VÝZNAM POPOUŠTĚNÍ SVAROVÝCH SPOJŮ NÍZKOLEGOVANÝCH ŽÁROPEVNÝCH OCELÍ AN IMPORTANCE OF PWHT ON LOW-ALLOYED CREEP- RESISTANT STEEL WELDS

Transcription

1 VÝZNAM POPOUŠTĚNÍ SVAROVÝCH SPOJŮ NÍZKOLEGOVANÝCH ŽÁROPEVNÝCH OCELÍ AN IMPORTANCE OF PWHT ON LOW-ALLOYED CREEP- RESISTANT STEEL WELDS Petr MOHYLA (1), Václav FOLDYNA (2) (1) CPIT, VŠB Technical University of Ostrava, Faculty of Mechanical Engineering, 17 listopadu 15, Czech Republic, (2) JINPO Plus, a.s., Ostrava, Czech Republic Abstrakt: Přednáška se zabývá vlivem popouštění svarových spojů nízkolegovaných žáropevných ocelí T23, T24 a na jejich mechanické vlastnosti. Oceli T23 a T24 byly vyvinuty pro svařování membránových stěn bez předehřevu a popouštění po svařování. Vlastní výzkum autorů článku v oblasti sekundárního vytvrzování ocelí obsahujících vanad ukazuje, že u nepopuštěných svarových spojů dochází k výraznému vytvrzování během expozice při provozní teplotě. Proces vytvrzování je doprovázen poklesem vrubové houževnatosti především v tepelně ovlivněné oblasti (TOO). Z výsledků vyplývá, že zařazení popouštění po svařování u těchto ocelí je nezbytné. Abstract: This paper deals with influence of PWHT of T23, T24 and welds on their mechanical properties. The T23 and T24 steel were designed for welding membrane water walls without preheating and without PWHT. Our investigations concerning secondary hardening of steels containing vanadium show, that non-tempered welded joints undergo a significant increase in hardness during exposure at operating temperature. This process is accompanied by decreasing of impact toughness especially in heat affected zone (HAZ). It follows that Post Weld Heat Treatment (PWHT) of T23 and T24 welds is necessary. 1. Introduction The worldwide effort in projecting of advanced coal-fired power plants leads to increase of their efficiency. The main development trend to improve the power plant efficiency is the increase of steam parameters. Construction of high efficiency power plants requires to obtain materials with improved high temperature strength, superior resistance to oxidation and resistance to high temperature corrosion. Increasing of temperature causes that the final superheaters must be made of material with superior high temperature strength. With the increase of pressure, not only the final superheater tubes, but all steel tubes including the economizer tubes and membrane water wall (MWW) tubes must provide improved high temperature strength [3]. Conventional world-wide used steel T22 (2,25Cr-1Mo) and 13CrMo4-4 don t have enough creep strength for the use as membrane water walls of ultra super critical boilers. This facts have led to development of new perspective low alloyed steels for membrane walls T23 (HCM2S) and T24 (7CrMoVTiB7-7). Both steels have excellent mechanical properties at high temperatures and improved weldability. 1

2 Development philosophy of T23 and T24 steels follows two main objectives. First, it is an increase of creep strength by optimization of the chemical composition. In comparison with conventional steel 10CrMo910 there is a small addition of V, Nb, B and especially W in T23 steel and V, Ti and B in T24 steel. Second, it is enhancement of weldability by reduction of C content below 0,10%. Membrane water walls (MWW) are very large components, so there is an effort to produce MWW without post weld heat treatment (PWHT). On larger components only local PWHT is possible, which can cause additional stresses and strains in the component. Further more PWHT is expensive. If PWHT is mandatory, also the repair of MWW will be more difficult and costly. 2. Previous studies There are some research works concerning the influence of PWHT on mechanical properties of low-alloyed creep resisting steel welds in the literature. Conclusions concerning necessity of PWHT are different. In the work [7] there were investigated thermal fatigue of new steels for ultra super critical power plants (T23). A testing water wall panel was welded without PWHT and exposed to app cycles, which comprises about 8000 hours of total heat exposure. Many cracks were found in some of tube to fin welds in HAZ after exposure. Authors conclude that the crack formation in the tube to weld HAZ region could not be clearly identified and they belive that it can be due to welding procedure or due to gab between the tube and the fin [7]. Thermal fatigue can certainly cause creation of cracks, but the basic precondition is to have a material, which is not brittle. An embrittlement of welded joints of T23 or T24 steel is related to PWHT. The mechanism, which can cause embrittlement is described in this article. In some research works [7, 8] authors conclude, that PWHT of T23/T24 welds is not necessary, nevertheless some recent works yield quite different conclusions. In the work [9] there was investigated creep behavior, hydrogen resistance and temper embrittlement behaviour of T23 and T24 steel welds. Within the experimental programme a wide variety of PWHT have been performed on many welded joints. Author concludes, that proper toughness values of weld metal can not be achieved when a PWHT is omitted. P23 and T24 materials can not be used in as welded condition and repair without PWHT is not a realistic option [9]. Author asserts, that for obtaining a sufficient toughness is necessary a hardness criterion of max. 300 HV. This can be achieved only with PWHT. The experimental part of the work [10] was focused on welding of piping made from steel P23 with several regimes of PWHT. Impact toughness and hardness testing of HAZ was performed. On the basis of achieved results authors recommend to apply PWHT in the range of C with holding time at least 60 minutes [10]. 3. Hardening Mechanism of Modern Low-alloy Steels The principle behind increased creep resistance of steels alloyed with vanadium, or titanium and niobium is dispersion of MX fine particles which have a significantly increased dimensional stability during extensive heat exposure in comparison to chromium carbide. Ultimately this means a great increase in creep resistance, represented by values for creep strength over 10 5 or hours. On the other hand the major effect of dispersion of MX particles is a decrease of plastic properties in steel in relation to so-called secondary hardening. 2

3 Even after typical heat treatment of the steel, which comprises normalisation and subsequent tempering at C, the structure of these steels is not absolutely in equilibrium. During subsequent extensive exposure at operating temperature, which is lower than the tempering temperature, partial precipitation of MX particles occurs as a result of solid solution saturation. This process is most significant around the weld joints, where due to the welding process the degree of dissolution of dispersed particles varies. Subsequently, the tempering temperature may not be maintained correctly, which causes imperfect precipitation of MX particles in weld metal and in the heat affected zone (HAZ). The result is an unbalanced microstructure of the weld which leads to the secondary hardening during subsequent exposure at operating temperatures. From this point of view thee preterition of PWHT is absolutely false. 4. Experimental work 4.1. Weld joints of CrMoV Steel Grade In the former Czechoslovakia, in the second half of the 20th century, engineers developed and introduced for practical use a 0,5%Cr-0,5%Mo-0,3%V type steel alloyed with vanadium (grade ), which achieved significantly higher creep strength values than 10CrMo9-10 grade steel. The disadvantage of steel is a high sensitivity of its mechanical properties to the welding process and parameters of heat treatment, including tempering. These technological obstacles were most likely the main reason why steel, even through its excellent creep-resistant properties, never became globally widespread. During the first part of an experimental programme testing weld joints were prepared on tubes ( steel) using welding process 111 and electrode EB 321 with matching chemical composition to parent material, see tab. 3. Three tempering temperatures were tested - 650, 680 and 715 C. These samples were then subject to long term heat exposure without stress at a temperature of 450 C. The affect of the temperature on weld metal hardness is illustrated in Fig. 1. Material tempered at 650 C shows, during exposure, a hardness from 60 to 70 HV10 higher than weld metal tempered at 715 C, which means an increase in hardness by 25% average. After reducing the tempering temperature from 715 C to 680 C, weld metal hardness increased by an average 15% during subsequent thermal exposure. Significant differences in weld metal hardness of weld joints subject to various tempering can be seen in the initial condition, i.e. prior to extensive thermal exposure. Very similar curve profiles were recorded for the normalization zone and overheated zone of the HAZ, all results are published in paper [1]. KCV impact toughness values were determined for selected samples. A curve made from these values has its minimum in the maximum hardness area and in the area of hardness decrease it has an increasing trend. Experimental results show that the impact toughness of weld joints is significantly dependent on the tempering temperature. The greatest difference was seen when the tempering temperature was dropped from 715 C to 680 C. In the initial condition the toughness of weld metal tempered at 715 C is 40% higher during sub-creep exposure and this difference increases locally to as much as 100%, see Fig. 2. 3

4 Fig. 1: Comparison of hardness profiles in the weld metal in dependence on weld joint tempering temperature, operating temperature 450 C Fig. 2: Comparison of weld metal (EB 321) impact toughness profiles in dependence on weld joint tempering temperature, operating temperature 450 C 4

5 Measurements of hardness and impact toughness show the presence of secondary hardening in weld joints of steel operated at higher temperatures. These results were supported by a microstructral analysis of dispersed particles using an transmission electron microscope (TEM). Image analysis was used to determine parameters of the dispersion phase of MX particles in investigated samples. Three samples, marked 1.0, 1.1 and 1.20, were selected, see figure 3. Sample 1.0 represents the initial condition, sample 1.1 represents maximum hardening after exposure, and sample 1.20 represents a decrease in hardness. Real and extrapolated duration of TEM samples ageing shows table 1. Calculated results for the equivalent particle diameter d ekv, number of particles per unit area n s, number of particles per unit volume n v, and mean inter-particle spacing λ are presented in Table 2. Fig. 3: Samples for TEM analysis of dispersed particles Table 1: Real and extrapolated duration of TEM samples ageing specimen No: real time and temperature of ageing h /550 C h h/550 C h extrapolated time of ageing for temperature 450 C The mean inter-particle spacing was calculated according to Ashby [5]: 5

6 2 λ= 0,69 3 ( n ) 1 2 v d ekv d ekv where: d ekv. equivalent particle diameter [nm] λ mean inter-particle spacing [nm] n v.. number of particles per unit volume (1) Sample No. Table 2: Calculated results for d ekv n s, n v, λ A X [nm 2 ] d ekv [nm] n s [m -2 ] n v [m -3 ] λ [nm] (Ashby) ,24 13,65 2, , , ,01 12,10 7, , , ,10 16,91 2, , ,46 Electron microstructural analysis together with image analysis confirmed that during extensive thermal exposure in the sub-creep zone, changes in dispersion phase of the welds of grade steel occur. To be specific, this is a process of supplementary precipitation and process of MX particles growth. The maximum hardness during thermal exposure corresponds to the area of supplementary precipitation of MX particles (sample 1.1). This is indicated by the highest count of particles per unit volume, smallest particle size and smallest mean inter-particle spacing. On the other side, hardening curves with a steep drop in hardness indicate an area of growth of secondary phase particles (sample 1.20). Proof of this is the largest mean particle area, marked drop in number of particles per unit volume and almost a double increase in the mean inter-particle spacing with respect to the supplementary precipitation area. The significant influence of the dispersion phase on mechanical properties of grade steel is confirmed by the fact that dislocation density was practically constant during extensive thermal exposure [2] Weld Joints of T23 and T24 Grade Steels As a result of the proposed alloying with carbide-forming elements and a reduced carbon content, T23 and T24 grade steels have excellent mechanical properties at elevated temperatures and are distinguished also by improved weldability. The essence of their high creep resistance is, just as for steel, very stable dispersed particles MX. (In the case of low-alloy steels these are vanadium, titanium or niobium carbides or carbo-nitrides.) For T23 and T24 grade steels we can expect similar secondary hardening of weld joints during tempering or extensive thermal exposure as for steel. Mean chemical composition of T23, T24 and grade steels is presented in Table 3. 6

7 Table 3: Chemical composition of 15128, T23, T24 steels and consumables Steel C Mn Si Cr Mo V W Nb Ti N B ,13 0,60 0,3 0,5 0,5 0, T23 0,08 0,40 Max. 0,50 2,25 0,15 0,25 1,6 0,05 - Max. 0,03 Max. 0,0060 T24 0,09 0,50 0,30 2,50 1,00 0, ,05 0,10 Max. 0,012 Max. 0,0070 E-B 321 (15 128) 0,08 0,70 0,30 0,60 0,50 0, Thyssen Cr2WV 0,50 0,20 2,35 0,06 0,25 1,70 0, (T23) 0,05 Thyssen Cromo3V (T24) 0,09 0,60 0,20 3,00 1,0 0,25-0, The experimental programme included fabrication of test weld joints on plates from T23 and T24 grade steels using welding process 111, electrode Thyssen Cr2VW, or Thyssen Chromo3V. A selection of the weld joints were tempered at 760, and 750 C respectively, and the remainder was left non-tempered (i.e. as welded condition). Samples were subject to simulated operation (heat exposure at 500 C without stress) and subsequently hardness and impact toughness was measured in individual zones of the weld joint. Fig. 5 shows hardness curves for the overheating zone of HAZ for T23 steel during simulated operation at 500 C. Fig. 6 shows the same for T24 steel. In the next stage selected samples were subject to measurement of impact toughness. The toughness profile has its minimum in the maximum hardness area and a rising trend in the area of decreasing hardness, just as in the case of grade steel. The affect of tempering after welding on impact toughness of the HAZ in T24 grade steel is depicted in Fig. 7. Figure 8 shows a comparison of hardness and notch toughness profiles in overheated zone of steel T24 at operating temperature. Measured values clearly show that non-tempered weld joints suffer a significant loss of plastic properties. The achieved results are in line with results measured in weld joints of grade steel [2]. Hardness profile curves show that welds which are not subjected to tempering undergo significant hardening in a relatively short time of 100, 2000 hrs respectively. Over longer times (over hrs) hardness decreases. Maximum of hardness values are 410 respectively 430 HV10, which are unacceptable values with respect to standard ČSN EN ISO [6], where allowed maximum is 350 HV10. Measured values show a presence of secondary hardening of weld joints of T23 and T24 steels. The hardening mechanism is, as in the case of grade steel, given by supplementary precipitation of dispersed particles MX. Fig. 8 shows a significant difference between the impact toughness of tempered and non-tempered weld joints of T24 steel. While the toughness of the overheated zone of HAZ in a tempered joint is about 180 J/cm 2 under high-temperature exposure (testing temperature of 20 C), in non-tempered condition KCV values are very low at around 20 J/cm 2. 7

8 Fig. 5: Hardness profiles in the overheated zone of HAZ for T23 grade steel, operating temperature 500 C Fig. 6: Hardness profiles in the overheated zone of HAZ for T24 grade steel, operating temperature 500 C 8

9 Fig.7: Impact toughness profiles in the overheated zone of HAZ for T24 grade steel, operating temperature 500 C Fig.8: Comparison of hardness and notch toughness profiles, overheated zone of steel T24, operating temperature 500 C, state as welded 9

10 5. Conclusion Results presented in this article demonstrate that weld joints of low-alloy creepresistant steels hardened by dispersed MX nano-particles are subject to a process of secondary hardening during extensive thermal exposure at elevated temperatures. It was clearly proved by microstructural analysis of MX particles of steel. This secondary hardening is common for all steels containing vanadium including T23 and T24 steel. Results of harness and impact toughness during high temperature exposure of T23 and T24 show, that secondary hardening is present and preterition of PWHT is non realistic, which is corresponding with [9]. The extent of this hardening depends on the temperature and duration of tempering after welding. Weld joint brittleness also depends on the operating temperature of the welded unit. In grade steel brittleness is seen at temperatures under 500 C, for T23 and T24 grade steels we can expect brittleness at even lower temperatures. From the achieved results we can conclude that PWHT of weld joints from , T23 and T24 grade steels is necessary especially in terms of obtaining sufficient plastic properties. For welds in grade steel the optimal tempering temperature is 715 to 730 C. While for T23, T24 and T25 grade steels can be recommended tempering temperatures in the range of 750 to 760 C in order to achieve permissible hardness and sufficient durability of weld joints. References [1] MOHYLA, P. Změny mechanických vastností svarových spojů oceli při dlouhodobé teplotní expozici v podcreepové oblasti, Sborník vědeckých prací VŠB TU Ostrava, 2001, část 2, str. 31 [2] MOHYLA, P. Změny mechanických vastností CrMoV svarových spojů při dlouhodobé teplotní expozici v podcreepové oblasti, doktorská disertační práce, VŠB TU Ostrava, 2001 [3] TOHYAMA, A., MINAMI, Y. Development of 2Cr-Mo-W-Ti-V-B Ferritic Steel for Ultra Duper Critical Boilers, In: Proceedings of COST Programe part I: Materials for Advanced Power Engineering 1998, Vol. 5, ISBN: [4] MOHYLA, P., KOUKAL, J. Contribution to Research of Weldability of Modern Low-Alloy Creep Resistant Steels. In: Acta Metallurgica Slovaca, 3/2003, ISSN: [5] ASHBY, M.F. Physics of Strength and Plasticity, A.S. Argon, MIT Press, 1969, str. 113 [6] ČSN EN ISO Stanovení a kvalifikace postupů svařování kovových materiálů. Praha: Český normalizační institut, květen 2005 [7] KARLSSON, A., RASMUSSEN, F., MONTGOMERY, M. On-Plant Test of TÜV HCM12 and ASME T23 Alloys for Use as Water Wall Materials.Key Engineering Materials Vols (2000), 2000 Trans Tech Publications, Switzerland, p [8] Vallourec&Mannesmann tubes. The T23/T24 Book [9] WORTEL, H. P23 and P24 Power Generation and Hydrogen Service Challenges and Limitations. In. 27. Vortragsveranstaltung, , Düsseldorf, Stahlinstitut VDEh, 2004, p [10] PECHA, J. MEŠKO, J. Weldability of New Creep Resistant Cr-Mo Steel with Tungsten Content. Materials Engineering, Vol. 13, 2006, No. 2, p