Study on the effect of microstructure change of 20CrNiV5 high strength and toughness steel on hardenability

High-pressure and high-load drilling environment in the deep soil requires high strength and toughness of the oil well pipe, so the oil well pipe steel must be heat-treated, and the hardenability of the steel is particularly important in manufacturing. In this paper, Gleeble-3500 thermal simulator is used to measure the CCT curves of 20CrNiV5 steel, and the influence of microstructure on hardenability is studied. Deformation inhibits the martensitic transformation of the steel and leads to the decrease of its hardenability; the inhomogeneity of the steel pipe after hot forming also has an important impact on the hardenability, and the multiphase microstructure of bainite and ferrite after hot rolling leads to austenitic heating. Austenite grains are small and uneven, which reduces the hardenability of the steel. The microstructure of the inner and outer layers of the hot-formed steel pipe is smaller than that of the middle layer. After heat treatment, its smaller austenite results in weak hardenability.


Introduction
In recent years, oil and gas development has shifted to deep oil and gas fields with harsh environments [1].The environment under high pressure and high load conditions requires oil well pipes to have high strength and toughness.An appropriate heat treatment process is a necessary condition for obtaining high strength and toughness, and hardenability is an important basis for formulating the heat treatment process.Good hardenability can ensure uniform surface and inner properties [2][3][4], so the effect of microstructure hardenability on steel properties is particularly significant.
The hardenability of steel mainly depends on its critical cooling rate, which is affected by many factors such as composition, heat treatment process, and austenite grain size [5][6][7].The main research methods mainly include the end quenching and CCT curve methods.Alloying elements can effectively improve the hardenability of steel [8][9][10][11][12], high heating temperature and long holding time in austenitizing can shift the C curve to the right and improve the hardenability [13,14], larger austenite grains also improve hardenability [15].In recent years, controlled rolling and cooling technology has been used in steel production to diversify the microstructure of steel.It is necessary to study the influence of various microstructures on the hardenability of steel.Therefore, based on measuring the CCT curves of 20CrNiV5 steel, the influence of hot-forming microstructure on its hardenability has been studied.

Experimental process and method
Experimental steel is 20CrNiV5, a kind of low alloy high strength steel, and its chemical composition is shown in Table 1.The steel cut from the hot-rolled slab is homogenized heat treated, and then the dynamic and static CCT curves are measured by Gleeble-3500 thermal simulator.When measuring static CCT curves, the steel was heated to 1150℃ at 5℃/s and held for 5 min for austenitization, then cooled to 890℃ at 10℃/s and held for 50 seconds, and then cooled to room temperature at 0.1, 0.3, 1, 5, 10, 20, 25, 30, 40, 50, 80℃/s.Dynamic CCT measurement parameters are the same, except that it is cooled from 1150 ℃ to 890℃, and then 20% compression was performed at a rate of 10℃/s after holding for 15 s.After holding for 35 s, it is also cooled to room temperature at the same cooling rate as the static CCT test.After thermal simulation, the samples were cut along the thermocouple welding point to make a metallographic sample and then eroded with a 4% nitric acid alcohol solution.The microstructure was analyzed by metallographic microscope (OM) and FEI Nova 400 field emission electron microscope (SEM), and the hardness was measured by an HV-1000 A microhardness tester.The phase transformation point was measured from the thermal simulation expansion curve of each sample by tangent method, and the dynamic and static CCT curves were drawn by combining the microstructure and microhardness, and the critical cooling rate of quenching was obtained.To compare the structural characteristics of different layers, the samples were cut from the outer, middle, and inner layer of the hot-formed steel pipe, and the thermal simulation test of the critical cooling rate was carried out.According to GB/T 6394-2017 (the Metal average grain size determination method) [16], the grain size of the original austenite, the grain size, and the volume fraction of ferrite and bainite were determined.

Critical cooling rate of 20CrNiV5 steel
Figure 1 shows the OM (a 1 -d 1 ) and SEM (a 2 -d 2 ) images of static CCT after thermal simulation with different cooling rates.The microstructure of steel with a cooling rate of 0.1℃/s is ferrite and pearlite.At 1℃/s, the microstructure is mainly bainite, and there is trace ferrite.At 5℃/s, the microstructure of the steel is bainite.When the cooling rate ≥ 20℃/s, the microstructure is martensite.Figure 2 shows the OM and SEM images of dynamic CCT at 25℃/s cooling rate, which is full martensite.It can be seen from Table 2 that the start temperature of ferrite, pearlite, and bainite phase transition measured by dynamic CCT experiment is higher than that measured by static CCT, while the start temperature of martensite phase transition measured by dynamic CCT experiment is lower than that measured by static CCT.The microstructure at a low cooling rate of 0.1℃/s ~ 0.3℃/s is mainly ferrite and pearlite, and the microhardness value is about 132 ~ 284 HV0.05.When the cooling rate increases to 1 ℃/s ~ 15 ℃/s, the microstructure of the steel is mainly composed of bainite and martensite, and the microhardness value is about 334 ~ 459 HV0.05.In addition, when the cooling rate is further increased to 20 ℃/s ~ 80 ℃/s, the microstructure is mainly martensite, and the microhardness value is the largest, about 458 ~ 499 HV0.05.By comparing the dynamic and static CCT curves, it can be concluded that 0.1℃/s to 1℃/s is the static CCT ferrite and pearlite phase zone, and the dynamic CCT transition temperature range shift to the upper left, the static CCT bainite phase zone extends from 0.3℃/s to 20℃/s, while the dynamic CCT phase zone extends to 25℃/s.The critical quenching cooling rate of static CCT is 20℃/s, while that of dynamic CCT is 25℃/s.
The deformation makes the ferrite and pearlite start to phase transition temperature increase, the bainite phase transition zone expands, and the B s temperature increase.At the same time, the deformation increases the critical quenching cooling rate and reduces the M s point temperature.This is because the thermal deformation of austenite will increase the crystal defects and strengthen the austenite microstructure, so that the martensite non-diffusion shear mechanism is blocked, resulting in a decrease in hardenability [17,18].

Microstructure of hot-formed steel pipe and its hardenability at the critical cooling rate
Fig. 4 shows the original microstructure and quenched microstructure of the hot-formed steel pipe.The original microstructure is composed of ferrite and bainite.The grain sizes of bainite lath and ferrite are slightly larger in the middle layer, and slightly smaller in the inner and outer edge layers.For the quarter layer, the bainite lath average size is 1.29 μm, and the ferrite grain average size is 10.54 μm, which is larger than the edge layer and smaller than the middle layer.Corresponding quantitative statistics are shown in Table 3.This is because the inner and outer edges are cooled by water, and the cooling rate is faster than that of the middle layer so that the grains are refined.In the inner and outer edge layers, the volume fraction of bainite is more than that in the middle layer, but for the ferrite, this situation is the opposite.This is also related to the cooling rate.When the cooling rate is lower, it is beneficial to the ferrite transformation.The increase in the cooling rate will inhibit the diffusion of iron and carbon atoms, thus reducing the phase transformation of ferrite and promoting the bainite transformation [19].thermal simulation.The microstructure of hot-forming steel after thermal simulation is mainly martensite, but based on the metallographic and SEM images, it can be seen that there are trace amounts of bainite and ferrite, and the bainite and ferrite in the inner and outer edge layers are more than those in the middle layer.After that, the inner edge layer sample has the weakest hardenability, and the corresponding cooling rate was increased to 26℃/s, full martensite was obtained under this thermal simulation condition.This microstructure is not completely hardenable, and its hardenability is weaker than that of steel after homogenization heat treatment.

Microstructural changes of the outer edge layer
Table 3. Quantitative statistics of the initial microstructure of hot-forming steel.Figure 5 shows the OM images of the original austenite grain boundary which is corroded by picric acid.The austenite grain size of the steel after homogenization heat treatment is larger than that of the hot-forming steel, and the uniformity is better.The average grain size of the quarter layer is 6.04 μm, which is larger than 5.66 μm of the edge layer and smaller than 6.36 μm of the intermediate layer.The uniformity of each layer of the hot-forming steel is the same.The hardenability of hot-forming steel is weaker than that of steel after homogenization heat treatment.This is because the microstructure of hot-forming steel is ferrite and bainite, and austenite preferentially nucleates at the phase boundary.The non-uniformity of the multi-phase microstructure leads to more nucleation cores of austenite, and ferrite will promote the austenitization of bainite, and the austenite grains generated by final heating are relatively fine and uneven.Due to the low nucleation rate, the austenite grain size formed by thermal simulation heating is large and uniform.The larger the grain size is, the smaller the grain boundary area per unit volume is, and the fewer nucleation sites for diffusion-type transformation, thereby delaying the supercooled austenite transformation; the more uniform the microstructure is, the lower the nucleation rate of austenite transformation, and thereby increasing Supercooled austenite stability.Both factors shift the C curves to the right, reducing the critical cooling rate of steel and improving hardenability [20,21].
The hardenability of the inner and outer edge layers of hot-forming steel is weaker than that of the middle layer, and the hardenability of the quarter layer is weaker than that of the middle layer but stronger than that of the edge layer.This is due to the thinner bainite lath size of the inner and outer edge layers, resulting in more nucleation sites per unit area, thereby increasing the austenite nucleation rate.And the ferrite in the inner and outer edge layers is also small, and the heating transformation will produce structural heredity, these all lead to the formation of relatively small austenite grains, reducing the steel hardenability [22].And the energy dispersive spectroscopy results also show no significant difference in layered elements, so the grain size is the core reason for the different hardenability of hotformed steel after layered quenching.

Conclusion
Based on the results and discussions presented above, the conclusions are obtained as below: (1) Dynamic CCT critical quenching cooling rate (25℃/s) is greater than the static CCT critical quenching cooling rate (20℃/s).Deformation inhibits martensitic transformation and increases the critical quenching cooling rate, resulting in a decrease in hardenability.
(2) Microstructure of hot stamping steel is bainite and ferrite, and it is not uniform.After heating, the austenite grains are relatively fine and uneven.After homogenization heat treatment, the microstructure of the steel is uniform tempered sorbite, and the austenite grains generated after heating are large and relatively uniform.Therefore, the hardenability of hot-forming steel is weaker than homogenization heat treatment ones.
(3) Grain size of bainite lath and ferrite in hot forming steel is slightly larger in the middle layer, slightly smaller in the inner and outer edge layers, and the bainite lath size and ferrite grain size of the quarter layer are larger than those of the edge layer but smaller than those of the middle layer.These lead to the smaller austenite grain size formed by recrystallization in the inner and outer edge layers after heating and holding, and it weaker hardenability of the edge layer.
(4) The original microstructure of steel has a decisive influence on its hardenability.The performance and toughness of unhardened steel components are insufficient and even fail in advance in use.Therefore, systematic research in this area has an important guiding role in the production and processing of steel.The follow-up research should focus on the effect of the original microstructure on the impact behavior of steel after heat treatment.

Figure 3
Figure 3 Static (solid line) and dynamic (dotted line) CCT curves of test steel.By comparing the dynamic and static CCT curves, it can be concluded that 0.1℃/s to 1℃/s is the static CCT ferrite and pearlite phase zone, and the dynamic CCT transition temperature range shift to the upper left, the static CCT bainite phase zone extends from 0.3℃/s to 20℃/s, while the dynamic CCT phase zone extends to 25℃/s.The critical quenching cooling rate of static CCT is 20℃/s, while that of dynamic CCT is 25℃/s.The deformation makes the ferrite and pearlite start to phase transition temperature increase, the bainite phase transition zone expands, and the B s temperature increase.At the same time, the deformation increases the critical quenching cooling rate and reduces the M s point temperature.This is because the thermal deformation of austenite will increase the crystal defects and strengthen the austenite microstructure, so that the martensite non-diffusion shear mechanism is blocked, resulting in a decrease in hardenability[17,18].

Figure 5 .
Figure 5.The original austenite grain boundaries of steel and hot-forming steel after homogenization heat treatment.

Table 2 .
Static and dynamic (left/right) CCT phase transition start and end temperature.