Czichos H., Saito T., Smith L.E. (Eds.) Handbook of Metrology and Testing

Подождите немного. Документ загружается.

718 Part D Materials Performance Testing

to be considered in addition to the hydrogen content.

A tensile test cannot always prove the embrittling ef-

fects of hydrogen caused by diffusion processes. With

high strain rates hydrogen-charged samples show nearly

the same reduction of area as samples free of hy-

drogen. It is only at very low straining rates that

hydrogen results in brittle behavior. The reason for

this phenomenon is that with high deformation rates

not enough hydrogen can be provided to cause a brit-

tle fracture. Ductile behavior can be observed, i. e.

a minimum time period is necessary for crack initiation

and crack propagation of a critical length [12.74, 86].

Fractography can provide proof of hydrogen embrittle-

ment even in those cases where there are only slight

changes in mechanical properties. Undeniable indica-

tions can be found in the surface area of the crack

breakthrough. Hydrogen-embrittled quenched and tem-

pered steels always have an intergranular crack zone.

This is clearly different from the behavior of notch-

based brittle fractures or ductile spontaneous fractures.

Apart from the crack area the other parts of the

fracture surface could look like notch-based brittle

fractures.

12.7.5 Test Methods

For testing the susceptibility of a material against

hydrogen-induced stress corrosion cracking or deter-

mining the effect of speciﬁc environmental conditions

on a metal with regard to hydrogen embrittlement

all test methods previously described in Sects. 12.5

and 12.6 are applicable. In addition to these tests also

the uptake of hydrogen in an electrochemical cell which

has been proposed by Devanathan can be investigated

and the data evolved may be used for determining the

diffusion coefﬁcient of hydrogen in a metal or, if the dif-

fusion coefﬁcient is known, provide data on hydrogen

activity at the surface of the metal.

12.8 High-Temperature Corrosion

In industrial high-temperature applications corrosion is

a major life-limiting factor for in-service components.

Such applications are e.g. fossil-fuel power stations,

petrochemical and chemical plants, exhaust ducts, land

based gas turbines and jet engines, etc. There exist

a number of tests aiming at evaluating the effect of

the high-temperature environment on material perfor-

mance for the different applications that in most cases

are laboratory tests, although some ﬁeld testing may

also be performed. In the laboratory tests it is usu-

ally intended to simulate conditions which are close

to the industrial situation with regard to environment,

temperature and mechanical stresses. As the situation

in high-temperature applications in most cases is very

complex it is, however, not easy to simulate the actual

conditions on a one-to-one basis. Therefore most of the

high-temperature corrosion tests in the laboratory are

based on a simpliﬁcation of the practical conditions or

on conditions leading to an acceleration of the test. In

the latter case it should, however, be kept in mind that

accelerated testing may lead to a change in the corro-

sion mechanism and, thus, possibly to misinterpretation

of the data for the material performance under practical

conditions. Furthermore extreme care should be taken if

the results of such accelerated laboratory tests are taken

for life time extrapolation.

12.8.1 Main Parameters

in High-Temperature Corrosion

Corrosion Testing

The basic principles for high-temperature corrosion

testing are relatively easy. Actually there are two types

of parameters which are usually measured. The ﬁrst is

parameters related to mass change that results from the

reaction of the material with its environment by which

either a mass gain or a mass loss can take place. A mass

gain will occur if the reaction leads to the formation of

a solid corrosion product on the surface and/or in the

metal subsurface zone. Mass losses will occur when ei-

ther the solid corrosion product spalls from the surface

(e.g. oxide bits ﬂake off the sample) or if volatile cor-

rosion products are formed which evaporate. Since the

mass change is proportional to the amount of species

from the environment taken up by the material in the ox-

idation reaction or to the amount of material evaporated

by formation of volatile corrosion products the high-

temperature corrosion kinetics can easily be determined

by continuous or discontinuous measurements of the

specimen weight or mass in tests. Such mass measure-

ments can of course also be performed with specimens

from ﬁeld exposure so that it becomes possible to deter-

mine corrosion kinetics also in ﬁeld testing.

Part D 12.8

Corrosion 12.8 High-Temperature Corrosion 719

In combination with mass measurements it is beneﬁ-

cial and even necessary to characterize additional dam-

age parameters resulting from the corrosion process.

These parameters are determined by post-experimental

sectioning of the specimens in the metallography and

measurement of geometrical data on surface scale thick-

ness and on internal corrosion depth. Again this method

can also be applied to specimens from ﬁeld testing. In

the present chapter it is intended to focus on laboratory

testing. Actually many of the aspects dealing with labo-

ratory testing (especially those on the evaluation of the

corrosion data) can also be applied to ﬁeld testing. With

regard to the speciﬁc aspects of ﬁeld testing, please refer

to [12.87].

12.8.2 Test Standards or Guidelines

Although high-temperature corrosion testing has been

performed in industry and scientiﬁc laboratories now

for almost a century there are very few guidelines or

standards describing the test procedures. In the follow-

ing the more important of these standards or guidelines

are listed.

•

ASTM G54Practice for simple static oxidation

testing [12.88]:

This standard covers the determination of the rela-

tive growth, scale, and microstructural characteris-

tics of an oxide on the surface of a pure metal or

alloy under isothermal condition in still air. This

standard was later withdrawn but is still regarded as

useful.

•

ASTM B76Method for accelerated life test of

nickel-chromium and nickel-chromium-iron alloys

for electrical heating [12.89]:

This standard covers the determination of the resis-

tance to oxidation of nickel-based electrical heating

alloys at elevated temperatures under intermittent

heating using a constant-temperature cycle test.

This test method is used for comparative purposes

only.

•

ASTM B78Method for accelerated life test of

iron-chromium-aluminum alloys for electrical heat-

ing [12.90]:

This standard covers the determination of the re-

sistance to oxidation of iron-chromium-aluminum

alloys for electrical heating at elevated temperatures

under intermittent heating similar to ASTM G 76.

This test is again used for comparative purposes

only.

•

ASTM G79Practice for evaluation of metals ex-

posed to carburization environments [12.91]:

This standard covers procedures for the identiﬁca-

tion and measurement of the extent of carburization

in a metal sample and for the interpretation and

evaluation of the effects of carburization. It applies

mainly to iron- and nickel-based alloys for high-

temperature applications.

•

Draft code of practice: TESTCORR – Discontinuous

corrosion testing in high-temperature gaseous envi-

ronment (ERA report 2000-0546) [12.92]:

This draft code of practice, which is the result

of a research programme supported by the CEC

standards, measurements and testing programme

describes in detail the procedures for the deter-

mination of specimen wastage using mass and

dimensional change. A major consideration is the

traceability of the measurements and the validity

of the statistical treatment of the data generated.

These procedures have been developed in con-

junction with specimen preparation guidelines and

recommendations for a statistical analysis of the

data.

•

JIS (Japanese Institute of Standards) Z 2281-

1993 Test method for continuous oxidation test

at elevated temperatures for metallic mater-

ials [12.93]:

This Japanese standard covers one of the most often-

used tests in high-temperature corrosion. A tentative

translation from the Japanese manuscript into En-

glish is available from JIS.

•

JIS Z 2282-1996 Method of cyclic oxidation test-

ing at elevated temperatures for metallic mater-

ials [12.94]:

This standard covers another most often used high-

temperature corrosion test in industry and is also

available as a tentative translation from Japanese

into English.

•

EFC publication no. 14 Guidelines for methods of

testing and research in high temperature corro-

sion [12.95]:

This publication contains the papers of a workshop

on this topic organized by the working party on cor-

rosion by hot gases and combustion products of the

European Federation of Corrosion where all major

aspects of high-temperature corrosion testing were

addressed.

•

EFC publication no. 27 cyclic oxidation of high tem-

perature materials – mechanisms, testing methods,

characterization and life time estimation [12.96]:

Part D 12.8

720 Part D Materials Performance Testing

This publication is based on the papers presented

at a workshop of the same title again organized by

the aforementioned working party of the European

Federation of Corrosion. The workshop was solely

devoted to thermal cycling oxidation testing.

Especially at the last mentioned workshop with strong

international participation from around the world it was

stated that the situation with regard to standardization

of high-temperature corrosion testing is rather unsat-

isfactory at present. As a consequence two initiatives

were started. One was in the form a European research

project named COTEST [12.97] which aimed at the de-

velopment of a code of practice for a test method for

thermal cycling oxidation testing. The results of this

work have been published as a book [12.98]. The second

initiative was started by Japanese industrial researchers

aiming at the development of an ISO standard based

on the existing Japanese standards mentioned above.

A workgroup was formed in the ISO technical commit-

tee 156 (corrosion of metals and alloys) which has the

designation WG 13, high-temperature corrosion. The

aim of this work group is to develop ISO standards in

the following ﬁelds.

•

Thermogravimetric testing:

ISO work group deﬁnition: in situ mass measure-

ments at elevated temperatures on a single specimen

without intermediate cooling.

•

Continuous isothermal exposure testing:

ISO work group deﬁnition: single post-exposure

Table 12.4 Draft documents of ISO TC 156 WG 13 on high temperature corrosion testing to become international

standards

Title Voting Stage (as of July 2010)

ISO/NP 21608: Corrosion of metals and alloys –

Test method for isothermal exposure oxidation testing under high temperature

corrosion conditions for metallic materials

30.99 (CD approved for registration as DIS)

ISO/CD 13573: Corrosion of metals and alloys –

Test method for thermal cycling exposure testing under high temperature corrosion

conditions for metallic materials

30.20 (CD study/ballot initiated)

ISO/CD 26146: Corrosion of metals and alloys –

Method for metallographic examination of samples after exposure to high

temperature corrosive environments

30.20 (CD study/ballot initiated)

New documents (under work as of July 2010):

1 Corrosion of Metals and Alloys Test method for high-temperature corrosion testing of metallic materials by fully embedding in

salt, ash, or other inorganic solids

2 Corrosion of Metals and Alloys Test method for high-temperature corrosion testing of metallic materials by partially embedding

in salt, ash, or other inorganic solids

3 Corrosion of Metals and Alloys Test method for high-temperature corrosion testing of metallic materials by immersing in molten

salt or other inorganic liquids

4 Corrosion of Metals and Alloys Test method for high-temperature corrosion testing of metallic materials by coating with salt,

ash, or other inorganic solids

mass measurement on a series of specimens without

intermediate cooling.

•

Discontinuous isothermal exposure testing:

ISO work group deﬁnition: series of mass mea-

surements on a single specimen with intermediate

cooling at predetermined times not necessarily reg-

ular.

•

Thermal cycling oxidation testing:

ISO work group deﬁnition: series of mass measure-

ments on a single specimen with repeated, regular

and controlled temperature cycles.

•

General guidelines for a post-exposure examination:

No deﬁnition by the ISO work group.

In the development of these work items the knowl-

edge existing in the form of the national standards and

European guidelines mentioned above is being used.

This particularly concerns the results from the TEST-

CORR document and the COTEST document as well

as the Japanese standards. These ISO standards are

expected to be the ﬁrst documents with international

acceptance in the ﬁeld of high-temperature corrosion

testing. The current documents of ISO WG13 are listed

in Table 12.4. As these documents are still under devel-

opment, in the present chapter the most recent situation

with regard to the working drafts of these standards will

be taken into account. This chapter on high-temperature

corrosion testing, therefore, cannot yet rely on generally

accepted existing standards but the content will be pre-

sented in a way which is supposed to be compatible with

the future ISO standards.

Part D 12.8

Corrosion 12.8 High-Temperature Corrosion 721

12.8.3 Mass Change Measurements

Principles

The basic idea behind the mass change measurements

is to have an easy method providing information about

the oxidation or corrosion kinetics. This kinetics al-

low an assessment of materials performance from the

viewpoint of oxidation or corrosion resistance accord-

ing to the schematic representation in Fig. 12.48. The

ideal situation with regard to high-temperature corro-

sion resistance is the formation of a very thin and

dense protective oxide scale on the surface of the mater-

ial. This behavior is usually characterized by parabolic

kinetics with slow scale growth rates, i. e. the mass in-

crease or the increase of the oxide scale thickness follow

a parabolic time law

(Δm/A)

t , (12.84)

where Δm is the mass increase, A is the specimen sur-

face area, k

is the parabolic oxidation rate constant and

t is oxidation time. A slow oxidation rate means that k

has a very low value. Equation (12.84) can also be writ-

ten in terms of the increase of the oxide scale thickness

according to



t , (12.85)

where x is the oxide scale thickness. As in many cases

oxidation does not strictly follow a parabolic rate law,

(12.84)and(12.85) can also be written in the more gen-

eral form including the temperature dependence [12.99]

Δm/A =

[

k(T)

]

1/n





1/n

, (12.86)

k(T ) =k



−E



(12.87)

with k(T) as a temperature-dependent oxidation rate

constant (for the parabolic case k(T) =k

), t

as the

time basis (e.g. 1 s if seconds are used for t), k

the general oxidation rate constant, E

as the activa-

tion energy, T as temperature and R as the general gas

constant. The exponent n characterizes the type of rate

law, e.g. n = 2 for the parabolic case. If oxidation or

high-temperature corrosion occur only via the forma-

tion of a surface layer the mass change results can be

converted into scale thickness data by

Δm/A = xf

(12.88)

with f

=ρ

MeX

(12.89)

Non-

protective

Temporary

protective

Protective

Δ m/A

–1

Protective

+ spalling

Evaporation

Scale growth with super-

imposed evaporation

Fig. 12.48 Schematic of the most common types of oxida-

tion/corrosion kinetics at high temperature

with ρ

MeX

being the density of the compound MeX and

MeX

being the molar mass of the same compound. M

and z are the molar mass and a stochiometric factor, re-

spectively, of the reacting species of the environment.

For several oxides the values of f

are given in ta-

bles [12.100]. This estimation, however, requires that

the surface scales are single phase and practically fully

dense and do not show any signiﬁcant degree of poros-

ity. Furthermore internal corrosion is neglected by this

conversion.

In the following, Fig. 12.48 shall be discussed in

some more detail as this schematic shows all the po-

tential results which may come out of the mass change

measurements. The following situations can be encoun-

tered.

Nonprotective Oxidation Behavior. In this case oxi-

dation/corrosion occurs at a very high rate, i. e. rapid

conversion of metal into the corrosion product layer on

top of the specimen is observed. This surface layer does

not have any protective character, which can be due to

cracks or pores in the scale or due to high diffusion rates

in the scale lattice. In most cases the corrosion kinetics

are linear (n = 1) but can also be of the parabolic type

(n = 2) with a high k

value if the growth of the corro-

Part D 12.8

722 Part D Materials Performance Testing

sion product layer is diffusion controlled. This case is

undesired in material service.

Temporary Protective Oxidation Behavior. In many

practical applications at the beginning of oxidation or

life time of the component a protective situation is

observed following slow parabolic kinetics or some-

times even cubic kinetics (n = 3) which also points

towards a protective behavior of the oxide scale formed.

However, after a certain period of time it may happen

that, due to scale cracking and/or depletion processes

in the oxide/subsurface-metal system, so-called break-

away oxidation starts, i. e. the formerly protective oxide

scale is converted into a nonprotective type that ends

up in a signiﬁcant increase of the oxidation rate (accel-

eration of oxidation). The characteristic values of this

type of curve are k

for the protective region, t

for the

moment at which accelerated oxidation starts (break-

away time) and k

for the oxidation rate constant in

the nonprotective post-breakaway time range. It should

be pointed out here that many laboratory tests for tech-

nical materials ended before t

had been reached so that

protective behavior of the material was assumed. How-

ever, under industrial application oxidation times are

usually much longer than laboratory tests so that t

may

easily be exceeded. Therefore laboratory results may

sometimes be questionable, which also illustrates the

problems which may arise from short-time accelerated

testing.

Protective Oxidation Behavior. This is the ideal situa-

tion which guarantees resistance of the material against

high-temperature corrosion and which is usually ob-

served when a thin and dense protective slowly growing

oxide scale is formed. In this case oxidation kinetics can

often be described by (12.88)and(12.89) with k

hav-

ing very low values. The value k

is therefore the key

value that characterizes the oxidation resistance and is

therefore determined in the oxidation tests.

Protective Oxidation Behavior but Spalling of the

Protective Scale. In particular when temperature

changes occur during oxidation, the protective oxide

scale may crack and spall due to different values of the

coefﬁcients of thermal expansion (CTE) of the oxide

scale and the substrate. As a consequence, starting from

a certain point of time, a sudden decrease in mass oc-

curs, followed by an increase of mass due to ongoing

oxidation. During the next spalling period mass de-

creases again and increases by oxidation until a further

spalling period starts, leading to a repeated sequence of

spalling and scale growth. Unless healing of the spalled

oxide occurs, and thus a restoration of the protective

effect becomes possible, spalling usually leads to a non-

protective situation.

Oxide Scale Growth with Superimposed Evaporation

at Corrosion Products. If a solid corrosion product scale

is formed during high-temperature corrosion, accompa-

nied by the formation of a volatile corrosion species,

then after an initial mass increase a continuous mass

decrease will occur. As an example such situations can

be encountered in oxidizing/chloridizing environments

where the oxide scale is reduced or destroyed by the

formation of volatile metal chlorides [12.101]. A sim-

ilar situation may also occur with other metal halogen

compounds. Kinetics in this case can be described by

Δm/A =

[

k(T)

]

1/n





1/n

−k

−





(12.90)

where k

−l

is the negative linear rate constant.

Corrosion only by Evaporation of Volatile Corrosion

Products. If neither solid nor liquid corrosion prod-

ucts are formed but only gaseous species result from

the corrosion reaction then a continuous linear decrease

in mass will be observed. This linear decrease is char-

acterized by n = 1 and a linear negative rate constant

of k

-l

Summarizing Fig. 12.48, it is actually only the pro-

tective case that can be accepted for industrial use of the

materials. For most technical materials the exponent n

in this case usually ranges between 2 and 3 and k(T)

takes low values. The temporary protective case can be

accepted if t

is not much shorter than the expected ex-

posure time in service and the spallation case can be

accepted if the amount of spalling is relatively small.

The same is valid for scale growth with superimposed

evaporation but all other cases lead to severe problems

in practical use.

The test methods described in the following serve

for the quantitative determination of the parameters

of Fig. 12.48 and also for the qualitative characteriza-

tion of oxidation/high-temperature corrosion behavior

according to these systematic cases. For further illustra-

tion in Fig. 12.49 examples of values of the parabolic

rate constant are given for protective oxides scales and

a temperature range 800–1400

◦

C [12.102]. This ﬁg-

ure shows that alumina surface scales are superior to

chromia surface scales with regard to the protective ef-

fect. Figure 12.49 furthermore can serve for orientation

Part D 12.8

Corrosion 12.8 High-Temperature Corrosion 723

Fig. 12.49 Oxidation rate constants k

for the alloy

groups Al

-scale formers and Cr

-scale formers (af-

ter [12.102]) 

whether the measured values determined in the experi-

ments characterize protective or nonprotective behavior

of the materials.

Thermogravimetric Testing. Mostly in scientiﬁc labo-

ratories the thermogravimetric method is a widely used

type of test to investigate the oxidation kinetics and

other types of high-temperature corrosion continuously

in situ. The principles of the equipment are given in

Fig. 12.50. The weight of the specimen is continuously

recorded by a laboratory balance, which usually sits

on top of a laboratory furnace. To hang the specimen

platinum wires or silica strings, which are supposed

to be inert in most environments, are normally used.

(It should however be mentioned here that under cer-

tain conditions even platinum and silica may evaporate

and, thus, lead to a decrease in weight of the string

which is also measured by the balance.) The speci-

men and the weighing system are usually encapsulated

in a retort to allow the use of deﬁned gas composi-

tions. The interior of the laboratory balance should be

shielded against penetration of the reactive gas in the

furnace by counterﬂow of inert gas (e.g. Ar). The gas

entrance should be opposite to the gas exit. In most

cases the retort is manufactured from quartz. Espe-

cially at very high temperatures, alumina retorts are

also used. For accelerated testing the measurements

are sometimes performed under thermal cycling con-

ditions (see later). The same equipment can be used

with a moveable furnace, the movement of which is

controlled by an electronic device that allows deﬁned

temperature cycles. The temperature at the specimen

is monitored by a thermocouple, which is placed close

to the specimen, although no friction between the ther-

mocouple and the specimen should occur. Furthermore

any friction between the quartz string or the platinum

wire leading to the balance and the retort or tube walls

should be avoided. Further recommendations for the

thermogravimetric tests are given below in the section

on continuous and discontinuous isothermal exposure

testing.

As indicated in Fig. 12.50 it is also possible to

combine this type of measurement, especially in the

case of thermal cycling testing, with acoustic emission

measurements. In this case the platinum wire that is

connected to the specimen can be used simultaneously

as an acoustic waveguide. Acoustic emission signals

Temperature (°C)

Parabolic

rate constant

/cm

–9

–10

–11

–12

–13

–14

1300 800

/T (K

–1

)

“Cr

”-

alloys

“Al

”-

alloys

796

900100011001200

Temperature

controlled

microbalance

Counter

weight

Ar-counter

stream

AE-sensor

Polymer seal

Thermocouple

Furnace

(moveable)

Quartz tube

Acoustic decoupler

(PTFE)

Gas exit

Pt-wire and

waveguide

Testing gas

Specimen

Gas inlet

Fig. 12.50 Schematic of a thermobalance for continuous in situ

mass measurements

Part D 12.8

724 Part D Materials Performance Testing

can reveal the extent to which cracking or spalling of

the oxide scale occurs either during isothermal or during

cyclic oxidation. The results of the thermogravimet-

ric measurements are the curves shown schematically

in Fig. 12.48. However, a major disadvantage of this

method is that only one sample can be tested at a time so

that generating a large number of comparative data for

different materials is a lengthy process. This is the main

reason why industrial laboratories prefer to test multiple

samples simultaneously. This issue will be addressed in

the next section.

Continuous and Discontinuous Isothermal

Exposure Testing

To accommodate a larger number of specimens in one

furnace a special arrangement of equipment is neces-

sary, as shown schematically in Fig. 12.51. This equip-

ment is again composed of a temperature-regulating

device for heating the test piece(s) uniformly at a con-

stant temperature and, ideally, a testing portion capable

of separating the test piece(s) from the outside air (i. e.,

a closed system). Since for many materials humidity in

the test environment may have a signiﬁcant effect on

oxidation behavior, a humidifying regulator should be

used to continuously supply a gas kept at a constant

humidity, which should be monitored with a hygrom-

eter. As for all closed systems, including those for

thermogravimetric testing, the gas supply should be

Gas supply

Va lve s

Heated device

containing catalyst

for non equilibrium

gas mixtures

(Hygrometer)

Humidifying

regulator:

Hot water or

electronic type

Heater

Test piece chamber Test piece supporter

Gas exhaust

Direction of

furnace movement

Measuring

instrument

for thermal cycling

Thermocouple

Thermocouples

Temperature

regulating

device

Heating device

Heating zone with

ribbon heater

Power

control

device

Gas flow

meter

Fig. 12.51 Basic design of a horizontal exposure furnace for discontinuous oxidation/corrosion testing

controlled by a gas-ﬂow meter. It is important that the

test-piece chamber is not composed of a material that

reacts with the test environment during the test to a de-

gree that changes the composition of the atmosphere.

If closed systems with a test-piece chamber cannot be

used, then tests may also be performed in an open sys-

tem with laboratory air. In this case it is recommended

that the humidity of the air should be recorded and

that the laboratory should be kept free from tempera-

ture changes and the inﬂuence of weather conditions as

far as possible.

The furnace has to be characterized prior to test-

ing to determine the length of the isothermal zone

inside the furnace and the set point of the apparatus.

A common method is by the use of an independent

movable thermocouple. The draft of the present ISO

work item recommends that the temperature-regulating

device should be capable of guaranteeing that the tem-

perature of the test piece is kept within a permissible

range, given in Table 12.5. The material of the ther-

mocouples has to withstand this test temperature. The

temperature has to be measured by a suitable device

according to ASTM E 633-00 [12.103]. Thermocou-

ples of type S (Pt-10% Rh/Pt), type R (Pt-Pt/13% Rh)

up to 1000

◦

C or type B (Pt/30% Rh-Pt/6% Rh) up to

1700

◦

C are preferred. A thermocouple should be po-

sitioned close to the test-piece surface and must be

calibrated in accordance with standards ASTM E 220-

Part D 12.8

Corrosion 12.8 High-Temperature Corrosion 725

Table 12.5 Permissible tolerance of temperature of test piece [12.97]

T ≤ 573 K 573 K < T ≤873 K 873 K < T ≤1073 K 1073 K < T ≤ 1273 K T > 1273 K

±2K ±3K ±4K ±5K ±7K

02, ASTM E 230-02 and ASTM E 1350-97 [12.104–

106]. If, however, the environment does not allow

the use of such thermocouples in this way the test-

piece temperature has to be deduced from the furnace

calibration using dummy test pieces and appropriate

thermometry in an inert environment. The thermo-

couple should be capable of conﬁrming that the test

temperature is within the range given in Table 12.5.It

should be located at a deﬁned, ﬁxed position as close to

the test pieces as possible.

When a humidifying regulator is used it should be

capable of adjusting to the desired humidity and the

space between the humidifying regulator and the test-

piece chamber should be kept above the dew point to

avoid condensation. The water-vapor content should be

measured, which can be achieved by, e.g., the use of

a hygrometer before the test-piece chamber or mea-

suring the amount of water after condensation of the

exhaust gases. The gas ﬂow has to be high enough to

ensure that no depletion of the reacting species will oc-

cur. At the same time the gas ﬂow must be slow enough

to allow the gas mixture to preheat and in some appli-

cations to reach reaction equilibrium. The TESTCORR

document recommends 1–10 mm/s.

It is recommended that the test pieces have a mini-

mum surface area of 500 mm

[12.92]. Their shape can

be a rectangular plate, a disc or a cylinder. Final surface

ﬁnishing of the test pieces should be 1200 grit according

to the Federation of European Producers of Abra-

sives (FEPA) standard 43-1984 R 1993 [12.107]and

ISO 6344 [12.108]. Deformation by marking, stamping

or notching of the surface should be avoided. Identiﬁ-

cation of the test pieces should be solely on the basis of

recording their relative position within the test chamber.

In particular for hanging specimens, e.g. for the ther-

mogravimetric test, holes for the test-piece support are

permissible. The test pieces must be dried after degreas-

ing in an ultrasonic bath using isopropanol or ethanol.

Before exposure the mass of the test pieces m

)has

to be determined by three independent measurements

with the difference between the measurements not ex-

ceeding 0.05 mg.

Several test pieces exposed for different times are

necessary to deﬁne the oxidation kinetics of the ma-

terial. Therefore it is recommended that duplicate test

pieces are used with (at least four) exposure times in-

creasing progressively (e.g., 10 h, 30 h, 100 h, 300 h,

...). The material used for the test-piece supports

should not react at the test temperature neither with the

environment nor with the specimen. Examples of rec-

ommended test-piece supports are shown in Fig. 12.52.

Where the possibility of depletion of active species in

the test atmosphere is a concern, the exchange of the

test atmosphere can be improved by the use of holes

or slots in the bottom part of the side walls of the

test-piece support. To remove volatile compounds it is

furthermore recommended that new test-piece supports

should be baked for at least 4 h at 1000

◦

Cinair.In

addition to the mass of the test pieces, that of the test-

piece supports m

) should be determined prior to

exposure.

Thermal Cycling Oxidation Testing

In the majority of industrial applications temperatures

are never constant. Therefore temperature cycling oxi-

dation/corrosion tests are most often used in industry.

The test equipment is rather similar to that for isother-

mal testing, with the major difference that either the

specimen or the furnace can be moved so that tem-

perature cycles are possible in the test. In Figs. 12.50

and 12.51 the test equipment shown for isothermal test-

ing already allowed for the possibility of moving the

furnace, so that in the same equipment cyclic oxida-

tion testing would also be possible. The advantage of

the equipment shown in Figs. 12.50 and 12.51 is that

these tests can be performed in a deﬁned environment.

In some of the more traditional thermal cycling oxida-

tion test rigs the specimens are moved by some device

out of and into the furnace, which usually requires

that these tests are performed in an open system. As

mentioned above, closed systems with deﬁned environ-

ments are preferred over open systems with laboratory

air.

For completeness it should be mentioned that, as

well as tests with small laboratory specimens, compo-

nent tests, e.g. with complete heat-exchanger tubes, can

even be performed under thermal cycling conditions, as

shown by the schematic in Fig. 12.53. In this case the

furnace consists of two half shells that can be moved

apart on a rail system so that the test-piece chamber is

at ambient temperature when the furnace is open. The

test piece, which in this example is a heat-exchanger

Part D 12.8

726 Part D Materials Performance Testing

Side view

Test piece

High purity alumina tube for

supporting test piece

Test piece support and basic layout of test piece arrangement – tube design

Test piece support and basic layout of test piece arrangement – U-shaped design

Test piece support and basic layout of test piece arrangement – rod supported design

Holes

Holes Alumina

rod

Cross sectional

view

Side view

Cross sectional

view

Cross sectional

view

Top view

Fig. 12.52 (a) Test-piece support and

basic layout of test-piece arrangement

– tube design; (b) Test-piece sup-

port and basic layout of test-piece

arrangement – U-shaped design;

layout of test-piece arrangement –

rod-supported design

Testing gas Testing gasQuartz tube

Inner heater

Current lead

Ceramic tube

Plastic bellows

Mobile half furnace

Controlling

thermocouple

Test tube

Measuring

thermocouple

AE sensor

Fig. 12.53 Equipment for thermal cyclic testing of heat-exchanger tubes (after [12.109])

Part D 12.8

Corrosion 12.8 High-Temperature Corrosion 727

tube, is placed inside a quartz tube and sealed at the

ends with ﬂanges. This design allows the possibility of

having deﬁned environments on the outer surface of the

tube as well as on the inner surface. These two envi-

ronments can even be different from each other, e.g.

combustion gas on the outside and steam on the in-

side of the tube. Component tests as in this example

are always desirable when geometrical effects of these

components on the oxidation behavior are to be inves-

tigated. Figure 12.53 again includes the possibility of

using the acoustic emission technique for the detection

of failure of the protective oxide scales due to thermal

cycling and spalling.

Generally all the recommendations given in the con-

tinuous and discontinuous isothermal exposure testing

section are also applicable to thermal cycling testing.

The difference compared to discontinuous isothermal

testing is that in thermal cycling testing the temper-

ature cycles always have the same shapes. Based on

a detailed scientiﬁc model calculation the parameters

characterizing such a thermal cycle have been deter-

mined as follows [12.110, 111]. The heating period of

the cycle starts when the test piece enters the furnace

and ends with the beginning of the hot dwell time. The

latter is deﬁned as the time when the actual tempera-

ture exceeds 97% of the desired hot dwell temperature

measured in kelvin. The hot dwell time ends and the

cooling time starts when the heating of the test piece is

stopped, e.g. by the removal of the test piece from the

furnace. The end of the cooling time is deﬁned as when

the actual test-piece temperature falls below 50

◦

C. The

cold dwell time starts after the test pieces have cooled

below 50

◦

C and ends when the test pieces are heated

again. The characteristic parameters of a thermal cycle

are summarized in Fig. 12.54.

Depending on the type of industrial application,

three general types of thermal cycles are regarded as

typical. The ﬁrst type is long-dwell-time testing, which

aims to simulate conditions in large-scale industrial

facilities encountered in applications such as power-

generation plants, waste-incineration plants or chemical

industry. In these applications the metallic components

are designed for extremely long-term operation, e.g.

typically for up to 100 000 h. Thermal cycling of the

materials occurs due to plant shutdowns, e.g. for reg-

ular maintenance, or due to unplanned shutdowns as

a result of offset conditions. Therefore, the time inter-

vals between various thermal cycles are relatively long

and the number of cycles is (related to the long op-

eration time of the components) relatively small, e.g.

typically around 50 cycles. A recommended test cycle

T (°C)

dwell

Hot dwell

(T >0.97 T

dwell

)

Cold dwell

(T <50 °C)

Heating time Cooling time t (min)

900

800

700

600

500

400

300

200

100

1000

0908070605040302010

Fig. 12.54 Schematic of a temperature cycle in thermal cycling ox-

idation testing

for this application consists of an overall time of 24 h

with a 20 h hot dwell time and 4 h period, which in-

cludes the cooling time, the cold dwell time and the

heating time.

The second type is thermal cycling with short

dwell times, which is typically experienced in ap-

plications such as jet engines, automotive parts, and

heat-treatment facilities. The intervals between start and

shutdown of the facilities are generally much shorter

than in applications with long dwell times. Also the life

times and/or the times until complete overhaul/repair

(typically 3000–30 000 h) are shorter and, depending

on the speciﬁc practical application, the number of

cycles is much higher than in the cases above. The rec-

ommended testing cycle in this case consists of one hour

hot dwell time and 15 min cold dwell time.

The third type of testing has ultrashort dwell times

and mainly addresses applications of high-temperature

alloys as heating elements in the form of wires or foils.

Another typical application in which such ultrashort

cycles prevail is catalyst foil carriers, e.g. in cars. In

such applications the number of cycles is related to the

overall design life (typically several hundred to a few

thousand hours) and can be extremely high, and the

time intervals between heating and cooling can be as

low as minutes or even seconds. Such conditions are

commonly also encountered in a number of other in-

dustrial applications such as burners and hot gas ﬁlters,

as well as in a large variety of domestic applications

where metallic heating elements are used e.g. in cook-

ing plates, toasters, boilers, dryers, and fryers. The

Part D 12.8