Fundamental Materials Science of the TRIP Phenomenon

Determination of the volume amount of retained austenite and ferrite in small specimens by magnetic measurements
Edwin Wirthl, Andreas Pichler, Reinhold Angerer, Peter Stiaszny, Karl Hauzenberger, Yuri F. Titovets, and Michael Hackl
The accurate determination of the volume fraction of retained austenite is essential for the optimisation of low alloyed TRIP steels. In this paper a magnetic method for the measurement of the volume amount of retained austenite is presented. Based on this method a new equipment was built, which allows the determination of the amount of austenite in an easy and quick way. Based on experimental investigations on samples of industrially produced TRIP steels the high accuracy of the method and the agreement with X-ray measurements is shown. This accurate determination of the amount of retained austenite allows a better understanding of the metallurgical processes and therefore an efficient optimisation of TRIP steel grades.

For the improvement of the passive safety of cars and a simultaneous reduction of the mass of the body in white the application of high strength thin sheet steel grades are of fundamental interest. To allow an innovative design an appropriate formability is necessary and grades with an excellent balance between the strength level and the formability are of particular interest. The development and application of BH, high strength IF and DP grades by the steel and automotive industry was a consistent step in this direction. A further improvement of this balance between strength and ductility which is characterized often by the product of tensile strength and uniform or total elongation is obtained by the application of a transformation induced plasticity. The development and application of low alloyed thin sheet TRIP steels [1; 2] (TRIP - transformation induced plasticity) is therefore a remarkable large step to solve the mentioned consumer demands. The outstanding elongation and formability of this TRIP steels at a very high strength level results from a strain induced martensitic transformation of metastable retained austenite during the deformation. Since Zackays paper [3] the beneficial phenomenon of transformation induced plasticity has been well known for high-alloyed steels, but for thin sheet applications austenite-stabilizing elements like Ni are too expensive. In the case of low-alloyed TRIP grades the retained austenite is therefore mainly stabilized by C and to some extent by Mn. The small size of the austenite grains and the surroundings support the stabilization of the retained austenite. The mechanical properties of TRIP steels are in turn determined by the amount and stability of the retained austenite. The precise, fast and cheap measurement of the amount of retained austenite is therefore an essential item for further research work, improvement and control of TRIP steels. Additional points are the determination of the stability of the retained austenite based on the measurement of the carbon content in the retained austenite, the grain size of the austenite and the characterization of the ausDr. Edwin Wirthl, Project Manager; Dr. Reinhold Angerer, General Manager; DI. Karl Hauzenberger, Project Manager; Michael Hackl, Project Manager, Voest Alpine Mechatronics GmbH, Linz; Dr. Andreas Pichler, Project Manager; Dr. Peter Stiaszny, General Manager, Voest Alpine GmbH, Linz, Austria; Prof. Dr. Yuri F. Titovets, Head of the Staff, Materialphysics & Computer Technology for Material Science Department, St. Petersburg University, St. Petersburg, Russia.

tenite surrounding. The stability of the retained austenite can be described on the base of the decrease of the amount of retained austenite as a function of plastic prestrain. In the field of the work on TRIP steels mainly X-ray diffraction measurements were used for the determination of the amount of retained austenite. Sometimes also classical metallography, Mssbauer and Neutron diffraction are used. A detailed description of this methods in comparison to magnetic measurements is given by Zhao et al. [4]. In some recent investigations in the field of the development of TRIP steels a magnetic method is used very successfully [2; 5; 6]. Based on these measurements a satisfying correlation was obtained between the processing parameters and the mechanical properties. In the present work a new equipment enabling an easy and fast determination of the fraction of retained austenite in TRIP steels via the magnetic method is presented. Based on a statistical evaluation the accuracy of the method is investigated in detail. Additionally, the method is critically compared with metallographic investigations and measurements based on X-ray diffraction. Finally, a useful application of the magnetic measurement method for material development is demonstrated.

Description of the method of measurement

The equipment consists of a magnetic yoke, which produces a high and homogenous magnetic field between its poles. A magnetic flux sensing coil, called the measurement coil, is mounted in the centre of this magnetic field. In figure 1 a general sketch of the equipment is shown.

Figure 1. General sketch of the measuring method

Int. Conf. on TRIP-Aided High Strength Ferrous Alloys

61

Fundamental Materials Science of the TRIP Phenomenon

The specimen to be measured is pushed through the measurement coil along its axis. The integral of the voltage pulse which is induced in the coil is measured. From this signal, the amount of the retained austenite in the sample can be calculated. Physical fundamentals. When a magnetic specimen is pushed through the measurement coil, the magnetization J of the specimen induces a voltage-pulse Uind in the coil as it is known from the law of induction. Uind can be calculated as U ind = N A dJ dt (1)

Determination of the amount of retained austenite in alloyed steel. Alloying elements decreases the saturation magnetization of steels. In formula 2 the saturation magnetization of pure ferrite has to be replaced by the saturation magnetization of the austenite-free saturation magalloy Fe netization of the alloy JS JS . alloy There are two ways to obtain JS : The first method is to measure the saturation magnetizaalloy of a reference specimen of the same composition JS tion as the interesting specimen with the guarantee that there is no austenite in the reference specimen. In this case the volume amount of austenite is calculated by:

where N is the number of loops of the measurement coil, A is the cross area of the measurement coil. H. Weberberger [7] has shown that the shape of the sample has no influence on the measured signal, when the length of the measurement coil is much greater than its diameter and the length of the specimen. In our case we use cylindrical shaped specimens with volume V that is determined from the mass and the density of the specimen. And so the magnetization J of the specimen can be calculated from Uind in the following way: J= U ind dt N V l (2)

J alloy J m 100% . Aaust = s J alloy

S

(4a)

If no reference specimen exists the influence of the alloying elements on the magnetic behaviour of the steel has to be considered. As mentioned above, the addition of alloys decreases the saturation magnetization of steel. Some authors have calculated the decreasing-factor for the most important alloys [7; 11...13]. The dimension of is T/%. In this case the volume amount of austenite is calculated as following: Fe JS ( n An ) J m n 100% Aaust = Fe ( n An ) JS
n

where V is the volume of the specimen and l is the length of the measurement coil. Determination of the amount of retained austenite in pure iron. Ferrite and austenite differ in their magnetic behaviour. And so only the ferritic amount of the sample contributes to the induced voltage pulse in the measurement coil. To guarantee that all the ferrite in the sample is magnetized, the magnetic field of the yoke must be high enough to bring the specimen in state of magnetic saturation. For pure ferrite the saturation magnetization was deFe termined as JS = 2.158T [8]. In the case of a mixture of ferrite and austenite the saturation magnetization of the specimen Js decreases with arising amount of austenite which could not be magnetized. The volume amount of austenite Aaust (in %) can now be calculated in the following way:

(4b)

where an and An (in %) are the decreasing factor and the amount of alloy n in the specimen. It must be mentioned that this method is not as exact as the first method using a reference specimen, because up to now there is no knowledge of the behaviour of for compositions of low alloyed TRIP steel grades.

Table 1. Results and mean values for the methods with (a) and without (b) reference sample Sample No. 960200A/124 960200A/125 960200A/127 960200A/128 %Ra -#1 10.2956 8.9203 12.2534 14.0964 11.0607 13.7751 10.2846 8.9198 12.2641 14.0867 11.0527 13.7706 %Ra -#2 %Ra -#3 %Ra -#4 %Ra -#5 %Ramean 10.3600 8.9669 12.4633 14.0516 11.0146 13.7603 10.3617 8.9585 12.4632 14.0519 11.0016 13.7590 10.3387 9.01329 12.3209 14.1113 11.0678 13.8372 10.3464 9.0119 12.3274 14.1027 11.0161 13.9393 10.4207 8.9481 12.3174 14.0486 11.0565 13.7004 10.4236 8.9447 12.3822 14.0257 11.0797 13.7049 10.4286 9.0671 12.4004 14.0112 11.1890 13.6143 10.4294 9.0771 12.4088 14.1021 11.2397 13.7120 10.37 8.98 12.35 14.06 11.08 13.74 10.37 8.98 12.37 14.07 11.08 13.78 ,% 0.056 0.058 0.082 0.040 0.066 0.084 0.060 0.063 0.076 0.034 0.096 0.095

a) with reference sample

Aaust =

JsFe

Jm
S

J Fe

100%

(3)

960200A/129 960200A/130 960200A/124 960200A/125 960200A/127 960200A/128 960200A/129 960200A/130

b) without reference sample

where Jm is the saturation magnetization of the specimen calculated from formula (2). This simple correlation is only valid in state of magnetic saturation. Otherwise the signal would show an exponential behaviour [9; 10].
62

Int. Conf. on TRIP-Aided High Strength Ferrous Alloys

Fundamental Materials Science of the TRIP Phenomenon

It can be seen that the accuracy is better than 0.1 % for the measured amount of retained austenite for all samples and the agreement between these two methods is also satisfying. Therefore all following magnetic measurements in the present paper are done by the method without a reference sample.

Comparison between X-ray measurement and the magnetic method

For comparison X-ray diffraction measurements were additionally performed on two industrial produced TRIP steels. A detailed description of the applied experimental procedure to determine the amount of retained austenite from the diffraction pattern is given in [14]. On the two samples the austenite volume fraction were determined with the X-ray measurement method as a function of the distance from the surface. For both samples an increasing austenite content is determined with an increasing distance from the surface, figure 2. In the centre of the samples a retained austenite content of 13.6 and 14.5 % was measured. The results from the magnetic measurement are shown in figure 2 as lines, too. As can be seen the magnetic method gives lower values than the X-ray measurement in the centre of the sample. The difference between the two samples however are in the same order for both measurement methods. Nevertheless the magnetic method measures the mean volume amount of retained austenite over the whole sample in contrast to the X-ray method which determines the local amount of retained austenite on the surface.

Figure 2. Comparison of the measured content of retained austenite of the samples S1 and S2

Comparison between the two methods. The most comfortable method to determine the amount of retained austenite is to use formula 4b, because, there is no need of any reference sample. Therefore it must be sure that both methods are in good agreement. We determined the amount of retained austenite in some samples with both methods. In table 1a and b the results are presented.

Application of the measurement

In the following part the useful application of the measurement of the retained austenite content based on the magnetic method will be demonstrated on a selected example. As already mentioned in the introduction a decisive parameter is the annealing treatment in the range of the bainitic transformation. Therefore in the following section the impact of the strip temperature between the cooling section and the zinc pot on the microstructure and mechanical properties is shown and critical discussed. Figure 3 shows the microstructure for overaging temperatures of 510, 470 and 430 C. Very small dark-brown regions are observed in combination with white retained austenite at the highest temperature of 510 C. Based on the optical microscopy these regions might be attributed to pearlite. Regions with the typical appearance of bainite were not seen at such high overaging temperatures. At the lower temperature of 470 C no pearlite is observed and a quite small fraction of bainite (light-brown) is seen. For the lower temperature of 430 C a remarkable increase of the amount of bainite is evident. The fraction of retained austenite measured with the magnetic method is shown in figure 4 as a function of the overaging temperature. An increase of the amount of retained austenite is observed for increasing overaging temperatures up to 490 C. Even higher temperatures result in a significant decrease.
63

Figure 3. Influence of the strip temperature in front of the zinc pot on the microstructure (TAN = 780 C, v = 70 m/min)

Figure 4. Influence of the strip temperature in front of the zinc pot on the amount of retained austenite (TAN = 780 C, v = 70 m/min)

Int. Conf. on TRIP-Aided High Strength Ferrous Alloys

Fundamental Materials Science of the TRIP Phenomenon

ter overaging at about 430 C in the hot dip galvanizing line due to the shorter overaging time. Higher, overaging temperatures results however in more but less stable retained austenite and therefore not as excellent mechanical properties. Due to the lower stability of the retained austenite a faster transformation of austenite occurs and results in more dual phase behaviour with lower yield strength and higher tensile strength. Even higher temperatures result in a loss of carbon to pearlite or cementite and therefore also in a decrease of the tensile strength.

Conclusions
a) Influence of overaging temperature on strength

In the present paper, a new method is presented, which allows to determine the amount of retained austenite in TRIP steel samples via magnetic measurements in a quick and easy way. The accuracy of the measurement is better than 0.1 % and is in good agreement with X-ray measurements. In addition the favourable support of the measurement of the amount of retained austenite for the development and improvement of TRIP steels is demonstrated.

References
[1] Sakuma, Y.; Matsumura, O.; Takechi, H.: Met. Trans. 22A (1991), p. 489. [2] Pichler, A.; Traint, S.; Pauli, H.; Mildner, H.; Szinyur, J.; Blaimschein, M.; Stiaszny, P.; Werner, E. A.: Processing and Properties of Cold-Rolled TRIP Steels, [in:] 43rd Mechanical Working and Steel Processing Conf. Proc., ISS, Warrendale, 2001, p. 411. [3] Zackay, V. F.; Parker, E. R.; Fahr, D.; Busch, R.: ASM Trans. Quart. 60 (1967), p. 252. [4] Zhao, L.; van Dijk, N.H.; Brck, E.; Sietsma, J.; van der Zwaag, S.: Mat. Sci. Eng. 313A (2001), p. 145. [5] Pichler, A.; Stiaszny, P.; Potzinger, R.; Tikal, R.; Werner, E. A.: TRIP Steels with Reduced Si Content, [in:] 40th Mechanical Working and Steel Processing Conf. Proc., ISS, Warrendale, 1998, p. 259. [6] Traint, S.; Pichler, A.; Tikal, R.; Stiaszny, P.; Werner, E. A.: Influence of Manganese, Silicon and Aluminum on the Transformation Behavior of Low Alloyed TRIP-Steels, [in:] 42nd Mechanical Working and Steel Processing Conf. Proc., ISS, Warrendale, 2000, p. 549. [7] Weberberger, H.: Magnetische Messungen an Proben beliebiger Form mit dem Fluxmeter, [in:] special print Elektrotechnik und Maschinenbau 87 (1970) No. 2. [8] Sachec, A.K.: Acta Met. 31 (1983), p. 2037/42. [9] Bastien, P.; Sulmo, A.: Rev. Mt. 61 (1959), p. 131/43. [10] Spiegelberg, K.; Beckmann, G.: Neue Htte 7 (1932) No. 3, p. 167/74. [11] Bozorth, R.M.: Ferromagnetism, D. van Nostrand Company Inc., Princtone, 1951. [12] Heck, C.: Magnetische Werkstoffe und ihre technische Anwendung, Hthig Verlag, Heidelberg, 1975. [13] Hoselitz, K.: Ferromagnetic properties of Metals and Alloys, Calderon Press, Oxford, 1952, p. 297. [14] Titovets, Y. V.: to be published. [15] Traint, S.; Pichler, A.; Stiaszny, P.; Werner, E. A.: Mechanical Properties and Phase Transformations of an Aluminium Alloyed TRIP-Steel, 43rd Mechanical Working and Steel Processing Conf. Proc, ISS, Warrendale, 2001, p. 449.

b) Influence of overaging temperature on elongation Figure 5. Influence of the strip temperature in front of the zinc pot on the mechanical properties (TAN = 780 C)

Figure 5 shows the mechanical properties as a function of the overaging temperature. The yield strength increases remarkably when lowering the overaging temperature from 470 to 450 C. A strong decrease is observed for the tensile strength in the same temperature range. The significant decrease of the tensile strength at the highest temperature is remarkable. The largest values for the uniform and total elongation are obtained at about 430 C. From the investigations in [2; 15] it can be deduced that a maximum of bainite and a minimum of retained austenite should be observed at an overaging temperature of roughly 400 C, because higher temperatures result in a reduced driving force and lower ones in a reduced transformation kinetics, and therefore less bainite, in the available time of the overaging treatment in the line. The strong decrease of the fraction of retained austenite seen at overaging temperatures as high as 510 C is a result of the precipitation of cementite or the formation of pearlite. Microstructures with a high fraction of bainite and therefore a lower fraction of very stable retained austenite have excellent mechanical properties due to the delayed transformation of the austenite to martensite. Following these arguments the best mechanical properties are expected af-

64

Int. Conf. on TRIP-Aided High Strength Ferrous Alloys