nanocrystalline materials, 2004, p.357

The authors ofthis monograph have taken the difficult task of presenting to thereader information on hundreds of original investigations of thenanocrystalline state, grouping these inves

MATERIALS

A.I Gusev, A.A Rempel

CAMBRIDGE INTERNATIONAL SCIENCE PUBLISHING

Preface vii

List of Main Notations xiii

1 INTRODUCTION 1

References 23

2 SYNTHESIS OF NANOCRYSTALLINE POWDERS 27

2.1 GAS PHASE SYNTHESIS 27

2.2 PLASMA CHEMICAL TECHNIQUE 32

2.3 PRECIPITATION FROM COLLOID SOLUTIONS 44

2.4 THERMAL DECOMPOSITION AND REDUCTION 49

2.5 MILLING AND MECHANICAL ALLOYING 52

2.6 SYNTHESIS BY DETONATION AND ELECTRIC EXPLOSION 59

2.7 ORDERING IN NON-STOICHIOMETRIC COMPOUNDS 65

2.8 SYNTHESIS OF HIGH-DISPERSED OXIDES IN LIQUID METALS 76

2.9 SELF-PROPAGATING HIGH-TEMPERATURE SYNTHESIS 78

References 79

3 PREPARATION OF BULK NANOCRYSTALLINE MATERIALS 89

3.1 COMPACTION OF NANOPOWDERS 90

3.2 FILM AND COATING DEPOSITION 101

3.3 CRYSTALLISATION OF AMORPHOUS ALLOYS 104

3.4 SEVERE PLASTIC DEFORMATION 108

3.5 DISORDER–ORDER TRANSFORMATIONS 114

References 124

4 EVALUATION OF THE SIZE OF SMALL PARTICLES 131

4.1 ELECTRON MICROSCOPY 132

4.2 DIFFRACTION 137

4.3 SUPERPARAMAGNETISM, SEDIMENTATION, PHOTON CORRELATION SPECTROSCOPY AND GAS ADSORPTION 151

References 156

RYSTALLINE POWDERS 159

5.1 STRUCTURAL AND PHASE TRANSFORMATIONS 159

5.2 CRYSTAL LATTICE CONSTANT 169

5.3 PHONON SPECTRUM AND HEAT CAPACITY 177

5.4 MAGNETIC PROPERTIES 190

5.5 OPTICAL PROPERTIES 207

References 214

6 MICROSTRUCTURE OF COMPACTED AND BULK NANOCRYSTALLINE MATERIALS 227

6.1 INTERFACES IN COMPACTED MATERIALS 228

6.2 STUDY OF NANOCRYSTALLINE MATERIALS BY MEANS OF POSITRON ANNIHILATION TECHNIQUE 234

6.3 STRUCTURAL FEATURES OF SUBMICROCRYSTALLINE METALS PREPARED BY SEVERE PLASTIC DEFORMATION 258

6.4 NANOSTRUCTURE OF DISORDERED SYSTEMS 268

References 275

7 EFFECT OF THE GRAIN SIZE AND INTERFACES ON THE PROPERTIES OF BULK NANOMATERIALS 284

7.1 MECHANICAL PROPERTIES 284

7.2 THERMAL AND ELECTRIC PROPERTIES 301

7.3 MAGNETIC PROPERTIES 311

References 330

8 CONCLUSIONS 340

References 345

SUBJECT INDEX 347

Success in Circuit lies Too bright for our infirm DelightThe Truth’s superb surprise”

After Emily Dickinson

Preface

In 1998 the monograph “Nanocrystalline Materials: Preparation andProperties” by A I Gusev was published by Ural Division of theRussian Academy of Sciences Publishing House (Yekaterinburg).The monograph was the first Russian and one of the first in theworld generalisation of experimental results and theoreticalconsiderations regarding the structure and properties of not onlydispersed but also bulk solids with the nanometer size of particles,grains, crystallites and other elements of the structure Themonograph was of considerable interest to readers and became,almost immediately after publishing, a bibliographic rarity not onlyfor readers but also for the majority of scientific and technicallibraries In more than 10 technical universities of Russia, this book

is used as a basis of a course of lectures “Nanocrystallinesubstances and materials” for students, specialising in advancedmaterials science Therefore, already in the year 2000 andsubsequently in 2001, the Nauka Publishing House (Moscow)published twice a supplemented edition of the book

“Nanocrystalline Materials” written by A I Gusev and A A.Rempel

The English edition of the monograph by A I Gusev and A A.Rempel, presented here to the reader, has been greatly refreshed,expanded and supplemented in comparison with the last Russianedition The monograph is concerned with one of the mostimportant current scientific problems, which is common formaterials science, solid state physics and solid state chemistry,namely the nanocrystalline state of matter It may be expected thatthe publication of the monograph in its new, expanded version will

be available to a considerably larger number of investigators andengineers concerned with the production and application ofnanocrystalline materials

state of matter were published in various scientific journals,conference proceedings and compilations of articles The authors ofthis monograph have taken the difficult task of presenting to thereader information on hundreds of original investigations of thenanocrystalline state, grouping these investigations in accordancewith the investigated materials and properties, describing thegeneral and special features in the results of these investigations,and focusing attention on the most interesting and practicallyimportant effects of the nanocrystalline state.

The term nano, which is derived from the Greek word nanos

which means dwarf, designates a milliardth (10-9) fraction of a unit.Thus, the science of nanostructures and nanomaterials deals withobjects in condensed matter physics on a size scale of 1 to 100 nm.The special physical properties of small particles have beenutilised by peoples for a very long time, although this has beencarried out unknowingly Suitable examples are ancient Egyptglasses, colored with colloidal particles of metals, dye pigmentsused in different historical periods The first scientific mention ofthe small particles is evidently the disordered movement of particles

of flower pollen, suspended in a liquid, discovered in 1827 by theScottish botanist R Brown This phenomenon is referred to asBrownian motion The article on this microscopic observation (R

Brown, Phil Mag 4, 161 (1828)) laid foundations to many

investigations The theory of Brownian motion, developedindependently by A Einstein and M Smoluchowski at thebeginning of the 20th century, is the basis of one of experimentalmethods of determining the size of small particles The scattering

of light by colloid solutions and glasses was studied by M Faradaybetween 1850 and 1860

The starting point of examination of the nanostructured state ofsubstance were the investigations in the area of colloid chemistry,which were already quite extensive since the middle of the 19th

century At the beginning of the 20th century, a significantcontribution to the experimental confirmation of the theory ofBrownian motion, to the development of colloid chemistry andexamination of dispersed substances, and to the determination of thesize of colloid particles was provided by the Swedish scientist T.Svedberg In 1919, he developed a method of separating colloidparticles from a solution using an ultracentrifuge In 1926, hereceived a Nobel Prize in chemistry for his work on dispersedsystems

heterogeneous catalysis, ultrafine powders and thin films Theseinvestigations raise question about the effect of the small size ofparticles (grains) on the properties of studied materials At present,the nanostructured materials include nanopowders of metals, alloys,intermetallics, oxides, carbides, nitrides, borides and thesesubstances in the bulk state with the grains of the nanometer size,together with nanopolymers, carbon nanostructures, nanoporousmaterials, nanocomposites, and biological nanomaterials Thedevelopment of nanomaterials is directly associated with thedevelopment and application of nanotechnology The examination

of nanomaterials has revealed a large number of grey areas in thefundamental knowledge of the nature of the nanocrystalline stateand its stability under different conditions On the whole, the field

of nanomaterials and nanotechnology is very wide and at presenthas no distinctive contours

The unique structure and properties of small atomic aggregationsare of considerable scientific and technical interest, because theyrepresent an intermediate state between the structure and properties

of isolated atoms and bulk solids However, the problem at whatstage of atom agglomeration the properties of bulk crystals isformed completely has not as yet been solved It is not clear howthe contributions of surface (associated with the interfaces) andbulk (associated with the size of the particles) effects to theproperties of the nanocrystalline materials can be separated Theinvestigations in this field were carried out for a long period oftime on isolated clusters, consisting from two atoms up to hundreds

of atoms, small particles with a size of more than 1 nm, andultrafine powders The transition from the properties of isolatednanoparticles to the properties of bulk crystalline substancesremained a grey area because the intermediate member, i e a bulksolid with the grains of the nanometer size, was not artificialcreated Only after 1985, when methods of preparation of bulknanocrystalline substances were developed, work was started to fillthis gap in the knowledge of solids

The spectrum of properties of matter can be enormouslyenhanced if nanometer-size particles are agglomerated to a bulkmaterial so that in addition to the crystallites with a nanometer-sizethey consist of a large portion of interfaces with a disorderedstructure and novel properties In very small crystallites of the size

of a few nanometers, this is a few millionth of a millimeter, newproperties appear due to quantum size effects or scaling laws, which

Indeed, the scientific interest to the nanocrystalline state of thesolids in the powdered and bulk form is associated mainly withexpectation of various size effects on the properties of thenanoparticles and nanocrystallites whose sizes are comparable orsmaller than the characteristic correlation scale of a specificphysical phenomenon or the characteristic length, which are present

in the theoretical description of some property or process Thesecharacteristic lengths are the free path of the electrons, the length

of coherence in superconductors, the wavelength of elasticoscillations, the size of the exciton in semiconductors, the size ofthe magnetic domain in ferromagnetics, etc

The industrial interest to the nanomaterials is caused by thepossibility of extensive modification and even principal changes inthe properties of the existing materials at transition to thenanocrystalline state, and by new possibilities offered bynanotechnology in the creation of materials and wares fromstructural elements of the nanometer size The essence ofnanotechnology is the possibility to work at the atomic andmolecular level, in the length scale between 1 and 100 nm, in order

to produce and use materials and devices characterised by newproperties and functions because of the small scale of theirstructure Thus, the term “nanotechnology” relates to the sizes ofstructural elements in particular Nanoproducts are already playing

an important role in almost all branches of industry The range ofapplication of these products is huge: more efficient catalysts, filmsfor microelectronics, new magnetic materials, protective coatings onmetals, plastics and glasses In the next couple of decades,nanostructured objects will operate in biological systems and will

be used in medicine The successes of nanotechnology may bemanifested most efficiently in electronics and computer technology

as a result of further miniaturisation electronic devices and thedevelopment of nanotransistors

In the content of this monograph, we have attempted to take intoaccount both purely scientific, fundamental interest in the problem

of the nanocrystalline state as a special non-equilibrium state ofmatter, and also some technical aspects of this problem, which are

of considerable importance for materials science and the practicalapplication of nanomaterials

The combined analysis of the structure and properties of isolatednanoparticles and nanopowders presented in the book, on the one side,and of bulk nanomaterials, on the other side, shows that the level of

aB1 lattice constant (period) of cubic unit cell

with B1 structure

Cp, Cv heat capacity at constant pressure and at

constant volume

D, <D> size or mean size (diameter) of particle

(grain, crystallite, domain)

Ddiff diffusion coefficient

FWHM full linewidth at half maximum

Tmelt melting temperature

isolated nanoparticles is considerably higher in comparison with bulknanocrystalline materials Evidently, this is a result of the considerablylonger (practically from the beginning of the 20th century) study ofhigh-dispersed systems and nanoclusters in comparison with bulknanomaterials which became the object of investigations only in thelast 10–15 years.

The monograph in the concentrated form includes a large part ofthe most important data on the nanocrystalline state of solids Inwriting the monograph, we used a very large number of originalinvestigations, starting from 1828 up to the year 2003, inclusive

It should be noted that more than 80 % of all references is made

to studies carried out since 1988 Thus, the monograph reflectsaccurately the current state of investigations of the nanocrystallinestate of solids We think it will be interesting and useful tospecialists in condensed state physics, solid state chemistry,physical chemistry and materials science

Yekaterinburg, March 2004

A I Gusev, A A Rempel

γe electronic heat capacity coefficient

η viscosity (liquid shear viscosity)

θ Bragg diffraction angle

θD characteristic Debye temperature

“ And freely men confess that this world’s spent,

When in the planets and the firmament They seek so many new; they see that this

Is crumbled out again to his atomies.

‘This all in pieces, all coherence gone, All just supply, and all relation ” After John Donne (An Anatomy of the World, 1611)

+D=FJAH

Introduction

The problem of production of ultrafine powders of metals, alloys andcompounds and submicrocrystalline materials for different areas oftechnology, has been discussed in literature for many years now

In the last couple of decades, the interest in this subject has greatlyincreased because it was found (primarily, in metals) that a decrease

in the size of crystals below some threshold value may result in alarge change of the properties [1–16] These effects form whenthe mean size of crystalline grains does not exceed 100 nm, and aremost evident when the grain size is smaller than 10 nm Whenexamining the properties of superfine materials, it is necessary totake into account not only their structure and composition but alsodispersion Polycrystalline superfine materials with a mean grain size

of 300 to 40 nm are referred to as submicrocrystalline, and thosewith a mean grain size of less than 40 nm as nanocrystalline The

conditional classification of materials on the basis of the size D of

particles (grains) is shown in Fig 1.1 Nanomaterials can also beclassified on the basis of their geometrical form and thedimensionality of structural elements from which they consist Themain types of nanocrystalline materials as regards the dimensionalityare cluster materials, fibrous materials, films and multilayermaterials, and also polycrystalline materials whose grains have

comparable size in all three mutually perpendicular directions (Fig.1.2) In this book, attention will be given mainly to the structure andproperties of bulk and powdered substances and materials with a

particle size of 5 to 200–300 nm, i.e nanocrystalline and

submicrocrystalline

Fig 1.1 Classification of substances and materials on the basis of particle (grain)

size D.

Fig 1.2 Types of nanocrystalline materials: 0D (zero-dimensional) clusters;

1D (one-dimensional) nanotubes, filaments and rods; 2D (two-dimensional) films and layers; 3D (three-dimensional) polycrystals.

,FRVDKHGUDO PHWDOOLF FOXVWHUV 1DQRFU\VWDOOLQH PDWHULDOV 6XEPLFURFU\VWDOOLQH PDWHULDOV

&RDUVH JUDLQHG PDWHULDOV

For the investigator and the engineer used to working withtraditional substances and materials in which the elements of themicrostructure have the size of approximately 1 µm or more, thefirst encounter with nanomaterials results in at least surprise Thissurprise is similar to that experienced by somebody who has seenfor the first time buildings constructed in the modern style: instead

of normal straight lines and angles, there are distorted planes andcomplicated broken contours (Fig 1.3) However, after some time

it becomes obvious that it is possible to live in such an unusualbuilding and that it may even be interesting This also relates to thenanocrystalline state, i.e it is unusual but it may be interesting towork with it

The difference between the properties of small particles and theproperties of bulk materials has been known for a relatively longperiod of time and has been utilised in different areas of technology.Suitable examples are widely used aerosols, dye pigments, glasscolored by colloidal particles of metals

Suspensions of metallic nanoparticles (usually iron or its alloys)with the size from 30 nm to 1–2 µm are used as additions to engineoil during service for the restoration of worn components ofautomobiles and other engines

The small particles and nanosized elements are used for theproduction of various aviation materials For example, radiowave-

Fig 1.3 The first encounter with nanomaterials causes the same surprise as these

buildings in the modern style, designed by F.O’Gehry in Düsseldorf (Germany).

absorbing ceramic materials, whose matrix is characterised by therandom distribution of fine-dispersion metallic particles, are used inaviation industry Whisker single crystals and polycrystals (fibres)are characterised by very high strength, for example, graphitewhiskers have a strength of ~24.5 GPa or ten times higher than thestrength of steel wire They are used as fillers for light compositematerials for aerospace applications Carbon fibres and graphitewhiskers are relatively thick (approximately 1–10 µm) and are notnanomaterials, but their production and applications were the firststep on the path to the development of carbon nanomaterials Afterdiscovery in 1984–1985 of a new allotropic modification of carbon,

i.e spherical fullerenes C60 [17, 18], attempts were made toproduce other topological forms of carbon nanoparticles One of theproposed possible forms of carbon nanoparticles was, in particular,

a quasi-one-dimensional tubular structure [19], referred to as thenanotube The nanotubes form as a result of rotation of (0001)basal planes of the hexagonal lattice of graphite and can be single

or multilayered In fact, in 1991 and in the following years of the20th century, researchers succeeded in detecting quasi-one-

dimensional tubular structures of carbon, i.e carbon nanotubes)

[20–22] As an example, Fig 1.4 shows a computer graphical model

of a double-shell carbon nanotube, and Fig 1.5 shows experimentallyproduced nanotubes [23] Carbon nanotubes with a diameter of

Fig 1.4 A computer graphic model for

a double-shell carbon nanotube showing

a helical arrangement of hexagons [23].

D > 5 nm, consisting of 2 to 50 coaxial tubes, were detected for

the first time by transmission electron microscopy in a condensate

in an electric arc discharge between graphite electrodes [20] Theresults of modelling of the structure and electronic properties ofcarbon nanotubes have been generalised in [24] The carbonnanotubes have high mechanical strength and may be used fordeveloping high-strength composites Nanotubes have been used indifferent mechanical nanodevices [25], like nanoindentors formicrohardness measurements Depending on the type of helicoidalordering of the carbon atoms in the walls of the carbon nanotubes,these nanotubes have semiconductor or metallic conductivity.Consequently, they are used as conducting elements in electronicnanotechnologies In atomic force microscopes, the carbonnanotubes have replaced the metallic probe [26]

By joining the carbon nanotubes it is possible to produce a largenumber of structures with differing properties Synthesis of thesestructures is very important for electronic technology T-connectednanotubes, which may operate as a contact device, were produced

in [27] The authors of [28, 29] grew Y-shaped carbon nanotubes(Fig 1.6); this structure is referred to as the Y-junction carbonnanotube Synthesis was carried out by chemical deposition fromthe gas phase (CVD): pyrolysis of acetylene with subsequent

Fig 1.5 Multi-shell carbon nanotubes [23].

growth of Y-nanotubes was carried out at a temperature of 920 K

in branching nanochannels of the aluminium matrix Cobalt, being

a growth catalyst, was deposited on the walls and the bottom of thenanochannels The diameter of the stem of the produced Y-nanotube was approximately 60 nm, the diameter of the branches

~40 nm The authors of [30] produce carbon Y-nanotubes bypyrolysis of organometallic precursors As a result of a defectivestructure in the area joining the prongs, the Y-nanotube passes

electric current only in one direction, i.e it operates as a diode

[29] If controlling voltage is additionally applied to one of theprongs of the Y-nanotube, the nanotube operates as a currentstabiliser The possibility of controlling the current leads to thepossibility of extensive application of Y-nanotubes in electronics.Recently, a group of investigators [31] from the Department ofMaterials Science and Engineering at the Rensselaer PolytechnicInstitute (Troy, USA) proposed a method of controlled growth ofcarbon nanotubes on a substrate coated with a layer of SiO2.Pattering of SiO2 was generated by photolithography and thensubjected to combined wet and dry etching in order to produceislands of SiO2 distributed in a specific fashion Subsequently,bundles of nanotubes, forming a unique nanostructure (Fig 1.7) weregrown in a gas mixture of xylene/ferrocene C8H10/Fe(C5H5)2 on

Fig 1.6 Model of a carbon Y-nanotube.

the islands of SiO2 by the CVD method In this process, the ironincluded in the composition of Fe(C5H5)2 plays the role of acatalyst Each bundle includes several tens of multiwalled nanotubeswith a diameter of 20–30 nm According to the authors of [31], suchnanostructures may be used in integrated systems of the nextgeneration and in microelectromechanical devices.

The first publications dealing with the production of boron nitridenanotubes appeared in 1995–1996 [32–34] Intensive research isbeing carried out into the synthesis of silicon carbide nanotubes.The range of applications of these nanotubes is even wider because

of higher hardness and high melting point of silicon carbide.Heterogeneous synthesis of silicon carbide fibres was described bythe authors of [35], the gas phase method of production of siliconcarbide nanofibres with a diameter of ~100 nm, produced fromsilicon and carbon powders, was described in [36], and the authors

of [37] reported on hollow silicon carbide nanostructures Thenanotubes and nanofilaments of silicon carbide and also of boroncarbide and SiO2, produced by these methods, were presented in

a lecture “Elongated structures of silicon carbide: nanotubes,nanofilaments and microfibres” by A I Kharlamov at the NATOAdvanced Study Institute “Synthesis, Functional Properties andApplications of Nanostructures” (July 26–August 4, 2002, Heraklion,Crete, Greece) At the same conference, in a lecture “The state-of-art synthesis techniques for carbon nanotubes and nanotubes-based architecture” P Ajayan told about silicon carbide nanotubesproduced on an Al2O3 substrate At the 2nd NASA Advanced

Fig 1.7 Controlled carbon nanotube growth on a silica-coated substrate [31].

Materials Symposium “New Directions in Advanced MaterialsSystems” (May 29–31, 2002, Cleveland, Ohio), D Larkin presented

a report “High temperature nano-technology: silicon carbidenanotubes synthesis”

The results of investigations and application of various nanotubesare presented in [38] The book starts with the playful amateurpoem “Material Ethereal” by Peter Butzloff (University of NorthTexas, Denton, USA) on a mystery nanotube and problems ofexamining it Only the first and final lines of this poem are givenhere:

We speculate but underrate

what mystery we wrangle,

a Nanotube from carbon crude

that nature did entangle.

Oh Nanotube of carbon crude

so cumbersome we trundle

through pass of phase, by time decays

and character to bundle!

The catalysis of chemical reactions is a very important and largearea of long-term and successful application of fine particles ofmetals, alloys and semiconductors Heterogeneous catalysis bymeans of high-efficiency catalysts produced from ultrafine powders

or ceramics with grains of the nanometer size is an independentand very large section of physical chemistry

Various problems of catalysis have been discussed in hundreds

of books and reviews and tens of thousands of articles Theextensive discussion of the problems of catalysis on fine particleswith respect to both the content and volume is outside theframework of this book and, consequently, certain generalassumptions relating to the catalytic activity of fine particles will

be discussed only briefly

Catalysis on fine particles plays an exceptionally important role

in industrial chemistry Catalysed reactions usually take place atlower temperatures than non-catalysed reactions and are moreselective In most cases, the catalysts are represented by isolatedfine particles of metals or alloys deposited on a carrier with a

developed surface (zeolites, silicagel, silica, pumice, glass, etc.) The

main task of the carrier is to support obtaining the smallest size of

deposited particles and prevent their spontaneous coalescence andsintering.

The high catalytic activity of fine particles is explained byelectron and geometrical effects, although this division is very

conventional because both effects have the same source, i.e the

small particles size The number of atoms in an isolated metallicparticle is small and, consequently, the distance between the energylevels δ ≈ EF/N (EF is Fermi energy, N is the number of atoms in

the particle) is comparable with thermal energy kBT At δ > kBT

the levels are discrete and the particle loses its metallic properties.The catalytic activity of the small metallic particles starts to beevident when the value of δ is close to kBT Consequently, it is

possible to evaluate the particle size at which the catalytic

properties become evident For metals, Fermi energy EF is

approximately 10 eV, at room temperature 300 K the value δ = kBT

= 0.025 eV and, consequently N ≈ 400; a particle consisting of 400atoms has a diameter of ~2 nm In fact, the majority of dataconfirm that the physical and catalytic properties start to changemarkedly when the particles reached the size of 2–8 nm In addition

to the discussed primary electron effect, there is a secondaryelectron effect This effect is caused by the fact that a largefraction of atoms is situated on the surface of small particles Theelectron configuration of these atoms is different from one of theatoms distributed inside the particle The secondary electron effecthas a geometrical source and also leads to changes in the catalyticproperties

The geometrical effect of catalysis depends on the number ofatoms distributed on the surface (on the faces), on the edges andtops of the small particle because these atoms have a differentcoordination The atoms situated on the faces have a highercoordination in comparison with the atoms on the tops and theedges If the atoms in the small coordination are catalytically mostactive, the catalytic activity increases with decreasing particle size

In another case, if the atoms located on the faces are catalyticallyactive, the rate of the catalysed reaction will be increased by largerparticles

A specific role in catalysis is played by the carrier because theatoms of the catalyst which are in direct contact with the carriermay change their electronic structure because of the formation ofbonds with the carrier It is evident that as the number of atomsthat are in contact with the carrier increases, the effect of thecarrier on catalytic activity becomes stronger It is clear that the

effect of the carrier is relatively small for large particles butincreases and becomes quite strong with a decrease in the particlesize.

Metallic alloys (for example, alloys of catalytically inert metals

of group I with metals of group VIII) are used as a catalystbecause of the dilution of the metal-catalyst in the alloy increasescatalytic activity This is similar to an increase in catalytic activitywith a decrease in the nanoparticle size To a first approximation,the similarity of the effects of a decrease in the particle size andmelting is caused by the fact that the valence electrons of everymetal in such alloys retain their affiliation and, consequently, acatalytically inert metal (for example, copper) acts as a diluent forthe particles of the catalytically active metal

Usually, the nanoparticles show catalytic activity in a verynarrow size range For example, Rh catalysts, produced by thedissociation of Rh6(CO)16 clusters, fixed to the surface of dispersesilica, catalyse the reaction of hydrogenation of benzene only when

the particle size is 1.5–1.8 nm, i.e only particles of Rh12 arecatalytically active in relation to this reaction The high selectivity

of the catalytic activity is also characteristic of nanoparticles ofwidely used catalysts such as palladium and platinum For example,the hydrogenation of ethylene was studied at a temperature of

520 K and a hydrogen pressure of 1 atm on a platinum catalystdeposited on SiO2 or Al2O3 A distinctive maximum of the reactionrate is observed when the size of Pt nanoparticles is about 0.6 nm.This high sensitivity of catalytic activity to the size of small particlesconfirms the importance of the development of selective methods

of production of nanoparticles with an accuracy to 1–2 atoms Thevery narrow size distribution of the nanoparticles is essential notonly for catalysis but also for microelectronics

A new area of catalysis of small particles is photocatalysis usingsemiconductor particles and nanostructured semiconductor films Forexample, this method is promising for photochemical purification ofeffluents to remove various organic contaminants by means of theirphotocatalytic oxidation and mineralisation

Detailed analysis of the effect of the size of small particles ofmetals and alloys, deposited on a carrier, can be found in [39] andalso in reviews [40, 41] concerned with catalysis using metallicalloys and palladium

Catalysis on small metallic particles may be regarded as thechemical size effect For example, nickel or palladium nanoparticles

on a SiO2 substrate are used as catalysists for hydrogenation of

benzene Nanoparticles are produced by decomposition oforganometallic complexes A decrease in the metallic particles size

is accompanied by an increase in specific catalytic activity, i.e the

activity related to 1 surface atom of the metal Let us consider thereaction of hydrogenation of benzene at a temperature of 373 Kand at benzene C6H6 and hydrogen H2 pressures of 6700 and

46700 Pa, respectively In this reaction, the specific catalyticactivity of nickel nanoparticles is increased 3–4 times when theparticle size become smaller than 1 nm and the dispersion tends tounity (Dispersion is the ratio of the number of atoms situated onsurface to the total number of the atoms in the particle.) Incatalysis on palladium nanoparticles with the dispersion close tounity, the identical effect in the same reaction is observed at 300

K A study of hydrogenolysis of ethane C2H6 at a temperature of

473 K and a pressure of C2H6 and H2 equal to 6700 and 26700 Pashowed that the very rapid increase in the specific catalytic activity

of the nickel nanoparticles is observed when their dispersion close

to unity

The rate of the reaction of hydrogenolysis of cyclopentane andmethylcyclopentane, related to 1 surface atom of the metal-catalyst,changes rapidly when the fraction of the surface atoms in thenanoparticle of the metal-catalyst (Pt, Ir, Pd, Rh deposited on glass,SiO2 or Al2O3) approaches unity [39]

Another chemical size effect is the shift of binding energy

3d5/2 of the internal level of palladium in relation to the size ofpalladium particles [39, 41] For palladium particles larger than 4–

5 nm, the binding energy of the 3d5/2 level is ~335 eV, i.e it is equal

to the value characteristic of bulk palladium A decrease in the size

of palladium nanoparticles from 4 to 1 nm is accompanied(irrespective of whether the material of the substrate is a conductor(carbon) or and insulator (SiO2, Al2O3, zeolites)) by an increase of

the binding energy of the 3d5/2 level The most probable reason forthe positive shift is the size dependence of the electronic structure

of palladium, namely the decrease of the number of valence electrons An identical shift of the binding energy of Pt 4f7/2 of theinternal level is recorded in the case of platinum nanoparticles [39].The increase of the chemical activity of thin-film hetero-structures is also a chemical size effect For example, in two-layeroxide heterostructures MgO/Nb2O5 the reaction of the type

takes place spontaneously at temperatures 800–1000 K lower thanthe temperature of the reaction between normal coarse-grainedoxides.

Hybrid nanocomposites of the metal–polymer type are produced

by forming nanoparticles in a specially prepared polymer matrix [42,43] Polymer composites with metallic nanoparticles are used aselectrically conducting film composite materials, and the amount ofthe filler in the matrix may reach 90 vol % The introduction ofmetal ions into polymer fibers makes it possible to produce coloredlightguides suitable for application in computer equipment Theoptical properties of polymers with fillers of nanoparticles of metals,alloys or semiconductors (CdS, CdSe, InP, InAs) are interesting.Because of light machining and the possibility of producing filmsfrom these polymer nanocomposites, they can be used for theproduction of optical elements and light filters

The nanoparticles and nanolayers are used widely in moderntechnology Multilayered nanostructures are used in the production

of microelectronic devices A suitable example are

layer-heterogeneous nanostructures, i.e superlattices in which superthin

layers (with a thickness from several to hundred of periods of thecrystal lattice or ~1–50 nm) of two different substances, forexample, oxides, alternate The structure represents a crystal inwhich in addition to the conventional lattice of periodicallydistributed atoms, there is a superlattice of repeating layers ofdifferent composition Owing to the fact that the thickness of thenanolayer is comparable with de Broglie wavelength of the electron,the quantum size effect is realised in superlattices in electronicproperties Utilising the effect of size quantisation in multilayerednanostructures enables the production of electronic devices withincreased operating speed and information capacity The simplestelectronic device of this type is, for example, the AlAs/GaAs/AlAstwo-barrier diode, consisting of a layer of gallium arsenide with athickness of 4–6 nm, distributed between two layers of aluminiumarsenide AlAs, with a thickness of 1.5–2.5 nm

Of special interest are magnetic nanostructures characterised bygiant magnetoresistance They are in the form of multilayered films

of alternating layers of ferromagnetic and non-magnetic metals, forexample, a ferromagnetic layer Co–Ni–Cu and a nonmagneticcopper layer alternate in the Co–Ni–Cu/Cu nanostructure Thethickness of the layers is of the order of the free path of the

electron, i.e several tens of nanometers Changing the strength of the applied external magnetic field from 0 to some value of H it

is possible to change the magnetic configuration of the multilayerednanostructure in such a manner that the electrical resistance willchange in a very wide range This makes it possible to utilise themagnetic nanostructures as detectors of the magnetic field Thehighest value of giant magnetoresistance in the Co–Ni–Cu/Cunanostructure is obtained for very thin layers of copper, thicknessapproximately 0.7 nm.

The development of electronics over a period of several decadeshas also progressed along the path of miniaturisation The first

‘jump’ in the development of electronic technology was thetransition from vacuum electron valves to the transistor The secondjump is associated with the application of integrated microcircuits.The transition to integrated microcircuits became possible afterunderstanding that all elements of the electronic circuit can beproduced from the same material of the semiconductor type, instead

of producing them from different materials Silicon is such amaterial The application of the material of the same type enabledthe construction of all elements of the electronic circuit directly inthe same specimen of this material and, connecting the elementstogether, produce an efficient microchip The first necessity fordecrease in the size of the electronic circuits came from militaryand space authorities of the USA, former USSR, Europeancountries, and Japan, who supported appropriate research projects

If the first simplest chips (1959) consisted of tens of elements,then in 1970, microcircuits included up to 10 000 elements Advances

in electronics were accompanied by a rapid decrease of the cost

of electronic devices (Fig 1.8) In 1958, the cost of one transistor

Fig 1.8 Decrease in the minimum characteristic size of electronic components and

growth of the volume of sales of electronic products [44].

was approximately 10 US dollars, and in the year 2000 this moneywould purchase a microcircuit with tens of millions of transistors[44] In currently available mass-produced microcircuits,approximately 1000 electrons are required for switching on/switching off a transistor At the end of the first decade of the 21stcentury, the required number of electrons will decrease to 10 as aresult of miniaturisation [44] and work is already been carried out

to develop a single-electron transistor [45]

Semiconductor heterostructures, produced from two or moredifferent materials, are of special interest for electronics In theseheterostructures, an important role is played by the transition layer,

i.e the interface between two materials In addition to this,

according to [46], the technical device in semiconductorheterostructures is the interface itself

Semiconductor heterostructures are fabricated from materialscontaining such elements as Zn, Cd, Hg, Al, Ga, In, Si, Ge, P, As,

Sb, S, Se, Te These elements belong to the groups II–VI of theperiodic table Silicon occupies the most important place in thetechnology of electronic materials, like steel in the production ofconstructional materials In addition to silicon, electronics requiresemiconductor compounds AIIIBV and their solid solutions, and also

AIIBVI compounds Of the compounds of the AIIIBV type, galliumarsenide GaAs is used most widely, and of the solid solutions it is

AlxGa1–xAs The application of solid solutions makes it possible toproduce heterostructures with a continuous but not jump-likevariation of composition The width of the forbidden band in theseheterostructures also changes continuously

In the production of the heterostructures, it is important to matchthe parameters of the crystal lattices of two contacting materials

If the two materials with greatly differing lattice constants grow oneach other, then an increase in the thickness of the layers results

in the formation of high strains at the interface and mismatchdislocations appear Strains appear irrespective of whether thetransition between the two layers is smooth or not To reduce thestrains, the lattice constants of the two materials should differ aslittle as possible Therefore, special attention in the study ofheterostructures is given to solid solutions of the AlAs–GaAssystem because the arsenides of aluminium and gallium have almostthe same lattice constants Single crystals of GaAs are an idealsubstrate for growing the heterostructures Another natural substrate

is indium phosphide InP which is used in combination with solidsolutions GaAs–InAs, AlAs–AlSb, and others

A breakthrough in making thin-layer heterostructures took placewith the development of a technology for the growth of thin layers

by molecular beam epitaxy and liquid-phase epitaxy It becamepossible to growth heterostructures with a very sharp interface.Consequently, it is possible to position the two heteroboundaries soclose to each other that the size quantum effects play thecontrolling role in this intermediate space The structures of thistype are referred to as quantum wells In quantum wells, the meannarrow-band layer has a thickness of several tens of nanometerswhich results in splitting of the electronic levels because of thesize quantisation effect This effect in the form of a characteristicstepped structure of optical spectra of absorption of the GaAs–AlGaAs semiconductor heterostructure with the superthin GaAslayer (quantum well) was detected for the first time by the authors

of [47] They also found the shift of characteristic energies with

a decrease in the thickness of the quantum well (GaAs layer) Inquantum wells, superlattices and other structures with very thinlayers and high strains may form without the formation ofdislocations and, consequently, it is not necessary to match thelattice parameters [48] Heterostructures, especially doubleheterostructures, including quantum wells, quantum wires andquantum dots, enable the control of various fundamental parameters

of semiconductor crystals, such as the width of the forbidden band(energy gap), the effective mass and mobility of charge carriers,the electron energy spectrum

The density of states N(E) in a three-dimensional (3D)

semi-conductor is a continuous function A decrease in the dimensionality

of the electron gas results in a change of the energy spectrum fromcontinuous to discrete as a result of its splitting (Fig 1.9) Thequantum well is a two-dimensional (2D) structure in which chargecarriers are restricted in the direction normal to layers, and maymove freely in the plane of the layer In quantum wires, the chargecarriers are already restricted in two directions and move freelyonly along the wire axis The quantum dot is a zero-dimensional(0D) structure where the charge carriers are already restricted inall three directions and are characterised by a completely discreteenergy spectrum

The size of the quantum dots produced by molecular beamepitaxy and lithography ranges from 1000 to 10 nm; smallerquantum dots (1–20 nm), whose surface is protected by organicmolecules preventing aggregation of the particles, may be produced

by colloidal chemistry methods [49]

The application of nanostructures in electronics will lead tofurther miniaturisation of electronic devices with transfer tonanosized elements for producing processors of a new generation.The i386T M processor, produced by Intel Corporation in 1983,contained 275000 transistors and performed more than 5 millionoperations per second; the i486TM processor, produced in 1989,already contained 1 200000 transistors The most widely usedprocessor at the end of the 20th century and the beginning of the21st century, Pentium Pro®, contains 5.5 million transistors andcarries 300 million operations per second The size of thetransistors reached the smallest value available for currenttechnologies and, consequently, a further decrease in the size may

be achieved only by the application of nanotechnology A practicaldifficulty which must be overcome in making quantum dots andsingle-electron transistors is the time instability of structures with

a small number of atoms The stability of these quantum-electronicelements is determined by the jump (diffusion) of already a smallnumber of atoms Since the diffusion processes on the surface and

Fig 1.9 Density of states N(E) for charge carriers as a function of the dimensionality

of the semiconductor: (3D) three-dimensional semiconductor, (2D) quantum well, (1D) quantum wire, (0D) quantum dot.

at the interface of quantum-electronic elements are very fast,processes of failure of the elements or even their movement on thesubstrate as an integral unit are already detected at roomtemperature [50] A problem of the stability of nanoelectroniccircuits can be solved only using multicomponent materials,including oxides, carbides and nitrides of metals These compoundshave a high melting point and low mobility of atoms and,consequently, have high thermal and time stability.

X-ray and ultraviolet optics uses special mirrors withmultilayered coatings of alternating thin layers of elements withhigh and low density, for example, tungsten and carbon,molybdenum and carbon, or nickel and carbon; the thickness of apair of layers is about 1 nm, and the layers should be atomicallysmooth The possibility of producing multilayered x-ray mirrors isone of the factors determining the application in certain areas ofnanotechnology, such as x-ray lithography, on the one hand, and inastronomical and astrophysical investigations, on the other hand.The formation of x-ray mirrors has been described in sufficientdetail in [51], where multilayered nickel–carbon Ni/C nanostructureswith a period of ~4 nm were investigated Other optical deviceswith nanosized elements, intended for application in x-raymicroscopy, are Frenel zone sheets with the smallest width of thezone of approximately 10 nm, and diffraction gratings with a periodsmaller than 100 nm

F/S heterostructures, formed by alterating thin layers of aferromagnetic and a superconductor, are very interesting In theF/S heterostructures, the superconducting and ferromagnetic regionsare divided in space but are linked together through the interfacebetween the layers In most cases, ferromagnetic interlayers F are

produced using Fe, Co, Gd, Ni whose Curie temperature TC is

considerably higher than the superconducting transition temperature

Tsc of metals (Nb, Pb, V), forming the layer S Experimental

examination of these heterostructures started in [52] in which themethod of rf sputtering was used to produce two-layer sandwichesF/Pb Other methods of production of superlattices of the F/S typeare molecular-beam epitaxy, electron beam evaporation, and dcmagnetron sputtering [53] Generally, superconductivity andferromagnetism are antagonistic phenomena Primarily, thisantagonism is reflected in relation to the magnetic field Thesuperconductor tends to push out the magnetic field (Meisnereffect), and the ferromagnetic concentrates the force lines of themagnetic field in its volume (magnetic induction effect) From the

viewpoint of microscopic theory, the antagonism is reduced to thefollowing: In a superconductor, the attraction force between theelectron generates Cooper pairs, whereas the volume interaction inthe ferromagnetic tends to align the electronic forces paralelly.Taking this into account, the coexistence of the superconductingand ferromagnetic order in a homogeneous system is unlikely, but

it can easily be achieved in artificial multilayered systems consisting

of alternating ferromagnetic and superconducting layers (Fig 1.10).The F/S type heterostructures with the layers of atomic thicknessmay be used in electronic devices of the next generation as logicelements and switches of superconducting current [54, 55], andsuperconductivity can be controlled by means of a weak externalmagnetic field [56] It should be mentioned that the properties ofthe F/S multilayered systems, including the superconductingtransition temperature, depend on the thickness of theferromagnetic and superconducting layers In most cases, thethickness of the ferromagnetic layer is smaller than 1 nm, thethickness of the superconducting layer is from 10 to 40–50 nm [56]

It is interesting to note that the superconducting transition

temperature Tsc in the F/S heterostructures may not only

monotonically decreases but also oscillates with increasing thickness

of the layer F For example, in a Fe/Nb/Fe three-layer system, with

Fig 1.10 Multilayer heterostructures ferromagnetic-superconductor F/S: (a) double

layers, (b) triple layers, (c) superlattices

an increase of the thickness of the iron layer dFe from 0.1 to

0.8 nm, the temperature Tsc initially decreases from 7 to 4.5 K and,

subsequently, with an increase of dFe to 1.0–1.2 nm, Tsc increases

to 5 K and with a further increase of dFe to 3 nm, thesuperconducting transition temperature decreases to 3.2–3.4 K [57].The layers of the metal and alloy, for example, NbxTi1–x/Co orV/FexV1–x, may alternate in the F/S heterostructures The hetero-structures of the superconductor–ferromagnetic semi-conductor type(for example, NbN/EuO/Pb or NbN/EuS/Pb) [58] with a Josephsontunnelling transition are also interesting In these heterostructures,the thickness of the layer of the ferromagnetic semiconductor (EuO,EuS) varies from 10 to 50 nm, and the thickness of thesuperconducting layers is greater than 200 nm

Engineering applications have no other parts, which are working

in complicated critical conditions, as blades of gas turbines in jet engines The transfer to a new generation of gas-turbine enginesrequires constructional materials, whose strength and hardnesswould be 20% higher, fracture toughness 50% better, and wearresistance twice as large Actual tests showed that the use ofheat-resistant nanocrystalline alloys in gas turbines provides at leastone-half of the required improvement in the properties Ceramicnanomaterials are widely used for fabrication of parts working atelevated temperatures, under nonuniform thermal loads, and inaggressive environment Thanks to their superplasticity, ceramicnanomaterials serve as materials of intricately shaped highlyprecision products used in the aerospace technology Nanoceramicsbased on hydroxy-apatite possesses biocompatibility and highstrength and therefore is used in orthopedy for the making ofartificial joints and in stomatology for the making of dentures Thenanocrystalline ferromagnetic alloys of the Fe–Cu–M–Si–B systems(M is the transition metal of the groups IV–VI) are used asexcellent transformer soft magnetic materials with a very lowcoercive force and high magnetic permeability

turbo-The small grain size determines a long length of the intergranularinterfaces: At a grain size from 100 to 10 nm the interfacescontained 10–50% of the atoms of the nanocrystalline solid Also,grains may have various atomic defects (vacancies or theircomplexes, disclinations, and dislocations), whose number anddistribution differ from those in coarse grains 5 to 30 µm in size

If the dimensions of a solid in one, two or three directions arecomparable with characteristic physical parameters having thelength dimensionality (the size of magnetic domains, the electron free

path, the size of excitons, the de Broglie wavelength, etc.), sizeeffects will be observed for the corresponding properties Thus, sizeeffects imply a set of phenomena connected with changes inproperties of substance, which are caused by (1) a change in theparticle size, (2) the contribution of interfaces to properties of thesystem, and (3) comparability of the particle size with some physicalparameters having the length dimensionality Because of thesespecific features of the structure, the properties of thenanocrystalline materials greatly differ from those of usualpolycrystals Therefore, the decrease in the grain size is viewed as

an efficient method for adjustment and modification of properties

of solids In fact, there are reports on the effect of thenanocrystalline state on the magnetic properties of ferromagnetics(Curie temperature, coercive force, saturation magnetisation) andmagnetic susceptibility of weak para- and diamagnetics, on theeffect of shape memory on the elastic properties of metals andlarge changes of their heat capacity and hardness, on the variation

of the optical and luminescence characteristics of semiconductors,

on the appearance of the plasticity of boride, carbide, nitride andoxide materials which are relatively brittle in the normal coarse-grained state In nanocrystalline materials, the combination of highhardness and plasticity is usually explained by difficulties in theactivation of dislocation sources as a result of the small dimensions

of crystals, on the one hand, and by the presence of grain boundarydiffusional creep, on the other hand [13] The nanomaterials arecharacterised by very high diffusibility of the atoms at the grainboundaries, which is 5–6 orders of magnitude higher than that inthe normal polycrystals However, the mechanisms of diffusionprocesses in nanocrystalline substances have not yet beencompletely explained and the literature contains contradictingexplanations of this problem The problem of the microstructure ofthe nanocrystals, i.e the structure of the interfaces and their atomicdensity, the effect of the nanovoids and other free volumes on theproperties of nanocrystals requires solution

Usually, when discussing the nonequilibrium metastable state, it

is assumed that this state may be compared with some equilibriumstate which exists in reality For example, the metastable glass-like(amorphous) state corresponds to the equilibrium liquid state (melt).The specific feature of the nanocrystalline state in comparison withother well-known nonequilibrium metastable state of matter is theabsence of the equilibrium state corresponding to this state in thestructure and a long length of the boundaries

The nanocrystalline materials represent a special state ofcondensed matter, i.e macroscopic ensembles of ultrafine particles

up to several nanometers in size The unusual properties of thesematerials are determined by both the specific features of separateparticles (crystallites) and by their collective behaviour, whichdepends on the interaction between nanoparticles

The main scope of investigations of the nanocrystalline state is

to answer the following questions:

(1) Is there a sharp, distinctive boundary between the bulk andthe nanocrystalline states?;

(2) Is there some critical grain (particle) size below which thecharacteristic properties of nanocrystals become observable, andabove which the material behaves as a bulk one? In other words,

is the transition from the bulk to the nanocrystalline state the phasetransformation of the first order from the thermodynamic viewpoint?The answer to this question is important for the correct procedure

of experimental investigations of the nanocrystalline state and forunderstanding the results

At the first sight, the transition to the nanocrystalline state is not

a phase transformation because the size effects increase graduallywith a decrease in the size of isolated particles or grain size in bulknanomaterials However, all experimental investigations have beencarried out on materials with a high dispersion of the particle orgrain size Obviously, one can assume that the dispersion of thedimensions leads to “bluring” of the phase transformation, if such

a transformation exists This could be confirmed by experimentswith the detection of the size effect Such experiment should becarried out on materials of the same chemical composition butdifferent dispersion in grain size The important condition of thisexperiment is equal size of the particles or grains of materialinvestigated Only in this experiment it will be possible to excludecompletely the effect of the dispersion of the size of the particlesand determine whether the size dependence of a specific property

is a continuous and smooth function or whether it contains jumps,inflection points and other special features Unfortunately, at thepresent time it is not possible to carry out such experiments

In solid-state mechanics, successes have been achieved inunderstanding of the nanocrystalline solid as an ensemble ofinteracting grain boundary defects This approach is most useful instudying bulk nanomaterials For analysis of the symmetricproperties of the polycrystals as a function of the changes of the

characteristics scale of structural heterogeneity, i.e as a function

of the grain size, the authors of [59] used the gauge field theory[60, 61] This theory was developed to describe the structural andphysical properties of materials with defects According to [59], adecrease in the grain size is accompanied by a topologicaltransition from solitary waves of orientation-shear instability, whichare characteristic of the polycrystalline state, to spatial-periodicstructures of defects The formation of these defects leads to thetransition to the nanocrystalline state This topological transition in

an ensemble of grain boundary defects is accompanied by a largechange of the characteristics of connectivity and scaling parameters.The main aim of the present monograph is to discuss the effects

of the nanocrystalline state, detected in the properties of metals andcompounds The structure and dispersion (the size distribution of thegrains) and, consequently, the properties of nanomaterials depend

on the method of production of these materials Therefore, in thesecond and third chapters of the book we discuss briefly the mainmethods of production of nanocrystalline powders and bulknanocrystalline materials It should be mentioned that significantadvances in studying of the nanocrystalline state of solids wereachieved after 1985 as a result of improvement of the available anddevelopment of new methods of production of both disperse andbulk nanocrystalline materials

The particle size has the strongest effect on the properties ofnanocrystalline substances Therefore, the fourth chapter considersthe main methods of determination of the particle size Specialattention is given to the diffraction method of determining theparticle size This method is widely available and used

The fifth chapter is concerned with the specific features of thephysical properties of isolated nanoparticles (nanoclusters) andnanopowders These properties are determined by the small particlesize The methods of production of powdered nanomaterials havebeen developed sufficiently and have been known for more than 50years A large amount of relatively reliable experimental materialhas been collected for the properties of isolated particles (in mostcases, metallic particles), and an efficient theoretical base has beendeveloped for understanding their properties and structure It should

be noted that the particles of the nanopowders occupy anintermediate position between the nanoclusters and bulk solid.The chapters 6 and 7, analysing the structure and properties ofbulk nanomaterials, contain the most recent experimental data.Almost all the results described in these chapters have beenobtained after 1988 A great majority of the investigations of bulk

nanocrystalline substances and materials have been concentrated onseveral problems One of this problems is the microstructure of thebulk nanomaterials and its stability, the state and relaxation of grainboundaries The microstructure is studied by different electronmicroscopy, diffraction and spectroscopic techniques Theseinvestigations are closed to the works on the study of the structure

of bulk nanomaterials by indirect techniques (investigation of phononspectra, calorimetry, measurement of the temperature dependence

of microhardness, elasticity moduli, and electrokinetic properties)

It is expected that the bulk nanomaterials will be used most widely

as constructional and functional materials in new technologies andalso as magnetic materials Consequently, chapter 7 pays specialattention to the mechanical and magnetic properties of bulknanomaterials Discussion of the structure and properties of isolatednanoparticles and bulk nanomaterials should lead to united views onthe current state of the investigations of this special state of matterand find common and specific features of isolated nanoparticles andbulk nanomaterials

R e f e r e n c e s

1 H Gleiter Materials with ultrafine microstructure: retrospectives and

perspectives Nanostruct Mater 1, 1-19 (1992)

2. R Birringer, H Gleiter Nanocrystalline materials In: Encyclopedia of

Material Science and Engineering Suppl Vol.1 Ed R W Cahn (Pergamon

5 H.-E Schaefer Interfaces and physical properties of nanostructured solids.

In: Mechanical Properties and Deformation Behavior of Materials Having

Ultrafine Microstructure Eds M A Nastasi, D M Parkin, H Gleiter.

(Kluwer Academic Press, Netherlands, Dordrecht 1993) pp.81-106

6 R W Siegel What do we really know about the atomic-scale structures of

nanophase materials? J Phys Chem Solids 55, 1097-1106 (1994)

7. I D Morokhov, L I Trusov, S P Chizhik Ultra-Dispersed Metallic

Substances (Atomizdat, Moscow 1977) 264 pp (in Russian)

8 I D Morokhov, V I Petinov, L I Trusov, V F Petrunin Structure and

properties of small metallic particles Uspekhi Fiz Nauk 133, 653-692 (1981)

(in Russian)

9. I D Morokhov, L I Trusov, V N Lapovok Physical Phenomena in

Ultra-Dispersed Substances (Energoatomizdat, Moscow 1984) 224 pp (in Russian)

10. Yu I Petrov Physics of Small Particles (Nauka, Moscow 1982) 360 pp (in

Russian)

11. Yu I Petrov Clusters and Small Particles (Nauka, Moscow 1986) 368 pp.

(in Russian)

12 L N Larikov Structure and properties of nanocrystaline metals and alloys.

Metallofizika 14, 3-9 (1992) (in Russian)

13 L N Larikov Diffusion processes in nanocrystalline materials Metallofizika

i Noveishie Tekhnologii 17, 3-29 (1995) (in Russian)

14 L N Larikov Nanocrystalline compounds of metals Metallofizika i

Noveishie Tekhnologii 17, 56-68 (1995) (in Russian)

15 R A Andrievski The synthesis and properties of nanocrystalline refractory

compounds Uspekhi Khimii 63, 431-448 (1994) (in Russan) (Engl transl.: Russ Chem Reviews 63, 411-428 (1994))

16 A I Gusev Effects of the nanocrystalline state in solids Uspekhi Fiz Nauk

168, 55-83 (1998) (in Russian) (Engl transl.: Physics - Uspekhi 41, 49-76

(1998))

17 E A Rohlfing, D M Cox, A Kaldor Production and characterization of

supersonic carbon cluster beams J Chem Phys 81, 3322-3330 (1984)

18 H W Kroto, J R Heath, S C O’Brien, R F Curl, R E Smalley C60:

buckminsterfullerene Nature 318, 162-163 (1985)

19 J W Mintmire, B I Dunlap, C T White Are fullerene tubules metallic?

Phys Rev Lett 68, 631-634 (1992)

20. S Iijima Helical microtubules of graphitic carbon Nature 354, 56-58 (1991)

21 S Iijima, P M Ajayan, T Ichihashi Growth model for carbon nanotubes.

Phys Rev Lett 69, 3100-3103 (1992)

22. P Calvert Strength in disunity Nature 357, 365-366 (1992)

23. S Iijima Carbon nanotubes MRS Bulletin 19, 43-49 (1994)

24 M S Dresselhaus Future directions in carbon science Ann Rev Mater Sci.

27, 1-34 (1997)

25 K Sohlberg, R E Tuzun, B Sumpter, D W Noid Application of rigidbody dynamics and semiclassical mechanics to molecular bearings Nanotechnology

8, 103-111 (1997)

26 L Delzeit, C V Nguyen, R M Stevens, J Han, M Meyyappan Growth

of carbon nanotubes by thermal and plasma chemical vapour deposition

processes and applications in microscopy Nanotechnology 13, 280-284

(2002)

27 M Menon, D Srivastava Carbon nanotube “T junctions”: Nanoscale

metal-semiconductor-metal contact devices Phys Rev Lett 79, 4453-4456 (1997)

28 J Li, C Papadopoulos, J Xu Growing Y-junction carbon nanotubes Nature

402, 253-254 (1999)

29 C Papadopoulos, A Rakitin, J Li, A S Vedeneev, J M Xu Electronic

transport in Y-junction carbon nanotubes Phys Rev Lett 85, 3476-3479

(2000)

30 B C Satishkumar, P J Thomas, A Govindaraj, C N R Rao Y-junction

carbon nanotube Appl Phys Lett 77, 2530-2532 (2000)

31 B Q Wei, R Vajtai, Y Jung, J Ward, Y Zhang, G Ramanath, P M Ajayan.

Organized assembly of carbon nanotubes Nature 416, 495-496 (2002)

32 N G Chopra, R J Luyken, K Cherrey, V H Crespi, M L Cohen, S G.

Louie, A Zettl Boron nitride nanotubes Science 269, 966-967 (1995)

33 A Loiseau, F Willaime, N Demoncy, G Hug, H Pascard Boron nitride nanotubes with reduced numbers of layers synthesized by arc discharge.

Phys Rev Lett 76, 4737-4740 (1996)

34 A K Cheetham, H Terrones, M Terrones, R Castillo, J P Hare, H W Kroto, D R M Walton, K Prassides, S Ramos, W K Hsu, J P Zhang Metal particle catalysed production of nanoscale BN structures Chem Phys.

Lett 259, 568-573 (1996)

35 A I Kharlamov, S V Loichenko, N V Kirillova, V V Fomenko Heterogeneous synthesis of silicon carbide filaments Teoret i Eksper.

Khimiya 38, 49-53 (2002) (in Russian)

36 A I Kharlamov, N V Kirillova Gas-phase reactions of forming of nanothread-like silicon carbide from powdery silicon and carbon Teoret i

Eksper Khimiya 38, 54-58 (2002) (in Russian)

37 A I Kharlamov, N V Kirillova, S V Kaverina Hollow silicon carbide

nanostructures Teoret i Eksper Khimiya 38, 232-237 (2002) (in Russian)

38. Science and Applications of Nanotubes Eds D Tomanek, R J Enbody.

(Kluwer Academic Publishers: New York – Dordrecht – Moscow 2002) 398

pp

39 M Che, C O Bennet The influence of particle size on the catalytic

properties of supported metals Advances in Catalysis 36, 55-172 (1989)

40 V Ponec Catalysis by alloys in hydrocarbon reactions Advances in Catalysis

32, 149-214 (1983)

41 Z Karpinski Catalysis by supported, unsupported, and electron–deficient

palladium Advances in Catalysis 37, 45-100 (1990)

42 A D Pomogailo Hybrid polymer-inorganic nanocomposites Uspekhi Khimii

69, 60-89 (2000) (in Russian) (Engl Transl.: Russ Chem Reviews 69,

53-80 (2000))

43. A D Pomogailo, A S Rosenberg, U E Uflyand Nanoscale Metal Particles

in Polymers (Khimiya, Moscow 2000) 672 pp (in Russian)

44 J S Kilby Turning potential into realities: The invention of the integrated

circuit (Nobel Lecture) Uspekhi Fiz Nauk 172, 1103-1109 (2002) (in

Russian)

45 D L Klein, R Roth, A K L Lim, A P Alivisatos, P L McEuen A single-electron transistor made from a cadmium selenide nanocrystal Nature

389, 699-701 (1997)

46 H Kroemer Nobel Lecture: Quasielectric fields and band offsets: teaching

electrons Rev Modern Phys 73, 783-793 (2001)

47 R Dingle, W Wiegmann, C H Henry Quantum states of confined carriers

in very thin AlxGa1−xAs - GaAs - AlxGa1−xAs heterostructures Phys Rev.

Lett 33, 827-830 (1974)

48 Zh I Alferov Nobel Lecture: The double heterostructure: concept and its

applications in physics, electronics and technology Rev Modern Phys 73,

52 J J Hauser, H C Theuerer, N R Werthamer Proximity effects between

superconducting and magnetic films Phys Rev 142, 118-126 (1966)

53 B Y Jin, J B Ketterson Artificial metallic superlattices Advances in

Physics 38, 189-366 (1989)

54 A I Buzdin, A V Vedyayev, N V Ryshanova Spin-orientation-dependent

superconductivity in F/S/F structures Europhysics Lett 48, 686-691 (1999)

55 L R Tagirov Low-field superconducting spin switch based on a

super-conductor/ferromagnet multilayer Phys Rev Lett 83, 2058-2061 (1999)

56 Yu A Izyumov, Yu N Proshin, M G Khusainov Competition between superconductivity and magnetism in ferromagnet/superconductor

heterostructures Uspekhi Fiz Nauk 172, 113-154 (2002) (in Russian) (Engl transl.: Physics - Uspekhi 45, 109-148 (2002))

57 Th Mühge, N N Garif ’yanov, Yu V Goryunov, G G Khaliullin, L R Tagirov, K Westerholt, I A Garifullin, H Zabel Possible origin for oscillatory superconducting transition temperature in superconductor/

ferromagnet multilayers Phys Rev Lett 77, 1857-1860 (1996)

58 A S Borukhovich Quantum tunneling multilayers and heterostructures with

ferromagnetic semiconductors? Uspekhi Fiz Nauk 169, 737-751 (1999) (in Russian) (Engl transl.: Physics - Uspekhi 42, 653-667 (1999))

59 O B Naimark Nanocrystalline state as a topological transition in an

ensemble of grain-boundary defects Fiz Metall Metalloved 84, 5-21 (1997) (in Russian) (Engl Transl.: Phys Metal Metallogr 84, 327-337 (1997))

60 E Kröner On gauge theory in defect mechanics trends in application of pure

mathematics to mechanics In: Lecture Notes in Physics Eds E Kröner and

K Kinchgassner (Springer, Heidelberg 1986) pp.281-296

61. A Kadic, D G B Edelen A Gauge Theory of Dislocations and Disclinations

(Springer, Berlin 1983) 186 pp.

+D=FJAH

2 Synthesis of Nanocrystalline Powders

2.1 GAS PHASE SYNTHESIS

Isolated nanoparticles are usually produced by evaporation of metal,alloy or semiconductor at a controlled temperature in theatmosphere of a low-pressure inert gas with subsequentcondensation of the vapour in the vicinity or on the cold surface.This is the simplest method of producing nanocrystalline powders

In contrast to vacuum evaporation, the atoms of the substance,evaporated in a rarefied inert atmosphere, lose their kinetic energymore rapidly as a result of collisions with gas atoms and formsegregations (clusters)

The first studies in this topic were carried out in 1912 [1,2]:examination of evaporation of Zn, Cd, Se and As in vacuum, andalso in hydrogen, nitrogen and CO2 showed that the size of theproduced particles depends on the pressure and atomic weight ofthe gas The authors of [3] evaporated gold from a heated tungstenfilament at a nitrogen pressure of 0.3 mm Hg (40 Pa), and produced

in the condensates spherical particles with a diameter of 1.5–10 nm(the mean diameter approximately 4 nm) They found that theparticle size depends on the gas pressure and, to a lesser degree,

on the rate of evaporation The particle size was determined by thehigh-resolution electron microscopy The condensation of vapours

of aluminium in H2, He and Ar at different gas pressures made itpossible to produce particles with a size of 20–100 nm [4] Later,the method of combined condensation of metal vapours in Ar and

He was used to produce Au–Cu and Fe–Cu highly dispersed alloys,formed by spherical particles with a diameter of 16–50 nm [5, 6]

A variant of the condensation of metal vapours in a gas atmosphere

is the method of dispersion of a metal by means of an electric arc

in a liquid with subsequent condensation of metallic vapours in liquidvapours, proposed as early as in the 19th century [7]; later, thismethod was improved by the authors of [8–10] The first extensivereview [11], concerned with detailed discussion of the condensationmethod and the formation of highly dispersed metal particles bycondensation of metallic vapours, was published in 1969 Severaltheoretical special features of condensation in supersaturatedvapours, which takes place by means of the formation and growth

of nuclei (clusters), were discussed in a review in [12]

The nanocrystalline particles with a size of ≤ 20 nm, produced

by evaporation and condensation, are spherical, and large particlesmay be faceted The size distribution of nanocrystals islogarithmico-normal and is described by the function

The systems using the principle of evaporation and condensation,differ in the method of input of evaporated material; the method ofsupplying energy for evaporation; the working medium; setup of thecondensation process; the system for collecting the producedpowder

Evaporationof ametal maytakeplacefrom a crucible or the metalmay be fed into the zone of heating and evaporation in the form ofwire, injected metallic powder or a liquid jet The metal may also bedispersed with a beam of argon ions.Energymaybe supplied directly

by heating, the passage of electric current through a wire, electric arcdischarge in plasma, induction heating with high- and superhighfrequency currents, laser radiation, and electron-beam heating

Evaporation and condensation may take place in vacuum, in astationary inert gas, or in a gas flow, including in a plasma jet.Condensation of the vapour–gas mixture with a temperature of5000–10000 K may take place during its travel into the chamber with

a large section and the volume filled with a cold inert gas; coolingtakes place both as a result of expansion and contact with the coldinert atmosphere There are systems in which two jets travel coaxiallyinto the condensation chamber: the vapour–gas mixture is suppliedalong the axis, and a circumferential jet of a cold inert gas travelsalong its periphery As a result of turbulent mixing, the temperature

of metal vapours decreases, supersaturation increases and rapidcondensation takes place Favourable conditions for the condensation

of metallic vapours are generated in adiabatic expansion in a Lavalnozzle, when rapid expansion results in the formation of a steeptemperature gradient and almost instantaneous vapour condensationtakes place

An independent task is the collection of the nanocrystalline powderproduced by condensation, because the individual particles of thispowder are so small that they are in constant Brownian motion andremain suspended in the gas, not settling under the effect of theforces of gravity The produced powders are collected using specialfilters and centrifugal deposition; in some cases, the particles aretrapped by a liquid film

The main relationships of the formation of nanocrystallineparticles by the method of evaporation and condensation are asfollows, [11, 14]:

1 The nanoparticles form during cooling of the vapours in the tion zone The size of this zone increases with a decrease of the gaspressure; the internal boundary of the condensation zone is in the vicin-ity of the evaporator, and its external boundary may, with a decrease ofgas pressure, extend outside the limits of the reaction vessel; at a pres-sure of several hundreds of Pa, the outer boundary of the condensationzone is situated inside the reaction chamber with a diameter of ≥ 0.1 mand convective gas flows play a significant role in the condensationprocess;

condensa-2 When the gas pressure is increased to several hundreds of Pa, the meanparticle size initially rapidly increases and then slowly approaches thelimiting value in the pressure range greater than 2500 Pa;

3 At the same gas pressure, the transition from helium to xenon, i.e from

a less dense inert gas to an inert gas with a higher density, is nied by a large increase in the particle size