Q-P PROCESSING OF HIGH-STRENGTH LOW-ALLOYED STEEL SHEETS

Transcription

1 Q-P PROCESSING OF HIGH-STRENGTH LOW-ALLOYED STEEL SHEETS Daniela HAUSEROVÁ a, Zbyšek NOVÝ b, Jaromír DLOUHÝ c, Petr MOTYČKA d a,b,c,d COMTES FHT a.s., Průmyslová 995, Dobřany, Czech Republic, comtesfht@comtesfht.cz Abstract Availability of high-strength sheets with good ductility at reasonable cost is not sufficient on the European market. In the present work, the QP effect (Quenching and Partitioning) was used to obtain thin sheets with high strength and good ductility. The experiment was performed on low-alloyed steel containing 0.2% carbon and a higher amount of silicon of about 1.5%. The material was rolled into 1 mm thick sheet. This grade of steel is a cost-effective material thanks to its low amount of alloying elements. This group of low-alloyed steels, if heat treated or thermomechanically treated in a suitable manner, offers a favourable combination of strength, elongation and toughness. The QP process consists in rapid quenching of the material between M s and M f temperatures to prevent the martensitic transformation from propagating through the entire volume of the workpiece. Subsequent heating causes diffusion of excess carbon from martensite to retained austenite, thereby increasing the stability of the austenite. Tempering of martensite can also occur. The aim of the QP process is to produce very fine martensite microstructure with retained austenite between martensite plates. Austenite becomes stabilised through higher content of silicon which suppresses carbide formation and keeps higher fraction of carbon in solid solution. Sheets from high-strength and ductile materials are expected to be used primarily in transport engineering and production of sports equipment. Keywords: QP (Quenching and Partitioning) process, high-strength sheet 1. INTRODUCTION The requirements on mechanical properties of steels are constantly increasing and new heat and thermomechanical treatment processes for such steels are being developed. High-strength low-alloyed steels, as a group of steels, offer favourable proportion of strength, elongation and toughness. The level of ultimate tensile strength required in these materials, upon suitable heat treatment, reaches 1,500 MPa. Their elongation should be around 15%, while the maximum content of alloying and residual elements should not exceed five weight per-cent. Another important requirement consists in good weldability. One of modern heat treatment techniques capable of meeting these requirements is the Q-P process which consists in rapid quenching of a material between the M s and M f temperatures in order to prevent full martensitic transformation. Subsequent heating to a temperature below M s initiates a diffusion flow of excess carbon from martensite to retained austenite. Cooling down to room temperature stabilizes the retained austenite thanks to prior diffusion of carbon from supersaturated martensite to the still untransformed austenite. The purpose of the Q-P process is to produce very fine martensite with retained austenite between martensite plates. (Fig. 1) [1]. After austenitizing, the steel should be quenched (Q-P) to a specific temperature calculated in such a way as to produce a pre-defined ratio of martensite and non-transformed austenite. Subsequently, the temperature of the material should be raised to the partitioning level (P. The carbon will diffuse to the existing austenite and increase its stability to the level where it does not transform upon cooling to Fig. 1 Schematic Q-P heat treatment [1] ambient temperature. As the austenite becomes

2 enriched in carbon during the partitioning stage, its actual M s -M f temperatures decrease. Full stabilisation requires that the M s temperature is depressed to or below room temperature to prevent martensitic or bainitic transformation of insufficiently stabilised austenite during final cooling [2-3]. Proposals for new chemical compositions of high-strength steels with the QP effect take into consideration, among other aspects, the requirement for austenite stabilization and cost of alloying additions. Stabilization of retained austenite in steels is achieved by adding manganese and nickel. Due to lower cost, alloying with manganese is preferred, normally in the amount of 1.5 to 2.5 wt. %. The key to stabilization of retained austenite is also its saturation with carbon during partitioning (or during the hold at the quenching temperature). The extent of carbon diffusion from martensite to austenite will only be sufficient if cementite is prevented from forming in martensite and at the martensite-austenite boundary. Cementite particles absorb carbon from their surroundings, forming a region of lower concentration, which in turn attracts carbon to diffuse from the surrounding material. This disrupts the desired diffusional flow of carbon from martensite to austenite. Elements which are reported to prevent cementite formation include silicon, aluminium and phosphorus. These elements do not dissolve in cementite and must escape by diffusion from the nucleus if a new cementite particle is to form. This makes cementite formation less favourable in energy terms. The impact of Si, Al and P on formation of transition carbides, such as Fe 2,4 C, is less obvious. Silicon content normally ranges from 0.3 to 2 wt. %. The amount of aluminium, if used, is no higher than 1.5 wt. %. [4]. Addition of carbide formers, such as Mo, Nb or V, in the cumulative amount of up to 0.3% leads to grain refinement in steel but is very likely to cause precipitation of these carbides in martensite during partitioning [5]. Alloying with these elements is therefore suitable for steels with carbon content of at least 0.2 wt. %. This level should prevent carbon from being used up in forming carbides of these elements during partitioning. The precipitated carbides are very fine and are densely and uniformly distributed within martensite laths. This is why they do not inhibit saturation of austenite with carbon as much as Fe 3 C particles do, where the latter are an order of magnitude coarser. Fine carbides do not absorb carbon from as large an area. They maintain, more or less, uniform carbon concentration along the entire martensite lath, though this is lower than in the state without such fine precipitates. The direction of diffusion of carbon from martensite to austenite remains. Fine precipitates also strengthen the martensite matrix of the Q-P steel. 2. EXPERIMENTAL Chemical composition of the experimental steel used in this study was proposed to stabilize austenite and suppress cementite formation and in order to support the Q-P process. However, its alloying included molybdenum for fine-grained microstructure and additional solid solution and precipitation strengthening. Only three additions were proposed: silicon and manganese with 1.5 wt. % levels and molybdenum in the amount of 0.25 wt. %. The carbon content was 0.2 wt. % (Table 1). As this is the first experimental melting, it appeared useful to keep the chemical composition simple without additional elements, such as aluminium, and thus obtain a reference material for comparing with optimized QP steels with more complex alloying [6]. Table 1 Chemical composition of the experimental steel [weight%]. C Si Mn P S Cr Mo Ni Al V Nb Ti N B <0.007 <0.001 <0.002 < The input stock for the heat treatment procedure was 0.9 mm thick rolled sheet. Initial bainite-martensite microstructure had a hardness of 428 HV.

3 The heat treatment procedure was carried out in the thermomechanical simulator MTS 810 with additional resistance heating capability (Fig. 2). It allowed the specimen temperature to be changed rapidly in a controlled manner. Specimens were parts of sheet metal with 0.9 mm thickness, 90 mm length and 20 mm width. They were held on both sides between symmetrical flat grips which supplied electrical current for resistance heating. Temperature was measured by K-type regulation thermocouple welded onto the specimen surface at its mid-length. The length of the specimen central section between grips was 60 mm. While heated and held at austenitizing temperature, the specimen cooled primarily through the transfer of heat to environment by radiation and also by heat transfer to the watercooled grips. Rapid quenching from austenitizing temperature was achieved by cooling the surface opposite of the attached regulation thermocouple with compressed air at the pressure of 0.7 MPa. The distance between grips was automatically and continuously adjusted throughout the experiment to avoid any external loads on the specimen. XRD phase analysis was performed in the automatic powder diffractometer AXS Bruker D8 Discover with a position-sensitive area HI-STAR detector and a cobalt X-ray source ( K = nm). The instrument was equipped with a polycapillary lens focusing the primary X-ray beam into a circular spot with a diameter of 0.5 mm Heat Treatment As mentioned above, the core of the Q-P process is partial quenching to a temperature between the martensite start and martensite finish temperatures and subsequent tempering close to the martensite start temperature. For this purpose, the temperature limits of martensitic transformation and the resulting volume fractions of martensite and austenite in the microstructure were determined using the dilatometer. The cooling rates used were 100, 150 and 200 C/s. The martensite start and martensite finish temperatures were found to be 370 C and 220 C, respectively. The volume fractions of martensite and austenite formed in the microstructure are shown in fig. Regression lines for austenite (pink) and martensite (blue) dilatation were constructed on the dilatometric curve for cooling at 150 C/s. They were used for estimating the dependency of the volume fraction of martensite on temperature during quenching (green curve) (Fig. 3): l( M( l ( [1 M( ] l (, or M l( la( M ( lm ( la( where l( is the experimentally measured total dilatation, M( is the volume fraction of martensite (0-1), l M ( is the martensite dilatation, l A ( is the austenite dilatation. A Fig. 2 Thermomechanical simulator MTS 810. Fig. 3 Dilatometric curve for evaluation of M s and M f and the dependence of the volume fraction of martensite on temperature. Cooling rate 150 C/s. The heat treatment schedules were proposed for Fig. 4 Q-P schedule /20 sec.

4 exploring the impact of the quenching temperature and the partitioning temperature on the microstructure, hardness and percentage of retained austenite in the microstructure. In all schedules, the austenitizing temperature was 900 C held for 60 seconds. The quenching temperatures of 250, 280, 310 and 330 C were used. The cooling rate was approximatly 130 C/s. Partitioning temperatures were chosen to be close to the martensite start temperature: 350 and 310 C. The schedules are detailed in Table 2 and an example of such schedule is shown in Fig. 4. Table 2 Heat treatment schedules Schedule Hardness HV5 Amount of retained austenite [%] Initial state Q-P /20 sec Q-P /20 sec Q-P /20 sec Q-P /20 sec Q-P /20 sec Q-P /20 sec Q-P /20 sec RESULTS AND DISCUSSION Fig. 5 Micrographs of the specimen Q-P /20 sec. The Q-P treatment was used for preparing very fine martensite-bainite microstructure with retained austenite (Fig. 5) In the course of the Q-P process, retained austenite became stabilized by carbon and manganese diffusing from martensite to retained austenite. The amount of retained austenite was measured by XRD analysis, yielding values between 6 and 9 % (Table 2). No retained austenite was detected in the initial microstructure. The amounts of retained austenite in the microstructure upon different schedules do not exhibit any considerable dependency on the quenching and partitioning temperatures. Retained austenite is

5 probably present in the form of laths between martensite laths. In some specimens, it also formed islands with the diameter of approximately 1 µm. Morphology and distribution of retained austenite will have to be mapped by means of a transmission electron microscope and an EBSD analyzer. Coarse needles or islands of lower bainite with plate-like fine carbides within bainite needles were embedded in fine martensite matrix. No texture was detected in the microstructure by metallographic or XRD analyses. Prior austenite boundaries are visible in the material. Hardness of individual specimens varies in dependence on selected quenching and partitioning temperatures. Hardness increases with decreasing quenching temperature, whereas it declines with increasing partitioning temperature. The specimen treated according to the /20 sec. schedule deviated from this trend. The highest hardness values were achieved by applying the schedule denoted as /20 sec.: 457 HV5. The amount of retained austenite in this specimen was 8 % (Table 2). At subsequent stages of the experiment, mechanical testing will be performed, which is expected to detect the impact of processing temperatures more effectively. 4. CONCLUSION The present experiment aimed at low-alloyed CMnSiMo steel with the carbon level of 0.21% involved the Q- P process and led to formation of martensite-bainite microstructure with retained austenite. Coarse needles or islands of lower bainite with plate-like fine carbides, which precipitated within bainite needles, were embedded in fine martensite matrix. X-ray diffraction revealed 9% of retained austenite in the microstructure of heat treated specimens. No austenite was detected in the initial microstructure. Hardness of individual specimens varied in dependence on selected quenching and partitioning temperatures. Hardness increased with decreasing quenching temperature, but declined with increasing partitioning temperature. The highest hardness value achieved was 457 HV5 in a sample with 8 % retained austenite which was treated according to a schedule with quenching temperature of 250 C and tempering temperature of 310 C with 20-second hold. Q-P microstructure should only contain fine martensite with retained austenite and no carbides. This is why this process will be further optimized to provide Q-P microstructure without carbides. The goal will be improvement in mechanical properties, i.e. achieving the highest possible strength of material combined with sufficient ductility. Acknowledgement This paper includes results achieved within the project GACR 106/09/1968: Development of New Grades of High-Strength Low-Alloyed Steels with Improved Elongation Values. References [1] Edmonds, D.V.,etc. Quenching and partitioning martensite A novel steel heat treatment. Materials Science and Engineering A (2006), [2] Gerdemann, F.L.H. Microstructure and hardness of 9260 steel heat-treated by the quenching and partitioning process, Aachen University of Technology, Germany, [3] Bhadeshia, H.K.D.H. High performance bainitic steels, Materials Science Forum A (2005) [4] D. H. Kim, J. G. Speer, H. S. Kim, B. C. de Cooman Observation of an Isothermal Transformation during Quenching and Partitioning Processing, Metallurgical and Materials Transactions A, 2009, vol. 40A, pp [5] N. Zhong, X.D.Wang, L. Wang, Y.H. Rong Enhancement of the mechanical properties of a Nbmicroalloyed advanced high-strength steel treated by quenching partitioning tempering process Materials Science and Engineering A 506, 2008, pp [6] De Moor, E.,etc. Quench and partitioning response of a Mo-alloyed CMnSi steel, Proceedings of New developments on metallurgy and applications of high strength steels, Vol. 1 and 2, 2008, pp