fry.standardisation in high temperature corrosion testing.ppt

Standardisation in High Temperature Corrosion TestingCorrosion Testing

Review of Task 4 of the UK-US Collaboration

23rd Annual Conference on Fossil Energy MaterialsPittsburgh Airport Marriott

May 12-14 2009May 12 14, 2009

Tony Fry National Physical Laboratory, Teddington, UK

Pittsburgh – 14th May 2009 UK-US Collaboration on Fossil Energy R&D - Advanced Materials

Overview

• Introduction to Task 4• Data collection and storage• Standardisation• Standardisation• Inter-comparison


What was task 4d ith?concerned with?

• To identify critical differences between standards for measurement of high temperature materials properties

1902• To identify where further standardisation for measurement of high temperature materials properties is required

1902NPL Opens

• To develop a common format for data exchange

• To investigate the use of commercial To investigate the use of commercial database software for collecting and maintaining materials properties data and micrographs 1906

Metallurgical


Metallurgical laboratory

The collaboration willgenerate a lot of data – how do

we manage and collect it?

• What data is being recorded?

• Can we agree on what should be • Can we agree on what should be reported?

What is the best method of capturing • What is the best method of capturing this data?

C• Can everyone use the selected method?


Data collation in a database, which includes metadatawhich includes metadata


How can we be sure thed t i bl ?data is comparable?

International Organisation for Standardisation

European Committee for Standardisation(CEN)

British Standrds Institure American National Standards Institure Deutsches Institut fur Normung

International Organisation for Standardisation(ISO)

British Standrds Institure(BSI)

American Society for Testing of Materials(ASTM)

American National Standards Institure(ANSI)

Deutsches Institut fur Normung(DIN)


How do the UK and USt d d ?standards compare?

Mechanical Tests

• Hardness

Corrosion Testing

• Laboratory exposures in steam and i d (i l di d it )• Tensile

• Creep• Low Cycle Fatigue (LCF)

High Cycle Fatigue (HCF)

mixed gases (including deposits)• Post-exposure evaluation of

environmental attack• Steam Loop exposures• High Cycle Fatigue (HCF)

Physical Testing

Steam Loop exposures• Coating Thickness• Burner rig testing• Thermal Cycling/Cyclic Oxidation

• Thermal Diffusivity• Dilatometry• Surface Area Measurement


Findings of the review –h i l t timechanical testing

Test Addressed by Standards Further work needed

Knoop Brinell Hardness ISO - TC164 Knoop, Brinell, Vickers, Rockwell No

Tensile ISO - TC164Ambient Temp.

High TempNo

High Temp.Creep ISO Uniaxial No

Standards exist for LCF ASTM, CEN, BSI strain controlled,

TMF.Not at this point

HCF Yes Ambient Temp. Yes, HT HCF


p ,

Findings of the review –h i l t tiphysical testing

Test Addressed by Standards Further work needed

Thermal Diffusivity ASTM Laser flash NoThermal Diffusivity ASTM Laser flash NoDilatometry Yes

Surface Area M t ISO Gas adsorption or

bilit NoMeasurement ISO permeability No


Findings of the review –i t ticorrosion testing

Corrosion Testing

• Laboratory exposures in steam and • Laboratory exposures in steam and mixed gases (including deposits)

• Post-exposure evaluation of environmental attack

At the time of the review there were no international US or Europeanenvironmental attack

• Steam Loop exposures• Coating Thickness

international, US or European standards existing for high temperature

corrosion of metallic materials.• Burner rig testing• Thermal Cycling/Cyclic Oxidation


But…

this area is being actively pursued by ISO TC156 WG13 and standards for:

• Test Method for Isothermal Exposure Testing under High Temperature • Test Method for Isothermal Exposure Testing under High Temperature Corrosion Conditions

• Method for Metallographic Examination of Samples after Exposure to High Temperature Corrosive EnvironmentsTemperature Corrosive Environments

• Thermal Cycling Exposure Testing Under High Temperature Corrosion Conditions

are in preparation.


No standards, so what…I t iInter-comparison

Seam Oxidation

• 3 materials (T92, 347HFG, IN740)

Boiler Corrosion

• 2 materials (T22 and P92)( , , )• Same material stock• Same temperatures

( )• Same material stock• Gas composition set

– 0.3% SO2 , 6.0% O2 , 14.6% CO2 , 74.2% H2

• Lab could prepare samples in their standard manner

• Tests conducted using their own

• Ash composition set– Na2SO4/K2SO4/Fe2O3 (1.5/1.5/1 on a molar basis)

• Temperatures setpreferred method

• Data analysed in their own preferred technique

• Lab left to prepare samples and expose using their preferred method


600°C

P92, 1000 h exposure

ALSD, H2O, 1 bar

ANSA, H20, I bar

CESI, H2O, 1 bar

600°C650°C

FZJ, Ar50%H2O, 1 bar

FZJ, H2O, 240 bar

INTA, Ar50%H2O, 1 bar

static steam

INTA, H2O, 1 bar

KEMA, H2O, 60 bar

NPL, H2O, 1 bar

0 100 200 300

SPGG, H2O, 1 bar

Willemshafen, 250 bar

static steam


0 100 200 300

oxide scale thickness, μm

Steam Oxidationi t l texperimental setup


Steam oxidationi t lexperimental

• Sample geometry– 10 x 10 x 3mm– 20 x 10 x 2mm

• Exposure procedure– Duplicate samples exposed for a set

time duration (no cycling)20 x 10 x 2mm– Semicircular section

• Surface Preparation

– Sample all exposed at the same time, thermal cycles introduced to remove samples

• Surface Preparation– Samples from bulk– Samples retained original surface

Surfaces prepared 600 grit SiC

– Ambient pressure & 17 bar

– Surfaces prepared 600-grit SiC


Steam oxidationlt T92 t 600 ºCresultsT92 at 600 ºC

25

-2

20

ge, m

gcm

-

10

15

ss C

hang

5

peci

fic M

as

NPL

Cranfield

00 1000 2000 3000 4000 5000 6000

Sp ORNL


Time, h

Steam oxidationlt T92 t 600 ºCresults T92 at 600 ºC

30

-2

20

25

e, m

gcm

-

15

20

ss C

hang

5

10

peci

fic M

as

NPL

Cranfield

00 1000 2000 3000 4000 5000 6000

Sp ORNL


Time, h

Steam oxidationlt T92 t 650 ºCresults T92 at 650 ºC

20

-2

141618

ge, m

gcm

-

81012

ass

Cha

ng

246

peci

fic M

a

NPL

Cranfield02

0 500 1000 1500 2000 2500 3000

Ti h

Sp

C a e d


Time, h

Steam oxidationT92 t t d dT92 temperature dependency

12

14

16

mgc

m -2

4

6

8

10

12

fic M

ass

Cha

nge,

m

NPL 600°C

0

2

4

0 500 1000 1500 2000 2500 3000Time, h

Spec

if 600 CNPL 650°C

101214161820

hang

e, m

gcm

-2

02468

10

Spe

cific

Mas

s C

Cranfield 600°CCranfield 650°C


00 500 1000 1500 2000 2500 3000

Time, h

Steam oxidationlt 347HFG t 650 ºCresults 347HFG at 650 ºC

6

2

3

45

mg,

cm

-2

01

2

s C

hang

e,

3

-2-1

0

ecifi

c M

ass

347HFG ORNL

347HFG NPL

34 HFG C fi ld

-4-3

0 1000 2000 3000 4000 5000

Spe 347HFG Cranfield


Time, h

Steam oxidationlt 347HFG t 700 ºCresults 347HFG at 700 ºC

5

3

4

mg,

cm

-2

1

2

s C

hang

e, m

-1

0

ecifi

c M

ass

347HFG NPL

-2

1

0 1000 2000 3000 4000 5000

Spe

347HFG Cranfield


Time, h

Steam oxidationlt IN740 t 750 ºCresults IN740 at 750 ºC

1.40

m -

2 IN740 ORNL

1.00

1.20

ge, m

g, c

m IN740 NPL

IN740 Cranfield

0.60

0.80

ass

Cha

ng

0.20

0.40

Spe

cific

Ma

0.000 1000 2000 3000 4000 5000

Time h

S


Time, h

What could be causingth diff ?the differences?

• Specimen Geometry

• Thermal Cyclingy g

• Orientation of grains

• Spallation


Seam Oxidation

• 3 materials (T92, 347HFG, IN740)

Boiler Corrosion

• 2 materials (T22 and P92)( , , )• Same material stock• Same temperatures

( )• Same material stock• Gas composition set

– 0.3% SO2 , 6.0% O2 , 14.6% CO2 , 74.2% H2

• Lab could prepare samples in their standard manner

• Tests conducted using their own

• Ash composition set– Na2SO4/K2SO4/Fe2O3 (1.5/1.5/1 on a molar basis)

• Temperatures setpreferred method

• Data analysed in their own preferred technique

• Lab left to prepare samples and expose using their preferred method


What about boiler corrosion, that’s OK isn’t?that s OK…isn t?

T22 at 425 ºC

50

60

30

40

oss,

mic

rons

10

20

Met

al L

o

NPL

Doosan

0

10

0 200 400 600 800 1000 1200

Doosan

NETL


Time, h

P92 at 675 ºC

1800

1200

1400

1600

ons

800

1000

1200

Loss

, mic

ro

200

400

600

Met

al

NPL

Doosan

0

200

0 100 200 300 400 500 600 700 800 900 1000 1100

Time h

NETL


Time, h

Why the differences?

• Specimen manufacture• Measurement accuracy• Fundamental differences in the apparatus• ?


What does all this mean?

No one is wrong the results are just differentNo one is wrong, the results are just different

• Results are self consistent with a laboratory

• Measurements are precise but there is scatter due to material effects (i.e. spalling)

• Ideally we would like high precision and good repeatability


Where do we go f h ?from here?

• Standard test methods for corrosion testing which address specimen manufacture and preparation as well as the actual test procedure and analysis.procedure and analysis.

• Need to addressS i t– Specimen geometry

– Surface preparation– Testing procedures

Measurement accuracy & uncertainty– Measurement accuracy & uncertainty

Phase 2 will be addressing some of these issues


Thank you for your attention

Any questions?y q


fry.standardisation in high temperature corrosion testing.ppt

Documents