Recrystallization of 30Cr2Ni4MoV ultra-super-critical rotor steel during hot deformation. Part I: Dynamic recrystallization

Fei Chen,Zhenshan Cui,Shijia Chen

DOI: https://doi.org/10.1016/j.msea.2011.03.008

2011-01-01

Abstract:Highlights ► The activation energy was 368.45 kJ/mol for the test steel. ► The critical stress can be expressed: σ C = 7.704 ln Z − 122.357. ► The kinetic equations have been proposed. ► The complete DRX grain size of the test steel can be described as D drx = 5.855 × 10 4 Z −0.243 . Abstract This paper presents an investigation that characterizes the evolution of the dynamically recrystallized structure of 30Cr2Ni4MoV ultra-super-critical rotor steel during hot deformation, as a starting point for studies of the static recrystallization (SRX) and the metadynamic recrystallization (MDRX) behaviors, by hot compression tests which are performed at the temperatures from 1243 K to 1543 K and strain rates from 0.001 s −1 to 0.1 s −1 on Gleeble-3500 thermo-mechanical simulator, and the corresponding flow curves are obtained. A third-order polynomial is then fitted to the work hardening region of each curve. The critical stress for initiation of dynamic recrystallization (DRX) can be calculated by setting the second derivative of the third order polynomial. By regression analysis, the activation energy in whole range of deformation temperature is determined to be Q = 368.45 kJ/mol. The complete DRX grain size ( D drx ) of the test steel is a function of Zener–Hollomon parameter ( Z ) and is independent of the true strain. The relationship of D drx and Z is found to be described in a form of power law function with an exponent of −0.24. Keywords Ultra-super-critical rotor steel Dynamic recrystallization Flow stress Hot deformation Microstructure 1 Introduction Power generation as a substantial part of the world's growth will be continually determined by coal at least in the next 20–30 years. However, there is a requirement for reduction of harmful emission from coal fired units and also an increasing demand for limiting greenhouse gas (CO 2 ) emission. The approach for meeting present requirements and demands is called “clean coal technology”. One of the most favoured solutions is the ultra-super-critical (USC) power cycle, which has demonstrated in several plants that it is a fully commercially developed process with excellent performance and availability records [1] . As a key equipment of the USC generator set, the steam turbine rotor should have good mechanical properties. Due to its good balance of strength, toughness and wear resistance, 30Cr2Ni4MoV (American grade: 3.5 Ni–Cr–Mo–V) ultra-super-critical rotor steel is widely used as the turbine rotor and disk material of the USC generation set. Low pressure (LP) of 1000 MW USC Steam Turbine rotor is usually forged through the hydraulic press. Because of many kinds of alloying elements existing in the steel and complicated microstructure evolution during the hot working, establishment of the hot working regime is difficult. In the past, many investigations have been carried out on the behavior of 30Cr2Ni4MoV rotor steel. Xuan et al. [2] investigated three different treatments, including heat treatment, deep cryogenic treatment and laser surface melting, on the corrosion resistance of 30Cr2Ni4MoV rotor steel in high temperature water. Rai et al. [3] studied the wear behavior of the Ni–Cr–Mo–V steel under varying loads and sliding velocities for its six different microstructural states. Maropoulos et al. [4] measured the fracture toughness, J Ic of the Ni–Cr–Mo–V steel using the ASTM method. Wang et al. [5] investigated stress corrosion cracking and blue brittleness of 30Cr2Ni4MoV rotor steel using high temperature aqueous solutions. Chen et al. [6,7] simulated the microstructural evolution of 30Cr2Ni4MoV rotor steel during the high-temperature austenitizing and dynamic recrystallization using cellular automaton (CA) method. Despite large amount of efforts invested into the behaviors of 30Cr2Ni4MoV rotor steel, the microstructure evolution during dynamic recrystallization (DRX) still need to be further investigated to study the workability and to optimize the hot forging processing parameters. In this article, the effects of thermomechanical parameters including strain rate and temperature on both the mechanical and microstructural aspects have been investigated. The critical stress for initiation of dynamic recrystallization (DRX) was calculated by setting the second derivative of the third order polynomial. The main focus has been on the evolution of the DRX microstructure at different strain rates and temperatures. These results in this research will be used as the basis for studies on the static recrystallization (SRX) and metadynamic recrystallization (MDRX) of 30Cr2Ni4MoV ultra-super-critical rotor steel in Part II and Part III of this work. In the following sections, experiment procedure is given in Section 2 . Some experimental results and discussion will be included in Section 3 . 2 Experimental material and procedures A commercial 30Cr2Ni4MoV rotor steel with compositions (wt.%) 0.280C–1.850Cr–3.350Ni–0.420Mo–0.089V–0.220Mn–0.0060Si–0.004P–0.002S–0.007Al–0.060Cu–0.004Sn–0.004As–(bal.)Fe was used in this investigation. Firstly, cylindrical specimens were machined with a diameter of 10 mm and a height of 15 mm. In order to minimize the frictions between the specimens and die during hot deformation, thin graphite flakes were laid between the punch head and the specimen head. The uniaxial hot compression tests were performed on a Gleeble-3500 thermo-mechanical simulator. It can be programmed to simulate both thermal and mechanical industrial process variables for a wide range of hot-deformation conditions. As shown in Fig. 1 , the specimens were heated to 1543 K at a heating rate of 10 K s −1 and held for 5 min. Then, the specimens were cooled to forming temperature at 10 K s −1 and held for 1 min to get a uniform temperature distribution. Four different forming temperatures (1243 K, 1343 K, 1443 K and 1543 K) and three different strain rates (0.001 s −1 , 0.01 s −1 and 0.1 s −1 ) were used in hot compression tests. To keep the austenite structure after DRX, the specimens were taken out quickly and quenched into cold water after each stage of compression when the true strain achieved was 0.8. Finally, the quenched specimens were sliced along the axial section. The sections were polished and etched with saturation picric and optical micrographs were recorded. 3 Results and discussion 3.1 Hot deformation flow curves A number of true stress–true strain curves obtained from the hot compression at different temperatures and strain rates are shown in Fig. 2 . All the true stress–true strain curves show a single peak, which implies the occurrence of the DRX phenomenon during hot deformation. The samples deformed at high temperatures ( i.e. , T ≧ 1343 K) and low strain rates ( i.e. , ε ˙ ≦ 0.01  s − 1 ) show typical DRX flow stress, including a sharp peak followed by a long steady state. At temperatures lower than 1343 K and strain rates higher than 0.01 s −1 , there is no clear steady state evident in the range of strains examined. From the experimental results (shown in Fig. 2 ), it can be seen that the effects of the temperature and strain rate on the flow stress are significant for all the tested conditions. Lower strain rate and higher temperature will decrease the flow stress of 30Cr2Ni4MoV rotor steel, while higher strain rate and lower temperature will increase the flow stress of 30Cr2Ni4MoV rotor steel. This is because at low temperatures, the number of slip systems is restricted, and the processes of recovery softening are not possible. Therefore, strain-hardening mechanisms, such as dislocation intersections and pileups, lead to the increasingly higher stresses required for continued deformation. As the deformation temperature increases, the rate for the vacancy diffusion, cross-slip of screw dislocations and climb of edge dislocations increases. Thus, recovery becomes easier to be activated and overcome hardening. The experimental results show that materials with a high deformation temperature can significantly reduce the flow stress. The peak strain ( ɛ P ) and the peak stress ( σ P ) obtained from the flow curves in Fig. 2 are listed in Table 1 . At a given strain rate, the higher the deformation temperature, the greater the dynamic softening extent and rate, the lower the peak stress. At a certain temperature, the higher the strain rate, the higher the rate of working hardening, the higher the peak stress. It is also found that the value of strain corresponding to the peak stress moves toward a higher strain-level with decreasing the deformation temperature. This is because lower strain rates provide longer time for energy accumulation and higher temperatures result in higher mobility at boundaries for the nucleation and growth of dynamic recrystallized grains and dislocation annihilation, and thus reduce the peak stress level [8] . Therefore, materials with a high deformation temperature are more prone to DRX. 3.2 Calculation of activation energy for hot deformation The Arrhenius equation is widely used to describe the relationship between the flow stress ( σ ), the deformation temperature ( T ) and the strain rate ( ε ˙ ), especially at high temperature. The hyperbolic law in Arrhenius type equation gives a better approximation between Zener–Hollomon parameter and stress [9] : (1) Z = ε ˙   exp Q R T (2) ε ˙ = A [ sinh ( α σ P ) ] n   exp − Q R T where the σ P is the peak stress, or the flow stress correspondent to the specified strain amount, Q is the activation energy for hot deformation, R is the gas constant (8.318 J mol −1 K −1 ), T is the absolute temperature at which deformation occurs, A , α and n are the material constants, and the range of α values is comparatively small (0.012–0.013). In the present work, the α is taken as 0.012 [10] . Taking the logarithm of both sides of Eq. (2) and rearranging we have: (3) ln   sinh ( α σ P ) = − 1 n ln   A + 1 n ln   ε ˙ + 1 n Q R T For the fixed temperature and strain, by differentiating Eq. (3) , the value of n can be expressed as (4) 1 n = ∂   ln   sinh ( α σ P ) ∂   ln   ε ˙ T We obtained n = 4.506 and A = 5.011 × 10 15 by substituting the values of the peak stress and strain rate for all the tested temperatures in to Eq. (3) and the regressed analysis results from Fig. 3 . For the constant strain and strain rate conditions, on taking partial derivative of both sides of Eq. (3) to 1/ T , the value of Q can be expressed as (5) Q = n R ∂   ln   sinh ( α σ P ) ∂ ( 1 / T ) ε ˙ So, by substituting the values of temperature and flow stress that are obtained from a fixed strain rate into Eq. (3) , the value of Q can be obtained as 368.45 kJ/mol from the slope in a plot of ln sinh( ασ P ) as a function of 1/ T ( Fig. 4 ). 3.3 Predicting the critical stress for initiation of DRX In true stress–true strain curves obtained from the hot compression, the onset of DRX can be recognized by a distinct peak in the flow stress in hot working data, due to the softening effect of recrystallization. However, it is well known that dynamic recrystallization actually initiates at a critical strain before the peak stress [11] . The onset of DRX process occurring in critical condition depends on the basic chemical composition of the materials and the hot deformation parameters. Various mathematics models have been put forward to predict the critical conditions for initiation of DRX in previous studies [12–17] . Barnett et al. [12,13] have established a mathematics model to identify the critical strain for initiation of DRX based on static recrystallization (SRX) nucleation mechanism. Gottstein et al. [14] predicted the critical strain for initiation of DRX on the basis of a dislocation work hardening model. Poliak and Jonas [15] put forward a unified description of the flow curve based on normalizing the stresses and strains by the respective peak or steady state values. Thus they speculate that DRX will be initiated satisfying two conditions: (1) maximum stored energy and (2) minimum dissipation rate. It can be expressed in the following formula: (6) ∂ ∂ σ − ∂ θ ∂ σ = 0 where θ is the work hardening rate (θ = ∂ σ /∂ ɛ ). As mentioned Eq. (6) , the critical points for the initiation of DRX can be determined directly from the inflections in − (∂θ/∂ σ ) − σ plots. Furthermore, the method of Poliak and Jonas was subsequently simplified by Najafizadeh and Jonas [16] . In their study, an appropriate third order polynomial equation was fitted to the strain hardening data. Recently, Mirzadeh and Najafizadeh [17] compared the following three methods for determining both the critical stress and strain for initiation of DRX using: (1) the strain hardening rate versus stress curve, (2) the natural logarithm of strain hardening rate versus strain curve, and (3) the constitutive equations, and they concluded that the θ – σ analysis and the ln θ – σ analysis were the best methods for determination of the critical stress and the critical strain, respectively. In this work, the method similar to Najafizadeh and Jonas [16,17] was used for determination of the critical stress for initiation of DRX. The inflection point is detected by fitting a third order polynomial to the θ – σ curves up to the peak point as follows: (7) θ = A 1 σ 3 + A 2 σ 2 + A 3 σ + A 4 where A 1 , A 2 , A 3 and A 4 are constants for a given set of deformation conditions. The second derivative of Eq. (8) with respect to σ can be expressed as: (8) d 2 θ d σ 2 = 3 A 1 σ C + A 2 where σ C is the critical stress. At critical stress for initiation of DRX, Eq. (8) equals to zero. Therefore, it can be rewritten as (9) 3 A 1 σ C + A 2 = 0 ⇒ σ C = − A 2 3 A 1 The experimental θ – σ data as well as its corresponding third order polynomials and curves that best fit these data are shown in Fig. 5 . It can be observed from the figure that the 3rd curve has a good fit as previously demonstrated by Poliak et al. [15–17] . According to this method, the critical points for the initiation of DRX can be obtained from the derivative of the third order polynomial versus true stress. The results are shown in Fig. 6 . All the third order polynomials for dynamic recrystallization of 30Cr2Ni4MoV rotor steel are presented in Table 2 . Fig. 7 illustrates the relation between the critical stresses calculated using Eq. (9) and the Zener–Hollomon parameter. As can be seen from Fig. 7 , there is a linear relationship between critical stress ( σ C ) and ln Z : (10) σ C = 7.704   ln   Z − 122.357 3.4 DRX microstructures The effects of forming temperature and strain rate on the microstructural evolution during dynamic recrystallization in hot deformed 30Cr2Ni4MoV rotor steel were investigated when the true strain achieved was 0.8. Fig. 8 (a–c) shows the optical microstructures of test steel after deformation at strain rate of 0.001 s −1 under the temperatures of 1243 K, 1343 K and 1443 K, respectively. It is obvious that the deformed austenite grain size decreases with decreasing the forming temperature. The average grain sizes are measured as 41.9 μm, 102.3 μm and 150.5 μm for the forming temperatures of 1243 K, 1343 K and 1443 K, respectively. From Fig. 8 (a–c), we can observe that the dynamically recrystallized grain size is refined with decreasing the deformation temperature. This is mainly because higher temperature can lead to higher deformation stored energy, therefore increasing the nucleation rate for dynamic recrystallization, refining dynamic recrystallization grain size. Fig. 8 (c–e) illustrates the optical deformed microstructures of test steel after deformation at 1443 K under strain rates of 0.001 s −1 , 0.01 s −1 and 0.1 s −1 , respectively. It can be easily found that the higher the strain rate, the finer the grains and the average grain sizes are measured as 150.5 μm, 92.0 μm and 76.3 μm for the strain rates of 0.001 s −1 , 0.01 s −1 and 0.1 s −1 , respectively. There are mainly two reasons for this tendency: one factor is that higher strain rate will increase the work hardening rate. It is well known that the driving force of grain nucleation and growth during dynamic recrystallization in metals is mainly related to the reduction of the deformation energy introduced by work hardening. With the increase of strain rate, there are more deformation energies stored in the deformed blocks, which will make the dislocation generation rate, the dislocation density and nucleation sites increase in the deformed microstructure. Then, more substructures can be generated in the initial grains when strain rate is higher, which will produce more nuclei per unit volume of the grains. This mechanism can make the grain finer when strain rate is higher [18] . The other reason is that higher strain rate will reduce the time for austenite recrystallization; therefore there is no sufficient time for austenite grain to grow up at a certain strain. Combining the above two factors, austenite grain size increases with decreasing the strain rate. 3.5 DRX kinetics Hot deformation in metals occurs by the movement of an increasing number of dislocations on specific crystal planes. The occurrence of work hardening therefore implies that, as the true strain is increased, it becomes more difficult either to generate or to move the relevant dislocations. It is generally accepted that only when dislocation density or strain reaches a critical value dynamic recrystallization can occur during hot deformation. With the strain increasing, when multiplication of dislocation due to continued deformation and annihilation of dislocation due to dynamic softening are in dynamic balance, flow stress would keep constant, and deformation of metals would comes to a steady stage in which complete dynamic recrystallization grains have equiaxed shape and keep constant size [19] . In general, whatever technique is used, studies of dynamic recrystallization show that it progresses in a sigmoidal manner with respect to time. The basic characteristic of DRX kinetics is that there is an initial incubation period which is followed by a rate of change which is initially slow, then accelerates and, finally, becomes slow again as the process is completed [20] . For a constant strain rate, Kinetics of DRX can be expressed as following [21] : (11) X drx = 1 − exp − k drx ε − ε c ε P n drx   ( ε ≥ ε c ) where X drx is the dynamic recrystallization fraction, k drx and n drx are dynamic recrystallization parameters depending on chemical composition and hot deformation conditions, ɛ , ɛ c and ɛ P are the true strain, the critical strain and the peak strain, respectively. The DRX volume fraction under different deformation conditions can be determined by either quantitative metallographic measurement or flow curve analysis. Flow curve analysis is a simpler and quicker method compared to quantitative metallographic measurement, as the latter needs a large quantity of samples to be examined and it can be difficult to precisely ascertain the new grains [22] . Here, to determine the progress of dynamic recrystallization fraction ( X drx ), the following expression is employed [23] : (12) X drx = σ WH − σ σ − s σ s s   ( ε ≥ ε c ) where σ WH denotes the flow stress if dynamic recovery is the only softening mechanism, σ s is the saturation stress, σ ss is the steady-state stress. From Eq. (13) , the amount of DRX is determined by further analysis of the true stress–true strain curves of DRX ( Fig. 2 ) under different deformation conditions. The specific methods for calculating the values of σ WH , σ s and σ ss can be found in Hodgson's work [21] . Combining Eqs. (11) and (12) , k drx and n drx are determined. As can be seen from Fig. 9 , there is a linear relationship between ln[ − ln(1 − X drx )] and ln[( ɛ − ɛ C )/ ɛ P ] under different deformation conditions, therefore the resulting equation for X drx is given by (13) X drx = 0 ε < ε C X drx = 1 − exp − 1.092 ε − ε C ε P 1.527 ε ≥ ε C 3.6 Grain size of complete DRX As shown in Fig. 8 , complete dynamic recrystallization grain size ( D drx ) depends on the deformation temperature and strain rate. In other words, the dynamic grain size is a function of Zener–Hollomon parameter ( Z ) alone and independent of the accumulated strain. Some of D drx and Z are listed in Table 3 . So D drx is a power-law function of Zener–Hollomon parameter: (14) D drx = B Z n where B and n are the materials constants. As shown in Fig. 10 , B and n are determined by regression analysis as B = 5.855 × 10 4 and n = −0.243. Therefore, the equation of complete dynamic recrystallization grain size for 30Cr2Ni4MoV rotor steel is given by: (15) D drx = 5.855 × 10 4 Z − 0.243 4 Conclusion The hot deformation behavior of 30Cr2Ni4MoV ultra-super-critical rotor steel has been investigated to characterize the evolution of the DRX structure by uniaxial hot compression tests on Gleeble-3500 thermo-mechanical simulator. The effect of different thermomechanical processing parameters such as temperature and strain rate has been investigated. From this study, the following conclusions are drawn: (1) The activation energy for hot deformation of 30Cr2Ni4MoV ultra-super-critical rotor steel is 368.45 kJ/mol. (2) The critical stress for initiation of DRX can be calculated by setting the second derivative of the third order polynomial versus true stress, and there is a linear relationship between critical stress ( σ C ) and ln Z σ C = 7.704   ln   Z − 122.357 (3) The dynamic recrystallization fraction ( X drx ) can be expressed as X drx = 0 ε < ε C X drx = 1 − exp − 1.092 ε − ε C ε P 1.527 ε ≥ ε C (4) The complete dynamic recrystallization grain size ( D drx ) of the test steel is a function of Zener–Hollomon parameter ( Z ) alone, and it can be described as D drx = 5.855 × 10 4 Z − 0.243 Acknowledgements Funding support from Science and Technology Commission of Shanghai Municipality (Project No. 08DZ1100302) and National Natural Science Foundation of China (Grant No. 51075270 ), it is gratefully acknowledged. References [1] http://www.bwe.dk/index.html [2] F.-Z. Xuan X.-Q. Huang S.-T. Tu Mater. Des. 29 2008 1533 1539 [3] D. Rai B. Singh J. Singh Wear 263 2007 821 829 [4] S. Maropoulos N. Ridley J. Kechagias S. Karagiannis Eng. Fract. Mech. 71 2004 1695 1704 [5] Y.-B. Wang W.-Y. Chu L.-J. Qiao J. Mater. Sci. Technol. 16 2000 631 633 [6] F. Chen Z.-S. Cui J. Liu X.-X. Zhang W. Chen Model. Simul. Mater. Sci. Eng. 17 1–19 2009 75015 [7] F. Chen Z.-S. Cui J. Liu W. Chen S.-J. Chen Mater. Sci. Eng. A 527 2010 5539 5549 [8] Y.-C. Lin M.-S. Chen J. Zhong Mech. Res. Commun. 35 2008 142 150 [9] C.M. Sellars S. Yue Proceedings of the International Conference on Mathematical Modeling of Hot Rolling of Steel CIMM, Hamilton 1990 1 [10] B.-Z. Wang W.-T.,Z.-Q.L. Fu P. Jiang W.-H. Zhang Y.-J. Tian Mater. Sci. Eng. A 487 2008 108 113 [11] E.I. Poliak J.J. Jonas Acta Mater. 44 1996 127 136 [12] M.R. Barnett G.L. Kelly P.D. Hodgson Scripta Mater. 43 2000 365 369 [13] M.R. Barnett G.L. Kelly P.D. Hodgson Metall. Mater. Trans. A 33 2002 1892 1900 [14] G. Gottstein M. Frommert M. Goerdeler N. Schafer Mater. Sci. Eng. A 387–389 2004 604 608 [15] E.I. Poliak J.J. Jonas ISIJ Int. 43 2003 684 691 [16] A. Najafizadeh J.J. Jonas ISIJ Int. 46 2006 1679 1684 [17] H. Mirzadeh A. Najafizadeh Mater. Des. 31 2010 1174 1179 [18] Y.-C. Lin M.-S. Chen Mater. Sci. Eng. A 501 2009 229 234 [19] G.-L. Ji F.-G. Li Q.-H. Li H.-Q. Li Z. Li Mater. Sci. Eng. A 527 2010 2350 2355 [20] P. Cotterill P.R. Mould Recrystallization and Grain Growth in Metals 1976 Wiley Edition [21] L.X. Kong P.D. Hodgson B. Wang J. Mater. Process. Technol. 89–90 1999 44 50 [22] A. Dehghan-manshadi M.R. Barnett P.D. Hodgson Metall. Mater. Trans. A 39A 2008 1359 1370 [23] A. Laasraoui J.J. Jonas Metall. Trans. A 22A 1992 1545 1558