Author’s Accepted Manuscript

Microstructural characterization, strengthening and

toughening mechanisms of a quenched and
tempered steel: Effect of heat treatment parameters

Bo Jiang, Meng Wu, Mai Zhang, Fan Zhao,

Zhigang Zhao, Yazheng Liu
www.elsevier.com/locate/msea

PII: S0921-5093(17)31220-0
DOI: http://dx.doi.org/10.1016/j.msea.2017.09.062
Reference: MSA35529
To appear in: Materials Science & Engineering A
Received date: 1 July 2017
Revised date: 13 September 2017
Accepted date: 14 September 2017
Cite this article as: Bo Jiang, Meng Wu, Mai Zhang, Fan Zhao, Zhigang Zhao
and Yazheng Liu, Microstructural characterization, strengthening and toughening
mechanisms of a quenched and tempered steel: Effect of heat treatment
p a r a m e t e r s , Materials Science & Engineering A,
http://dx.doi.org/10.1016/j.msea.2017.09.062
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
 Microstructural characterization, strengthening and toughening mechanisms of a
quenched and tempered steel: Effect of heat treatment parameters
Bo Jianga, *, Meng Wua, Mai Zhangb, Fan Zhaoa, Zhigang Zhaoc, Yazheng Liua, *
a
School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing 100083, China

b
School of Advanced Engineering, University of Science and Technology Beijing, Beijing 100083, China

c
Fushun Special Steel Co. Ltd., Fushun 113001, China

*
Corresponding author. Tel.: +86 010 62333174. E-mail address: [email protected] (Bo Jiang)

*
Corresponding author. Tel.: +86 010 62334939. E-mail address: [email protected] (Ya-zheng Liu)

Abstract: A quenched and tempered steel for a large bearing ring was investigated. The heat
treatment experiments were designed by using the L9 (34) type orthogonal form. Based on these
conditions, a better combination of mechanical properties was obtained. The results showed that
the quenching and the tempering temperatures were the most influential factors on the strength
and toughness. The dislocation strengthening and the solid solution strengthening of the dissolved
alloying carbides are the main mechanisms of increasing the strength by decreasing the tempering
temperature and increasing the quenching temperature, respectively. The stripped carbides and
long chain carbides strongly influence the differences in the tensile strength of the steels under
different processes. The toughness AKv at -20 ºC was increased by 42.2 J when the quenching
temperature increased from 800 to 900 ºC. The stripped undissolved carbides at lower quenching
temperature promoted crack propagation and cleavage fracture and thus decreased the toughness
of the steel. The AKv was increased by 47.4 J when the tempering temperature increased from 550
to 650 ºC. The long chain carbides distributed along the grain boundary and the martensitic laths
with a high density of dislocations at the lower tempering temperature decreased the toughness.
Oil quenching can improve both the strength and toughness by refining the martensitic
microstructure.
Key words: Quenching, Tempering, Strength, Toughness, Carbide.
1. Introduction
An excellent combination of strength, toughness and hardenability of steel has always been
desired for use in large equipment, for example, the large bearing ring of a tunneling boring
machine (TBM). A number of studies have been conducted on the microalloying method in order
to improve the comprehensive properties [1-3]. It has been reported that the combined addition of
the C and Ni elements can not only improve the strength and hardenability but also guarantees the
toughness. In addition, the heat treatment parameters including the quenching temperature, the
quenching agent, the tempering temperature and time are also vital for the final microstructure and
the mechanical properties. Lee’s report [4] on the 4340 steel showed that the distribution of the
particles was of primary importance in controlling the grain size and the toughness. Woei-Shyan
[5] studied the mechanical properties and microstructure evolution of 4340 steel under different
tempering conditions. The results showed that strength, hardness and ductility varied with
increasing tempering temperature and holding time and were directly correlated to the appearance
of the carbides and the dislocation density of martensitic laths. Tomita [6] investigated the effect
of the full martensite quenching rate on the microstructure and tensile properties of 4340 steel and
suggested that the beneficial effect on the strength of the slowly quenched steels is caused by a
dispersion hardening effect due to carbon segregation or fine carbide precipitation in the
martensite during quenching. Generally, many studies have been performed on the separate effect
of heat treatment parameters on the strengths or toughness of steel. However, few investigations
have focused on the effect of the order of the heat treatment parameters. The research on the effect
of the order is quite meaningful for the engineers in the factory because these results provide
guidance offers for modification of the parameters and thus an improvement of the mechanical
properties. Furthermore, previous studies only focused on the correlation of the microstructure and
the properties, with only a few investigations concentrating on the discussions of the strengthening
or toughening mechanisms.
In this work, a quenched and tempered steel with the combined addition of the C and Ni for a
large bearing ring will be studied. The heat treatment experiments are designed by using L9 (34)
type orthogonal form in order to achieve the best properties. The effects of heat treatment
parameters on the microstructure evolution will also be studied. Based on these experiments, the
strengthening and toughening mechanisms will be discussed.

2. Experimental Section
2.1. Materials and processes
The experimental steel was taken from the edge of the annealed forged billet with a diameter
of approximately 1200 mm and was the commercial steel designed for the bearing ring of a TBM
with a diameter larger than 6 m. The forged steel was produced from a 30-ton electro-slag ingot
with a height of 3 m in a steel mill. In the forging process, three upsetting and cogging processes
were conducted in order to eliminate the defects and refine the casting microstructure. The
chemical composition (in wt.%) of the steel was determined as
0.45C-0.28Si-0.73Mn-1.12Cr-0.24Mo-1.43Ni and was Fe balanced. Specimens for the toughness
tests were machined in the form of a bar with dimensions of 12 mm×12 mm×60 mm. Specimens
for the strength, elongation and reduction of area tests were in the form of a cylinder with a
diameter of 12 mm and a length of 80 mm. The specimens were then heated to different quenching
temperatures, held for 30 min prior to quenching to room temperature, and then tempered at
different temperatures with soaking for different times and cooled in air. The heat treatment
protocol is shown in Fig. 1.
The effect of many different parameters on the overall performance characteristic in a
condensed set of experiments can be examined by using the orthogonal experimental design. Once
the parameters affecting a process that can be controlled have been determined, the levels at which
these parameters should be varied must be determined. Determining what levels of a variable to
test requires an in-depth understanding of the process, including the minimum, maximum, and
interval for each level [7-8]. In this paper, the heat treatment was designed by using the orthogonal
experiment method with the orthogonal form where the following four factors (parameters):
quenching temperature (factor A), tempering temperature (factor B), tempering time (factor C)
and quenching agent (factor D), and three levels were analyzed as shown in Tables 1 and 2. For
each factor, three levels were designed; for example, the quenching temperatures are 800 ºC, 850
ºC and 900 ºC according to the factory. As long as the number of factors and the number of levels
are decided, the proper orthogonal table could be selected. In this study, a L9 (34) type orthogonal
table was selected as shown in Table 2. The cooling rates of quenching processes with different
agents were measured by placing a thermocouple inside the specimens. The temperature was
recorded as a function of time. Oil cooling has the highest cooling rate at the same temperature.
Besides, the Jmat-pro software was employed to test the continuous cooling curve (CCT) of the
investigated steel in order to know the cooling behavior. The composition, the austenitizing
temperature and the grain size of the steel were input to the software, then the CCT will be
calculated.
Additionally, a verification experiment was subsequently conducted based on the optimum
isothermal process obtained from the orthogonal experiments. In addition, trials 10-12 were also
conducted as a contrast to study the influence of different factors on the microstructure and
properties.
 Fig. 1. Schematic of heat treatment process.

Table 1. Factors and levels of the L9 (34) type orthogonal heat treatment experiments

Factor

A B C D
Level
Quenching Tempering Tempering Quenching

temperature/ ºC temperature/ ºC time/h agent

1 800 550 1 Air

2 850 600 3 Wind

3 900 650 5 Oil

Table 2. Mechanical properties under different heat treatment processes

Process Mechanical property

Quenching Tempering Reduction

Trial No. Tempering Quenching Tensile Yield -20 ºC
temperature/ temperature/ Elongation/% of
time/h agent strength/MPa strength/MPa AKV/J
ºC ºC area/%

1 800 550 1 Air 919.0 734.1 9.1 14.5 54.1

2 800 600 3 Wind 920.6 803.2 29.9 17.2 61.5

3 800 650 5 Oil 784.0 650.3 46.6 25.2 65.4

4 850 550 3 Oil 1150.2 1042.6 31.5 15.0 57.1

5 850 600 5 Air 963.4 824.0 33.9 18.0 64.6

6 850 650 1 Wind 937.7 802.4 59.7 18.1 63.5

7 900 550 5 Wind 1126.3 1003.5 50.2 15.4 58.1

8 900 600 1 Oil 1037.4 907.2 64.8 15.9 62.0

9 900 650 3 Air 916.9 770.1 71.2 19.3 66.8

Verification
900 650 1 Oil 985.4 850.5 83.7 20.5 65.2
experiment

10 800 650 1 Oil 873.4 748.3 42.2 21.4 63.2

11 900 550 1 Oil 1221.2 1100.5 36.3 15.0 54.8

12 900 650 1 Air 962.3 814.5 67.7 20.4 67.6

2.2. Properties tests and microstructural investigation
After the heat treatments, Charpy V-notch samples with dimensions of 10 mm×10 mm×55
mm were prepared. The V-notch was perpendicular to the rolling direction. Charpy impact tests
were performed at -20 ºC on an SANS-ZBC2452-B instrumented impact machine according to the
working condition of the steel. Specimens for the tensile tests with 5 mm diameter and 25 mm
gauge length were prepared according to GB/T 228-2010 [9]. Three specimens were tested for
each heat treatment process in order to guarantee reproducibility.
Samples for metallographic analysis were prepared by wire cut electrical discharge
machining. The martensite and bainite microstructures were observed by ZEISS ULTRA 55 Field
Emission Scanning Electron Microscopy (FESEM) and FEI QUANTA 250 Scanning Electron
Microscope (SEM) after grinding, polishing, and erosion with 4% nital solution. The average
grain size was measured by quantitative image analyzer. A JEM-2100(HR) transmission electron
microscope (TEM) was also used to observe the fine structure and the particles. The chemical
compositions the particles were analyzed using the energy dispersive X-ray spectrum (EDS) in the
TEM. The crystal structures of the particles were identified by Digital Micrograph software. The
fracture surfaces for each heat treatment were examined in the SEM to determine the fracture
mode. Furthermore, the surface perpendicular to the fracture was also observed by FESEM to
investigate the relationship between the fracture initiation and the microstructure in the different
deformation processes.

3. Results and Discussions

3.1. Orthogonal experiments
As shown in Table 2, nine experiments (Nos.1-9) were carried out based on the L9 (34) type
orthogonal form, and the mechanical properties including tensile strength, yield strength, impact
toughness AKv, elongation and reduction of area were obtained. As shown in Fig. 2, relations
between the factors and indexes were obtained according to the range analysis method. The range
value indicates the significance of the factors’ effect. According to reference [10], a larger range
value means that the factor has a larger effect on the mechanical properties.
Examination of Figs. 2 shows that the tempering temperature is the most influential factor
and that the quenching temperature is the secondary influential factor for the tensile and yield
strengths. These properties reach their maximum values at the A3B1C2D3 combination.
Meanwhile, it can be seen from Fig. 2 that the quenching and tempering temperatures both have a
stronger influence on the AKv value at -20 ºC. The highest value of AKv can be obtained at the
A3B3C1D3 combination. For the index of elongation shown in Fig. 2, the tempering temperature
is the most influential factor. However, the value of elongation only changes from 15% to 21%
with the factors. Therefore, generally speaking, the elongation is not sensitive to these factors, and
its highest value can be obtained at the A1B3C3D3 combination. The relationship between the
factors and the area reduction is quite similar to that of the elongation. The highest value of the
area reduction is at the A3B3C3D3 combination. Consequently, to obtain the comprehensive
mechanical properties with high toughness, elongation, area reduction and modest strength (higher
than 900 MPa), the optimum heat treatment process was determined: quenching at 900 ºC and
tempering at 650 ºC, holding for 1 h and using oil as the quenching agent.
 Fig. 2. Relations between factors and indexes.

3.2. Verification experiment

The optimum heat treatment process was conducted as shown in Table 2 in order to confirm
that the comprehensive mechanical properties can be obtained. The results are presented in Table 2.
It can be seen that the tensile strength of 985.4 MPa, yield strength of 850.5 MPa, AKv at -20 ºC
of 83.7 J, elongation of 20.5% and area reduction of 65.2% were obtained in the verification
experiment. This shows that a better combination of strength, toughness and plasticity was in fact
achieved.

3.3. Effect of factors on the mechanical properties and microstructure evolution

3.3.1 Effect of quenching temperature
The mechanical properties of trials 10-12, conducted to study the effect of different factors,
were obtained and are shown in Table 2. Comparison of the obtained properties for trial No. 10
and the optimum process shows that when the quenching temperature decreased from 900 to 800
ºC, the tensile strength, the yield strength and the toughness AKv were greatly decreased, while
the reduction of area and the elongation remained almost unchanged. Figure 3(a) shows TEM
micrographs of the microstructures of the steel obtained using the optimum process. It is clear that
the typical tempered martensitic structure was achieved, consisting of martensitic laths, spherical
carbides and short rod-like carbides. The TEM micrographs of the microstructures of the steel for
trial No. 10 are shown in Figure 3(b). It can be seen that many long-stripped carbides are present
in the steel in addition to the spherical carbides. The stripped carbides are due to the dissolved
carbides inherited from the original annealed microstructure because of the lower quenching
temperature in the process. The EDS results, which show the chemical compositions of the
carbides, are listed in Table 3. The electron diffraction patterns presented in Figure 3(b) show that
these carbides are all of the M23C6 type, in complete agreement with all previous studies [2-3].

(a) (b)

Fig. 3. TEM micrographs showing the quenched and tempered microstructures: (a) for the optimum

process; (b) for trial No. 10 (quenching temperature 800 ºC).

Table 3. Composition and distribution of carbides

Total alloying
Trial Carbide C/wt.% Cr/wt.% Mn/wt.% Si/wt.% Ni/wt.% Mo/wt.%
elements/wt.%
 The
optimum Spherical 11.10 9.98 3.45 9.30 0.19 - 36.35
process
The
Short
optimum 9.12 5.18 1.38 13.08 0.73 - 20.37
rod-like
process

No. 10 Stripped 10.97 13.09 3.88 - 0.86 2.66 20.49

3.3.2 Effect of tempering temperature

The comparison of properties for trial No. 11 and the optimum process showed that when the
tempering temperature was lowered from 650 to 550 ºC, the tensile and yield strengths were
increased markedly, but the toughness AKv, the reduction of area and the elongation were all
greatly decreased. Comparison of the microstructures shown in Figs. 3 and 4 shows that the
carbides in the microstructure for trial No. 11 are all long chain carbides distributed along the
grain boundaries. In addition, due to the lower tempering temperature, many martensitic laths with
high density of dislocations are still present, as shown in Fig. 4(b).

(a) (b)

Fig. 4. TEM micrographs showing the quenched and tempered microstructures for trial No. 11 (tempering

temperature 550 ºC).

3.3.2 Effect of quenching agent

Comparison of the results for trial No. 12 and the optimum process shows that when the
quenching agent was changed from oil to air, the tensile and yield strengths were slightly
decreased, and both the elongation and the reduction of area did not change much, whereas the
toughness AKv decreased more significantly from 83.7 to 67.7 J. Figure 5 shows that there are
also some regions with aggregate spherical carbides other than the typical martensitic tempered
laths. The microstructures in these regions are similar to that of the bainite after tempered. To
confirm this assumption, further quenching experiments were conducted. The quenched
microstructures with different quenching agents were observed. Examination of Fig. 6 shows that
the oil-quenched microstructure is full martensite, while the air-quenched microstructure is
martensite and bainite. The bainite carbides in Fig. 6(b) show the same morphology as the
aggregate carbides in the tempered microstructure for trial No. 12. Additionally, the CCT curve of
the steel was calculated by the Jmat-pro software and is shown in Fig. 7. Examination of the oil
and air cooling curves in Fig. 7 shows that the oil cooling was effective in the area of full
martensite, while the air cooling dropped in the area of martensite and bainite. Consequently, it
can be concluded that the microstructure with aggregate spherical carbides for trial No. 12 with air
as the quenching is the bainite after tempered. In addition, it can be seen that the oil-quenched
microstructure is obviously finer than that air-quenched microstructure. The sizes of the martensite
packets in the oil and air microstructures shown in Figs. 6(a) and 6(b) were measured to be 18 μm
and 23 μm, respectively. The width of the martensitic tempered laths in the air cooling
microstructure is also larger.

Fig. 5. TEM micrographs showing the quenched and tempered microstructures for trial No. 12 (air is the

quenching agent).

(a) (b)

Fig. 6. FESEM micrographs showing the quenched microstructures with the same quenching temperature

of 900 ºC: (a) oil as the quenching agent; (b), (c) air as the quenching agent.
 Fig. 7. CCT curves of the investigated steel calculated using the JMatPro software.

3.4. Tensile fractography analysis and the strengthening mechanism

3.4.1. Fractography observation
For all tested specimens, the observation of the fracture surface shows that heavy necking
occurred during tensile loading. The typical ductile ‘cup-and-cone’ fracture surface could be seen
in Fig. 8, in which the fracture surface of steel under the optimum process was shown. It can be
concluded that the fracture surface mainly includes three parts [5]: the fiber region in the center;
the outer smaller shear lip; and the larger radial region between the two. The fiber region in the
center is the initiation of the fracture, and thus, the morphology of this region reflects the tensile
properties [11]. A detailed view of the fiber region of steel is shown in Fig. 8. The fracture surface
exhibits mostly ductile dimples resulting from the coalescence of microvoids but also a certain
number of secondary cracks. However, there are more secondary cracks on the fracture surface of
the sample for trial No. 10. The fracture surface of the sample for trial No. 11 is different because
the fracture mainly consists of cleavages as well as some portion of the dimples.
(a) (b)

Fig. 8. Tensile fracture : (a) the optimum process specimens; (b) trial No.10.

The surfaces perpendicular to the fracture were also observed and are shown in Fig. 9. As is
(a)
well-known, tensile fracture begins with the formation of the (b)
microvoids and follows the crack
initiation. The microvoids are easily formed between the matrix and the secondary particles such
as the inclusions and the carbides [12]. As can be seen from Fig. 9(a), the surface of the sample
under the optimum process shows that some microvoids exist that were formed between the
matrix and the carbides. The microvoids will then be transformed into the secondary crack as
shown in Fig. 9(a). It can be seen that the crack propagates along the martensite block boundary
because it requires more energy to cross a boundary [13]. This phenomenon is confirmed in Fig.
9(b), which shows the surface of the sample for trial No. 10. More stripped or spherical carbides
are present near the fracture surface, which is in agreement with the microstructure observation
discussed in section 3.3. It is clear that many microvoids are present between the matrix and the
carbides. However, no microvoids can be observed in Fig. 9(c), which shows the surface of the
sample for trial No. 11. The fracture occurred along the martensite block, martensite packet or the
grain boundaries. For the surface of the sample of trial No. 12, the morphology is nearly the same
as that of the optimum process.
(a) (b) (c) (d)

Fig. 9. Surfaces perpendicular to the fracture of specimens under different experiments: (a) optimum

process; (b) trial No. 10; (c) trial No. 11: (d) trial No. 12.

3.4.2. Effect of quenching temperature on the strengthening mechanism

For the tempered martensitic steel, the yield strength can be explained by several
strengthening mechanisms including solid solution strengthening, dislocation strengthening, grain
boundary strengthening and precipitation strengthening by hindering dislocation movement [14-15,
2-3]. For the optimum process and trial No. 10 with different quenching temperatures, the
dislocation strengthening should be nearly the same due to the same tempering temperature and
time. Thus, solid solution strengthening, grain boundary strengthening and precipitation
strengthening can give rise to the difference in the strengths between the two processes. According
to the Hall-Petch relationship, σG=σi+kGd-1/2 was obeyed. kG was nearly 24.7 MPamm1/2, and
consequently, the finer microstructure of trial No. 10 can increase the strength due to the lower
quenching temperature. Additionally, there are more carbides in the microstructure of trial No. 10
as shown in Figs. 2, 3 and 9(b). The particles can improve the strength by blocking the movement
of dislocations and preventing the movement of the interfaces of subgrains [16]. Thus, the M23C6
carbides can also contribute to the strengthening of the steel for trial No. 10. However, it can be
seen in Table 3 that the alloying elements contents of spherical and stripped carbides are all very
high compared to that of the matrix as shown in section 2.1. This means that more carbon and
alloying atoms were dissolved in the steel under the optimum process due to the higher quenching
temperature. The solid solution strengthening here can be explained that the ferritic matrix is
strengthened by the formation of alloy solid solution. For the strong solid solution strengthening
elements, such as carbon, the yield strength increase can be calculated as △σs=ks[C]1/2. For the
weak solid solution strengthening elements, the yield strength increase can be calculated as
△σs=ks[M]. Here, ks is the coefficient, and [M] is the weight percentage of the element. Therefore,
the solid solution strengthening can contribute to the increase of the strength of the steel under the
optimum process. According to the results shown in Table 2, the yield strength of the steel of the
optimum process is 102.2 MPa higher that of trial No. 10 steel. Therefore, it can be concluded that
solid solution strengthening is the main effect of the quenching temperature on the strengthening
mechanism .
For the tempered martensitic steel, the tensile strength can also be explained by the
strengthening mechanisms that control the yield strength through the control of the initiation and
propagation of the microcracks [17]. In addition to the above-mentioned mechanisms, the factors
that influence the initiation and propagation of the microcracks, such as inclusions or carbides,
also have a great effect on the tensile strength. The solid solution strengthening, grain boundary
strengthening and precipitation strengthening are the mechanisms that can influence the tensile
strength of the steels of the optimum process and trial No. 10. While the relationship between the
tensile strength and the grain size also obeys the Hall-Petch relationship, TSG=TSi+kGd-1/2, the
coefficient kG is smaller than that of the yield strength. Thus, the finer microstructure in the steel
of trial No. 10 increases the tensile strength less than the yield strength. The carbides in the steel
of the trial No. 10 can also increase the tensile strength by preventing the propagation of
microcracks similar to the grain boundary. This effect is nearly the same as the effect on the yield
strength. However, only coherent and semicoherent carbide particles with high distribution density
may be taken as effectively preventing the crack propagation, as we can see from Fig. 9(b), the
microvoids were easily formed between the matrix and the carbides, and more secondary cracks
were formed on the fracture surface of the sample of trial No. 10. The applied stress for the
formation of microvoids around the second phase particle was inferred by LeMay [18-19] to be
given by the following equation:

( )( ) ( )( ) (1)

where q is the average stress concentration coefficient; γS is the specific surface energy of the
matrix; E is the elastic modulus of the matrix; a is the size of second phase particle; σ is the yield
strength of the matrix; V is the volume of all the second phase particles and △V is the volume of
the plastic deformation zone around the second phase particles. It can be concluded from the
equation that the value of SVF is decreased when the size of the second phase particle increases,
and thus, the microcracks are easily formed. As can be seen from Figs. 3 and 9(b), the size of the
carbides in the steel of trial No. 10 is substantially larger than that in the steel under the optimum
process, and the microvoids are thus more easily formed and are larger. In the steel, the critical
size of microcracks can be defined as [20-21]:

(2)

where γP is the plastic energy consumed by the formation of the microcracks per unit area; SC is
the fracture strength; and v is the Poisson’s ratio. It can be inferred from this equation that
increasing the size of the microcracks in the steel can effectively decrease the tensile strength.
Based on this, the M23C6 carbides with large size in the steel of trial No. 10 decreased the tensile
strength by promoting the initiation of the microcracks. For the solid solution strengthening, the
higher contents of carbon and alloying elements in the steel of the optimum process can also
increase the tensile strength even though the increase is lower than that of the yield strength.
According to the results shown in Table 2, the tensile strength of the steel of the optimum process
is 112.0 MPa higher that of trial No. 10. The tensile strength difference is larger than the yield
strength difference. It can be concluded that the solid solution strengthening is the main tensile
strengthening mechanism of the steel under the optimum process, while the effects of grain
boundary strengthening and the stripped carbides on the tensile strength are the key mechanisms
that increase the difference of the tensile strength of the steels of the two processes.

3.4.3. Effect of tempering temperature on the strengthening mechanism

For the optimum process and trial No. 11 with different tempering temperatures, the grain
boundary strengthening and solid solution strengthening should be nearly the same due to the
same quenching temperature and agent. According to the Ashby-Orowan equation, when the
precipitates are non-deformable, the precipitation strengthening is given by [17]:

(3)

where M is the Taylor factor for bcc materials and is approximately equal to 2; G is the shear
modulus (80 GPa at room temperature); b is the Burgers vector equals to 0.248 nm; is
Poisson’s ratio (0.291); d is the diameter of the particle in the steel of the optimum process which
was measured as 77.8 nm; f is the volume fraction of precipitates which can be calculated as 0.8%
according to the previous study [22]. Then, the contribution of the precipitation strengthening to
the yield strength in the steel under the optimum process can be obtained as 153 MPa. According
to the Ashby-Orowan equation, the precipitation strengthening should be smaller in the steel of
trial No. 11 due to the large length of the carbides. However, the yield strength of the steel of trial
No. 11 is 250 MPa higher than that of the steel of the optimum process, as can be seen from the
data in Table 2. Therefore, the effect of the tempering temperature on the strengthening
mechanism is due to dislocation strengthening. A higher tempering temperature can promote the
recrystallization of the martensite and thus lowers the dislocation density and softens the matrix.
The high dislocation density regions can also be found in the microstructure of trial No. 11 as
shown in Fig. 4(b). The results can also be confirmed by many previous studies which showed that
the tempering temperature can effectively influence the strength [23-25].
As discussed above, the dislocation strengthening and precipitation strengthening are also the
mechanisms which can influence the tensile strength of the steels of the optimum process and trial
No. 11 with different tempering temperatures. The precipitation strengthening can increase the
tensile strength of the steel under the optimum process nearly as much as the yield strength. The
dislocation strengthening is also the main strengthening mechanism which improve the strength of
the steel under the trial No. 11. However, the tensile strength increase is smaller than that of the
yield strength because a massive dislocation pile-up can cause the initiation of the microcracks. In
addition, as we can see from Fig. 3(b), long carbides are distributed along the martensite packet
boundaries. As discussed in section 3.4.2, the long carbides also accelerate the initiation of
microcracks along the boundaries, promote the crack initiation and decrease the tensile strength.
Consequently, the tensile strength difference between the two processes is 235.8 MPa, which is
lower than that for the yield strength. Of course, dislocation strengthening is still the mechanism
improving the tensile strength of the steel of trial No. 11. However, the effects of the long chain
carbides and the dislocation strengthening are both the mechanisms which decrease the difference
of the tensile strengths of the steels of the two processes with different tempering temperatures.

3.4.4. Effect of quenching agent on the strengthening mechanism

According to the microstructural observations, the dislocation strengthening, precipitation
strengthening and solid solution strengthening for the optimum process and trial No. 12 with
different quenching agents should be nearly the same. Therefore, the grain boundary strengthening
is the main mechanism that causes the difference in the strengths between the two processes. As
discussed above, the Hall-Petch relationship, σG=σi+kGd-1/2, was obeyed. kG was nearly 24.7
MPamm1/2 for the yield strength and was nearly 17.0 MPamm1/2 for the tensile strength. d is the
martensite packet size. It can be calculated that the yield strength difference is 24.7 MPamm1/2
(d2-1/2-d1-1/2), while the tensile strength difference is 17.0 MPamm1/2 (d2-1/2-d1-1/2). The ratio of the
difference of the tensile strength and the difference of the yield strength is 0.688. According to
Table 2, it can be seen that the tested value of the ratio was 0.642 which is quite similar to the
calculated value. This means that the grain boundary strengthening is the strengthening
mechanism which increases the strengths of the steels for the processes with oil quenching.

3.5. Charpy impact fractography analysis and the toughening mechanism

The toughness of steel mainly depends on its ability to absorb the energy during the process
of deformation or fracture, namely, the ability to resist the formation and expansion of the
crack[17]. In the fracture process, the toughness mainly depends on the crack driving force. For
the optimum process and trial No. 10 with different quenching temperatures, there exist the
differences of the solid solution, grain boundary and precipitation which may influence the driving
force. The stronger solid solution and coarser grain in the steel under the optimum process were
both detrimental to the toughness, as shown in previous studies [17]. However, the toughness of
the steel under the optimum process is 83.7 J, which is 41.5 J higher than that of the trial No. 10
steel. As shown in Fig. 3, many stripped undissolved carbides exist in the steel of trial No. 10 with
lower quenching temperature. It can be seen in Fig. 10(c), which shows the surface perpendicular
to the fracture of the specimen of trial No. 10, that the deformation of the carbides was not
compatible with that of the matrix, and thus, many microvoids were formed. These microvoids
promote the crack propagation and reduce the consumed energy. Figures 10(a) and 10(b) show the
impact fractures of the specimens at the crack initiation under different experiments. It can be seen
that the fracture of the specimen of the optimum process mainly consists of dimples and small
amounts of quasi-cleavage surfaces. However, there are more cleavage surfaces and cleavage
steps on the facture of specimen of trial No. 10, as shown in Fig. 10(b). Consequently, it can be
concluded that the carbides promote the crack propagation and the cleavage fracture, and thus
decrease the toughness of the steel for trial No. 10.
(a) (b)
 (c)

Fig. 10. Morphology of the specimens for different experiments: SEM impact fracture (a) optimum process,

(b) trial No. 10; (c) surface perpendicular to the fracture of specimen for trial No. 10.

For the optimum process and trial No. 11 with different tempering temperatures, the main
differences influencing the toughness are the long chain carbides distributed along the grain
boundary and the martensitic laths with high dislocation density in the steel of trial No. 11 as
shown in Fig. 4. In the fracture process, the massive dislocation pile-up can initiate the cleavage
crack by Cottrell dislocation reaction, and the long chain carbides along the boundary can promote
the intergranular crack [23, 26-27]. As shown in Fig. 11(a), the fracture of the steel of trial No. 11
mainly consists of the cleavage facets and the cracks. Comparison of trial No. 12 and the optimum
process steels with different quenching agents shows that the microstructure of the steel in the
former case is coarser and consists of some bainite. Bainite has a higher toughness than the
martensite. Consequently, it can be deduced that the grain size of the microstructure plays the
main role in determining the toughness. Many reports have confirmed that finer microstructure
can improve the impact energy by preventing the crack propagation in the fracture process [28-30].
It can be seen in Figs. 6(a), 6(b), 3(a) and 5 that the martensitic packets size and the laths width in
the steel of the optimum process are both finer than that of the trial No. 12. Therefore, the impact
fracture surface of the specimen of trial No. 12 shown in Fig. 11(b) is obviously coarser and
consists of larger proportion of quasi-cleavage facets compared with that of the optimum process.
 (a) (b)

Fig. 11. SEM impact fractures of the specimens under different experiments: (a) trial No. 11, (b) trial No. 12.

4. Conclusions
(1) The tempering temperature is the most influential factor and the quenching temperature
the secondary influential factor on the tensile and yield strengths of the investigated steel. The
quenching and tempering temperatures both have a stronger influence on the AKv value at -20 ºC.
The elongation and the reduction of area are not sensitive to these factors.
(2) The optimum heat treatment process for the investigated steel was determined and
consists of quenching at 900 ºC and tempering at 650 ºC, holding for 1 h, and the quenching agent
is oil. The steel produced by the optimum heat treatment process exhibits a better combination of
mechanical properties, namely, tensile strength of 985.4 MPa, yield strength of 850.5 MPa, the
AKv at -20 ºC of 83.7 J, the elongation of 20.5% and the reduction of area of 65.2%.
(3) Increasing the quenching temperature from 800 to 900 ºC increased the tensile strength by
112.0 MPa and the yield strength by 102.2 MPa. The solid solution strengthening of the dissolved
alloying carbides is the main mechanism. The effects of grain boundary strengthening and the
stripped carbides on the tensile strength are the key mechanisms that increase the difference in the
tensile strength of the steels.
(4) Increasing the tempering temperature from 550 to 650 ºC decreased the tensile strength
by 235.8 MPa and the yield strength by 250.0 MPa. The dislocation strengthening is the main
mechanism. The long chain carbides and the dislocation strengthening both decrease the
difference in the tensile strength of the steels. Replacement of air quenching by oil quenching
slightly increased the strengths. The grain boundary strengthening is the mechanism of this effect.
(5) The toughness value AKv at -20 ºC was increased by 42.2 J when the quenching
temperature increased from 800 to 900 ºC. The stripped undissolved carbides decreased the
toughness of the steel. The AKv was increased by 47.4 J when the tempering temperature
increased from 550 to 650 ºC. The long chain carbides and the martensitic laths with high
dislocation density decreased the toughness. Compared to air quenching, oil quenching increased
the AKv by 16 J by refining the martensitic microstructure.
Acknowledgements
The authors appreciate the financial support by the Fundamental Research Funds for the
Central Universities of China (No.FRF-TP-16-032A1), China Postdoctoral Science Foundation
(No.2016M600915) and Student Research Training Program of University of Science and
Technology Beijing (No.20170374).

References
[1] Jiang B, Zhou L, Wen X, et al. Heat treatment properties of 42CrMo steel for bearing ring of varisized
shield tunneling machine[J]. Acta Metallurgica Sinica (English Letters), 2014, 27(3): 383-388.
[2] Jiang B, Mei Z, Zhou L, et al. Microstructure evolution, fracture and hardening mechanisms of quenched
and tempered steel for large sized bearing rings at elevated quenching temperatures[J]. Metals and
Materials International, 2016, 22(4): 572-578.
[3] Jiang B, Dong Z, Zhou L, et al. Microstructural characterization and hardening mechanism of steel for
large size bearing ring under fast heating and short soaking time condition[J]. Steel Research International,
2015, 87(9): 1127-1136.
[4] Lee S, Majno L, Asaro R J. Correlation of microstructure and fracture toughness in two 4340 steels[J].
Metallurgical and Materials Transactions A, 1985, 16(9): 1633-1648.
[5] Lee W S, Su T T. Mechanical properties and microstructural features of AISI 4340 high-strength alloy steel
under quenched and tempered conditions[J]. Journal of Materials Processing Technology, 1999, 87(1):
198-206.
[6] Tomita Y, Okabayashi K. Effect of quench rate on microstructure and tensile properties of ALSL 4320 and
4340 steels[J]. Metallurgical and Materials Transactions A, 1987, 18(1): 115-121.
[7] Roy R K. Design of Experiments Using Taguchi Approach: 16 Steps to Product and Process Improvement,
John Wiley & Sons, New York, NY, USA, 2001.
[8] Ross P J. Taguchi Techniques for Quality Engineering, Mc-Graw Hill, New York, NY, USA, 1988.
[9] GB/T 228-2010, Metallic Materials-Tensile Testing-Part 1: Method of Test at Room Temperature, 2011.
Beijing, China.
[10] Wu X, Leung D Y C. Optimization of biodiesel production from camelina oil using orthogonal
experiment[J]. Applied Energy, 2011, 88(11): 3615-3624.
[11] Kao A S, Kuhn H A, Richmond O, et al. Tensile fracture and fractographic analysis of 1045 spheroidized
steel under hydrostatic pressure[J]. Journal of Materials Research, 1990, 5(1): 83-91.
[12] Ceschini L, Marconi A, Martini C, et al. Tensile and impact behaviour of a microalloyed medium carbon
steel: Effect of the cooling condition and corresponding microstructure[J]. Materials & Design, 2013,
45(6):171-178.
[13] Wang C, Wang M, Shi J, et al. Effect of microstructure refinement on the strength and toughness of low
alloy martensitic steel[J]. Journal of Materials Science and Technology, 2007, 23(5):659-664.
[14] Yang G, Sun X, Li Z, et al. Effects of vanadium on the microstructure and mechanical properties of a high
strength low alloy martensite steel[J]. Materials & Design, 2013, 50: 102-107.
[15] Rancel L, Gómez M, Medina S F, et al. Measurement of bainite packet size and its influence on cleavage
fracture in a medium carbon bainitic steel[J]. Materials Science and Engineering: A, 2011, 530: 21-27.
[16] Wang S S, Peng D L, Chang L, et al. Enhanced mechanical properties induced by refined heat treatment
for 9Cr–0.5 Mo–1.8 W martensitic heat resistant steel[J]. Materials & Design, 2013, 50: 174-180.
[17] Q. Yong, Second Phases in Structural Steels, Metallurgical Industry Press, Beijing, P. R. China, 2006.
[18] Le May I. Microstructural Analysis and Fractography[M]//Microstructural Analysis. Springer US, 1973:
153-168.
[19] Le May I. Failure mechanisms and metallography: a review[M]//Metallography in failure analysis.
 Springer US, 1978: 1-31.
[20] Griffith A A. The Phenomena of rupture and flow in solids[J]. Philosophical Transactions of the Royal
Society of London. Series A. 1920, 221: 163-198.

[21] Orowan E. Discussion in symposium on internal stresses in metals and alloys,

Institute of Metals, Lodon, United Kingdom, 1948.

[22] Götz G, Blum W. Influence of thermal history on precipitation of hardening phases in tempered martensite
10% Cr-steel X12CrMoWVNbN 10-1-1[J]. Materials Science and Engineering: A, 2003, 348(1): 201-207.
[23] Jundan Chen, Wenlin Mo, Pei Wang, et al. Effects of tempering temperature on the impact toughness of
steel 42CrMo[J]. Acta Metallurgica Sinica, 2012, 48(10): 1186-1193. (in chinese)
[24] Dhua S K, Ray A, Sarma D S. Effect of tempering temperatures on the mechanical properties and
microstructures of HSLA-100 type copper-bearing steels[J]. Materials Science and Engineering: A, 2001,
318(1): 197-210.
[25] Zou D, Ying H, Zhang W, et al. Influence of tempering process on mechanical properties of 00Cr13Ni4Mo
supermartensitic stainless steel[J]. Journal of Iron and Steel Research, International, 2010, 17(8): 50-54.
[26] Dhua S K, Ray A, Sarma D S. Effect of tempering temperatures on the mechanical properties and
microstructures of HSLA-100 type copper-bearing steels[J]. Materials Science and Engineering: A, 2001,
318(1): 197-210.
[27] Mayer S, Leitner H, Scheu C, et al. Changes in tempering and its effect on precipitation behaviour in a
hot-work tool steel[J]. Steel Research International, 2009, 80(1): 89-95.
[28] Petch N J. The influence of grain boundary carbide and grain size on the cleavage strength and impact
transition temperature of steel[J]. Acta Metallurgica, 1986, 34(7): 1387-1393.
[29] Hwang B, Kim Y G, Lee S, et al. Effective grain size and Charpy impact properties of high-toughness X70
pipeline steels[J]. Metallurgical and Materials Transactions A, 2005, 36(8): 2107-2114.
[30] Morito S, Yoshida H, Maki T, et al. Effect of block size on the strength of lath martensite in low carbon
steels[J]. Materials Science and Engineering: A, 2006, 438: 237-240.