- Email: [email protected]
Influence of temperature in arc-activated plasma nitriding of maraging steel in solution annealed and aged conditions
Surface & Coatings Technology 258 (2014) 754–762
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Influence of temperature in arc-activated plasma nitriding of maraging steel in solution annealed and aged conditions J. Fernández de Ara a,⁎, E. Almandoz a, J.F. Palacio a, G.G. Fuentes a, R.J. Rodríguez b, J.A. García b a b
Asociación de la Industria Navarra, Carretera Pamplona 1, 31191 Cordovilla, Spain IMEM Department, Universidad Pública de Navarra, Campus de Arrosadía s/n, 31006 Pamplona, Spain
a r t i c l e
i n f o
Article history: Received 15 May 2014 Accepted in revised form 29 July 2014 Available online 7 August 2014 Keywords: Maraging steel Plasma nitriding Aging Wear resistance Hardness Corrosion
a b s t r a c t Maraging alloys are a group of high performance steels that combine high strength and good fracture toughness. Their excellent combination of properties is due to their microstructure, formed by nanometer-sized intermetallic precipitates embedded in a ductile martensite matrix. Their hardening treatments consist of aging processes that are compatible with thermochemical treatments such as nitriding or nitrocarburizing. An arc-activated plasma nitriding technique has been used in this study to investigate the effect of simultaneous aging and ion nitriding of maraging steel grade 300 at temperatures from 450 °C to 510 °C. The paper compares the results obtained for solution annealed and aged hardened samples in order to verify the compatibility of both processes, aging and nitriding. Chemical composition and micro-hardness profiling, glancing incidence X-ray diffraction (GIXRD), metallographic tests, corrosion studies and wear resistance tests were carried out. The results demonstrate the feasibility of arc-activated ion nitriding to achieve significant improvements of the surface properties of maraging steels and to merge together thermochemical and aging processes, thus reducing their processing time and energy consumption. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Maraging steels are a kind of alloys belonging to ultra-high strength steels that present an excellent strength to toughness ratio, good machinability and weldability and ease of heat treatment [1–4]. The most extended maraging alloys are those containing 18% of nickel and high cobalt levels. Alloys with high titanium and molybdenum contents combine strength values up to 2400 MPa with fracture toughness (KIC) around 70 MPa·m1/2 [2]. Hardening treatment of maraging steel consists of a solution annealing step aiming to get a homogeneous martensitic matrix followed by an aging process at around 500 °C during 1–4 h. The mechanical properties are achieved by precipitation of nanometer-sized intermetallic compounds during the aging stage. Ni3Ti is formed during the first steps – usually expressed as (Ni,Fe,Co)3(Ti,Mo) – followed by Fe2Mo – (Fe,Co, Ni)2(Ti,Mo) – and Fe7Mo6. Some important advantages of maraging steels with respect to other steels are the already abovementioned great resistance to crack propagation and dimension stability during aging treatment, due to their good thermal conductivity. The exceptional properties of maraging steels make them to be considered as strategic materials, although the costly alloy elements that are used in their ⁎ Corresponding author at: AIN, Carretera Pamplona, 1. Cordovilla. E-31191 (Spain). Tel.: + 34 948421190; fax: + 34 948421100. E-mail address: [email protected] (J. Fernández de Ara).
http://dx.doi.org/10.1016/j.surfcoat.2014.07.084 0257-8972/© 2014 Elsevier B.V. All rights reserved.
production explain why they are only used for added value applications. Moreover, maraging alloys present some other limitations, especially their low wear resistance compared to conventional tool steels due to the absence of carbides, as carbon content is below 0.03 wt.%. Surface properties of maraging steels can be improved through the utilization of thermochemical treatments [5–8] and the compatibility of temperature and time can be exploited to make simultaneous aging and nitriding. In particular, plasma assisted nitriding has become one of the most interesting techniques for these materials. Conventional processes such as gas nitriding are usually applied at 500–550 °C, which can affect the mechanical properties of the alloys due to overaging [1,2]. On the contrary, plasma assisted nitriding can be carried out at temperatures lower than 500 °C with good efficiency, avoiding overaging inconveniences and showing good effectiveness for increasing wear resistance, surface hardness and corrosion resistance [6,9–13]. In this study, a plasma assisted nitriding technique was carried out in a conventional PVD reactor. It is based on the generation of a cathodic arc, similar to that of PVD coating processes, to enhance the ionization of the plasma. N2–Ar gases are used as precursors during ion nitriding. The application of the glow discharge is produced by a medium–high current on the metallic cathodes usually employed for the deposition of PVD films. A secondary anode is used as an electron collector in the chamber to increase the ionization of the plasma. Resistive heaters installed at the walls of the reactor allow an accurate control of the nitriding temperature and therefore, of the aging treatment. In this
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762 Table 1 Chemical composition of the main elements in the alloy (wt.%).
755
Table 2 Experimental parameters of the processes.
Element
GDOES
Specified
C Ni Co Mo Ti Al Fe
0.008 17.31 9.05 4.90 0.66 0.11 Balance
b0.03 17–19 8.5–9.5 4.7–5.2 0.6–0.7 b0.15 Balance
work, temperatures of 450 °C, 475 °C, 500 °C and 515 °C during 90 min have been used to investigate the effect of nitriding on maraging steel specimens in annealed condition and also on specimens subjected to prior aging treatment, with special attention to hardness, wear and corrosion resistance. This ion nitriding also allows the development of duplex treatments where the nitriding may be followed by a PVD deposition.
2. Experimental Grade 300 annealed and aged maraging steels were used in this work. Specimens in the annealed condition were denoted as “S” (=solubilized), whereas aged samples were labeled with a “T” (=thermal treated). Table 1 gathers the nominal chemical composition for grade 300 and that averaged on various specimens after a chemical analysis using Glow Discharge Optical Emission Spectroscopy (GDOES). A number of samples of this alloy were prepared in the form of mirror polished discs of 5 mm thick and 35 mm in diameter. The average hardness of the substrates in the annealing condition before plasma nitriding was 30 HRc and 52–53 HRc for the age hardened specimens.
Process temperature (°C)
Intensities at heater (A)
Bias (V)
Time (min)
Ar/N2 mass flow rate
Total pressure (mbar)
450 475 500 515
50 150 350 450
−600 −600 −600 −600
90 90 90 90
3:1 3:1 3:1 3:1
1 1 1 1
× × × ×
10−2 10−2 10−2 10−2
2.1. Arc activated plasma nitriding technique The treatment presented here consists of an arc-activated plasma nitriding technique [14]. The schematic diagram of the arrangement is shown in Fig. 1. The processes were carried out in a commercial PVD equipment (METAPLAS IONON MZR 323). The capacity of the vacuum chamber is 400 mm × 400 mm × 500 mm. The procedure starts with the use of heating resistances and the final cleaning of the samples with the assistance of argon ion bombardment at 400 °C. For the arc-activated plasma nitriding, three chromium cathodes were used. In the treatment, a current of 85 A is applied to the metal cathodes to get a discharge. At the other side of the chamber, a separated secondary anode is polarized to act as an electron collector, gathering the electrons resulting from the cathodic arc electronic thermal emission and those generated in the glow-discharge. Opposite to the cathodes, a shield is placed so the generated vapor stream is blocked and does not reach the specimens. Thanks to this arrangement, only electrons can cross the chamber, which increases the ionization of the gases during their path to the secondary anode. Table 2 summarizes the experimental parameters utilized in this study. Ohmic currents varied from 50 A to 450 A. Samples were biased
Fig. 1. Scheme of the PVD equipment used for the nitriding process: 1, arc cathodes; 2, heating resistance; 3, substrate holder assembly; 4, cathode shield; 5, electron collector (secondary anode).
756
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
a)
b)
10 450 ºC - S 450 ºC - T
9
8
7
7
6
6
5 4
5 4
3
3
2
2
1
1
0
0
10
20
30
40
50
60
70
475 ºC - S 475 ºC - T
9
wt. % N
wt. % N
8
10
0
80
0
10
20
30
Depth μm
c)
d)
10 500 ºC - S 500 ºC - T
9
10
7
6
6
5 4
3 2
1
1 30
40
80
4
2
20
70
5
3
10
60
50
60
70
515 ºC - S 515 ºC - T
8
7
0
50
9
wt. % N
wt. % N
8
0
40 Depth μm
80
Depth μm
0
0
10
20
30
40
50
60
70
80
Depth μm
Fig. 2. Nitrogen diffusion depth profiles of nitrided specimens S and T obtained by GDOES. a) 450 °C; b) 475 °C; c) 500 °C; d) 515 °C.
to −600 V. High purity Ar and N2 in a ratio of 3:1 were fed into the chamber until the working pressure was set at 1 × 10−2 mbar. Process temperatures have been estimated using the tempering hardness–temperature curve of 100Cr6 steel, by putting hardened samples of this alloy in the vacuum chamber during all the processes. This temperature control method was previously designed by performing a series of tests using the same process times and temperatures by using a conventional furnace (error b 0.3%). Thereby, an experimental tempering curve of a quenched batch of 100Cr6 steel samples representing temperature reached vs. hardness was plotted. For the elaboration of this curve, temperatures ranged from 300 °C to 600 °C. Hardness HRc was measured at the back side of the samples after a soft fine-grain polishing to ensure that surface effects are negligible. Concurrently, this method for the measurement of the temperature was also realized with DIN 1.2767 steel obtaining similar and consistent results, which confirmed the reliability of the method. To perform this method, 100Cr6 steel was chosen due to the fact that small variation of temperature leads to big changes of bulk hardness. This latter considering small dispersion of the HRc measurements (0.1–0.2 SD) and the precision of the muffle used for the test, the accuracy of the temperature measuring method can be estimated to be about 4–6 °C for the given range.
2.2. Sample characterization Nitrogen diffusion profiles were studied using Glow Discharge Optical Emission Spectroscopy (GDOES). The analyses were carried out in a JY 10000RF spectrometer (HORIBA-JOBIN YVON) equipped with a RF source working at constant pressure and constant applied power. The optical part of the spectrometer consists of a polychromator of 1 m focal length which led to very reliable and precise time resolved depth profiles. The equipment was calibrated for all the elements of interest
in maraging steels (Ni, C, Mo, Ti, Fe, Co, Al, N and O), in order to translate the measured light intensities into chemical composition [15]. Glancing incidence X-ray diffraction (GIXRD) analyses of the specimens were carried out to study the formation of nitrides on the compound layer. A D8 Advance Bruker diffractometer, using Cr Kα1 radiation (λ = 2.2897 Å) and an incidence angle of 3°, was used. The microstructural characterization was completed using field emission scanning electron microscopy (HITACHI S4800) by evaluating the cross-sections of the nitrided and un-nitrided samples. These samples were fractures starting from previously notched specimens after 10 min of immersion in liquid nitrogen. The tribological characterization included roughness measurements and wear resistance tests at room temperature. Surface finish was measured before and after nitriding in a WYCO-RST 500 profilometer using the vertical scanning interferometry (VSI) mode. For the sliding tests, a CSM THT 8-153 tribometer under ball-on-disc configuration was used and set for 20,000 rotating cycles at normal loads of 10 N using alumina counterballs of 6 mm in diameter. The wear rates and the worn-out tracks were evaluated by measuring the volume losses of the specimens by optical profilometry. Rockwell-C hardness (Instron Testor 930-250) was tested to measure the hardness of the bulk material. The evaluation of micro-hardness profiles was carried out through Knoop indentation measurements at 50 g in a BUEHLER Micromet 2103 on mirror polished cross-sections of the studied specimens. Corrosion tests were performed using a potentiostat device (GAMRY Reference 600) in a three electrode cell. The maraging steel specimens act as working electrodes, whereas a Ag/AgCl electrode and a platinum electrode are used as the reference and the counter electrode, respectively. The tests were performed in aqueous NaCl 0.6 M solution. Corrosion behavior was evaluated by polarization curves which were recorded at a scan rate of 0.5 mV/s after 900 s for the stabilization of the system. The sweep was established from 300 mV below the Eoc to
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
S
16000
α-Fe
(111) (111)
10000
800
γ '-FeNi3N (200) (200)
(110)
12000
N
4
ε -Fe2N1-x
(200)
o
515 C - S
8000
o
500 C - S
6000
o
475 C - S
4000
S
Ref S 450 ºC - S 475 ºC - S 500 ºC - S 515 ºC - S
1000
1
γ '-Fe
14000
Counts
a)
(110)
E (mV) vs Ag/AgCl
18000
(111)
a)
757
600 400 200 0 -200 -400
o
450 C - S
2000
-600
Ref S
0
-7
30
40
50
60
70
80
90
100
-6
-5
-4
-3
α -Fe
600
o
515 C - T
8000
E (mV) vs Ag/AgCl
800
γ '-FeNi3N (200)
(200) (111)
(110)
(101)
(010)
Counts
10000
N
4
ε -Fe2N1-x (200)
14000 12000
1
2
3
4
2
10
3
10
5
o
500 C - T
6000
o
475 C - T
4000
o
450 C - T
2000
T
Ref T 450 ºC - T 475 ºC - T 500 ºC - T 515 ºC - T
1000
1
γ '-Fe
(110)
(111)
16000
0
10 10 10 10 10 10
i (A/cm )
b)
T
(111)
18000
-1
2
Counts
b)
-2
10 10 10 10 10 10 10
110
400 200 0 -200 -400 -600
Ref T
0 30
40
50
60
70
80
90
100
110
Angle 2θ Fig. 3. Glancing incidence angle X-ray diffractogram of the nitrided specimens and the substrate references at various temperatures. a) Solution annealed specimens; b) prior aged specimens.
a potential up to 1.5 V. The corroded area of the specimens was 0.5 cm2. The results of the corrosion tests were assessed based on the information extracted from the polarization curves of a minimum of three repetitions per sample. Corrosion scars were evaluated by optical microscopy. 3. Results and discussion 3.1. Chemical composition in depth profiles and X-ray diffraction Fig. 2 shows GDOES analyses obtained for specimens S and T at different temperatures. The thicknesses of the nitrided layers vary from 45 μm (at 450 °C) to 75 μm (at 515 °C). Nitrogen diffusion presents slight differences between specimens S and T. Thicker layers have been obtained for S samples, especially at temperatures of 500– 515 °C. However, the surface concentration of nitrogen up to the first 5 μm is observed to be greater for the T specimens. This fact is associated to the presence of intermetallic precipitates in the matrix, which block the diffusion of the nitrogen into the substrate [16]. Glancing incidence X-ray diffractograms are shown in Fig. 3, which shows separately the diffraction patterns of S and T samples. The diffractograms of S and T reference specimens show the characteristic α-Fe peaks at 68° (110) and 106° (200). After plasma nitriding, the apparition of new peaks due to the formation of a compound layer is observed. The dominant peaks at 63° (111) and 74° (200) correspond to
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
10
4
5
10
2
i (A/cm ) Fig. 4. Potentiodynamic polarization curves of the nitrided specimens and the un-treated substrates. a) Solution annealed specimens; b) prior aged specimens.
γ′-Fe4N phase (or Fe3NiN, whose peaks overlap with the Fe4N ones), and they are in good agreement with the stoichiometric Fe4N measured at the surface [17]. These phases have been observed at every nitriding temperature in this work, for S and T nitrided specimens. No significant variation of the peak positions was observed, due to the thermodynamic stability of stoichiometric γ′-Fe4N phase in the range of temperature of the present treatments. Furthermore, peaks corresponding to ε-Fe2N1 − x phases can be also observed. For S specimens, a clear dependence of these phases with the temperature can be established. At 450 °C, ε-Fe2N phases were found whereas peaks corresponding to ε-Fe3N phases were observed at a treatment temperature of 475 °C. At 500 °C and 515 °C the ε-phases decreased significantly. On the other hand, the intensity of the γ′-Fe4N phase peaks becomes higher when the temperature increases. These results are consistent with the reported studies [18–20] that explain the evolution of the stoichiometry of these compound layer phases from ε-Fe2N (ε-Fe2N1 − x with x near zero) at low temperatures to ε-Fe3N (ε-Fe2N1 − x with x up to 1/3) while increasing temperature up to 500 °C. For T specimens, a behavior with some substantial differences was noticed. γ′-Fe4N phases can be observed at every nitriding temperature, and the intensity of these peaks becomes greater as the treatment temperature increases. On the other hand, ε-Fe2N1 − x phases can be observed at 450 °C and 475 °C, and also a peak corresponding to the substrate's contribution appears at 475 °C. This may be explained by
758
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
Fig. 5. Corrosion scars. a) Un-treated maraging steel (T); b) 500-T.
the lower amount of nitrogen in the surface for the samples treated at 475 °C and because it is a relatively high temperature to form stable εphases, so the signal of the substrate is observable. At 500 °C a weak peak corresponding to ε-Fe2N1 − x was detected. At 515 °C no presence of this phase was noticed. In general, it can be observed that ε-Fe2N1 − x phases are more stable at higher temperatures in S specimens, which were simultaneously aged and nitrided. This fact will have an important effect on the mechanical properties of the nitrided maraging steels. Especially important will be the effect on the surface micro-hardness, due to the higher hardness of ε-phases compared to γ′-Fe4N. The formation of other phases due to the combination of strong nitride forming elements such as Al, Ti or Mo present in the matrix can also be expected, especially for the specimens in solution annealed condition, where theoretically all these elements were dispersed in the matrix in the initial condition. Aluminum, titanium and molybdenum nitrides would have a strong influence on the mechanical properties of the compound layer and the diffusion zone in S specimens and their bulk hardening. However, some considerations should be taken into account. In the case of aluminum, there is a low concentration of this element (0.11 wt.%), and it does not have a significant role in the formation of intermetallic precipitates, not contributing to relevant differences between S and T nitrided specimens. Titanium (0.66 wt.%) is one of the most active elements in the matrix, and a competitive effect between nitride formation and the precipitation of intermetallics is expected. Nevertheless, it has to be considered that the nitriding process begins after 30 min of pre-heating, where a temperature of 400 °C is reached. It is known from kinetic calculations [2] that Ni3Ti is one of the first precipitates to be formed, thus diminishing the presence of this element in the form of solubilized atoms in the matrix. Consequently, the low amount of titanium and its rapid impoverishment from the matrix limit the formation of titanium nitrides and approach the
Table 3 Ecorr, icorr, Epit and ipit obtained through the polarization curves. Specimens
Ecorr (mV) vs . Ag/AgCl
icorr (A/cm2) × 109
Epit (mV) vs. Ag/AgCl
ipit (A/cm2) × 109
Ref — T Ref — S 450 °C — T 450 °C — S 475 °C — T 475 °C — S 500 °C — T 500 °C — S 515 °C — T 515 °C — S
−408 −396 −56 −119 −68 −119 −67 −181 −72 −140
42.1 272.0 5.33 4.61 5.77 7.55 5.23 5.45 9.21 3.93
– – 448 373 239 158 677 743 940 185
– – 50 63 30 20 736 323 1093 107
response of S and T materials. Molybdenum is less active than aluminum and titanium, but the enthalpy of formation of Mo2N is similar to that of iron nitrides [21], so the presence of molybdenum nitrides would be expected, and accordingly in a higher concentration for S samples than for T samples. GIXRD did not reveal the presence of Mo2N in the compound layer for S nor T samples. Only a slight increase in the hardness of the nitrided zone in S specimens was detected that may correspond to the contribution of ε phases and other nitrides. This point will be subject of further studies. It is interesting to point out that X-ray diffraction patterns show significantly broader peaks for the nitrided phases in S specimens with respect to those observed for the T specimens, especially in the case of γ′-Fe4N phases. It may be explained owing to the formation of inhomogeneous micro-strained structures resulting from the incorporation of small nitrides, and to the contribution of γ′-FeNi3N phases, whose diffraction peaks can be found next to γ′-Fe4N, displaced 0.2° to higher angles. Actually, small shoulders can be observed on the right side of the γ′-Fe4N peaks at 63° for S specimens treated at 450 °C and 475 °C, which can be attributed to γ′-FeNi3N. This is consistent with the expected higher content of Ni in the matrix taking into account the lower amount of Ni-rich precipitates due to the lower treatment temperatures and the fact that S specimens were not previously precipitation hardened. It is also remarkable that no signs of overaging were evidenced by means of X-ray diffraction, as no peaks of reverse austenite were found. It is also correlated with the HRc results shown in Table 4, where a loss of base hardness was not appreciated. 3.2. Corrosion tests The corrosion resistance of the nitrided specimens was studied and compared to the untreated material through potentiodynamic tests. Fig. 4 shows the polarization curves of plasma nitrided specimens S (a) and T (b). An important enhancement of the corrosion resistance of the maraging surfaces after plasma nitriding can be observed [6]. Actually, a change in the corrosion behavior can be observed. Fig. 4 displays a uniform corrosion behavior for un-nitrided maraging steel, typically observed for mild steels. After ion nitriding treatments, a pseudo-passive layer is formed in both S and T specimens, offering a better protection to the material regardless of the treatment temperature. Fig. 5a–b shows the corrosion tested surfaces of the un-treated maraging steel T and the nitrided one at 500 °C after the tests. Uniform and massive corrosion can be observed for the reference materials S and T, whereas the appearance of pitting corrosion is common to all the nitrided specimens, both S and T (c.f. Fig. 5b). Li and Bell [22] reported similar corrosion results in their study of plasma nitriding of AISI 410 martensitic stainless steel. The values of corrosion potential (Ecorr),
Fig. 6. SEM cross-sections of the reference materials and the nitrided specimens. a) Un-treated maraging steel S; b) un-treated maraging steel T; c) 450-S; d) 450-T; e) 475-S; f) 475-T; g) 500-S; h) 500-T; i) 515-S; j) 515-T.
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
a)
b)
c)
d)
e)
f)
g)
h)
i)
j)
759
760
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
a)
Table 4 Roughness, hardness HRc and HK0.05, and wear rate.
Ref — T Ref — S 450 °C — T 450 °C — S 475 °C — T 475 °C — S 500 °C — T 500 °C — S 515 °C — T 515 °C — S
Roughness
Hardness
Ra nm
HRc
7 8 179 202 208 202 221 284 200 266
± ± ± ± ± ± ± ± ± ±
2 2 2 4 4 5 2 3 4 4
52.3 30.2 53.0 47.9 53.3 51.4 53.1 51.3 53.3 52.9
± ± ± ± ± ± ± ± ± ±
0.3 0.2 0.2 0.3 0.4 0.3 0.3 0.2 0.4 0.2
K (m3 × N−1 × m−1)
– – 745 752 639 649 730 781 674 759
3.81 1.62 1.98 1.61 1.09 1.12 1.43 2.11 9.25 1.02
48 36 37 36 27 44 18 21
± ± ± ± ± ± ± ± ± ±
3.11 0.18 1.04 0.98 0.11 0.07 0.29 1.30 0.25 0.06
o
450 C - S o 475 C - S o 500 C - S o 515 C - S
850
Surface HK0.05
± ± ± ± ± ± ± ±
S
900
Wear rate
× × × × × × × × × ×
10−13 10−12 10−14 10−14 10−14 10−14 10−14 10−14 10−15 10−14
800 750 700
HK0.05
Specimens
650 600 550 500 450 400 0
10
20
30
40
50
60
70
80
90
Depth μm
b)
T
900
o
450 C - T o 475 C - T o 500 C - T o 515 C - T
850 800 750 700
HK0.05
critical current density (icorr), pitting potential (Epit) and pitting current (ipit) are listed in Table 3. An important displacement to higher corrosion potential values after plasma nitriding is also observed (from − 400 V to around − 60 V for T specimens and between − 120 V and −180 V for S specimens). The T and S samples treated at 475 °C show worse behavior than the rest, presenting significant lower Epit values. This can be understood observing the compound layer of the specimens nitrided at 475 °C. For both S and T samples at this temperature a higher proportion of mixtures of different phases can be observed, penalizing the corrosion resistance. It is well known that a compound layer with a mono-phasic structure of γ′-Fe4N present a better behavior than that containing ε-Fe2N1 − x phases or biphasic structures [23]. In general, for T specimens, corrosion resistance increases as the temperature increases, as the result of the formation of a greater content of γ′-Fe4N phase over ε-Fe2N1 − x.
650 600 550
3.3. Microstructure, hardness, roughness and wear tests SEM micrographs of the un-treated and nitrided specimens in cross-section are depicted in Fig. 6a–j. The typical martensitic structure corresponding to maraging steels can be observed for S and T specimens (c.f. Fig. 6a–b) [2,8]. No differences can be detected between S and T specimens owing to the small size of the precipitates in the aged samples, well below the detection limits of FE-SEM instruments. Nevertheless, the presence of these intermetallic compounds can be taken for granted considering the base hardness of both types of samples, later explained (Table 4). In the micrographies of the cross-sections shown in Fig. 6c–j, a ductile fracture behavior of the maraging steel substrate can be observed, whereas the nitrided zones, whose thickness agrees with the measured one using GDOES, present a fragile fracture behavior owing to the iron nitrides (γ′-Fe4N and ε-Fe2N1 − x). A ductile behavior of the substrates can be noticed where the diffusion profile of nitrogen ends, with the same microstructure than that of the un-treated specimens. Roughness measurements (Table 4) show that the Ra values increase from below 10 nm to more than 200 nm due to the effect of the intensive ion bombardment. Table 4 also shows the HRc values for all studied samples. Used temperatures and nitriding times are compatible with aging treatments for maraging steel hardening, as it was suggested by GIXRD measurements due to the absence of austenite. Considerable increases of HRc values are observed for S specimens after treatment, which confirms that plasma nitriding can also be used for simultaneous age hardening of maraging steel, as it has been previously reported for other authors studying simultaneous thermochemical treatments and bulk hardening of martensitic steels [24]. A raise of hardness Rockwell C up to 48–53 HRc was measured, according to the treatment temperature. However, at 450 °C bulk hardness is smaller than that obtained at higher temperatures, indicating that the aging treatment was not fully completed. No
500 450 400 0
10
20
30
40
50
60
70
80
90
Depth μm Fig. 7. Knoop hardness at 50 g load depth profiles. a) Solution annealed specimens (S); b) prior aged specimens (T).
evidence of overaging was observed for any of the pre-aged specimens in light of the Rockwell C values (Table 4), in agreement with the XRD and also considering the microstructure images (Fig. 6). Fig. 7 shows the Knoop micro-hardness (50 g load) profiles measured on the mirror polished cross-section of S and T specimens after nitriding. Greater surface hardness followed by a sharper fall of the values was observed for the S specimens in comparison with the T ones. As it is shown in Table 4, in all the cases surface hardness is greater for the S specimens, especially at higher temperatures. This is consistent with the presence of ε-Fe2N phases in the compound layer, which are harder than those of γ′-Fe4N. For T specimens, the presence of ε-Fe2N phases is bigger at lower temperatures, so higher surface hardness values at those temperatures were actually expected. Maximum surface hardness does not vary significantly for the S specimens, with the next exception. Low hardness values were obtained in the surface (up to the first 5–10 μm) for the samples treated at 475 °C, in which a lower nitrogen amount in the surface was detected by GDOES analyses. Both S and T specimens presented considerable low surface hardness values. At this temperature, GIXRD showed peaks belonging to the signal of the steel for T specimens and weak peaks of ε-phases, so a low surface hardness is consistent with the reported data. However, for S specimens GIXRD revealed peaks corresponding to ε-Fe3N phases in good agreement with
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762 1800 1500
Wear Rate
1200 900 300 35
12
3
-1
-1
K * 10 (m ·N ·m )
600
30 25 20 15 10 5 0
Ref T Ref S 450 ºC 450 ºC 475 ºC 475 ºC 500 ºC 500 ºC 515 ºC 515 ºC
Fig. 8. Comparison of the wear rate of the un-treated maraging steel and the specimens treated at various temperatures.
761
that they were not detected through the analysis techniques being used in this work it will be necessary further studies to sustain this point. Wear resistance was also studied with the help of a pin-on-disc tribometer at room temperature. Fig. 8 shows the wear coefficients of the nitrided specimens. A great improvement of the wear resistance was obtained after plasma nitriding in all the cases (one order of magnitude for T specimens and two orders for S specimens). Although small differences were found between the specimens, a good correlation between surface Knoop hardness at 50 g and wear rate can be observed. In general, specimens which present a greater surface hardness exhibit greater wear rate, therefore suggesting third-body wear effects during the tests. This fact can be explained owing to the fragile behavior of the ε phases present in the specimens with higher hardness. The surface topography of the tracks of the un-treated material and the nitrided specimens (S and T) at 500 °C is displayed in Fig. 9, where abrasive wear can be observed. Deeper tracks and greater volume losses are observed for the un-nitrided samples. 4. Conclusions
that stoichiometry at 475 °C. To explain this fact, the nature of the nitrides present in the compound layer should be taken into account, considering that the hardness of ε-Fe2N phase (ε-Fe2N1 − x with x near zero) is higher than that of ε-Fe3N (ε-Fe2N1 − x with x near 1/3). The formation of these phases is influenced by the treatment temperature [18–20] and the nitriding activity and, consequently, the concentration of nitrogen that incorporates to the material. On the other hand, the fact that other nitrides such as Mo2N might be present at a higher concentration in S specimens and the fact that they could also contribute to the higher hardness have to be considered. However, taking into account
Arc-activated plasma nitriding is a feasible technology for the nitriding of maraging steel in both solution annealed and aged conditions. On top of this, simultaneous aging and nitriding processes are possible on solution annealed specimens of maraging steel grade 300 at temperatures from 475 °C to 515 °C, whereas at 450 °C the aging process did not lead to the specified bulk hardness that corresponds to complete aging process. Arc-activated plasma nitriding significantly enhanced corrosion properties of maraging steel and also led to important improvements in wear resistance. After the plasma nitriding treatments
65 µm
120 µm
a)
b)
8 µm
c)
9 µm
d)
Fig. 9. 3D-images and 2D-profiles of the wear tracks measured by optical profilometry. a) Un-treated maraging steel (S); b) un-treated maraging steel (T); c) 500-S; d) 500-T.
762
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
in both conditions of the maraging steel, the formation of γ′-Fe4N and εFe2N1 − x phases was observed in the compound layer. In general, εphases were preferably found at lower treatment temperatures, while the amount of γ′-Fe4N phase increased with the temperature. Furthermore, ε-Fe2N1 − x phases were also detected at the highest nitriding temperatures in the specimens that were not previously aged. This fact explains the greater surface hardness found for solution annealed nitrided maraging steel.
[7] [8] [9] [10] [11] [12] [13] [14] [15]
Acknowledgments [16]
The authors acknowledge the financial support from Navarra's Government (IIQ1404) through ‘Supermaraging’ project and the Ministry of Economy and Competitiveness of Spain through the Consolider-Ingenio project ‘Funcoat’ (2010-CSD2008-00023). References [1] S. Floreen, Metall. Rev. 13 (1968) 115–128. [2] W. Sha, Z. Guo, Maraging steels, Modeling of Microstructure, Properties and Applications, first ed., Woodhead Publishing Limited, New Cambridge, 2009. (Cambridge). [3] B. Gupta, V. Gopalakrishna, J.S. Yadav, A. Kumar, B. Saha, Aerospace Materials, S. Chang and Company, New Delhi, 1996. 695. [4] J. Rezek, I.E. Klein, J. Yahalom, Appl. Surf. Sci. 108 (1997) 159–165. [5] Kishora Shetty, S. Kumar, P. Raghothama Rao, J. Phys. Conf. Ser. 100 (2008) 062013. [6] Kishora Shetty, S. Kumar, P. Raghothama Rao, Surf. Coat. Technol. 203 (2009) 1530–1536.
[17] [18] [19] [20] [21] [22] [23]
[24]
Yuji Ohue, Koji Matsumoto, Wear 263 (2007) 782–789. K. Hussain, A. Tauqir, A. ul Haq, A.Q. Khan, Int. J. Fatigue 21 (1999) 163–168. T. Spalvins, ‘Ion Nitriding’ Conference Proceedings, 1, ASM International, 1986. A. Leyland, K.S. Fancey, A.S. James, A. Matthews, Surf. Coat. Technol. 41 (1990) 295. W. Möller, S. Parascandola, T. Telbizova, R. Günzel, E. Richter, Surf. Coat. Technol. 136 (2001) 73–79. M.B. Karamiş, E. Gerçekcioĝlu, Wear 243 (2000) 76–84. H. Michel, T. Czerwiec, M. Gantois, D. Ablitzer, A. Ricard, Surf. Coat. Technol. 72 (1995) 103–111. G.G. Fuentes, R. Rodríguez, J.A. García. European Patent EP2369028 (A1), assigned to Asociación de la Industria Navarra (AIN). J.A. García, R.J. Rodríguez, R. Martínez, C. Fernández, A. Fernández, R. Payling, Appl. Surf. Sci. 235 (2004) 97. H. Selg, Nitriding of Fe–Mo alloys and maraging steel: structure, morphology and kinetics of nitride precipitation, PhD. thesis Max-Planck-Institut für Intelligente Systeme (ehemals Max-Planck-Institut für Metallforschung), Stuttgart, 2012. (University of Stuttgart). K.H. Jack, Proc. R Soc. A 195 (1948) 34. M.K. Lei, Z.L. Zhang, Surf. Coat. Technol. 91 (1997) 25–31. A.M. Vredenberg, C.M. Perez-Martin, J.S. Custer, D.O. Boerma, L. de Wit, F.W. Saris, N. M. van der Pers, Th.H. de Keijser, E.J. Mittemeijer, J. Mater. Res. 7 (1992) 2689. R. Hutchings, G.A. Collins, J. Tendys, Surf. Coat. Technol. 51 (1992) 489. D.A. Porter, K.E. Easterling, Phase Transformations in Metals and Alloys, CRC Press, Taylor & Francis Group, Finland, 2004. C.X. Li, T. Bell, Corros. Sci. 48 (2006) 2036–2049. J.C. Díaz-Guillén, G. Vargas-Gutiérrez, E.E. Granda-Gutiérrez, J.S. Zamarripa-Piña, S.I. Pérez-Aguilar, J. Candelas-Ramírez, L. Álvarez-Contreras, S.I. Pérez-Aguilar, J. Candelas-Ramírez, L. Álvarez-Contreras, J. Mater. Sci. Technol. 29 (3) (2013) 287–290. R.B. Frandsen, T. Christiansen, M.A.J. Somers, Surf. Coat. Technol. 200 (2006) 5160–5169.
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Influence of temperature in arc-activated plasma nitriding of maraging steel in solution annealed and aged conditions J. Fernández de Ara a,⁎, E. Almandoz a, J.F. Palacio a, G.G. Fuentes a, R.J. Rodríguez b, J.A. García b a b
Asociación de la Industria Navarra, Carretera Pamplona 1, 31191 Cordovilla, Spain IMEM Department, Universidad Pública de Navarra, Campus de Arrosadía s/n, 31006 Pamplona, Spain
a r t i c l e
i n f o
Article history: Received 15 May 2014 Accepted in revised form 29 July 2014 Available online 7 August 2014 Keywords: Maraging steel Plasma nitriding Aging Wear resistance Hardness Corrosion
a b s t r a c t Maraging alloys are a group of high performance steels that combine high strength and good fracture toughness. Their excellent combination of properties is due to their microstructure, formed by nanometer-sized intermetallic precipitates embedded in a ductile martensite matrix. Their hardening treatments consist of aging processes that are compatible with thermochemical treatments such as nitriding or nitrocarburizing. An arc-activated plasma nitriding technique has been used in this study to investigate the effect of simultaneous aging and ion nitriding of maraging steel grade 300 at temperatures from 450 °C to 510 °C. The paper compares the results obtained for solution annealed and aged hardened samples in order to verify the compatibility of both processes, aging and nitriding. Chemical composition and micro-hardness profiling, glancing incidence X-ray diffraction (GIXRD), metallographic tests, corrosion studies and wear resistance tests were carried out. The results demonstrate the feasibility of arc-activated ion nitriding to achieve significant improvements of the surface properties of maraging steels and to merge together thermochemical and aging processes, thus reducing their processing time and energy consumption. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Maraging steels are a kind of alloys belonging to ultra-high strength steels that present an excellent strength to toughness ratio, good machinability and weldability and ease of heat treatment [1–4]. The most extended maraging alloys are those containing 18% of nickel and high cobalt levels. Alloys with high titanium and molybdenum contents combine strength values up to 2400 MPa with fracture toughness (KIC) around 70 MPa·m1/2 [2]. Hardening treatment of maraging steel consists of a solution annealing step aiming to get a homogeneous martensitic matrix followed by an aging process at around 500 °C during 1–4 h. The mechanical properties are achieved by precipitation of nanometer-sized intermetallic compounds during the aging stage. Ni3Ti is formed during the first steps – usually expressed as (Ni,Fe,Co)3(Ti,Mo) – followed by Fe2Mo – (Fe,Co, Ni)2(Ti,Mo) – and Fe7Mo6. Some important advantages of maraging steels with respect to other steels are the already abovementioned great resistance to crack propagation and dimension stability during aging treatment, due to their good thermal conductivity. The exceptional properties of maraging steels make them to be considered as strategic materials, although the costly alloy elements that are used in their ⁎ Corresponding author at: AIN, Carretera Pamplona, 1. Cordovilla. E-31191 (Spain). Tel.: + 34 948421190; fax: + 34 948421100. E-mail address: [email protected] (J. Fernández de Ara).
http://dx.doi.org/10.1016/j.surfcoat.2014.07.084 0257-8972/© 2014 Elsevier B.V. All rights reserved.
production explain why they are only used for added value applications. Moreover, maraging alloys present some other limitations, especially their low wear resistance compared to conventional tool steels due to the absence of carbides, as carbon content is below 0.03 wt.%. Surface properties of maraging steels can be improved through the utilization of thermochemical treatments [5–8] and the compatibility of temperature and time can be exploited to make simultaneous aging and nitriding. In particular, plasma assisted nitriding has become one of the most interesting techniques for these materials. Conventional processes such as gas nitriding are usually applied at 500–550 °C, which can affect the mechanical properties of the alloys due to overaging [1,2]. On the contrary, plasma assisted nitriding can be carried out at temperatures lower than 500 °C with good efficiency, avoiding overaging inconveniences and showing good effectiveness for increasing wear resistance, surface hardness and corrosion resistance [6,9–13]. In this study, a plasma assisted nitriding technique was carried out in a conventional PVD reactor. It is based on the generation of a cathodic arc, similar to that of PVD coating processes, to enhance the ionization of the plasma. N2–Ar gases are used as precursors during ion nitriding. The application of the glow discharge is produced by a medium–high current on the metallic cathodes usually employed for the deposition of PVD films. A secondary anode is used as an electron collector in the chamber to increase the ionization of the plasma. Resistive heaters installed at the walls of the reactor allow an accurate control of the nitriding temperature and therefore, of the aging treatment. In this
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762 Table 1 Chemical composition of the main elements in the alloy (wt.%).
755
Table 2 Experimental parameters of the processes.
Element
GDOES
Specified
C Ni Co Mo Ti Al Fe
0.008 17.31 9.05 4.90 0.66 0.11 Balance
b0.03 17–19 8.5–9.5 4.7–5.2 0.6–0.7 b0.15 Balance
work, temperatures of 450 °C, 475 °C, 500 °C and 515 °C during 90 min have been used to investigate the effect of nitriding on maraging steel specimens in annealed condition and also on specimens subjected to prior aging treatment, with special attention to hardness, wear and corrosion resistance. This ion nitriding also allows the development of duplex treatments where the nitriding may be followed by a PVD deposition.
2. Experimental Grade 300 annealed and aged maraging steels were used in this work. Specimens in the annealed condition were denoted as “S” (=solubilized), whereas aged samples were labeled with a “T” (=thermal treated). Table 1 gathers the nominal chemical composition for grade 300 and that averaged on various specimens after a chemical analysis using Glow Discharge Optical Emission Spectroscopy (GDOES). A number of samples of this alloy were prepared in the form of mirror polished discs of 5 mm thick and 35 mm in diameter. The average hardness of the substrates in the annealing condition before plasma nitriding was 30 HRc and 52–53 HRc for the age hardened specimens.
Process temperature (°C)
Intensities at heater (A)
Bias (V)
Time (min)
Ar/N2 mass flow rate
Total pressure (mbar)
450 475 500 515
50 150 350 450
−600 −600 −600 −600
90 90 90 90
3:1 3:1 3:1 3:1
1 1 1 1
× × × ×
10−2 10−2 10−2 10−2
2.1. Arc activated plasma nitriding technique The treatment presented here consists of an arc-activated plasma nitriding technique [14]. The schematic diagram of the arrangement is shown in Fig. 1. The processes were carried out in a commercial PVD equipment (METAPLAS IONON MZR 323). The capacity of the vacuum chamber is 400 mm × 400 mm × 500 mm. The procedure starts with the use of heating resistances and the final cleaning of the samples with the assistance of argon ion bombardment at 400 °C. For the arc-activated plasma nitriding, three chromium cathodes were used. In the treatment, a current of 85 A is applied to the metal cathodes to get a discharge. At the other side of the chamber, a separated secondary anode is polarized to act as an electron collector, gathering the electrons resulting from the cathodic arc electronic thermal emission and those generated in the glow-discharge. Opposite to the cathodes, a shield is placed so the generated vapor stream is blocked and does not reach the specimens. Thanks to this arrangement, only electrons can cross the chamber, which increases the ionization of the gases during their path to the secondary anode. Table 2 summarizes the experimental parameters utilized in this study. Ohmic currents varied from 50 A to 450 A. Samples were biased
Fig. 1. Scheme of the PVD equipment used for the nitriding process: 1, arc cathodes; 2, heating resistance; 3, substrate holder assembly; 4, cathode shield; 5, electron collector (secondary anode).
756
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
a)
b)
10 450 ºC - S 450 ºC - T
9
8
7
7
6
6
5 4
5 4
3
3
2
2
1
1
0
0
10
20
30
40
50
60
70
475 ºC - S 475 ºC - T
9
wt. % N
wt. % N
8
10
0
80
0
10
20
30
Depth μm
c)
d)
10 500 ºC - S 500 ºC - T
9
10
7
6
6
5 4
3 2
1
1 30
40
80
4
2
20
70
5
3
10
60
50
60
70
515 ºC - S 515 ºC - T
8
7
0
50
9
wt. % N
wt. % N
8
0
40 Depth μm
80
Depth μm
0
0
10
20
30
40
50
60
70
80
Depth μm
Fig. 2. Nitrogen diffusion depth profiles of nitrided specimens S and T obtained by GDOES. a) 450 °C; b) 475 °C; c) 500 °C; d) 515 °C.
to −600 V. High purity Ar and N2 in a ratio of 3:1 were fed into the chamber until the working pressure was set at 1 × 10−2 mbar. Process temperatures have been estimated using the tempering hardness–temperature curve of 100Cr6 steel, by putting hardened samples of this alloy in the vacuum chamber during all the processes. This temperature control method was previously designed by performing a series of tests using the same process times and temperatures by using a conventional furnace (error b 0.3%). Thereby, an experimental tempering curve of a quenched batch of 100Cr6 steel samples representing temperature reached vs. hardness was plotted. For the elaboration of this curve, temperatures ranged from 300 °C to 600 °C. Hardness HRc was measured at the back side of the samples after a soft fine-grain polishing to ensure that surface effects are negligible. Concurrently, this method for the measurement of the temperature was also realized with DIN 1.2767 steel obtaining similar and consistent results, which confirmed the reliability of the method. To perform this method, 100Cr6 steel was chosen due to the fact that small variation of temperature leads to big changes of bulk hardness. This latter considering small dispersion of the HRc measurements (0.1–0.2 SD) and the precision of the muffle used for the test, the accuracy of the temperature measuring method can be estimated to be about 4–6 °C for the given range.
2.2. Sample characterization Nitrogen diffusion profiles were studied using Glow Discharge Optical Emission Spectroscopy (GDOES). The analyses were carried out in a JY 10000RF spectrometer (HORIBA-JOBIN YVON) equipped with a RF source working at constant pressure and constant applied power. The optical part of the spectrometer consists of a polychromator of 1 m focal length which led to very reliable and precise time resolved depth profiles. The equipment was calibrated for all the elements of interest
in maraging steels (Ni, C, Mo, Ti, Fe, Co, Al, N and O), in order to translate the measured light intensities into chemical composition [15]. Glancing incidence X-ray diffraction (GIXRD) analyses of the specimens were carried out to study the formation of nitrides on the compound layer. A D8 Advance Bruker diffractometer, using Cr Kα1 radiation (λ = 2.2897 Å) and an incidence angle of 3°, was used. The microstructural characterization was completed using field emission scanning electron microscopy (HITACHI S4800) by evaluating the cross-sections of the nitrided and un-nitrided samples. These samples were fractures starting from previously notched specimens after 10 min of immersion in liquid nitrogen. The tribological characterization included roughness measurements and wear resistance tests at room temperature. Surface finish was measured before and after nitriding in a WYCO-RST 500 profilometer using the vertical scanning interferometry (VSI) mode. For the sliding tests, a CSM THT 8-153 tribometer under ball-on-disc configuration was used and set for 20,000 rotating cycles at normal loads of 10 N using alumina counterballs of 6 mm in diameter. The wear rates and the worn-out tracks were evaluated by measuring the volume losses of the specimens by optical profilometry. Rockwell-C hardness (Instron Testor 930-250) was tested to measure the hardness of the bulk material. The evaluation of micro-hardness profiles was carried out through Knoop indentation measurements at 50 g in a BUEHLER Micromet 2103 on mirror polished cross-sections of the studied specimens. Corrosion tests were performed using a potentiostat device (GAMRY Reference 600) in a three electrode cell. The maraging steel specimens act as working electrodes, whereas a Ag/AgCl electrode and a platinum electrode are used as the reference and the counter electrode, respectively. The tests were performed in aqueous NaCl 0.6 M solution. Corrosion behavior was evaluated by polarization curves which were recorded at a scan rate of 0.5 mV/s after 900 s for the stabilization of the system. The sweep was established from 300 mV below the Eoc to
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
S
16000
α-Fe
(111) (111)
10000
800
γ '-FeNi3N (200) (200)
(110)
12000
N
4
ε -Fe2N1-x
(200)
o
515 C - S
8000
o
500 C - S
6000
o
475 C - S
4000
S
Ref S 450 ºC - S 475 ºC - S 500 ºC - S 515 ºC - S
1000
1
γ '-Fe
14000
Counts
a)
(110)
E (mV) vs Ag/AgCl
18000
(111)
a)
757
600 400 200 0 -200 -400
o
450 C - S
2000
-600
Ref S
0
-7
30
40
50
60
70
80
90
100
-6
-5
-4
-3
α -Fe
600
o
515 C - T
8000
E (mV) vs Ag/AgCl
800
γ '-FeNi3N (200)
(200) (111)
(110)
(101)
(010)
Counts
10000
N
4
ε -Fe2N1-x (200)
14000 12000
1
2
3
4
2
10
3
10
5
o
500 C - T
6000
o
475 C - T
4000
o
450 C - T
2000
T
Ref T 450 ºC - T 475 ºC - T 500 ºC - T 515 ºC - T
1000
1
γ '-Fe
(110)
(111)
16000
0
10 10 10 10 10 10
i (A/cm )
b)
T
(111)
18000
-1
2
Counts
b)
-2
10 10 10 10 10 10 10
110
400 200 0 -200 -400 -600
Ref T
0 30
40
50
60
70
80
90
100
110
Angle 2θ Fig. 3. Glancing incidence angle X-ray diffractogram of the nitrided specimens and the substrate references at various temperatures. a) Solution annealed specimens; b) prior aged specimens.
a potential up to 1.5 V. The corroded area of the specimens was 0.5 cm2. The results of the corrosion tests were assessed based on the information extracted from the polarization curves of a minimum of three repetitions per sample. Corrosion scars were evaluated by optical microscopy. 3. Results and discussion 3.1. Chemical composition in depth profiles and X-ray diffraction Fig. 2 shows GDOES analyses obtained for specimens S and T at different temperatures. The thicknesses of the nitrided layers vary from 45 μm (at 450 °C) to 75 μm (at 515 °C). Nitrogen diffusion presents slight differences between specimens S and T. Thicker layers have been obtained for S samples, especially at temperatures of 500– 515 °C. However, the surface concentration of nitrogen up to the first 5 μm is observed to be greater for the T specimens. This fact is associated to the presence of intermetallic precipitates in the matrix, which block the diffusion of the nitrogen into the substrate [16]. Glancing incidence X-ray diffractograms are shown in Fig. 3, which shows separately the diffraction patterns of S and T samples. The diffractograms of S and T reference specimens show the characteristic α-Fe peaks at 68° (110) and 106° (200). After plasma nitriding, the apparition of new peaks due to the formation of a compound layer is observed. The dominant peaks at 63° (111) and 74° (200) correspond to
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
10
4
5
10
2
i (A/cm ) Fig. 4. Potentiodynamic polarization curves of the nitrided specimens and the un-treated substrates. a) Solution annealed specimens; b) prior aged specimens.
γ′-Fe4N phase (or Fe3NiN, whose peaks overlap with the Fe4N ones), and they are in good agreement with the stoichiometric Fe4N measured at the surface [17]. These phases have been observed at every nitriding temperature in this work, for S and T nitrided specimens. No significant variation of the peak positions was observed, due to the thermodynamic stability of stoichiometric γ′-Fe4N phase in the range of temperature of the present treatments. Furthermore, peaks corresponding to ε-Fe2N1 − x phases can be also observed. For S specimens, a clear dependence of these phases with the temperature can be established. At 450 °C, ε-Fe2N phases were found whereas peaks corresponding to ε-Fe3N phases were observed at a treatment temperature of 475 °C. At 500 °C and 515 °C the ε-phases decreased significantly. On the other hand, the intensity of the γ′-Fe4N phase peaks becomes higher when the temperature increases. These results are consistent with the reported studies [18–20] that explain the evolution of the stoichiometry of these compound layer phases from ε-Fe2N (ε-Fe2N1 − x with x near zero) at low temperatures to ε-Fe3N (ε-Fe2N1 − x with x up to 1/3) while increasing temperature up to 500 °C. For T specimens, a behavior with some substantial differences was noticed. γ′-Fe4N phases can be observed at every nitriding temperature, and the intensity of these peaks becomes greater as the treatment temperature increases. On the other hand, ε-Fe2N1 − x phases can be observed at 450 °C and 475 °C, and also a peak corresponding to the substrate's contribution appears at 475 °C. This may be explained by
758
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
Fig. 5. Corrosion scars. a) Un-treated maraging steel (T); b) 500-T.
the lower amount of nitrogen in the surface for the samples treated at 475 °C and because it is a relatively high temperature to form stable εphases, so the signal of the substrate is observable. At 500 °C a weak peak corresponding to ε-Fe2N1 − x was detected. At 515 °C no presence of this phase was noticed. In general, it can be observed that ε-Fe2N1 − x phases are more stable at higher temperatures in S specimens, which were simultaneously aged and nitrided. This fact will have an important effect on the mechanical properties of the nitrided maraging steels. Especially important will be the effect on the surface micro-hardness, due to the higher hardness of ε-phases compared to γ′-Fe4N. The formation of other phases due to the combination of strong nitride forming elements such as Al, Ti or Mo present in the matrix can also be expected, especially for the specimens in solution annealed condition, where theoretically all these elements were dispersed in the matrix in the initial condition. Aluminum, titanium and molybdenum nitrides would have a strong influence on the mechanical properties of the compound layer and the diffusion zone in S specimens and their bulk hardening. However, some considerations should be taken into account. In the case of aluminum, there is a low concentration of this element (0.11 wt.%), and it does not have a significant role in the formation of intermetallic precipitates, not contributing to relevant differences between S and T nitrided specimens. Titanium (0.66 wt.%) is one of the most active elements in the matrix, and a competitive effect between nitride formation and the precipitation of intermetallics is expected. Nevertheless, it has to be considered that the nitriding process begins after 30 min of pre-heating, where a temperature of 400 °C is reached. It is known from kinetic calculations [2] that Ni3Ti is one of the first precipitates to be formed, thus diminishing the presence of this element in the form of solubilized atoms in the matrix. Consequently, the low amount of titanium and its rapid impoverishment from the matrix limit the formation of titanium nitrides and approach the
Table 3 Ecorr, icorr, Epit and ipit obtained through the polarization curves. Specimens
Ecorr (mV) vs . Ag/AgCl
icorr (A/cm2) × 109
Epit (mV) vs. Ag/AgCl
ipit (A/cm2) × 109
Ref — T Ref — S 450 °C — T 450 °C — S 475 °C — T 475 °C — S 500 °C — T 500 °C — S 515 °C — T 515 °C — S
−408 −396 −56 −119 −68 −119 −67 −181 −72 −140
42.1 272.0 5.33 4.61 5.77 7.55 5.23 5.45 9.21 3.93
– – 448 373 239 158 677 743 940 185
– – 50 63 30 20 736 323 1093 107
response of S and T materials. Molybdenum is less active than aluminum and titanium, but the enthalpy of formation of Mo2N is similar to that of iron nitrides [21], so the presence of molybdenum nitrides would be expected, and accordingly in a higher concentration for S samples than for T samples. GIXRD did not reveal the presence of Mo2N in the compound layer for S nor T samples. Only a slight increase in the hardness of the nitrided zone in S specimens was detected that may correspond to the contribution of ε phases and other nitrides. This point will be subject of further studies. It is interesting to point out that X-ray diffraction patterns show significantly broader peaks for the nitrided phases in S specimens with respect to those observed for the T specimens, especially in the case of γ′-Fe4N phases. It may be explained owing to the formation of inhomogeneous micro-strained structures resulting from the incorporation of small nitrides, and to the contribution of γ′-FeNi3N phases, whose diffraction peaks can be found next to γ′-Fe4N, displaced 0.2° to higher angles. Actually, small shoulders can be observed on the right side of the γ′-Fe4N peaks at 63° for S specimens treated at 450 °C and 475 °C, which can be attributed to γ′-FeNi3N. This is consistent with the expected higher content of Ni in the matrix taking into account the lower amount of Ni-rich precipitates due to the lower treatment temperatures and the fact that S specimens were not previously precipitation hardened. It is also remarkable that no signs of overaging were evidenced by means of X-ray diffraction, as no peaks of reverse austenite were found. It is also correlated with the HRc results shown in Table 4, where a loss of base hardness was not appreciated. 3.2. Corrosion tests The corrosion resistance of the nitrided specimens was studied and compared to the untreated material through potentiodynamic tests. Fig. 4 shows the polarization curves of plasma nitrided specimens S (a) and T (b). An important enhancement of the corrosion resistance of the maraging surfaces after plasma nitriding can be observed [6]. Actually, a change in the corrosion behavior can be observed. Fig. 4 displays a uniform corrosion behavior for un-nitrided maraging steel, typically observed for mild steels. After ion nitriding treatments, a pseudo-passive layer is formed in both S and T specimens, offering a better protection to the material regardless of the treatment temperature. Fig. 5a–b shows the corrosion tested surfaces of the un-treated maraging steel T and the nitrided one at 500 °C after the tests. Uniform and massive corrosion can be observed for the reference materials S and T, whereas the appearance of pitting corrosion is common to all the nitrided specimens, both S and T (c.f. Fig. 5b). Li and Bell [22] reported similar corrosion results in their study of plasma nitriding of AISI 410 martensitic stainless steel. The values of corrosion potential (Ecorr),
Fig. 6. SEM cross-sections of the reference materials and the nitrided specimens. a) Un-treated maraging steel S; b) un-treated maraging steel T; c) 450-S; d) 450-T; e) 475-S; f) 475-T; g) 500-S; h) 500-T; i) 515-S; j) 515-T.
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
a)
b)
c)
d)
e)
f)
g)
h)
i)
j)
759
760
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
a)
Table 4 Roughness, hardness HRc and HK0.05, and wear rate.
Ref — T Ref — S 450 °C — T 450 °C — S 475 °C — T 475 °C — S 500 °C — T 500 °C — S 515 °C — T 515 °C — S
Roughness
Hardness
Ra nm
HRc
7 8 179 202 208 202 221 284 200 266
± ± ± ± ± ± ± ± ± ±
2 2 2 4 4 5 2 3 4 4
52.3 30.2 53.0 47.9 53.3 51.4 53.1 51.3 53.3 52.9
± ± ± ± ± ± ± ± ± ±
0.3 0.2 0.2 0.3 0.4 0.3 0.3 0.2 0.4 0.2
K (m3 × N−1 × m−1)
– – 745 752 639 649 730 781 674 759
3.81 1.62 1.98 1.61 1.09 1.12 1.43 2.11 9.25 1.02
48 36 37 36 27 44 18 21
± ± ± ± ± ± ± ± ± ±
3.11 0.18 1.04 0.98 0.11 0.07 0.29 1.30 0.25 0.06
o
450 C - S o 475 C - S o 500 C - S o 515 C - S
850
Surface HK0.05
± ± ± ± ± ± ± ±
S
900
Wear rate
× × × × × × × × × ×
10−13 10−12 10−14 10−14 10−14 10−14 10−14 10−14 10−15 10−14
800 750 700
HK0.05
Specimens
650 600 550 500 450 400 0
10
20
30
40
50
60
70
80
90
Depth μm
b)
T
900
o
450 C - T o 475 C - T o 500 C - T o 515 C - T
850 800 750 700
HK0.05
critical current density (icorr), pitting potential (Epit) and pitting current (ipit) are listed in Table 3. An important displacement to higher corrosion potential values after plasma nitriding is also observed (from − 400 V to around − 60 V for T specimens and between − 120 V and −180 V for S specimens). The T and S samples treated at 475 °C show worse behavior than the rest, presenting significant lower Epit values. This can be understood observing the compound layer of the specimens nitrided at 475 °C. For both S and T samples at this temperature a higher proportion of mixtures of different phases can be observed, penalizing the corrosion resistance. It is well known that a compound layer with a mono-phasic structure of γ′-Fe4N present a better behavior than that containing ε-Fe2N1 − x phases or biphasic structures [23]. In general, for T specimens, corrosion resistance increases as the temperature increases, as the result of the formation of a greater content of γ′-Fe4N phase over ε-Fe2N1 − x.
650 600 550
3.3. Microstructure, hardness, roughness and wear tests SEM micrographs of the un-treated and nitrided specimens in cross-section are depicted in Fig. 6a–j. The typical martensitic structure corresponding to maraging steels can be observed for S and T specimens (c.f. Fig. 6a–b) [2,8]. No differences can be detected between S and T specimens owing to the small size of the precipitates in the aged samples, well below the detection limits of FE-SEM instruments. Nevertheless, the presence of these intermetallic compounds can be taken for granted considering the base hardness of both types of samples, later explained (Table 4). In the micrographies of the cross-sections shown in Fig. 6c–j, a ductile fracture behavior of the maraging steel substrate can be observed, whereas the nitrided zones, whose thickness agrees with the measured one using GDOES, present a fragile fracture behavior owing to the iron nitrides (γ′-Fe4N and ε-Fe2N1 − x). A ductile behavior of the substrates can be noticed where the diffusion profile of nitrogen ends, with the same microstructure than that of the un-treated specimens. Roughness measurements (Table 4) show that the Ra values increase from below 10 nm to more than 200 nm due to the effect of the intensive ion bombardment. Table 4 also shows the HRc values for all studied samples. Used temperatures and nitriding times are compatible with aging treatments for maraging steel hardening, as it was suggested by GIXRD measurements due to the absence of austenite. Considerable increases of HRc values are observed for S specimens after treatment, which confirms that plasma nitriding can also be used for simultaneous age hardening of maraging steel, as it has been previously reported for other authors studying simultaneous thermochemical treatments and bulk hardening of martensitic steels [24]. A raise of hardness Rockwell C up to 48–53 HRc was measured, according to the treatment temperature. However, at 450 °C bulk hardness is smaller than that obtained at higher temperatures, indicating that the aging treatment was not fully completed. No
500 450 400 0
10
20
30
40
50
60
70
80
90
Depth μm Fig. 7. Knoop hardness at 50 g load depth profiles. a) Solution annealed specimens (S); b) prior aged specimens (T).
evidence of overaging was observed for any of the pre-aged specimens in light of the Rockwell C values (Table 4), in agreement with the XRD and also considering the microstructure images (Fig. 6). Fig. 7 shows the Knoop micro-hardness (50 g load) profiles measured on the mirror polished cross-section of S and T specimens after nitriding. Greater surface hardness followed by a sharper fall of the values was observed for the S specimens in comparison with the T ones. As it is shown in Table 4, in all the cases surface hardness is greater for the S specimens, especially at higher temperatures. This is consistent with the presence of ε-Fe2N phases in the compound layer, which are harder than those of γ′-Fe4N. For T specimens, the presence of ε-Fe2N phases is bigger at lower temperatures, so higher surface hardness values at those temperatures were actually expected. Maximum surface hardness does not vary significantly for the S specimens, with the next exception. Low hardness values were obtained in the surface (up to the first 5–10 μm) for the samples treated at 475 °C, in which a lower nitrogen amount in the surface was detected by GDOES analyses. Both S and T specimens presented considerable low surface hardness values. At this temperature, GIXRD showed peaks belonging to the signal of the steel for T specimens and weak peaks of ε-phases, so a low surface hardness is consistent with the reported data. However, for S specimens GIXRD revealed peaks corresponding to ε-Fe3N phases in good agreement with
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762 1800 1500
Wear Rate
1200 900 300 35
12
3
-1
-1
K * 10 (m ·N ·m )
600
30 25 20 15 10 5 0
Ref T Ref S 450 ºC 450 ºC 475 ºC 475 ºC 500 ºC 500 ºC 515 ºC 515 ºC
Fig. 8. Comparison of the wear rate of the un-treated maraging steel and the specimens treated at various temperatures.
761
that they were not detected through the analysis techniques being used in this work it will be necessary further studies to sustain this point. Wear resistance was also studied with the help of a pin-on-disc tribometer at room temperature. Fig. 8 shows the wear coefficients of the nitrided specimens. A great improvement of the wear resistance was obtained after plasma nitriding in all the cases (one order of magnitude for T specimens and two orders for S specimens). Although small differences were found between the specimens, a good correlation between surface Knoop hardness at 50 g and wear rate can be observed. In general, specimens which present a greater surface hardness exhibit greater wear rate, therefore suggesting third-body wear effects during the tests. This fact can be explained owing to the fragile behavior of the ε phases present in the specimens with higher hardness. The surface topography of the tracks of the un-treated material and the nitrided specimens (S and T) at 500 °C is displayed in Fig. 9, where abrasive wear can be observed. Deeper tracks and greater volume losses are observed for the un-nitrided samples. 4. Conclusions
that stoichiometry at 475 °C. To explain this fact, the nature of the nitrides present in the compound layer should be taken into account, considering that the hardness of ε-Fe2N phase (ε-Fe2N1 − x with x near zero) is higher than that of ε-Fe3N (ε-Fe2N1 − x with x near 1/3). The formation of these phases is influenced by the treatment temperature [18–20] and the nitriding activity and, consequently, the concentration of nitrogen that incorporates to the material. On the other hand, the fact that other nitrides such as Mo2N might be present at a higher concentration in S specimens and the fact that they could also contribute to the higher hardness have to be considered. However, taking into account
Arc-activated plasma nitriding is a feasible technology for the nitriding of maraging steel in both solution annealed and aged conditions. On top of this, simultaneous aging and nitriding processes are possible on solution annealed specimens of maraging steel grade 300 at temperatures from 475 °C to 515 °C, whereas at 450 °C the aging process did not lead to the specified bulk hardness that corresponds to complete aging process. Arc-activated plasma nitriding significantly enhanced corrosion properties of maraging steel and also led to important improvements in wear resistance. After the plasma nitriding treatments
65 µm
120 µm
a)
b)
8 µm
c)
9 µm
d)
Fig. 9. 3D-images and 2D-profiles of the wear tracks measured by optical profilometry. a) Un-treated maraging steel (S); b) un-treated maraging steel (T); c) 500-S; d) 500-T.
762
J. Fernández de Ara et al. / Surface & Coatings Technology 258 (2014) 754–762
in both conditions of the maraging steel, the formation of γ′-Fe4N and εFe2N1 − x phases was observed in the compound layer. In general, εphases were preferably found at lower treatment temperatures, while the amount of γ′-Fe4N phase increased with the temperature. Furthermore, ε-Fe2N1 − x phases were also detected at the highest nitriding temperatures in the specimens that were not previously aged. This fact explains the greater surface hardness found for solution annealed nitrided maraging steel.
[7] [8] [9] [10] [11] [12] [13] [14] [15]
Acknowledgments [16]
The authors acknowledge the financial support from Navarra's Government (IIQ1404) through ‘Supermaraging’ project and the Ministry of Economy and Competitiveness of Spain through the Consolider-Ingenio project ‘Funcoat’ (2010-CSD2008-00023). References [1] S. Floreen, Metall. Rev. 13 (1968) 115–128. [2] W. Sha, Z. Guo, Maraging steels, Modeling of Microstructure, Properties and Applications, first ed., Woodhead Publishing Limited, New Cambridge, 2009. (Cambridge). [3] B. Gupta, V. Gopalakrishna, J.S. Yadav, A. Kumar, B. Saha, Aerospace Materials, S. Chang and Company, New Delhi, 1996. 695. [4] J. Rezek, I.E. Klein, J. Yahalom, Appl. Surf. Sci. 108 (1997) 159–165. [5] Kishora Shetty, S. Kumar, P. Raghothama Rao, J. Phys. Conf. Ser. 100 (2008) 062013. [6] Kishora Shetty, S. Kumar, P. Raghothama Rao, Surf. Coat. Technol. 203 (2009) 1530–1536.
[17] [18] [19] [20] [21] [22] [23]
[24]
Yuji Ohue, Koji Matsumoto, Wear 263 (2007) 782–789. K. Hussain, A. Tauqir, A. ul Haq, A.Q. Khan, Int. J. Fatigue 21 (1999) 163–168. T. Spalvins, ‘Ion Nitriding’ Conference Proceedings, 1, ASM International, 1986. A. Leyland, K.S. Fancey, A.S. James, A. Matthews, Surf. Coat. Technol. 41 (1990) 295. W. Möller, S. Parascandola, T. Telbizova, R. Günzel, E. Richter, Surf. Coat. Technol. 136 (2001) 73–79. M.B. Karamiş, E. Gerçekcioĝlu, Wear 243 (2000) 76–84. H. Michel, T. Czerwiec, M. Gantois, D. Ablitzer, A. Ricard, Surf. Coat. Technol. 72 (1995) 103–111. G.G. Fuentes, R. Rodríguez, J.A. García. European Patent EP2369028 (A1), assigned to Asociación de la Industria Navarra (AIN). J.A. García, R.J. Rodríguez, R. Martínez, C. Fernández, A. Fernández, R. Payling, Appl. Surf. Sci. 235 (2004) 97. H. Selg, Nitriding of Fe–Mo alloys and maraging steel: structure, morphology and kinetics of nitride precipitation, PhD. thesis Max-Planck-Institut für Intelligente Systeme (ehemals Max-Planck-Institut für Metallforschung), Stuttgart, 2012. (University of Stuttgart). K.H. Jack, Proc. R Soc. A 195 (1948) 34. M.K. Lei, Z.L. Zhang, Surf. Coat. Technol. 91 (1997) 25–31. A.M. Vredenberg, C.M. Perez-Martin, J.S. Custer, D.O. Boerma, L. de Wit, F.W. Saris, N. M. van der Pers, Th.H. de Keijser, E.J. Mittemeijer, J. Mater. Res. 7 (1992) 2689. R. Hutchings, G.A. Collins, J. Tendys, Surf. Coat. Technol. 51 (1992) 489. D.A. Porter, K.E. Easterling, Phase Transformations in Metals and Alloys, CRC Press, Taylor & Francis Group, Finland, 2004. C.X. Li, T. Bell, Corros. Sci. 48 (2006) 2036–2049. J.C. Díaz-Guillén, G. Vargas-Gutiérrez, E.E. Granda-Gutiérrez, J.S. Zamarripa-Piña, S.I. Pérez-Aguilar, J. Candelas-Ramírez, L. Álvarez-Contreras, S.I. Pérez-Aguilar, J. Candelas-Ramírez, L. Álvarez-Contreras, J. Mater. Sci. Technol. 29 (3) (2013) 287–290. R.B. Frandsen, T. Christiansen, M.A.J. Somers, Surf. Coat. Technol. 200 (2006) 5160–5169.