Abstract
The focus towards the design of light environment friendly fuel efficient vehicles has led to an increase in the demand for high strength steels. As it is true for any developing country, the extra challenge for Indian steel industries is to produce such high strength steels at minimum cost. The aim of the present work was to develop two high strength steel grades with minimum tensile strengths of 540 and 590 MPa, in combination with good ductility and stretch flangeability, through the thin slab casting and rolling route. The present paper describes the alloy and process design methodology used for achieving the target properties. Several mill trials were taken to test the proposed design, wherein the target microstructure and mechanical properties were successfully achieved.
Introduction
During the past few decades, the automobile industry has shifted its focus towards the design of light weight environment friendly, fuel efficient vehicles. This has necessitated the replacement of conventional low strength steels with higher strength steels so that thinner gauges can be used and reduction in the weight of the auto components is possible without compromising on safety and functional requirements. Among the various autocomponents, the weight reduction achievable by substituting conventional mild steel with high strength steels is found to be the largest for autowheels 1 wherein the energy savings is estimated to be 1.2–1.3 times the amount possible for non-rotating parts. 2 Apart from this, there is also an increasing demand for high strength steels for automotive structural parts like suspension and chassis components. The typical tensile strength requirement for such applications is 540–590 MPa. Moreover, considering that these parts are quite complicated in shape and are manufactured mainly by press forming, apart from the high strength requirements, it is also essential that such steels have good formability and especially stretch flangeability [which is measured as the hole expansion ratio (HER)].
In general, high strength formable grades, such as those described above, are produced through the conventional continuous casting hot strip rolling route. These alloys rely on addition of very high amounts of Mn (1.7–2.0 wt-) and costly alloying elements like Cr/Ni/Mo/B and/or high amounts of microalloying elements like Nb/Ti3,4 to achieve the target properties. In order to minimise the cost, not only should the alloy addition be optimised, but the processing cost also needs to be reduced. Of late, thin slab casting and rolling (TSCR) technology has become a major process for hot strip production in the world due to its minimal energy consumption, higher productivity and reduced cost of production.
5
However, there are certain unique features of this production route due to which the production of high strength grades of consistent quality through this route is not very straightforward. Some of these features are listed below:
austenitisation takes place through the δ → γ polymorphic transformation; hence, the austenitisation temperature is quite high unlike in the conventional hot strip rolling process wherein austenite forms at the reheating temperature through phase transformation; as a result, the primary austenite grains in this route are quite coarse (500–2000 μm)6,7 cast slabs are soaked in a tunnel furnace at 1100–1150°C for around 20–30 min primarily for temperature equalisation,
8
unlike in conventional mills where the slabs are reheated to temperatures >1200°C for 2–3 h; therefore, the TSCR process requires judicious selection of microalloying elements to ensure their complete dissolution during temperature equalisation, so that they can be effectively utilised during the subsequent processing steps there is no rough rolling stage as in the conventional mill, and hence, the as cast dendritic microstructure has to be rolled and transformed into a homogeneous microstructure in the finishing train.
8
Considering the factors mentioned above, it is necessary that chemistry and processing parameters specifically suited to TSCR conditions be developed, so that properties similar to those obtained by traditional routes can be achieved.
The aim of the present work was therefore to design the chemistry and processing parameters required for producing two high strength steel grades, one with a minimum tensile strength of 540 MPa for wheel rim applications (steel 1) and another with a minimum tensile strength of 590 MPa for wheel rim as well as structural applications (steel 2), commercially through the TSCR mill at Tata Steel, Jamshedpur works. Additionally for both steels, a minimum HER of 70 was aimed for.
Alloy design: Metallurgical aspects
At the outset, the various possible strengthening mechanisms were critically examined so that they could be effectively utilised to identify the appropriate chemistry and processing parameters required for achieving the target properties. The same are discussed below.
Solid solution strengthening
The most effective solid solution strengthening elements are Mn and Si. However, Mn levels beyond 1.5 enhance centreline segregation during continuous casting and also cause microstructural heterogeneities. 9 On the other hand, Si levels beyond 0.5 wt- lead to formation of surface scales due to oxides of silicon. 10 Hence, the Mn level of the steel was fixed at 1.0–1.5 wt-, whereas the Si level was fixed at 0.1–0.35 wt for the current development. The carbon level was restricted to < 0.07 wt-, in order to ensure that the steel composition was outside the peritectic range so as to guarantee adequate castability of the steel in the thin slab caster. 11 Lower carbon content also results in good weldability, which is necessary for automotive grades. 11
Precipitation strengthening
Precipitation strengthening is brought about by the formation of fine precipitates of microalloyed carbides, nitrides or carbonitrides. In order to form such fine precipitates, it is necessary that the microalloying elements remain in solid solution till the time rolling commences. However, since the TSCR process involves very short duration soaking at relatively low temperatures before rolling, the choice of the microalloying elements must be judiciously made so as to ensure complete dissolution during temperature equalisation. The most commonly used microalloying elements are Ti, Nb and V. Among these, Ti is generally avoided in TSCR grades because the dissolution temperature of TiN precipitates (if formed during the prior processing steps) is very high, and hence, these precipitates are likely to remain undissolved even after temperature equalisation. In fact, the tunnel furnace conditions may only lead to coarsening of the TiN precipitates, rendering them ineffective for precipitation strengthening. 8 On the other hand, the dissolution temperature of niobium is much lower than titanium. Moreover, apart from precipitation hardening, Nb retards recrystallisation of austenite and lowers the austenite–ferrite transformation temperature, both factors leading to enhanced grain refinement of ferrite in Nb microalloyed steels.8,12 The dissolution temperature of vanadium is even lower than both Nb and Ti. Further, in the presence of adequate amount of nitrogen (>0.008 wt-), enhanced precipitation of V(CN) takes place, which not only contributes to precipitation hardening but also retards recovery and softening of the lower transformation products like bainite during prolonged holding at relatively higher temperatures after coiling. 11 In view of the above, two microalloying strategies were adopted, one with only Nb for steel 1, and the other with Nb+V for steel 2 (to account for its higher strength requirement). In order to determine the maximum permissible amount of Nb which would completely dissolve in austenite during the temperature equalisation process, the dissolution temperatures of carbonitrides of Nb and Nb+V for various Nb and N contents were calculated using Thermo-Calc software using the TCFE7 database. 13 The base composition of the steels used for the calculations was 0.06 wt-C, 1.3 wt-Mn, 0.25 wt-Si, 0.005 wt-S, 0.012 wt-P, 0.03 and 0.05 wt-V (for the Nb+V steels). Considering that AlN has a close packed hexagonal structure and its nucleation kinetics in austenite is very slow,14,15 it was not included in the calculations. Figure 1 shows the dissolution temperatures thus calculated, with the average tunnel furnace temperature of 1125°C indicated as a solid line. From the figure, it is evident that when N content is 0.0085 wt- or more, the Nb content has to be restricted to 0.025 wt- to ensure complete dissolution under tunnel furnace conditions. Even when the N content is as low as 0.005 wt-, Nb content beyond 0.03 wt- cannot be used. Since for Nb+V steels, N has to be >0.008 wt- to ensure enhanced precipitation of V(CN), the Nb content of both steels was restricted to < 0.025 wt-.
Dissolution temperature of Nb(CN) or (Nb,V)(CN) in austenite for various Nb and N contents calculated using Thermocalc; 11 solid line represents average tunnel furnace temperature of 1125°C
Grain refinement
One of the most effective strengthening mechanisms is grain refinement of ferrite. 16 Grain refinement is generally brought about using the controlled rolling technology, which is possible with proper use of microalloying elements. The minimum ferrite grain size achievable by this method is 3–5 μm. 17
The contribution of each of the mechanisms mentioned above, to the yield strength of ferrite, was calculated using standard empirical equations applicable to microalloyed steels, 18 and the same is illustrated in Fig. 2. The figure also shows the maximum attainable yield strength of ferrite calculated as the sum of the contributions from all the strengthening mechanisms. Considering that the target yield strength for steel 1 was ∼500 MPa and that for steel 2 was ∼550 MPa (as indicated by dotted lines in the figure), it is evident that the above mentioned strengthening mechanisms are not sufficient to achieve the targeted yield strength, hence indicating that a single phase precipitation strengthened ferritic microstructure would not be adequate. Therefore, the only way forward was to tailor the microstructure so as to include lower temperature transformation products like bainite (or acicular ferrite) and/or martensite to get the desired strength.
Strengthening contribution of various mechanisms towards yield strength (YS) of ferrite in steels 1 and 2; dotted lines in figure represent target yield strengths of steels, i.e. 500 MPa for steel 1 and 550 MPa for steel 2
Dual phase steels with ferrite–martensite microstructure have found wide applications in the automotive sector due to their superior strength formability combination.19,20 However, these steels have relatively low yield strength and low stretch flangeability. 21 Additionally, these steels soften in the heat affected zone. 21 On the other hand, steels with ferrite+bainite microstructure have much better strength–stretch flangeability balance and fracture toughness. 22 Additionally, problems associated with welding are not encountered in these steels. 21 Hence, bainite was chosen as the second phase microstructure for both steels.
Taking into consideration all the factors mentioned above, a microstructure consisting of precipitation strengthened ferrite as matrix and bainite as the second phase was targeted for both grades.
Process design
After hot rolling in a finishing mill, the steel undergoes two stages of cooling; the first is laminar cooling on the runout table (RoT), and the second is slow cooling during coiling of the strip. In general, the total time spent in the RoT is around 10–12 s. Hence, the challenge of producing a ferrite+bainite microstructure in the RoT is that, in this short time, the austenite has to transform to ferrite and bainite without any concomitant pearlite formation. At the same time, martensite formation during cooling or coiling also has to be prevented.
In order to establish the appropriate RoT cooling strategy, the continuous cooling transformation (CCT) diagrams for the selected alloy chemistries were generated using a neural network based CCT prediction model. 23 For these calculations, only the solid solution strengthening elements, namely, C, Mn and Si, were considered. The effect of niobium was not considered because it influences the CCT diagram in contrary ways depending on its state of precipitation. When in solid solution, niobium retards the austenite–ferrite transformation kinetics, whereas when NbC exists as precipitates, they act as nucleation sites for ferrite resulting in an enhanced transformation kinetics. 8 Vanadium, on the other hand, does not significantly affect the ferrite transformation kinetics, but it retards pearlite formation to some extent. 23 However, vanadium was also not included in the calculation, as it was not added for steel 1. The CCT diagram shown in Fig. 3 was constructed using steel 1 chemistry, but since the compositions of both the steels with respect to the solid solution strengthening elements were nearly equal, the CCT diagram for Steel 2 was more or less identical to the one shown in Fig. 3.
Continuous cooling transformation diagram for selected chemistries generated using neural network based CCT prediction model 21
On the basis of the generated CCT diagram a single stage RoT cooling strategy with rapid cooling (between 30 and 100°C s− 1) from finish rolling temperature to an intermediate temperature in the bainitic bay followed by air cooling to the coiling temperature was proposed so that the desired microstructure consisting of ferrite+bainite could be produced.
Mill trials
Mill trials were carried out using the proposed chemistries as given in Table 1. The finish rolling temperature TFRT was selected such that Ae3 − 50°C ≤ TFRT ≤ Ae3+50°C, where Ae3 is the temperature at which the transformation of austenite to ferrite starts at equilibrium. After finishing, the strip was fast cooled at the proposed cooling rate until an intermediate temperature TINT, given by Ae3 − 320°C ≤ TINT ≤ Ae3 − 300°C was reached. This was followed by natural cooling till the coiling temperature given by 450°C < TCT < 500°C was attained, and then the strip was coiled at TCT. Figure 4 shows a schematic diagram of the proposed cooling profile.
Chemical compositions of trial grades with Fe to balance/wt-
Schematic diagram of cooling profile
Tensile test specimens of 50 mm gage length and 12.5 mm gage width (ASTM E8) were machined parallel to the transverse direction of the rolled strips. For each steel, a minimum of three samples were prepared from the tail end of the strips. Tensile tests were carried out using an electromechanical tensile testing machine (Instron 5582) at a crosshead speed of 1.67 mm s− 1. From the stress–strain data, yield strength (0.2 offset), tensile strength and total elongation were determined.
Hole expansion tests were carried out according to ISO 16630 standards on a 60 tonne hydraulic press. For this, 100 × 100 mm square samples with a 10 mm diameter hole punched at the centre of the specimens were used. For each steel, a minimum of four samples were prepared from the tail end of the strips. The punched hole was expanded using a 50 mm diameter conical punch with a 60° apex angle. All tests were conducted in the burr up position, and no lubricant was applied to the specimen. The test was stopped as soon as a through thickness crack was detected. The HER (in ) was calculated according to equation (1), where do and df are the initial and final diameters of the punched hole, which were determined by taking the mean of three measurements made roughly at 60° to each other
Transmission electron microscopy (TEM) of carbon extraction replicas was carried out to determine the nature of the precipitates. For preparing the extraction replicas, polished samples were first etched with 2 nital. The etched samples were then coated with a carbon film following which they were immersed in concentrated HCl till the carbon film/extraction replicas started peeling off. The replicas were then floated in distilled water from where they were collected on copper grids. The replicas were observed using a JEOL JEM2200F transmission electron microscope operated at 200 kV. The composition of the precipitates was determined using energy dispersive X-ray spectrometry (EDS).
Results
Microstructure
Figures 5 and 6 show the typical optical and scanning electron micrographs obtained from the hot rolled steel samples of steels 1 and 2 respectively. In both steels, the microstructures were found to have a ferritic matrix and bainite as second phase. Le Pera 24 etching and SEM confirmed that no measurable martensite was present in the microstructures. In steel 2, slight banding was observed, which could be due to the moderately higher level of manganese used in this steel. In both steels, the average area fraction of bainite was determined to be 0.25 from standard point counting method. The average grain size of ferrite was found to vary between 3.5 and 3.7 μm in both steels.
a optical and b scanning electron micrographs of hot rolled steel 1 grade (αB: bainite; α: ferrite)
a optical and b scanning electron micrographs of hot rolled steel 2 grade (αB: bainite; α: ferrite)
Precipitation state
The precipitation state in steels 1 and 2 are shown in Figs. 7 and 8 respectively. From EDS analysis, the precipitates were found to consist of either only Nb, or Nb+Ti in steel 1, whereas in steel 2, the precipitates were found to consist of either Nb+V or Nb+V + Ti. Precipitates containing only Ti were not detected in either steel. Furthermore, in steel 2, precipitates containing either only Nb, or only V or V+Ti were not detected. The average atomic per cent of Nb in the Nb+Ti containing precipitates in steel 1 was found to be 54, indicating that they were only slightly richer in Nb. In steel 2, on the other hand, the average atomic per cent of Nb in the Ti containing precipitates was found to be 63, with the average atomic per cent of vanadium and titanium being 19 and 22 respectively. In the non-Ti precipitates, also the average atomic per cent of Nb was 61, clearly indicating that, in all precipitates of steel 2, Nb was the dominant component. In both steels, the Ti containing precipitates were found to be larger with an average size of ∼30 nm, while the size of the non-Ti precipitates was found to be ∼17 nm.
a transmission electron micrograph of carbon extraction replica showing precipitation state in steel 1, b size distribution of precipitates and EDS spectrum of precipitates marked c A and d B
a transmission electron micrograph of carbon extraction replica showing precipitation state in steel 2, b size distribution of precipitates and EDS spectrum of precipitates marked c A and d B
Mechanical properties
The mechanical properties (yield strength, ultimate tensile strength and total elongation) of the hot rolled steels are listed in Table 2. In both steels, the mechanical properties with respect to ultimate tensile strength and HER were achieved.
Mechanical properties*
*YS: yield strength; UTS: ultimate tensile strength; TEL: total elongation; HER: hole expansion ratio.
Discussion
Traditional high strength steels for wheel rim and structural applications are produced through the conventional continuous casting, i.e. hot strip mill route, which, due to their high alloy addition, are not only expensive but also very difficult to produce through the TSCR route. This work successfully demonstrates that the same grades can be produced through the TSCR route using much leaner chemistries by appropriately tailoring the microstructure to achieve the target properties. However, there are some critical factors pertaining to the design philosophy, which are described in this section.
Microstructure evolution and importance of cooling profile
Continuous cooling lowers the austenite–ferrite transformation temperature, whereas, when austenite is deformed, the transformation temperature increases. Hence, the finish rolling temperature was set at Ae3 ± 50°C to ensure that there is no rolling in the intercritical region. This was confirmed by the absence of elongated ferrite grains, typical of intercritical rolling, in the final microstructure. After finish rolling, the deformed austenite was cooled on the RoT according to the prescribed cooling strategy. On the basis of several mill trials taken for both steel grades, the cooling profile was identified to be one of the most critical parameters for achieving the desired microstructure and mechanical properties. In the first stage of cooling, austenite transforms to ferrite. The high cooling rate at this stage not only prevents any pearlite formation but also results in lowering the ferrite start temperature, which leads to refinement of the ferrite grain size. 25 However, if the cooling rate is too high, it will prevent ferrite formation also, which is not desirable.
The first stage of cooling continues till an intermediate temperature (which is below the bainite start temperature Bs) is reached. The air cooling step beyond this stage results in the transformation of the remaining austenite to bainite. The intermediate temperature is, therefore, a very crucial parameter because if the correct intermediate temperature is not attained, apart from insufficient bainite formation, undesired microstructural components like pearlite and/or degenerate pearlite may form, which not only affects the tensile properties but also lowers the HER. 26 A few examples of the above conditions are described here to highlight the significance of attaining the correct cooling profile.
Figure 9 shows the optical and scanning electron micrographs of steel 1 produced with a slower than prescribed cooling rate and a higher than prescribed coiling temperature. The mechanical properties and the ferrite grain size thus obtained are listed in Table 3. From the optical micrograph (Fig. 9a), the microstructure seems to contain only ferrite and pearlite. However, a few bainite islands could be detected at much higher magnifications using SEM. The ferrite grain size (Table 3) was also much higher than the targeted value. Failure to achieve the proper cooling rate led to the inappropriate microstructure as a result of which the targeted tensile properties could not be achieved.
a optical and scanning electron micrographs of steel 1 produced using improper cooling profile (α: ferrite; P: pearlite; αB: bainite)
Mechanical properties and grain size of steel 1 grade produced using improper cooling profile
Figure 10 shows the scanning electron micrographs of steel 2 grade produced with the correct initial cooling rate but a high intermediate temperature. The micrographs reveal that the second phase consists of a mixture of bainitic regions, pearlitic regions and what appears to be ferrite/cementite aggregates that could be bainite and/or non-lamellar pearlite or degenerate pearlite. The corresponding mechanical properties and grain size obtained are listed in Table 4. Failure to attain the correct intermediate temperature might have led to the formation of the non-bainitic components in the microstructure, which not only led to inferior tensile properties but also low HER. 26 However, due to the correct initial cooling rate, the target ferrite grain size could be achieved.
Scanning electron micrograph of steel 2 produced using high intermediate temperature (1: bainitic region; 2: pearlitic region; 3: bainite and/or non-lamellar pearlite)
Mechanical properties and grain size of steel 2 grade produced using high intermediate temperature
Precipitation behaviour
As mentioned earlier, addition of titanium is not preferred in TSCR grades because of the high dissolution temperature of TiN precipitates. Moreover, the presence of titanium also leads to premature precipitation (before rolling) of Nb and/or V leading to reduction in their overall effectiveness and hence reduction in the yield strength. 27 Since, in the present work, titanium was not intentionally added, the detection of titanium containing precipitates in both steels suggests that the scrap used to make the steel contained traces of titanium. However, the fact that no significant effect on the yield strength was recorded implies that the extent of premature precipitation was quite low.
In steel 2, all precipitates had Nb as the dominant component, which is expected as Nb has a higher dissolution temperature and hence starts precipitating earlier compared to V, and the extent of precipitation is also higher. Although the existence of precipitates was confirmed in these steels, a detailed TEM study after various processing stages is required for accurately determining the exact mechanism of precipitation.
In general, the contribution of precipitation strengthening is determined as a function of the volume fraction and radius of precipitates using the Ashby–Orowan relationsip. 16 However, for design purposes, this relationship is not very useful, as determination of volume fraction and radius of precipitates beforehand requires in-depth modelling of the precipitation process, which is even more complicated in the case of thermomechanically processed steels wherein precipitation can take place during hot rolling (in the interpass), in the austenite during cooling after finish rolling, during phase transformation to ferrite or homogeneously in ferrite during coiling. Alternatively, some amount of microalloying elements may still remain in solid solution, which also contributes to the strength. Hence, in the present work, instead of the Ashby–Orowan relationship, a more simplified approach, which takes into account the total content of the precipitating alloy, was used, 18 and the estimated strength was found to be quite accurate.
Conclusions
Two high strength hot rolled automotive steel grades with minimum tensile strengths of 540 and 590 MPa with excellent ductility and stretch flangeability have been developed through the TSCR mill at Tata Steel Jamshedpur works.
The present work primarily focused on exploring the various possible strengthening mechanisms that could be utilised effectively in order to achieve the target properties. Based on the present study, a microstructure consisting of precipitation strengthened ferrite as matrix and bainite as the second phase was targeted. Through literature survey, thermodynamic calculations, empirical strength calculations and process simulation, suitable microalloyed steel chemistries and a single stage RoT cooling strategy were proposed.
Several successful mill trials for both grades were taken wherein the target mechanical and microstructural requirements were met.
Footnotes
Acknowledgements
The project team wishes to thank the management of Tata Steel for supporting this work. We are especially grateful to various departments of Tata Steel for facilitating the trials and arranging for the samples for analysis. Very special thanks are due to Mr Yadav, Mr Sashi, Mr N. Mahto, Mr Nitish, Mr Hilal, Mr Mohit Lal and Dr Pampa Ghosh for assisting in carrying out the characterisation work.