J. Cent. South Univ. (2012) 19: 1189-1195

DOI: 10.1007/s11771-012-1127-x

Effects of pre-precipitation of Cr₂N on microstructures and properties of high nitrogen stainless steel

LI Jing-yuan(李静媛)¹, LIU Hui-nan(刘慧男)², HUANG Pei-wu(黄佩武)³

1. School of Materials Science and Engineering, University of Science and Technology Beijing,Beijing 100083, China;

2. Materials Science and Engineering Program, University of California, Riverside 92521, USA;

3. School of Metallurgical Engineering, Hunan University of Technology, Zhuzhou 412007, China

? Central South University Press and Springer-Verlag Berlin Heidelberg 2012

Abstract: Aging precipitation and solid solution heat treatment were carried out on three steels which have chromium content of 18%, manganese content of 12%, 15%, 18%, and nitrogen content of 0.43%, 0.53%, 0.67%, respectively. The mechanisms of precipitation and solid solution of high nitrogen austenitic stainless steel were studied using the scanning electron microscopy, transmission electron microscopy, electron probe micro analysis and mechanical testing. The results show that, Cr₂N is the primary precipitate in the tested stainless steels instead of Cr₂₃C₆. Cr₂N nucleates at austenitic grain boundaries and grows towards inner grains with a lamellar morphology. By means of pre-precipitation of Cr₂N at 800 ℃, the microstructure of the steels at solid solution state can be refined, thus improving the strength and plasticity. After the proposed treatment, the tensile strength, the proof strength and the elongation of the tested steel reach 881 MPa, 542 MPa and 54%, respectively.

Key words: Cr₂N; precipitation; refinement; high nitrogen stainless steel (HNSS)

1 Introduction

High nitrogen stainless steels (HNSS) attracted much interest due to their low nickel content, high strength, high toughness, and excellent resistance to pitting and crevice corrosion [1-3]. However, many issues presented in the production process, such as the low nitrogen content, poor deformability, and low ductility, require thorough research on the microstructure, phase transformation and precipitation mechanisms during hot forming process and heat treatment [4-7].

Research on the precipitates is one of the most important fields [8-9]. Many studies have shown that chromium nitride (Cr₂N) is the primary precipitate in high nitrogen austenitic stainless steels. However, the morphologies of Cr₂N have not yet determined consistently. SCHAEVE et al [10] studied high nitrogen Cr-Mn austenitic stainless steel and reported that Cr₂N precipitated along the grain boundaries intermittently, and then grew into the crystals with cellular morphology. LEE et al [11] only found some granular, rod-like precipitates and did not find any lamellar nitrides in Fe-22Cr-21Ni-6Mo-0.3N austenitic stainless steel even after aging at 900 ℃ for 168 h.

Additionally, most of the researches were focused on the precipitation of Cr₂N, but rarely mentioned its solid solution and its influence on the refinement of grain of HNSS. BLINOV et al [12] reported the effect of heat treatment on the fracture toughness of hot-rolled 0.4Cr20Ni6Mn11Mo2N0.5 steel and analyzed the formation processing of quasi-cleavage; however, the behavior of precipitation and solid solution was not stated.

In this work, a high-nitrogen austenitic stainless steel is focused on, which contains Cr 18%, Mn 12%-18%, N 0.43%-0.67%, C less than 0.08% and Ni free (mass fraction). The morphology and nasal nip temperature of Cr₂N precipitate, and the effects of pre-precipitated Cr₂N on the microstructural evolution and mechanical properties of HNSS during solid solution treatment are reported.

2 Experimental

Three different alloy ingots of high nitrogen austenitic stainless steel were prepared. They were made in an experimental induction furnace with chemical compositions listed in Table 1. The size of the ingot was 90 mm×90 mm×80 mm.

Table 1 Chemical compositions of tested steels (mass  fraction, %)          

Hot rolling was carried out in an experimental 4-high strip rolling mill with the roll width of 350 mm. Ingots were heated and kept at 1 280 ℃ for 2 h and rolled into plates through 9 passes into a final thickness of 5.1 mm. The finishing rolling temperature was 950 ℃ and the plates were subsequently cooled in water. The plates were cut into samples for heat treatment experiments, microstructure investigation and mechanical testing.

The samples were heated up to a temperature of 600-1 100 ℃ with an interval of 50 ℃, and were kept at that temperature for 0.5, 2 and 4 h followed by water quenching to room temperature. The precipitates and microstructural change were investigated for each condition using optical microscope (OM).

The chemical composition of the precipitates in the samples aged at 800 ℃ for 4 h was analyzed using electron probe micro analyzer (EPMA). Transmission electron microscopy (TEM) was employed to determine the crystal structure of precipitates.

In order to study the effects of various heat treatments on mechanical properties of high nitrogen austenitic stainless steel, solid solution treatment was carried out at 1 050 ℃ and 1 100 ℃, respectively, after aging treatment at 800 ℃. The effects of precipitates on microstructure and properties were analyzed by comparing the results of solid solution treatment with or without aging at 800 ℃.

3 Results and discussion

3.1 Microstructures and properties of rolling plates

Microstructures of the tested steel ingots were greatly refined after 9-pass hot rolling. Figure 1 shows the optical microscopy image of longitudinal section of steel plate B. Since the finishing rolling temperature was higher than recrystallization temperature, a fine microstructure of equiaxial recrystallized crystal with an average grain size of 30 μm was obtained. Mechanical properties of steel plate were tensile tested at room temperature, as presented in Table 2. The ultimate tensile strength (i.e. R_m) and the proof strength at 0.2% non-proportional extension (i.e. R_p0.2) increased while elongation (i.e. δ) decreased as the Mn and N contents increased. R_m, R_p0.2 and δ of steel plate A composed of Fe-18%Cr-18%Mn-0.67%N were 1 160 MPa, 1 070 MPa and 23%, respectively.

Fig. 1 Microstructure of Fe-18%Cr-15%Mn-0.53%N steel plate

Table 2 Mechanical properties of tested steel plates

3.2 Precipitation of Cr₂N at elevated temperatures

3.2.1 Thermodynamic analysis of precipitation process

Thermodynamic phase diagram is the foundation of analyses of phase transformation and precipitation versus temperature, and thus, it is essential for developing the heat treatment processing. Fe-N pseudo binary phase diagrams of the tested steel were studied based on the Thermo-calc phase diagram analysis system. The results showed that, the temperature interval for Cr₂N precipitation became narrower when the Mn content increased. It is due to the role of Mn in expanding austenitic phase region and facilitating the solution of N in the γ-Fe matrix. Figure 2 shows the Fe-N pseudo binary phase diagram of Fe-18%Cr-18%Mn alloy. It can be seen that σ phase, M₂₃C₆ and Cr₂N precipitate as the temperature increases, but the precipitates are re-dissolved into γ-Fe matrix when the temperature increases above 1 000 ℃. So, it is suggested that the temperature of solid solution should be higher than     1 000 ℃. As N content increases, the solid solution temperature of Cr₂N phase increases and that of σ and M₂₃C₆ phase decreases. This indicates that, as N content increases, Cr₂N becomes more stable while the precipitation of σ phase and M₂₃C₆ is inhibited.

3.2.2 Morphology of precipitates and nasal nip temperature of precipitation

According to the analysis of thermodynamic phase diagrams and further consideration on kinetics conditions, the aging heat treatment temperatures were determined to be 600-1 000 ℃ with each interval of 50 ℃. The holding periods were 0.5, 2 and 4 h in order to obtain various amounts of precipitates.

Fig. 2 Fe-N pseudo ternary phase diagram of Fe-18%Cr- 18%Mn alloy

The aging experiment results indicated that a few granular and short rod-like precipitates started to appear in austenite grain boundaries discontinuously when the steels were treated at 700 ℃. As a result, the boundaries became serrated and not straight. When treated at 800 ℃ for 0.5 h, precipitates were still mainly concentrated at the grain boundaries, but began to protrude towards the inner grain. Figure 3(a) shows the microstructure of plate A treated at 800 ℃ for 0.5 h. The precipitates increased obviously and began to replace the austenite grain when the holding time was up to 2 h. The precipitates continued to grow and increase, and their area percent was about 30% when the holding time was extended to  4 h. However, when the aging temperature rose to 900 ℃, only some fine particles but no protruding precipitates were observed in the grains and grain boundaries even after being held for 4 h. No precipitates were found at higher temperatures. This indicates that the nasal nip temperature of precipitation should be 800 ℃.

As shown in Fig. 3(b), the precipitates nucleated at grain boundaries and protruded towards inner grains. The grown precipitates had a lamellar morphology and were very similar to pearlite. The thickness of lamellae was 100-200 nm with a space width of 1-3 μm. The lamellar precipitates were brittle and had weak adhesion with matrix. As a result, they could fall off matrix easily during grinding or eroding in a strong boundary erosion agent. Therefore, it can be deduced that this kind of precipitate deteriorates the plasticity and toughness, and can be considered as harmful precipitate.

Microscopic component analysis can directly determine the change of element content in a micro-field using EPMA. The line and area scanning of microzone, which is a typical method in EPMA, can tell us the chemical compositions of precipitates distinctly. The area scanning of a microzone including the lamellar precipitates was carried out and the results are shown in Fig. 4. The analyzed elements included C, N, Mn and Cr. Cr and N contents in the precipitate layer were obviously higher than those in layer space. However, due to the fact that the diffusion of N was significantly higher than that of other elements, N content in layer space was equivalent to the matrix while Cr content in layer space was much lower than that in matrix. Thus, it was speculated that the lamellar precipitate was chromium nitride.

Fig. 3 Microstructures of plate A after aging at 800 ℃ for 0.5 h obtained by EPMA (a) and for 4 h obtained by SEM (b)

The same precipitation and nasal nip temperature were observed in A, B and C plates, which indicated that the precipitates were the same in high manganese and high nitrogen austenitic stainless steels. The quantity of precipitates decreased as nitrogen content decreased. The average amounts of lamellar precipitates in A, B and C plates were 30%, 20% and 5%, respectively, after heat treated at 800 ℃ for 4 h.

3.2.3 Type of precipitate

The crystal structure of material can be determined using TEM. Figure 5 shows the typical morphology and electron diffraction of the precipitates in steel plate A aged at 800 ℃ for 4 h. It can be seen that the thickness of lamellae was 150 nm with a lamellar space of 1 μm. It was determined that the precipitates had hexagonal close-packed structure (HCP) with the lattice constant a=0.480 5 nm and c=0.447 9 nm through the diffraction spot calibration. The lamellar precipitate was determined to be Cr₂N by combining TEM and EPMA analyses.

Fig. 4 EPMA area scanning of precipitates in steel plate A after aging at 800 ℃ for 4 h

Fig. 5 TEM micrograph (a) and diffraction pattern (b) of Cr₂N precipitate in plate A treated at 800 ℃ for 4 h

Other precipitates except Cr₂N were not found using OM, SEM and EPMA. Only a few cubic structural M₂₃C₆ particles with an average diameter of 100 nm were observed under TEM in the samples prepared by carbon extraction replica technique. It was suggested that Cr₂N was the main precipitate and the precipitation of M₂₃C₆ and σ phase was greatly inhibited. The lattice structure of nitride matched with the matrix more than that of carbide and the cohesion of nitrides with matrix was stronger than that of carbides. At the same time, the nucleation of nitrides was easier than that of carbides, as a result of higher diffusion coefficient of N than that of C and subsequent preferential N segregation near grain boundaries [13-14]. Additionally, owing to the solution of N in austenite, diffusion coefficient of Cr decreased, lattice parameters of M₂₃C₆ reduced and interfacial dislocation of M₂₃C₆ increased [15-16]. The phenomena described above reduced the driving force for Cr₂₃C₆ growth, impeded its nucleation and growth and consequently limited the precipitation. A small amount of Cr₂₃C₆ could precipitate on the grain boundary, when a lot of Cr₂N precipitated. Subsequently, the solid solubility of supersaturated N in the matrix decreased and the driving force for the precipitation of carbides increased.

3.3 Growth of austenitic grain and change of mechanical properties during solid solution

In order to eliminate work hardening, improve plasticity for subsequent cold working, the solid solution treatment was usually required for hot rolled austenite stainless steel plates. Due to the solid solution treatment, distorted lattice was recovered, elongated grains recrystallized, internal stress was relieved and the strength decreased. But at the same time, the austenite grains tended to coarsen at elevated temperature, which decreased ductility, especially for steels with high contents of Mn and N. When Mn and N contents are high, the binding strength between Fe atoms decreases, thus facilitating Fe atom diffusion and rapid grain growth of austenite.

Figure 6 shows optical images of steel A after solid solution treatment at 950-1 100 ℃ for 4 h. When the grains were separated by annealing twins, they appeared to have a smaller average diameter of 20 μm. Therefore, no growth was observed as compared with that at rolling state. The grain was coarsened to over   70 μm at 1 000 ℃ abruptly and it grew further with increasing the temperature. According to the phase diagram described above, Cr₂N still existed in austenitic matrix below 1 000 ℃ and it inhibited the growth of austenite crystal. When the temperature was up to 1 100 ℃, the average diameter of austenite grain was close to 150 μm. It could be deduced that the coarsening temperature of Fe-18%Cr- 18%Mn-0.67%N steel was 1 000 ℃. However, in order to make the precipitates dissolve totally and obtain the more homogenous austenitic single phase and higher deformability, it is suggested that the solid solution temperature should be higher than 1 050 ℃.

When the austenite grain grows by grain boundary movement, the stacking sequence of atomic layer will occasionally dislocate on the (111) crystal face at the corner of grain boundary. The stress concentration at the grain boundary was resulted from the plastic deformation, providing necessary energy for the initiation of twins. Thus, a coherent twin boundary originated in dislocated place and then twins formed at the corner of grain boundary. The twins grew across the grain and became bigger as the grain grew. The grain boundary became less apparent and eventually disappeared with temperature increasing and holding time extending, so that the twins boundary became the new grain boundary. This may explain why very straight austenitic grain boundary appeared after solid solution treatment.

Solid solution treatment can improve the deformability of alloys. As shown in Fig. 7, the strength of steel plate A decreased and its elongation increased as the solid solution temperature increased. It can be seen that the R_m, R_p0.2 and δ changed from 1 160 MPa,      1 070 MPa and 23% in rolling state to 815 MPa, 500 MPa and 43%, respectively, after treated at 1 050 ℃ for 4 h. The strength decreased to a great extent, especially R_p0.2, which dropped by half. However, the elongation only increased to 43%, which is less than 50%. The grain coarsening at high temperature may contribute to this phenomenon. As mentioned above, high contents of manganese and nitrogen caused the inherent coarsening of austenite grain.

Fig. 6 Microstructures of steel plate A after solid solution treatment at different temperatures for 4 h: (a) 950 ℃, (b) 1 000 ℃,     (c) 1 050 ℃, (d) 1 100 ℃

Fig. 7 Mechanical properties of steel plate A after solid solution

3.4 Effect of pre-precipitation of Cr₂N on refinement of solid solution microstructure

Since the grain boundary is relatively unstable, it tends to reduce the free energy of the alloy system and transform to a stable state. This results in the fact that the surface area of grain boundary reduces and the grain grows up. The growth of austenite grain is driven by the grain boundary energy and is a thermal activation process controlled by the diffusion of alloy atoms. This process is accomplished through the movement and diffusion of atom at grain boundary. The precipitates provide the resistance to grain boundary migration and can inhibit the grain growth due to the pinning effect at the grain boundary. The more the precipitate phase, the stronger the pinning effect is [17]. Therefore, the elements which are easy to form refractory carbides, such as Ti, V, and Nb, are usually considered as the elements preventing austenitic grain growth.

In this work, a novel treatment of combining aging at 800 ℃ for Cr₂N pre-precipitation with subsequent solid solution treatment was developed and characterized for the purpose of refining the austenite grain of the high manganese and high nitrogen steels. Figure 8 shows the optical microscopy of rolling plate A solid solution treated at 1 050 ℃ and 1 100 ℃ for 4 h after aging pre-precipitation at 800 ℃ for 4 h. It could be seen that the grain sizes at 1 050 ℃ and 1 100 ℃ were less than 50 μm and 100 μm, respectively, which were remarkably smaller than those without pre-precipitation aging treatment. It should be due to the inhibition effects of lamellar Cr₂N on the migration of austenite grain boundary.

On the other hand, even the precipitate dissolved, it was still rich in chromium at the original precipitates place for a long time, which could be considered as chromium micro-segregation. It was resulted from the difficult diffusion of chromium in austenite. It is well-known that activation energy for diffusion of chromium is high up to 335 kJ/mol in austenite matrix, which is much higher than other alloy elements such as Ni, Mn, C, and N. Therefore, the micro-segregation of chromium was hard to be eliminated in this work. The inhomogenous distribution of substitutional atoms would affect the diffusion of Fe atoms, and thereby retard the migration of grain boundary and growth of austenite grain. It can be seen that the higher the activation energy for self diffusion of Fe atoms, the finer the grain can be obtained from the following classical mechanics formula [18]:

where D_A is the average diameter of growing austenite; K is a constant; τ is the holding time; Q is activation energy for self diffusion of Fe; T is temperature.

Fig. 8 Microstructures of steel plate A after aging at 800 ℃ for 4 h followed by solid solution treatment at 1 050 ℃ (a) and   1 100 ℃

The microstructure refinement certainly leads to the improvement of property. The strength increases mainly due to the elevation of deformation resistance at grain boundary, which is resulted from the increasing of the number of grains and the enlargement of the total area of grain boundary. In addition, grain refinement reduces the dislocation density at boundaries and decreases the local stress concentration. Therefore, it is not easy for the dislocations to cross a grain boundary to the adjacent grains and consequently the yield strength (or proof strength of 0.2% non-proportional extension) increases. At the same time, small grain size can improve the compatibility of deformation so that the plasticity and toughness can be improved. As listed in Table 3, both strength and elongation of the tested high nitrogen austenite steel plates were improved when they were aged at 800 ℃ for 4 h and subsequent solid solution treated at 1 050 ℃ for 4 h.

Table 3 Mechanical properties of tested plates after various heat treatments

R_m, R_p0.2 and δ of steel plate A, increased by 45 MPa, 44 MPa and 11%, respectively. This is mainly because steel A contains the highest nitrogen content of 0.67%, thus it precipitates the most Cr₂N at 800 ℃ and provides the strongest resistance effect on the grain boundary migration and grain coarsening.

4 Conclusions

1) The ultimate tensile strength, the proof strength at 0.2% non-proportional extension and the elongation of Fe-18%Cr-18%Mn-0.67%N steel at rolling state are    1 160 MPa, 1 070 MPa and 23%, respectively.

2) The precipitation of M₂₃C₆ and σ phase is greatly inhibited in the high manganese and high nitrogen austenitic stainless steels. Additionally, Cr₂N, the main precipitate, nucleates at austenitic grain boundary and grows towards inner grains with a lamellar morphology. The thickness of lamellae is 100-200 nm with a space width of 1-3 μm. The nasal nip temperature of precipitation of Cr₂N is 800 ℃. The quantity of Cr₂N increases as nitrogen content increases.

3) Due to the high contents of Mn and N, the austenite grains tend to coarsen at elevated temperature. The grain coarsening temperature of Cr-Mn-N austenitic stainless steel is 1 000 ℃.

4) By means of pre-precipitation of Cr₂N at 800 ℃, the microstructure at solid solution state can be refined, and the strength and plasticity can be improved at the same time. In addition, as the nitrogen content increases, the grain becomes more refined. After this processing, R_m, R_p0.2 and δ of Fe-18%Cr-18%Mn-0.67%N steel plate change to 881 MPa, 542 MPa and 54%, respectively.

References

[1] LIU Zhi-hong, MA Xiao-bo, ZHU De-qing, LI Yu-hu, LI Qi-hou. Preparation of ferronickel from laterite ore in reduction smelting process [J]. Journal of Central South University: Science and Technolog, 2011, 42(10): 2905-2910. (in Chinese)

[2] ROKANOPOULOU A,PAPADIMITRIOU G D. Production of high nitrogen surfaces on 2205 duplex stainless steel substrate using the PTA technique [J]. Materials Science and Technology, 2011, 27(9): 1391-1398.

[3] YANG Ke, REN Yi-bin. Nickel-free austenitic stainless steels for medical applications [J]. Science and Technology of Advanced Materials, 2010, 11(1): 1-13.

[4] WAN Peng, REN Yi-bin, ZHENG Bing-chun, YANG Ke. Analysis of magnetism in high nitrogen austenitic stainless steel and its elimination by high temperature gas nitriding [J]. J Mater Sci Tech., 2011, 27(12): 1139-1142.

[5] PARK W I, JUNG S M, SASAKI Y. Fabrication of ultra high nitrogen austenitic stainless steel by NH₃ solution nitriding [J]. ISIJ International, 2010, 50(11): 1546-1551.

[6] HONG C M, SHI J, SHENG LY, CAO W C, HUI W J, DONG H. Effects of hot-working parameters on microstructural evolution of high nitrogen austenitic stainless steel [J]. Materials & Design, 2011, 32(7): 3711-3717.

[7] LI Ji-chao, CHU Shao-jun. Effect of sampling methods on N detection in high nitrogen steel [J]. Steel Research International, 2011, 82(11): 1273-1277.

[8] BAI Guang-hai, LI Jin-shan, HU Rui. Effect of thermal exposure on the stability of carbides in Ni-Cr-W based superalloy [J]. Materials Science and Engineering A, 2011, 528: 2339-2344.

[9] CAROSI A, AMATI M, GREGORATTI L, KACIULIS S, MEZZI A, MONTANARI R, ROVATTI L, UCCIARDELLO N. Heating modification of an austenitic steel with high-nitrogen content [J]. Surface and Interface Analysis, 2010, 42(6/7): 726-729.

[10] van der SCHAEVE F, TAILLARD R, FOCT J. Discontinuous precipitation of Cr₂N in a high nitrogen, chromium-manganese austenitic stainless steel [J]. Materials Science, 1995, 30: 6035-6045.

[11] LEE T H, KIM S J, JUNG Y C. Crystallographic details of precipitates in Fe-22Cr-21Ni-6Mo-(N) superaustenitic stainless steels aged at 900 ℃ [J]. Metallurgical and Materials Transactions A, 2000, 31(7): 1713-1718.

[12] BLINOV V M, BOTVINA L P, TYUTIN M P, BLINOV E V, MUSHNIKOVA S Y and ZHARKOVA N A. Effect of heat treatment on the fracture toughness of hot-rolled corrosion-resistant austenitic high-nitrogen 0.4Cr20Ni6Mn11Mo2N0.5 steel [J]. Russian Metallurgy (Metally), 2011(9): 831-836.

[13] BRIANT C L, HALL E L. A comparison between grain boundary chromium depletion in austenitic stainless steel and corrosion in the modified Strauss test [J]. Corrosion, 1986, 42(9): 522-531.

[14] BRIANT C L. Nitrogen segregation to grain boundaries in austenitic stainless steel [J]. Scripta Metallurgica, 1987, 21(1): 71-74.

[15] BABA H, KATADA Y. Effect of nitrogen on crevice corrosion in austenitic stainless steel [J]. Corrosion Science, 2006, 48(9): 2510-2524.

[16] BABA H, KODAMA T, KATADA Y. Role of nitrogen on the corrosion behavior of austenitic stainless steel [J]. Corrosion Science, 2002, 44(1): 2393-2407.

[17] HU De-lin, LIU Zhi-en. Research on the migration of austenite grain boundary [J]. Acta Metallurgica Sinica, 1986, 22(4): 58-62 (in Chinese)

[18] WANG Kun-lin. Fundamentals of materials engineering [M]. Beijing: Tsinghua University Press, 2009: 116-117. (in Chinese)

(Edited by YANG Bing)

Foundation item: Project(50974014) supported by the National Natural Science Foundation of China

Received date: 2011-06-26; Accepted date: 2011-10-09

Corresponding author: LI Jing-yuan, Associate Professor; Tel: 86-10-82376939; E-mail: jerranlee@yahoo.com.cn