Preparation and characterization of highly amorphous HVOF stainless steel coatings

Abstract

A partially amorphous FeCrMoWBCSi powder has been HVOF sprayed in order to produce highly amorphous
coatings. The extinction or retention of crystalline phases due to the spraying process is discussed.
Amorphicity in coatings is associated with a high melting degree. The latter is attained by a high particle
temperature and sufficient residence time in the flame. Coating properties, such as porosity, microhardness
and adhesion strength are evaluated. The lowest coating porosity corresponds to the most
amorphous coating. The least crystalline coating presents the highest corrosion resistance in 3.5% NaCl.

Full text

Journal of Alloys and Compounds 504S (2010) S283–S287
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Preparation and characterization of highly amorphous
HVOF stainless steel coatings
D. Zoisa,b, A. Lekatoua,∗, M. Vardavouliasb
aDepartment of Materials Science and Engineering, University of Ioannina, 45111 Ioannina, Greece
bPyrogenesis SA, Technological Park of Lavrion, 19500 Lavrion, Greece
article info
Article history:
Received 2 July 2009
Received in revised form 16 January 2010
Accepted 10 February 2010
Available online 18 February 2010
Keywords:
Nanostructured steel
Coating materials
Thermal spraying
Corrosion
X-ray diffraction
abstract
A partially amorphous FeCrMoWBCSi powder has been HVOF sprayed in order to produce highly amor-
phous coatings. The extinction or retention of crystalline phases due to the spraying process is discussed.
Amorphicity in coatings is associated with a high melting degree. The latter is attained by a high particle
temperature and sufficient residence time in the flame. Coating properties, such as porosity, micro-
hardness and adhesion strength are evaluated. The lowest coating porosity corresponds to the most
amorphous coating. The least crystalline coating presents the highest corrosion resistance in 3.5% NaCl.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
A method of reducing the overall manufacturing costs of a
Bulk Metallic Glass (BMG) component with superior properties
is the deposition of metallic amorphous layers on metal sub-
strates. The fabrication of amorphous coatings requires rapid
quenching from the melt at cooling rates fast enough to sup-
press the nucleation and growth of competing crystalline phases
[1]. One way to achieve this is to employ thermal spray pro-
cesses. For example, Sampath [2] investigated the application of
Air Plasma Spraying (APS) and Vacuum Plasma Spraying (VPS)
to deposit Ni–Mo based metallic glass coatings (MGCs). The APS
coatings contained defects such as porosity, insufficient interlamel-
lar bonding, oxide inclusions, etc.; in the VPS coatings, crystalline
structures had been developed, due to the absence of convective
cooling.
It is conceivable that minimal oxide presence, minimal intercon-
nected porosity and high amorphicity can significantly improve the
corrosion performance of a coating. High Velocity Oxy-Fuel (HVOF)
spraying may lead to lower oxidation than plasma spraying due to
lower temperatures, and low coating porosity due to the super-
sonic velocities of the propelled particles [3]. However, complete
amorphization during quenching might not be accomplished. For
instance, HVOF NiCrMoB MGCs have been reported to consist of
∗Corresponding author. Tel.: +30 26510 07309; fax: +30 26510 07034.
E-mail address: alekatou@cc.uoi.gr (A. Lekatou).
crystalline precipitates – Cr2O3, NiCr2O4and M3B2(M = Mo or Cr)
– within an amorphous Ni-rich matrix [1].
Cost-wise, there is a growing interest in the employment of
Fe and Cu-based BMGs, as potential replacements for the much
more expensive Zr-, Ti-, Ni- and Ln-based alloys [4]. To date, the
vast majority of the works on the corrosion behavior of Fe-based
MGs refer to bulk materials [e.g. 5–9]. Regarding the corrosion
behavior of high entropy coatings, an HVAF (High Velocity Air Fuel
spraying) NiNbTiZrCoCu coating exhibited high corrosion resis-
tance in aqueous HCl, though lower than that of the respective
amorphous ribbons [10]. The corrosion resistance of an HVOF FeCr-
MoCBY coating in 1 M HCl was comparable to that of the bulk
amorphous counterpart and superior to the corrosion resistance of
electroplated Cr-based coatings [11]. Sputtered Fe–Ni–Cr–W MGCs
presented excellent pitting resistance in neutral and acidic solu-
tions, which was preserved after a crystallization heat treatment
[12].
This work is part of a wider investigation on stainless steel high
entropy coatings, aiming at identifying the HVOF spraying parame-
ters under which the amorphous fraction is maximized and coating
oxidation is minimized. The coatings are free of expensive rare
earth metals, such as Y, Ga and lanthanides, which are usually
added to amorphous steels as glass forming ability enhancers (GFA)
[13,14].
2. Experimental
A gas atomized Fe–18.3Cr–7.7Mo–1.6W–1.8Mn–14.9B–3.6C–2.6Si powder
(at.%) of nominal particle size (−53 + 10) ␮m, manufactured by the NanoSteel Co.,
0925-8388/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2010.02.062

S284 D. Zois et al. / Journal of Alloys and Compounds 504S (2010) S283–S287
Table 1
Spraying parameters.
Spraying series Oxygen-to-fuel ratio Total gas
supply (l/min)
Spraying
distance (mm)
Feed rate
(g/min)
Power output
(kW)
Particle
temperature
(◦C)
Particle
velocity (m/s)
NS13.6 720 250 38 85 1583 ±5 557 ±29
NS25.0 725 250 38 103 1723 ±7 714 ±19
NS36.4 735 250 38 108 1658 ±5 725 ±26
Fig. 1. Cross-section of the powder feedstock and elemental mapping.
was HVOF sprayed on 304 stainless steel coupons. Spraying was performed by a
Sulzer Metco 2700 Hybrid torch. During spraying, particle temperatures and veloc-
ities were recorded by the Oseir Spraywatch 3i system. Three spraying series were
conducted, where the principal variable was the oxygen-to-fuel (propane) ratio. The
spraying parameters are given in Table 1.
The powder and the coatings were microstructurally characterized by SEM-EDX
(Philips XL 40 SFEG). Microhardness was measured by a Shimadzu tester (14 mea-
surements per coating). The coating adhesion strength (3 measurements per torch
power) was measured by a portable elcometer (110 P.A.T.T.I.®) according to ASTM
C633-01. XRD patterns were obtained by a Bruker D8 Advance diffractometer.
Potentiodynamic polarization tests were performed on coupons ground to 1000
grit and encapsulated in epoxy, so that a surface area of ∼1cm
2was exposed to
aerated 3.5% NaCl at ambient temperature (Gill AC potentiostat by ACM Instruments,
scan rate: 10 mV/min). A standard three-electrode cell was employed, with Ag/AgCl
as the reference electrode and a platinum gauge as the counter electrode. Corrosion
current densities were determined by Tafel extrapolation [15].
3. Results and discussion
3.1. Feedstock characterization
Fig. 1a shows that the powder particles have a spherical shape
due to the gas atomization processing. Some particles are hollow,
a typical consequence of the atomization process. Precipitates rich
in W, Mo and Si are often observed within the coarsest particles
(Fig. 1b, c and d).
The XRD spectrum of the powder shows a hump typical of amor-
phicity (Fig. 2). Peaks corresponding to crystalline phases are also
observed: martensite (Fe1.86C0.14 ); borides of stoichiometries simi-
lar to Cr1.65Fe0.35 B0.96, t-Fe3B, Fe23B6; borocarbide of stoichiometry
similar to Fe23(B,C)6. Mo and Cr may also be contained in the borides
due to their high combining tendency with B [16]. In addition,
Cr, Mo and Si (to a lesser extent) may form substitutional solid
solutions with ␣-Fe in a wide range of compositions [17]; hence,
the possibility of participating in all the detected phases is strong.
The existence of borides could be harmful to the amorphous state,
Fig. 2. XRD patterns of the powder and the coatings. 1: Fe3O4,2:Fe
1.86C0.14 ,3:Fe
23B6,
4: Fe23(B,C)6,5:Fe
15Si3B2,6:Fe
3B, 7: Cr1.65Fe0.35 B0.96. The above compositions are
only indicative. The actual phases probably contain and other elements (Cr, Mo, Si,
W) in solid solution with iron.

D. Zois et al. / Journal of Alloys and Compounds 504S (2010) S283–S287 S285
Table 2
Characteristic properties of the coatings.
Coating Coating
thickness (␮m)
HV300 Porosity (%) Adhesion
(MPa)
NS1150 ±12 870 ±52 1.0 ±0.4 >60
NS2143 ±20 826 ±93 0.8 ±0.3 >60
NS3157 ±16 796 ±93 1.1 ±0.3 >60
because they lead to a reduced B content in the supercooled liquid,
and, consequently, to a reduced GFA [18].
3.2. Spraying parameters
As shown in Table 1, the highest temperature is recorded at the
intermediate oxygen-to-fuel ratio (NS2); it would be expected that
the highest temperature would be recorded at the highest ratio
producing the highest power output (NS3). It is postulated that
ratios greater than 5 lead to reduced flame temperatures because
of oxygen overload; on the other hand, ratios lower than 5 also
lead to reduced flame temperatures, because of insufficient oxygen
content. Critical oxygen-to-fuel ratios for maximum flame temper-
atures have also been noticed on spraying white cast iron [19] and
NiTiZrSiZn amorphous coatings [20].
Table 1 shows that the particle velocities follow the order
NS1<NS
2∼
=NS3in proportion to the power output.
3.3. Coating characterization
Table 2 shows low coating porosities. (It should be emphasized
that the employed image analysis techniques can only identify
the coarsest pores and give an overall view of the coating poros-
ity.) NS2coating presents the lowest porosity due to the highest
melting degree achieved at the highest particle temperatures. At
high particle temperatures, molten material of high diffusivity
fills the asperities and gaps of the previously deposited layers,
leading to low porosities [21]. All the coatings present similar
hardness values, which are substantially higher than the ones
reported for stainless steel 304s [8] and stainless steel thermally
sprayed coatings [3]. All the coatings present satisfactory adhesion
strength.
Fig. 3 illustrates a cross-section of NS1coating. In all the coatings,
a few coarse unmolten or semi-molten particles are discerned pre-
senting dispersions of rectangular particles. EDX analysis revealed
that these precipitates are (Mo,Cr,Fe,W) borocarbides or carbides.
The main reasons for the presence of coarse unmelted particles in
the coatings can be low particle temperatures (NS1,Table 1) and/or
Fig. 3. Cross-section of NS1coating; the arrow points at a coarse unmelted particle
with retained precipitates.
Fig. 4. Carbide particles (arrow pointed) along a splat layer (cross-section of NS2
coating).
short residence time in flame caused by high particle velocities
(NS2,NS
3,Table 1). Isolated rectangular particles are also discerned
in all the coatings. EDX analysis revealed that they are (Mo,Cr,Fe,W)
borocarbides or carbides. All the coatings present inter-splat oxide
stringers (dark contrast lines) running parallel to the substrate
(Fig. 4).
Fig. 2 shows that the humps in the XRD spectra of the coat-
ings are ampler than the humps in the powder spectrum. Also, the
“crystalline” peaks of the coatings are notably weaker than those of
the powder. The lower crystallinity of the coatings, as compared to
the powder feedstock, can be explained by the quenching of fully
molten particles to the glass state from their highest temperature,
through the action of flattening onto the substrate surface [22].
Comparison of the XRD scans of the three coatings, shows that
NS2coating exhibits minimal presence of ‘crystalline’ peaks, whilst
its broad halo peak (2= 40–50◦), is almost free of small ‘crystalline’
peaks. It seems that at the intermediate oxygen-to-fuel ratio, the
temperature was high enough to melt most of the particles; at the
same time, the particle velocity was high enough to ensure short
particle residence in the flame and, consequently, low oxidation.
NS3coating seems to present a higher crystalline fraction than NS1
coating, despite the higher particle temperature. The higher crys-
talline fraction of NS3, as compared to NS1, is mostly indicated
by the higher intensity peaks corresponding to Cr1.65Fe0.35 B0.96,
Fe23(B,C)6and Fe3O4, as well as the more distinct ‘crystalline’ peaks
Fig. 5. Potentiodynamic polarization behavior of the coatings in 3.5% NaCl, at ambi-
ent temperature.

S286 D. Zois et al. / Journal of Alloys and Compounds 504S (2010) S283–S287
Table 3
Corrosion properties extracted from the voltammograms of Fig. 5 (Ecorr: corrosion potential, icorr : corrosion current density, ip: current density at the first passive stage, Eb:
breakdown potential after the first passive stage).
Coating Ecorr (mV vs. Ag/AgCl) icorr (mA/cm2)ip(mA/cm2)Eb(mV vs. Ag/AgCl)
NS1−368 3.72 ×10−30.004–0.007 −52
NS2−352 1.60 ×10−30.002–0.006 41
NS3−387 4.74 ×10−30.007–0.02 −18a
aMetastable pitting starts at −150 mV.
within the hump. The most oxidizing environment during spraying
of NS3coating may account for its highest oxide content. The lower
amorphicity of NS3, as compared with NS1, despite the higher parti-
cle temperature, is attributed to the higher particle velocity causing
shorter particle residence in flame and, hence, less time for melt-
ing. A complementary reason for the relatively low amorphicity
of NS3coating could be the insufficient GFA and/or cooling rate in
regions of altered chemical compositions, such as the oxide stringer
vicinity.
The precipitates of the type Cr1.65Fe0.35 B0.96,Fe
23(B,C)6(also
detected in the pristine powder) have probably been retained in
the coarsest powder particles, which were deposited semi-molten.
Moreover, Fe23(B,C)6is reported as a principal phase in the system
Fe–B–C, at room temperature [23]. In the system Fe–B–X (X: any
one or a combination of C and one or two of the most common of
the first and second series transition elements), the main phases
at room temperature are (Fe,X) and (X,Fe)2B – almost pure X2B–
[23]; this justifies the detection of Cr1.65Fe0.35 B0.96 in the coatings.
The detection of Fe15Si3B2in the coatings is compatible with the
suggestion that Fe3(Si,B) is a Si-stabilized phase originating from
one of the metastable forms of Fe3B[24].
M3B and M23B6, which were present in the powder feed-
stock, are missing from the coatings or their peak intensity is
notably reduced. The notable reduction or extinction of t-M3B and
fcc-M23B6can be explained by their instability [16,25,26]. Upon
spraying, the borides were dissolved in the Fe-melt. During quench-
ing, the cooling rate was fast enough to suppress B segregation. The
oversaturated Fe was cooled as martensite and/or glass.
The XRD results can justify the similar hardness values for the
three coatings, given in Table 2. The hardness gained by the lower
porosity [27] and the higher amorphicity of the NS2coating, as com-
pared with NS1and NS3, is offset by the hardness lost owing to the
lower content of carbides, borides and oxides.
3.4. Corrosion behavior
The voltammograms of the deposited coatings in 3.5% NaCl,
are illustrated in Fig. 5. The extracted corrosion values are given
in Table 3. All the coatings exhibit passive behavior following the
initial active state. (NS3coating does not really show a current stabi-
lization trend, but rather a current limiting stage.) The subsequent
sharp increase in current density at relatively low potentials, sug-
gests breakdown of passivity by pitting. A second stage of current
stabilization follows pitting; however, the high currents indicate
the formation of highly unstable deposits in the pits.
The most amorphous coating (NS2) presents the noblest cor-
rosion potential, the lowest corrosion current density, the lowest
passive current densities and the highest breakdown–pitting
potential. The superior corrosion performance of NS2coating can
be attributed to its higher amorphous fraction, in comparison with
NS1and NS3, since the homogeneous glass phase is expected to
allow the growth of a uniform protective film. Furthermore, NS2
coating presents the lowest possibility amongst the three coatings
for galvanic effects between adjacent phases of different composi-
tions and crystal structures. For instance, the presence of carbides
in coatings reportedly accelerates microgalvanic corrosion [28].A
third reason for the corrosion resistance of NS2coating is its low
porosity that reduces the chances for interconnected porosity. On
the other hand, the lowest corrosion resistance is exhibited by the
least amorphous and most porous coating, namely NS3coating.
4. Conclusions
The main conclusions drawn from the characterization of coat-
ings manufactured by HVOF spraying of a FeCrMoWBCSi partially
amorphous powder under three different oxygen-to-fuel ratios
(3.6, 5 and 6.4), are:
1. The coatings present a higher amorphous fraction than the pow-
der feedstock, owing to substantial melting during spraying and
subsequent quenching to the glass state.
2. Spraying and quenching result in a significant decrease in the
metastable boride phases.
3. Amorphicity in coatings is associated with a high melting degree.
The latter is achieved by a high particle temperature (attained
at a critical oxygen-to-fuel ratio: in this case around 5) and suf-
ficient residence time in the flame (depending on the spraying
particle velocity).
4. The highest oxidation levels correspond to the highest oxygen-
to-fuel ratio.
5. The lowest coating porosity is attained at the highest tempera-
ture (i.e. intermediate oxygen-to-fuel ratio).
6. During potentiodynamic polarization in 3.5% NaCl, all the coat-
ings show passivation disrupted by pitting.
7. A high amorphous fraction combined with low porosity benefits
the corrosion resistance of a coating.
References
[1] A.H. Dent, A.J. Horlock, D.G. McCartney, S.J. Harris, Mater. Sci. Eng. A283 (2000)
242–250.
[2] S. Sampath, Mater. Sci. Eng. A167 (1993) 1–10.
[3] L. Pawloski, The Science and Engineering of Thermal Spray Coatings, Wiley,
1995.
[4] W.H. Wang, C. Dong, C.H. Shek, Mater. Sci. Eng. R 44 (2004) 45–89.
[5] H.X. Li, S. Yi, Mater. Chem. Phys. 112 (2008) 112–114.
[6] A. Lekatou, A. Marinou, P. Patsalas, M. Karakasides, J. Alloys Compd. 483 (2009)
514–518.
[7] J. Jayaraj, Y.C. Kim, K.B. Kim, H.K. Seok, E. Fleury, Sci. Technol. Adv. Mater. 6
(2005) 282–289.
[8] Y.Y. Chen, T. Duval, U.D. Hung, J.W. Yeh, H.C. Shih, Corros. Sci. 47 (2005)
2257–2279.
[9] M. Magrini, P. Matteazzi, A. Frignani, Mater. Chem. Phys. 13 (1985) 71–83.
[10] A.P. Wang, Z.M. Wang, J. Zhang, J.Q. Wang, J. Alloys Compd. 440 (2007) 225–228.
[11] H.S. Ni, X.H. Liu, X.C. Chang, W.L. Hou, W. Liu, J.Q. Wang, J. Alloys Compd. 467
(2009) 163–167.
[12] R. Wang, J. Non-Cryst. Solids 61&62 (1984) 613–618.
[13] Z.H. Gan, H.Y. Yi, J. Pu, J.F. Wang, J.Z. Xiao, Scripta Mater. 48 (2003) 1543–1547.
[14] Q.J. Chen, H.B. Fan, J. Shen, J.F. Sun, Z.P. Lu, J. Alloys Compd. 407 (2006) 125–128.
[15] M. Stern, A.L. Geary, J. Electrochem. Soc. 104 (1957) 56–61.
[16] L. Yijian, H. Jian, J. Mater. Sci. 26 (1991) 2833–2840.
[17] ASM Handbook: Alloy Phase Diagrams, vol. 3, ASM Intern., 1997.
[18] Y.P. Wu, P. Lin, C. Chu, Z. Wang, M. Cao, J. Hu, Mater. Lett. 61 (2007) 1867–1872.
[19] O. Maranho, D. Rodrigues, M. Boccalini Jr., A. Sinatora, Surf. Coat. Technol. 202
(2008) 3494–3500.
[20] H. Choi, S. Lee, B. Kim, H. Jo, C. Lee, J. Mater. Sci. 40 (2005) 6121–6126.
[21] D. Zois, A. Lekatou, M. Vardavoulias, I. Panagiotopoulos, A. Vazdirvanidis, J.
Therm. Spray Technol. 17 (2008) 887–894.
[22] D. Shin, F. Gitzhofer, C. Moreau, J. Therm. Spray Technol. 16 (2007) 118–127.

D. Zois et al. / Journal of Alloys and Compounds 504S (2010) S283–S287 S287
[23] K.S. Narasimhan, F.J. Semel, in: Proceedings of the 2007 International Con-
ference on Powder Metallurgy and Particulate Materials, May 13–16, Denver,
Colorado, Metal Powder Industries Federation, Paper No. 2007-01-0145.
[24] V. Raghavan, J. Phase Equilib. Diffus. 28 (2007) 380–381.
[25] A. Inoue, B.L. Shen, C.T. Chang, Intermetallics 14 (2006) 936–944.
[26] Q. Wei, B. Yang, A. Jiang, Chin. Phys. Lett. 8 (1991) 296–299.
[27] H. Luo, D. Goberman, L. Shaw, M. Gell, Mater. Sci. Eng. A346 (2003) 237–
245.
[28] M. Belkhaouda, L. Bazzi, A. Benlhachemi, R. Salghi, B. Hammouti, S. Kertit, Appl.
Surf. Sci. 252 (2006) 7921–7925.