coatings

Article
Microstructure and Wear of (CrN/CrAlN)/(CrAlN/VN)
and (CrN/TiAlN)/(TiAlN/VN) Coatings for Molds
Used in High Pressure Casting of Aluminum
Aneta Wilczek 1 , Jerzy Morgiel 2, *, Łukasz Rogal 2 , Wojciech Maziarz 2 and Jerzy Smolik 3
1 Limatherm S.A., 34-600 Limanowa, Poland; wilczek.a@limatherm.pl
2 Institute of Metallurgy and Materials Science, Polish Academy of Sciences, 30-059 Krakow, Poland;
l.rogal@imim.pl (Ł.R.); w.maziarz@imim.pl (W.M.)
3 Institute for Sustainable Technologies, Łukasiewicz Research Network, 26-600 Radom, Poland;
jerzy.smolik@itee.radom.pl
* Correspondence: j.morgiel@imim.pl




Received: 5 February 2020; Accepted: 9 March 2020; Published: 11 March 2020


Abstract: Molds made of tool steels used in aluminum high-pressure die casting should routinely
withstand tens of thousands of injection cycles, but repeated loading and temperature spikes result in
their frequent premature wear. Extending their lifetime could be sought by nitriding or application of
coatings of even higher hardness or both. Therefore, in the present experiment the arc-deposited
Cr/(CrN)/nx (CrN/CrAlN)/mx (CrAlN/VN) or Cr/(CrN)/nx (CrN/TiAlN)/mx (TiAlN/VN) nano-multilayer
stacks were deposited on glow discharge nitrided X40CrMoV5.1 steel. The scanning and transmission
electron microscopy backed by Energy Dispersive X-ray Spectroscopy measurements of local chemical
composition helped to confirm that the coatings are built of nanolayers of respective nitrides of period
less than 10 nm. They also showed that droplets being characteristic for arc deposition method were
enriched either in chromium, aluminum or vanadium but not in titanium. Both coatings presented
comparable hardness of ~25 GPa, but the one covered with TiAlN/VN was roughly twice as wear
resistant as the CrAlN/VN. Simultaneously, they were ~200 and ~100 more wear resistant than
X40CrMoV5.1reference steel.

Keywords: hybrid surface treatment; nano-multilayers; microstructure; wear resistance

1. Introduction
Molds used for high pressure casting of aluminum are made of tool steel, as they have to withstand
long-term loading and temperature cycling. These harsh working conditions cause formation of local
fissures and voids, both increasing in number and growing in size with time [1]. During the casting
process, all the mold defects are inherent to the produced parts adding to their surface roughness,
which eventually exceeds a level acceptable for customers. The latter situation enforces immediate
production breaks, which can be restarted only after mold refurbishing or substitution with a new one.
Any of these two actions is both a time-consuming and expensive operation. That is why the foundries
are interested in development of molds capable of withstanding significantly longer production runs
than those being presently in use.
The extension of steel molds’ lifetime was at first sought by carburizing, nitriding or carbo-nitriding
their surface, producing a relatively thick diffusive zone [2,3]. All these treatments turned out to be
only partly successful as they raise not only the surface hardness, but also the harmful tensile strain
level, accelerating development of a net of surface cracks. The experiments from other related areas,
i.e., covering forging steel dies with PVD TiN/Ti(N,C)/TiC coatings, indicated that proper adjustment
of interlayer thickness and chemical composition allows to tailor the surface stresses, improving

Coatings 2020, 10, 261; doi:10.3390/coatings10030261 www.mdpi.com/journal/coatings

Coatings 2020, 10, 261 2 of 11

the wear resistance as compared with that of the base material [4]. However, due to much lower
thickness of Physical Vapor Deposition (PVD) coatings as compared with diffusive zones formed
during carburizing or nitriding, the gains were generally on a par. Later, it was found out that
a hybrid treatment merging nitriding of the substrate with the following deposition of PVD CrN
monolithic coatings with a Cr buffer layer is the most efficient way to extend the lifetime of hot forging
tools [5]. Recently, it was shown that the development of even better die protection could be achieved
by deposition of nano-multilayer surface coatings with a combination of double or ternary nitrides
like CrVN/TiN [6] or Cr/CrN/(CrN/CrAlN)/(CrAlN/VN) [7]. This idea was based on the assumption
that such nano-multilayers offer not only higher hardness than the monolithic coatings, but also
secure some self-lubricating capabilities due to the presence of vanadium nitride. The topmost VN
nanolayers would be eventually exposed to air at higher temperatures resulting in their reaction with
oxygen and formation of shearable vanadium oxides, helping to relieve surface stresses and serving as
self-lubricating agent.
Protection of molds with such advanced hybrid coatings is a complicated process especially
due to the problems with the deposition of PVD nano-multilayers of precisely controlled thickness
and chemical composition. Additionally, the experiences from large-scale industrially produced TiN
coated drills and similar indicated that significant extension of their lifetime was achieved as long
as deposited coatings were at least several micrometers thick [8]. The latter requirement is hard to
fulfill in practice with most PVD techniques, except magnetron or arc sputtering. The effectiveness of
the arc technique is the highest, but on the other hand marred by its tendency for droplet deposition
of diameter exceeding many times the nanolayers period [9,10]. The latter situation leads to local
deformation of the nanolayers geometry compromising the coatings quality. Additionally, both of
these methods are liable to intermixing of deposited materials and producing multilayers of chemical
composition being far from the nominal values. Therefore, independently of the PVD method used
for deposition of hybrid coatings, their development should be based on detailed microstructure
characterization serving as a feedback for the proper control of this process.
The aim of the presently described experiment was to estimate the effectiveness of a
protection of the heat resistant steel through the hybrid treatment including substrate nitriding
followed by arc deposition of Cr/(CrN)/nx (CrN/CrAlN)/mx (CrAlN/VN) or Cr/(CrN)/nx (CrN/TiAlN)
/mx (TiAlN/VN) nano-multilayer coatings. It was realized through detailed scanning and transmission
electron microscopy (SEM/TEM) microstructure characterization backed by hardness, adhesion and
wear measurements.

2. Materials and Methods

The DIN 1.2344, EN X40CrMoV5.1 tool steel containing (in wt. %): C–0.39, Si–1.10, Mn–0.40,
Cr–5.20, Mo–1.40, V–0.95, Fe–balance was used as a substrate for the developed coatings. The steel
substrate was hot rolled at approximately 1050 ◦ C, then annealed at 840 ◦ C for 4 h and cooled with
a furnace. Next, the complex heat treatment consisting of austenitization at 950 ◦ C for 0.5 h and
tempering at 550 ◦ C for 2 h were conducted. The glow discharge nitriding was performed at 520 ◦ C
(Ubias = 560–580 V, Ibias = 7–9 A). Two types of coatings, i.e., Cr/(CrN)/nx (CrN/CrAlN)/mx (CrAlN/VN)
and Cr/(CrN)/nx (CrN/CrTiN)/ mx (CrTiN-VN), assigned as P1 and P2, respectively, were deposited
with a MZ383 Metaplas Ionon system. It consists of the vacuum chamber equipped with seven arc
sources arranged as shown in Figure 1. In the case of Cr/(CrN)/nx (CrN-TiAlN)/mx (TiAlN/VN) coatings,
the Cr70Al30 targets were substituted with Ti70Al30 ones. Substrates were fixed to rotating table in a
way that their surfaces were located opposite to the arc targets.
 Coatings 2020,
2019, 10,
9, x261
FOR PEER REVIEW 33of
of 12
11
Coatings 2019, 9, x FOR PEER REVIEW 3 of 12

Figure 1. Schematic presentation of location of Cr, V and Cr70Al30 targets within the vacuum
Figure 1.
Figure
chamber Schematic
1.ofSchematic presentation of of
presentation
the arc deposition location
system. of Cr,
location of VCr,
andVCr70Al30 targets targets
and Cr70Al30 within the vacuum
within the chamber
vacuum
of the arc deposition system.
chamber of the arc deposition system.
Deposition of the chromium layer was realized by activating the arc on all the chromium targets
Deposition of
Deposition of the
the chromium
chromium layer
layer was
was realized
realized by
by activating
activating the arc
arc on
on all
all the
the chromium
chromium targets
targets
under argon atmosphere, while chromium nitride was obtained bythe substituting argon with nitrogen.
under
under argon atmosphere,
argon atmosphere, while chromium nitride was obtained by substituting argon with nitrogen.
The CrN-CrAlN stack waswhile chromium
obtained nitride was
by activating two obtained
Cr and two by substituting argonunder
Cr70Al30 targets with nitrogen.
nitrogen
The CrN-CrAlN
The CrN-CrAlN stack was
stack was obtained by activating two Cr and two Cr70Al30 targets under nitrogen
atmosphere and rotating theobtained
substratesbyat
activating two Cr and
5 rpm. Deposition of two Cr70Al30
VN-CrAlN wastargets under
carried out innitrogen
similar
atmosphere and
atmosphere and rotating
rotating the
the substrates
substrates at
at 55 rpm.
rpm. Deposition of
Deposition of VN-CrAlN
VN-CrAlN waswas carried
carried out
out in
in similar
similar
way, i.e., by staying with activated discharge at Cr70Al30 targets and simultaneously switching arcs
way, i.e., by
way, chromium staying
i.e., by staying with activated
with activated discharge
discharge at Cr70Al30
at Cr70Al30 targets
targets and simultaneously
and simultaneously switching arcs
from to vanadium targets. Parameters of the realized deposition process are switching arcs
given in Table
from chromium to
to vanadium targets.
targets. Parameters
Parametersof ofthe
therealized
realizeddeposition
depositionprocess
processare
aregiven
givenininTable
Table1
1from chromium
(more details are vanadium
given in [7]).
(more details are given in [7]).
1 (more details are given in [7]).
Table 1. Deposition parameters of Cr/nxx(CrN)/(CrN/CrAlN)/mxx(CrAlN/VN) multilayers.
Table 1. Deposition parameters of Cr/n (CrN)/(CrN/CrAlN)/m (CrAlN/VN) multilayers.
Table 1. Deposition parameters of Cr/nx(CrN)/(CrN/CrAlN)/mx(CrAlN/VN) multilayers.
Target Substrate Substrate Deposition
Deposited
Deposited Process Pressure
Process Pressure ActiveActive Target Current Substrate Bias Substrate Deposition
Material Atmosphere (mbar) Targets IArc
Target UBIAS
Current
(A)
Substrate
(V)
Substrate
Bias Temperature
Temperature
T (◦ C) Time
Deposition
Time
t (min)
Deposited
Material Process
Atmosphere Pressure
(mbar) Active
Targets
−3 Current
IArc (A) Bias
UBIAS (V) Temperature
T (°C) t Time
(min)
Cr
Material 100% Ar
Atmosphere (mbar) 3 × CrTargets
5.0 × 10 ~50 10
Cr CrN 100% Ar 5.0 × 10 3 × Cr 3 × Cr
−3 IArc (A) UBIAS~50 (V) T (°C) t (min)
10
55 400 22
Cr
CrN 100%NAr
100% 2 3.5 × 5.0
10 −2× 10 −3
3 × Cr ~150 ~50 10
22
CrN 2 × Cr
CrN 33 × Cr 30 22
/CrAlN 2 × Cr70 Al2
CrN /CrAlN 55 400 30
CrAlN 100% N2 ×23Cr
3.5 × 102−2× Cr702Al × Cr
70Al3 ~150
CrN /CrAlN
/VN
55 400 54 30
100% N2 3.5 × 10−2 2 × 2V × Cr70 70Al33 ~150
CrAlN /VN 54
2 ×2Cr VAl3
× 70
CrAlN /VN 54
An arrangement of the buffer layers 2 ×and
V a sequence of the stacks of respective multilayers for
An arrangement
the Cr/(CrN)/n of the buffer
x (CrN/CrAlN)/m layers and coating
x (CrAlN-VN) a sequence of the stacks
is presented of respective
in Figure 2 (in themultilayers for
case of second
the An arrangement
Cr/(CrN)/n of the
x(CrN/CrAlN)/mbuffer layers
x(CrAlN-VN) and a sequence
coating is of the
presented
coating the CrAlN nanolayers were substituted with CrTiN ones). stacks
in of
Figurerespective
2 (in the multilayers
case of for
second
the Cr/(CrN)/n x(CrN/CrAlN)/mx(CrAlN-VN) coating is presented in Figure 2 (in the case of second
coating the CrAlN nanolayers were substituted with CrTiN ones).
coating the CrAlN nanolayers were substituted with CrTiN ones).

Figure 2.
Figure Scheme of
2. Scheme of multilayers
multilayers arrangement
arrangement in Cr/(CrN)/nxx(CrN/CrAlN)/m
in Cr/(CrN)/n (CrN/CrAlN)/mxx(CrAlN/VN)
(CrAlN/VN) coating.
coating.
Figure 2. Scheme of multilayers arrangement in Cr/(CrN)/n (CrN/CrAlN)/m (CrAlN/VN) coating.
x x
Coatings 2020, 10, 261 4 of 11

Coatings 2019, 9, x FOR PEER REVIEW 4 of 12

Measurements of hardness and Young modulus of the multilayer coating were performed
usingMeasurements
Berkovich indenter of hardness
on NHT andnanotester
Young modulus produced of the multilayer
by CSM coating(Peuseux,
Instruments were performed using
Switzerland).
Berkovich indenter on NHT nanotester produced by CSM Instruments
Indentation depth was limited to ~300 nm (<30% of the coating thickness) with the aim to eliminate the (Peuseux, Switzerland).
Indentation depth wasThe
effect of the substrate. limited to ~300 nm
final hardness was (<30%
takenofasthe
an coating
average thickness)
of up to 15with the aim
individual to eliminate
measurements.
the effect was
Adhesion of the substrate.
assessed in the The final test
scratch hardness was taken
with Rockwell as an average
penetrator (Peuseux, of up to 15 individual
Switzerland) on the
measurements. Adhesion was assessed in the scratch test with
REVETEST CSM. The loading (from 0 to 200 N) was proportional to penetrator distance (∆F = 10 N/mm). Rockwell penetrator (Peuseux,
Switzerland) on the REVETEST
The start of cohesion cracks was CSM.markedThe loading (from first
as Fc1 , while 0 to 200 N) was proportional
delamination events were to marked
penetrator as
distance (ΔF = 10 N/mm). The start of cohesion cracks was marked
Fc2 . The force at which a total removal of the coating from the under the Rockwell penetrator takes as Fc 1 , while first delamination
events
place waswere marked
denoted as asFc3Fc 2. was
. It The estimated
force at which basing a total
on theremoval of the coating
sharp changes in the value fromof the under
both the
friction
Rockwell
coefficientpenetrator
and acoustic takes place was
emission denoted
as well as Fc3. It was
as observation of the estimated
indenter basing
line using on the sharp microscopy
scanning changes in
the value of both friction
(Hitachi TM3000, Tokyo, Japan) method. coefficient and acoustic emission as well as observation of the indenter line
usingThe
scanning microscopy (Hitachi TM3000, Tokyo, Japan) method.
coatings’ microstructure was characterized using XL30 (30 kV) FEI scanning electron
The coatings’
microscope (SEM) (FEI, microstructure was characterized
Eindhoven, Netherlands) usingG2XL30
and Tecnai (30 kV)
F20 200kV FEG FEI scanning electron
transmission electron
microscope (SEM) (FEI, Eindhoven, Netherlands) and Tecnai
microscope (Eindhoven, Netherlands). The latter was equipped with Fischione High Angle Annular G2 F20 200kV FEG transmission
electron
Dark Field microscope
(HAADF)(Eindhoven,detector for STEM Netherlands). The latter
observations. was chemical
The local equippedcomposition
with Fischione wasHigh Angle
determined
Annular Dark Field (HAADF) detector for STEM observations.
with an integrated EDAX Energy Dispersive X-ray Spectroscope (EDS) (Berwyn, PA, USA). Thin The local chemical composition was
determined with an integrated EDAX Energy Dispersive X-ray
lamellas for TEM observations were prepared using FEI Quanta 200 3D Dual Beam Focused Ion Beam Spectroscope (EDS) (Berwyn, PA,
USA). Thin lamellas
(FIB) system. It involved for covering
TEM observations
coatings with were prepared
a ~300 using FEI
nm amorphous Quanta
carbon layer200 3D Dual
serving as a Beam
buffer
Focused Ion Beam (FIB) system. It involved covering coatings
between them and platinum protecting stripe (1 mm wide, ~up to 20 mm long) was deposited on top with a ~300 nm amorphous carbon
layer serving
of them. Next,asalla material
buffer betweenat the stripethemsidesand wasplatinum
removed protecting
with thestripe
Ga+ ion(1 mm beam. wide,
Thus,~up thetoobtained
20 mm
long) was
platelet was deposited
lifted out onwith
top of thethem. Next, allattachment
Omniprobe material atand the welded
stripe sides was
to FIB removed
copper grid.with the Ga+
Finally, the
ion beam. Thus, the obtained platelet was lifted out with the Omniprobe
platelet was thinned down to ~100 nm with Ga+ ions producing a lamella presenting the multilayer attachment and welded to
FIB copper
coating in agrid. Finally, the platelet was thinned down to ~100 nm with Ga+ ions producing a lamella
cross-section.
presenting the multilayer coating in a cross-section.
3. Results
3. Results
The SEM image acquired in Back Scattered Electrons (BSE) mode presenting cross-section of the
hybridTheprocessed
SEM image toolacquired
steel, i.e., inaBack Scattered
nitrided Electronscovered
base material (BSE) modewith presenting
multilayer cross-section of the
protective coatings,
hybrid
showedprocessed
that depositiontool steel,of the i.e., a nitrided
Cr/CrN buffer base material
double layercovered
securedwith multilayer
formation protective coatings,
of a non-porous interface
showed
between that themdeposition
(Figure 3).ofThe theSEM/EDS
Cr/CrN buffer profilesdouble layer secured
documenting formation
distribution of a non-porous
of elements across the
interface
Cr/(CrN)/n x (CrN/CrAlN)/m
between them (Figure x (CrAlN/VN)
3). The coating
SEM/EDS profiles
confirms documenting
stepped decreasedistribution of elements
of Cr concentration from
across
bottom theto top Cr/(CrN)/n
as well as(CrN/CrAlN)/m
x
partial substitution x(CrAlN/VN)
of Al withcoating
V in theconfirms stepped
near surface decrease3a).
area (Figure of The
Cr
concentration
EDS profile obtained from bottom to top as well
from Cr/(CrN)/n x (CrN/TiAlN)/m x (TiAlN/VN)
as partial substitution of Alcoating
with Vshowsin the roughly
near surface area
the same
(Figure
stepped3a). The EDS
relation profile obtained
as it concerns from Cr/(CrN)/n
the diminishing x(CrN/TiAlN)/mx(TiAlN/VN) coating shows
concentration of Cr (raised Cr intensity is an artifact
roughly the same stepped
due to superposition of Cr-Kα relation
with V-Kβ as itlines)
concerns theasdiminishing
as well concentration
raising concentration of V inoftheCrnear-surface
(raised Cr
intensity
area (Figureis an3b).artifact
Thedue to superposition
surface of the multilayers of Cr-Kα waswith V-Ktolines)
found as well aswith
be decorated raising concentration
droplets of sizes
of V in the
starting from near-surface
tens of nanometersarea (Figure and 3b).
ending Theinsurface of therange.
micrometer multilayers was found to be decorated
with droplets of sizes starting from tens of nanometers and ending in micrometer range.

Figure 3. SEM/BSE images of cross-section and EDS profiles for P1 (a) and P2 (b) hybrid coatings.
Figure 3. SEM/BSE images of cross-section and EDS profiles for P1 (a) and P2 (b) hybrid coatings.
Coatings 2020, 10, 261 5 of 11

The 2019,
Coatings STEM/HAADF microstructure images helped to establish that coatings are built
9, x FOR PEER REVIEW 5 of 12 of
pseudo-columnar grains and the droplets observed previously on the coatings surface are also
present The STEM/HAADF
within them (Figuremicrostructure
4a,c). The EDS images
profileshelped to establish that
are characterized by coatings
loweringare built ofofpseudo-
number acquired
columnar grains and the droplets observed previously on the
counts when approaching the surface of the coating due to reduced thickness of the TEMcoatings surface are also present within in
lamellas
them (Figure 4a,c). The EDS profiles are characterized by lowering number
such areas (Figure 4b,d). The EDS signal acquired from the Cr/(CrN)/nx (CrN/CrAlN)/mx (CrAlN/VN) of acquired counts when
approaching the surface of the coating due to reduced thickness of the TEM lamellas in such areas
coating documented high chromium concentration in the first buffer layer and its lowering in
(Figure 4b,d). The EDS signal acquired from the Cr/(CrN)/nx(CrN/CrAlN)/mx(CrAlN/VN) coating
CrN and the following stacks of nitride/nitride multilayers (Figure 4b). In the case of the
documented high chromium concentration in the first buffer layer and its lowering in CrN and the
Cr/(CrN)/nx (CrN-TiAlN)/mx (TiAlN-VN) multilayer the situation is similar with an exception of
following stacks of nitride/nitride multilayers (Figure 4b). In the case of the Cr/(CrN)/nx(CrN-
much smallerx difference between counts from Cr and CrN caused by oversaturation of the EDS detector
TiAlN)/m (TiAlN-VN) multilayer the situation is similar with an exception of much smaller
due to large foil thickness. Simultaneously, the diminishing concentration of chromium is substituted
difference between counts from Cr and CrN caused by oversaturation of the EDS detector due to
firstlarge
by aluminum
foil thickness.starting at the middlethe
Simultaneously, of diminishing
the multilayer and, later, also
concentration by vanadium
of chromium located closer
is substituted first to
theby surface parts. The activation of the aluminum arc target is accompanied
aluminum starting at the middle of the multilayer and, later, also by vanadium located closer toby a strong local enrichment
of multilayer
the surface in this The
parts. element, i.e., its
activation of concentration
the aluminumrises above is
arc target theaccompanied
average for by theaCrN/CrAlN
strong localor
CrN/TiAlN
enrichment stack. However, ainswitching
of multilayer on of i.e.,
this element, Ti orits
V targets proceeds
concentration through
rises abovegradual approaching
the average for the to
theCrN/CrAlN
average concentration of the involved elements. As it took place during
or CrN/TiAlN stack. However, a switching on of Ti or V targets proceeds through SEM/EDS measurements,
thegradual
presence of a false chromium
approaching to the averagesignal within the of
concentration mostthe top TiAlN/VN
involved elements.results
As from
it tooka place
superposition
during
of vanadium and chromiumthe
SEM/EDS measurements, lines, i.e., V-Kβ
presence of a is taken
false for Cr-Kα
chromium (Figure
signal within 4d).theThemostSTEM/EDS
top TiAlN/VN profiles
results at
obtained from a superposition
higher of vanadium
spatial resolution from the and chromium lines,
near-surface i.e., V-K
area helped to is taken for Cr-Kα
differentiate (Figure
the multilayer
4d). TheofSTEM/EDS
character this coating profiles
using obtained
vanadium at signal
higher (Figure
spatial resolution from the near-surface
5). The complementary changesarea helped
of Al and Ti
to differentiate
concentration werethe multilayer
weaker but stillcharacter
evident, of this changes
while coating in using
Cr invanadium
that area weresignaltoo (Figure
noisy to 5).ascribe
The
them to any specific set of nanolayers (the same situation was with the signal from nitrogen, which in
complementary changes of Al and Ti concentration were weaker but still evident, while changes was
Cr in that
therefore area were
omitted fromtoobothnoisy to ascribe
graphs). The them to any
contrast specificseen
variation set of
in nanolayers
STEM/HAADF (the same
images situation
and the
was with
changes the signal signal
in STEM/EDS from nitrogen, which was
strength helped therefore the
to determine omitted from both
characteristic graphs). of
parameter The thecontrast
coatings
variation seen
as given in Table 2. in STEM/HAADF images and the changes in STEM/EDS signal strength helped to
determine the characteristic parameter of the coatings as given in Table 2.

Figure
Figure 4. Scanning
4. Scanning Transmission
Transmission ElectronElectron Microscopy/
Microscopy/High AngleHigh Angle
Annular DarkAnnular Dark Field
Field STEM/HAADFI
STEM/HAADFI mages of cross-section of all the P1 (a, b) and P2(c, d) hybrid coatings
mages of cross-section of all the P1 (a,b) and P2 (c,d) hybrid coatings with accompanying EDS profiles with
accompanying EDS profiles presenting distribution
presenting distribution of Cr, Al, Ti, V, N and Fe. of Cr, Al, Ti, V, N and Fe.
Coatings 2020, 10, 261 6 of 11
Coatings 2019, 9, x FOR PEER REVIEW 6 of 12

Figure STEM/HAADF
5. 5.
Figure STEM/HAADFimages
imagesofof cross-section microstructureofoftop
cross-section microstructure topofof
P1P1
(a,(a,b) and
b) and P2(c,d)
P2(c, hybrid
d) hybrid
coatings with
coatings accompanying
with accompanyingprofiles
profilespresenting
presenting distribution of Cr,
distribution of Cr, Al,
Al, Ti
Ti and
andV. V.

Table 2. P1 and P2 multilayer coatings characteristics measured from STEM/HAADF images and
Table 2. P1 and P2 multilayer coatings characteristics measured from STEM/HAADF images and EDS
EDS profiles.
profiles.
(P1) x x (CrAlN/VN) (nm)
(P1)Cr/CrN/n
Cr/CrN/nx(CrN/CrAlN)/m
(CrN/CrAlN)/mx(CrAlN/VN) (nm)
CrN/CrAlN CrN/CrTiN
CrN/CrTiN CrAlN/VN
CrAlN/VN CrAlN/VN
CrAlN/VN
Cr CrN
Cr CrN/CrAlN
CrN Stack
Stack period
Periodλλ Stack Stack Period λ Period λ
~120 ~700 ~1000 ~7 ~1700 ~5
~120 ~700 ~1000 ~7 ~1700 ~5
(P2) Cr/CrN/nx(CrN/TiAlN)/mx( TiAlN/VN) (nm)
(P2) Cr/CrN/nx (CrN/TiAlN)/m x (TiAlN/VN) (nm)
CrN/TiAlN TiAlN/VN TiAlN/VN
Cr CrN CrN/TiAlN Stack
Period λ
CrN/TiAlN CrN/TiAlN TiAlN/VNStack TiAlN/VNPeriod λ
Cr CrN
~100 ~700 ~1000 Stack ~6 λ
Period Stack~1600 Period λ ~6
~100 ~700 ~1000 ~6 ~1600 ~6
STEM/EDS maps with local distribution of coating forming elements indicated that the
deposition of droplets takes place during the growth of all successive parts of the coatings, i.e., they
STEM/EDS
bear Cr, Al, Ti maps
and Vwith
ones.local distribution
However, of coating
the biggest forming
droplets elements
were found to beindicated that the deposition
built of vanadium (Figure
of droplets
6). takes place during the growth of all successive parts of the coatings, i.e., they bear Cr, Al, Ti
and V ones. However, the biggest droplets were found to be built of vanadium (Figure 6).
The mechanical testing of deposited coatings helped to established that they both present nearly
the same hardness of 25.0 ÷ 26.0 GPa (Table 3). However, the P2 was characterized by ~10% higher
stiffness over the P1, i.e., with the average value of the Young modulus E of 400 and 370 GPa for the
former and the latter, respectively.

Table 3. Hardness, Young modulus and values of FC1 , FC2 and FC3 obtained for the investigated coatings.

hmax Fmax Hardness Young Modulus Fc1 (*) Fc2 (**) Fc3 (***)
Sample
(nm) (mN) H (GPa) E (GPa) (N) (N) (N)
P1 300 49 25.4 ± 1.2 373 ± 28 56 133 176
P2 300 51 25.6 ± 1.4 400 ± 37 64 95 160
* FC1 —load on indenter under which show up first cracks (start of coating decohesion). ** FC2 —load on indenter
under which starts locating delamination. *** FC3 —load on indenter under which all the coating is removed from
the scratch line
Coatings 2020, 10, 261 7 of 11
Coatings 2019, 9, x FOR PEER REVIEW 7 of 12

Figure
Figure 6. Images of
6. Images of cross-section
cross-section of
of P1
P1 (a)
(a) and
and P2
P2 (b)
(b) hybrid
hybrid coatings
coatings with
with accompanying
accompanying EDS
EDS maps
maps
presenting distribution of Cr, Al, Ti, V, N.
presenting distribution of Cr, Al, Ti, V, N.

Results of scratch
The mechanical testsofofdeposited
testing the P1 and P2 coatings
coatings helped to areestablished
presentedthat in the
theyFigure 7. Thenearly
both present SEM
observations of the path covered by Rockwell indenter helped to establish
the same hardness of 25.0 ÷ 26.0 GPa (Table 3). However, the P2 was characterized by ~10% higher that in both coatings the
first microcracks
stiffness over thedenoted as thethe
P1, i.e., with Fc1average
show up at a distance
value of the Youngcorresponding
modulus Etoof ~60
400N.andUp370to that
GPaloading,
for the
also the friction coefficient
former and the latter, respectively. remains practically negligible for both coatings. The measurements of
accompanying acoustic emission from the P2 coating gave couple of sound pulses clearly above the
noiseTable
level 3.
indicating
Hardness,thatYoungin this case the
modulus andvalue of Fc
values of 1Fcould
C1, FC2 be
andhalf
FC3 of this determined
obtained through SEM
for the investigated
observations.
coatings. Applying the force exceeding 60 N caused a development of series of evident adhesive
cracks accompanying by steadily raising friction coefficient as well as a series of acoustic emission
Young
pulses of medium hmaxstrength
Fmaxin both the P1and P2 coatings. Only after
Hardness Fc1(*)the load onFcthe
2(**) indenter reached
Fc3(***)
Sample Modulus
the values of ~130 (nm)N or(mN)~100 N for H the
(GPa)P1 and P2E coating, (N) (N) (N)
(GPa) respectively, the scratch path showed first
signsP1of removal of small49
300 parts of25.4
the±coatings
1.2 visibly,
373 ± 28both at the 56 bottom of the 133indenter path 176 and at
theirP2edges (the300 value of the
51 second critical
25.6 ± 1.4 force is marked
400 ± 37 as Fc ).
2 64 The above processes
95 continue
160 till
the 160–170 *F NC1at which a total removal of the coating from under the indenter
–load on indenter under which show up first cracks (start of coating decohesion) starts, followed by a
steep raise in friction force as well as multiple strong acoustic pulses.
**FC2–load on indenter under which starts locating delamination

***FC3–load on indenter under which all the coating is removed from the scratch line

Results of scratch tests of the P1 and P2 coatings are presented in the Figure 7. The SEM
observations of the path covered by Rockwell indenter helped to establish that in both coatings the
 for both coatings, even if this for the P2 coating was slightly higher than for P1 throughout most of
the test, i.e., ~0.7 against ~0.6, respectively. The wear track on reference nitrided WCL steel was not
only significantly wider after comparable times, but also bears the marks of deep groves along it. As
a consequence of the above, the friction coefficient measured for this sample was not only the highest
but also showed frequent instabilities. The quantitative assessment of the wear test helped to establish
Coatings 2020, 10, 261 8 of 11
that, as compared with the reference plasma nitrided tool steel, the P1 and P2 multilayer coatings
present ~100 times and ~200 times better wear resistance, respectively (Figure 9).

Figure 7. SEM images of paths results of scratch-test accompanied by graphs of friction coefficient,
friction force and acoustic emission in relation to applied load for P1 (a) and P2 (b) coatings.
Figure 7. SEM images of paths results of scratch-test accompanied by graphs of friction coefficient,
friction force and acoustic emission in relation to applied load for P1 (a) and P2 (b) coatings.
The wear test of the P1 and P2 multilayer coatings, done in a ball-on disc stand at ambient
temperature, showed that the tracks forming as a result of ball action have smooth edges, sides and
bottom (Figure 8). The brighter spots within the wear track represent droplets cut by the passing ball,
while the darker ones represent holes left by broken out droplets. The friction coefficients are similar
for both coatings, even if this for the P2 coating was slightly higher than for P1 throughout most of
the test, i.e., ~0.7 against ~0.6, respectively. The wear track on reference nitrided WCL steel was not
only significantly wider after comparable times, but also bears the marks of deep groves along it. As a
consequence of the above, the friction coefficient measured for this sample was not only the highest
but also showed frequent instabilities. The quantitative assessment of the wear test helped to establish
that, as compared with the reference plasma nitrided tool steel, the P1 and P2 multilayer coatings
Coatings 2019, ~100
present 9, x FOR PEER
times REVIEW
and ~200 times better wear resistance, respectively (Figure 9). 9 of 12

Figure 8. Wear track and friction coefficient measured for (a) P1, (b) P2 samples and (c) X40CrMoV5.1
Figure 8. Wear track and friction coefficient measured for (a) P1, (b) P2 samples and (c) X40CrMoV5.1
tool steel.
tool steel.
 Figure 8. Wear track and friction coefficient measured for (a) P1, (b) P2 samples and (c) X40CrMoV5.1
Coatings 2020, 10, 261 9 of 11
tool steel.

Figure 9. Wear resistance of P1, P2 multilayer coatings and reference nitrided tool steel.
Figure 9. Wear resistance of P1, P2 multilayer coatings and reference nitrided tool steel.
4. Discussion
4. Discussion
Reactive arc plasma sputtering of monolithic chromium, vanadium or sintered alloy targets like
CrAlReactive
or TiAl atarc plasma sputtering
substrates rotating inoffront
monolithic
of them, chromium, vanadium
as in the present or sintered
experiment, alloyus
allows targets like
to obtain
CrAl or TiAl
the stacks at substrates
of alternating rotating in front
nitride/nitride of them,
nanolayers ofas intotal
the the present
thickness experiment, allows us toplacing
of a few micrometers, obtain
the stacks of alternating nitride/nitride nanolayers of the total thickness of a
it among the most potent PVD methods. The high capability of this method in the deposition of few micrometers, placing
it among
hard the most
coatings waspotent PVD methods.
documented Thewith
in the past highthe
capability of this
production of method
nanoperiodin thenitride/metallic
deposition of hard
and
coatings was documented in the past with the production of nanoperiod
nitride/nitride multilayers by Wiecinski et al. [10] and Chen et al. [11], respectively. The so muchnitride/metallic and
nitride/nitride multilayers by Wiecinski et al. [10] and Chen et al. [11], respectively.
prized high deposition rate of the arc-plasma method is, however, tarnished by the unavoidable The so much
prized high
presence deposition rate
of droplets—as longofasthe
onearc-plasma
wants to staymethod
with is,
thehowever, tarnished
high deposition rateby the unavoidable
available with this
presence of droplets—as
method—exceeding long as one wants
the nano-multilayer to stay
period andwith the highofdeposition
intermixing the depositedrate material.
available with this
method—exceeding
The arc dropletsthe nano-multilayer
compromise period and intermixing
the multilayers’ of the deposited
geometry, increasing material.
the surface roughness as
The arc droplets compromise the multilayers’ geometry, increasing
well as overshadowing their surroundings and blocking of an in-flow of a new material the surface roughness as
in this area.
well as overshadowing their surroundings and blocking of an in-flow of a new
The latter situation leads to formation of voids, which under stress could act as crack nucleation material in this area.
The
sites.latter
The situation leads to formation
EDS measurements of voids,chemical
of nanolayers’ which under stress could
composition act astocrack
pointed nucleation
a strong sites.
intermixing
The EDS measurements
of material coming from of nanolayers’
different targetschemical composition
at their interfaces. pointed
It falls down to away
a strong
fromintermixing
them, but of in
the case of multilayer periods <10 nm it is still significant even in centers of the individual layers (as
documented in Figure 4). The intermixing experimentally determined through EDS measurements is
definitely strongly overestimated by unavoidable probe broadening in the specimen (~100 nm thick),
but more precise measurements with atom probe tomography of PVD CoFe/Cu/CoFe/Ni nanolayers [12]
indicated the presence of gradient composition extending up to 2 nanometers. Chang et al. [11] showed
that the fcc-nitride/fcc-nitride multilayer interfaces are at least locally coherent, what explains a raise
in hardness with the lowering of the multilayer period. However, the presence of intermixing of
deposited material relieving stresses at such boundaries places a natural limit on their period of ~10 nm,
below which no more gains in multilayer hardness are possible.
The local chemical composition, analyzed through a 2D mapping with EDS method, showed
that the arc temperature is so high that the discussed droplets are ejected from both elemental and
alloyed targets at approximately same rate independently from the strong difference between their
melting temperature. Additionally, switching on the CrAl or TiAl targets results in short (extending up
to 20 periods of CrN/CrAlN nanolayers) but relatively strong enrichment in aluminum of respective
alloyed nitride layers (50% above the nominal). It is significant that somewhat similar activation of still
another target later on, i.e., taking place during substitution of the Cr target with the V one, starting
with the deposition of VN/CrAlN nano-multilayer allows us to deposit the material of designed
chemical composition from the very beginning. The above may only mean that the CrAl or AlTi
Coatings 2020, 10, 261 10 of 11

targets compacted from the elemental powders carry higher surface content of the aluminum oxide.
During the start of the arc this surface material is the first to be transferred to the substrates causing the
observed local enrichments in aluminum or titanium. But in such a case these parts of multilayers
will have to be enriched in the oxygen. Its absence at the acquired spectra results just from the high
absorption of this relatively soft X-ray radiation by metallic atoms in the thick FIB specimen.
Hardness measurements of monolithic bi-metallic nitride CrAlN arc coatings were found to be
significantly (by ~30%) harder than CrN ones, i.e., achieving close to 23 GPa, but partial substitution of
Cr with V (being of high interest as its oxide forms an excellent solid lubricant) has practically no effect
on it [13]. Magnetron deposition of metal-nitride/metal-nitride stacks, like in case of the TiAlN/CrN
system as shown by Barshilia et al. in 2005 [14], turned out more effective than simple alloying of
chromium nitride, as they presented hardness as high as ~38 GPa for the multilayer period of ~7 nm.
Later experiments done by Chen et al. [9] showed that also the arc CrN/CrAlN/ (CrAlN/AlTiSiN)
multilayer coatings could achieve comparable high hardness (~35 GPa) as long as a proper bias voltage
sieving away at least part of droplets will be applied. Merging the knowledge from these experiments
opened a field for development of coatings with bi-metallic multilayer nitrides from the Cr-Al-Ti-V
system. As a consequence, the hardness of 38 GPa was achieved for AlCrN/TiVN (λ~14 nm) arc
multilayers [11]. The lower hardness (25–26 GPa) of the presently arc deposited stacks of nitrides with
CrAlN/VN or TiAlN/VN at their top (elaborated within the present experiment) means that even the
nano-multilayer engineering cannot make up for hardness loss due to introduction of relatively soft
VN nitride.
Wear of materials protected by the coatings mostly depends on their adhesion and hardness as
well as the type of wear debris. The latter could, under special conditions, serve as an additional
solid lubricant. Therefore, the progress in this area is gained through development of the coatings
characterized by improved micromechanical properties but also containing material lowering the
friction. In this application the most popular is vanadium, as at higher temperatures and in the presence
of oxygen from the atmosphere it turns into easily shearable vanadium oxide particles. Uchida et
al. [13] showed that the wear of softer monolithic CrVN coating on AiSi52100 steel is four times lower
(even more on cemented carbide substrates) than that of the much harder CrAlN. The first experiments
with CrAlN/VN multilayer coatings (λ~20 nm) brought less spectacular, but also significant gains in
wear resistance, i.e., improvement by nearly two times was achieved [15]. The following experiments
on arc AlCrN/TiN multilayers coatings (λ~14 nm) deposited on WC-Co substrates also gave roughly
two times higher wear resistance [11]. The combination of stacks of hard multilayer nitrides covered
with stack of only slightly softer but bearing mono-metallic VN layers (in both case with λ < 10 nm) on
nitrided X40CrMoV5 steel elaborated throughout the present experiment again helped to obtain very
high increase in wear resistance, i.e., from 100 to 200 times, over the nitrided WCL reference material.
The above success should be partly attributed to a deposition of vanadium carrying multilayers of
very low period what helped to preserve the overall hardness of this coatings exceeding those proper
for monolithic CrN. The other parameter contributing toward the low wear must have been the
nitriding of the WCL steel substrate, which gave the coating a proper support and adhesion through
the intermediate Cr/CrN bi-layer buffer.

5. Summary
This experiment proved that hybrid protection of tool steels in the form of its nitriding followed
by deposition of dedicated PVD nanolayered coatings helped to increase its wear resistance from 100 to
200 times as compared with the customary single step protection involving the former treatment. The
idea of a hybrid treatment comes from the observations that the newly developed nearly super-hard
PVD coatings could indeed withstand longer loading only in the case they are backed with a substrate
with a relatively high surface hardness like a nitrided one. Inserting the metallic/metallic-nitride
Cr/CrN bi-layer as buffer is a second step in building up gradient hardness support for the nearly
super-hard stack of metallic-nitride/metallic-nitride CrN/CrAlN or CrN/TiAlN nanolayered coatings.
Coatings 2020, 10, 261 11 of 11

Substituting CrN with softer VN in the upper stacks of multilayers causes a drop of harness, but as
the multilayer period will be kept <10 nm, the inter-layer interfaces will secure most of its previous
strength. Simultaneously, the presence of vanadium in the upper stack nullifies a negative effect of
the presence of arc-deposition inflicted macro-defects in the form of droplets, as they are immediately
sheared and transferred to wear-debris acting as solid lubricant.

Author Contributions: A.W. was involved in planning of the experiment and helped with the wear and hardness
measurements, as well as was co-writer of the manuscript. J.M., who performed the TEM investigations and
prepared first draft of manuscript. Ł.R. helped with the wear and hardness measurements and acted as a co-writer
of the manuscript. W.M. help with the TEM investigations as well as acted as co-writer of the manuscript. J.S. was
responsible for elaboration of the optimum nanolayers sequence and their deposition. All authors have read and
agreed to the published version of the manuscript.
Funding: The research was supported by Polish science financial resources The National Centre for Research and
Development, in frame of “Fast Path” program, project No. POIR.01.01.01-00-0890/17.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Hawryluk, M.; Gronostajski, Z.; Kaszuba, M.; Polak, S.; Widomski, P.; Smolik, J.; Ziemba, J. Analysis of the
wear of forging tools surface layer after hybrid surface treatment. Int. J. Mach. Tools Manuf. 2017, 114, 60–71.
[CrossRef]
2. Souchard, J.P.; Jacquot, P.; Buvron, M. Plasma overcarburizing of chromium steels for hot working and wear
applications. Mater. Sci. Eng. A 1991, 140, 454–460. [CrossRef]
3. Cao, H.; Liu, Y.; Shen, C.; Tang, H.; Huang, Y.; Lei, J. Erosion of mold-steel surface oxidation and nitriding
layers due to a die-casting aluminum alloy. Mater. Tech. 2018, 52, 619–624. [CrossRef]
4. Smolik, J.; Zdunek, K. Investigation of the influence of chemical composition of Ti(Cx N1−x ) layer on the stresses
value in the multilayer coating TiC/Ti(Cx N1−x )/TiN. Surf. Coat. Tech. 1999, 116–119, 398–403. [CrossRef]
5. Mazurkiewicz, A.; Smolik, J. Development of novel nano-structure functional coatings with the use of the
original hybrid device. Mater. Sci. Forum 2011, 674, 1–9. [CrossRef]
6. Contreras, E.; Galindez, Y.; Rodas, M.A.; Bejarno, G.; Gomez, M.A. CrVN/TiN nano-scale multilayer coatings
deposited by DC unbalanced magnetron sputtering. Surf. Coat. Tech. 2017, 332, 214–222. [CrossRef]
7. Smolik, J.; Mazurkiewicz, A.; Bujak, J.; Paćko, D.; Rogal, Ł. Analysis of fatigue strength of multi-layer coatings
type Cr/CrN/(CrN-Me1Me2N)/(Me1Me2N-VN). J. Mach. Constr. Main. 2019, 114, 31–38.
8. Zhang, S. TiN coating of tool steels: A review. J. Mater. Proc. Tech. 1993, 39, 165–177. [CrossRef]
9. Chen, W.; Lin, Y.; Zheng, J.; Zhang, S.; Liu, S.; Kwon, S.C. Preparation and characterization of CrAlN/TiAlSiN
nano-multilayers by cathodic vacuum arc. Surf. Coat. Tech. 2015, 265, 205–211. [CrossRef]
10. Wieciński, P.; Smolik, J.; Garbacz, H.; Kurzydłowski, K.J. Failure and deformation mechanisms during
indentation in nanostructured Cr/CrN multilayer coatings. Surf. Coat. Tech. 2014, 240, 23–31.
11. Chang, Y.; Weng, S.; Chen, C.; Fu, F. High temperature oxidation and cutting performance of AlCrN, TiVN
and multilayered AlCrN/TiVN hard coatings. Surf. Coat. Tech. 2017, 332, 494–503. [CrossRef]
12. Kelly, T.F.; Miller, M.K. Invited Review Article: Atom probe tomography. Rev. Sci. Instr. 2007, 78, 031101.
[CrossRef] [PubMed]
13. Uchida, M.; Nihira, N.; Mitsuo, A.; Toyoda, K.; Kubota, K.; Aizawa, T. Friction and wear properties of CrAlN
and CrVN films deposited by cathodic arc ion plating method. Surf. Coat. Tech. 2004, 177–178, 627–630.
[CrossRef]
14. Barshilia, H.C.; Prakash, M.S.; Jain, A.; Rajam, K.S. Structure, hardness and thermal stability of TiAlN and
nano-layered TiAlN/CrN multilayer films. Vacuum 2005, 77, 169–179. [CrossRef]
15. Qiu, Y.; Li, B.; Lee, J.; Zhao, D. Self-lubricating CrVN Coating Strengthened via multi-layering with VN. J.
Iron Steel Res. Inter. 2014, 21, 545–550. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).