GB2335439A - Improved stainless steels - Google Patents

Abstract

A stainless steel alloy consisting essentially of 10 to 20wt% chromium, 0.025 to 0.2wt% ruthenium, 0 to 2wt% of one or both of nickel and molybdenum the balance being iron, and which displays improved corrosion resistance compared to previously known ferritic stainless steels.

Description

JRL1368 1 IMPROVED STAINLESS STEELS 2335439 The present invention relates

to a novel class of stainless steels which show irrproved corrosion resistance over cornmercially available 430 type stainless steels.

Sorne rrv--tals and alloys, in acidic environrne-nts, display what is termed "passive" behaviour. That is, in environments that one would expect to bring about active dissolution, certain metals and alloys are apparently passive, remaining undamaged by the acid. This is due to the formation of a thin film of oxide on the metal surface which inhibits the dissolution of the underlying metal. The tendency of the metal or alloy to corrode has not been eliminated, but merely suppressed. The most widely known example of this is stainless steel - iron alloyed with chromiurn. The thermodynamically less stable chromium oxidises to produce a passivating layer on the surface of the steel, 2Cr + 311,0 - Cr,0, + 611' + 6e-.

The iron also oxidises to produce an oxide layer, 2Fe + 31120- Fe,03 + 611+ + 2e-.

The range of environments in which a stainless steel will form a passive film is dependent on its composition and microstructure. As a general rule, a higher degree of corrosion resistance will be displayed by stainless steels containing large concentrations of chron-dum. Other elements, such as nickel and molybdenum, are also added to irrprove the corrosion resistance. The microstructure also affects the corrosion properties as well as the mechanical properties, for example inclusions, intern-ietaRic phases and areas depleted in a certain element may be susceptible to localised corrosion.

Although thermal and mechanical treatments can allow the mechanical properties of a stainless steel to be varied, they are ultimately dependent on the composition. For this reason, the stainless steel whose composition lends it maximum corrosion resistance in a certain environment cannot always be used as it may lack certain mechanical properties vital to the application. For economic reasons also, stainless steels 2 can rarely be manufactured with the ideal corrosion resistant composition or formed so as to be free from defects. The addition of alloying elements increases the cost of a stainless steel, as does production by ultra-high purity routes.

The commonly used stainless steels include 430 ferritic stainless steels and 316 austenitic stainless steels. The ferritic 430 provides moderate corrosion resistance in rnildly oxidising and organic acids. It finds application in food handling industries and food preparation areas such as equipment, sinks and working surfaces.

Within the automotive industry, 430 is used in areas such as windscreen wiper arms, mounts for wing mirrors, wheel covers, general trirn, fasteners and sections of exhaust pipes. The austenitic 316, while being more expensive than 430, possesses far superior corrosion resistance. In conjunction with the resistance to general corrosion in atmosphere and aqueous media, it shows resistance to pitting h] acetic acid, phosphoric acids and environments containing low concentrations of chloride ions. The high is corrosion resistance and case of fabrication (if 3 16 has allowed it to become the principal stainless steel for use in the chemical industry, being used for the manufacture of reaction vessels, piping and beat exchangers. However, it remains expensive.

It has been known for some time that the range of environments in which an alloy will display passivity can be expanded by increasing the rate of the cathodic process. This would appear to be counter-intuitive as increasing the cathode potential should increase the rate of dissolution. However, in an alloy that shows passive behaviour, increasing the rate of cathodic reaction can push the corrosion potential into the passive range. This is known as cathodic modification. The efficiency of the cathodic reaction can be increased by the addition of alloying elements more noble than the base alloy. According to Streicher (Plat. Met. Review, 21, 51 (1977)), the earliest report of this phenomenon was by Monnartz who, in 1911, reported that the corrosion resistance of Fe-Cr alloys could be increased in certain environments by the addition of platinum either by alloying or by contact with a platinum wire.

3 In studies of Fe-Cr alloys, Tornashov et al (Prot. Met. 11, 379 (1975)) reported tests carried out on Fe-Cr alloys with Cr content varying between 25 and 60wt% and PGM additions of 0.1 - 0.2wtc7,-. The corrosion rates were calculated by measuring the weight loss of the sample alloys in solution of sulphuric acid with concentrations of 10, 20, 40 and 50% at 1 00T. The tirne to passivation was determined by monitoring the open circuit potential. Palladium additions of 0. 15 - 0.2wt% were found to give the shortest times to spontaneous passivation, and the lowest corrosion rate in the passive region was found with 40wtc/,- Cr. Above this leve the passivation potential increased only slightly, while the current density increased substantially above 60wt% Cr.

In a subsequent study, Chernova et a/ (Corrosion, 34, 445 (1978)) determined that an iron alloy containing 0.2wt17r palladiurn and 40wt% chromium has optimal passivation characteristics.

Higginson et al (Corrosion Sci. 29, 1293 (1989)) studied the effects of alloying 0. 1 % and 0.2% ruthenium with an Fe-40Cr alloy on the time to passivation and temperature dependence in solutions of sulphuric acid. It was shown that Fe-40Cr 0. 1Ru and Fe-40Cr-0.2Ru spontaneously pasivateel in both 0.5M 14C1 and 0. 5M H2S04.

With increasing temperature, the time to passivation decreased.

Various further studies have been carried out looking at the effect of n-dnor additions of PGMs or other metals to Fe-40Cr stainless steels, for example Tomashov et al (Prot. Met., 9, 289 (1973)) and Tjong (App. Surf. Science, 44, 7 (1990) and App. Surf. Science, 45, 301 (1990)).

All previous studies indicate that a chromium content of approximately 40% is required for the addition of small quantities of PGM to have an effect on the corrosion resistance of the stainle,:.-. steel. The inventors have now discovered that it is possible, contrary to previous thinking and reported studies, to provide a corrosion 4 resistant stainless steel comprising the addition of ruthenium to a low chromium content ferritic stainless steel, ie a stainless steel with a chromium content of less than 20wt%. Small quantities of molybdenum or nickel or both rnay be added with beneficial results.

is Accordingly, the present invention provides a stainless steel alloy consisting essentially of 10 to 20wt% chromium 0.025 to 0.2wt% rutheniunj 0 to 2wt% of one or both of nickel and molybdenum, the balance being iron. The alloy may also contain up to 0.4wt% each of other metals or elements which are present in the alloy as trace metals or impurities.

A first embodiment of the invention provides a stainless steel alloy consisting essentially of 10 to 20wt% chromIurn, 0.025 to 0.2wt% ruthenium, the balance being iron.

A second embodiment of the invention provides a stainless steel alloy consisting essentially of 10 to 20wt% chromium, 0.025 to 0.2wt% ruthenium and 0.1 to 2wt% of one or both of molybdenum and nickel the balance being iron.

The stainless steels of the present invention demonstrate improved corrosion resistance over currently avalluble ferritic stainless steels (for example 430), the corrosion resistance being on a level with that shown by the more expensive austenitic steels. The stainless steels of the invention are however cheaper and more easily manufactured than austenitic stainless steels. The steels of the invention may be used in all the applications described hereinbefore for ferritic and for austenitic steels.

A further aspect of the present invention provides a process for the preparation of a stainless steel of the invention.

The present invention will now he described by way of example only, which is not intended to be lin-iiting thereof.

Preparation of Alloys Commercial type-430 stainlesssteel was acquired in 2 mm sheet forrrL For each melt, 10kg was weighed out and ultrasonically cleaned in acetone. The afloying addition of ruthenium for each charge was weighed out and pressed into a cylindrical pellet. The pellets were then sintered at 900'C for two hours in a nitrogen atmosphere.

The steel was melted in a vacuum induction furnace at the melting temperature of 160WC for approximately 30 minutes. The ruthenium pellets were added approximately five minutes before casting. Alloys of the following nominal compositions were cast.

1. 430 without alloying additions. (0 Ru) 2. 430 plus 0. 1 wtc/c, ruthenium (0. 1 Ru) 3. 430 plus 0. 16wtc/r ruthenium (0. 16 Ru) Samples were cut for analysis before the ingots were soaked at 120WC in air for 1 hour and then rofied down to 1Onirn thickness in seven passes. Analysis of the alloys was carried out by Are Ernission Spectrography (AES) and Inductively Coupled Plasma (ICP) AES; these result,, are reported in Table 1.

Table 1 - Composition of the fourstainless steels as analysed by sparkemission spectrography (Figures in italics are frorn 1CP-AES analysis) 430 0% R u 0. 1 % Ru 0.2% Ru Fe 81.381.3 81.7 81.481.6 81.678.1 Ru - - 0.093 0.162 Cr 16.4916.2 16.62 16.8216.3 16.5915.6 c 0.051 0.032 0.065 0.045 si 0.312 0.387 0.340 0.338 Mn 0.32 0.29 0.28 0.28 p 0.0307 0. 0 325 0.0315 0.0340 6 NI 0.144 0.133 0.143 0.135 M0 0.093 0.126 0.123 0.126 v 0.103 0.108 0.110 0.109 C0 0.025 0.026 0.026 0.026 Ti 0.007 O.W6 0.007 0.006 Cu 0.062 0.077 0.060 0.071 W 0.084 0.093 0.083 0.083 Nb 0.017 0.016 0.015 0.016 Sn 0.005 O.W5 0.005 0.005 AI 0.020 0.02 1 0.027 0.021 B 0.0007 WM7 08 0.0005 As 0.0132 ().() 10 0.0383 0.0029 N 0.000 0.000 0.028 0.000 Sections were cut frorn the rolled strips and annealed at 75TC for two hours followed by air cooling, in accordance with commercial practice as reported in the literature (J. R. Davis (Editor), ASM Specialty Handbook: Stainless Steels, ASM International, Ohio (1994)). During hot rolling, raplol cooling through 925'C can allow c austenite to be retained, transforming to iiiai,teii..,.lte at lower temperatures. The anneal was to transform any austenite/marten.site tk) fenlite, Two comn-iercial stainless steels were used for comparisons of the corrosion properties. The 316 stainless steel was supplied as 1 = sheet and the 904L was supplied as a 30x10x5 mm section cut frorn a much larger plate. No heat treatment was given to these steels as they were to be tes.ted in the 'as received' condition.

7 Age Hardening Age hardening tests were carried out on four alloys: as received 430, 0 Ru, 0. 1 Ru and 0. 16 Ru. Three ageing temperatures were used: 30TC, 40TC and 475C AD four alloys were aged at these temperatures for 10, 20 and 40 minutes, 1, 2, 4, 8 and 16 hours, 1, 2, 4 and 7 days. Each sample was approximately Icrn' in volume; these were cut before the ingots were rolled and annealed. After ageing, the hardness of the samples was measured using a Vickers Pyramid Hardness machine. For each alloy at each temperature, the change in mean hardness was calculated over the seven day period as this gives an indication of the increase in hardness with ageing time. The results are given in Table 2.

Table 2: Deviation betweeti initial and final hardness as a percentage of initial value.

30WC 4M C 4750C 430 5% 2 7 9, 15% 0 Ru 3% - 2 c/( 16% 0.1 Ru 3%, 1 C7t 12% 0. 16 Ru -3% 617f 13% At each of the temperatures measured and up to 168 hours (seven days), the addition of ruthenium has no detectable effect on the thermal ageing of the rolled and annealed alloys.

Grain Growth Studies Samples of 'as received' 430 stainless steel were annealed at 50C intervals from 10OWC to 120CC for 30 minutes wid then quenched. The samples were then prepared for metallography and etched with 109- oxalic acid for 75 seconds. The grain size of each sample was measured using Magiscan produced by Applied Irmging, 8 linked to a nkroscope fitted with a CCD camera. The data were plotted against temperature. The SO'C interval which showed the largest increase in grain size was analysed flulher, at ITC intervals. Samples of 0 Ru, 0. 1 Ru and 0. 16 Ru were annealed in the same manner over this temperature range, and their grain size measured. The results of the grain growth studies are given in Table 3.

Table 3: Results of grain size measurements after armealing at temperatures between 10OTC and 1200'C 430 0 Ru O 1 Ru 0.16 Ru 10000C 0.0 14 mm - 10500C 0.0 16 mni - - - 1 1000C 0.22 mm 0.19111111 0.12 nim 0.15 mm 11100C - 0.095 nini 0.090 nun 0.18 Min 1120'C 0. 15 nini 0. 13 nim 0.44 mm 11300C 0.44 nini 0.36 nini 0.37 Imm 11400C - 0.44 nim 0.42 nim 0.51 nun 11500C 0.67 nim 0.50111111 0.41 nim 0.51 MM 120WC 1.0 nim - - - The addition of rutheniuni shows a sniall decrease of the temperature at which grain growth occurs.

Electrochernical Testing Samples approximately 30x 1 0x2 i-ni-n were cut from the rolled and annealed sections and ground down to 1200---rit SIC finish (particles size = 3 pm). A c final abrasion was given just prior to the comniencenient of each test. This comprised abrasion with 1200 grit SiC paper in trichloroet byte ne (BDH Laboratory Supplies Ltd., Poole, England) followed by thirty seconds of ulti-aoiiic agitation whilst rsed in trichloroethylene.

9 The tests were carried out in a standard five-port glass cell with a platinum counter electrode and either a saturated calomel reference electrode (SCE) or a mercury/mercury sulphate electrode, connected via a Luggin probe. The solutions were prepared from analytical grade reagents (13DH Laboratory Supplies Ltd., Poole, England) and de-ionised water of unknown purity. It was de-aerated by bubbling "oxygen-free nitrogen" (BOC Ltd., Guildford, England) through a glass sparge pipe for minutes prior to the start of each test. After 20 minutes of de-aeration, the alloy sample was removed from the trichloroethylene an(] immediately placed in the cell, where it remaffied, above the surface of the solution for the final ten minutes of de-aeration.

Care was taken to ensure that the alloy was held above the splash zone.

The nitrogen flow rate was then reduce(] to a slight flow and the sparge pipe was moved to above the surface of the acid before the sample was immersed, to prevent any disturbance of the solution that rnight affect the measurements. The nitrogen flow remained on to prevent the irigress of air through any imperfect seals in the cell.

The top of the sarnple was. held in a crocodile clip and the lower portion immersed in the acid. The depth to which each sample was immersed was controlled in order to keep the sample area, of 2.()ciii2, constant between specimens.

For testing at higher temperatures, the cell was placed in a water bath.

The bath was heated by a thermostatically controlled immersion coil and constantly stirred to ensure a uniform temperature throughout. In order to allow the temperature of the cell and contents to equilibrate with the warer bath, the cell was left in the bath for one hour before the commencement of each test.

Two potentiostats were employed in the testing: an ACM Instruments Field Machine and an ACM Instruments AutoDC potentiostat.

1 () Poladsation Smeej)s: lm-nediately after immersion in the solution, the sample was cathodically pre-treated by polarisation to -0.8 V vs SCE for 10 minutes.

Then a sweep was started from -0.8 V vs SCE to 1.8 V vs SCE at a rate of 1 mV s-1, whilst recording the current density. The tests were repeated twice to check the reproducibility of the results.

Q2en Circuit Measurements: The samples were prepared as in the above tests with a cathodic pre-treatment at -0.8 V vs SCE for 10 minutes. The samples were then left at open circuit and the potential rneasured for up to four hours.

Corrosion in De-aei-(itel 0.1.44 Sulphuric Acid: The open circuit behaviour of the alloys as a function of time is shown in Figure 1. The rise from -0-48V vs SCE to -0.36V vs SCE of the 0. 1 Ru alloy, and the rise from -0.43V vs SCE to -0.37V vs SCE of the 0. 16 Ru alloy indicates that the alioy have passivated. Polarisation curves for the alloys are given in Figure 2. The various electrochernical pararrieters from Figures 1 and 2 are tabulated in Table 4.

Table 4: Results of the open eircuit potential (ocp) measurements and potentiodynamic sweeps in de-aerated 0. 1 M sulphuric acid.

ocp after 1pm. E,p (,,,A enj-2) (mA cm"2) E, 1000s (S) (V vs SCE) (V vs SCE) (V vs SCE) 430 -0.45 2.9 0.008 0.9 0 Ru -0.56 - -0.45 3.3 0.019 0.9 0.1 Ru -0.31 100 -0.33 0.09 0.024 0.9 -0.32 -20 -0.30 0.04 0.012 0.9 Corrosion in De-aerated 1.44 Sulphuric Acid: The open circuit behaviour of the alloys is shown in Figure 3. Tests were also carried out on 316 stainless steel and 904L stainless steel to give a comparison with commercially available austenitic stainless steels with corrosion resistance thought to be comparable to that of the 0.16 Ru 11 alloy. Potentiodynamic curves for all the alloys are shown in Figure 4. The electrochemical parameters from Figures 3 and 4 are tabulated in Table 5.

Table 5: Results of the open circuit potential (ocp) measurements and potentiodynamic sweep in de-acrated 1 M sulphuric acid.

ocp after 100s Epp iell, i E, (V vs SCE) (V vs SCE) (niA eni-2) (mA cm-1) (V vs SCE) 430 -0.34 19.5 0.01 0.95 0 Ru -0.54 -0.36 19 0.04 0.95 0.1 Ru -0.47 -0.34 17.5 0.12 0.95 0. 16 Ru -0.28 -0.23 0.15 0.04 0.95 316 -0.32 -0.26 0.03 0.02 0.90 904 -0.27 -0.26 0.07 0.02 0.90 Q12en Circuit Measurements awl Siib,,;eqlieiit Potentiodynamic SweO:

The samples were cathodically polarised for one nilnute at - 1.0 V vs SCE, the shaft of the working electrode was lightly tapped to reniove bubbles from the surface and then it was left at open potential for 30 minute,., Open Circuit Measurements wid Subsequent Potentiodynamic Sweeps in De-aerated M Sulphia-ic Acid: The open circult potential was measured for 30 rninutes and then a potentiodynamic sweep was initiated from the open circuit potential in an anodic direction at 1 mV s' to determine ",hether the alloy had passivated. The absence of an anodic peak in the subsequent potentiodynan-lic sweep indicated that the alloy had already passivated at open circuit. The open circuit behaviour of the 0 Ru, 0. 1 Ru, 0. 16 Ru and 316 alloys in de-acrated 1 NI sulphuric acid at room temperature is shown in Figure 5. The subsequent polart.zitioji curve are shown in Figure 6. The electrocherTfical parameters from Figures 5 and 0 are tabulated in Table 6.

12 Table 6: Results of open circuit potential measurement (ocp) and subsequent potentiodynan-dc sweeps in de-aerated IM sulphuric acid at room temperature.

ocp after 100 s active/passive (V vs SCE) 0 Ru -0.55 active 0.1 Ru -0.47 active 0. 16 Ru -0.30 passive 316 -0.45 passive The open circuit measurei-nents of this test allow a determination of the influence of cathodic pre-treatment on the corrosion behaviour of the alloys.

The open circuit measurernents and subsequent potentiodynamic:

sweeps were also carried out in de-aerated 1 M sulphuric acid with an elevated temperature of 6WC. The open circuit behaviour is shown in Figure 7. The subsequent potentiodynan-fic curves are shown in Figure 8. The electrochen-lical values from Figures 7 and 8 are tabulated in Table 7.

Table 7: Results of open circuit potential (ocp) measurements and subsequent potentiodynamic sweeps in de-aerated 1 M sulphuric acid at WC ocp after 100\ active/passive (V vs SCE) 0 Ru -0,55 active 0.1 Ru -0.43 active 0.16 Ru -0.29 passive 316 -0.41 active 0 13 Open Circuit Potential Measurements and Subsequent Potentiodynamic Sweeps in De-aerated 0.5M Hydrochloric Acid. Open circuit measurements were carried out for 30 minutes followed by a potentiodynan-fic sweep firom the corrosion potential in a positive direction at a rate of 10 mV min' to determine whether the alloys had passivated. Figure 9 shows the open circuit behaviour of the alloys. Figure 10 shows the subsequent potent io(l ynamic sweeps of the four alloys. The electrochemical parameters from Figures 9 zinc] 10 are tabulated in Table 8.

Table L Results of open circuit potential (ocp) measurements and subsequent potentiodynank sweeps in (]c-aerate(] 0.5 M hydrochloric acid op after I M.)s (V vs SCE) (niA ern-2) IP.

(V vs SCE) (mA ern-2) 0 Ru -0.51 - - - 0.1 Ru -0.49 -0.32 4.8 0.24 0.16 Ru -0.45 -0.31 206 0.10 316 -0.46 -0.31 0.48 0.07 Pitting Studies in 1M Sodiuni Chlori: Samples of the alloys, of area 2.0 cm ', were abraded with 1200 grit SIC paper and rinsed with de-ionised 2G water before immersion in de-aterated 1 M sodiuni chloride solution. Immediately after immersion, they were polarised to -500 mV vs SCE for 60 seconds and then the potential was increased at a positive sweep rate of 10 inV rnin' until the current reached 100PA.

At this point the sweep was reversed at the same rate until the current entered the cathodic region. Figure 11 shows the results of the pitting studies at room temperature.

Due to the high level of resistance to chloride pitting of 904L reported in the literature (J. R. Davies (Editor), ASM Specialty Handbook: Stainless Steels, ASM International Ohio (1994)), only 430, 0 Ru, 0. 1 R u, 0. 16 R u ant] 316 alloys were tested in this environment. The results are tabulated in Table 9.

V 14 chloride Tgbie 9: The results of pittint, studies in de-aerated 1 M sodium c Epi, (V vs SCE) EP (V vs SCE) 430 -0.23 0.05 -0.22 0.02 0 Ru -0.02 0.01 -0.13 0.002 0.1 Ru 0.10 0.04 -0.07 0.03 0. 16 Ru 0.11 0.001 0.02 0.02 316 -0.09 0.02 -0.07:t 0.005 Open Circuit Measurenients in 0.5M Sodilini Siilj)liate: Samples of the 0 Ru and 0. 16 Ru alloys were prepared as in previous tests. Each was immersed in a beaker containing naturally aerated 0.5M sodiuni sulphate with a reference electrode. The open circuit potential was periodically i- neasured with a digital voltrrx-.ter over eight days. The results of the open circuit potential iiiea,,.ui.ci-nciit, in naturaBy aerated 0.5M sodium sulphate solution are shown in Figure 12.

Claims (3)

1. A stainless steel alloy consisting of 10 to 20wt% chron-durn, cl 0.025 to 0.2wt% ruthenium, 0 to 2wtc/(- of one or both of nickel and molybdenum, the balance being iron.

2. A stainless steel alloy according to claim 1 consisting essentially of 10 to 20wt% chron-dum, 0.025 to 0.2wt% ruthejiiui-n, the balance being iron.

3. A stainless steel alloy accordin- to claim 1 consisting essentially of 10 1 & to 20wt% chromium, 0.025 to 0.2wtc/e- rutheniuni and 0. 1 to 2wt% of one or both of molybdenum and nickel the balance being iron.