New Italian standard for the creep assessment of martensitic steels

Abstract

For the evaluation of the fraction of life consumed due to creep damage in martensitic materials, unlike what happens for ferritic materials for which there is a consolidated method, there are currently several usable methods. Therefore, laboratory tests and in-depth microstructural analyses were carried out on Grade 91 and 92 steels for the realization of a reference atlas on microstructural modification and precipitates state evolution during isothermal ageing and thermal creep.

Keywords

Martensitic steel metallographic replica XRD SFM neutron diffraction

1. Introduction

In the last 10 to 15 years a lot of work has been carried out to investigate the microstructure evolution of steel Grades 91 and 92 to understand the degradation mechanisms. In the range of service temperature up to 600 °C, the most dangerous microstructural modification seems to be the formation of Laves phase associated with the nucleation of voids and the drop in ductility [14]. As a matter of fact, grain boundary coarse Laves phase particles probably enhance growth and coalescence of multiple cavities at grain boundaries.

Creep behaviour and creep damage by the nucleation, growth and coalescence of multiple cavities, are currently under deep investigation on the steel Grades 91 and 92. For these alloys, long term creep behaviour is governed by several microstructural modifications, such as M23C6 coarsening, nucleation and coarsening of the intermetallic Laves phase (Fe2Mo and Fe2W respectively for Grade 91 and 92) and further recovery of martensite that make difficult to define the correlations between microstructure evolution and creep strength.

In Italy the use of martensitic steels in power plants started only a few years ago. Currently the Italian Thermotechnical Committee is drafting a new standard for the life assessment of creep operated pressure equipment, including modern steam boilers.

For the evaluation of the spent life ratio several methods are available, even if each of them is not exhaustive. For martensitic materials, even more than ferritic materials, for a correct assessment it is important to know and collect all possible information on the product:

–
chemical composition of the steel;
–
technological process of manufacturing;
–
heat treatments;
–
any welding qualifications;
–
anything else useful for defining the initial delivering conditions, before it begins its operational life;
–
a detailed description of the operating conditions.

It is equally important to collect information, as detailed as possible, relating to the operating conditions.
2. Materials and methods

The design concepts for steel Grades 91 and 92 are to achieve stable microstructure, high strength by precipitation strengthening and solution hardening through elements as Mo or Mo/W, high corrosion resistance with high Cr content in order to ensure long term service.

Steel Grades 91 and 92 gain their creep strength mainly by precipitation hardening. A fine precipitate distribution is achieved through a two-step final heat treatment consisting of normalizing and tempering.

Normalizing followed by air cooling results in martensitic transformation, which produces a high number of internal high and low angle boundaries as well as a high density of dislocations, all of which can act as nucleation sites for precipitate particles. During tempering subgrains (polygonised ferrite), with high dislocation density, nucleate within martensite laths, resulting in martensite recovery (tempered martensite), and M23C6 (Cr, Fe, Mo or W) carbides and fine MX (V, Nb) M carbo-nitrides precipitate on boundaries as well as on dislocations.

The main methods for component analysis are listed below:

–
Extraction replicas;
–
Thin foils;
–
Metal replica;
–
Small scale mechanical tests;
–
Scanning force microscope;
–
X-ray diffraction;
–
Neutron diffraction;
–
Electro-magnetic methods;
–
Hardness tests.

It should be noted that the methods described must be considered in combination with the NDEs and the hardness tests. Furthermore, it is recommended to combine the methods among them.
2.1. Extraction replicas

Extraction replicas, studied using TEM (Transmission Electron Microscopy), is a metallographic method aimed at the mechanical extraction of particles and carbides.

The support, on which the particles are extracted, is normally made by a resin. With this method it is possible to underline an evolution in terms of size, morphology and chemical composition of precipitates and phases (Vanadium and Niobium carbonitrides (MX), metal carbides of Iron, Chromium and Molybdenum (M23C6), modified Z phase, Laves phases), (Figs 1 and 2). Therefore, this method gives, for each category of precipitates, the chemical composition and size, without removing material and with a higher resolution of TEM compared to SEM.

Fig. 1.

Extraction replicas after 10,000 h at 600 °C.

Fig. 2.

Extraction replicas after 25,000 h at 600 °C.

Typical problems are related to the development of the extractive replica process, the potential risk of pollution during sampling and the TEM management (from acetate support to support for TEM use). Furthermore, in addition to limited availability of TEM equipment, the analysis is usually performed on a few particles (<200) and it requires a slow and expensive analysis for each position [9].

2.2. Thin foils

This method requires the withdrawal of sample by mechanical methods (e.g. scoop sampler). The withdrawal has to be carried out with great care to avoid deformation and temperature increases. The removal of material from the component allows to operate directly on the metal sample, which is analysed through TEM. TEM has higher resolution compared to SEM, but as already specified for the extraction replica method, it requires a slow and expensive analysis for each position and the analysis is usually performed on a few particles (<200), so on a very localized survey area. The thin foil method provides, for each category of precipitates, the chemical composition and size, allows the study of the distribution of particles in situ and the analysis of the dimensions of the sub-grain, (Figs 3, 4). A disadvantage of this method is the interference between the metal matrix and the particles [9].

Fig. 3.

Thin foils after 10,000 h at 600 °C.

Fig. 4.

Thin foils after 25,000 h at 600 °C.

2.3. Metal replica

It is an extractive replica on a metal support, consisting of a metal with low hardness, high ductility, simple chemical composition and different from that of the component to be examined, without removing material, (Figs 5, 6).

Fig. 5.

Morphological reproduction of the microstructure with metal replica.

Fig. 6.

Portable apparatus for metal replica preparation.

With this method an evaluation of the size and number of particles takes place, by means of image analysis through SEM/SEM EDX, so with a larger survey area than the TEM analysis and the statistic is based on more particles than the TEM (>800). In addiction there is an increased availability of SEM equipment and the analysis times and costs are lower than the TEM but has a lower resolution.

Compared to extractive replication, with this method one obtains morphological reproduction of the microstructure, with the possibility of analysing the subgrain and to evaluate the microstructure. Typical problems are related to the development of the metalloreplic process and to the potential risk of pollution during sampling [9,18].

2.4. Small scale mechanical tests (small punch, impression creep, miniaturized samples)

Small scale mechanical tests are test methodologies using miniaturized samples to obtain the mechanical properties of the component, (Figs 7, 8).

Fig. 7.

Graphs of small punch tests.

Fig. 8.

Graphs of small punch tests.

In some cases, it is possible to carry out the examination without compromising the future operation of the components, after checking the stability of the residual thickness.

The instrumentation used for these tests is a specific tool/equipment according to the type of test, (Figs 9, 10, 11).

Fig. 9.

Small punch machinery.

Fig. 10.

INAIL small punch machinery.

Fig. 11.

INAIL – New SPCT machinery.

This kind of test provides information on mechanical behaviour, but it has the disadvantage to be a semi-destructive test, with difficulty in sampling and the necessary correlations between micro and macro tests [8,10,12].

2.5. Scanning force microscope

Scanning force microscope is a microscope that allows to produce a three-dimensional profile of the analysed surface with related microscopic images (Figs 12, 13, 14). It allows a resolution in the order of tens of nm; it detects very small early-stage creep cavities and it is not necessary to operate under vacuum (which is required using the SEM) so an application is possible in situ using a portable SFM.

Fig. 12.

SFM.

Fig. 13.

SFM test - Graph 2d.

Fig. 14.

SFM test - Graph 3d.

This method, therefore, allows the determination of the distribution and size of the cavities and possibly the evolution in terms of size, morphology and chemical composition of the precipitates and phases.

However, this microscope has an extreme sensitivity to vibrations (the feasibility of use in the field must be verified from time to time) and a slow scan speed. The image size is smaller than other microscopic techniques and the image quality strongly depends on the type of probe used. In addition it is necessary to carry out a post-processing treatment of the image, after its acquisition, because there is the possibility of having artifacts [15,19–21].

2.6. X-ray diffraction

The method allows the study of phase transitions involving structural variations, using as instrumentation the diffractometer. It is possible to operate both on massive samples and on powders. It possible to see the presence and quantity of non-metallic phases and to measure the deformation in fraction of nanometers. This method is applicable in situ (applies only to the portable instrument).

The technique allows the recognition and quantitative study of the sample phases; each phase has an arrangement of the atoms, a characteristic which is reflected in a typical diffraction profile (pattern) (Fig. 15). It detects local plastic deformation and texture.

Fig. 15.

XRD patterns of the powders of the sample t0 and aged at 600 °C.

However, the measurement takes from a few to a few tens of minutes per point, during which neither component nor equipment must move or vibrate. Preparation for measurement depends on the necessary precision.

Analysis of diffraction profiles (patterns) is very complex, in fact it requires high specific training. In case of massive samples, a routine software offers only for a few seconds the concentration of phases present at 5–10%; for higher sensitivity, more time and specific applications are required and it is necessary to operate in several steps if you want to study the second phases [4–7].

2.7. Neutron diffraction

It is a non-destructive testing method used to determine the residual stresses and to establish the value of the applied stress by measuring the elastic deformations. The available neutron techniques are ND (Neutron Diffraction) and SANS (Small Angle Neutron Scattering). They also allow the identification and quantification of the crystallographic phases of the compounds.

This method indicates the presence and quantity of non-metallic phases and it measures the deformation in fraction of nanometers. It allows the recognition and quantitative study of the phases of the sample. Each phase has an arrangement of the atoms, a characteristic which is reflected in a typical diffraction profile (pattern). It detects local plastic deformation and texture.

Measurements can only be carried out using neutron sources (nuclear reactors, research centers, laboratories with neutron sources); it is not applicable in situ and it does not detect deformations present in the sub-superficial part of the product that are less than a few microns. Analysis of diffraction profiles (patterns) requires high specific training and there is a poor availability of tools (neutron sources) [25] (Fig. 16).

Fig. 16.

EN ISO 21432.

Fig. 17.

Electro-magnetic method.

2.8. Electro-magnetic methods

These methods, through electromagnetic sensors, provide cyclic magnetization by detecting the hysteresis flow with respect to the intensity (Fig. 17). Momentary permeability and Barkhausen magnetic noise are obtained, both correlated with the microstructure and it happens Barkhausen magnetic noise increase and magnetic permeability modification.

These methods are applicable as usual NDT, simple to use, but it would require adequate NDT qualification; in addiction it can be used on large dimensions; it allows to identify P91 and it measures the effects in the laboratory during a complete creep test. The relevant changes in the material can be detected only after 80% of the spent life [2,26,27].

There is a low availability of laboratories with available know-how.

2.9. Hardness tests

Hardness tests (in Vickers or Brinell), carried out with the Portable hardness tester, are quick and simple using in rapidly hardness decaying situations, they give a good warning, also having reference curves available. It is possible to note a hardness decay if creep damage increases. However, these tests have a great dispersion of the results and moreover they depend on several factors: on the hardness of origin (zero point), on the operator and on the instrument, on the preparation of the measuring point and on the position of the component [11,16].

3. Martensitic steels microstructure atlas

Steel Grades 91, 92 microstructure evolution atlas aims to realize a reference atlas on microstructural modification and precipitate state evolution during isothermal ageing and thermal creep of steel Grades 91 and 92, both in base material (BM) and heat affected zone (HAZ) of similar welded joints, providing long term aging in furnace, creep tests and advanced microstructural analyses.

The atlas aims to be a reference of the microstructural modification of steel Grades 91 and 92 for monitoring martensite recovery, hardness evolution, growth of creep cavities and evolution of precipitates in relation to size, morphology of the main phases (MX, M23C6, and Laves) during isothermal ageing and thermal creep.

For the realization of the Steel Grades 91, 92 microstructure evolution atlas, a project with seven specific objectives was carried out:

Study of the evolution of the microstructure and analysis of the precipitates of Grade 91 steel, artificially aged in the furnace, from 40,000 to 65,000 h at 550 °C, from 50,000 to 75,000 at 600 and 650 °C;

Study of the effect of mechanical stress on the evolution of the microstructure and of the state of Grade 91 steel precipitation at 550 °C, 600 °C and 650 °C, with a creep test for each temperature and value of the load corresponding to the failure for creep at 50,000 h (according to ECCC datasheet);

Study of the microstructural evolution of Grade 91 in the Thermally Altered Zone (HAZ) of welded joints subjected to creep failure tests at temperatures of 550 °C and 600 °C (with failure times of the order of 10,000, 20,000 and 50,000 h);

Creation of an isotherm for Grade 92 steel at a temperature between 600 and 650 °C, with 5 load conditions corresponding to creep failure at 20,000, 40,000, 60,000, 80,000 and 100,000 h (according to ECCC datasheet);

Study of the evolution of the microstructure and of the precipitation state of Grade 92 steel at temperatures of 600 and 650 °C with load and in absence of load, for times of 5,000, 10,000 and 20,000 h; as regards the tests with load, interrupted creep tests at 5,000, 10,000 and 20,000 h, with load corresponding to creep failure at 50,000 h (according to ECCC datasheet);

Study of the microstructural evolution in HAZ of Grade 92 steel artificially aged in a furnace at 600 and 650 °C, for 5,000, 10,000 and 20,000 h;

Comparison of the microstructural evolution of the two Grade 91 and 92 steels for aging times from 3,000 to 75,000 h.

3.1. Project N°1: GR 91 BM ageing

With this project, Study of microstructure and precipitate status evolution was carried out, after ageing isothermal thermal treatments in furnace of steel Grade 91 samples at 550 °C for time from 40,000 up to 65,000 h, at 600 °C and 650 °C for time from 50,000 up to 75,000 h.

The main results achieved are described as follows:

–
Microstructural features by SEM-EDS on steel Grade 91 aged samples, after total annealing time of about 62,000 h, revealed tempered martensite laths within original prior austenitic grains and polygonal ferritic sub-grains on martensite lath boundaries and within recovered martensite laths. Prior austenitic grain boundaries and polygonal ferrite sub-grain boundaries are covered by coarse M23C6 (Cr, Fe, Mo)-rich carbides and (Fe, Mo)-rich Laves phase particles, with size in the range 0.5–1.0 μm.
–
Precipitation and coarsening of M23C6 carbides have taken place, on prior austenitic and polygonal grain boundaries (sensibilization) as well as on martensite lath boundaries within recovered martensite laths and polygonal ferritic sub-grains, giving evidence of an advanced stage of the recovery process of the original tempered martensite in steel Grade 91 during annealing.
–
Different amounts of Laves phase and particle size distributions have been assessed on steel Grade 91 aged samples at temperature 550 °C, 600 °C and 650 °C for total annealing time of about 62,000 h. The highest Laves phase fraction area has been assessed in sample aged at 550 °C, approximately two orders of magnitude higher than that of sample aged at 650 °C. Laves phase fraction area of sample aged at 600 °C ranks between the two samples aged at 550 °C and 650 °C. The comparison for each ageing temperature between interrupted after 16,000 h aged sample and the previous interrupted after 4,000, 8,000, 12,000 h aged samples give evidence of a slight increase of fraction area of Laves phase within 12,000 h reaching almost steady value within 16,000 h.
–
The largest Laves phase particles are observed in aged sample at 650 °C for total annealing time of about 62,000 h, associated to the lower fraction area [3,24]. The comparison for each ageing temperature between interrupted after 16,000 h aged sample and the previous interrupted after 4,000, 8,000 and 12,000 h aged samples gives evidence of a slight coarsening of Laves phase. The evolution of Laves phase particle distribution produces a slight shift on the right towards coarser class size and broadening of distribution at ageing temperature 550 °C and 600 °C.

3.2. Project N°2: GR 91 BM creep

Creep tests were launched on steel Grade 91 base material at 550 °C, 600 °C and 650 °C with various stress levels (from 48 to 175 MPa). Microstructural analyses which include microstructural analysis by SEM-SE on morphological replica and precipitate state analysis by SEM-EDS on extractive replica have been carried out.

Microstructural modification and precipitate evolution on specimens after 4,000, 8,000 and 12,000 h have been evaluated by comparing the results of analysis on morphological and extractive replica taken on Grade 91 in the as-received normalized and tempered condition [22]:

Microstructures revealed by SEM on morphological replica taken on crept specimens are constituted by tempered martensite laths within original prior austenitic grains. Few polygonal ferritic sub-grains have been observed, especially at gauge length respect to normalized and tempered condition, giving evidence of a limited recovery of martensite during 12,000 h long creep tests in the tested conditions, namely at 550 °C, 600 °C and 650 °C under loading conditions, 175 MPa, 99 MPa and 54 MPa respectively.

–
Creep cavities have not been detected in the crept specimens at gauge length, giving evidence that creep damage in the experienced conditions is not yet observable.
–
Precipitate analysis by SEM-EDS on extractive replica taken on crept specimens revealed the presence M23C6 (Cr, Fe, Mo)-rich carbides on prior austenite grain boundaries, lath boundaries and within recovered martensite laths. Laves phase (Fe, Mo)-rich particles, not observed in normalized and tempered condition, have been detected on both gauge length and clamping head of the specimens, Laves phase precipitates are positioned on prior austenite grain boundaries lath boundaries and within recovered martensite laths.
–
The qualitative analyses of precipitate status on extractive replica from Grade 91 creep specimens after 4,000, 8,000 and 12,000 h creep tests, give evidence that the most important modification of precipitate status is the formation of Laves phase, originally not present in the material after normalization and tempering heat treatment. Laves phase precipitation starts just during the initial 4,000 h and developed during 12,000 h creep at 550 °C, 600 °C and 650 °C.
–
After 12,000 h creep test at 650 °C, Laves phase evolution is in accordance with the previous assessment on Grade 91 bulk samples after thermal ageing in furnace at 650 °C [3,24].

3.3. Project N°3: GR 91 HAZ creep

Creep tests were launched on steel Grade 91 weld at 550 °C, 600 °C and 650 °C with various stress levels (from 54 to 200 MPa). The total number of original creep test program have broken before the expected time to rupture (premature failures). Investigation on creep cavities after creep rupture has been carried out on different positions of gauge length and fractured zones in the HAZ of steel Grade 91 failed crept cross weld specimens after creep at 550 °C for long term >10,000 h. Investigation includes analysis of macro creep cavities by using light optical microscope (OM), SEM-EDS analysis of micro creep cavities in relation to their position on the microstructural features (along grain boundaries, lath boundaries and non-metallic inclusions) on different positions of gauge length and fractured zone in the HAZ of metallographic sections after etching to highlight microstructure and assessments of size distribution, geometrical parameters and fractional area of micro creep cavities and non-metallic inclusions by SEM-EDS-AIA on scattered area that covers different positions of gauge length and fractured zone in the HAZ on polished metallographic specimens. The main achieved results are summarized as follows:

–
Macro cavities and cracks are positioned within a distance of 0.5 mm from fracture surface in the HAZ of steel Grade 91 failed cross weld specimen.
–
A very small number of micro creep cavities have been revealed on the gauge length close to fractured zones positioned on the HAZ. Micro creep cavities are positioned in the vicinity of non-metallic inclusions along grain and lath boundaries.
–
Similar amounts of overall non-metallic inclusions and creep cavities have been assessed on the gauge length close to fractured zones on the HAZ of steel Grade 91 failed crept cross weld specimens of after creep tests at 500 °C, 600 °C and 650 °C.

The comparison of creep cavity state considering the combined information on microstructural characterization and the quantitative assessment of overall non-metallic inclusions and creep cavities of steel Grade 91 premature failed crept cross weld specimens gives evidence that the evolution of creep cavities has taken place in steel Grade 91 mostly in the heat affected zones.

Compared analysis of creep behavior of steel Grade 91 HAZ and predicted creep strength has been performed. Isothermal creep strength of Grade 91 HAZ at 550 °C is positioned slightly below the predicted creep strength of steel Grade 91. Isothermal creep strength of steel Grade 91 HAZ at 600 °C is positioned between the predicted creep strength of steel Grade 91 and the allowable stress assessed for steel Grade 91 welds.
3.4. Project N°4: GR 92 BM isothermal creep

Isothermal creep tests were launched on Grade 92 base material at 620 °C with various stress levels (from 86 to 114 MPa). One test has broken before the expected time to rupture (premature failure). Investigation on creep cavities after creep rupture has been carried out on different positions of gauge length and fractured zone within the gauge length of steel Grade 92 failed crept specimen. Investigation includes analysis of macro creep cavities by using light optical microscope (OM) and SEM-EDS analysis micro creep cavities in relation to their position on the microstructural features on different positions of gauge length and fractured zone positioned within the gauge length of metallographic sections after etching to highlight microstructure and assessment of size distribution, geometrical parameters and fractional area of micro creep cavities and non-metallic inclusions by SEM-EDS-AIA on scattered area that covers different positions of gauge length and fractured zone positioned within the gauge length of polished metallographic specimen [17,28].

The main achieved results are summarized as follows:

–
Quite few cavities with a limited size up 2 μm are positioned within a distance of 0.5 mm from fracture surface within the gauge length of steel Grade 92 failed crept specimen.
–
A very small number of micro creep cavities have been revealed on different positions of gauge length and fractured zone positioned within the gauge length of steel Grade 92 failed crept specimen. Micro creep cavities are positioned in the vicinity non-metallic inclusions along grain and lath boundaries.
–
The amount of overall non-metallic inclusions and creep cavities on the gauge length close to fractured zone of steel Grade 92 failed crept specimen has been assessed considering the fraction area and number density
–
Size distribution of overall non-metallic inclusions and creep cavities on the gauge length close to fractured zone in failed crept specimen is mainly characterized by a peak centred at 0.3 μm class size with a frequency around 35% and mean particles size around 0.35 μm.

3.5. Project N°5: GR 92 BM creep

Creep tests were launched on Grade 92 base material at 600 °C and 650 °C with various stress levels. Sampling and preparation of morphological and extraction replica taken on clamping head and gauge length of two selected interrupted after 12,000 h crept specimens have been performed. Microstructural analyses which include microstructural analysis by SEM-SE on morphological replica and precipitate state analysis by SEM-EDS on extractive replica have been carried out.

Microstructural modification and precipitate evolution specimens after 4,000, 8,000 and 12,000 have been evaluated by comparing the results of analysis on morphological and extractive replica taken on Grade 92 in the as-received normalized and tempered condition:

–
Microstructures revealed by SEM on morphological replica taken on crept specimens are tempered martensite laths within original prior austenitic grains. Few polygonal ferritic sub-grains have been observed respect to normalized and tempered condition specially at gauge length, giving evidence of a limited recovery of martensite during 12,000 h long creep tests namely at 600 °C and 650 °C under loading conditions, 125 MPa, and 65 MPa respectively.
–
Creep cavities have not been detected in crept specimens at gauge length, giving evidence that creep damage in the experience conditions is not yet observable.
–
Precipitate analysis by SEM-EDS on extractive replica taken on crept specimens revealed the presence M23C6 (Cr, Fe, W, Mo)-rich carbides on prior austenite grain boundaries, lath boundaries and within recovered martensite laths. Laves phase (Fe, W, Mo)-rich particles, not observed in Grade 92 in the as-received normalized and tempered condition, have been detected on both gauge length and clamping head of the crept specimens. Laves phase precipitates are positioned on prior austenite grain boundaries lath boundaries and within recovered martensite laths, especially at 600 °C creep temperature.
–
The qualitative analysis of precipitate status on extractive replica from Grade 92 crept specimens after 4,000, 8,000 and 12,000 h long creep tests, give evidence that the most important modification of precipitate status of Grade 92 is the formation of Laves phase, originally not present in the material after normalization and tempering heat treatment. Laves phase precipitation starts just during the initial 4,000 h and developed during 12,000 h long creep at 600 °C and 650 °C.

3.6. Project N°6: GR 92 HAZ creep

Creep tests were launched on steel Grade 92 weld at 600 °C and 650 °C with various stress levels (from 65 to 152 MPa). The total number of original creep test programme have broken before the expected time to rupture (premature failures).

Investigation on creep cavities after creep rupture has been carried out on different positions of gauge length and fractured zones in the HAZ of steel Grade 92 failed crept cross weld specimen after creep at 600 °C whose rupture occurred later. Investigation includes analysis of macro creep cavities by using light optical microscope (OM) and SEM-EDS analysis micro creep cavities in relation to their position on the microstructural features (along grain boundaries, lath boundaries and non-metallic inclusions) on different positions of gauge length and fractured zone in the HAZ of metallographic section after etching to highlight microstructure and assessments of size distribution, geometrical parameters and fractional area of micro creep cavities and non-metallic inclusions by SEM-EDS-AIA on scattered area that covers different positions of gauge length and fractured zone in the HAZ on polished metallographic specimen.

The main achieved results, which include the total number of premature failed crept cross weld specimens of steel Grade 92 after creep tests at 600 °C and 650 °C are summarized as follows:

–
Macro cavities and cracks are positioned within a distance of 0.3 mm from fracture surface in the HAZ of steel Grade 92 failed cross weld specimens. Macro creep cavities and cracks size varies from 2 μm up to 80 μm.
–
A very small number of micro creep cavities with size in the range from 0.2–2 μm has been revealed on the gauge length close to fractured zones positioned on the HAZ. Few large cavities with size 1–2 μm, are positioned close to non-metallic inclusions along grain and lath boundaries.
–
Similar amounts of overall creep cavities and non-metallic inclusions have been assessed on the gauge length close to fractured zones in failed crept cross weld specimens of steel Grade 92 after creep tests at 600 °C and 650 °C.

The comparison of creep cavity state considering the combined information on microstructural characterization and the quantitative assessment of overall non-metallic inclusions and creep cavities of steel Grade 92 premature failed crept cross weld specimens gives evidence that the evolution of creep cavities has taken place in steel Grade 92 mostly in the heat affected zones [17,28].

Compared analysis of creep behaviour of steel Grade 92 HAZ and predicted creep strength for steel Grade 92 has been performed:

–
Isothermal creep strength at 600 °C and 650 °C as function of creep rupture times of steel Grade 92 HAZ are compared with the predicted creep strength of Grade 92, allowable stress assessed for Grade 92 welds. Isothermal creep stress of Grade 92 HAZ at 650 °C is positioned below the predicted creep strength of Grade 92 [1,13,23].

3.7. Project N°7: Technical reports on steel grades 91 and 92 microstructure and precipitate evolution atlas

The last project includes the overall technical reports on microstructure and precipitate evolution of steel Grades 91 and 92, especially on activities and output data of laboratory tests and in-depth microstructure analyses, including a detailed description of the procedure, typology of specimens and methodologies used, in order to identify relationships of microstructure and precipitation state parameters between temperature-time-applied load, or evolution relationships (phase 1). Evolution relationships are further processed for each steel Grade 91 and 92 for a compared analysis of material behavior, base material and heat affected zone, under single o simultaneous effect of isothermal annealing and applied load (phase 2).

4. Conclusions

The methodologies to identify the evolution of creep damage of martensitic steels cannot be aligned with the methodologies suitable for conventional carbon steels, since the damage shows a particular trend that cannot be standardized in the method identified in UNI 11374;

It is necessary to identify the most suitable NDT techniques to determine creep damage.

There are currently several methods that appear to give very promising results.

The work on the new Italian standards is aimed at analyzing the fundamental mechanisms of creep damage in martensitic steels, identifying the limits and characteristics, in order to identify the most promising methodologies to detect the damage itself, also in compliance with safety.

Footnotes

Conflict of interest

None to report.

References

Allen

, Test programmes – P92 cross-welds (WG3A Autumn 2016 P92 welds).

Allen

D.J.

Wilson

J.W

Peyton

Shibli

and Hasegawa

, The development of the electromagnetic (EM) sensor technique for on-site creep damage detection and assessment, in: 1st EPERC Conference, Rome, 2019.

Atlante delle Microstrutture per l’Acciaio ASTM GRADO T/P91, 2013–2015.

Berti

and De Marco

, Experimental creep degradation curve from P92 grade steel by on site X-ray diffraction, in: “3rd International ECCC- Creep & Fracture Conference”, 5–7 maggio, Roma, 2014.

Berti

De Marco

Gariboldi

and Naumenko

, Sensitivity of on-site X-ray diffraction technology to detect creep phenomenon, in: “3rd International ECCC- Creep & Fracture Conference”, 5–7 maggio, Roma, 2014.

Berti

De Marco

Pinciroli

and Gariboldi

, Osservazioni da Diffrazione a raggi X e da durezza nello studio dei fenomeni a creep, in: “Sicurezza e Affidabilità delle Attrezzature a Pressione (SAFAP 2012)”, 14–15 giugno, Napoli, 2012.

Berti

De Marco

and Tolle

, Tecnologie diffrattometriche alla nanoscala: nuove frontiere di sviluppo e normazione, in: “Sicurezza e Affidabilità delle Attrezzature a Pressione (SAFAP 2010)”, 13–14 maggio, Venezia, 2010.

Bruchhausen

Altstadt

Austin

Dymacek

Holmström

Jeffs

, European standard on small punch testing of metallic materials, Ubiquity Proceedings 1(S1) (2018), 11. doi:10.5334/uproc.11.

Casiraghi

and Merckling

, Ipotesi di correlazione quantitativa di risultati microstrutturali con la durata di esposizione per acciai al 9%Cr, in: SAFAP, 2018.

10.

CWA 15627 Small Punch Test Method, 2007.

11.

Dutoit

DeBruycker

Huysmans

and Sarens

, Life assessment of Grade 91 Components Based on Hardness and Creep Void Density (HAVOC Method), in: 3rd ECCC Conference, Roma, 2014.

12.

Folgarait

Cipolla

Di Gianfrancesco

Lombardi

and Tiberi

, Criteri avanzati di progettazione e indagine di acciai e componenti soggetti a creep per impianti USC, in: GIORNATA DI STUDIO SUL CREEP, ISPESL, Roma - 2 aprile, 2009.

13.

Gotti

, Italian Creep Group activity update for ECCC WG3A, in: ECCC WG3A Web Meeting, 14th April 2020.

14.

Hald

Tassa

and Matera

, Creep behavior and precipitate evolution of P92 after long-term creep, in: Proc. 4th International ECCC Conference Creep & Fractur Düsseldorf, Germany, 2017.

15.

Klenk

and Maile

K.-H.

, Life Assessment of high temperature components on the basis of VGB guidelines, in: ETD New High Temperature Materials Seminar, 2008, London, OMMI 2008, Volume 5, Issue 3.

16.

Masuyama

, Life Assessment issues for P91 components, in: ETD New High Temperature Materials Seminar, 30/11/2009, London.

17.

Parker

J.D.

and Siefert

, Creep fracture in tempered martensitic steels, International Journal of Petrochemical Science & Engineering 3(1) (2018).

18.

Salvo

and Merckling

, Metallografia non distruttiva innovativa per la previsione della vita residua di materiali metallici, in: SAFAP, 2018.

19.

Shibli

, Remaining life assessment issues in high Cr martensitic steels and development of new innovative tools for damage monitoring and integrity assessment, Transactions of the Indian Institute of Metals 63 (2010), 339–348.

20.

Shibli

, Inspection and life assessment challenges for the high temp. Plant P91/P92 components, in: Parsons Conference, University of Portsmouth.

21.

Shibli

, Developments in the inspection, monitoring and life assessment of aberrant P91 and P92 component damage, in: 4th ECCC Conference, Dusseldorf, 2017.

22.

Steel Grades 91 and 92 microstructure evolution atlas, Second Progress Semiannual Technical Report, RINA Technical Report No. 20984R Rev., 1st April 2020.

23.

Tassa

and Matera

, ECCC WG3A Long term creep tests for P92 steel at CSM (ECCC WG3A 2015).

24.

Tonti

Lega

Ramires

Alleva

Di Gianfrancesco

Matera

and Augugliaro

, Microstructure atlas and Life Assessment Tools, in: Proc. 3rd International ECCC Conference Creep & Fracture, Rome Italy, 2014.

25.

UNI CEN ISO/TS 21432 “Prove non distruttive - Metodo di prova normalizzato per la determinazione delle tensioni residue mediante diffrazione a neutroni”.

26.

Wilson

J.W.

Allen

D.J.

Peyton

A.J.

and Shibli

, Detection of creep degradation during pressure vessel testing using electromagnetic sensor technology, Materials at High Temperatures (2017).

27.

Wilson

J.W.

Allen

D.J.

Peyton

A.J.

Shibli

and Davis

, Detection of creep degradation during pressure vessel testing using electromagnetic sensor technology, Materials at High Temperatures 34(5–6) ,448–457.

28.

Yadav

Sonderegger

, Characterization and quantification of cavities in 9Cr martensitic steel for power plants, Materials Science and Technology (2015).