Nitrides: Transition Metal Solid-State Chemistry · Nitrides: Transition Metal Solid-State Chemistry Walter Lengauer TU-Wien—Laboratory of Physical Metallurgy, Vienna, Austria 1

Nitrides: Transition MetalSolid-State Chemistry

Walter LengauerTU-Wien—Laboratory of Physical Metallurgy, Vienna, Austria

1 Introduction 12 Structure and Bonding 13 Preparation 44 Characterization 55 Thermodynamics and Specific Structural Features 66 Properties of FCC Transition Metal Nitrides 137 Uses 188 References 21

1 INTRODUCTION

Together with the transition metal carbides and borides, thetransition metal nitrides belong to a family of materials thatfeatures an unusual combination of outstanding properties,among which are exceptional hardness, high melting point,metallic luster with sometimes vivid colors, and simplemetallic structures combined with excellent electrical andthermal conductivities. This combination of these propertieshas attracted considerable attention and has resulted innumerous technical applications.1–4 Lengauer5 has compiledseveral data and features of transition metal nitrides andcarbonitrides. Many of the nitride phases closely resemblemetallic alloys, with broad ranges of homogeneity, particularlyif the structure of the metallic host lattice is one of the typesencountered in typical metals. They often form solid solutionswith structurally related compounds, such as other transitionmetal nitrides or carbides, and are easily wetted by liquidmetals of the iron group.

The chemical stability of transition nitrides is considerable;they are not readily attacked by dilute acids except by oxidizingacids and hydrofluoric acid or alkaline solutions. The thermalstability is determined by their free energies of formation;the stability decreases with increasing group number. WhileTiN, ZrN, and HfN can be melted without decomposition atambient pressure, the nitrides of the other groups decomposeby liberating nitrogen before melting points are reached.

Table 1 gives a brief survey of the most important propertiesof the well-characterized binary transition metal nitrides with amaximum nitrogen content of 50 mol% N. It should, however,

Dedicated to Professor Peter Ettmayer on the occasion of his 80th birthday.

Update based on the original article by Walter Lengauer and Alexander Eder,Encyclopedia of Inorganic Chemistry © 2005 John Wiley & Sons, Ltd.

be emphasized that practically all properties are significantlyinfluenced by the nitrogen to metal ratio (Section 6) and byimpurities.

2 STRUCTURE AND BONDING

2.1 Structures: General Features

Transition metal nitrides, together with the structurallyclosely related transition metal carbides, belong to the familyof ‘‘interstitial compounds’’ or alloys. The common featuresof this class of materials are the very simple metallicstructures with the smaller nitrogen atoms in the interstitialvoids of the frequently close-packed host lattice. The simpleconceptual Hagg model, in which the basically metalliccharacter of the host metal is thought to be only modifiedby insertion of nonmetal atoms into the voids, had to besubstantially restructured in the light of modern bondingtheories in order to account for the strong binding energiesobserved between transition metal atoms and nonmetalatoms. Nevertheless, the rather approximate interstitial alloyconcept has proved helpful for gaining insight into the basicprinciples governing the structures of interstitial alloys andcompounds.

Because of their structural similarities, transition metalnitrides usually form solid solutions with transition metalcarbides. However, there are important differences, possiblyowing to the different bonding character of nitrogen ascompared to that of carbon. Transition metal carbides andnitrides generally have face-centered cubic (fcc) or hexagonalclose-packed (hcp) metal lattices with nonmetal atomsmore or less randomly distributed on the interstitial sites;deviations from this simple concept are much more frequentlyencountered with nitrides. If the immediate neighborhood ofthe nonmetal X (X = C, N) is considered, one can define thestructural elements of interstitial alloys.6 In the group 4 and 5transition metals, the common structural element of the nitridesand carbides is the T6X octahedron. With increasing radiusratio rX/rT, the trigonal prismatic group T6X is favored overthe octahedral group. Therefore, the transition metal carbidesof groups 6–9 feature trigonal prismatic T6X elements indifferent crystallographic arrangements. In transition metalnitrides, a third structural element is occasionally observed,particularly in group 5 compounds, consisting of T5X squarepyramids. These prevail even in the complex nitrides, suchas the Z-phases NbCrN, TaCrN, NbMoN, and TaMoN withthe filled �-CuTi-type structure. With this square pyramid,which has lower symmetry than an octahedron, such nitrideshave more complicated structures than the correspondingcarbides.

Because of the formation of structural elements otherthan T6N octahedra, the fcc structure of δ-nitrides isdestabilized increasingly with increasing group numberand, within a group, with increasing periodic number. For

Encyclopedia of Inorganic and Bioinorganic Chemistry, Online © 2011–2015 John Wiley & Sons, Ltd.This article is © 2015 John Wiley & Sons, Ltd.This article was published in the Encyclopedia of Inorganic and Bioinorganic Chemistry in 2015 by John Wiley & Sons, Ltd.DOI: 10.1002/9781119951438.eibc0146.pub2

2 NITRIDES: TRANSITION METAL SOLID-STATE CHEMISTRY

Tab

le1

Bul

kpr

oper

ties

ofim

port

antt

rans

itio

nm

etal

nitr

ides

Nit

ride

Hom

ogen

eity

Str

uctu

reL

atti

ceD

ensi

tyC

olor

Mic

roha

rdne

ssM

elti

ngH

eatc

ondu

ct.(b

)T

herm

alex

p.E

lect

r.S

uper

cond

.ra

nge(a

)(B

rava

ispa

ram

eter

(gcm

−3)

(roo

mte

mp.

)po

int(b

)(r

oom

tem

p.)

coef

fici

enta

t100

0K

resi

stan

cetr

ansi

tion

tem

p.1

−x

atla

ttic

e)(r

oom

tem

p.)

(GP

a)(K

)(W

m−1

(106

K−1

)(r

oom

tem

p.)(b

)T

c(K

)17

00K

(nm

)K

−1)

com

p.(1

−x)

(μ�

cm)

TiN

1−x

0.50

fcc

0.42

15–

Met

alli

cgr

ay23

––

––

<1

1.00

0.42

425.

39(c

)G

olde

nye

llow

1730

5029

9.5

275.

8Z

rN1−

x0.

72fc

c0.

4585

–Y

ello

w–

––

––

–1.

000.

4570

7.32

(c)

Yel

low

1330

0025

7.8

2410

.47

HfN

1−x

0.85

fcc

0.45

23–

Lig

htye

llow

13.9

(0.9

4)–

––

–8.

7(0

.85)

1.06

0.45

1613

.83(c

)D

ark

gree

nye

llow

1633

3031

8.5

276.

92V

N1−

x0.

70fc

c0.

4060

6.05

Bro

wn

yell

ow13

.0–

–11

.0–

2.7

(0.7

6)1.

000.

4138

6.04

Bro

wn

5.7

2350

1110

.865

8.9

NbN

1−x

0.70

fcc

0.43

808.

24(0

.92)

Pal

eye

llow

13.0

––

9.5

(0.9

2)–

13.8

(0.8

4)1.

000.

4392

8.16

Pal

eye

llow

11.0

(d)

3.8

10.2

(0.9

5)60

17.2

TaN

1−x

0.72

(220

0K

)fc

c0.

4345

––

––

––

––

1.00

0.42

3815

.9(c

)G

ray

yell

ow32

(d)

–8.

0–

8.9

(0.9

4)T

aN1.

00he

xa0

.519

0–

––

(d)

9.5

–12

8–

c0.2

911

CrN

1.00

(150

0K

)fc

c0.

4148

6.14

Gra

y11

1300

(d)

11.7

2.3

640

(f)

CrN

1−x

0.31

(e)

hcp

a0.4

750(g

)6.

51G

ray

12(g

)–

––

–(f

)

c0.4

430(g

)

0.50

a0.4

796

––

14(b

)18

00–

–81

–c0

.447

0M

oN1−

x0.

40fc

c0.

4139

––

––

––

––

0.62

(120

0K

)0.

4162

9.84

(c)

Gra

y17

(b)

––

9.3

–5.

08M

oN–

hex

a0.5

745

9.1(c

)G

ray

––

––

–15

.1c0

.612

2

(a)N

itro

gen-

rich

phas

ebo

unda

ryde

pend

son

nitr

ogen

pres

sure

.(b

)E

xact

com

posi

tion

unkn

own,

prob

ably

clos

eto

2:1

(CrN

1−x,

MoN

1−x)

or1:

1(a

llot

hers

)st

oich

iom

etry

.(c

)X

-ray

dens

ity.

(d)D

ecom

pose

s.(e

)C

alcu

late

d.(f

)N

otsu

perc

ondu

ctin

g.(g

)E

xact

com

posi

tion

unkn

own.


NITRIDES: TRANSITION METAL SOLID-STATE CHEMISTRY 3

example, the nitrides of group 4, δ-TiN1−x, δ-ZrN1−x, andδ-HfN1−x crystallize in the sodium chloride structure. No low-temperature modifications of these stoichiometric nitrides areknown. In group 5, all three metals form stoichiometric fccnitrides also, but δ-VN1.00 transforms below 205 K into atetragonal modification,7 δ-NbN1.00 changes below around1320 ◦C into hexagonal η-NbN,8 and δ-TaN1.00 convertsbelow around 1920 ◦C congruently into hexagonal ε-TaN.9

In group 6, these fcc nitrides are increasingly less stable: fccCrN is nearly a line compound (i.e., one with a very narrowrange of stoichiometry) at low temperature with a reducedthermochemical stability; it transforms below approximately280 K into a tetragonal antiferromagnetic compound.10 fccγ-MoN1−x apparently does not occur with a nitrogen contenteven approaching the ideal composition MoN1.00 (exceptperhaps at very high pressures) but it is rather in the vicinityof MoN0.50. fcc MoN0.50 also transforms into a tetragonalmodification with an ordered arrangement of nitrogen atomsin the interstitial sites. This tendency proceeds further in theMn–N and Fe–N systems, where ferromagnetic fcc phaseswith the stoichiometry T4N are observed with an orderedarrangement of N atoms.11 Ordered manganese nitride phaseswith respect to the arrangements of nitrogen and a concurrenttetragonal distortion of the metal host lattice are found. An fccMnN1−x can be prepared by high-rate reactive sputtering.12

An extensive compilation of nitride phases is contained inPearson’s handbook13 and in the ASM binary phase diagramcompilation.11

The different stacking of close-packed metal layers can bedescribed by the Jagodzinski–Wyckoff notation, where close-packed stacking ABC, ABC, . . . is designated as ‘‘c’’ and hcpstacking AB, AB, . . . is designated as ‘‘h’’. Figure 1 showsthese structures and lists several metal nitride and carbidephases that can be described with this formalism.

2.2 Bonding

The bonding in transition metal carbides and nitrides canbe described as a mixture of metallic, covalent, and ioniccomponents.

The metallic character is immediately noticed in the highelectrical conductivities of these compounds. The bondingmechanism was described extensively by a variety ofapproaches for calculating the density of states (DOS) and,hence, the electron density, in fcc transition metal nitrides,carbides, and oxides.14 In the DOS of these compounds, thereis a minimum at a valence electron concentration (VEC) ofeight that corresponds to the stoichiometric composition ofgroup 4 carbides TiC, ZrC, and HfC. Figure 2 compares thedecomposed and the total DOS of TiN and TiC15 with respectto the s, p, and d bands. At low energies of <0Ryd, the sband is characterized by a large contribution of the nonmetal2s state. Above 0Ryd, the p band follows and there is asignificant contribution of d states to this band, too. At higherenergies, the d band composed of d states follows with the

(hhc)3

(hc)2h c

(hhcc)3

Figure 1 Close-packed structure types observed for various tran-sition metal nitrides. Large circles designate metal atoms and smallcircles designate nitrogen atoms. In h sequences, only 50% of theinterstices can be occupied. The following transition metal nitridescrystallize in these structures: h: β-V2N, β-Nb2N, β-Ta2N, ε-Cr2N,ζ -Mn2N, ε-Fe2N; c: δ-TiN1−x, δ-ZrN1−x, δ-HfN1−x, δ-VN1−x, δ-NbN1−x, δ-TaN1−x, γ-MoN1−x; (hc)2: ScNbN, ScTaN, Mn3TaN4−x;(hc)2 with metal and nonmetal sites exchanged: η-NbN, η-TaN;(hhc)3: η-Ti3N2−x, η-Hf3N2−x; (hhcc)3: ζ -Ti4N3−x, ζ -Hf4N3−x

Fermi energy EF located well within this band. The DOSat EF is higher for TiN than for TiC, consistent with theobserved superconducting properties (TiN, TC = 5.4 K; TiC,TC < 1.2 K) and room-temperature electrical conductivities(Section 6.3).

An illustrative picture of covalent bonding in thesecompounds can be drawn in the molecular orbital (MO)scheme16 where atomic orbitals of atoms are combined to formbonding and antibonding states. Figure 3 shows a schematicdrawing of these bonds between metal d and nonmetal porbitals in the (100) plane of the fcc structure, which canbe decomposed into two different bonding symmetries, t2gand eg. It can be shown that the calculated electron densitiesin this plane reflect the change of eg bonding symmetry inTiC to t2g bonding symmetry in TiN. Thus, the exchangein bonding strength from a strong Ti–C interaction in TiCtoward a more pronounced Ti–Ti interaction in TiN is to beexpected.



00

20

40

60

0−0.8 −0.4 0 0.4

E (Ryd)

EF

1.2EF

20

40

60

TiC

TiN

92,5

164

gt

(E)

gt

(E)

Figure 2 Density of states (DOS) for TiC and TiN. (Reproducedfrom Ref. 15. ‘Nitrides and Borides, Electronic Structure ofStoichiometric and Non-Stoichiometric Titanium Carbides andNitrides’ in The Physics and Chemistry of Carbides, 1990, p. 485,Ed. R. Freer. With kind permission of Springer Science+BusinessMedia.)

The ionic contribution to the binding mechanism oftransition metal nitrides and carbides can be estimated fromthe charge transfer from the metal atom to the nonmetal atom.It is about half an electron per atom, which contributes to theelectrostatic interaction of metal and nonmetal.

The bonding of 4d transition metal nitrides and carbidescan be treated in a very similar way; the nonmetal contributionto the electron density is about the same, whereby thecontribution from 4d electrons causes a greater spatialextension of electron density than in 3d compounds. In 5dtransition metal nitrides and carbides, relativistic effects owingto the heavy cores have to be taken into account.

3 PREPARATION

Transition metal nitrides can be prepared by allowing themetal or the metal hydride in powdered or compact form toreact with molecular or atomic nitrogen or flowing ammonia(equations 1–3):

Ti + 1

2N2 → TiN (1)

ZrH2 + 1

2N2 → ZrN + H2 (2)

Mo + NH3 → MoN + 3

2H2 (3)

eg

TiC

2 2

3

3 3

116 4

TiN

pdσ(a)

− −

− −

− −

−

−

−

−

−

+

+

+

+

+

+

+

+

+

−

−−−

−

+

+

+ + + +

(b)

pdπ pdσ

t2g

Figure 3 Bonding schemes (a) of different symmetries in the (100)plane and calculated electron densities and (b) of TiN and TiCshowing the change around Ti from eg symmetry TiC to t2g bondingsymmetry in TiN. (Reproduced with permission from Ref. 16. ©John Wiley & Sons, Inc.)

The pressure of molecular nitrogen at a given temperaturethat is necessary to obtain a specific nitride phase is defined bythe nitrogen potential (partial free energy of nitrogen) of thenitride. In flowing ammonia, very high nitrogen potentials canbe maintained.17 However, the decomposition of ammonia bythe reaction (equation 4)

2 NH3 → N2 + 3 H2 (4)

is very fast at increased temperatures, which are necessary fordiffusion of nitrogen in the nitrides, if bulk samples have to beprepared. Hence, the amount of decomposed ammonia is high.The nitriding potential of ammonia according to equation 4 is(equation 5)

RT• ln(pN2) = −DG–3RT• ln p

(3

2x

)+ 2RT• ln p(1 − x)

(5)where �G is the free energy of the above decompositionreaction and x is the portion of ammonia that has decomposed.The smaller the portion x, the higher is the nitriding potentialof ammonia.

The formation of nitrides from oxides is performed inthe presence of carbon as a reducing agent, according to theoverall reaction (equation 6)

TiO2 + 2C + 1

2N2 → TiN + 2CO (6)



This reaction usually proceeds over several intermediateproducts and the resulting nitride may contain oxygen andcarbon.

For the preparation of nitrides from metal powders, theexothermic heat of formation can be utilized in the self-sustaining high-temperature synthesis (SHS) process.18 Ifthe reaction is exothermic enough, that is, if the heatgenerated is sufficient to reach the reaction temperature,then the starting material is converted to the nitride andthe heat is dissipated. For the preparation of large crystals,zone-annealing techniques were developed.19 Similar to thezone-melting process, a heated zone is passed along a rodin a nitrogen atmosphere. The starting materials are eithersolid-state sintered nitride powders or metal powders pressedinto rods. The applied temperatures must be very high(about 3000 ◦C) in order to overcome the low diffusivityof nitrogen in nitrides, but they must not exceed themelting temperatures of the compounds in order to avoidincongruent melting and decomposition. Crystals on theorder of 1 cm3 were produced by this technique. Whilethey still may show some concentration gradients, theyare suitable for solid-state investigations such as neutrondiffractometry.

A great variety of methods were applied to the preparationof nitride layers. These can be divided into three main groups:

• Chemical vapor deposition (CVD)—a volatile transitionmetal compound such as a chloride is reacted inthe gas phase with ammonia, nitrogen/hydrogen, orammonia/hydrogen/nitrogen mixtures. The high-meltingnitride that forms as a reaction product in the gasphase nucleates on any solid present, for example,hardmetals, and forms compact strongly adhering layersat temperatures around 1000 ◦C.

• Physical vapor deposition (PVD)—the transition metalis evaporated by electron beam or plasma heating orsputtered by glow discharge and reacted with low-pressurenitrogen (typically in the range of 10−4 mbar). Numerousmodifications of PVD methods have been developed. Thenitride layers that are deposited on the substrate oftenhave high lattice defect concentrations and hence exhibitsignificant deviations from generally known bulk values(lattice parameters, hardness, composition, etc.).

• Surface nitridation is widely used for surface hardeningand can be achieved by subjecting the material surfaceto the action of ammonia, molecular or atomic nitrogen,or nitrogen ions. Casehardening techniques also make useof cyanide salt baths to form nitrides or carbonitrides onmaterial surfaces. Reactive plasma techniques and high-pressure techniques were developed to keep the workpiecetemperature low and to avoid thermally induced changesin the workpiece. Laser irradiation together with an N2jet directed on the heated spot is employed for locallylimited heating of the sample. This nitridation method isapplicable to titanium alloys (e.g., Ti6Al4V) used to createwear- and corrosion-resistant layers on steam turbine

blades for protection against droplets having supersonicvelocity.

4 CHARACTERIZATION

4.1 Chemical Analysis

Several procedures can be applied for the analysis ofnitrogen and impurities such as oxygen and carbides innitrides.

• Combustion analysis20 according to the Dumasmethod—the sample is heated at temperatures of up to1200 ◦C in oxygen and/or with oxidizing agents suchas V2O5, CuO, or Cu2O. The evolved N2 is measuredeither volumetrically in an azotometer or via a gaschromatograph21 equipped with a thermoconductivitydetector. With the latter equipment, C and H can also bedetermined simultaneously via the combustion productsCO2 and H2O. The accuracy of the method is better than1 rel%N for finely powdered samples if combustion canbe completed within a rather short time span.

• Carrier gas hot extraction—the sample is heated to morethan 2000 ◦C in the presence of carbon and an auxiliarybath of metal such as platinum and in a stream of a carriergas such as He. Nitrogen is liberated as N2 and is usuallymeasured in a thermoconductivity cell. Oxygen combineswith C to give CO and can thus be analyzed. Because ofthe high temperatures involved, the method is suitable forpowder samples with larger grain size or even for smallsolid pieces. The accuracy of the method is on the orderof 1–2 rel%N.

• Kjehldahl analysis—this involves the dissolution ofnitrides in acids whereby the nitrogen is convertedinto ammonium ions. The dissolution process may betime consuming as some nitrides, such as δ-TiN1−x,are chemically very stable against acids and must bedissolved using severe conditions such as by the actionof hydrofluoric acid in polytetrafluoroethylene (PTFE)containers at 100–200 ◦C. It is doubtful whether all of thenitrogen can be converted into NH4

+, so systematic errorsmay arise.

4.2 Physical Analysis

Among the microphysical methods used for the analysisof nitrogen in nitrides, for example, electron probe mi-croanalysis (EPMA), Auger electron spectroscopy (AES),ultraviolet X-ray spectroscopy (UPS), X-ray spectroscopy(XPS), secondary ion-mass spectrometry (SIMS) or sec-ondary ion-neutral mass spectrometry (SNMS), and Ruther-ford backscattering (RBS), EPMA is considered to have thehighest reproducibility and accuracy, provided that the lat-eral resolution of only a few microns is sufficient. The



main problem in EPMA is the rather low energy of NKα

radiation. Even with windowless detector systems, a highbackground-to-peak ratio together with low net intensi-ties is the result.22 High-accuracy EPMA requires externalor internal standardization22,23 together with a wavelength-dispersive spectrometer equipped with high-quality crystals(e.g., W–Si multilayer or lead stearate crystals). For stan-dardization, diffusion couples were used in which diffusionlayers are broadened.24–26 Such samples can even be pre-pared in single-phase form upon increase of diffusion timeso that the gross chemical composition is the same as thephase composition, a necessary prerequisite for application ofdifferent analyzing techniques and for true standardization.27

By such a method, single-phase standards of ε-TaN,28 δ-CrN,29 and ζ -Hf4N3−x

30,31 were prepared. These phaseshave a very narrow homogeneity range, at least at certaintemperatures.

Neutron activation analysis (NAA) making use of thenuclear reaction 14N(n,2n)13N can be applied for bulkdetermination of nitrogen.

5 THERMODYNAMICS AND SPECIFICSTRUCTURAL FEATURES

5.1 Nitrogen Partial Pressure

The nitrogen partial pressure of the transition metal nitridesis a function of temperature and increases substantiallywith increasing group number. This is reflected in theRichardson diagram shown in Figure 4. The lower (themore negative) the free energy of formation, the lower isthe equilibrium nitrogen pressure of the nitride. Group 4nitrides can easily be obtained by nitridation of the metalswith molecular nitrogen under low pressures even at hightemperatures, whereas for the group 5 nitrides, the nitrogenpressure must already exceed normal pressure at temperaturesabove 1400 ◦C, and even higher pressures or more energeticnitriding conditions must prevail for the preparation of group6 nitrides.

The nitrogen pressure determines the compositionof the nitride if equilibrium is attained. Thermochem-ical models were established that correctly predictpressure–temperature–composition (p–T–x) relationshipsfor nitride phases.32

Nitride phases with a nonmetal to metal ratio greaterthan 1 were obtained with high-pressure nitrogen or athigh nitrogen potentials, such as those prevailing in flowingammonia or ionic nitrogen atmospheres. Either an extensionof the mononitride phase field to compositions N/T > 1 wasobserved, such as in the Nb–N, Hf–N, and Ta–N systems, ordistinct nitride phases were found such as in the Zr–N, Hf–N,Nb–N, and Ta–N systems.33–35 Thermodynamic models werealso used for these phases to establish the stability regions

and to supply data for preparation at extremely high nitrogenpressures and high temperatures.33

Ion implantation techniques can lead to nitrides havingextremely high nitrogen to metal ratios, such as TiN1.5, butthese phases are probably metastable.

5.2 Phase Equilibria and Structural Features

Several of the transition metal–nitrogen systems are notyet fully established. This is especially the case for some of themetals in groups 6–9, probably because of the lower technicalrelevance of the nitrides and the high nitrogen pressuresneeded for equilibrium investigations.

For nitrides of a composition exceeding [N]/[Me] = 1, thatis, containing more than 50 mol% N, which can be obtainedat increased nitrogen activity,33,34 an introduction into a phasediagram is not possible. Such nitrides, for example, cubicZr3N4 and Hf3N4

36 as well as Ta2N3,37 were preparedin small bulk samples by diamond anvil cell preparation;some others, for example, orthorhombic Zr3N4, Ta3N5,and Mo5N6, were prepared with high-pressure nitridationor reaction with ammonia,38–40 and several niobium andtantalum nitrides were generated in the form of a band indiffusion couples35 or in the form of thin films.41 Somefurther details on some of these nitrides are given inSection 6.8.

5.2.1 The Ti–N System

The distinctive features of the Ti–N system42 are aconsequence of the high solubility of nitrogen in α-Ti; itconsists of the three subnitride phases η-Ti3N2−x, ζ -Ti4N3−x,and ε-Ti2N and the fcc phase δ-TiN1−x with the sodiumchloride structure. η-Ti3N2−x and ζ -Ti4N3−x were found tobe isostructural with the hafnium subnitride phases and arecharacterized by metal atom layers comprising cubic andhexagonal stacking sequences (Figure 1). Figure 5 shows thephase diagram of the Ti–N system.

In δ-TiN0.50, the random distribution of nitrogen atomsundergoes an ordering process43 below about 880 ◦C, witha concurrent tetragonal distortion of the metal lattice. Theconsequently formed δ′-TiN0.5 phase is only formed fromδ-TiN1−x by lattice distortion; it is not an equilibrium phase(e.g., it is not formed in isothermal diffusion couples inthe form of a phase band). Figure 6 shows a diffusioncouple44 that was prepared by nitridation of Ti metalin a first step and then annealed in order to form theδ′-TiN0.5 phase within the nitrogen-poor region of δ-TiN1−x. The metallographic appearance of a hatched regionconfirms the above-mentioned formation of the δ′-TiN0.5phase.

A possible Ti3N4 phase has so far only been proposedby calculation, contrary to Zr3N4 and Hf3N4, which wereobtained in an anvil cell.36



−700

−600

−500

−400

−300

−200

−100

0

ΔG°

KJ

mol

−1 N

2

100

200

300

400

500

600

7000 500 1000 1500

Temperature (°C)2000 2500

500 1000

T

TTM

M

MT

TN

TN

T

M

MN

M

B

B

B

B

B

M

M

T

T

T

T M

M

M

T

M

M

M

M

M

2Zr(s) + N2(g) = 2ZrN(s)

2Ti(s) + N2(g) =

2AI(s) + N2(g) = 2AIN(s)

2Y(s) + N2(g) = 2YN(s)

2La(s) + N2(g) = 2LaN(s)

2B(s) + N2(g) = 2BN(s)

2Ta(s) + N2(g) = 2TaN(s)

2Mo2N

4Cr(s) + N2(g) = 2Cr2N(s)

8Fe(I) + N2(g

) = 2Fe2N(s)

3Ca(

g) +

N 2(g

) = C

a 3N 2

(s)

3Mg(

g) +

N 2(g

) = M

g 3N 2

(s)

16Fe(I) + N 2

(g) = 2Fe 8

N(s)

3H 2(g) +

N 2(g) =

2NH 3(g)

4Mo(s) + N2(g) = 2Mo2N(s)

2Fe2N

2NH 3(g)

2Fe 8N(s)

2Cr(s) + N2(g) = 2CrN(s)

1/2Si3N4Mn5N2

Mn3N2

2VN

3/2Si(s) + N2(g) = 1/2Si3N4

(s)

2V(s) + N2(g) = 2VN(s)

Mg 3N 2

Ca 3N 2

2NbN

2BN

2TiN(s)

2CeN(s)2Ce(s) + N2(g) =

1500Temperature (K)

2000 2500 3000

TM

Figure 4 Richardson diagram of various transition metal mononitrides (together with other nitrides) giving the free energy of formationas a function of temperature. (Reproduced from Ref. 4. P. Ettmayer and W. Lengauer, ‘Nitrides’ in Ullmann’s Encyclopedia of IndustrialChemistry. pp. 341. 1991. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)

5.2.2 The Zr–N System

The solid solution of nitrogen in Zr, α-Zr(N), exists up toa composition of about 33 mol% N. The phase diagram ofthe Zr–N system is characterized by the presence of only onenitride phase, δ-ZrN1−x, up to a composition of [N]/[Zr] =1. It should be mentioned that none of the subnitride phasesthat exist in the neighboring systems Ti–N and Hf–N occur

here. In Massalski’s compilation11 (Figure 7), the Zr3N4

phase36,45 has not been included. This phase was prepared

at high nitrogen pressure and temperature (15.6–18 GPa,

2500–2800 K) in an anvil cell and could thus be characterized

in more detail36. It crystallizes in space group I4 3d (no. 220)

with a lattice parameter of 0.6740 nm. Further properties are

given in Section 6.8.



5000 5 10 15

0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.901.00

20 25

at% N

[N] / [Ti]

30 35 40

24

1060

1080

1100

1280

1300

28 32

α

ε

δ

δ-TiN1−x

ζ

at% N

ε-Ti2N

α-Ti(N)

L

β-T

i(N) η-

Ti 3

N2−

x

36

1340

1360

1380

1550T (K)1570

0.40 0.50 [N]/[Ti]

45 50

1000

1500

2000

T (

°C)

T (

°C)

2500

3000

3500

Figure 5 Phase diagram of the Ti–N system. (Reproduced from Ref. 42. © Elsevier, 1991.)

δ–TiN1−x ε–Ti2Nδ′–TiN1−x

100 μm

Figure 6 Reannealed diffusion couple showing the formation of theδ′-TiN0.5 phase44

5.2.3 The Hf–N System

Besides the fcc δ-HfN1−x, the Hf–N system containstwo subnitride phases,46 η-Hf3N2−x and ζ -Hf4N3−x. Onlya few phase equilibria studies were performed for thissystem. Figure 7 shows the compiled phase diagram togetherwith results from diffusion couples.30,31 The homogeneityranges of the subnitride phases could be established; thedecomposition temperatures of these phases, however, needto be reinvestigated.

An Hf3N4 phase with the same structure as Zr3N436 but

with a lattice parameter of 0.6701 nm, and thus smaller than thecorresponding Zr nitride, was synthesized at 18 GPa at 2800

K in a diamond anvil cell by laser heating. Some properties ofHf3N4 are given in Section 6.8.

5.2.4 The V–N System

In the V–N system, the phases β-V2N and δ-VN1−xhave cph and fcc structures, respectively. Carlsson et al.47

have reviewed the V–N system and reinvestigated thevanadium-rich portion. Compiled data and results ofa phase diagram study based on diffusion couples48

were used to construct the phase diagram shown inFigure 8.

Stoichiometric VN undergoes a phase transition into atetragonal phase below 205 K.7 At a composition of aroundVN0.80, a transition into a tetragonal phase occurs at 500 ◦C.Two different versions of the phase diagram at temperaturesbelow 500 ◦C were reported,51,52 one with a eutectic and onewith a peritectic phase reaction between tetragonal VN0.83 andδ-VN1−x.

5.2.5 The Nb–N System

Brauer et al.53,54 made careful studies on p–T–x re-lationships in the Nb–N system. In recent years, thephase reactions in this system were investigated by high-temperature X-ray diffraction as well as by use of diffusioncouples.35,49,55,56 The phase reactions could be assumed as



0700

1100

1500

1900

2300

T (

°C)

T (

°C)

2700

3100

10 20 30

at% N(a)

(b)

α-Zr(N)

α-Hf(N)

L

β

βδ-ZrN1−x

L

40 50 60

0500

1000

1500

2000

2500

3000

3500

10 20 30

at% N

40 50 60

η-H

f 3N

2−x

ζ-H

f 4N

3−x

δ-H

fN1−

x

Figure 7 Phase diagrams of group 4 transition metal–nitrogensystems. (a) Zr–N (Ref. 11. Reproduced by permission of MaterialsInformation Society/ASM International), (b) Hf–N. (Reproducedfrom Ref. 30. © Elsevier, 1993.)

established (Figure 8). Interestingly, γ-Nb4N3−x is formedby a martensitic transformation from δ-NbN1−x; hence, aninterface between δ-NbN1−x and γ-Nb4N3−x does not form indiffusion couples.49

Near the composition NbN, at least three phases wereobserved. For η-NbN, two different structures were proposedwith respect to the nitrogen positions. It contains trigonalprismatic and octahedral T6N structure elements and featuresa metal layer sequence AABB. η-NbN undergoes a phasetransition at about 1320 ◦C (depending on the nitrogenpressure) into fcc δ-NbN1−x with a sodium chloride structure.Another phase, δ′-NbN0.97, is most probably oxygen stabilizedor occurs as a metastable phase with phase transitionsand is not included in the phase diagram. By reacting Nbpowder with nitrogen under high pressure, the composition

of the δ-NbN1−x phase can be extended beyond N/Nb =1. At even higher nitrogen pressure, the phase Nb5N6 canbe prepared.35 Nb5N6 is isotypic with Ta5N6 and featuresT5N square pyramidal structural elements. These structuralelements were not observed in the analogous carbide systemsand are probably linked to the higher electron concentrationin nitrides.

5.2.6 The Ta–N System

The lower stability of the tantalum nitride phases ascompared to the nitrides of the group 4 metals is reflected intheir substantially higher equilibrium pressures (Figure 8).50

In order to prepare the high-temperature fcc phase δ-TaN1−x,the nitrogen pressure must exceed 10 bar, but it can bestabilized at lower pressures and temperatures by minuteamounts of carbon. It is the hardest known transition metalnitride phase. The Ta2N phase belongs to the ε-Fe2N type atlow temperatures and undergoes an order–disorder transitionto an L′3 type between 1600 and 2000 ◦C. The often-citedmelting point of TaN of 2950 ◦C is probably incorrect andhas to be attributed to Ta2N. TaN is expected to melt atabout 2800–3000 ◦C and nitrogen equilibrium pressures of1–3 kbar. The �-TaN phase with a WC-type structure isoften observed in the reaction products of Ta powder withnitrogen or ammonia at temperatures below 1100 ◦C and isbelieved by Brauer et al.57 to be a high-pressure phase that ismetastable under normal conditions. ε-TaN has a structure notencountered in carbides, featuring an arrangement of Ta5Nsquared pyramids. There are three stoichiometric phases in theTa–N system with decreasing metallic character. Red Ta3N5is nonmetallic and was synthesized by the reaction of TaCl5with NH3. In this rhombohedral phase, the nitrogen atomsare situated in the centers of tetrahedra, while the tantalumatoms are still sixfold coordinated by nitrogen atoms in theform of slightly distorted octahedra. Orthorhombic η-Ta2N3was prepared in different types of a high-pressure apparatusat 9–20 GPa nitrogen pressures and temperatures up to 2000K.37,58–60 This compound can contain some small amount ofnitrogen. Because of the closure of pores after metallographicpreparation, a self-healing mechanism was reported.37 Furtherdata are given in Section 6.8.

5.2.7 The Cr–N System

The Cr–N system is characterized by the nitride phasesβ-Cr2N1−x and δ-CrN. The former has an appreciablehomogeneity range. The arrangement of the metal atomsis cph and the nitrogen atoms are distributed in an orderedarrangement in the interstitial voids corresponding to theε-Fe2N-type structure. CrN crystallizes in the sodium chloride-type structure and its homogeneity range at low temperatureis rather narrow, whereas it is substantially broadened aboveabout 1230 ◦C.



0V(a)

(b)

(c)

Nb

0

1000

1400

1600

1800

2000

α-Ta(N)

β-Ta2N

ε-TaN

δ-TaN1−x

0

0 10Ta

20 30

at% N

40 50

10 20 30

at% N

40 50 60

η-NbN

30 bar N2

1 bar N2

γ-N

b 4N

3

β-Nb2N

δ -NbN1-x

α-Nb(N)

1200

1400

1600

1800

2000

500

1000

1500α-V(N)

β-V2N

δ ′-VN1−x

δ -VN1-x

L

T (

°C)

T (

°C)

T (

°C)

2000

2500

10 20 30

at% N

40 50

Figure 8 Phase diagram of group 5 transition metal–nitrogen systems. (a) V–N (Ref. 48. Reproduced by permission of Materials InformationSociety/ASM International), (b) Nb–N (Reprinted from Ref. 49. © 2000, with permission from Elsevier), and (c) Ta–N. (Reproduced fromRef. 50. © Elsevier, 1998.)

Dilute acids readily dissolve Cr2N1−x, whereas CrNappears to withstand acid attack for prolonged peri-ods. CrN is paramagnetic at room temperatures andundergoes a paramagnetic–antiferromagnetic first-order

transition at temperatures between 276 and 286 K, de-pending on the composition.10 The phase equilibria inthe Cr–N system are strongly dependent on the ni-trogen pressure. A phase diagram based on literature



0

Cr(a)

(b)

500

2000 10 20

10−1

10

100α

500

50010010

1

1

30

at% N

40 50

500

800

1100

1400

1700

2000

2300

2600

2900

1000

1500

T (

°C)

T (

°C)

2000

10 20 30

at% N

40

−2

−2

−1

−1 0

−1

0

β-Cr2N

δ-CrN1-x

γ-Mo2N1-x

β-MoN1-x

δ-MoN

L

Cr

50 60

Figure 9 Phase diagrams of group 6 transition metal–nitrogensystems. (a) Cr–N system (Reproduced from Ref. 29. J. PhaseEquilib, vol. 20, 1999, p. 35, ‘Phase equilibria and multiphase reactiondiffusion in the Cr–C and Cr–N systems’, W. Mayr, W. Lengauer, D.Rafaja, J. Bauer, and M. Bohn. With kind permission from SpringerScience and Business Media); (b) Mo–N, isobar pressures are givenin bar. (Reproduced from Ref. 61. ‘Phase Equilibria and CrystalStructures of Transition Metal Nitrides’ in Science of Hard Materials,p. 47, Eds: R. K. Viswanadham, D. J. Rowcliffe, J. Gurland. Withkind permission of Springer Science+Business Media.)

compilation and a study by diffusion couples is shown inFigure 9.29

5.2.8 The Mo–N and W–N Systems

In the Mo–N system (phase diagram61 in Figure 9), threenitride phases were reported. An fcc phase, originally namedγ-Mo2N because of its composition around 33 at% N, shouldinstead be written as δ-MoN1−x to be consistent with thenotation for other fcc nitrides. Diffusion couple studies gavea homogeneity range at 1000 ◦C of 28.5–38.2 mol% N.62

As fcc–Mo2N1−x was reported to exist at nitrogen contentsabove 33 at% and as it forms solid solutions with fcc δ-CrN, it can be assumed to be basically of the fcc δ-nitridetype. This compound undergoes an ordering transition ofthe nitrogen atoms and a concurrent tetragonal distortionof the metal lattice into β-Mo2N between 400 and 850 ◦C,depending on the composition. A hexagonal phase, Mo2N,which would be expected by analogy with the other transitionmetal nitride systems, appears to be thermodynamically lessstable than the cubic phase, because in the ternary systemMo–C–N, the compositional range of the hexagonal solidsolution Mo2(C,N) extends from Mo2C to approximatelyMo2(C0.35N0.65).63

Another phase, δ-MoN, with a structure similar to that ofWC, was prepared by various methods including nitridationof Mo powder with flowing ammonia and reaction of MoCl5with NH3.17 It was structurally characterized to have severaltypes of ordered arrangements of nitrogen in the centers oftrigonal prismatic structure elements.64

While at least 15 different nitride phases with compositionsranging between W2N and WN2 have been reported, none ofthem appears to be very well characterized. Some theoreticalcalculation of phase stabilities of tungsten nitrides havebeen carried out.65 As for the Mo–N system, an fcc nitrideWN1−x appears to exist, as nitridation experiments66 underpressure in the Mo–W–N system have shown that the fccMoN1−x phase is able to form extended solid solutionswith the isotypic phase WN1−x. The pressure required tobring molecular nitrogen to reaction with tungsten at suchtemperatures, where kinetics are favorable, appears to bepossible in diamond anvil cells. It is quite probable thatanother phase, WN with a WC-type structure, exists; at leastit was shown that nitrogen stabilizes the quaternary solidsolution (Mo,W)(C,N) with a WC-type structure. Except forthe solubility of nitrogen in bcc WN, no reliable data for thissystem are available.

5.2.9 The Mn–N, Fe–N, Ni–N, and Co–N Systems

The Mn–N system11 bears some resemblance to the Cr andFe nitride systems. The correspondence to the Cr–N systemmanifests itself by the existence of the Mn2N1−x phase witha cph arrangement of Mn atoms as well as two nearly cubicphases with compositions η-Mn4N3 and δ-Mn6N5.

Similarity to the Fe–N system is reflected by the extensivesolid solubility of nitrogen, particularly in the fcc manganese(austenite) phase and the occurrence of a ferromagnetic Mn4Nphase. The nitrides with higher nitrogen contents, η-Mn4N3and δ-Mn6N5, can only be prepared by the reaction of ammoniawith manganese powder at rather low temperature, suggestingthat the equilibrium pressures of nitrogen are very high evenat moderate temperatures. η-Mn4N3 resembles in its structurethe ordered phases Mn4N3 and β-Mn2N, where ordering ofthe nitrogen atoms at the interstitial sites is accompanied by atetragonal distortion of the cubic lattice. The �-Mn6N5 phase



400

0 10

α

20

Fe4N

at% N

30

ζ -Fe2N

ε -Fe2N1−x

γ

40

600

800

1000

T (

°C) 1200

1400

1600

102 103 104 105

L

1800

Figure 10 Phase diagram of the Fe–N system: isotherm pressures are in bar; the region around the ζ -Fe2N phase was modified. (Reproducedfrom Ref. 68. © Elsevier, 1991.)

comes very close to the stoichiometric composition MnNbut still has a clearly visible tetragonality of the basicallyfcc lattice. CrN apparently stabilizes an fcc solid solutionextending from CrN up to (Cr0.2Mn0.8)N1−x at 1000 ◦C and800 bar N2.67

The solubility of nitrogen in austenite (γ-Fe) is much higherthan that in α-Fe, and in this respect, nitrogen behaves similarlyto carbon. Upon cooling of nitrogen-containing austenite, astructure develops that is similar to pearlite and is consequentlycalled nitrogen pearlite. Unlike carbon, nitrogen forms anordered Fe4N phase that is ferromagnetic below 480–508 ◦C.At about 680 ◦C, Fe4N transforms into ε-Fe2N1−x, and in thatrespect, the close similarity to the Mn–N system becomesapparent. Besides ε-Fe2N with an ordered arrangement ofnitrogen atoms on the interstitial sites between the hexagonallyarranged metal atoms, the ζ -Fe2N phase with a slightly highernitrogen content and an orthorhombic distortion of the cphmetal lattice was reported. Figure 10 presents the Fe–N phasediagram with pressure isobars.68

Various iron nitrides and carbonitrides are formed duringcase hardening of steels and alloyed steels.69 These ironnitrides are hard, highly wear resistant, and usually do notlead to dimensional changes in the case-hardened items.

While there are no phase diagrams available for the Co–Nand Ni–N systems, it can be supposed that they are similar tothe Fe–N system but with even higher equilibrium pressures.Several phases were prepared by the reaction of flowingammonia with Co and Ni powder.

5.2.10 Ternary and Higher Order Nitride Systems

Binary transition metal nitrides can form ternary solidsolutions with each other70 if they are isostructural and iftheir lattice parameters differ by no more than about 10–15%.δ-VN1−x and δ-ZrN1−x do not form extensive solid solutionswith each other and the intersolubility of δ-VN1−x and δ-HfN1−x is not complete, either.

Ternary phases with structures different from those ofthe phases of the binary boundary systems are more theexception than the rule. Such phases were reported inthe systems Nb–Mo–N, Ta–Mo–N, Nb–Ta–N, Zr–V–N,Nb–Cr–N, and Ta–Cr–N. Information about ternary tran-sition metal–nitrogen systems is often available for specifictemperatures only.57 This is even more the case for quaternarynitride systems, which play a role in the production of carboni-tride cermets (see Section 7) where quaternary compounds ofthe types (Ti,Mo)(C,N) and (Ti,W)(C,N) are of interest (seeCarbides: Transition-Metal Solid-State Chemistry), as wellas in layer technology where titanium nitride-based coatingsof the type Ti(C,B,N) are prepared by magnetron sputter-ing. Layers consisting of ternary compounds in the systemsTi–Al–N, Cr–Al–N, Ti–V–N, Ti–Si–N, and Ti–Cr–Nalso have favorable properties with respect to performance incutting tools71–78 (see Section 7).

A further group of ternary nitrides is of the type R2Fe17Nx,where R is a rare earth element such as Sm, Ce, or Nd andx indicates the variable nitrogen content. These phases havebeen prepared by nitridation at relatively low temperaturesand up to 15 MPa N2, under which conditions the nitrogencontent can reach x = 3. They are closely related to



borides and carbides with similar formulas and have attractedconsiderable interest for hard-magnetic applications.79 It wasreported that these phases can reach compositions wherex = 6, using a combination of hydrogen and nitrogentreatments.

Additional generally interesting phases from the crystalchemistry viewpoint are the H-phases,80 for example, Ti2AlN,nitrogen perovskites, filled β-manganese-type phases, andNowotny6 phases.

6 PROPERTIES OF FCC TRANSITION METALNITRIDES

For carbonitrides, see Carbides: Transition-Metal Solid-State Chemistry1.

6.1 Melting Points

Melting points of various nitrides81 were measured undernitrogen pressures up to 300 bar. The nitrogen pressure hasto be chosen in such a way as to be equal to the nitrogenequilibrium pressure of the nitride at its melting point. Themelting point temperatures of the various nitrides can reachmore than 3000 ◦C and can be read from the phase diagramsgiven in Section 5.2.

6.2 Color

Many mononitrides exhibit vivid colors of a bright metallicappearance that depends on the N/T ratio. Table 1 lists thecolors of the mononitrides and gives some idea of their color-composition interdependency. These colors can be observed inbulk samples and have an important role when thin films77,82

are produced for ornamental purposes.

6.3 Thermal and Electrical Properties

Interestingly, the transition metal carbides and nitridesdo not show the expected decrease in heat conductivitieswith temperature but instead display an increase.83 Figure 11illustrates this property for the group 4 nitrides and showsa comparison with group 4 carbides and carbonitrides. Thesubstitution of nitrogen by carbon in group 4 transition metalnitrides significantly reduces the heat conductivities.

The electrical conductivities of transition metal carbidesand nitrides are greatly influenced by the nonmetal/metalratio; as this ratio approaches unity, the electrical conductivityreaches a maximum. This is shown for δ-TiN1−x in Figure 12.85

A similar behavior can be observed for other carbides andnitrides. The electrical conductivities of these compoundsdecrease linearly with increasing temperature.84

The phenomenon of superconductivity is common for allfcc transition metal nitrides of groups 4 and 5 and is also

15

20

0 200 400 600 800 1000

ZrC

HfCTiC

TiC0.2

ZrC0.2

N0.8

N0.8

Temperature (°C)

Hea

t con

duct

ivity

(W

m−1

K−1

)

2000

25

30

35

40

45

50

TiN

ZrN HfN

Figure 11 Heat conductivities of nearly stoichiometric δ-TiN1−x, δ-ZrN1−x, and δ-HfN1−x compared with δ-TiN1−x, δ-ZrC1−x, δ-HfC1−x,and δ-Ti(C0.2N0.8) and δ-Zr(C0.2N0.8) as a function of temperature.83

δ-Hf(C0.2N0.8) coincides closely with the curve for TiN84

0.500.40

0.80

1.20

1.60

2.00

2.40

2.80

3.20

3.60

4.00

Ele

ctric

al c

ondu

ctiv

ity /

104

Ω−1

cm

−1

0.60 0.70 0.80 0.90 1.00

Composition [N]/[Ti]

Figure 12 Electrical conductivity at room temperature of δ-TiN1−xas a function of stoichiometry. (Reproduced from Ref. 85. © Elsevier,1992.)

observed in fcc MoN1−x and hexagonal MoN. δ-NbN1−xshows the highest superconducting transition temperature of17 K, which was the highest known of any material for severaldecades. It is noteworthy that hexagonal δ-MoN also has one ofthe highest superconducting transition temperatures (15 K) ofthese materials,64 and fcc γ-MoN1.00, although never preparedwith a stoichiometric composition, was predicted theoreticallyto have a Tc of up to 30 K. All transition metal nitrides aretype II (hard) superconductors. For several fcc nitrides, suchas δ-TiN1−x, δ-ZrN1−x, and δ-VN1−x, Tc increases linearlywith increasing nitrogen content up to [N]/[T] = 1;86 δ-NbN1−x shows a much weaker increase. The Tc data reportedfor δ-HfN1−x are contradictory as to whether an increase



0.20.8 0.85 0.9

HfN(b)

HfN(a)

TiN

ZrN

VN

NbN

0.95Composition ([N]/[Ti])

1

0.4

0.6

T/T

c(m

ax)

0.8

1.0

Figure 13 Superconducting transition on temperatures (normalizedrelative to maximum Tc) as a function of stoichiometry for varioustransition metal nitrides. For δ-HfN1−x, different functions werereported (a,b). (Reproduced with permission from Ref. 86. © JohnWiley & Sons, Ltd., 1990.)

or a decrease occurs. This is probably because δ-HfN1−xcan apparently exist with a hyperstoichiometric composition.Figure 13 shows the normalized Tc of several fcc nitrides(where the highest Tc of each compound is equal to 1) andTable 1 shows the highest Tc for each compound for the givencomposition.

6.4 Thermal Expansion

The thermal expansion of various transition metal ni-trides was measured at both low temperatures87 (8–300 K)and high temperatures88 (up to 1500 K). According tothe polynomial expression fit of the low-temperature lat-tice parameters, δ-VN1−x exhibits negative thermal ex-pansion at temperatures below 55 K for the composi-tion VN0.96 and below 41 K for the composition VN0.72.At these temperatures, the lattice parameter is a mini-mum. An even more pronounced minimum was found forScN at 107 K.87 The high-temperature thermal expansiondata are summarized in Figure 14.88 Interestingly, con-sidering its low-temperature thermal expansion behavior,δ-VN1−x has the highest high-temperature expansion co-efficient among these compounds, up to 11×10−6 K−1 at1100 K.

6.5 Nitrogen Diffusivities

In transition metal nitrides, nitrogen diffusion proceedsvia a vacancy mechanism in the nonmetal sublattice. Thevacancy concentration is a function of composition within thehomogeneity range of the fcc phases, and hence, the nitrogendiffusivity is most probably composition dependent as well.89

3005.0

6.0

7.0

8.0

9.0

10.0

11.0

12.0

500 700 900 1100

TaN1.17

ZrN0.93HfN1.01

ZrN0.91

MoN0.53

TiN0.95NbN0.953

NbN0.99NbN0.924

NbN0.975

VN0.93VN0.707

TiN0.89VN0.996

1300

Temperature (K)

Coe

ffici

ent o

f lin

ear

ther

mal

exp

ansi

on (

10−6

K−1

)

1500 1700

Figure 14 High-temperature thermal expansion of various transi-tion metal nitrides. (Reproduced from Ref. 88. Monatsh. Chem.,117, 1986, p. 713, ‘Lattice parameters and thermal expansion ofδ-VN1−x from 298–1000 K’, W. Lengauer and P. Ettmayer. Withkind permission from Springer Science and Business Media.)

In this respect, the Ti–N90, Nb–N,91 and Cr–N29 systemswere investigated in detail.

Figure 15 shows the nitrogen diffusivities in titaniumnitride at three different temperatures determined from high-precision EPMA line scans27 in diffusion couples (nitridedTi sheet).90 Because of the identical slope of the threelines, the concentration dependency of the diffusivity wasfound to be temperature independent in this temperatureinterval.

Figure 16 shows the concentration-dependent diffusivityin CrN29. Generally, diffusivity studies were conducted byinvestigating nitride layer growth as a function of time, whichgives the average nitrogen diffusivity in the various phases.A parabolic relationship, d = k·t0.5, holds for this layergrowth, where d is the layer thickness, k a constant, and tthe diffusion time. The parabolic behavior indicates that thenitride layer growth is indeed diffusion controlled. By useof wedge-type samples, the thickness of diffusion layers isincreased if the sample thickness increases. Such a deviationfrom the parabolic law (if different thicknesses are compared)can be used to establish the nitrogen diffusivities in all thephases at a given temperature.24,26 An example of such awedge-type sample is shown in Figure 17 for the Hf–Nsystem.30

Nitrogen diffusion in nitrides such as δ-TiN1−x, δ-ZrN1−x,and δ-HfN1−x has an activation energy of 2–4 eV. Althoughthe metal diffusivity in transition metal nitrides is not yetinvestigated in detail, the activation energy of that process is



0.5−16

−15

−14

−13

−12

0.6

1

2

3

0.7 0.8

x in TiNx

log

D/m

2 s−1

0.9 1

Figure 15 Concentration dependence of the nitrogen diffusion coefficient in titanium nitride90; (1) 1900, (2) 1600, (3) 1350 ◦C

Table 2 Diffusion data for nitrogen in transition metal nitrides

Compound D0 (cm2·s−1) E (eV) A (cm−3 mol−1)

δ-TiN1−x 0.03 2.4 6.5δ-HfN1−x 0.02 2.70 —ζ -Hf4N3−x 0.14 2.74 —η-Hf3N2−x 0.54 2.73 —δ-VN1−x 12.7 2.93 —β-V2N 13.6 2.92 —δ-NbN1−x 0.68 – 1.10 3.20 20β-Nb2N 120 4.0 —δ-TaN1−x 3.4 3.6 —β-Ta2N 1.4 3.3 —δ-CrN1−x 0.18 2.92 184β-Cr2N 3.51 2.68 —

probably much higher and can be estimated to be about 7–8eV. Table 2 shows diffusivity data of transition metal nitrides.

The diffusivity versus composition was investigatedfor δ-TiN1−x

90, δ-HfN1−x,30 and δ-NbN1−x.91 A modifiedBoltzmann–Matano analysis, carried out on nitrogen diffusionprofiles measured by EPMA, yielded a composition-dependentdiffusion coefficient for N diffusion expressed as (equation 7)

D = f ν2d exp

( −E

kBT

)exp(A(c+ − c))cm2s−1 (7)

where f = 1/12, for fcc compounds, ν is the jump frequencyof the N atom, d the mean distance between lattice planes,E the activation energy, kB the Boltzmann constant, A aparameter for the steepness of concentration dependency, cthe concentration, and c+ is the maximum concentration;(c+ – c) = x in the formula MeNx.

6.6 Elastic Properties

Transition metal nitrides exhibit high Young’s moduli,appreciably higher than the elastic moduli of the transitionmetals. As can be seen from Figure 18,92 the Young’s moduliof δ-TiN1−x and δ-ZrN1−x increase with increasing nitrogento metal ratio, whereas for δ-HfN1−x, the modulus remainsapproximately constant. Several of these data were obtainedfrom thin layers. It is well known that in layers large deviationsfrom bulk properties (lattice parameter, composition) mayoccur, due probably to the high concentration of defectsparticularly in PVD layers, while in bulk samples, pores candrastically influence the results.93

The elastic properties of a variety of transition metal nitrideswere critically reviewed by Kral et al.93 The temperaturedependence of δ-TiN1−x and of TiC0.6N0.4 is shown inFigure 19. Up to the maximum investigation temperatures,the moduli lower only slightly upon increasing temperature,both indicating the stiffness of the materials and that softeningof the lattice occurs at even higher temperatures.

6.7 Hardness

Some transition metal nitrides are among the hardest ma-terials known. Figure 20 shows hardness versus compositionfor δ-TiN1−x

85 and for the group 5 transition metal nitrides.The hardness decreases with increasing nitrogen content forthe group 5 nitrides.35,88,95 This behavior can probably be ex-plained by electronically induced lattice softening as a resultof the generation of antibonding states. It is known from bandstructure calculations that in fcc transition metal compoundswith more than eight valence electrons, antibonding states aresuccessively occupied. Indeed, in δ-TiN1−x, a maximum in



5.810−13

10−12

10−11

10−10

10−9

10−8

10−7

1450 1400 1350 1300

Temperature (°C)

1250 1200 1150

6.0 6.2 6.4

Reciprocal temperarure (10−4 K−1)

Diff

usio

n co

effic

ient

(cm

2 s−1

)

6.6 6.8 7.0

Figure 16 Temperature dependence of diffusion coefficients forthe Cr–N system obtained from layer growth and concentrationprofiles,29 �: δ-CrN1−x from layer growth (different concentrationranges for different temperatures), �: δ-CrN1−x from concentrationprofile fitting (lower dashed line CrN1.00, upper dashed line CrN0.67),©: β-Cr2N from concentration profile fitting, •: β-Cr2N from layergrowth, �: α-Cr(N) from layer growth. (Reproduced from Ref. 29. J.Phase Equilib, vol. 20, 1999, p. 35, ‘Phase equilibria and multiphasereaction diffusion in the Cr–C and Cr–N systems’, W. Mayr, W.Lengauer, D. Rafaja, J. Bauer, and M. Bohn. With kind permissionfrom Springer Science and Business Media.)

microhardness is found at the composition δ-TiN0.67, corre-sponding to 7.3 valence electrons. This is also reflected inthe hardness of titanium and zirconium carbonitrides with50 mol% C+N, which show a maximum for [C]/([C]+[N])= 1, that is, VEC = 8 for the pure carbide, whereas the

00.4 0.5

TiNZrNHfN

0.6 0.7 0.8

[N]/[T]

0.9 1 1.1 1.2

200

400

You

ng’s

mod

ulus

(G

Pa)

600

800

1000

Figure 18 Young’s modulus of bulk and layered group 4 transitionmetal nitrides. (Reproduced from Ref. 92. © Elsevier, 1987.)

hardness is lowered as carbon is replaced by nitrogen.96

An increasing VEC arising from increasing nitrogen contentwould then lead to a decrease in hardness when the numberof valence electrons exceeds eight. For δ-HfN1−x, a smoothincrease in microhardness with increasing nitrogen content isobserved. This, however, is probably a result of the differencein electronic states owing to 5f electrons. Also, the carboni-tride Hf(C,N) does not show such a pronounced maximum at[C]/([C]+[N]) = 1 as do Ti and Zr carbonitrides.96

6.8 Properties of Nitrides with a Nitrogen ContentExceeding 50 mol%

In recent years, bulk nitrides with a nitrogen contentsignificantly exceeding 50 mol% N have been prepared.33,34

They have very interesting properties of which Table 3 givesan overview.37,59,97–100 Even the preparation in form of thinvery hard layers was suceesful.41,101 Tests showed that suchlayers outperform conventional coatings41; hence, a technicalapplication would seem likely to occur in due course.

200 μm

δ-HfN1−x

δ-HfN1−x

ζ-Hf4N3−x

ζ-Hf4N3−x

ε-Hf3N2−x

ε-Hf3N2−x

α-Hf(N)

Figure 17 Wedge-type sample of the Hf–N system with various diffusion layers. (Reproduced from Ref. 30. © Elsevier, 1993.)



0

G

B

E

E

B

G

150

250

350

Ela

stic

con

stan

ts E

, B, a

nd G

(G

Pa)

Ela

stic

con

stan

ts E

, B, a

nd G

(G

Pa)

450

550

200 400

T (°C)(a) (b)

600 800 0150

250

350

450

550

200 400

T (°C)

600 800

Figure 19 Elastic constants E, B, and G as a function of temperature (a) for TiN5 and (b) for TiC0.6N0.494

160.50 0.60 0.70 0.80

Composition ([N]/[Ti ])(a)

(b)

0.90 1.00

δ-TaN1−x

δ-NbN1−x

δ-VN1−x

171819H

V0.

3 (G

Pa)

20212223242526

0.75

10

15

20

25

30

35

0.8

Composition ([N ]/[Me])

Mic

roha

rdne

ss H

V0.

1 (G

Pa)

0.9 1

Figure 20 (a) HV0.3 Microhardness of δ-TiN (Reproduced from Ref. 85. © Elsevier, 1992) and HV0.01 microhardness (b) of group 5transition metal nitrides as a function of composition. (Reproduced from Ref. 5. Books: W. Lengauer in R. Riedel: Handbook of CeramicHard Materials. pp. 202. 2000. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)



Table 3 Structure data and solid-state properties of bulk transition metal nitrides with [N]/[Me] > 1 and some low amount of oxygen

Zr3N4 [U3] Hf3N4 Ta2N3

Space group 220 cubic 220 cubic 62 orthorhombicStructure type Th3P4 Th3P4 U2S3Lattice parameter(s) a = 0.6740(6) nm a = 0.6701(6) nm a = 8.1911(17)

b = 8.1830(17)c = 2.9823(3)

Density 7.159 g cm−3 13.058 g cm−3 13.42 g cm−3

Hardness HV(1)=12.0(6)(e)97 16 GPa(a),ca. 35 GPa(f) ca. 30 GPa(b)

Young’s modulus 394(24)(d) GPa98 223(2) GPa(c)37

322(3) GPa(d)37

Bulk modulus 217(20)(d)GPa98 227(7) GPa99 176(6) GPa(c)37

219(13) GPa98 281(15) GPa(d)37

319(6) GPa59

Shear modulus 163(9)(d) GPa98 87(1) GPa(c)37

123(2) GPa(d)37

Poisson ratio 0.20(4)(d)98 0.29(1)(c)37

0.31(1)(d)37

(a)Porous sample (10% porosity).(b)Estimated from elastic properties.(c)Measured for a porous sample with the volume fraction porosity (VFP) of 18%.(d)For dense material, derived from experimental data for porous sample using a theoretical approach taking into account the porosityinfluence.(e)Porous sample: VFP = 25–30%.(f)For dense material, estimated from the dependence of HV on porosity and from the elastic shear modulus for dense material.98

7 USES

Nitrides and carbonitrides in form of powders play animportant role in hardmetals (cemented carbides). The readeris also referred to the article on Carbides of this Encyclopedia(see Carbides: Transition-Metal Solid-State Chemistry).

7.1 Functional-Gradient Hardmetals withCarbonitrides, Interaction of Nitrogen withHardmetals

The interaction of nitrogen with hardmetals (cementedcarbides) containing nitride and carbonitride-forming metalssuch as Ti, Ta, and Nb can be exploited to form near-surface gradients in order to enhance the performance ofthe hardmetals and to supply specific surface conditionsfor a better adherence of coatings.102–104 By means ofthe appropriate application of nitrogen atmospheres uponsintering, this interaction leads to various forms of near-surface microstructures in functional-gradient hardmetals (seealso Carbides: Transition-Metal Solid-State Chemistry). Themicrostructures can be tailored in order to adjust hardness andtoughness of the materials.104,105

Upon appropriate application of nitrogen pressure,nitrogen-enriched or nitrogen-depleted near-surface regionscan be established. Figure 21 shows these microstructures.

These surface regions have a higher hardness than thebulk if the nitrogen is enriched because of the formation

of hard nitrides. Contrarily, the hardness is lowered buttoughness is increased in nitrogen-depleted surface regionfree of fcc carbonitrides.106 Figure 22 shows this behaviorfor the two types (compare with Figure 21). In fact, avariety of different near-surface microstructures is possible,depending on the composition and process parameters ofthe sintering regime.103 Even nitrogen enrichment belowa tough surface layer is possible.105 Functional-gradienthardmetals show excellent performance in some specificcutting applications.107

In hexagonal tungsten carbide (WC), nitrogen can beintroduced in order to form a carbonitride W(C,N).108,109

Upon synthesis starting from W + C powders and applicationof nitrogen pressures up to 160 MPa at temperaturesup to 1500 ◦C, W(C,N) powders with 20 mol% N couldbe prepared. W(C,N) can also be prepared by ammoniasynthesis with and without addition of CH4 at ambientpressure and temperatures up to 1300 ◦C with nitrogencontents up to 5 mol% N. Exploiting the ability of WCto form W(C,N) by nitridation at high nitrogen pressures,WC powders with particles covering a nitrided rim haverecently successfully employed to obtain fine-grained WC–Cohardmetals. Nitrogen serves as a transient grain-growthinhibitor as long as the nitrided surface exists. Upon liquid-phase sintering, this nitrided rim is dissolved in the eutectic,nitrogen is outgassed, and a pore-free hardmetal body can beobtained.110,111



Ti(C,N) enriched

(a) (b)50 μm 20 μm

WC enriched

Figure 21 Microstructures of functional-gradient hardmetals with (a) a nitrogen-enriched (WC-depleted) zone (light-optical microscopeimage) and (b) a nitrogen-depleted (WC-enriched) zone (SEM image)

140 10 20 30

Depth (μm)(a) (b)

40 50 0 10 20 30

Depth (μm)

40 50

16

18HV

0.02

5 (G

Pa)

20

22

24

26

28

14

16

18

20

22

24

26

28

Regular gradient

fcc-free zone

Inverse gradient

Figure 22 Hardness HV0.025 of surface zones of functional-gradient hardmetals as a function of distance. (a) Hardness of the nitrogen-enriched surface region (regular gradient) of a functional-gradient hardmetal and (b) hardness of a nitrogen-depleted WC-rich surface zone(fcc-free zone) and hardness of a WC-rich surface zone with inner Ti(C,N) enrichment, ‘‘inverse gradient’’. (Reproduced from Ref. 106. ©Plansee, 2005.)

7.2 Cermets—Cemented Carbonitrides

Cermets112–116 are cemented carbonitrides in which onlya cubic hard phase without free hexagonal WC is present(see also Carbides: Transition-Metal Solid-State Chemistry).However, there is no boundary between a typical cementedcarbide (hardmetal), that is, pure WC–Co, on the one handand a typical cermet on the other: the microstructures cancontain both free WC and fcc phases. In some hardmetalgrades, Ti-based fcc carbides and carbonitrides are added inorder to improve the performance, for example, in interruptedcutting processes. Cermets can be produced by liquid-phasesintering of blends of a variety of compounds such as TiN, TiC,WC, Mo2C, Ti(C,N), (Ti,W)C, (Ti,W)(C,N), (Ti,Mo)(C,N),(Ta,Nb)C, VC, Cr3C2, and so on together with bindermetals such as Co, Ni, or Co/Ni. After sintering, quaternaryhard phases based on (Ti,W)(C,N) and (Ti,Mo)(C,N) or acombination thereof forms. Further metal constituents of the

starting formulation such as Ta, Nb, Cr, and V are introducedinto the hard-phase particles upon sintering. Cr also dissolvesin substantial amount in the binder phase.

Figure 23 shows a microstructure of a modern cermet. InFigure 23(a), the grain size of the hard carbonitride particles isbelow 1 μm, whereas the microstructure of the cermet shownin 23(b) is substantially coarser. The hard-phase particlesof cermets often exhibit a so-called core-rim-type structure,which means that in principal two quaternary carbonitridesexist. The core of these particles is Ti and N enriched andthe rim is C- and W- or Mo-rich relative to the core. Oftenan inner rim is found that has a substantial higher amountof W. All these constituents have an fcc structure with alattice parameter very close to each other so they cannotbe distinguished by X-ray diffraction. In scanning electronmicroscopy images that are atomic mass sensitive, they appearas dark or even black core (light elements—Ti), white inner



1 μm 10 μm

(a) (b)

Figure 23 Microstructure of cermets with a hard-phase grain size below 1 μm (a) and 2–3 μm (b). (Reproduced by permission of Ceratizit,Luxemburg S.a.r.l.)

Table 4 Some characteristics and accompanying advantages ofcermets used for metal-cutting operations

Characteristics Advantage

Low specific density High tool rotationLow friction, low tendency tostick

High surface quality of workpiece

Low thermal conductivity Dry machiningHigh heat resistance High cutting speedHigh performance withoutcoating

Sharp cutting edge

Long tool life Lower tooling costsLow flank wear High dimensional accuracyLow-risk raw materials Safe supply

rim (heavy elements—W), and gray outer rim (mixture oflight and heavy elements).

The characteristics and the following advantages of cermetsused for metal-cutting operations (turning, milling, anddrilling) can be summarized as follows (Table 4).

The workpiece materials for which cermets are employedare mainly high and low alloyed and unalloyed steels, ferriticand martensitic corrosion-resistant steels, ferritic/martensiticand duplex stainless steels, cast iron grades (with theexception of gray cast iron), as well as pearlitic andferritic cast iron grades. Cermets are much less suitedfor machining of hardened steels and noniron materialssuch as aluminum alloys, brass, Inconel, and titaniumalloys. Examples of products in which cermets are usedas cutting materials are connecting rods and shafts forengines. Because of their increased brittleness as comparedto WC–Co hardmetals, cermets should only be used onhighly stable machines without vibration and by applicationof a low depth of cut. The workpieces should be free ofscale. Interrupted cutting is a problem for cermets. Theirspecific density is less than a half of the specific weight ofhardmetals; hence, high rotation speeds of cermet cutters arepossible.

Table 5 contains some important data for industrial gradesof cermets. The hard phase is Ti(C,N) based, and for thestarting formulation, tungsten, molybdenum, niobium, and/ortantalum carbides are added.

7.3 Wear-Resistant Layers

In order to increase the service life of hardmetal toolsfor cutting and milling operations, they are coated withCVD- and PVD-produced transition metal nitride layers.The coatings are composed mainly of δ-TiN1−x, Ti(C,N),(Ti,Al)N, (Ti,Si)N, and (Ti,Cr,Si)N, occasionally with someother elements such as V and B added. Nitride-coatedtool tips result in a better surface finish of the workpieceowing to both a decrease in diffusion welding betweentool tip and workpiece and an increase in thermal ox-idation resistance. The total layer thickness is generally5–20 μm. Often, layers of the above-mentioned nitridesare combined with layers of Al2O3. Figure 24 shows amultilayer coating including δ-TiN1−x layers on a WC–Cohardmetal.

In recent years, very interesting coatings have beendeveloped in the Ti–Al–N system. Previously, only PVDwas capable of producing thermodynamically unstable fcc(Ti,Al)N coatings.71 However, it was shown73 that CVDmethods working around 800–1000 ◦C can also be usedto form coatings of fcc Ti1−xAlxN with x up to 0.9.Such layers have excellent performance in turning andmilling operations of steel. Figure 25 shows a fractographof a Ti1−xAlxN coating on a hardmetal substrate. Uponintroduction of carbon, Ti1−xAlx(C,N) coatings74 can beprepared by CVD at temperatures of 800–850 ◦C. The nitridehas only a low carbon content of 2 mol%, but the layercovers some portion of amorphous carbon. These layers,too, show excellent cutting performance in milling tests.The hardness of these nitride and carbonitride layers is



Table 5 Important solid-state properties of industrial cermet grades.

Grade 1 Grade 2 Grade 3 Grade 4 Unit

Binder phase content Co + Ni 6.0 12.2 14.1 15.4 wt%Specific density 6.40 6.60 6.40 7.00 g·cm−3

Vickers hardness 1780 1620 1630 1580 HV30Transverse rupture strength 1500 1800 2000 2000 MPaCompressive strength 4000 5000 4700 4600 MPaYoung’s modulus 448 445 425 430 GPaFracture toughness 10.3 8.0 8.0 10.0 MPa·m1/2

Thermal conductivity 11 19 15 14 W·m−1K−1

Thermal expansion coefficient 8.3 9.0 7.6 9.4 10−6 K−1

Courtesy of Kennametal Technologies, Mistelgau and Ceratizit, Luxemburg.

10 μm

TiN

Ti(C,N)TiN

Al2O3

Figure 24 CVD multilayer coating on WC–Co hardmetal withyellow TiN1−x as top and bottom layer. Following the bottom layer,a Ti(C,N) layer (gray) was deposited, in turn followed by a blackAl2O3 layer (invisible in the light-optical microscope)

2 μm

Figure 25 Fractograph of a TixAl1−xN coating on a hardmetal(WC–Co). (Courtesy of Dr. Ingolf Endler, Fraunhofer IKTSDresden, Germany.)

around 3000 HV0.01. Even higher hardnesses of up to4100 HV0.01 can be obtained if Al is replaced by Si toform Ti1−xSixN and Ti1−xSix(C,N) coatings.75 These are nolonger single phase, but contain in fact crystalline titaniumnitride and titanium silicide and amorphous silicon nitride,whereas in the C-containing coatings, Ti(C,N) and Si(C,N)coexist. The increased hardness stems from the partially

nanocrystalline nature of the phases. A problem of the Si-containing coatings is their weaker adherence to the hardmetalsubstrate.

Bright golden yellow δ-TiN1−x is also used for ornamentallayers; for example, on watch cases where the combination ofcolor with increased scratch resistance is exploited.

By PVD methods, it is possible to prepare a huge varietyof multielement nitride layers. The metals are introducedupon appropriate choice of the sputter targets. The layers canbe single phase, multiphase, and multiphase nanostructuredas well as amorphous, for instance, in the Ti–Si–N systemwhere nanocrystalline TiN and amorphous Si3N4 coexist.76

Other examples of layers prepared with PVD are Cr- andV-containing nitride-based coatings.77,78

7.4 Diffusion Barriers

Because of their chemical inertness, high electricalconductivities, and the ease with which they can be depositedas layers, transition metal nitrides, for example, titanium,zirconium, tantalum, niobium, and tungsten nitrides, andalso their ternary forms such as (Ti,Zr)N and (Ti,Ta)Nare excellent candidates for use in multilayer metallizationsof integrated circuits.117–120 In these devices, interdiffusionbetween the different metals, compound layers, or thin metalwires during thermal load with soldering and packagingoperations must be avoided, otherwise degradation of themultilayer arrangement will occur. In order to protect theselayers against interdiffusion, very thin nitride layers (<100nm) can be deposited by PVD methods. These layers are stableup to 600 ◦C (depending on the time and the substrate) andprevent the layers from undergoing structural changes.

8 REFERENCES

1. P. Schwarzkopf and R. Kieffer, Refractory Hard Metals,Macmillan, New York, 1953.

2. H. J. Goldschmidt, Interstitial Alloys, Butterworths, London,1967.



3. L. E. Toth, Transition Metal Carbides and Nitrides, AcademicPress, New York-London, 1971.

4. P. Ettmayer and W. Lengauer, in Ullmann’s Encyclopedia ofIndustrial Chemistry, Verlag Chemie, Weinheim, 1991, Vol.A17, Chap. Nitrides, p.341.

5. W. Lengauer, in Handbook of Ceramic Hard Materials, ed.R. Riedel, Wiley-VCH, Weinheim, 2000, Vol. I, p.202.

6. H. Nowotny and F. Benesovsky, in Phase Stability in Metalsand Alloys, eds P. S. Rudman, J. Stringer and R. I. Jaffee,McGraw-Hill, New York, 1967, p.319.

7. F. Kubel, W. Lengauer, K. Yvon, A. Junod, and K. Knorr,Phys. Rev. B, 1988, 38, 12908.

8. G. Brauer and H. Kirner, Z. Anorg. Allg. Chem., 1964, 328,34.

9. J. Gatterer, G. Dufek, P. Ettmayer, and R. Kieffer, Monatsh.Chem., 1975, 106, 1137.

10. J. D. Browne, P. R. Lidell, R. Street, and T. Mills, Phys. StatusSolidi, 1970, A1, 715.

11. T. B. Massalski (ed), Binary Alloy Phase Diagrams 2ndedition, ASM International, Metals Park, OH, 1990.

12. K. Suzuki, T. Suzuki, Y. Fujinaga, T. Kaneko, H. Yoshida, Y.Obi, and S. Tomiyoshi, J. Alloys Compd., 2002, 360, 34.

13. P. Villars and L. D. Calvert eds, Pearson’s Handbook ofCrystallographic Data for Intermetallic Phases, 2nd edn,ASM, Metals Park, OH.

14. K. Schwarz, CRC Crit. Rev. Solid State Mater. Sci., 1987,13(3), 211.

15. A. Neckel, in The Physics and Chemistry of Carbides, Nitridesand Borides, ed. R. Freer, Kluwer Academic Publishers,Dordrecht, 1990, p.485.

16. P. Blaha and K. Schwarz, Int. J. Quantum Chem., 1983, 23,1535.

17. W. Lengauer, J. Cryst. Growth, 1988, 87, 295.

18. A. G. Merzhanov, Ceram. Intl., 1995, 21(5).

19. N. Christensen and S. Fregerslev, Acta Chem. Scand. A, 1978,A31, 861.

20. W. Lengauer, Talanta, 1991, 38(6), 659.

21. R. Taubler, S. Binder, M. Groschner, W. Lengauer, and P.Ettmayer, Mikrochim. Acta I , 1992, 107, 337.

22. G. F. Bastin and H. J. M. Heijligers, Quantitative ElectronProbe Microanalysis of Nitrogen, University of Technology,Eindhoven, 1988.

23. W. Lengauer, J. Bauer, A. Guillou, J.-P. Bars, M. Bohn, E.Etchessahar, J. Debuigne, and P. Ettmayer, Mikrochim. ActaI , 1992, 107, 303.

24. W. Lengauer, J. Alloys Compd., 1995, 229, 80.

25. D. Rafaja, W. Lengauer, and P. Ettmayer, Acta Mater., 1996,44(12), 4835.

26. D. Rafaja and W. Lengauer, Mater. Struct. Chem., Biol. Phys.Technol., 1997, 4(2), 43.

27. W. Lengauer, J. Bauer, M. Bohn, H. Wiesenberger, and P.Ettmayer, Mikrochim. Acta, 1997, 126, 279.

28. M. Dopita, B. Wollein, D. Rafaja, W. Gruner, and W.Lengauer, Defect Diff. Forum, 2001, 194–199, 1613; Proc.DIMAT 2000, Scitec Publications, Uetikon-Zurich, Paris,2000, Part 2, 17.

29. W. Mayr, W. Lengauer, D. Rafaja, J. Bauer, and M. Bohn, J.Phase Equilib., 1999, 20, 35.

30. W. Lengauer, D. Rafaja, R. Taubler, C. Kral, and P. Ettmayer,Acta Metall. Mater., 1993, 41(12), 3505.

31. W. Lengauer, D. Rafaja, G. Zehetner, and P. Ettmayer, ActaMater., 1996, 44(8), 3331.

32. M. Hoch, in Phase Stability in Metals and Alloys, eds P.S. Rudman, J. Stringer and R. I. Jaffee, McGraw-Hill, NewYork, 1967, p.419.

33. E. Horvath-Bordon, R. Riedel, A. Zerr, P. F. McMillan, G.Auffermann, Y. Prots, W. Bonger, R. Kniep, and P. Kroll,Chem. Soc. Rev., 2006, 36, 987.

34. A. Zerr, R. Riedel, T. Sekine, J. E. Lowther, W.-Y. Ching,and I. Tanaka, Adv. Mater., 2006, 18, 2933.

35. C. Politis and G. Rejman, Kernforschungszentrum KarlsruheGmbH, Karlsruhe (D), KfK Report Ext6/78-1, 1978.

36. A. Zerr, G. Miehe, and R. Riedel, Nat. Mater., 2003, 2, 185.

37. A. Zerr, N. Chigarev, O. Brinza, S. M. Nikitin, A. M.Lomosonov, and V. Gusev, Phys. Stat. Solidi, 2012, 6(12),484.

38. M. Lerch, E. Fuglein, and J. Wrba, Z. Allg. Anorg. Chem.,1996, 622, 367.

39. G. Brauer, J. Weidlein, and J. Strahle, Z. Allg. Anorg. Chem.,1966, 348, 298.

40. R. Marchand, F. Tessier, and F. J. DiSalvo, J. Mater. Chem.,1999, 9, 297.

41. M. Chowalla and H. E. Unalan, Nat. Mater., 2005, 4, 317.

42. W. Lengauer, Acta Metall. Mater., 1991, 39, 2985.

43. G. Lobier and J. P. Marcon, Compt. Rend. Acad. Sci. Ser. C(Paris), 1969, 268, 1132.


45. P. Kroll, Phys. Rev. Lett., 2003, 90(12), 125501.

46. E. Rudy, Metall. Trans., 1970, 1, 1249.

47. O. N. Carlsson, J. F. Smith, and R. H. Nafziger, Metall. Trans.A, 1986, A17, 1647.

48. C. Teichmann, W. Lengauer, P. Ettmayer, J. Bauer, and M.Bohn, Metall. Mater. Trans. A, 1997, 28A, 837.

49. W. Lengauer, M. Bohn, B. Wollein, and K. Lisak, Acta Mater.,2000, 48(6), 2633.

50. H. Wiesenberger, W. Lengauer, and P. Ettmayer, Acta Mater.,1998, 46(2), 651.

51. T. Onozuka, J. Appl. Crystallogr., 1978, 11, 132.

52. T. Onozuka, Trans. Jap. Inst. Met., 1982, 23, 315.

53. G. Brauer and W. Kern, Z. Anorg. Allg. Chem., 1983, 507,127.

54. G. Brauer and W. Kern, Z. Anorg. Allg. Chem., 1984, 512, 7.



55. R. Berger, W. Lengauer, and P. Ettmayer, J. Alloys Compd.,1997, 259, L9.

56. M. Joguet, W. Lengauer, M. Bohn, and J. Bauer, J. AlloysCompd., 1998, 269, 233.

57. G. Brauer, E. Mohr, A. Neuhaus, and A. Skokan, Monatsh.Chem., 1972, 103, 794.

58. A. Zerr, G. Miehe, J. Li, D. A. Dzivenko, V. K. Bulatov, H.Hofer, N. Bolfan-Casanova, M. Fialin, G. Brey, T. Watanabe,and M. Yoshimura, Adv. Funct. Mater., 2009, 19, 2282.

59. A. Friedrich, B. Winkler, L. Bayarjargal, E. A. J. Arellano,W. Morgenroth, J. Biehler, F. Schroeder, J. Y. Yan, and S. M.Clark, J. Alloys Compd., 2010, 502, 5.

60. C. Jiang, Z. J. Lin, and Y. Zhao, Phys. Rev. Lett., 2009, 103,185501.

61. P. Ettmayer and A. Vendl, in Science of Hard Materials, edsR. K. Viswanadham, D. J. Rowcliffe and J. Gurland, Plenumpress, New York (NY), 1983, p.47.

62. V. Ucakar, F. Chevire, F. Tessier, N. Krendelsberger,R. Marchand, and W. Lengauer, Defect Diff. Forum,2001, 194–199, 1607; Proc. DIMAT 2000, Paris, ScitecPublications, Uetikon-Zurich, 2000, Part 2, 17.

63. P. Ettmayer, Monatsh. Chem., 1970, 101, 1720.

64. A. Bezinge, K. Yvon, J. Muller, W. Lengauer, and P. Ettmayer,Solid State Commun., 1987, 63, 141.

65. P. Kroll, T. Schroeter, and M. Peters, Angew. Chem. Intl. Ed.,2005, 44, 4249.

66. G. Banik, P. Ettmayer, A. Vendl, and R. Kieffer, High Temp.High Press., 1979, 11, 349.

67. P. Ettmayer, A. Vendl, E. Horvath, and R. Kieffer, Monatsh.Chem., 1978, 109, 1277.

68. K. Frisk, Calphad, 1991, 15(1), 79.

69. S. Lampman, ASM Handbook, Heat Treating, ASM Interna-tional, Metals Park (OH), 1991, Vol. 4.

70. H. Holleck, Inst. Mater. Festkoerperforsch., Kern-forschungszent. Karlsruhe G.m.b.H., Karlsruhe, Fed. Rep.Ger. Avail. INIS. Report 1981, (KFK-3087-B) 353 pp.,From: INIS Atomindex 1982, 13(8), Abstr. No. 664151.Report written in German. CAN 97:79764 AN 1982:479764CAPLUS.

71. O. Knotek, W. D. Munz, and T. Leyendecker, J. Vac. Sci.Technol., 1987, A5, 2173.

72. K. Bobzin, E. Lugscheider, R. Nickel, and P. Immrich,Materialwiss. Werkstofftech., 2006, 37(10), 83.

73. I. Endler, M. Hohn, M. Herrmann, R. Pitonak, S. Ruppi, M.Schneider, H.van den Berg, and H. Westphal, Surf. Coat.Technol., 2008, 203, 530.

74. I. Endler, M. Hohn, M. Herrmann, H. Holzschuh, R. Pitonak,S. Ruppi, H.van den Berg, H. Westphal, and L. Wilde, Surf.Coat. Technol., 2010, 205, 1307.

75. I. Endler, M. Hohn, J. Schmidt, S. Scholz, M. Herrmann, andM. Knaut, Surf. Coat. Technol., 2013, 215, 133.

76. S. Veprek and M. J. G. Veprek-Heijman, Surf. Coat. Technol.,2008, 202, 5063.

77. C. Mitterer, P. H. Mayrhofer, W. Waldhauser, E. Kelesoglu,and P. Losbichler, Surf. Coat. Technol., 1998, 108, 230.

78. R. Franz and C. Mitterer, Surf. Coat. Technol., 2013, 228, 1.

79. K. H. J. Buschow, Rep. Prog. Phys., 1991, 54, 1123.

80. R. Ward, in MTP International Rewiev of Science, InorganicChemistry Series One, ed. D. W. A. Sharp, Butterworths,London, 1972, Vol. 5.

81. P. Ettmayer, R. Kieffer, and F. Hattinger, Metall(Berlin),1974, 28, 1151.

82. J. Perry, M. Georgson, and W. D. Sproul, Thin Solid Films,1988, 157, 255.

83. W. Lengauer, S. Binder, K. Aigner, P. Ettmayer, A. Guillou,J. Debuigne, and M. Bohn, J. Alloys Compd., 1995, 217, 137.

84. V. S. Eremeyev, Y. M. Iwanov, and A. S. Panov, Izv. Akad.Nauk. SSSR, Metall., 1969, 4, 262.


86. W. Lengauer, Surf. Interface Anal., 1990, 15, 377.

87. W. Lengauer, P. Ettmayer, and K. Yvon, J. Less-CommonMet., 1991, 168, L7.

88. W. Lengauer and P. Ettmayer, Monatsh. Chem., 1986, 117,713.

89. H. Matzke, in The Physics and Chemistry of Carbides, Nitridesand Borides, ed. R. Freer, Kluwer Academic Publishers,Dordrecht, 1990, p.357.

90. D. Rafaja and W. Lengauer, unpublished.

91. D. Rafaja, W. Lengauer, H. Wiesenberger, and M. Joguet,Metall. Mater. Trans. A Phys. Metall. Mater. Sci., 1998,29A(2), 439.

92. E. Torok, A. J. Perry, L. Chollet, and W. D. Sproul, Thin SolidFilms, 1987, 153, 37.

93. C. Kral, W. Lengauer, D. Rafaja, and P. Ettmayer, J. AlloysCompd., 1998, 265(1–2), 215.

94. B. A. Mathe, J. D. Comins, A.G. Every, and W. Lengauer,Int. J. Refract. Met. Hard Mater., 2014, 45, 212.

95. W. Lengauer and P. Ettmayer, J. Less-Common Met., 1985,109, 351.

96. Q. Yang, W. Lengauer, T. Koch, M. Scheerer, and I. Smid, J.Alloys Compd., 2000, 309, L5.

97. D. A. Dzivenko, E. Horvath-Bordon, A. Zerr, G. Miehe, P.Kroll, R. Boehler, P. McMillan, and R. Riedel, J. Phys.: Conf.Ser., 2008, 121, 062003.

98. A. Zerr, N. Chigarev, R. Brenner, D. A. Dzivenko, and V.Gusev, Phys. Stat. Solidi - Rapid Res. Lett., 2010, 4(353).

99. D. A. Dzivenko, A. Zerr, R. Boehler, and R. Riedel, SolidState Commun., 2006, 139, 255.

100. D. A. Dzivenko, A. Zerr, E. Schweitzer, M. Goken, R. Boehler,and R. Riedel, Appl. Phys. Lett., 2007, 90, 191910.

101. Q. N. Meng, M. Wen, C. Q. Qu, C. Q. Hu, and W. T. Zheng,Surf. Coat. Technol., 2011, 205, 2865.



102. K. Tsuda, A. Ikegaya, K. Isobe, N. Kitagawa, and T. Nomura,Powder Metall., 1996, 39, 296.

103. W. Lengauer and K. Dreyer, J. Alloys Comp., 2002, 338(1–2),194.

104. W. Lengauer and K. Dreyer, Int. J. Refract. Met. Hard Mater.,2006, 24, 155.

105. W. Lengauer, Materialwiss. Werkstofftechn., 2005, 36(10),460.

106. A. Eder, W. Lengauer, K. Dreyer, H. van den Berg, H.-W.Daub, and D. Kassel Proc. 16th Plansee Seminar, 2005, 3,120.

107. J. Gluhmann, M. Schneeweiß, H.van den Berg, D. Kassel, K.Rodiger, K. Dreyer, and W. Lengauer, Int. J. Refract. Met.Hard Mater., 2013, 36, 38.

108. N. Asada, Y. Yamamoto, T. Igarashi, Y. Doi, and K. Hayashi,J. Jap. Soc. Powder & Powder Metall., 1999, 46(4), 373.

109. K. Tanaka and K. Hayashi, J. Jap. Soc. Powder & PowderMetall., 1999, 46(11), 1154.

110. M. Wetzel, S. Wawrzik, C. Buchegger, C. Gierl, W. Lengauer,2012, Proc. EuroPM 2012, Basel (CH), Session 3, paper 507,CD ROM Edition, EPMA, Shrewsbury, UK, ISBN 978-1-1899072-31-1

111. S. Wawrzik, C. Buchegger, M. Wetzel, W. Lengauer, 2013,Proc. 18th Plansee Seminar 2013, paper HM17, PlanseeGroup, Reutte (A).

112. P. Ettmayer, Annu. Rev. Mater. Sci., 1989, 19, 145.

113. P. Ettmayer, H. Kolaska, W. Lengauer and K. Dreyer, Int. J.Refract. Met. Hard Mater., 1995, 13, 343.

114. J. Kim, M. Kim, M. Kang, and S. Kang, Int. J. Refract. Met.Hard Mater., 2013, 36, 278.

115. A. Demoly, C. Veitsch, W. Lengauer, and K. Rabitsch, Int. J.Refract. Met. Hard Mater., 2011, 29(6), 716.

116. A. Demoly, C. Veitsch, W. Lengauer, and K. Rabitsch, in:Advances in Sintering Science and Technology II: CeramicTransactions, 2012, Vol. 232, pp. 57–69, eds.: S.-J. L. Kang,R. Bordia, E. A. Olevsky, D. Bouvard, John Wiley & Sons,Hoboken NJ, ISBN 978-1-118-27375-3

117. S. P. Murarka, Defect Diffus. Forum, 1988, 59, 99.

118. H.von Seefeld, N. W. Cheung, M. Maenpaa, and M.-A.Nicolet, IEEE Trans. Electron. Devices, 1980, ED-27, 873.

119. H. Kim, J. Vac. Sci. Technol., 2003, 21, 2231.

120. C. Lee and Y. L. Kuo, JOM , 2007, 59, 44.


Documents

Nitrides: Transition Metal Solid-State Chemistry · Nitrides: Transition Metal Solid-State Chemistry Walter Lengauer TU-Wien—Laboratory of Physical Metallurgy, Vienna, Austria 1