.74 kJ/mol) is significantly lower than those of pure MgH2 (∼100 kJ/mol) [21, 22]

Fig. 5 (a) Isothermal hydriding kinetic curves obtained at different temperatures, (b) JMAK plots, and (c) Arrhenius plot for determining the activation energy of the CaMg 1.9Ni0.1 alloy. 3.4. Hydrolysis properties of CaMg2, CaMg1.9Ni0.1 alloys, and their hydrides

ACCEPTED MANUSCRIPT

Owing to their room-temperature hydrogen-absorption capacity, CaMg1.9Ni0.1 alloys can store hydrogen released from the factory. CaMg2 and CaMg1.9Ni0.1 alloys have

RI PT

theoretical hydrogen-storage capacities of 6.4 wt.% and 5.9 wt.%, respectively. However, hydrolysis-induced hydrogen generation is twice as high as the storage capacity, and values of up to 12.8 wt.% (1434 ml/g H2) and 11.8 wt.% (1327 mL/g H2) have been realized in

SC

this work. We investigated the hydrolysis properties of CaMg2-based alloys and hydrides in order to determine the effect of hydrogen addition on the CaMg2 system. The hydrolysis

M AN U

behavior of the as-milled CaMg2 and CaMg1.9Ni0.1 alloys, at 298 K, is described by the hydrogen evolution curves shown in Fig. 6(a). A comparison of the hydrolysis kinetics of these alloys reveals that, in the first 30 min, the hydrogen-generation rate of CaMg2 was higher than that of CaMg1.9Ni0.1 and slower thereafter. The hydrolysis of CaMg2 yielded (at

TE D

a constant rate) 131 mL/g, 179.3 mL/g, and 351.7 mL/g of hydrogen in 20, 30, and 72 min, respectively. During the hydrolysis of CaMg1.9Ni0.1, 100 mL/g of hydrogen was generated in 20 min. However, the hydrolysis rate increased thereafter and hence 171.4 mL/g, 442.9

EP

mL/g, and 571 mL/g of hydrogen were generated in 30, 72, and 120 min, respectively. The increased reactivity of CaMg1.9Ni0.1, after the first 30 min, stems from the addition of Ni,

AC C

which has a low hydrogen overpotential and (as a cathode material) induces severe galvanic corrosion of alloys [26]. Fig. 5(b) shows the hydrogen generation curves that describe the hydrolysis behavior

of the H-CaMg2 and H-CaMg1.9Ni0.1 hydrides. The hydrolysis rate of H-CaMg1.9Ni0.1 is higher than that of H-CaMg2. In fact, 872 mL/g, 968 mL/g, and 1053 mL/g (9.4 wt.%) of hydrogen were generated in 3 min, 5 min, and 12 min, respectively, in a sustained and

ACCEPTED MANUSCRIPT

rapid hydrogen-release process. The hydrolysis of H-CaMg2 was initially very rapid, lasting for only 50 s and producing 800 mL/g (7.14 wt.%) of hydrogen. The max

RI PT

hydrolysis rate achieved (2640 mL/(min·g)) is, to the best of our knowledge, the fastest hydrolysis rate ever reported for the Ca-Mg-H system. The addition of Ni significantly

enhanced the hydrolysis properties of the CaMg2 alloy. Therefore, in 12 min,

SC

H-CaMg1.9Ni0.1 produces 1053 mL/g of hydrogen, which is 94.7% of its theoretical

TE D

M AN U

hydrogen yield.

Fig. 6 Hydrogen evolution curves associated with hydrolysis of (a) as-prepared CaMg2 and CaMg1.9Ni0.1 alloys, (b) H-CaMg2 and H-CaMg1.9Ni0.1 hydrides at 298 K. A comparison of the hydrolysis properties and phase content of H-CaMg2 and

EP

H-CaMg1.9Ni0.1 reveals that the hydrolysis properties of the Ca5Mg9H28 phase are far superior to those of the Ca4Mg3H14 phase. Complete hydrogenation of CaMg2 yields ∼65%

AC C

of Ca4 Mg3H14. Furthermore, the theoretical hydrogen generation yield of hydrogenated CaMg2 (come from 65% Ca4Mg3H14 is ∼825 mL/g hydrogen) is similar to that of H-CaMg2

(800 mL/g). The hydrolysis curves revealed that the hydrolysis of H-CaMg2 is incomplete. However, the addition of Ni resulted in significantly enhanced hydrolysis properties, as evidenced by the hydrogen yield (1053 mL/g in 12 min) of H-CaMg1.9Ni0.1. The rapid and constant hydrolysis rate of this alloy can exceed 365 mL· min-1· g -1 in the first 2 min.

ACCEPTED MANUSCRIPT

Table 1 shows list of the technologies for the preparation of Mg-Ca based materials and the hydrogen yields in previous works. Compared to the methods used in those studies,

RI PT

the method employed in this work involves milder reaction conditions, a simpler reaction process, and higher hydrogen yields.

Table 1 Hydrolysis-generated hydrogen yields of CaMg2-based materials Reactants

Conditions

Max hydrogen yield (298 K)

atm Ar

magnetic stirrer

SC

Spex 8000 shaker mill/10 h/1

MgH2 and 20.3 mol% Ca [27]

Spex 8000 shaker mill/10 h/1

MgH2 and 20.3 mol% CaH2 [27]

76% of hydrolyzed fraction in 30 min

M AN U

atm Ar

76% of hydrolyzed fraction in 90 min

magnetic stirrer

50% of hydrolyzed fraction in

MgH2-25 wt% Ca [28]

Spex 8000 shaker mill/15 h

30 min, 57% in 45min and 62% in 80min

65% of hydrolyzed fraction in

MgH2-25 wt% CaH2 [28]

TE D

Spex 8000 shaker mill/15 h

201 mL/g, 400 mL/g and 732.8 mL/g in 10 min, respectively. 320.6 mL/g, 549.6 mL/g and 755.7 mL/g in 60 min, respectively.

Resistance furnace melting/broken/300-mesh sieved

179.3 mL/g in 30 min, and 351.7 mL/g H2 in 72 min

Induction melting/broken/300 mesh sieved

442.9 mL/g in 72 min, and 571 mL/g (80.9%)H2in 120 min

EP

AC C

CaMg1.9Ni0.1 alloy (in this work)

CaMg2 hydride (in this work)

in 80 min

Resistance furnace melting/Spex 8000 shaker mill/2 h/hydrogenated at 673 K and 4-5 MPa H2.

x wt.% Mg-Ca (x=10, 20 and 30 ) alloy hydride [29]

CaMg2 alloy (in this work)

20 min, 70% in 30 min and 76%

Resistance furnace melting/broken/300 mesh

800 mL/g in 1 min

sieved/hydrogenated at 573 K and 3.8 MPa H2.

CaMg1.9Ni0.1 hydride (in this work)

Induction melting/broken/300 mesh sieved/hydrogenated at 298 K and 3.8 MPa H2.

968 mL/g in 5 min, and1053 mL/g (94.7%) hydrogen in 12 min

3.5. The hydrolysis mechanisms of H-CaMg2 and H-CaMg1.9Ni0.1

ACCEPTED MANUSCRIPT

The hydrogen

evolution mechanism for the hydrolysis of H-CaMg2 and

H-CaMg1.9Ni0.1 alloys at 298 K was also analyzed by using the Avrami–Erofeev equation

RI PT

[30]. Fig. 7(a) and (b) show the hydrogen evolution curves by hydrolysis of H-CaMg2 and H-CaMg1.9Ni0.1 alloys at 298 K, respectively. The hydrogen production is expressed as conversion yield (%), which is defined as the volume of produced hydrogen over the

SC

theoretical volume. As shown in Fig. 7, the experimental curves describing the hydrolysis

of H-CaMg 2 and H-CaMg1.9Ni0.1 are well-fitted to the Avrami–Erofeev equation (Eq. (6)),

M AN U

which is deduced from the nucleation and growth process: -. = 1 − exp −3 ∗ . 5 

(6)

where 67 is the reaction rate, i.e., the ratio of reacted material to total material, B and m are constants, and t is the reaction time. The values of B and m, which were obtained via

TE D

fitting, estimated error S (i.e., the standard error), and r, the correlation coefficient, are shown in Fig. 7. The low values of S and r confirm the close correspondence between the fitted and experimental data. The fitting confirms that the hydrolysis reaction of H-CaMg2

EP

and H-CaMg1.9Ni0.1 is governed by the nucleation growth mechanism. However, owing to

AC C

the difference in their m values, the rate-controlling steps of the nucleation and growth reaction are different for different alloys. For example, m value of 0.62 is the one-dimensional diffusion process, while 1.07 indicates the three-dimensional interface reaction process [31]. The m value of H-CaMg2 (0.54) is close to 0.62, indicating that the

reaction is basically a one-dimensional diffusion process. While, an m value of 0.9 is close to 1.07 (see Fig. 6(b)), indicating that the hydrolysis reaction of H-CaMg1.9Ni0.1 is a three-dimensional interface reaction process. This may result from the catalytic effect of

ACCEPTED MANUSCRIPT

fine Ni particles, which uniformly distribute in the matrix of the hydrides, and promote the

SC

RI PT

hydrolysis reaction around the interface region.

M AN U

Fig. 7 Kinetic curves associated with hydrogen evolution of the (a) H-CaMg2 and (b) H-CaMg1.9Ni0.1 composites at 298 K.

4. Conclusions

Hydrogen storage and hydrogen generation properties of CaMg2-based alloys were

TE D

investigated. CaMg2 alloy did not absorb hydrogen at 298 K, but, at 573 K, absorbed 4.48 wt.% of hydrogen and transformed to Ca4Mg3H14 and MgH2 under 3.8 MPa of hydrogen. The addition of Ni enabled room-temperature absorption of hydrogen. In fact, the CaMg1.9Ni0.1 alloy had a

EP

room-temperature hydrogen-absorption capacity of 5.65 wt.% with an apparent activation energy of 41.74 kJ/mol. Furthermore, in the case of Ca5Mg9H28, values of 94.80 kJ/mol H2 and 160.2

AC C

J/(K·mol) H2 were obtained for the enthalpy and entropy, respectively, of dehydrogenation;

corresponding values of 97.34 kJ/mol H2 and 168.2 J/(K·mol) H2 were obtained for MgCaH3.72. Hydrolysis of the H-CaMg1.9Ni0.1 alloy occurs rapidly at room temperature, even without the addition of a catalyst, leading to a hydrolysis yield of 94.7%, in 12 min, and 1053 mL/g of hydrogen. The reaction mechanism of H-CaMg1.9Ni0.1 is governed by a three-dimensional

interface reaction process. CaMg1.9Ni0.1 alloy can be industrially synthesized via

ACCEPTED MANUSCRIPT

conventional induction melting of abundant and cheap Mg, Ca, and a small ratio of Ni. This alloy can absorb hydrogen at room-temperature and mild hydrogen pressure. These

RI PT

attributes suggest that the CaMg1.9Ni0.1 alloy, a new low-cost hydrogen storage and highly efficient hydrolysis material, has significant potential for use in portable fuel cells or other commercial hydrogen-production systems.

SC

Acknowledgements

M AN U

This work is financially supported by the National Natural Science Foundation of China (Nos. 51431001, 51271078 and U120124), by International Science & Technology Cooperation Program of China (2015DFA51750) and by the Project Supported by Guangdong Natural Science Foundation (2014A030311004, 2014GKXM011). Author Ouyang also thanks Guangdong Province Universities and Colleges Pearl River Scholar

References

TE D

Funded Scheme (2014).

EP

[1] T.Z. Si, Y. Cao, Q.G. Zhang, D.L. Sun, L.Z. Ouyang, M. Zhu, Enhanced hydrogen storage properties of a Mg-Ag alloy with solid dissolution of indium: a comparative study, J. Mater. Chem. A, 3 (2015) 8581-8589.

[2] Z. Zhao, Y. Zhu, L. Li, Efficient catalysis by MgCl2 in hydrogen generation via hydrolysis of

AC C

Mg-based hydride prepared by hydriding combustion synthesis, Chem. Commun., 48 (2012) 5509-5511.

[3] M. Zhu, Y. Lu, L. Ouyang, H. Wang, Thermodynamic Tuning of Mg-Based Hydrogen Storage Alloys: A Review, Materials, 6 (2013) 4654-4674. [4] J. Huot, G. Liang, R. Schulz, Magnesium-based nanocomposites chemical hydrides, J. Alloys Compd., 353 (2003) L12-L15. [5] C. An, G. Liu, L. Li, Y. Wang, C. Chen, Y. Wang, L. Jiao, H. Yuan, In situ synthesized one-dimensional porous Ni@C nanorods as catalysts for hydrogen storage properties of MgH2, Nanoscale, 6 (2014) 3223-3230. [6] X.Z. Xiao, C.C. Xu, J. Shao, L.T. Zhang, T. Qin, S.Q. Li, H.W. Ge, Q.D. Wang, L.X. Chen, Remarkable hydrogen desorption properties and mechanisms of the Mg2FeH6@MgH2 core-shell nanostructure, J. Mater. Chem. A, 3 (2015) 5517-5524.

ACCEPTED MANUSCRIPT

[7] Y. Liu, H. Pan, M. Gao, Y. Zhu, Y. Lei, Q. Wang, The effect of Mn substitution for Ni on the structural and electrochemical properties of La0.7Mg0.3Ni2.55−xCo0.45Mnx hydrogen storage electrode alloys, Int. J. Hydrogen Energy, 29 (2004) 297-305. [8] L.Z. Ouyang, F.X. Qin, M. Zhu, The hydrogen storage behavior of Mg3 La and Mg3 LaNi0.1, Scripta Mater., 55 (2006) 1075-1078. Compd., 446 (2007) 124-128.

RI PT

[9] L.Z. Ouyang, H.W. Dong, M. Zhu, Mg3Mm compound based hydrogen storage materials, J. Alloys [10] L.Z. Ouyang, X.S. Yang, H.W. Dong, M. Zhu, Structure and hydrogen storage properties of Mg3Pr and Mg3PrNi0.1 alloys, Scripta Mater., 61 (2009) 339-342.

[11] N. Terashita, K. Kobayashi, T. Sasai, E. Akiba, Structural and hydriding properties of (Mg1−xCax)Ni2 Laves phase alloys, J. Alloys Compd., 327 (2001) 275-280.

[12] P. Chiotti, R.W. Curtis, P.F. Woerner, Metal hydride reactions: II. Reaction of hydrogen with

SC

CaMg2 and CaCu5 and thermodynamic properties of the compounds, J. Less Common Met., 7 (1964) 120-126.

[13] D. Lupu, A. Biris, E. Indrea, R.V. Bucur, Effects of Ca additions on some Mg-alloy hydrides, Int. J.

M AN U

Hydrogen Energy, 8 (1983) 701-703.

[14] N. Terashita, E. Akiba, Hydrogenation Properties of CaMg2 Based Alloys, Mater. Trans., 45 (2004) 2594-2597.

[15] L. Ouyang, M. Ma, M. Huang, R. Duan, H. Wang, L. Sun, M. Zhu, Enhanced Hydrogen Generation Properties of MgH2-Based Hydrides by Breaking the Magnesium Hydroxide Passivation Layer, Energies, 8 (2015) 4237-4252.

[16] P.S. Rudman, Hydriding and dehydriding kinetics, J. Less Common Met., 89 (1983) 93-110. [17] K.F. Kelton, Analysis of crystallization kinetics, Mater. Sci. Eng. A, 226–228 (1997) 142-150.

TE D

[18] Y. Liu, F. Wang, Y. Cao, M. Gao, H. Pan, Q. Wang, Mechanisms for the enhanced hydrogen desorption performance of the TiF4-catalyzed Na2 LiAlH6 used for hydrogen storage, Energ. Environ. Sci., 3 (2010) 645-653.

[19] O. Kircher, M. Fichtner, Kinetic studies of the decomposition of NaAlH4 doped with a Ti-based catalyst, J. Alloys Compd., 404–406 (2005) 339-342. [20] J.W. Christian, The Theory of Transformations in Metals and Alloys, Mater. Today, 6 (2003) 53.

EP

[21] R.K. Singh, T. Sadhasivam, G.I. Sheeja, P. Singh, O.N. Srivastava, Effect of different sized CeO2 nano particles on decomposition and hydrogen absorption kinetics of magnesium hydride, Int. J. Hydrogen Energy, 38 (2013) 6221-6225.

AC C

[22] Z. Yuan, T. Yang, W. Bu, H. Shang, Y. Qi, Y. Zhang, Structure, hydrogen storage kinetics and thermodynamics of Mg-base Sm5Mg41 alloy, Int. J. Hydrogen Energy, 41 (2016) 5994-6003.

[23] C.A. Chung, C.-S. Lin, Prediction of hydrogen desorption performance of Mg2Ni hydride reactors, Int. J. Hydrogen Energy, 34 (2009) 9409-9423. [24] S. Hayashi, Distribution and dynamics of hydrogen in the low-temperature phase of Mg2NiH4

studied by solid-state NMR, Inorg. Chem., 41 (2002) 2238-2242. [25] Q. Luo, X.-H. An, Y.-B. Pan, X. Zhang, J.-Y. Zhang, Q. Li, The hydriding kinetics of Mg–Ni based hydrogen storage alloys: A comparative study on Chou model and Jander model, Int. J. Hydrogen Energy, 35 (2010) 7842-7849. [26] M.H. Grosjean, M. Zidoune, L. Roue, Hydrogen production from highly corroding Mg-based materials elaborated by ball milling, J. Alloys Compd., 404 (2005) 712-715. [27] J.P. Tessier, P. Palau, J. Huot, R. Schulz, D. Guay, Hydrogen production and crystal structure of

ACCEPTED MANUSCRIPT

ball-milled MgH2–Ca and MgH2–CaH2 mixtures, J. Alloys Compd., 376 (2004) 180-185. [28] L.X. Hu, E.D. Wang, Hydrogen generation via hydrolysis of nanocrystalline MgH2 and MgH2-based composites, Trans. Nonferrous Met. Soc. China, 15 (2005) 965-970. [29] P. Liu, H. Wu, C. Wu, Y. Chen, Y. Xu, X. Wang, Y. Zhang, Microstructure characteristics and hydrolysis mechanism of Mg–Ca alloy hydrides for hydrogen generation, Int. J. Hydrogen Energy, 40 (2015) 3806-3812.

RI PT

[30] L.Z. Ouyang, J.M. Huang, H. Wang, Y.J. Wen, Q.A. Zhang, D.L. Sun, M. Zhu, Excellent hydrolysis performances of Mg3RE hydrides, Int. J. Hydrogen Energy, 38 (2013) 2973-2978.

[31] R. Aiello, J.H. Sharp, M.A. Matthews, Production of hydrogen from chemical hydrides via

AC C

EP

TE D

M AN U

SC

hydrolysis with steam, Int. J. Hydrogen Energy, 24 (1999) 1123-1130.

ACCEPTED MANUSCRIPT

Highlights


The hydrogen storage properties and transition mechanism of CaMg2-based alloys are clarified. The hydrolysis performance of hydrogenated CaMg2 hydrides is discussed for the first time.




CaMg1.9Ni0.1 alloys absorb 5.65 wt.% hydrogen at 25 kJ/mol.

with Ea being 41.74

An undefined hydride in Mg-Ca system has been determined to be Ca5Mg9H28.

AC C

EP

TE D

M AN U

SC




RI PT