Effect of Tungsten on Mechanical Properties of Reduced Activation Ferritic-Martensitic Steel Subjected to Intercritical Heat Treatment Academic research paper on "Materials engineering"

Available online at www.sciencedirect.com

SciVerse ScienceDirect

Procedía Engineering 55 (2013) 277 - 283

Procedía Engineering

www.elsevier.com/locate/procedia

6th International Conference on Creep, Fatigue and Creep-Fatigue Interaction [CF-6]

Effect of Tungsten on Mechanical Properties of Reduced Activation Ferritic-Martensitic Steel Subjected to Intercritical

Heat Treatment

C.S.Sasmala*, K.S.Chandravathib, M.Nandagopalb, S.Panneer Selvib, P.Parameswaranb, K.Lahab, M. D.Mathewb, T.Jayakumarb, E.Rajendra Kumara

aInstitute for Plasma Research, Gandhinagar - 382428, India bIndira Gandhi Centre for Atomic Research, Kalpakkam - 603102, India

Abstract

Fusion welded joint of RAFM steel as in other ferritic steels possesses lower creep rupture strength than the base steel. The failure in the joint is associated with the soft intercritical region of heat affected zone, which is subjected to peak temperatures in between Ac! and Ac3 during weld thermal cycle. In this investigation effort has been initiated to understand the effect of tungsten in softening the 9Cr-W RAFM steels on intercritical temperature exposure. Metallographic investigations were carried out on the heat-treated steels by optical and transmission electron microscopes. The steels at all heat treated conditions had tempered martensitic structure. Refinement of prior austenitic grain of the steels was noticed on soaking in the intercritical temperature range. The grain size of the steels increased with soaking at temperatures well above the Ac3 transformation temperature. TEM investigation revealed that the martensitic laths were decorated with M23C6 type of carbides and the presence of MX type of (Ta,V)C carbides inside the laths. Hardness, tensile and creep tests were carried on the steels to elucidate the effects of soaking at temperatures in and around Ac! and Ac3 transformation temperatures. All the steels suffered reduction in mechanical strength after soaking at temperatures in the intercritical temperature range. Different softening tendency of the RAFM steel having different tungsten content on intercritical annealing has been explained based on the microstructural investigation.

© 2013 The Authors. Published ty Elsevier Lte.

Selection and peer-review under responsibility of the Indira Gandhi Centre for Atomic Research. Keywords: RAFM; HAZ; creep; tensile; precipitates

1. Introduction

Materials development is a key issue for the realization of future fusion reactor. Reduced activation 9Cr-ferritic-martensitic steels are being presently considered as main candidate materials for the first wall

Corresponding author: Email address: cssasmal@gmail.com

1877-7058 © 2013 The Authors. Published by Elsevier Ltd.

Selection and peer-review under responsibility of the Indira Gandhi Centre for Atomic Research. doi: 10.1016/j.proeng.2013.03.254

components of future fusion reactor. Ferritic-martensitic steels have been considered because of their inherent void swelling resistance and better thermo-physical and thermo-mechanical properties compare to those of the austenitic stainless steels. However, they suffer from reduction in toughness induced by low temperature neutron irradiation, characterized by a shift in Ductile to Brittle Transition Temperature (DBTT) to higher temperature. These steels have chemical compositions similar to that of modified 9Cr-1Mo (9Cr-1Mo-0.1C-0.22V-0.06Nb) steel with the difference that the Mo and Nb being substituted with W and Ta respectively from residual radioactive point of view [1]. The presence of highly residual radioactive elements (Ni, Co, Cu, Al and Si etc.) is also kept as low as possible levels. International efforts to develop RAFM steel have focused on varying tungsten in the range 1 to 2 wt. % and tantalum in the range 0.02 to 0.18 wt. %. Tungsten addition increases creep rupture strength but decreases toughness properties. Tantalum in the RAFM steel plays an important role in lowering DBTT through its effect on prior-austenitic grain size refinement. However, higher tantalum decreases the weldability [2-5].

Being a ferritic steel, the microstructure of the RAFM steel depends on the normalizing and tempering heat treatments. A variety of microstructure having appreciably different mechanical properties can be developed in the steel depending upon the normalizing temperature due to phase transformation. Heat Affected Zone (HAZ) of fusion welded joint of ferritic steel is a typical example where a variety of microstructures developed because of the exposure to different peak temperatures ranging from below Ac1 to above Ac3 and Ac4 transformation temperatures during weld thermal cycle. A knowledge of the microstructures developed in the RAFM steel on normalization is required to estimate the mechanical property of the steel especially its weld

In this work, correlation between microstructure and mechanical properties of RAFM steel having different tungsten content has been investigated. Different microstructures in the steels have been produced on subjecting the steels different normalization at temperatures in the range below Ac1 to above Ac3 transformation temperatures.

2. Experimental

Three heats of RAFM steels having different tungsten content in the range 1 - 2 wt. % have been produced in collaboration with MIDHANI, Hyderabad. The steels were subjected to final normalizing (980oC for 30 minutes) and tempering (760oC for 90 minutes) heat treatments and delivered in the form of plate with dimension 12mm x 500mm x 1000mm. The chemical compositions of the steels are given in the Table 1. The steels are designated as 9Cr-1W-0.06Ta, 9Cr-1.4W-0.06Ta and 9Cr-2W-0.06Ta based on their tungsten content.

Samples of dimensions around 10 x 10 x 10 mm from each steel were soaked for 5 minutes at various peak temperatures in the range of 750oC to 1200oC followed by oil quenching. The temperature range was below Ac1 temperature (830oC) to well above the Ac3 transformation temperature ( 875oC) of the RAFM steel. The quenched steel samples were subsequently tempered at 760oC for 1 hour. Specimens in the as-quenched and tempered conditions were metallographically prepared for microstructural and hardness investigations. An immersion etching in Villela's reagent (1gram picric acid and 5 ml hydrochloric acid in 95 ml ethanol) was found adequate to reveal the microstructure of the RAFM steels. Vickers hardness measurement was carried out at room temperature under the load of 10 kg. The prior austenite grain size of the heat treated steels was measured by linear intercept method employing optical and scanning electron microscopes.

Selected heat treatments with normalizing temperatures in between below Ac1 to above Ac3 and subsequent tempering at 760 oC for 1 hour were performed on the bigger sample of the steels for carrying out tensile and creep tests. Tensile tests have been carried out at room temperature at a nominal strain rate 3x10"4s_1. Tensile tests at 550 oC were carried out at the same strain rate. Creep tests have been carried out with stress 240 MPa at 550 oC.

Table.1. Chemical composition (wt. %) of three heats of 9Cr-W-0.06Ta RAFM steel.

Element 9Cr-1.0W-0.06Ta 9Cr-1.4W-0.06Ta 9Cr-2.0W-0.06Ta

Cr 9.05 9.03 8.99

C 0.08 0.12 0.12

Mn 0.56 0.56 0.65

V 0.23 0.24 0.24

W 1.0 1.38 2.06

Ta 0.06 0.06 0.06

N 0.02 0.03 0.02

O 0.002 0.002 0.002

P 0.002 < 0.002 0.002

S 0.001 <0.001 0.001

B 0.0005 <0.0005 <0.0005

Ti 0.002 <0.005 <0.005

Nb 0.003 <0.001 <0.001

Mo 0.009 <0.002 <0.002

Ni 0.007 0.005 0.004

Cu 0.005 0.002 0.002

Al 0.003 0.003 0.003

Si 0.05 0.06 0.06

Co 0.004 <0.005 0.005

As+Sn+Sb <0.002 <0.004 <0.001

3. Results and discussion

3.1. Microstructure and hardness of the steel

The 9Cr-ferritic steels are generally used in normalized and tempered condition. It has tempered martensitic structure. Typical optical microstructures of the steel soaked for 5 minutes at a temperature in the range of 750 -1150 oC are shown in Fig. 1 for the 9Cr-1W-0.06Ta steel. Similar microstructures were obtained for the other two steels. The prior austenitic grain size of the steel was found to vary appreciably with the soaking temperature Fig. 2. Figures 3 displays the variations in hardness of the steels with soaking temperature in the as-quenched and tempered conditions respectively. Soaking the steel at temperatures below Ac1 (830oC) for 5 minutes could cause a little change in initial microstructure (Fig.1(a)) and hardness of the steel (Fig.3). Intercritical (between Ac1 (830oC) and Ac3 (875oC)) soaking resulted in decrease of the prior austenitic grain size (Fig. 1(b) and Fig. 2) and increase in the hardness of the as-quenched steel (Fig. 3(a)). Larger amount of freshly formed martensite derived from the transformation of austenite-developed during intercritical soaking and increased the hardness of the steel. Soaking the steel above Ac3 temperature resulted in coarsening the prior austenitic grain (Fig. 1(c)). Both the grain size and the hardness increased on soaking at temperatures beyond Ac3. With the increase in soaking temperature above Ac3, prior existing alloy carbonitrides dissolved along with the austenitization. The dissolution of carbonitrides increased with the increase in soaking temperature.

This would enrich the alloy content in solution of the austenite. On cooling, the austenite would transform to finer martensite due to depression of Ms temperature because of solute enrichment, leading to increase in hardness with soaking temperature above Ac3[6]. The prior austenitic grain size increased progressively with soaking at temperatures from close to Ac3 upto around 1050 oC followed by a rapid increase. Persistence of the prior-existing carbonitrides at soaking temperatures below 1050 oC impeded grain growth of the austenite. The impeding action was lost due to the complete dissolution of alloy carbonitrides at temperatures beyond 1050 oC and the grain size increased with the increase in soaking temperature. The increase in tungsten content appeared to have no appreciable effect on the grain size of the steel for normalizing temperatures below 1050 oC. However, above 1050 oC increase tungsten content decrease the grain size.

(c) (d)

Fig. 1. Microstructure of 9Cr-1W-0.06Ta RAFM steel after 5 minutes soaking at the indicated temperature followed by oil quenching. (a) 750oC, (b) 850oC, (c) 1050oC, (d) TEM micrograph indicate MX and M23C6 precipitate.

Tempering at 760 oC for 1 hour decreased the hardness of the steel in all microstructural conditions (Fig. 3(b)). On tempering, a noticeable reduction in hardness of the steel was seen for soaking at temperatures in the intercritical temperature range, resulting in a hardness trough. In comparison, the steel suffered a very marginal decrease in hardness on tempering when soaked at temperatures below Ac1. The RAFM steel derives its strength from the complex microstructures consisting of a high dislocation density, cell and subgrain boundaries decorated with M23C6 carbides and very fine coherent intragranular Ta-V carbides of the MX type. TEM investigation confirmed the precipitation of M23C6 and MX type carbides on (Fig. 1(d)). The reduction in hardness due to intercritical heating has been reported to be due to (i) replacement of martensite laths with high dislocation density by large subgrains with low dislocation density, and (ii) the coarsening of M23C6 carbides [7]. Soaking the steel at temperatures much higher than the Ac3 resulted in martensitic structure with high dislocation, with the consequence of increased in hardness (Fig.3). The tungsten content had quite significant

effect on the hardness of the steel on intercritical annealing. The reduction in hardness of the RAFM steel was comparatively less in the higher tungsten contained steel (Fig. 3).

Soaking Temperatur

Fig. 2. Variation in prior austenite grain size of RAFM steel with 5 minutes soaking temperature followed by oil quenching.

in (/> 01 c ■n

—r-800

-■-1 .ow -

-A-1.4W-

—i-■-1-r-

1000 1100

Soaking temperature, °C

-"t— 900

Fig.3. Variation in hardness of RAFM steels with 5 minutes soaking temperature. (a) in the as-quenched condition and (b) in tempered condition.

3.2. Tensile properties

Tensile tests were carried out on the steel in different microstructural state in 760 OC for 1 hour tempered condition. The steel at all microstructural state conditions showed monotonic smooth stress-strain curves for the tensile tests. The variations of 0.2% offset yield strength and the ultimate tensile strength with soaking temperature as shown in Fig. 4. The strength was found to increase with soaking temperature above Ac3 at all test temperatures. The higher strength of coarse grain martensite (Fig. 1(c)) (grain size above Ac3) may be attributed to its fine martensite structure with smaller lath and packetsize and fine distribution of carbides [6]. The rearrangement of the martensitic lath with high dislocation density into equiaxed cell, the replacement of needle shape particles with round smaller particles and reduction of misfit of VN particles within the matrix may be the reason for the reduction in tensile strength of the steel subjected to annealing in intercritical range.

-■-8Cr-1.DW-D.D6Ta -i-8Cr-1.4W-D.D6Ta -*-8Cr-2.DW-D.D6Ta

Soaking Ten^ienitijre, C

(a) (b)

Fig.4. Variation in (a) ultimate tensile strength and (b) yield strength (b) of RAFM steels with 5 minutes soaking temperature in the tempered condition.

3.3. Creep properties

Figure 5 shows the creep rupture life and creep strain curve for all steels at 550OC. The creep rupture life increases with the increasing in tungsten concentration or by the addition of MC carbide forming elements V and Ta. The creep deformation of this steel was characterized by a small instantaneous strain on loading, a transient primary stage, an apparent secondary stage, followed by a prolonged tertiary creep regime. It has been revealed that, in normalized and tempered 9Cr-martensitic steel, the transient creep is a consequence of the movement and annihilation of excess dislocations and that the acceleration creep is a consequence of gradual loss of creep strength due to the microstructural instability of the material on creep exposure [7]. The creep rupture life was found to increase with soaking temperature above Ac3 for all steels at a temperature 550 oC.

550C 24QMPa

9Gr-1 OW-O 06Ta —A— 9Cr-1.4W-0.06Ta —A— 9Cr-2.0W-0.06Ta

Soaking Temperature, C

Fig.5. Variation in creep rupture life of RAFM steels with 5 minutes soaking temperature (a) in the tempered condition and (b) creep

curve of RAFM steel tested at 550OC.

4. Conclusions

• The prior austenite grain size and hardness (after tempering) of the steel decreased when heated in and around the Ac3 transformation temperature. Further heating above Ac3, resulted in increase in the grain size and hardness of the steels.

• Subjecting the steel to inter-critical heating reduced its hardness and tensile strength. The tensile strength of the steel increased with soaking temperature above the Ac3. The reduction in tensile strength was less for steel having higher tungsten content.

• The creep rupture life of this RAFM steels increases with increasing in W concentration. The creep rupture life is less in inter-critical soaking temperature and it increases as the soaking temperature increases above Ac3 temperature.

• The sub-boundary hardening is the most important strengthening mechanism in the creep of these steels and is enhanced by the distribution of precipitate along boundaries.

Acknowledgement

Authors wish to acknowledge M/s MIDHANI, Hyderabad, the special public sector steel making company of the India, for producing the RAFM steel. They also thank the team members, involved in the experimental setup for heat-treatment and metallographic sample preparation.

References

[1] KJ.Harrelson, S.H.Rou, R.C.Wilcox, J. Nucl. Mater. 141-143 (1986) 508-512.

[2] Baldev Raj, T.Jayakumar, J. Nucl. Mater. Volume 417, issue 1-3(2011)72-76.

[3] R.Lindau, M.Schirra, Fusion Eng. Des. 58-59(2001)781-785.

[4] F.Abe, T.Noda, H.Araki, S.Nakazawa, J. Nucl. Mater. 179-181(1991)663-666.

[5] A.Khoyama, A.Hishinuma, D.S.Gelles, R.L.Klueh, W.Dietz, K.Ehrlich, J. Nucl. Mater. 233-237(1996)138-147.

[6] K.S.Chandravathi, K.Laha, K.Bhanu Sankar Rao, S.L.Mannan, Material Science and Technology, Vol. 17 (2001)559-565.

[7] Y.Tsucida, K.Okamoto, Y.Tokunaga, Study of creep rupture strength in heat affected zone of 9Cr-1Mo-V-Nb-N steel by welding thermal cycle, Welding International, Vol. 60 (1996) 27-33.