Ferritic-austenitic stainless steel and use of the steel .
The present invenition relates to a ferritic- austenitic steel with a high content of Cr and N and a low content of Ni, primarily intended for high-pressure components in plants for the synthesis of urea. However, it is also suited for other purposes where where good corrosion resistance or high strength is required. Duplex stainless steels are steels being characterized by a ferritic-austenitic structure, where the two phases have different compositions. Modern duplex stainless steels are mainly alloyed with Cr, Mo, Ni and N. The duplex structure implies that Cr and Mo will be enriched in the ferrite and Ni and N in the austenite. Most of the modern duplex steels contain 22- 27% Cr, 4-7% Ni, 0-4% Mo and 0,1-0,3% N. This gives the materials a duplex structure with 30-70% ferrite and the rest austenite. Other elements, such as Mn, Cu, Si and , also occur in order to give the alloys special properties.
The duplex stainless steels are often used as alternatives to austenitic stainless steels, but to a lower price, inter alia because of the Ni content in duplex stainless steels. Thus, it is often possible to find a duplex stainless steel with a corrosion resistance corresponding to the austenites. An example o of this is the austenitic 254 SMO (UNS S 31254) , with 20% Cr, 18% Ni, 6% Mo and 0,2% N, which has a corrosion resistance in chloride-containing environments of the same level as the duplex steel SAF 2507 (UNS S 32750) , with 25% Cr, 7% Ni, 4% Mo and 0,3% N.
However, some austenitic steels, such as Sandvik 2RE69, which was developed specially for the
use in urea processes, with 25% Cr, 22% Ni and 2% Mo, has so far lacked an adequate correspondence among the duplex stainless steels, for the use in urea processes. This problem has been solved by the present invention. The alloying levels in duplex stainless steels are restricted upwards by the structure stability. The ferritic-austenitic structure implies that the material is sensitive to embrittlement at 475°C and separation of intermetallic phase in the temperature range 600- 1000°C. Separation of intermetallic phase is enhanced primarily by high contents of Cr and Mo, but it can be suppressed by the inclusion of N. The effect of N on structure stability means that higher contents of Cr may be alloyed into the material without any deterioration of the structure stability. However, the N-content is limited upwards by its solubility in the melt, which gives rise to porosity at too high percentages, and by the solid solubility in the alloy, which may cause nitride precipitation. In order to increase the solubility of N in the melt, the Mn and Cr contents may be increased. However, Mn increases the risk for separation of intermetallic phase, wherefore Mn should be restricted. Since N is a strong austenite promotor, the Ni content can be lowered considerably by an increased N content and still maintaining a ferritic-austenitic structure.
Plants for the synthesis of urea constitute an interesting application for austenitic and duplex stainless steels. Urea is produced by a synthesis of ammonia and carbon dioxide under high pressure and high temperature. The process solution in the high-pressure part is very corrosive towards carbon steels. Therefore, special steels are used to a large extent, but also titanium and zirconium are used. However, the
latter are very costly in purchase and manufacture, thus restricting their use.
Austenitic stainless steels dominate today as material in the high-pressure part of the urea process . A frequently occurring steel is Sandvik 3R60R U.G., which is a modified AISI 316L (UNS S 31603), characterized by 18% Cr, 14% Ni and 2,7% Mo and a carefully controlled ferrite content . In the most demanding applications, steels of the type 25% Cr - 22% Ni - 2% Mo (UNS S 31050) are used. A requirement for the use of stainless steels is that the passivity of the steel can be upheld. Therefore, oxygen is added to the process solution in the urea synthesis. Thus, this addition is only necessary because of a material- technical point of view, while however simultaneously causing energy and yield losses, as well as it is a potential safety risk at too high contents. Therefore, out of a process-technical point of view, there is a desire to reduce the addition of oxygen, if possible completely eliminating it. However, in today's processes it is difficult to guarantee that the required amount of oxygen be present in the process solution. This is the case at, e.g., the boiling of the solution, which takes place in the stripper, this being the most critical heat exchanger. Some corrosion also occurs on steels of the type Cr25-Ni22-Mo2 (UNS S 31050) under certain conditions. Corrosion on AISI 316L (UNS S 31603) mainly takes place under condensing conditions. Thus, adequate passivity cannot be upheld in all parts of the process.
Inferior material quality also causes corrosion in the urea process, which results in attacks in connection to weldings. Inhomogenous material is another reason for corrosion. These factors show that
good structure .stability is a prerequisite for good corrosion endurance in the urea solution, or at other applications where good corrosion endurance is required. In relation with the composition of urea steels, it is well known that Cr has a beneficial influence on the corrosion resistance. A number of investigations has also shown that Ni in austenitic steels is detrimental under conditions when low contents of oxygen occur in the process solution. This results in a pronounced increase of the corrosion speed with increasing content of Ni in the steel . On the other hand, ferritic steels containing low contents of Ni have a very low corrosion under these conditions. However, the ferritic steels have large limitations as a construction material because of bad structure stability, which results in problems in connection with welding and manufacturing.
Ferritic-austenitic stainless steels are very interesting in view of several aspects, primarily as a material in the urea process. The high strength of these steels can be well used in the high-pressure part, and the moderate nickel content gives this steel type a better resistance to corrosion under oxygen-free conditions. Thus, a ferritic-austenitic steel should have a high Cr content and a low Ni content in order to have a good resistance in a urea environment at oxygen- free conditions.
The present invention relates to a duplex ferritic-austenitic steel with a high content of Cr and N and a low content of Ni, primarily intended for high pressure parts in the urea synthesis, but which may also find applications in other environments where high-alloy austenitic stainless steels are used for
attaining corrosion resistance. Particularly good properties have been obtained in the following applications:
- stripper tubes in urea plants - ferrules in urea plants
- condensation tubes in urea plants
- welded-on materials in urea plants
- high pressure tubes in urea plants
- nitric acid process (e.g., in the cooler-condenser) - paper and pulp industry (e.g., in white liquor environment)
- welding wire
Other suitable applications for the steel alloy according to the invention are as manufacturing material for seamless tubes, welded tubes, flanges, couplings and sheet-metal.
The alloy is characterized by good structure stability, which has been attained at the high Cr content by the inclusion of high contents of N. Underneath follows a presentation of the appending diagrams.
Diagram 1 shows the influence of Cr on the corrosion according to the Huey test.
Diagram 2 shows the influence of Mn on the corrosion according to the Huey test.
Diagram 3 shows the influence of Mo on the corrosion according to the Huey test.
Diagram 4 shows the influence of N on the corrosion according to the Huey test. Diagram 5 shows the influence of Cr on the corrosion according to the Streicher test.
Diagram 6 shows the influence of Mn on the corrosion according to the Streicher test.
Diagram 7 shows the influence of Mo on the
corrosion according to the Streicher test .
Diagram 8 shows' the influence of N on the corrosion according to the Streicher test.
Diagram 9 shows the influence of N on the distribution coefficient %Crα/%Cry.
Diagram 10 shows the influence of N on the distribution coefficient %Niα/%Niy.
The alloy contains in % by weight:
- C max. 0, 05 - Si max. 0,8
- Mn 0,3 - 4
- Cr 28 - 35
- Ni 3 - 10
- Mo 1,0 - 4,0 - N 0,2 - 0,6
- Cu max. 1,0
- max. 2,0
- S max 0,010
- Ce 0 - 0,2 the remainder being Fe and impurities in normal amounts. Ferrite content: 30-70 %.
Carbon is to be considered rather as an impurity element in the present invention and has a limited solubility in both ferrite and austentite. This limited solubility implies that a risk for carbide precipitations exists at too high percentages, with decreased corrosion resistance as a consequence. Therefore, the C-content should be restricted to maximally 0,05%, preferrably to maximally 0,03%, and most preferrably to max. 0,02%.
Silicon is used as a desoxidation additive at steel
manufacture and increases the flowability at production and welding. Too high Si contents increase the tendency for precipitations of intermetallic phases and decreases the solubility of N. Of this reason the Si content should be restricted to max. 0,8%, preferrably to max. 0,5%.
Manganese is added to the alloy to increase the solubility of N in the melt and for replacing Ni as an alloying element, since Mn is considered to be austenite-stabilizing. Investigations which are presented further down in this description show that Mn influences the corrosion resistance negatively and, moreover, deteriorates the structure stability, which results in a greater risk of precipitations of intermetallic phase. However, alloying with Mn has anyway been considered as advantageous due to increased solubility of N and a possibility of lower contents of Ni with maintained ferrite content. However, it has surprisingly been shown that Mn has a negligeable austenite-stabilizing effect. Without being bound by any theoretical relations, the reason for this is believed to be that the high N content gives a redistribution of Mn from the austenite to the ferrite, so that the Mn content differs insignificantly between ferrite and austenite when the N content is high. Moreover, it has been shown that at the Cr contents in question in the alloy, the N solubility in the melt is so high that it does not motivate any alloying of Mn to any high degrees. Suitably, a Mn content of between 0,3 and 4 % b.w. is chosen, preferrably between 0,3 and 1% b.w.
Chromium is the most active element for increasing the
resistance against most types of corrosion. At urea synthesis the Cr content is of great importance for the resistance, wherefore the Cr content should be maximized as far as possible out of a structure stability point of view. An increase of the Cr content implies that the ferrite, which always has a higher Cr content than the austenite, will be sensitive to precipitations at too high percentages, which implies that precipitation, mainly of intermetallic phase, may happen in the ferrite and at ferrite-ferrite and ferrite-austenite grain borders. Therefore, it is desirable to redistribute Cr from the ferrite to the austenite, so that the weaker austenite gets increased resistance, and that a higher total amount of Cr can be alloyed in without deteriorating the structure stability. Furthermore, Cr increases the solubility of N in the melt and the solid solubility of N in the alloy. At a sufficiently high Cr content, the melt production is also enhanced by the fact that it is then also possible to add all the nitrogen in the AOD
(Argon-Oxygen Decarburization) , which implies that the addition of expensive Fe-Cr-N at the ladle treatment can be dispensed with. In order to attain sufficient corrosion resistance in the austenite, the Cr content should be at least 25%, preferrably at least 27%, in the austenite phase. Therefore, a total Cr content of at least 28% is desirous, preferrably at least 29%. However, in the combination with Mo, W, Si and Mn, chromium increases the risk of precipitations of intermetallic phase, wherefore the Cr content should be limited to max. 35%, preferrably max. 33%.
Nickel is mainly used as an austenite stabilizing element and according to the present invention its
content should be kept as low as possible. An important reason for the bad resistance of austenitic stainless steels in urea environments with low contents of oxygen is supposed to be their relatively high content of Ni. The lower content of Ni in ferritic-austenitic stainless steels is assumed to be a primary reason to the better resistance of this type of alloy in urea environments with low oxygen contents. Alloying with N implies that lower contents of Ni may be added with maintained content of ferrite. However, the Ni content in the austenite phase is restrictive for the resistance in a urea environment. The Ni content is higher in the austenite phase in comparison with the ferrite phase. Investigations presented further down surprisingly show that an increase of the N content not only makes possible a decrease of the Ni content, but also implies that the Ni is redistributed from the austenite phase to the ferrite phase. Thus, this means a further improved resistance in urea environment, since the weaker austenite phase obtains a lower Ni content than expected when the N content was increased. In the present invention a content of 3-10% Ni is required, preferrably 3-7% Ni, in order to attain a ferrite content in the range 30-70%.
Molybdenum is a very active element and is used to improve the passivity of the alloy. Mo together with Cr and N are those elements that most effectively increase the resistance against pitting and crevice corrosion. Investigations presented further down surprisingly show that an increasing Mo content in the alloy according to the present invention improves the resistance in corrosive environments, which results in a decreased corrosion at the Huey tests when the Mo content is
increased. Furher, Mo diminishes the tendency for precipitations of nitrides by increasing the solid solubility of N. Therefore, a Mo content of at least 1% is desirable. However, too high contents of Mo involve the risk of precipitations av intermetallic phases, especially if at the same time the Cr content is high. Therefore, the Mo content should be limited to max. 4%, preferrably max. 3%, in particular max. 2%.
Nitrogen is a very active element in the alloy. N is a strong austenite former and enhances the reconstitution of austenite in the heat-influenced zone after welding. N influences the distribution of Cr and Mo, so that higher contents of N increase the relative share of Cr and Mo in the austenite. On the one hand, this means that the austenite becomes more resistant to corrosion, and on the other hand that higher contents of Cr and Mo may be included into the alloy while the structure stability is maintained, since the share of Cr and Mo in the ferrite, which is more apt to precipitations than austenite, becomes lower at determined percentages of Cr and Mo. Since the austenite phase is the weaker phase in the alloy, this phase will preferentially be attacked in a urea environment. Therefore, inclusion of high N contents is highly advantageous, because it decreases the risk for precipitations in the intermetallic phase in the ferrite, at the same time as it increases the corrosion resistance. Moreover, it is well known that N suppresses the formation of intermetallic phase, also in wholly austenitic steels.
Thus, in duplex steels a lowered sensibility to precipitations is obtained due to element redistribution, as well as a decreased sensibility to precipitations due to the nitrogen in itself as an
alloying element. Therefore, N should be added to a content of at least "0,20%, preferrably at least 0,30%, and in particular at least 0,36%. Upwards, the N content is limited by the solubility in the melt and by the solid solubility in the alloy, which involves that precipitation of nitrides may happen at too high N contents. Further, too high contents of N may cause formation of pores at welding. Therefore, the N content should be limited to max. 0,60%, preferrably max. 0,55%.
At welding nitrogen enhances the reconstitution of austenite, which drastically improves the toughness and the corrosion resistance of the weld joint.
Copper improves the general corrosion resistance in acid environments, such as sulphuric acid. However, high contents of Cu decrease the pitting and crevice corrosion resistance. Further, in the alloy type in question the solid solubility of Cu is limited. Therefore, Cu should be restricted to max. 1,0%
Tungsten increases the resistance against pitting and crevice corrosion. However, high contents of W increases the risk for precipitation of intermetallic phases, particularly in combination with high contents of Cr and Mo. Moreover, the inclusion of W results in a more costly handling in the steel plant. Therefore, the amount of W should be limited to max. 2,0%.
Sulphur influences the corrosion resistance negatively by the formation of easily soluble sulfides. Furthermore, the hot-workability is deteriorated by high contents of S. Therefore, the content of S should
be restricted to max. 0,010%, preferrably to max. 0,005%, in particular to max. 0,001%.
Cerium may be added to the alloy, for instance in the form of a misch-metal, in order to improve the hot- workability of the alloy. Contrary to manganese sulfides, Ce forms ceriumoxy-sulfides which do not deteriorate the corrosion resistance. Therefore, Ce may be added to the alloy in percentages up to max. 0,2%, preferrably max. 0,1%. If Ce is added the content should be at least 0,03%.
The ferrite content of the alloy is important for guaranteeing the phase composition, the structure stability, the hot-workability and the corrosion resistance. Therefore, the ferrite content should be in the range 30-70 % by volume, preferrably in the range 30-55 % b.v.
The invention aims at providing an alloy with good corrosion properties, primarily in urea environment. Since Cr is the most effective element for improving the resistance, which is a consequence of the fact that the passivity properties improve with increasing Cr content, the Cr content should be as high as possible. However, at a certain content of Cr the structure stability deteriorates. The reason for this is that the ferrite, having higher contents of Cr and Mo than the austenite, becomes apt to precipitations, whereby mainly intermetallic phases will precipitate in the ferrite, or at the grain borders between ferrite- ferrite or ferrite-austenite. However, by including N into the alloy, Cr and Mo will be redistributed into the austenite phase.
In order to optimize the Cr content in the austenite phase, preferrably the following conditions should be fulfilled (as usual, all percentages are % by weight) : %Cr + 15* %N > 31, in particular > 33
In order to further improve the resistance agains oxidizing environments, the following condition is advantageously fulfilled:
%Cr + %Mo - 8- %N > 22 In order to minimize the risk of precipitations of intermeallic phases at the production, the composition should preferrably be reflected by the following relation:
%Cr + 4,l*%Mo + 1,4* %Ni - 6,9-%N < 55; and in particular by the following relation:
%Cr + 0,9-%Mn + 4,5*%Mo - 12,9*%N < 35
In order to further improve the conditions for attaining a sufficient nitrogen solubility in the melt at the AOD production, preferrably the following condition is fulfilled:
4,5-%Cr + 2-%Mn + l,l*%Mo - %Ni > 114
Example
A number of test charges were provided by casting ingots of 170 kg which were hot-forged and extruded to bars, whereof material for most of the corrosion tests also were cold-rolled before solution heat treatment.
Table 1 shows the composition of the test charges in % by weight.
Table 1
Steel Charge C Si Mn Cr Ni Mo N
654604 0,015 0,18 7,07 30,94 6,56 1,95 0,37
654593 0,016 0,20 6,12 30,46 8,08 1,55 0,28 ( invention) 654594 0,017 0,22 1,13 30,42 9,92 1,26 0,18
654617 0,015 0,23 8,00 29,14 6,56 1,03 0,27
654597 0,016 0,19 5,87 29,01 8,89 1,96 0,18 (invention) 654598 0,015 0,21 1,09 29,31 6,08 1,56 0,38
654600 0,015 0,22 8,19 27,46 7,47 1,34 0,18
654601 0,021 0,22 6,04 27,65 4,07 1,01 0,38 ( invention) 654602 0,015 0,22 1,16 27,50 6,70 2,02 0,28 ( invention 654605 0,015 0,27 1,02 29,03 8,86 1,06 0,19
654606 0,017 0,23 7,45 29,61 5,24 1,52 0,37
654611 0,015 0,19 7,46 30,77 6,35 1,55 0,36
654614 0,015 0,20 1,70 28,81 9,17 1,07 0,13
654596 0,017 0,22 8,00 28,85 7,10 1,95 0,28
0 The alloys 1 - 9 were provided according to the principle for statistical test planning (STP) , whereby Cr, Mn, Mo and N were varied in a systematic way, which gives information about the influence of the above mentioned four elements upon the different result parameters. The ferrite content was held within the range 40-60% for all the alloys by balancing with Ni .
Mechanical Properties
The mechanical properties of the materials were found good. One example is given in table 2, where the mechanical properties of alloy no. 10 are presented.
Table 2
Alloy Yield point Ultimate A5 Impact Vickers in tension strength (%) strength hardness Rp0,2 (MPa) Rm (MPa) (J/cm2 ) HVIO 20°C -50°C
10 471 715 37 260 260 250
It is clear from this table that the material has a good ductility, which is illustrated by the high value on rupture extension (A5) . The material also has a good impact strength, also at low temperatures. The reversion temperature is lower than -50°C.
Structure Stability
It is very important that the alloy be sufficiently structure-stable in order to be capable of being produced and welded without any precipitation of intermetallic phase or nitrides . The high N content in the alloy makes the material, in spite of its high alloying level, to manage the requirements on structure stability. The reason for this is that the nitrogen suppresses the formation of intermetallic phase, at the same time as nitrogen redistributes Cr and Mo from the ferrite to the austenite.
Table 3 shows the influence of the full annealing procedure on the structure of some test charges, the heating temperature being 150°C/min, the full annealing temperature 1020°C, the holding time 3 min, with varying cooling rates. The share of intermetallic phase was measured by pit counting
Table 3
Alloy Cooling temperature Amount of Intermetallic (°C/min) Phase (%)
3 140 0
3 17,5 46
10 17,5 0
13 17,5 28
Obviously, alloy 13 is more sensitive to precipitations than alloy 10, which depends on the fact that the nitrogen content is lower in alloy 13. Alloy 3 manages a cooling rate of 140°C/min without any precipitation of intermetallic phase, but not 17,5°C/min. The reason for this is that the nitrogen content is too low at the higher Cr content in this alloy to be equally structure-stable as alloy 10. However, alloy 3 is producible since the cooling rate of 140°C/min corresponds to the lowest feasible cooling rate at the manufacturing.
Corrosion Testing
Huey testing according to ASTM A262, practice
C, is an established method for measuring the resistance of alloys in oxidizing environments and for testing if the material has been heat-treated in a correct way. Possible precipitations, e.g. chromium
carbides in the grain borders, increases the corrosion rate at Huey testing. Materials for urea environments are often delivered to specifications requiring low corrosion rates at Huey testing, and are primarily considered to give a view of the material's behaviour in oxidizing process environments. Table 4 shows the corrosion rate at Huey testing of the alloys in solution heat treated condition (1040°C/20 min/H20) .
Table 4
Huey testing of alloys in solution heat treated codition.
The alloys 1-9 may be used for calculating the influence of Cr, Mn, N and Mo on the corrosion rate at Huey testing. The diagrams 1-4 show graphically the influence of the elements. The following relation is obtained:
Corrosion
(mm/year) = 0,285 - 0,0080% Cr - 0,0080% Mo +
in intervals of 5°C until pitting arises, is a common method for testing corrosion resistance in sea water. The temperatures at which pitting start, may be seen below in Table 5. In this table the critical pitting temperature (CPT) is given in 6-percent FeCl3.
Table 5
Alloy CPT ( ° C)
4 45 10 60 13 45
The higher N content of alloy 10 gives a better pitting resistance than that of alloy 13. Alloy 4 has a worse resistance than alloy 10 because of a higher Mn content. Mn can lower the pitting resistance by the fact that easily soluble manganese sulfides are formed.
The balancing of the included elements according to the present invention has been done very carefully in order to attain a good structure stability, good mechanical properties and high corrosion resistance. Diagram 9 shows the influence of N on the distribution of Cr in the ferrite (Crα) and Cr in the austenite (Cry) . It can be seen that Cr is transferred from the ferrite to the austenite at increased N content . Diagram 10 shows the influence on the distribution of Ni between the ferrite (Niα) and the austenite (NiV) . Surprisingly it has been shown that Ni is transferred from the austenite to the
ferrite at increasing N content.
-.Table 6 shows the phase composition of two alloys according to the invention, the phase compositions being expressed in % by weight.
Table 6
It can be seen that a high N content gives a high Cr content in the austenite at the same time as the Ni content is low.