ww.sciencedirect.com

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e8

Available online at w

ScienceDirect

journal homepage: www.elsevier .com/locate/he

Enhanced hydrogen storage in accumulative rollbonded Mg-based hybrid

Mohammad Faisal, Anshul Gupta, Suboohi Shervani, Kantesh Balani,Anandh Subramaniam*

Department of Materials Science and Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India

a r t i c l e i n f o

Article history:

Received 17 January 2015

Received in revised form

17 March 2015

Accepted 19 March 2015

Available online xxx

Keywords:

Accumulative roll bonding

Mg-based hybrids

Hydrogen storage

Pressure-composition-isotherms

* Corresponding author. Tel.: þ91 (512) 259 7E-mail address: anandh@iitk.ac.in (A. Su

http://dx.doi.org/10.1016/j.ijhydene.2015.03.00360-3199/Copyright © 2015, Hydrogen Ener

Please cite this article in press as: FaisalInternational Journal of Hydrogen Energy

a b s t r a c t

Magnesium based hybrids have potential applications for hydrogen storage in the solid

state. Although magnesium can store a high amount of hydrogen (7.6 wt.%), high tem-

peratures (~300 �C) and pressures (0.3e1 MPa) are required for the same. Additionally, the

kinetics of absorption of hydrogen in bulk magnesium is slow. In the current work, Mg

eLaNi5-soot hybrids are synthesized by the accumulative roll bonding (ARB) process (30 roll

passes, 50% reduction per pass). It is observed that the hybrid absorbs 5 wt.% hydrogen at

250 �C at a pressure of ~0.33 MPa (4.5 wt.% hydrogen at a plateau pressure of less than

0.08 MPa). After 30 ARB passes, the kinetics of absorption of the hybrid was 4.0 wt.%

hydrogen in 30 s at 2 MPa, which is 3500% faster than the Mg (ARB) sample and 500% faster

than MgeLaNi5 hybrid. This combination of operating parameters and enhanced hydrogen

storage properties (high capacity at lower temperatures and pressures combined with

rapid kinetics) offer exciting prospects towards applications, given that bulk samples can

be synthesized in large quantities using the process developed. After 25 ARB passes there

are more than 105 layers in the hybrid and the layer thickness becomes less than ~24 nm.

Hence, intimate mixing of the components of the hybrid, along with a fine microstructure

and increased defect density in the hybrid, seems to play an important role in the

enhancement of the hydrogen absorption properties.

Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

Introduction

Hybrids are emerging as promising materials for solid state

hydrogen storage [1e4]. It is expected that hybrids will help to

overcome some of the limitations found in monolithic mate-

rials used for hydrogen storage. These limitations specifically

relate to a preferred combination of thermodynamic (tem-

perature and pressure of absorption and desorption) and ki-

netic parameters (rate of absorption and desorption) [5].

215.bramaniam).95gy Publications, LLC. Publ

M, et al., Enhanced hyd(2015), http://dx.doi.org

Multiple techniques have been used by investigators to

synthesize these hybrids. These include ball milling [6], sin-

tering [7], sputtering [8] and conventional mechanical mixing

[9]. Some of these processing routes (e.g. ball milling) not only

help in the synthesis of the hybrid, but also help in micro-

structural engineering of the sample (i.e. achieving finer grain

size, fine scale distribution of the second phase, higher

dislocation density, etc.) [10,11].

One of the widely used methods for the synthesis of

monolithic and composite materials for hydrogen storage is

ished by Elsevier Ltd. All rights reserved.

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e82

ball milling. MgeLaNi5 composites have been synthesized by

ball milling by Terzieva et al. [12]. They have reported the

absorption of 1.9 wt.% hydrogen in ~30 s at 300 �C and 1 MPa

pressure in a Mg-30 wt.% LaNi5 sample. Liang et al. [6] studied

hydrogen storage properties in ball milled MgH2-30 wt.%

LaNi5. They observed hydrogen storage capacity of ~3.7 wt.%

at a plateau pressure of 0.2 MPa at 310 �C. They also reported

~1.4 wt.% hydrogen absorption in 30 s (1 MPa, 150 �C). To

emphasize the effect of finer grain size on the hydrogen ab-

sorption, Fu et al. [13] synthesized Mg-5 wt.% LaNi5 compos-

ites by ball milling under hydrogen atmosphere and reported

higher hydrogen storage capacity (~4.6 wt.% at a plateau

pressure of 0.04 MPa at 245 �C and ~4.8 wt.% at a plateau

pressure of >0.1 MPa at 285 �C), coupled with reasonably good

kinetics (>3.5 wt.% in 30 s a ~285 �C at 2 MPa). A point note-

worthy of consideration is that the kinetic data, inferred from

plots in literature, should be read with a degree of caution

(given that the data set in most cases is in a time scale of

60 min). With increasing wt.% of LaNi5 (5%, 15%, 35%) in the

composite, a decrease in the amount of hydrogen absorbed is

observed at the plateau pressure. It is to be noted that in all

cases a 'strict' plateau is not observed (i.e. the plateau is

'sloping'). Sun et al. [1] synthesized Mg-X wt.% LaNi5 (X ¼ 20,

30, 40, 50) by ball milling followed by sintering. They have

achieved best hydrogen storage capacity of ~4.3 wt.% (0.2 MPa

plateau pressure at 300 �C) for the Mg-20 wt.% LaNi5 com-

posite. Liu et al. [14] prepared Mg-20 wt.% LaNi5 composite

using a laser sintering technique and they have observed

appreciable hydrogen storage capacity (~4 wt.% at a plateau

pressure of 0.15 MPa at 300 �C).Some literature exists on the hydrogen storage character-

istics of MgeC composites as well. Zhou et al. [15] have pre-

paredMg-30wt.% carbon (crystallitic form obtained from coal)

composite by ball milling under hydrogen pressure of 1 MPa.

They observed hydrogen storage capacity of ~4.2 wt.% at a

plateau pressure of 0.28 MPa (at 300 �C). They have explained

enhanced hydrogen absorption in MgeC composites by the

presence of carbon 'dangling bonds', which can absorb

hydrogen. Konarova et al. [16] have observed reasonable

hydrogen storage kinetics (2.1 wt.% in 30 s at 250 �C) in a

MgH2eC porous composite, synthesized by decomposition of

an organomagnesium precursor under hydrogen. Jeloaica

et al. [17] have reported that during interaction of hydrogen

atom with graphite (0001) surface, both physiosorption and

chemisorption of hydrogen take place. This leads to the

elimination of the energy barrier required for the diffusion of

hydrogen and thereby enhances kinetics. Wu et al. [18] have

prepared MgH2-5wt.% AP (as prepared SWNT containing

metallic particles). They observed fast kinetics absorbing

5 wt.% hydrogen in ~30 s at 200 �C. They have identified the

role of CNTs in acting as diffusion channels into the Mg ma-

trix, thus enhancing the kinetics. Amirkhiz et al. [19] prepared

SWNT-MgH2 composite by co-milling for 1 h. They proposed

that SWNT functions as a “hydrogen pump” by penetrating

into the thin surface hydroxide shell on the surface Mg. A

general conclusion on the role of carbon enhancing the ki-

netics is related to the unique electron characteristics of car-

bon and the morphology effect [19]. Wu and Cheng [20]

studied the effect of carbon on hydrogen storage properties.

They have stated that small radius curvature carbon show

Please cite this article in press as: Faisal M, et al., Enhanced hydInternational Journal of Hydrogen Energy (2015), http://dx.doi.org

appreciable catalytic effect when mechanically milled with

Mg due to the presence of sp and sp2 hybridization and thus

the delocalized p electronsmay interact with hydrogen atoms

and molecules.

In the past decade, many severe plastic deformation

techniques, including ARB [21e23], have gained prominence

as synthesis techniques to fabricate hydrogen storage mate-

rials [24]. Dufour and Huot [25] prepared an MgePd composite

by ARB. The hydrogen storage characteristics of these com-

posites are not as good as some of the LaNi5 containing

composites discussed earlier. The MgePd composite absorbs

~1 wt.% hydrogen at a plateau pressure of 0.4 MPa at 350 �Cwith a sloping plateau. Mohsen et al. [26] have synthesized

Mg-22 at.% Ti layered composites via accumulative roll

bonding. These composites show hydrogen storage capacity

of ~3 wt.% at a plateau pressure of 0.6 MPa at 350 �C and

reasonable kinetics of ~0.8 wt.% in 60 s (at 2 MPa and 350 �C).Botta et al. [27] have synthesized an Mg sample using several

cold rolling passes. It showed limited absorption of 0.5 wt.%

hydrogen in 30 s (350 �C, 2MPa). Large amounts of strain due to

severe plastic deformation techniques lead to refinement in

grain size [28] and creation of non-metallurgical bonded in-

terfaces [26]. Also, a number of defects such as dislocations,

vacancies are also increased during the process [11]. There-

fore, the effect of grain refinement, defects and synergistic

combination of chemisorptions and physiosorption help in

achieving enhanced kinetics of hydrogen storing material via

ARB in MgeLaNi5-Carbon(soot) hybrids.

The current work aims at the following: (i) synthesis of Mg

based hybrids (containing LaNi5 and soot) by accumulative roll

bonding (ARB), (ii) characterization of the material for

hydrogen absorption and desorption properties via pressure-

composition-isotherms (PCI) and wt.%H versus time plots.

The overall goal is to: (i) push the amount of hydrogen stored

in the hybrid to that obtained by routes like ball milling, (ii)

reduce the absorption pressure and temperature (with respect

to other bulk samples), (iii) increase the desorption pressure

(to >0.1 MPa), (iv) enhance the absorption and desorption ki-

netics (to bring it on par with ball milled samples). One syn-

thesized sample will also be tested for cyclability (i.e. storage

of Hydrogen with multiple absorptionedesorption cycles).

An important point to note with respect to the hydrogen

absorption capacity is that, often in the literature, the

maximum hydrogen absorption capacity is quoted, instead of

the hydrogen absorption at the plateau pressure. From a

perspective of applications of these materials for hydrogen

storage, the absorption at the plateau pressure is the relevant

parameter, which should be considered. Hence, in the current

work, the hydrogen absorption capacity at the plateau pres-

sure is quoted. For cases in the literature, where the absorp-

tion at plateau pressure is not quoted, the value is deduced

from their reported experimental curves.

Experimental details

MgeLaNi5-Soot hybrids were synthesized by an accumulative

roll bonding (ARB) process (as schematically illustrated in

Fig. 1). Magnesium sheets (0.8 mm uniform thickness) were

prepared from starting sheets of 5 mm thickness (cut from

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095

Fig. 1 e Schematic showing the steps involved in the

accumulative roll bonding process to synthesize

MgeLaNi5-soot hybrids. Sample is trimmed after each pass

to removed cracked areas.

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e8 3

99.9% pure Mg ingots) by three pass rolling (50% reduction per

pass), using 7.8 HP driven 5” diameter high carbon steel rolls

rotating at 72 rpm. Intermediate annealing (400 �C for 10 min)

was carried out to retain the ductility of the sample. LaNi5 was

prepared by melting Lanthanum ingots (99.9% purity) and

Nickel shots (99.9% purity, 3e5 mm diameter and 2 mm

length) in a radio frequency vacuum induction furnace. After

evacuation the melting was done under argon atmosphere.

LaNi5 (ground to powder using a mortar and pestle) and soot

(from a kerosene lamp) were incorporated between the Mg

layers, before the ARB process. The severe plastic deformation

introduced during cold rolling (50% reduction per pass), not

only introduces strain in the material, but also bonds the

sheets of Mg. After rolling, the sample is cut into two halves,

stacked and rolled again, with further incorporation of LaNi5(for the first 10 ARB passes) and soot (25 passes).

A total of ~8 wt.% of LaNi5 (after 10 passes) and ~1.25 wt.%

of soot (after 25 passes) was incorporated in the hybrid. The

sample was subjected to a total of 30 ARB passes. The process

was carried out without any protective atmosphere. Mg

(without incorporation of interlayer materials) and MgeLaNi5(fraction of LaNi5 being 8 wt.%) samples were also prepared by

ARB (keeping the other process parameters constant) as

reference samples for the comparison of the hydrogen ab-

sorption characteristics, with MgeLaNi5-soot hybrids. The

amount of material incorporated per pass is restricted by the

Please cite this article in press as: Faisal M, et al., Enhanced hydInternational Journal of Hydrogen Energy (2015), http://dx.doi.org

requirement that a good bond is obtained between the sheets

on cold rolling. The total number of passes in ARB is limited by

the fact that the edges of the sample tend to crack upon cold

rolling. The thickness of the final sample is about 0.3e0.5 mm.

Phase formation of the hydrogenated sample was studied

by X-ray diffraction (Bruker D8 Focus) using Cu Ka radiation

(l ¼ 1.54 A), with a scan rate of 0.5�/min (step size 0.01�).Elemental analysis of the sample was performed on MgeLa-

Ni5esoot (30 ARB passes) using energy dispersive spectros-

copy (EDS) in a JEOL JSM-7100F Field Emission Scanning

Electron Microscope (FESEM).

A standard Sievert's apparatus (Gas Reaction Controller,

Advanced Materials Co., Pittsburgh, USA) was used to obtain

pressure-composition-isotherms (PCI) and to study the ab-

sorption and desorption kinetics of hydrogen (99.99% pure). A

K-type thermocouple (connected to Omega Controllers

(CN8241)) was used to measure the temperature of the system

with an accuracy of ±0.2 �C and with a stability of better than

±0.05 �C. Pressure transducers (Honeywell TJE) were used to

measure pressure, with a resolution of 50 Pa for measure-

ments up to 1 MPa and 0.001 MPa for measurements up to

20 MPa. In order to establish the equilibrium of the gas in the

sample chamber, the values of change in the slope of tem-

perature with time (0.002 �C/s) and change in the slope of

pressure with time (0.00245 psi/s) together were ensured to

reach near zero for at least 2 min. The activation procedure

consists of cycling the sample under the hydrogen pressure of

2 MPa for 1 h, followed by evacuation to a vacuum of 0.01 MPa

for half an hour (at 300 �C). Sampleswere completely desorbed

under a vacuum of 0.01 MPa for 1 h (at 350 �C) before per-

forming absorption and desorption measurements. All mea-

surements of absorption kinetics were carried out at 2 MPa

and that of desorption kinetics at 0.01MPa pressure. Hydrogen

absorption characteristics were studied using 100 mg of the

sample, which was loaded in a stainless steel chamber

(SS316L, 2cc volume).

Results and discussions

Fig. 2 shows the PCI curves obtained by hydrogenation carried

out on Mg, MgeLaNi5 and MgeLaNi5-soot ARB samples at

300 �C. The ARB processed Mg sample (reference sample in

which LaNi5 or soot has not been incorporated) has a

hydrogen storage capacity of 3 wt.% hydrogen at a pressure of

~0.6 MPa (~2 wt.% at a plateau pressure of ~0.45 MPa) whereas

an enhanced capacity of 4 wt.% for MgeLaNi5 at 0.4 MPa

(3 wt.% at a plateau pressure of 0.2 MPa) and 5 wt.% for

MgeLaNi5-soot hybrid at 0.33 MPa (4.5 wt.% at a plateau

pressure of 0.2 MPa) is observed. Inset to the Fig. 2 shows PCI

(absorption) curves for the MgeLaNi5-soot sample after

cycling up to six cycle at 300 �C and kinetic curves (wt.%

hydrogen absorbed with time) up to six cycles at 2 MPa and

300 �C. To test the utility of the hybrid for repeated storage

(absorption followed by desorption), the MgeLaNi5-soot

sample was subjected to six absorptionedesorption cycles

using pressure composition isotherms at 300 �C. The PCI

curves for six absorption cycles are as shown in the inset to

Fig. 2. It is clearly seen that the sample shows repeatable

hydrogen storage capacity of ~5 wt.% at ~0.33 MPa (~4.5 wt.%

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095

Fig. 2 e PCI curves obtained for three different samples synthesized by the ARB process (30 passes): (a) Mg, (b) MgeLaNi5, (c)

MgeLaNi5-soot at 300 �C. Inset to the figure shows PCI (absorption) curves for the MgeLaNi5-soot sample after cycling up to

six cycle at 300 �C and kinetic curves (wt.% hydrogen absorbed with time) up to six cycles at 2 MPa and 300 �C.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e84

at the plateau pressure of 0.2 MPa). The PCI curve for samples,

which have undergone 2e6 cycles, is slightly higher (wt.%

hydrogen) than the initial sample. Absorption kinetic curves

at 300 �C and 2 MPa pressure (shown in the inset up to six

cycles) show that there is a slight increase in the capacity after

the fifth cycle.

Clearly, there is a significant benefit of synthesizing the

three component hybrid, in terms of the amount of hydrogen

absorbed at the plateau pressure. The increased interfacial

area (with concomitant increase in the sub-interface area) and

soot are expected to be contributing to this enhanced ab-

sorption of hydrogen at the plateau pressure [15]. It is to be

noted that unlike the ball milled powder samples, the ARB

samples can be considered as 'bulk'.Fig. 3 shows the PCI absorption and desorption curves

obtained for the MgeLaNi5-soot hybrids at three different

temperatures (250 �C, 270 �C, and 300 �C). It is seen that with

a decreasing temperature (in the range of 300 �Ce250 �C) thehydrogen storage capacity at the plateau pressure remains

practically unaffected; however, there is a significant

reduction in the plateau pressure with temperature. At

250 �C the hybrid absorbs 5 wt.% at ~0.33 MPa hydrogen

pressure (4.5 wt.% hydrogen at a plateau pressure < 0.08

MPa). This result compares favorably with the results of Fu

et al. [13] at 245 �C (~4.6 wt.% at a plateau pressure of

0.04 MPa). During desorption a slight rise in the pressure is

observed, before attaining a nearly constant value (plateau).

This arises due to the inherent limitation of the instrument

used and indicates non-equilibrium conditions during

desorption (due to very slow kinetics at the beginning and

end of desorption [29]).

Please cite this article in press as: Faisal M, et al., Enhanced hydInternational Journal of Hydrogen Energy (2015), http://dx.doi.org

The significant difference is the advantages of the pro-

cessing methodology adopted in the current investigation.

These are: (i) there is no need of a hydrogen atmosphere

during processing, (ii) there is considerable saving in pro-

cessing time (ball milling was done for 20 h ormore, ARB takes

less than 30minmanually) and (iii) the process is amenable to

easy scale up [24].

The inset to Fig. 3 shows van't Hoff plot (ln (P) vs. 1000/T)

generated using the plateau pressure and temperature from

pressure-composition isotherms. The enthalpy of hydride

formation (DHabs ¼ �69 kJ/mol) and decomposition

(DHdes ¼ 79 kJ/mol H2) is calculated from the van't Hoff plot

and is close to the value in the literature for the formation of

MgH2 [30]. Further work needs to be carried out to understand

the underlying mechanisms responsible for the enhanced

absorption due to the addition of LaNi5 and soot. The entropy

of hydride formation is calculated to be 126 J/K.mol H2 and

decomposition is calculated to be 135 J/K.mol H2.

The inset to the Fig. 3 also shows PCI curves (absorption) for

Mg (30 passes) sample at (i) 300 �C, (ii) 380 �C& (iii) 400 �C along

with the van't Hoff plot generated from the PCI data obtained

at the three temperatures for Mg (30 passes). It is evident from

the enthalpy data that Mg is forming MgH2 phase during

hydrogen absorption. Higher plateau pressure has been

ascribed to sloping nature of the curve at 300 �C and slower

kinetics leading to non-equilibrium nature of the curve [29].

The information related to the kinetics of absorption is

shown in Fig. 4 (for Mg, MgeLaNi5 & MgeLaNi5-soot samples).

It is seen that at 300 �C (& 2 MPa pressure) the MgeLaNi5-soot

sample can absorb ~4 wt.% hydrogen in 30 s and 4.7 wt.% in

5min. Hence, thewt.% hydrogen absorbed in 30 s at 300 �C and

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095

Fig. 3 e PCI curves (absorption and desorption) for the MgeLaNi5-soot hybrid (30 passes) at different temperatures: (i) 250 �C,(ii) 270 �C & (iii) 300 �C. Inset shows PCI curves (absorption) for Mg (30 passes) sample at (i) 300 �C, (ii) 380 �C & (iii) 400 �Calong with the van't Hoff plot generated from the PCI data obtained at the three temperatures for Mg (30 passes) and

MgeLaNi5-soot hybrid (30 passes).

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e8 5

2 MPa pressure is superior to that of the ball milled MgeLaNi5sample of Fu et al. [13] (>3.5 wt.% in 30 s at ~285 �C at 2 MPa)

and significantly increased over that of Mg-30 wt.% LaNi5composites of Terzieva et al. [12]. The inset in Fig. 4 shows

Fig. 4 e Plot of wt.% hydrogen absorbed with time for Mg and M

different temperatures ((i) 250 �C, (ii) 270 �C, (iii) 300 �C) and deso

constant at 2 MPa during absorption and 0.01 MPa during deso

Please cite this article in press as: Faisal M, et al., Enhanced hydInternational Journal of Hydrogen Energy (2015), http://dx.doi.org

enhanced kinetics of absorption (plot of wt.%H with time) of

MgeLaNi5-soot samples with increasing temperatures

(compared at 250 �C, 270 �C & 300 �C) at 2 MPa. The inset also

shows the desorption curves at 0.01 MPa. It is to be noted that

g based hybrids. Inset shows absorption kinetics at three

rption at ((i) 250 �C, (ii) 270 �C, (iii) 300 �C). Pressure was kept

rption.

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095

Fig. 6 e XRD pattern from (a) MgeLaNi5-soot (30 ARB

passes) sample (b) MgeLaNi5-soot (hydrogenated at 300 �Cand 0.2 MPa plateau pressure, with 4.5 wt.%H capacity) (c)

MgeLaNi5-soot (hydrogenated at 300 �C and 2 MPa

pressure with 5.9 wt.%H capacity). Inset to (a) shows

zoomed in view of peaks (200 and 111) of second phase

LaNi5. Inset (towards right) shows weight fraction of

phases calculated using pseudo voigt function with peak

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e86

the desorption kinetics is extremely sluggish at 250 �C and

270 �C. The good kinetics of absorption observed can be

attributed to the fine-scale microstructure created by the ARB

process, along with a high defect density including disloca-

tions and interfaces [26].

Fig. 5 shows SEM micrographs for MgeLaNi5-soot hybrid

undergone (a) 30ARBpasses (b) hydrogenatedat 300 �Candat a

plateau pressure of 0.33 MPa. The inset to the figure (Fig. 5)

shows zoomed in view overlaid with EDS data plot obtained

froma line segment.As canbe seen fromFig. 5(a), inaddition to

the confirmation of the presence of LaNi5 between the mag-

nesium layers, it is seen that there is a scatter in the particle

size. Fig. 5(b) shows the formation of hydride phase at the

interface. Line scan along the interface suggest a drop in per-

centage of magnesium in the dark region. This could suggest

formation of MgH2 with lower concentration of magnesium.

Formation of MgH2 has also been confirmed by XRD (Fig. 6).

Fig. 6 shows X-ray diffraction patterns for (a) MgeLaNi5-

soot (30 pass) sample (b) MgeLaNi5-soot (hydrogenated at

0.33 MPa plateau pressure at 300 �C, having 4.5 wt.% hydrogen

storage capacity) (c) MgeLaNi5-soot (hydrogenated at 2 MPa

and 300 �C, having 5.9 wt.% hydrogen storage capacity). The

formation of MgH2 can be identified in the hydrogenated

sample. Additionally, from the relative intensity of theMg and

MgH2 peaks, it can be inferred that most of the Mg has been

converted intoMgH2. This correlateswell with the observation

Fig. 5 e Cross-sectional SEM micrographs of MgeLaNi5-

soot undergone (a) 30 ARB passes (b) partially

hydrogenated upto 4.5 wt.% hydrogen during PCI

absorption at 300 �C and at a plateau pressure of 0.33 MPa.

The inset to the figure (Fig. 5) shows zoomed in view

overlaid with EDS data plot obtained from a line segment.

matching index >95%.

Please cite this article in press as: Faisal M, et al., Enhanced hydInternational Journal of Hydrogen Energy (2015), http://dx.doi.org

of a single plateau in the PCI curve of the hybrid (i.e. it is only

the Mg in the hybrid, which is forming a hydride). Phase

fraction (in wt.%) have been calculated by fitting pseudo voigt

function in the XRD pattterns with peakmatching index >95%using JADE 2.6.3. The weight fraction of LaNi5 calculated

matches closely with the added amount. It is evident from the

phase fraction data that the amount of LaNi5 in the hybrid

does not change on hydrogenation.

Table 1 shows consolidated phase analysis (excluding soot)

of MgeLaNi5-soot hybrid undergone (a) 30 ARB Passes (b) hy-

drogenation at 300 �C and 0.33 MPa plateau pressure (4.5 wt.%

Table 1 e Phase fraction (excluding soot) in MgeLaNi5-soot hybrid after (a) 30 ARB passes (b) hydrogenation at300 �C and 0.33 MPa plateau pressure having 4.5 wt.%hydrogen storage capacity (c) hydrogenation at 300 �Cand 2 MPa pressure having 5.9 wt.% hydrogen storagecapacity.

Sample Wt.% H Phases Phase fraction(wt.%)

SEM XRD

ARB hybrid None Mg 90 92.3

LaNi5 8.9 7.7

Hydrogenated (at plateau

pressure and 300 �C)4.5 wt.%H Mg 40.5 42.6

MgH2 49.7 48.9

LaNi5 8.5 8.4

Hydrogenated (at 2 MPa

and 300 �C)5.9 wt.%H Mg 5 8.7

MgH2 88 82.9

LaNi5 7 8.4

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e8 7

H capacity) (c) hydrogenated at 300 �C and 2 MPa pressure

(5.9 wt.% H capacity) using SEM (point countmethod) and XRD

(pseudo-voigt fit).

The scalability of the technique developed is an important

advantage of themethod [31,21]. TheAbovementioned results

are significant improvements, keeping in view the bulk nature

of the sample and the easy scalability of the process employed

to synthesize the hybrids. The processing time to synthesize

the hybrid is also considerably reduced as compared to ball-

milling. Additionally, the ARB processing route does not

need environment control (like ball milling in hydrogen at-

mosphere), which is a major advantage from an application

standpoint.

From the foregoing discussions an optimum set of oper-

ating parameters for the hybrid can be arrived at: (i) absorp-

tion at 250 �C (the kinetic aspect is offset by the low plateau

pressure and the advantage of a lower temperature), (ii)

desorption at 300 �C (desorption at 250 �C is sluggish and at

300 �C reaches 95% of its capacity in 30 min).

Summary and conclusions

In this work, MgeLaNi5(~8 wt.%)-soot(~1.25 wt.%) hybrids

were synthesized by accumulative roll bonding process (30

roll passes with 50% reduction in each pass). A brief summary

and conclusions of the current work are:

(1) The hybrid sample (MgeLaNi5-soot) showed enhanced

hydrogen storage properties, 5 wt.% hydrogen at

~0.33 MPa hydrogen pressure at 250 �C (4.5 wt.%

hydrogen absorption at a plateau pressure of

<0.08 MPa). These figures compare satisfactorily with

ball milled MgeLaNi5 composite powders and consid-

erably better than ARB synthesized MgeLaNi5 hybrids.

(2) The sample showed absorption of ~4 wt.% hydrogen in

30 s at 300 �C and 2 MPa (~4.7 wt.% in 5 min). The ki-

netics of absorption is also considerably enhanced over

bulk Mg (ARB) and compares reasonably with samples

synthesized by other techniques like ball milling

(3.5 wt.% in 30 s, Mg-5 wt.% LaNi5, 285 �C) [13].(3) Soot seems to play an important role both in terms of

the increased absorption capacity and enhanced ki-

netics, without adding much weight to the hybrid.

Further studies are required to determine the precise

mechanisms responsible for the same. The hydrogen

storage characteristics of the hybrid are significantly

better than other bulk samples (keeping in view ca-

pacity, reversibility and kinetics).

(4) The absorption characteristics of the sample (as

measured from PCI curves) were found to be stable till

six cycles.

Acknowledgments

The authors would like to thank Prof. Maiya, Dr. Dattatreyan,

Dr.Rajalakshmi, Dr. P.Muthukumar, Prof. Vishwanathan, Prof.

Please cite this article in press as: Faisal M, et al., Enhanced hydInternational Journal of Hydrogen Energy (2015), http://dx.doi.org

Raj Pala, Prof. Muralidhar, Dr. Sankar and Dr. Satoru for

invaluable discussions and inputs related to hydrogen storage.

The authors would like to thank DST for funding of the project

under the Technology Systems Development Programme.

r e f e r e n c e s

[1] Sun D, Gingl F, Enoki H, Ross DK, Akiba E. Phase componentsof the sintered Mg-X wt % LaNi5 (X¼20e50) composites andtheir hydrogenation properties. Acta Mater 2000;48:2363e72.

[2] Zhu M, Wang H, Ouyang L, Zeng M. Composite structure andhydrogen storage properties in Mg-base alloys. Int JHydrogen Energy 2006;31:251e7. http://dx.doi.org/10.1016/j.ijhydene.2005.04.030.

[3] Zhu M, Zhu W, Chung C, Che Z, Li Z. Microstructure andhydrogen absorption properties of nano-phase compositeprepared by mechanical alloying of MmNi5�x(CoAlMn)x andMg. J Alloys Comp 1999;295:531e5.

[4] Ueda TT, Tsukahara M, Kamiya Y, Kikuchi S. Preparation andhydrogen storage properties of MgeNieMg2Ni laminatecomposites. J Alloys Comp 2005;386:253e7. http://dx.doi.org/10.1016/j.jallcom.2004.04.154.

[5] Targets for onboard hydrogen storage systems for light-dutyvehicles. US Department of Energy, Office of EnergyEfficiency and Renewable Energy and the FreedomCAR andfuel partnership; 2009. Revision 4.0, p.9.

[6] Liang G, Huot J, Boily S, Van Neste A, Schulz R. Hydrogenstorage in mechanically milled MgeLaNi5 and MgH2eLaNi5composites. J Alloys Comp 2000;297:261e5.

[7] Nagai H, Tomizawa H, Ogasawara T, Shoji K-I. Hydridingcharacteristics of Mg-xwt.%LaNi5 sintered alloys. J LessCommonMet 1990;157:15e24. http://dx.doi.org/10.1016/0022-5088(90)90402-6.

[8] Kalisvaart WP, Kubis A, Danaie M, Amirkhiz BS, Mitlin D.Microstructural evolution during hydrogen sorption cyclingof MgeFeTi nanolayered composites. Acta Mater2011;59:2083e95. http://dx.doi.org/10.1016/j.actamat.2010.12.010.

[9] Orimo S, Fujii H, Tabata M. Synthesis of fine compositeparticles for hydrogen storage, starting from Mg-YNi2mixture. J Alloys Comp 1994;210:37e43. http://dx.doi.org/10.1016/0925-8388(94)90112-0.

[10] Skripnyuk VM, Rabkin E, Estrin Y, Lapovok R. Improvinghydrogen storage properties of magnesium based alloys byequal channel angular pressing. Int J Hydrogen Energy2009;34:6320e4. http://dx.doi.org/10.1016/j.ijhydene.2009.05.136.

[11] Valiev RZ, Islamgaliev RK, Alexandrov LV. Bulknanostructured materials from severe plastic deformation.Pergamon; 2000pp.103e89.

[12] Terzieva M, Khrussanova M, Peshev P. Hydriding anddehydriding characteristics of Mg-LaNi5 composite materialsprepared by mechanical alloying. J Alloys Comp1998;267:235e9.

[13] Fu Y, Groll M, Mertz R, Kulenovic R. Effect of LaNi5 andadditional catalysts on hydrogen storage properties of Mg. JAlloys Comp 2008;460:607e13. http://dx.doi.org/10.1016/j.jallcom.2007.06.008.

[14] Liu D, Si T, Wang C, Zhang Q. Phase component,microstructure and hydrogen storage properties of the lasersintered Mge20wt.% LaNi5 composite. Scr. Mater2007;57:389e92. http://dx.doi.org/10.1016/j.scriptamat.2007.05.011.

[15] Zhou S, Chen H, Ding C, Niu H, Zhang T, Wang N, et al.Effectiveness of crystallitic carbon from coal as milling aid

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 5 ) 1e88

and for hydrogen storage during milling with magnesium.Fuel 2013;109:68e75. http://dx.doi.org/10.1016/j.fuel.2012.09.002.

[16] Konarova M, Tanksale A, Beltramini JN, Lu GQ. Porous MgH2/C composite with fast hydrogen storage kinetics. Int JHydrogen Energy 2012;37:8370e8. http://dx.doi.org/10.1016/j.ijhydene.2012.02.073.

[17] Jeloaica L, Sidis V. DFT investigation of the adsorption ofatomic hydrogen on a cluster-model graphite surface. ChemPhys Lett 1999:157e62.

[18] Wu CZ, Wang P, Yao X, Liu C, Chen DM, Lu GQ, et al.Hydrogen storage properties of MgH2/SWNT compositeprepared by ball milling. J Alloys Comp 2006;420:278e82.http://dx.doi.org/10.1016/j.jallcom.2005.10.028.

[19] Amirkhiz BS, Danaie M, Mitlin D. The influence of SWCNT-metallic nanoparticle mixtures on the desorption propertiesof milled MgH2 powders. Nanotechnology 2009;20:204016.http://dx.doi.org/10.1088/0957-4484/20/20/204016.

[20] Wu C, Cheng HM. Effect of carbon on hydrogen storageperformance of hydrides. J Mater Chem 2010;20:5290e400.http://dx.doi.org/10.1039/b926880d.

[21] Saito Y, Utsunomiya H, Tsuji N, Sakai T. Novel ultra-highstraining process for bulk materials-development of theaccumulative roll-bonding (ARB) process. Acta Mater1999;47:579e83. http://dx.doi.org/10.1016/S1359-6454(98)00365e6.

[22] Tsuji N, Saito Y, Lee S-H, Minamino Y. ARB (Accumulativeroll-bonding) and other new techniques to produce bulkultrafine grained materials. Adv Eng Mater 2003;5:338e44.http://dx.doi.org/10.1002/adem.200310077.

[23] Estrin Y, Vinogradov A. Extreme grain refinement by severeplastic deformation: a wealth of challenging science. AdvEng Mater 2013;61:782e817. http://dx.doi.org/10.1016/j.actamat.2012.10.038.

Please cite this article in press as: Faisal M, et al., Enhanced hydInternational Journal of Hydrogen Energy (2015), http://dx.doi.org

[24] Huot J, Skryabina NY, Fruchart D. Application of severeplastic deformation techniques to magnesium for enhancedhydrogen sorption properties. Metals 2012;2:329e43. http://dx.doi.org/10.3390/met2030329.

[25] Dufour J, Huot J. Study of Mg6Pd alloy synthesized by coldrolling. J Alloys Comp 2007;446e447:147e51. http://dx.doi.org/10.1016/j.jallcom.2006.11.191.

[26] Danaie M, Mauer C, Mitlin D, Huot J. Hydrogen storage inbulk MgeTi and Mgestainless steel multilayer compositessynthesized via accumulative roll-bonding (ARB). Int JHydrogen Energy 2011;36:3022e36. http://dx.doi.org/10.1016/j.ijhydene.2010.12.006.

[27] Botta WJ, Jorge AM, Veron M, Rauch EF, Ferrie E, Yavari AR,et al. H-sorption properties and structural evolution of Mgprocessed by severe plastic deformation. J Alloys Comp2013;580:S187e91. http://dx.doi.org/10.1016/j.jallcom.2013.03.013.

[28] Lowe T, Valiev R. The use of severe plastic deformationtechniques in grain refinement. JOM 2004;56:64e8. http://dx.doi.org/10.1007/s11837-004-0295-z.

[29] Long S, Zou J, Liu Y, Zeng X, Ding W. Hydrogen storageproperties of a MgeCe oxide nano-composite preparedthrough arc plasma method. J Alloys Comp2013;580:S167e70. http://dx.doi.org/10.1016/j.jallcom.2013.02.063.

[30] Griessen R, Riesterer T. Heat of formation models. In:Schlapbach L, editor. Hydrogen in intermetallic compoundsI: electronic, thermodynamic, and crystallographicproperties, preparation. Springer series topics in appliedphysics; 1988. p. 219e84.

[31] Jamaati R, Toroghinejad MR. High-strength and highly-uniform composite produced by anodizing and accumulativeroll bonding processes. Mater Des 2010;31:4816e22. http://dx.doi.org/10.1016/j.matdes.2010.04.048.

rogen storage in accumulative roll bonded Mg-based hybrid,/10.1016/j.ijhydene.2015.03.095