Development of TBF Steels with 980 MPa Tensile Strength for Automotive Applications: Microstructure
and Mechanical Properties
A. Bachmaier, K. Hausmann, D. Krizan, A. Pichler
voestalpine Stahl Linz GmbH, voestalpine-Str. 3, 4020 Linz, Austria
Key words: TBF grades, influence of the processing parameters¸ continuous annealing line, mechanical properties, microstructure
In this paper, an overview over the development of industrially processed advanced high strength sheet steels for automotive
applications with a tensile strength of 980MPa and an improved ductility compared to the widely used high-strength dual-phase and
complex phase automotive steel grades is given. For these advanced high strength steels with tensile strength above 980MPa which
enable lighter and safer car-bodies a steadily increasing demand is observed. Optimal processing parameters are chosen according to
different heat treatment schedules in lab trials. Phase transformations are investigated by dilatometric measurements. Furthermore, the
mechanical properties are determined. The microstructure of these materials plays a key role in their mechanical properties. Therefore,
the parameters of the heat treatment cycles are adapted to obtain the microstructure which lead to the desired mechanical properties
during processing via continuous annealing lines.
As a response to the steadily increasing requirements of the automotive sector regarding lighter and safer car bodies, advanced high
strength sheet steels with a tensile strength above 980 MPa and improved formability (ductility and bendability) compared to
conventional dual-phase and complex-phase steel grades with the same strength level are currently developed by the steel industry .
TRIP bainitic ferrite (TBF) and Quench-and-Partitioning (Q&P) steels are such steel grades which are belonging to the group of so
called “third generation of advanced high strength steels” . The microstructure of TBF steels consists of a matrix of bainitic ferrite
with retained austenite inclusions and is produced by isothermal holding in the bainitic regions after fast cooling from fully austenitic
microstructures . Q&P steels exhibit a matrix which consists of martensite, tempered martensite or/and lower bainite with retained
austenite inclusions produced by isothermal holding after partial martensite transformation by short cooling or by isothermal holding
below the martensite start after fast cooling from fully austenized microstructures . During straining, the TRIP (TRansformation
Induced Plasticity) effect of retained austenite known from conventional TRIP steels is used as well which leads to excellent
formability at room temperature. By the combination of the different phases in TBF and Q&P steels, the properties can be tailored for
specific applications. TBF steels (“dual-phase type” with high strain hardening) which can be used for deep drawing applications or
Q&P steels (“complex-phase type” exhibiting higher yield ratios, reduced elongations and reduced n-values) with an excellent
bendability can be used for bending operations or roll forming.
The most important step during processing is the annealing of the as-cold rolled material via continuous annealing line. Typical “base”
chemical compositions of TBF steels contain C, Si and Mn as major alloying elements. Alloy modifications include variations of the
Al, Nb and Cr content . Si suppresses the formation of cementite during bainitic transformation which enhances the C content in
retained austenite and thus allows the austenite to be stabilized by carbon [6, 7]. High Si contents of 1.5wt% are usually used in these
types of steels . As a consequence, the transformation of retained austenite into martensite upon deformation and/or thermally
produced martensite during final cooling is prevented. Although Si is of major importance to prevent carbide precipitation during
annealing of the cold rolled material, it causes problems during processing via continuous annealing lines. Si alloyed steels exhibit
selective oxidation at the steel surface resulting, for example, in deteriorated galvanizability . Therefore, other alloying elements
having a similar effect of suppressing carbide formation have to be considered. The influence of reduced Si contents (<1wt%) in
combination with the addition of other alloying elements completely or partially substituting Si has been studied recently . Based
on the results, commercially produced cold rolled material was manufactured, investigated in detail by different lab trials to obtain
ideal processing parameters for the final annealing via continuous annealing to obtain TBF and Q&P steel grades and produced on
industrial scale to obtain the desired mechanical properties.
The results in this paper are obtained from a commercially produced cold rolled material which was cold rolled to a final thickness of
1.4 mm. The chemical composition is given in Table 1.
Table 1: Chemical composition of the investigated steel grade in wt. %.
To optimize annealing parameters and to study the transformation behavior, annealing simulations in the laboratory with the Multi-
Purpose Annealing Simulator (MULTIPAS) and dilatometric investigations on a Bähr dilatometer DIL805 A/D were performed on
specimens prepared from the commercially processed cold rolled material. Dilatometric investigations with different cooling rates and
varying overaging temperatures and time (I-IV) were conducted. Volume fractions of transformed phases during the dilatometric
investigations are derived from the dilatometric changes using the lever rule.
Fig.1: Schematics of the applied laboratory annealing series.
Furthermore, annealing series on the basis of actual continuous annealing line layouts were conducted in the laboratory with the
MULTIPAS (III-IV). Summarized, the following annealing series were applied:
I) Specimens were cooled with different cooling rates (3K/s - 80K/s) to room temperature after annealing for 60s at 900°C.
II) Specimens were heated to an annealing temperature of 900°C, held at the annealing temperature for 60s to obtain fully austenitic
microstructures and subsequently quenched with a cooling rate of 70K/s to different isothermal holding temperatures between 325°C
and 500°C for an isothermal holding time of 600s.
III) Specimens were heated to an annealing temperature of 850°C, held at the annealing temperature for 60s to obtain fully austenitic
microstructures and subsequently quenched with a cooling rate of 30K/s and 50K/s to three isothermal holding temperatures (400°C,
425°C, and 450°C) and for different isothermal holding times of 30-600s. Dilatometric investigations were only performed with an
overaging time of 600s.
IV) Q&P heat treatment cycles according to industrial processing and annealing layouts were performed. Specimens were heated to
annealing temperatures of 850°C, held for 60s, cooled with two different cooling rates (30K/s and 50K/s) to three different quench
temperatures (350°C- 380°C) and two different overaging temperatures (400°C and 440°C).
A schematic of the annealing series applied in dilatometric investigations as well as in the annealing simulations is shown in Fig. 1.
The most important annealing parameters for the applied annealing series are summarized in Table 2.
Samples for light optical metallography were prepared by standard metallographic preparation (etched with LePera’s etchant) .
Volume fraction of retained austenite was measured using the saturation magnetization method . The mechanical properties with
tensile specimens machined with their tensile axis parallel to the rolling direction were determined according to testing procedure DIN
EN ISO 6892-1.
Additionally, cold rolled sheets were industrially processed via continuous annealing line by choosing the optimal processing
parameters obtained by the lab trials and subsequent microstructural characterization was performed as described above. Furthermore,
the mechanical properties with tensile specimens machined with their tensile axis parallel and perpendicular to the rolling direction
Table 2: Annealing parameters (Tan...annealing temperature, tan…annealing time, TOA…overaging temperature, tOA…overaging time,
TQ…quenching temperature and CR…cooling rate) for dilatometric investigations (I-IV) and MULTIPAS simulations (III-IV).
Annealing series T
(s) CR (K/s) T
III 850 60
IV 850 60
RESULTS AND DISCUSSION
In order to optimize annealing conditions to obtain TBF and Q&P steel grades, dilatometric measurements were performed. The onset
of austenite formation during heating (Ac1=753°C), the temperature of the completion of austenite formation (Ac3=831°C) and the
martensite transformation start and finish temperature (Ms=371°C and Mf=289°C, respectively) were calculated from dilatometric data
during continuous cooling (I). Furthermore, the isothermal transformation behavior in the bainitic holding range was investigated
which is an essential step during the production of TBF grades via continuous annealing lines. The overaging temperature influences
not only the transformation behavior and the kinetics of transformation, it also determined the amount of austenite transformed during
isothermal holding. A phase transformation during cooling from the annealing temperature could not be observed for cooling rates
>50 K/s in the continuous cooling experiments. Therefore, a cooling rate of 70K/s and holding temperatures well above Ms were
chosen to solely investigate the transformation behavior in the bainitic formation range (II). Fig.2a shows the transformation kinetics
of the austenite during the isothermal holding at different holding temperatures from 400°C to 500°C. For the highest isothermal
holding temperature a two-step reaction is observed. The first step of transformation which is the bainitic reaction is completed very
fast (tOA<60s). Carbide precipitation is subsequently observed. At lower overaging temperatures, a one-step transformation behavior is
observed but the transformation kinetic is decelerated with decreasing overaging temperature due to a lower diffusion. Nevertheless an
increasing amount of bainite is formed with decreasing overaging temperature. Transformation kinetics is rather slow at 400°C and
425°C, but the transformation is completed within the overaging times which are typical of a continuous annealing line (~600s) and
the overall amount of formed bainite is maximized. The temperature dependence of the bainite formation can be explained by the
well-known To-concept . If a displacive growth mechanism of bainitic ferrite is assumed, the transformation stops if the C content
of the remaining austenite reaches the To-boundary which is given by the intersection point between Gibb’s energy curves of ferrite
and austenite having an identical composition at a certain temperature. Lower overaging temperatures allow higher C contents in the
untransformed austenite. For transformation temperatures of 400°C and 425°C, carbide precipitation is not observed within the
holding times investigated and high amounts of retained austenite are stable at room temperature. At higher holding temperatures
(450°C and 500°C) the amount of austenite which transforms into bainite becomes less. If carbide precipitation additionally occurs
during isothermal holding (transformation temperature of 500°C), the austenite transforms nearly completely to martensite during final
cooling to room temperature and no or very less retained austenite is stabilized at room temperature which is shown in Fig.2b.
Fig.2: a) Dilatation-time curves obtained from isothermal holding and b) content of retained austenite as a function of the different
In subsequent annealing simulations on the basis of actual continuous annealing line layouts (III), the influence of the overaging
temperature (400-450°C) and time in the bainitic range on the microstructure and mechanical properties with two different cooling
rates (30K/s and 50K/s) was studied (Fig.3a-c). Comparing the different cooling rates, yield and tensile strengths are slightly enhanced
whereas the uniform elongation values are lower for higher cooling rates (Fig.3a-b). Faster cooling rates lead to lower amounts of pro-
eutectoid ferrite and/or bainitic ferrite during cooling. In Fig.3d, the transformed phase fractions before reaching the overaging
temperature for both cooling rates as a function of temperature are plotted. A significant higher amount of transformation at a cooling
rate of 30K/s before reaching the final overaging temperature results in lower measured tensile strength values. In general, a cooling
rate of 30 K/s provides a better combination of strength and elongation values. Comparing different overaging temperatures at this
cooling rate, there is no significant influence of the overaging temperature on the tensile strength for low bainitic temperatures (400°C
or 425°C) if the overaging time is >300s. For short overaging times (<300s), an influence of the slower transformation kinetics at
lower overaging temperatures shown in Fig.2a are reflected in the mechanical properties as well. The tensile strength at these
overaging temperatures decreases due to the increasing amount of bainite formed during longer isothermal holding at the overaging
temperature. As a consequence, the stability of retained austenite is increased which leads to a lower amount of retained austenite
which transforms into martensite upon deformation and/or thermally produced martensite during final cooling. Due to the fastest
transformation kinetics and the lowest amount of bainite formed at an overaging temperature of 450°C, a certain amount of retained
austenite transforms to martensite during final cooling and the highest tensile strength values are measured of specimens annealed at
this overaging temperature which are not influenced by the overaging time.
Yield strengths are significantly influenced by the overaging temperatures whereby increasing amounts of bainite formed results in
increasing yield strength values. Only the specimens annealed at 400°C exhibit higher yield strengths than 700 MPa. The lowest yield
strength are obtained for an overaging temperature of 450°C independent of overaging time which fits quite well to the transformation
A maximum of the uniform elongation is observed for each overaging temperature. For overaging temperatures of 425°C and 450°C,
the highest uniform elongation values are obtained at an overaging time of 120s. At the lowest overaging temperature, the maximum is
shifted to longer holding times. At high overaging times, nearly no difference between uniform elongation values for the samples
annealed at 450°C and 425°C with a cooling rate of 30K/s is visible. A higher amount of retained austenite stabilized at room
temperature results in higher uniform elongations values. The stability of retained austenite is a combined effect of a chemical
stabilization due to an optimum concentration of carbon in austenite and a size effect [14, 15]. At high overaging temperatures,
austenite to bainite transformation is quite fast but the austenite carbon concentration is reduced for longer overaging times by carbide
precipitation which lowers the stability of the retained austenite. Due to the slower kinetics of austenite transformation at lower
overaging temperature, application of longer overaging times results in higher uniform elongation values. An ideal combination of
yield strength, tensile strength and uniform elongation for a TBF steel is obtained for a cooling rate of 30K/s and an overaging
temperature of 400°C.
Fig.3: a-c) Influence of the overaging temperature and overaging time as well as the cooling rate on the mechanical properties. d)
Influence on the cooling rate on the transformed phase fraction upon cooling from the annealing temperature until the final overaging
temperature is reached.
Additionally, Q&P heat treatment annealing simulations according to industrial processing and annealing layouts were performed to
obtain parameters for industrial processing of Q&P steels (IV). The influence of three different quench temperatures (350°C, 360°C
and 380°C) on the microstructure and mechanical properties with two different cooling rates (30K/s and 50K/s) and two different
overaging temperatures was studied (Fig.4a-d). Two quench temperatures were chosen to be under and one quench temperature to be
slightly above the martensite transformation temperature (Ms=371°C) determined from continuous cooling experiments. In general,
higher yield and tensile strength values are obtained at the higher cooling rate which are steadily increasing with decreasing overaging
temperature. The cooling rate determines the tensile strength values which are nearly unaffected from the different overaging
temperatures. In contrast, the yield strength values are significantly reduced at higher overaging temperatures. The overaging
temperature has also a certain influence on the uniform and total elongation which is enhanced for higher overaging temperatures.
Regarding the influence of the cooling rate, higher values for uniform and total elongation are obtained for the lower cooling rate.
In Fig.5, transformation maps illustrating the transformed phase fractions during annealing obtained from dilatometric investigations
with the same annealing parameters used for the annealing simulations described above are shown. The map provides an overview
about the phase transformations which occur during the heat treatment cycle. Before the final quench temperature is reached the
occurring phase transformations are related to ferrite or bainitic ferrite formed during cooling. Phase transformations directly at the
quench temperature correspond to the athermal formation of martensite which starts rapidly if the quenching temperatures are below
the martensite start temperature. Phase transformations during isothermal holding at the overaging temperature are referred to the
formation of bainite. Very small additional dilation of the dilatometric specimen can be further caused by C partitioning from
martensite/bainite to austenite during isothermal holding at the overaging temperature . Furthermore, the amount of measured
retained austenite at room temperature is given in the maps.
Depending on the final quench temperature, up to 8% more phase transformation occurs until the quench temperature is reached at the
slower cooling rate. As a consequence, the amount of transformed phase fractions directly at the quench temperature is significantly
decreased. The cooling rate has also a small influence on the amount of retained austenite which is slightly higher at lower cooling
Fig.4: Influence of the quenching temperature and overaging temperature as well as the cooling rate on the mechanical properties.
Yield and tensile strength for cooling rates of 30K/s and 50K/s for a) an overaging temperature of 400°C and b) 440°C. Uniform and
total elongation for cooling rates of 30K/s and 50K/s for c) an overaging temperature of 400°C and d) 440°C.
Assuming that the transformed phase during cooling is ferrite and/or bainitic ferrite and martensite at the quench temperature, the
lower yield and tensile strength values as well as the higher uniform and total elongation values at the slower cooling rate can be
explained by the lower fraction of “hard” martensitic phase and higher fraction of “soft” ferrite and/or bainitic ferrite phase formed
during cooling in the microstructure. The significantly lower yield strengths obtained at higher overaging temperatures for both
cooling rates are a consequence of the higher fraction of upper bainite which is obtained by transformation at high overaging
temperatures. Higher uniform and total elongation values at the higher overaging temperature are due to enhanced amounts of retained
austenite and due increased amount of upper bainite in the microstructure as well.
Comparing the tensile strength differences between both overaging temperatures at a certain cooling rate, the difference is nearly
constant for all quenching temperatures and the strength can be mainly controlled by choosing a suitable amount of ferrite and/or
bainitic ferrite and martensite. Overaging temperature differences of 40°C between an overaging temperature of 400°C and 440°C are
obviously too small to result in a significant softening effect due to tempered martensite. Although higher elongation values are
obtained for higher overaging temperatures, yield strength values are below 800 MPa. Good combination of yield strength, tensile
strength and uniform elongation for a Q&P steel is given at a cooling rate of 30K/s, the entire range of quench temperatures
investigated and an overaging temperature of 400°C.
Fig.5: Transformation maps obtained from dilatometric investigations for an overaging temperature of 400°C for a cooling rate of
30K/s (a) and 50K/s (b) as well as for an overaging temperature of 440°C for a cooling rate of 30K/s (c) and 50K/s (d). Phase
transformations before reaching the quench temperature (Ferrite and/or bainite formed during cooling), at the quench temperature
(Martensite formed during cooling) and during isothermal holding (Bainite formed during austempering) are shown. The amount of
retained austenite (ret) is also given.
Production on industrial scale
Based on the results from the annealing simulations and dilatometric measurements on industrially cold-rolled material the optimal
annealing parameters were selected to produce TBF and Q&P steel grades. An overaging temperature of 400°C provides an optimal
combination of sufficiently fast transformation kinetics with an optimal amount of bainite formed to obtain the desired mechanical
properties for TBF steels (yield strength>700 MPa, tensile strength>980 MPa, total elongation>14%). The microstructure of the
continuously annealed TBF steel is shown in Fig.6a which consists of bainite and retained austenite. The amount of retained austenite
is ~12%. Typical mechanical properties of the continuously annealed TBF steel which are obtained from longitudinal (L) and
transversal (T) specimens are shown in Fig.6b. The yield strength and the tensile strength are about 762-771 MPa and 1021-1024
MPa, respectively. The resulting yield strength ratio is, therefore, ~0.75. The values of the total elongation are about 15%,
respectively. At strain levels between 4 and 6%, the n-value is 0.16. n-values about 0.14 are measured at strain levels between 6 and
10 %. The excellent properties can be also described by the products of tensile strength times total elongation which is about 15.300
MPa% for the TBF steel grade.
Fig. 6: a) Microstructure and b) obtained mechanical properties of a continuously annealed TBF steel obtained from longitudinal (L)
and transversal (T) specimens.
A quench temperature below 380°C and an overaging temperature of 400°C provides an optimal combination of an optimal amount of
martensite and of bainite formed during the isothermal overaging to obtain the desired mechanical properties for Q&P steels (yield
strength>800 MPa, tensile strength>980 MPa, total elongation>10%). The microstructure of the continuously annealed Q&P steel is
shown in Fig.7a which consists of lower bainite, small fractions of martensite and retained austenite. The amount of retained austenite
is ~9%. Typical mechanical properties obtained from longitudinal (L) and transversal (T) specimens are shown in Fig.7b.
Fig. 7: a) Microstructure and b) obtained mechanical properties of a continuously annealed Q&P steel obtained from longitudinal (L)
and transversal (T) specimens.
The yield strength and the tensile strength are about 870-880 MPa and 1080-1090 MPa, respectively. The resulting yield strength ratio
is, therefore, ~0.81. The values of the total elongation are about 11-12 %. At strain levels between 2 and 4%, the n-value is 0.12. n-
values about 0.13 are measured at strain levels between 4 and 6 %. The excellent properties can be also described by the products of
tensile strength times total elongation which is about 13.000 MPa%.
The influence of typical continuous annealing line processing parameters on the microstructure and mechanical properties of advanced
high strength steel grades was investigated for industrially processed cold-rolled material with a Si content <1wt%. Based on
experimental results optimized annealing cycles were selected to successfully industrially produce TBF steel grades with a minimum
yield strength of 700 MPa (minimum tensile strength 980 MPa) as well as Q&P steel grades with a minimum yield strength of 800
MPa (minimum tensile strength 980 MPa) via continuous annealing line. The total elongations are clearly above 14% and 10% for
TBF steel grades and Q&P steel grades, respectively. Compared to conventional dual-phase and complex phase automotive steel
grades with tensile strengths of 980 MPa, an enhanced ductility is reached. The strength and ductility can be simply varied by
adjusting the amount of “hard” phase i.e. martensite or lower bainite in combination with suitable amounts of “soft” phase i.e. ferrite,
upper bainite and retained austenite due to appropriate annealing conditions.
1. L. Samek and D. Krizan, “Steel – Material of choice for automotive lightweight applications”, Conference Proceeding of
Metal 2013, Brno, Czech Republic, 2013.
2. E. De Moor, P.J. Gibbs, J.G. Speer and D.K. Matlock, “Strategies for third-generation advanced high-strength steel
development”, Iron and Steel Technology, 7/11, 2010, pp.133-144.
3. K. Sugimoto, M. Tsznezawa, T. Hojo and T. Ikeda, “Ductility of 0.1-0.6C-1.5Si-1.5Mn ultra high strength TRIP aided sheet
steel with bainitic ferrite matrix”, ISIJ International, Vol.44, 9, 2004, pp.1608-1614.
4. B.C. De Cooman and J. Speer, “Quench and partitioning steel: New AHSS concept for automotive anti-intrusion
applications”, Steel Research International, Vol. 77, 9-10, 2006, pp.364-640.
5. S. Traint, A. Pichler, R. Tikal, P. Stiaszny and E.A. Werner, “Influence of manganese, silicon and aluminium on the
transformation behavior of low-alloyed TRIP steels”, 42nd Mechanical Working and Steel Processing Conference, Toronto,
6. E. Kozeschnik and H. K. D. H. Bhadeshia, “Influence of Silicon on Cementite Precipitation in Steels”, Materials Science and
Technology, Vol. 24, 2008, pp.343-347.
7. H. K. D. H. Bhadeshia, “The Bainite Transformation: Unresolved Issues”, Materials Science and Engineering A, Vol. A273-
275, 1999, pp.58-66.
8. S.A. Khan and H.K.D.H. Bhadeshia, “Kinetics of Martensitic transformation in partially bainitic 300M steel”, Materials
Science and Engineering A 129, 1990, pp.257-272.
9. J. Mahieu, S. Claessens and B.C. Cooman, “Galvanizability of High-Strength steels for automotive applications”,
Metallurgical and Materials Transactions A, Vol.32A, 2001, pp.2905-2908.
10. K. Hausmann, D. Krizan, K. Spiradek-Hahn, A. Pichler and E. Werner, Steel Research International, in preparation.
11. F.S. LePera, Journal of Metals, 1980, pp. 32-38.
12. E. Wirth, A. Pichler, R. Angerer, P. Stiaszny, K. Hauzenberger, Y.F. Titovets and M. Hackl, “Determination of the volume
amount of retained austenite and ferrite in small specimens by magnetic measurements”, International Conference on TRIP-
Aided High Strength Ferrous Alloys, Ghent, Belgium, June 2002, pp.61-64.
13. H. K. D. H. Bhadeshia, “Bainite: The Incomplete Reaction Phenomenon and the Approach to Equilibrium, Proceedings
Conference Solid Phase Transformations”, Pittsburgh, USA, 1981, pp.1041-1048.
14. P J Jacques, “Transformation-induced plasticity for high strength formable steels”, Current Opinion in Solid State &
Materials Science, 8(3-4), 2004, pp.259–265.
15. J.Wang and S. Van Der Zwaag, “Stabilization mechanisms of retained austenite intransformation-induced plasticity steel”,
Metallurgical and Materials Transactions A-Physical Metallurgy and Materials Science, Vol.32A, 2001, pp.1527–1539.
16. M.J. Santofimia, L. Zhao, R. Petrov, C. Kwakenraak, W.G. Sloof and J. Sietsma, “Microstructural development during the
quenching and partitioning process in newly designed low-carbon steel”, Acta Materialia, Vol.59, 2011, pp.6059-6068.