Duplex Stainless Steels and Chloride Stress Corrosion Cracking

thank you Dan uh those of you who know me uh probably associate me still with L electric go back further with but uh I retired in 2007 so uh I've been out circuit and a year ago I relocated from the fand area to Chapel Hill North Carolina so that shoveling snow is no longer than my job could we kill the lights in the front row I make okay so in logical order this would have probably been the first presentation of the day but it wasn't for my uh anyway uh what I'd like to start is why Earth dlex St Fields exist in the first place well the answer is a chloride stress corrosion PR we talked about all kinds of corrosion tests this morning and it was always pitting the real reason they exist isn't pity it's chloride stress corrosion P the problem is that the tintic stainless Fels like 304l and 316l are highly susceptible to and CRA the partic stainless Steels are highly resistant to Chloride stress C Chlor the fidic stainless Steels however are really ugly Critters for what reason is green the toughness vanishes when you w done C of graen duplex stainless Steels then coming to the picture as a way to combine properties of Austin stand and the properties and people now you guys all know Murphy's Law right anything can go wrong will go wrong and if you apply Murphy's Law to alloy design what would Murphy's Law tell you about this situation it would tell you that if you combine two alloy systems you will get the worst of f but that's not what happens with duplex duplex stainless steel is a clear violation of mer La so anyway in order to get chlorine exess corrosion cracking you need chlorides you need tensil stress uh tensil stress stress is automatic with Wells so you can always always always anticipate yield Point residual stresses around a well better so you're always going to be susceptible to Chloride stress corrosion cracking in the Alloys like real 4 316 that's why duplexes are interesting this is a a picture of chloride stress corrosion crack or a whole lot of chloride stress corrosion cracks this is in 304 stainless 304l on the left side you see the unetched um appearance and the signature of chlorid St and cracking is crack branching so you have a crack at the surface that becomes a multitude of crack subsurface and on the right hand side you have the same picture but uh it's in this area here and it's um enlarged and the samples the etch and the reason for the etching is to show you that the chloride Express corrosion crack doesn't care about grain boundaries you can see here you have grain boundaries and you have a crack going right through a grain it jumps from grain to grain to grain it doesn't care about grain boms this has nothing to do with sensitization it has everything to do with chlorides and chloride stress corrosion C now you got two ways to cure chloride stress corrosion crackless steel one way is lots of fair and that originally LED people into all kinds of advancements in what I call third generation ftic stainless Steels these are the ftic stainless Steels like 44 18 Chrome 2 mly very low carbon very low nitrogen um or 261 is another alloy 26 Chrome one Molly and very low nitrogen very low carbon they're ugly to fabricate in any kind of thickness Beyond sheet metal so we try not toat the other alternative um was to go with lots of get away from um compositions that were susceptible and that takes you into ni based Alloys or the soall super atic the about 25 nickel 20 Crome or something like that duplex turns out to be the cheapest alternative and the most friendly because you get those fabrication properties that are like Austin stain St in particular grain growth doesn't hurt as it does with okay so in a duplex stainless steel you got approximately equal amounts of fite rate some say 50% 5050 or something these things give you resistance to Chloride stress corrosion cracking like the peridic stainless Steels they give you fabrication properties that are more like atic stainless Steels and they give you a bonus doubling of the yield that was already mentioned now how do duplex stainless Steels come out come about what you see here is a um section through a turnery iron chromium nickel diagram at a constant iron level okay so it's 70% iron and what you're doing as you go from left to right across this diagram is you're going from no nickel up to 20% nickel and from 30% Chrome down to 10% Chrome and then up at the top here we have liquid and in the center here we have have this triangle tic triangle that's called that's where liquid plus ferite plus tinite coexists it's a very small ring where they closes all three phases at once but the important thing about it is that an ordinary tic stainless steel like 304 L for example is going to have a composition very close to the edge of this triangle so that the metal is going to solidify as liquid going to Liquid Plus f going to liquid plus ferite plus tinite then into ferite plus tinite and eventually down to 100% tinite that's why your 304 base metal doesn't contain any fa however if you take that 304 base metal and you melt it with a gas tungsten AR torch no filament just melt and then measure the fite you will find there's some f and that's because during solidification you get some formation of ferite and some of that ferite survives to room temperature to get rid of it you need to cause chronium to diffuse out of the fite and nickel to diffuse into the fite fite and the and the tinite have different chemical compositions all right now a duplex stainless field starts out further to the left in the diagram someplace over here where solidification goes from liquid to fite the Liquid Plus fite to 100% fite and 100% fite exists for an appreciable temperature range and then as you cool further it enters into the two phase field where ferite and tinite coexist and that ferite plus tinite will coexist until the cooling has gone so far that nothing else happens and that's what you freeze in at room temperature now how easily the austinite forms depends upon how high the temperature is where the first aite if the first austinite forms at a high temperature then diffusion is pretty rapid chromium can partition to the ferite nickel can partition to the tinite and it's not very dependent upon the fuid but if the first a doesn't show up until a rather low temperature now the diffusion of chromium and nickel is very limited the end result is that you don't partition chromium and nickel and that's where you can get into trouble with these things because you're locking in very high Farah you don't get anywhere near equilibrium transformation laate and the end result is you get very poor mechanical properties okay that leads us to starting to talk about the duplex stainless Steels in terms of generations um you know people talk about dlex stainless steel today as if this is really new technology but two Flex stainless Steels are older than I am those things existed already in the 1930s you see these two allo CD4 MCU which is kind of Forerunner of 255 that alloy has been around since the 1930s that alloy was a cast Al c4c was a cast designation castings are always kned afterwards so if you had to do any repair welding on it you would use any more or less match FL material and you wouldn't yal it then it was okay but in the as welding condition I properties off 329 likewise the few that was a rock Mater very popular in Japan not so popular in the United States um but as a rock material has some very interesting properties uh again in the Asal condition you would consider unweld because the mechanical properties and the corrosion properties are very important it's only after kneeing that it develops the properties problem is that you get in the as welded condition virtually 100% all right and then the properties are those of AIC stainess steel with very coose grains in other words very poor tpic uh also because there was a little bit of residual nitrogen in it you got Chum nitrite precipitates so the corrosion resistance was also they had to AAL at a temperature around 1100° C or higher in order to get a good world and what you were really doing was adjusting the balance of aite versus second generation happened by accident I believe it happened um in the late 70s and became popular in the 80s the thing that triggered it was the development of the Argon oxygen decarbonization system for refining stainless fields in the Argon oxygen decarbonization system you BL argon um with a small amount of oxygen in it into the liquid metal the oxygen reacts with the carbon takes the carbon out and the Argon is the carrier to help the bubbles remove get removed from the puddle from the U melt but um when you read the books about when aod was developed what you find out is that people right from the get-go with aod were using air as a source of oxygen and air is 80% nitrogen so they accidentally got nitrogen into the metal and voila the weldability improved improved market so that in the second generation of duplex stainless Steels the ones that proliferated in the 1980s uh typified by 20304 and 2205 um those materials had about 12 points of nitrogen with 12 points of nitrogen in them the heat affected Zone was now somewhat sensitive to heat input if you had high heat input you would get good heat effective Zone proper if you had low heat input you would not get good its own properties it all depending on how much time that temperature the material had in order to produce a decent amount of O Okay and when this generation of of U base materials came about in the 1980s it became very popular to adjust the th metal composition to 9% nickel instead of 5% so that you would get more tinite in the well metal that didn't do a thing for you for the heat effect itself so you had to control the heat input in order to get a decent heat effect now I want to point out one thing here 20205 in the literature was always associated with UNS number s31803 and that's very important because it allowed for a rather minimum nitrogen cont 8% minimum nitrogen in order to satisfy UNS 3180 now um two Japanese gentlemen uh professors oawa and koseki um back in the 1990s did a remarkable study that was published in the Wilding uh about alloy partitioning in duplex St speak the trouble with the publication in the welding Journal is that in those days the welding Journal did not publish in color in the recent supplement so that I don't think very many people appreciated the tremendous discovery that oala and keki had made but they had a second publication of it that appeared in the IW commission 9 and I saw that and it was one of those aha I get it so let me go through this uh in a little detail to to point out what they discovered and how important it is for us okay so what you have here is a rock 2205 type Bas material uh in the upper left corner you see the base material in the rock form and it you can see the layered appearance the grayer material is very the white material is aome and then what they did was use an electron microb to look at the distribution of the aloine elements in that BR material and each of the maps is color coded so that you can see where the high concentration of an element is and where the low concentration is so if you look to the fite area here and you look to the chromium map that corresponds to it you see the red color in the fair and that corresponds to 25% chrom the alloy only contains 22% conversely if you look through the austinite area which is blue here that corresponds to 20% so there's a partitioning chromium promotes ferite so it's segregates under equilibrium conditions conversely if you look at the nickel now what you see over here in the ferite area is low nickel and now the blue color corresponds to 5% nickel the alloy is nominally 6% nickel so you got 5% Nickel in the fite and over in the austinite area you have the red color which corresponds to about s or 7 and 12% so the nickel has also partition if you look to the mum map the mum map looks just like the chromium map Rich mum area is the fite the low mum area is the A and now we're saying the alloy is 3% mum but the fite contains almost 4% mum and the aite contains 22% so there's a partitioning ofel of the LI the most interesting alloy element however is the nitrogen now in the nitrogen color coding blue corresponds to zero red corresponds to about 253 something in that order and if you look in the farad areas it's zero you look in the tinite areas it's around 2.25 for an alloy that contains 12% nitrogen so it's an almost complete partitioning of the nitrogen to the tinite and away from the now if you take the same material and you make an autogenous well done this is what you get myogenous guest same material and this this relates not only to the L metal but to the hottest part of the heat effect and it's critical that I bring in the hottest part of the heat effect Zone although o c showed this as well metal not as heat Zone but it applies to the heat effective Zone as well when you look at the micro structure in the upper left corner now what you see is a big fite grain out lined by little plates of austinite on the grain B and then in the center of the ferite grain there is all this black specs the chromium nitrites which I think I hav showed this morning and if you look very closely you can see next to the austinite ples along the grain boundary there's an area that's clean doesn't have black no call me okay now if we look at the partitioning alloy elements you look at the chromium map there's no partitioning evidence if you look at the nickel map there's no partitioning evidence what you can see a little of in the nickel map is these little hexagonal areas that are slightly more yellow in the outline and green in the center okay that tells you solidification for doesn't tell you anything about the transformation when fite transformed a this has to do with solidification if you look at the malum map again you can see a little bit of the kind of coring or the yellow hexagon hexagonal shapes in there but you don't see anything that suggests the micro structure up here however when you look at the nitrogen distribution you can see the ferite grain very clearly outlined by nitrogen which is inite plates on the bra you can see in the center of the grain an essentially nominal nitrogen content that's caused by a mixture of chromium nitrites and Par and next to the grain boundary you can see the zero blue color zero nitrogen blue color which means the nitrogen had a chance to to get out of there so nitrogen is diffusing towards the aite out of the fah near the grain boundaries the nitrogen can get out but in the center of the grain it can't doesn't have enough now you might say well what if I wanted to get rid of those nitrides what would I do I might take the nitrogen out but then I have a peric stainless Ste 100% turn a better way go and this was referred to by own again this morning is put in more nitrogen instead of 12 points now we have 18 points in nit otherwise it's the same composition 225 now when you look at the base metal micr structure you can still see a big grain outlined by aite plates but there are all kinds of aite plates crisscrossing the grain internally now when you look at the alloy element mapping you can see a little bit of the chromium partitioning if you got really good imagination here I think you can see there's a tendency for a blue outline here that's a farag grain boundary which is completely covered with Austin this tells you that the transformation of the first formation of austinite took place at a higher temperature where nickel or where where chromium had a chance to diffuse out of the A and if you look at the nickel map you still can't see anything but the little hexagonal shapes nickel is the slowest diffusing of all the elements if you look at the malum map you can again see in fact you can see it better now that blue outline light green where the mum is bailing out of the areas that transformed aate and if you look to the nitrogen map you can see the same big pair of grain outlined by high nitrogen on the grain B you've got all these ostanite plates inside the grain which are also full of nitrogen and then you've got ferite areas that are completely clear of nitrogen now you got good properties it's the nitrogen that makes the alloy Welling that promotes formation of tinite at high temperatures and diffusion can go on so one of the things we like to do now is as a result of this we try adding nitrogen to the shielding gas to keep it in the weld metal we increase the nickel content in the filler metal at 9% to help the weld metal and with this addition of nickel We Now call the Chlor metal 20209 instead of 2205 to tell you that it's 9% instead of 5% but this brings us to the third generation of duplex th seals and 2205 appears in this generation as well but now it has a new UN number instead of 31803 it's you see 32205 the asme code in year 2000 redefined the 2205 Bas CH by this UNS number 32205 the minimum nitrogen content in that alloy is now 14 points instead of eight and that has a big effect on the wabil now the heat effective zone is not very sensitive to heating you can go 12 K per mm to 22 K per mm and still get good and Ne effective Zone c big Advance partly for historical reasons but partly also to improve the toughness of the W metal filler Metals still tend to be high in. and now we start looking at pitting resistance but it's only after we got to the third generation of duplex stainless Steels that we thought at all about pting as I said earlier there was chloride stress corrosion cracking that was the main reason and now we can look at this pit resistance index which includes nitrogen and I'll give you a slightly different version of the fitting resistance index because there's a factor for Tungsten in it I think when you saw the one this morning you saw it without tungsten tungsten is considered to be half as effective as mum in terms of dividing resistance to pting otherwise from for Pon free alloy pitting resistance Index calcula this way is the same as the way you saw it this morning but for 0 on 100 which contains deliberate additions of tungsten this matters so the third generation of duplexes have rather modest pitting resistance they got pitting resistance index numbers in the low to mid-30s what you would like is to get up to 40 because then you can be competitive with the so-called super Austin these are the Alloys of uh 20 Chrome 25 nickel like 254 Smo or al6xn and that brought us to the fourth generation of H like stainless Steels and these are the so-called copics there are two main Alloys in the r form 2507 and 01 100 and here you can see 01 100 contain in addition of tungsten which gets its pity resistance index up over 40 and then the cast version CD3 MN is the cast version 2507 and CD3 M WC anyway um that's those are the cast equivalents uh for 257 and 0100 so that's your fourth generation of uh pric stainless or duplex stainless Steels mechanical properties of the super reflex stainless Steels are really outstanding this is like hy80 steel with corrosion resistance you're looking at 8 C I minimum yield strength in the super um you will notice that the uh yield strength numbers for cast materials are low and that's caused by the fact that when you unal the material the micr structure TS topiz and part of the strength in a duplex stainless steel comes from the layered structure of parite and when you lose that layered structure you lose some structure you have to be used to that notion where do we apply these things I was run through a bit this morning there's a lot of environments including with these super duplexes sea they're resistant to seaw W that's quite impres people make things out of 316 stainless and they put them in C1 and understand why it Corrs CU gosh it's stainless it's not stainless in SE water okay so that's just kind of a list of things that uh you'll find um one of the interesting applications for the super duplexes is the blue Gast dlers blue Gast desulfurizers have high temperature areas and low temperature areas you don't use super duplex in the high temperat part of the gas but you do use it in the low temperature and it's quite resistant to the acids and the chlorides and whatnot that exist in that part seems a little odd to think about the way this happened but after the fourth generation became popular there was a kind of a relook at the low alloy the socaled Lees now originally I had put 2304 in the um second generation of D but I think today we would reassign it to the lean duplexes which come really came after the fourth generation but the 2101 alloy is probably the most prolific today of the lean duplexes this alloy is very low in nickel only one and half% but it's got 5% mes now the function of the manganese is not to promote a you have to get that notion out of your head manganese does not promote a those of you who know the shuer diagram and see manganes in there as an atiz it get that out of your head it doesn't work but what mainly does do there's two things one is if you get austinite it enhances the stability of the austinite with respect to transformation to Martin site at low temperature so that 5% manganes prevents formation of Martin side and low but the other thing that it does is it enhances the solubility of the metal for nitrogen and it's nitrogen that really makes the oite form in the first place anyway the idea of this alloy and the other lean duplex with 2304 and 244 I think mentioned in some of you the idea with those is to get corrosion resistance that's more or less competitive with 304 L but do it with very little me here you're talking about 1 1/2% nickel instead of roughly 9% nickle in a 30 or four hour now an interesting thing about these things I I told you with third generation duplexes that they were not sensitive to heating with normal aring processes I should have ADD I didn't Del the Third generation of duplexes can be welded by The Arc processes they cannot be welded by resistance weld they cannot be welded by laser be because the metal cools too fast even with however 2101 is different and mic did some work on Resistance welding of 21101 and you have to do some tricks with 21101 if you're going a resistance welding the idea here again is to get some transformation to austinite in the as welded condition in the normal resistance well you get a little austinite on the grain boundaries and a little austinite inside and a lot of chromium nitrates but but in a resistance welding machine after the resistance welding cycle itself is done you know you've made a resistance weld you can pulse the resistance welder a second time a 2C puls of energy 2 seconds drops the ferite content from 83% to 74% and what it particularly does I'm not sure that you can see it as well here as you might see it in a paper version but the amount of the uh chromium nitrides in the dark areas here is greatly reduced they're discontinuous you can live you can get useful mechanical properties out of a resistance weld that is post heated with this little 2C post heat treatment you can get useful mechanical properties and useful corosion resistance so you can make things like dimple jacket Heat exchangers by resistance color Metals for these things are classified both by AWS and in the ISO standards so what you see in AWS is uh 2209 which is directed at 2205 type base material the 2307 is intended for the lean defens the alloy was originally developed for 2101 but it also works 24 um that alloy is quite different in composition it has quite a bit of nickel compared to the 211 base material um the reason for that is in the ASW condition if you have a a 21101 composition and you pick up oxygen from a flux shielder process you get lousy mechanical properties particular particular very poor toughness and even very poor t but by depending on nickel to help you out you get better confidence so the 2037 composition or 237 NL as is stand is adequate for 2553 exists only in AWS it doesn't exist in um European or is but this is essentially a matching filler material for the pellan 255 which shows up back the this is the 255 255 this is the rock composition now if we go to the filler metal the 2553 matches that composition exactly and you would use it primarily on cast material not on rock material because with that Niel it's going to have l properties a has good the 2593 which is the next one down that is the filler metal that would be recommended for the ASW condition so you actually have two fill models one for the for the As for the castings and one for the okay and then 2594 is a filler material intended for 257 and the last 12595 is for the Zer 100 now in most practical application of super dupes people use 2594 and 2595 interchangeably for Z 100 257 and I don't think anyone has ever shown any convincing evidence that one or the other is better now we talked about hyper duplex that was mentioned again this morning uh there are two hyper duplex Alloys that have seen some appearance in the literature um 25 or 2707 which is a um 27% chromium 7 5 m and4 nitrogen has a PR that doesn't quite 50 but with it then there's a second one this 3207 which is 32 Chrome 7 Niel but it's not so high and that actually does have a above 50 the filler material that is proposed for these things by sandic is the Pryor of the material is 27 Chrome 9 Nickel 5 Molly 24 nitrogen and it's the same filler material for that filling material is only available as gas tunks that AR fing material that's the only welding method that's ever been used on those things as far as I know in any to any extent and I have at the moment I don't know that there's any commercially significant application of them manufacturer was very aggressive towards par that's the only one they okay all right thanks but anyway this is sort of like maybe that's what's in the future maybe not if it's not a commercial success soon I think it will matter okay now C of the fide phase in the newes there is a lot of possibility for precipitation these were mentioned also this morning uh the alpha Prime phase which forms in a temperature range of about 400 to 540 C the sigma phase which forms between 500 and 1070 C and the kai phase which can go up to 1100 C1 so that essentially prohibits you from using the alloy anywhere from below 400 to 1100 well that alloy doesn't have any useful properties above 1100 see um it's sort of like peanut butter terms of mechanical property so the real interest is in below 400 C now here you will see this is the temperature range where it forms where the alpha Prime pH will form in few hundred hours you go to 10,000 hours that temperature comes down and so people will say 300 see is the maximum service temperature for these Alloys or maybe even 250 so those are all limitations now the alpha Prime and brittle occurs when not an inel compound but when farite partitions into two ferites an iron rich ferite and a chromium Rich fite and if you look at the binary iron chromium phase diagram what you will see is that that two-phase decomposition of fah into an iron F and a chromium F when that occurs the chromium Rich phah contains over 90% chromium and that horribly depres chromium content in The Matrix and so the Matrix becomes very susceptible to corrosion the guy named Tavaris uh published a paper on the internet from which I borrowed this material what he's looking at here is stin wall pip so it's not sharpen impact Val you shouldn't allow you too much um but what he's looking at here is the um mechanical properties of uh 2205 material using the old 31803 number so I'm not sure exactly what the nitrogen number is but U he's looking at time at 475° C on the horizontal axis and then on the vertical axis on one side we're looking at impact energy and jewels and these are very subsize sharpy specimens because um of the fact that it was thin all material so what you see in the uh impact results um Sharpies start out at a bit over 30 Jews and they drop in something on the order of100 hours to less than half and if you also look at the hardness of the material he's using Bell hardness so the hardness starts at a number in the the 220 area and it goes up with increase in time and temperature so that you're up in area of 350 after 500 hours so there's a serious Harding of the alloy and a serious emling of the alloy and if you look at corrosion what it was done here was to use a electrochemical uh pitting cell to um just increase the potential until you got the first surge of current in the electr chemical cell and that's the initiation of pitting then you stop the process immediately so instead of getting big pits that you can stick your finger into or stick a a ballpoint pen into or something like that you get little bitty pits and you can see here all the pits are in the dark etching material which is the P there are no pits whatsoever in the Aus you can't see the alpha Prime you need a scanning transmission electron microscope see that but you can see the pits that result of the CH depletion next to the AL so that's very destructive to the properties heat treatment of duplex stainless Steels especially when they've been well it's a little bit of tricky distance you know cuz we're saying well you can go you've got a whole temperature in from 400° C up to something over 1050 1070 de seat where bad phases for C phase p phase and the ASM standard a240 used to call for a kneeling of 2205 and a minimum temperature of 10 40° and that was just fine for the basic it was just fine for matching composition well done but it turns out it wasn't just fine for 9% nickel fill this was illustrated by a company that gave a paper at one of the duplex World conferences back in the Netherlands uh in the 1990s uh in which they they reported on fabrication of a very large dis head for 2205 the head was large enough that it could not be fabricated out of one plate of 2205 mat so they welded a number of plates together using 9% Niel filler and then they co-formed it to make this dish t uh and after they cold formed it they saw some damage to the mechanical properties in terms of L so they went to a kneeling the kneeling said 1040 DEC minimum so they went up to 1050 or 1060 I don't remember the exact water quench well the head wared slightly bring the analing so they needed to in it up a little bit so they put it back in the forming press and they were just going to give it a little bit of nudge to chew it up and it fractured the welds unzipped they literally unzipped and the and the plates came apart they investigated and found that the weld metal was loaded with signal interestingly enough I gave another paper at the same conference in which I talked about deep treatment in group like stainless fields and to begin with on that paper I referred back to an old diagram that was developed by a gentleman named Paul grer that worked for climax M company back in the 1970s and his diagram which he shows here which I'll show here um is looking at the phases that are stable as a function of temperature in a 25% chromium 35% lium alloy with varying amounts of nickel so what you have on the horizontal axis here is nipp content increasing from 0 to 30% and all of these Alloys contain the same amount of chromium andum so what you have here now our temperature feels for 100% fite farite plus tinite tinite farite plus Sigma fite plus austinite plus Sigma and austinite plus Sigma the important feature of this diagram is the sigma solvus boundary that's the boundary above which Sigma B and here you can see that as the nickel content increases the temperature the sigma Solis Rises so when we went from a 5% nickel base material to a 9% nickel thermal we raised the sigma salvus temperature by about 100° C Rob's diagram was there but no we saw well I was fooling around with duplex ferms at the same time and I got to looking into heat treating Heat Treating of these things because I was making some filler Metals for a purveyor of dlex casting so I made a weld metal of a 2209 composition IED it at 10 65° which is 25° higher than the as recommended minimum so I should have been safe and I wanted question and I had Sharpie specimens tensil specimens machine from it and a metal graphic well the first specimen that got addressed in my laboratory was Sharpie impact so I have a technician in the we call it affectionately the buing lab I was broke things speci and um I asked him to break specimens at room temperature to start out so he got this sample he broke it he broke the first couple impact specimens at room temperature then he came running down the hall of my office he goes doc doc you got to see this you got to see this what so I go on the busing landb and he says now watch this he puts the specimen in the impact specimen in the impact machine at room temperature hits the hammer now when you got a good impact specimen it's going to go th this one went pink three jewels fulls size impact not a subale fre then we measured the T elongation and it was about 10 then I looked at the micro structure and here was the answer all these white Islands islands that are outlin but not fetch signif H dark stuff of course that's the and the white stuff would be a in the as welded condition I measured the ferite content of that and it was 45 number I use magnetic 45 in this condition was 22 Fad in other words half of The Fad was six so I said well maybe we just didn't a meal it long enough so instead of 2 hours at 1065 I went to 72 hours and 65 and I got exactly the same thing it wasn't that I hadn't heated long enough to dissolve the signal it was that the phase was stable at 1065 then I took it up to 1100° 111 but anyway after that AAL I got good impacts I got good ductility and I measured the fite number again and you know what it's was 45 again the reaction was exact was completely reversible I could get the sigma to go back into solution and I got far now that was kind of interesting to me because when you start out with um High fite glob uh let's say you get a 60 or 70 fite number in the ASW condition and you anal it you will see the fite number drops drops appreciated but if you start out with a fite number around 45 doesn't Dr and the reason is because the alloy started to form tinite at higher temperature and you got very close to equilibrium partitioning very close the equilibrium out of a So when you say is the um fite stable question really is how much fite is there you got a lot of fite no it's not all stable but if you only got a little bit of fite yeah it's stable okay so as a result of that asdm RIS ASM a480 to change the ining temperatures for the 2205 and similar now it was mentioned this morning that um you never want to make an autogenous well in a duplex stain because you will have horrible mechanical properties um when you make the root pass in a pipe you generally do that by gas T to AR and welders don't like open Roots as a matter of fact they hate them and they hate you for telling them you have to make a well with an open roof but the conventional wisdom is you got to use open roots and here we're talking about an open a root Gap that might be um 2 to 3 mm uh some people say 3 to 4 mm depending upon which U ring process you're using now this stt that's a surface pensent transfer that's a Lincoln craving for um gas metal AR short circuit and transfer with a controll BL and there are other people make similar material and they all come for the 3 to4 mm recommendation but anybody in this room I'm very big at least three on okay I'm going to be careful all right I'm going to tell I'm going to tell you two things nobody believes filler metal producers because everybody knows filler metal producers why so they always check on us I'm a filal producer I've been checked this time they don't believe us when we tell us how much F and something they don't believe us when we tell them how much pencil strength they're going to get or how much they checkup well shouldn't believe welders either they cheat and when you tell them that they got to have a 2 to three or 3 to 4 mm root G you're not looking they might squeeze it down to make their job easier it's much easier to be Bridge closed gr than open am I telling the truth so far you guys always kind of okay so what are you going to do about this as an engineer are you going to let the welders Chief in the next thing the way a lot of fabrication is done particularly around the North Sea where there's a lot of duplex in walks is to put a little dent on either side of the joint preparation on the surface and the dent is put there fixed distance away from the edge of the joint so that it won't be melted by the welding it'll still be there afterwards that's called a witness then you measure the distance between those two dents after the welding is done and you can catch the welder mating except that a do well duplex sh 3 4 mm sometimes even more that's all figured in though that's all figured in those things are called witness Mars I don't know if they're still popular in the north CA area Fred probably knows more about that than I do but I know they were very popular at one time they still using them Fred okay all right so I had to tell you about color metal color metal producers BL I Told You chea So share the blame all right okay now High nitrogen works well for control th the heat effective Zone but nitrogen tends to be lost from the weld pool I think Ellen referred to that this morning so it's common to add some nitrogen to the shielding gas particularly gas and art Welding which has the greatest tendency to lose nitrogen and I will confess to you right now that if you use even 2% nitrogen in the shielding and gas tungston un your tungsten is going to deteriorate more rapidly than it would without the nitrogen emiss so that I can say come deal with cuz that's the only way you're going to get good properties in the super reflex St you got to get the nitrogen in the filler Metals we still use the 9% I mention this is the kind of micr structure you can expect if you add nitrogen to the shielding gas and a super duplex well this is4 257 base material here you can see the heat affected Zone has a little more ferite in it than the base material has a little more F but it's enough austinite there to prevent any extensive nit nitrite precipitation I'll go back I'll go back to Julia's Julian's talk this morning you will never see Zero inner metallics or zero nitr I agree with you completely going to be some there question is are that's the kind of micr structure you should have um this is looking at gas in our quality with pure argon versus 5% nitrogen and with pure cure argon you'll still get pretty good impacts but you've got these pretty large fade areas which are at risk for getting nitrate crucification whereas when you add the nitrogen to the shielding gas you have very little in the way of large F areas and therefore a much less tendency to get the now another thing that was referred to this morning but I want to mention it again perhaps the reinforcement um the worst pipe welder for duplex stainless steel is a guy who learned to weld pipe on carbon steel and the reason is because when he learned to weld pipe on carbon steel he learned cold fast root pass followed by Hots and that's a long thing to do with and especially with this is a um heat affected Z from a um I'm sorry this is this is root came from a um customer complaint that is failing toughness in the root put in a thin root pass followed by a hot and the end result was that the roof pass solidifi fast enough that some nitrogen was held in solution in the some cold Nitro don't know but the hot pass reheated it and caused secondary tinite to precipitate and sign pretty hard to tell which is the secondary tinite which is the signaries in the center of those High areas but that's a mixture of secondary oate and Sy so the right way to do things is to have the heat input of the root path be at least as high as the heat input for the second pass we have to protect the root pass cuz after all in a pipe that's what's going to be exposed to the cor and I will also tell you that if you have some interner metallics and some chromium nitrites in the inter interior of the material where it's not exposed to the corroding it doesn't hurt your impacts it doesn't hurt let it be but if it's at the surface is a different story at the surface is a very different story and it's the root pass that's at the surface now one other thing that hasn't been mentioned yet is the fusable hydrogen stainless not stainless you don't have to worry about diffusable hogen the hell you don't you have to worry about hydrogen and duplex stainless steel especially when the fite content gets high you know the specifications will tell you oh you can use 30 to 70% F if you got 70% F and you got appreciable diffusable hydrogen you're going to have a problem there was a 2205 pipeline FR prob remember this one in Malaysia back in the um early 80s I think it was 20205 welded with covered electrodes it was welded above ground on stands because the area where was being laid was sloy so they didn't put it into the trench until after all the Ws were made and um they went to lift the pipe off the stands with a whole series of cranes and lay it into the ditch they very slightly deformed because not all the cranes lifted at the same rate and Wells broke they got to looking at it it was hydrogen it was a high fite weld I admit to that it was a very high fite W but but it was also well made with covered electrodes that had been exposed to the very dry environment you had in the Malaysian pen it got kind of sck and they produced hyrogen cracks and um oddly enough I managed to do that in the lab when a fo wi before I realize what was going on I was looking at high so this was a uh 2550 F Type L it had a fa number of about 80 and what you could see on this is the fracture end of the tcil speci a half in diam pencil speci right here you have a very large fish eye you guys work with low alloy Steels you know what I mean by fish they are crystallin fractures caused by diffusable H and the tensil elongation of that stuff is only about 10 to 12% stainless steel should be treated as when you're welding it as if it were something requiring low high that means you keep your filler Metals dry especially once shielded filler metals and that includes some I'm going to keep it dry no different from welding High Str now if you buy filler metal that's at the low end of the fite range in the 30 to 40 range you're pretty much immune to this but the problem is the specifications are saying 30 to 70% when you get up to the highend of that range you're susceptible one thing I want to tell you you never want to do is use gas Thon Arc and put hydrogen in the shielding Cas it's okay to put it in the root gas when it doesn't get into the AR don't put it in the shielding gas CU hydrogen will get into the L and then if you have a high F Well metal you're susceptible to exactly okay so we get to the uh I just want to EMP reemphasize one more thing and that is if you remel duplex stainless steel by G by ging you're going to get high fairing if you try to dress the surface with gas thps and AR without thr metal you're asking for trouble you don't want to do that I thinken said that this morning and I completely agree with her there so the rules of the game to well duplex stainless High nickel filling material avoid High dilution welding or if you are going to do high dilution welding you're going to want to anal afterwards you know the longitudinal seam in pipe is often times welded by plasma art Keyhole with no filler metal or very little FAL or by submerged art with very high dilution process and as long as you knel the pipe afterwards that's just fine you also have to avoid the entire temperature range from 300 to about 1100° C or you got to heat treat at 11:20 to be sure you going dissolve all the undesirable things and water clinch and you got a water clinch more or less directly from 1120 uh or at least above um 14 now the purveyors of the ferium 255 BH matal have insisted for a long time that a kneeling had to be done at 1040 de shouldn't KN at a higher temperature because you'll still have some residual nitrogen inite and that might cause Ching nitrates and so one of the things I worked out in my my uh years of Decay was a call a step anal where the initial analing temperature was 1120 and then we drop the temperature in the furnace to 10:40 hel it at 10:40 for only a few minutes and then you a get out water of Quench and then we could get good meal properties in the LA 255 well met and we could follow the recommendation so that stepan I know is being used other devices I assume it refer to so with that I'll conclude by saying as was said this morning loing of D stainless is different not follow the rules everything will now one one I would disagree with one that was said this morning about Distortion I think uh you you said it and I think uh maybe Ellen said it too that you get less Distortion with DX well that depends one of the things you have to appreciate is that duplex is a whole lot stronger than Austin if you allow it to free move tintic stainless will move more more than D but if you do a half-ass job of restraint duplex will move more than El if you're going to restrain duplex from W you don't want to use bungee CS now you want to to use massive restraint to keep it fromy because it develops strength more rapidly and is stronger not itics it will pull more if you do a light restraint job you got to do heavy restraint so with that I'll close and open the Flor questions you have

thank you Dan uh those of you who know me uh probably associate me still with L electric go back further with but uh I retired in 2007 so uh I&#39;ve been out circuit and a year ago I relocated from the fand area to Chapel Hill North Carolina so that shoveling snow is no longer than my job could we kill the lights in the front row I make okay so in logical order this would have probably been the first presentation of the day but it wasn&#39;t for my uh anyway uh what I&#39;d like to start is why Earth dlex St Fields exist in the first place well the answer is a chloride stress corrosion PR we talked about all kinds of corrosion tests this morning and it was always pitting the real reason they exist isn&#39;t pity it&#39;s chloride stress corrosion P the problem is that the tintic stainless Fels like 304l and 316l are highly susceptible to and CRA the partic stainless Steels are highly resistant to Chloride stress C Chlor the fidic stainless Steels however are really ugly Critters for what reason is green the toughness vanishes when you w done C of graen duplex stainless Steels then coming to the picture as a way to combine properties of Austin stand and the properties and people now you guys all know Murphy&#39;s Law right anything can go wrong will go wrong and if you apply Murphy&#39;s Law to alloy design what would Murphy&#39;s Law tell you about this situation it would tell you that if you combine two alloy systems you will get the worst of f but that&#39;s not what happens with duplex duplex stainless steel is a clear violation of mer La so anyway in order to get chlorine exess corrosion cracking you need chlorides you need tensil stress uh tensil stress stress is automatic with Wells so you can always always always anticipate yield Point residual stresses around a well better so you&#39;re always going to be susceptible to Chloride stress corrosion cracking in the Alloys like real 4 316 that&#39;s why duplexes are interesting this is a a picture of chloride stress corrosion crack or a whole lot of chloride stress corrosion cracks this is in 304 stainless 304l on the left side you see the unetched um appearance and the signature of chlorid St and cracking is crack branching so you have a crack at the surface that becomes a multitude of crack subsurface and on the right hand side you have the same picture but uh it&#39;s in this area here and it&#39;s um enlarged and the samples the etch and the reason for the etching is to show you that the chloride Express corrosion crack doesn&#39;t care about grain boundaries you can see here you have grain boundaries and you have a crack going right through a grain it jumps from grain to grain to grain it doesn&#39;t care about grain boms this has nothing to do with sensitization it has everything to do with chlorides and chloride stress corrosion C now you got two ways to cure chloride stress corrosion crackless steel one way is lots of fair and that originally LED people into all kinds of advancements in what I call third generation ftic stainless Steels these are the ftic stainless Steels like 44 18 Chrome 2 mly very low carbon very low nitrogen um or 261 is another alloy 26 Chrome one Molly and very low nitrogen very low carbon they&#39;re ugly to fabricate in any kind of thickness Beyond sheet metal so we try not toat the other alternative um was to go with lots of get away from um compositions that were susceptible and that takes you into ni based Alloys or the soall super atic the about 25 nickel 20 Crome or something like that duplex turns out to be the cheapest alternative and the most friendly because you get those fabrication properties that are like Austin stain St in particular grain growth doesn&#39;t hurt as it does with okay so in a duplex stainless steel you got approximately equal amounts of fite rate some say 50% 5050 or something these things give you resistance to Chloride stress corrosion cracking like the peridic stainless Steels they give you fabrication properties that are more like atic stainless Steels and they give you a bonus doubling of the yield that was already mentioned now how do duplex stainless Steels come out come about what you see here is a um section through a turnery iron chromium nickel diagram at a constant iron level okay so it&#39;s 70% iron and what you&#39;re doing as you go from left to right across this diagram is you&#39;re going from no nickel up to 20% nickel and from 30% Chrome down to 10% Chrome and then up at the top here we have liquid and in the center here we have have this triangle tic triangle that&#39;s called that&#39;s where liquid plus ferite plus tinite coexists it&#39;s a very small ring where they closes all three phases at once but the important thing about it is that an ordinary tic stainless steel like 304 L for example is going to have a composition very close to the edge of this triangle so that the metal is going to solidify as liquid going to Liquid Plus f going to liquid plus ferite plus tinite then into ferite plus tinite and eventually down to 100% tinite that&#39;s why your 304 base metal doesn&#39;t contain any fa however if you take that 304 base metal and you melt it with a gas tungsten AR torch no filament just melt and then measure the fite you will find there&#39;s some f and that&#39;s because during solidification you get some formation of ferite and some of that ferite survives to room temperature to get rid of it you need to cause chronium to diffuse out of the fite and nickel to diffuse into the fite fite and the and the tinite have different chemical compositions all right now a duplex stainless field starts out further to the left in the diagram someplace over here where solidification goes from liquid to fite the Liquid Plus fite to 100% fite and 100% fite exists for an appreciable temperature range and then as you cool further it enters into the two phase field where ferite and tinite coexist and that ferite plus tinite will coexist until the cooling has gone so far that nothing else happens and that&#39;s what you freeze in at room temperature now how easily the austinite forms depends upon how high the temperature is where the first aite if the first austinite forms at a high temperature then diffusion is pretty rapid chromium can partition to the ferite nickel can partition to the tinite and it&#39;s not very dependent upon the fuid but if the first a doesn&#39;t show up until a rather low temperature now the diffusion of chromium and nickel is very limited the end result is that you don&#39;t partition chromium and nickel and that&#39;s where you can get into trouble with these things because you&#39;re locking in very high Farah you don&#39;t get anywhere near equilibrium transformation laate and the end result is you get very poor mechanical properties okay that leads us to starting to talk about the duplex stainless Steels in terms of generations um you know people talk about dlex stainless steel today as if this is really new technology but two Flex stainless Steels are older than I am those things existed already in the 1930s you see these two allo CD4 MCU which is kind of Forerunner of 255 that alloy has been around since the 1930s that alloy was a cast Al c4c was a cast designation castings are always kned afterwards so if you had to do any repair welding on it you would use any more or less match FL material and you wouldn&#39;t yal it then it was okay but in the as welding condition I properties off 329 likewise the few that was a rock Mater very popular in Japan not so popular in the United States um but as a rock material has some very interesting properties uh again in the Asal condition you would consider unweld because the mechanical properties and the corrosion properties are very important it&#39;s only after kneeing that it develops the properties problem is that you get in the as welded condition virtually 100% all right and then the properties are those of AIC stainess steel with very coose grains in other words very poor tpic uh also because there was a little bit of residual nitrogen in it you got Chum nitrite precipitates so the corrosion resistance was also they had to AAL at a temperature around 1100° C or higher in order to get a good world and what you were really doing was adjusting the balance of aite versus second generation happened by accident I believe it happened um in the late 70s and became popular in the 80s the thing that triggered it was the development of the Argon oxygen decarbonization system for refining stainless fields in the Argon oxygen decarbonization system you BL argon um with a small amount of oxygen in it into the liquid metal the oxygen reacts with the carbon takes the carbon out and the Argon is the carrier to help the bubbles remove get removed from the puddle from the U melt but um when you read the books about when aod was developed what you find out is that people right from the get-go with aod were using air as a source of oxygen and air is 80% nitrogen so they accidentally got nitrogen into the metal and voila the weldability improved improved market so that in the second generation of duplex stainless Steels the ones that proliferated in the 1980s uh typified by 20304 and 2205 um those materials had about 12 points of nitrogen with 12 points of nitrogen in them the heat affected Zone was now somewhat sensitive to heat input if you had high heat input you would get good heat effective Zone proper if you had low heat input you would not get good its own properties it all depending on how much time that temperature the material had in order to produce a decent amount of O Okay and when this generation of of U base materials came about in the 1980s it became very popular to adjust the th metal composition to 9% nickel instead of 5% so that you would get more tinite in the well metal that didn&#39;t do a thing for you for the heat effect itself so you had to control the heat input in order to get a decent heat effect now I want to point out one thing here 20205 in the literature was always associated with UNS number s31803 and that&#39;s very important because it allowed for a rather minimum nitrogen cont 8% minimum nitrogen in order to satisfy UNS 3180 now um two Japanese gentlemen uh professors oawa and koseki um back in the 1990s did a remarkable study that was published in the Wilding uh about alloy partitioning in duplex St speak the trouble with the publication in the welding Journal is that in those days the welding Journal did not publish in color in the recent supplement so that I don&#39;t think very many people appreciated the tremendous discovery that oala and keki had made but they had a second publication of it that appeared in the IW commission 9 and I saw that and it was one of those aha I get it so let me go through this uh in a little detail to to point out what they discovered and how important it is for us okay so what you have here is a rock 2205 type Bas material uh in the upper left corner you see the base material in the rock form and it you can see the layered appearance the grayer material is very the white material is aome and then what they did was use an electron microb to look at the distribution of the aloine elements in that BR material and each of the maps is color coded so that you can see where the high concentration of an element is and where the low concentration is so if you look to the fite area here and you look to the chromium map that corresponds to it you see the red color in the fair and that corresponds to 25% chrom the alloy only contains 22% conversely if you look through the austinite area which is blue here that corresponds to 20% so there&#39;s a partitioning chromium promotes ferite so it&#39;s segregates under equilibrium conditions conversely if you look at the nickel now what you see over here in the ferite area is low nickel and now the blue color corresponds to 5% nickel the alloy is nominally 6% nickel so you got 5% Nickel in the fite and over in the austinite area you have the red color which corresponds to about s or 7 and 12% so the nickel has also partition if you look to the mum map the mum map looks just like the chromium map Rich mum area is the fite the low mum area is the A and now we&#39;re saying the alloy is 3% mum but the fite contains almost 4% mum and the aite contains 22% so there&#39;s a partitioning ofel of the LI the most interesting alloy element however is the nitrogen now in the nitrogen color coding blue corresponds to zero red corresponds to about 253 something in that order and if you look in the farad areas it&#39;s zero you look in the tinite areas it&#39;s around 2.25 for an alloy that contains 12% nitrogen so it&#39;s an almost complete partitioning of the nitrogen to the tinite and away from the now if you take the same material and you make an autogenous well done this is what you get myogenous guest same material and this this relates not only to the L metal but to the hottest part of the heat effect and it&#39;s critical that I bring in the hottest part of the heat effect Zone although o c showed this as well metal not as heat Zone but it applies to the heat effective Zone as well when you look at the micro structure in the upper left corner now what you see is a big fite grain out lined by little plates of austinite on the grain B and then in the center of the ferite grain there is all this black specs the chromium nitrites which I think I hav showed this morning and if you look very closely you can see next to the austinite ples along the grain boundary there&#39;s an area that&#39;s clean doesn&#39;t have black no call me okay now if we look at the partitioning alloy elements you look at the chromium map there&#39;s no partitioning evidence if you look at the nickel map there&#39;s no partitioning evidence what you can see a little of in the nickel map is these little hexagonal areas that are slightly more yellow in the outline and green in the center okay that tells you solidification for doesn&#39;t tell you anything about the transformation when fite transformed a this has to do with solidification if you look at the malum map again you can see a little bit of the kind of coring or the yellow hexagon hexagonal shapes in there but you don&#39;t see anything that suggests the micro structure up here however when you look at the nitrogen distribution you can see the ferite grain very clearly outlined by nitrogen which is inite plates on the bra you can see in the center of the grain an essentially nominal nitrogen content that&#39;s caused by a mixture of chromium nitrites and Par and next to the grain boundary you can see the zero blue color zero nitrogen blue color which means the nitrogen had a chance to to get out of there so nitrogen is diffusing towards the aite out of the fah near the grain boundaries the nitrogen can get out but in the center of the grain it can&#39;t doesn&#39;t have enough now you might say well what if I wanted to get rid of those nitrides what would I do I might take the nitrogen out but then I have a peric stainless Ste 100% turn a better way go and this was referred to by own again this morning is put in more nitrogen instead of 12 points now we have 18 points in nit otherwise it&#39;s the same composition 225 now when you look at the base metal micr structure you can still see a big grain outlined by aite plates but there are all kinds of aite plates crisscrossing the grain internally now when you look at the alloy element mapping you can see a little bit of the chromium partitioning if you got really good imagination here I think you can see there&#39;s a tendency for a blue outline here that&#39;s a farag grain boundary which is completely covered with Austin this tells you that the transformation of the first formation of austinite took place at a higher temperature where nickel or where where chromium had a chance to diffuse out of the A and if you look at the nickel map you still can&#39;t see anything but the little hexagonal shapes nickel is the slowest diffusing of all the elements if you look at the malum map you can again see in fact you can see it better now that blue outline light green where the mum is bailing out of the areas that transformed aate and if you look to the nitrogen map you can see the same big pair of grain outlined by high nitrogen on the grain B you&#39;ve got all these ostanite plates inside the grain which are also full of nitrogen and then you&#39;ve got ferite areas that are completely clear of nitrogen now you got good properties it&#39;s the nitrogen that makes the alloy Welling that promotes formation of tinite at high temperatures and diffusion can go on so one of the things we like to do now is as a result of this we try adding nitrogen to the shielding gas to keep it in the weld metal we increase the nickel content in the filler metal at 9% to help the weld metal and with this addition of nickel We Now call the Chlor metal 20209 instead of 2205 to tell you that it&#39;s 9% instead of 5% but this brings us to the third generation of duplex th seals and 2205 appears in this generation as well but now it has a new UN number instead of 31803 it&#39;s you see 32205 the asme code in year 2000 redefined the 2205 Bas CH by this UNS number 32205 the minimum nitrogen content in that alloy is now 14 points instead of eight and that has a big effect on the wabil now the heat effective zone is not very sensitive to heating you can go 12 K per mm to 22 K per mm and still get good and Ne effective Zone c big Advance partly for historical reasons but partly also to improve the toughness of the W metal filler Metals still tend to be high in. and now we start looking at pitting resistance but it&#39;s only after we got to the third generation of duplex stainless Steels that we thought at all about pting as I said earlier there was chloride stress corrosion cracking that was the main reason and now we can look at this pit resistance index which includes nitrogen and I&#39;ll give you a slightly different version of the fitting resistance index because there&#39;s a factor for Tungsten in it I think when you saw the one this morning you saw it without tungsten tungsten is considered to be half as effective as mum in terms of dividing resistance to pting otherwise from for Pon free alloy pitting resistance Index calcula this way is the same as the way you saw it this morning but for 0 on 100 which contains deliberate additions of tungsten this matters so the third generation of duplexes have rather modest pitting resistance they got pitting resistance index numbers in the low to mid-30s what you would like is to get up to 40 because then you can be competitive with the so-called super Austin these are the Alloys of uh 20 Chrome 25 nickel like 254 Smo or al6xn and that brought us to the fourth generation of H like stainless Steels and these are the so-called copics there are two main Alloys in the r form 2507 and 01 100 and here you can see 01 100 contain in addition of tungsten which gets its pity resistance index up over 40 and then the cast version CD3 MN is the cast version 2507 and CD3 M WC anyway um that&#39;s those are the cast equivalents uh for 257 and 0100 so that&#39;s your fourth generation of uh pric stainless or duplex stainless Steels mechanical properties of the super reflex stainless Steels are really outstanding this is like hy80 steel with corrosion resistance you&#39;re looking at 8 C I minimum yield strength in the super um you will notice that the uh yield strength numbers for cast materials are low and that&#39;s caused by the fact that when you unal the material the micr structure TS topiz and part of the strength in a duplex stainless steel comes from the layered structure of parite and when you lose that layered structure you lose some structure you have to be used to that notion where do we apply these things I was run through a bit this morning there&#39;s a lot of environments including with these super duplexes sea they&#39;re resistant to seaw W that&#39;s quite impres people make things out of 316 stainless and they put them in C1 and understand why it Corrs CU gosh it&#39;s stainless it&#39;s not stainless in SE water okay so that&#39;s just kind of a list of things that uh you&#39;ll find um one of the interesting applications for the super duplexes is the blue Gast dlers blue Gast desulfurizers have high temperature areas and low temperature areas you don&#39;t use super duplex in the high temperat part of the gas but you do use it in the low temperature and it&#39;s quite resistant to the acids and the chlorides and whatnot that exist in that part seems a little odd to think about the way this happened but after the fourth generation became popular there was a kind of a relook at the low alloy the socaled Lees now originally I had put 2304 in the um second generation of D but I think today we would reassign it to the lean duplexes which come really came after the fourth generation but the 2101 alloy is probably the most prolific today of the lean duplexes this alloy is very low in nickel only one and half% but it&#39;s got 5% mes now the function of the manganese is not to promote a you have to get that notion out of your head manganese does not promote a those of you who know the shuer diagram and see manganes in there as an atiz it get that out of your head it doesn&#39;t work but what mainly does do there&#39;s two things one is if you get austinite it enhances the stability of the austinite with respect to transformation to Martin site at low temperature so that 5% manganes prevents formation of Martin side and low but the other thing that it does is it enhances the solubility of the metal for nitrogen and it&#39;s nitrogen that really makes the oite form in the first place anyway the idea of this alloy and the other lean duplex with 2304 and 244 I think mentioned in some of you the idea with those is to get corrosion resistance that&#39;s more or less competitive with 304 L but do it with very little me here you&#39;re talking about 1 1/2% nickel instead of roughly 9% nickle in a 30 or four hour now an interesting thing about these things I I told you with third generation duplexes that they were not sensitive to heating with normal aring processes I should have ADD I didn&#39;t Del the Third generation of duplexes can be welded by The Arc processes they cannot be welded by resistance weld they cannot be welded by laser be because the metal cools too fast even with however 2101 is different and mic did some work on Resistance welding of 21101 and you have to do some tricks with 21101 if you&#39;re going a resistance welding the idea here again is to get some transformation to austinite in the as welded condition in the normal resistance well you get a little austinite on the grain boundaries and a little austinite inside and a lot of chromium nitrates but but in a resistance welding machine after the resistance welding cycle itself is done you know you&#39;ve made a resistance weld you can pulse the resistance welder a second time a 2C puls of energy 2 seconds drops the ferite content from 83% to 74% and what it particularly does I&#39;m not sure that you can see it as well here as you might see it in a paper version but the amount of the uh chromium nitrides in the dark areas here is greatly reduced they&#39;re discontinuous you can live you can get useful mechanical properties out of a resistance weld that is post heated with this little 2C post heat treatment you can get useful mechanical properties and useful corosion resistance so you can make things like dimple jacket Heat exchangers by resistance color Metals for these things are classified both by AWS and in the ISO standards so what you see in AWS is uh 2209 which is directed at 2205 type base material the 2307 is intended for the lean defens the alloy was originally developed for 2101 but it also works 24 um that alloy is quite different in composition it has quite a bit of nickel compared to the 211 base material um the reason for that is in the ASW condition if you have a a 21101 composition and you pick up oxygen from a flux shielder process you get lousy mechanical properties particular particular very poor toughness and even very poor t but by depending on nickel to help you out you get better confidence so the 2037 composition or 237 NL as is stand is adequate for 2553 exists only in AWS it doesn&#39;t exist in um European or is but this is essentially a matching filler material for the pellan 255 which shows up back the this is the 255 255 this is the rock composition now if we go to the filler metal the 2553 matches that composition exactly and you would use it primarily on cast material not on rock material because with that Niel it&#39;s going to have l properties a has good the 2593 which is the next one down that is the filler metal that would be recommended for the ASW condition so you actually have two fill models one for the for the As for the castings and one for the okay and then 2594 is a filler material intended for 257 and the last 12595 is for the Zer 100 now in most practical application of super dupes people use 2594 and 2595 interchangeably for Z 100 257 and I don&#39;t think anyone has ever shown any convincing evidence that one or the other is better now we talked about hyper duplex that was mentioned again this morning uh there are two hyper duplex Alloys that have seen some appearance in the literature um 25 or 2707 which is a um 27% chromium 7 5 m and4 nitrogen has a PR that doesn&#39;t quite 50 but with it then there&#39;s a second one this 3207 which is 32 Chrome 7 Niel but it&#39;s not so high and that actually does have a above 50 the filler material that is proposed for these things by sandic is the Pryor of the material is 27 Chrome 9 Nickel 5 Molly 24 nitrogen and it&#39;s the same filler material for that filling material is only available as gas tunks that AR fing material that&#39;s the only welding method that&#39;s ever been used on those things as far as I know in any to any extent and I have at the moment I don&#39;t know that there&#39;s any commercially significant application of them manufacturer was very aggressive towards par that&#39;s the only one they okay all right thanks but anyway this is sort of like maybe that&#39;s what&#39;s in the future maybe not if it&#39;s not a commercial success soon I think it will matter okay now C of the fide phase in the newes there is a lot of possibility for precipitation these were mentioned also this morning uh the alpha Prime phase which forms in a temperature range of about 400 to 540 C the sigma phase which forms between 500 and 1070 C and the kai phase which can go up to 1100 C1 so that essentially prohibits you from using the alloy anywhere from below 400 to 1100 well that alloy doesn&#39;t have any useful properties above 1100 see um it&#39;s sort of like peanut butter terms of mechanical property so the real interest is in below 400 C now here you will see this is the temperature range where it forms where the alpha Prime pH will form in few hundred hours you go to 10,000 hours that temperature comes down and so people will say 300 see is the maximum service temperature for these Alloys or maybe even 250 so those are all limitations now the alpha Prime and brittle occurs when not an inel compound but when farite partitions into two ferites an iron rich ferite and a chromium Rich fite and if you look at the binary iron chromium phase diagram what you will see is that that two-phase decomposition of fah into an iron F and a chromium F when that occurs the chromium Rich phah contains over 90% chromium and that horribly depres chromium content in The Matrix and so the Matrix becomes very susceptible to corrosion the guy named Tavaris uh published a paper on the internet from which I borrowed this material what he&#39;s looking at here is stin wall pip so it&#39;s not sharpen impact Val you shouldn&#39;t allow you too much um but what he&#39;s looking at here is the um mechanical properties of uh 2205 material using the old 31803 number so I&#39;m not sure exactly what the nitrogen number is but U he&#39;s looking at time at 475° C on the horizontal axis and then on the vertical axis on one side we&#39;re looking at impact energy and jewels and these are very subsize sharpy specimens because um of the fact that it was thin all material so what you see in the uh impact results um Sharpies start out at a bit over 30 Jews and they drop in something on the order of100 hours to less than half and if you also look at the hardness of the material he&#39;s using Bell hardness so the hardness starts at a number in the the 220 area and it goes up with increase in time and temperature so that you&#39;re up in area of 350 after 500 hours so there&#39;s a serious Harding of the alloy and a serious emling of the alloy and if you look at corrosion what it was done here was to use a electrochemical uh pitting cell to um just increase the potential until you got the first surge of current in the electr chemical cell and that&#39;s the initiation of pitting then you stop the process immediately so instead of getting big pits that you can stick your finger into or stick a a ballpoint pen into or something like that you get little bitty pits and you can see here all the pits are in the dark etching material which is the P there are no pits whatsoever in the Aus you can&#39;t see the alpha Prime you need a scanning transmission electron microscope see that but you can see the pits that result of the CH depletion next to the AL so that&#39;s very destructive to the properties heat treatment of duplex stainless Steels especially when they&#39;ve been well it&#39;s a little bit of tricky distance you know cuz we&#39;re saying well you can go you&#39;ve got a whole temperature in from 400° C up to something over 1050 1070 de seat where bad phases for C phase p phase and the ASM standard a240 used to call for a kneeling of 2205 and a minimum temperature of 10 40° and that was just fine for the basic it was just fine for matching composition well done but it turns out it wasn&#39;t just fine for 9% nickel fill this was illustrated by a company that gave a paper at one of the duplex World conferences back in the Netherlands uh in the 1990s uh in which they they reported on fabrication of a very large dis head for 2205 the head was large enough that it could not be fabricated out of one plate of 2205 mat so they welded a number of plates together using 9% Niel filler and then they co-formed it to make this dish t uh and after they cold formed it they saw some damage to the mechanical properties in terms of L so they went to a kneeling the kneeling said 1040 DEC minimum so they went up to 1050 or 1060 I don&#39;t remember the exact water quench well the head wared slightly bring the analing so they needed to in it up a little bit so they put it back in the forming press and they were just going to give it a little bit of nudge to chew it up and it fractured the welds unzipped they literally unzipped and the and the plates came apart they investigated and found that the weld metal was loaded with signal interestingly enough I gave another paper at the same conference in which I talked about deep treatment in group like stainless fields and to begin with on that paper I referred back to an old diagram that was developed by a gentleman named Paul grer that worked for climax M company back in the 1970s and his diagram which he shows here which I&#39;ll show here um is looking at the phases that are stable as a function of temperature in a 25% chromium 35% lium alloy with varying amounts of nickel so what you have on the horizontal axis here is nipp content increasing from 0 to 30% and all of these Alloys contain the same amount of chromium andum so what you have here now our temperature feels for 100% fite farite plus tinite tinite farite plus Sigma fite plus austinite plus Sigma and austinite plus Sigma the important feature of this diagram is the sigma solvus boundary that&#39;s the boundary above which Sigma B and here you can see that as the nickel content increases the temperature the sigma Solis Rises so when we went from a 5% nickel base material to a 9% nickel thermal we raised the sigma salvus temperature by about 100° C Rob&#39;s diagram was there but no we saw well I was fooling around with duplex ferms at the same time and I got to looking into heat treating Heat Treating of these things because I was making some filler Metals for a purveyor of dlex casting so I made a weld metal of a 2209 composition IED it at 10 65° which is 25° higher than the as recommended minimum so I should have been safe and I wanted question and I had Sharpie specimens tensil specimens machine from it and a metal graphic well the first specimen that got addressed in my laboratory was Sharpie impact so I have a technician in the we call it affectionately the buing lab I was broke things speci and um I asked him to break specimens at room temperature to start out so he got this sample he broke it he broke the first couple impact specimens at room temperature then he came running down the hall of my office he goes doc doc you got to see this you got to see this what so I go on the busing landb and he says now watch this he puts the specimen in the impact specimen in the impact machine at room temperature hits the hammer now when you got a good impact specimen it&#39;s going to go th this one went pink three jewels fulls size impact not a subale fre then we measured the T elongation and it was about 10 then I looked at the micro structure and here was the answer all these white Islands islands that are outlin but not fetch signif H dark stuff of course that&#39;s the and the white stuff would be a in the as welded condition I measured the ferite content of that and it was 45 number I use magnetic 45 in this condition was 22 Fad in other words half of The Fad was six so I said well maybe we just didn&#39;t a meal it long enough so instead of 2 hours at 1065 I went to 72 hours and 65 and I got exactly the same thing it wasn&#39;t that I hadn&#39;t heated long enough to dissolve the signal it was that the phase was stable at 1065 then I took it up to 1100° 111 but anyway after that AAL I got good impacts I got good ductility and I measured the fite number again and you know what it&#39;s was 45 again the reaction was exact was completely reversible I could get the sigma to go back into solution and I got far now that was kind of interesting to me because when you start out with um High fite glob uh let&#39;s say you get a 60 or 70 fite number in the ASW condition and you anal it you will see the fite number drops drops appreciated but if you start out with a fite number around 45 doesn&#39;t Dr and the reason is because the alloy started to form tinite at higher temperature and you got very close to equilibrium partitioning very close the equilibrium out of a So when you say is the um fite stable question really is how much fite is there you got a lot of fite no it&#39;s not all stable but if you only got a little bit of fite yeah it&#39;s stable okay so as a result of that asdm RIS ASM a480 to change the ining temperatures for the 2205 and similar now it was mentioned this morning that um you never want to make an autogenous well in a duplex stain because you will have horrible mechanical properties um when you make the root pass in a pipe you generally do that by gas T to AR and welders don&#39;t like open Roots as a matter of fact they hate them and they hate you for telling them you have to make a well with an open roof but the conventional wisdom is you got to use open roots and here we&#39;re talking about an open a root Gap that might be um 2 to 3 mm uh some people say 3 to 4 mm depending upon which U ring process you&#39;re using now this stt that&#39;s a surface pensent transfer that&#39;s a Lincoln craving for um gas metal AR short circuit and transfer with a controll BL and there are other people make similar material and they all come for the 3 to4 mm recommendation but anybody in this room I&#39;m very big at least three on okay I&#39;m going to be careful all right I&#39;m going to tell I&#39;m going to tell you two things nobody believes filler metal producers because everybody knows filler metal producers why so they always check on us I&#39;m a filal producer I&#39;ve been checked this time they don&#39;t believe us when we tell us how much F and something they don&#39;t believe us when we tell them how much pencil strength they&#39;re going to get or how much they checkup well shouldn&#39;t believe welders either they cheat and when you tell them that they got to have a 2 to three or 3 to 4 mm root G you&#39;re not looking they might squeeze it down to make their job easier it&#39;s much easier to be Bridge closed gr than open am I telling the truth so far you guys always kind of okay so what are you going to do about this as an engineer are you going to let the welders Chief in the next thing the way a lot of fabrication is done particularly around the North Sea where there&#39;s a lot of duplex in walks is to put a little dent on either side of the joint preparation on the surface and the dent is put there fixed distance away from the edge of the joint so that it won&#39;t be melted by the welding it&#39;ll still be there afterwards that&#39;s called a witness then you measure the distance between those two dents after the welding is done and you can catch the welder mating except that a do well duplex sh 3 4 mm sometimes even more that&#39;s all figured in though that&#39;s all figured in those things are called witness Mars I don&#39;t know if they&#39;re still popular in the north CA area Fred probably knows more about that than I do but I know they were very popular at one time they still using them Fred okay all right so I had to tell you about color metal color metal producers BL I Told You chea So share the blame all right okay now High nitrogen works well for control th the heat effective Zone but nitrogen tends to be lost from the weld pool I think Ellen referred to that this morning so it&#39;s common to add some nitrogen to the shielding gas particularly gas and art Welding which has the greatest tendency to lose nitrogen and I will confess to you right now that if you use even 2% nitrogen in the shielding and gas tungston un your tungsten is going to deteriorate more rapidly than it would without the nitrogen emiss so that I can say come deal with cuz that&#39;s the only way you&#39;re going to get good properties in the super reflex St you got to get the nitrogen in the filler Metals we still use the 9% I mention this is the kind of micr structure you can expect if you add nitrogen to the shielding gas and a super duplex well this is4 257 base material here you can see the heat affected Zone has a little more ferite in it than the base material has a little more F but it&#39;s enough austinite there to prevent any extensive nit nitrite precipitation I&#39;ll go back I&#39;ll go back to Julia&#39;s Julian&#39;s talk this morning you will never see Zero inner metallics or zero nitr I agree with you completely going to be some there question is are that&#39;s the kind of micr structure you should have um this is looking at gas in our quality with pure argon versus 5% nitrogen and with pure cure argon you&#39;ll still get pretty good impacts but you&#39;ve got these pretty large fade areas which are at risk for getting nitrate crucification whereas when you add the nitrogen to the shielding gas you have very little in the way of large F areas and therefore a much less tendency to get the now another thing that was referred to this morning but I want to mention it again perhaps the reinforcement um the worst pipe welder for duplex stainless steel is a guy who learned to weld pipe on carbon steel and the reason is because when he learned to weld pipe on carbon steel he learned cold fast root pass followed by Hots and that&#39;s a long thing to do with and especially with this is a um heat affected Z from a um I&#39;m sorry this is this is root came from a um customer complaint that is failing toughness in the root put in a thin root pass followed by a hot and the end result was that the roof pass solidifi fast enough that some nitrogen was held in solution in the some cold Nitro don&#39;t know but the hot pass reheated it and caused secondary tinite to precipitate and sign pretty hard to tell which is the secondary tinite which is the signaries in the center of those High areas but that&#39;s a mixture of secondary oate and Sy so the right way to do things is to have the heat input of the root path be at least as high as the heat input for the second pass we have to protect the root pass cuz after all in a pipe that&#39;s what&#39;s going to be exposed to the cor and I will also tell you that if you have some interner metallics and some chromium nitrites in the inter interior of the material where it&#39;s not exposed to the corroding it doesn&#39;t hurt your impacts it doesn&#39;t hurt let it be but if it&#39;s at the surface is a different story at the surface is a very different story and it&#39;s the root pass that&#39;s at the surface now one other thing that hasn&#39;t been mentioned yet is the fusable hydrogen stainless not stainless you don&#39;t have to worry about diffusable hogen the hell you don&#39;t you have to worry about hydrogen and duplex stainless steel especially when the fite content gets high you know the specifications will tell you oh you can use 30 to 70% F if you got 70% F and you got appreciable diffusable hydrogen you&#39;re going to have a problem there was a 2205 pipeline FR prob remember this one in Malaysia back in the um early 80s I think it was 20205 welded with covered electrodes it was welded above ground on stands because the area where was being laid was sloy so they didn&#39;t put it into the trench until after all the Ws were made and um they went to lift the pipe off the stands with a whole series of cranes and lay it into the ditch they very slightly deformed because not all the cranes lifted at the same rate and Wells broke they got to looking at it it was hydrogen it was a high fite weld I admit to that it was a very high fite W but but it was also well made with covered electrodes that had been exposed to the very dry environment you had in the Malaysian pen it got kind of sck and they produced hyrogen cracks and um oddly enough I managed to do that in the lab when a fo wi before I realize what was going on I was looking at high so this was a uh 2550 F Type L it had a fa number of about 80 and what you could see on this is the fracture end of the tcil speci a half in diam pencil speci right here you have a very large fish eye you guys work with low alloy Steels you know what I mean by fish they are crystallin fractures caused by diffusable H and the tensil elongation of that stuff is only about 10 to 12% stainless steel should be treated as when you&#39;re welding it as if it were something requiring low high that means you keep your filler Metals dry especially once shielded filler metals and that includes some I&#39;m going to keep it dry no different from welding High Str now if you buy filler metal that&#39;s at the low end of the fite range in the 30 to 40 range you&#39;re pretty much immune to this but the problem is the specifications are saying 30 to 70% when you get up to the highend of that range you&#39;re susceptible one thing I want to tell you you never want to do is use gas Thon Arc and put hydrogen in the shielding Cas it&#39;s okay to put it in the root gas when it doesn&#39;t get into the AR don&#39;t put it in the shielding gas CU hydrogen will get into the L and then if you have a high F Well metal you&#39;re susceptible to exactly okay so we get to the uh I just want to EMP reemphasize one more thing and that is if you remel duplex stainless steel by G by ging you&#39;re going to get high fairing if you try to dress the surface with gas thps and AR without thr metal you&#39;re asking for trouble you don&#39;t want to do that I thinken said that this morning and I completely agree with her there so the rules of the game to well duplex stainless High nickel filling material avoid High dilution welding or if you are going to do high dilution welding you&#39;re going to want to anal afterwards you know the longitudinal seam in pipe is often times welded by plasma art Keyhole with no filler metal or very little FAL or by submerged art with very high dilution process and as long as you knel the pipe afterwards that&#39;s just fine you also have to avoid the entire temperature range from 300 to about 1100° C or you got to heat treat at 11:20 to be sure you going dissolve all the undesirable things and water clinch and you got a water clinch more or less directly from 1120 uh or at least above um 14 now the purveyors of the ferium 255 BH matal have insisted for a long time that a kneeling had to be done at 1040 de shouldn&#39;t KN at a higher temperature because you&#39;ll still have some residual nitrogen inite and that might cause Ching nitrates and so one of the things I worked out in my my uh years of Decay was a call a step anal where the initial analing temperature was 1120 and then we drop the temperature in the furnace to 10:40 hel it at 10:40 for only a few minutes and then you a get out water of Quench and then we could get good meal properties in the LA 255 well met and we could follow the recommendation so that stepan I know is being used other devices I assume it refer to so with that I&#39;ll conclude by saying as was said this morning loing of D stainless is different not follow the rules everything will now one one I would disagree with one that was said this morning about Distortion I think uh you you said it and I think uh maybe Ellen said it too that you get less Distortion with DX well that depends one of the things you have to appreciate is that duplex is a whole lot stronger than Austin if you allow it to free move tintic stainless will move more more than D but if you do a half-ass job of restraint duplex will move more than El if you&#39;re going to restrain duplex from W you don&#39;t want to use bungee CS now you want to to use massive restraint to keep it fromy because it develops strength more rapidly and is stronger not itics it will pull more if you do a light restraint job you got to do heavy restraint so with that I&#39;ll close and open the Flor questions you have

Transcript for:Duplex Stainless Steels and Chloride Stress Corrosion Cracking

Transcript for:
Duplex Stainless Steels and Chloride Stress Corrosion Cracking