Martienssen W., Warlimont H. (Eds.). Handbook of Condensed Matter and Materials Data

Подождите немного. Документ загружается.

322 Part 3 Classes of Materials

Table 3.1-114 Summary of fatigue data of refractory metals, pretreatment: Aw = as worked, Sr = stress relieved, Rxx =

recrystallized, cont.

Test Material Production Dimension (mm) Pretreatment Test conditions, S

(MPa) S

Ref.

mode

process plate: thickness, T(

◦

C)/t

(h) temperature,

bar: diameter frequency

High-frequency push–

pull fatigue, R =−1,

fatigue limit S

for 50%

fracture probability at

N = 10

Mo P/M bar 11 Rxx 1300/4 RT, 20 kHz 278 0.56 [1.176]

TZM P/M bar 11 Rxx 1600/2 RT, 20 kHz 383 0.69 [1.176]

Ta EB bar 3 Aw

RT, 20 kHz

383

286

0.72

0.92

[1.126]

Ta2.5W P/M plate 2 Rxx RT, 20 kHz 300 [1.126]

Ta10W P/M bar 12 Rxx RT, 20 kHz 580 [1.126]

Nb bar 3 Aw

Rxx

RT, 20 kHz

230

220

0.80

0.65

[1.177]

–1

–2

Cycles to failure N

Total strain range (%)

1088 K

As-received

Recrystallized

Room temperature

As-received

Recrystallized

∆

Fig. 3.1-172 Low-cycle fatigue data of as-received and re-

crystallized W at room temperature and 815

◦

C [1.186]

Low-cycle fatigue test data of Mo at high test tem-

peratures were reported in [1.187]. The inﬂuence of

microstructural changes in cold-worked molybdenum

on the low-cycle fatigue behavior was reported for

test temperatures between 300

◦

C and 950

◦

C [1.188].

Deformation experiments under low-cycle fatigue con-

ditions between room temperature and 100

◦

C showed

that recrystallized Mo, in spite of the low tempera-

ture and stress level, exhibits considerable plastic

strains which depend sensitively on the loading fre-

quency [1.189]. Data on the high temperature (350

◦

and 500

◦

C) isothermal mechanical fatigue behavior of

TZM were reported and a model for lifetime prediction

was proposed [1.183].

Fig. 3.1-173 K

of forged Mo rods (ϕ = 74%) versus de-

gree of recrystallization [1.147]

Fracture Mechanics Properties

Fracture Toughness. Fracture toughness properties are

affected by many parameters (thermomechanical pre-

treatments, microstructure, specimen and crack plane

orientation; testing procedures as well as the prepara-

tion of the starting notch and of the fatigue precrack).

Due to the peculiarities of the P/M production process,

it is frequently not possible to introduce sufﬁcient de-

formation into products of larger dimension in order to

completely eliminate sinter pores which may affect the

dynamic properties. The increase of fracture toughness

with increasing degree of hot working of disc-shaped

compact tension specimens, cut from a hot forged Mo

bar is shown in Fig. 3.1-173 and the decrease of frac-

ture toughness with increasing fraction of recrystallized

microstructure in Fig. 3.1-174 [1.147].

Fracture toughness data for Mo, TZM, and

W materials are summarised in Table 3.1-115. Data

taken at elevated temperature for Mo and W are

shown in Fig. 3.1-175 [1.192], and for TZM in

Fig. 3.1-176 [1.193]. Data for Nb could only be

0 20406080100

Fracture toughness K

(MPa m

)

Degree of recrystallization (%)

Part 3 1.9

Metals 1.9 Refractory Metals and Alloys 323

Table 3.1-115 Fracture mechanical data for various refractory metals, pretreatment: Aw = as worked, Sr = stress relieved, Rxx =

recrystallized. RT = room temperature

Fracture toughness data Specimen type: CT compact tension, SCT center surface cracked

tension, DCT disk-shaped compact tension, SNB side notched bend

CNT center through-thickness notched tension

Material Production Dimension (mm) Pretreatment Specimen Crack plane Test Fracture Ref.

process plate: thickness, T (

◦

C)/t

(h) type ASTM 399 temperature toughness

bar: diameter (MPa m

1/2

)

Mo P/M plate 6.5 Rxx 1300/2 CNT L-T RT 9.5 [1.126]

Mo P/M bar 50 Sintered DCT R-C RT 6 [1.147]

Hot forged DCT R-C RT 12

Rxx 1400/2 DCT R-C RT 5

Mo VA C plate 6.4 Sr CT L-T/T-L RT 20/21 [1.195]

Rxx CT L-T/T-L RT 18/19

Sr CT L-T/T-L 300

◦

C 74/70

Rxx CT L-T/T-L 300

◦

C 60/67

Mo5Re bar 54 Rxx 1300/1 DCT R-C RT/450

◦

C 18/27 [1.193]

Mo–0.3wt% P/M plate Sintered CT L-T RT 25 [1.196]

TZM VA C plate 6.4 Sr CT T-L/L-T RT 19/15 [1.195]

Rxx CT T-L/L-T RT 15/18

Sr CT T-L/L-T 300

◦

C 85/89

Rxx CT T-L/L-T 300

◦

C 59/68

TZM P/M bar 54 Aw DCT R-C RT 19 [1.159, 197]

Machined to 12 Aw SCT L-C RT 37

Swaged to 12 Rxx 1400/2 SCT L-C RT 19

W bar 10 SNB C-R/L-R −196

◦

C 11/6 [1.198]

SNB C-R/L-R RT 13/8

W bar 25 Aw SNB R-L RT/150

◦

C/ 9/10/13/14 [1.199]

300

◦

C/500

◦

W Single crystal bar 16 (111)[011] SNB −196

◦

C/RT 9/31 [1.198]

bar 16 (011)[100] SNB −196

◦

C/RT 4/20

bar 16 (100)[001] SNB −196

◦

C/RT 3/9

W Aw SNB RT/400/600/ 15/21/26/39 [1.200]

800

W5Re Aw SNB RT/400

◦

C/ 11/26/58 [1.200]

600

◦

W-1 wt% P/M bar 25 Aw SNB R-L RT/150

◦

C/ 9/10/11/13 [1.199]

300

◦

C/500

◦

Nb plate 1.25 Sr CNT L-T −253

◦

C 40 [1.192]

Rxx CNT L-T −253

◦

C 72

Nb Rxx SNB −196

◦

C 37 [1.194]

Rxx SNB RT 37

determined below −200

◦

C [1.192]. Data on the

impact and dynamic toughness of Nb between

−196

◦

C and 25

◦

C were reported in [1.194]. The dy-

namic cleavage fracture toughness was shown to be

37 MPa m

1/2

, relatively independent on grain size and

test temperature.

Part 3 1.9

324 Part 3 Classes of Materials

Table 3.1-115 Fracture mechanical data for various refractory metals, pretreatment: Aw = as worked, Sr = stress relieved, Rxx = recrystallized. RT = room

temperature, cont.

Fatigue crack growth and threshold data Specimen type: SCT center surface cracked tension

CNT center through-thickness notched tension SNT side notched tension

CCT corner cracked tension

Material Production Dimension Pretreatment Specimen Crack Test Stress da/ dN- K

K

th,eff

ASTM Ref.

process (mm) T (

◦

C)/t

(h) type plane conditions, ratio R range,

plate: thickness, ASTM temperature, (m/cycle)

bar: diameter 399 frequency

Mo P/M plate 6.5 Rxx 1300/2 CNT L-T RT, 20 kHz −1 10

−13

–10

−9

5.8 E647 [1.126]

Mo P/M bar 12 Aw SCT L-C RT, 20 kHz −1 10

−13

–10

−9

11.3 11 E647 [1.159]

Rxx 1400/2 10.7 11

Mo P/M bar 12 Sr 850/1 SNT C-R RT, 20 kHz −1 10

−13

–10

−9

10.3 7.0 E647 [1.201]

Rxx 1400/2 8.5 6.6

TZM P/M bar 12 Rxx 1700/2 SNT C-R RT, 20 kHz −1 10

−13

–10

−9

10.1 6.3 E647 [1.201]

Mo5Re P/M bar 12 Rxx 1300/1 SCT L-C RT, 20 kHz −1 10

−13

–10

−9

11.4 9.9 E647 [1.126]

Ta10W Rxx 1450/1 CCT RT 0.4 10

−9

–10

−5

[1.202]

20 kHz resonance test method, crack growth monitored by traveling light microscope,

∆K

corresponding to a crack growth rate of da/ dN < 10

−13

m/cycle,

∆K

th,eff

calculated from crack closure measurements based on strain gauge method

0 0.5 1 1.5 2

Fracture toughness K

(MPa m

)

In (A

/A)

Fig. 3.1-174 K

of forged Mo rods versus ln( A

/A)

[1.147]

100

–273

–200 –100 0 100 200 300 400 500

Test temperature (°C)

× 1000 (psi in.

)

Wrought,

stress relieved

Recrystal-

lized

Tungsten

Molybdenum

Columbium

Molyb-

denum

Tungsten

Fig. 3.1-175 Effect of test temperature on fracture

toughness of unalloyed W, Mo, and Nb sheet speci-

mens [1.192]

0 100 200 300 400 500

T(°C)

(MPa m

)

TZM

Mo5Re

Fig. 3.1-176 Effect of test temperature on static fracture

toughness of TZM and Mo5Re specimens [1.193]

Part 3 1.9

Metals 1.9 Refractory Metals and Alloys 325

Fatigue Crack Growth. Few data on the linear region of

crack growth were published for Ta10W (Fig. 3.1-177),

and a Nb

−

Zr alloy [1.202,203].

Threshold Stress Intensity for Fatigue Crack Growth.

The fatigue crack growth behavior of Mo, TZM,

and W in the region near the threshold stress inten-

sity, ∆K

, considered to correspond to fatigue crack

growth rates of da/dN < 10

−13

m/cycle, is shown in

Fig. 3.1-178 [1.126]. The available crack growth and

threshold data are included in Table 3.1-115.

An effective threshold value for fatigue crack

growth, ∆K

th,eff

, can be computed. Methods for the

determination of this effective threshold stress inten-

sity range are described in [1.197]. The available

data on ∆K

th,eff

of Mo and TZM are listed in

Table 3.1-115.

Short Fatigue Crack Growth Behavior. The nucle-

ation and growth behavior of short fatigue cracks is of

considerable practical and theoretical signiﬁcance. Dif-

ferences in the growth behavior exist between initial

short cracks (length comparable with microstructural

features) and long cracks of macroscopic dimen-

sions. The irregular growth rate of such short cracks

–3

–4

–5

–6

–7

510 50

∆K (MPa m

)

Crack growth rate da/dN (mm/cycle)

Ta–10W

Condition:

recrystallized and aged,

tests at RT,

stress ratio R=0.4

Eroded

notch

Fatigue

crack

8 mm

Fig. 3.1-177 Crack propagation behavior of recrystallized

P/M Ta10W specimens at room temperature at a stress

ratio of R = 0.4 [1.202]

–9

–10

–11

–12

–13

1 10 100

∆ K (MPa m

)

dc/dN (m/cycle)

= 11.4 MPa m

Fig. 3.1-178 Crack growth curve near threshold stress

intensity of a center surface cracked specimen ma-

chined from a recrystallized Mo5Re rod, tested at

a stress ratio of R =−1, test temperature = 50

◦

and 20 kHz cyclic frequency; open symbols test under

increasing load, solid symbols tests under decreasing

load [1.126]

(Fig. 3.1-179) may pose problems for a conservative

prediction of fatigue life [1.201]. Microscopic obser-

vations during fatigue exposure of specimens, loaded

at stress amplitudes slightly above the fatigue limit,

show initially a surface deformation very early in fa-

tigue life, followed by short crack initiation and growth

up to the ﬁnal long crack growth. The number of

fatigue cycles of the various stages depends on the mi-

crostructure and the presence of second phase particles

(Fig. 3.1-180) [1.204].

It is known that fatigue failures occur in defect-

containing materials after a high number of loading

cycles (N > 10

) at stresses considerably below the

fatigue limit determined by conventional test proce-

dures (N < 10

). A fracture mechanics approach to

this problem was proposed by Kitagawa et al. [1.205].

Based on a diagram relating a cyclic stress am-

plitude with crack length, a critical defect size

) can be deduced which, when exceeded, causes

a reduction of the fatigue strength. A modiﬁca-

tion of this diagram by introducing the value of

the effective stress intensity range is shown in

Fig. 3.1-181 [1.201]. Good agreement between pre-

dicted values and experimental results are obtained for

hemispherical surface notches of various sizes [1.206,

207].

Part 3 1.9

326 Part 3 Classes of Materials

3×10

2×10

1×10

0 100 200 300 400

1×10

–10

1×10

–11

1×10

–12

1×10

–13

0 100 200 300 400

Number of loading cycles N

Crack length c (µm)

Crack growth rate dc/dN (m/cycle)

Crack length c (µm)

1000

500

100

100 1000 10000

∆K

th, eff

(MPa)

Cyclic stress amplitude

Crack length c (µm)

Region of microcracks

Short crack growth behavior

(Stage I propagation)

Region of short cracks

with crystallographic

stage II growth

Region of long cracks

Mo stress relieved

Mo recrystallized

TZM recrystallized

Region of non-propagating

microcracks

Fatigue limit (

)

∆K

Fig. 3.1-181 Effect of crack length on stress amplitude for crack growth in specimens of stress relieved Mo, recrystallized

Mo, and recrystallized TZM (modiﬁed Kitagawa diagram) [1.201]

Fig. 3.1-179a,b Growth behavior of short surface cracks

in a TZM specimen, tested at a stress amplitude of

375 MPa [1.201,204].

(a) Crack length as function of num-

ber of loading cycles.

(b) Crack growth rate as function of

crack length

100

abcd

N/N

(%)

Macro

crack growth

SC-growth

Microcrack

initiation

Surface

deformation

No visible damage

Failure

Fig. 3.1-180 Fraction of total fatigue life spent on damage

accumulation and crack growth in specimens of stress re-

lieved Mo, recrystallized Mo, TZM with ﬁne particles, and

TZM with coarse particles [1.204]

Part 3 1.9

Metals 1.9 Refractory Metals and Alloys 327

Creep Properties

Creep-rupture data available up to 1970 were collected

in [1.107]. The fairly wide scatter results from minor

differences in microstructure, thermomechanical pre-

treatment, and impurity levels, but possibly also from

impurities picked up from the environment in the high-

temperature test systems. A summary of 10 000 h/1%

creep data for Mo, W, and TZM is given in Fig. 3.1-182;

the 100 h creep-rupture data for Mo, W, Nb, Ta, W25Re

and TZM are given in Fig. 3.1-183, based on [1.107]. Up

to 1100

◦

C the carbide-precipitation-hardened material

TZM reveals the highest creep strength, only outper-

formed by Mo and W alloys precipitation hardened with

hafnium carbide, which are not considered in these ﬁg-

ures. Comparing the stress causing a steady state creep

rate of 1 ×10

−4

/h, as illustrated in Fig. 3.1-184, it can

be concluded that precipitation hardening is effective up

to 1600

◦

C. At higher temperatures, the creep strength

of TZM deteriorates to the level of Mo or even be-

low. Above 1600

◦

C, W-based materials offer the best

performance.

The inﬂuence of the microstructure on the creep

mechanisms of Mo is illustrated in Fig. 3.1-185

for 1450

◦

C [1.208]. For a test stress of 35 MPa the

steady state creep rate is almost independent of the

grain size. In this stress regime a stress exponent of

5.4 was obtained, indicating dislocation creep as the

rate-controlling mechanism. Lowering the test stress to

14 MPa and 7 MPa, grain-size-dependent creep mech-

anisms, such as diffusion creep and grain boundary

sliding, become active, and as a consequence, the

steady state creep rate increases with decreasing grain

size [1.208].

The steady state creep rates of Mo rods made of

VAC ingots, VAC and P/M Mo sheets, and P/M Mo–

0.7wt%La

sheets are summarized in Fig. 3.1-186.

Stress exponents of 4.3 (P/M Mo sheet/1800

◦

C),

4.5–4.7(VACMosheet/1600

◦

C), 4.6(VACMo

rod/1600

◦

C), 5.0 (Mo–0.7wt%La

/1800

◦

C), and

5.5(VACMosheet/2200

◦

C) indicate that dislocation-

controlled creep is rate-controlling in the stress regime

investigated.

For AKS–W wires, used as lamp ﬁlaments, creep

resistance is one of the most important require-

ments. The ﬁne, potassium ﬁlled bubbles act as

an effective barrier against dislocation movement,

thereby reducing the deformation rate in the power-

Fig. 3.1-184 Stress for steady state creep rate of 1 × 10

−4

versus 1/T for Mo and TZM, deformed samples, various

shapes [1.208]

Temperature (°C)

600 800 1000 1200 1400 1600 1800 2000 2200

Stress for 1% elongation in 10 000h (MPa)

0.1

100

1000

W TZM Mo

Fig. 3.1-182 Comparison of 10 000 h/1% creep data for Mo, TZM,

and W based on [1.107]

100

1000

Stress (MPa)

Test temperature (°C)

1000 1100 1200 1300 1400 1500 1600 1700 1800 1900

TZM

W-25Re

Fig. 3.1-183 Comparison of 100 h of rupture data for selected re-

fractory metals based on [1.107]

0.1

100

1000

Stress for steady-state creep rate of 10

–4

0.0004

0.00045

0.0005

0.00055

0.0006

0.00065

0.0007

0.00075

0.0008

0.00085

0.0009

1/T (1/K)

Mo TZM

Part 3 1.9

328 Part 3 Classes of Materials

–1

–2

–3

–4

–5

–6

–7

Grain size (mm)

Steady state creep rate (1/h)

0.01 0.1 1 10

3.5 MPa

7 MPa

14 MPa

35 MPa

Fig. 3.1-185 Steady state creep rate of molybdenum sheets ver-

sus grain size. Sheet thickness = 2mm/6 mm; test temperature =

1450

◦

C; test atmosphere = hydrogen [1.208]

–1

–2

–3

–4

–5

–6

Steady-state creep rate of (1/h)

110

100

Test stress (MPa)

T=1600°C; arc-cast Mo sheet, th.=0,5mm

T=1600°C; arc-cast Mo (M9) sheet, th.=0,5mm

T=1600°C; arc-cast Mo rod, dia.=4,1mm

T=2200°C; arc-cast Mo sheet, th.=0,5mm

T=2200°C; arc-cast Mo (M9) sheet, th.=0,5mm

T=2200°C; arc-cast Mo rod, dia.= 4,1mm

T=1800°C; PM-Mo sheet, th.=6mm, grain size=0,1 mm

T=1800°C, MLR sheet,th.=1mm

Fig. 3.1-186 Steady state creep rate at various test temperatures

of vacuum-arc-cast (VAC) Mo sheet (thickness = 0.5mm), VAC

Mo rod (diameter = 4.1 mm), P/M Mo sheet (thickness = 6mm),

and Mo–0.7 wt% La

sheet (thickness = 1mm) versus test

stress [1.107,208]

law creep regime. Nabarro–Herring and/or Coble creep

is suppressed because of the large diffusion dis-

tances in a structure with large, highly elongated

grains. Grain boundaries resist sliding because of the

interlocking structure. A comparison of creep rup-

ture data of pure tungsten and two AKS–W grades

with different grain aspect ratios (GAR) is given in

Fig. 3.1-187.

In the high stress regime (> 60 MPa) and at tempera-

tures between 2500

◦

C and 3000

◦

C, stress exponents

between 8 and 25 were found [1.162,209,210]. This high

stress dependence led to the introduction of a threshold

stress (σ

) below which a component does not revealany

measurable creep deformation under usual service con-

ditions. For this threshold stress, which is lower than the

Orowan stress, the detachment of the dislocations from

the second phase particles or bubbles is the controlling

factor [1.211,212].

In the second phase particle/metal matrix interface,

the dislocation line energy is lower compared to the

dislocation line energy in the metal matrix. Of all dis-

persion strengthened materials investigated, potassium

bubbles in AKS–W exert the most attractive interaction

on dislocations [1.213].

For material produced in the 1970s with a mean,

but strongly scattered, grain aspect ratio of around

35, dislocation creep dominated at stresses > 60 MPa

(stress exponent = 25), as can be seen in Fig. 3.1-188.

For material produced 20 years later with a similar grain

aspect ratio of 31, a stress exponent of 1.2 was found

in the stress regime from 30 to 80 MPa, indicating

a diffusion-controlled creep process [1.214]. The eval-

uation of the strain rate/stress dependence of the values

10 100

Stress (MPa)

Time to rupture (min)

AKS-W (GAR = 4)

pure W

AKS-W (GAR > 12)

Fig. 3.1-187 Creep rupture data for AKS–W wires in com-

parison with pure tungsten. Wire diameter = 0.183 mm;

4 < GAR < 12; test temperature = 2527

◦

C; atmosphere

= vacuum better than 7 ×10

−5

Pa; heating rate = approxi-

mately 2000

◦

C/s [1.162]

Part 3 1.9

Metals 1.10 Noble Metals and Noble Metal Alloys 329

–6

–7

–8

–9

10 100

Stress (MPa)

Strain rate (1/s)

Zilberstein 1998

Wright 1978

Stress exponent= 25

Stress exponent= 1.2

Fig. 3.1-188 Strain rate versus stress for AKS–W wires

tested at 2527

◦

C. Wright 1978 [1.162]: AKS–W wire with

a diameter of 0.183 mm; atmosphere =vacuum better than

7×10

−5

Pa; heating rate =approximately 2000

◦

C/s; GAR

=35±10; pre-recrystallizedat 2527

◦

C/10 min. Zilberstein

1998 [1.215]: AKS–W wire with a diameter of 0.178 mm;

atmosphere =vacuum; GAR =31±1; pre-recrystallized at

2527

◦

C/15 min

Fig. 3.1-189 Time-to-rupture versus grain aspect ratio for

AKS–W wires with a diameter of 0.183 mm. Test tem-

perature = 2527

◦

C; test stress = 73.6 MPa; atmosphere

= vacuum better than 7 ×10

−5

Pa; heating rate = approxi-

mately 2000

◦

C/s [1.162]

generated under vacuum also reveals a stress exponent

close to 1 [1.215]. Data for the stress exponents are

summarized in [1.113].

By lowering the grain aspect ratio, the transi-

tion temperature between dislocation and diffusion

creep is shifted towards lower stresses. In the low

1000

100

020

Time to rupture (min)

Grain aspect ratio

GAR regime a strong dependence of the creep re-

sistance on microstructural features can be observed,

as grain boundary related phenomena, such as grain

boundary sliding and diffusion creep resulting in cav-

itations become rate-controlling. The inﬂuence of the

GAR on time to creep rupture is demonstrated in

Fig. 3.1-189.

3.1.10 Noble Metals and Noble Metal Alloys

The noble metals Ag, Au, Pd, Pt, Rh, Ir, Ru,

and Os are characterized by their positive reduction

potentials against hydrogen, high densities, high melt-

ing temperatures, high vapor pressures (Fig. 3.1-190),

high electrical and thermal conductivities, optical re-

ﬂectivity (Fig. 3.1-191), and catalytic properties. The

electronic density of states (DOS) near the Fermi

surface is nearly the same for all noble metals. Indi-

vidual differences of electrical conductivity, magnetic,

and optical behavior are related to different posi-

tions of the Fermi level relative to the DOS function

(Fig. 3.1-192). Small energy differences between their

outer s and d electronic states result in multiple oxidation

states.

Silver, Au, Pd, and Pt are comparatively soft and

ductile. Their hardness increases in the order Rh <

Ir < Ru < Os. Strengthening of the alloys is affected

by solid solution and dispersion hardening. The corro-

sion resistance against different agents decreases in the

order Ir > Ru > Rh > Os > Au > Pt > Pd > Ag.

Part 3 1.10

330 Part 3 Classes of Materials

–1

–2

–3

–4

–5

–6

–7

–8

–9

–10

–11

Vapour pressure (Torr)

Temperature (°C)

48001600

2000 2800

3200

3600

4000

4400400

800

1200

2400

Ag Au Pd Rh Pt Ru Ir Os

Fig. 3.1-190 Vapour pressures of the noble metals [1.216, p. 43]

0.9

0.8

0.7

0.6

0.5

0.4

400

460 520 580 640 700 760

Reflectivity R

Wavelength (nm)

Fig. 3.1-191 Optical reﬂectivity in the visible spectral range [1.217,

p. 173]

The purity grades of the elements are standard-

ized according to ASTM (American Society for Testing

and Materials) standards from 99.8to99.999 wt%:

Ag (B 413–69), Au (B 562–86), Pd (B 589–82), Pt

(B 561–86), Rh (B 616–78), Ir (B 671–91), Ru (B 717).

–8 –6 –4 –2 0 2 4

Density of state N(E)

Electronic energy E (eV)

d-band Fermi level

Fermi level

Hole

Electron

s-band

Ru, Os

Rh, Ir

Pd, Pt

Au Ag

Fig. 3.1-192 Schematic density of states (DOS) curve of

the noble metals [1.218, p. 30]

The pure elements and their alloys are key ma-

terials in electronics and electrical engineering (Ag,

Au, Pd, Pt, and Ru) and serve to manufacture high

strength, corrosion-resistant, high temperature, and

highly oxidation-resistant structural parts (Pt, Au, Rh,

and Ir). The platinum group metals, Ag, and Au, in both

the metallic state and in the form of chemical com-

pounds are effective heterogeneous or homogeneous

catalysts for a wide variety of chemical reactions. Tra-

ditional applications of noble metals and their alloys are

in dentistry (Au, Pt, Ag, Pd, and Ir), jewellery (Au, Ag,

Pt, Pd, Rh, and Ir), and in coins and medals (Au and

Ag).

3.1.10.1 Silver and Silver Alloys

Application

Silver and silver alloys are used for electrical contacts,

connecting leads in semiconductor devices, solders and

brazes, corrosion-resistant structural parts, batteries, ox-

idation catalysts, optical and heat reﬂecting mirrors,

table ware, jewellery, dentistry, and coins. Silver halides

are base components in photographic emulsions.

Production

Silver is extracted from ores through lead melts and

precipitation with zinc by the Parkes process. Zinc

is removed by distillation, while the remaining lead

Part 3 1.10

Metals 1.10 Noble Metals and Noble Metal Alloys 331

and base metals are removed by oxidation (cupella-

tion) up to ≈99% Ag. True silver ores are extracted

by cyanide leaching. High purity grades are pro-

duced by electrolysis. Bars, sheets, and wires are

produced by classical metallurgical processing, pow-

der by chemical and by electrolytic precipitation from

solutions, and nano-crystalline powder grades by dis-

persion in organic solutions. Coatings and laminate

structures are produced by cladding, by electroplat-

ing, in thick ﬁlm layers by applying pastes of silver

or in silver alloy powder with organic binder and glass

frits onto ceramic surfaces and ﬁring, in thin ﬁlm

coatings by evaporation, and by sputtering Compos-

ite materials are made by powder technology, or by

inﬁltration of liquid Ag into sintered refractory met-

als skeletons. Commercial grades of Ag are listed in

Table 3.1-116. Standard purities of crystal powder and

bars range from 99.9% to 99.999 wt% (ASTM B 413-

69) [1.217].

Phases and Phase Equilibria

Selected phase diagrams are shown in Figs. 3.1-193–

3.1-198 [1.219, 220]. Silver forms continuous solid

solutions with Au and Pd, with miscibility gaps occur-

ing in alloy systems with Mn, Ni, Os, P, Rh. Data for the

solubility of oxygen are given in Table 3.1-117 [1.216].

Thermodynamic data are given in Tables 3.1-118–

3.1-121. The entropy of fusion (L/T ) of completely

disordered intermetallic phases can generally be cal-

culated by fractional addition from those of the

components. For the completely ordered state the term

−19.146 (N

log N

) is to be added to

the calculated entropy of fusion [1.216–218, 221, 222].

The molar heat capacity of the homogeneous alloy

phases and intermetallic compounds, as calculated ap-

proximately from the atomic heat capacities of the

components using Neumann–Kopps’ rule, is obeyed

to within ±3% in the temperature range 0–500

◦

in the Ag

−

Au, Ag

−

Al, Ag

−

Al, and Ag

−

Mg al-

loy systems. The heat capacities of heterogeneous

alloys may be calculated by fractional addition from

those of the components by the empirical relation

= 4.1816(a +10

−3

bT +10

−2

)J/(Kmol) to sat-

isfactory accuracy.

Designation Grade (wt%) Impurity Maximum content (ppm)

“Good delivery” > 99.9 any, Cu 1000

Fine silver > 99.97 Cu/Pb/Bi/Se/Te 300/10/10/5/5

Fine silver 999.9 > 99.99 Cu/Pb/Bi/Se/Te 100/10/10/5/5

F. silver high pure < 99.999 Fe/Pb/Au/Cu/Cd/Bi/Se/Te 2/1/1/1/0.5/0.5/0.5/0.5/0.5

Table 3.1-116 Speciﬁcations

of ﬁne silver grades [1.217,

p. 51]

80 9010 20 30 40 50 60 70

1400

1300

1200

1100

1000

900

800

700

600

500

T (K)

Cu (at.%)

Cu (wt %)

51015202530354045506070 9080

Ag-Cu

1233.5 K

14.1 39.9

1052 K

95.1

1356K

(Ag)

(Cu)

Fig. 3.1-193 Binary phase diagram: Ag

−

Cu [1.219, p. 38]

80 9010 20 30 40 50 60 70

1900

1800

1700

1600

1500

1400

1300

1200

1100

T (K)

Ni (at.%)

Ni (wt %)

510 20 30 40506070 9080

1600

1500

1400

1300

1200

Ni (at. %)

T (K)

123Ag

Ag-Ni

Two liquids

1708 K

1233 K

1728 K

≈ 97

Liquid L

L + (Ni)

1233 K

(Ni)

1233.5 K

Fig. 3.1-194 Binary phase diagram: Ag

−

Ni [1.219, p. 88]

Part 3 1.10