BACKGROUND OF THE INVENTION Over the recent years, the environments in which corrosion resistant metallic materials were used, became more aggressive, the requirements on the corrosion properties as well as on their mechanical properties increased. Duplex steel alloys, which were established as an alternative to until that used steel grades, as for example high alloyed austenitic steels, nickel-base alloys or other high alloyed steels, are not exempted from this development.

An established measure for the corrosion resistance in chloride-containing environments is the so-called Pitting Resistance Equivalent (abbreviated PRE), which is defined as PRE=% Cr+3,3% Mo+16% N where the percentages for each element allude to weight-percent.

A higher numerical value indicates a better corrosion resistance in particular against pitting corrosion. The essential alloying elements, which affect this property, are according to the formula Cr, Mo, N. An example for such a steel grade is evident from EP0220141, which hereby through this reference is included in this description. This steel grade with the denotation SAF2507 (UNS S32750) was mainly alloyed with high contents of Cr, Mo and N. It is consequently developed against this property with above all good resistance to corrosion in chloride environments.

In recent times also the elements Cu and W have shown to be efficient alloying additions for further optimization of the steel's corrosion properties in chloride environments. The element W has by then been used as substitute for a portion of Mo, as for example in the commercial alloy DP3W (UNS S39274) or Zeron100, which contain 2,0% respectively 0,7% W. The later contains even 0,7% Cu with the purpose to increase the corrosion resistance of the alloy in acid environments.

The alloying addition of tungsten led to a further development of the measure for the corrosion resistance and thereby the PRE-formula to the PREW-formula, which also makes the relationship between the influence of Mo and W on the alloys corrosion resistance clearer : PREW=% Cr+3,3 (% Mo+0,5% W) +16% N, such as described for example in EP 0 545 753. This publication refers to a duplex stainless alloy with generally improved corrosion properties.

The above-described described steel grades have a PRE-number, irrespective method of calculation, which lies above 40.

From the alloys with good corrosion resistance in chloride environments also SAF 2906 shall be mentioned, which composition appears from EP 0 708 845. This alloy, which is characterized by higher contents of Cr and N compared to for example SAF2507, has shown being especially suitable for use in environments, where resistance to intergranular corrosion and corrosion in ammonium carbamate is of importance, but it has also a high corrosion resistance in chloride-containing environments.

US-A-4 985 091 describes an alloy intended for use in hydrochloric and sulfuric acid environments, where mainly intergranular corrosion arises. It is primarily intended as alternative to recently used austenitic steels.

US-A-6 048 413 describes a duplex stainless alloy as alternative to austenitic stainless steels, intended for use in chloride-containing environments.

The disadvantage with the above-described alloys, all with high PRE-numbers, is the appearance of hard and brittle intermetallic precipitations

in the steel, as for example sigma phase, especially after heat-treating, such as for example by welding under later processing. That results in a harder material with poorer workability and finally a deteriorated corrosion resistance.

In order to further improve among others the pitting corrosion resistance of duplex stainless steels, an increase of the PRE-number in both the ferrite phase and the austenite phase is required, without for that sake jeopardizing the structural stability or workability of the material. If the composition in the two phases is not equivalent with regard to the active alloying components, one phase becomes more sensitive for pitting and crevice corrosion. Consequently, the phase which is more sensitive to corrosion decreases the resistance, while the structural stability is decreased by the highest alloyed phase.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a duplex stainless steel alloy, which shows high corrosion resistance in combination with improved mechanical properties and which is most appropriate for use in environments where a high resistance to general corrosion and localized corrosion is required, such as for example in chloride-containing environments.

It is another object of the present invention to provide a duplex stainless steel alloy with a content of ferrite in the range of 40 to 65 volume-% and a PRE-number of at least between 46 and 50 in both the austenite and ferrite phase and with an optimum relationship between PRE austenite and PRE ferrite in the range of 0,90 to 1,15 ; preferably between 0,9 and 1,05.

It is a further object of the present invention to provide a duplex stainless steel alloy with a Critical Pitting Corrosion Temperature (henceforth abbreviated CPT) value higher than 90°C, preferably higher than 95°C and a Critical Crevice-corrosion Temperature (henceforth abbreviated CCT) value of lowest 50°C in 6% Fecal3, preferably at least 60°C in 6% Fecal3.

It is a further object of the present invention to'provide an alloy with impact strength of at least 100 J at room temperature and an elongation after tensile test of at least 25% at room temperature.

For its high alloying content, the material according to the present invention shows remarkably good workability, in particular hotworkability and shall thereby be very suitable to be used for example the production of bars, tubes, such as welded and seamless tubes, plate, strip, wire, welding wire, constructive parts, such as for example pumps, valves, flanges and couplings.

These objects are fulfilled according to the present invention with duplex stainless steel alloys, which contain (in weight-%) up to 0,03% C, up to 0,5% Si, 24,0-30, 0% Cr, 4,9-10, 0% Ni, 3,0-5, 0% Mo, 0,28-0, 5% N, 0-3,0% Mn, 0-0,0030% B, up to 0,010% S, 0-0,03% Al, 0-0,010% Ca, 0-3,0% W, 0-2,0% Cu, 0-3,5% Co, 0-0,3% Ru, balance Fe and inevitable impurities.

SHORT DESCRIPTION OF THE DRAWINGS Figure 1 shows CPT-values from tests of the test heats in the modified ASTM G48C test in"Green Death"-solution compared with the duplex steels SAF2507, SAF 2906 as well as the high alloyed austenitic steel 654SMO.

Figure 2 shows CPT-values attained with the help of the modified ASTM G48C test in"Green Death"-solution for the test heats compared with the duplex steel SAF2507 as well as the austenitic steel 654SMO.

Figure 3 shows the average amount of erosion in mm/year in 2% HCI at a temperature of 75°C.

Figure 4 shows the results from hot ductility testing for most of the heats.

DETAILED DESCRIPTION OF THE INVENTION A systematic development work has surprisingly shown that one by means of a well-balanced combination of the elements Cr, Mo, Ni, N, Mn and Co can obtain optimal dispensation of the elements in the ferrite and austenite, which enables a very corrosion resistant material with only an insignificant amount of sigma phase in the material. The material obtains also good workability, which enables extruding to seamless tubes. It shows that with the intention to obtain a combination of high corrosion resistance in connection with good structural stability a much narrow combination of the alloying elements in

the material is required. The alloy according to the invention contains (in weight- %): C Max 0,03% Si Max 0,5% Mn 0-3,0% Cr 24,0-30, 0% Ni 4,9-10, 0% Mo 3,0-5, 0% N 0,28-0, 5% B 0-0,0030% S max 0,010% Co 0-3,5% W 0-3,0% Cu 0-2,0% Ru 0-0,3% Al 0-0,03% Ca 0-0,010% balance Fe and normal occurring impurities and additions, whereby the content of ferrite is 40-65 volume-%.

Carbon (C) has limited solubility in both ferrite and austenite. The limited solubility implies a risk of precipitation of chromium carbides and the content should therefore not exceed 0,03 weight-%, preferably not exceed 0,02 weight- %.

Silicon (Si) is utilized as desoxidation agent in the steel production as well as it increases the flowability during production and welding. However, too high contents of Si lead to precipitation of unwanted intermetallic phase, wherefore the content is limited to max 0,5 weight-%, preferably max 0,3 weight-%.

Manganese (Mn) is added in order to increase the N-solubility in the material.

However, it has shown that Mn only has a limited influence on the N-solubility in the type of alloy in question. Instead there are found other elements with higher influence on the solubility. Besides, Mn in combination with high contents of sulfur can give rise to formation of manganese sulfides, which act as initiation- points for pitting corrosion. The content of Mn should therefore be limited to between 0-3,0 weight-%, preferably 0,5-1, 2 weight-%.

Chromium (Cr) is a much active element in order to improve the resistance to a majority of corrosion types. Furthermore, a high content of chromium implies that one gets a very good N-solubility in the material. Thus, it is desirable to keep the Cr-content as high as possible in order to improve the corrosion resistance. For very good amounts of corrosion resistance the content of chromium should be at least 24,0 weight-%, preferably 27,0-29, 0 weight-%.

However, high contents of Cr increase the risk for intermetallic precipitations, for what reason the content of chromium must be limited up to max 30,0 weight-%.

Nickel (Ni) is used as austenite stabilizing element and is added in suitable contents in order to obtain the desired content of ferrite. In order to obtain the desired relationship between the austenitic and the ferritic phase with between 40-65 volume-% ferrite, an addition of between 4,9-10, 0 weight-% nickel, preferably 4,9-8, 0 weight-%, is required.

Molybdenum (Mo) is an active element which improves the resistance to corrosion in chloride environments as well as preferably in reducing acids. A too high Mo-content in combination with that the Cr-contents are high, implies that the risk for intermetallic precipitations increases. The Mo-content in the present invention should lie in the range of 3,0-5, 0 weight-%, preferably 3,6-4, 7 weight- %, in particular 4,0-4, 3 weight-%.

Nitrogen (N) is a very active element, which increases the corrosion resistance, the structural stability as well as the strength of the material. Further, a high N-

content improves the recovering of the austenite after welding, which gives good properties within the welded joint. In order to obtain a good effect of N, at least 0,28 weight-% N should be added. At high contents of N the risk for precipitation of chromium nitrides increases, especially when simultaneously the chromium content is high. Further, a high N-content implies that the risk for porosity increases because of the exceeded solubility of N in the smelt For these reasons the N-content should be limited to max 0,5 weight-%, preferably >0,35-0, 45 weight-% N is added.

Boron (B) is added in order to increase the hotworkability of the material. At a too high content of Boron the weldability as well as the corrosion resistance could deteriorate. Therefore, the content of boron should be limited to 0,0030 weight-%.

Sulfur (S) influences the corrosion resistance negatively by forming soluble sulfides. Further, the hot workability detoriates, for what reason the content of sulfur is limited to max 0,010 weight-%.

Cobalt (Co) is added in order to improve foremost the structural stability as well as the corrosion resistance. Co is an austenite-stabilizing element. In order to obtain effect should at least 0,5 weight-%, preferably at least 1,5 weight-% be added. Because cobalt is a relatively expensive element, the addition of cobalt is therefor limited to max 3,5 weight-%.

Tungsten increases the resistance to pitting-and crevice corrosion. But the addition of too high contents of tungsten in combination with that the Cr- contents as well as Mo-contents are high, means that the risk for intermetallic precipitations increases. The W-content in the present invention should lie in the range of 0-3,0 weight-%, preferably between 0,5 and 1,8 weight-%.

Copper is added in order to improve the general corrosion resistance in acid environments such as sulfuric acid. At the same time Cu influences the

structural stability. However, high contents of Cu imply that the solid solubility will be exceeded. Therefor the Cu-content should be limited to max 2,0 weight- %, preferably between 0,5 and 1,5 weight-%.

Ruthenium (Ru) is added in order to increase the corrosion resistance. Because ruthenium is a very expensive element, the content should be limited to max 0,3 weight-%, preferably more than 0 and up to 0,1 weight-%.

Aluminum (Al) and Calcium (Ca) are used as desoxidation agents at the steel production. The content of Al should be limited to max 0,03 weight-% in order to limit the forming of nitrides. Ca has a favorable effect on the hotductility.

However, the Ca-content should be limited to 0,010 weight-% in order to avoid an unwanted amount of slag.

The content of ferrite is important in order to obtain good mechanical properties and corrosion properties as well as good weldability. From a corrosion point of view and a point of view of weldability a content of ferrite between 40-65% is desirable in order to obtain good properties. Further, high contents of ferrite imply that the impact strength at low temperatures as well as the resistance to hydrogen-induced brittleness risks deteriorating. The content of ferrite is therefore 40-65 volume-%, preferably 42-60 volume-%, in particular 45-55 volume-%.

DESCRIPTION OF PREFERED EMBODIEMENTS In the examples below the composition of a number of test heats is presented, which illustrate the effect of different alloying elements on the properties. Heat 605182 represents a reference composition and is consequently not a part of the field of this invention. Neither shall the remaining heats be considered limiting the invention, without only specifying examples of heats, which illustrate the invention according to the claims.

The specified PRE-numbers or-values consider always amounts calculated according to the PREW-formula, even though this is not explicitly mentioned.

EXAMPLE 1 The test heats according to this example were produced by casting of 170kg ingots in the laboratory, which were hotforged to round bars. Those were hotextruded to bars (round bars as well as flat bars), where test material was taken out from the round bars. Further on the flat bars were annealed before cold rolling took place, whereafter further test material was taken out.

From a material engineering point of view, the process can be considered to be representative for the preparation in bigger scale, for example for the production of seamless tubes by the extrusion method, followed by cold rolling. Table 1 shows the composition of the first batch of test heats.

Table 1. Composition for test heats, weight-%. Heat Mn Cr Ni Mo W Co V La Ti N 605193 1,03 27,90 8,80 4,00 0,01 0, 02 0,04 0,01 0,01 0,36 605195 0,97 27,90 9, 80 4,00 0,01 0,97 0,55 0,01 0,35 0,48 605197 1,07 28,40 8,00 4,00 1,00 1, 01 0,04 0,01 0,01 0,44 605178 0,91 27,94 7,26 4,01 0,99 0,10 0,07 0,01 0,03 0,44 605183 1,02 28,71 6,49 4,03 0,01 1,00 0,04 0,01 0,04 0,28 605184 0,99 28,09 7,83 4,01 0,01 0,03 0,54 0,01 0,01 0,44 605187 2,94 27,74 4,93 3,98 0,01 0,98 0,06 0,01 0,01 0,44 605153 2,78 27,85 6,93 4,03 1,01 0,02 0,06 0,02 0,01 0,34 605182 0, 17 23,48 7,88 5,75 0, 09 0,05 0,04 0, 01 0, 10 0,26 In purpose to investigate the structural stability samples from every heat were annealed at 900-1150°C with steps of 50°C as well as they were quenched in air, respective water. At the lowest temperatures intermetallic phase was formed. The lowest temperature, where the amount of intermetallic phase became insignificant, was determined with the help of studies in light optical microscope. New samples from respective heat were annealed afterwards at said temperature under five minutes, whereafter the samples were cooled down with the constant cooling rate of-140°c/min to room temperature.

Subsequently, the area fraction of sigma phase in the materials was determined with digital scanning of the pictures with back-scattering electrons in a scanning electron microscope. The results appear from Table 2.

Tmax sigma was calculated with Thermo-Calc (TC version N thermodynamic database for steel TCFE99) based on characteristic amounts for all specified elements in the different variations. Tmax sigma is the dissolving temperature for the sigma phase, where high dissolving temperatures indicate lower structural stability.

Table 2. Heat Heat treatment Amount a [vol-%] Tmax 6 605193 1100°C, 5min 7, 5% 1016 605195 1150°C, 5min 32% 1047 605197 1100°C, 5min 18% 1061 605178 1100°C, 5min 14% 1038 605183 1050°C, 5min 0,4% 997 605184 1100°C, 5min 0,4% 999 605187 1050°C, 5min 0, 3% 962 605153 1100°C, 5min 3,5% 1032 605182 1100°C, 5min 2,0% 1028 The purpose of this investigation is to be able to rank the material with regard to the structural stability, i. e. this is not the real content of sigma phase in the samples, which were heat treated and quenched before for example the corrosion testing. One can see that Tmax sigma, which was calculated with Thermo-Calc does not directly coincide with the measured amounts of sigma phase, however it is distinct that the test heats with the lowest calculated Tmax sigma contain the lowest amount sigma phase during this investigation.

The pitting corrosion properties of all heats were tested for ranking in the so-called"Green Death"-solution, which consists of 1 % FeCl3, 1 % CuCI2, 11% H2S04, 1,2% HCI. The test procedure is equivalent to the pitting corrosion testing according to ASTM G48C, however, it will be carried out in the more aggressive"Green Death"-solution. Further, some of the heats were tested according to ASTMG48C (2 tests per heat). Also the electrochemical testing in 3% NaCI (6 tests per heat) was carried out. The results in form of the Critical Pitting Temperature (CPT) from all tests appear from Table 3, such as the PREW-number (Cr+3,3 (Mo+0,5W) +16N) for the total composition of the alloy as well as for austenite and ferrite. The indexing alpha refers to the ferrite and gamma refers to the austenite.

Table 3.

Heat PRE PRE PREy/PRE CPT°C CPT°C CPT°C PRE Modified ASTM G48 C 3% NaCI ASTM G48C 6% FeCI3 Green death 605193 51,3 49,0 0,9552 46,9 90/90 64 605195 51,5 48,9 0,9495 48,7 90/90 95 605197 53, 3 53,7 1, 0075 50, 3 90/90 >95 >95 605178 50,7 52,5 1,0355 49,8 75/80 94 605183 48,9 48,9 1, 0000 46,5 85/85 90 93 605184 48,9 51, 7 1, 0573 48,3 80/80 72 605187 48, 0 54, 4 1, 1333 48, 0 70/75 77 605182 54,4 46,2 0,8493 46,6 75/70 85 62 654SMO 90/85 SAF2507 70/70 SAF2609 60/50 It is established that there exists a linear ratio between the lowest PRE-number in the austenite or ferrite and the CPT-value in the duplex steel, but the results in Table 3 show that the PRE-number not solely explains the CPT-values. In Figure 1 the CPT-values from test in the modified ASTM G48C test are shown diagrammatically. The duplex steels SAF2507, SAF2906 as well as the high alloyed austenitic steel 654SMO are included as reference. It is distinct from these results that all test materials show better CPT in the modified ASTM G48C than SAF2507 as well as SAF2906. Furthermore some of the test materials show CPT results in the modified ASTM G48C at the same level as or in excess of 654SMO. The test heat 605183, alloyed with cobalt shows good structural stability at a controlled cooling rate of (-140°C/min) in spite that it contains high contents of chromium as well as of molybdenum, shows better results than SAF2507 and SAF2906. It appears from this investigation that a high PRE does not solely explain the CPT values, without the relationship PRE austenit/PRE ferrite is of extreme weight for the properties of the higher alloyed duplex steels, and a very narrow and exact leveling between the alloying elements is required in order to obtain this optimum ratio, which lies between 0,9-1, 15; preferably 0,9-1, 05 and simultaneously obtain PRE values of above

46. The relationship PRE austenit/PRE ferrite against CPT in the modified ASTM G48C test for the test heats is given in Table 3.

The strength at room temperature (RT), 100°C and 200°C and the impact strength at room temperature (RT) have been determined for all heats and is shown as average amount for three tests.

Tensile test specimen (DR-5C50) were manufactured from extruded bars, 0 20mm, which were heattreated at temperatures according to Table 2 in 20 minutes followed by cooling down in either air or water (605195, 605197,605184). The results of the tests are presented in Table 4 and 5. The results of the tensile test show that the contents of chromium, nitrogen and tungsten strongly influence the impact strength of the material. Besides 605153, all heats fulfill the requirement of a 25% elongation at tensile testing at room temperature (RT).

Table 4. Impact strength Heat Temperature Rpo2 RP1 O Rm |A5 IZ (MPa) (MPa) (MPa) (%) (%) 605193 RT 652 791 916 29,7 38 100°C 513 646 818 30, 4 36 200°C 511 583 756 29,8 36 605195 RT 671 773 910 38, 0 66 100°C 563 637 825 39,3 68 200°C 504 563 769 38,1 64 605197 RT 701 799 939 38,4 66 100°C 564 652 844 40,7 69 200°C 502 577 802 35,0 65 605178 RT 712 828 925 27,0 37 100°C 596 677 829 31,9 45 200°C 535 608 763 27, 1 36 605183 RT 677 775 882 32, 4 67 100°C 560 642 788 33, 0 59 200°C 499 578 737 29,9 52 605184 RT 702 793 915 32,5 60 100°C 569 657 821 34,5 61 200°C 526 581 774 31, 6 56 605187 RT 679 777 893 35, 7 61 100°C 513 628 799 38,9 64 200°C 505 558 743 35,8 58 605153 RT 715 845 917 20,7 24 100°C 572 692 817 29, 3 27 200°C 532 611 749 23, 7 31 605182 RT 627 754 903 28,4 43 100°C 493 621 802 31,8 42 Table 5. Impact Strength Heat Annealing Cooling Impact Annealing Cooling Impact [°C/min] strength [°C/min] strength J J 605193 1100/20 Air 35 1100/20 Water 242 605195 1150/20 Water 223 605197 1100/20 Water 254 1130/20 Water 259 605178 1100/20 Air 62 1100/20 Water 234 605183 1050/20 Air 79 1050/20 Water 244 605184 1100/20 Water 81 1100/20 Air 78 605187 1050/20 Air 51 1100/20 Water 95 605153 1100/20 Air 50 1100/20 Water 246 605182 1100/20 Air 22 1100/20 Water 324

This investigation shows very distinct that water quenching is certainly necessary in order to obtain the best structure and consequently good values for the impact strength. The requirement is 100J at test at room temperature and all heats pass this, except heat 605184 and 605187, where certainly the latter lies very near the requirement.

Table 6 shows the results from the Tungsten-Inert-Gas remelting test (henceforth-abbreviated TIG), where the heats 605193,605183, 605184 as well as 605253 show a good structure in the heat affected zone (Heat Affected Zone, henceforth-abbreviated HAZ). The Ti-containing heats show Tin in HAZ.

A too high chromium-and nitrogen content results in precipitation of Cr2 N, which shall be avoided because it detoriates the properties of the material.

Table 6. Heat Precipitations Protective gas Ar (99, 99%) 605193 HAZ: OK 605195 HAZ: Large amounts of TIN and 6-phase 605197 HAZ: Small amounts of Cr2N in 8-grains, but not much 605178 HAZ: Cr2N in 8-grains, otherwise OK 605183 HAZ: OK 605184 HAZ: OK 605187 HAZ: CrN quite near the meltingbond, no precipitations farther out 605153 HAZ: OK 605182 HAZ: TiN and decorated grainboundaries 8/8

EXAMPLE 2 In the below-mentioned example the composition of a further number of test heats produced with the purpose to find the optimum composition is given. These heats are modified starting out from the properties of the heats with good structural stability as well as high corrosion resistance, from the results, which were shown in example 1. All heats in Table 7 are included in the composition according to the present invention, where heats 1-8 are included into a statistical test model, while the heats e to n are additional test alloys within the scope of this invention.

A number of test heats were produced by casting of 270kg ingots, which were hotforged to round bars. Those were extruded to bars, wherefrom test samples were taken. Afterwards the bar was annealed before cold rolling to flat bars was executed, after that further test material was taken out. Table 7 shows the composition for these test heats.

Table 7. Heat Mn Cr Ni Mo W Co Cu Ru B N 1 605258 1,1 29,0 6,5 4, 23 1, 5 0, 0018 0,46 2 605249 1,0 28,8 7,0 4, 23 1, 5 0, 0026 0,38 3 605259 1,1 29,0 6,8 4, 23 0, 6 0, 0019 0,45 4 605260 1,1 27,5 5, 9 4, 22 1, 5 0, 0020 0,44 5 605250 1,1 28,8 7,6 4, 24 0, 6 0, 0019 0,40 6 605251 1,0 28,1 6,5 4, 24 1, 5 0, 0021 0,38 7 605261 1,0 27,8 6,1 4, 22 0, 6 0, 0021 0,43 8 605252 1,1 28, 4 6, 9 4, 23 0, 5 0, 0018 0,37 e 605254 1,1 26, 9 6, 5 4, 8 1, 0 0, 0021 0,38 f 605255 1,0 28,6 6, 5 4, 0 3, 0 0, 0020 0,31 g605262 27 27,6 6, 9 3,9 1,0 1, 0 0, 0019 0,36 h 605263 1,0 28,7 6, 6 4,0 1, 0 1, 0 0, 0020 0,40 in 605253 1,0 28, 8 7, 0 4, 16 1, 5 0, 0019 0,37 i 605266 1,1 30,0 7, 1 4, 02 0, 0018 0,38 k 605269 1, 0 28,5 7,0 3,97 1,0 1, 0 0, 0020 0,45 1 605268 1,1 28,2 6, 6 4,0 1, 0 1, 0 1, 0 0, 0021 0,43 m 605270 1,0 28,8 7, 0 4, 21, 50, 1 0,0021 0, 41 n 605267 1,1 29,3 6, 5 4, 23 1, 5 0, 0019 0,38 Table 8. Thermo-Calc Variant a-formula a T-C PRE total PRE a PRE y Tmax Tmax empirical sigma Cr2N 1 46 50 50,2 47, 8 50, 5 1006 1123 2 52 50 49,1 48,4 49,8 1019 1084 3 45 50 50,2 47,9 52,6 1007 1097 4 46 50 49,2 46,5 49,8 986 1121 5 47 50 49,1 48,5 49,7 1028 1038 6 52 50 48,1 47,1 49,2 998 1086 7 44 50 49,2 46,6 52,0 985 1081 8 46 50 48,1 47,2 49, 1 1008 1044 e 46 53 49,3 48, 4 49,5 1010 1099 f 65 52 46,7 47,2 46, 1 1008 1090 9 48 51 48, 4 48, 4 48,3 1039 979 h 50 53 50,0 48,4 51,7 1035 1087 i 52 50 49, 1 48, 4 49, 8 1019 1084

Thermo-Calc-values according to Table 8 (T-C version N thermodynamic database for steel TCFE99) are based on characteristic amounts for all specified elements in the different variations. The PRE-number for the ferrite and austenite is based on their equilibrium composition at 1100°C.

Tmax sigma is the dissolving temperature for the sigma phase, where high dissolving temperatures indicate lower structural stability.

The distribution of the alloying elements in the ferrite-and austenite phase was examined with microprobe analysis, the results appear from Table 9.

Table 9. Heat Phase Cr Mn Ni Mo W Co Cu N 605258 Ferrite 29, 8 1, 3 4,8 5, 0 1,4 0, 11 Austenite 28, 3 1, 4 7,3 3, 4 1,5 0, 60 605249 Ferrite 29,8 1, 1 5,4 5, 1 1,3 0, 10 Austenite 27, 3 1, 2 7,9 3, 3 1, 6 0, 53 605259 Ferrite 29,7 1, 3 5,3 5, 3 0,5 0,10 Austenite 28,1 1, 4 7,8 3, 3 0,58 0, 59 605260 Ferrite 28, 4 1, 3 4,4 5, 0 1, 4 0,08 Austenite 26,5 1, 4 6,3 3, 6 1, 5 0,54 605250 Ferrite 30,1 1, 3 5,6 5, 1 0, 46 0,07 Austenite 27,3 1, 4 8, 8 3, 4 0, 53 0,52 605251 Ferrite 29, 6 1, 2 5,0 5, 2 1, 3 0,08 Austenite 26, 9 1, 3 7, 6 3,5 1,5 0, 53 605261 Ferrite 28, 0 1, 2 4, 5 4, 9 0, 45 0, 07 Austenite 26,5 1,4 6,9 3, 3 0, 56 0, 56 605252 Ferrite 29,6 1, 3 5,3 5, 2 0, 42 0, 09 Austenite 271 1,4 8,2 3, 3 0, 51 0, 48 605254 Ferrite 28, 1 1,3 4,9 5, 8 0, 89 0, 08 Austenite 26,0 1,4 7,6 3, 8 1, 0 0, 48 605255 Ferrite 30,1 1,3 5,0 4, 7 2,7 0, 08 Austenite 27,0 1,3 7,7 3,0 3,3 0,45 605262 Ferrite 28,8 3,0 5,3 4,8 1,4 0,9 0,08 Austenite 26,3 3,2 8,1 3,0 0,85 1, 1 0, 46 605263 Ferrite 29,7 1,3 5,1 5,1 1,3 0, 91 0, 07 Austenite 27,8 1,4 7,7 3,2 0,79 1, 1 0, 51 605253 Ferrite 30,2 1,3 5,4 5, 0 1, 3 0, 09 Austenite 27, 5 1,4 8,4 3, 1 1, 5 0, 48 605266 Ferrite 31,0 1,4 5,7 4, 8 0, 09 Austenite 29,0 1,5 8,4 3, 1 0, 52 605269 Ferrite 28, 7 1,3 5,2 5,1 1,4 0, 9 0, 11 Austenite 26,6 1,4 7,8 3,2 0,87 1, 1 0, 52 605268 Ferrite 29,1 1,3 5,0 4,7 1,3 0,91 0,84 0,12 Austenite 26,7 1,4 7,5 3,2 0,97 1,0 1,2 0,51 605270 Ferrite 30,2 1,2 5,3 5,0 1,3 0,11 Austenite 27,7 1,3 8,0 3, 2 1, 4 0, 47 605267 Ferrite 30,1 1,3 5,1 4, 9 1, 3 0,08 Austenite 27,8 1,4 7,6 3, 1 1, 8 0,46

The pitting corrosion properties of all heats have been tested in the "Green Death"-solution (1 % FeCl3, 1 % CuCl2, 11 % H2S04, 1,2% HCI) for ranking. The test procedures are the same as pitting corrosion testing according to ASTM G48C, but the testing will be executed in a more aggressive solution than 6% FeCI3, the so-called"Green Death"-solution. Also the general corrosion testing in 2% HCI (2 tests per heat) was executed for ranking before the dewpoint testing. The results from all tests appear from Table 10, Figure 2 and Figure 3. All tested heats perform better than SAF2507 in"Green Death"- solution. All heats lie within the identified range of 0,9-1, 15; preferably 0,9-1, 05 applicable for the ratio PRE austenit/PRE ferrite at the same time as PRE in both austenite and ferrite is in excess of 44 and for most of the heats even considerable in excess of 44. Some of the heats attain even the limit of total PRE 50. It is very interesting to note that heat 605251, alloyed with 1,5 weight- % cobalt, performs almost equivalent with heat 605250, alloyed with 0,6 weight- % cobalt, in"Green Death"-solution in spite of the lower chromium content in heat 605251. It is particularly surprising and interesting because heat 605251

has a PRE-number of ca. 48, which is in excess of some of today's commercial superduplex alloys simultaneously as the Tmax sigma-value below 1010°C indicates a good structural stability based on the values in Table 2 in Example 1.

In Table 10 even PREW-number (% Cr+3,3% (Mo+0,5% W) +16% N) for the total composition of the alloy and PRE in austenite as well as in the ferrite (rounded off) based on composition of the phases are specified as measured with micro probe. Content of ferrite was measured after heat-treating at 1100°C followed by water quenching.

Table 10 Heat a-halt PREW PRE PRE PREy/CPT °C Total a y PREa Green death 605258 48, 2 50,3 48,1 49,1 1,021 605249 59, 8 48, 9 48, 3 46, 6 0,967 75/80 605259 49, 2 50,2 48,8 48,4 0,991 605260 53, 4 48,5 46,1 47,0 1,019 605250 53, 6 49,2 48,1 46,8 0,974 95/80 605251 54, 2 48,2 48,1 46,9 0, 976 90/80 605261 50, 8 48,6 45,2 46,3 1, 024 605252 56, 6 48,2 48,2 45,6 0,946 80/75 605254 53, 2 48,8 48,5 46,2 0,953 90/75 605255 57, 4 46,9 46, 9 44,1 0,940 90/80 605262 57, 2 47,9 48,3 45,0 0,931 605263 53, 6 49,7 49, 8 47, 8 0,959 605253 52, 6 48,4 48,2 45,4 0,942 85/75 605266 62,6 49,4 48,3 47,6 0,986 605269 52,8 50,5 49,6 46,9 0,945 605268 52, 0 49,9 48,7 47,0 0,965 605270 57, 0 49,2 48,5 45,7 0,944 605267 59, 8 49,3 47,6 45,4 0,953 In order to examine the structural stability in detail the samples were annealed for 20 minutes at 1080°C, 1100°C and 1150°C, whereafter they were quenched in water. The temperature, where the amount of intermetallic phase became insignificant was determined with help of investigations in a lightoptical microscope. A comparison of the structure of the heats after annealing at 1080°C followed by water quenching indicates which of the heats are more suspect to contain undesired sigma phase. The results are shown in Table 11. Control of the structure shows that the heats 605249,605251,

605252,605253, 605254,605255, 605259,605260, 605266 as well as 605267 are free from unwanted sigma phase. Moreover, heat 605249, alloyed with 1,5 weight-% cobalt, is free from sigma phase, while heat 605250, alloyed with 0,6 weight-% cobalt, contains a very small amount of sigma phase. Both heats are alloyed with high contents of chromium, approximately 29,0 weight-% and the molybdenum content of approximately 4,25 weight-%. If one compares the compositions of the heats 605249,605250, 605251 and 605252 with thought on the content of sigma phase, it is very distinct that the range of composition for that optimum material is very narrow, in this case with regard to the structural stability. It further shows that the heat 605268 contains only sigma phase compared to heat 605263, which contains much sigma phase. What mainly distinguishes these heats from each other is the addition of copper to heat 605268. Heat 605266 and also 605267 are free from sigma phase, despite of a high content of chromium the later heat is alloyed with copper. Further, the heats 605262 and 605263 with addition of 1,0 weight-% tungsten show a structure with much sigma phase, while it is interesting to note that heat 605269, also with 1,0 weight-% tungsten but with higher content of nitrogen than 605262 and 605263 shows a considerable smaller amount of sigma phase. Consequently, a very well-leveled balance between the different alloying elements at these high alloying contents is required of for example chromium and molybdenum in order to obtain good structural properties.

Table 11 shows the results from the light optical examination after annealing at 1080°C, 20min followed by water quenching. The amount of sigma phase is specified with values from 1 to 5, where 1 represents that no sigma phase was detected in the examination, while 5 represents that a very high content of sigma phase was detected in the examination.

Table 11. Heat Sigma Cr Mo W Co Cu N Ru phase 605249 1 28 8 4, 23 1, 5 0 38 605250 2 28, 8 4, 24 0, 6 0, 40 605251 1 28,1 4, 24 1, 5 0, 38 605252 1 28,4 4, 23 0, 5 0, 37 605253 1 28,8 4, 16 1, 5 0, 37 605254 1 26,9 4, 80 1, 0 0, 38 605255 1 28, 6 4, 04 3, 0 0, 31 605258 2 29,0 4, 23 1, 5 0, 46 605259 1 29,0 4, 23 0, 6 0, 45 605260 1 27,5 4, 22 1, 5 0, 44 605261 2 27,8 4, 22 0, 6 0, 43 605262 4 27,6 3,93 1,0 1, 0 0, 36 605263 5 28,7 3,96 1,0 1, 0 0, 40 605266 1 30, 0 4, 02 0, 38 605267 1 29, 3 4, 23 1, 5 0, 38 605268 2 28,2 3,98 1,0 1,0 1,0 0,43 605269 3 28,5 3,97 1,0 1, 0 0, 45 605270 3 28,8 4, 19 1, 5 0, 41

In Table 12 the results from the impact strength testing of some of the heats are shown. The results are very good, which indicates a good structure after annealing at 1100°C followed by water quenching and the requirement of 100J will be managed with large margin of all tested heats.

Table 12. Heat Annealing Quenching Impact Impact Impact [°C/min] strength strength strength J J J 605249 1100/20 Water >300 >300 >300 605250 1100/20 Water >300 >300 >300 605251 1100/20 Water >300 >300 >300 605252 1100/20 Water >300 >300 >300 605253 1100/20 Water 258 267 257 605254 1100/20 Water >300 >300 >300 605255 1100/20 Water >300 >300 >300

Figure 4 shows the results from the hot ductility testing of the most of the heats. A good workability is of course of vital importance in order to be able to produce the material to product forms such as bars, tubes, such as welded and seamless tubes, plate, strip, wire, welding wire, constructive elements, such as for example pumps, valves, flanges and couplings. The heats 605249,605250, 605251,605252, 605255,605266 as well as 605267,

the most with nitrogen content around 0,38 weight-% show somewhat improved hot ductility values.

Summary of the test results In order to obtain good corrosion properties, simultaneously as the material shows good structural stability, hotworkability and weldability the material should be optimized according to the following : PRE-number in ferrite should exceed 45, but preferably be at least 47.

PRE-number in austenit should exceed 45, but preferably be at least 47.

PRE-number for the entire alloy should preferably be at least 46.

Relationship PRE austenit/PRE ferrite should lie in the range of 0,9-1, 15; preferably in the range 0,9-1, 05.

The content of ferrite should lie in the range preferably 45-55 volume-%.

Tmax sigma should not exceed 1010°C.

The content of nitrogen should lie in the range 0,28-0, 5 weight-%, preferably in the range 0,35-0, 48 weight-%, but preferably 0,38-0, 40 weight-%.

The content of cobalt should lie in the range 0-3,5 weight-%, preferably 1,0-2, 0 weight-%, but preferably 1,3-1, 7 weight-%.

In order to ensure the high nitrogen solubility, i. e. if the content of nitrogen is in the range 0,38-0, 40 weight-% should at least 29 weight-% Cr be added as well as at least 3,0 weight-% Mo, thus the total content of the elements Cr, Mo and N fulfills said requirements on the PRE-number.