- Email: [email protected]
Improvement of impact toughness by modified hot working and heat treatment in 13%Cr martensitic stainless steel
Author’s Accepted Manuscript Improvement of Impact toughness by modified hot working and heat treatment in 13%Cr Martensitic Stainless Steel Kulkarni Srivatsa, Perla Srinivas, G Balachandran, V Balasubramanian www.elsevier.com/locate/msea
PII: DOI: Reference:
S0921-5093(16)31113-3 http://dx.doi.org/10.1016/j.msea.2016.09.045 MSA34130
To appear in: Materials Science & Engineering A Received date: 6 July 2016 Revised date: 10 September 2016 Accepted date: 13 September 2016 Cite this article as: Kulkarni Srivatsa, Perla Srinivas, G Balachandran and V Balasubramanian, Improvement of Impact toughness by modified hot working and heat treatment in 13%Cr Martensitic Stainless Steel, Materials Science & Engineering A, http://dx.doi.org/10.1016/j.msea.2016.09.045 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.
Improvement of Impact toughness by modified hot working and heat treatment in 13%Cr Martensitic Stainless Steel Kulkarni Srivatsa*, Perla Srinivas, G Balachandran, V Balasubramanian Kalyani Carpenter Special Steels Pvt. ltd, Mundhwa, Pune 411036, Maharashtra, India [email protected] [email protected] Abstract Improvement of the general mechanical properties and in particular sub-zero impact toughness in a 0.2%C-13%Cr martensitic stainless steel has been explored by varying the hot deformation and heat treatment conditions. The deformation conditions include hot rolling an ingot in one case and cogging the ingot to a semis followed by hot rolling in another case. The bars made from both routes were subjected to a single hardening heat treatment at 980oC and 1040oC oil quenched and a double hardening heat treatment at 1040 oC followed by 980oC oil quenched. The hardened steels were subjected to a standard two stage tempering at 710oC followed by 680oC. The impact toughness was found to be doubled in the cogged and rolled steel in double hardened condition. Other processing conditions show varying impact toughness levels. The toughness observed was correlated to the grain size and the carbide distribution in the matrix and the fractography features.
Keyword: Martensitic stainless steel deformation, 13%Cr steel, single hardening, double hardening, grain boundary carbide, grain refinement
1.0 Introduction There are several applications which demand superior strength-toughness and corrosion resistance in industrial sectors such as oil and gas, power generation and petrochemical. The 13%Cr stainless steel is a popular grade in the oil & gas industry [1] due to its lower cost with a combination of strength, toughness and corrosion resistance compared to the other grades of stainless steel family. The popularly used 13%Cr martensitic stainless steel grade contains about 0.2%C content, which on heat treatment exhibit a tempered lath martensitic structure with fine distribution of carbide that strengthens the matrix. The size, shape and distribution of the carbides and the grain size affect the mechanical properties especially sub-zero impact toughness of the steel. The carbide distribution and grain size is influenced by hot deformation and the heat treatment. Straight Cr martensitic stainless steels are easily forgeable and the hot working and deformation characteristics of the 0.2%C–13%Cr steel involves deformation in the fully austenite regime at temperatures between 1050°C to 1230°C, where there is complete absence of delta ferrite [2]. Post hot working the steel is slow cooled to room temperature leading to the formation of lath martensite microstructure. The hot worked steel is softened for fabrication by full annealing between 830 to 885°C [2], where the microstructure obtained has spheroidized carbides in a ferritic matrix. The carbides in annealed condition show up M23C6 type carbide along grain boundary along with presence of M 7C3 type carbide [3-7]. Increasing the annealing time is reported to convert M3C to M7C3, which after long time exposure form M23C6 carbides. The annealed steel after fabrication is strengthened by hardening and tempering. Hardening is usually carried out by austenitising in the range 965°C to 1050°C followed by air cooling, oil or polymer quenching [2]. At low austenitizing temperature (965°C - 1000°C), the dissolution of carbides in austenite is not complete and residual
1
carbides along grain boundary impairs the mechanical properties [8-10]. At high austenitization temperature (1000°C - 1040°C), there can be severe grain growth which affects the mechanical properties. Desirable mechanical properties are achieved by tempering the steel between 200 to 750oC avoiding the temper embrittlement range between 350 to 550oC [11]. Tempering below embrittlement range retains good strength with poor toughness and tempering above embrittlement shows high toughness with good strength and the carbide formed during tempering is reported to be M23C6. Presence of grain boundary carbides in all the above processing impairs the sub- zero toughness and the corrosion resistance [12-13]. Increase in the austenitizing temperature or soaking time that ensures complete dissolution of carbides without grain growth improve the toughness of the steel. The 13%Cr steel with higher carbon content (0.45%), austenitized at high temperature greater than 1040°C showed lower carbide fraction but had grain growth that deteriorated the strength and toughness [14]. In general, austenitizing condition (1000°C - 1040°C) that dissolves the grain boundary M23C6 carbides in austenite without grain growth is desired [15-16]. In the high carbon 13%Cr (D2 steel), the influence of sub-zero cryogenic treatment (-196 and -73oC) on secondary carbides by accelerated martensite decomposition has been characterized for estimation of carbide volume fraction and correlated it with impact toughness and strength [17]. The influence of microstructure on fracture toughness of cryogenically treated D2 steel has been studied and the toughness observed to decrease with lowering of retained austenite and enhanced secondary carbides. The presence of primary carbide is reported to have significant fall in the toughness and deep cryogenic treatment improved toughness due to carbide distribution [18]. The carbide distribution as a function of various types of low and subzero temperature treatment in D2 type steel and its influence on fracture toughness and wear resistance has been correlated to the microstructural modification associated with retained austenite and secondary carbide distribution [19]. In another study on sub-zero treated vanadis tool steel, the carbide distribution have been characterised by SEM, TEM and XRD techniques. The tempering response shows no secondary hardening in the sub-zero treated samples [20]. In an extended study on vanadis 6 tool steel, the effect of refining the secondary carbides and retained austenite and their influence on fracture toughness and wear was determined [21]. In the present study, a combination of thermo-mechanical and heat treatment processing in 0.2%C-13%Cr steel have been developed to improve the sub-zero impact toughness of the 13%Cr-0.2%C martensitic stainless steel by modifying the carbide distribution. 2.0 Experimental details The steel was melted in a 35 MT electric arc furnace followed by ladle refining and vacuum degassing. Liquid steel was then cast in a four metric ton ingot casting (avg ingot size of 570 mm) through bottom pour up-hill technique. Two typical ingots were chosen for study. One ingot was directly hot rolled through a blooming mill to 235 X 215 mm semis followed by finish rolling mill to 170 mm diameter bar. The second ingot was initially cogged to 250 X 270 mm semis and then it was further rolled to 170 mm diameter bar. The overall reduction ratio from cast ingot was maintained at 14 in both the conditions. The hot working parameters are summarized in Table 2. After hot rolling, the bars were subjected to annealing at 850°C for 4h followed by furnace cool up to 590°C. The annealed microstructure was characterized using SEM and XRD. XRD was carried out on the annealed steel on a Bruker D8 machine, with step size of 0.02° and scan speed of 3s per step with Mo Kα filter. Cut bars of 170 mm length, were chosen for heat treatment studies in a laboratory scale muffle furnace. Three cut
2
bars, each in conventional rolled bars and cogged + rolled bars, were subjected to the following heat treatment. One set (conventional roll and cogged + rolled) was conventionally hardened by austenizing at 980°C and 1040°C soaked for 3h 15m followed by oil quenching. Another set of bars (conventional rolled and cogged + rolled) were initially austenitized at 1040°C for 3h 15m followed by oil quench. Post quenching, the steel bars were kept in the quenchant for 45 m. The bars were then subjected to a second hardening treatment by austenizing at 980 oC for 3h 15m followed by oil quenching. All the samples post hardening were subject to double tempering followed by oil cool. First tempering was done at 710°C for 5h 15m followed by a second tempering at 680°C for 5h 15m. The steel chemistry was determined using ARL 3460 spectrometer and gas content was obtained using LECO gas analysers. The hardness was measured using Vickers hardness machine under 200 g with diamond indenter in various conditions. Mechanical testing was carried out at mid radius location using Shimadzu tensile testing machine and the test sample conformed to ASTM E8 round sample of gage diameter 12.5 mm and gage length 50 mm. The impact properties was evaluated at sub-zero temperatures (-10°C) for various heat treated conditions using an FIE testing machine. Samples for impact test were prepared as per ASTM E23. The microstructure was evaluated using optical microscope and select samples were examined for SEM. Etchant used was Villella reagent. The SEM microstructure as typically shown in Fig.1, for the carbide distribution was quantified using Image J software at 1000X magnification to capture the finest carbides in the microstructure. 10 fields were examined for carbide fraction. The areal fraction was determined for each field by colouring the different phases from which the percentage of the individual phases are automatically calculated by the image analysis software. The standard deviation was also determined. The prior austenite grain size was determined by oxidising the grain boundaries in a furnace as per ASTM E112. Fractography images were captured after double tempering using SEM under two set of processing conditions. Brittle area fracture was measured using Olympus stereomicroscope as per ASTM A370. Table 1 Hot working parameters used in the present study
Condition
Semis Cogging/Rolling Temp, oC
Conventional rolling
Rolling to 235 X 215 mm at 1210oC
Cogged+ Rolled
Cogged to 250 X 270 mm at 1175oC
Deformation finish temperature o C
Post forging /rolling processing
Final Rolling cycle
980
Deep vermiculite cooling
Rolling to 170 mm dia at 1210oC
950
Stress relieved at 750°C
Rolling to 170 mm dia at 1210oC
3
Post rolling processing Deep vermiculite cooling; followed by annealing at 850oC Deep cooling in a vermiculite bed and annealing at 850°C
Fig. 1 Typical SEM microstructure at 1000X showing the carbides (white) and matrix (black) for quantifying carbide fraction. 3.0 Results and Discussion The present study examines the mechanical properties especially the sub-zero impact toughness in a 13%Cr0.2%C steel. The study uses ingot cast steel subjected to deformation routes (i) conventional hot rolling (ii) cogged + rolled. Rolled bar samples from both the wrought steel products, were subjected to a (i) Single hardening heat treatments at 980oC and 1040°C followed by oil quenching and (ii) a double hardening heat treatment involving (1040oC/ oil quench + 980oC/ oil quench) heat treatments. The hardened steels were subject to a standard two stage tempering heat treatment at 710oC followed by 680oC [2]. The mechanical properties have been examined with a focus on the low temperature impact toughness. The properties have been correlated to the composition and microstructure evolution during the processing. The tempering temperature of 710oC was chosen to achieve optimum combination of strength, toughness and corrosion resistance satisfying some technical end uses. 3.1 Effect of alloying addition on martensite formation The chemical composition of the present steel is shown in Table 2. Typically, the 13%Cr steel has carbon content between 0.15 to 0.3% where adequate strength, corrosion resistance and toughness is realized. In the present steel, the carbon was maintained at 0.18%. At this carbon, the martensite start (Ms) temperature is 297oC & the finish temperature (Mf) is 197oC, which is well above room temperature as per formula given below [22-23], this in turn reduces the presence of retained austenite on hardening. In addition lower carbon content produces lesser carbide fraction in matrix, which improves the toughness of the steel at moderate expense of strength. Lower carbon promotes predominantly M23C6 type carbide, while the higher carbon has potential to promotes M 7C3 (M = Fe, Cr) type carbides as well [10]. Ms (oC) = 540 – (497*C + 6.3*Mn + 36.3*Ni + 10.8*Cr + 46.6*Mo)
--------(1)
Mf (oC) = Ms – 100
--------(2)
As per the above equation, the Ms temperature is 297oC and the Mf temperature is 197oC, which is matching with data reported by Nakazawa for 0.2C-13%Cr steel [24]. Considering the effect of austenitising temperature on Ms
4
and Mf temperature, Nakazawa has reported a fall in Ms and Mf temperature with austenitising temperature below 1000oC, where the carbide dissolution is complete. In the present study, using the data of Nakazawa, the difference in Ms temperature is about 10oC and the difference in Mf temperature is about 15oC for the austenitising temperatures of 980 and 1040oC. Hence, this low variation in M s and Mf temperature is not expected to significantly alter the microstructural features of the retained austenite or the carbide fraction. Lower sulphur is maintained to get more isotropic properties especially improved transverse impact toughness which decreases with elongated sulphides. The phosphorous is maintained at low level to avoid temper embrittlement. Inherently, the 13% Cr content in steel gives good corrosion resistance and it also ensures good hardenability and the CCT diagram is shifted to the right, which enables martensite formation in the through thickness over a wide range of cooling rate including cooling in air even in thick sections [25]. The Al/N ratio was maintained above 2.5 to ensure N fixing by residual Al to form AlN which refine austenite grains during deformation. Table 2 Chemical composition (in wt.%) of the steel examined. Element Actual
C 0.18
Mn 0.85
Si 0.30
S 0.002
P 0.020
Cr 12.95
Ni 0.18
Mo 0.03
Al 0.028
N 0.0070
MS(°C) 297
Mf (°C) 197
3.2 Characterization of the annealed microstructure The 13%Cr steel is usually full annealed after hot working in the temperature range from 830 to 885oC to soften the matrix required for improved fabrication by machining. The microstructure obtained after annealing shows spheroidized carbides in ferrite matrix as reported in literature [3-7]. In the present study, the microstructure obtained after annealing of the bars showed fine grain boundary precipitates and coarse grain boundary chain precipitates in the ferrite matrix as shown in Fig. 2. The composition of precipitates was assessed using SEMEDAX [Table 3]. The EDAX spot analysis on chain precipitates in Fig. 2 shows a peak for carbon content compared to the matrix implying that it is carbide. As it is difficult to precisely determine composition of the carbide in SEM-EDAX due to the low atomic no. of carbon, alternate methods have been used to confirm the carbide. Analysis of Cr/Fe ratio was used to confirm the carbide [26]. The Cr/Fe ratio for the matrix where carbon is lean was estimated to be 0.27, while the chain type precipitate phases showed Cr/Fe ratio of 0.97 which was reported to imply the M23C6 type carbide [26]. The higher Cr/Fe ratio is observed in the precipitate carbide as more Cr is fixed by carbon than Fe atom compared to the matrix.
5
Fig. 2 SEM microstructure in the annealed condition. Annealed steel samples were subjected to XRD analysis as in Fig. 3. The results confirmed presence of ferrite peaks and absence of Cr7C3 and presence of some fine peaks of Cr23C6 type carbide. The peaks compared well with literature data [27-31]. Table 3 The EDAX spot analysis of the carbide in the steel (At%). Location Spot 1
Phase Matrix (ferrite)
C
Cr
Fe
Cr/C
(Cr+Fe)/C
Cr/Fe
14.04
*
17.20
63.62
1.22
5.756
0.27
*
16.25
61.30
1.174
5.607
0.265
Spot 2
Matrix (ferrite)
13.83
Spot 3
Matrix (ferrite)
14.30*
16.85
62.37
1.178
6.718
0.27
Spot 4
Precipitate
38.32
30.46
31.22
0.795
1,610
0.975
Spot 5
Precipitate
35.11
32.33
33.55
0.921
1.876
0.96
Spot 6
Precipitate
40.32
31.78
32.78
0.788
1.601
0.97
*Solubility of C in ferrite is less and atomic no of C is 6, so actual carbon in ferrite matrix may be less
Fig. 3 XRD analysis showing ferrite peaks with minor Cr23C6 type carbide peaks in the annealed samples of conventionally rolled sample. All the peaks of Cr23C6 were indexed at the expected peak location although only peaks at planes (511), (660) and (800) could show up due to low volume fraction. This means that annealed condition has Cr23C6 carbides predominantly.
3.3 Effect of heat treatment on prior austenite grain size The conventionally rolled bar and the cogged + rolled bar subject to heat treatment conditions, were examined at various conditions of hardening for prior austenite grain sizes as in Fig. 4. The prior austenite grain size of the conventional rolled and cogged + rolled steel samples hardened at 980 oC showed considerable variations. The grain size of the conventional rolled steel had large patches of coarse grain size with an ASTM No 6 as shown in Fig. 4(a), while the cogged + rolled steel showed a more uniform grain size ASTM No.6.5 as in Fig. 4(b). The
6
reason for fine grain size in cogged + rolled condition may be attributed to higher deformation strains towards the core that enables higher degree of recrystallization compared to hot rolling where deformation is concentrated towards the surface. Hardening at 1040°C resulted in coarsening of the prior austenite grains in both the conventional rolled [ASTM no 4.5] and cogged + rolled [ASTM 5.5] samples as seen in the Fig. 4(c-d). It is observed that the higher temperature of heat treatment has led to the coarsening of the grains. The results obtained were close to the values reported in literature [32]. Double hardening heat treatment at 1040°C/OQ + 980°C/OQ resulted in a refinement of the prior austenite grains in conventional rolled material (ASTM 7.0) and cogged + rolled material (ASTM7.5) as shown in Fig. 4(e-f). The probable reason for finer grain in the double hardening treatment may be attributed to generation of defect structure in the matrix during martensite formation in the first hardening at 1040oC/OQ and the defect structure generated in this stage gets recrystallized too at the second hardening condition at 980oC/OQ. A similar studied on double hardening was carried out on 16Cr-2Ni steel [33-34], where double hardening refined the grain size and improved impact properties. However the result did not match with present study but trend observed to be same.
Fig. 4 Grain size of (a) Conventional rolled sample hardened at 980°C [Grain size: ASTM No.6] (b) Cogged + Rolled sample hardened at 980oC [Grain size: ASTM No.6.5] (c) Conventional rolled sample hardened at 1040° C showing [Grain size: ASTM No.4.5] (d) Cogged + Rolled samples hardened for 1040oC showing [Grain size: ASTM No.5.5] (e) Conventional rolled sample hardened at 1040 + 980°C [Grain size: ASTM No.7] (f) Cogged + Rolled samples hardened for 1040oC + 980°C [Grain size: ASTM No.7.5] 3.4 Characterization of the as-quenched hardness and microstructure:
7
The samples examined for the grain sizes at various hardening heat treatment were subsequently evaluated for their as-quenched hardness, microstructures and carbide fraction [Fig. 5(a-f) and Table 4]. It is seen from Table 4 that the as-quenched hardness at 8 different locations on each sample varied between 458 and 567 HV. The conventional hot rolled sample, hardened at 980oC showed the least hardness of 458 ± 15 HV, while hardening it at 1040oC showed a hardness of 474 ± 12 HV. The conventional rolled material subject to double hardening [1040°C/OQ + 980°C/OQ] showed further improved hardness up to 531 ± 8 HV. The steel processed in the cogged + rolled condition shows about 30 to 60 HV more hardness than the conventional rolled material at the corresponding double hardened condition. The high hardness in the steel implies that carbon dissolution in the matrix and finer grain sizes. Similar results were reported in literature [35]. But the discrepancy in hardness value was attributed to minor chemical compositional changes. The microstructures of the as-quenched samples examined using SEM showed carbides in a matrix with martensitic lath [Fig. 5(a-d) & Table 4]. The SEM microstructure of the conventional rolled and cogged + rolled sample subjected to single hardening at 980oC show coarse residual chain type carbides along with matrix carbides [Fig. 5(a-b)].The matrix has coarse lath martensite probably due to the coarse austenite grain structure as shown in Fig. 4(a-b). Single hardening at 1040oC for both the processing condition further decreased the matrix carbides and thinning down of chain carbides took place [Fig. 5(c-d)]. There was further coarsening of lath martensite observed due to grain coarsening as compared to single hardening at 980°C shown in Fig. 4(c-d). This dissolution of carbides at 1040°C led to increased as-quench hardness compared to 980oC hardening. The microstructure of the samples subjected to double hardening heat treatment [1040°C/OQ + 980°C/OQ] in conventional rolled and cogged+ rolled conditions show complete absence of grain boundary chain carbides [Fig. 5(e-f)] and the lath martensitic microstructure is finest corresponding to finer grain sizes [Fig. 4(e-f)]. The carbide content and distribution is least in the cogged and rolled condition comparison to the conventional rolled condition after double hardening condition. The present study shows that the deformation involving cogged + rolled condition followed by double hardening redistributes the carbides (both residual and grain boundary) more effectively than that in other conditions. In spite of the tempering condition remaining the same, the amount and distribution of the precipitated carbides during tempering varied with the deformation processing and hardening conditions. This shows that careful selection of deformation and hardening condition can redistributes the carbides that influences the impact toughness.
8
Table 4 As-quench hardness correlated with microstructure and carbide fraction at various hardening conditions. Process Route Rolled
Hardening condition (oC)
As quenched hardness (HV) 458 ± 15
Microstructure observation Carbides sparsely distributed in the matrix and there is the presence of grain boundary carbide chain. Remnants of lath martensite structure are observed. Carbide fraction (Std. dev.) = 4.36% (±0.21)
980°C/OQ Cogged+ rolled
474 ± 12
Rolled
490 ± 10
Carbides in matrix have dissolved while no remnant grain boundary carbide chain seen. Strong remnant of lath martensitic microstructure. Carbide fraction (Std. dev.) = 2.32% (±0.15) Carbides in the matrix is very coarse and dissolution of carbide chains seen. Martensite laths coarsening seen. Carbide fraction (Std. dev.) = 1.22% (±0.06) Carbides are totally dissolved and the coarsening of lath martensite; Carbide fraction (Std. dev) = 0.84%(±0.06) Microstructure was completely free of K1 carbides and fine lath martensite was uniformly present throughout the matrix; Carbide fraction (Std. dev.) = 0.56% (±0.08) Carbide fraction (Std. dev) = 0.52% (±0.05)
1040°C/OQ Cogged+ rolled
505 ± 6
Rolled
531 ± 8
Cogged+ rolled
1040°C/OQ+ 980°C/OQ
567 ± 6
Fig. 5 SEM microstructure of as-quenched samples (a) Rolled subject to 980oC/OQ (b) Cogged + Rolled subject to 980oC /OQ (c) Rolled subject to 1040oC/OQ (d) Cogged + Rolled subject to 1040oC/OQ (e) Rolled subject to (1040oC/OQ+980oC/OQ ) (f) Cogged + Rolled subject to (1040oC/OQ+980°C/OQ).
3.5 Tempered Microstructure and Mechanical Properties The mechanical properties of the steels in the conventionally rolled and cogged + rolled steels were examined in the various hardening treatment followed by tempering. The tempering involved standard tempering at 710oC followed by a second tempering of 680oC in all the samples [2]. At this tempering temperature, a good
9
combination of strength and sub zero toughness is realized for a wide range of applications. This tempering temperature is above the secondary hardening range and the expected phases are a ferritic matrix phase free of carbon and carbide precipitated out that favours high toughness.
The mechanical properties of the steel under the various hot working conditions after the various hardening followed by standard tempering are shown in Table 5. The conventional rolled and cogged + rolled steel hardened and tempered condition showed comparable strength and ductility properties. The conventional rolled steel show impact toughness lower than the cogged + rolled steel at similar tempering condition. The conventional rolled and cogged + rolled samples subjected to double hardening heat treatment [1040oC/OQ + 980oC/OQ] followed by the tempering treatment showed higher strength and higher toughness compared to the single hardening heat treatment. The impact toughness of cogged + rolled steel is found to be doubled along with enhanced strength. Although a double hardening process may be expensive, the properties are attractive. When the toughness is very high, the tempering temperatures can be lowered to lower the toughness, which consequently results in enhanced strength. Increased strength enables weight reduction of the structure, savings in material and cost. In order to meet the sub-zero impact toughness, micro alloying of the steel with expensive Ni and Mo adopted by some of the end users can be avoided by the use of the double hardening heat treatment.
The optical microstructure of the conventional rolled and cogged + rolled steels after tempering at low and high magnification is shown in Fig. 6(a-h). Tempered martensite microstructure with carbides precipitated on the ferrite lath boundaries along with carbides chain along grain boundaries can be observed. The carbide formed for these grades at this temperature range was reported to be M23C6 carbides [11,36-37]. The conventional rolled steel hardened at 980oC after tempering showed fine laths containing trans-granular carbides along with grain boundary carbides as in Fig. 6(a-b). The cogged + rolled steel hardened at 980oC after tempering showed more ferritic zones with some dispersed grain boundary carbides [Fig. 6(c-d)]. The conventional rolled steel and the cogged + rolled steel hardened at 1040oC followed by tempering showed that in spite of higher hardening temperature the carbides have not dissolved in some zones [Fig. 6(e-h)]. The ferrite laths are coarser due to the coarser austenite grains are coarser as in Fig. 4(c-d). The extent of thickening of grain boundary carbides hardened at 1040 oC was less in the cogged + rolled sample compared to the rolled sample. The rolled steel sample subjected to the double hardened [1040 oC/OQ + 980°C/OQ] treatment followed by tempering showed much coarser ferritic lath matrix with highly diffused grain boundary carbides as in Fig. 6(i-j). The cogged + rolled sample with double hardening [1040oC/OQ + 980°C/OQ] followed by tempering showed high dispersion of ferritic lath with very fine carbides [Fig. 6(k-l)]. The improved toughness in the cogged + rolled steel in the double hardened may be attributed to the finer grain size [Fig. 4(e-f)], absence of grain boundary carbide and fine distribution of matrix carbide [Fig. 5(e-f)].
The hardened and tempered samples used for optical microscopy were further examined for finer details using SEM. The conventional rolled steel and the cogged + rolled steel hardened at 980 oC followed by tempering show coarse carbide distribution in the matrix [Fig. 7(a-b)]. The conventional rolled samples show prominent grain boundary carbides in Fig. 7(a) than the cogged + rolled sample in Fig. 7(b). The toughness in rolled sample was
10
less [CVN = 16 to 18 J] compared to cogged + rolled sample after single hardened at 980°C and 1040oC tempered [CVN = 20 to 23 J]. This may be attributed to the presence of higher amount of grain boundary carbides. In contrast, the cogged + rolled material shows reduced carbides along the grain boundaries, lower amount of dispersed carbides and a coarser ferrite lath in the matrix [Fig. 7(b)] at the same conditions of heat treatment that led to high toughness compared to the rolled steel. This implies that during cogging and subsequent hardening there is favourable carbide dissolution and redistribution in the matrix. The cogging process introduces higher effective strains in the core and the deformation penetration creates more sub angle and high angle grain boundaries during recrystallization. In addition, there is higher adiabatic rise in temperature due to deeper strains in cogging. All these factors enable carbides to effectively dissolve in the matrix and there is coarsening of laths indicative of grain refinement.
Table 5 Mechanical properties of the steel evaluated at the various processing routes tempered at the standard tempering at 710°C/5h 15m/oil cooled + 680°C/5h 15m/oil cooled Mechanical properties Process Route
Heat treatment
Cogged + Rolled
Single hardened at 980°C/3h15m/ oil quenched; Single hardened at 980°C/3h 15m/ oil quenched.
Conventional rolled
Single hardened at 1040°C/3h 15m/ oil quenched +
Cogged + Rolled
Single hardened at 1040°C/3h 15m/oil quenched
Conventional rolled
% Mean carbide (Std dev)
Hardness (HV)
YS (MPa)
UTS (MPa)
%E
%RA
8.32 ( ±0.16)
218 ± 6
570
712
20.3
63.5
18
5.34 (±0.14)
220 ± 4
565
715
20.6
62.6
23
210 ± 5
556
714
21.2
64.2
16
2.65 (±0.09)
218 ± 4
575
725
21.3
63.5
20
3.78 (±0.11)
CVN (J)
Conventional rolled
Double hardened at 1040°C/3h 15m/ oil quenched + 980°C/3h 15m/oil quenched
3.05 (±0.08)
227 ± 5
620
752
26.7
66.1
29
Cogged + Rolled
Double hardened at 1040°C/3h 15m/ oil quenched + 980°C/3h 15m/oil quenched
2.16 (±0.06)
225 ± 6
631
765
26.4
66.6
41
The toughness value is least in the conventional rolled sample subject to single hardening at 1040°C followed by tempering, compared to the cogged + rolled sample at same tempering condition. Single hardened samples in rolled and cogged + rolled routes show impact toughness value between 16 and 23 J which is closer to specification and reports in literature [38]. Least toughness of 16 J was observed at 1040°C followed by tempering
11
and it could be attributed to the significant grain coarsening and also presence of residual carbides in the grain boundary [Fig. 4(c-d) & Fig. 5(c-d)]. The SEM microstructure of the samples subjected to double hardening heat treatment [1040oC/OQ + 980°C/OQ] in the rolled and the cogged + rolled samples show that the carbides are almost completely dissolved and are re-precipitated as very fine carbide along with coarser ferrite laths [Fig. 7(e-f)] due to finer prior austenite grain sizes [Fig. 4(e-f)]. The conventional hot rolled steel still shows some remnant grain boundary carbides [Fig. 7(a)] which results in lower toughness than in the double hardened steel [Fig. 7(e)]. The cogged and rolled steel in the double hardened condition shows complete absence of grain boundary carbide and a very fine carbide distribution in the matrix and this has given almost double the impact toughness [Fig.
7(e-f)].
12
Fig. 6 Optical microstructure of the steels subject to tempering (a)-(b) Conventional rolled and single hardened at 980oC [CVN = 18J]; (c)-(d) (Cogged + rolled) steel subjected to single hardening at 980oC [CVN = 23J]; (e) – (f) Conventional rolled and single hardened at 1040 oC [CVN = 16J]; (g)-(h) (Cogged + rolled) steel subjected to single hardening at 1040oC [CVN = 20J]; (i)-(j) Conventional rolled steel double hardenened [1040°C/OQ + 9800C/OQ] [CVN = 29J] (k-l) (Cogged + rolled) steel double hardenened [1040°C/OQ + 9800C/OQ] [CVN = 41J].
Fig. 7 SEM microstructure after double tempering tempering (a) Conventional rolled product hardened at 980°C [CVN = 18J] (b) Cogged + rolled product hardened at 980°C [CVN = 23J] (c) Conventional rolled product hardened at 1040°C [CVN = 16J] (d) Cogged + rolled product hardened at 1040°C [CVN = 18J] (e) Rolled product hardened at 1040°C + 980°C [CVN = 29J] (d) Cogged + Rolled product hardened at 1040°C +980°C [CVN = 41J]. The ferrite laths in the double hardened cogged + rolled steel are extremely fine compared to the single hardened cogged and rolled steel. The cogged and rolled steel with double hardening and tempering revealed the best impact toughness which can be the bench mark microstructure. Thus, the fine carbides in cogged and rolled steel
13
is due to dissolution and re-precipitation of the carbide in deeply worked steel compared to conventional rolled steel. The lath structure of the ferrite in the single hardening is further refined by double hardening.
3.6 Effect of carbide distribution on mechanical properties of 13%Cr steels The carbide distribution observed in the SEM microstructures was further quantified in the single hardened and double hardened followed by the standard tempering. The percentage of the carbide in the microstructure and the hardness in the various processing conditions was estimated using the SEM images using the image analysis software. The carbide fraction was calculated at five different locations and the average carbide fraction is reported along with standard deviation results are shown in Tables 4 & 5 and Fig. 8. The results show that the carbide content in the tempered condition is much higher than in the hardened condition, as expected. The conventional rolled sample austenitized at 980oC, in the as-quenched condition show a carbide volume fraction of ~ 4%, which enhances to ~ 8% on tempering. At this condition, both the as-quenched and tempered hardness are the lowest compared to other heat treatment condition as shown in Fig. 8. The volume fraction of carbide is ~2% for cogged + rolled sample in the as-quenched condition hardened at 980oC, which on tempering enhances to ~ 5%. The volume fraction of carbide is ~1% for both conventional rolled and (cogged + rolled) sample in the asquenched condition hardened at 1040oC, which on tempering enhances to ~ 3.5 and 3% respectively.
For the
double hardened steel, the volume percentage of carbide in the as-hardened condition in conventional rolled sample is 0.5% which increases to about 3% where the tempered hardness is the highest. The cogged + rolled samples after double hardened in the as-quenched condition show about ~ 0.5% carbide, which on tempering becomes ~ 2%. High deformation and double hardening create more defect structure in the matrix carbides are fine on tempering, which enhances the strength and the toughness.
Fig. 8 The carbide fraction estimated in SEM and the hardness at various processing conditions.
The carbide content in the microstructure as a function of the hardness obtained under after the two processing condition was examined as in Fig. 9. The as-quenched hardness at various conditions was found to decrease as the carbide volume in the microstructure increased. This means that as carbide content is increased less carbon is in the martensite leading to lowering of hardness. In the tempered condition, the hardness values are almost similar but the carbide volume fraction show considerable variation with processing condition. This is important because
14
lower carbide content will ensure that there is lesser carbide precipitation at the same hardness levels. This lowering of carbide content in the double hardened and conventional rolled and cogged + rolled sample gives higher hardness as more carbon is available in the matrix to harden the martensite and toughen with less brittle particles in the matrix and in the grain boundaries. The mechanical properties in the tempered condition was plotted as a function of the total carbide content irrespective of the processing condition in Fig. 10 and it is seen that as the carbide content increases there is moderate loss in strength and ductility, while there is significant loss in toughness. Hence, to achieve highest strength and good toughness level, thermo mechanical processing and heat treatment condition should ensure finer and low volume carbides in the matrix and the grain boundary. Tempering should not precipitate coarse carbides. Thus, this study has brought out the best microstructure for 13%Cr steel that can give improved strength and subzero impact toughness.
Fig. 9 Carbide fraction at various conditions of heat treatment compared with matrix hardness.
Fig. 10 Effect of carbide content in the tempered martensitic microstructure on the mechanical properties of the 13% Cr steel.
3.7 Fractography of the broken impact sample
15
Fractography was carried out on the broken impact test samples to understand the fracture mode under various processing conditions subjected to single hardening and double hardening condition followed by tempering. The fractography results are shown in Fig. 11(a-f).
It is seen that, in conventional rolled sample single hardened at 980°C and 1040°C followed by tempering shows higher degree of cleavage facets as in Fig. 11(a&c) the corresponding impact toughness is the least. The cogged + rolled steel samples subject to single hardening at 980°C followed by tempering showed mixed mode failure with lower cleavage facets [Fig. 11(b)] than the conventional rolled sample. Hence, a moderate improvement in toughness was observed. The conventional rolled double hardened [1040 oC/OQ + 980°C/OQ] followed by tempering still higher degree of dimples but some isolated spots of cleavage facets [Fig. 11(e)]. The dimples were coarse. The cogged and rolled sample after double hardening [1040 oC/OQ + 980°C/OQ] and tempering showed fully dimple microstructure without cleavage facets [Fig. 11(f)]. This has resulted in highest toughness in the steel. The fractographic features are attributed to the grain size and carbide morphologies. The cogged + rolled samples exhibit fine grains compared to conventional rolled samples. Secondly, there is absence of grain boundary carbides in the cogged + rolled sample compared to the conventional rolled steel which results in less crack initiation sites and lower propagation rate due to fineness of the carbide. .
Fig. 11 Fractography images of the impact test samples hardenede and tempered steels at (a) Conventional rolled steel austenitized at 980°C/OQ/tempered [CVN = 18J] (b) Cogged + Rolled steel austenitized at 980°C/OQ / tempered [CVN = 23J] (c) Conventional rolled steel austenitized at 1040°C/OQ/tempered [CVN = 16J] (d) Cogged + Rolled steel austenitized at 1040°C/OQ / tempered [CVN = 18J] (e) Conventional rolled steel double hardened [1040°C/OQ + 980°C/OQ]/tempered [CVN = 29J] (f) Cogged + Rolled steel double hardened [1040°C/OQ + 980°C/OQ]/tempered [CVN = 41J]
16
The relation between the brittle area fracture, carbide fraction and impact toughness was evaluated. The brittle area fracture was measured on cross section of three broken impact samples using stereomicroscope. The standard deviation was also calculated and plotted in the Fig. 12. The typical stereo micrograph of the fractured sample at each condition showing the brittle area is also shown in Fig. 12. As the carbide fraction increases there is less ductile area compared to brittle area. The shear lip is prominent at lowest carbide density. From the Fig. 12 it is observed that, as the carbide fraction increases, the brittle area fracture increases and toughness decreases. In the Fig. 12 Point c, d represent the high brittle fracture area where the corresponding impact fracture observed is low. This is due to grain growth occurring at 1040°C for both the processing conditions.
Fig. 12 Tempered carbide volume fraction (%) as a function of mean brittle area fracture (mm 2) and impact toughness (J). Pictures in the graph show the stereo image of (a) Rolled 980°C + tempering [CVN = 18J] (b) Cogged + Rolled 980°C + tempering [CVN = 23J] (c) - Rolled 1040°C + tempering [CVN = 16J] (d) Cogged + Rolled 1040°C + tempering [CVN = 18J] (e) Rolled 1040+ 980°C + tempering [CVN = 29J] (f) Cogged + Rolled 1040 + 980°C + tempering [CVN = 41J] As a future activity, it is of interest to extend the influence of thermo-mechanical processing, double hardening and double tempering on other high carbon stainless tool steels such as D2, AISI 440 etc., where refinement of carbides are likely to improve their toughness properties and can be combined with sub-zero treatment as well [17-21]. In summary this study has given the best process condition in terms of deformation and heat treatment along with the best microstructure towards achieving excellent improvement in strength and toughness in 13%Cr martensitic stainless steel. A fine grained sample obtained by cogging or cogged + rolled route, with double hardening will refine the grain structure, the lath morphology, the extent of carbide precipitation, and distribution of carbides precipitation that ultimately improves the strength and toughness.
4.0 Conclusions
1. The improvement in the mechanical properties with a focus on impact toughness in 13%Cr martensitic stainless steel has been examined in the steel processed through conventional hot rolling and cogged + rolled
17
conditions followed by subsequent heat treatments involving single hardening at 980oC and 1040oC and also a double hardening at 1040oC/OQ followed by 980oC/OQ. The hardening was followed by a two stage tempering at 710°C and 680°C. 2. The hot worked and annealed steel shows ferrite matrix with Cr 23C6 type carbides in grain boundary and matrix. 3. The samples that were subject to single hardening at 980oC, 1040°C or double hardened at 1040oC/OQ + 980oC/OQ which influenced the carbide distribution in the steel leading to a strong influence on impact toughness of the steel. 4. Highest toughness was realized in the cogged and rolled steel subjected to double hardening followed by standard tempering which is attributed to the lowest carbide content and fineness along the grain boundary and the matrix which gave a fractography feature of fine dimples without cleavage facets in the matrix. 5. Lowest toughness was obtained in the conventionally rolled steel subjected to single hardening at 980 oC, 1040°C followed by standard tempering which led to highest cleavage facets associated with highest carbide content with highest distribution along the matrix and the grain boundary. 6. Moderate improvement of toughness sample is observed in cogged + rolled samples subject to single hardened [980oC and 1040°C] followed by tempering than the conventionally rolled samples subjected to double hardening [1040oC/OQ + 980oC/OQ] and tempered conditions. The toughness could be correlated to the presence of carbides along grain boundary and matrix that led to some cleavage facets in the fractography. 7. The present study brings out the best thermo-mechanical processing condition, heat treatment condition, the bench mark microstructure to be targeted to improve the impact toughness.
Acknowledgments The authors wish to thank the constant encouragement and support provided by Dr Baba Kalyani, Chairman & Managing Director and Mr R K Goyal, Managing Director, Kalyani Carpenter Special Steels Pvt. Ltd., Pune. Thanks are due to Department of Scientific and Industrial Research for their support of technical activities at KCSSPL R&D centre.
References [1] Cássio Barbosa, Ibrahim Abud, Recent Pat. Corr. Sci., 3 (2013), 27-38 [2] H. Chandler, Heat Treaters Guide: Practice and Procedure for Iron and Steel, 2 nd ed., ASM international, (1994), 775 [3] A. Hede and B. Aronson, J. Iron Steel Inst., (1969), 1241-51 [4] T. Fujita and N. Takahashi, Trans. ISIJ, 18 (1978), 702-11 [5] A. Bja¨rbo: Scand. J. Metall., 23 (1994), 103-12 [6] A. Bjärbo, Mats Hättestrand, Metall. Mater. Trans. A, 32 (2001), 19-27 [7] B. Mintz, J. M. Arrowsmith, ASTM, STP 706 (1980), 313-335 [8] R. Gullberg, J. Iron Steel Res Int., (1973), 59 – 65 [9] L. D. Barlow, J. Mater. Eng. Perform., 21 (2012) issue no 7, 1327 – 1336 [10] H. Bhadeshia, Steels: Microstructure and Properties, 1st ed., Butterworth – Heinemann, (2006), 15-20 [11] V. Anil Kumar, R. K. Gupta, M. K. Karthikeyan, P. Ram Kumar, P. P. Sinha, Mater. Sci. forum , 710 (2012), 489-494 [12] A. Nasery Isfahany, H. Saghafian, G. Borhan, J. Alloys Compd, 509 (2011), issue no 9, 3931-3936 [13] A.F. Candelaria, C. E. Pinedo, J. Mater. Sci. Lett., 22 (2003), 1151-1153 [14] C. Garcia de Andres, L.F. Alvarez, V. Lopez, J. Mater. Sci. Lett., 33 (1988), 4095–4100.
18
[15] C. Garcia - De - Andris, L. F. Lavarez, First International Conference on Phase Transformation, (1996), 82 [16] Ming Laura Xu: Secondary carbide dissolution and coarsening in 13%Cr Martensitic Stainless steels during austenization, PhD dissertation, (2012), 1-156 [17] D. Das, A.K. Dutta, K.K. Ray, Mater. Sci. Engng., A527(2010), 2182 - 2193. [18] D. Das, R. Sarkar, A.K. Dutta, K.K. Ray, Mater. Sci. Engng., A528 (2010), 589 – 603 [19] D. Das, K.K.Ray, Mater. Sci. Engng., A541 (2012),45 – 60 [20] P. Jurci, M. Domankova, L. Caplovic, J. Ptacinova, J. Sobotova, P. Salabova, O. Prikner, B. Sustarsic,D. Jenko, D. Vacuum, 111(2015),92-101. [21] J. Sobotova, P. Jurci, I .Dloughy, Mater. Sci. Engng., A652 (2016), 192-204. [22] Steve Lampman: Weld Integrity and Performance: A Source Book Adapted from ASM International, 1st ed, ASM International, (1997), 417 - 442 [23] J. E. Ramirez, Weld J, 86 (2007), 125s – 134s [24] K. Nakazawa, Trans Jpn Inst Met, 4 (1963), 200 - 205 [25] M. Atkins, Atlas of Continuous Cooling Transformation Diagrams for Engineering Steels, 1 st ed., ASM, (1993), 145-150 [26] K. Wieczerzak, P. Bała, M.Stępień, G. Cios, T. Kozieł, Arch. Metall. Mater., 60 (2015), issue no 2, 779-782 [27] E. Jartych, T. Pikula, D. Oleszak, Hyperfine interact, 192 (2009), 117-124 [28] Märt Kolnesa, Jüri Pirsoa, Jakob Kübarseppa, Mart Viljusb, Rainer Traksmaab, Proceedings of the Estonian Academy of Sciences, 65 (2016), issue no 2, 138-143 [29] Arūnas Baltusnikas, Rimantas Levinskas, Mater. Sci., 12 (2006), issue no 3, 192-198 [30] M. Hillert et al. J. Iron steel inst., 209 (1971), 49 – 66 [31] Mehmet Erdem, Mehmet Altuğ,Mustafa Karabulut, Int. J. Adv. Manuf. Technol., (2015), 1-15 [32] Toshihiro Tsuchiyamas, Etsuo Takaki and Sadavuki Nakamura, ISIJ Int., 81 (1995), issue no 2, 59-64 [33] J. K. Lia, Stainless steels and Introduction and their recent development, 1 st ed., Bentham EBooks, (2012) 41 – 51 [34] K.P. Balan, A. Venugopal Reddy, D.S. Sarma, J. Mater. Eng. Perform, 8 (1999), issue no 3, 385-393 [35] Toshihiro Tsuchiyamas, Etsuo Takaki, Sadavuki Nakamura, ISIJ Int. 80 (1994), issue no 12, 68-73 [36] B Maio, D.O. Northwood, L.C. Lim, M.O. Lai, Mater. Sci. Eng. A, 171 (1993), 21-33 [37]Gopa chakraborty, C. R. Das, S. K. Albert, A. K. Bhaduri, V. Thomas Paul, G. Panneerselvam, Arup Dasgupta, Mater. Character., 100 (2015), 81-87 [38] G. V. Prabhu Gaunkar, A. M. Huntz, and P. Lacombe, Met. Sci., 14 (1980), issue no 7, 241-251
19
PII: DOI: Reference:
S0921-5093(16)31113-3 http://dx.doi.org/10.1016/j.msea.2016.09.045 MSA34130
To appear in: Materials Science & Engineering A Received date: 6 July 2016 Revised date: 10 September 2016 Accepted date: 13 September 2016 Cite this article as: Kulkarni Srivatsa, Perla Srinivas, G Balachandran and V Balasubramanian, Improvement of Impact toughness by modified hot working and heat treatment in 13%Cr Martensitic Stainless Steel, Materials Science & Engineering A, http://dx.doi.org/10.1016/j.msea.2016.09.045 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.
Improvement of Impact toughness by modified hot working and heat treatment in 13%Cr Martensitic Stainless Steel Kulkarni Srivatsa*, Perla Srinivas, G Balachandran, V Balasubramanian Kalyani Carpenter Special Steels Pvt. ltd, Mundhwa, Pune 411036, Maharashtra, India [email protected] [email protected] Abstract Improvement of the general mechanical properties and in particular sub-zero impact toughness in a 0.2%C-13%Cr martensitic stainless steel has been explored by varying the hot deformation and heat treatment conditions. The deformation conditions include hot rolling an ingot in one case and cogging the ingot to a semis followed by hot rolling in another case. The bars made from both routes were subjected to a single hardening heat treatment at 980oC and 1040oC oil quenched and a double hardening heat treatment at 1040 oC followed by 980oC oil quenched. The hardened steels were subjected to a standard two stage tempering at 710oC followed by 680oC. The impact toughness was found to be doubled in the cogged and rolled steel in double hardened condition. Other processing conditions show varying impact toughness levels. The toughness observed was correlated to the grain size and the carbide distribution in the matrix and the fractography features.
Keyword: Martensitic stainless steel deformation, 13%Cr steel, single hardening, double hardening, grain boundary carbide, grain refinement
1.0 Introduction There are several applications which demand superior strength-toughness and corrosion resistance in industrial sectors such as oil and gas, power generation and petrochemical. The 13%Cr stainless steel is a popular grade in the oil & gas industry [1] due to its lower cost with a combination of strength, toughness and corrosion resistance compared to the other grades of stainless steel family. The popularly used 13%Cr martensitic stainless steel grade contains about 0.2%C content, which on heat treatment exhibit a tempered lath martensitic structure with fine distribution of carbide that strengthens the matrix. The size, shape and distribution of the carbides and the grain size affect the mechanical properties especially sub-zero impact toughness of the steel. The carbide distribution and grain size is influenced by hot deformation and the heat treatment. Straight Cr martensitic stainless steels are easily forgeable and the hot working and deformation characteristics of the 0.2%C–13%Cr steel involves deformation in the fully austenite regime at temperatures between 1050°C to 1230°C, where there is complete absence of delta ferrite [2]. Post hot working the steel is slow cooled to room temperature leading to the formation of lath martensite microstructure. The hot worked steel is softened for fabrication by full annealing between 830 to 885°C [2], where the microstructure obtained has spheroidized carbides in a ferritic matrix. The carbides in annealed condition show up M23C6 type carbide along grain boundary along with presence of M 7C3 type carbide [3-7]. Increasing the annealing time is reported to convert M3C to M7C3, which after long time exposure form M23C6 carbides. The annealed steel after fabrication is strengthened by hardening and tempering. Hardening is usually carried out by austenitising in the range 965°C to 1050°C followed by air cooling, oil or polymer quenching [2]. At low austenitizing temperature (965°C - 1000°C), the dissolution of carbides in austenite is not complete and residual
1
carbides along grain boundary impairs the mechanical properties [8-10]. At high austenitization temperature (1000°C - 1040°C), there can be severe grain growth which affects the mechanical properties. Desirable mechanical properties are achieved by tempering the steel between 200 to 750oC avoiding the temper embrittlement range between 350 to 550oC [11]. Tempering below embrittlement range retains good strength with poor toughness and tempering above embrittlement shows high toughness with good strength and the carbide formed during tempering is reported to be M23C6. Presence of grain boundary carbides in all the above processing impairs the sub- zero toughness and the corrosion resistance [12-13]. Increase in the austenitizing temperature or soaking time that ensures complete dissolution of carbides without grain growth improve the toughness of the steel. The 13%Cr steel with higher carbon content (0.45%), austenitized at high temperature greater than 1040°C showed lower carbide fraction but had grain growth that deteriorated the strength and toughness [14]. In general, austenitizing condition (1000°C - 1040°C) that dissolves the grain boundary M23C6 carbides in austenite without grain growth is desired [15-16]. In the high carbon 13%Cr (D2 steel), the influence of sub-zero cryogenic treatment (-196 and -73oC) on secondary carbides by accelerated martensite decomposition has been characterized for estimation of carbide volume fraction and correlated it with impact toughness and strength [17]. The influence of microstructure on fracture toughness of cryogenically treated D2 steel has been studied and the toughness observed to decrease with lowering of retained austenite and enhanced secondary carbides. The presence of primary carbide is reported to have significant fall in the toughness and deep cryogenic treatment improved toughness due to carbide distribution [18]. The carbide distribution as a function of various types of low and subzero temperature treatment in D2 type steel and its influence on fracture toughness and wear resistance has been correlated to the microstructural modification associated with retained austenite and secondary carbide distribution [19]. In another study on sub-zero treated vanadis tool steel, the carbide distribution have been characterised by SEM, TEM and XRD techniques. The tempering response shows no secondary hardening in the sub-zero treated samples [20]. In an extended study on vanadis 6 tool steel, the effect of refining the secondary carbides and retained austenite and their influence on fracture toughness and wear was determined [21]. In the present study, a combination of thermo-mechanical and heat treatment processing in 0.2%C-13%Cr steel have been developed to improve the sub-zero impact toughness of the 13%Cr-0.2%C martensitic stainless steel by modifying the carbide distribution. 2.0 Experimental details The steel was melted in a 35 MT electric arc furnace followed by ladle refining and vacuum degassing. Liquid steel was then cast in a four metric ton ingot casting (avg ingot size of 570 mm) through bottom pour up-hill technique. Two typical ingots were chosen for study. One ingot was directly hot rolled through a blooming mill to 235 X 215 mm semis followed by finish rolling mill to 170 mm diameter bar. The second ingot was initially cogged to 250 X 270 mm semis and then it was further rolled to 170 mm diameter bar. The overall reduction ratio from cast ingot was maintained at 14 in both the conditions. The hot working parameters are summarized in Table 2. After hot rolling, the bars were subjected to annealing at 850°C for 4h followed by furnace cool up to 590°C. The annealed microstructure was characterized using SEM and XRD. XRD was carried out on the annealed steel on a Bruker D8 machine, with step size of 0.02° and scan speed of 3s per step with Mo Kα filter. Cut bars of 170 mm length, were chosen for heat treatment studies in a laboratory scale muffle furnace. Three cut
2
bars, each in conventional rolled bars and cogged + rolled bars, were subjected to the following heat treatment. One set (conventional roll and cogged + rolled) was conventionally hardened by austenizing at 980°C and 1040°C soaked for 3h 15m followed by oil quenching. Another set of bars (conventional rolled and cogged + rolled) were initially austenitized at 1040°C for 3h 15m followed by oil quench. Post quenching, the steel bars were kept in the quenchant for 45 m. The bars were then subjected to a second hardening treatment by austenizing at 980 oC for 3h 15m followed by oil quenching. All the samples post hardening were subject to double tempering followed by oil cool. First tempering was done at 710°C for 5h 15m followed by a second tempering at 680°C for 5h 15m. The steel chemistry was determined using ARL 3460 spectrometer and gas content was obtained using LECO gas analysers. The hardness was measured using Vickers hardness machine under 200 g with diamond indenter in various conditions. Mechanical testing was carried out at mid radius location using Shimadzu tensile testing machine and the test sample conformed to ASTM E8 round sample of gage diameter 12.5 mm and gage length 50 mm. The impact properties was evaluated at sub-zero temperatures (-10°C) for various heat treated conditions using an FIE testing machine. Samples for impact test were prepared as per ASTM E23. The microstructure was evaluated using optical microscope and select samples were examined for SEM. Etchant used was Villella reagent. The SEM microstructure as typically shown in Fig.1, for the carbide distribution was quantified using Image J software at 1000X magnification to capture the finest carbides in the microstructure. 10 fields were examined for carbide fraction. The areal fraction was determined for each field by colouring the different phases from which the percentage of the individual phases are automatically calculated by the image analysis software. The standard deviation was also determined. The prior austenite grain size was determined by oxidising the grain boundaries in a furnace as per ASTM E112. Fractography images were captured after double tempering using SEM under two set of processing conditions. Brittle area fracture was measured using Olympus stereomicroscope as per ASTM A370. Table 1 Hot working parameters used in the present study
Condition
Semis Cogging/Rolling Temp, oC
Conventional rolling
Rolling to 235 X 215 mm at 1210oC
Cogged+ Rolled
Cogged to 250 X 270 mm at 1175oC
Deformation finish temperature o C
Post forging /rolling processing
Final Rolling cycle
980
Deep vermiculite cooling
Rolling to 170 mm dia at 1210oC
950
Stress relieved at 750°C
Rolling to 170 mm dia at 1210oC
3
Post rolling processing Deep vermiculite cooling; followed by annealing at 850oC Deep cooling in a vermiculite bed and annealing at 850°C
Fig. 1 Typical SEM microstructure at 1000X showing the carbides (white) and matrix (black) for quantifying carbide fraction. 3.0 Results and Discussion The present study examines the mechanical properties especially the sub-zero impact toughness in a 13%Cr0.2%C steel. The study uses ingot cast steel subjected to deformation routes (i) conventional hot rolling (ii) cogged + rolled. Rolled bar samples from both the wrought steel products, were subjected to a (i) Single hardening heat treatments at 980oC and 1040°C followed by oil quenching and (ii) a double hardening heat treatment involving (1040oC/ oil quench + 980oC/ oil quench) heat treatments. The hardened steels were subject to a standard two stage tempering heat treatment at 710oC followed by 680oC [2]. The mechanical properties have been examined with a focus on the low temperature impact toughness. The properties have been correlated to the composition and microstructure evolution during the processing. The tempering temperature of 710oC was chosen to achieve optimum combination of strength, toughness and corrosion resistance satisfying some technical end uses. 3.1 Effect of alloying addition on martensite formation The chemical composition of the present steel is shown in Table 2. Typically, the 13%Cr steel has carbon content between 0.15 to 0.3% where adequate strength, corrosion resistance and toughness is realized. In the present steel, the carbon was maintained at 0.18%. At this carbon, the martensite start (Ms) temperature is 297oC & the finish temperature (Mf) is 197oC, which is well above room temperature as per formula given below [22-23], this in turn reduces the presence of retained austenite on hardening. In addition lower carbon content produces lesser carbide fraction in matrix, which improves the toughness of the steel at moderate expense of strength. Lower carbon promotes predominantly M23C6 type carbide, while the higher carbon has potential to promotes M 7C3 (M = Fe, Cr) type carbides as well [10]. Ms (oC) = 540 – (497*C + 6.3*Mn + 36.3*Ni + 10.8*Cr + 46.6*Mo)
--------(1)
Mf (oC) = Ms – 100
--------(2)
As per the above equation, the Ms temperature is 297oC and the Mf temperature is 197oC, which is matching with data reported by Nakazawa for 0.2C-13%Cr steel [24]. Considering the effect of austenitising temperature on Ms
4
and Mf temperature, Nakazawa has reported a fall in Ms and Mf temperature with austenitising temperature below 1000oC, where the carbide dissolution is complete. In the present study, using the data of Nakazawa, the difference in Ms temperature is about 10oC and the difference in Mf temperature is about 15oC for the austenitising temperatures of 980 and 1040oC. Hence, this low variation in M s and Mf temperature is not expected to significantly alter the microstructural features of the retained austenite or the carbide fraction. Lower sulphur is maintained to get more isotropic properties especially improved transverse impact toughness which decreases with elongated sulphides. The phosphorous is maintained at low level to avoid temper embrittlement. Inherently, the 13% Cr content in steel gives good corrosion resistance and it also ensures good hardenability and the CCT diagram is shifted to the right, which enables martensite formation in the through thickness over a wide range of cooling rate including cooling in air even in thick sections [25]. The Al/N ratio was maintained above 2.5 to ensure N fixing by residual Al to form AlN which refine austenite grains during deformation. Table 2 Chemical composition (in wt.%) of the steel examined. Element Actual
C 0.18
Mn 0.85
Si 0.30
S 0.002
P 0.020
Cr 12.95
Ni 0.18
Mo 0.03
Al 0.028
N 0.0070
MS(°C) 297
Mf (°C) 197
3.2 Characterization of the annealed microstructure The 13%Cr steel is usually full annealed after hot working in the temperature range from 830 to 885oC to soften the matrix required for improved fabrication by machining. The microstructure obtained after annealing shows spheroidized carbides in ferrite matrix as reported in literature [3-7]. In the present study, the microstructure obtained after annealing of the bars showed fine grain boundary precipitates and coarse grain boundary chain precipitates in the ferrite matrix as shown in Fig. 2. The composition of precipitates was assessed using SEMEDAX [Table 3]. The EDAX spot analysis on chain precipitates in Fig. 2 shows a peak for carbon content compared to the matrix implying that it is carbide. As it is difficult to precisely determine composition of the carbide in SEM-EDAX due to the low atomic no. of carbon, alternate methods have been used to confirm the carbide. Analysis of Cr/Fe ratio was used to confirm the carbide [26]. The Cr/Fe ratio for the matrix where carbon is lean was estimated to be 0.27, while the chain type precipitate phases showed Cr/Fe ratio of 0.97 which was reported to imply the M23C6 type carbide [26]. The higher Cr/Fe ratio is observed in the precipitate carbide as more Cr is fixed by carbon than Fe atom compared to the matrix.
5
Fig. 2 SEM microstructure in the annealed condition. Annealed steel samples were subjected to XRD analysis as in Fig. 3. The results confirmed presence of ferrite peaks and absence of Cr7C3 and presence of some fine peaks of Cr23C6 type carbide. The peaks compared well with literature data [27-31]. Table 3 The EDAX spot analysis of the carbide in the steel (At%). Location Spot 1
Phase Matrix (ferrite)
C
Cr
Fe
Cr/C
(Cr+Fe)/C
Cr/Fe
14.04
*
17.20
63.62
1.22
5.756
0.27
*
16.25
61.30
1.174
5.607
0.265
Spot 2
Matrix (ferrite)
13.83
Spot 3
Matrix (ferrite)
14.30*
16.85
62.37
1.178
6.718
0.27
Spot 4
Precipitate
38.32
30.46
31.22
0.795
1,610
0.975
Spot 5
Precipitate
35.11
32.33
33.55
0.921
1.876
0.96
Spot 6
Precipitate
40.32
31.78
32.78
0.788
1.601
0.97
*Solubility of C in ferrite is less and atomic no of C is 6, so actual carbon in ferrite matrix may be less
Fig. 3 XRD analysis showing ferrite peaks with minor Cr23C6 type carbide peaks in the annealed samples of conventionally rolled sample. All the peaks of Cr23C6 were indexed at the expected peak location although only peaks at planes (511), (660) and (800) could show up due to low volume fraction. This means that annealed condition has Cr23C6 carbides predominantly.
3.3 Effect of heat treatment on prior austenite grain size The conventionally rolled bar and the cogged + rolled bar subject to heat treatment conditions, were examined at various conditions of hardening for prior austenite grain sizes as in Fig. 4. The prior austenite grain size of the conventional rolled and cogged + rolled steel samples hardened at 980 oC showed considerable variations. The grain size of the conventional rolled steel had large patches of coarse grain size with an ASTM No 6 as shown in Fig. 4(a), while the cogged + rolled steel showed a more uniform grain size ASTM No.6.5 as in Fig. 4(b). The
6
reason for fine grain size in cogged + rolled condition may be attributed to higher deformation strains towards the core that enables higher degree of recrystallization compared to hot rolling where deformation is concentrated towards the surface. Hardening at 1040°C resulted in coarsening of the prior austenite grains in both the conventional rolled [ASTM no 4.5] and cogged + rolled [ASTM 5.5] samples as seen in the Fig. 4(c-d). It is observed that the higher temperature of heat treatment has led to the coarsening of the grains. The results obtained were close to the values reported in literature [32]. Double hardening heat treatment at 1040°C/OQ + 980°C/OQ resulted in a refinement of the prior austenite grains in conventional rolled material (ASTM 7.0) and cogged + rolled material (ASTM7.5) as shown in Fig. 4(e-f). The probable reason for finer grain in the double hardening treatment may be attributed to generation of defect structure in the matrix during martensite formation in the first hardening at 1040oC/OQ and the defect structure generated in this stage gets recrystallized too at the second hardening condition at 980oC/OQ. A similar studied on double hardening was carried out on 16Cr-2Ni steel [33-34], where double hardening refined the grain size and improved impact properties. However the result did not match with present study but trend observed to be same.
Fig. 4 Grain size of (a) Conventional rolled sample hardened at 980°C [Grain size: ASTM No.6] (b) Cogged + Rolled sample hardened at 980oC [Grain size: ASTM No.6.5] (c) Conventional rolled sample hardened at 1040° C showing [Grain size: ASTM No.4.5] (d) Cogged + Rolled samples hardened for 1040oC showing [Grain size: ASTM No.5.5] (e) Conventional rolled sample hardened at 1040 + 980°C [Grain size: ASTM No.7] (f) Cogged + Rolled samples hardened for 1040oC + 980°C [Grain size: ASTM No.7.5] 3.4 Characterization of the as-quenched hardness and microstructure:
7
The samples examined for the grain sizes at various hardening heat treatment were subsequently evaluated for their as-quenched hardness, microstructures and carbide fraction [Fig. 5(a-f) and Table 4]. It is seen from Table 4 that the as-quenched hardness at 8 different locations on each sample varied between 458 and 567 HV. The conventional hot rolled sample, hardened at 980oC showed the least hardness of 458 ± 15 HV, while hardening it at 1040oC showed a hardness of 474 ± 12 HV. The conventional rolled material subject to double hardening [1040°C/OQ + 980°C/OQ] showed further improved hardness up to 531 ± 8 HV. The steel processed in the cogged + rolled condition shows about 30 to 60 HV more hardness than the conventional rolled material at the corresponding double hardened condition. The high hardness in the steel implies that carbon dissolution in the matrix and finer grain sizes. Similar results were reported in literature [35]. But the discrepancy in hardness value was attributed to minor chemical compositional changes. The microstructures of the as-quenched samples examined using SEM showed carbides in a matrix with martensitic lath [Fig. 5(a-d) & Table 4]. The SEM microstructure of the conventional rolled and cogged + rolled sample subjected to single hardening at 980oC show coarse residual chain type carbides along with matrix carbides [Fig. 5(a-b)].The matrix has coarse lath martensite probably due to the coarse austenite grain structure as shown in Fig. 4(a-b). Single hardening at 1040oC for both the processing condition further decreased the matrix carbides and thinning down of chain carbides took place [Fig. 5(c-d)]. There was further coarsening of lath martensite observed due to grain coarsening as compared to single hardening at 980°C shown in Fig. 4(c-d). This dissolution of carbides at 1040°C led to increased as-quench hardness compared to 980oC hardening. The microstructure of the samples subjected to double hardening heat treatment [1040°C/OQ + 980°C/OQ] in conventional rolled and cogged+ rolled conditions show complete absence of grain boundary chain carbides [Fig. 5(e-f)] and the lath martensitic microstructure is finest corresponding to finer grain sizes [Fig. 4(e-f)]. The carbide content and distribution is least in the cogged and rolled condition comparison to the conventional rolled condition after double hardening condition. The present study shows that the deformation involving cogged + rolled condition followed by double hardening redistributes the carbides (both residual and grain boundary) more effectively than that in other conditions. In spite of the tempering condition remaining the same, the amount and distribution of the precipitated carbides during tempering varied with the deformation processing and hardening conditions. This shows that careful selection of deformation and hardening condition can redistributes the carbides that influences the impact toughness.
8
Table 4 As-quench hardness correlated with microstructure and carbide fraction at various hardening conditions. Process Route Rolled
Hardening condition (oC)
As quenched hardness (HV) 458 ± 15
Microstructure observation Carbides sparsely distributed in the matrix and there is the presence of grain boundary carbide chain. Remnants of lath martensite structure are observed. Carbide fraction (Std. dev.) = 4.36% (±0.21)
980°C/OQ Cogged+ rolled
474 ± 12
Rolled
490 ± 10
Carbides in matrix have dissolved while no remnant grain boundary carbide chain seen. Strong remnant of lath martensitic microstructure. Carbide fraction (Std. dev.) = 2.32% (±0.15) Carbides in the matrix is very coarse and dissolution of carbide chains seen. Martensite laths coarsening seen. Carbide fraction (Std. dev.) = 1.22% (±0.06) Carbides are totally dissolved and the coarsening of lath martensite; Carbide fraction (Std. dev) = 0.84%(±0.06) Microstructure was completely free of K1 carbides and fine lath martensite was uniformly present throughout the matrix; Carbide fraction (Std. dev.) = 0.56% (±0.08) Carbide fraction (Std. dev) = 0.52% (±0.05)
1040°C/OQ Cogged+ rolled
505 ± 6
Rolled
531 ± 8
Cogged+ rolled
1040°C/OQ+ 980°C/OQ
567 ± 6
Fig. 5 SEM microstructure of as-quenched samples (a) Rolled subject to 980oC/OQ (b) Cogged + Rolled subject to 980oC /OQ (c) Rolled subject to 1040oC/OQ (d) Cogged + Rolled subject to 1040oC/OQ (e) Rolled subject to (1040oC/OQ+980oC/OQ ) (f) Cogged + Rolled subject to (1040oC/OQ+980°C/OQ).
3.5 Tempered Microstructure and Mechanical Properties The mechanical properties of the steels in the conventionally rolled and cogged + rolled steels were examined in the various hardening treatment followed by tempering. The tempering involved standard tempering at 710oC followed by a second tempering of 680oC in all the samples [2]. At this tempering temperature, a good
9
combination of strength and sub zero toughness is realized for a wide range of applications. This tempering temperature is above the secondary hardening range and the expected phases are a ferritic matrix phase free of carbon and carbide precipitated out that favours high toughness.
The mechanical properties of the steel under the various hot working conditions after the various hardening followed by standard tempering are shown in Table 5. The conventional rolled and cogged + rolled steel hardened and tempered condition showed comparable strength and ductility properties. The conventional rolled steel show impact toughness lower than the cogged + rolled steel at similar tempering condition. The conventional rolled and cogged + rolled samples subjected to double hardening heat treatment [1040oC/OQ + 980oC/OQ] followed by the tempering treatment showed higher strength and higher toughness compared to the single hardening heat treatment. The impact toughness of cogged + rolled steel is found to be doubled along with enhanced strength. Although a double hardening process may be expensive, the properties are attractive. When the toughness is very high, the tempering temperatures can be lowered to lower the toughness, which consequently results in enhanced strength. Increased strength enables weight reduction of the structure, savings in material and cost. In order to meet the sub-zero impact toughness, micro alloying of the steel with expensive Ni and Mo adopted by some of the end users can be avoided by the use of the double hardening heat treatment.
The optical microstructure of the conventional rolled and cogged + rolled steels after tempering at low and high magnification is shown in Fig. 6(a-h). Tempered martensite microstructure with carbides precipitated on the ferrite lath boundaries along with carbides chain along grain boundaries can be observed. The carbide formed for these grades at this temperature range was reported to be M23C6 carbides [11,36-37]. The conventional rolled steel hardened at 980oC after tempering showed fine laths containing trans-granular carbides along with grain boundary carbides as in Fig. 6(a-b). The cogged + rolled steel hardened at 980oC after tempering showed more ferritic zones with some dispersed grain boundary carbides [Fig. 6(c-d)]. The conventional rolled steel and the cogged + rolled steel hardened at 1040oC followed by tempering showed that in spite of higher hardening temperature the carbides have not dissolved in some zones [Fig. 6(e-h)]. The ferrite laths are coarser due to the coarser austenite grains are coarser as in Fig. 4(c-d). The extent of thickening of grain boundary carbides hardened at 1040 oC was less in the cogged + rolled sample compared to the rolled sample. The rolled steel sample subjected to the double hardened [1040 oC/OQ + 980°C/OQ] treatment followed by tempering showed much coarser ferritic lath matrix with highly diffused grain boundary carbides as in Fig. 6(i-j). The cogged + rolled sample with double hardening [1040oC/OQ + 980°C/OQ] followed by tempering showed high dispersion of ferritic lath with very fine carbides [Fig. 6(k-l)]. The improved toughness in the cogged + rolled steel in the double hardened may be attributed to the finer grain size [Fig. 4(e-f)], absence of grain boundary carbide and fine distribution of matrix carbide [Fig. 5(e-f)].
The hardened and tempered samples used for optical microscopy were further examined for finer details using SEM. The conventional rolled steel and the cogged + rolled steel hardened at 980 oC followed by tempering show coarse carbide distribution in the matrix [Fig. 7(a-b)]. The conventional rolled samples show prominent grain boundary carbides in Fig. 7(a) than the cogged + rolled sample in Fig. 7(b). The toughness in rolled sample was
10
less [CVN = 16 to 18 J] compared to cogged + rolled sample after single hardened at 980°C and 1040oC tempered [CVN = 20 to 23 J]. This may be attributed to the presence of higher amount of grain boundary carbides. In contrast, the cogged + rolled material shows reduced carbides along the grain boundaries, lower amount of dispersed carbides and a coarser ferrite lath in the matrix [Fig. 7(b)] at the same conditions of heat treatment that led to high toughness compared to the rolled steel. This implies that during cogging and subsequent hardening there is favourable carbide dissolution and redistribution in the matrix. The cogging process introduces higher effective strains in the core and the deformation penetration creates more sub angle and high angle grain boundaries during recrystallization. In addition, there is higher adiabatic rise in temperature due to deeper strains in cogging. All these factors enable carbides to effectively dissolve in the matrix and there is coarsening of laths indicative of grain refinement.
Table 5 Mechanical properties of the steel evaluated at the various processing routes tempered at the standard tempering at 710°C/5h 15m/oil cooled + 680°C/5h 15m/oil cooled Mechanical properties Process Route
Heat treatment
Cogged + Rolled
Single hardened at 980°C/3h15m/ oil quenched; Single hardened at 980°C/3h 15m/ oil quenched.
Conventional rolled
Single hardened at 1040°C/3h 15m/ oil quenched +
Cogged + Rolled
Single hardened at 1040°C/3h 15m/oil quenched
Conventional rolled
% Mean carbide (Std dev)
Hardness (HV)
YS (MPa)
UTS (MPa)
%E
%RA
8.32 ( ±0.16)
218 ± 6
570
712
20.3
63.5
18
5.34 (±0.14)
220 ± 4
565
715
20.6
62.6
23
210 ± 5
556
714
21.2
64.2
16
2.65 (±0.09)
218 ± 4
575
725
21.3
63.5
20
3.78 (±0.11)
CVN (J)
Conventional rolled
Double hardened at 1040°C/3h 15m/ oil quenched + 980°C/3h 15m/oil quenched
3.05 (±0.08)
227 ± 5
620
752
26.7
66.1
29
Cogged + Rolled
Double hardened at 1040°C/3h 15m/ oil quenched + 980°C/3h 15m/oil quenched
2.16 (±0.06)
225 ± 6
631
765
26.4
66.6
41
The toughness value is least in the conventional rolled sample subject to single hardening at 1040°C followed by tempering, compared to the cogged + rolled sample at same tempering condition. Single hardened samples in rolled and cogged + rolled routes show impact toughness value between 16 and 23 J which is closer to specification and reports in literature [38]. Least toughness of 16 J was observed at 1040°C followed by tempering
11
and it could be attributed to the significant grain coarsening and also presence of residual carbides in the grain boundary [Fig. 4(c-d) & Fig. 5(c-d)]. The SEM microstructure of the samples subjected to double hardening heat treatment [1040oC/OQ + 980°C/OQ] in the rolled and the cogged + rolled samples show that the carbides are almost completely dissolved and are re-precipitated as very fine carbide along with coarser ferrite laths [Fig. 7(e-f)] due to finer prior austenite grain sizes [Fig. 4(e-f)]. The conventional hot rolled steel still shows some remnant grain boundary carbides [Fig. 7(a)] which results in lower toughness than in the double hardened steel [Fig. 7(e)]. The cogged and rolled steel in the double hardened condition shows complete absence of grain boundary carbide and a very fine carbide distribution in the matrix and this has given almost double the impact toughness [Fig.
7(e-f)].
12
Fig. 6 Optical microstructure of the steels subject to tempering (a)-(b) Conventional rolled and single hardened at 980oC [CVN = 18J]; (c)-(d) (Cogged + rolled) steel subjected to single hardening at 980oC [CVN = 23J]; (e) – (f) Conventional rolled and single hardened at 1040 oC [CVN = 16J]; (g)-(h) (Cogged + rolled) steel subjected to single hardening at 1040oC [CVN = 20J]; (i)-(j) Conventional rolled steel double hardenened [1040°C/OQ + 9800C/OQ] [CVN = 29J] (k-l) (Cogged + rolled) steel double hardenened [1040°C/OQ + 9800C/OQ] [CVN = 41J].
Fig. 7 SEM microstructure after double tempering tempering (a) Conventional rolled product hardened at 980°C [CVN = 18J] (b) Cogged + rolled product hardened at 980°C [CVN = 23J] (c) Conventional rolled product hardened at 1040°C [CVN = 16J] (d) Cogged + rolled product hardened at 1040°C [CVN = 18J] (e) Rolled product hardened at 1040°C + 980°C [CVN = 29J] (d) Cogged + Rolled product hardened at 1040°C +980°C [CVN = 41J]. The ferrite laths in the double hardened cogged + rolled steel are extremely fine compared to the single hardened cogged and rolled steel. The cogged and rolled steel with double hardening and tempering revealed the best impact toughness which can be the bench mark microstructure. Thus, the fine carbides in cogged and rolled steel
13
is due to dissolution and re-precipitation of the carbide in deeply worked steel compared to conventional rolled steel. The lath structure of the ferrite in the single hardening is further refined by double hardening.
3.6 Effect of carbide distribution on mechanical properties of 13%Cr steels The carbide distribution observed in the SEM microstructures was further quantified in the single hardened and double hardened followed by the standard tempering. The percentage of the carbide in the microstructure and the hardness in the various processing conditions was estimated using the SEM images using the image analysis software. The carbide fraction was calculated at five different locations and the average carbide fraction is reported along with standard deviation results are shown in Tables 4 & 5 and Fig. 8. The results show that the carbide content in the tempered condition is much higher than in the hardened condition, as expected. The conventional rolled sample austenitized at 980oC, in the as-quenched condition show a carbide volume fraction of ~ 4%, which enhances to ~ 8% on tempering. At this condition, both the as-quenched and tempered hardness are the lowest compared to other heat treatment condition as shown in Fig. 8. The volume fraction of carbide is ~2% for cogged + rolled sample in the as-quenched condition hardened at 980oC, which on tempering enhances to ~ 5%. The volume fraction of carbide is ~1% for both conventional rolled and (cogged + rolled) sample in the asquenched condition hardened at 1040oC, which on tempering enhances to ~ 3.5 and 3% respectively.
For the
double hardened steel, the volume percentage of carbide in the as-hardened condition in conventional rolled sample is 0.5% which increases to about 3% where the tempered hardness is the highest. The cogged + rolled samples after double hardened in the as-quenched condition show about ~ 0.5% carbide, which on tempering becomes ~ 2%. High deformation and double hardening create more defect structure in the matrix carbides are fine on tempering, which enhances the strength and the toughness.
Fig. 8 The carbide fraction estimated in SEM and the hardness at various processing conditions.
The carbide content in the microstructure as a function of the hardness obtained under after the two processing condition was examined as in Fig. 9. The as-quenched hardness at various conditions was found to decrease as the carbide volume in the microstructure increased. This means that as carbide content is increased less carbon is in the martensite leading to lowering of hardness. In the tempered condition, the hardness values are almost similar but the carbide volume fraction show considerable variation with processing condition. This is important because
14
lower carbide content will ensure that there is lesser carbide precipitation at the same hardness levels. This lowering of carbide content in the double hardened and conventional rolled and cogged + rolled sample gives higher hardness as more carbon is available in the matrix to harden the martensite and toughen with less brittle particles in the matrix and in the grain boundaries. The mechanical properties in the tempered condition was plotted as a function of the total carbide content irrespective of the processing condition in Fig. 10 and it is seen that as the carbide content increases there is moderate loss in strength and ductility, while there is significant loss in toughness. Hence, to achieve highest strength and good toughness level, thermo mechanical processing and heat treatment condition should ensure finer and low volume carbides in the matrix and the grain boundary. Tempering should not precipitate coarse carbides. Thus, this study has brought out the best microstructure for 13%Cr steel that can give improved strength and subzero impact toughness.
Fig. 9 Carbide fraction at various conditions of heat treatment compared with matrix hardness.
Fig. 10 Effect of carbide content in the tempered martensitic microstructure on the mechanical properties of the 13% Cr steel.
3.7 Fractography of the broken impact sample
15
Fractography was carried out on the broken impact test samples to understand the fracture mode under various processing conditions subjected to single hardening and double hardening condition followed by tempering. The fractography results are shown in Fig. 11(a-f).
It is seen that, in conventional rolled sample single hardened at 980°C and 1040°C followed by tempering shows higher degree of cleavage facets as in Fig. 11(a&c) the corresponding impact toughness is the least. The cogged + rolled steel samples subject to single hardening at 980°C followed by tempering showed mixed mode failure with lower cleavage facets [Fig. 11(b)] than the conventional rolled sample. Hence, a moderate improvement in toughness was observed. The conventional rolled double hardened [1040 oC/OQ + 980°C/OQ] followed by tempering still higher degree of dimples but some isolated spots of cleavage facets [Fig. 11(e)]. The dimples were coarse. The cogged and rolled sample after double hardening [1040 oC/OQ + 980°C/OQ] and tempering showed fully dimple microstructure without cleavage facets [Fig. 11(f)]. This has resulted in highest toughness in the steel. The fractographic features are attributed to the grain size and carbide morphologies. The cogged + rolled samples exhibit fine grains compared to conventional rolled samples. Secondly, there is absence of grain boundary carbides in the cogged + rolled sample compared to the conventional rolled steel which results in less crack initiation sites and lower propagation rate due to fineness of the carbide. .
Fig. 11 Fractography images of the impact test samples hardenede and tempered steels at (a) Conventional rolled steel austenitized at 980°C/OQ/tempered [CVN = 18J] (b) Cogged + Rolled steel austenitized at 980°C/OQ / tempered [CVN = 23J] (c) Conventional rolled steel austenitized at 1040°C/OQ/tempered [CVN = 16J] (d) Cogged + Rolled steel austenitized at 1040°C/OQ / tempered [CVN = 18J] (e) Conventional rolled steel double hardened [1040°C/OQ + 980°C/OQ]/tempered [CVN = 29J] (f) Cogged + Rolled steel double hardened [1040°C/OQ + 980°C/OQ]/tempered [CVN = 41J]
16
The relation between the brittle area fracture, carbide fraction and impact toughness was evaluated. The brittle area fracture was measured on cross section of three broken impact samples using stereomicroscope. The standard deviation was also calculated and plotted in the Fig. 12. The typical stereo micrograph of the fractured sample at each condition showing the brittle area is also shown in Fig. 12. As the carbide fraction increases there is less ductile area compared to brittle area. The shear lip is prominent at lowest carbide density. From the Fig. 12 it is observed that, as the carbide fraction increases, the brittle area fracture increases and toughness decreases. In the Fig. 12 Point c, d represent the high brittle fracture area where the corresponding impact fracture observed is low. This is due to grain growth occurring at 1040°C for both the processing conditions.
Fig. 12 Tempered carbide volume fraction (%) as a function of mean brittle area fracture (mm 2) and impact toughness (J). Pictures in the graph show the stereo image of (a) Rolled 980°C + tempering [CVN = 18J] (b) Cogged + Rolled 980°C + tempering [CVN = 23J] (c) - Rolled 1040°C + tempering [CVN = 16J] (d) Cogged + Rolled 1040°C + tempering [CVN = 18J] (e) Rolled 1040+ 980°C + tempering [CVN = 29J] (f) Cogged + Rolled 1040 + 980°C + tempering [CVN = 41J] As a future activity, it is of interest to extend the influence of thermo-mechanical processing, double hardening and double tempering on other high carbon stainless tool steels such as D2, AISI 440 etc., where refinement of carbides are likely to improve their toughness properties and can be combined with sub-zero treatment as well [17-21]. In summary this study has given the best process condition in terms of deformation and heat treatment along with the best microstructure towards achieving excellent improvement in strength and toughness in 13%Cr martensitic stainless steel. A fine grained sample obtained by cogging or cogged + rolled route, with double hardening will refine the grain structure, the lath morphology, the extent of carbide precipitation, and distribution of carbides precipitation that ultimately improves the strength and toughness.
4.0 Conclusions
1. The improvement in the mechanical properties with a focus on impact toughness in 13%Cr martensitic stainless steel has been examined in the steel processed through conventional hot rolling and cogged + rolled
17
conditions followed by subsequent heat treatments involving single hardening at 980oC and 1040oC and also a double hardening at 1040oC/OQ followed by 980oC/OQ. The hardening was followed by a two stage tempering at 710°C and 680°C. 2. The hot worked and annealed steel shows ferrite matrix with Cr 23C6 type carbides in grain boundary and matrix. 3. The samples that were subject to single hardening at 980oC, 1040°C or double hardened at 1040oC/OQ + 980oC/OQ which influenced the carbide distribution in the steel leading to a strong influence on impact toughness of the steel. 4. Highest toughness was realized in the cogged and rolled steel subjected to double hardening followed by standard tempering which is attributed to the lowest carbide content and fineness along the grain boundary and the matrix which gave a fractography feature of fine dimples without cleavage facets in the matrix. 5. Lowest toughness was obtained in the conventionally rolled steel subjected to single hardening at 980 oC, 1040°C followed by standard tempering which led to highest cleavage facets associated with highest carbide content with highest distribution along the matrix and the grain boundary. 6. Moderate improvement of toughness sample is observed in cogged + rolled samples subject to single hardened [980oC and 1040°C] followed by tempering than the conventionally rolled samples subjected to double hardening [1040oC/OQ + 980oC/OQ] and tempered conditions. The toughness could be correlated to the presence of carbides along grain boundary and matrix that led to some cleavage facets in the fractography. 7. The present study brings out the best thermo-mechanical processing condition, heat treatment condition, the bench mark microstructure to be targeted to improve the impact toughness.
Acknowledgments The authors wish to thank the constant encouragement and support provided by Dr Baba Kalyani, Chairman & Managing Director and Mr R K Goyal, Managing Director, Kalyani Carpenter Special Steels Pvt. Ltd., Pune. Thanks are due to Department of Scientific and Industrial Research for their support of technical activities at KCSSPL R&D centre.
References [1] Cássio Barbosa, Ibrahim Abud, Recent Pat. Corr. Sci., 3 (2013), 27-38 [2] H. Chandler, Heat Treaters Guide: Practice and Procedure for Iron and Steel, 2 nd ed., ASM international, (1994), 775 [3] A. Hede and B. Aronson, J. Iron Steel Inst., (1969), 1241-51 [4] T. Fujita and N. Takahashi, Trans. ISIJ, 18 (1978), 702-11 [5] A. Bja¨rbo: Scand. J. Metall., 23 (1994), 103-12 [6] A. Bjärbo, Mats Hättestrand, Metall. Mater. Trans. A, 32 (2001), 19-27 [7] B. Mintz, J. M. Arrowsmith, ASTM, STP 706 (1980), 313-335 [8] R. Gullberg, J. Iron Steel Res Int., (1973), 59 – 65 [9] L. D. Barlow, J. Mater. Eng. Perform., 21 (2012) issue no 7, 1327 – 1336 [10] H. Bhadeshia, Steels: Microstructure and Properties, 1st ed., Butterworth – Heinemann, (2006), 15-20 [11] V. Anil Kumar, R. K. Gupta, M. K. Karthikeyan, P. Ram Kumar, P. P. Sinha, Mater. Sci. forum , 710 (2012), 489-494 [12] A. Nasery Isfahany, H. Saghafian, G. Borhan, J. Alloys Compd, 509 (2011), issue no 9, 3931-3936 [13] A.F. Candelaria, C. E. Pinedo, J. Mater. Sci. Lett., 22 (2003), 1151-1153 [14] C. Garcia de Andres, L.F. Alvarez, V. Lopez, J. Mater. Sci. Lett., 33 (1988), 4095–4100.
18
[15] C. Garcia - De - Andris, L. F. Lavarez, First International Conference on Phase Transformation, (1996), 82 [16] Ming Laura Xu: Secondary carbide dissolution and coarsening in 13%Cr Martensitic Stainless steels during austenization, PhD dissertation, (2012), 1-156 [17] D. Das, A.K. Dutta, K.K. Ray, Mater. Sci. Engng., A527(2010), 2182 - 2193. [18] D. Das, R. Sarkar, A.K. Dutta, K.K. Ray, Mater. Sci. Engng., A528 (2010), 589 – 603 [19] D. Das, K.K.Ray, Mater. Sci. Engng., A541 (2012),45 – 60 [20] P. Jurci, M. Domankova, L. Caplovic, J. Ptacinova, J. Sobotova, P. Salabova, O. Prikner, B. Sustarsic,D. Jenko, D. Vacuum, 111(2015),92-101. [21] J. Sobotova, P. Jurci, I .Dloughy, Mater. Sci. Engng., A652 (2016), 192-204. [22] Steve Lampman: Weld Integrity and Performance: A Source Book Adapted from ASM International, 1st ed, ASM International, (1997), 417 - 442 [23] J. E. Ramirez, Weld J, 86 (2007), 125s – 134s [24] K. Nakazawa, Trans Jpn Inst Met, 4 (1963), 200 - 205 [25] M. Atkins, Atlas of Continuous Cooling Transformation Diagrams for Engineering Steels, 1 st ed., ASM, (1993), 145-150 [26] K. Wieczerzak, P. Bała, M.Stępień, G. Cios, T. Kozieł, Arch. Metall. Mater., 60 (2015), issue no 2, 779-782 [27] E. Jartych, T. Pikula, D. Oleszak, Hyperfine interact, 192 (2009), 117-124 [28] Märt Kolnesa, Jüri Pirsoa, Jakob Kübarseppa, Mart Viljusb, Rainer Traksmaab, Proceedings of the Estonian Academy of Sciences, 65 (2016), issue no 2, 138-143 [29] Arūnas Baltusnikas, Rimantas Levinskas, Mater. Sci., 12 (2006), issue no 3, 192-198 [30] M. Hillert et al. J. Iron steel inst., 209 (1971), 49 – 66 [31] Mehmet Erdem, Mehmet Altuğ,Mustafa Karabulut, Int. J. Adv. Manuf. Technol., (2015), 1-15 [32] Toshihiro Tsuchiyamas, Etsuo Takaki and Sadavuki Nakamura, ISIJ Int., 81 (1995), issue no 2, 59-64 [33] J. K. Lia, Stainless steels and Introduction and their recent development, 1 st ed., Bentham EBooks, (2012) 41 – 51 [34] K.P. Balan, A. Venugopal Reddy, D.S. Sarma, J. Mater. Eng. Perform, 8 (1999), issue no 3, 385-393 [35] Toshihiro Tsuchiyamas, Etsuo Takaki, Sadavuki Nakamura, ISIJ Int. 80 (1994), issue no 12, 68-73 [36] B Maio, D.O. Northwood, L.C. Lim, M.O. Lai, Mater. Sci. Eng. A, 171 (1993), 21-33 [37]Gopa chakraborty, C. R. Das, S. K. Albert, A. K. Bhaduri, V. Thomas Paul, G. Panneerselvam, Arup Dasgupta, Mater. Character., 100 (2015), 81-87 [38] G. V. Prabhu Gaunkar, A. M. Huntz, and P. Lacombe, Met. Sci., 14 (1980), issue no 7, 241-251
19