2017 2nd International Conference on Applied Mechanics, Electronics and Mechatronics Engineering (AMEME 2017) ISBN: 978-1-60595-497-4

Effect of Different Heat Treatment Processes on Properties of P92 Steel

Peng-fei MA1, Qian XIE1, Zhi-fang PENG1 and Wen-kai XIAO1,* 1Wuhan University, School of Mechanical and Power, Engineering, Wuhan, China

*Corresponding author

Keywords: Heat treatment, Microstructure, Hardness, δ-ferrite.

Abstract. In this paper, P92 steel billets produced in a steel mill are chosen as experimental samples

and the effects of different heat treatment parameters on the microstructure and properties of P92 steel

are studied. The performance of the core and the outer circle of P92 steel samples were also

compared. Based on the test results of normalizing and tempering process, an optimized heat

treatment scheme for P92 steel is proposed.

Introduction

P92 steel has now become an ideal steel for ultra supercritical thermal power generation units [1-3].

A steel mill changed the content of some elements in P92 heat-resistant steel, improved smelting and

forging methods, and produced a new batch of P92 steel forging billets. In this paper, the P92 steel

produced by the steel mill is chosen as the research material, and the tests of different normalizing

temperature and different tempering time are carried out. The microstructure and hardness of the

treated samples were examined. By comparing and analyzing the experimental results, the reasons for

this phenomenon are investigated and the better heat treatment scheme is proposed.

Preparation of Experimental Materials

The raw material’s chemical composition is shown in Table 1. The specimens were sampled at the

center and the surface of the raw material by wire cutting, and the specimen size is 10mm * 15mm *

20mm.

Table 1. Chemical composition (mass fraction) of forging test piece for P92 steel.

Experiment Approach

This experiment used heat treatment equipment for BTF-1400 Beiyike 1400 °C vacuum tube furnace.

The normalizing temperature are 1060 °C and 1070 °C and the normalizing time is 1h. The tempering

temperature is 760 °C, and the hold time is 1h, 3h and 5h respectively. The cooling mode after

tempering is air cooling.

Sample Test Result and Analysis

Effect of Normalizing Temperature on Properties of P92 Steel

Microstructures of specimens near the outer circle as shown in Fig 1, Fig 2 and Fig 3. The lath

martensite structure can be observed as shown in Fig 2 and Fig 3. When the temperature increases, the

grain size and the lath size have grown, but not obvious. As shown in Fig 4 and Fig 5, compared to the

normalized microstructure, the tempering of the back plate decreases and the lath is not obvious

C S Si Mn P Ni Cr Mo Cu Al

0.11 0.002 0.18 0.45 0.013 0.25 8.91 0.50 0.04 0.005

W V Ti As Sn Pb Sb Bi Nb B

1.71 0.21 0.002 0.009 0.009 0.001 0.005 0.007 0.072 0.0021

289

before, and a small amount of carbide is precipitated. The change of grain size is not obvious. When

P92 steel is normalized, austenitizing occurs and then changes to lath martensite structure after

cooling. At this point, more C and N elements are dissolved in the martensite lattice gap.Two methods

of Brinell hardness and microhardness were used to test the original state of the sample. The test

results are shown in Table 2.

Table 2. Original hardness of sample.

Figure 6 shows that with the extension of tempering time, the normalizing temperature at 1070 °C

can reflect higher hardness than 1060 °C. As shown in Fig 7, solid solution strengthening makes the

hardness obviously increase, from about 200 HB of the original state to more than 400HB. It also can

HB HV

Surface 202 194

Core 182 196

Figure 1. Original sample. Figure 2. Normalizing at 1060 °C. Figure 3. Normalizing at 1070 °C. Figure 4. Tempering for 3h after

normalizing at 1060 °C. Figure 5. Tempering for 3h after

normalizing at 1070 °C.

Figure 6. The microhardness of the sample after

normalizing at different temperatures. Figure 7. The Brinell hardness of samples after normaling at

different temperatures. 290

be seen that when the normalizing temperature increases from 1060 °C to 1070 °C, the hardness of

P92 steel increases.

Influence of Tempering Time on Properties of P92 Steel

The core sample is tempered at 760 °C after normalizing at 1070 °C for 1 hour. As shown in Figure 8,

Fig 9, Fig 10, there are still more lath martensite in the tempering for 1h, and the lath is relatively

small when tempering for 3h and 5h. It is obvious that there is precipitation of carbide. After

tempering with P92 steel, the microstructure will occur during the tempering process, and lath

martensite will gradually change into tempered martensite with carbide precipitation. As shown in Fig

11, when the other parts are the same, the hardness of the core decreases at first and then increases

with the increase of time. At the early stage of tempering, the martensite recovery is not enough, and

the hardness of the matrix remains high. During the middle of tempering, the martensite recovered

sufficiently, resulting in a decrease in dislocation density. The precipitation of carbide gradually

decreases and the hardness of matrix decreases obviously, but the precipitation strengthening is not

obvious, which leads to the decrease of overall hardness. In the later stage of tempering, carbides are

fully dispersed and precipitated, and the strengthening of precipitation is obviously manifested. At the

same time, martensite matrix is still relatively stable, thus showing higher hardness [4-5].

Comparison of Properties between Core and Near Outer Circle of P92 Steel Forging Specimen

It can be seen from Fig 12 that the hardness of the P92 steel forging specimen near the outer circle is

the highest, while the core hardness is smaller. In actual production, because of the large size and

large thickness of the workpiece, the phenomenon of near surface and core unevenness may exist in

the forging process.

Figure 8. Temper for 1h. Figure 9. Temper for 3h. Figure 10. Temper for 5h.

Figure 11. The microhardness of the sample tempering at

different time.

Figure 12. The distribution of hardness with the distance

from the outer part of the specimen to the

center of the sample.

291

In order to study the effect of this uneven phenomenon on the performance, the samples were

sampled at the heart and near the outer circle for normalizing and tempering heat treatment. As shown

in Figure 13, Figure 14, the grain near the foreign aid is thinner and the heart is larger. This

phenomenon is due to the dynamic recrystallization degree of the near surface relative to the core in

the forging process, and the cooling rate is larger, which leads to the grain of the outer circle is smaller

than that of the heart. The grain near the outer circle is still smaller than that of the core, indicating that

the original microstructure is hereditary. As show in Fig 15, Fig 16, the microstructure of the outer

circle is still finer and the heart is thicker than that of the core after heat treatment.

Figure 16. The normalizing

temperature of the core is

1060 °C for 1h.

Figure 17. Tempering for 3h after

the outer is normalized.

Figure 18. Tempering for 3h

after the core is normalized.

As show in Figure 17, Figure 18, the microstructure near the outer circle is still finer and the heart is

thicker. A small amount of δ-ferrite was found in the microstructure of the heart. Because the content

of Cr in the sample is as high as 8.91%, if the composition is not properly controlled, it may lead to the

presence of delta ferrite in the matrix during casting, forging or heat treatment.The mechanical

properties of δ-ferrite in P92 steel is extremely harmful, which leads to the notch sensitivity greatly

reduce the impact toughness of the material[6-7]. The strength, hardness and toughness of the metal

decrease because of the aggregation of the carbides on the interface between the ferrite and the

Figure 13. The `original state near

the outer.

Figure 14. Core primitive state. Figure 15. The normalizing temperature of

the outer is 1060°C for 1h.

Figure 19. Microhardness of the sample. Figure 20. Microhardness of the sample.

292

surrounding martensite matrix. Because of the difference in solubility between the W and the

martensitic group, Laves phase will preferentially form in the delta iron region, and the δ-ferrite will

also promote the growth of the Laves phase, which will decrease the fatigue life of P92 steel under

high temperature service[8-9].

As shown in Figure 19, Figure 20, the hardness of the near outer circle is not significantly different

from that of the core specimen. With the extension of tempering time, the hardness near the outer

circle can be higher than that of the center. In general, the hardness of specimens near the outer circle

is higher than that of the core specimens. After the normalizing and tempering , the presence of

δ-ferrite in the core is one of the causes of the hardness of the heart lower than that of the outer circle.

However, it is not possible to determine whether there is delta iron in the original and normalized

states and to reduce the hardness of the core sample.

Conclusions

(1) Hardness test data show that 1070 °C normalization can achieve higher hardness. With the

extension of tempering time, the normalizing temperature at 1070 °C can reflect higher hardness than

1060 °C.

(2) When the other parts are the same, the hardness of the core decreases at first and then increases

with the increase of tempering time.This is related to the reaction of the microstructure during the

tempering process and the precipitation of carbides.

(3) After the normalizing and tempering , the presence of δ-ferrite in the core is one of the causes of

the hardness of the heart lower than that of the outer circle.

References

[1] Shen Q, Liu H G. Application of New Type Heat-resistant Steel T/P92 and T/P122 in

Ultra-supercritical Unit and Quality Control [J]. Electric Power Construction, 2010.

[2] Shen Q, Liu H G. Application of New Type Heat-resistant Steel T/P92 and T/P122 in

Ultra-supercritical Unit and Quality Control [J]. Electric Power Construction, 2010.

[3] Yin Z, Hui C, Liu H G. Performance of heat-resistant steel P92 used in 1000 MW ultra

supercritical unit after long-term service at high temperature[J]. Transactions of Materials & Heat

Treatment, 2012, 33(11):105-110.

[4] Baek J W, Nam S W, Kong B O, et al. Key Engineering Materials, 2005, 297-300(22):463-470.

[5] Jing H M, Wang B, Xiong D S, et al. Advanced Materials Research, 2011,

194-196(12):1280-1286.

[6] Baek J W, Nam S W, Kong B O, et al. Key Engineering Materials, 2005, 297-300(22):463-470.

[7] Shi R X, Liu Z D. Iron & Steel, 2012.

[8] Baek J W, Nam S W, Kong B O, et al. The Effect of Delta-Ferrite in P92 Steel on the Formation of

Laves Phase and Cavities for the Reduction of Low Cycle Fatigue and Creep-Fatigue Life [J]. Key

Engineering Materials, 2005, 297-300(22):463-470.

[9] Shi R X, Liu Z D. Growth Behavior of Laves Phase of δ-Ferrite in P92 Steels [J]. Iron & Steel,

2012.

293