Fabrication and characterization of Fe-based amorphous coatings prepared by high-velocity arc spraying

Materials and Design 78 (2015) 118–124

Contents lists available at ScienceDirect

Materials and Design journal homepage: www.elsevier.com/locate/matdes

Fabrication and characterization of Fe-based amorphous coatings prepared by high-velocity arc spraying Wenmin Guo, Jianfeng Zhang ⇑, Yuping Wu, Sheng Hong, Yujiao Qin College of Mechanics and Materials, Hohai University, Nanjing 210098, China

a r t i c l e

i n f o

Article history: Received 18 November 2014 Revised 14 April 2015 Accepted 18 April 2015 Available online 18 April 2015 Keywords: Thermal spray Coating Microstructure Amorphous Crystallization

a b s t r a c t Fe-based coatings with a high amorphous content were firstly developed by the traditional twin wires arc spray technology. In consideration of empirical rules, including the multi-component system, an optimal concentration of small atoms, negative heat of mixing and an appropriate atom size mismatch among the main components, the cored wires were designed to contain eight elements, which have an optimized atomic volume strain criterion kn, in range of 0.14–0.21, to render the coatings a high glass forming ability. Then the coatings were prepared using the above-designed cored wires through a rapid arc spray melting and solidification process. Crystalline phases could not be identified from the XRD patterns within the XRD resolution limits, suggesting that the as-sprayed coatings were approximately comprised of fully amorphous phases. With a dense structure and a low porosity of only 2%, the amorphous Fe-based coatings exhibited an attractive combination of high hardness (900–1100 HV0.3) and superior bonding strength (44.9–54.8 MPa). The coating at kn = 0.21 had the lowest Gibbs free energy difference DG, exhibited the largest super-cooled liquid region DTx, Lu’s criterion factor c value and the heat of crystallization (DH) values, which indicating the highest GFA. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction With a combination of high mechanical strength, superior corrosion resistance and excellent soft magnetism, the development of bulk metallic glasses (BMGs) has attracted much attention in the recent decades to meet the ever-growing requirements of the industry [1,2]. However, the wide applications of BMGs are still limited due to their inherent drawbacks, such as apparent roomtemperature brittleness [2], high fabrication cost and the technical difficulties in fabrication of large-size products. Thermal spray has been recognized as one of the most promising approaches to reduce the preparation cost, improve the toughness, and thus widen the industrial application field of the Febased amorphous materials [3]. Recently, significant efforts have been conducted to develop high performance Fe-based amorphous coatings by high velocity oxy-fuel (HVOF) spray, plasma spray (PS), twin-wire arc spray (TWAS), etc [4,5]. These thermal spray technologies show promising applications in ships in marine environment, containers for the spent nuclear fuel, oil and gas industries and power stations [3].

⇑ Corresponding author at: College of Mechanics and Materials, Hohai University, 1-Xikang Road, Nanjing 210098, China. E-mail address: [email protected] (J. Zhang). http://dx.doi.org/10.1016/j.matdes.2015.04.027 0261-3069/Ó 2015 Elsevier Ltd. All rights reserved.

However, during thermal spray processes, the formation and retention of a non-equilibrium phase, i.e. an amorphous phase, is still a challenging task. It requires not only a high glass forming ability of the powder feed stock, but also a sufficiently rapid cooling rate of the melted droplets in the thermal spray process. The cooling rate of the molten droplets in the PS was reported to be 107–108 K/s [3]. Even that in HVOF process (104 K/s [3]) or arc spray process (105 K/s [3]) was orders of magnitude lower, it is still high enough to deposit many alloy compositions above the respective critical cooling rate, thereby maintaining a vitreous state. In recent years, HVOF and PS technologies have been considered as the two most attractive approaches for the deposition of amorphous coatings [4]. Otsubo et al. [6] and Kishitake et al. [7] reported that coatings with 100% amorphous phase content were successfully fabricated by HVOF and low-pressure plasma spraying (LPPS). However, in order to get a high amorphous phase content and chemical composition uniformity in amorphous coatings, powder feedstock for the HVOF or PS processes should be generally prepared by gas or water atomization of BMGs, which is quite tedious to increase the cost greatly. In addition, the amorphous phase content and composition homogeneity of the as-obtained powder feedstock cannot be guaranteed since they are extremely sensitive to various atomization approaches and operating parameters.

119

W. Guo et al. / Materials and Design 78 (2015) 118–124

In the present study, we firstly reported that Fe-based coatings with a high amorphous content were fabricated by a more simple and efficient coating process, i.e., the traditional twin wires arc spray (AS) technology. Compared to HVOF and PS processes, AS exhibits many unique advantages, such as a high spray rate, a short treating time, a low production cost, and on-site operation ability [8]. By this route, the wide applications of Fe-based amorphous coatings would come into reality to protect industrial components from deterioration in various aggressive environments. Meanwhile the restrictions in the practical application caused by the size limitation of bulk metallic glass would be released. 2. Composition design concept for the cored wires In order to obtain the high amorphous content coatings, the feedstock used for the thermal spray processes should have a high glass forming ability (GFA). As proposed by Inoue [9] and Johnson [10], alloys with high GFA generally possessed the following compositional features: (I) a multi-component system, (II) significant atomic size ratios above 12%, (III) negative heat of mixing and (IV) deep eutectic rule based on the Trg criterion. Therefore, the following four criteria were adopted for the composition design of the cored wires in the present study. 2.1. Multi-component Based on the ‘‘Confused principle’’ [11], the GFA of glass formers would be enhanced by a multi-component alloy system, i.e., the more elements involved, the lower the chance to form a viable crystal structure, and the greater the chance of glass formation. In the present study, eight elements (Fe, Cr, Si, B, Nb, Mo, Ni and Al) were selected to form the Fe-based alloy system. 2.2. Topological models For the bulk metal-metalloid materials, the best composition to form a glass is to contain about 75–85 at.% of the metal component and 25–15 at.% of the metalloid component according to the topological models [12]. In consideration of the burning loss and inevitable oxidation of the metalloid during the arc spray process, the composition in the present study was designed to contain 20 at.%, 25 at.%, and 30 at.% of the metalloid elements (B, C, Si.), respectively.

where RA and RB are the radii of the solvent and solute atoms, respectively. CB (In atomic percent) is the solute concentration of element B. VA is the atomic volume of A. DVAB is the difference in atomic volumes between A and B. Thus, in order to obtain a high amorphous content, three feedstock materials were designed with special kn values of 0.14, 0.18, 0.21, which were labeled as F20, F25, F30 coating, respectively.

3. Experiment procedures 3.1. Cored wires Fig. 1 shows the process for the preparation of cored wires used for AS. Initially, Chromium (99%), Boron carbide (B 85%, C 8%), Ferrosilicon (72%), Niobite (Nb 55%), ferromolybdenum (Mo 55%) and a trace amount of aluminum (Al 97%) were mixed by ball milling as the core (Fig. 1(a)). The ball milling was carried out at 180 rpm in ethanol using a stainless steel vial and balls with a ball/powder weight ratio of 5:1. The mixed powders were dried in a rotary evaporator at 60 °C for 24 h at a negative air pressure of 5 Pa, and then surrounded by an outer skin of 304L austenitic stainless steel to form cored wires, as shown in Fig. 1(b). The cross-sectional image of the as-prepared cored wire was shown in Fig. 1(c). The composition of the cored wires, as shown in Table 1, was designed according to the analysis in part 2. 3.2. Development of coatings AISI 1020 steel was used as the substrate in the present investigation. The nominal chemical composition of the AISI 1020 steel is 0.23 wt.% C, 0.24 wt.% Si, 0.016 wt.% S, 0.012 wt.% P, 0.4 wt.% Mn and balance Fe. Prior to the arc spray process, the substrate was cut into a dimension of 30 mm  10 mm  8 mm, subsequently polished by SiC papers down to 150-grit, ultrasonically degreased in acetone, and then grit blasted with alumina powders. The prepared

2.3. Negative heat of mixing A large negative heat of mixing among the constituent elements is known to strengthen the interaction among the constituent elements and promote the chemical short range ordering in the liquid [13]. In the represent study, the mixing enthalpy calculated by Miedema’s model for the Fe–B, Cr–B, Nb–B, Fe–Si, Nb–Si, Cr–Si, Fe–C, Nb–C, Cr–C, Nb–Cr and Fe–Nb pair is 26, 31, 54, 35, 56, 37, 50, 102, 61, 7, and 16 kJ/mol, respectively [3]. 2.4. Atomic volume strain criterion For a multi-component alloy system, effects on the GFA of atomic volume strain was proposed to be evaluated by an empirical criterion kn [2], which in agreement with the experimental data of various alloy systems. The kn for the best glass-forming alloys of a multi-component system is calculated to be about 0.18 by Eq. (1).

kn ¼

n1 n1   X X   3 jDV AB =V A j  C B ¼ ðRB =RA Þ  1  C B B¼1

B¼1

ð1Þ

Fig. 1. (a) Morphology of the mixed raw powders, (b) preparation process of the core-wires. (c) Cross-sectional morphology of the as-prepared cored wires.

Table 1 The nominal composition of the prepared cored wires.

F20 F25 F30

Cr

B

C

Si

Nb

Mo

Ni

Fe

<24.2 <21.5 <18.8

<16.5 <21.0 <25.5

<1.5 <2.0 <2.5

<2.0 <2.0 <2.0

<2.5 <2.5 <2.5

<1.0 <1.0 <1.0

<5.0 <5.0 <5.0

Bal. Bal. Bal.

120

W. Guo et al. / Materials and Design 78 (2015) 118–124

Table 2 The optimized arc spraying parameters.

F20 F25 F30

Spraying voltage, V

Spraying current, A

Compressed air pressure, MPa

The standoff distance, mm

40 40 36

180 180 160

0.65 0.65 0.65

200 200 200

Fe-based cored wires of 2 mm diameter were used as feedstock. The coating was deposited on the as-prepared substrate with a High-Jet DZ-Arc-500E high velocity arc spray system under optimized spray conditions as shown in Table 2. 3.3. Characterization The microstructure of the coatings was characterized by a Philips XL30 scanning electron microscope (SEM) equipped with energy dispersive spectrometer (EDS) and a FEI Tecnai F20s transmission electron microscopy (TEM). Porosity of the coatings was evaluated using image analyzer of DT2000 software. Phases were identified by the X-ray diffraction that was carried out in a D8 Advance diffractometer (Bruker AXS, Germany) with a Cu Ka target and operated at 40 kV and 30 mA. Scans were run with a step size of 0.01° from 20° to 90° (2h). Differential scanning calorimeter (DSC) measurement of all the as-sprayed coatings was conducted from room temperature to 1400 °C in an argon gas atmosphere using a STA 449 DSC instrument (NETZSCH Instruments Co., Ltd., Germany) at a heating rate of 10–40 K/min. Vickers microhardness was measured on the cross section of the as-sprayed coatings using a HXD-1000TC microhardness tester at a load of 2.94 N for 15 s. Adhesion testing was done according to ASTM C633-13 standard using adhesive E7 epoxy (Provided by Shanghai Research Institute of Synthetic Resins, China.) with a cross-head speed of 0.05 in./min.

interface between the lamellar structures or within the splats. The porosity of the as-sprayed F20, F25, and F30 coatings was calculated to be 1.9%, 2.1%, and 2.2%, respectively. It is apparent from Fig. 2(a)–(c) that the lamellae structure of these coatings is more and more distinguishable due to the formation of pores and oxides at the boundaries between splats. In general, the pores located between flattened droplets were proposed to be caused by the loose packed layer structure or gas porosity phenomenon, while other pores within the flattened particles by volume shrinkage [14]. 4.2. XRD and TEM analysis of as-sprayed coatings Various Fe-based amorphous coatings have been fabricated by AS in literature, such as SHS 8000 [15], FeBSiNbCr [16], FeCrBSiMnNbY [16], and Fe60B22Si3Cr10(Nb,W)5 [17]. However, XRD analysis results indicated that these coatings were partially amorphous combined with a few crystalline phases. Fig. 3 shows the XRD patterns of the as-sprayed coatings in the present research. The presence of the diffraction halo around 2h  45° indicated all the as-sprayed coatings were mainly composed of an amorphous structure. Crystalline phases could hardly be identified within the XRD resolution limits. Thus, the self-designed Fe-based alloy system in the present study was proved to be an excellent glass former with a wide range of selectable chemical composition. In order to give more detailed information, the microstructure of the as-sprayed F30 coating was studied using TEM, as shown in Fig. 4. The diffused halo rings in the selected area diffraction patterns confirmed that the as-sprayed F30 coating was mainly

4. Results 4.1. Microstructure Fig. 2(a)–(c) illustrates the typical cross-sectional morphologies of the as-sprayed Fe-based coatings. All the as-sprayed coatings were consisted of numerous flattened lamellae parallel to the substrate. A trace amount of dark lines, composed of iron and chromium oxides by EDS analysis (Fig. 2(d)), was observed on boundaries of splats. Micro-pores were also observed along the

Fig. 3. XRD patterns of the as-sprayed (a) F20, (b) F25, and (c) F30 Fe-based amorphous coatings.

Fig. 2. Cross-sectional morphologies of the arc sprayed (a) F20, (b) F25, (c) F30 Fe-based amorphous coatings, with (d) EDS analysis results for spots A, B, C.

W. Guo et al. / Materials and Design 78 (2015) 118–124

121

rate, the interval between the first and the second exothermic peaks narrowed gradually, and the exothermic peak shifted to a higher temperature due to the thermodynamic effect [18]. The peak temperature (Tp) is high and the enthalpy change of the exothermic peaks is large when the heating rate is increased, indicating that the glass transition and crystallization exhibited a dependence on the heating rate during continuous heating [19]. The first peak in the DSC traces was ascribed to primary crystallization and the second peak arose from the crystallization of the residual amorphous phase [20]. Activation energy (Eac), related to the magnitude of energy barrier which the atoms should overcome for rearrangement from disorder to order states, is an important parameter to represent the thermal stability of an amorphous phase against crystallization. The Eac for crystallization of the F20 coating was evaluated by Kissinger equation (Eq. (2)) [21] and Ozawa equation (Eq. (3)) [22] as follows.

Fig. 4. Bright-field TEM image with the embedded selected area electron diffraction (SAED) patterns of the as-sprayed F30 coating.

lnðr=T 2p Þ ¼ Eac =ðR  T p Þ þ C

ð2Þ

lnðrÞ ¼ 1:0516Eac =ðR  TpÞ þ C

ð3Þ

where Tp is the peak temperature at heating rate ‘‘r’’, R is the real gas constant 8.3145 J/mol K, and C is a constant. Kissenger and Ozawa plots of the as-sprayed F20 coating were drawn in Fig. 6(b) in straight lines with a slope of Eac/R = B, where B is a constant. Putting the values of R and B (determined from the Ozawa plots), the activation energies for the first stage and second stage crystallization were calculated to be 347.9 and 533.1 kJ/mol. According to the Kissinger plots, the activation energies for the first stage and second stage crystallinzation were calculated to be 350.2 and 544.3 kJ/mol, much higher than 375.0 and 333.0 kJ/mol for that of HVOF sprayed SHS7170 coating [23] and 201.9 kJ/mol for the Fe61Co5Zr8Y2Cr2Mo7B15 BMG [24]. The high activation energy suggested a high thermal stability of the amorphous phase in F20 coating. 4.4. Mechanical properties Fig. 5. DSC traces of the as-sprayed (a) F20, (b) F25 and (c) F30 amorphous coatings at a heating rate of 10 K/min.

composed of amorphous phase. Crystallines could hardly be observed in Fig. 4. 4.3. Thermal analysis In order to evaluate the GFA of these arc-sprayed Fe-based coatings, the DSC traces of these as-sprayed coatings at a heating rate of 10 K/min were examined, as plotted in Fig. 5. The initial crystallization of the F20 coating started at 903.3 K and ended at 948.0 K, whereas that of the F25 coating started at 891.8 and ended at 923.0 K. The first exothermic peak of F25 coating was lower than that of F20, and the temperature range was narrowed. Moreover, it even disappeared in the F30 coating, namely the first exothermic peak was superposed to the second one. Similarly, the second crystallization temperature of these coatings also shifted toward a lower temperature with increasing kn. In conclusion, the crystallization behavior of the coatings changed from two stages to one. It is interesting to note that this unique kn-related crystallization behavior was not reviewed in literature, which calls for further attentive investigation. Fig. 6(a) shows the DSC traces of the as-sprayed F20 amorphous coating obtained at a heating rate from 10 to 40 K/s, in which two exothermic peaks were all observed. With increasing the heating

Table 3 compares the microhardness and bonding strength of the thermal sprayed coatings in the present study with that reported in literature. The microhardness of the as-sprayed F20, F25, and F30 coatings in the present study is about 1000 HV0.3, much higher than that of metallic coatings (Fe–26Cr–6Al [25]) and amorphous coatings (Fe–10Cr–10Mo–8P–2C [6], Fe48Cr15Mo14C15B6Y2 [26]), even close to that of ceramic reinforced coatings (WC–10Co–4Cr [27]). The bonding strength of the assprayed coatings (F20/F25/F30) with a thickness of 500 lm was located in a range from 44.9 to 54.8 MPa, about 10% higher than that of HVOF sprayed Fe48Mo14Cr15Y2C15B6 coating (41 MPa) [28]. It should also be pointed out here that the glue mostly broke before the coatings did, indicating that the intrinsic bonding strength of the coating with the substrate should be higher than the reported values. The fracture morphologies of the as-sprayed coatings after the bonding strength tests were illustrated in Fig. 7. Only a very small proportion of the F20 and F25 coatings has been tore apart from the out layer of the as-sprayed coatings, as shown in Fig. 7(a)– (d), and the cracks were distributed throughout the lamellar structure. The surface morphologies of the damaged coatings were smooth, indicating that the bonding strength of the as-sprayed coatings has been primarily deteriorated from the interface between the lamellar structures. As shown in Fig. 7(b), a whole block of the lamellar structure has been torn up. This flexuous lamellar structure suggested that F20 coating could possess a good toughness. As shown in Fig. 7(e), it is clear that a relatively larger

122

W. Guo et al. / Materials and Design 78 (2015) 118–124

Fig. 6. (a) DSC traces of the as-sprayed F20 coating at a heating rate of 10–40 K/min and (b) the related Kissenger/Ozawa plots.

Table 3 Mechanical properties of the as-sprayed Fe-based amorphous coatings with other thermal spray coatings reported in the literature. Coatings

Spray methods

Main phase

Microhardness

Bonding strength

F20 F25 F30 Fe–10Cr–10Mo–8P–2C [6] Fe48Mo14Cr15Y2C15B6 [28] Fe48Cr15Mo14C15B6Y2 [26] Fe–26Cr–6Al [25] WC–10Co–4Cr [27]

AS AS AS HVOF HVOF HVOF AS HVOF

Amorphous Amorphous Amorphous Amorphous Amorphous Amorphous, a-Fe, Fe2C, Cr2O3 FeCr, AlFe, Al2O3 /

985 ± 50 HV0.3 (9.7 ± 0.5 GPa) 1041 ± 63 HV0.3 (10.2 ± 0.6 GPa) 971 ± 124 HV0.3 (9.5 ± 1.2 GPa) 5.5 GPa / 7.0 GPa 3.9–5.5 GPa 10.6 ± 0.5 GPa

P54.8 MPa P51.8 MPa P44.9 MPa / 25.0–41.0 MPa / / /

Fig. 7. The fracture morphology of the arc sprayed Fe-based amorphous coatings (a–b) for F20 coating, (c–d) for F25 coating, (e–f) for F30 coating.

123

W. Guo et al. / Materials and Design 78 (2015) 118–124

Table 4 The glass transition temperature Tg, onset crystallization temperature Tx, enthalpy of crystallization DH, melting point Tm, enthalpy of fusion DHf, liquidus temperature Tl, supercooled liquid region DTx, reduced glass transition temperature Trg, and c value for the F20, F25, F30 as-sprayed coatings determined via DSC. (Numbers (1 and 2) in the subscript of Tx and DH represent the first and second crystallinzation stage in the DSC traces as shown in Fig. 5).

F20 F25 F30

Tg, K

Tx1, K

DH1, J/g

Tx2, K

DH2, J/g

Tm, K

DHf, J/g

Tl, K

DTx, K

Trg

c

863.4 854.7 843.9

909.3 891.8 –

6.4 – –

953.0 933.6 915.4

42.5 41.8 85.4

1475.0 1459.4 1440.2

648.7 518.2 320.7

1508.0 1496.6 1482.6

45.9 37.1 71.5

0.573 0.571 0.569

0.383 0.379 0.393

part of the F30 coating has been damaged, which was in consistence with the low bonding strength as shown in Table 3. The initial pores have been stuck together by cracks and the crosssectional fracture surface of the lamellar structure demonstrated a typical brittle fracture features (Fig. 7(f)).

5. Discussion From the perspective of devitrification, a large super-cooled liquid region (DTx, DTx = Tx–Tg [9]) indicates that the amorphous phase could exist in a wide temperature range without crystallization in the heating process, i.e., the amorphous phase has a high thermal stability and then represents a high GFA [29]. From the perspective of amorphization, when a liquid alloy is cooled from the molten state down to a temperature below the glass transition temperature (Tg), the viscosity of the melt would increase as a higher reduced glass transition temperature (Trg, Trg = Tg/Tl [12]) and then a glass would likely form. Typically, a minimum value of Trg  0.4 was necessary for an alloy to become a glass, and the homogeneous nucleation of the crystalline phase in the supercooled liquid is completely suppressed when Trg P 2/3 [12]. In the light of both of the considerations above, Lu and Liu [30] proposed a new c criterion (c = Tx/(Tg + Tl)) that seems to be applicable in a majority of the alloy systems. Therefore, these three criteria DTx, Trg, and c, which were most extensively used for the GFA assessment of BMGs, were selected to evaluate the GFA of the assprayed coatings in the present study. Table 4 presents the parameters such as the glass transition temperature Tg, the onset crystallization temperature Tx, the melting point Tm, the liquidus temperature Tl, and the heat of crystallization DH of the as-sprayed F20, F25 and F30 coatings determined from the DSC plots (Fig. 5). Using these values, DTx, Trg, and c criteria were also calculated and summarized here. Fig. 8 demonstrates the effects of kn on variation tendency of DTx, Trg, and c criteria. As observed, F30 coating (with a kn value equal to 0.21) presents the highest DTx and c values than those of F20 and F25 coatings, indicating that the former has the highest GFA. F25 coating, which has the most optimized kn value equaled

Fig. 8. Effects of atom volume strain criterion (kn) on the variation tendency of the super-cooled liquid region (DTx), reduced glass transition temperature (Trg), and Lu’s criterion factor (c).

to 0.18, unexpectedly shows the lowest DTx and c values. Therefore, the GFA prediction results according to the DTx and c criterion seem to contradict with the kn criterion as discussed in Section 2.4. The Trg values of these as-sprayed coatings was in a range from 0.569 to 0.573 (Table 4), much higher than the minimum value of Trg (0.4) necessary for the amorphization of alloys [12]. Therefore, all the coatings in the present study should possess a high glass forming ability. However, the Trg values of these coatings showed a continuously linear decrease with the increased kn values (Fig. 8), indicating that the F20 coating showed a highest GFA with a largest Trg values equaled to 0.573. This GFA prediction result also showed a tendency opposite to that by the DTx, c and kn criteria. The above analysis result indicates that, those criteria widely used for the bulk metallic glasses are inapplicable for evaluating the GFA of the as-sprayed coatings. This could be presumably attributed to the differences between the approaches for amorphous coatings and BMGs. Firstly, due to the burning loss and oxidation of elements during the thermal spray process, the chemical composition of the amorphous coatings is much more inhomogeneous than that of BMGs [3]. Secondly, considering that thermal spray coatings was fabricated by a successive deposition of molten droplets, thus the pre-deposited amorphous structure could be partially devitrified by the localized reheating caused by the heat transmission from the droplets deposited later [31]. Therefore, the as-sprayed coatings were composed of an amorphous structure with a very small portion of nanocrystallines and oxides. In order to further investigate the GFA of the as-sprayed coatings, the Gibbs free energy difference (DG) between the liquid and crystalline phases in the super-cooled liquid region was calculated using the following simplified formula Eq. (4)[32]:

DG ¼ ðDHf  DTÞ=T m  c  DSf  ½DT  T  lnðT m =TÞ

ð4Þ

where DT = Tm–T, DSf = DHf/Tm and c is the proportionality coefficient, taken as 0.8 for metallic under-cooled liquids. DG, which reflects the stability of under-cooled liquid for alloys, has been reported to be a contributing factor for GFA [32]. A smaller DG means a larger embryo critical size for nucleation, and hence larger chemical fluctuations are required. Fig. 9 illustrates the DG curves

Fig. 9. The Gibbs free energy difference curves of the as-sprayed (a) F20, (b) F25 and (c) F30 Fe-based amorphous coatings.

124

W. Guo et al. / Materials and Design 78 (2015) 118–124

for the F20, F25 and F30 as-sprayed coatings. It is apparent that F30 coating exhibited the lowest DG, suggesting the highest GFA. The results showed a similar tendency with that analyzed by DTx and c criteria as well as the enthalpies associated with the crystallization (85.4 J/g for the F30 coating). 6. Conclusions In consideration of various empirical rules for the composition design for glass formers, Fe-based cored wires with high glass forming ability were well designed to render the arc-sprayed coating a high content of amorphous phase. Then, Fe-based coatings with a high amorphous content were firstly fabricated on the AISI 1020 steel substrate by a twin-wire arc spray technology. With a dense structure and a low porosity of only approximately 2%, the coatings exhibited an attractive combination of high hardness (900–1100 HV0.3) and superior bonding strength (44.9– 54.8 MPa). The glass forming ability of these as-sprayed coatings was evaluated by five criteria of atomic volume strain criterion kn, the super-cooled liquid region DTx, the reduced glass transition temperature Trg, Lu’s criterion factor c value and the Gibbs free energy difference DG. The coating with the highest kn of 0.21 was proposed to possess the highest GFA based on the analysis of DTx, c, DG, and the enthalpy of crystallization DH. Although the composition inhomogeneous and formation of oxides in arcsprayed coatings is generally inevitable, the TEM analysis results showed that the Fe-based amorphous coatings in the present study were approximately composed of a homogeneous amorphous structure. Acknowledgements The authors would like to acknowledge the financial supports from Fundamental Research Funds for the Central Universities (2013B34414, 2013B22814), National Natural Science Foundation of China (50979028), Natural Key Foundation of Jiangsu Provience (BK2011025) and National 973 Plan Project (2012CB719804, 2015CB057803). References [1] E. Axinte, Metallic glasses from ‘‘alchemy’’ to pure science: present and future of design, processing and applications of glassy metals, Mater. Des. 35 (2012) 518–556. [2] C. Suryanarayana, A. Inoue, Iron-based bulk metallic glasses, Int. Mater. Rev. 58 (3) (2013) 131–166. [3] W.M. Guo, Y.P. Wu, J.F. Zhang, S. Hong, G.Y. Li, G.B. Ying, et al., Fabrication and characterization of thermal-sprayed Fe-based amorphous/nanocrystalline composite coatings: an overview, J. Therm. Spray Technol. 23 (7) (2014) 1157–1180. [4] V. Varadaraajan, R.K. Guduru, P.S. Mohanty, Synthesis and microstructural evolution of amorphous/nanocrystalline steel coatings by different thermalspray processes, J. Therm. Spray Technol. 22 (4) (2013) 452–462. [5] J.R. Lin, Z.H. Wang, P.H. Lin, J.B. Cheng, X. Zhang, S. Hong, Effects of post annealing on the microstructure, mechanical properties and cavitation erosion behavior of arc-sprayed FeNiCrBSiNbW coatings, Mater. Des. 65 (2015) 1035– 1040. [6] F. Otsubo, H. Era, K. Kishitake, Formation of amorphous Fe–Cr–Mo–8P–2C coatings by the high velocity oxy-fuel process, J. Therm. Spray Technol. 9 (4) (2000) 494–498.

[7] K. Kishitake, H. Era, F. Otsubo, Thermal-sprayed Fe–10Cr–13P–7C amorphous coatings possessing excellent corrosion resistance, J. Therm. Spray Technol. 5 (4) (1996) 476–482. [8] V.R.S.S. Brito, I.N. Bastos, H.R.M. Costa, Corrosion resistance and characterization of metallic coatings deposited by thermal spray on carbon steel, Mater. Des. 41 (2012) 282–288. [9] A. Inoue, Stabilization of metallic supercooled liquid and bulk amorphous alloys, Acta Mater. 48 (1) (2000) 279–306. [10] W.L. Johnson, Bulk glass-forming metallic alloys: Science and technology, Mrs Bull. 24 (10) (1999) 42–56. [11] Y. Zhang, T.T. Zuo, Z. Tang, M.C. Gao, K.A. Dahmen, P.K. Liaw, et al., Microstructures and properties of high-entropy alloys, Prog. Mater. Sci. 61 (2014) 1–93. [12] C. Suryanarayana, A. Inoue, Bulk metallic glasses, CRC Press, New York, 2011. [13] G. Wang, H.B. Fan, Y.J. Huang, J. Shen, Z.H. Chen, A new TiCuHfSi bulk metallic glass with potential for biomedical applications, Mater. Des. 54 (2014) 251– 255. [14] X.B. Zhao, Z.H. Ye, Microstructure and wear resistance of molybdenum based amorphous nanocrystalline alloy coating fabricated by atmospheric plasma spraying, Surf. Coat. Technol. 228 (2013) S266–S270. [15] J. Zhou, J.K. Walleser, B.E. Meacham, D.J. Branagan, Novel in situ transformable coating for elevated-temperature applications, J. Therm. Spray Technol. 19 (5) (2010) 950–957. [16] X.B. Liang, J.B. Cheng, J.Y. Bai, B.S. Xu, Erosion properties of Fe based amorphous/nanocrystalline coatings prepared by wire arc spraying process, Surf. Eng. 26 (3) (2010) 209–215. [17] J.B. Cheng, Z.H. Wang, B.S. Xu, Wear and corrosion behaviors of FeCrBSiNbW amorphous/nanocrystalline coating prepared by Arc spraying process, J. Therm. Spray Technol. 21 (5) (2012) 1025–1031. [18] J.L. Wu, Y. Pan, J.D. Huang, J.H. Pi, Non-isothermal crystallization kinetics and glass-forming ability of Cu–Zr–Ti–In bulk metallic glasses, Thermochim. Acta. 552 (2013) 15–22. [19] J.L. Cárdenas-Leal, J. Vázquez, D.G. Barreda, R. González-Palma, P.L. LópezAlemany, P. Villares, Analysis of the glass-crystal transformation kinetics by means of the theoretical method developed (TMD) under both non-isothermal and isothermal regimes. Application to the crystallization of the Ag0.16As0.34Se0.50 glassy alloy, J Alloys Compd. 622 (2015) 610–617. [20] K. Lu, Nanocrystalline metals crystallized from amorphous solids: nanocrystallization, structure, and properties, Mater. Sci. Eng. R 16 (4) (1996) 161–221. [21] H.E. Kissinger, Reaction kinetics in differential thermal analysis, Anal. Chem. 29 (11) (1957) 1702–1706. [22] M. Zhu, J.J. Li, L.J. Yao, Z.Y. Jian, F.E. Chang, G.C. Yang, Non-isothermal crystallization kinetics and fragility of (Cu46Zr47Al7)97Ti3 bulk metallic glass investigated by differential scanning calorimetry, Thermochim. Acta 565 (2013) 132–136. [23] K. Chokethawai, D.G. McCartney, P.H. Shipway, Microstructure evolution and thermal stability of an Fe-based amorphous alloy powder and thermally sprayed coatings, J. Alloys Compd. 480 (2) (2009) 351–359. [24] Q.J. Chen, H.B. Fan, J. Shen, J.F. Sun, Z.P. Lu, Critical cooling rate and thermal stability of Fe–Co–Zr–Y–Cr–Mo–B amorphous alloy, J. Alloys Compd. 407 (1) (2006) 125–128. [25] G.M. Liu, K. Rozniatowski, K. Kurzydlowski, Quantitative characteristics of FeCrAl films deposited by arc and high-velocity arc spraying, Mater. Charact. 46 (2–3) (2001) 99–104. [26] H.S. Ni, X.H. Liu, X.C. Chang, W.L. Hou, W. Liu, J.Q. Wang, High performance amorphous steel coating prepared by HVOF thermal spraying, J. Alloys Compd. 467 (1) (2009) 163–167. [27] G. Bolelli, V. Cannillo, L. Lusvarghi, T. Manfredini, Wear behaviour of thermally sprayed ceramic oxide coatings, Wear 261 (11–12) (2006) 1298–1315. [28] Y. Peng, C. Zhang, H. Zhou, L. Liu, On the bonding strength in thermally sprayed Fe-based amorphous coatings, Surf. Coat. Technol. 218 (2012) 17–22. [29] T.D. Shen, B.R. Sun, S.W. Xin, Effects of metalloids on the thermal stability and glass forming ability of bulk ferromagnetic metallic glasses, J. Alloys Compd. 631 (2015) 60–66. [30] Z.P. Lu, C.T. Liu, A new glass-forming ability criterion for bulk metallic glasses, Acta Mater. 50 (13) (2002) 3501–3512. [31] Y.P. Wu, P.H. Lin, G.Z. Xie, J.H. Hu, M. Cao, Formation of amorphous and nanocrystalline phases in high velocity oxy-fuel thermally sprayed a Fe–Cr– Si–B–Mn alloy, Mater. Sci. Eng. A 430 (1) (2006) 34–39. [32] U.R. Pedersen, Direct calculation of the solid–liquid Gibbs free energy difference in a single equilibrium simulation, J. Chem. Phys. 139 (10) (2013) 104102-1–104102-9.