Quenching and partitioning (Q&P) process: A critical review of the competing reactions

Abstract

The quenching and partitioning (Q&P) process has been proposed recently as an alternative route to obtain a unique microstructure consisting of decarburised martensite and retained austenite that results in a superior balance between strength and ductility. The stabilisation of austenite after the Q&P processing is mainly achieved through its carbon enrichment as a result of carbon partitioning from primary martensite formed during initial quenching. However, carbon partitioning is usually accompanied by complex competing reactions which are known to affect the austenite retention at room temperature. This review is a critical assessment of these competing reactions during the Q&P process and their implication on the evolution of microstructure.

Keywords

Quenching and partitioning microstructure evolution phase transformation carbide precipitation martensite and bainite transformation

Introduction

Steel is a widely used engineering material for structural and functional applications. One of the major applications is in automobiles where a major portion of the vehicle body structures ( %) is made up of steel. Extensive research is being carried out in academia and industry to develop advanced high strength steels (AHSS) in a cost-effective manner that offer high strength in combination with good formability to meet the increasing demands of light-weight automotive vehicles with enhanced passenger safety. Based on the simple composite models, it was argued that multiphase microstructures consisting of martensite along with a significant amount of retained austenite possessing moderate stability exhibit superior combinations of strength and ductility than that observed for the so-called first and second-generation AHSS [1 -3]. While martensite contributes to the overall strength, retained austenite benefits both the strength and ductility through the transformation-induced plasticity (TRIP) effect during external deformation which delays the onset of necking and maintains a high work-hardening rate [4 -6].

Quenching and partitioning (Q&P) is a relatively new processing route to develop multiphase microstructures consisting of martensite and retained austenite [7,8]. The Q&P process, in general, involves three stages: (i) austenitisation followed by quenching to an interrupted quenching temperature (QT) between the (martensite-start) and (martensite-finish) temperatures at which austenite partially transforms to martensite leaving some untransformed austenite (Stage-I), (ii) partitioning step at the QT itself (‘one-step’ Q&P [9]) or at some higher temperature below or above the (‘two-step’ Q&P) where partitioning of carbon from supersaturated martensite to untransformed austenite occurs (Stage-II) [7,10] and (iii) final quenching from partitioning temperature (PT) to room temperature (RT) (Stage-III). When temperature of the carbon-enriched untransformed austenite after partitioning is above RT (i.e. ), it undergoes a partial transformation to fresh martensite during final quenching; otherwise (if ) it is stabilised fully at RT. The fresh martensite obviously has higher carbon than the primary one and its is designated as . A schematic illustration of the Q&P treatment along with the evolution of microstructure at intermediate stages is shown in Figure 1. Since the partitioning treatment during one-step and two-step Q&P processes is performed at a constant temperature (Figure 1), these processes are also referred to as isothermal Q&P processes. Some other variants of the Q&P process such as the non-isothermal Q&P process [2,11] and the Q&P process integrated with the hot stamping process [12 -14] have also been investigated recently.

Figure 1

Schematic illustration of the Q&P process along with schematic microstructure evolution during different intermediate steps [31,46]. and t, austenitisation temperature and time, respectively; QT, quenching temperature; PT, partitioning temperature; , partitioning time; , carbon content of the bulk alloy; and , carbon contents of primary martensite and untransformed austenite at QT, respectively; and , carbon contents of primary martensite and untransformed austenite after the partitioning treatment, respectively; , carbon content of fresh martensite formed during quenching from PT to RT; , carbon content of retained austenite at RT; , austenite formed at or remained untransformed at QT; , primary martensite formed at QT; and , decarburised martensite and carbon-enriched austenite after partitioning, respectively; , fresh martensite and , retained austenite.

The stabilisation of austenite after the Q&P treatment is mainly achieved through its carbon enrichment as a result of carbon partitioning from primary martensite [7,8,15]. This is the basic operating mechanism of austenite stabilisation after Q&P processing which was suggested long ago by Matas and Hehemann [16], however it was explored in detail by Speer and co-workers [7] in the last decade. The carbon partitioning from primary martensite to untransformed austenite has been verified using X-ray [15,17 -20] and in situ neutron diffraction [21,22] measurements, nano-secondary ion mass spectroscopy (SIMS) [23] and atom probe tomography (APT) analysis [24 -26]. Thermodynamic models such as constrained carbon equilibrium (CCE) [8,27 -29] and CCE incorporating carbide precipitation (CCE ) [24] have also been proposed to estimate the carbon content of untransformed austenite after partitioning. One of the key design aspects of the Q&P process is the stabilisation of a higher austenite fraction possessing optimum stability to exhibit a beneficial TRIP effect [3]. After Q&P processing, the resultant amount of retained austenite depends mainly on the QT [30]. The QT that maximises the amount of retained austenite is commonly referred to as the optimum quenching temperature (QT^opt) [8,31]. A simple model to predict the QT^opt and the corresponding amount of retained austenite was proposed by Speer et al. [31].

It has been reported that the competing reactions such as carbide ( and ) formation and bainite transformation, in general, may occur simultaneously during the partitioning treatment [19,25,30,32,33]. These reactions are found detrimental to the resultant austenite fraction and its stability [33,34]. Therefore, control over these reactions is necessary to make the Q&P process effective. Silicon is used extensively in the Q&P steels because of its ability to retard the cementite formation [9,15,35,36]. Even Al and P produce similar retardation effects on cementite formation but to a lower extent [37 -39]. The lower QTs and higher PTs were found beneficial to minimise the extent of bainite transformation during the partitioning treatment [33,40]. In addition to these competing reactions, the formation of carbides within martensite during initial quenching [33,35,41], loss of carbon in the segregation of carbon atoms at dislocations and defects in martensite and at interfaces [2,17,18,33,42 -44], incomplete carbon partitioning from primary martensite during partitioning [45] are also possible which have not been considered in the published literature on the Q&P process.

The Q&P process is still at an early stage of development and some fundamental aspects of phase transformation, microstructure evolution and the microstructure–property relationship are yet to be established. Additionally, the competing reactions as discussed above need to be explored further to understand the mechanism of austenite stabilisation clearly. This review mainly focuses on analysing the possible competing reactions that occur during the Q&P process and their impact on the microstructure evolution.

Q&P concept: Theoretical analysis

Thermodynamics of carbon partitioning

Speer et al. [7], assuming that the competing reactions such as carbide formation and bainite transformation are completely suppressed, defined a metastable condition representing the completion of carbon partitioning between martensite and austenite which was later referred to as ‘Constrained Carbon Equilibrium (CCE)’ [8,27 -29]. Under CCE condition, the ‘end-point’ of carbon partitioning can be given by two conditions which define its unique thermodynamic and mass balance constraints:

Carbon partitioning from supersaturated martensite to untransformed austenite progresses until carbon attains equal chemical potential in ferrite (decarburised martensite) and austenite.

Interface between martensite and austenite ( ) is immobile at the partitioning temperature, thereby, even short-range movement of iron and substitutional atoms is precluded.

To satisfy the thermodynamic condition for CCE, the tangents to the ferrite and austenite free energy curves must intersect the carbon axis at the same point. This condition can be satisfied by an infinite set of phase compositions and two such conditions where the chemical potential of carbon is equal in both the phases are shown in Figure 2. To calculate the applicable phase compositions of ferrite (

) and austenite (

) under the CCE condition after partitioning, the following equations were proposed by Speer et al. [7]: (1)

(2)

(3)

(4)

where

represents the carbon content, in mole fraction, of the parent austenite,

is the partitioning temperature (PT) in Kelvin,

is the universal gas constant,

is the fraction of untransformed austenite at QT and

and

represent the fractions of ferrite and austenite, respectively, under the CCE condition. The values of

and

are calculated for a Fe–0.24C–1.87Mn–1.31Si–0.34Al–0.32Mo–0.66Ni (wt-%) steel [40,45] by solving Equations (1)–(4) and the results are plotted in Figure 3. The CCE calculations show that almost all the carbon partitions from primary martensite to untransformed austenite during partitioning, because of the low solubility of carbon in ferrite.

Figure 2

Schematic illustration of the Gibbs free energy versus composition diagram at temperature showing two possible sets of phase compositions representing the constrained carbon equilibrium (CCE) condition in a Fe–C–M system (M, substitutional element) [7]. , chemical potential of carbon in a phase ‘ ’ under the CCE condition and , carbon concentration (in mole fraction) of a phase ‘ ’ under the CCE condition.

Figure 3

Variation of the carbon content of (a) decarburised martensite and (b) untransformed austenite with partitioning temperature (PT) for different values under the CCE condition evaluated for a Fe–0.24C–1.87Mn–1.31Si–0.34Al–0.32Mo–0.66Ni (wt-%) steel [40].

The experimental carbon content of retained austenite reported in literature (estimated using XRD analysis) and the corresponding calculated values under the CCE condition are summarised in Table 1. Some of the important observations that can be drawn from this data in Table 1 are as follows:

In most of the cases, the experimental carbon concentration of retained austenite is lower than the calculated value under the CCE condition (Table 1). This deviation was attributed to the formation of carbide inside martensite during partitioning [25,46] which is not accounted for in the CCE model. It is also noteworthy that the carbon activity relation between ferrite and austenite used in the CCE calculations (Equation (1)) was proposed for binary Fe–C alloys, but different alloying additions might affect the activity and the distribution of carbon in ferrite and austenite. Toji et al. [24] suggested a modification in this CCE model to account for cementite formation inside martensite assuming carbon equilibria between austenite, ferrite and cementite under a constrained condition. Even then, a deviation between the experimental and predicted carbon content of austenite under the CCE condition was observed by Toji et al. [24]. They suggested that for a more accurate prediction of the carbon content of untransformed austenite after partitioning, change in the free energy of carbide with progress of partitioning should also be considered. However, this deviation further points out that other competing reactions might also be involved.

It can be seen that with an increase in the QT, the deviation between actual and predicted carbon concentration of retained austenite decreases [45,47] or even in some cases, the experimental carbon concentration is higher than the predicted values [47,48] (Table 1). This is because at higher QT, a small amount of primary martensite is formed during initial quenching up to QT which does not cause sufficient carbon enrichment of untransformed austenite during partitioning. This unstable austenite partially transforms to bainite during partitioning and to fresh martensite during final quenching from PT to RT. The formation of bainite also contributes to some degree of carbon enrichment of untransformed austenite during partitioning [42] which is not accounted for in the CCE model.

Table 1

A comparison of the experimental carbon content of decarburised martensite ( ) and retained austenite ( ) reported in the literature (estimated by XRD analysis) with the calculated values under the CCE condition.

Composition (wt-%)		Q&P treatment				XRD (wt-%)		CCE (wt-%)
C	Other elements	QT (°C)	PT (°C)	(s)						Ref.
0.30	3.50Mn, 1.60Si	180	400	200	0.21	–	0.96	0.0008	1.41	[33]
0.59	2.9Mn, 2.0Si	–	400	1000	–	–	0.91	–	5.05	[25]
0.25*	1.59Mn,1.63Si, 0.04Al	200	400	10	0.11	–	1.09	0.002	2.21	[34]
				30		–	1.22
				100		–	1.28
				1000		–	1.18
0.38	1.54Mn, 1.48Si	225	400	300	0.26	–	1.42	0.0008	1.44	[35]
0.38	1.54Mn, 1.48Si	250	450	10	0.32	–	1.04	0.001	1.18	[49]
				30			1.19
				60			1.07
				300			0.79
0.20	1.54Mn, 1.30Si, 1.48Cr, 0.07Ni	210	400	300	0.07	–	0.86	0.003	2.75	[47]
		245			0.15	–	0.98	0.0007	1.32
		282	450		0.31	–	0.85	0.0005	0.64
0.19	1.52Mn, 1.32Si, 0.01Cr, 1.53Ni	267	450	300	0.16	–	1.03	0.0012	1.17
		300	400		0.33	–	1.12	0.0002	0.57
		317			0.51	–	1.15	0.0001	0.37
0.24	1.87Mn, 1.31Si, 0.34Al, 0.32Mo, 0.66Ni	230	400	300	0.10	0.04	0.98	0.002	2.33	[45]
		240			0.12	0.04	0.97	0.0015	1.96
		260			0.19	0.03	0.94	0.0007	1.25
		280			0.26	0.02	0.93	0.0004	0.92
0.32	1.78Mn, 0.64Si, 1.75Al, 1.20Co	210	375	60	0.41	–	1.38	0.0002	0.77	[48]
0.31	1.91Si, 0.24Mo, 13.95Ni	23	400	30	–	0.24	0.74	–	–	[26]
				60	–	0.26	0.67	–	–
				100	–	0.23	0.70	–	–
				1000	–	0.23	0.60	–	–

is the amount of untransformed austenite at QT, and indicate the carbon contents of decarburised martensite and untransformed austenite at partitioning temperature (PT), respectively, calculated under the CCE condition after solving Equations (1)–(4) and is the partitioning time.

*The original carbon content of the steel used in this work was 0.19 wt-%. This value of 0.25 wt-% C is taken here since intercritical annealing was performed at 820°C for 180 s to form 25% intercritical ferrite resulting in ∼0.25 wt-% C in the intercritical austenite [34].

Kinetics of the carbon partitioning

The models discussed above do not incorporate carbon partitioning kinetics, they rather assume complete partitioning under ideal conditions as dictated by thermodynamics alone. However, in reality, the carbon content of retained austenite after the Q&P treatment was found to depend on the progress of partitioning treatment [26,34,49] (Table 1). Therefore, to control the resultant microstructure after the Q&P processing, the timescale over which carbon rejection from primary martensite and its homogenisation within untransformed austenite take place should also be considered. Rizzo et al. [50] simulated the kinetics of carbon partitioning in a Fe–0.19C–1.59Mn–1.63Si (wt-%) steel at 400°C using DICTRA software assuming idealised partitioning conditions such as complete suppression of the competing reactions and immobile martensite/austenite ( ) interface. It has been observed by them that the martensite rejects its carbon in less than 0.1 s while it takes ∼10 s to homogenise inside untransformed austenite at 400°C. The DICTRA software has also been employed to simulate the kinetics of carbon partitioning in some other studies [51,52]. However, the carbon content of decarburised martensite in a Fe–0.19C–1.59Mn–1.63Si–0.04Al (wt-%) steel sample quenched to the QT of 220°C and partitioned at 400°C for 10 s is measured to be around 0.04 and 0.10 wt-% using 3D APT analysis [34]. HajyAkbary et al. [53] have also reported a value of 0.05 wt-% for the carbon concentration of decarburised martensite in a sample quenched to 180°C and partitioned at 400°C for 200 s in a Fe–0.3C–3.5Mn–1.6Si (wt-%) steel, using 3D APT analysis. Results such as these point out that during actual partitioning, the carbon depletion may be slower than that predicted using DICTRA. It should be recognised here that during DICTRA simulation, idealised partitioning conditions were assumed. Additionally, in DICTRA simulations, martensite is usually treated as supersaturated ferrite without incorporating the effect of tetragonal lattice distortion, carbon trapping at dislocations and/or defects and carbide formation inside martensite [54].

HajyAkbary et al. [33] used a numerical model proposed by Santofimia et al. [55] to simulate the kinetics of carbon partitioning at 400°C in a Fe–0.3C–3.5Mn–1.6Si (wt-%) steel. It was observed that completion of the carbon partitioning took longer times than that predicted by the numerical simulation [33]. It was suggested that decomposition of the -carbides, which were formed during initial quenching, is essential for completion of carbon partitioning from martensite to austenite and the slow kinetics of decomposition of -carbides retards the carbon partitioning process [33]. Additionally, the models developed so far use published values of carbon diffusivity in ferrite to represent the mobility in martensite which does not account for the influence of the martensite dislocation substructure on the carbon mobility [2]. A detailed kinetic analysis of the Q&P process, after incorporating the issues discussed above, is still due.

Optimum quenching temperature (QT^opt)

As stated above, Speer's model [31] is generally employed to estimate the optimum quenching temperature (QT^opt), for a given alloy. This model presumes that nearly all the carbon is partitioned from primary martensite to untransformed austenite, as was observed by the CCE calculation (Figure 3) and the competing reactions are suppressed during partitioning. In this model, the extent of martensite transformation during initial quenching to the QT (below ) and final quenching from PT to RT is described using Koistinen and Marburger (KM) equation [56]: (5) where is the amount of martensite formed upon quenching to which is below the temperature; during initial quenching, = QT. The temperature of the alloy can be estimated using published empirical relations [57 -61]. The carbon content of austenite remaining untransformed after the partitioning treatment is then calculated using carbon mass balance. The amount of fresh martensite (if any) can also be calculated using Equation (5) after substituting = RT and the with , which is the martensite-start temperature of carbon-enriched austenite.

Figure 4(a) shows a typical example of the QT^opt calculations using Speer's methodology [8,31]. The results showed that when initial quenching after austenitisation is done up to QT^opt, a substantial amount of retained austenite can be stabilised at RT after the Q&P processing. A comparison between the calculated, using Speer's model, and the actual amount of retained austenite estimated using XRD analysis as a function of QT is shown in Figure 4(b) [62]. The variation of the actual amount of retained austenite with QT is in qualitative agreement with trends predicted using Speer's model. However, deviations between the experimental and calculated QT^opt and the corresponding retained austenite fraction are observed in this case; the measured amount of retained austenite is lower than the predicted value (Figure 4(b)). Similar deviations between predicted and experimentally measured results have also been reported in literature [8,15,34,48] which are attributed mainly to the violation of the ideal partitioning conditions such as complete carbon partitioning, suppression of the bainite formation and carbide precipitation during partitioning [34,63]. In addition, segregation of some carbon atoms might take place at dislocations in martensite and at lath boundaries [18,45] and these carbon atoms no longer remain available to enrich the untransformed austenite during partitioning. The application of the KM equation [56] which is commonly used to predict the kinetics of martensite transformation in this model, also needs to be reanalysed since it is applicable to a specific range of chemical composition.

Figure 4

(a) Prediction of the optimum quenching temperature (QT^opt) for a Fe–0.19C–1.96Al–1.46Mn–0.02Si (wt-%) steel containing 50% intercritical ferrite [8,31] and (b) comparison between the QT^opt and the corresponding retained austenite fraction predicted using Speer's methodology and the experimental values obtained using XRD analysis [62]. The symbols used here have already been defined earlier in this discussion.

The observed discrepancies between the experimental (literature data) and calculated (by CCE and Speer's models) amount of retained austenite and its carbon content after the Q&P processing have been attributed mainly to the carbide precipitation inside martensite and bainite formation during partitioning. In some studies on the Q&P process, the amount of retained austenite and its carbon content was found to decrease when the duration of the partitioning treatment was relatively long [26,34,49] (Table 1). This variation cannot be explained simply by the formation of carbide precipitates and bainite transformation since these reactions are expected to take place over a shorter time duration. As pointed out earlier, some additional competing reactions might also take place during quenching and partitioning.

Competing reactions: A comprehensive analysis

A brief discussion regarding some of the experimental observations on competing reactions is presented in this section.

Segregation of carbon atoms at dislocations and defects

It is expected that some carbon atoms remain trapped at dislocations and defects in martensite and at lath boundaries during partitioning [2,17,18,33,42 -44]. The segregation of carbon atoms to lattice defects such as lath boundaries and dislocations during quenching or aging of martensite has been reported extensively in the tempering literature [64 -71]. Atomic-scale evidences of the segregation of carbon atoms at dislocations inside martensite [25] and carbon atom clustering [26] in the full quenched samples have also been provided earlier. Therefore, at the QT or just at the beginning of partitioning, the carbon concentration of primary martensite ( ) may be less than the bulk carbon concentration of the alloy ( ), i.e. < , unlike in the theoretical analysis where it was assumed that (Figure 1).

However, during the partitioning treatment, recovery of dislocation structure is known to take place which results in a decrease in dislocation density [2,18]. Therefore, the carbon atoms trapped at these dislocations in martensite are expected to release. This carbon can lead to carbide precipitation or enrichment of the untransformed austenite [72]. However, owing to the fact that the addition of Si delays the cementite formation [73,74], the enrichment of untransformed austenite is more likely in the presence of Si.

Incomplete carbon partitioning from primary martensite

The carbon content of decarburised martensite away from the carbon-enriched regions was measured to be 0.063 ± 0.005 wt-% using 3D APT analysis of a Fe–0.24C–1.87Mn–1.31Si–0.34Al–0.32Mo–0.66Ni (wt-%) steel quenched to 260°C and partitioned at 400°C for 300 s [45], which is quite high than the corresponding CCE value (Table 1). HajyAkbary et al. [33] have estimated the carbon concentration of primary martensite to be 0.05 wt-% after partitioning at 400°C for 200 s using 3D APT analysis. These results indicate that complete decarburisation of primary martensite does not occur even during a longer duration partitioning treatment. However, the reasons for this behaviour are not understood well and require further analysis.

Carbide formation

The formation of carbides ( and ) inside martensite laths has been reported in several studies on the Q&P processing [5,15,33 -35,49], even in alloys containing relatively high Si [12,15]. It has been reported that these carbides usually form during partitioning [12,15,24,25,75]. It is believed by some researchers that carbides incorporate Si as a solute, thereby, Si addition is found ineffective to suppress their formation [19]. Recent work has however shown that -carbide has a kinetic advantage and therefore it is not suppressed even in the presence of Si [76].

The formation of carbides within martensite has also been observed during initial quenching before partitioning [33,35,41,49]. This process, which is commonly known as autotempering [77], depends on the temperature, the carbon content of the alloy and cooling rate after austenitisation [41,78]; higher temperatures and lower cooling rates provide favourable conditions for autotempering [5,79]. Pierce et al. [35,49], using Mössbauer spectroscopy and TEM analysis, performed quantitative analysis of carbides formed during water quenching and Q&P processing of a Fe–0.38C–1.54Mn–1.48Si (wt-%) steel. The amount of -carbides was estimated to be ∼1.0 at.-% in the water quenched sample, which consumed around 26% of total carbon in the bulk alloy [35,49]. However, Kalish and Cohen [80] have argued that in the presence of dislocations, carbon atoms are more likely to segregate at dislocations inside martensite rather than precipitating as carbides.

It has been argued by Edmonds et al. [15] that the carbides formed during initial quenching or at lower QT might re-dissolve during higher temperature partitioning, providing a source of carbon either for cementite formation or for further enrichment of untransformed austenite. Based on the SEM and TEM investigation of a Fe–0.3C–3.5Mn–1.6Si (wt-%) steel, HajyAkbary et al. [33] qualitatively observed that the apparent number density of -carbides decreased when partitioning time ( ) is increased from 5 to 200 s at 400°C which was attributed to the dissolution of carbides during partitioning. Based on the thermodynamic calculations, HajyAkbary et al. [33] have pointed out that -carbide is unstable at 400°C and therefore re-dissolves during partitioning.

In contrast to the observations made by HajyAkbary et al. [33], Pierce et al. [35] observed an increase in the amount of η-carbides from 1.4 to 2.4 at.-% when the was increased from 10 to 300 s at the PT of 400°C. At the same time, they argued that dissolution of some -carbides during partitioning treatment cannot be ignored completely and argued that an increase in the amount of -carbides with increasing at 400°C is possible because of the change in the dynamics between carbon partitioning from martensite to austenite and -carbide precipitation [35]. If it is assumed that the carbon rejected as a result of the dissolution of carbides enriches the untransformed austenite, then the carbon concentration of retained austenite is expected to increase with progress of partitioning which was validated experimentally by HajyAkbary et al. [33]. In contrast, the carbon content of retained austenite has also been reported to decrease for a longer duration partitioning treatment [26,49,81] which cannot be explained by the dissolution of carbides.

It should be recognised that if the partitioning treatment is performed for a long time, untransformed austenite might also decompose to ferrite and cementite in order to achieve equilibrium microstructure at the partitioning temperature. However, Pierce et al. [35] argued that carbide precipitation inside austenite is thermodynamically unfavourable and was not observed by them. Bigg et al. [21,22], based on the in situ neutron diffraction study, have observed a decrease in the austenite fraction and its carbon content for longer duration partitioning treatment, similar to that observed by Clarke et al. [34]. This data reported by Bigg et al. [21,22] was further analysed by Speer et al. [2] and the observed decrease in the retained austenite fraction and carbon content was attributed to the cementite formation [2]. Based on the TEM and Mössbauer spectroscopic analysis of the Q&P treated Fe–0.38C–1.54Mn–1.48Si (wt-%) steel at 400 and 450°C, Pierce et al. [49] provided experimental evidence of the decomposition of austenite to ferrite and cementite during the partitioning treatment at 450°C. The decrease in the amount of austenite and its carbon content beyond the partitioning time of 30 s at 400 and 450°C was attributed to the increase in the amount of cementite [49].

It is also important to note that the conversion of some carbides to cementite is also possible for a longer duration partitioning treatment, much like the third stage of tempering [71]. Toji et al., after APT analysis of a Fe–0.59C–2.9Mn–2.0Si (wt-%) alloy partitioned at 400°C for 300 s, observed plate-shaped (thickness ∼5 nm) carbon-enriched features with a carbon content of ∼25 at.-%; the stoichiometric carbon content of cementite [25]. Furthermore, as shown in Figure 5, Seo et al. [17] also observed a carbon-enriched region of around 25 nm diameter in the vicinity of retained austenite. The carbon content of this region was estimated to be 23 at.-% (6 wt.-%). It can also be seen that this carbon-enriched region is depleted with Si (Figure 5). Toji et al. [25] argued that this slight Si depletion perhaps indicates that carbides that were formed during the initial stage of partitioning are in the transition stage from metastable carbide to stoichiometric cementite.

Figure 5

Reconstruction of the APT data of a Fe–0.2C–4.0Mn–1.6Si–1.0Cr (wt-%) steel quenched to 170°C and partitioned at 450°C for 300 s: (a) carbon atom map and (b) 1D concentration profile along the solid arrow indicated in (a) [17]. Produced with permission of Elsevier.

Bainite transformation

In some recent studies, experimental evidences of the bainite transformation during partitioning are provided [12,33,44,82,83]; even when the partitioning treatment is performed below the temperature [48,84 -86]. The dilatation curve of a Fe–0.24C–1.87Mn–1.31Si–0.34Al–0.32Mo–0.66Ni (wt-%) steel observed during partitioning at 400°C, indicates a small volume expansion (Figure 6) [40]. This small volume expansion is usually ascribed to bainite transformation [41,45,87 -90] because martensite in this class of steels forms athermally. However, a contraction in this dilatation can also be seen towards the end of the partitioning treatment. It is well-known from the tempering literature that conversion of carbides into cementite leads to a volume contraction [71]. Therefore, it is believed that this contraction is because of the conversion of some previously formed carbides into cementite. However, direct experimental evidence for this has not been provided yet.

Figure 6

Dilatation curve observed during the partitioning treatment at 400°C for 600 s, when initial quenching was done at 280°C [40].

Although bainite transformation during partitioning is expected to reduce the resultant amount of retained austenite after the Q&P process, it has been invoked as an alternative mechanism of the carbon enrichment of untransformed austenite during partitioning [42]. The carbon content of bainitic-ferrite formed during partitioning is usually assumed to be very small [91 -93]. However, some latest studies indicate that it depends on the carbon content of the parent austenite and transformation temperature with respect to the Zener ordering temperature [93 -96]. The bainitic-ferrite is expected to be of low carbon (∼0.03 wt-%) [93,97,98] if the transformation temperature is higher than the Zener ordering temperature and the carbon content of parent austenite is low, otherwise, the carbon content of bainitic-ferrite could be much higher than expected from paraequilibrium between austenite and ferrite [99]. It is obvious that the carbon content of austenite remaining untransformed at QT from which bainite forms during partitioning changes rapidly due to decarburisation of martensite. Therefore, the carbon content of bainitic-ferrite also needs to be examined carefully with reference to the recent results [96,100]. This aspect is not explored yet concerning the Q&P process.

A summary of the microstructural constituents formed during the Q&P process, after considering all the possible competing reactions, is given in Table 2. All the possible phase transformations/competing reactions during quenching and partitioning stages are summarised in Table 3. In recent work, Speer's model was revised after incorporating some of these competing reactions such as incomplete carbon partitioning from martensite [45], bainite formation [33,40], carbide precipitation [19,32,49,101] and segregation of carbon atoms at dislocations and defects [18,26]. The revised model showed a better match between the experimental and calculated QT^opt and the corresponding retained austenite fraction (Figure 7).

Figure 7

Comparison of the experimental QT^opt and the corresponding retained austenite fraction ( ) with the values calculated using the revised model with = 0, 0.03 and 0.06 wt-% [45]. , carbon content of decarburised martensite after the partitioning treatment.

Table 2

The possible microstructural constituents at intermediate stages of the Q&P process, their composition, phase fraction and interrelation between them after incorporating the competing reactions.

Stage of thermal treatment	Microstructure	Composition	Amount	Remarks
Austenitisation, T_A>			1	–
Stage-I: QT <	+ +carbide	, ,	,	, ,
Stage-II: PT, = 0	+ +carbide	, ,	,	, ,
Stage-II: PT, =	+ + +carbide+	, ,	, ,	, , +
Stage-III: RT	+ + + +carbide+	, , ,	, , ,	If , If ,

and represent the amounts of primary martensite formed up to QT during initial quench and austenite remaining untransformed at QT, respectively, and , , and denote the amounts of bainitic-ferrite formed during the partitioning treatment, untransformed austenite at the end of partitioning, fresh martensite formed during final quenching after partitioning and retained austenite at RT, respectively, is the carbon content of bainitic-ferrite, is the amount of carbon trapped at dislocations inside martensite and and represent the bainitic-ferrite and cementite, respectively.

It is important to note here that the formation of carbide precipitates will affect the amount of other constituent phases and their interrelationship. However, owing to the fact that only a small amount of carbide precipitates usually forms during the Q&P treatment, this aspect is ignored in this table.

Table 3

A summary of the possible phase transformations/competing reactions that are possible during the quenching and partitioning stages.

Q&P stages	Phase transformations/ competing reactions	Remarks
Stage-I: Initial quenching up to QT (QT < )	Formation of primary martensite leaving some untransformed austenite, increase in dislocation density Segregation of carbon atoms at dislocations and lath boundaries Carbide precipitation inside martensite (autotempering)	A higher and lower cooling rate provide favourable conditions for autotempering [5,79] In the presence of dislocations, segregation of carbon atoms at dislocations may be favoured over carbide precipitation [80,102]
Stage-II: Partitioning at PT (PT = QT or PT > QT above or below )	Carbon rejection from primary martensite to untransformed austenite Carbide precipitation and/or dissolution Bainite transformation Recovery of dislocation in martensite and therefore C atoms trapped at these dislocations may be available for partitioning or precipitation. Decomposition of untransformed austenite to ferrite and cementite Conversion of previously formed carbides to cementite	The amount of carbon in solid solution in martensite has been observed to be higher than the value calculated under CCE condition even after prolonged partitioning treatment [33,34,45]. This signifies incomplete carbon partitioning from primary martensite The formation of cementite inside austenite becomes thermodynamically feasible when its carbon content exceeds the extrapolated paraequilibrium line

Conclusion

The occurrence of the competing reactions during quenching and partitioning has been critically reviewed based on the available literature on the Q&P process and tempering of martensite. A comparison between the experimental (literature) and calculated (by CCE and Speer's model) amount of retained austenite and its carbon content was made. Some of the important observations that can be drawn from this review are as follows:

A substantial discrepancy between the actual and calculated carbon partitioning behaviour is observed which has been attributed mainly to the competing reactions such as carbide precipitation and bainite formation during partitioning. Some additional competing reactions like carbide precipitation even during initial quenching below the martensite-start temperature, segregation of carbon atoms at dislocations and defects inside martensite, incomplete carbon partitioning from primary martensite are also possible during the Q&P treatment.

The formation of carbide precipitates, segregation of carbon atoms at defects or interfaces and incomplete carbon partitioning reduce the carbon available for partitioning. The formation of bainite, on the other hand, results in some degree of carbon enrichment of untransformed austenite but it also consumes a part of this untransformed austenite which otherwise could have been stabilised to RT.

With progress of the partitioning treatment, the extent of decarburisation of primary martensite increases. In addition, dissolution of previously formed carbide precipitates and recovery of dislocation in martensite is also possible. These processes result in the release of carbon atoms from primary martensite which may lead to either carbide precipitation or enrichment of the untransformed austenite. However, the enrichment of untransformed austenite is more likely in the presence of Si which is known to delay the cementite formation.

The formation of cementite inside austenite becomes thermodynamically feasible when the carbon concentration of untransformed austenite exceeds the extrapolated paraequilibrium line and may in fact take place when the duration of the partitioning treatment is relatively long. Transformation of some of the previously formed transition carbides to cementite is also possible when the partitioning treatment is performed for a longer time. The formation of cementite leads to a significant decrease in the carbon content of retained austenite.

Footnotes

Acknowledgements

The author is grateful to Professor Shiv Brat Singh for his encouragement, suggestions and critical review of the work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

References

Matlock

Speer

De Moor

Recent developments in advanced high strength sheet steels for automotive applications: an overview. JESTECH. 2012; 15:1–12.

Speer

De Moor

Findley

Analysis of microstructure evolution in quenching and partitioning automotive sheet steel. Metall Mater Trans A. 2011; 42:3591–3601.

Matlock

Speer

JG.

Third generation of AHSS: microstructure design concepts. In: Halder

Suwas

Bhattacharjee

, editors. Microstructure and texture in steels and other materials. London: Springer; 2008. p. 185–205.

Kwon

Lee

Kim

New trends in advanced high strength steel developments for automotive application. Mater Sci Forum. 2010; 638–642:136–141.

Yan

Liu

Comparison on mechanical properties and microstructure of a C-Mn-Si steel treated by quenching and partitioning (Q&P) and quenching and tempering (Q&T) processes. Mater Sci Eng A. 2015; 620:58–66.

Gui

Gao

Guo

Effect of bainitic transformation during BQ&P process on the mechanical properties in an ultrahigh strength Mn-Si-Cr-C steel. Mater Sci Eng A. 2017; 684:598–605.

Speer

Matlock

De Cooman

Carbon partitioning into austenite after martensite transformation. Acta Mater. 2003; 51:2611–2622.

Speer

Edmonds

Rizzo

Partitioning of carbon from supersaturated plates of ferrite, with application to steel processing and fundamentals of the bainite transformation. Curr Opin Solid State Mater Sci. 2004; 8:219–237.

Streicher

Speer

Matlock

Quenching and partitioning response of a Si-added TRIP sheet steel. In: Speer

, editor. Proc Int Conf Adv high-strength sheet steels automot appl. Warrendale (PA): AIST; 2004. p. 51–62.

10.

Matlock

Brautigam

Speer

JG.

Application of the quenching and partitioning (Q&P) process to a medium-carbon, high-Si microalloyed bar steel. Mater Sci Forum. 2003; 426–432:1089–1094.

11.

Thomas

Speer

Matlock

DK.

Quenched and partitioned microstructures produced via Gleeble simulations of hot-strip mill cooling practices. Metall Mater Trans A. 2011; 42:3652–3659.

12.

Seo

Cho

De Cooman

BC.

Application of quenching and partitioning (Q&P) processing to press hardening steel. Metall Mater Trans A. 2014; 45A:4022–4037.

13.

Liu

Jin

Dong

Martensitic microstructural transformations from the hot stamping, quenching and partitioning process. Mater Charact. 2011; 62:223–227.

14.

Liu

Jin

Enhanced mechanical properties of a hot stamped advanced high-strength steel treated by quenching and partitioning process. Scr Mater. 2011; 64:749–752.

15.

Edmonds

Rizzo

Quenching and partitioning martensite - a novel steel heat treatment. Mater Sci Eng A. 2006; 438–440:25–34.

16.

Matas

Hehemann

RF.

Retained austenite and the tempering of martensite. Nature. 1960; 187:685–686.

17.

Seo

Cho

Estrin

Microstructure-mechanical properties relationships for quenching and partitioning (Q&P) processed steel. Acta Mater. 2016; 113:124–139.

18.

Microstructure and mechanical properties of an ultrahigh-strength 40SiMnNiCr steel during the one-step quenching and partitioning process. Metall Mater Trans A. 2010; 41A:1284–1300.

19.

De Moor

Lacroix

Clarke

Effect of retained austenite stabilized via quench and partitioning on the strain hardening of martensitic steels. Metall Mater Trans A. 2008; 39:2586–2595.

20.

Speer

Rizzo

Matlock

Proceedings of 59th Annual Congress of ABM; Sao Paulo; 2004. p. 4824–4836.

21.

Bigg

Edmonds

Eardley

ES.

Real-time structural analysis of quenching and partitioning (Q&P) in an experimental martensitic steel. J Alloys Compd. 2013; 577S:S695–S698.

22.

Bigg

Matlock

Speer

Dynamics of the quenching and partitioning (Q&P) process. Solid State Phenom. 2011; 172–174:827–832.

23.

Choi

Zhu

Sun

Determination of carbon distributions in quenched and partitioned microstructures using nanoscale secondary ion mass spectroscopy. Scr Mater. 2015; 104:79–82.

24.

Toji

Miyamoto

Raabe

Carbon partitioning during quenching and partitioning heat treatment accompanied by carbide precipitation. Acta Mater. 2015; 86:137–147.

25.

Toji

Matsuda

Herbig

Atomic-scale analysis of carbon partitioning between martensite and austenite by atom probe tomography and correlative transmission electron microscopy. Acta Mater. 2014; 65:215–228.

26.

Thomas

Danoix

Speer

Carbon atom re-distribution during quenching and partitioning. ISIJ Int. 2014; 54:2900–2906.

27.

Hillert

Årgen

On the definitions of paraequilibrium and orthoequilibrium. Scr Mater. 2004; 50:697–699.

28.

Speer

Matlock

De Cooman

Comments on “On the definitions of paraequilibrium and orthoequilibrium” by M. Hillert and J. Ågren, Scripta Materialia, 50, 697-9 (2004). Scr Mater. 2005; 52:83–85.

29.

Hillert

Årgen

Reply to comments on “On the definition of paraequilibrium and orthoequilibrium”. Scr Mater. 2005; 52:87–88.

30.

De Moor

Speer

JG.

Bainitic and quenching and partitioning steels. In: Rana

Singh

, editors. Automot steels des metall process appl. Duxford, UK: Woodhead Publishing; 2017. p. 289–316.

31.

Speer

Matlock

Streicher

Quenching and partitioning: a fundamentally new process to create high strength TRIP sheet microstructures. In: Damm

Merwin

, editors. Austenite form decompos. Warrendale (PA): TMS/ISS; 2003. p. 505–522.

32.

Speer

JG.

Phase transformations in quenched and partitioned steels. In: Pereloma

Edmonds

, editors. Phase transform steels diffus transform high strength steels, model adv anal tech. Vol. 2. Cambridge, UK: Woodhead Publishing Limited; 2012. p. 247–270.

33.

HajyAkbary

Sietsma

Miyamoto

Interaction of carbon partitioning, carbide precipitation and bainite formation during the Q&P process in a low C steel. Acta Mater. 2016; 104:72–83.

34.

Clarke

Speer

Miller

Carbon partitioning to austenite from martensite or bainite during the quench and partition (Q&P) process: a critical assessment. Acta Mater. 2008; 56:16–22.

35.

Pierce

Coughlin

Williamson

Characterization of transition carbides in quench and partitioned steel microstructures by Mössbauer spectroscopy and complementary techniques. Acta Mater. 2015; 90:417–430.

36.

Thomas

De Moor

Speer

In: De Moor

Jun

Speer

, editors. Tensile properties obtained by Q&P processing of Mn-Ni steels with room temperature quench temperatures, Proceedings of the International Symposium on New Developments in Advanced High-strength Sheet Steels. Warrendale (PA): AIST; 2013. p. 153–165.

37.

Gallagher

Speer

Matlock

. Microstructure development in TRIP-sheet steels containing Si, Al, and P. In: Proc 44th Mech Work Steel Process Conf; Warrendale (PA): Iron and Steel Society; 2002. p. 153–172.

38.

Mahieu

Claessens

De Cooman

BC.

Galvanizability of high-strength steels for automotive applications. Metall Mater Trans A. 2001; 32:2905–2908.

39.

Grajcar

Kuziak

Zalecki

Third generation of AHSS with increased fraction of retained austenite for the automotive industry. Arch Civ Mech Eng. 2012; 12:334–341.

40.

Kumar

Singh

SB.

, Understanding of the effect of quenching and partitioning (Q&P) parameters on the stabilization of austenite, Department of Metallurgical and Materials Engineering, Indian Institute of Technology Kharagpur, India, Unpublished research. 2021.

41.

Clarke

Miller

Field

Atomic and nanoscale chemical and structural changes in quenched and tempered 4340 steel. Acta Mater. 2014; 77:17–27.

42.

Speer

De Moor

Clarke

AJ.

Critical assessment 7: quenching and partitioning. Mater Sci Technol. 2015; 31:3–9.

43.

Seo

Cho

De Cooman

BC.

Kinetics of the partitioning of carbon and substitutional alloying elements during quenching and partitioning (Q&P) processing of medium Mn steel. Acta Mater. 2016; 107:354–365.

44.

Seo

Cho

De Cooman

BC.

Modified methodology for the quench temperature selection in quenching and partitioning (Q&P) processing of steels. Metall Mater Trans A. 2016; 47A:3797–3802.

45.

Kumar

Singh

SB.

Evolution of microstructure during the “quenching and partitioning (Q&P)” treatment. Materialia. 2021; 18:101135.

46.

Speer

Rizzo Assunção

Matlock

The “quenching and partitioning” process: background and recent progress. Mater Res. 2005; 8:417–423.

47.

Pierce

Coughlin

Clarke

Microstructural evolution during quenching and partitioning of 0.2C-1.5Mn-1.3Si steels with Cr or Ni additions. Acta Mater. 2018; 151:454–469.

48.

Samanta

Das

Chakrabarti

Development of multiphase microstructure with bainite, martensite, and retained austenite in a Co-containing steel through quenching and partitioning (Q&P) treatment. Metall Mater Trans A. 2013; 44:5653–5664.

49.

Pierce

Coughlin

Williamson

Quantitative investigation into the influence of temperature on carbide and austenite evolution during partitioning of a quenched and partitioned steel. Scr Mater. 2016; 121:5–9.

50.

Rizzo

Edmonds

. Carbon enrichment of austenite and carbide precipitation during the quenching and partitioning (Q&P) process.

Phoenix, Arizona, USA: TMS,

International conference on solid-solid phase transformations in inorganic materials, 2005. p. 535-544.

51.

Linke

Gerber

Hatscher

Impact of Si on microstructure and mechanical properties of 22MnB5 hot stamping steel treated by quenching & partitioning (Q&P). Metall Mater Trans A. 2018; 49:54–65.

52.

Peng

The relationships of microstructure-mechanical properties in quenching and partitioning (Q&P) steel accompanied with microalloyed carbide precipitation. Mater Sci Eng A. 2018; 723:247–258.

53.

HajyAkbary

Sietsma

Miyamoto

Analysis of the mechanical behavior of a 0.3C-1.6Si-3.5Mn (wt%) quenching and partitioning steel. Mater Sci Eng A. 2016; 677:505–514.

54.

Clarke

Speer

Matlock

Influence of carbon partitioning kinetics on final austenite fraction during quenching and partitioning. Scr Mater. 2009; 61:149–152.

55.

Santofimia

Zhao

Sietsma

Model for the interaction between interface migration and carbon diffusion during annealing of martensite-austenite microstructures in steels. Scr Mater. 2008; 59:159–162.

56.

Koistinen

Marburger

RE.

A general equation prescribing the extent of the austenite-martensite transformation in pure iron-carbon alloys and plain carbon steels. Acta Metall. 1959; 7:59–60.

57.

Steven

Haynes

AG.

The temperature of formation of martensite and bainite in low-alloy steels. J Iron Steel Inst. 1956; 183:349–359.

58.

Andrews

KW.

Empirical formulae for the calculation of some transformation temperatures. J Iron Steel Inst. 1965; 203:721–727.

59.

Capdevila

Caballero

De Andrés

CG.

Determination of martensite-start temperature in steels: Bayesian neural network model. ISIJ Int. 2002; 42:894–902.

60.

van Bohemen

SMC.

Bainite and martensite start temperature calculated with exponential carbon dependence. Mater Sci Technol. 2012; 28:487–495.

61.

Lee

Park

KS.

Prediction of martensite start temperature in alloy steels with different grain sizes. Metall Mater Trans A. 2013; 44:3423–3427.

62.

Gerdemann

Microstructure and hardness of 9260 steel heat treated by the quenching and partitioning process [dipl. thesis]. Aachen University of Technology (RWTH); 2004.

63.

Santofimia

Speer

Clarke

Influence of interface mobility on the evolution of austenite–martensite grain assemblies during annealing. Acta Mater. 2009; 57:4548–4557.

64.

Jung

Lee

Microstructural and dilatational changes during tempering and tempering kinetics in martensitic medium-carbon steel. Metall Mater Trans A. 2009; 40A:551–559.

65.

Uwakweh

ONC

Génin

JMR

Silvain

J-F.

Electron microscopy study of the aging and first stage of tempering of high-carbon Fe-C martensite. Metall Trans A. 1991; 22:797–806.

66.

Cheng

van der Pers

Böttger

Lattice changes of iron-carbon martensite on aging at room temperature. Metall Trans A. 1991; 22:1957–1967.

67.

Sherman

Eldis

Cohen

The aging and tempering of iron-nickel-carbon martensites. Metall Trans A. 1983; 14A:995–1005.

68.

Nagakura

Hirotsu

Kusunoki

Crystallographic study of the tempering of martensitic carbon steel by electron microscopy and diffraction. Metall Trans A. 1983; 14:1025–1031.

69.

Speich

GR.

Tempering of low-carbon martensite. Trans TMS-AIME. 1969; 245:2553–2564.

70.

van Rooyen

Mittemeijer

EJ.

Differential thermal analysis of iron-carbide martensites. Scr Metall. 1982; 16:1255–1260.

71.

Cheng

Brakman

Korevaar

The tempering of iron-carbon martensite; dilatometric and calorimetric analysis. Metall Trans A. 1988; 19A:2415–2426.

72.

Edmonds

Miller

Microstructural features of ‘quenchnig and partitioning’: a new martensitic steel heat treatment. Mater Sci Forum. 2008; 539–543:4819–4825.

73.

Krauss

Tempering and structural change in ferrous martensitic structures. In: Proc Int Conf Phase Transform Ferr Alloy; Warrendale (PA): Metallurgical Society of AIME; 1984. p. 101–123.

74.

Bhadeshia

HKDH

Honeycombe

Tempering of martensite. Steels: microstructure and properties. 4th ed. Oxford, UK: Butterworth-Heinemann; 2017. p. 237–270.

75.

Kim

Speer

Kim

Observation of an isothermal transformation during quenching and partitioning processing. Metall Mater Trans A. 2009; 40:2048–2060.

76.

Jang

Kim

Bhadeshia

HKDH.

ϵ-Carbide in alloy steels: first-principles assessment. Scr Mater. 2010; 63:121–123.

77.

Krauss

Martensite in steel: strength and structure. Mater Sci Eng A. 1999; 273–275:40–57.

78.

Hennessy

Sharma

Ansell

GS.

Tempering kinetics of alloy steels as a function of quench rate and M _s temperature. Metall Trans A. 1983; 14:1013–1019.

79.

Williamson

Nakazawa

Krauss

A study of the early stages of tempering in an Fe-1.2 Pct alloy. Metall Trans A. 1979; 10:1351–1363.

80.

Kalish

Cohen

Structural changes and strengthening in the strain tempering of martensite. Mater Sci Eng. 1970; 6:156–166.

81.

Kim

Lee

De Cooman

BC.

Microstructure of low C steel isothermally transformed in the M _s to M _f temperature range. Metall Mater Trans A. 2012; 43:4967–4983.

82.

Santofimia

Nguyen-Minh

Zhao

New low carbon Q&P steels containing film-like intercritical ferrite. Mater Sci Eng A. 2010; 527:6429–6439.

83.

Santofimia

Zhao

Petrov

Microstructural development during the quenching and partitioning process in a newly designed low-carbon steel. Acta Mater. 2011; 59:6059–6068.

84.

Navarro-López

Sietsma

Santofimia

MJ.

Effect of prior athermal martensite on the isothermal transformation kinetics below M _s in a low-C high-Si steel. Metall Mater Trans A. 2016; 47:1028–1039.

85.

Samanta

Biswas

Giri

Formation of bainite below the M _s temperature: kinetics and crystallography. Acta Mater. 2016; 105:390–403.

86.

van Bohemen

SMC

Santofimia

Sietsma

Experimental evidence for bainite formation below M _s in Fe-0.66C. Scr Mater. 2008; 58:488–491.

87.

Celada-Casero

Kwakernaak

Sietsma

The influence of the austenite grain size on the microstructural development during quenching and partitioning processing of a low-carbon steel. Mater Des. 2019; 178:107847.

88.

Kim

Lee

SJ.

Effect of Ni addition on the mechanical behavior of quenching and partitioning (Q&P) steel. Mater Sci Eng A. 2017; 698:183–190.

89.

Navarro-López

Hidalgo

Sietsma

Characterization of bainitic/martensitic structures formed in isothermal treatments below the M _s temperature. Mater Charact. 2017; 128:248–256.

90.

Jimenez-Melero

van Dijk

Zhao

The effect of aluminium and phosphorus on the stability of individual austenite grains in TRIP steels. Acta Mater. 2009; 57:533–543.

91.

Goodenow

Barkalow

Hehemann

RF.

Bainite transformations in hypoeutectoid steels. Phys prop martensite bainite, Spec Rep 93. Scarborough: The Iron and Steel Institute; 1965. p. 135–141.

92.

Cottrell

SA.

The formation of bainite. J Iron Steel Inst. 1952; 172:307–313.

93.

Hasan

Kumar

Chakrabarti

Understanding the effect of prior bainite/martensite on the formation of carbide-free bainite. Philos Mag ISSN. 2020; 100:797–821.

94.

Rementeria

Capdevila

Nguez-reyes

Carbon clustering in low-temperature bainite. Metall Mater Trans A. 2018; 49:5277–5287.

95.

Garcia-Mateo

Jimenez

Yen

Low temperature bainitic ferrite: evidence of carbon super-saturation and tetragonality. Acta Mater. 2015; 91:162–173.

96.

Ranjan

Singh

SB.

Isothermal bainite transformation in low-alloy steels: mechanism of transformation. Acta Mater. 2021; 202:302–316.

97.

Udyansky

Pezold

Bugaev

Interplay between long-range elastic and short-range chemical interactions in Fe-C martensite formation. Phys Rev B. 2009; 79:1–5.

98.

Zener

Kinetics of the decomposition of austenite. Trans Am Inst Min Met Eng. 1946; 167:550–595.

99.

Jang

Bhadeshia

HKDH

Suh

D-W.

Solubility of carbon in tetragonal ferrite in equilibrium with austenite. Scr Mater. 2013; 68:195–198.

100.

Rementeria

Poplawsky

Urones-Garrote

Carbon supersaturation and clustering in bainitic ferrite at low temperature. In: Proc 5th Int Symp Steel Sci, Vol. 1; Kyoto, Japan; 2017. p. 29–34.

101.

Kähkönen

Pierce

Speer

Quenched and partitioned CMnSi steels containing 0.3 and 0.4 wt.% carbon. J Mater. 2016; 68:210–215.

102.

Bhadeshia

HKDH.

Other Fe-C carbides. Theory of transformations in steels. Boca Raton and Abingdon: CRC Press, Taylor & Francis Group; 2021. p. 419–444.

Quenching and partitioning (Q&P) process: A critical review of the competing reactions

Abstract

Keywords

Introduction

Q&P concept: Theoretical analysis

Thermodynamics of carbon partitioning

Kinetics of the carbon partitioning

Optimum quenching temperature (QTopt)

Competing reactions: A comprehensive analysis

Segregation of carbon atoms at dislocations and defects

Incomplete carbon partitioning from primary martensite

Carbide formation

Bainite transformation

Conclusion

Footnotes

Acknowledgements

References

Optimum quenching temperature (QT^opt)