Increasing the lifespan of high-pressure die cast molds subjected to severe wear

V. Nunes, F.J.G. Silva, M.F. Andrade, R. Alexandre, A.P.M. Baptista

PII: S0257-8972(17)30883-6
DOI: doi:10.1016/j.surfcoat.2017.05.098
Reference: SCT 22635

To appear in: Surface & Coatings Technology

Received date: 29 March 2017

Revised date: 27 May 2017
Accepted date: 28 May 2017

Please cite this article as: V. Nunes, F.J.G. Silva, M.F. Andrade, R. Alexandre, A.P.M.
Baptista, Increasing the lifespan of high-pressure die cast molds subjected to severe wear,
Surface & Coatings Technology (2017), doi:10.1016/j.surfcoat.2017.05.098

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

Increasing the Lifespan of High-Pressure Die Cast Molds Subjected to Severe

Wear

T
IP
V. Nunes 1,2, F. J. G. Silva1*, M. F. Andrade1, R. Alexandre3, A. P. M. Baptista4

R
SC
1ISEP – School of Engineering, Polytechnic of Porto

NU
Rua Dr. António Bernardino de Almeida, 431

4200-072 Porto, PORTUGAL

MA

2SONAFI, Sociedade Nacional de Fundição Injetada, S.A.

D

Rua Santos Dias, 1052

TE

4466-901 São Mamede Infesta, PORTUGAL

P
CE

3TEandM – Technology, Engineering and Materials, S.A.

AC

Parque Industrial do Taveiro, Lotes 41/42

3045-508 Taveiro, PORTUGAL

4INEGI – Instituto de Ciência e Inovação em Engenharia Mecânica e Engenharia Industrial

Rua Dr. Roberto Frias, Campus da FEUP

4200-465 Porto, PORTUGAL

 ACCEPTED MANUSCRIPT

Abstract

Regardless the increasingly incorporation of composite materials on vehicle components, high

T
pressure die casting still remains one of the most useful manufacturing techniques to obtain

IP
automotive parts with a complex shape in a cost effective way. It is well known that automotive

R
industry requires high production cadence as well as high products quality. Thus, systematic

SC
approaches are permanently being done leading to optimize all the production and management

aspects.

NU
The aluminum alloys commonly used in automotive parts such as fuel pumps bodies, throttle
MA
bodies, EGR valves, support brackets and so on usually contain Silicon which presents high

abrasively. The aluminum flow at high temperature and high speed into the mold induces severe
D

wear, sometimes due to a combination of abrasion and erosion effects.

TE

In this study, two molds with typical severe wear problems were selected and the wear

mechanisms involved were deeply studied. After that, a careful selection of the best coating for
P
CE

this purpose was done and some of the most critical parts of the mold were coated in order to

test possible effective advantages of the coating application, analyzing the wear resistance
AC

behavior and wear mechanisms involved. In parallel, tribological tests were also carried out in

order to study if a correlation between laboratory and industrial tests can be drawn. Scanning

Electron Microscopy (SEM) and Energy Dispersive Spectroscopy were intensively used to

characterize the coatings and the wear mechanisms observed. Laboratory tribological tests have

involved ball scattering and block-on-ring tests, trying to impose low and medium loads on the

contact, respectively. Promising results were obtained allowing conclude that certain coatings

present a better behavior than other ones in this field of application.

Keywords: Wear, Abrasion, Erosion, High-pressure die casting, Mold wear, Wear mechanisms,

Mold lifespan
 ACCEPTED MANUSCRIPT

1. Introduction

Regardless of the increasingly incorporation of composite materials on vehicle components,

T
high pressure die casting still remains one of the most useful manufacturing technique to obtain

IP
automotive parts with a complex shape in a cost effective way. However, the molds that are

R
used to produce those parts are constantly exposed to highly severe conditions, such as high

SC
pressure, rapid temperature fluctuations and erosion from fast-moving molten metal. The

following steps can be deemed in the high-pressure die-casting process: filling the shot sleeve,

NU
high-velocity cavity filling, application of a supplementary high pressure, cooling and
MA
solidification, mold opening and part ejection, die cooling and corresponding lubrication for a

new injection cycle [1]. The usual molten metal input speed is comprised between 20-60 m/s

and the temperature, depending on aluminum alloy type is around 700°C [2]. The maintenance
D

or replace of these molds require a huge cost which implies that producers need to find the best
TE

solution to increase their lifespan. The industrial environmental and working conditions increase
P

the capacity to induce some failure mechanisms on hot work tool steel, such as erosion,
CE

corrosion, wear and thermal fatigue [3]. Over the last years, some researches have been

developed to understand the different types of failure mechanisms [2, 4-7].

AC

The aluminum injection into the die consists on one of the most severe processes due to

aluminum soldering mechanisms. The molten aluminum creates chemical reactions with the die

surface and leads to failure mechanisms on aluminum die casting [4, 8]. For this reason, a large

number of studies have been developed to build coatings to prevent the mechanisms that

decrease the mold lifespan [1, 3, 9-16]. Ceramic coatings are usually used to avoid some failure

mechanisms, in particular, heat checking occurrence. However, other coatings can also

contribute to improve the mold lifespan and the cost may not be the most relevant factor in this

kind of application [11].

 ACCEPTED MANUSCRIPT

In 1997, Wang [14] presented an extensive work about how coatings can improve the mold

lifespan, having studied current coatings at that time, namely TiN, TiAlN and CrN, using

different hot work tool steel and maraging steels, analyzing different strands such as molten

aluminum corrosion, toughness resistance, hardness and heat changes. The study allows

T
IP
realizing that TiN is not a good solution due to its low oxidation temperature, concluding as

well that H13 or Marlok steel, when coated, can be considered for high-pressure injection

R
molding, mainly if an improvement of the impact toughness and corrosion and erosion

SC
behaviors are desired. This study was corroborated by Park and Kim [16], who conclude that

NU
TiN starts to oxidize at 500oC, while other studied coatings (TiAlN and TiSiN) showed a much

better oxidation resistance up to 700oC. Furthermore, it is well-known that TiN tends to

MA
dissociate at temperatures above 600oC [17, 18]. These last coatings exhibited as well good

mechanical properties, but the TiSiN is better for medium temperatures, being the TiAlN

coating the best for higher temperatures. However, tests carried out by Dobrzanski et al. [10]
D
TE

allowed state that TiN coating led achieving a wear resistance five times greater than

X37CrMoV5-1 type hot work tool steel under the same pin-on-disc test conditions at room
P

temperature, as well as in the same tests carried out at 500oC. Similar studies were also
CE

performed by Tentardini et al. [8] with the same TiN and CrN coatings regarding the aluminum

die casting, but this time using H13 steel and Anviloy® 1150 as substrates. These researchers
AC

found that CrN coating presented a better behavior than TiN one regarding soldering

mechanisms with the aluminum alloy in the casting process. Moreover, Guzilia et al. [1] had

also investigated soldering phenomenon, concluding that the use of TiN, CrN and TiCn coatings

avoid soldering between the aluminum alloy and the steel mold, being possible to observe a

built-up layer of cast aluminum alloy, which causes less damages to the mold, avoiding its rapid

degradation, because the coatings act as a physical barrier that prevents any reaction between

the molten aluminum and the mold steel surface. Furthermore, Heim et al. [12] also studied the

soldering phenomenon in aluminum die casting, concluding as well that coatings such as TiN,

TiCN, TiBN and TiAlCN prevent that nefarious problem.

 ACCEPTED MANUSCRIPT

Based yet on the same coatings, some new coating architecture has been tested, namely using

multi-layered coatings [19]. The different layers intended to assume different roles such as

thermal barrier (outer layer – rare earth oxide coating), a diffusion barrier (middle layer – TiAlN

coating) and thin adhesive layer (inner layer – Ti coating). By this way, thermal fatigue

T
IP
resistance of the substrate was significantly improved, as can be observed after 4000 thermal

cycles with liquid aluminum. A similar approach was carried out by Bobzin et al. [20], testing a

R
multilayered CrN/AlN/Al2O3 coating on AISI H11 steel substrate, being the Al2O3 the top layer.

SC
Also, two industrial coatings were used in that study under the same test conditions, including

NU
industrial tests with 5884 aluminum shots in an aluminum die casting machine. The

CrN/AlN/Al2O3 showed an interesting behavior comparatively to the other coatings, improving

MA
significantly the mold lifespan. Phase transformations in the coating were reported being

attributed to the temperature reached in the mold. However, a slightly different approach was

also studied by Muller [15], who used a plasma nitriding pre-treatment of the mould surface,
D
TE

followed by PVD or PACVD coatings such as TiBN, CrN and W-C:H films, concluding that

plasma nitriding pre-treatment improves the surface macro-hardness and critical load relatively
P

to surfaces just doted of the same PVD or PACVD coatings. On the other hand, Rodríguez-
CE

Baracaldo [21] studied a combination of nitriding pre-treatments with (Ti0.7Al0.3)N coating,

comparing it with two nitriding pre-treated steel and other steel just provided with (Ti0.6Al0.4)N
AC

coating, carrying out the wear tests at room and high temperature (600oC). While at room

temperature there were no significant differences regarding the wear behavior, the same is not

true at elevated temperature, where the nitriding pre-treated steel provided with the (Ti0.6Al0.4)N

coating showed the best wear behavior, comparatively to the pre-treated steel and not pre-

treated steel coated with (Ti0.6Al0.4)N coating, being the last one that performed the worst wear

behavior. Also, a different approach was investigated by Tomaslewski et al. [22], co-depositing

Mo together with TiAlN using AISI M2 equivalent high-speed steel as substrates. The quasi-

multilayer obtained is TiAlN/(TiAl)1-x - MoxN giving way to a Mo content of about 8%,

improving the friction coefficient relative to steel from 0.8 to 0.5 and decreasing the wear rate to

values below 1015 m3/N·m. This work was preceded by similar approaches carried out by other
 ACCEPTED MANUSCRIPT

authors regarding pure multilayered systems based on distinct layers of TiAlN interspersed with

Mo ones [23-25]. More recently, another study was performed by Reenwinkel et al. [26]

regarding the elastic properties of TiAlN-WNx coatings using different concentrations of TiAlN

and WNx. This study showed that the higher is the concentration of W, the lower is Young’s

T
IP
modulus, what can be extrapolated to wear behavior.

R
More recently, other advanced coatings have been investigated regarding the improvement of

SC
the wear resistance property. Park et. al. [27] studied the influence of nitriding pre-treatment of

H13 steel before the deposition of TiB2. They concluded that the adhesion of the coating to the

NU
substrate increased to values higher than 30 N by this nitriding process but the hardness

decreased to values between 20-30 GPa. However, this pre-treatment cannot be applied to
MA
coatings with high hardness (>60 GPa) due to the difference between the hardness of this

coating and the substrate, thus, the adhesion is difficult to improve for high coating hardness.
D

TiB2 similar coatings have been also studied by other authors [28], concluding that these
TE

coatings present a very good adhesion and wear resistance properties regarding the test

conditions used in that work (micro-abrasion wear test configuration). Moreover, Maaza et al.
P

[29, 30] also tried to decrease the diffusion of some other elements or compounds such as Ni
CE

and B4C on Ti by using different surface treatments regarding the its use on some surface
AC

multilayered systems in other applications.

Molds for plastic injection also represents some challenges for coating science, mainly those are

used for Glass-Fiber Reinforced Plastics (GFRP) injection, taking into account the abrasion

promoted by the glass-fiber tips. However, in these cases, there are not the same problems as in

the aluminum injection because the temperatures usually used to melt aluminum alloys are

significantly higher comparatively to plastic materials [31]. In this field, other more complex

coatings have been also investigated using micro-abrasion wear tests, such as TiAlSiN [32] and

TiAlCrSiN [33], showing promising properties in terms of adhesion and wear resistance.

However, in these cases the working temperature was not a concern take into account.
 ACCEPTED MANUSCRIPT

In this work, two different TiAlN coatings have been investigated (Ti0.4Al0.6N and Ti0.5Al0.5N)

regarding their possible application on high-pressure die casting molds mobile parts and

cavities. These films were deeply characterized and tested in order to evaluate their ability to

resist to wear under different conditions.

T
R IP
2. Materials and methods

SC
According to high-pressure die casting process is important to understand the mold cavity

NU
surface behavior when the aluminum is filling it. Thus, some usual wear tests were performed in

order to evaluate the surface wear resistance, trying to simulate the coating behavior under the
MA
high-pressure die castings conditions, mainly the aluminum flow close into the runners’ area.

The Bluet tribometer tests intend to evaluate the relative motion between some moving parts of

the mold over the solidified aluminum part, as well as slide cores movement, extractor pins, and
D
TE

the other mold parts. This kind of wear test configuration was also used by other authors as well

to test the surface wear resistance of parts subjected to medium loads [34-36].
P
CE

2.1. Materials
AC

2.1.1. Mold and substrate material

The current investigation was developed using as substrate AISI H13 steel (DIN X40CrMoV5-

1) (W. nr. 1.2344) which presents an average hardness of 42-48 HRC and 210 GPa Young’s

modulus. This substrate material is a hot work tool steel and it is used for certain types of

manufacturing tools, namely aluminum high-pressure die casting molds, being used for the

samples that intend to simulate the die casting mold.

The steel is supplied in the hardened and tempered state and the hardness work value and others

mechanical properties are optimized by the supplier for each type of process application within
 ACCEPTED MANUSCRIPT

a previous set range. The chemical composition is described in Table 1, following the

information provided by the supplier.

Table 1 – AISI H 13 (DIN X 40 CrMoV5-1) chemical composition provided by die steel supplier (wt. %)

T
C Si Mn Cr Mo V P S

IP
0,35 – 0,42% 0,80 – 1,20% 0,25 – 0,50% 4,80 – 5,50% 1,20 – 1,50% 0,85 – 1,15% 0,030% 0,020%

R
SC
2.1.2. Aluminum characterization

NU
Aluminum is a non-ferrous metal that is used for several processes, including die casting

manufacturing. On this paper, they were used two different types of aluminum-silicon-copper
MA
alloy for aluminum high-pressure die casting, A380 (AlSi9Cu3(Fe)) and A13 (AlSi12Cu1(Fe)).

The main difference between those alloys is regarding silicon content. A13 is a eutectic alloy for

aluminum high-pressure die casting. It means that the range between liquid and solid state is too
D
TE

short, for that reason this alloy is injected at a higher temperature than A380 alloy to avoid the

solidification of aluminum before filling the mold. A380 alloy is widely used on parts produced
P

by high-pressure die casting due to a good relationship between material properties and
CE

production feasibility. According to the company procedures, these alloys are poured at 630°C

and 680°C, respectively to A380 and A13. In Table 2 are represent the chemical composition of
AC

each aluminum alloy.

Table 2 - Aluminum-silicon-copper chemical composition provided by aluminum supplier (wt. %)

Elements
Alloy
Si Fe Cu Mn Mg Cr Ni Zn Pb Sn Ti

A380 8.0 – 11.0% 0.6 – 1.1% 2.0 – 4.0% 0.55% 0.15 – 0.55% 0.15% 0.55% 1.2% 0.35% 0.15% 0.20%
A13 10.5 – 13.5% 0.6 – 1.1% 0.7 – 1.2% 0.55% 0.35% 0.10% 0.30% 0.55% 0.20% 0.10% 0.15%
 ACCEPTED MANUSCRIPT

2.1.3. Laboratory samples geometric characterization

Block on ring tribometer configuration usually uses a block in contact with a ring, allowing to

measure the friction coefficient and analyze the wear behavior. In order to reproduce the wear

T
conditions as close as possible relatively to the mold work, a block in each aluminum alloy

IP
previously referred was used and a ring in hot work tool steel was utilized, which was coated

R
with 𝑇𝑖40 𝐴𝑙60 𝑁 or 𝑇𝑖50 𝐴𝑙50 𝑁, with the showed dimensions in Figure 1. The selected contact

SC
was conform, enlarging the contact area and distributing by this way the load by greater surface.

NU
MA
D
P TE
CE

Figure 1 - Block (a) and ring (b) samples geometry

AC

Due to usual porosity entrapped on cast aluminum, the block was manufactured by a milling

process, performed with care, leaving the natural roughness, which was improved by a

mechanical ground process with 1200 particles/square inch abrasive sandpaper. In the other

hand, the rings were produced by turning and drilling processes, performed with care, leaving

the natural roughness that characterizes this kind of process.

 ACCEPTED MANUSCRIPT

2.1.4. Coatings used in this work

In order to improve the surface wear resistance, different TiAlN compositions were used in this

work. These coatings were obtained by physical vapor deposition (PVD) process, as described

T
later in this paper. Concerning the literature, this kind of coatings has been previously

IP
investigated in high-pressure die casting process as well as other manufacturing processes,

R
having shown very interesting properties when subjected to high working temperatures, as

SC
previously described in the introduction section.

2.2. Methods

2.2.1. Coating process

NU
MA
𝑇𝑖40 𝐴𝑙60 𝑁 and 𝑇𝑖50 𝐴𝑙50 𝑁 were produced using an industrial Cemecon CC800/9ML PVD

Magnetron Sputtering system, which uses unbalanced magnetron sputtering technique. Prior to
D

film deposition, samples were cleaned using an ultrasonic degreasing bath for 15 minutes
TE

followed by a demineralized water bath in order to avoid surface contamination. Samples were

assembled in a PVD machine stand rotating at 1 rpm during the deposition process, which
P

intends to encourage a better film uniformity. Two different types of TiAlN targets were used in
CE

order to reproduce the films, regarding the different characteristics intended for the selected

coatings. The parameters used for each coating can be seen in Table 3.
AC

Table 3 - PVD sputtering deposition parameters

𝑻𝒊𝑨𝒍𝑵 𝟓𝟎/𝟓𝟎 𝑻𝒊𝑨𝒍𝑵 𝟒𝟎/𝟔𝟎

Time deposition (min) 240 240
Chamber gases Ar, N, Kr
Pressure (mPa) 500
Temperature (°C) 480
Bias (V) 80
Target Current (A) 20
Holder rotational speed 1
(rpm)
 ACCEPTED MANUSCRIPT

2.2.2. Morphology and thickness coatings characterization

In order to evaluate the coatings morphology and thickness, samples were cut using a

STRUERS MINITOM disk sawing machine to show a cross-section view of the coatings,

T
followed by metallurgical preparation. The cut samples were assembled into a cylinder prepared

IP
for grinding and polishing operations using a STRUERS PREDOPRESS machine with a

R
thermoset resin. Grinding process was performed taking into account the evaluation of the

SC
former finishing process in order to remove the different grooves at each grinding task, using a

conventional set of abrasive sandpapers (order: 220, 500, 800, 1200 particles/square inch).

NU
However, grinding operation was not enough to completely eliminate the grooves to get the

most accurate results for these analyses and it was needed a polishing task with two different
MA
diamond grit solutions of 3 µm and 1 µm, respectively, in order to get a better definition

between the coating and the substrate. Thickness measurement, morphology characterization
D

and wear scars were evaluate using FEI Quanta 400FEG scanning electron microscope (SEM)
TE

provided with EDAX Genesis X-ray spectroscope (EDS).

P
CE

2.2.3. Coatings chemical composition analysis

AC

In order to evaluate the coating composition with the needed accuracy, a CAMECA SX-50

Electron Probe Micro-Analysis (EPMA) system equipped with Wavelength Dispersive

Spectroscopy (WDS) system was used, letting to confirm the coatings chemical composition

and the effective difference between two TiAlN films.

2.2.4. Adhesion analysis

Adhesion between coatings and substrate was evaluated by scratch test, following the BS EN

1071-3:2005 standard. Thus, a CSM REVETEST scratch tester equipment was used imposing

10 N·min-1 as load increasing rate per unit of time, within the range of 0 – 100 N. Scratch-tests
 ACCEPTED MANUSCRIPT

were performed using a diamond sliding indenter speed of 10 mm·min-1. According to

previously referred standard, the test was repeated five times on each sample, in order to

improve the results truthfulness. The scratches produced were also investigated using FEI

Quanta 400FEG scanning electron microscope (SEM) in order to quantify the critical loads

T
IP
those originated cohesive and adhesive failures, as well as the identification of the phenomena

registered when the failures occurred.

R
SC
2.2.5. Ultra-Micro Hardness Test

NU
Ultra-micro hardness tests were performed using Fisherscope H100 equipment, assessing to the
MA
coating hardness using a Vickers indenter. The normal load selected was 50 mN which was kept

constant during 30 seconds, avoiding creep phenomenon. This load was selected taking into
D

account that indentations depth cannot exceed 10% of the coating thickness, avoiding by this
TE

way substrate influence on the acquired values. This equipment gives rise to ‘load-depth’

curves, which allows for assessing to the hardness (H), reduced (Er) and revised (E’) Young’s
P

modulus. This test was carried out 10 times in different areas of each sample in order to get an
CE

average value dotted with the required exactness. Hardness 6.00 software was used to draw the

load – penetration depth curves and calculate the hardness and Young’s modulus values.
AC

Hardness tests were performed according to ISO 14577-1:2015 standard.4

2.2.6. Surface roughness characterization

Ring surface roughness was assessed using a profilometer MAHR PERTHOMETER M2

following the established standard procedure DIN EN ISO 4288 / ASME b461, and the

measurements were performed using seven segments of 0,8 mm each one (cut-off), disregarding

the first and the last ones due to the accelerating and deaccelerating processes. The evaluation of

the roughness was performed taking into account the following parameters: arithmetic mean
 ACCEPTED MANUSCRIPT

surface roughness (Ra), maximum surface roughness (Rmax) and R-profile. The roughness was

confirmed using also a VEECO Multimode atomic force microscope equipped with a 7 nm

radius probe tip and provided with the NanoScope 6.13 software. The area of analysis selected

was 50 x 50 m2, letting to achieve the larger area possible with high accuracy. Main roughness

T
IP
parameters and 3D surface scan were achieved, letting to understand the surface morphology

pattern of the analyzed coatings.

R
SC
2.2.7. Wear behavior characterization

NU
Block on ring tribometer was used trying to simulate the wear and friction behavior on contact,

being the typical geometry of the samples shown before, in Figure 1. Each sample was produced
MA
in different materials: rings on hot work tool steel (AISI H13) and blocks on aluminum-silicon-

copper alloys (A380 and A13). A view of the home-made block on ring tribometer used on this
D

work can be seen in Figure 2, which was already used for many other works already published
TE

[34 - 36]. Steel rings were assembled on a shaft provided with two pins those avoid the relative

motion between the shaft and the ring. This set was animated with a rotational speed electrically
P

controlled by frequency inverter device. The tribometer’s head section is free to move in
CE

ascendant and descendent direction, empowering block assembly in the equipment stand and

corresponding adjustment to the ring surface allowing a conformal contact with relatively large
AC

area. The tests were carried out in dry conditions using 70 N as normal load and 33.9 m ·

min−1 linear speed.

In Table 4 are described the parameters used on the block on ring tribometer. The load usually

used on this type of equipment depends on the total disks weight that is placed over the holder

plate. In this work, there were not used any disks to increase the natural load on ring and block

surface contact because the weight associated with the holder structure, pads, and guides (70 N)

was considered the most appropriated to produce the accurate and uniform results in order to

evaluate wear surface behavior.

 ACCEPTED MANUSCRIPT

T
R IP
SC
Loadand
Load andLoad-holder
Load-holder

NU
MA
D
P TE
CE

Figure 2 - View of the block on ring tribometer used in tribological tests

AC

Table 4 - Tribological test parameters

Load [N] 70

Speed [rpm] 90
Linear speed [mm/min]
33 912

Ring Diameter [mm] 120

Contact Area [mm2] 255

Contact pressure [MPa] 0,27

Number of cycles 100

Total sliding distance per test [m] ~37,7

 ACCEPTED MANUSCRIPT

Industrial tests were developed on the real die-casting casting process, proceeding to 30,000

aluminum alloy injection cycles. Core pins previously coated were assembled in critical areas of

the injection die, taking part of the overall die motion, where die and slide cores are moving

before the molten aluminum is filled. Different core pins with Ti40Al60N and Ti50Al50N were

T
IP
assembled in the mold cavity taking into account that molten aluminum flow and temperature

are the same for the whole set. The coated core pins were subjected to the mold movements and

R
thermal cycle, being directly subjected to the molten aluminum alloy flow during the mold

SC
filling process. After 30,000 injection cycles, the pin cores were extracted and evaluated at room

NU
temperature, in order to study the wear promoted by the aluminum flow during the high-

pressure injection cycle over the core pins. These tests intended to evaluate the wear behavior
MA
and friction coefficient on solid aluminum and coated surfaces contact. Furthermore, wear

mechanisms were also studied attending the scars left on the surface.
D
P TE
CE
AC

Figure 3 - View of the mold where the core pins were assembled for industrial tests
 ACCEPTED MANUSCRIPT

3. Results and Discussion

3.1. Coating thickness and composition analysis

Figure 4 presents the thickness analysis made on the cross-section of the coatings after

T
metallurgical preparation. Regarding the images, it can be concluded that Ti40 Al60 N and

IP
Ti50 Al50 N films are formed by monolayer thin film around 3 µm and 2 µm, respectively. A

R
careful observation of the coatings along the prepared samples permits to observe that films

SC
follow the substrate surface, keeping their thickness almost constant. During SEM

observations, EDS spectrum was obtained to verify the coating composition and try to detect

NU
any impurity (Figure 5). Also, the structure of the coatings was analyzed, allowing conclude that

the films are homogenous. The coating thickness values evaluated on this paper are very similar
MA
with other values reported in other studies [37, 38, 41] regarding TiAlN coatings.
D
P TE
CE
AC

Figure 4 - Cross-section views of the coatings (𝑇𝑖40 𝐴𝑙60 𝑁 (a) and 𝑇𝑖50 𝐴𝑙50 𝑁 (𝑏)) allowing measure the coatings
thickness
 ACCEPTED MANUSCRIPT

T
R IP
SC
Figure 5 - EDS spectra collected during SEM observations allowing confirm the composition coatings: (a)

NU
𝑇𝑖40 𝐴𝑙60 𝑁 and (b) 𝑇𝑖50 𝐴𝑙50 𝑁 MA
3.2. Surface morphological analysis

The coatings surface morphology was also evaluated during SEM observations because this is
D

an important parameter regarding the tribological analysis. Attending to the pictures depicted in
TE

Figure 6, it can be concluded that Ti50 Al50 N coating is smoother than Ti40 Al60 N. Taking into

account that these coatings need to be able to contact easily with other surfaces having relative
P

motion between them, the Ti50 Al50 N surface morphology is more suitable to produce a lower
CE

friction behavior on the block and ring contact due to droplets absence on the coated surface, in
AC

order to avoid slug formations that will be propagated along the contact. On Ti40 Al60 N surface,

it is possible to perceive some droplets that are embedded into the coating surface, allowing

concludes that there are some chances that these particles stand out from the matrix where they

are embedded, acting as hard third-body and promoting by this way more severe wear

consequences.
 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
Figure 6 - Coatings morphological characterization (a) 𝑇𝑖40 𝐴𝑙60 𝑁 and (b) 𝑇𝑖50 𝐴𝑙50 𝑁 (top view)
MA

3.3. Roughness analysis

D

In order to assess the surface roughness, it was used two different methods as mentioned above
TE

in 2.2.6. section. Firstly, atomic force microscope (AFM) analysis was carried out in order to
P

make a scan of the surface topography regarding each coating surface, as can be seen in
CE

FigureFigure 7. These scans permit to observe once again that Ti50 Al50 N coating surface is

smoother than Ti40 Al60 N one. At the same time, roughness parameters as the arithmetic mean
AC

surface roughness (Ra) and maximum roughness (Rmax) were collected, letting to compare the

surfaces morphology. For tribological purposes, the arithmetic mean surface roughness (Ra)

should be considered in the longitudinal alignment due to the rotation direction.

Regarding industrial application, the coatings were also evaluated using an adequate

profilometer, collecting the same parameters formerly gathered but using a linear analysis in an

extensive area (5.6 mm of linear analysis) comparing with the AFM analysis (50 x 50 m2). The

comparing results can be observed in Table 5, including the substrate values.

 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
MA

Figure 7 - AFM topography analysis on 𝑇𝑖40 𝐴𝑙60 𝑁 (a) and 𝑇𝑖50 𝐴𝑙50 𝑁 (b) coated ring surface
D
TE

Table 5 - Roughness analysis using MAHR PERTHOMETER profilometer and AFM

P

Samples TiAlN 50/50 TiAlN 40/60 Substrate

CE

Alignment Longitudinal Transverse Longitudinal Transverse Longitudinal Transverse

Ra (µm) 0.292 ± 0.010 0.576 ± 0.096 0.336 ± 0.025 0.571 ± 0.066 0.270 ± 0.069 0.806 ± 0.040
AC

𝑹𝒎á𝒙 (µm) 4.020 ± 0.542 6.257 ± 0.780 3.830 ± 0.457 4.710 ± 0.627 2.440 ± 0.102 7.200 ± 0.922

After Table 5 analysis, it is possible to conclude that the difference between roughness substrate

material and after PVD deposition is characterized by a lower value, more precisely 0.02 µm for

Ti50 Al50 N and 0,066 µm for Ti40 Al60 N. According to the references [37,39], the arithmetic

mean surface values are higher than expected probably due to machining operations (tools and

parameters) that cause an increase of final roughness surface.

 ACCEPTED MANUSCRIPT

3.4. Coatings chemical composition analysis

The coatings chemical composition was evaluated in order to study the coatings behavior

according to a specific ratio. In this paper, it was needed to consider the different titanium and

T
aluminum ratios, because it is needed to conclude about what is the best coating for aluminum

IP
high-pressure die casting or to find the straight way to achieve that. On Table 5 is show the

R
chemical composition results for both used coatings obtained by EPMA.

SC
Table 1 - Coatings chemical composition results

𝑻𝒊𝟓𝟎 𝑨𝒍𝟓𝟎 𝑵 coating 𝑻𝒊𝟒𝟎 𝑨𝒍𝟔𝟎 𝑵 coating

NU
Element Wt% Wt%
NK 52,01 52,45
Al K 25,01 28,43
MA
Ti K 22,98 19,11
D

3.5. Adhesion between coatings and substrate

TE

Being the adhesion a critical parameter regarding tribological applications, scratch-tests were

performed in order to evaluate the bonding strength between substrate and coatings, using the
P

equipment and parameters referred previously. Figure 8 allows observe and determine the first
CE

event occurred on Ti40 Al60 N coating at 30 N normal load (Figure 8a), which can be considered
AC

as a cohesive failure. The first event that can be connected to an adhesive failure occurred at 47

N normal load, as can be seen in Figure 8b. On the other hand, Ti50 Al50 N coating showed

similar behavior but providing normal load values for failure events slightly higher than the

other coating, namely 33 N and 49 N (Figure 9), respectively to cohesive and adhesive failures.

In the case of the image 9b), it is not clear that we are in presence of a detachment but,

regarding the dimension of the event, it was considered that there are no conditions to consider

the film completely attached to the substrate after this event. Table 6 intends to resume all

values obtained in these tests.

 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
Figure 1 - Scratch-test critical loads (a) cohesive and (b) adhesive failure events on 𝑇𝑖40 𝐴𝑙60 𝑁 coating
MA
D
P TE
CE

Figure 9 - Scratch-test critical loads (a) cohesive and (b) adhesive failure events on 𝑇𝑖50 𝐴𝑙50 𝑁 coating
AC

Table 7 - Scratch-test analysis on 𝑇𝑖40 𝐴𝑙60 𝑁 and 𝑇𝑖50 𝐴𝑙50 𝑁 coated surface

Scratch-test
Critical Load 𝐓𝐢𝟒𝟎 𝐀𝐥𝟔𝟎 𝐍 𝐓𝐢𝟓𝟎 𝐀𝐥𝟓𝟎 𝐍
𝑳𝒄𝟏 31 N 33 N
𝑳𝒄𝟐 47 N 49 N
 ACCEPTED MANUSCRIPT

3.6. Ultra-micro hardness analysis

The hardness and Young’s modulus of the coatings were assessed by Vickers indentation, as

described on 2.2.5. Concerning that thin films were being used in this work and their thickness

T
was already known, the maximum indentation depth should not exceed 10% coating thickness,

IP
avoiding substrate influence on the values obtained. Thus, the load selected was 50 mN, as

R
mentioned before. Regarding the results exposed on Table 7 for hardness and Young’s modulus,

SC
one can conclude that both coatings show values whose are according to the literature [14, 21,

37, 39-41], being possible to state that Ti40Al60N show higher hardness and Young’s modulus

NU
than its competitor.
MA
Table 8 - Ultra-micro hardness and Young’s modulus of the used coatings

Coating Hardness Young Modulus

H [GPa]
D

Er [GPa] E’ [GPa]
Ti50Al50N 23,57 ± 7,15 271,60 ± 43,70 320,05 ± 37,82
TE

Ti40Al60N 33,63 ± 4,76 368,10 ± 62,60 469,88 ± 37,66

P
CE

3.7. Wear and tribological behavior analysis

The tribological tests were carried out to evaluate the wear behavior of each coating. These tests
AC

were performed using a block on ring tribometer as stated before and they were used two blocks

on each ring with two different aluminum alloys usually used in high-pressure die casting, as

mentioned in section 2.1.2. In order to assess the friction force on contact area, it was collected

and recorded data for further treatment regarding the coefficient of friction evaluation during

each test. A chart was drawn for each test regarding the tangential load values achieved by the

data-acquisition board, amplified and computed by a personal computer dotted with MatLab®

software, indicating them as a function of the time/sliding distance, as described previously in

section 2.2.7. Soldering mechanisms between steel and molten aluminum were usually founded

on aluminum high-pressure die casting producing intermetallic connections, due to some

metallurgical compatibility between steel and aluminum, which in the practical case will
 ACCEPTED MANUSCRIPT

damage the mold surface and decrease its lifespan. For that reason, quantitative evaluation of

mass loss was not measured due to the significant hardness difference between samples, being

mainly evaluated the material transfer between block and ring. Thus, it was studied the

evaluation of aluminum transfer and how it evolved during the test.

T
IP
In order to accurately achieve the results on tribological tests, each block and ring used sample

R
were carefully identified to keep the traceability for further data analysis. Hence, for SEM

SC
analysis, the samples were judiciously cut using a thin diamond disk to assemble them properly

on SEM holder and ensuring the preservation of all grooves and scars that were produced during

NU
wear tests. It was used carbon adhesive strip to hold the samples for SEM assessment.
MA
In Figure 10 and Figure 11 are depicted SEM images collected from ring coated surfaces. It is

possible to see two different colored areas with a periodic distance between them. This distance

corresponds to the turning feed rate and the difference of tonality means that aluminum alloy
D

was transferred from the block to the ring just in the peaks of the ring, preserving the steel
TE

surface of the ring without aluminum on the valleys. Regarding these SEM observations, it can
P

be stated that it has a homogeneous transfer of aluminum alloy but there are some areas where
CE

the transfer seems a little more intense. However, apparently, there is no remarkable difference

between the behavior presented by A380 and H13 alloys in contact with the steel ring.
AC
 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
Figure 2 - SEM images collected of 𝑇𝑖40 𝐴𝑙60 𝑁 coated ring surface in contact with different aluminum alloy ((a)
MA
A380 and (b) A13) after wear tribological tests on block on ring tribometer. Sliding direction represented by arrows
D

Figure10 depicts Ti40 Al60 N coated ring surface after contact with each type of aluminum alloy
TE

during the wear tests. As can be seen, some aluminum areas are darker than others that it can be

related to the aluminum layer thickness transferred from the block to the ring surface. EDS
P

spectra were taken during SEM analysis allowing evaluate and characterize the transferred
CE

material. Figure 11 corresponds to Ti50 Al50 N coated ring surface in the same conditions as

described regarding Figure10. In this case, it is possible to observe a more intense transfer from
AC

the block to the ring for A13 aluminum alloy than its competitor, A380. Thus, it seems to have

higher metallurgical affinity between the A13 alloy with the Ti50 Al50 N coating than with the

Ti40 Al60 N alloy, meaning that the contents of aluminum and titanium in the coating produce

effects on the amount of material transferred in the contact.

 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
Figure 11 - SEM images collected of 𝑇𝑖50 𝐴𝑙50 𝑁 coated ring surface in contact with different aluminum alloy ((a)
A380 and (b) A13) during wear tribological tests on block on ring tribometer. Sliding direction represented by
arrows
MA

Figure 12 corresponds to a detailed analysis regarding one of the dark areas of Figure 10. In
D

Figure 12a) can be seen a relatively thin layer of aluminum transferred from the block to the
TE

ring, which is dispersed by the peaks of the Ti40 Al60 N coating, where the contact during the
P

wear tests effectively is made. Micro-analyses made by EDS in two different colored areas are
CE

shown in Figures 12b) and 12c, being possible to observe that spectrum of the Zone 1 (lighter

one) corresponds to the coating composition remaining almost clear of other block particles or
AC

contaminants, whereas the spectrum of the Zone 2 (darker one) corresponds essentially to

aluminum alloy transferred from the block to the ring surface. The aluminum layer is thick

enough to avoid the coating detection through it even using a 15 kV electron beam. In this case,

the aluminum alloy is deposited on the ring in a certain area and, after that, it seems that there is

a preferable aluminum transfer always in this area, promoting the layer thickness increase in this

area, while the other areas remain clear of aluminum transfer effect. Thus, the distance selected

for tribological tests was enough to promote a significant amount of aluminum from the block to

the ring, but the last one does not show any evidence of wear because the coating does not show

any kind of visible damage.

 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
MA
D
P TE
CE

Figure 12 - SEM observation and spectra of transferred A13 aluminum alloy from the block to 𝑇𝑖40 𝐴𝑙60 𝑁 coated
ring surface (a, b, c). Sliding direction represented by arrow
AC

Figure 13 corresponds to a similar wear test, this time conducted over Ti50 Al50 N coating and

under the same conditions. Analogous dark stains can be observed in the coating surface but, in

this case, the distribution is randomly made covering a significant area visible in the picture. As

in the former case, the lighter areas correspond to coating free of aluminum alloy transferred

from the block as well as almost free of contaminants. The thickness of the aluminum alloy

deposited over the coating is thick as well, which does not allow detect the coating and steel

substrate (Figure 13c), just presenting significant peaks of O (oxides), Al and Si (block alloy).

The different behavior relatively to Ti40 Al60 N coating can be attributed to the different

morphology of the coatings (greater peaks/agglomerates on the Ti40 Al60 N surface relatively to
 ACCEPTED MANUSCRIPT

Ti50 Al50 N), promoting a localized transfer on the prominent surface peaks and helping that

material is being accumulated preferably in these areas. On the other hand, the smoother surface

presented by the Ti50 Al50 N coating encourages a greater distribution of the load on the contact,

being also the material transferred equally better distributed over a larger area. Due to the

T
IP
different coating composition, the affinity between the block aluminum alloy can assume as

well its influence, having a greater readiness to support adhered transferred films.

R
SC
NU
MA
D
P TE
CE
AC

Figure 13 - SEM observation and spectra of transferred A380 aluminum alloy on 𝑇𝑖40 𝐴𝑙60 𝑁 coated ring surface (a,
b, c). Sliding direction represented by arrow
 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
MA
D
P TE
CE

Figure 14 - SEM observation and spectra of transferred A13 aluminum alloy on 𝑇𝑖50 𝐴𝑙50 𝑁 coated ring surface (a, b,
c). Sliding direction represented by arrow
AC

Figure 14 represents a detail view of the extended area affected by the transferred aluminum

alloy. It seems clear that some prominent areas of the coating assume the contact with the block

and the material transfer acts preferably in these areas, promoting the thickness increase of this

layer, mainly because it assumes particular relevance in the contact, being those areas the ones

responsible for carrying the load between through the contact. Regardless its thickness, the

aluminum layer seems do not disturb the coating surface, which remains intact. It is expected

that these aluminum layers have distinct behaviors on the bottom and top: in the bottom, the

material in carrying permanently the load, being predictable that a cold hardening effect can be

produced in this area. At the top of the layer, it is expected that shear loads are stronger than the
 ACCEPTED MANUSCRIPT

adhesion ones, being a certain renewal of that layer along the test duration. Despite that, the

linkage between aluminum adhered layer and ring coating seems does not be strong enough to

promote coating detachments regarding the test duration considered in this work.

T
Figure 15 depicts the same situation regarding the Ti50 Al50 N coating, being possible to observe

IP
that there is a thick layer in the central area of the image but there are other small stains of

R
aluminum alloy at a different level, whose show a great affinity to the coating surface, even in

SC
areas where the coating seems to be in a lower level comparatively to the average.

NU
MA
D
P TE
CE
AC

Figure 15 - SEM observation and spectra of transferred A13 aluminum alloy on 𝑇𝑖50 𝐴𝑙50 𝑁 coated ring surface (a, b,
c). Sliding direction represented by arrow
 ACCEPTED MANUSCRIPT

Indeed, the same does not occur in Figure 14a), where the aluminum alloy tends to adhere in the

peaks left by the turning process and that coating process cannot disguise. Based on this

observation, the affinity of the Ti50 Al50 N to aluminum alloy seems to be greater than that one

showed by Ti50 Al50 N coating. The conclusion extracted from Figures 14 and 15 are a little

T
IP
corroborated by the average coefficient of friction assessed in the same tribological tests.

Effectively, the average coefficient of friction registered under the same condition was always

R
higher when the ring was coated with Ti50 Al50 N than Ti40 Al60 N, as can be seen in Table 9.

SC
NU
Table 9 - Friction coefficient values reached in tribological tests performed on Block on Ring tribometer

Coating 𝐓𝐢𝟒𝟎 𝐀𝐥𝟔𝟎 𝐍 𝐓𝐢𝟓𝟎 𝐀𝐥𝟓𝟎 𝐍

Aluminum alloy A380 A13 A380 A13
MA
Friction coefficient (µ) 0,159 0,047 0,188 0,114
D

Industrial tests were also performed in order to confirm the observations made in laboratory
TE

tests. Thus, core pins usually subjected to severe wear due to the natural motion and also due to

the temperatures developed in the high-pressure die casting process were coated and inserted
P

into the mold, trying to establish a correlation between the tribological tests carried out at room
CE

temperature and laboratory loads with the real service developed by the core pins into the mold
AC

and under the usual working conditions used in high-pressure die casting. After 30,000 shots,

the core pins were extracted in order to evaluate the state of the surface. It is noteworthy that

these core pins are moving two times for each injection and their tip is directly exposed to the

molten aluminum flow during the mold filling process.

In Figure 16 is possible to observe (a) the core pin as-produced (before PVD coating process),

(b) the core pin already coated (in this case, with Ti40 Al60 N), (c) the tip of the core pin coated

with Ti50 Al50 N after 30,000 shots and (d) the tip of the core pin coated with Ti40 Al60 N after

30,000 shots. As can be seen in that picture, the wear under real working conditions was also

more severe in the pin coated with Ti50 Al50 N than its competitor, Ti40 Al60 N, as pointed out by
 ACCEPTED MANUSCRIPT

the arrows in Figure 16c). Moreover, the adhesion seems to be greater in the Ti50 Al50 N than in

the competitor, attending the amount of aluminum alloy deposited on tip surface.

(c) (d)

T
R IP
SC
NU
MA
Figure 16 - Core pin samples used on die-casting mold (a) uncoated pin, (b) 𝑇𝑖40 𝐴𝑙60 𝑁 coated pin, (c) after 30,000
shots with 𝑇𝑖50 𝐴𝑙50 𝑁 coating, (d) after 30,000 shots with 𝑇𝑖40 𝐴𝑙60 𝑁 coating
D
TE

It is noteworthy that these core pins when used in the uncoated state are not able to make more

than 20,000 shots, occurring mechanical seizing. Indeed, the Ti50 Al50 N shows some evidence
P

of coating failure in the lighter arrow of the Figure 16c), comparing with the surface of its
CE

competitor, which presents just a little contact mark. The tip of the Ti40 Al60 N core pin also

presents less wear and aluminum adhesion than the Ti50 Al50 N coated core pin, corroborating by
AC

this way the observations made before in the laboratory tests.

Thus, the Ti40 Al60 N coating shows better properties to be applied in inserts or movable parts of

molds used in high-pressure die casting process than the Ti50 Al50 N coating, regardless both

contribute to a drastic increase of the mold parts lifespan.

 ACCEPTED MANUSCRIPT

4. Conclusions

This work addressed the comparison between two similar coatings in order to evaluate their

suitability to be used in movable parts or cavities of molds utilized in high-pressure die

T
casting. Comparative studies comprised laboratory and practical tests, subjecting the

IP
coatings to hard work conditions. Coatings were deeply characterized, allowing understand

R
the phenomena felt during the wear tests. After all, tests carried out and corresponding

SC
evaluations, it is possible to state that:

 Ti50 Al50 N coating presented good wear resistance in laboratory tests but revealed a

NU
higher propensity to capture material from the counterface, aluminum alloy. The

material from the block adheres easily to the Ti50 Al50 N coating surface and pursues
MA
thickening the surface aluminum layers until the shear load promoted on the contact

being higher than the intrinsic strength of the aluminum layer, letting to go out the
D

more recent adhered material;

TE

 Ti50 Al50 N coating, when used in industrial tests, showed poorer wear resistance,

letting to realize some evidence of wear relatively to its competitor;

P

 Ti40 Al60 N showed a very good wear behavior in laboratory tests, presenting lower
CE

friction coefficient and letting to realize that its more irregular surface induces that
AC

contact is made preferably in these points around them the formation of adhered

aluminum is more evident. Thus, this coating presents two favorable aspects: the

rougher surface avoids a stronger contact between the coating and the aluminum

block and the metallurgical affinity seems to be lower in this pair of materials;

 When tested in industrial environment, Ti40 Al60 N also showed a better wear

behavior, revealing lesser aluminum adhesion and presenting just a little mark of

contact;

 It also can be stated that lifespan of the mold parts tested was not tested to the limit

due to lack of time. However, after 30,000 injection shots the parts show a very
 ACCEPTED MANUSCRIPT

healthy aspect, which is a promising indicator regarding the maximum number of

cycles they will be able to produce;

 No significant differences were felt between the two aluminum alloys used as block

material. However, the hardness difference between the coatings and aluminum

T
IP
alloy is so high that no wear evidence was felt on the coatings surface.

Face to the results obtained, the tested Ti40 Al60 N coating revealed to be a better solution than

R
Ti50 Al50 N one, due to having presented less friction coefficient and better wear resistance.

SC
5. Acknowledgements NU
MA
The authors of this work would like to thank TEandM due to its collaboration in providing all

the coatings necessary to carry out this study. SONAFI is also acknowledged due to its
D

collaboration in this project, availability to provide samples and materials, as well as for the
TE

time given to Vitor Nunes in order to perform all the tests and analyses. The authors would like
P

to extend their gratitude to Dr. Rui Rocha (CEMUP) for his help and criticism in SEM analysis, as
CE

well as Prof. Monteiro Baptista and Prof. Jorge Seabra for their support and for allowing access

to the INEGI facilities. Those responsible for the Mechanical Engineering Department are also
AC

acknowledged due to the permission was given to use all the laboratories. Fátima Andrade is

also acknowledged for her support and help in laboratory tasks carried out at ISEP.
 ACCEPTED MANUSCRIPT

6. References

[1] S. Gulizia, M.Z. Jahedi, E.D. Doyle, Performance evaluation of PVD coatings for high pressure die casting,

T
Surf. Coatings Technol., 140(3) (2001) 200–205. DOI: 10.1016/S0257-8972(01)01037-4.

IP
[2] A. Persson, S. Hogmark, J. Bergström, Thermal fatigue cracking of surface engineered hot work tool steels,

R
Surf. Coatings Technol., 191(2–3) (2005) 216–227. DOI: 10.1016/j.surfcoat.2004.04.053.

SC
[3] I. Peter, M. Rosso, and F.S. Gobber, Study of protective coatings for aluminum die casting molds, Appl.

Surf. Sci., 358 (2015) 563–571. DOI: 10.1016/j.apsusc.2015.08.013.

NU
[4] Z.W. Chen, M.Z. Jahedi, Die erosion and its effect on soldering formation in high-pressure die casting of

aluminum alloys, Mater. Des., 20(6) (1999) 303–309. DOI: 10.1016/S0261-3069(99)00035-7.

MA
[5] A. Persson, S. Hogmark J. Bergström, Simulation and evaluation of thermal fatigue cracking of hot work

tool steels, Int. J. Fatigue, 26(10) (2004) 1095–1107. DOI: 10.1016/j.ijfatigue.2004.03.005.

D

[6] D. Klobčar, J. Tušek, Thermal stresses in aluminum alloy die casting dies, Comput. Mater. Sci., 43(4) (2008)
TE

1147–1154. DOI: 10.1016/j.commatsci.2008.03.009.

P

[7] D. Klobčar, L. Kosec, B. Kosec, J. Tušek, Thermo fatigue cracking of die casting dies, Eng. Fail. Anal., 20
CE

(2012) 43–53. DOI: 10.1016/j.engfailanal.2011.10.005.

[8] E.K. Tentardini, A.O. Kunrath, C. Aguzzoli, M. Castro, J.J. Moore, and I.J.R. Baumvol, Soldering
AC

mechanisms in materials and coatings for aluminum die casting, Surf. Coatings Technol., 202(16) (2008)

3764–3771. DOI: 10.1016/j.surfcoat.2008.01.019.

[9] O. Knotek, F. Löffler B. Bosserhoff, PVD coatings for diecasting moulds, Surf. Coatings Technol., 62(1–3)

(1993) 630–634. DOI: 10.1016/0257-8972(93)90310-K.

[10] L.A. Dobrzanski, M. Polok, P. Panjan, S. Bugliosi, M. Adamiak, Improvement of wear resistance of hot

work steels by PVD coatings deposition, J. Mater. Process. Technol., 155–156(1–3) (2004) 1995–2001.

DOI: 10.1016/j.jmatprotec.2004.04.405.

[11] B. Navinšek, P. Panjan, I. Urankar, P. Cvahte, F. Gorenjak, Improvement of hot-working processes with

PVD coatings and duplex treatment, Surf. Coatings Technol., 142–144 (2001) 1148–1154. DOI:

10.1016/S0257-8972(01)01098-2.
 ACCEPTED MANUSCRIPT

[12] D. Heim, F. Holler, C. Mitterer, Hard coatings produced by PACVD applied to aluminium die casting, Surf.

Coatings Technol., 116–119 (1999) 530–536. DOI: 10.1016/S0257-8972(99)00104-8.

[13] C. Mitterer, F. Holler, F. Üstel, D. Heim, Application of hard coatings in aluminium die casting - soldering,

erosion and thermal fatigue behaviour, Surf. Coatings Technol., 125(1–3) (2000) 233–239. DOI:

T
10.1016/S0257-8972(99)00557-5.

IP
[14] Y. Wang, A study of PVD coatings and die materials for extended die-casting die life, Surf. Coatings

R
Technol., 94–95(97) (1997) 60–63. DOI: 10.1016/S0257-8972(97)00476-3.

SC
[15] K.B. Müller, Deposition of hard films on hot-working steel dies for aluminium, J. Mater. Process. Technol.,

130–131 (2002) 432–437. DOI: 10.1016/S0924-0136(02)00715-X.

[16]
NU
I.-W. Park, K.H. Kim, Coating materials of TiN, Ti–Al–N, and Ti–Si–N by plasma-enhanced chemical
MA
vapor deposition for mechanical applications, J. Mater. Process. Technol., 130-131 (2002) 254-259. DOI:

10.1016/S0924-0136(02)00807-5.

[17] N.J.M Carvalho, E. Zoestbergen, B.J Kooi, J.Th.M De Hosson, Stress analysis and microstructure of PVD
D

monolayer TiN and multilayer TiN/(Ti,Al)N coatings, Thin Solid Films, 429(1-2) (2003)179-189. DOI:
TE

10.1016/S0040-6090(03)00067-1.

[18] J. Musil, H. Hrubý, Superhard nanocomposite Ti1−xAlxN films prepared by magnetron sputtering, Thin
P

Solid Films, 365 (2000) 104-109. DOI: 10.1016/S0040-6090(00)00653-2.

CE

[19] A. Srivastava, V. Joshia, R. Shivpuria, R. Bhattacharya, S. Dixit, A multilayer coating architecture to reduce
AC

heat checking of die surfaces, Surf. Coatings Technol., 163–164 (2003) 631-636. DOI: 10.1016/S0257-

8972(02)00690-4.

[20] K. Bobzin, T. Brögelmann, U. Hartmann, N.C. Kruppe, Analysis of CrN/AlN/Al2O3 and two industrially

used coatings deposited on die casting cores after application in an aluminum die casting machine, Surf.

Coatings Technol., vol. 308 (2016) 374-382. DOI: 10.1016/j.surfcoat.2016.09.040.

[21] R. Rodríguez-Baracaldo, J.A. Benito, E.S. Puchi-Cabrera, M.H. Staia, High temperature wear resistance of

(TiAl)N PVD coating on untreated and gas nitrided AISI H13 steel with different heat treatments, Wear, 262

(2007) 380-389. DOI: 10.1016/j.wear.2006.06.010.

[22] Ł. Tomaszewski, W. Gulbinski, A. Urbanowicz. T. Suszko, A. Lewandowski, W. Gulbinski, TiAlN based

wear resistant coatings modified by molybdenum addition, Vacuum 121 (2015) 223-229. DOI:

10.1016/j.vacuum.2015.08.027.
 ACCEPTED MANUSCRIPT

[23] C.J. Tavares, L. Rebouta, M. Andritschky, F. Guimaraes, A. Cavaleiro, Mechanical and surface analysis of

Ti0.4Al0.6N/Mo multilayers, Vacuum 60 (2001) 339-346. DOI: 10.1016/S0042-207X(00)00413-9.

[24] C.J. Tavares, L. Rebouta, J.P. Rivière, J. Pacaud, H. Garem, K. Pischow, Z. Wang, Microstructure of

superhard (Ti,Al)N/Mo multilayers, Thin Solid Films, 398-399 (2001) 397-404. DOI: 10.1016/S0040-

T
6090(01)01349-9.

IP
[25] C.J. Tavares, C. Vidrago, L. Rebouta, J.P. Rivière, E. Le Bourhis, M.F. Denanot, Optimization and thermal

R
stability of TiAlN/Mo multilayers, Surf. Coat. Technol. 200 (2005) 288-292. DOI:

SC
10.1016/j.surfcoat.2005.02.099.

[26] T. Reeswinkel, D.G. Sangiovanni, V. Chirita, L. Hultman, J.M. Schneider, Structure and mechanical

NU
properties of TiAlNeWNx thin films, Surf. Coat. Technol. 205 (2011) 4821-4827. DOI:

10.1016/j.surfcoat.2011.04.066.
MA
[27] B. Park, D.-H. Jung, H. Kim, K.-C. Yoo, J.-J. Lee, J. Joo, Adhesion properties of TiB2 coatings on nitrided

AISI H13 steel, Surf. Coat. Technol. 200 (2005) 726-729. DOI: 10.1016/j.surfcoat.2005.01.064.
D

[28] F.J.G. Silva, R.C.B. Casais, R.P. Martinho, A.P.M. Baptista, Mechanical and Tribological Characterization
TE

of TiB2 Thin Films, J. Nanosci. Nanotechnol., 12 (2012) 9187–9194. DOI: 10.1166/jnn.2012.6759.

[29] M. Mâaza, C. Sella, J. P. Ambroise, M. Kâabouchi, M. Milôche, F. Wehling and M. Groos, Determination of
P

diffusion coefficient D and activation energy Qa of nickel into titanium in Ni-Ti multilayers by grazing-
CE

angle neutron reflectometry, J. Appl. Cryst. 26 (1993) 334-342. DOI: 10.1107/S0021889892010355.

AC

[30] M. Mâaza. A. Menelle. J.P. Chauvineau. B. Pardo. A. Raynal. F. Bridou. C. Sella. T. Megademini.

Monochromation and anodization with Ti-B4C multilayers in neutron optics, Physica B, 198(1-3) (1994)

231-234. DOI: 10.1016/0921-4526(94)90167-8.

[31] K. Bobzin, T. Brögelmann, R.H. Brugnara, N.C. Kruppe, S. Chromy, Influence of HPPMS pulse parameters

on the reactive gas N2 and on the properties of (Cr, Al)N coatings, Surf. Coat. Technol. 293 (2016) 28–34.

DOI: 10.1016/j.surfcoat.2015.12.072.

[32] M.F.C. Andrade, R.P. Martinho, F.J.G. Silva, R.J.D. Alexandre, A.P.M. Baptista, Influence of the abrasive

particles size in the microabrasion wear tests of TiAlSiN thin coatings, Wear 267 (2009) 12–18. DOI:

10.1016/j.wear.2008.12.114.

[33] R.P. Martinho, M.F.C. Andrade, F.J.G. Silva, R.J.D. Alexandre, A.P.M. Baptista, Microabrasion wear

behaviour of TiAlCrSiN nanostructured coatings, Wear 267 (2009) 1160–1165. DOI:

 ACCEPTED MANUSCRIPT

10.1016/j.wear.2008.12.063.

[34] Francisco J. G. Silva, Revestimentos duros ultra-finos executados por P.V.D. (in Portuguese), Master Thesis,

Faculty of Engiineering, University of Porto, Porto, 1997.

T
[35] Arnaldo M. G. Pinto, Influência do Comportamento Dinâmico da Máquina de Ensino nos Resultados de

IP
Atrito e Desgaste, (in Portuguese), Master Thesis, Faculty of Engineering, University of Porto, Porto, 2001.

R
[36] F. Silva, A. T. Ribeiro, L. A. Ferreira, A comparative study of the tribological behaviour of TiN and ZrN

SC
PVD coatings, Lubrication at the frontier, D. Dowson et al. (Eds), Tribology Series, Elsevier, Amsterdan, pp

141-147.

NU
[37] P. Terek, L. Kovacevic, A. Miletic, P. Panjan, S. Balos, B. Skoric, Effects of die core treatments and

surfaces finishes on the sticking and galling tendency of Al-Si alloy casting during ejection, Wear, 356-357
MA
(2016) 122-134. DOI: 10.1016/j.wear.2016.03.016.

[38] E. Lugscheider, C. Barimani, S. Guerreiro, K. Bobzin, Corrosion tests of PVD coatings with die lubricant

used for Al high-pressure die-casting dies, Surf. Coat. Technol., 108-109 (1998) 408-412. DOI:
D

10.1016/S0257-8972(98)00624-0.
TE

[39] J.L. Mo, M.H. Zhu, B. Lei, Y.X. Leng, N. Huang, Comparison of tribological behaviours of AlCrN and

TiAlN coatings - Deposited by physical vapor deposition, Wear, 263 (2007) 1423-1429. DOI:
P

10.1016/j.wear.2007.01.051.
CE

[40] A. Drnovsek, P. Panjan, M. Panjan, M. Cekada, The influence of growth defects in sputter-deposited TiAlN
AC

hard coatings on their tribological behavior, Surf. Coat. Technol., 288 (2016) 171-178. DOI:

10.1016/j.surfcoat.2016.01.021.

[41] B. Wang, G.R. Bourne, A.L. Korenyi-Both, A.K. Monroe, S.P. Midson, M.J. Kaufman, Method to evaluate

the adhesion behavior of aluminum-based alloys on various materials and coatings for lube-free die casting ,

J. Mater. Process. Technol., 237 (2016) 386-393. DOI: 10.1016/j.jmatprotec.2016.06.031.

 ACCEPTED MANUSCRIPT

List of Captions

Figure 1 - Block (a) and ring (b) samples geometry

T
Figure 2 - View of the block on ring tribometer used in tribological tests

IP
Figure 3 - View of the mold were the core pins were assembled for industrial tests

R
Figure 4 - Cross-section views of the coatings (Ti40Al60N (a) and Ti50Al50N (b)) allowing

SC
measure the coatings thickness

Figure 5 - EDS spectra collected during SEM observations allowing confirm the composition

NU
coatings: (a) Ti40Al60N and (b) Ti50Al50N

Figure 6 - Coatings morphological characterization (a) Ti40Al60N and (b) Ti50Al50N (top view)
MA
Figure 7 - AFM topography analysis on Ti40Al60N (a) and Ti50Al50N (b) coated ring surface

Figure 8 - Scratch-test critical loads (a) cohesive and (b) adhesive failure events on
D

Ti40Al60N coating
TE

Figure 9 - Scratch-test critical loads (a) cohesive and (b) adhesive failure events on Ti50Al50N

coating
P

Figure 10 - SEM images collected of Ti40Al60N coated ring surface in contact with different
CE

aluminum alloy ((a) A380 and (b) A13) during wear tribological tests on block on ring
AC

tribometer. Sliding direction represented by arrows

Figure 11 - SEM images collected of Ti50Al50N coated ring surface in contact with different

aluminum alloy ((a) A380 and (b) A13) during wear tribological tests on block on ring

tribometer. Sliding direction represented by arrows

Figure 12 - SEM observation and spectra of transferred A13 aluminum alloy on Ti40Al60N

coated ring surface (a, b, c). Sliding direction represented by arrow

Figure 13 - SEM observation and spectra of transferred A380 aluminum alloy on Ti40Al60N

coated ring surface (a, b, c). Sliding direction represented by arrow

Figure 14 - SEM observation and spectra of transferred A13 aluminum alloy on Ti50Al50N

coated ring surface (a, b, c). Sliding direction represented by arrow

 ACCEPTED MANUSCRIPT

Figure 15 - SEM observation and spectra of transferred A13 aluminum alloy on 𝑇𝑖50𝐴𝑙50𝑁

coated ring surface (a, b, c). Sliding direction represented by arrow

Figure 16 - Core pin samples used on die-casting mold (a) uncoated pin, (b) Ti40Al60N coated

pin, (c) after 30,000 shots with Ti50Al50N coating, (d) after 30,000 shots with Ti40Al60N coating

T
IP
List of Captions

R
SC
Table 1 – AISI H 13 (DIN X 40 CrMoV5-1) chemical composition provided by die steel

NU
supplier (wt. %)

Table 2 - Aluminum-silicon-copper chemical composition provided by aluminum supplier (wt.

MA
%)

Table 3 - PVD sputtering deposition parameters

D

Table 4 - Tribological test parameters

TE

Table 5 - Roughness analysis using MAHR PERTHOMETER profilometer and AFM

Table 6 - Coatings chemical composition results

P
CE

Table 7 - Scratch-test analysis on Ti40Al60N and Ti50Al50N coated surface

Table 8 - Ultra-micro hardness and Young’s modulus of the used coatings

AC

Table 9 - Friction coefficient values reached in tribological tests performed on block on ring

tribometer

Highlights

 High-pressure die casting molds require high maintenance due to its hard work;
 Coatings have been already applied in this kind of tool subjected to high temperature;
 Two different TiAlN were tested in order to overcome molds cavity wear;
 Ti40Al60N showed better performance than Ti50Al50N in laboratory tests;
 Industrial tests confirmed the best performance of Ti40Al60N used in core pins of molds.