ASM Metals HandBook Vol. 8 - Mechanical Testing and Evaluation

Подождите немного. Документ загружается.

fatigue crack growth rates are measured under conditions of a cyclic stress intensity (ΔK) at subcritical levels

(K < K

). Fatigue failures start at points of stress concentration and can be considered as flaws in the material.

As these flaws grow during the fatigue process, they can reach the critical size and lead to catastrophic failure

by rapid fracture. For this purpose, fatigue crack growth testing and analysis are used to determine the number

of cycles to reach the critical flaw size for a given material (with a fracture toughness, K

). These tests are

described in more detail in the article “Fatigue Crack Growth Testing” in this Volume.

References cited in this section

28. M.E. Fine and Y.-W. Chung, Fatigue Failure in Metals, Fatigue and Fracture, Vol 19, ASM Handbook,

ASM International, 1996, p 63

29. M.R. Mitchell, Fundamentals of Modern Fatigue Analysis for Design, Fatigue and Fracture, Vol 19,

ASM Handbook, ASM International, 1996, p 227–249

30. N.E. Dowling, Estimating Fatigue Life, Fatigue and Fracture, Vol 19, ASM Handbook, ASM

International, 1996, p 250–262

31. D.L. McDowell, Multiaxial Fatigue Strength, Fatigue and Fracture, Vol 19, ASM Handbook, ASM

International, 1996, p 263–273

32. D. Woodford, Design for High-Temperature Applications, Materials Selection and Design, Vol 20,

ASM Handbook, ASM International, 1998, p 580

Overview of Mechanical Properties and Testing for Design

Howard A. Kuhn, Concurrent Technologies Corporation

Creep

Many design applications involve materials and components that are subjected for extended periods of time to

high-temperature environments. For example, the tie bar shown in Fig. 1 could be part of a hanger supporting

steel parts in a furnace for heat treatment. Other examples include turbine blades in jet engines and pressure

vessels in high-temperature refinery operations. In these cases, failure of the material occurs by complex

diffusion-controlled phenomena leading to cavitation, creep elongation, and eventual rupture of the material.

Testing of materials in its simplest forms involves subjecting a tensile specimen to constant load or constant

stress within a high-temperature environment and measuring the elongation with time. A typical curve of creep

elongation versus time is shown in Fig. 36. After initial rapid growth in creep strain, the rate of creep strain

reaches a steady state, followed again by rapid growth to rupture. Increasing stress on temperature increases the

rate of creep strain and decreases the time to creep rupture. Figure 37(a) shows the influence of temperature and

stress on the time to rupture. Figure 37(b) shows the temperature dependence of yield strength, tensile strength,

and stress for creep rupture at 1,000 h and 100,000 h. Typically, the stresses for creep rupture are less than the

yield strength of the material, even for relatively short rupture times.

Fig. 36 Typical result for creep strain as a function of time

Fig. 37(a)

Fig. 37(b)

Typical data used for designs where materials are exposed to high temperatures for extended periods. (a)

Creep stress versus time to rupture for Astroloy. (b) Temperature dependence of yield strength, tensile

strength, and creep rupture strength at two different times for a nickel-based superalloy

In design applications to avoid creep rupture, the stresses on the material can be calculated as in the case of

static loads such as Eq 2, 12, and 13. The failure stress, σ

, used in these equations, however, would be the

rupture strength of the material determined from curves such as Fig. 37(a) at the operating temperature of the

components in the design application. Depending on the expected life of the part, then, the operating

temperature will determine the maximum stress that can be applied. Alternatively, for a given operating stress

and expected lifetime, the maximum operating temperature can be determined from data such as Fig. 37(a) (Ref

32).

In design applications where creep occurs, elongation of the material often becomes the limiting failure

parameter rather than creep rupture strength. A major example is the elongation of turbine blades during turbine

engine operation. If the blades elongate too much, they contact the internal parts of the engine leading to

catastrophic failure. In this case, elongation would be described by the creep strain profile in Fig. 36, for

example, rather than the elastic deflection calculated in Eq 4. Thus, for a given operating temperature and load,

the lifetime for successful operation of a component would be obtained from data such as that provided in Fig.

36 for the specified limit on elongation.

Because of the importance of creep failure in many applications, refined test procedures have been developed

for accurate measurement of creep behavior under complex thermal and stress histories, as well as various

corrosive environments as described in the Section “Creep and Stress-Relaxation Testing” in this Volume.

Reference cited in this section

32. D. Woodford, Design for High-Temperature Applications, Materials Selection and Design, Vol 20,

ASM Handbook, ASM International, 1998, p 580

Overview of Mechanical Properties and Testing for Design

Howard A. Kuhn, Concurrent Technologies Corporation

Environmental Effects on Mechanical Properties

The mechanical properties of metals can be adversely affected when the metal is exposed to a corrosive

environment while being simultaneously stressed. Even in only mildly corrosive environments, the

consequences can be unexpected and serious. These mechanisms that cause adverse affects are collectively

known as environmentally assisted cracking and include some of the more common mechanisms by which

metals actually fail in service. The most important of the phenomena are stress corrosion, hydrogen

embrittlement, and corrosion fatigue.

Stress-corrosion cracking (SCC) of metals occurs in certain environments when cracks initiate and propagate

under conditions where neither the stress nor the environment acting alone would have caused cracking. The

propagation of SCC may eventually lead to structural failure or at least to ineffective performance of a

component (due to problems such as leaks and distortion). Stress-corrosion cracks initiate only at a surface—

and only at a surface that is exposed to the damaging environment. Once initiated, they propagate laterally into

the section thickness, by either transgranular cracking (across grains, Fig. 38) or intergranular cracking (along

grain boundaries, Fig. 39 ). Most of the surface of a stress-cracked metal is essentially unattacked. Stress

cracking may occur at relatively low stress levels compared with stress needed for failure (tensile strength) and

at relatively low concentrations of chemicals (such as Cl

for austenitic stainless steels).

Fig. 38 Transgranular stress-corrosion cracking (SCC) in annealed 310 stainless steel after prolonged

exposure in a chloride-containing environment. Electrolytic: 10% chromic acid etch. 150×

Fig. 39 Branched intergranular SCC in a lightly drawn tube of 12200 copper after exposure to an amine

boiler-treatment compound. Potassium dichromate etch. (a) 100×. (b) 500×

Although the mechanism of stress corrosion is not completely understood, SCC appears to develop only when

the stress (applied or residual) is tensile in nature and when its magnitude exceeds a threshold value (Fig. 40), a

value that is near the yield strength of the metal. The rate of crack propagation thereafter increases rapidly with

increasing stress. Once initiated, stress-corrosion cracks extend laterally and grow into the underlying section in

a plane that is normal to the principal tensile stress. They tend to branch extensively in the process (Fig. 38 and

39 ), but there are occasional exceptions to this. Only a limited number of separate cracks are likely to initiate

and then tend to be confined to a limited area of a component. This is in contrast to the more widespread and

random nature of intercrystalline corrosion.

Fig. 40 Stress corrosion cracking threshold examples. (a) Stainless steels in boiling 42% magnesium

chloride solution. (b) Comparison of K

ISCC

of AISI 4340 steel (tensile yield strength, 1515 MPa, or 220

ksi) in methanol and salt water at room temperature

The environments that induce SCC are specific to particular metals, and only a limited range of environments

can cause cracking in any one metal. Some examples of cracking environments are listed in Table 7, most of

them being at worst only mildly corrosive in a general sense. The presence of oxygen is important in most of

them. Many of these environments are also likely to be encountered in everyday usage, and some are virtually

impossible to avoid. Moisture containing chlorides that will cause cracking of aluminum alloys is a case in

point, because moisture and traces of chlorides are ubiquitous. Traces of ammonia that can cause cracking of

brasses are also frequently present in the atmosphere due to the decomposition of organic matter and the

presence of animal waste products (Ref 33).

Table 7 Alloy/environment systems exhibiting stress-corrosion cracking

Alloy

Environment

Carbon steel

Hot nitrate, hydroxide, anhydrous ammonia, and carbonate/bicarbonate solutions

High-strength steels

Aqueous electrolytes, particularly when containing H

Austenitic stainless

Hot, concentrated chloride solutions; caustics, saline solution, and chloride-

steels

contaminated steam

High-nickel alloys

High-purity steam, hot caustics

Aluminum alloys

Aqueous Cl

, Br

, and I

solutions, including contaminated water vapor

Titanium alloys

Aqueous Cl

, Br

, and I

solutions; methanol organic liquids; N

; hydrochloric

acid

Magnesium alloys

Aqueous Cl

solutions

Zirconium alloys

Aqueous Cl

solutions; organic liquids; I

at 350 °C (660 °F)

Copper alloys

Ammonia and amines for high-zinc brasses; ammoniacal solutions for α brass;

range of solutions for other specific alloys

Gold alloys

(a)

Chlorides, particularly ferric chloride; ammonium hydroxide; nitric acid

(a) Alloys containing less than 67% gold

Prevention of SCC. Stress cracking may be reduced or prevented by the following practices:

• Decreasing the stress level by annealing, design, and so forth

• Avoiding the environment that leads to stress cracking

• Changing the metal if the environment cannot be changed

• Adding inhibitors or applying cathodic protection to reduce the rate of corrosion

In principal, the easiest way to prevent SCC is to specify the load and geometry for a given measured or

assumed initial flaw size such that K < K

ISCC

(where K

ISCC

is the critical stress intensity for stress-corrosion

cracking). For some alloys, however, the value of K

ISCC

may be so low that impossible initial flaw sizes or

impractically low stresses must be specified. Alternatively, attention may be focused on the use of coatings,

alternate materials, or other means of corrosion protection. It is also important to understand the nature of crack

growth and avoid the conditions leading to fast fracture; that is, the stress intensity factor due to flaws

developed by corrosion must be kept below K

, the fracture toughness of the material, by limiting the stress or

time of exposure so that cracks do not grow to the critical size.

Measuring and testing of SCC behavior is a complex subject, but one approach is to subject a material

specimen containing a prescribed defect to stress in the chemical environment (Fig. 41) (Ref 19). The increase

in K (usually resulting from increase in crack length a) is then measured as a function of time. As a increases,

the applied stress, σ, may change, and the factor Y in Eq 37 may also change. Their combined effects contribute

to the increase in K. Figure 42 shows a typical result (Ref 19). For each initial stress intensity value, K

, the

stress intensity increases until it reaches K

, and fast fracture occurs. Below a certain value of K

, crack growth

does not occur; this level of K

is denoted K

ISCC

, or the critical stress intensity for stress corrosion cracking in

the environment under which the test was conducted. For example, in saltwater, K

ISCC

for heat treated 2000 and

7000 series aluminum alloys are approximately 80% of their K

values. For heat treated 4340 and 300M steels,

ISCC

is about one-third of the K

values, and for titanium alloys, K

ISCC

varies widely from 25 to 40% of their K

values (Ref 19).

Fig. 41 Modified compact tension specimen for environmentally assisted cracking measurement.

Fig. 42 Stress intensity growth with time in a corrosive environment

Hydrogen Embrittlement and Cracking. Hydrogen embrittlement is another form of environmentally assisted

crack growth. Only a comparatively few metals are susceptible to this phenomenon, but prominent among them

are the high-strength steels, that is, steels having tensile strengths above about 1000 MPa (145 ksi). In some

aspects, hydrogen embrittlement is similar to stress-corrosion cracking. Hydrogen-induced cracking (HIC) has

been proposed as the SCC mechanism for carbon and high-strength ferritic steels, nickel-base alloys, titanium

alloys, and aluminum alloys (Ref 34).

Cracking resistance of steels is a major concern in refining and petrochemical industries where aqueous H

S is

present. The generally accepted theory of the mechanism for hydrogen damage in wet H

S environments is that

monatomic hydrogen is charged into steel as a result of sulfide corrosion reactions that take place on the

material surface. The primary source of atomic hydrogen available at internal surfaces of pipeline and vessel

steels is generally the oxygen-accelerated dissociation of the H

S gas molecule in the presence of water. The

basic reaction is:

The FeS formed on the surface of the steel is readily permeated by atomic hydrogen, which diffuses further into

the steel.

This diffusion of atomic hydrogen into steel is associated with three distinct forms of cracking:

• Hydrogen-induced cracking

• Stress-oriented hydrogen-induced cracking

• Hydrogen stress cracking (also known as sulfide stress cracking and sulfide stress-corrosion cracking)

Hydrogen-induced cracking (HIC) and stress oriented hydrogen-induced cracking (SOHIC) are both caused by

the formation of hydrogen gas (H

) blisters in steel. Hydrogen-induced cracking, also called stepwise cracking

or blister cracking, is primarily found in lower-strength steels, typically with tensile strengths less than about

550 MPa (80 ksi). It is primarily found in line pipe steels.

In contrast, hydrogen stress cracking does not involve blister formation, but it does involve cracking from the

simultaneous presence of high stress and hydrogen embrittlement of the steel. Hydrogen stress cracking occurs

in higher-strength steels or at localized hard spots associated with welds or steel treatment. As a general rule of

thumb, hydrogen stress cracking can be expected to occur in process streams containing in excess of 50 ppm

S (although cracking has been found to occur at lower concentrations).

The basic factors of these cracking modes include temperature, pH, pressure, chemical species and their

concentration, steel composition and condition, and welding or the condition of the weld heat-affect zone.

These types of cracking and important variables for failure control are described in more detail in Ref 35 and

36.

Corrosion Fatigue. As previously noted and shown in Fig. 35, fatigue is affected seriously in the presence of a

corrosive environment. Another consequence is that even those alloys that have definite fatigue endurance

limits no longer do so. The presence of a particular environment is not required for the deterioration in

properties, as it is for SCC. The sole requirement is that the environment be sufficiently corrosive, although

there is not necessarily a direct correlation between general corrosiveness and effect on corrosion fatigue.

For steels, the corrosion endurance limit ranges from about 50 to 10% of the limit in air. The corrosion

endurance limit also is independent of metallurgical structure and thus shows little correlation with strength.

Therefore, the endurance limit of steels, even under mildly corrosive conditions, is much less than that in air

and does not increase with an increase in the tensile strength of the steel (Ref 33). The combination of corrosion

with a cycling stress eliminates the benefits of all efforts made to improve the strength of steels as assessed by

static mechanical tests.

References cited in this section

19. R.W. Hertzberg, Deformation and Fracture Mechanics of Engineering Materials, 2nd ed., John Wiley

& Sons, 1983, p 240, 287, 288, 436–477

33. L. Samuels, Metals Engineering: A Technical Guide, ASM International, 1988, p 161, 168

34. G. Koch, Stress-Corrosion Cracking and Hydrogen Embrittlement, Fatigue and Fracture, Vol 19, ASM

Handbook, 1996, p 486

35. P. Timmins, Failure Control in Process Operations, Fatigue and Fracture, Vol 19, ASM Handbook,

1996, p 479

36. P. Timmins, Solutions to Hydrogen Attack in Steels, ASM International, 1997

Overview of Mechanical Properties and Testing for Design

Howard A. Kuhn, Concurrent Technologies Corporation

Shock Loading

Another nonstatic loading condition often found in machine parts involves shock, or impact forces. This

condition occurs if the time duration of the load is less than the natural period of vibration of the part or

structure. Failure of a part under shock loading, as with other types of loading, depends on material parameters

and geometric factors.

To illustrate this condition, consider the tie bar (Fig. 1) under impact tensile loading. If the bar is used to stop

the motion of another part, then the kinetic energy of the moving part is absorbed by elongation of the tie bar

and converted into elastic strain energy in the bar. Then, the maximum stress in the bar will be:

σ = V

(Eq 37)

where V is the velocity of the mass, m, when it impacts the bar; A and L are the cross-sectional area and length

of the bar, respectively; and E is the elastic modulus of the bar material.

To prevent failure under a shock load, the stress in Eq 37 must be less than the strength of the material, σ

Then, the parameters in Eq 37 can be regrouped into:

V < (σ

/ )( )

(Eq 38)

which shows that the combination of impact velocity and mass that can be tolerated depends not only on the

material strength but also on its elastic modulus. In other words, selecting a material that maximizes the

parameter /E will maximize the impact energy that can be absorbed. Equation 38 also shows that increasing

the volume of material in the bar, AL, increases its shock resistance. For example, increasing the length of the

bar does not affect its ability to carry static axial load but does increase its ability to resist an impact load.

The description of shock resistance of a tie bar can be extended to other forms of loading, such as bending. For

example, for a beam subjected to a lateral impact load (Fig. 6), the kinetic energy of a mass that is stopped by

the beam is converted into strain energy in the beam. Then, the maximum stress that is in the beam is given by:

σ = V H

(Eq 39)

where V is the velocity of mass m when it impacts the beam; L and H are the beam length and height,

respectively; I is the moment of inertia of the beam cross section; and E is the elastic modulus of the beam

material. Again, σ must be less than the strength of the material, σ

, and Eq 39 can be rearranged to:

V < (σ

/ )( )

(Eq 40)

As with the bar under axial impact loading, the material parameter that maximizes shock resistance is /E, but

the geometry parameter is now LI/H

. In the case of rectangular beam, from Eq 18, I = bH

/12. Then, the

geometry parameter becomes LbH/12. Since bH is the cross-sectional area of the beam, the geometry parameter

becomes AL/12, so the impact resistance of the beam is increased by increasing the beam volume, similar to the

case of the bar under tensile impact. A longer beam, for example, will not carry as much static load as a shorter

beam, but the longer beam will be more resistant to failure by impact loading.

More complicated geometries can be analyzed through finite element models, which give the distribution of

stresses due to impact loads throughout the part. In these cases the same general effects of the material

parameter ( /E) and geometry parameter (volume) on shock resistance apply.

Overview of Mechanical Properties and Testing for Design

Howard A. Kuhn, Concurrent Technologies Corporation

Acknowledgments

Portions of the section “Environmental Effects on Mechanical Properties” were adapted from L.E. Samuels,

Metals Engineering: A Technical Guide, ASM International, 1988.