WO2024100433A1

WO2024100433A1 - Ferritic stainless steel sheet and associated production method

Info

Publication number: WO2024100433A1
Application number: PCT/IB2022/060735
Authority: WO
Inventors: Coralie PARRENS; Pierre-Olivier Santacreu; Malo CARRADOT; Pierre-Emmanuel LEGER
Original assignee: Aperam
Priority date: 2022-11-08
Filing date: 2022-11-08
Publication date: 2024-05-16

Abstract

The invention relates to a ferritic stainless steel sheet, the composition of which comprises, the contents being expressed by weight: C ≤ 0.03% 0.25% ≤ Mn ≤ 1%, preferably 0.3% ≤ Mn ≤ 0.5% 0% < Si ≤ 0.20%, preferably Si ≤ 0.15% with Mn/Si ≥ 1.2 S ≤ 0.005% P ≤ 0.04% 19.0% ≤ Cr ≤ 24.0% Ni ≤ 0.5% Mo ≤ 0.10% N ≤ 0.03% Cu ≤ 0.20% 0.40% ≤ Nb ≤ 1.0% 0.05% ≤ Ti ≤ 0.2%, preferably 0.05% ≤ Ti ≤ 0.15% Zr ≤ 0.02% Al ≤ 0.02% V ≤ 0.2% Co ≤ 0.05% Sn ≤ 0.05%, rare earths ≤ 800 ppm, it being understood that: V + Zr + Al ≤ 0.2% Ti + V + Zr + Al ≤ 0.30% Ti + Nb ≤ 1.0% Ni + Cu + Co ≤ 0.60% 2xNb-7xC ≥ 0.8% 29 0% ≤ Ti - 4xN ≤ 0.15% 0.2 ppm ≤ Ca ≤ 20 ppm 1 ppm ≤ O ≤ 60 ppm, the remainder of the composition consisting of iron and unavoidable impurities resulting from the preparation, the sheet being an annealed and pickled sheet, the sheet comprising a volume fraction of Fe₂Nb Laves phases of less than 0.2%.

Description

Ferritic stainless steel sheet and associated manufacturing process

The present invention relates to a ferritic stainless steel sheet, as well as a method of manufacturing the same.

The development of high temperature electrochemical applications (fuel cells or solid oxide electrolyzers) to transform hydrogenated compounds into electricity or conversely produce hydrogenated products from carbon-free electricity requires the use of new materials which must function between 500°C and 1000°C for tens of thousands of hours with a few dozen on-off cycles. The anodes, cathodes and electrolytes constituting the cell of these systems are made from ceramics or cermet, the nature of which can vary depending on the design. These cells are supported or linked to metal parts called interconnectors whose role is to distribute the reactive gases in the cell and to collect the electrons. The low thermal expansion coefficient of cells whose electrolyte is often based on yttriated zirconia (10.5x10" ⁶ K ^-1 ) requires the use of high chromium ferritic stainless steels whose thermal expansion coefficient is very close and which are capable of resisting oxidation at these high temperatures. Furthermore, the interconnectors must not cause degradation of the performance of the cell by contamination of the anode and the cathode, in particular by diffusion of chromium or by. too high electrical resistivity of the oxide forming on its surface.

The ferritic stainless steels used today are very high chromium, with control of certain residuals such as silicon kept very low and contain rare earths (especially lanthanum). Their vacuum production from pure raw materials is therefore expensive. The weight share of the interconnectors being very high in these systems, the competitiveness of these electrolyzers is affected. The idea is therefore to use more traditional ferritic stainless steels produced by an electric furnace such as AISI 441 or 444 steels which have good hot properties. However, experiments show that, under certain conditions of use, the performance of such steels degrades more quickly despite an oxide layer comparable in composition and thickness. The origin of this more rapid degradation of performance is a segregation of silicon at the metal-oxide interface which forms a more or less continuous thin film of silica (silicon oxide) which is very electrically resistive, with a resistivity of order of 10 ⁶ Q.cm at 850°C. This value should be compared to the resistivity of chromine CrsOs, which is of the order of 100 Q.cm at 850°C. THE silica film is very thin in thickness, and typically has a thickness of the order of a few tens of nanometers at 850°C after 1000 hours of operation. However, given the electrical resistivity of silica, it contributes to greatly increasing the specific surface resistance (Area specific resistance or ASR in English). However, silicon cannot be completely eliminated by current production processes at a reasonable cost.

An aim of the invention is therefore to provide a ferritic stainless steel sheet having conductivity properties comparable to those of very high chromium steels, even after use for long periods at high temperature.

Preferably, we also seek to improve the creep resistance of the sheet because the system is also subjected to thermomechanical loads which can generate deformations modifying the contacts and the functions of the interconnectors.

For this purpose, the subject of the invention is a ferritic stainless steel sheet, the composition of which comprises, the contents being expressed by weight:

C ≤ 0.03%

0.25% ≤ Mn ≤ 1%, preferably 0.3% ≤ Mn ≤ 0.5% 0% ≤ Si ≤ 0.20%, preferably Si ≤ 0.15% with Mn/Si > 1.2

S ≤ 0.005%

P ≤ 0.04%

19.0% ≤ Cr ≤ 24.0%

Ni ≤ 0.5% Mo ≤ 0.10% N ≤ 0.03% Cu ≤ 0.20% 0.40% ≤ Nb ≤ 1.0% 0.05% ≤ Ti ≤ 0.2%, preferably 0 .05% ≤ Ti ≤ 0.15% Zr ≤ 0.02% Al ≤ 0.02%

V ≤ 0.2%

Co ≤ 0.05%

Sn ≤ 0.05%,

Rare earths ≤ 800 ppm provided that:

V + Zr +AI ≤ 0.2% Ti + V + Zr + AI ≤ 0.30%

Ti + Nb ≤ 1.0%

Ni + Cu + Co s 0.60%

2xNb-7xC > 0.8%

0% ≤ Ti - 4xN ≤ 0.15%

0.2 ppm ≤ Ca ≤ 20 ppm

1 ppm ≤ O ≤ 60 ppm the rest of the composition being made up of iron and inevitable impurities resulting from the preparation, the sheet being an annealed and pickled sheet, the sheet comprising a lower volume fraction of Laves Fe ₂ Nb phases at 0.2%.

The ferritic steel sheet according to the invention may also comprise one or more of the following characteristics, taken individually or in any technically possible combination:

- the alloy has a rare earth content of between 50 ppm and 800 ppm.

- the niobium content of the alloy respects the following relationship: Nb - 10x(C+N) > 0%.

- the sheet metal has a thickness of between 0.1 mm and 2.5 mm;

- the sheet has an average grain size of between 30 micrometers and 80 micrometers when the sheet has a thickness of between 1.2 mm and 2.5 mm and an average grain size of between 15 micrometers and 80 micrometers when the sheet has a thickness greater than or equal to 0.1 mm and less than 1.2 mm;

- the sheet metal is cold-rolled and annealed sheet metal;

- when subjected to heat treatment at a temperature of 850°C for a period of 1000 hours, the sheet comprises a volume fraction of Fe ₂ Nb Laves phases greater than or equal to 0.8%;

- when subjected to heat treatment at a temperature of 850°C for a period of 1000 hours, the sheet comprises a volume fraction of cubic phases Fe ₃ Nb ₃ X less than 0.05%;

- when subjected to heat treatment at a temperature of 850°C for a period of 1000 hours, the sheet comprises, on each of its faces, a layer of oxides and, at the interface between the steel the sheet metal and the oxide layer, silicon oxide precipitates, such that the surface fraction of the silicon oxide precipitates at the interface between the steel of the sheet metal and the oxide layer is less than or equal to at 0.35; And - the oxide layer has a thickness less than or equal to 10 μm.

The invention also relates to a method of manufacturing a ferritic stainless steel sheet comprising the following steps:

- a steel is produced having the composition as described above;

- we proceed to the casting of a semi-product from this steel;

- the semi-finished product is brought to a temperature greater than or equal to 1150°C and less than or equal to 1260°C for a period of between 40 minutes and 60 minutes and the semi-finished product is hot rolled to obtain a hot-rolled sheet hot with a thickness between 2.5 mm and 6 mm;

- the hot-rolled sheet is annealed,

- the hot-rolled and annealed sheet is stripped,

- said hot-rolled sheet is cold rolled, at a temperature between room temperature and 300°C, in a single step or in several steps separated by intermediate annealing;

- a final annealing of the cold-rolled sheet is carried out, at a temperature between 1000°C and 1100°C and for a period of between 10 seconds and 6 minutes to obtain a completely recrystallized structure.

The manufacturing process according to the invention may also comprise one or more of the following characteristics, taken individually or in any technically possible combination:

- the annealing of the hot-rolled sheet is carried out at a temperature between 1000°C and 1100°C, for a period of 30 seconds to 6 minutes;

- the intermediate anneal(s) are carried out at a temperature between 950°C and 1100°C for a period of 30 seconds to 6 minutes; And

- the final annealing is carried out at a temperature between 1050°C and 1090°C.

The invention will be well understood and other aspects and advantages will appear more clearly on reading the detailed description which follows of an exemplary embodiment given with reference to the appended figures, in which:

Figure 1 schematically illustrates a sheet according to the invention seen in a cross section after an aging heat treatment;

Figure 2 schematically illustrates the determination of the surface fraction of silica precipitates at the metal-oxide interface after such an aging treatment.

The invention relates to a ferritic stainless steel sheet, the composition of which comprises, the contents being expressed by weight: C ≤ 0.03%

0.25% ≤ Mn ≤ 1%, preferably 0.3% ≤ Mn ≤ 0.5%

0% ≤ Si ≤ 0.20%, preferably Si ≤ 0.15% with Mn/Si > 1.2

S ≤ 0.005%

P ≤ 0.04%

19.0% ≤ Cr ≤ 24.0%

Ni ≤ 0.5%

MB ≤ 0.10%

N ≤ 0.03%

Cu ≤ 0.20%

0.40% ≤ Number ≤ 1.0%

0.05% ≤ Ti ≤ 0.2%, preferably 0.05% ≤ Ti ≤ 0.15%

Zr ≤ 0.02%

Al ≤ 0.02%

V ≤ 0.2%

Co ≤ 0.05%

Sn ≤ 0.05%,

Rare earths ≤ 800 ppm provided that:

V + Zr +AI ≤ 0.2%

Ti + V + Zr + AI ≤ 0.30%

Ti + Nb ≤ 1.0%

Ni + Cu + Co ≤ 0.60%

2xNb-7xC ≥ 0.8%

0% ≤ Ti - 4xN ≤ 0.15%

0.2 ppm ≤ Ca ≤ 20 ppm

1 ppm ≤ O ≤ 60 ppm the rest of the composition being made up of iron and inevitable impurities resulting from the production.

Regarding the chemical composition of steel, carbon increases the mechanical characteristics at high temperatures, especially creep resistance. However, due to its very low solubility in ferrite, carbon tends to precipitate in the form of M23C6 or M7C3 carbides at a temperature below approximately 900°C. This precipitation generally located at grain boundaries can lead to a depletion of chromium in the vicinity of these joints and therefore sensitization to intergranular corrosion. This awareness can be encountered in particular in Heat Affected Zones in welding which have been heated to very high temperatures. The carbon content must therefore be limited to at most 0.03% to obtain satisfactory resistance to intergranular corrosion as well as not to reduce formability. Additionally, the carbon content must satisfy a relationship with the niobium, as will be explained later.

Chromium is an essential element for stabilizing the ferritic phase and increasing resistance to oxidation. In conjunction with the other elements of the composition, its minimum content must be greater than or equal to 19.0% in order to obtain a ferritic structure at any temperature and good resistance to cyclic oxidation, particularly when the thickness of the sheet metal is thin (less than or equal to 0.5 mm) and that this thickness limits the reservoir of chromium available during oxidation. Indeed, during aging, and as illustrated in Figure 1, an oxide layer 2 comprising an internal layer 4 of chromine Cr ₂ O3 and possibly an external layer 5 of chromium oxide rich in manganese forms on the surface of the metal substrate 1 and protects the steel over very long periods and at high temperatures. The maximum chromium content must not, however, exceed 24.0%, otherwise the mechanical resistance at room temperature will be excessively increased, generating significant brittleness and degrading the formability.

In the context of this invention, the terms “internal” and “external” are used in relation to the proximity to the metal substrate 1, an internal layer being closer to the metal substrate 1 than an external layer.

The alloy has a manganese content of between 0.25% by weight and 1% by weight. At these levels, manganese increases the mechanical characteristics of the alloy and also allows the formation of an external layer 5 of chromium oxide rich in manganese and capable of containing spinel type iron (Mn,Fe)Cr ₂ O ₄ . The good thermodynamic stability of these oxides makes it possible to limit the evaporation of chromium at high temperatures in the presence of water vapor. This outer layer 5 of chromium oxide rich in manganese is also favorable to the good adhesion of a protective coating, for example a coating of the LSM type (lanthanum manganite doped with strontium) or of the MCO type (cobalt manganese oxide spinel). ). These spinels also have very good electrical conduction, with a resistivity of around 20 Q.cm at 850°C. However, beyond 1% by weight, the hot oxidation kinetics becomes too rapid and a thick, very adherent oxide layer develops, making difficult stripping operations during sheet metal manufacturing. Manganese is also a gammagenic element that should be limited like Ni, Cu, Co in ferritic steels. The manganese content is therefore limited to 1%.

Preferably the manganese content is between 0.3% and 0.5%.

Silicon is a very effective element for increasing resistance to oxidation. However, the silicon oxide (silica) 3 which forms at the metal-oxide interface has a very low expansion coefficient, of the order of 1.10 ^-6 K' ¹ , ten times lower than that of the base metal. and chromium oxide Cr ₂ O ₃ , reducing the adhesion of the oxide layer 2 as a whole. However, poor adhesion of the oxide layer 2 leads to degraded conductivity properties. In addition, silicon has a high electrical resistivity, which is very harmful in the intended application since the oxidized metal must have good electrical conductivity. Silicon is also a cold-hardening element in ferrite, reducing its ductility and its ability to be cold-formed. The silicon content must therefore be limited to a minimum and must not exceed 0.20% by weight with a preference for a maximum of 0.15%. The silicon content is greater than 0%, due to the inevitable presence of trace silicon. The silicon content generally remains greater than or equal to 0.05%. Indeed, reducing the Si content below this value requires expensive production processes.

In addition, the Mn/Si ratio must be greater than or equal to 1.2, to promote the formation of the outer layer 5 of chromium oxide rich in manganese of type (Mn,Fe)Cr2C>4 described above to the detriment silicon oxide.

Sulfur and phosphorus are impurities that reduce hot ductility and formability. Phosphorus easily segregates at grain boundaries and reduces their cohesion. Sulfur is also harmful to oxidation by segregating at the metal-oxide interface and reducing its adhesion to the metal. As such, the sulfur and phosphorus contents must be respectively less than or equal to 0.005% and 0.04% by weight.

Nickel is a gammagenic element that increases the ductility of steel. In order to maintain a single-phase ferritic structure, its content is limited. Furthermore, nickel does not improve the targeted properties and its voluntary addition would increase the cost of production given its high price. Its content must be as low as possible and less than or equal to 0.5% by weight.

Molybdenum increases not only high temperature resistance but also oxidation resistance. However, in steels with a high chromium content, containing titanium and niobium, it generates fragility of the ferritic matrix, particularly in hot-rolled strips with a thickness of between 2.5 mm and 6 mm. Molybdenum excessively reduces ductility and formability, and is an expensive additive. Its content must be less than or equal to 0.10%, preferably strictly less than 0.10%.

Like carbon, nitrogen increases mechanical characteristics. However, nitrogen tends to precipitate at grain boundaries in the form of nitrides, thereby reducing corrosion resistance. In order to limit the problems of sensitization to intergranular corrosion, the nitrogen content must be less than or equal to 0.03%.

Copper has a hot hardening effect. In excessive quantities, however, it reduces ductility during hot rolling. Like nickel, it is also a gammagenic element that must be limited. As such, the copper content must therefore be less than or equal to 0.20% by weight.

Niobium is an important element of the invention. Usually, this element can be used as a stabilizing element in ferritic stainless steels: in fact, the phenomenon of sensitization to intergranular corrosion mentioned above can be avoided by the addition of elements forming very thermally stable carbides or carbonitrides. . In this way, the carbon and nitrogen in solution are reduced as much as possible and thus a subsequent precipitation of carbides and chromium nitrides is avoided. Niobium, as well as titanium and, to a lesser extent, zirconium and vanadium, therefore stably fixes carbon and nitrogen.

But niobium also combines with iron to form certain intermetallic compounds in the interval 650°C-950°C: the inventors have demonstrated that an intergranular precipitation of hexagonal Fe ₂ Nb occurring at high temperature could be brought to light. advantage to increase the mechanical properties when hot, particularly in creep, therefore in the targeted conditions of use.

Furthermore, the inventors have discovered that the precipitates of Fe ₂ Nb compounds with a hexagonal structure (called Laves phases) capture part of the silicon, and therefore minimize the formation of silicon oxides, which is sought for the application. In particular, compared _to the cubic phases Fe ₃ Nb ₃ . Adjusting the initial silicon content and capturing part of the silicon by such phases makes it possible to very significantly reduce the segregation of silicon at the metal-oxide interface and the formation of very resistive silica. In addition, the nature and spatial distribution of these intergranular precipitates, that is to say at grain boundaries, are very favorable for resisting creep up to 1000°C. In order to form Fe ₂ Nb phases under the targeted conditions of use, several conditions must be respected: the niobium content must be between 0.40% and 1.0% and such that 2xNb-7xC > 0.8% , and the composition further comprises titanium at a content of between 0.05% and 0.2%, and such that 0% ≤ Ti - 4xN ≤ 0.15%. Preferably, the niobium, carbon and nitrogen contents are also such that Nb - 10x(C+N) > 0%.

If the total Nb content of the steel is less than 0.40%, the steel is insufficiently stabilized and the quantity of Fe ₂ Nb precipitates formed at high temperature is insufficient to obtain the targeted properties at high temperature. To obtain this favorable precipitation of niobium, the inventors have also highlighted the importance of the effective niobium content: the effective niobium designates the quantity of niobium in solid solution available to precipitate with the iron, making the hypothesis that the carbon and nitrogen completely precipitated with niobium and titanium in the form of carbonitrides TiN, TiC and Nb(C,N). To ensure a sufficient effective niobium content, the niobium content must respect the relationship 2xNb-7xC ≥ 0.8%, and preferably respect the relationship Nb - 10x(C+N) > 0%.

Conversely, a significant excess of Nb is weakening, particularly in a very high chromium ferrite. This excess greatly increases the hardness and slows down recrystallization, which also limits the cold processing capacity of the metal. We therefore limit the niobium content to Nb ≤ 1.0%.

To promote the formation of Fe ₂ Nb phases, to the detriment of Fe ₃ Nb ₃ X phases, the titanium content must be greater than or equal to 0.05%, but less than or equal to 0.2% to limit excessive internal oxidation at high temperature. Preferably, the titanium content is greater than or equal to 0.05% and/or less than or equal to 0.15%.

Furthermore, titanium must respect a relationship with nitrogen 0% ≤ Ti-4xN ≤ 0.15%. Otherwise, the niobium precipitates, not in the form of Fe ₂ Nb with a hexagonal structure from 650°C, but in the form of cubic Fe ₃ Nb ₃ X compounds, less effective in trapping silicon.

Furthermore, the sum of the Nb and Ti contents must be less than or equal to 1.0% to avoid excessive fragility of the metal and guarantee satisfactory resilience of the welds.

Vanadium, zirconium and aluminum are nitrogen stabilizing elements, which increase mechanical resistance at high temperatures, but it is however advisable to limit the total content of these elements to at most 0.2% so as not to decrease trainability. Furthermore, the aluminum content is limited to at most 0.02% because aluminum is susceptible to oxidation to form very electrically resistive alumina. In particular, the electrical resistivity of alumina is of the order of 10 ⁷ Q.cm at 850°C, alumina is therefore ten times more resistive than silica (SiO ₂ ).

The zirconium content is at most 0.02% so as not to reduce formability and limit the risk of surface defects.

The vanadium content is at most 0.2% so as not to reduce formability.

In addition, the inventors have demonstrated that the contents of titanium, aluminum, vanadium and zirconium must be jointly limited in order to limit the fragility of the metal. The sum of their contents must be such that: Ti+AI+V+Zr ≤ 0.30%.

Cobalt is a heat-hardening element which degrades formability. Therefore, its content must be less than or equal to 0.05% by weight.

In addition, the content of cobalt added to that of copper and nickel must be as low as possible to avoid an increase in electrical resistivity detrimental to the application. To this end, the total Ni+Cu+Co content must be less than or equal to 0.60%.

In order to avoid hot forgeability problems, the tin content must be less than or equal to 0.05%.

The alloy also contains calcium at a level between 0.2 ppm and 20 ppm. In fact, calcium is involved in the production process to reduce the sulfur and oxygen contents of the steel. The concentration should be limited to avoid internal oxidation of the metal during use at high temperatures.

The oxygen content of steel is between 1 ppm and 60 ppm, its content being for example around 20 ppm. As with calcium, the concentration should be limited to avoid the presence of oxide inclusions or internal oxidation of the metal during use at high temperatures.

Rare earth elements (REE) can be added to improve the adhesion of the oxide layers which make the steel resistant to corrosion. However, the rare earth content must not exceed 800 ppm. Beyond this content, the production of the metal could be made difficult due to the reactions of the rare earths with the refractories lining the ladle. These reactions would lead to the notable formation of REE oxides which would degrade the inclusionary cleanliness of the steel. Furthermore, the effectiveness of REEs is sufficient at the proposed contents, and going beyond would only unnecessarily increase the cost of production due to the high price of REEs, and also the accelerated wear of the refractories that this would entail. . When rare earths are added, their content is preferably at least 50 ppm. Preferably, the rare earths are chosen from cerium, lanthanum and yttrium or combinations thereof. In particular, rare earths include a mixture of cerium and lanthanum. Preferably, rare earths consist of a mixture of cerium and lanthanum.

The steel according to the invention is generally in the form of cold-rolled and annealed sheet and the thickness of the sheet is generally between 0.1 mm and 2.5 mm.

According to the invention, the structure of the steel in the delivered state, that is to say cold-rolled and annealed sheet metal, is completely recrystallized.

Bending creep of sheet metal at high temperature depends on the thickness of the sheet metal. For an identical mechanical load and metallurgical state, a sheet of thinner thickness deforms more in bending.

When the thickness of the cold-rolled and annealed sheet is between 1.2 mm and 2.5 mm, the average grain size of the steel is preferably between 30 micrometers and 80 micrometers, i.e. i.e. ASTM index (ASTM E-112 standard) between 4 and 7. In this strip thickness range, a grain size greater than or equal to 30 micrometers (ASTM ≤ 7) is advantageous, because it guarantees a low deformation by flexural creep of the sheet at high temperature relative to its thickness. Furthermore, in this strip thickness range, the grain size is preferably less than or equal to 80 micrometers (ASTM > 4). Indeed, a grain size greater than 80 micrometers (ASTM ≤ 4) leads to the appearance of unsightly surface irregularities, known as “orange peel” appearance, during shaping at room temperature and harmful to the good adhesion of the protective coating.

When the thickness of the cold-rolled and annealed sheet is greater than or equal to 0.1 mm and less than 1.2 mm, the average grain size of the steel is preferably between 15 micrometers and 80 micrometers, c that is to say with an ASTM index between 4 and 9. In this band thickness range, a grain size greater than or equal to 15 micrometers (ASTM ≤ 9) is advantageous, because it guarantees low deformation by creep of the sheet at high temperature relative to its thickness. Furthermore, in this band thickness range, the grain size is preferably less than or equal to 80 micrometers (ASTM > 4). Indeed, a grain size greater than 80 micrometers (ASTM ≤ 4) leads to the appearance of unsightly surface irregularities, known as “orange peel” appearance, during shaping at room temperature, and harmful to the good adhesion of the protective coating.

The microstructure of the cold-rolled and annealed sheet in the delivered state includes precipitates which consist mainly of titanium and niobium carbonitrides intragranular. In fact, the annealing carried out on the sheet has the effect of dissolving the intermetallic precipitates of the Laves Fe ₂ Nb type with a hexagonal structure and the Fe ₃ Nb ₃ X type with a cubic structure present in the microstructure.

In particular, the volume fraction of Laves Fe ₂ Nb phases in the cold-rolled and annealed sheet in the delivered state is less than 0.2%.

The volume fraction of Fe ₂ Nb Laves phases is determined as follows.

Firstly, standard polishing is carried out with abrasives with a grain size of up to 1 μm, followed by an electrolytic attack in 60% nitric acid on a cross section of a sample, taken in a direction orthogonal to the direction of rolling.

The observation is made using an electron microscope in backscattered electron mode. A minimum magnification of x1000 is used to obtain an overall view and accurately determine particle size. Five images are produced per sample and condition. The backscattered electron mode generates a chemical composition contrast on a scale of 256 levels called gray levels ranging from white (255) to black (0).

The volume fraction of the Fe ₂ Nb and Fe ₃ Nb ₃ X intermetallic phases is determined by image analysis, for example using Image J software, from the images thus obtained.

Firstly, we process the images obtained by a thresholding method so as to keep only the whitest precipitates corresponding to the intermetallic phases Fe ₂ Nb and Fe ₃ Nb ₃ X. Indeed, in backscattered electron mode, the precipitates rich in niobium and iron, due to their atomic number, appear as the lightest compounds in the images. Therefore, the intermetallic phases Fe ₂ Nb and Fe ₃ Nb ₃ During this thresholding step, for each image, the threshold is chosen so as to distinguish the intermetallic precipitates Fe ₂ Nb and Fe ₃ Nb ₃ X from the rest of the image. After thresholding, the intermetallic precipitates are represented in white in a black matrix.

Manual filtering of the images is then carried out to eliminate artifacts such as holes or impurities.

We then determine the surface fraction of intermetallic phases Fe ₂ Nb and Fe ₃ Nb ₃ X from this thresholded and filtered image. In this context, it is accepted that the volume fraction is equal to the surface fraction. We therefore obtain the volume fraction of intermetallic phases Fe ₂ Nb and Fe ₃ Nb ₃ X. Furthermore, measurements of the Fe/Nb ratio are carried out by energy dispersive spectroscopy (EDS) in each of the images. This ratio being different in _each of these intermetallic phases, it makes it possible to distinguish _between cubic intermetallic Fe ₃ Nb ₃ of intermetallic phases Fe ₂ Nb and Fe ₃ Nb ₃ X.

In cold-rolled and annealed sheet metal, niobium is mainly in solid solution. In particular, the mass content of Nb in solid solution in the cold-rolled and annealed sheet in the delivered state is at least 0.3%.

The sheet according to the invention is further characterized in that, when it is subjected to heat treatment at a temperature between 650°C and 1000°C for a time greater than or equal to 30 minutes, the structure of the sheet comprises, in addition to the carbonitrides of titanium and niobium mentioned above, a homogeneous and intergranular precipitation of Fe ₂ Nb compounds with a hexagonal structure (Laves phases). The structure of the sheet subjected to such heat treatment may further include an intraganular precipitation of Fe ₂ Nb compounds, which, however, decreases as the duration of the heat treatment increases.

In particular, at the end of an aging heat treatment in air at a temperature of 850°C for a period of 1000 hours, the volume fraction of Fe ₂ Nb Laves phases in the sheet is at least 0, 8%. Furthermore, these phases of Fe ₂ Nb lavas are predominantly intergranular.

The aging heat treatment in air at a temperature of 850°C for a period of 1000 hours is considered representative of the conditions of use of steel and is generally used for the qualification of steels.

According to the invention, the Fe ₂ Nb precipitates are the majority among the intergranular precipitates. As mentioned above, these precipitates have the advantage of capturing part of the silicon to reduce its content in the solid solution, which is then less likely to segregate at the metal-oxide interface.

In particular, after such heat treatment, the volume fraction of Fe ₃ Nb ₃ X precipitates remains less than 0.05%.

Furthermore, together with the grain size, the nature and distribution of these Fe ₂ Nb precipitates are very favorable for resisting creep up to 1000°C.

Due to the formation of Fe ₂ Nb precipitates, the segregation of silica at the metal-oxide interface is very limited compared to the steels of the prior art. In particular, after such an aging treatment in air at a temperature of 850° C. for 1000 hours, the surface fraction of silica segregation at the metal-oxide interface remains limited.

In particular, at the end of such an aging treatment, as illustrated in Figure 1, the sheet comprises, on each of its faces, a layer of oxides 2 above the base metal 1. The layer of oxides 2 comprises an internal layer 4 of chromium oxide Cr ₂ O3 (chromine) and an external layer 5 of chromium oxide rich in manganese and which may contain spinel type iron (Mn, Fe) Cr ₂ O4.

At the interface between the oxide layer 2 and the steel of the sheet, also called base metal 1, the sheet includes precipitates of silicon oxide 3 or silica. In the sheets according to the invention, after such an aging treatment, the surface fraction of the silica precipitates at this metal-oxide interface is less than or equal to 0.35, which means that the silica covers at most 35% of the metal-oxide interface. The electrical conductivity of silica, of the order of 10 ^-6 S.cm ^-1 at 850°C, is 10,000 times lower than that of chromine, of the order of 10 ^-2 S.cm ^-1 at 850°C, and its coefficient of thermal expansion is 10 times lower than those of chromine, the base metal and zirconia, which makes up the electrolyte of the electrochemical cell. This surface fraction of silica at the metal-oxide interface is directly related to the resistivity, the specific surface resistance and the conductivity of the metal-oxide interface. Thus, when the surface fraction of silica at the interface is equal to 35%, the electrical conductivity of the interface is reduced by 35% compared to a configuration without silica, going from 10 ^-2 S.cm ^-1 in l The absence of silica at 6.5.10 ^-3 S.cm ^-1 where the resistivity of the interface is increased by approximately 50%, going from 100 Q.cm to 150 Q.cm. Furthermore, the specific surface resistance of the interface, equal to the resistivity multiplied by the thickness of the silica film, is increased equivalently.

The surface fraction of the segregation of silicon oxides 3 can be determined from an image of a cross section of a sample, in a direction orthogonal to the rolling direction, obtained by scanning electron microscope with magnification of x10,000 in backscattered electron mode.

The horizontal edges of the image are parallel to the surface of the sheet metal. The length L of the observed cross section, in a plane transverse to the sheet, is equal to 12 pm.

The backscattered electron mode generates a chemical composition contrast on a scale of 256 levels called gray levels ranging from white (255) to black (0). In the images obtained, silica, by its composition and the atomic number of the elements constituent, appears as the darkest phase in contrast with the base metal which appears very light and the chromium oxide with an intermediate shade of gray.

The image thus obtained corresponds to the image noted (a) in Figure 2.

This image is then analyzed using image analysis software, for example using Image J software.

More particularly, the image is, initially, transformed by the image analysis software via automated processing in order to increase the contrast between the elements of the image: the base metal 1, the layer d oxides 2 and the segregations of silicon oxides 3. The objective of this step is that all the acquired images have an identical contrast and highlight the segregations of silicon oxides 3 independently of the metal-oxide interface observed and oxidized metal analyzed. At the end of this processing, we obtain the image noted (b) in Figure 2.

The image is then thresholded so as to distinguish the silicon oxide 3 segregations from the rest of the image. Thus, we only keep two levels, black for the segregations of silicon oxides 3 and white for the rest of the image, by setting a threshold adapted to distinguish the segregations of silicon oxides 3. As an example , the threshold is chosen equal to 70, pixels with a gray level greater than 70 being represented in white 255 and pixels with a gray level less than or equal to 70 being represented in black. Then, possible artifacts are filtered and the processing quality of each image is checked manually.

We then determine, from this image, the sum of the projected lengths Li, on a longitudinal axis, of the regions of the interface in which the silicon oxide precipitates 3 are present in this measurement field, as illustrated on the Image (c) of Figure 2.

The surface fraction is then calculated as the ratio between the sum of the projected lengths and the length L of the measurement field Z Li/L. In the example illustrated in Figure 2, the surface fraction is equal to 77%.

According to the invention, after heat treatment in air at 850°C for 1000 hours, this Z Li/L ratio remains less than or equal to 0.35, i.e. 35%.

Furthermore, taking into account the chromium content of the steel according to the invention, the total thickness of the oxide layer 2 generally remains less than or equal to 10 μm after such heat treatment.

The sheet metal according to the invention can in particular be obtained by the following process:

- a steel having the previous composition is produced;

- we proceed to the casting of a semi-product from this steel; - the semi-finished product is brought to a temperature greater than or equal to 1150°C and less than or equal to 1260°C for a period of between 40 minutes and 60 minutes, and the semi-finished product is hot rolled to obtain a rolled sheet hot with a thickness between 2.5 mm and 6 mm;

- the hot-rolled sheet is annealed, for example at a temperature between 1000°C and 1100°C for a period of 30 seconds to 6 minutes;

- the hot-rolled and annealed sheet is stripped;

- said hot-rolled sheet is cold rolled, at a temperature between ambient and 300°C, in a single step or in several steps, the sheet being annealed and pickled following each step. It should be understood that, by the term "stage", we designate here cold rolling comprising either a single pass, or a succession of several passes (for example five passes) which are not separated by any intermediate annealing; we can envisage, for example, a cold rolling sequence comprising a first series of five passes, then an intermediate annealing, then a second sequence of five passes; typically, the intermediate anneals separating the stages are carried out between 950°C and 1100°C for 30 seconds to 6 minutes;

- a final annealing of the cold-rolled sheet is carried out, at a temperature between 1000°C and 1100°C, preferably between 1050°C and 1090°C, and for a period of between 10 seconds and 6 minutes, for obtain a completely recrystallized structure with an ASTM average grain size preferably between 4 and 7 if the sheet has a thickness of between 1.2 mm and 2.5 mm and preferably between 4 and 9 if the sheet is thick. thickness greater than or equal to 0.1 mm and less than 1.2 mm. This heat treatment allows the niobium to be put into solid solution.

The volume fraction of Laves phases, ie of Fe ₂ Nb compounds with a hexagonal structure, in the structure of the sheet is very low and less than 0.2% in the delivered state, ie after this final annealing.

We will now describe a series of experiments demonstrating the interest of the invention. Laboratory castings were studied, the chemical analyzes of which are given in Table 1.

(&ÿns) \neafqei

For each of the steels in Table 1, the remainder is iron, as well as inevitable impurities resulting from production.

The cast samples were transformed according to the following process:

- the cast semi-finished products were heated to a temperature of 1220°C for 40 minutes, and hot rolled to obtain a 5 mm thick sheet;

- the sheets were annealed at 1080°C for 6 minutes and pickled;

-the hot-rolled sheets were cold-rolled at room temperature to obtain a sheet with a thickness of 1.5 mm;

- a final annealing was carried out on the sheets at a temperature of 1080°C for 4 minutes.

We determined the volume fraction of Laves phases, ie of Fe ₂ Nb compounds of hexagonal structure, in the structure of each sheet in the delivery state, that is to say at the end of the annealing of the sheet cold rolled. In Table 2, the volume fraction of Laves phases thus determined in the sheet metal in the delivered state is indicated in the column “Phases of Laves Fe ₂ Nb in the delivered state”.

The sheets were then subjected to heat treatment at 850°C for a period of 1000 hours.

Following this heat treatment, the precipitates present in the structure were determined. In table 2 below, in the column “Fe ₂ Nb lava phases after treatment at 850°C for 1000 hours”, the volume fraction of lava phases at the end of this heat treatment.

We _also determined the volume fraction of _{intermetallic} precipitates Fe ₃ Nb ₃ hours” in Table 2.

We then determined for each of the sheets the surface fraction of the segregation of silicon oxides 3 from an image obtained by electron microscope, as described above. The surface fraction thus determined and reported in the column “Surface fraction of silica after treatment at 850°C for 1000 hours” in Table 2.

The resilience of the hot, annealed and pickled strip obtained at the end of hot rolling at 1220°C up to a thickness of 5 mm was also determined by a resilience test on a KCV specimen using the Charpy test following the NF EN ISO 148-1 standard (March 2017 version) during which the energy absorbed by the specimen broken by bending impact as a function of temperature (between -10°C and 80°C). The strip is considered to be ductile if its resilience is greater than 30 J/cm ² for a temperature of 20°C (temperature closest to ambient temperature).

Table 2

In Table 2, the comparative tests are underlined.

We note that, for the tests according to the invention (E1 to E3), in which:

- the niobium and titanium contents comply with the conditions described above, namely:

- Nb between 0.40% and 1.0% and 2xNb-7xC > 0.8%, and

- Ti between 0.05% and 0.2%, and 0% ≤ Ti - 4xN ≤ 0.15%, and

- the Mn/Si ratio is greater than 1.2, and

- the sum Ni+Cu+ Co is between 0 and 0.60%, the volume fraction of Fe ₂ Nb intermetallic precipitates with hexagonal structure is greater than or equal to 0.8% after treatment at 850°C for 1000 hours and the fraction surface fraction of silicon oxide 3 after treatment at 850°C for 1000 hours represents a surface fraction less than or equal to 35%.

Thus, we obtain sheets with good performance in terms of electrical conductivity under conditions of use, and in particular comparable to those of very high chrome steels.

On the contrary, in the case of comparative tests E5 to E8, the titanium and/or niobium contents do not comply with the above conditions. In these tests, the volume fraction of Fe ₂ Nb intermetallic precipitates with hexagonal structure is less than 0.8% after treatment at 850°C for 1000 hours. Silicon is therefore less likely to be trapped by the Laves Fe ₂ Nb phases and to limit silica segregation even when the silicon is less than 0.2% as in the K5 alloy. Furthermore, in these comparative tests, the Mn/Si ratio does not comply with the above condition. We note that, in these tests, the surface fraction of silicon oxide 3 after treatment at 850°C for 1000 hours represents a surface fraction greater than 35%, which leads to degraded performance in terms of electrical conductivity, with a interface resistivity increased by more than 50%.

Furthermore, in tests E1 to E3, the hot strip is ductile, since it has a resilience greater than 30 J/cm ² . On the contrary, in the case of comparative tests E4 and E5, in which the steel has a molybdenum content greater than the limits described in the context of the invention, the hot strip obtained is not ductile, since it presents a resilience of less than 30 J/cm ² .

As explained above, according to an optional aspect:

- the average grain size of the steel is between 30 micrometers and 80 micrometers, that is to say with an ASTM index between 4 and 7, when the thickness of the cold-rolled and annealed sheet is between between 1.2 mm and 2.5 mm; And

- the average grain size of the steel is between 15 micrometers and 80 micrometers, that is to say with an ASTM index between 4 and 9, when the thickness of the cold-rolled and annealed sheet is greater or equal to 0.1 mm and less than 1.2 mm.

In order to confirm the technical effect of this optional property, the inventors subjected sheets having compositions according to the invention to a creep test under their own weight at 850°C for 200 hours.

These sheets were obtained, from hot-rolled and pickled sheets obtained by the process described above, by cold rolling at room temperature to a final thickness, reported in Table 3 for each test, followed by a final annealing under annealing conditions reported in Table 3. The sheets are then pickled and the grain sizes measured by the circular intercept method as described in ASTM E112. Table 3 also specifies the average grain size for each test.

Creep was measured using a creep test called “Sag Test”. The Sag Test is not standardized but is used to characterize creep. This essay is described in the article Faria, Geraldo Lucio de; Melo, Denilson Pereira of; Moreira, Paulo Sérgio. UTILIZAÇÀO DA METODOLOGIA SAG TEST PARA AVALIAR O COMPORT AMENTO EM FLUÊNCIA DOS AÇOS INOXIDÀVEIS AISI 321 E AISI 441, p. 34-44. In: 75° Congresso Anual da ABM, Sâo Paulo, 2022. ISSN: 2594-5327, DOI 10.5151/2594-5327-34135.

For the creep test, metal strips of length 205 mm taken in the rolling direction, width 25 mm and thickness corresponding to the final thickness of the strip (1.5 mm or 0.5 mm depending on the test considered).

These flat strips are then suspended in the oven at 850°C on two 200 mm air gap supports for a determined period. Deflection measurements, which characterize creep deformation, are carried out regularly at 1 hour, 25 hours, 50 hours, 100 hours and 200 hours. The deflection of the specimen is measured at room temperature on a flat surface, typically a marble, using a comparator with a precision and resolution of less than 0.05 mm. For each test, three test pieces were tested.

The average values of the deflection of the test piece after 200 hours of exposure are reported in Table 3 below.

Table 3

These tests demonstrate that a grain size less than 30 μm for sheets with a thickness of 1.5 mm degrades the creep properties. Indeed, a deflection greater than 3 mm is observed after 200 hours at 850°C. Likewise, an average grain size less than 15 m for sheets with a thickness of 0.5 mm degrades the creep properties. Indeed, a deflection greater than 9 mm is observed after 200 hours at 850°C.

The embodiment in which the average grain size complies with the conditions specified above is therefore particularly advantageous in terms of creep resistance.

Claims

1. Ferritic stainless steel sheet, the composition of which includes, the contents being expressed by weight:

C ≤ 0.03% 0.25% ≤ Mn ≤ 1%, preferably 0.3% ≤ Mn ≤ 0.5% 0% ≤ Si ≤ 0.20%, preferably Si ≤ 0.15% with Mn/ If > 1.2

S ≤ 0.005% P ≤ 0.04%

19.0% ≤ Cr ≤ 24.0% Ni ≤ 0.5%

Mo ≤ 0.10% N ≤ 0.03%

Cu ≤ 0.20% 0.40% ≤ Nb ≤ 1.0%

0.05% ≤ Ti ≤ 0.2%, preferably 0.05% ≤ Ti ≤ 0.15%

Zr ≤ 0.02%

Al ≤ 0.02%

V ≤ 0.2% Co ≤ 0.05% Sn ≤ 0.05%,

Rare earths ≤ 800 ppm provided that:

V + Zr +AI ≤ 0.2% Ti + V + Zr +AI ≤ 0.30%

Ti + Nb ≤ 1.0%

Ni + Cu + Co s 0.60% 2xNb-7xC > 0.8% 0% ≤ Ti - 4xN ≤ 0.15% 0.2 ppm ≤ Ca ≤ 20 ppm

1 ppm ≤ O ≤ 60 ppm the rest of the composition being made up of iron and inevitable impurities resulting from the preparation, the sheet being an annealed and pickled sheet, the sheet comprising a volume fraction of phases of Laves Fe ₂ Nb less than 0.2%.

2. Ferritic stainless steel sheet according to claim 1, in which:

50 ppm ≤ Rare earths ≤ 800 ppm.

3. Ferritic stainless steel sheet according to one of claims 1 or 2, in which:

Nb - 10x(C+N) > 0%.

4. Ferritic stainless steel sheet according to any one of claims 1 to 3, characterized in that the sheet has a thickness of between 0.1 mm and 2.5 mm.

5. Ferritic stainless steel sheet according to claim 4, characterized in that the sheet has an average grain size of between 30 micrometers and 80 micrometers when the sheet has a thickness of between 1.2 mm and 2.5 mm and an average grain size of between 15 micrometers and 80 micrometers when the sheet metal has a thickness greater than or equal to 0.1 mm and less than 1.2 mm.

6. Ferritic stainless steel sheet according to any one of claims 1 to 5, characterized in that the sheet is a cold-rolled and annealed sheet.

7. Ferritic stainless steel sheet according to any one of claims 1 to 6, characterized in that, when subjected to heat treatment at a temperature of 850°C for a period of 1000 hours, the sheet comprises a volume fraction of Fe ₂ Nb Laves phases greater than or equal to 0.8%.

8. Ferritic stainless steel sheet according to any one of claims 1 to 7, characterized in that, when subjected to heat treatment at a temperature of 850°C for a period of 1000 hours, the sheet comprises a volume fraction of Fe ₃ Nb ₃ X phases less than 0.05%.

9. Ferritic stainless steel sheet according to any one of claims 1 to 8, characterized in that, when subjected to heat treatment at a temperature of 850°C for a period of 1000 hours, the sheet comprises , on each of its faces, a layer of oxides (2) and, at the interface between the steel of the sheet and the layer of oxides (2), precipitates of silicon oxide (3), such that the surface fraction of the silicon oxide precipitates (3) at the interface between the steel of the sheet and the oxide layer (2) is less than or equal to 0.35.

10. Ferritic stainless steel sheet according to claim 9, characterized in that the oxide layer (2) has a thickness less than or equal to 10 pm.

11. Process for manufacturing a ferritic stainless steel sheet characterized in that:

- a steel is produced having the composition according to any one of claims 1 to 3;

- we proceed to the casting of a semi-product from this steel;

- the hot rolled sheet is annealed,

- the hot-rolled and annealed sheet is stripped,

- said hot-rolled sheet is cold rolled, at a temperature between ambient and 300°C, in a single step or in several steps separated by intermediate annealing;

12. Method according to claim 11, characterized in that the annealing of the hot-rolled sheet is carried out at a temperature between 1000°C and 1100°C, for a period of 30 seconds to 6 minutes.

13. Method according to one of claims 11 or 12, characterized in that the intermediate annealing(s) are carried out at a temperature between 950°C and 1100°C for a period of 30 seconds to 6 minutes.

14. Method according to any one of claims 11 to 13, characterized in that the final annealing is carried out at a temperature between 1050°C and 1090°C.