Accepted Manuscript

CrAlON CAE-PVD coatings for oxidation and wear protection of TZM alloys in
FAST sintering applications Author names and affiliations

Eluxka Almandoz, Jonathan Fernández de Ara, Javier Martínez de Bujanda, José

Fernández Palacio, Rafael José Rodríguez, Zhenxue Zhang, Hanshan Dong, Yi Qin, Gonzalo
García Fuentes

PII: S0254-0584(18)30054-3

DOI: 10.1016/j.matchemphys.2018.01.056

Reference: MAC 20328

To appear in: Materials Chemistry and Physics

Received Date: 21 March 2017

Revised Date: 11 January 2018

Accepted Date: 14 January 2018

Please cite this article as: Eluxka Almandoz, Jonathan Fernández de Ara, Javier Martínez de
Bujanda, José Fernández Palacio, Rafael José Rodríguez, Zhenxue Zhang, Hanshan Dong, Yi
Qin, Gonzalo García Fuentes, CrAlON CAE-PVD coatings for oxidation and wear protection of TZM
alloys in FAST sintering applications Author names and affiliations, Materials Chemistry and
Physics (2018), doi: 10.1016/j.matchemphys.2018.01.056

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Title

CrAlON CAE-PVD coatings for oxidation and wear protection of TZM alloys in FAST

sintering applications

1
 ACCEPTED MANUSCRIPT

Author names and affiliations

Eluxka Almandoz*a, Jonathan Fernández de Araa, Javier Martínez de Bujandaa,b, José

Fernández Palacioa, Rafael José Rodríguezc,d, Zhenxue Zhange, Hanshan Donge, Yi

Qinf, Gonzalo García Fuentesa.

aAsociación de la Industria Navarra, Carretera Pamplona 1, 31191 Cordovilla, Spain.
bEscuela Técnica Superior de Ingeniería, Universidad del País Vasco, Alameda Urquijo

s/n, 48013 Bilbao, Spain.

cUniversidad Pública de Navarra, Campus de Arrosadía s/n, 31006, Pamplona, Spain.
dInstitute for Advanced Materials - INAMAT, Campus de Arrosadía s/n, 31006,

Pamplona, Spain.
eSchool of Metallurgy and Materials, University of Birmingham, Birmingham B15 2TT,

UK.
fDepartment of Design, Manufacturing and Engineering Management, Strathclyde

University, Glasgow, G1 1XJ, UK.

Corresponding author: Eluxka Almandoz - [email protected]

Jonathan Fernández de Ara – [email protected]

Javier Martínez de Bujanda – [email protected]

José Fernández Palacio – [email protected]

Rafael José Rodríguez – [email protected]

Zhenxue Zhang – [email protected]

Hanshan Dong – [email protected]

Yi Qin – [email protected]

Gonzalo García Fuentes – [email protected]

2
 ACCEPTED MANUSCRIPT

Abstract

In this work CrAlO and CrAlO/CrAlN multilayers deposited by cathodic arc

evaporation are evaluated as protective films in metal and ceramic powder FAST

sintering tool dies fabricated in titanium-zirconium-molybdenum allows (TZM). The

films have been characterised in terms of their composition, microstructure, mechanical

properties and thermal stability in air at high temperatures between 800 ºC and 1100 ºC;

in addition the tribological performance has been analysed at room temperature and at

400 ºC.

The crystalline structure and composition of the CrAlO based coatings are compatible

with the formation of a mixture of α-corundum and a cubic fcc (Cr,Al)2O3. The

crystalline structure of the multilayer is, on the other hand, dominated by the cubic fcc

lattice plane reflections of the CrAlN.

The deposited specimens have high hardness, between 25 and 30 GPa, which are stable

even after annealing at 1000 ºC. Even more, the multilayer coating also exhibited good

mechanical stability at 1100 ºC. The multilayer coating also exhibited an excellent

behaviour against wear at 400 ºC.

Sintering trials using coated TZM dies have been carried out using Ti90Sn10 and Al2O3

high energy ball milled powders. The experimental results show that the oxide based

coating formulations are potentially able to protect the tools from wear, sticking and

oxidation of their surfaces. This may allow the use of TZM material as an alternative to

other substrates such as graphite.

3
 ACCEPTED MANUSCRIPT

Keywords

Cathodic arc evaporation; oxide coating; wear resistance; hardness; sintering.

4
 ACCEPTED MANUSCRIPT

Main text

1. Introduction

Titanium-Zirconium-Molybdenum alloys (TZM) have excellent strength and creep

resistance at elevated temperatures. These properties make TZM alloy a candidate

material for tooling die in electric field sintering techniques (FAST). FAST is a recently

developed technique for multimaterial complex shape component manufacturing,

including the production of parts of small sizes and complex geometries [1]. Applying

heat by external electric current, the plasticity induced in the sintered material (also

called electroplasticity) paves the way [2], to enable the forming of difficult-to-process

materials such as titanium alloys and hard ceramics. FAST process is particularly

suitable for mass production of miniature or micro-parts due to its rapid heating nature

as described in the literature [2].

Today, graphite is the most used material for FAST tool dies, due to its mechanical

stability at high temperatures; required to be up to 800-900 ºC for metal powders and up

to 1300 ºC for ceramic powder sintering. However, graphite is brittle and difficult to

machine. Given that the FAST tool dies should withstand pressures of up to 150 MPa,

under high temperatures [3], TZM seems an appropriate alternative tool material for this

specific application. However, TZM alloy shows poor oxidation resistance at

temperatures above 400 ºC [4]. In this case, when the size of the tools and components

is scaled down to millimetres, soldering due to severe adhesion between the mould and

the working material would occur and therefore, the demoulding may become a major

challenge. For this reason, and as a part of the EU research project FP7-MicroFast-

608720, this work focuses on the investigation of high oxidation protection coatings to

enable the use of this alloy in FAST processes, specifically those deposited by vacuum

physical vapour deposition (PVD) techniques [5], [6].

5
 ACCEPTED MANUSCRIPT

High oxidation resistance coatings deposited by PVD or CVD techniques have been

extensively investigated in the literature [7]–[9]. Chromium or titanium nitrides

containing either Al (ternary) or Al and Si (quaternary) have demonstrated excellent

thermal stability and oxidation resistance at temperatures of up to 900 ºC [10]–[12].

Above these temperature ranges (i.e. 1000 ºC and higher) the coatings decompose under

different mechanisms and evolve to oxides with different compositions and

microstructures, thus experiencing volumetric changes which cause film structural

failures due to cracking.

Cr-Al-O coating systems deposited by PVD have been investigated in the last decade as

a coating material able to approach the mechanical performance -Al2O3 phases [13],

[14]. These coating can form solid solutions of the type -(Cr,Al)2O3 for certain

composition ranges at temperatures accessible by PVD techniques of 400-550 ºC [15],

due to the fact that its precursors, α-Cr2O3 and α-Al2O3, crystallize in the same spatial

group R-3c. On the other hand it has been reported that solid solutions of cubic fcc-

(CrAl)2O3 lattice structure can also be formed and are mechanically valid for

applications on cutting tools. Various scientific papers [16]–[21] have reported the

composition, microstructure, lattice and mechanical/wear properties of Cr-Al-O and Cr-

Al-O-N coating systems. Hirai et al. [22] reported on Cr-Al-O synthesized by pulsed

laser deposition at a substrate temperature of 400 °C, obtaining fcc cubic structures and

a metal non-metal ratio of 1 and an oxygen content of 25 at.%. Pedersen et al [15] and

Diechle et al [23] reported the formation of corundum-like structures in DC and r.f.

sputtered Cr-Al-O films.

Khatibi et al [18] studied the system (AlxCr1-x)2+yO3-y for various deposition conditions

in cathodic arc evaporation. The corundum phase was favoured at high Cr content and

at high O2 flows, while the cubic phase was observed mostly for high Al content and

6
 ACCEPTED MANUSCRIPT

low O2 flow. Diechle et at [23] found that the sputtered coatings developed corundum

structures α-(Al1−x,Crx)2O3 in the form solid solutions. Najafi et al [24] found that the

coatings Cr-Al-O could dynamically evolve from a fcc cubic crystalline structure to a -

(CrAl)2O3 during the deposition process when the coating thickness exceeded 2

microns. Ramm et al [25] developed different coating microstructures and compositions

based on the pulsed electron emission P3e arc evaporation of Cr/Al cathodes in oxygen

atmospheres. They showed that the produced coatings developed corundum-like lattice

structure forming a solid solution of the type (Al1 − xCrx)2O3.

The thermal stability and oxidation resistance of Cr-Al-O coatings have also been

addressed in different studies. Khatibi et al [18] showed that the thermal annealing of

fcc-(Al0.83Cr0.17)2.3O2.7 leads to the transformation from fcc to a corundum structure

which initiates at 900 ºC and is totally completed at temperatures above 1000 ºC. In the

case of films with an original corundum structure in the as deposited state, the thermal

annealing did not lead to any structural modifications up to 1100 ºC. Similar results

were found by Edlmayr et al [26] on vacuum annealed arc deposited coatings.

Even if the composition of a coating defines its performance, it only does it partially. Its

structure or architecture design plays also an important role. Multilayered PVD coatings

with different compositions provides the possibility to obtain enhanced hardness,

toughness and thermal stability [27]. In this paper a CrAlO monolayer and a

CrAlO/CrAlN multilayer PVD coatings have been studied as possible surface

modification solutions to enable the use of TZM alloy in high temperature, >1000 ºC,

FAST sintering. Most of the existing literature references reports on single coating

structures [18], [23], [24] and their properties. Only one recent, to the best of our

knowledge, reports on the properties of CrAlO/CrAlN nanomultilayers [28].

7
 ACCEPTED MANUSCRIPT

The “in service” performance of these substrate-coating systems was finally tested by

the realization of sintering trials, and the corresponding visual inspection of the sintered

parts and the coating integrity.

2. Experimental

2.1 Coating deposition

The coatings have been produced by cathodic arc evaporation in a commercial

METAPLAS MZR-323 PVD reactor. The reactor is equipped with 2 opposing

evaporation panels, each of them hosting 3 circular cathodes (63 mm diameter)

vertically aligned. The effective coating volume is about 400 mm × 400 mm × 500

mm. For the preparation of the coatings, cathodes of two types were used: cathodes of

Cr (99.8 % purity) for the bonding layer and of CrAl (50 at.% Cr - 50 at.% Al) for the

rest of the coating. All the relevant process parameters are depicted in Table 1.

Table 1: Experimental parameters of the deposition processes.

Coating Material A·h No. of Gases Total

bilayers Pressure (Pa)
Bonding layer Cr 20 - Ar 1.2
CrN 40 - N2 1.8
Monolayer CrAlON 100 - N2+O2 (63:37) 1.8
Multilayer CrAlON 50 N2+O2 (63:37) 1.8
5
CrAlN 50 N2 1.8

For the coating deposition, high purity Ar, N2 and/or O2 gases were introduced in the

chamber. A continuous DC bias of -50 V was applied on the substrates during the

8
 ACCEPTED MANUSCRIPT

deposition process. During all the processes, the substrates were rotated around the

central vertical axis of the reactor chamber at a speed of 5 rpm. A Cr+CrN adhesion

layer was deposited for all the coatings. The deposition temperature of 470 °C for the

monolayer process and 500 ºC for the multilayer process was achieved by a combined

plasma and electrical resistance heating process. Process temperatures have been

monitored using the tempering hardness–temperature curve of 100Cr6 steel, by putting

hardened samples of this material in the vacuum chamber during all the processes as

reported by Fernández de Ara et al. [29].

2.2 Substrate materials

Four different substrate materials were used for the study: 40 mm diameter discs of H13

hot work steel (53-54 HRc), with a chemical composition in weight of 0.40 % C, 1.2 %

Si, 0.30 % Mn, 5.5 % Cr, 1.2 % Mo and 0.86% V; 40 mm diameter discs of Inconel 718

with a chemical composition in weight of 51.7 % (Ni+Co), 19.7 % Cr, 4.8 % (Nb+Ta),

3.1 % Mo, 1.0 % Ti, 0.40 % Al; punches of Titanium-Zirconium-Molybdenum (TZM)

alloy with a chemical composition of 0.50 % Ti, 0.08 % Zr, 0.02 % C and Mo

(balance); and Boron doped Si wafers with a thickness of 525 ± 15 µm from Si-Mat.

Discs of H13 steel and of Inconel 718 were mirror polished (Ra < 15 nm) and cleaned in

ultrasonic bath using de-oiling agents and de-ionised water. All the substrates were

vacuum-heated and ion-bombarded before the coating process.

2.3 Characterization techniques

9
 ACCEPTED MANUSCRIPT

Glow Discharge Optical Emission Spectroscopy (GDOES) was used to obtain the

chemical composition depth profiles of the films. The GDOES analyses were done in a

Jovin-Yvon JY 1000 RF optical spectrometer equipped with 40 detection channels.

Glancing incidence X-ray diffraction (GI-XRD) diffractograms of the coatings were

acquired in a D8 Advance Bruker diffractometer using a Cr Kα1 radiation source (λ =

0.22897 nm) at an incidence angle of 1º. The coating thickness was measured with the

help of the Calotest (CSM instruments, ball diameter 30 mm). The surface roughness

was measured before and after coating deposition with a WYCO-RST 500 white light

profilometer using the vertical scanning interferometry mode (VSI). For the wear

resistance analysis, CSM HT tribometer under ball-on-disc configuration was used. The

sliding tests were made using alumina counterballs of 6 mm in diameter at normal loads

of 5 N when carried out at room temperature and at 2 N when carried out at high

temperature (200 ºC and 400 ºC). The wear rates were calculated after measuring the

volume losses of the specimens by white light profilometry. A Field Emission Scanning

Electron Microscope (FESEM) HITACHI S-4800, coupled with an Energy Dispersive

X-ray Spectroscopy (EDX) detector for elemental analysis, was used to inspect the

cross section of the films.

Nanoindentation tests were performed on as-deposited samples using a Hysitron

Triboindenter© 950 fitted with a Berkovich indenter with a tip end radius of around 150

nm. Forty indentations, separated enough between them not to influence each other,

were made on the samples at 5 mN of maximum load. The load function consisted on 5

seconds of loading segment, followed by 2 seconds of holding time to account for

material creep and 5 seconds of unloading.

10
 ACCEPTED MANUSCRIPT

The Oliver and Pharr method [30] was used to extract the hardness and Young’s

modulus values from the curves. At 5 mN, the maximum penetration of the indenter in

the surface was around 120 nm for a 1.8 µm thick for the monolayer coating, and

around 105 nm for a 2.4 µm thick multilayer coating. Those penetrations are well below

the 10 % of the total thickness of the coating, value widely accepted for the measuring

of the mechanical properties of the layer with no interference of the substrate.

High temperature annealing tests in ambient air were carried out in a Nabertherm LT

24/11/B180 muffle oven from 800 ºC up to 1100 ºC in steps of 100 ºC. A heating rate of

20 ºC·min-1, a holding time of 2 hours was used for each step sample and natural oven

cooling. Nanoindentation tests and SEM-EDX analyses were carried out after the high

temperature annealing tests.

2.4 Sintering trials

A Gleeble-3800 thermal simulation machine was used for the Micro-FAST sintering

tests of Al2O3 (99.9% purity) and Ti90Sn10 (99.9%) powders with averaged particle

size of 0.18 µm and 10-45 µm respectively were used for the sintering experiments. The

powders were synthesized by high energy ball milling at MBN nanomaterialia S.r.l. The

tooling rig consists of a TZM punch / graphite die couple, as shown in Figure 5. Two

TZM punches, coated with a commercial PVD CrAlSiN and PVD AlCrO/AlCrN

multilayer respectively were prepared for the sintering of a cylindrical sample of Φ4.0

mm × 4.0mm aiming to examine the feasibility of the PVD coating on TZM punch.

The deposition conditions and properties of the CrAlSiN PVD coating are described by

R. Ji [31] and dimensions of the die and punches by Hijji et al. [2]. The pre-determined
11
 ACCEPTED MANUSCRIPT

amount of powders were poured into the graphite die and then closed by the pre-coated

TZM punch. A heating rate of 50 °C/s and a holding time of 240 s were used to sinter

the powders under a pressure of 75 MPa in vacuum (45 Pa). Two sintering temperatures

of 900 °C and 1100 °C were used for Ti90Sn10 powders and of 1200 °C and 1300 °C

for Al2O3 powders.

3. Results and discussion

3.1 Chemical composition and microstructure

Figure 1 shows the chemical in-depth profiles of both coatings obtained by GD-OES.

12
 ACCEPTED MANUSCRIPT

100
Fe
90
80 Al
C
70 Cr
Atomic %

O Fe
60
N
50 O
40
30
20 Cr
Al
10 N
C
0
0 1 2 3 4
Depth (microns)
a)

100
Fe
90
80
Al
70 C
Atomic %

60 Cr
Fe
50 O N
40 O
30
20 Cr
Al
10 N
C
0
0 1 2 3 4
Depth (microns)
b)

Figure 1: GD-OES analyses of the a) monolayer CrAlO coating, and b) multilayer

CrAlO/CrAlN coating.

Both coatings present a Cr/Al atomic ratio close to 1/1, in agreement with the nominal

composition of the cathodes. In the case of the monolayer, the overall chemical

composition responds to the stoichiometric formula (Cr,Al)2O3. In the case of the

multilayer the alternation of the oxide and the nitride coatings hinders a totally reliable

quantification of the composition for the individual sub-layers. In any case, it can be

seen that the nitrogen composition of the nitride sublayers is smaller than the sum of the

Cr plus Al. On the other hand, the oxide sublayers seem to retain the chemical

composition of the corresponding monolayer, as the deposition conditions are similar in

both cases. The total thicknesses of the monolayer and the multilayer coatings on H13

13
 ACCEPTED MANUSCRIPT

were 1.8 microns and 2.4 microns respectively. The difference in thickness is due to a

higher coating growth speed of the nitride sublayers than that of the oxides, as reported

by Khatibi et al [32]. The bilayer period of the multilayer (λ) is of 250 nm on silicon

substrate and the thickness ratio of the CrAlO/CrAlN is approximately 2:1.

Figure 2 show the cross-section images of the coatings as deposited on Si substrates (the

silicon substrate appears darker than the coating). The Cr+CrN bonding layer, of around

200 nm, can be clearly distinguished.

Si
a)

Si
b)

Figure 2: SEM images of the cross-section of as-deposited coatings on Si waflers, a)

monolayer CrAlO, b) multilayer CrAlO/CrAlN.

14
 ACCEPTED MANUSCRIPT

The monolayer presents a compact structure in which some columnar growth can be

distinguished. On the other hand the multilayer exhibits a dense columnar structure in

the nitride layer (bright), which is barely replicated in the oxide sublayers (dark). The

monolayer coatings in general exhibit a higher number of internal droplets than the

multilayer. This excess of droplets is caused by the poisoning of the Cr/Al in the

presence of oxygen.

Figure 3 shows the diffractograms of the deposited coatings on Inconel substrates.

*
a.u.

(104)(110) (113) Multilayer

* *
(024) (116)
Monolayer

50 60 70 80 90 100 110 120

/2

Figure 3: GI-XRD diffractograms of both untreated coatings.

15
 ACCEPTED MANUSCRIPT

The diffractogram of the monolayer coating shows peaks centered at 51.8º, 55.8º, 65.9º,

79.3º and 88.4º, which are all located in between the theoretical positions of the planes

(104), (110), (113), (024) and (116) respectively, of the phases α-Cr2O3 eskolaite and α-

Al2O3 corundum, (lower and upper limits in 2θ indicated as green dotted lines). This

group of peaks exhibits large widths, indicating the presence of small size crystalline

domains. No double peaks are observed in this set of diffractions, which may indicate

the formation of a lattice structure of the ‘solid solution’ type. At the 2 range where the

plane (113) of eskolaite-corundum diffracts, there are also other possible contributions

giving rise to reflections such as AlCr intermetallics. Therefore, the peak at 65.9º cannot

be invariably assigned to the plane -(113).

The peak at 69.1º, marked with an asterisk, can be attributed to a diffraction of a lattice

plane (002) of a cubic fcc phase, likely of (Cr,Al)2O3 stoichiometry, as shown by

GDOES. This is compatible with Khatibi et al [18] and more recently by different

authors [24], who observed the presence of cubic fcc lattice structures in arc deposited

coatings of stoichiometry close to (Cr,Al)2O3. This phase has been described as a cubic

fcc lattice with a 33% cation vacancies. The corresponding lattice parameter obtained

using the Bragg equation and the miller indexes (200) is 0.402 nm, close to that found

by Khatibi et al. [33] for the cubic fcc lattice phases (0.404 nm) of stoichiometry

(CrxAl1-x)2O3 with 0.6 < x < 0.7. On the other hand, the peak at 107.3º is compatible

with the reflections of the planes (022) of the same cubic structure. In fact, the estimated

lattice parameter using the miller indexes (022) is also of 0.402 nm.

The diffraction pattern of the CrAlO/CrAlN multilayer exhibits mainly the reflections of

the planes (111), (200) and (022) of the fcc CrAlN at 57.8º, 67.7º and 104.8º, which

appear in Figure 3 marked with a black circle. This is in agreement with the larger

volume fraction of nitride sub-layers with respect to the total film volume, as observed
16
 ACCEPTED MANUSCRIPT

by SEM cross sections. On the other hand the diffraction pattern of the multilayer

sample also exhibits diffraction peaks at 65.9º, from the (113) planes of the α-

(Cr,Al)2O3 phase and at 107.3º attributed to the (022) planes of the cubic fcc-

(Cr,Al)2O3, in the form of shoulders of the (200) and (022) fcc-CrAlN peaks

respectively. These peaks stem from the oxide sublayers of the multilayer. None of the

other peaks present in the monolayer coating could be observed in the multilayer

formulation.

3.2 Roughness, hardness and wear tests

Table 2 shows the roughness (Ra: arithmetic average height in the profile and Rq:

geometric average height in the profile) and thickness of the coatings. The roughness

values of the monolayer are higher than these typically obtained for transition metal

nitrides deposited by CAE-PVD as in the cases of Antonov et al [34] and Bayón et al

[35] where they obtained Ra roughness values in the range of the 30 to 100 nm.

Table 2: Roughness (Ra and Rq) and thickness of the monolayer and multilayer

coatings.

Coating Ra (nm) Rq (nm) Coating thickness (µm)

Monolayer 135.2 ± 15.9 231.8 ± 33.1 1.8 ± 0.1

Multilayer 58.2 ± 2.6 120.8 ± 9.2 2.4 ± 0.2

17
 ACCEPTED MANUSCRIPT

This is due to the presence of oxygen in the chamber during the deposition. This fact

was also reported by Khatibi et al. [32] and Sjölén et al. [36], who found that the

introduction of O2 in the chamber as a reactive gas results in the extension of arcs over

the target surface compared to the locally dense arcs formed in a pure nitrogen

atmosphere. On the contrary, the roughness values of the multilayer coating are

significantly smaller than these of the monolayer (cf. Table 2), due to the absence of

oxygen gas during the nitride deposition cycles. During these cycles, the oxide layer of

the target is removed, leading to a lesser ejection of microdroplets.

Table 3 shows the indentation hardness and modulus of the coatings, as deposited and

after a thermal annealing in air at 1000 ºC for the monolayer; and 1000 ºC and 1100 ºC

in the case of the multilayer. The multilayer coating exhibits an indentation hardness of

29.8 GPa and a modulus of 219 GPa; whereas these for the monolayer are 25.2 GPa and

193 GPa respectively. Different values of indentation hardness are reported in the

literature for Cr-Al-N, Cr-Al-O and a multilayer CrAlO/CrAlN [28]. Arc evaporated

CrAlN and CrAlSiN hardness as high as 35-40 GPa has been reported by Endrino et al

[37] and Polcar et al [38]. Khatibi et al [18] reported values from 24 to 30 GPa for arc

deposited CrAlO films depending on the Cr/Al ration, and Najafi et al [21] in the range

33 to 25 GPa. Cr-Al-O sputtered coatings, on the other hand, exhibited hardness

between 24–27 GPa [15]. Raab et. al [28] found that the indentation hardness of

CrAlO/CrAlN multilayers as the bilayer period increases. The hardness/modulus ratio

H3/E’2 has also been calculated for the two coatings, as an indicator of the resistance to

plastic deformation [38]. In this case, the H3/E’2 of the multilayer coating is higher than

that of the monolayer.

18
 ACCEPTED MANUSCRIPT

Table 3: Hardness, Young Modulus and H3/E’2 values of the CrAlO monolayer coating

and of the CrAlO/CrAlN multialyer coating in the untreated state and after the high

temperature resistance tests at 1000 ºC and 1100 ºC.

Coating Monolayer Multilayer

As After annealing As After annealing After annealing
State
deposited 1000 ºC deposited 1000 ºC 1100 ºC
H (GPa) 25.2 ± 2.7 25.8 ± 3.4 29.8 ± 2.6 27.9 ± 3.8 21.3 ± 3.6
E’ (GPa) 193 ± 13 191 ± 13 219 ± 8 207 ± 20 194 ± 9
H3/E’2 0.430 0.471 0.552 0.507 0.257

Table 4: Wear rate (in units of m3/Nm) at room temperature, 200 ºC and 400 ºC.

Sample Room temperature 200 ºC 400 ºC

Monolayer (1.85 ± 0.52) × 10-16 (7.39 ± 1.10) × 10-15 Catastrophic wear

Multilayer (3.45 ± 1.30) × 10-17 No measurable wear No measurable wear

or build-up or build-up

Table 4 shows the wear rates of the coatings measured at RT, 200 ºC and 400 ºC. At RT

the monolayer coating shows a wear rate of 1.85  10-16 m3/Nm, and the multilayer of

3.45 × 10-17 m3/Nm, which is one order of magnitude smaller than that of the

monolayer. At 200 ºC, the wear resistance of the monolayer coating decreases down to

7.39 × 10-15 m3/Nm (cf. Table 4), and is strongly worsened at 400 ºC. The multilayer

coating shows better wear resistance at 200 ºC and 400 ºC than at room temperature.

Instead of revealing a wear track during the wear tests, a slight material growth could be

19
 ACCEPTED MANUSCRIPT

observed in the tracks. Polcar et al [38] showed values of wear rate for arc deposited

AlCrN, and AlCrSiN films of around 510-16 m3/Nm, well stable between RT and

400ºC, using similar testing conditions of 5N load with an Al2O3, 6 mm diameter

counterball.

3.3 High temperature annealing tests

The morphology of the films after the annealing is shown in Figure 4 in plain view.

Temperature (ºC) Monolayer Multilayer

As deposited

800

1000

20
 ACCEPTED MANUSCRIPT

1100

1100

higher

magnification

Figure 4: SEM images of the samples’ surface untreated and after high temperature

annealing tests.

In the range of 800 ºC to 1000 ºC, the samples showed features similar to those in the

untreated state, and no major degradation signs could be observed in this temperature

range. Annealing at 1100 ºC produced a transformation on the surface morphology of

both coatings. The surface of the monolayer shows a large number of patches indicating

a partial degradation of its integrity. In the case of the multilayered coating, the signs of

thermal degradation are much less visible.

EDX semi-quantitative analyses have been done in different zones, and shown in Table

5, to confirm the coating detachment from the observed patches of Fig. 4 after

annealing. In the insert of Table 5, the SEM plain view pictures of the patches for the

monolayer and the multilayer is shown. Zones 1, correspond to the areas where the

coating integrity is preserved, and Zones 2 where the bare substrate is visible. The Ni,

Fe and other element signals, from the substrate, increases in Zones 2 with respect to

21
 ACCEPTED MANUSCRIPT

these measured on Zones 1, whereas the signal of the coating constituents, such as Cr,

and Al decrease, though do not totally vanished.

Table 5: Semi-quantitative elemental analysis on the surface of the monolayer and the

multilayer coated Inconel after annealing of 1100 ºC in air. Data in at%.

Monolayer Zone 1 Zone 2

N - -

O 63.83 57.75

Al 4.26 1.50

Ti 2.25 1.80

Cr 28.74 11.78

Mn 0.31 0.08

Fe 0.31 3.84

Ni 0.00 9.83

Nb 0.30 13.41

Multilayer Zone 1 Zone 2

N - -

O 52.59 69.32

Al 5.03 4.21

Ti 1.47 0.43

Cr 37.32 15.51

Mn - -

Fe 1.10 0.94

Ni 2.49 6.62

Nb 0.00 2.96

22
 ACCEPTED MANUSCRIPT

With regards to the mechanical properties after annealing, the hardness of the annealed

monolayer coating (cf. Table 3) remains at the same value at 1000 ºC as this measured

at RT, whereas the hardness of the multilayer slightly decreases to 27.9 GPa. After the

annealing at 1100 ºC the monolayer coating is not stable anymore and the

nanoindentation curves do not provide reliable hardness values. Interestingly the

multilayer coating shows a high stable hardness value of 21.3 GPa even after an

annealing at 1100 ºC. The mechanical stability of Cr-Al-O coatings has also been

reported by only few authors. For example, Khatibi et al. [18] reported that the

indentation hardness of arc evaporated Cr-Al-O coatings dropped 3 - 5 GPa after

annealing at 1100 °C compared with the untreated in the fcc cubic state. Tien et al [39],

reported on the decrease of CrAlSiN arc deposited coatings from 30.2GPa to 28.3 GPa

after 1 h thermal annealing at 900ºC. At 1100ºC, the degradation of the coating hindered

reliable measurements of indentation hardness.

3.4 Sintering trials

For the FAST sintering trials, the multilayer coating performance has been qualitatively

compared with a benchmark CrAlSiN coating deposited by arc evaporation. It is

noteworthy to highlight that, with the bare TZM punch uncoated, the sintering trials

were unsuccessful because the powders stuck on the tool, making the part ejection not

possible.

Figure 5 shows the graphite die and the TZM punch after sintering of a) Ti90Sn10

powders at 1100 ºC and b) Al2O3 powders at 1300 ºC. After sintering Ti90Sn10

powders at temperature of 900 °C and 1100 °C, the TZM punch coated with the

23
 ACCEPTED MANUSCRIPT

CrAlSiN benchmark coating can be easily withdrawn from the graphite die without

sticking. The TZM punch coated with CrAlO/CrAlN coating can be also slipped out of

the die and the sintered Ti90Sn10 part, with a relative density of 95 %, could be easily

separated from the punch (Figure 5a).

a) b)

Figure 5: PVD coated TZM punches and graphite die assembly after sintering a)

Ti90Sn10 powders at 1100 ºC and b) Al2O3 powders at 1300 ºC. The punch on the left

of both photos is CrAlO/CrAlN coated and the one in the right CrAlSiN coated.

The PVD coated TZM punches / graphite die assemblies were further tested by sintering

Al2O3 powders at 1200 °C and 1300 °C (Figure 5b). After the sintering, the TZM punch

with the CrAlO/CrAlN coating can be easily pulled out of the graphite die without

sticking. The sintered Al2O3 sample has a relative density over 90 % and it could be

ejected out of the graphite die by standard procedures. After four Al2O3 powder

sintering tests, the punches were sectioned to inspect the integrity of the coatings. The

TZM punch coated with CrAlSiN coating can be taken off the die but the punch surface

became rougher indicating that the CrAlSiN coating was degraded at such high

temperatures. As observed in Figure 6a, some Al2O3 and graphite debris remained stuck

on the surface of the CrAlSiN coated TZM punch, which could be confirmed by the

EDX analysis, and the coating was partially damaged after sintering Al2O3 powders at

1300 °C.
24
 ACCEPTED MANUSCRIPT

coating
coating

substrate substrate

a) b)

Figure 6: Cutaway views of the TZM punches after 4 sinterings a) CrAlSiN coated and

b) CrAlO/CrAlN coated.

On the other hand, the amount of adhered Al2O3 sintered powders on the punch coated

with the multilayer is much smaller, as shown in Figure 6b; in addition the integrity of

the CrAlO/CrAlN multilayer coating on the TZM punch is mostly preserved after 4

sintering cycles.

All the observed results indicate that the multilayering structure of the CrAlO/CrAlN

coating system produces a superior mechanical and tribological performance over the

CrAlO at room and at high temperatures. In fact the benefit of multilayering has been

well reported in the literature [40-43]. The presence of various dissimilar material

interfaces in a coating system can block the propagation of cracks at the interface of the

tougher part [40]. The alternation of two different layer also forces the re-nucleation

during the film growth allowing to obtain more dense and compact growth [42]. The

multilayering is an effective strategy for the blockage of pinholes formation through to

the substrate [43], thus hindering the oxidation kinetics and the consequent degradation

of the coatings and substrate system [42].

25
 ACCEPTED MANUSCRIPT

4. Conclusions

Monolayer CrAlO and multilayer CrAlO/CrAlN coatings have been deposited by

industrial-scale cathodic arc evaporation system in N2/O2 atmosphere. The CrAlO

exhibited an undetermined volume fraction mixture of cubic and corundum phases of

stoichiometry close to (Cr,Al)2O3. The multilayer showed the prevalence of the cubic

structure of the CrAlN and some volume fractions of the cubic and corundum oxide

phases from the oxide sublayers. The structural, mechanical and tribological results

evidence a strong dependence on the film structure. The multilayer CrAlO/CrAlN

coating exhibited higher mechanical stability after thermal annealing at 1000ºC and

1100ºC than these of the monolayer. The multilayer retained the high hardness up to

1100ºC of annealing, whereas it did not show hints of structural degradation until

1100ºC. In addition, the multilayer has been also shown to enhance wear resistance

performance of the monolayer CrAlO at RT and at 400ºC.

In the sintering trials, both CrAlSiN monolayer and CrAlO/CrAlN multilayer coating

have been proved to be adequate to coat punches used for the sintering of Ti90Sn10

allowed powders. On the other hand, the CrAlO/CrAlN multilayer coating qualitatively

showed protection properties to the TZM punch tool during the sintering of Al2O3, with

respect to that show by the CrAlSiN coating.

26
 ACCEPTED MANUSCRIPT

Acknowledgements

We acknowledge the financial support of the European Commission through the FP7

Grant Agreement 608720 (Micro-FAST), and the MINECO (Spain) in the frame of the

PROTEOX project (MAT2013-45391-P). The authors want to thank Alberto Colella

and Prof. P. Matteazzi from MBN nanomaterialia S.r.l. for the supply of the Ti90Sn10

and Al2O3 powders.

27
 ACCEPTED MANUSCRIPT

References

[1] D. Lu, Y. Yang, Y. Qin, and G. Yang, “Forming microgears by micro-FAST

technology,” J. Microelectromechanical Syst., vol. 22, no. 3, pp. 708–715, 2013.

[2] H. Hijji, Y. Qin, K. Huang, M. Bin Zulkipli, S. Yang, and J. Zhao, “Forming

alumina (Al2O3) by micro-FAST,” in Advances in Manufacturing Technology

XXX, Y. M. Goh and K. Case, Eds. 2016, pp. 61–66.

[3] Z. A. Munir, D. V. Quach, and M. Ohyanagi, “Electric current activation of

sintering: A review of the pulsed electric current sintering process,” Journal of

the American Ceramic Society, vol. 94, no. 1. pp. 1–19, 2011.

[4] M. E. Cura et al., “Friction behavior of alumina/molybdenum composites and

formation of MoO3-x phase at 400 ºC,” Tribol. Int., vol. 87, pp. 23–31, 2015.

[5] S. P. Chakraborty, S. Banerjee, K. Singh, I. G. Sharma, A. K. Grover, and A. K.

Suri, “Studies on the development of protective coating on TZM alloy and its

subsequent characterization,” J. Mater. Process. Technol., vol. 207, no. 1–3, pp.

240–247, 2008.

[6] Z. Zhang, X. Li, E. Almandoz, G. G. Fuentes, and H. Dong, “Sliding friction and

wear behaviour of Titanium-Zirconium-Molybdenum (TZM) alloy against Al2O3

and Si3N4 balls under several environments and temperatures,” Tribol. Int., vol.

110, 2017.

[7] J. Lin, J. J. Moore, J. Wang, and W. D. Sproul, “High temperature oxidation

behavior of CrN/AlN superlattice films,” Thin Solid Films, vol. 519, no. 8, pp.

2402–2408, 2011.

[8] J. Musil, “Hard nanocomposite coatings: Thermal stability, oxidation resistance

and toughness,” Surf. Coatings Technol., vol. 207, pp. 50–65, 2012.

[9] Y. P. Feng, L. Zhang, R. X. Ke, Q. L. Wan, Z. Wang, and Z. H. Lu, “Thermal

28
 ACCEPTED MANUSCRIPT

stability and oxidation behavior of AlTiN, AlCrN and AlCrSiWN coatings,” Int.

J. Refract. Met. Hard Mater., vol. 43, pp. 241–249, 2014.

[10] M. Witthaut, R. Cremer, A. von Richthofen, and D. Neuschütz, “Improvement of

the oxidation behavior of Ti1-xAlxN hard coatings by optimization of the Ti/Al

ratio,” Fresenius. J. Anal. Chem., vol. 361, no. 6, pp. 639–641, 1998.

[11] Y. C. Chim, X. Z. Ding, X. T. Zeng, and S. Zhang, “Oxidation resistance of TiN,

CrN, TiAlN and CrAlN coatings deposited by lateral rotating cathode arc,” Thin

Solid Films, vol. 517, no. 17, pp. 4845–4849, 2009.

[12] D. Chaliampalias et al., “Compositionally gradient PVD CrAlSiN films:

structural examination and oxidation resistance,” Surf. Eng., vol. 844, no. July,

pp. 1–7, 2016.

[13] K. Bobzin, “High-performance coatings for cutting tools,” CIRP J. Manuf. Sci.

Technol., no. 2016, 2016.

[14] B. P. Dhonge et al., “Wear and oxidation resistance of combustion CVD grown

alumina films,” Surf. Coatings Technol., vol. 206, no. 22, pp. 4574–4579, 2012.

[15] K. Pedersen, J. Bøttiger, M. Sridharan, M. Sillassen, and P. Eklund, “Texture and

microstructure of Cr2O3 and (Cr,Al)2O3 thin films deposited by reactive

inductively coupled plasma magnetron sputtering,” Thin Solid Films, vol. 518,

no. 15, pp. 4294–4298, 2010.

[16] L. de Abreu Vieira et al., “Approaches to influence the microstructure and the

properties of Al-Cr-O layers synthesized by cathodic arc evaporation,” Surf.

Coatings Technol., vol. 204, no. 11, pp. 1722–1728, 2010.

[17] M. Pohler, R. Franz, J. Ramm, P. Polcik, and C. Mitterer, “Cathodic arc

deposition of (Al,Cr)2O3: Macroparticles and cathode surface modifications,”

Surf. Coatings Technol., vol. 206, no. 6, pp. 1454–1460, 2011.

29
 ACCEPTED MANUSCRIPT

[18] A. Khatibi, A. Genvad, E. Göthelid, J. Jensen, P. Eklund, and L. Hultman,

“Structural and mechanical properties of corundum and cubic (AlxCr1-x)2+yO3-y

coatings grown by reactive cathodic arc evaporation in as-deposited and annealed

states,” Acta Mater., vol. 61, no. 13, pp. 4811–4822, 2013.

[19] M. Pohler, R. Franz, J. Ramm, P. Polcik, and C. Mitterer, “Influence of pulsed

bias duty cycle variations on structural and mechanical properties of arc

evaporated (Al,Cr)2O3 coatings,” Surf. Coatings Technol., vol. 282, pp. 43–51,

2015.

[20] J. Nohava, P. Dessarzin, P. Karvankova, and M. Morstein, “Characterization of

tribological behavior and wear mechanisms of novel oxynitride PVD coatings

designed for applications at high temperatures,” Tribol. Int., vol. 81, pp. 231–

239, 2015.

[21] H. Najafi, A. Karimi, P. Dessarzin, and M. Morstein, “Correlation between

anionic substitution and structural properties in AlCr(OxN1-x) coatings deposited

by lateral rotating cathode arc PVD,” Thin Solid Films, vol. 520, no. 5, pp. 1597–

1602, 2011.

[22] M. Hirai, H. Saito, T. Suzuki, H. Suematsu, W. Jiang, and K. Yatsui, “Oxidation

behavior of Cr-Al-N-O thin films prepared by pulsed laser deposition,” in Thin

Solid Films, 2002, vol. 407, no. 1–2, pp. 122–125.

[23] D. Diechle, M. Stueber, H. Leiste, S. Ulrich, and V. Schier, “Combinatorial

approach to the growth of alpha-(Al1-x,Crx)2O3 solid solution strengthened thin

films by reactive r.f. magnetron sputtering,” Surf. Coatings Technol., vol. 204,

no. 20, pp. 3258–3264, 2010.

[24] H. Najafi, A. Karimi, P. Dessarzin, and M. Morstein, “Formation of cubic

structured (Al1-xCrx)2+δO3 and its dynamic transition to corundum phase during

30
 ACCEPTED MANUSCRIPT

cathodic arc evaporation,” Surf. Coatings Technol., vol. 214, pp. 46–52, 2013.

[25] J. Ramm, M. Ante, T. Bachmann, B. Widrig, H. Brändle, and M. Döbeli, “Pulse

enhanced electron emission (P3eTM) arc evaporation and the synthesis of wear

resistant Al-Cr-O coatings in corundum structure,” Surf. Coatings Technol., vol.

202, no. 4–7, pp. 876–883, 2007.

[26] V. Edlmayr, M. Pohler, I. Letofsky-Papst, and C. Mitterer, “Microstructure and

thermal stability of corundum-type (Al0.5Cr0.5)2O3 solid solution coatings grown

by cathodic arc evaporation,” Thin Solid Films, vol. 534, pp. 373–379, 2013.

[27] U. Helmersson, S. Todorova, S. A. Barnett, J. E. Sundgren, L. C. Markert, and J.

E. Greene, “Growth of single-crystal TiN/VN strained-layer superlattices with

extremely high mechanical hardness,” J. Appl. Phys., vol. 62, no. 2, pp. 481–484,

1987.

[28] R. Raab, C. M. Koller, S. Kolozsvári, J. Ramm, and P. H. Mayrhofer, “Interfaces

in arc evaporated Al-Cr-N/Al-Cr-O multilayers and their impact on hardness,”

Surf. Coatings Technol., 324, pp. 236–242, 2017.

[29] J. Fernández de Ara, E. Almandoz, J. F. Palacio, G. G. Fuentes, R. J. Rodríguez,

and J. A. García, “Influence of temperature in arc-activated plasma nitriding of

maraging steel in solution annealed and aged conditions,” Surf. Coatings

Technol., vol. 258, pp. 754–762, 2014.

[30] C. Oliver and M. Pharr, “An improved technique for determining hardness and

elastic modulus using load and displacement sensing indentation experiments,” J.

Mater. Res., vol. 7, no. 11, pp. 1564–1583, 1992.

[31] R. Ji, “Development and Characterisation of Nano-multilayer CrAlSiN Coating

Systems for Cutting Tools.”

[32] A. Khatibi, J. Sjölen, G. Greczynski, J. Jensen, P. Eklund, and L. Hultman,

31
 ACCEPTED MANUSCRIPT

“Structural and mechanical properties of Cr-Al-O-N thin films grown by cathodic

arc deposition,” Acta Mater., vol. 60, no. 19, pp. 6494–6507, 2012.

[33] A. Khatibi et al., “Face-centered cubic (Al1−xCrx)2O3,” Thin Solid Films, 519, pp.

2426–2429, 2011.

[34] M. Antonov, I. Hussainova, F. Sergejev, P. Kulu, and A. Gregor, “Assessment of

gradient and nanogradient PVD coatings behaviour under erosive, abrasive and

impact wear conditions,” Wear, vol. 267, no. 5–8, pp. 898–906, 2009.

[35] R. Bayón et al., “Corrosion-wear behaviour of PVD Cr/CrN multilayer coatings

for gear applications,” Tribol. Int., vol. 42, no. 4, pp. 591–599, 2009.

[36] J. Sjölén, L. Karlsson, S. Braun, R. Murdey, A. Hörling, and L. Hultman,

“Structure and mechanical properties of arc evaporated Ti-Al-O-N thin films,”

Surf. Coatings Technol., vol. 201, no. 14, pp. 6392–6403, 2007.

[37] J. L. Endrino and V. Derflinger, “The influence of alloying elements on the phase

stability and mechanical properties of AlCrN coatings,” Surf. Coatings Technol.,

vol. 200, no. 1–4 SPEC. ISS., pp. 988–992, 2005.

[38] Tomas Polcar, Albano Cavaleiro. High-temperature tribological properties of

CrAlN, CrAlSiN and AlCrSiN coatings. Surface & Coatings Technology 206

(2011) 1244–1251

[39] Tien, S.-K., Lin, C.-H., Tsai, Y.-Z., Duh, J.-G. Microstructural evaluation of

high oxidation resistant CrAlSiN hard coating at elevated temperature in air

atmosphere. Key Engineering Materials. 373-374 (2008) 446-451

[40] J. Musil, F. Kunc, H. Zeman, and H. Poláková, “Relationships between hardness,

Young’s modulus and elastic recovery in hard nanocomposite coatings,” Surf.

Coatings Technol., vol. 154, no. 2–3, pp. 304–313, 2002.

[41] C. H. R. V. Kumar, P. K. Nair, and B. Ramamoorthy, “Characterization of

32
 ACCEPTED MANUSCRIPT

multilayer pvd nanocoatings deposited on tungsten carbide cutting tools,” Int. J.

Adv. Manuf. Technol., vol. 38, no. 5–6, pp. 622–629, 2008.

[42] H. Riedl, E. Aschauer, C. M. Koller, P. Polcik, M. Arndt, and P. H. Mayrhofer,

“Ti-Al-N/Mo-Si-B multilayers: An architectural arrangement for high

temperature oxidation resistant hard coatings,” Surf. Coatings Technol., vol. 328,

pp. 80–88, 2017.

[43] W. Y. Ho, C. H. Shen, C. L. Chang, and D. Y. Wang, “Corrosion behaviors of

Cr(N,O)/CrN multi-layered coatings by cathodic arc deposition,” Surf. Coatings

Technol., vol. 202, no. 4–7, pp. 745–749, 2007.

33
 ACCEPTED MANUSCRIPT

Figure captions

Figure 1: GD-OES analyses of the a) monolayer CrAlO coating, and b) multilayer

CrAlO/CrAlN coating.

Figure 2: SEM images of the cross-section of as-deposited coatings on Si waflers, a)

monolayer CrAlO, b) multilayer CrAlO/CrAlN.

Figure 3: GI-XRD diffractograms of both untreated coatings.

Figure 4: SEM images of the samples’ surface untreated and after high temperature

annealing tests.

Figure 5: PVD coated TZM punches and graphite die assembly after sintering a)

Ti90Sn10 powders at 1100 ºC and b) Al2O3 powders at 1300 ºC. The punch on the left

of both photos is the CrAlO/CrAlN coated and the one in the right the CrAlSiN coated.

Figure 6: Cutaway views of the TZM punches after 4 sinterings a) CrAlSiN coated and

b) CrAlO/CrAlN coated.

34
 ACCEPTED MANUSCRIPT

Table captions

Table 1: Experimental parameters of the deposition processes.

Table 2: Roughness (Ra and Rq) and thickness of the monolayer and multilayer

coatings.

Table 3: Hardness, Young Modulus and H3/E’2 values of the CrAlO monolayer coating

and of the CrAlO/CrAlN multilayer coating in the untreated state and after the high

temperature resistance tests at 1000 and 1100 ºC.

Table 4: Wear rate (in m3/Nm) at room temperature, 200 ºC and 400 ºC.

Table 5: EDX results on the surface of Inconel samples coated with the monolayer and

the multilayer coatings affter the high temperature resistance tests at 1100 ºC.

35
 ACCEPTED MANUSCRIPT

Highlights

• Deposition of CrAlO monolayer and CrAlO/CrAlN multilayer coatings by

cathodic arc.

• Protection of TZM alloys in FAST sintering applications.

• Excellent wear and high temperature resistance.