- Email: [email protected]
Design of wear resistant boron-modified supermartensitic stainless steel by spray forming process
Materials & Design 83 (2015) 214–223
Contents lists available at ScienceDirect
Materials & Design journal homepage: www.elsevier.com/locate/matdes
Design of wear resistant boron-modified supermartensitic stainless steel by spray forming process G. Zepon a,⇑, A.R.C. Nascimento a, A.H. Kasama b, R.P. Nogueira c,d, C.S. Kiminami e, W.J. Botta e, C. Bolfarini e a
Programa de Pós-Graduação em Ciência e Engenharia de Materiais, Universidade Federal de São Carlos, Rod. Washington Luiz, km 235, 13565-905 São Carlos, SP, Brazil Centro de Pesquisa & Desenvolvimento Leopoldo Américo Miguêz de Mello (Cenpes), Avenida Horácio Macedo, 950, 21941-915 Rio de Janeiro, RJ, Brazil Univ. Grenoble Alpes, LEPMI, F-38000 Grenoble, France d CNRS, LEPMI, F-38000 Grenoble, France e Departamento de Engenharia de Materiais, Universidade Federal de São Carlos, Rod. Washington Luiz, km 235, 13565-905 São Carlos, SP, Brazil b c
a r t i c l e
i n f o
Article history: Received 19 January 2015 Revised 5 May 2015 Accepted 1 June 2015
Keywords: Spray forming Supermartensitic stainless steel Borides Wear resistance Corrosion resistance
a b s t r a c t In this paper the chemical composition of the supermartensitic stainless (SM) was modified with the addition of small boron contents (0.3, 0.5 and 0.7 wt.%) and processed by spray forming aiming the development of functionalized stainless steel with higher wear resistance. The addition of boron to the SM leads to the formation of continuous network of M2B type borides uniformly distributed in the refined microstructure promoted by the spray forming process. The wear resistance was evaluated by two different methodologies: (1) the standardized dry sand/rubber wheel test (ASTM G65); and (2) a plate-on-cylinder (POC) wear test which was designed to simulate in laboratorial scale the tribosystems found in wear of risers and casings. It was shown that the wear mechanisms that take place in both tests are quite different, but in all cases increasing the boron content is always accompanied by an increase in the wear resistance. Electrochemical analyses were performed to evaluate the corrosion resistance of the designed alloys. It could be seen that corrosion properties similar to the commercial SM can be achieved in the SM modified with 0.7 wt.% of boron if an over content of chromium is added to the chemical composition. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction The production of oil in the pre-salt layers in the coast of Brazil brought new challenges concerning the development of materials which withstand the severe work conditions (mainly wear and corrosion issues) found during drilling and exploitation operations. These extreme conditions motivate the development of new materials with improved wear and corrosion properties, aiming their application as coatings for pipes and parts used in such operations. Drilling risers (steel pipes that links the well head on the sea floor to the platform) and casings (steel pipes used to protect the wall of wells during the drilling operations), for example, are frequently subjected to severe wear conditions due to the contact with rotating tool joints of the drill pipes. Moreover, these pipes work in contact with rock debris resulting from the drilling process and drilling fluids (water or oil based fluids containing mainly bentonite) which are usually rich in chlorides, increasing the susceptibility ⇑ Corresponding author. E-mail address: [email protected] (G. Zepon). http://dx.doi.org/10.1016/j.matdes.2015.06.020 0264-1275/Ó 2015 Elsevier Ltd. All rights reserved.
of failure caused by wear and corrosion. Since mid 1990’s the weldable 13Cr martensitic stainless steel grades, also called supermartensitic stainless steel (SM), has been increasingly applied in seamless pipes for drilling, casing, and tubing for application in oil and gas fields presenting higher corrosive environments [1–3]. Such stainless steel grades are based on the Fe–Cr–Ni–Mo system with up to 13 wt.% of Cr, 4–6 wt.% of Ni, 0.5–2.5 wt.% of Mo, low amounts of carbon, nitrogen, phosphorus and sulfur (C 6 0.02 wt.%, N, P, S 6 0.03 wt.%). The optimized microstructure of the material is free from d-ferrite and offers good corrosion resistance in environments containing CO2 and H2S. Typical values of mechanical properties of the SM grades are: 25–32 HRC, 650–750 MPa of 0.2% yield strength, 880–950 MPa of tensile strength, elongation at rupture up to 20% and impact energy up to 100 J [4,5]. In spite of their good mechanical and corrosion properties, stainless steel grades usually presents low wear resistance. Wear resistance is not an intrinsic property of the materials, but a tribosystem property where the materials in contact, type of reciprocal movement, relative speed, load levels, environmental conditions, presence of abrasive particles and lubricants can play
G. Zepon et al. / Materials & Design 83 (2015) 214–223
important roles in the wear behavior of a component. Based on this, from an engineering point of view, it is advisable testing the wear properties of materials reproducing closely the real operational condition. Spray forming process is an advanced casting process which allows the manufacture of near net shape preforms of alloy compositions that are problematical in conventional casting processes [6,7]. It has been reported that spray-forming is a useful tool for generating high-alloyed materials, such as, for instance, AISI D2 and AISI M3 high-alloyed tool steels, with reduced macro-segregation, fine microstructure and uniformly distribution of carbides and second phases [8]. Spray forming has been used to develop several materials, such as aluminum alloys (Al–Si, Al–Mg alloys) [9–12], iron-based alloys (high chromium cast iron and high speed steel) [13–15] and metal-matrix-composites (MMC’s) [16,17] with improved wear resistances. In all cases the wear properties are improved by the microstructural refinement and uniformity promoted by the spray forming process. Kasama et al. [14] and Matsuo et al. [15], for instance, showed that sprayed-formed high chromium cast irons present a substantial refinement in the microstructure scale when compared to the same conventionally cast alloy. The presence of fine M7C3 type carbides uniformly dispersed in an austenitic/martensitic matrix was effective to improve considerably both the abrasive and the sliding wear resistances. Since 1960’s it is well known that boriding is one of the most effective methods for increasing the wear resistance of steel parts due to the formation of hard borides such as Fe2B (1600 HV) and FeB (1800 HV) [18,19]. In the last years several authors reported the development of boron modified iron-based alloys to be used in coatings or parts subjected to severe wear conditions [20–24]. Recently, we have reported that addition of small quantities of boron in the chemical composition of spray-formed supermartensitic stainless steel led to the formation of M2B type borides (where M = Fe, Cr, Ni, Mo), which was very beneficial to the abrasive wear resistance of the alloy [22]. Moreover, since boron is a low-cost alloying element (iron–boron alloys), its use in the chemical composition of steels to generate enhanced properties is also economically interesting. In this paper we report the complete design of the wear resistant alloy based on the chemical composition of the supermartensitic stainless steel modified with boron additions and produced by spray forming process, aiming the achievement of the advantageous microstructure characteristics of spray-formed materials, i.e., refined microstructure and high level of microstructural homogeneity. The target composition of the alloy was designed aiming to achieve the best combination of wear and corrosion properties. Bearing in mind that wear is a tribosystemic property, two different types of wear tests were performed. The dry sand/rubber wheel abrasive wear test (ASTM G65-04), and a plate-on-cylinder-POC wear test, which simulates in laboratorial scale the tribosystem found when drilling risers and casings are worn by the contact of the rotating drill pipe. The last wear test performed and the models used to determine the material’s wear properties were based on real scale wear tests reported by petroleum engineers [25,26].
2. Experimental procedures 2.1. Spray forming Discs of approximately 250 mm-diameter and 15 mm-thick were spray formed in a close coupled spray forming equipment. Four different compositions of supermartensitic stainless steel modified with boron additions, hereinafter named SM-0.3B, SM-0.5B, SM-0.7B and SM-14Cr-0.7B, were spray formed. Commercial supermartensitic stainless steel bar (VSM 13 supplied
215
by Villares Metals S/A.), iron–boron alloy with 16 wt.% of boron, iron–molybdenum alloy with 62 wt.% Mo, commercial pure chromium and nickel were used as raw materials. In each spray forming run, approximately 4 kg of raw materials were melted in an induction furnace and sprayed, using N2 as atomization gas, onto a rotating carbon steel disc substrate. The pouring temperature of all materials was 1650 °C, the spray distance 460 mm and the substrate speed 45 rpm. The average melt flow and gas flow of all spray forming runs were approximately 133 g s1 and 170 g s1, respectively, resulting in a gas-to-metal ratio (GMR) of approximately 1.2. 2.2. Characterization The chemical compositions of the final alloys were determined by inductively coupled plasma atomic emission spectrometry-ICP-AES, excepting C and S that were analyzed by direct combustion. Phases identification was performed by XRD analysis using a Rigaku Geigerflex ME210GF2 model diffractometer with Cu Ka radiation. The microstructures were characterized by scanning electron microscopy (SEM) using a FEI Inspect S50 Scanning Electron Microscope. In order to reveal the microstructures the polished samples were etched with a 3HCl:1HNO3 solution. Deep etching with 10 mL HCl, 3 ml HNO3, 5 mL FeCl3 and 82 mL ethyl alcohol solution were also performed in order to reveal the borides morphologies. Vickers microhardness was measured in accordance with ASTM E384 standards. 2.3. Wear tests Two different wear testing were performed in order to evaluate the wear resistance of the spray formed materials. Dry sand against rubber wheel abrasive wear test was performed in accordance with the procedure A of ASTM G65-04 standard. A home-made wear testing machine was used to perform a plate-on-cylinder wear test. Fig. 1 presents a schematic illustration of the wear test performed. The wear test machine consists of three separate chambers inside each one machined samples with dimensions of 25 90 10 mm are forced against a rotating quenched and tempered AISI 1040 steel axis with hardness of 55 HRC. Both the samples and axis were surface grinded to average roughness Ra = 1.7 lm. A normal force is applied to the sample by an arm system. In this work an initial normal force of 539.34 N was applied, but during the wear tests these initial normal force is slightly increased due to changing in the relative position of the mass center of the arms as consequence of the thickness loss of the sample. This increase of the normal force can be accurately described as a function of the loss thickness (h) by the following polynomial equation (FN = 539.34 + 7.1267h + 0.0865h2 + 0.0004h3, where the units of FN and h are N and mm, respectively), and these increments were considered to perform the calculus of contact pressure. In all tests the rotation speed used was 252 rpm. In order to simulate the wear conditions found in risers and casing, these chambers were filled with 6 l of drilling fluid donated by System Mud ltda. with the composition shown in Table 1. Three samples of each material were tested and the values presented here are the mean value and standard deviation of these samples. When the cylinder slides in contact with the sample, a crescent worn groove is formed as shown in Fig. 1. The groove volume, or the accumulated worn volume, was measured at each 30 min along 10 h and plotted as function of the sliding distance. The obtained curves were fitted using Eq. (1), which is similar to the empirical model presented by Hall and Malloy [25] to describe the wear behavior of casings and risers in real scale wear tests.
V ¼ Af1 exp½Bðsc Þg
ð1Þ
216
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 1. Schematic representation of the plate-on-cylinder wear test.
Table 1 Composition of the drilling MUD used in the plate-on-cylinder wear tests. Content (g/L) Bentonite KCl Viscosifier polymer Sand (AFS 50/70)
50 40 3.3 85
where V is accumulated worn volume in m3, s is the sliding distance in m and A, B and C are constants parameters that describe the evolution of wear with the sliding distance. As reported in [25] this function in Eq. (1) can represent a remarkable variety of shapes, all of which have one characteristic in common: As the sliding distance increases, the accumulated worn volume approaches to the limiting value A. In other words, this function shows that the wear rate decreases with the increasing of the sliding distance. Physically, reaching the limit A value does not mean that the wear completely stops, but that the wear rate reached very low values, which can be neglected. After fitting the experimental data with Eq. (1), the wear rate (dV/ds) can be calculated by using the Eq. (2).
dV ¼ A B C exp½Bðsc Þ sðc1Þ ds
ð2Þ
According to the authors [25], the decrease of the wear rate is associated with the reduction of the contact pressure caused by the increase of the contact area when the groove grows. When the limiting A value is reached the contact pressure also reach a constant value, so called threshold contact pressure (TCP). The TCP is now an important wear property, and the lower the TCP values obtained and the faster they are reached in a system, the better is the materials wear resistance. The commercial supermartensitic stainless, whose the microstructure is shown in Fig. 2, was used as reference alloy in both wear tests, and the results were compared with the boron modified spray-formed alloys. 2.4. Corrosion test Electrochemical analyses were performed using a conventional three electrodes set-up. The working electrodes were the spray formed boron modified supermartensitic stainless steel, with an immersed average area of 1 cm2, the counter electrode was a platinum sheet and an Ag/AgCl electrode was used as reference electrode. The analyses were carried out in an acid solution using deionized water, 0.6 mol/L of NaCl and addition of H2SO4 until pH = 4.0, simulating a sea water environment. Controlling of pH was performed using a pHmeter instrument. Open circuit analyses were carried out in a potentiostat, model Autolab, and the software NOVA 10.10. Measurements were launched 30 min after the
Fig. 2. Microstructure of the commercial supermartensitic stainless steel bar used as reference alloy for the wear tests. Etching: Vilela.
sample immersion to allow steady corrosion potential conditions to be reached. The potentiodynamic polarization curves were obtained by sweeping the potential from 150 mV below the corrosion potential to a maximum potential corresponding to a current of 0.1 mA/cm2 at a scan rate of 1 mV/s. For comparison purposes, electrochemical analyses were also carried out under the same conditions on the commercial supermartensitic stainless steel, the same used as reference in the wear tests, with chemical composition shown in Table 2.
3. Results 3.1. Microstructural characterization Table 3 shows the chemical compositions of the spray-formed alloys. One can see that the boron contents of 0.3, 0.5 and 0.7 wt.% were successively achieved in all alloys, as well as the Cr, Ni and Mo contents are in the range of the medium alloy supermartensitic stainless steel grade. The SM-0.5B presented a lower Cr
Table 2 Chemical composition of the commercial supermartensitic stainless steel (wt.%). %C 0.007
%Cr 11.86
%Ni 5.87
%Mo 2.00
%Si 0.25
%S 0.001
%P 0.016
%Mn 0.45
%Nb 0.01
%W 0.01
%Co 0.03
%N 0.0095
%Cu 0.06
%V 0.03
Ti 0.135
%Fe 79.26
217
G. Zepon et al. / Materials & Design 83 (2015) 214–223 Table 3 Chemical composition of the spray-formed boron modified supermartensitic stainless steel. wt.%
%C
%Cr
%Ni
%Mo
%B
%Ti
%Mn
%S
%Fe
SM-0.3B SM-0.5B SM-0.7B SM-14Cr-0.7B
0.066 ± 0.005 0.068 ± 0.001 0.096 ± 0.004 0.039 ± 0.004
12.00 ± 0.17 10.85 ± 0.52 11.88 ± 0.13 13.56 ± 0.26
5.68 ± 0.03 5.38 ± 0.25 5.88 ± 0.03 5.42 ± 0.03
2.09 ± 0.01 1.90 ± 0.09 2.09 ± 0.01 1.82 ± 0.04
0.37 ± 0.03 0.54 ± 0.04 0.69 ± 0.02 0.71 ± 0.08
0.044 ± 0.004 0.044 ± 0.004 0.041 ± 0.006 0.040 ± 0.006
0.40 ± 0.06 0.38 ± 0.05 0.49 ± 0.06 0.52 ± 0.05
0.0014 ± 0.0001 0.0041 ± 0.0001 0.001696 ± 0.0003 0.00151 ± 0.0003
Bal. Bal Bal. Bal.
content than the other alloys, 10.85 wt.%, which can reduce its corrosion resistance. On the other hand the SM-14Cr-0.7B presents a chromium content of 13.56 wt.%, higher than all supermartensitic grades (maximum Cr content of 13 wt.%). This high chromium content, as will be seen in Section 3.3, was proposed in order to improve the corrosion properties of the designed alloy. As can be seen, the carbon content in all alloys were in the range of 0.06–0.09 wt.%, also higher than the conventional SM grades (maximum of 0.03 wt.%). This increase in the carbon contents came from impurities of the raw materials and as result of contamination during the melt process. It can be seen in the XRD patterns of the spray-formed alloys, Fig. 3(a), that all compositions presented high intensity peaks related to low carbon martensite. By a detailed analysis of the
XRD patterns of the SM-0.7B and SM-14Cr-0.7B, Fig. 3(b), it can be clearly seen low intensity peaks of the orthorhombic M2B type boride. Fig. 4(a), (c), (e) and (g) show the microstructure of the SM-0.3B, SM-0.5B, SM-0.7B and SM-14Cr-0.7B, respectively. In all cases, the microstructure is composed by equiaxed martensitic grains with the hard M2B type borides, salient phases, at the grain boundaries. As can be seen in Fig. 4(b), (d), (f) and (h) these borides present morphologies of an interconnected eutectic net extended to all microstructure. This sort of microstructure of equiaxed grains with a refined and uniformly distributed eutectic net could be only achieved due to the solidification features of the spray forming process that happens in very small regions with average size of the droplets formed in the atomization stage, avoiding the macrosegregation of boron, which has a very low solubility in all iron’s solid phases [27,28]. It is also important to point out the very low porosity levels observed in the spray-formed deposits, being in all cases lower than 1%. These low porosity levels is attribute to the low viscosity of the remaining liquid during the deposition stage, which is effective to fill the interstices or cavities between the deposited solid particles. The chemical composition of each phase measured by EDS analysis can be seen in Table 4. The M2B type boride is composed mainly by Fe and Cr, with smaller amounts of Ni and Mo. It can be seen that the Mo content is higher in the lower boron contents alloys. In all cases the M2B present higher chromium content than the matrix, which can be deleterious for the corrosion properties once the chromium content of the matrix reach lower values than the supermartensitic stainless steel grades. In the case of the SM-14Cr-0.7B chemical composition of the matrix is in the range of the medium alloy supermartensitic stainless steel grades, which can be appropriate for the corrosion properties of the alloy. An important aspect of the microstructure of the spray-formed boron-modified supermartensitic stainless steels is that the increase of the boride fraction with the increasing of boron content is also accompanied by a reduction of the average grain size. Both the increase in the boride fraction and the reduction of grain size results in an increasing of the hardness, as can be seen in Fig. 5, which can be also benefit for the wear properties. 3.2. Wear resistance
Fig. 3. (a) XRD patterns of the spray-formed boron modified supermartensitic stainless steels. (b) Zoom of the SM-0.7B and SM-14Cr-0.7B XRD patterns.
In order to evaluate the effect of boron content in the wear resistance of the spray-formed supermartensitic stainless steel, samples of SM-0.3B, SM-0.5B, SM-0.7B and commercial SM were tested. The results of volume loss in the dry sand rubber wheel abrasive wear test are presented in Fig. 6. It can be seen that hardness of the commercial SM (284 HV) is much lower than the boron-modified SM stainless steels. But even with higher hardness, the SM-0.3B (440 HV) showed a similar volume loss value than the commercial SM, 45.2 mm3 and 44.9 mm3, respectively. However, with the increase of the boron content in the SM-0.5B (498 HV) and the SM-0.7B (549 HV), the increase of hardness is also accompanied by the reduction of volume loss, 36.9 mm3 and 29.3 mm3, respectively. By observing Fig. 7, one can see the presence of several holes in the worn surfaces of all materials, which suggest that a three-body wear mechanism is happening. In this sort of wear mechanism the abrasive sand particles can roll between the rubber
218
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 4. Microstructures and borides morphology of (a) and (b) SM-0.3B, (c) and (d) SM-0.5B, (e)and (f) SM-0.7B and (g) and (h) SM-14Cr-0.7B observed by SEM images (secondary electrons).
Table 4 EDS microanalyses of the martensitic matrixes and the M2B type borides of the sprayformed boron modified supermartensitic stainless steels. Matrix
M2B
wt.%
%Fe
%Cr
%Ni
%Mo
%Si
%Fe
%Cr
%Ni
%Mo
SM-0.3%B SM-0.5%B SM-0.7%B SM-14Cr-0.7B
80.8 82.0 81.7 78.3
11.6 11.0 10.7 13.4
6.0 5.7 6.1 6.1
1.3 1.1 1.0 0.9
0.3 0.1 0.4 0.5
56.2 66.0 67.1 60.6
30.8 14.2 23.8 33.5
1.7 4.0 3.0 2.4
11.0 15.4 5.8 3.0
and sample surfaces and the surface material is removed by subsequent indentations of the sharp corners of the abrasive particles. In this case, as the grain size and consequently the distance between the hard borides in the SM-0.3B are larger than those presented by the SM-0.5B and SM-0.7B, the protection against the indentations promoted by the borides in the material with lower boron content is less effective, once the area of exposed matrix is higher in this material. This is a reasonable explanation for the similar results of loss volume of the SM-0.3B and the commercial SM. On the
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 5. Grain size and hardness of the spray-formed boron modified supermartensitic stainless steel.
Fig. 6. Hardness and volume loss of the commercial SM, SM-0.3B, SM-0.5B and SM0.7B in the dry sand against rubber wheel abrasive wear test (ASTM G65-04 procedure A).
219
other hand, when the boron content is increased, the volume fraction the hard M2B type borides increases and the grain size decreases so that the distance between borides is reduced, which results in an effective protection of the martensitic matrix against the abrasive indentations reducing the volume loss values. Fig. 8(a) and (b) show the experimental results of the POC wear tests for the SM-0.3B, SM-0.5B, SM-0.7B and commercial SM. It can be seen that the accumulated worn volume after 25,000 m of sliding distance reduces from 5.2 107 m3 for the SM-0.3B to 4.4 107 m3 for the SM-0.5 and to 4.0 107 m3 for the SM-0.7B, showing that the increase of boron content also presented positive effects in the plate-cylinder wear test, see Fig. 8(a). However, it can be seen that at the same sliding distance the accumulated worn volume of the commercial SM is 3.9 106 m3, i.e., one order of magnitude higher than the spray-formed boron modified SM, see Fig. 8(b). This means that even small additions as 0.3 wt.% of boron are effective to improve the wear resistance of the supermartensitic stainless steel in this tribosystem. It can be seen that, even with the small increase of the normal force caused by the sample thickness loss, all materials presented the tendency of reducing the wear rate with the increase of sliding distance, as reported by Hall and Malloy [25]. In Fig. 8 and Table 5 can be seen the fitted curves and the fitted parameters of the experimental data with the Eq. (1). The experimental data of all materials were well fitted by the empirical equation purposed as can be seen by the R squared values. The most important value of this equation is the limiting value A, which indicates the maximum accumulated worn volume until the wear rate becomes negligible. By fitting the experimental data it was found the maximum accumulated worn volume being 8.3 107 m3 for the SM-0.3B, 6.0 107 m3 for the SM-0.5B, 4.7 107 m3 for the SM-0.7B and 5.9 106 m3 for the commercial SM. By using this model to evaluate the wear resistance of the material in the POC wear test, the better is the material’s wear resistance, the lower is its maximum accumulated wear volume and the faster it reaches this value. Fig. 9(a) shows the wear rate versus
Fig. 7. SEM images (secondary electrons) of the worn surfaces of (a) commercial SM, (b) SM-0.3B, (c) SM-0.5B and (d) SM-0.7B after the dry sand against rubber wheel abrasive wear test.
220
G. Zepon et al. / Materials & Design 83 (2015) 214–223 Table 5 Fitted parameters from experimental data using Eq. (1). V = A {(1 exp[B (sC)]} A SM-0.3B SM-0.5B SM-0.7B Commercial SM
B 7
8.276 10 5.972 107 4.701 107 5.908 106
3
4.67 10 7.99 103 2.71 103 1.78 104
C
R2
0.527 0.506 0.872 0.867
0.99447 0.99772 0.98043 0.9998
Fig. 8. Experimental data and fitted curves of accumulated worn volume versus sliding distance in the plate-on-cylinder wear test.
sliding distance, calculated by Eq. (2), of all materials. As it can be seen, in all cases by increasing the sliding distance the wear rate approaches to zero, but very high sliding distances are necessary in fact to reach the null value. However, it can be clearly seen that the wear rate of the boron modified SMs decays faster as higher is the boron content of the sample. As expected, for shorter sliding distances the wear rate of the commercial SM is much higher than the boron-modified alloys, but for very high sliding distances it also decays approaching the null value. One can see that for 150 km of sliding distance the wear rate of the commercial SM is almost the same of the SM-0.3B, approximately 1.0 1012 m3/m (or 1 mm3/km), which is a very low wear rate. At the same distance the wear rate of the SM-0.5B is 2.5 1013 m3/m (or 0.25 mm3/km) and of the SM-0.7B practically reached zero. Fig. 9(b) shows the evolution of the contact pressure versus the sliding distance. The contact pressure of all materials decays with the increasing of the sliding distance approaching to the value of the TCP. As the wear rate, the TCP of all materials would be reached only for very high sliding distances. The TCP values found for each material are: 1.02 MPa, 1.13 MPa, 1.23 MPa and 0.56 MPa for the SM-0.3B, SM-0.5B, SM-0.7B and commercial SM, respectively. Based on this, one can see that adding small amounts of boron in the spray formed supermartensitic stainless steel clearly improved the wear resistance in the plate-on-cylinder wear test by reducing the wear rates as well as the maximum accumulated worn volumes and, consequently, increasing the TCPs.
Fig. 9. (a) Wear rate and (b) contact pressure versus sliding distance in the plateon-cylinder wear test.
It is interesting to observe that in the case of the dry sand/ rubber wheel abrasive wear test almost no difference between the wear resistance of the commercial SM and the SM-0.3B could be seen, while in the plate-on-cylinder wear test the behavior of the SM-0.3B was very similar to the SM-0.7B and much superior to the commercial SM. Such different results can be explained by looking to the worn surfaces and the collected debris of the plate-on-cylinder wear test samples in Fig. 10. It can be seen that when the hard cylinder slides against the commercial SM, the asperities and the abrasive particles present in the fluid push the surface material forward in a process that involves considerable plastic deformation. During the sliding the material is detached from the surface in form of debris with plate morphology, as can be clearly seen in Fig. 10(a) and (b). Once the presence of the hard M2B type borides in the microstructure restrict the plastic
G. Zepon et al. / Materials & Design 83 (2015) 214–223
221
Fig. 10. SEM images (secondary electrons) of the worn surfaces and plate-like debris of (a) and (b) commercial SM, (c) and (d) SM-0.3B, (e) and (f) SM-0.5B and (g) and (h) SM0.7B after the plate-on-cylinder wear test sand against rubber wheel abrasive wear test.
deformation of the material, the wear mechanism observed in the commercial SM is less effective in the boron modified alloys. Through the analysis of Fig. 9(c)–(h) it can be noted that the same wear mechanism of pushing and removing plate-like debris occurs in the SM-0.3B, SM-0.5B and SM-0.7B, but in a very reduced scale. When the boron content and the borides fraction are increased, the restriction to plastic deformations also increases resulting in higher wear resistances.
3.3. Corrosion resistance Electrochemical analyses were performed in order to evaluate the effect of boron addition in the chemical composition of the spray-formed boron modified SM. In addition to the SM-0.3B, SM-0.5B and SM-0.7B the SM-14Cr-0.7B was also analyzed to evaluate the effect of the increase in the chromium content in the alloy and, consequently, the increase in the remaining Cr content in the
222
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 11. Polarization curves in acid solution (pH = 4) for the SM-0.3B, SM-0.5B, SM-0.7B, SM-14Cr-0.7B and commercial SM.
Table 6 Electrochemical properties obtained from the polarization curves in acid media (pH = 4).
SM-0.3B SM-0.5B SM-0.7B SM-14Cr-0.7B Commercial SM
Ecorr (mV)
Ecrit (mV)
DE (mV)
icorr (A/cm2)
503 589 427 405 157
423 530 300 142 +92
80 59 127 263 249
6E6 5E6 8E7 3E7 1E07
martensitic matrix. Fig. 11 shows the polarization curves for all alloys tested in the acid solution (pH = 4) and Table 6 shows the main information obtained from the polarization curves; corrosion potential (Ecorr), corrosion current (icorr), critical potential (Ecrit, defined as the potential where the current density reaches the level of 10 lA/cm2) and potential width before reaching the critical value (DE = Ecrit Ecorr). One can see that the SM-0.5B presented the lower performance in the polarization test, presenting the lower value of Ecorr, 589 mV, a high value of icorr, 5 106 A/cm2, and not showing the presence of a passive plateau. This result can be attributed to the low Cr content present in this alloy, only 10.85 wt.% The Ecorr and icorr measured for the SM-0.3B were -503 mV and 6 106 A/cm2, respectively. The SM-0.3B alloy also presented a passive plateau, but this plateau occurred in very high currents, approximately 2 105 A/cm2, indicating a bad quality of the passive layer. The SM-0.7B has presented better values of Ecorr and icorr than the SM-0.3B, 427 mV and 8 107 A/cm2, but no passive plateau could be seen in this sample, probably due to the higher impoverishment in the matrix Cr content caused by the formation of higher fraction of borides. The SM-14Cr-0.7B presented much better corrosion properties than the other boron modified alloys, presenting Ecorr and icorr values of 405 mV and 3 107 A/cm2, respectively, and the formation of a passive plateau. Although the Ecorr of the SM-14Cr-0.7B be considerably lower than the commercial SM, Ecorr = 157 mV, the icorr values are very close, for the commercial SM icorr = 1 107 A/cm2, as well as the DE, 263 mV and 249 mV, for the SM-14Cr-0.7B and commercial SM, respectively. This better performance of the SM-14Cr-0.7B when compared to the other boron modified SMs can be explained by the increase in the remaining Cr content in the matrix, approximately 13 wt.% for this alloy, which increase the tendency to form a more efficient passive layer. This result shows that increasing the chromium content close to 14 wt.% in the alloy modified with 0.7%B results in a corrosion resistance more similar to the commercial
SM, while maintaining the same microstructure of martensitic matrix with a continuous net of eutectic M2B type borides. 3.4. Final remarks Dense deposits of boron modified supermartensitic stainless steel with refined microstructures composed by equiaxed martensitic grains (with average grain sizes ranging from 10 to 20 lm) with a network of eutectic M2B type borides uniformly distributed at the grain boundaries were successfully produced by spray forming process. The equiaxed, refined and homogeneous microstructures are consequence of the spray forming features which allows the solidification to occur in very small regions avoiding the macrosegregation of boron [6,7]. It was shown that by increasing the boron content from 0.3 wt.% to 0.7 wt.% two effects can be observed: (1) increase of the volume fraction of M2B type borides; and (2) reduction of the average grain size. Both effects contribute to reduce the free mean distance between the adjacent borides, which is very beneficial for the improvement of the abrasive wear resistance in the dry sand against rubber wheel wear test. By analyzing the worn surfaces it was possible to note that the material removal mechanism that takes place in this case is the subsequent indentations of the rolling abrasive particles. In this case, the increase of boride volume fractions as well as the reduction of the distance between then improve the wear resistance by reducing the exposed matrix area and, consequently, prevent the abrasive particle to indent and remove material from the matrix. The same effect on the abrasive wear resistance of cast white irons with high volume of M7C3 type carbides were already reported in the literature [14,29,29]. This can explain the reason for the SM-0.3%B, which has the higher average grain size among the B-containing samples, to present a similar wear resistance compared to the commercial SM, even having a higher hardness. However, very different wear behaviors were observed in the plate-on-cylinder wear test. In this tribosystem, the material is removed from the surface through a mechanism that involves considerably high amounts of surface plastic deformation. In this case, the eutectic M2B boride network formed in the materials, even with only 0.3 wt.%, was effective to restrain the surface plastic deformation, resulting, in all cases, in accumulated worn values one order of magnitude lower than the commercial SM. It is worth to point out that the wear behavior of all materials were accurately described by the empirical model proposed by Hall and Malloy [25] in which the wear rate decreases with the increase of the sliding distance and decrease of the contact pressure. It was clearly shown that increasing the boron content in the chemical composition of
G. Zepon et al. / Materials & Design 83 (2015) 214–223
the spray formed supermartensitic stainless steel is beneficial to improve the wear properties, reducing the wear rates and increasing the TCP. In spite of the improvement in the wear resistance, the addition of boron content in the chemical composition of the conventional medium alloy supermartensitic stainless steel, reduces the corrosion resistance of the alloy due to the impoverishment of the remaining chromium content in the martensitic matrix caused by the formation of Cr-rich borides. The reduction in the chromium content of the matrix can be a serious drawback for the formation and growth of highly protective passive films, thus considerably reducing the corrosion properties of the spray formed boron modified supermartensitic stainless steel. It was however shown that by adding an over content of chromium in the chemical composition of the SM modified with 0.7 wt.% of B, in this case 13.56 wt.% of Cr, it is possible to maintain the same microstructure, with almost same martensitic grain size, volume fraction of M2B borides, and consequently maintaining the same wear properties, but with a higher chromium content in the matrix. This increase in the matrix’s chromium content improves the tendency of formation and the quality of the passive film, clearly improving the corrosion properties of the alloys, which retrieved the corrosion resistance level of the commercial SM in spite of lower corrosion potential values. 4. Conclusions Dense deposits of boron modified supermartensitic stainless steel with 0.3, 0.5 and 0.7 wt.% of boron can be successfully produced by spray forming with equiaxed and refined microstructure composed by martensitic matrix and a continuous network of eutectic M2B type borides at the grain boundaries. The increase in the boron content increases the volume fraction of M2B borides and reduces the martensitic average grain size. The abrasive wear resistance measured by dry sand against rubber wheel is as better as higher is the volume fraction of borides and smaller the distance between then. The empirical model proposed by the petroleum engineers to evaluate the wear of risers and casings can be successfully applied to describe the wear behavior of the materials in the plate-on-cylinder wear test. The presence of the continuous network of M2B in the microstructure of the spray formed boron modified SMs improves the wear resistance of the alloys in the plate-on-cylinder wear test reducing the wear rates and increasing the threshold contact pressure. In this case the wear properties are also improved by increasing the boron content, but all boron modified alloys present maximum accumulated worn volume ten times lower than the commercial SM. The formation of M2B type borides rich in chromium leads to a reduction of the remaining matrix’s chromium content of the boron modified alloys, reducing their corrosion properties. The
223
addition of an over content of chromium, in this case 13.56 wt.%, increases the remaining matrix’s chromium content improving the corrosion properties while maintaining the same microstructural characteristics of the wear resistant spray formed boron modified supermartesitic stainless steel.
Acknowledgements The authors would like to thank the Brazilian institutions FAPESP, CNPq, CAPES and Petrobras for the financial support. References [1] C.P. Linne, F. Blanchard, G.C. Guntz, Corrosion/1997 1997: Paper No. 28. [2] H. Widen, P. Toussaint, L. Courdreuse, Past, present and future of weldable supermartensitic alloys, in: Supermartensitic Stainl. Steel 2002, vol. 1. Brussels, Belgium, 2002. [3] A.G. Haynes, Some factors governing the metallurgy and weldability of 13%Cr and newer Cr–Ni martensitic stainless steel, in: Supermartensitic Stainl. Steel 1999, vol. 1. Brussels, Belgium, 1999. [4] P. Toussaint, J.J. Dufrane, Advances in the making and base material properties of supermartensitic stainless steels, in: Supermartensitic Stainl. Steel 2002, vol. 1. Brussels, Belgium, 2002. [5] C.a. Della Rovere, C.R. Ribeiro, R. Silva, L.F.S. Baroni, N.G. Alcântara, S.E. Kuri, Mater. Sci. Eng., A 586 (2013) 86. [6] P.S. Grant, Prog. Mater. Sci. 39 (1995) 497. [7] P.S. Grant, Metall. Mater. Trans. A Phys. Metall. Mater. Sci. 38 A (2007) 1520. [8] A. Schulz, V. Uhlenwinkel, C. Escher, R. Kohlmann, A. Kulmburg, M.C. Montero, R. Rabitsch, W. Schützenhöfer, D. Stocchi, D. Viale, Mater. Sci. Eng., A 477 (2008) 69. [9] D.M. Goudar, K. Raju, V.C. Srivastava, G.B. Rudrakshi, Mater. Des. 51 (2013) 383. [10] D.M. Goudar, K. Raju, V.C. Srivastava, G.B. Rudrakshi, Mater. Des. 47 (2013) 489. [11] R. Mittal, A. Tomar, D. Singh, J. Mater. Eng. Perform. 23 (2014) 975. [12] Y. Jia, F. Cao, S. Scudino, P. Ma, H. Li, L. Yu, J. Eckert, J. Sun, Mater. Des. 57 (2014) 585. [13] D.N. Hanlon, W.M. Rainforth, C.M. Sellars, Wear 225–229 (1999) 587. [14] a.H. Kasama, a.J. Mourisco, C.S. Kiminami, W.J. Botta Fo, C. Bolfarini, Mater. Sci. Eng., A 589 (2004) 375–377. [15] T.T. Matsuo, C.S. Kiminami, W.J.B. Fo, C. Bolfarini, Wear 259 (2005) 445. [16] K. Kaur, O.P. Pandey, J. Mater. Eng. Perform. 22 (2013) 3101. [17] V.C. Srivastava, G.B. Rudrakshi, V. Uhlenwinkel, S.N. Ojha, J. Mater. Sci. 44 (2009) 2288. [18] E.V. Shadrichev, A.E. Ivanov, Met. Sci. Heat Treat. 26 (1984) 235. [19] E.V. Shadrichev, S.I. Rumyantsev, Met. Sci. Heat Treat. 24 (1982) 495. [20] J.W. Li, G.S. Zhang, S.Z. Wei, Q. Zhao, Appl. Mech. Mater. 52–54 (2011) 1718. [21] H.G. Fu, J.D. Xing, Y.P. Lei, L.M. Huang, J. Mater. Eng. Perform. 20 (2011) 1665. [22] G. Zepon, C.S. Kiminami, W.J. Botta, C. Bolfarini, Mater. Res. J. Mater. 16 (2013) 642. [23] N. Yüksel, S. S ß ahin, Mater. Des. 58 (2014) 491. [24] S. Mahathanabodee, T. Palathai, S. Raadnui, R. Tongsri, N. Sombatsompop, Mater. Des. 46 (2013) 588. [25] K.P. Hall, R.W.M., Soc. Pet. Eng., 2005, SPE 9430:1. [26] D. Gao, L. Sun, J. Lian, Pet. Sci. 7 (2010) 494. [27] P.E. Busby, M.E. Warga, C. Wells, Trans. Am. Inst. Min. Metall. Eng. 197 (1953) 1463. [28] N.J. Calos, E. Graham, D.R. Cousens, P. Christodoulou, C.H.L. Kennard, L.K. Bekessy, S.F. Parker, Mater. Trans. 42 (2001) 496. [29] C.P. Tabrett, I.R. Sare, M.R. Ghomashchi, Int. Mater. Rev. 41 (1996) 59.
Contents lists available at ScienceDirect
Materials & Design journal homepage: www.elsevier.com/locate/matdes
Design of wear resistant boron-modified supermartensitic stainless steel by spray forming process G. Zepon a,⇑, A.R.C. Nascimento a, A.H. Kasama b, R.P. Nogueira c,d, C.S. Kiminami e, W.J. Botta e, C. Bolfarini e a
Programa de Pós-Graduação em Ciência e Engenharia de Materiais, Universidade Federal de São Carlos, Rod. Washington Luiz, km 235, 13565-905 São Carlos, SP, Brazil Centro de Pesquisa & Desenvolvimento Leopoldo Américo Miguêz de Mello (Cenpes), Avenida Horácio Macedo, 950, 21941-915 Rio de Janeiro, RJ, Brazil Univ. Grenoble Alpes, LEPMI, F-38000 Grenoble, France d CNRS, LEPMI, F-38000 Grenoble, France e Departamento de Engenharia de Materiais, Universidade Federal de São Carlos, Rod. Washington Luiz, km 235, 13565-905 São Carlos, SP, Brazil b c
a r t i c l e
i n f o
Article history: Received 19 January 2015 Revised 5 May 2015 Accepted 1 June 2015
Keywords: Spray forming Supermartensitic stainless steel Borides Wear resistance Corrosion resistance
a b s t r a c t In this paper the chemical composition of the supermartensitic stainless (SM) was modified with the addition of small boron contents (0.3, 0.5 and 0.7 wt.%) and processed by spray forming aiming the development of functionalized stainless steel with higher wear resistance. The addition of boron to the SM leads to the formation of continuous network of M2B type borides uniformly distributed in the refined microstructure promoted by the spray forming process. The wear resistance was evaluated by two different methodologies: (1) the standardized dry sand/rubber wheel test (ASTM G65); and (2) a plate-on-cylinder (POC) wear test which was designed to simulate in laboratorial scale the tribosystems found in wear of risers and casings. It was shown that the wear mechanisms that take place in both tests are quite different, but in all cases increasing the boron content is always accompanied by an increase in the wear resistance. Electrochemical analyses were performed to evaluate the corrosion resistance of the designed alloys. It could be seen that corrosion properties similar to the commercial SM can be achieved in the SM modified with 0.7 wt.% of boron if an over content of chromium is added to the chemical composition. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction The production of oil in the pre-salt layers in the coast of Brazil brought new challenges concerning the development of materials which withstand the severe work conditions (mainly wear and corrosion issues) found during drilling and exploitation operations. These extreme conditions motivate the development of new materials with improved wear and corrosion properties, aiming their application as coatings for pipes and parts used in such operations. Drilling risers (steel pipes that links the well head on the sea floor to the platform) and casings (steel pipes used to protect the wall of wells during the drilling operations), for example, are frequently subjected to severe wear conditions due to the contact with rotating tool joints of the drill pipes. Moreover, these pipes work in contact with rock debris resulting from the drilling process and drilling fluids (water or oil based fluids containing mainly bentonite) which are usually rich in chlorides, increasing the susceptibility ⇑ Corresponding author. E-mail address: [email protected] (G. Zepon). http://dx.doi.org/10.1016/j.matdes.2015.06.020 0264-1275/Ó 2015 Elsevier Ltd. All rights reserved.
of failure caused by wear and corrosion. Since mid 1990’s the weldable 13Cr martensitic stainless steel grades, also called supermartensitic stainless steel (SM), has been increasingly applied in seamless pipes for drilling, casing, and tubing for application in oil and gas fields presenting higher corrosive environments [1–3]. Such stainless steel grades are based on the Fe–Cr–Ni–Mo system with up to 13 wt.% of Cr, 4–6 wt.% of Ni, 0.5–2.5 wt.% of Mo, low amounts of carbon, nitrogen, phosphorus and sulfur (C 6 0.02 wt.%, N, P, S 6 0.03 wt.%). The optimized microstructure of the material is free from d-ferrite and offers good corrosion resistance in environments containing CO2 and H2S. Typical values of mechanical properties of the SM grades are: 25–32 HRC, 650–750 MPa of 0.2% yield strength, 880–950 MPa of tensile strength, elongation at rupture up to 20% and impact energy up to 100 J [4,5]. In spite of their good mechanical and corrosion properties, stainless steel grades usually presents low wear resistance. Wear resistance is not an intrinsic property of the materials, but a tribosystem property where the materials in contact, type of reciprocal movement, relative speed, load levels, environmental conditions, presence of abrasive particles and lubricants can play
G. Zepon et al. / Materials & Design 83 (2015) 214–223
important roles in the wear behavior of a component. Based on this, from an engineering point of view, it is advisable testing the wear properties of materials reproducing closely the real operational condition. Spray forming process is an advanced casting process which allows the manufacture of near net shape preforms of alloy compositions that are problematical in conventional casting processes [6,7]. It has been reported that spray-forming is a useful tool for generating high-alloyed materials, such as, for instance, AISI D2 and AISI M3 high-alloyed tool steels, with reduced macro-segregation, fine microstructure and uniformly distribution of carbides and second phases [8]. Spray forming has been used to develop several materials, such as aluminum alloys (Al–Si, Al–Mg alloys) [9–12], iron-based alloys (high chromium cast iron and high speed steel) [13–15] and metal-matrix-composites (MMC’s) [16,17] with improved wear resistances. In all cases the wear properties are improved by the microstructural refinement and uniformity promoted by the spray forming process. Kasama et al. [14] and Matsuo et al. [15], for instance, showed that sprayed-formed high chromium cast irons present a substantial refinement in the microstructure scale when compared to the same conventionally cast alloy. The presence of fine M7C3 type carbides uniformly dispersed in an austenitic/martensitic matrix was effective to improve considerably both the abrasive and the sliding wear resistances. Since 1960’s it is well known that boriding is one of the most effective methods for increasing the wear resistance of steel parts due to the formation of hard borides such as Fe2B (1600 HV) and FeB (1800 HV) [18,19]. In the last years several authors reported the development of boron modified iron-based alloys to be used in coatings or parts subjected to severe wear conditions [20–24]. Recently, we have reported that addition of small quantities of boron in the chemical composition of spray-formed supermartensitic stainless steel led to the formation of M2B type borides (where M = Fe, Cr, Ni, Mo), which was very beneficial to the abrasive wear resistance of the alloy [22]. Moreover, since boron is a low-cost alloying element (iron–boron alloys), its use in the chemical composition of steels to generate enhanced properties is also economically interesting. In this paper we report the complete design of the wear resistant alloy based on the chemical composition of the supermartensitic stainless steel modified with boron additions and produced by spray forming process, aiming the achievement of the advantageous microstructure characteristics of spray-formed materials, i.e., refined microstructure and high level of microstructural homogeneity. The target composition of the alloy was designed aiming to achieve the best combination of wear and corrosion properties. Bearing in mind that wear is a tribosystemic property, two different types of wear tests were performed. The dry sand/rubber wheel abrasive wear test (ASTM G65-04), and a plate-on-cylinder-POC wear test, which simulates in laboratorial scale the tribosystem found when drilling risers and casings are worn by the contact of the rotating drill pipe. The last wear test performed and the models used to determine the material’s wear properties were based on real scale wear tests reported by petroleum engineers [25,26].
2. Experimental procedures 2.1. Spray forming Discs of approximately 250 mm-diameter and 15 mm-thick were spray formed in a close coupled spray forming equipment. Four different compositions of supermartensitic stainless steel modified with boron additions, hereinafter named SM-0.3B, SM-0.5B, SM-0.7B and SM-14Cr-0.7B, were spray formed. Commercial supermartensitic stainless steel bar (VSM 13 supplied
215
by Villares Metals S/A.), iron–boron alloy with 16 wt.% of boron, iron–molybdenum alloy with 62 wt.% Mo, commercial pure chromium and nickel were used as raw materials. In each spray forming run, approximately 4 kg of raw materials were melted in an induction furnace and sprayed, using N2 as atomization gas, onto a rotating carbon steel disc substrate. The pouring temperature of all materials was 1650 °C, the spray distance 460 mm and the substrate speed 45 rpm. The average melt flow and gas flow of all spray forming runs were approximately 133 g s1 and 170 g s1, respectively, resulting in a gas-to-metal ratio (GMR) of approximately 1.2. 2.2. Characterization The chemical compositions of the final alloys were determined by inductively coupled plasma atomic emission spectrometry-ICP-AES, excepting C and S that were analyzed by direct combustion. Phases identification was performed by XRD analysis using a Rigaku Geigerflex ME210GF2 model diffractometer with Cu Ka radiation. The microstructures were characterized by scanning electron microscopy (SEM) using a FEI Inspect S50 Scanning Electron Microscope. In order to reveal the microstructures the polished samples were etched with a 3HCl:1HNO3 solution. Deep etching with 10 mL HCl, 3 ml HNO3, 5 mL FeCl3 and 82 mL ethyl alcohol solution were also performed in order to reveal the borides morphologies. Vickers microhardness was measured in accordance with ASTM E384 standards. 2.3. Wear tests Two different wear testing were performed in order to evaluate the wear resistance of the spray formed materials. Dry sand against rubber wheel abrasive wear test was performed in accordance with the procedure A of ASTM G65-04 standard. A home-made wear testing machine was used to perform a plate-on-cylinder wear test. Fig. 1 presents a schematic illustration of the wear test performed. The wear test machine consists of three separate chambers inside each one machined samples with dimensions of 25 90 10 mm are forced against a rotating quenched and tempered AISI 1040 steel axis with hardness of 55 HRC. Both the samples and axis were surface grinded to average roughness Ra = 1.7 lm. A normal force is applied to the sample by an arm system. In this work an initial normal force of 539.34 N was applied, but during the wear tests these initial normal force is slightly increased due to changing in the relative position of the mass center of the arms as consequence of the thickness loss of the sample. This increase of the normal force can be accurately described as a function of the loss thickness (h) by the following polynomial equation (FN = 539.34 + 7.1267h + 0.0865h2 + 0.0004h3, where the units of FN and h are N and mm, respectively), and these increments were considered to perform the calculus of contact pressure. In all tests the rotation speed used was 252 rpm. In order to simulate the wear conditions found in risers and casing, these chambers were filled with 6 l of drilling fluid donated by System Mud ltda. with the composition shown in Table 1. Three samples of each material were tested and the values presented here are the mean value and standard deviation of these samples. When the cylinder slides in contact with the sample, a crescent worn groove is formed as shown in Fig. 1. The groove volume, or the accumulated worn volume, was measured at each 30 min along 10 h and plotted as function of the sliding distance. The obtained curves were fitted using Eq. (1), which is similar to the empirical model presented by Hall and Malloy [25] to describe the wear behavior of casings and risers in real scale wear tests.
V ¼ Af1 exp½Bðsc Þg
ð1Þ
216
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 1. Schematic representation of the plate-on-cylinder wear test.
Table 1 Composition of the drilling MUD used in the plate-on-cylinder wear tests. Content (g/L) Bentonite KCl Viscosifier polymer Sand (AFS 50/70)
50 40 3.3 85
where V is accumulated worn volume in m3, s is the sliding distance in m and A, B and C are constants parameters that describe the evolution of wear with the sliding distance. As reported in [25] this function in Eq. (1) can represent a remarkable variety of shapes, all of which have one characteristic in common: As the sliding distance increases, the accumulated worn volume approaches to the limiting value A. In other words, this function shows that the wear rate decreases with the increasing of the sliding distance. Physically, reaching the limit A value does not mean that the wear completely stops, but that the wear rate reached very low values, which can be neglected. After fitting the experimental data with Eq. (1), the wear rate (dV/ds) can be calculated by using the Eq. (2).
dV ¼ A B C exp½Bðsc Þ sðc1Þ ds
ð2Þ
According to the authors [25], the decrease of the wear rate is associated with the reduction of the contact pressure caused by the increase of the contact area when the groove grows. When the limiting A value is reached the contact pressure also reach a constant value, so called threshold contact pressure (TCP). The TCP is now an important wear property, and the lower the TCP values obtained and the faster they are reached in a system, the better is the materials wear resistance. The commercial supermartensitic stainless, whose the microstructure is shown in Fig. 2, was used as reference alloy in both wear tests, and the results were compared with the boron modified spray-formed alloys. 2.4. Corrosion test Electrochemical analyses were performed using a conventional three electrodes set-up. The working electrodes were the spray formed boron modified supermartensitic stainless steel, with an immersed average area of 1 cm2, the counter electrode was a platinum sheet and an Ag/AgCl electrode was used as reference electrode. The analyses were carried out in an acid solution using deionized water, 0.6 mol/L of NaCl and addition of H2SO4 until pH = 4.0, simulating a sea water environment. Controlling of pH was performed using a pHmeter instrument. Open circuit analyses were carried out in a potentiostat, model Autolab, and the software NOVA 10.10. Measurements were launched 30 min after the
Fig. 2. Microstructure of the commercial supermartensitic stainless steel bar used as reference alloy for the wear tests. Etching: Vilela.
sample immersion to allow steady corrosion potential conditions to be reached. The potentiodynamic polarization curves were obtained by sweeping the potential from 150 mV below the corrosion potential to a maximum potential corresponding to a current of 0.1 mA/cm2 at a scan rate of 1 mV/s. For comparison purposes, electrochemical analyses were also carried out under the same conditions on the commercial supermartensitic stainless steel, the same used as reference in the wear tests, with chemical composition shown in Table 2.
3. Results 3.1. Microstructural characterization Table 3 shows the chemical compositions of the spray-formed alloys. One can see that the boron contents of 0.3, 0.5 and 0.7 wt.% were successively achieved in all alloys, as well as the Cr, Ni and Mo contents are in the range of the medium alloy supermartensitic stainless steel grade. The SM-0.5B presented a lower Cr
Table 2 Chemical composition of the commercial supermartensitic stainless steel (wt.%). %C 0.007
%Cr 11.86
%Ni 5.87
%Mo 2.00
%Si 0.25
%S 0.001
%P 0.016
%Mn 0.45
%Nb 0.01
%W 0.01
%Co 0.03
%N 0.0095
%Cu 0.06
%V 0.03
Ti 0.135
%Fe 79.26
217
G. Zepon et al. / Materials & Design 83 (2015) 214–223 Table 3 Chemical composition of the spray-formed boron modified supermartensitic stainless steel. wt.%
%C
%Cr
%Ni
%Mo
%B
%Ti
%Mn
%S
%Fe
SM-0.3B SM-0.5B SM-0.7B SM-14Cr-0.7B
0.066 ± 0.005 0.068 ± 0.001 0.096 ± 0.004 0.039 ± 0.004
12.00 ± 0.17 10.85 ± 0.52 11.88 ± 0.13 13.56 ± 0.26
5.68 ± 0.03 5.38 ± 0.25 5.88 ± 0.03 5.42 ± 0.03
2.09 ± 0.01 1.90 ± 0.09 2.09 ± 0.01 1.82 ± 0.04
0.37 ± 0.03 0.54 ± 0.04 0.69 ± 0.02 0.71 ± 0.08
0.044 ± 0.004 0.044 ± 0.004 0.041 ± 0.006 0.040 ± 0.006
0.40 ± 0.06 0.38 ± 0.05 0.49 ± 0.06 0.52 ± 0.05
0.0014 ± 0.0001 0.0041 ± 0.0001 0.001696 ± 0.0003 0.00151 ± 0.0003
Bal. Bal Bal. Bal.
content than the other alloys, 10.85 wt.%, which can reduce its corrosion resistance. On the other hand the SM-14Cr-0.7B presents a chromium content of 13.56 wt.%, higher than all supermartensitic grades (maximum Cr content of 13 wt.%). This high chromium content, as will be seen in Section 3.3, was proposed in order to improve the corrosion properties of the designed alloy. As can be seen, the carbon content in all alloys were in the range of 0.06–0.09 wt.%, also higher than the conventional SM grades (maximum of 0.03 wt.%). This increase in the carbon contents came from impurities of the raw materials and as result of contamination during the melt process. It can be seen in the XRD patterns of the spray-formed alloys, Fig. 3(a), that all compositions presented high intensity peaks related to low carbon martensite. By a detailed analysis of the
XRD patterns of the SM-0.7B and SM-14Cr-0.7B, Fig. 3(b), it can be clearly seen low intensity peaks of the orthorhombic M2B type boride. Fig. 4(a), (c), (e) and (g) show the microstructure of the SM-0.3B, SM-0.5B, SM-0.7B and SM-14Cr-0.7B, respectively. In all cases, the microstructure is composed by equiaxed martensitic grains with the hard M2B type borides, salient phases, at the grain boundaries. As can be seen in Fig. 4(b), (d), (f) and (h) these borides present morphologies of an interconnected eutectic net extended to all microstructure. This sort of microstructure of equiaxed grains with a refined and uniformly distributed eutectic net could be only achieved due to the solidification features of the spray forming process that happens in very small regions with average size of the droplets formed in the atomization stage, avoiding the macrosegregation of boron, which has a very low solubility in all iron’s solid phases [27,28]. It is also important to point out the very low porosity levels observed in the spray-formed deposits, being in all cases lower than 1%. These low porosity levels is attribute to the low viscosity of the remaining liquid during the deposition stage, which is effective to fill the interstices or cavities between the deposited solid particles. The chemical composition of each phase measured by EDS analysis can be seen in Table 4. The M2B type boride is composed mainly by Fe and Cr, with smaller amounts of Ni and Mo. It can be seen that the Mo content is higher in the lower boron contents alloys. In all cases the M2B present higher chromium content than the matrix, which can be deleterious for the corrosion properties once the chromium content of the matrix reach lower values than the supermartensitic stainless steel grades. In the case of the SM-14Cr-0.7B chemical composition of the matrix is in the range of the medium alloy supermartensitic stainless steel grades, which can be appropriate for the corrosion properties of the alloy. An important aspect of the microstructure of the spray-formed boron-modified supermartensitic stainless steels is that the increase of the boride fraction with the increasing of boron content is also accompanied by a reduction of the average grain size. Both the increase in the boride fraction and the reduction of grain size results in an increasing of the hardness, as can be seen in Fig. 5, which can be also benefit for the wear properties. 3.2. Wear resistance
Fig. 3. (a) XRD patterns of the spray-formed boron modified supermartensitic stainless steels. (b) Zoom of the SM-0.7B and SM-14Cr-0.7B XRD patterns.
In order to evaluate the effect of boron content in the wear resistance of the spray-formed supermartensitic stainless steel, samples of SM-0.3B, SM-0.5B, SM-0.7B and commercial SM were tested. The results of volume loss in the dry sand rubber wheel abrasive wear test are presented in Fig. 6. It can be seen that hardness of the commercial SM (284 HV) is much lower than the boron-modified SM stainless steels. But even with higher hardness, the SM-0.3B (440 HV) showed a similar volume loss value than the commercial SM, 45.2 mm3 and 44.9 mm3, respectively. However, with the increase of the boron content in the SM-0.5B (498 HV) and the SM-0.7B (549 HV), the increase of hardness is also accompanied by the reduction of volume loss, 36.9 mm3 and 29.3 mm3, respectively. By observing Fig. 7, one can see the presence of several holes in the worn surfaces of all materials, which suggest that a three-body wear mechanism is happening. In this sort of wear mechanism the abrasive sand particles can roll between the rubber
218
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 4. Microstructures and borides morphology of (a) and (b) SM-0.3B, (c) and (d) SM-0.5B, (e)and (f) SM-0.7B and (g) and (h) SM-14Cr-0.7B observed by SEM images (secondary electrons).
Table 4 EDS microanalyses of the martensitic matrixes and the M2B type borides of the sprayformed boron modified supermartensitic stainless steels. Matrix
M2B
wt.%
%Fe
%Cr
%Ni
%Mo
%Si
%Fe
%Cr
%Ni
%Mo
SM-0.3%B SM-0.5%B SM-0.7%B SM-14Cr-0.7B
80.8 82.0 81.7 78.3
11.6 11.0 10.7 13.4
6.0 5.7 6.1 6.1
1.3 1.1 1.0 0.9
0.3 0.1 0.4 0.5
56.2 66.0 67.1 60.6
30.8 14.2 23.8 33.5
1.7 4.0 3.0 2.4
11.0 15.4 5.8 3.0
and sample surfaces and the surface material is removed by subsequent indentations of the sharp corners of the abrasive particles. In this case, as the grain size and consequently the distance between the hard borides in the SM-0.3B are larger than those presented by the SM-0.5B and SM-0.7B, the protection against the indentations promoted by the borides in the material with lower boron content is less effective, once the area of exposed matrix is higher in this material. This is a reasonable explanation for the similar results of loss volume of the SM-0.3B and the commercial SM. On the
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 5. Grain size and hardness of the spray-formed boron modified supermartensitic stainless steel.
Fig. 6. Hardness and volume loss of the commercial SM, SM-0.3B, SM-0.5B and SM0.7B in the dry sand against rubber wheel abrasive wear test (ASTM G65-04 procedure A).
219
other hand, when the boron content is increased, the volume fraction the hard M2B type borides increases and the grain size decreases so that the distance between borides is reduced, which results in an effective protection of the martensitic matrix against the abrasive indentations reducing the volume loss values. Fig. 8(a) and (b) show the experimental results of the POC wear tests for the SM-0.3B, SM-0.5B, SM-0.7B and commercial SM. It can be seen that the accumulated worn volume after 25,000 m of sliding distance reduces from 5.2 107 m3 for the SM-0.3B to 4.4 107 m3 for the SM-0.5 and to 4.0 107 m3 for the SM-0.7B, showing that the increase of boron content also presented positive effects in the plate-cylinder wear test, see Fig. 8(a). However, it can be seen that at the same sliding distance the accumulated worn volume of the commercial SM is 3.9 106 m3, i.e., one order of magnitude higher than the spray-formed boron modified SM, see Fig. 8(b). This means that even small additions as 0.3 wt.% of boron are effective to improve the wear resistance of the supermartensitic stainless steel in this tribosystem. It can be seen that, even with the small increase of the normal force caused by the sample thickness loss, all materials presented the tendency of reducing the wear rate with the increase of sliding distance, as reported by Hall and Malloy [25]. In Fig. 8 and Table 5 can be seen the fitted curves and the fitted parameters of the experimental data with the Eq. (1). The experimental data of all materials were well fitted by the empirical equation purposed as can be seen by the R squared values. The most important value of this equation is the limiting value A, which indicates the maximum accumulated worn volume until the wear rate becomes negligible. By fitting the experimental data it was found the maximum accumulated worn volume being 8.3 107 m3 for the SM-0.3B, 6.0 107 m3 for the SM-0.5B, 4.7 107 m3 for the SM-0.7B and 5.9 106 m3 for the commercial SM. By using this model to evaluate the wear resistance of the material in the POC wear test, the better is the material’s wear resistance, the lower is its maximum accumulated wear volume and the faster it reaches this value. Fig. 9(a) shows the wear rate versus
Fig. 7. SEM images (secondary electrons) of the worn surfaces of (a) commercial SM, (b) SM-0.3B, (c) SM-0.5B and (d) SM-0.7B after the dry sand against rubber wheel abrasive wear test.
220
G. Zepon et al. / Materials & Design 83 (2015) 214–223 Table 5 Fitted parameters from experimental data using Eq. (1). V = A {(1 exp[B (sC)]} A SM-0.3B SM-0.5B SM-0.7B Commercial SM
B 7
8.276 10 5.972 107 4.701 107 5.908 106
3
4.67 10 7.99 103 2.71 103 1.78 104
C
R2
0.527 0.506 0.872 0.867
0.99447 0.99772 0.98043 0.9998
Fig. 8. Experimental data and fitted curves of accumulated worn volume versus sliding distance in the plate-on-cylinder wear test.
sliding distance, calculated by Eq. (2), of all materials. As it can be seen, in all cases by increasing the sliding distance the wear rate approaches to zero, but very high sliding distances are necessary in fact to reach the null value. However, it can be clearly seen that the wear rate of the boron modified SMs decays faster as higher is the boron content of the sample. As expected, for shorter sliding distances the wear rate of the commercial SM is much higher than the boron-modified alloys, but for very high sliding distances it also decays approaching the null value. One can see that for 150 km of sliding distance the wear rate of the commercial SM is almost the same of the SM-0.3B, approximately 1.0 1012 m3/m (or 1 mm3/km), which is a very low wear rate. At the same distance the wear rate of the SM-0.5B is 2.5 1013 m3/m (or 0.25 mm3/km) and of the SM-0.7B practically reached zero. Fig. 9(b) shows the evolution of the contact pressure versus the sliding distance. The contact pressure of all materials decays with the increasing of the sliding distance approaching to the value of the TCP. As the wear rate, the TCP of all materials would be reached only for very high sliding distances. The TCP values found for each material are: 1.02 MPa, 1.13 MPa, 1.23 MPa and 0.56 MPa for the SM-0.3B, SM-0.5B, SM-0.7B and commercial SM, respectively. Based on this, one can see that adding small amounts of boron in the spray formed supermartensitic stainless steel clearly improved the wear resistance in the plate-on-cylinder wear test by reducing the wear rates as well as the maximum accumulated worn volumes and, consequently, increasing the TCPs.
Fig. 9. (a) Wear rate and (b) contact pressure versus sliding distance in the plateon-cylinder wear test.
It is interesting to observe that in the case of the dry sand/ rubber wheel abrasive wear test almost no difference between the wear resistance of the commercial SM and the SM-0.3B could be seen, while in the plate-on-cylinder wear test the behavior of the SM-0.3B was very similar to the SM-0.7B and much superior to the commercial SM. Such different results can be explained by looking to the worn surfaces and the collected debris of the plate-on-cylinder wear test samples in Fig. 10. It can be seen that when the hard cylinder slides against the commercial SM, the asperities and the abrasive particles present in the fluid push the surface material forward in a process that involves considerable plastic deformation. During the sliding the material is detached from the surface in form of debris with plate morphology, as can be clearly seen in Fig. 10(a) and (b). Once the presence of the hard M2B type borides in the microstructure restrict the plastic
G. Zepon et al. / Materials & Design 83 (2015) 214–223
221
Fig. 10. SEM images (secondary electrons) of the worn surfaces and plate-like debris of (a) and (b) commercial SM, (c) and (d) SM-0.3B, (e) and (f) SM-0.5B and (g) and (h) SM0.7B after the plate-on-cylinder wear test sand against rubber wheel abrasive wear test.
deformation of the material, the wear mechanism observed in the commercial SM is less effective in the boron modified alloys. Through the analysis of Fig. 9(c)–(h) it can be noted that the same wear mechanism of pushing and removing plate-like debris occurs in the SM-0.3B, SM-0.5B and SM-0.7B, but in a very reduced scale. When the boron content and the borides fraction are increased, the restriction to plastic deformations also increases resulting in higher wear resistances.
3.3. Corrosion resistance Electrochemical analyses were performed in order to evaluate the effect of boron addition in the chemical composition of the spray-formed boron modified SM. In addition to the SM-0.3B, SM-0.5B and SM-0.7B the SM-14Cr-0.7B was also analyzed to evaluate the effect of the increase in the chromium content in the alloy and, consequently, the increase in the remaining Cr content in the
222
G. Zepon et al. / Materials & Design 83 (2015) 214–223
Fig. 11. Polarization curves in acid solution (pH = 4) for the SM-0.3B, SM-0.5B, SM-0.7B, SM-14Cr-0.7B and commercial SM.
Table 6 Electrochemical properties obtained from the polarization curves in acid media (pH = 4).
SM-0.3B SM-0.5B SM-0.7B SM-14Cr-0.7B Commercial SM
Ecorr (mV)
Ecrit (mV)
DE (mV)
icorr (A/cm2)
503 589 427 405 157
423 530 300 142 +92
80 59 127 263 249
6E6 5E6 8E7 3E7 1E07
martensitic matrix. Fig. 11 shows the polarization curves for all alloys tested in the acid solution (pH = 4) and Table 6 shows the main information obtained from the polarization curves; corrosion potential (Ecorr), corrosion current (icorr), critical potential (Ecrit, defined as the potential where the current density reaches the level of 10 lA/cm2) and potential width before reaching the critical value (DE = Ecrit Ecorr). One can see that the SM-0.5B presented the lower performance in the polarization test, presenting the lower value of Ecorr, 589 mV, a high value of icorr, 5 106 A/cm2, and not showing the presence of a passive plateau. This result can be attributed to the low Cr content present in this alloy, only 10.85 wt.% The Ecorr and icorr measured for the SM-0.3B were -503 mV and 6 106 A/cm2, respectively. The SM-0.3B alloy also presented a passive plateau, but this plateau occurred in very high currents, approximately 2 105 A/cm2, indicating a bad quality of the passive layer. The SM-0.7B has presented better values of Ecorr and icorr than the SM-0.3B, 427 mV and 8 107 A/cm2, but no passive plateau could be seen in this sample, probably due to the higher impoverishment in the matrix Cr content caused by the formation of higher fraction of borides. The SM-14Cr-0.7B presented much better corrosion properties than the other boron modified alloys, presenting Ecorr and icorr values of 405 mV and 3 107 A/cm2, respectively, and the formation of a passive plateau. Although the Ecorr of the SM-14Cr-0.7B be considerably lower than the commercial SM, Ecorr = 157 mV, the icorr values are very close, for the commercial SM icorr = 1 107 A/cm2, as well as the DE, 263 mV and 249 mV, for the SM-14Cr-0.7B and commercial SM, respectively. This better performance of the SM-14Cr-0.7B when compared to the other boron modified SMs can be explained by the increase in the remaining Cr content in the matrix, approximately 13 wt.% for this alloy, which increase the tendency to form a more efficient passive layer. This result shows that increasing the chromium content close to 14 wt.% in the alloy modified with 0.7%B results in a corrosion resistance more similar to the commercial
SM, while maintaining the same microstructure of martensitic matrix with a continuous net of eutectic M2B type borides. 3.4. Final remarks Dense deposits of boron modified supermartensitic stainless steel with refined microstructures composed by equiaxed martensitic grains (with average grain sizes ranging from 10 to 20 lm) with a network of eutectic M2B type borides uniformly distributed at the grain boundaries were successfully produced by spray forming process. The equiaxed, refined and homogeneous microstructures are consequence of the spray forming features which allows the solidification to occur in very small regions avoiding the macrosegregation of boron [6,7]. It was shown that by increasing the boron content from 0.3 wt.% to 0.7 wt.% two effects can be observed: (1) increase of the volume fraction of M2B type borides; and (2) reduction of the average grain size. Both effects contribute to reduce the free mean distance between the adjacent borides, which is very beneficial for the improvement of the abrasive wear resistance in the dry sand against rubber wheel wear test. By analyzing the worn surfaces it was possible to note that the material removal mechanism that takes place in this case is the subsequent indentations of the rolling abrasive particles. In this case, the increase of boride volume fractions as well as the reduction of the distance between then improve the wear resistance by reducing the exposed matrix area and, consequently, prevent the abrasive particle to indent and remove material from the matrix. The same effect on the abrasive wear resistance of cast white irons with high volume of M7C3 type carbides were already reported in the literature [14,29,29]. This can explain the reason for the SM-0.3%B, which has the higher average grain size among the B-containing samples, to present a similar wear resistance compared to the commercial SM, even having a higher hardness. However, very different wear behaviors were observed in the plate-on-cylinder wear test. In this tribosystem, the material is removed from the surface through a mechanism that involves considerably high amounts of surface plastic deformation. In this case, the eutectic M2B boride network formed in the materials, even with only 0.3 wt.%, was effective to restrain the surface plastic deformation, resulting, in all cases, in accumulated worn values one order of magnitude lower than the commercial SM. It is worth to point out that the wear behavior of all materials were accurately described by the empirical model proposed by Hall and Malloy [25] in which the wear rate decreases with the increase of the sliding distance and decrease of the contact pressure. It was clearly shown that increasing the boron content in the chemical composition of
G. Zepon et al. / Materials & Design 83 (2015) 214–223
the spray formed supermartensitic stainless steel is beneficial to improve the wear properties, reducing the wear rates and increasing the TCP. In spite of the improvement in the wear resistance, the addition of boron content in the chemical composition of the conventional medium alloy supermartensitic stainless steel, reduces the corrosion resistance of the alloy due to the impoverishment of the remaining chromium content in the martensitic matrix caused by the formation of Cr-rich borides. The reduction in the chromium content of the matrix can be a serious drawback for the formation and growth of highly protective passive films, thus considerably reducing the corrosion properties of the spray formed boron modified supermartensitic stainless steel. It was however shown that by adding an over content of chromium in the chemical composition of the SM modified with 0.7 wt.% of B, in this case 13.56 wt.% of Cr, it is possible to maintain the same microstructure, with almost same martensitic grain size, volume fraction of M2B borides, and consequently maintaining the same wear properties, but with a higher chromium content in the matrix. This increase in the matrix’s chromium content improves the tendency of formation and the quality of the passive film, clearly improving the corrosion properties of the alloys, which retrieved the corrosion resistance level of the commercial SM in spite of lower corrosion potential values. 4. Conclusions Dense deposits of boron modified supermartensitic stainless steel with 0.3, 0.5 and 0.7 wt.% of boron can be successfully produced by spray forming with equiaxed and refined microstructure composed by martensitic matrix and a continuous network of eutectic M2B type borides at the grain boundaries. The increase in the boron content increases the volume fraction of M2B borides and reduces the martensitic average grain size. The abrasive wear resistance measured by dry sand against rubber wheel is as better as higher is the volume fraction of borides and smaller the distance between then. The empirical model proposed by the petroleum engineers to evaluate the wear of risers and casings can be successfully applied to describe the wear behavior of the materials in the plate-on-cylinder wear test. The presence of the continuous network of M2B in the microstructure of the spray formed boron modified SMs improves the wear resistance of the alloys in the plate-on-cylinder wear test reducing the wear rates and increasing the threshold contact pressure. In this case the wear properties are also improved by increasing the boron content, but all boron modified alloys present maximum accumulated worn volume ten times lower than the commercial SM. The formation of M2B type borides rich in chromium leads to a reduction of the remaining matrix’s chromium content of the boron modified alloys, reducing their corrosion properties. The
223
addition of an over content of chromium, in this case 13.56 wt.%, increases the remaining matrix’s chromium content improving the corrosion properties while maintaining the same microstructural characteristics of the wear resistant spray formed boron modified supermartesitic stainless steel.
Acknowledgements The authors would like to thank the Brazilian institutions FAPESP, CNPq, CAPES and Petrobras for the financial support. References [1] C.P. Linne, F. Blanchard, G.C. Guntz, Corrosion/1997 1997: Paper No. 28. [2] H. Widen, P. Toussaint, L. Courdreuse, Past, present and future of weldable supermartensitic alloys, in: Supermartensitic Stainl. Steel 2002, vol. 1. Brussels, Belgium, 2002. [3] A.G. Haynes, Some factors governing the metallurgy and weldability of 13%Cr and newer Cr–Ni martensitic stainless steel, in: Supermartensitic Stainl. Steel 1999, vol. 1. Brussels, Belgium, 1999. [4] P. Toussaint, J.J. Dufrane, Advances in the making and base material properties of supermartensitic stainless steels, in: Supermartensitic Stainl. Steel 2002, vol. 1. Brussels, Belgium, 2002. [5] C.a. Della Rovere, C.R. Ribeiro, R. Silva, L.F.S. Baroni, N.G. Alcântara, S.E. Kuri, Mater. Sci. Eng., A 586 (2013) 86. [6] P.S. Grant, Prog. Mater. Sci. 39 (1995) 497. [7] P.S. Grant, Metall. Mater. Trans. A Phys. Metall. Mater. Sci. 38 A (2007) 1520. [8] A. Schulz, V. Uhlenwinkel, C. Escher, R. Kohlmann, A. Kulmburg, M.C. Montero, R. Rabitsch, W. Schützenhöfer, D. Stocchi, D. Viale, Mater. Sci. Eng., A 477 (2008) 69. [9] D.M. Goudar, K. Raju, V.C. Srivastava, G.B. Rudrakshi, Mater. Des. 51 (2013) 383. [10] D.M. Goudar, K. Raju, V.C. Srivastava, G.B. Rudrakshi, Mater. Des. 47 (2013) 489. [11] R. Mittal, A. Tomar, D. Singh, J. Mater. Eng. Perform. 23 (2014) 975. [12] Y. Jia, F. Cao, S. Scudino, P. Ma, H. Li, L. Yu, J. Eckert, J. Sun, Mater. Des. 57 (2014) 585. [13] D.N. Hanlon, W.M. Rainforth, C.M. Sellars, Wear 225–229 (1999) 587. [14] a.H. Kasama, a.J. Mourisco, C.S. Kiminami, W.J. Botta Fo, C. Bolfarini, Mater. Sci. Eng., A 589 (2004) 375–377. [15] T.T. Matsuo, C.S. Kiminami, W.J.B. Fo, C. Bolfarini, Wear 259 (2005) 445. [16] K. Kaur, O.P. Pandey, J. Mater. Eng. Perform. 22 (2013) 3101. [17] V.C. Srivastava, G.B. Rudrakshi, V. Uhlenwinkel, S.N. Ojha, J. Mater. Sci. 44 (2009) 2288. [18] E.V. Shadrichev, A.E. Ivanov, Met. Sci. Heat Treat. 26 (1984) 235. [19] E.V. Shadrichev, S.I. Rumyantsev, Met. Sci. Heat Treat. 24 (1982) 495. [20] J.W. Li, G.S. Zhang, S.Z. Wei, Q. Zhao, Appl. Mech. Mater. 52–54 (2011) 1718. [21] H.G. Fu, J.D. Xing, Y.P. Lei, L.M. Huang, J. Mater. Eng. Perform. 20 (2011) 1665. [22] G. Zepon, C.S. Kiminami, W.J. Botta, C. Bolfarini, Mater. Res. J. Mater. 16 (2013) 642. [23] N. Yüksel, S. S ß ahin, Mater. Des. 58 (2014) 491. [24] S. Mahathanabodee, T. Palathai, S. Raadnui, R. Tongsri, N. Sombatsompop, Mater. Des. 46 (2013) 588. [25] K.P. Hall, R.W.M., Soc. Pet. Eng., 2005, SPE 9430:1. [26] D. Gao, L. Sun, J. Lian, Pet. Sci. 7 (2010) 494. [27] P.E. Busby, M.E. Warga, C. Wells, Trans. Am. Inst. Min. Metall. Eng. 197 (1953) 1463. [28] N.J. Calos, E. Graham, D.R. Cousens, P. Christodoulou, C.H.L. Kennard, L.K. Bekessy, S.F. Parker, Mater. Trans. 42 (2001) 496. [29] C.P. Tabrett, I.R. Sare, M.R. Ghomashchi, Int. Mater. Rev. 41 (1996) 59.