nanomaterials for solid state hydrogen storage, 2009, p.346

2, we review hydrogen storage properties of selected simple metal and inter-metallic hydrides with the most emphasis on magnesium hydride MgH2 which now can be treated as a model hydride

Series Editor: Narottam P Bansal

NASA Glenn Research CenterCleveland, OH 44135

narottam.p.bansal@nasa.gov

Aims and Scope of the Series

During the last couple of decades, notable developments have taken place in the science and technology of fuel cells and hydrogen energy Most of the knowledge developed in this field is contained in individual journal articles, conference pro-ceedings, research reports, etc Our goal in developing this series is to organize this information and make it easily available to scientists, engineers, technologists, designers, technical managers and graduate students The book series is focused to ensure that those who are interested in this subject can find the information quickly and easily without having to search through the whole literature The series includes all aspects of the materials, science, engineering, manufacturing, modeling, and applications Fuel reforming and processing; sensors for hydrogen, hydrocarbons and other gases will also be covered within the scope of this series A number of volumes edited/authored by internationally respected researchers from various countries are planned for publication during the next few years

Titles in this series

Nanomaterials for Solid State Hydrogen Storage

R.A Varin, T Czujko, and Z S Wronski

ISBN 978-0-387-77711-5, 2009

Modeling Solid Oxide Fuel Cells: Methods, Procedures and Techniques

R Bove and S Ubertini, eds.

ISBN 978-1-4020-6994-9, 2008

Zbigniew S Wronski

Nanomaterials for Solid State Hydrogen Storage

University of Waterloo University of Waterloo

Zbigniew S Wronski

CANMET Energy Technology Centre

Hydrogen Fuel Cells and

Library of Congress Control Number: 2008929618

© Springer Science+Business Media, LLC 2009

All rights reserved This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York,

NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis Use in connection with any form of information storage and retrieval, electronic adaptation, computer software,

or by similar or dissimilar methodology now known or hereafter developed is forbidden

The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject

to proprietary rights.

Printed on acid-free paper

springer.com

Although hydrogen as a chemical element has been known to the humankind and used in various capacities for a very long time, only in the past 15 years its importance

to the world population as an energy vector has gradually emerged A long-term reliance of humanity on the energy derived solely from fossil fuels, such as coal in the nineteenth and crude oil and natural gas in the twentieth century, has led to a number of new challenges facing all of us in the twenty-first century, such as sharp reduction in the world crude oil and eventually coal supply, global warming and following climate changes due to the release of growing amounts of greenhouse gas

CO2, and poor urban air quality Hydrogen is essentially the only viable remedy for the growing world energy problems Hydrogen is a very attractive alternative energy vector for replacing fossil fuel-based economy The future Hydrogen Economy offers a potential solution to satisfying the global energy requirements while reducing (and eventually eliminating) carbon dioxide and other greenhouse gas emissions and improving energy security Hydrogen is ubiquitous, clean, efficient, and can be produced directly from sunlight and water by biological organ-isms and using semiconductor-based systems similar to photovoltaics Hydrogen can also be produced indirectly via thermal processing of biomass or fossil fuels where the development of advanced technological processes combined with a CO2sequestration is emerging

However, this rosy picture, as it usually happens in a real life, is marred by a number of obstacles which must be overcome before the Hydrogen Economy becomes a reality One of these obstacles is safe and efficient storage of hydrogen particularly for mobile/automotive applications where hydrogen gas will be supplied to fuel cells that, in turn, will power the transport vehicles in a clean, inexpensive, safe, and efficient manner From all possible solutions to hydrogen storage the one which relies upon storage in solid media (hydrides) is the most attractive one The fast emerging nanoscience/nanotechnology will allow fabricating nanomaterials for solid-state hydrogen storage that can, in a long run, revolutionize hydrogen storage

This book is our modest contribution to this innovative area of hydrogen storage Wherever possible we tried to illustrate the hydrogen storage behavior by our own results In Chap 1, we introduce the reader to the motivation for the transformation

to the Hydrogen Economy In a number of following sections/subsections, we

v

provide a comprehensive synchronic history of development of hydrides and materials including the existing fabrication methods with a special emphasis on ball (mechanical) milling in high-energy mills Important hydride properties and experi-mental techniques for assessing hydrogen storage behavior are also discussed In Chap 2, we review hydrogen storage properties of selected simple metal and inter-metallic hydrides with the most emphasis on magnesium hydride (MgH2) which now can be treated as a model hydride whose hydrogen storage properties in nanos-tructured form can be used as a benchmark for comparing the properties of other hydrides Chapter 3 brings a thorough review of the properties of complex hydrides whose high volumetric and gravimetric capacities make them most attractive for the vehicular solid-state hydrogen storage in transportation Chapter 4 provides infor-mation on carbons and nanocarbons as alternative means of hydrogen storage to solid hydrides This includes diamond and nanodiamond, graphene, ordered graph-ites and nanographites, disordered and active carbons, fullerenes, carbon nano-tubes, and other nanoshapes Chapter 5 is a sort of an executive summary where we provide a critical assessment of the present state of knowledge and make predic-tions for the future developments.

1 Introduction 1

1.1 Motivation: The Hydrogen Economy 1

1.2 Brief, Synchronic History of Development of Hydrides and Nanomaterials 7

1.2.1 Early Investigations of Metal–Hydrogen Systems and Hydrides 7

1.2.2 Early Routes to Nanomaterials 10

1.2.3 Historical Development of Classical Hydrogen Storage AB5 Alloys 13

1.2.4 Historical Development of Interstitial Hydrides in Other Intermetallic Systems 15

1.2.5 Historical Development of Nanophase AB2 Intermetallic Hydrides 16

1.2.6 New Routes to Nanomaterials: Mechanical Alloying and Mechanochemical Activation 17

1.2.7 Historical Development of Lightweight Metal Hydrides and Hydride Complexes 18

1.2.8 Early Studies of Noninterstitial Transition Metal Ternary Hydrides 20

1.2.9 Toward Chemical/Complex Hydrides 21

1.2.10 Historical Development of Nanocarbons and Carbon Nanotubes 23

1.2.11 New Materials and Techniques 25

1.3 Nanoprocessing in Solid State in High-Energy Ball Mills 27

1.3.1 Processes for the Synthesis of Nanostructured Materials 27

1.3.2 Milling Processes and Equipment 28

1.3.3 Nanoprocessing Methods and Mechanisms 37

1.3.3.1 Mechanical Milling 38

1.3.3.2 Mechanical Alloying 39

vii

1.3.3.3 Mechanochemical Activation 40

1.3.3.4 Mechanochemical Synthesis (Mechanosynthesis) of Nanohydrides 52

1.3.3.5 Mechanical Amorphization 55

1.4 Important Hydride Properties and Experimental Techniques 56

1.4.1 Thermodynamics 56

1.4.1.1 Pressure–Composition–Temperature (PCT) Properties 56

1.4.1.2 Calculation of Activation Energy 60

1.4.2 PCT and Kinetic Curves Determination by Volumetric Method in a Sieverts-Type Apparatus 65

1.4.3 Microstructural Characterization of Ball-Milled Hydrides 71

1.4.4 Weight Percent of a Hydride Phase and Hydrogen by DSC Method 73

References 74

2 Simple Metal and Intermetallic Hydrides 83

2.1 Mg/MgH2 83

2.1.1 Crystallographic and Material Characteristics 83

2.1.2 Hydrogen Storage Characteristics of Commercial Mg and MgH2 87

2.1.2.1 Absorption 87

2.1.2.2 Desorption 93

2.1.3 Hydrogen Storage Characteristics of Mechanically (Ball) Milled MgH2 102

2.1.3.1 Microstructural Evolution During Milling and Subsequent Cycling of Commercial MgH2 Powders 103

2.1.3.2 Hydrogen Absorption of Ball-milled Commercial MgH2 Powders 112

2.1.3.3 Hydrogen Desorption of Ball-milled Commercial MgH2 Powders 115

2.1.4 Hydrogen Storage Characteristics of MgH2 Synthesized by Reactive Mechanical (Ball) Milling of Mg 129

2.1.5 Aging Effects in Stored MgH2 Powders 146

2.1.6 Other Methods of Synthesis of Nanostructured MgH2 than Ball Milling 147

2.2 MgH2 with Catalytic Additives 151

2.2.1 Mg/MgH2–Metals and Intermetallics 152

2.2.1.1 Desorption in Vacuum 152

2.2.1.2 Desorption at Atmospheric Pressure of Hydrogen 153

2.2.2 Mg/MgH2–Metal Oxides 165

2.2.3 Mg/MgH–Carbon / Graphite and Carbon Nanotubes 169

2.3 Other Metal Hydrides Containing Mg 170

2.4 AlH3 174

2.5 Other Metal and Intermetallic-based Hydrides: New Developments 177

2.5.1 Metal Hydrides 179

2.5.2 Rare-Earth AB5 Compounds 181

2.5.3 Titanium–Iron AB Compounds 182

2.5.4 Titanium and Zirconium AB2 Compounds 183

2.5.5 Other Novel Intermetallic Hydrides 183

References 183

3 Complex Hydrides 195

3.1 Ternary Transition Metal Complex Hydrides 196

3.1.1 Mg2NiH4 196

3.1.2 Mg2FeH6 198

3.1.3 Mg2CoH5 204

3.2 Alanates 206

3.2.1 NaAlH4 206

3.2.2 LiAlH4 213

3.2.3 Mg(AlH4)2 and Ca(AlH4)2 223

3.3 Amides 231

3.4 Metal Borohydrides 240

3.5 Destabilization of High Desorption Temperature Hydrides by (Nano)Compositing 253

3.5.1 MgH2–LiAlH4 Composite System 255

3.5.2 MgH2–NaAlH4 Composite System 265

3.5.3 MgH2–NaBH4 Composite System 270

References 281

4 Carbons and Nanocarbons 291

4.1 Diamond and Nanodiamonds 291

4.2 Graphene, Ordered Graphite, and Nanographites 294

4.2.1 Graphene 294

4.2.1.1 In-Plane σ and Out-of-Plane π Bonding 295

4.2.1.2 Van der Walls Interplanar and Intermolecular Interactions 296

4.2.1.3 Physisorption of Hydrogen on Carbons 297

4.2.1.4 Chemisorption of Hydrogen on Carbons 298

4.2.2 Graphitic Nanofibers, Whiskers, and Polyhedral Crystals 299

4.2.3 Graphite 299

4.3 Disordered and Active Carbons 301

4.3.1 Disordered Graphites and Mechanically-Activated Carbons 301

4.3.2 Active Carbons and Chemically Activated Carbons 303

4.3.3 Amorphous Carbon 304

4.4 Highly Ordered Fullerenes, Carbon Nanotubes, and Carbon Nanohorns 305

4.4.1 Fullerenes and Hydrofullerenes 305

4.4.2 Carbon Nanotubes 308

4.4.3 Carbon Nanohorns 312

4.4.4 Nanostructured Carbon Shells and Carbon Onions 314

References 317

5 Summary 321

5.1 Metal/Intermetallic Hydrides 322

5.2 Complex Hydrides 323

5.3 Nanocarbons and Others 324

Index 327

Introduction

1.1 Motivation: The Hydrogen Economy

The energy supply to the humankind in the last two centuries was solely based on fossil fuels such as coal in the nineteenth century and crude oil and natural gas in the twentieth century Unfortunately, this fossil fuel-based economy has led to a number of new challenges facing all of us in the twenty-first century such as global warming and following climate changes due to the release of growing amounts of greenhouse gas CO 2 , poor urban air quality, and reduction in the world crude oil supply, which could reach so-called Hubbert’s Peak around year 2011–2020 It is also noted that no major oil field has been discovered since 1970 [1] Since the mid-1970s a concept of the ecologically clean Hydrogen Economy has been gain-ing momentum as essentially the only viable remedy for the growing world energy problems The Hydrogen Economy offers a potential solution to satisfying the global energy requirements while reducing (and eventually eliminating) carbon dioxide and other greenhouse gas emissions and improving energy security Hydrogen is a very attractive alternative energy vector It is ubiquitous, clean, efficient, and can be produced directly from sunlight and water by biological organisms and using semiconductor-based systems similar to photovoltaics Hydrogen can also be pro-duced indirectly via thermal processing of biomass or fossil fuels where the devel-opment of advanced technological processes combined with a CO 2 sequestration [2] have the potential to produce essentially unlimited quantities of hydrogen in a sustainable manner For example, electricity produced by wind turbines or nuclear power plants during off-peak periods [3] can be used for the electrolysis of water into hydrogen [4] and the latter stored for future distribution to places of use [1] When hydrogen burns, it releases energy as heat and produces water: 2H 2 +

O 2→2H 2 O Since no carbon is involved then using hydrogen produced from renewable

or nuclear energy as energy resource would eliminate carbon monoxide and CO 2 emissions and reduce greenhouse warming However, if air is used for flame com-bustion of hydrogen, small amounts of NO x may be produced In a fuel cell, hydro-gen is converted directly to electricity in a similar reaction to the earlier one for burning hydrogen, and in essence just electricity and water are produced Pure hydrogen enters the anode channel in a fuel cell and diffuses through a porous

DOI: 10.1007/978-0-387-77712-2_1 © Springer Science + Business Media, LLC 2009

anode toward the catalyst (Pt) where the hydrogen molecules H 2 are stripped of

their electrons and become positively charged ions (protons) (H 2 → 2H + + 2e − )

Protons then migrate through the proton-permeable polymeric (Nafion) membrane

(Proton Exchange Membrane also called Polymer Electrolyte Membrane – PEM)

and the electrons generated during oxidation pass through the external circuit to the

cathode, thereby creating electric current On the cathode side, humidified air

enters the cathode channel and diffuses toward the cathode-side catalyst layer At

the catalyst Pt surface, hydrogen protons recombine with electrons and oxygen

molecules in air to produce water and heat according to the overall reaction ½O 2 +

2H + + 2e − → H 2 O + heat This waste heat gives a PEM fuel cell an operating

tem-perature of 60–80°C [5] And last but not least is the fact that a hydrogen fuel-cell

car can convert hydrogen energy into motion about 2–3 times as efficiently as a

normal car converts gasoline energy into motion: depending on how it is designed

and run, a good fuel-cell system is about 50–70% efficient in converting hydrogen

to electricity, while a typical car engine’s efficiency from gasoline to output shaft

averages to only about 15–17% [6]

Scott [7] succinctly summarized the inevitability of hydrogen becoming the fuel

of the future by two rationales: depletion-based rationale and climate-based

ration-ale Driven by depletion, civilization must move from fossil fuels to sustainable

energy sources Realistically, the only way sustainable sources can be harvested to

make chemical fuels is via hydrogen Otherwise, how else can we get energy from

wind, solar, or nuclear power to fuel an airplane? On the other hand, atmospheric

CO 2 growth is such that the concentration of CO 2 in the atmosphere has increased

from 280 to 370 ppm over the past 150 years CO 2 emissions can only be slowed

by the extensive use of hydrogen and can only be stopped with the supremacy of

sustainable-derived H 2 among chemical fuels

The realization of the enormous benefits of the Hydrogen Economy has

trig-gered over the last 15 years intense activities in the development of

hydrogen-related technologies There are three major technological obstacles to the full

implementation of the Hydrogen Economy in the next few decades The first is the

cost of safe and efficient production of hydrogen gas At present, 48% of hydrogen

is produced from methane steam reforming, 30% from oil/naphtha reforming, 18%

from coal gasification, and only 3.9% from the electrolysis of water [8] Apparently,

the bulk of hydrogen production still relies upon fossil fuels, a by-product of which

is CO 2 However, the fossil fuel-based processes are much cheaper than the

elec-trolysis of water The efforts are underway to reduce the price of elecelec-trolysis-

electrolysis-derived hydrogen to $2–3 per kg The second obstacle is the further development

of PEM fuel cell (PEMFC), which is the primary cell most suitable for

transporta-tion Most importantly, the research is focused on the extension of the usable

serv-ice life, water flooding, dynamics, and reliability The present cost of energy

derived from a PEMFC is around $200 per kW and this must be reduced to around

$30 per kW before PEMFC could be fully commercialized The third obstacle is

hydrogen storage for supplying PEMFC There are three major competing

tech-nologies for hydrogen storage: compressed gas cylinders, liquid hydrogen tanks, and

metal hydrides [9, 10] Their comparison is shown in Table 1.1 A major drawback

of compressed hydrogen storage for transportation applications is the small amount

of hydrogen that may be stored in a reasonable volume (volumetric capacity/density)

As can be seen in Table 1.1 compressed hydrogen gas technologies, even at such enormous pressures as ~80 MPa, also suffer from low volumetric densities not exceeding ~40 kgH 2 m −3 As pointed out by Sandi [10] even at such a high pressure

as 70–80 MPa, the energy content of compressed hydrogen is significantly less than that for the same volume of gasoline: 4.4 MJ L −1 (at 70 MPa) for hydrogen com-pared with 31.6 MJ L −1 for gasoline Even though considered to be quite simple and inexpensive, the high pressure of 80 MPa involved in hydrogen gas cylinders raises safety concern There is also some cost involved with compression to such high pressures Another consideration is the large pressure drop during use In addition, because most of the system parts exposed to hydrogen will be metallic, there is a concern of well-known hydrogen embrittlement [10]

The liquid hydrogen tank for its part offers almost twice as high storage capacity

by volume as pressurized hydrogen; however, this is still less than half that required

by the Department of Energy (D.O.E FreedomCAR goal) (the requirements will be discussed later) A major drawback of liquid storage is a big cost of liquefaction, which today can add as much as 50% to the cost of H 2 [6, 9, 10] There are also safety issues associated with the handling of cryogenic liquids and the problem of evaporative loss

Solid-state hydrides that include metal/intermetallic and complex (chemical) hydrides are characterized by the highest volumetric capacities, and they do not suffer drawbacks as those experienced by compressed and liquid hydrogen Because of the low pressures involved in metal hydride technologies and the fact that the release of hydrogen takes place via an endothermic process, this method of hydrogen storage is, however, the safest of all Moreover, the hydrogen released from a metal hydride is of very high purity and, therefore, can be used directly to feed a PEM fuel cell

The U.S Department of Energy (DOE) introduced a number of targets for board hydrogen storage systems within the frame work of its FreedomCAR pro-gram for the years 2007, 2010, and 2015 [11, 12] , which are listed in Table 1.2

It is now appropriate to discuss solid state hydrogen storage in hydrides in the context of the targets shown in Table 1.2

Table 1.1 Comparison of three major competing technologies for hydrogen storage (based on [9, 10] )

Storage system

Volumetric gen capacity (kgH 2 m −3 ) Drawbacks Compressed hydrogen gas

hydro-under 80 MPa pressure

are required; cost of pressurization; large pressure drop during use; hydro- gen embrittlement of storage tanks Liquid hydrogen at cryogenic

tank at −252°C (21 K)

safety; cost of liquefaction

Conventional metal hydrides based on metals such as V, Nb, Pd, Li, Na, etc

have gravimetric capacities too low for any commercial consideration in mobile

hydrogen storage with the exception of LiH, which has high capacity but

extremely high desorption temperature [13] A notable exception of a metal

hydride is Mg, which has a relatively high gravimetric capacity and can desorb

at 300°C and slightly below after nanostructuring treatment (it will be

dis-cussed later) In essence, none of the metal hydrides can meet the DOE targets

[13, 14] Similarly, hydrides based on intermetallic compounds AB (FeTi,

ZrNi), AB 2 (ZrMn 2 /TiMn 2 /TiCr 2 ), AB 5 (LaNi 5 or MmNi 5 where Mm –

misch-metal), and A 2 B (Mg 2 Ni) have relatively low gravimetric storage capacities as

shown in Table 1.3 , which are unsuitable for transportation storage although a

number of them desorb hydrogen within the temperature range targeted by

DOE at the desorption pressure ( P desorption ) of 1 atm (which is more or less an

operating pressure of a PEM FC)

However, there exist many complex hydrides having high and very high

gravi-metric storage capacities some of which are shown in Table 1.4 [13– 22] Their

theoretical capacity is calculated as the ratio of the atomic mass of hydrogen in

Table 1.2 US DOE FreedomCAR hydrogen storage system targets [11, 12]

Table 1.3 Hydrogen storage properties of intermetallic compounds (from [12] )

Maximum hydrogen capacity

the hydride formula to the molecular mass of hydride Some of these hydrides

called complex hydrides (like borohydrides) decompose in a multistage sequence

such as, for example, LiBH 4 that decomposes in the first stage into LiH + B + (3/2)H 2 and in the second stage LiH decomposes into Li and H However, only the first reaction that releases about 13.8 wt% H 2 (1.5 mol of H per 1 mol of LiBH 4 ) is potentially reversible [23] Therefore, the fourth column in Table 1.4 includes so-called theoretical reversible gravimetric capacity , which is the amount of hydrogen that is potentially feasible to be reversibly desorbed/absorbed from such a complex hydride Unfortunately, as shown in the last column in Table 1.4 the major problem of complex hydrides is that their desorption temperatures are not even close to the operating temperature range required by the DOE targets (Table 1.2 ) Some of the borohydrides such as Zn(BH 4 ) 2 that start desorbing around 80°C, close to the operating temperature of PEM FC, release a toxic borane gas B 2 H 6 together with hydrogen [24, 25] , which can quickly destroy a PEM FC membrane

Therefore, the major focus of research in the last decade is on finding the means to substantially reduce the desorption/absorption temperature of high-temperature complex hydrides and in addition to improve their absorption/des-orption kinetics where applicable As can be clearly seen in Table 1.4 only high-capacity hydrides are important for onboard hydrogen storage for vehicular applications As pointed out in [11] depending on the storage material and on the system design, material capacities may need to be a factor of up to two times higher than system capacity targets If such a rule is to be applied to the high-capacity hydrides in Table 1.4 than vis-à-vis extremely restrictive DOE targets

Table 1.4 Hydrogen storage properties of selected high-capacity hydrides [13– 22]

Metal–

hydrogen

Theoretical mum gravimetric

maxi-H 2 capacity) (wt%)

Theoretical ible gravimetric capacity (wt%)

Approx tion temperature range (°C)

in Table 1.2 most of the high-capacity hydrides should not be even considered

as on-board storage materials for PEMFC vehicles This of course is very

unrea-sonable and it seems that sooner or later the DOE targets must be updated to be

in line with the reality The second parameter of an utmost importance is

desorp-tion/absorption temperature In reality, it may be extremely difficult to find a

material desorbing in exactly the same temperature range as required by the

DOE targets in Table 1.2 However, materials with higher desorption

tempera-tures cannot be completely ruled out of hand as viable storage media First, there

is a recent trend to increase the working temperature of the PEM fuel cell by

using an electrolyte membrane of polybenzimidazole (PBI) doped with

phos-phoric acid [26] Second, a nanostructured MgH 2 with a catalyst that reduces

desorption temperature to below 300°C could be used on board in a recently

proposed two-stage reservoir [27]

Another very restrictive and debatable DOE requirement for onboard

hydro-gen storage is the reversibility of solid state hydrides (e.g., onboard refueling

with the target rate of 1.5 kgH 2 min −1 for 2010 as shown in Table 1.2 ) However,

as pointed out by Sandrock et al [28] it is an immense problem to remove the

exothermic heat at that charging rate For example, if we charged H 2 at 1 5 kg min −1

into a vehicular storage tank based on NaAlH 4 ( Δ H = 37 kJ mol −1 H 2 ), we would

have to remove heat at the rate of 450 kW! This would require very substantial

and costly heat exchangers and in practice would completely eliminate the

pos-sibility of quick onboard recharging Another point of importance is an

enor-mous cost attached to building an entire infrastructure of a hydrogen gas

refueling station such as steel piping, vessels, etc Last but not least is the

hydro-gen embrittlement phenomenon in which various metals, such as high-strength

steel, aluminum and titanium alloys become brittle and eventually crack under

load following exposure to hydrogen This process could be a problem for

hydrogen steel piping and other components of a hydrogen gas refueling station

In our opinion much easier solution is off-board recharging in which the depleted

solid state hydride container is repleted with nanostructured solid state storage

materials manufactured in a dedicated plant as schematically shown in Fig 1.1

According to this new vision, the refueling station is just a retail station for

hydrogen storage containers where refueling could become eventually a fully

automated process reduced to a quick replacement of containers without any

complicated infrastructure This could also lead to the growth of new businesses

for the Hydrogen Economy Also, an off-board recharging solves the problem of

chemical reversibility, which creates a difficulty for a number of hydrides

other-wise quite attractive as storage materials Ball milling can be used to induce a

mechanical reversibility carried out in a ball milling device instead of applying

high pressure and temperature for the chemical synthesis Such hydride storage

technology based on mechanochemical reactions and nanostructuring processes

conducted in high-energy ball mills constitutes the core of this book Alternative,

nonhydride storage, as exemplified in carbon-based nanomaterials, is discussed

in the second part of this book.

1.2 Brief, Synchronic History of Development of Hydrides and Nanomaterials

1.2.1 Early Investigations of Metal–Hydrogen Systems

and Hydrides

Although not considered to be stored and delivered from metal-hydrogen systems, hydrogen has been used for energy generation since the 1800s Being a major

constituent, up to 50 vol.%, of so-called syngas , synthetic gas manufactured by

gasification of coal, wood, or waste, it was widely used in homes in Eastern USA

and Europe from circa 1850 until the Second World War The water gas reaction utilized in production of syngas was the first in the man’s endeavor to generate energy from hydrogen , the most energetic and the most abundant ingredient of our

geosphere and biosphere

Fuel cell car (Neckar, Daimler Chrysler) or hydrogen internal combustion engine

Hydrogen refuelling station* Depleted container Container with

At about same time when the hydrogen-rich syngas had been combusted for

electricity generation in gas-fueled electric plants, some studies driven by pure

scientific curiosity were conducted, which led to observation of a direct conversion

of hydrogen chemical energy into electrical energy Sir Humphrey Davy (1802)

experimented on a cell that utilized a carbon anode in aqueous nitric acid, and

Sir William Grove built a gaseous voltaic battery, what is considered the first hydrogen

fuel cell Obviously, flameless burning of hydrogen in such a fuel cell as to generate

electric current was not in the ranks with direct combustion of hydrogen in syngas

for generation of electric energy in just launched electromagnetic generators

Obviously, this was also long before concerns have become raised about pollution

of air and the change to the environment and the climate brought by burning

coal-bearing fuels for heat, electricity, and mass transportation Simply, both the

eco-nomical drivers and the technology to directly convert hydrogen energy into electric

energy were not in place

Some 100 years after Cavendish’s discovery of hydrogen, and only 3 years after

it was realized that hydrogen sorbed from chemical or electrochemical sources

causes blistering and embrittlement to steel vessels, Graham [29] observed the

abil-ity of palladium to absorb hydrogen and wrote in Philosophical Transactions of the

Royal Society of London:

“Hence palladium has taken up a large volume of gas when the temperature of the metal

never exceeded 245°C,” and again

“1 vol palladium held 643.3 vols hydrogen

By the care of my zealous assistant, Mr W.C Roberts, the hydrogen employed in these

experiments was purified to the highest degree by passing it in succession through alcohol,

water, caustic potash, and tubes of 0.7 meter each, filled with broken glass impregnated

with nitrate of lead, sulphate of silver, and oil of vitriol The gas was inodorous, and burned

with a barely visible flame.”

The reversible absorption was observed to proceed in presence of either metallic

palladium or in palladium–silver alloys; much less hydrogen was absorbed in Cu

sponge (1 vol Cu: 0.6 vol H.), and not at all in Os–Ir The reaction of hydrogen

with palladium, so being described by Graham, was:

Palladium hydride is not a stoichiometric chemical compound but simply a metal in

which hydrogen is dissolved and stored in solid state, in space between Pd atoms of

crystal lattice of the host metal Relatively high solubility and mobility of H in the

FCC (face-centered-cubic) Pd lattice made the Pd–H system one of the most transparent,

and hence most studied from microstructural, thermodynamic, and kinetic points of

view Over the century that followed many metal-hydrogen systems were investigated

while those studies were driven mostly by scientific curiosity Researchers were

inter-ested in the interaction of hydrogen molecule with metal surfaces adsorption and

diffusion into metals Many reports on absorption of H 2 in Ni, Fe, Ni, Co, Cu, Pd, Pt,

Rh, Pd–Pt, Pd–Rh, Mo–Fe, Ag–Cu, Au–Cu, Cu–Ni, Cu–Pt, Cu–Sn, and lack of

absorption in Ag, Au, Cd, Pb, Sn, Zn came from Sieverts et al [30– 33]

At the time when Sieverts et al reported first studies on absorption of hydrogen

by metals and alloys, a series of seminal papers had been published by Avrami [34– 36] on the kinetics of phase transformations driven by nucleation and growth

of nuclei and leading to microstructural granulation , viz grain size distribution in

solid material Therefore, foundations for hydrogen storage and for formation of

fine and ultrafine-grained microstructures in solid materials had been developing

concurrently

At that time when world emerged from the Second World War, and became falling into another Cold War with the Soviet Union, liquid hydrogen has been intensively studied as a rocket fuel In contrast to the dangers of a new world deal, and in the climate of overwhelming optimism triggered by the end of war, hydrogen was also contemplated as the fuel of the future for a civil supersonic aircraft flying between Europe and North America, with Canadian Montreal’s supermodern Mirabel airport being considered for accommodating overseas flights from Europe

After World War II, investigations of hydrides were driven mostly by nuclear

reactor applications as to understand hydride-caused embrittlement of reactor

met-als, such as Zr, or to take advantage of high populations of hydrogen atoms in hydrides to scatter, moderate, or shield from energetic neutrons in high-temperature, mobile nuclear reactors Zirconium alloys in water-cooled reactors, in particu-lar, were known to pick up hydrogen, and precipitate zirconium hydrides on cooling Therefore, zirconium–hydrogen system was the one studied in close reference to development of nuclear reactors The first intermetallic hydride reported was ZrNiH 3 [37]

On historical milieu we would like to bring attention of our readers to a curious

synchronic development of the discipline of hydrides and a new discipline of rials science and technology Concurrently to the reporting of first intermetallic

mate-hydrides came postwar major advances in examination of microstructures in metal alloys and compounds First, it was an appropriate interpretation of data from X-ray diffraction for crystalline and polycrystalline materials and realization that line broadening in the diffraction pattern can be caused by both small grain sizes and/

or internal strains, while the separation of these two contributions can be achieved

using plots for Williamson-Hall method [38] Then came the first edition of a

clas-sic book on X-ray diffraction procedures for polycrystalline and amorphous

materi-als (term nanomaterimateri-als was still not coined) [39] The late 1960s brought the Rietveld method for profile refinement of a whole diffraction pattern, composed of

several phases distinct in both chemical and grain size characteristics [40] A new dislocation theory and a transmission electron microscopy were applied in studies

of dislocations and dilatational misfit of zirconium hydrides precipitated in Zr metal matrix [41] These early metallographic investigations quite often were giving evidence to formation of nanometric precipitates, as metastable hydride phases in metals under nonequilibrium conditions imposed either by high dose of ionizing radiation or fairly rapid cooling (>10 °C s −1 )

Now, let us come back to our other historical thread: the development of hydrided materials Large body of results was coming in the 1960s from research

on apparently nanometric structures observed in fine metal powders and deposits,

like dispersed Fe, and cathodically charged Ni layers prepared in electrochemical

cells [42] Yet, the thermodynamic information extracted from such

electrochemi-cally charged metals concerned inherently nonequilibrium processes, and was

considered doubtful and less of value In fact, an important experiment on direct

synthesis of nickel hydride, NiH, at 18 kbar of H 2 gas in a high-pressure cell,

car-ried out by Baranowski in Poland’s Institute of Physical Chemistry in Warsaw in

1966 [43] was driven by the desire to achieve true equilibrium conditions between

H 2 gas and solid nickel hydride phase At that time a common approach in

investi-gations of metal–hydrogen systems was to conduct studies on well-ordered

poly-crystalline metals or, if possible, single crystals Obviously, ordered metals used in

experiments let investigators to construct the most transparent and instructive models

for a metal-hydrogen system

Although by the end of 1960s many metals were reported to form hydrides,

those were either too stable, like ZrH 2 , or too unstable, as NiH to attract interest in

respect to reversible hydrogen sorption

1.2.2 Early Routes to Nanomaterials

There is a strong interplay between science and technology New science often

cre-ates new technological opportunity, and reciprocally, new technology development

often triggers new opportunity to advance science Such interplay is well illustrated

in the tread in which the science of new materials for hydrogen storage has always

been strongly intertwined with the parallel development of innovative materials

processing techniques and technologies Initially, those were techniques of

high-pressure physics and chemistry, and instrumental techniques in electrochemistry

Later development was fueled by an emerging discipline of materials science The

advent of nanomaterials , as one of a few founding blocks of future nanotechnology ,

came from materials science laboratories around the world Nanotechnology is a

technology that owes its name to the prefix nano , a Greek word for dwarf , as

applied to objects that exhibit billionth (10 −9 ) meter dimensions The first reports on

unusual new metallic phases, where observed structures exhibit partitioning into

nanosized grains, or are simply grainless, have been coming from groups in the

USA and other countries worldwide A new process of rapid quenching of metals,

with the cooling rate of 10 6 K s −1 , first applied by Pol Duvez and his colleagues [44]

in the California Institute of Technology in Pasadena in 1960, produced unexpected

new fine-grained and grainless, metastable alloy phases, the first nanometals and

amorphous metals Using a quench technique capable of cooling metal melts to

ambient temperatures with such extraordinary cooling rates, through just spraying

and splashing of milligrams of melt alloys on a chill surface [sic!], the process of

nucleation and growth, as described by Avrami 20 years earlier, was kinetically

bypassed to yield a configuration of frozen liquid or amorphous metal Duvez et al

realized this possibility by reporting complete solid solubility in Cu–Ag and

Ga–Sb–Ge systems and formation of new nonequilibrium phases in Ag–Ge and

Au–Si – these being the first nanocrystalline alloys and the first amorphous alloys

Cohen and Turnbull [45] were quick to point out that amorphous alloys can exhibit glass-to-crystal transition Indeed, when processed by low-temperature annealing

they formed first metal alloys with nanometric grains, hence, the first nanoalloys

The new science of amorphous and noncrystalline solids would not have developed

so rapidly if Chen and Polk had not invented useful amorphous ferromagnetic rials and this technological breakthrough had not been vigorously pursued by the big US company, the Allied Chemical Corporation (later Allied Co) Indeed, in

mate-1972, a new chapter in metallurgical technology has been opened, when a new

rapid quenching process of melt spinning , viz., casting of a stream of molten metal

on a copper drum, rotating with speeds ranging from 5 to 20 m s −1 , was used by Chen and Polk that worked for the Allied Chemical Corporation in USA [46] to spin the first noncrystalline ferrous (and ferromagnetic) metal ribbon, Fe 80 B 20 Further to this a new process to form alloy ribbons with structures ranging from nanometric (then called microcrystalline) to amorphous was invented by several researchers, among them Bedell, Narashiman, and Ray working for corporate research in companies in the USA It came as a shock to many physicists that amor-phous iron alloy can be ferromagnetic in spite of absolute lack of crystalline lattice (and indeed, absence of a theory of amorphous magnetism) With the invention of

a new class of soft magnetic materials it became clear that applications could be found that will utilize the outstanding combination of physical and chemical prop-erties as being the direct consequence to the lack of structural crystalline long-range order (LRO), and the presence of short-range order (SRO); the latter being lim-ited to nanometric distances between atoms Among those new materials were binary Fe–B, Fe–Zr, Cu–Zr, Ni–Zr, Pd–Si, Mg–Zn, and many ternary or even quater-nary amorphous systems Many patents were filed in the USA and Japan, and by many groups worldwide, on amorphous alloys and intermetallics that were termed

metallic glasses or glassy metals [47] and had been manufactured by a new rapid solidification technology This technology became intensively developed by MIT’s

Center for Materials Science and Engineering in the USA [48] and Japan’s Tohoku University, among many other materials research laboratories worldwide

From the historical point of view it is important to realize that many if not all of these amorphous alloys, i.e., metallic glasses, when carefully annealed at low tem-

peratures, change to devitrified , nanostructured alloy phases Those were the first nanomaterials engineered on purpose They were produced in form of thin ribbons

and wires via rapid solidification processing of melt alloys They allowed led exploration of chemical, mechanical, magnetic, and other properties as arising from nanostructures Shortly, the most prominent benefits of new nanostructures were achieved in the outstanding improvement in the energy product of new neo-dymium supermagnets produced by crystallization of amorphous Fe–Nd–B alloys (as reviewed in [49] ) Shortly after, among research stimulated by the outstanding combination of chemical and magnetic properties in metallic glasses were the first papers on the action of hydrogen on amorphous intermetallics and magnetic metallic

control-glasses Absorption and diffusion of hydrogen in amorphous alloys was studied in

UK by Harris [50] and by Cantor with his colleagues [51] , in Germany by

Kirchheim et al [52, 53] , and in the USA by Bowman [54] , among few others In

Canada, Wronski et al [55] studied hydrogenation of F 90 Zr 10 amorphous

interme-tallic via cathodic polarization, and concluded that hydrogen can be reversibly

released by low-temperature annealing at less than 100°C The hydrogenation was

accompanied by a concurrent drastic change of magnetic properties, from the

para-magnetic nonhydrogenated phase to the ferropara-magnetic hydrogenated amorphous

solid solution ; during those transformations, the F 90 Zr 10 remained amorphous in its

bulk, yet Fe magnetic nanocrystals precipitated on the surface layer of the

hydro-genated amorphous F 90 Zr 10 [56] In the USA, Johnson and colleagues in Caltech

observed quite the opposite: the first example of a hydrogen-induced

crystal-to-glass transformation in metallic alloys resulting in the formation of amorphous

hydrides [57] :

0.75 0.25 2 0.75 0.25 1.14

Reaction (1.3) proceeded only at a relatively low temperature just above 150°C and

although amorphous alloy exhibited large H/M (hydrogen-to-metal) atom ratio, the

charging and discharging of amorphous phase proceeded in a range of pressures,

while a crystalline alloy could have been cycled at almost constant pressure [58]

Bowman [59] explained that the strong exothermic reactions between hydrogen and

most amorphous intermetallics release excess energy, which causes crystallization

of amorphous phase He also reported that many metallic glasses tend to absorb less

hydrogen then their nanocrystalline counterparts; he explained this behavior by the

random order of atoms in amorphous phase, which might restrict the number of

interstitial sites that are favorable for hydrogen occupancy

While interest in amorphous alloys was becoming vane over the time, the

para-digm shift in materials science brought about by the discovery of amorphous metals

and rapid solidification cannot be underestimated The interest of physicists,

chem-ists, materials scientchem-ists, and engineers became refocused from well-ordered

crys-talline materials to disordered and nanocrystalline phases Shortly, substantial

R&D programs were initiated in the USA, Japan, and worldwide both at

universi-ties and national government laboratories Foundations have been established for

the advance of the science of nanomaterials

However, the transition from nanoscience to nanotechnology had to come from

yet a concurrent innovation in tools used by scientists This was the invention of the

first scanning tunneling microscope (STM) in 1981 [60] , followed by the invention

of the atomic force microscope (ATM) in 1986 [61]

While STM and ATM microscopes provided powerful tools to advance new

science of solids at nanometric region materials scientists were also greatly

inspired by papers coming from Gleiter and his group at Saarbrücken University

in Germany Those were the papers that redefined nanocrystalline materials as

such where the fraction of atoms located at grain boundaries of polycrystalline

material is comparable to the fraction of atoms at the core of grains when the grain

size reaches nanometers [62] A claim was put forward that such materials are to

be fundamentally different from, and often superior to, those of the conventional

polycrystals and the amorphous solids Gleiter observed that nanometer-sized crystalline materials being polycrystals with very small crystallite sizes of about 2–10 nm in diameter are composed of randomly oriented high-angle grain bound-

aries The first such nanocrystalline phases produced on purpose came from

Saarbrücken about 1984 by evaporation of the material in a high-purity inert gas atmosphere followed by condensation and compaction in ultrahigh vacuum [63]

By this time chemists working in the field of metal catalysts were well aware of importance of nanometric features, such as edge sites and surface clusters on chemical catalysis Molecular orbital calculations performed 10 years earlier predicted that the ionization potential of nanostructured metals should increase as their grain size decreases [64] The high surface-to-volume ratio observed in nanostructured materials not only appeared important to the number of active sites in a catalyst, but also seemed to influence the defect chemistry (like oxygen and other anion defects) Siegel [65] computed that the percentage of metal atoms

on the surface of grain increases from a few percent in a 100-nm particle to about 90% in a 1-nm crystallite These simple experiments by Gleiter and equally sim-ple calculation by Siegel were eye opening for others in research community in material science Immediately, a new consensus reached among materials scien-tists was that extending our understanding of structure–property relationship in solid materials down to the nanometer regime should be attractive for develop-ment of engineering materials with an outstanding combination of properties or novel properties Among materials that became studied were nanophases pro-

duced by mechanical alloying [66] and nanohydrides where the size effect

deter-mines hydrogen storage properties [67]

1.2.3 Historical Development of Classical Hydrogen

Storage AB 5 Alloys

Development of first practical hydrogen storage alloys, AB 5 -type intermetallics, had its beginning in a typical accidental laboratory experimentation, although it was predetermined by a vigorous development of a new discipline of materials science and engineering in late 1960s The outstanding hydrogen sorption properties of rare-earth AB 5 intermetallics were accidentally discovered in Philips Laboratories

in Eindhoven, Netherlands at about 1969 in a program to develop a new permanent magnet alloy [68] In the work on Sm–Co magnet alloy it was observed that a loss

of coercivity occurs, when magnets were aged in humid air, which was related to two concurrent reactions: corrosion of SmCo 5 intermetallic with release of hydro-gen, and sorption of the released hydrogen by still uncorroded alloy:

Zijlstra et al [69] demonstrated that the reaction of (1.5) is reversible at several bars

of pressure at ambient temperature In the following studies on the origin of

mag-netic coercivity in these new magnets, Phillips laboratories came with the hydride

of LaNi 5 , lanthanum–nickel hydride , which reversibly bonded more than six atoms

of H per one formula unit (H/M ratio > 1) [70]

Reversibility of hydrogen charging was very good at room temperature This is

because the hexagonal lattice of the metal host, like this of LaNi 5 , does not undergo

major transformation as hydrogen is inserted interstitially This was the first AB 5

-type interstitial hydride in which hydrogen is stored between metal atoms

Those rare-earth AB 5 -type hydrides were quickly utilized in rechargeable nickel

metal hydride batteries where electrochemical hydrogen chargin g and discharging

take place at ambient temperature Such electrochemical hydrogen storage is

reversible, when the negative hydride electrode (anode) is combined with the positive

Ni electrode (cathode) in the battery cell

As a matter of fact, the first hydrides with practical hydrogen storage capacities

were realized in rechargeable nickel metal hydride batteries For more information

on electrochemical hydrogen storage in rechargeable batteries a reader can be

referred to several recent reviews on this subject [71– 73]

Renewed interest in solid-state hydrogen storage came with first Oil Crisis in

1970s By early 1973 a patent was granted to US Brookhaven National Laboratory,

where Reilly had produced first MmNi 5 alloy for hydrogen storage [74] MmNi 5 ,

where Mm – mischmetal – is a unrefined (cerium free) mixture of rare-earth metals,

mostly La and Nd, obtained from Bastanite from California (also found in large

deposits in Inner Mongolia, China), was much less expensive than La and has already

been used as a deoxidizer in steel industry Soon after, the potential for energy-related

application for nickel and nickel alloys attracted the major Ni metal and alloy

pro-ducer in Canada and the USA, INCO Garry Sandrock, then with INCO, achieved

optimization of MmNi 5 compositions for gas hydrogen storage through careful partial

substitutions for A and B in AB 5 formula [75] The compositional changes have

reduced the raw materials costs to about 30% of that for the LaNi 5 alloy

LaNi 5 has an attractive value of the equilibrium hydrogen desorption pressure

dropping in the range between 1 and 2 atm at 25°C In the MmNi 5 this pressure

rises to a much less attractive 30 atm, which can imply difficult engineering design

for gas hydrogen storage, and was completely impractical in MmNi 5 used for the

anode in a NiMH cell; obviously, battery cells operate at near-atmospheric

pres-sure The substitutions of MmNi 5 with Ca for Mm and Al for Ni were quite

effec-tive in lowering the equilibrium pressure of hydrogen back to about 1–2 atm at

25°C, which was desired for easy hydrogen charging and discharging at ambient conditions Approaches along the same line came in the same time from Japan [76] The alloying of polycrystalline AB 5 alloy was an excellent (and very practical) result of new materials design involving adjustment of the strength of chemical bonds between metallic elements in AB 5 hexagonal lattice and its interstitial hydro-gen atoms The bond strength in the metallic host lattice was further optimized with cosubstitutions, which was effective in reducing the volume change on hydriding

of the hexagonal AB 5 unit [77] ; the density of material can, indeed, be related to the

bond lengths through an approximate formula: R o = 0.0115 A av / a 2 c , where A av is an average atomic weight for the AB 5 formula unit, and a and c are the hexagonal lat-

tice parameters in nanometers [78] The lower dilatation of the AB 5 cage on placing hydrogen in it was equivalent to better cohesiveness of atomic structure in this intermetallic, hence improved resistance to alloy decrepitation on repetitive (cyclic) charging and discharging, hence better corrosion resistance of MmNi 5 alloys in an aggressive caustic battery electrolyte

Although substitution with Al came at the price of the lowered hydrogen capacity, another Mn substitution ameliorated this problem [79] With compliance with the well-observed 10–15-year shift from new material development to its full commer-cialization, the Al and Mn substitutions have been combined in multielement MmNi(Co, Mn, Al) 5 battery alloy and optimized for the use in anodes in NiMH commercial batteries According to recent review a typical commercial AB 5 alloy can consist of many elements, such as in hydroalloy F from GfE Metallen und Materialen GmbH: La 0.64 Nd 0.36 Ni 0.95 Cr 0.19 Mn 0.41 Co 0.15 (in wt%) [71] Again, one can observe how the development of materials for gas hydrogen storage for fuel cells has been intertwined with a parallel development of alloys for electrochemical hydrogen storage in rechargeable battery cells AB 5 alloys used in either hydride

battery cell or in a hydrogen storage tank for fuel cell are not per se nanocrystalline

but the microstructural design process, which led to their development in 1980s and 1990s, reached nanoscale phase partitioning

1.2.4 Historical Development of Interstitial Hydrides

in Other Intermetallic Systems

The first systematic research into gas hydrogen storage for practical applications began in the early 1970s at the Brookhaven National Laboratory (BNL) on New York’s Long Island The investigations were conducted in a program that explored possibility of storing electric energy produced in off-peak hours by nuclear power stations through generation of hydrogen, and then the storage of hydrogen in metals instead of storing electric current So, at the same time as the AB 5 -group hydrides

were intensively investigated in the BNL, new AB-type hydrides came into view Iron–Titanium , FeTi, hydride was discovered in 1974 by Reilly and Wiswall [80] in

the same Brookhaven National Laboratory This metal hydride formed from a 50–50 atom ratio of titanium and iron, the lowest cost and the most convenient

hydride, offered a reversible capacity of almost 1.5 wt% H 2 operating at ambient

temperatures A desirable feature of this new hydride was that the hydriding–

dehydriding reaction occurs near ambient temperature and during the charging

pressure ranges between 500 and 15 psi The prototype 400-kg FeTi hydride

stor-age unit was built by a New Jersey’s electric utility The charge–discharge reactions

were observed to be very rapid but strongly limited by the rate of heat transfer FeTi

material was not losing hydrogen storage capacity even after thousands of cycles

when high-purity hydrogen was used Yet, the alloy was very sensitive to O 2 and

CO poisons, and even to traces of H 2 O Obviously, while meeting the expectations

of stationary storage of nuclear energy in the form of solid hydride it was too heavy

for storage of hydrogen for transportation energy Anyway, the needs for cars

pro-pelled by hydrogen fuel were not in place at the end of the 1970s

1.2.5 Historical Development of Nanophase AB 2

Intermetallic Hydrides

A new venue for hydride research has been evolving from the study of multielement

Ti–Ni alloy thin films for hydrogen oxidation in fuel cells conducted by Stanford

Ovshinsky in the USA in the early 1980s The new disordered multielement,

mul-tiphase AB 2 -type alloys for hydrogen storage exhibited a large degree of structural

disorder and submicron partitioning of phases Ovshinky has been pioneering a

view [81] that structural disorder at nanoscale results in outstanding combination of

hydrogen storage in crystalline grains, while the disordered grain-boundary phase

provides channels for hydrogen delivery and release from nanocrystalline grains A

first, quintessential tool of the coming nanotechnology , a scanning tunneling

micro-scope, was then used to reveal that in the TiVZrCrNi multiphase alloy the hydrogen

is preferentially sorbed in the Ti-rich phase, while V-rich phase was likely to

cor-rode preferentially to expose Ni nanoparticles [82] This shift of view was

congru-ent with enormous intellectual potcongru-ential, which was brought about by research on

amorphous and rapidly quenched phases (amorphous selenium, chalcogenide

glasses, and metallic glasses) in the 1960s and 1970s US Ovonics patents came just

in time when interest has been growing in Europe, Japan, and North America in

chemical properties of amorphous metals and metallic glasses prepared by rapid

solidification technology The materials challenge was to optimize hydrogen capacity

and corrosion properties in multiphase nanostructures for the anode in nickel metal

hydride (NiMH) cell Shortly, the outcome of international race began toward

development of a second generation of multiphase and multielement NiVZrTiCr

hydrides The nanostructured battery materials were developed applying new

prin-ciples of materials science established in previous decade and reflecting on the role

of disorder and nonequilibrium processing in overcoming the conflicting

require-ments for materials properties In the early 1990s sealed nickel metal hydride

bat-teries became available in Japan and elsewhere for consumer goods Shortly

afterward, the new nanopowders, which were proven in electrochemical hydrogen

storage, began to be investigated for gas hydrogen storage This came when new

drivers emerged for growth of hydrogen energy in the 1990s The concern now was for clean environment, and mitigation of greenhouse gases, and appeared in the context of evidence for accelerating global climate change

1.2.6 New Routes to Nanomaterials: Mechanical Alloying

and Mechanochemical Activation

Like the discovery of nonequilibrium quenching of metal melts in the 1960s and 1970s led to the development of disordered multiphase AB 2 hydrides in the 1980s,

an interest in nonequilibrium milling of metal powders, in the 1980s, has been

set-ting foundations for the development of the nanohydrides of the following decade

The nonequilibrium synthesis of new metal alloys by high-energy ball milling of powders was developed almost 20 years earlier by John Benjamin and his coworkers

at the International Nickel Company (INCO) [83] The development came from research on new alloys intended for gas turbines, and was an outcome of an effort to produce nickel superalloys, which were mechanically hardened through an inocula-tion with fine refractory metal oxides, Al 2 O 3 , Y 2 O 3 , and ThO 2 , in so-called oxide dispersion-strengthened (ODS) process In his research Benjamin pointed out that

mechanical alloying ( MA ), a new process induced by intensive milling of metal and

compound powders, could make new materials with unique properties [84] (In a

historical note like this, the credit for coining the term mechanical alloying must be

given to a Patent Attorney for INCO company.) However, it was not until the early 1980s that the discovery by Koch and coworkers [85] of amorphous alloys made by

MA renewed interest in intermetallic compounds, which were difficult to synthesize

by orthodox metallurgy Those authors demonstrated that, during mechanical alloying

of elemental metal powders of nickel and niobium in a special high-energy ball mill (SPEX™ model), the final product becomes an amorphous alloy This was demon-strated by the complete disappearance of the sharp Bragg reflections in X-ray diffrac-tion pattern, which are characteristic of any crystalline phase, and appearance of a broad maximum, characteristic of amorphous alloys; the latter was previously seen in metals that were rapidly quenched from melt The amorphous structure was con-firmed in the exothermic glass-to-crystal transformation peak revealed by differential scanning calorimetry (DSC) Two years before the publication by Koch, Yermakov

et al [86] reported on the amorphization of a number of Y–Co intermetallic pounds by grinding of ingots, which were prealloyed by casting This process was

com-later termed mechanical milling (MM) process Further early investigations were

car-ried out by Schwarz et al in the US Los Alamos National Laboratories [87] , as well

as by groups in Germany and the Netherlands Already, in these early stages of new research, it was well recognized that by ball milling the average size of grains in powder particulates was reduced to nanometric dimensions

Either MA or MM processes drop under more general class of solid-state phization reactions , SSAR Amorphization by irradiation of solids was observed yet

amor-in the era of study of materials for nuclear reactors In 1962, Bloch [88] amorphized

U Fe by exposing it to fluxes of nuclear fission fragments Others observed amorphization

of intermetallic compounds by high-voltage electrons [89] Schwarz and Johnson

[90] demonstrated the phenomenon and a new process of amorphization by diffusion

between thin-film sandwiches of crystalline La and Au In the late 1970s, Johnson

and coworkers [57] in Caltech demonstrated that in addition to techniques of

amor-phization by rapid solidification of molten alloys and amoramor-phization by mechanical

alloying of solid alloys, the process can be conducted by repetitive exposure of an

alloy to hydrogen pressure and vacuum This provided a new solid-state processing

route to amorphization by hydrogenation To this class also belong reactions of

hydrogen decrepitation of metals and alloys, the method proposed by Harris [50] for

preparation of fine powders for subsequent sintering into new generation of

neodym-ium supermagnets Through cycling of a cast metal ingot of Fe–Nd–B in hydrogen

the ingot was effectively changed into fine powder; however, nanometric hydride

phases that must have formed concurrently to the decrepitation were not considered

for hydrogen storage So it took another thread in materials chemistry history tracks

to bring mechanical processing for advancing research on hydrogen sorption in solid

media This thread can be attributed to studies of effects of mechanical grinding on

activation of inorganic solids for chemical reactions

As a matter of fact, the approach to apply mechanical energy, in form of grinding

and milling, as to activate or bring about new paths for chemical reactions was much

older In a historical review on mechanochemistry , as a branch of solid state chemistry

dealing with mechanically activated chemical reactions, pioneering role is attributed to

photochemist M Carey-Lea who as early as 1892 [91] began publishing reports of

systematic studies on decomposition of silver and mercury halides, then dozens of

other stable compounds, under applied mechanical pressure, using so simple a method

as grinding compounds in a mortar Extracting mercury from its sulfide while rubbing

it in a copper mortar was known yet in ancient Greece [92] However, in modern times

it was a large body of work on mechanochemical activation of inorganic compounds

that was coming from laboratories in Russia and the USA that demonstrated the role

of defects in activation of materials for chemical reactions conducted in a solid state or

at the interface of solids and gases Researchers from the US Naval Civil Engineering

Laboratory engineered mechanical alloys as composites where the second phase of Fe

(or Ni, Ti, or Cu) has been dispersed at nanometric manner in the Mg phase; these

alloys were intended as supercorroding alloys to react rapidly with seawater to

pro-duce heat and hydrogen gas in underwater welding of naval vessels [93] Shortly after,

in years 1984–1997, Ivanov et al [94] from Russia were the first to explore mechanical

alloying to obtain magnesium alloys for hydrogen storage

1.2.7 Historical Development of Lightweight Metal Hydrides

and Hydride Complexes

Formation of magnesium and calcium hydride compounds was first recognized by

German chemists yet at the end of nineteenth century [95] but it took more than a

half a century before a substantial yield of Mg hydride was obtained from direct

synthesis of the elements and the first determination of values for the tion pressure on temperature, the enthalpy of formation, and the activation energy were obtained [96] Since then, Mg metal with its hexagonal lattice was well researched for hydrogen storage.

Magnesium dihydride can store several times as much hydrogen per unit weight

as AB-type TiFe hydride However, the first and the most limiting problem with lightweight Mg alloys for hydrogen storage was always the very poor kinetics of hydrogenation Indeed, metallic magnesium does not react with hydrogen at low temperatures, and reacts only very slowly at high temperatures Because of the relative inertness with respect to hydrogen, Mg was even recommended as a structural metal suitable for contact with hydrogen up to the melting temperature [sic!] Not surprisingly, many controversial and poor reproducing data have been published on the hydrogenation/dehydrogenation kinetics The natural inertness

of Mg to molecular hydrogen can be attributed to well-known factors: (1) thick surface oxides that cover magnesium metal do not provide sites catalytically active for dissociation of hydrogen molecule, which is known to occur on contact with a metallic surface (clean metallic surface often provides active sites for catalytic breaking of very strong H–H bond), (2) surface-adsorbed impurity gases further slow the kinetics of hydrogen absorption into the bulk grains (the process that may be related to the hydrolysis of Mg with water vapor and causing even more thick oxide/hydroxide surface barriers; Sect 2.1.5) On the other hand hydrogen dissolved in magnesium metal is readily removed by heating it in vacuum up to 300–400°C The stable Mg dihydride was prepared as early as 1951 by direct heating of Mg in gaseous hydrogen under high pressures; however, the reaction was slow and difficult to complete [97] While the chemical reactivity of Mg metal to hydrogen is poor it was early realized in works by Dymova et al [98] that its dihydride can be a strong chemical reductor, and its increased reactivity depends strongly on small grain size So the effect of grain size was pointed on

as being one of high importance

Mg 2 Ni alloys and magnesium–nickel A 2 B-type complex hydrides , Mg 2 NiH 4 were well known, being first developed in the US Brookhaven National Research Laboratory [80] , and studied in Nordic countries [99] , where the interest in Mg has been stimulated by availability of hydropower required to produce Mg metal

in the first place (Even earlier, Pebler studied A 2 B-type hydride of Zr 2 Ni [100] ) The new Mg 2 Ni alloy alleviated the problem inherent to Mg metal: very poor kinetics of hydrogenation reaction Reilly and Wiswall [80] noticed that by the presence of intermetallic magnesium compounds, Mg 2 Ni and Mg 2 Cu, on the Mg surface this reaction can be greatly accelerated The hydrogen release can pro-ceed without major crystallographic transformation as the hydride is converted back to the Mg 2 Ni:

The next alkali earth in the periodic table is Ca and its ternary hydride was found to be

less stable than the Mg ternary hydride Indeed, the desired reaction for CaNi 5 is:

and this reaction was observed to work repeatedly and reversibly at near room

temperature Yet, the following disproportionation reaction is a preferred reaction

path as long as the temperature is slightly increased (to 200°C) and the diffusion of

the metal atoms becomes significant in comparison to the diffusion of smaller H

atoms The reaction product observed in the disproportionation was the mixture of

more stable dihydride and metallic nickel:

Although in later studies most of the intermetallic compounds were found to be

thermodynamically unstable and falling into the disproportionation reaction path

easily their actual tendency to do so varied markedly from system to system, as

later it was learnt from studies on transition metal ternary hydrides Many of these

hydrogen-storing hydrides became available by the end of 1970s, most of them

under the trademark of commercial HY-STOR hydrogen storage alloys : 300 Series

(Mg-base), 200 Series (Ni-base, including both CaNi 5 -type and MmNi 5 type), and

100 Series (FeTi-based)

The first and the most important alloy, for which the H storage properties clearly

benefited from nanostructure and nanoprocessing, was Mg 2 Ni, an A 2 B-type

interme-tallic Although yet in 1961 Dymova et al [98] pointed to the small grain size as a

primary factor in improving hydrogen sorption in Mg, and other Russians, Ivanov et

al [94] , demonstrated the high-energy ball milling route to hydrogen storage in

magnesium, it took another four decades and series of studies reported from the

McGill University in Canada to bring again the benefits of fine grains to the attention

of the community of researchers in hydrogen storage The clear demonstration of the

effect of nanostructuring came in works of Zaluskis’ [101] A study of ball milled

powders showed that when the smallest X-ray diffracting metal grains in milled

particles reach the range 5–50 nm the kinetics of both absorption and desorption is

improved by an order of magnitude These were nanostructured hydrides or

nano-hydrides It must, however, be pointed out that they completely ignored the effects

of simultaneous decrease of particle size during milling (Sect 2.1.3)

1.2.8 Early Studies of Noninterstitial Transition Metal

Ternary Hydrides

With the beginning of the 1990s the interest in hydrides was somehow refocused

Firstly, interest evolved in other hydrides of Mg with transition metals, beyond

already well-researched Mg Ni stoichiometry Mg and Fe do not alloy to form

Mg 2 Fe binary intermetallic by either ingot casting or metal powder sintering dures; however, when sintering is done in hydrogen, a very stable Mg 2 FeH 6 , nonin- terstitial, ternary magnesium–iron hydride , forms [102] The Fe and Co transition metal hydrides (in reality they are complex hydrides) with Mg were firstly synthe-

proce-sized in the University of Geneva in 1991 in the same group that observed tion of this hydride during sintering of powders While presenting a formal derivative of A 2B stoichiometry, the hydrides were, in essence, Fe–metal and

forma-Co–metal hydride anionic complexes [103] The new anionic complexes of Fe and Co

were stable rendering the desorption temperature too high When desorbed, the hydrides could not change to the hydrogen-free A 2 B alloys because Mg 2 Fe and Mg 2 Co

do not exist in the binary Mg–Fe and Mg–Co systems (e.g., iron and magnesium are immiscible in the solid state and form a large miscibility gap in the liquid state) Indeed, the following reaction (1.13) is in striking contrast to reactions (1.2–1.4) and (1.5–1.6) where the dehydrogenated phase was converted back to stable host alloy (LaNi 5 or Mg 2 Ni), and indeed similar to the disproportionation reaction (1.12)

One can notice that in (1.13) the Fe 3+ in the (FeH 6 ) complex is reduced to lic Fe, the process that requires a major rearrangement of atoms Therefore it is not surprising that the kinetics of desorption was poor while being controlled by diffu-sion of the metal atoms

metal-Although more research had been coming, particularly from groups in the University of Geneva and in Stockholm, research on Mg 2 FeH 6 and similar transition-metal complex hydrides was put aside for some time This, perhaps, was unjustified

as some of those ternary hydrides formed with alkali earths and transition metal complexes, viz., BaReH 6 [104] , exhibited interesting storage properties: desorption below 100°C and 2.7 wt% H 2 capacity at very high volumetric density of 134 g H 2

L −1 Although too expensive for storage in cars, such hydrides would find a niche

in fuel cell power sources in portable electronics, if only the desorption temperature

is lowered However, the needs were not in place, at the time of great success of hydrides in rechargeable NiMH

A stable aluminum hydride , AlH 3 , has been known to exist since some time

It has been prepared by various techniques including the direct reaction of metallic

Al with atomic hydrogen by Siegel [105] AlH 3 is soluble in ether when freshly synthesized but polymerizes while aged and forms insoluble precipitate whose structure depends on the degree of polymerization of –(AlH 3 )– monomer

1.2.9 Toward Chemical/Complex Hydrides

With the coming of a New Millennium, and in the dramatic context of the September 11,

2001 terrorist attacks other drivers for hydride research became evident: energy security as to secure uninterrupted supply of energy for still unsustainable world

economies, and ever-growing needs for fuels for mass transportation Also, the

rapid economic growth in China and Asia put big pressure on markets for fossil

fuels USA Congress and President issued a “Great Challenge” in respect to

devel-opment of hydrogen storage for future fuel cell cars The interest has been renewed

for storage of hydrogen in chemical media other than metal and alloy powders So

far, there are about 170 hydrides listed on the website hydpark.ca.sandia.gov Of

particular interest became complex hydrides with participation of alkaline metal

ions and metal complexes and in mixed ionic and covalent coordination

com-pounds This class of hydrides has become known as chemical or complex hydrides

Among them are nontransition metal sodium and lithium alanates and

borohy-drides The two most investigated are sodium aluminum hydride (sodium alanate) ,

NaAlH 4 , and sodium borohydride , NaBH 4 The former, alanate, was synthesized

almost two decades earlier in Russia by Dymova et al [106] Those complex

hydrides were not thought to be suitable for solid-state hydrogen storage media as

they release hydrogen irreversibly at conditions that are outside the temperature–

pressure range desirable for fuel cell propelled cars In particular, their hydriding

and dehydriding characteristics have been creating significant challenges This

opinion has been changed with a breakthrough work by Bogdanovic´ and Schwickardi

[107] , who discovered that the addition of Ti compound (chloride) to the complex

hydride NaAlH 4 causes reversible hydrogen storage in the range up to 3.7 wt% H 2

under moderate conditions of temperature and pressure However, the reaction has

long been known to proceed in two steps [108]

NaA1H ↔1 / 3Na A1H + 2 / 3H →NaH + Al + 3 / 2H (1.14)

and again, like in the reaction of (1.13), the completely discharged phase did not

form an alloy, NaAl, from the original metals; instead alanate is reduced to metallic

Al, in fact nanoaluminum , that does not easily recharge with H 2

In 1997 a study conducted by Ford Corp (USA) compared the weights and

volumes of various hydrogen storage media in the context of their use for a

light-weight four- to five-passenger fuel-cell car with a 500-km range per one hydrogen

charge [109] Similar studies were solicited by other major car manufacturers, and

government organizations in several G7 countries The interest was raised by the

theoretical electrical conversion efficiency for an ideal hydrogen–oxygen fuel cell

being an impressive 83%, and up to 60% of the hydrogen chemical energy

con-verted to electric energy (electric current) This compared very favorably with circa

45% achieved in hydrogen internal combustion engines

Since then, search for new reversible lightweight complex hydrides was heating

up Interest aroused in other complexes of hydrogen with Al and B that have been

known since 1950s Alkali metal and magnesium tetrahydroborides were

synthe-sized [110, 111] and investigated by Germans and Russians for thermal

decomposi-tion from the late 1950s to the early 1970s Lithium tetrahydroboride, LiBH 4 , which

has already been known as a strong reducing agent in organic synthesis, has

received renewed attention after Züttel [112] reported the onset of hydrogen

des-orption at approximately 200°C promoted by SiO admixed to this borohydrides

Among the alanates, the sodium alanate and the lithium alanates have been well

studied Magnesium aluminum hydride (magnesium alanate) , Mg(AlH 4 ) 2 , with the total 9.3 wt% H 2 , has been known since its synthesis by Wiberg in the 1950s; how-ever, there have been difficulties with good reaction yield to produce this hydride until more than 50 years later Fichtner and colleagues [113] produced a sufficiently pure compound for structural investigations Again, the reversibility of the reaction was questionable as the experiments so far demonstrated that after discharge of hydrogen the alanate yields Al phase, not unlike sodium alanate in the reaction of (1.14) Nevertheless, the intensive research on complex hydrides has established itself solidly in place in early years of the new Millennium

1.2.10 Historical Development of Nanocarbons and Carbon

Nanotubes

Concurrent stream of the development of nanomaterials for solid-state hydrogen storage comes from century-old studies of porous materials for absorption of gas-ses, among them porous carbon phases, better known as activated carbon Absorption of gases in those materials follows different principles from just dis-

cussed absorption in metals Instead of chemisorption of H 2 gas into the crystalline

structure of metals, it undergoes physisorption on crystalline surfaces and in the

porous structure formed by crystals The gases have also been known to be sisorbed on fine carbon fibers

Reports on growth of fine filaments of carbon have been coming since 1890, when filamentous carbon was obtained by passing cyanogens over red-hot porce-lain Hollow carbon fibers were reported in a Russian journal in 1952 [114] An interesting view point about the discovery of carbon-nanotube-like materials can be

found in Guest Editorial to the journal Carbon [115] Fine fibers showing graphitic

structures were frequently encountered as unwanted phases in both steel industry and in industrial catalysis, where they deposited on walls of metallurgical furnaces and chemical reactors Since then development of carbon filaments was driven by their use as reinforcement phase in lightweight composite materials for space and aerospace industries Research focused on fabrication of carbon fibers from polymer-based precursors such as rayon, polyacrylonitrile (PAN), or mesophase pitch

In the 1970s Oberlin et al [116] published images of what can be now called

single- or double-wall carbon nanotubes (SWNT, DWNT) , although at this time,

images obtained from transmission electron microscopes cannot clearly reveal the number of concentric hollow tubes in the fiber In the 1980s he produced hollow carbon filaments by chemical vapor deposition (CVD) in a floating catalyst reactor where they were nucleated and grown on 10-nm metal catalyst nanoparticles Such CVD method to produce fine hollow filaments became with time the most versatile

production method for carbon nanotubes (CNT) At the time, it was mechanical

property that was of primary interest, as it was realized early that well bulk carbon fibers resist crack propagation Therefore, little attention was paid toward the

Oberlin’s small and hollow objects, which only latter were termed carbon

nano-tubes, and were becoming studied around the world for other than mechanical,

mostly electronic, and later hydrogen storage properties The situation changed

diametrically in 1991 when Iijima [117] published a breakthrough paper entitled

“Helical microtubules of graphitic carbon.” Iijima’s rediscovery of CNTs came

dur-ing the research conducted to follow on a breakthrough research on another carbon

nanophase, fullerene , and spherical C 60 carbon atom configuration, which was

dis-covered 6 years earlier by Kroto et al [118] and published in a seminal paper on

C-60 buckminsterfullerene The latter molecule consisted of 60 carbons that

satis-fied a mathematical theorem by Leonhard Euler requiring a spherical surface, as to

be entirely built up from pentagons and hexagons, the molecule must have exactly

12 pentagons like the geodesic domes of the architect R Buckminster Fuller, and

not unlike a soccer ball Shortly after, the now clear CNT images were observed by

powerful high-resolution transmission electron microscope (HRTEM) in soot

pro-duced by the just published Kratschmer-Huffman method on the laboratory scale

production of C 60 [119] CNTs can be seen as single, one atom-thick, graphitic

planes, viz graphene planes, rolled to form nanoscale cylinders, which are often

closed at the end by half a sphere of the earlier discovered fullerene These

struc-tures, while comprising of only one single-wall cylinder were termed single-walled

nanotubes, SWNTs, whereas those consisting of two or more concentric graphene

cylinders became known as multiwalled nanotubes (MWNTs) Depending of how

the graphene sheet is rolled up, the three forms of CNTs, zigzag, armchair, and

chiral, change their electronic properties from metallic, to superconducting, to

insulating In 1992 it was shown that MWNTs can be produced in DC arc discharge

in He with use of two graphite electrodes [120] , and in the following year the

SWNTs were also produced in arc-discharge using alloy catalysts (like Fe–Co)

[121 – 123] Recently, unrolled single-layer graphene phase was also prepared by

Novoselov et al [124]

In 1997 Dillon and coworkers [125] reported the first experimental result of high

hydrogen uptake by carbon nanotubes; their estimate was 5 – 10 wt% H 2 This was

followed by hot race to claim by many researchers ever-larger potency of carbon

nanotubes to absorb and retain hydrogen The most known became the claim by

Baker et al [126] that some nanofibers can absorb over 40 – 65 wt% H 2 However,

these experimental results and their interpretation were strongly criticized and have

not yet been able to be reproduced More careful studies, that follow, have shown

that only 0.7 – 1.5 wt% of hydrogen gas is adsorbed in nanofibers under ambient

temperature and pressures slightly above 100 bar [127] Since then, the area of

nanocarbons and carbon nanotubes for hydrogen storage has seen continuing

inter-est from many research groups worldwide

Carbon nanotubes, as graphene and graphite, are highly ordered carbon phases

However, a separate line can be drawn for historical development of disordered

carbon phases; among them is an amorphous carbon (am-C) In it, strong bonding

between carbons did not allow for completely chaotic distribution of carbon atoms

in solid-state phase Instead, amorphous carbon exhibits random distribution of

three possible coordinations of carbon atoms in a planar sp 2 , tetrahedral sp 3 , and

even linear sp 1 configurations of electronic orbitals Some of these phases were

shown to be greatly stabilized by hydrogenation Hydrogenated amorphous carbon has been observed to contain as much as 40 – 60 at.% H 2 in a structure consisting of more than 70% tetrahedrally bonded atoms in sp 3 configuration [128] At lower

hydrogen content at 25 – 30 at.% H 2 the sp 3 fraction is high At 70 at.% H 2 , the

hydrogenated tetrahedral amorphous carbon, known better as diamond-like carbon (DLC) , has been developed for many very useful applications [128] These amor-

phous phases have only been produced as thin films in deposition from high-density

plasma Recently, a new spherical graphitic nanophase, termed carbon nanoshell ,

was found by Wronski and Carpenter [129] in carbon residues obtained by acid leaching of Ni from commercial carbonyl nickel powders produced in century-old CVD carbonyl nickel process in INCO refineries in Wales, UK and Canada However, these and other new nanocarbons are still waiting to be investigated for hydrogen storage As for the year 2007, the recent Viewpoint Paper by Chahine and Bénard [130] states clearly that these new nanocarbons and nanostructures that store hydrogen by physisorption will still require development to bring about quali-tative changes; means are still waiting for looking at them from a new angle

1.2.11 New Materials and Techniques

In the past several years, some old compounds were put into perspective from a new angle keeping still in mind that to meet hydrogen storage targets in hydrogen storage for vehicular applications such storage must be reversible in near-ambient tempera-tures (less than 100°C) and pressure range One of the systems that gained great inter-

est was lithium nitride Li 3 N has been known since the 1910 study by Ruff in Germany

to compound with hydrogen, and release it under strong heating However, it was only

in recent investigations by Ping Chen and coworkers in National University of Singapore [131] that the Li–N–H system was reassessed and demonstrated hydrogen absorption in as low temperature as 100°C, and desorption that is rapid, yielding as much as 9.3 wt% H 2 in half an hour when temperature is raised to 255°C What attracted wide interest in this research was that about 6.3 wt% H 2 was desorbed below 200°C, and the desorption reaction follows the reverse of absorption:

Li N + 2H ↔Li NH + LiH H+ ↔LiNH + 2LiH (1.15) However, the desorption proceeds under 10 −5 bar vacuum, and traces of ammonia that is toxic for fuel cells were detected

While lithium nitride was an old material revisited, zeolites are well-researched

materials for their catalytic and physisorption properties These microporous ganic frameworks were known to trap hydrogen; however, their potential for hydro-gen storage was not great as they show only moderate storage capacities On the

inor-other hand, a new microporous metal–organic frameworks (MOFs) ignited return of

interest in hydrogen storage based on physisorption rather then chemisorption

In 2003, Rossi et al [132] reported a MOF material with high hydrogen sorption

capacity, 4.5 wt% H 2 at cryogenic temperature of 78 K The promising hydrogen

sorption properties of MOF materials were discovered only 4 years from the time

the first synthesis of this class of materials was performed [133] Much more

mod-est, 1 wt% H 2 storage, yet at room temperature and 20-bar pressure was observed

This started another vigorous research in materials, in which scaffolding-like nature

leads to extraordinarily high surface area of 2,500–4,000 m 2 g −1 , i.e., more than four

times the specific surface area of zeolites In the following research the capacity of

systems such as IRMOF-6 and -8 in room temperature exceeded that of carbon

nanotubes in cryogenic temperatures Since then, the area of mesoporous hydrogen

storage media , where hydrogen is sorbed in very-high porosity materials, has been

experiencing great interest from the community of researchers in hydrogen

storage

The intertwining of the development of nanomaterials and methods of

nanoprocess-ing is still ongonanoprocess-ing While development of new MOF-type materials is an example of

a quintessential bottom–up nanotechnology , an opposite route, top–down

nanotechnol-ogy approach, realized through reduction in the grain and particle size has also been

furthered New techniques have become developed, which have potential to take

mate-rials research beyond metals and carbons, and beyond the limitations imposed by the

choice between Scylla and Charybdis of a strong chemisorption in nanohydrides and

a weak physisorption in nanocarbons New explored materials are nanocomposites or

hybrid hydrogen storage media where chemical bonds are optimized for reversible

storage of hydrogen in solids One of these new techniques having potential for

prepa-ration of such materials is the controlled reactive hydrogen ball milling/alloying

(CRMM/MA) of metals, nonmetallic elements, and compounds in hydrogen-filled ball

mills This approach to preparation of new hydrides and hydride nanocomposites has

been developed at the University of Waterloo and CANMET’s government

laborato-ries in Ottawa, Canada The milling is conducted in specialty, high-energy ball mills,

where trajectories of balls are well controlled, and H 2 gas can be supplied and absorbed

during milling The complex Mg 2 FeH 6 hydride was prepared in this way by Varin et

al [134] in the direct, mechanically driven synthesis, viz direct mechanosynthesis

reaction by reacting inexpensive elemental Mg and Fe metals:

The reaction of (1.16) follows reverse path to the thermal decomposition reaction

of (1.13) and proceeds at room temperature and only slight overpressure of

hydro-gen supply This presents a new mechanical activation route to manufacturing of

nanomaterials for hydrogen storage

This brief history of century-old investigations toward hydrogen interaction with

solid materials and nanomaterials brings us to the current state of affairs when the

hydrogen storage for fuel cell systems still remains to be solved Indeed, in the first

decade of the new Millennium, and at the advent of the Hydrogen Economy, fuel

cell stacks for use in mass transportation, like those developed by Ballard Power

Systems based in Canada, are ready for mass commercialization Also, hydrogen

production systems, as those widely used in the chemical and fossil-fuel refining industries are ready to supply hydrogen fuel to fuel cell cars Sources of hydrogen are numerous More than 90% of hydrogen is today produced through

(oil)-thermochemical reforming of methane gas, CH 4 , at high-temperature (800–1,700°C) reaction with steam or oxygen [(1.17) and (1.18)] to make the syngas…not unlike the gas produced by gasification of coal in the nineteenth century (e.g., (1.1))

hydrogen challenge is waiting solution Can advances of nanotechnology and materials meet the hydrogen challenge ?

1.3 Nanoprocessing in Solid State in High-Energy Ball Mills

1.3.1 Processes for the Synthesis of Nanostructured Materials

Since the interest arouse in unusual chemical, physical, and mechanical properties

of nanostructured materials, progress in the synthesis of such materials has only been accelerating Already established wet chemistry methods, with use of aqueous and nonaqueous solutions, have been reexamined with respect to feasibility to con-duct chemical precipitation of nanocrystalline phases under conditions that hinder growth of customary crystalline phases These methods involve precipitation of metals from its salts with use of strong reductors as borohydrides In general, meth-ods have been explored that accelerate the rate of reaction and thus limit the time available for ions to diffuse and contribute to growth of the nucleating crystals Among the advantages of the established wet chemistry methods, like sol–gel processing, are good control over microstructure and particle morphology, and no need for special equipment Sol–gel approaches were delivering good results even when a synthesis of complex structures was required (e.g., in complex multimetal oxides) Generation of nanometer-sized phases through deposition of metals and metal oxides from gas phase was among the first dry-chemical methods [62] Many

of new processes required development of special equipment or methods as to conduct

physical vapor deposition (PVD) or chemical vapor deposition (CVD) under

thermodynamically nonequilibrium conditions Among these old and new processes

for preparation of nanocrystalline alloys and compounds are the following:

• Solution-precipitation methods, e.g., those with use of strong borohydride reductors

• Sol–gel technologies

• Reverse-micelle synthesis

• Polymer-mediated synthesis

• Protein-templated methods

• Rapid solidification and devitrification of amorphous metals and metallic glasses

• Combustion-flame chemical vapor condensation processes (Kear)

• Induction-heating chemical vapor condensation processes

• DC and RF magnetron sputtering, inclusive of the method of thermalization

• Laser ablation methods

• Supercritical fluid processing

• Sonochemical synthesis and microwave hydrodynamic cavitation synthesis

Many if not all of these processes have already been developed and well established

even before the term nanomaterials was coined; they remain further refined with focus

on manufacturing materials for particular properties, electronic, etc All of them are well

covered in books and monographs, which are too many to make good recommendation

to our reader For recent progress the reader can consult Nanomaterials Handbook ,

where the editor, Yury Gogotsi, assembled impressive body of authors to write chapters

on processing and properties of many nanomaterials, particularly well covered being

carbon-based nanomaterials, metallic semiconductors, and ceramic nanomaterials

[135] All of these methods stem from a natural property of atoms to self-organize into

ordered structures and crystallites as being driven by energy benefits when atoms join

into structures Short-range ordering in metal alloys may be formed during deposition

of atoms or atom clusters from gas phases, and survive nonequilibrium processing of

metals via solidification from rapidly quenched melts The short-range order of atoms

may even be preserved in amorphous metallic alloys The approach to build up

nanoc-rystals through such self-organization of atoms is well recognized in the preparation of

nanomaterials and termed the bottom–up nanotechnology approach

Bottom–up processing of nanomaterials relies upon use of either liquid solvents

and vapor gas phases, or solidification from molten metals and compounds

However, in this book we concentrate on nanoprocessing conducted entirely in a

solid state This is a class of methods based on processes conducted in high-energy

ball mills Such methods are based on high-energy grinding and milling of materials

They provide top–down approach toward manufacturing nanomaterials

1.3.2 Milling Processes and Equipment

The milling, grinding, and pulverizing of materials have always been one of unit

operations in chemical engineering They are well described in chemical engineering

handbooks along with heat transfer, mass transfer, fluid flow, and thermodynamic processes Each of these unit operations gather knowledge of physical laws and practice that is necessary for reliable engineering design in many industries, viz., mineral, ceramic, and powder metallurgy The primary objectives of milling have always been mixing or blending, change in particle shape (morphology), and first

of all: size reduction The use of the milling and pulverizing for causing the change in fundamental mechanical, chemical, and physical properties of the mate-rials itself was not the target of milling or grinding Use of mechanical milling for synthesis of new alloys, compounds, and nanomaterials was not envisioned until recent decades

Equipment used for milling may be classified according to the way in which mechanical forces are applied: (1) between two solid surfaces (crushing, shearing), and (2) at one solid surface (impact) The milling in ball mills combines both crush-ing/shearing and impact forces combined in various proportions, depending on the equipment used There are many different designs of ball mills, which can be used for processing of advanced materials Among them are the following:

• Tumbler, jar, drum , or cannon ball mills

• Szegvari attritor vertical mills and other vertical stirred ball mills

• Planetary Fritsch and Retsch model mills

• Shaker (vibratory) SPEX model mills

• A.O.C magnet-controlled mechanical model mill (Uni-Ball magnetomill)

• A.O.C electric discharge-assisted mechanical mill

• ZOZ continuous-fed horizontal mill, and other horizontal high-energy bead mills Ball or tube mills have a cylindrical or conical shell, rotating on a horizontal, verti-cal axis The ball mill differs from the tube mill by being short, and having the crucible length and diameter almost identical The typical ball mill, used as much

in laboratory like in industry, has been the tumbler ball mill (in laboratory practice also known as the jar mill and in industry drum mill) The grinding balls (usually

in large numbers) impact upon the powder charge when cylindrical container placed on rollers rotates along horizontal axis The balls may roll down the inside wall surface, which produces shear forces on powder trapped between the wall and the ball, but mostly they fall freely accelerated only by gravitation force, then impacting the powders (and other balls) beneath them To maximize the impact

forces imposed on powder the criterion of critical speed may be applied, where N c expressed in rotations per minute (RPM) is the theoretical speed at which the cen-trifugal force on a ball (at the height of its orbital path) becomes equal to the force

where D is diameter of the mill in meters, and the ball diameter is kept small with

respect to the mill diameter

The milling in ball mills can be described as kinetic processing when the contact

of the balls with the material is the main event of kinetic energy transfer from the