Accepted Manuscript

New environmentally friendly coatings for hot forging tools

A. Dubois, M. Dubar, C. Debras, K. Hermange, C. Nivot, C.

Courtois

PII: S0257-8972(18)30306-2
DOI: doi:10.1016/j.surfcoat.2018.03.055
Reference: SCT 23235
To appear in: Surface & Coatings Technology
Received date: 2 October 2017
Revised date: 13 February 2018
Accepted date: 19 March 2018

Please cite this article as: A. Dubois, M. Dubar, C. Debras, K. Hermange, C. Nivot,
C. Courtois , New environmentally friendly coatings for hot forging tools. The address
for the corresponding author was captured as affiliation for all authors. Please check if
appropriate. Sct(2017), doi:10.1016/j.surfcoat.2018.03.055

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
New environmentally friendly coatings for hot forging tools

A. DUBOISa, M. DUBARa, C. DEBRASa, K. HERMANGEb, C. NIVOTb, C. COURTOISb

a Univ. Valenciennes, CNRS, UMR 8201 - LAMIH - Laboratoire d'Automatique de Mécanique et

d'Informatique Industrielles et Humaines, F-59313 Valenciennes, France
b Univ. Valenciennes, EA 2443 - LMCPA - Laboratoire des Matériaux Céramiques et Procédés Associés, F-
59313, Valenciennes, France

Abstract

Two alumina coatings are developed to protect tool surface in hot forging: pure γ-alumina matrix loaded

PT
with α-alumina particles and pure γ-alumina matrix loaded with h-BN particles. The coatings are deposited
on stainless tools by using a sol-gel procedure. Adherent coatings are obtained with an average thickness
ranging from 200 to 300 nm. The coatings are tested according to the conditions of contact encountered in

RI
hot forging processes: coated tools slide against the surface of hot specimens where they generate a plastic
strain. The design of experiment involves specimens heated to 1100°C, lubricated and unlubricated contact,

SC
small and large specimen plastic strain, uncoated and sol-gel coated tools. Tests results are the Coulomb’s
coefficient of friction, SEM-EDS and roughness measurement of the tool surfaces after a sliding length of 40
mm.
NU
Results show that the tested sol-gel coatings remain firmly bonded to the tool surface. Compared to
uncoated tools, alumina sol-gel coatings limit the amount of material transfer from specimen surface to tool
surface. Loading the alumina sol-gel coating with h-BN particles significantly improves its friction behaviour:
the friction coefficient is lower and sticking phenomena are reduced.
MA

keywords: sol-gel coating, α-alumina, h-BN, friction, wear, hot forging

E D
PT
CE
AC

1/24
 ACCEPTED MANUSCRIPT

1 Introduction

Hot forging is an efficient way to preform metal parts quickly and at low cost. Hot forging processes
commonly involve two dies that compress a hot billet to the desire shape. The main advantage of hot
forming is that the yield stress of the billet is strongly lowered due to its high temperature. Hot forging
allows the forming of complex geometry with a reduced amount of mechanical energy. Hot forging has two
main drawbacks. The first drawback is the deformed billet that cannot be considered as a finish product due
to poor surface quality and geometrical tolerance (surface oxidation, thermal contraction, possibility of
phase transformation for some alloys, etc.). The second is the cost of the forming tools. The cost of the dies
is generally estimated as 8 to 15% of the total production costs. In extreme cases, with small production

PT
series, it can even reach 30% [1-2]. In usual working conditions, forging dies undergo only elastic strains.
Their lifetime is then mainly conditioned by surface defect occurring during manufacturing. The conditions
of contact between the dies and the hot billet are very severe: high contact pressures, high sliding
velocities, billet surfaces heated to temperature up to 1200°C, presence of oxide scales, etc. These

RI
conditions of contact generate complex wear patterns, with combined action of mechanical and thermal
stresses, adhesion, abrasion and plastic strain [3-7]. As friction and wear reduce tool life [8-12], the

SC
improvement of wear resistance of hot forming tools directly contributes to enhancing their lifetime and, by
consequence, reduces the production cost of the finished product.
NU
Many technologies have been developed to improve wear resistance. Most of them are based on the fact
that wear is inversely proportional to material hardness, so increasing the surface hardness by appropriate
surface treatments leads to a higher wear resistance. Two main approaches are used: surface treatments
and surface coatings. Surface treatments increase hardness by modifying the metallurgical state in the
MA

vicinity of the treated surface. They consist in a heat treatment combined with an input of hardening
element, such as bore, carbon or nitrogen. Nitriding is still a surface treatment widely used in hot forging to
improve hot forging die lifetime [13]. Surface coating consists in the deposition of a hard layer on the
surface. The layer can be obtained by different techniques, such as PVD, CVD or PACVD. These techniques
D

condition the layer ability to reduce friction and wear [14-18].

E

Surface coatings have to fulfil two constraints in order to be used in hot forming:
i.) the temperature of the deposition should not involve the softening of tool material [11][16],
PT

ii.) the coating must resist to the thermal fatigue induced by the picks of temperature occurring
when the billets come in contact with the tool. Considering hot forming of steel, picks of
temperature may reach 400°C, leading to a surface temperature close to 750°C [19].
CE

For example, PVD TiN coatings which are commonly used in cold forming cannot be employed in hot forging
because the working temperature is too high and quickly leads to a decrease of its working efficiency [20].
AC

On the contrary, alumina coatings are stable at such high working temperature, but these coatings cannot
be easily processed at low temperature unless using expensive Plasma assisted CVD [21][22].

Sol gel route is known to produce reactive powder or coatings to enhance high temperature die forming
tools [23]. Sol-gel techniques are easily practicable low temperature chemistry, environmentally friendly,
and are shown as a promising way to chemically modify surfaces [24].

Extensive work has been published on oxide coatings deposited by the sol-gel process to enhance corrosion
resistance [25-31] and to improve hardness of metallic substrates [32-35]. Among the various characterized
oxide coatings, alumina is one of the most promising materials because of its thermal stability, corrosion
and wear resistance and chemical inertness [30][36-41].
The aim of the present work is to study the possibility of using Al2O3 sol-gel coatings to improve lifetime of

2/24
 ACCEPTED MANUSCRIPT
hot forging tools. Two sol-gel coatings are analysed: pure -alumina matrix loaded with -alumina particles
and pure -alumina matrix loaded with h-BN particles. The Warm and Hot Upsetting-Sliding Test is
introduced first. This test bench is involved to simulate conditions of contact closed to industrial hot forging
ones. Then the coating procedure is presented and the main characteristics of the two studied are given.
Finally tests are performed and the sol-gel coatings are ranked in terms of their abilities to protect hot
forming tool surface and to reduce friction stresses.

2 Experimental procedure

2.1 The Warm and Hot Upsetting Sliding Test

The Warm and Hot Upsetting Sliding Test (WHUST) is a friction test developed to simulate in a laboratory
environment the conditions of contact encountered in the industry of hot forming [10][13]. The test bench

PT
involves a contactor which slides along the surface of a hot specimen with given contact pressure (Fig. 1).

Input parameters of the WHUST are the contactor penetration within the specimen surface p, the sliding

RI
velocity Vs, the materials, temperatures and shapes of contactors and specimens [42]. The design of the test
bench allows the use of contactors and specimens directly machined from industrial tools and workpieces.
This particularity guarantees that the chemical and physical properties of surfaces in contact during the

SC
tests are representative of those of industrial processes: they possess the same roughness, the same
chemical reactivity, the same surface pollution, etc. A heating cartridge is embedded in the contactor and
allows the contactor surface to be heated to a temperature up to 300°C. An inductor furnace is used to heat
NU
specimens to temperatures up to 1200°C.
The temperatures, the sliding velocity, the shape and the penetration of the contactor are chosen in order
to impose a contact pressure, a plastic strain, an interface temperature similar to those of industrial
processes [43]. In the present study, the contactor has a cylindrical shape, the axis of the cylinder being
MA

perpendicular to the sliding direction. This geometry has been chosen because it leads to a better control of
the imposed contact conditions and to a better accuracy on the identification of the coefficient of friction
[44].
E D
PT
CE
AC

Figure 1: schematic views and main parameters of the Warm and Hot Upsetting Sliding Test (WHUST).

WHUST results can be divided in two main types: mechanical and physical results. Mechanical raw results
are the forces acting in directions normal and tangential to the sliding velocity. Physical raw results are
surface analyses and measurements performed on specimens and contactors: roughness, micrographs,
SEM-EDS. These raw results are processed and expressed as “performance indicators” in order to qualify
and to quantify the efficiency of a lubricant to reduce friction and to protect tool and workpiece surfaces

3/24
 ACCEPTED MANUSCRIPT
[45]. Main performance indicators are coefficients of friction and the variation of roughness due to wear
and material transfer. The Coulomb’s coefficient of friction is often used as performance indicator. Its value
is given by:

q. Ft − p. Fn
𝜇= (1)
p. Ft + q. Fn

where p is the penetration of the contactor within the specimen, q is length of the contact surface and Ft
and Fn are respectively the tangential and normal forces recorded during the friction test [Dub96b].
Nonetheless, the contact pressures encountered in metal forming can be very high and, in that case, the
friction stresses are no more proportional to the contact pressure. The constant friction model (a.k.a.
Tresca’s friction law) provides better results [46]:

PT
 = m.k0 (2)

where  is the friction stress, m is the friction factor and k0 is the yield stress in pure shear in the vicinity of

RI
the contact zone. Assuming that the material respect the von Mises yield criterion, the friction factor is
identified by:

SC
√3(q. Ft − p. Fn )
m= (3)
σ0 . L(q2 + p2 )
NU
where L is the mean width of the surface of contact and 0 is the yield stress of the specimen (in MPa). The
main difficulty here is to obtain a reliable estimation of the yield stress in the vicinity of the contact zone.
MA

This property can be estimated by hardness measurement at room temperature [47] but no experimental
measurement at working temperature is available.

The variation of roughness due to material transfer is quantified by profilometric measurements performed
D

on the contactor surface. A profile of the surface perpendicular to the sliding direction is measured. The
profile starts and ends on the unused areas of the surface (surface of the contactor that does not come into
E

contact with the specimen). The reference line of the profile is then defined as the mean plane of the
PT

unused zones. The material transfer on the worn area of the contact surface is quantified by the two criteria
defined by Boher et al. [3, 12]. Profiles with a height superior to the reference line represent adhesive
material and are quantified by the adhesive criterion (Stad); profiles with a height below the reference line
CE

represent abrasion and are quantified by the abrasive criterion (Stab). Stad and Stab are given by:

1 L
Stad = ∫0 max⁡(z, 0)dx (5)
L
AC

1 L
Stab = L ∫0 min⁡(z, 0)dx (6)

where z is the height of the profile, L the length of the worn zone and x the variable along the worn zone
(fig. 2). Stad and Stab are expressed in µm. The abrasive criterion (Stab) represents a loss volume per unit of
surface: its value is negative.

4/24
 ACCEPTED MANUSCRIPT

Fig. 2. Example of profile. Definition of the reference line and of the adhesive and abrasive criteria.

PT
2.2 Starting materials

RI
2.2.1 As received contactors
AISI H11 steel is one of the most used tool materials in hot metal forming. Nonetheless, the deposition

SC
procedure involves heating the tool material to a temperature equalled to 600°C for an hour. The hardness
of AISI H11 steel starts decreasing when its temperature reaches 500°C. The loss of hardness reaches
28% when heated to 600°C for one hour [48]. This make the AISI H11 steel unappropriated to undergo
NU
sol-gel coating. The study has then been carried out on contactors made of AISI 304 stainless steel. Even if
stainless steel is not a common material for tools in hot forging this material has showed very good ability
to increase sol-gel coating adhesion [49] and it can undergo heat treatment at 600°C without strong
MA

modifications of its mechanical properties [50]. The chemical composition of this stainless steel is given in
Table 1.
D

Table 1: Chemical composition of the AISI 304 Stainless Steel according to AISI norm
E

Wt.% C Wt.% Cr Wt.% Ni Wt.% Mn Wt.% S Wt.% Si Wt.% Fe

PT

<0.08 18-20 8 - 10.5 <2 <0.03 <1 Balance

CE

Fifty four steel contactors were prepared for the study. Each stainless steel contactor was polished and
cleaned in alcohol in an ultrasonic bath in order to achieve a clean surface with a controlled roughness.
Surface roughness was measured by using a white light non-contact profilometer ZYGO NewView 7300. The
AC

polishing lead to orthotropic surfaces, with a mean roughness Ra equal 0.13 µm is the polishing direction
and 0.23 µm in the transverse direction (Fig.3). The polishing direction was adjusted to match with the
sliding direction of the WHUST.

5/24
 ACCEPTED MANUSCRIPT

Figure 3: 3D non-contact profilometry on a 4 by 4 mm² AISI 304 substrate surface

PT
2.2.2 Coating procedure

RI
The alumina Sol-Gel coating is prepared by controlled gelification of Aluminum tri-sec butoxide
precursor in an alcoholic based solvent [51]. For this work, alumina gels are loaded with ceramic powders.

SC
Characteristics of starting materials are listed in table 2.-alumina is chosen due to its high hardness and
was successfully used as a load powder in alumina sol-gel coating [52]. Hexagonal-Boron nitride is
introduced in the sol-gel film due to its well-known lubricant effect [53-56]. The flowchart of the gel
preparation is presented on figure 4. Amounts of reactants are given by:
NU
 molar ratio water/aluminium precursor: 2/1,
 molar ratio AcAc/aluminum precursor: 0.6,
MA

 final concentration of aluminum precursor: 0.5mol/l,

 nitric acid: in sufficient amount to reach pH=4. For ceramic powder loaded gel,
 the loading of ceramic powder is fixed to Vol%50 of the obtained composite coating.
E D

Table 2: Characteristics of starting materials

Formula or mineral
PT

Reagent or mineral provider purity

phase
Aluminum tri-sec-
Al(O(CH2)3CH3)3) Alfar Aesar 97%
butoxide
CE

Sol-gel reagents
Ac-ac Acetyl acetone Carlo Erba RPE
Solvent isopranol Fisher 97%
Alumina -Al2O3 ALCOA CT3000SG
AC

Load powders Saint Gobain

Boron nitride h-BN 98% (2%B2O3)
Ceramics

6/24
 ACCEPTED MANUSCRIPT

Al(But)3 + iPrOH mixing

Aluminum Tri-sec Butoxide (precursor) and Isopropanol
(solvent) are mixed during one hour

+ AcAc
complexant Acetylacetona addition, and stirring during 2
hours

+H2O

PT
Dropwisely addition of Water during 20 minutes

+ HNO3 (65%)

RI
Nitric acid addition until pH reach 4

SC
Gentle stirring during 24 hours
NU
Optional : addition of load ceramic particles under
ultrasonic assistance

Figure 4: Sol-Gel preparation flowchart

MA

The coating of the contactor operates in three steps:

D

i. Sol-gel and composite sol-gel preparation procedure: aluminium precursor and isopropanol are
mixed together during 1 hour. AcAc is slowly added to regulate the gelification rate. The solution is
E

stirred during 2 hours in a closed vessel. Deionised water is then dropwisely added during 40min in
order to obtain a clear solution. At the end, pH is adjusted to 4 by slow addition of concentrated
PT

nitric acid. The gel is ageing in a close vessel during 12 hours under slow agitation. At this time, a
stable gel is obtained with a viscosity close to 4 10-3 Pa.s. This gel can be manipulated at this
viscosity value for two up to three days. If a loaded gel is needed, the ceramic powders are added
CE

just after the ageing step under ultrasonic agitation. The dip procedure is performed after 2 min of
agitation.
ii. Dip coating procedure: the samples are dipped at a constant withdrawal rate of 60mm/min.
AC

Samples are placed vertically to ensure a homogeneous coating all over the tool without any
excessive thickness.
iii. Drying and thermal treatment: Coated tools are dried in air at 110°C. A first thermal treatment is
performed in air with the following conditions: heating up to 400°C at 1°C/min, holding time 1hour
at this temperature, then cooling down to room temperature at 5°C/min. This thermal treatment
transforms the gel into amorphous aluminium phase [37]. This procedure is repeated to obtain the
target coating thickness. The final thermal treatment conditions in air are as follow: heating rate
1°C/min up to 600°C, holding time 1 hour, and cooling down to room temperature at 5°C/min. At
this step, a water free and amorphous alumina based coating is obtained. Figure 5 presents
contactors before and after treatment.

7/24
 ACCEPTED MANUSCRIPT

PT
Figure 5: AISI 304 contactors before (left) and after (right) the sol-gel coating process

RI
SC
2.2.3 Characteristics of coated tools NU
Two characterization methods were performed to estimate the final thickness of the sol-gel coating. First
the contactors were coated according to the procedure presented in figure 4 and were observed by SEM in
the direction normal to the coating cross section (Fig. 6). SEM observations exhibit that the coating is thin,
MA

adherent and homogeneously distributed onto the substrate.

Second, 3D confocal analysis was used along a straight line of the contactor surface. Measurements started
on a non-coated part of the contactor and ended on the coating so that the variation in height was
corresponding to the coating thickness (Fig. 7).
E D
PT
CE
AC

Figure 6: Measurement of the thickness of a non-loaded sol-gel coating using SEM.

8/24
 ACCEPTED MANUSCRIPT

PT
Figure 7: Measurement of the thickness of a non-loaded sol-gel coating at transition using confocal
microscopy

RI
Both methods of measurement estimate the average coating thickness to be in the range 200 - 300 nm. This
result is consistent with the thickness of other alumina Sol-Gel depositions encountered in literature

SC
[37][41][57][58].

Figures 8 and 9 present SEM observations of sol-gel coatings loaded with -Alumina and with h-BN
NU
particles. These observations are made on the lateral flat surfaces used to manipulate the contactors. This
explains the presence of some polishing scratches that do not exist on the working surface (cylindrical
surface). The ceramic powder is clearly detected in the sol-gel. The load for both ceramic powders appears
to be finely and homogeneously dispersed onto the substrate.
MA
E D
PT
CE

Figure 8: Sample coated with an Alumina Sol Gel Figure 9: Sample coated with an Alumina Sol Gel
loaded with -Alumina particles loaded with h-BN
AC

2.3 Design of experiments

The WHUST was involved to test the behaviour of three surface preparations: non-coated surface, alumina
sol-gel loaded with  alumina coating, and alumina sol-gel loaded with hexagonal Boron Nitride coating.

The testing conditions were chosen to be representative of hot forming processes. They corresponded to
the conditions of contact encountered near the flash zone at the die/workpiece interface of close die
forging operations [13]. This die zone is subject to moderate contact pressure (mean value around 100 MPa

9/24
 ACCEPTED MANUSCRIPT
with local value up to 300 MPa) but undergo an untimely tool wear due to the large sliding length of the
workpiece material along the tool (larger than 10 mm), to the high value of workpiece plastic strain and to
possible lacks of lubricant (Fig.10).

PT
RI
Figure 10: example of wear localisation in the flash zone of a hot forging die (AISI H11 steel without
coating). After 15000 strokes with a graphite based lubricant (left). After 6500 strokes with a white

SC
lubricant (right).

Four different contact configurations were investigated on each surface preparation. These configurations
NU
correspond to two lubrications and two mechanical loadings. The two types of lubrication were without
lubricant and with a commercial dispersion of 5% of graphite particles in water. The mean size of graphite
particles was 12 µm. The thickness of the graphite layer remaining on contactor surface after water
evaporation was equalled to 30 µm [59]. The two mechanical loadings were a mean contact pressure close
MA

to 100 MPa, a sliding velocity equalled to 60 mm.s-1 and an effective plastic strain equalled to 0.63 or to
2.33. Each configuration was tested three times leading to a total of 36 tests. Each test was performed with
new contactors and specimens.
D

Specimens were cylinders of 30 mm diameter made of AISI 4820 steel. They were heated to 1200°C in an
induction furnace and cool down in air to 1100°C just before testing. The heating and cooling kinetic was
E

adjusted so that the thickness of oxide scale was lower than 20 µm, which is the value encountered on the
PT

industrial process [59]. Contactors were made of AISI 304 stainless steel. Their active surface was heated to
200°C by a heating cartridge embedded into the contactor.
CE

WHUST results are given by the friction factor m identified from equation (4), SEM-EDS analyses of
contactors surface and the abrasive and adhesive criteria given by equations (5) and (6).
AC

3 Results and discussion

3.1 Numerical simulations

Finite element simulations of the WHUST have been carried out. The aim of these computations was first to
identify the WHUST input parameters needed to simulate the mechanical contact conditions given in
section 2.3. The computations also provided helpful information not accessible by experiments, such as the
local distribution of contact pressure, the specimen temperature and the yield stress in the vicinity of the
contact area (required to identify the friction factor m).

10/24
 ACCEPTED MANUSCRIPT
Specimens and contactors were modelled with 3D thermomechanical elements with full integration.
Element size was chosen in order to achieve a density of 4 elements by square millimetre at the contact
surfaces (Fig. 11). The meshing density was designed to obtain accurate results at the interface and through
the volume concerned by the plastic deformation. The heat transfer coefficient was set to 5000 W.m-2.°C-1
[10]. A particular attention was paid to the description of the bulk behaviour of the AISI 4820 steel. Indeed,
most commonly used behaviour laws are not able to manage correctly the coupling between the strong
decrease in temperature and the strain rate increase that occurs in the contact vicinity. The model proposed
by Puchi-Cabrera et al. was then chosen for its ability to model the variation of the behaviour due to
instantaneous strain rate and temperature changes [60].
Even if particles and oxide scales transferred from the specimen to the contactor lead to a rough
nonhomogeneous surface that requires to take into account the uncertainty of the response to the external
loading [61], the present work assumes that the evolution of roughness does not affect the stability of the

PT
sliding contact. Computations are then performed with the constant friction law under quasi static loading
conditions. The coefficient of friction used for the numerical simulations was identified from equation (3)
and trial experimental tests.

RI
SC
NU
MA
E D
PT
CE

Figure 11. Finite element mesh, contactor in green, specimen in blue.

AC

Table 3 presents the main results of the finite element simulations of the WHUST. Even though the plastic
strain strongly increases from 0.66 to 2.33 when tests are performed with penetration p increasing from 0.1
to 0.4 mm, the increase of the strain rate and the decrease of the temperature lead the yield stress to
remain almost constant and equal to 245 MPa in the vicinity of the contact surface.
Except a maximum value observed in the front of the contact surface, the distribution of contact pressure is
quite homogeneous all along the contact surface (Fig. 12).

11/24
 ACCEPTED MANUSCRIPT
Table 3. Design of experiments, main mechanical data for simulations performed with a contactor radius R0
equalled to 20 mm.
Yield stress near
Mean Maximum Mean strain
Yield stress Maximum Interface the contact zone
Penetration contact contact rate near the
at 1100°C plastic temperature at interface
p (mm) pressure pressure contact zone
(MPa) strain (°C) temperature
(MPa) (MPa) (s-1)
(MPa)
0.1 95 235 142 0.66 905 19 235

0.4 105 285 152 2.33 905 35 245

PT
RI
SC
NU
MA
D

Figure 12: Top view of the contactor. Distribution of the contact pressure along the surface for a penetration
p equal to 0.4 mm.
E
PT

3.2 Surface observations of the tool

Surface observations are performed in two steps: optical observations are first carried out to provide a
CE

global view of the tool surface after the tests. SEM micrographs are taken in a second step and lead to more
local characterization (size of the transferred particles and thickness of the transferred layers). Energy
dispersive X-Ray Spectroscopy is involved to determine the chemical nature of transferred materials onto
AC

the contactor.

Material transfer is noticed on contactor surface after each experiment (Figure 13). The adhesive layer is
mainly made of iron oxide torn off the specimen. The width concerned with adhesion correspond to the
width of the contact surface, i.e. transfer occurs partially or totality all over the contact zone. Two main
areas are observed: an adhesive area and an abrasive area. The adhesion area is located on the front of the
contact surface where the contact pressure is maximum (figure 12). The transferred material is mainly
perpendicular to the sliding direction and presents a lot of large asperities. The abrasive area is located at
the rear of the contact zone and shows scratches parallels to the sliding direction.

12/24
 ACCEPTED MANUSCRIPT

PT
Figure 13. Front view of the contact zone. Friction mark left on surfaces after one test without lubricant.

RI
SC
SEM-EDS analyses were performed on contact area of the contactors used for the tests with a plastic strain
of 0.66. These analyses were done in order to have a first quantification of transferred materials. Analyses
were performed along a line perpendicular to the sliding direction in the middle of the contact area (figures
NU
14-19). Three pictures were taken along the contact zone on each contactor. As each test was performed
three times, nine images were analysed by Gimp software in order to estimate the ratio between adhesive
area and the contact area.
MA

Figures 14 and 15 are related to tests performed without coating. SEM-EDS analyses on these pictures are
targeting chromium in order to distinguish stainless steel of the contactor from the oxide scale transfer from
the specimens. Material transfer from specimen to contactor surface is large and covers slightly more than
80% of the contact area when test are performed without lubricant (figure 14). Even though the use of
D

graphite based lubricant strongly reduces the area subjected to material transfer, more than 55% of the
contact area remains affected by adhesion when no coating is used (figure 15).
E

Figures 16 to 19 are related to trials performed with sol-gel coatings. SEM-EDS analyses in these pictures are
targeting aluminium in order to distinguish the coating of the contactor from the steel transfer from the
PT

specimens. Significant reductions of adhesion phenomena are observed. Compared to uncoated contactors,
alumina coating loaded with -alumina and alumina coating loaded with h-BN respectively lead to a
decrease of 30% and 57% of the area affected by material transfer when no lubricant is used (figures 16 and
CE

18). This result is quite similar when the contact is lubricated with a graphite in water solution: -alumina
loaded coating lead to a reduction of 35% of the adhesion area when h-BN loaded coating lead to a
reduction of 54% (figures 17 and 19).
AC

13/24
 ACCEPTED MANUSCRIPT

PT
Figure 14: material adhesion on uncoated surface tested without lubricant and a plastic strain of 0.66.

RI
SC
NU
MA
D

Figure 15: material adhesion on uncoated surface tested with lubricant and a plastic strain of 0.66.
E
PT
CE
AC

Figure 16: material adhesion on an alumina coated surface without lubricant. The coating is loaded with
Alpha Alumina and a plastic strain of 0.66.

14/24
 ACCEPTED MANUSCRIPT

PT
Figure 17: material adhesion on an alumina coated surface with lubricant. The coating is loaded with Alpha
Alumina and a plastic strain of 0.66.

RI
SC
NU
MA
D

Figure 18: material adhesion on an alumina coated surface without lubricant and a plastic strain of 0.66. The
coating is loaded with hexagonal Boron Nitride.
E
PT
CE
AC

Figure 19: material adhesion on an alumina coated surface with lubricant. The coating is loaded with
hexagonal Boron Nitride and a plastic strain of 0.66.

The material transfer phenomenon is present all over the contact surface for trials performed under a 2.33
plastic strain. SEM micrographs highlight accumulations of several layers (Figures 20-21) when smoother

15/24
 ACCEPTED MANUSCRIPT
adhesion areas were observed under smaller plastic strain. An assumption is that when performing test
under small plastic strain, transfer material mainly comes from the oxide scale present at the surface of the
specimens. These scales are transfer from the specimen surface to the contactor surface at the beginning of
the contact. Scales are then upset under a pressure of 100 MPa, leading to the rather smooth adhesive area
presented in figures 14 to 19. Under larger plastic strain, in addition to the oxide scale already present on
specimen surface, chips of fresh steel are torn from the specimen and adhere to contactor surface. Chips
oxide when the surface is cooling down, leaving place to the oxide layers displayed in figures 20 and 21.

Nonetheless, conclusion about the positive effect of sol-gel coating cannot be drawn from these optical
observations as adhesive material convers the whole contact area.

PT
RI
SC
NU
Figure 20: Sticking material on a stainless tool without any coating and without lubricant for a plastic strain
MA

of 2.33
E D
PT
CE
AC

Figure 21: Sticking material on a stainless tool with an Alumina coating loaded with h-BN particle and
without lubricant for a plastic strain of 2.33

16/24
 ACCEPTED MANUSCRIPT

3.3 Quantification of adhesion

Roughness measurements have been performed to quantify the material transfer as no conclusion about
the positive effect of sol-gel coating were drawn from optical observations under high plastic strain.
Roughness measurements were performed on contactor surface after each test. 2D surface profiles were
measured on the main adhesive area along a line perpendicular to the sliding direction (Figure 22). Surface
profiles with a height superior to the reference line represent adhesive material and are quantified by the
adhesive criterion Stad (eq.5); profiles with a height below the reference line represent abrasion and are
quantified by the abrasive criterion Stab (eq.6).

PT
RI
SC
NU
MA

Figure 22: surface profile measurement for WHUSTs performed without lubricant.
D

Adhesive material is present almost all over the contact line of contactors used without coating or used with
alumina loaded with -alumina sol-gel coatings (Figure 19, green surface). Abrasion seems to appear on
E

coatings loaded with hexagonal boron nitride (red surface). Nonetheless, SEM-EDS clearly show that
PT

aluminium is still present on contactor surfaces after the tests meaning that the coating hasn’t been
removed from the surface (figures 18-19). The negative profiles observed in the present study are then due
to a plastic deformation of the contactors and not to abrasion. As a consequence only the adhesive criterion
CE

Stad is analysed in the following (figure 23).

An increase of plastic strain from 0.63 to 2.33 leads to an increase of the thickness of transfer material
whatever the surface coating or the lubrication involved. So the plastic strain that undergo the specimens
AC

affects both the dimension of the adhesion area and the thickness of transferred material. Nonetheless, the
ranking of the coating is not really affected by this parameter: uncoated contactors always present the
largest amount of adhesion, contactors with sol-gel coating loaded with h-BN always present the lowest.
The thickness of the adhesive layer is reduced up to 75% when h-BN loaded coatings are used.
As one can expect, lubrication reduces the amount of material transferred from specimen to contactor
surfaces. Graphite particle present in the lubricant reduces friction. Consequently the tangential stresses in
the area of contact are smaller and can no more tear so much material off specimen surfaces. Nonetheless
the adhesive criterion is not very sensitive to the presence of lubricant when sol-gel coatings are used. The
thickness of the adhesive material is strongly reduced when lubricant is used in addition to the h-BN coating
under small plastic strain. But no big differences are noticed for higher plastic strains. This means that the
efficiency of the lubricant is sensitive to the mechanical properties of contact. Only tests performed with
the -alumina loaded sol-gel coating are not significantly influenced by lubricant. For this coating, the

17/24
 ACCEPTED MANUSCRIPT
amount of transfer remains equalled to 8 µm for low plastic strain tests and 21 µm for high plastic strain
tests.

PT
RI
SC
Figure 23: Effect of coating, lubrication and plastic strain on the adhesive criterion Stad
NU
3.4 Friction coefficients

Equation (3) was used to identify the friction factor m according to the tested conditions of contact (table
MA

4). The normal and tangential forces were recording during the tests and the yield stress in the vicinity of
the surface where identified from the finite element results presented in table 3. Even though the
adjustments of the WHUST allow test with a constant contact pressure and sliding velocity to be performed,
friction evolves during the experiments. This is mainly due to the consumption of the lubricant or the
accumulation of oxide scale at the contactor/specimen interface [10]. All friction factors presented in table
D

4 were then identified after a sliding length of 15 mm so that all results are comparable.
E

The friction factor m is equalled to one when no coating and no lubricant is used. This signifies the friction
PT

stress is maximum and equalled to the yield stress in pure shear in the vicinity of the contact zone. The
maximum of material transfer is logically encountered in these conditions of contact.
For tests performed without lubricant, friction is reduced by at least 30% when sol-gel coating are used. The
CE

reduction of friction is up to 50% for tests performed under small plastic strains and is equalled to 45% for
tests under high plastic strain and sol-gel loaded with h-BN.
The use of graphite base lubricant reduces friction. The beneficial effect of lubrication is more pronounced
under low plastic strain (reduction of 30% of the friction stress). Lubricant has almost no effect of friction
AC

factor for tests under high plastic strain (reduction lower than 5%).

Table 4: values of friction factor m according to the conditions of contact and to the contactor coating.
Friction factor m dispersion of graphite
no lubricant
particles in water
contactor coating p = 0.66 p = 2.33 p = 0.66 p = 2.33
non-coated surface 1 1 0.71 0.98
alumina sol-gel coating loaded with  alumina 0.5 0.72 0.55 0.69
alumina sol-gel coating loaded with h-BN 0.5 0.55 0.35 0.55

18/24
 ACCEPTED MANUSCRIPT

3.5 Discussion

All results coming from the Warm and Hot Upsetting-Sliding Tests, i.e. surface analyses, adhesive criterion
Stad and friction factors m, follow the same trend and lead to similar results and ranking of sol-gel coatings.
Test performed without coating and without lubrication lead to the maximum of material transfer.
Contactor surface is not protected: oxide scale and direct metal-metal occurring during sliding lead to
material transfer from specimen to contactor surface. This transfer results in an increase of friction.

The use of a graphite based lubricant protects contactor surface and delays the occurrence of adhesion. The

PT
coefficient of friction is then reduced. As the graphite lubricant is sprayed on the contactor surface, its
graphite layers are consumed during the sliding of the contactor along the specimen [10]. The consumption
of lubricant is more effective when test are performed under higher plastic strain since the contact area is
larger and fresh material comes up to the surface due to surface enlargement. As a consequence, the

RI
benefit of lubrication is more significant for tests performed at 0.63 of plastic strain than for test performed
at 2.33.

SC
As expected, the proposed sol-gel coatings remain firmly bonded to contactor surfaces and are then able to
protect them. They limit the amount of material transfer in terms of surface affected by adherence and in
NU
terms of thickness of the transfer layer. As a direct consequence, sol-gel coatings also reduce friction and
can be used even when no lubricant is involved. As it was expected, the reduction of friction is more
significant when the sol-gel coating is loaded with h-BN particles [53-56].
MA

4 Conclusion
D

Two alumina sol-gel coatings are developed in order to increase tool life in hot forming processes. The
coatings are made of a gamma alumina matrix loaded with alpha alumina or h-BN particles. They are
E

applied on stainless steel tools using a sol-gel procedure. The coatings are thin, with homogeneous
thickness of 200 to 300 nm.
PT

The coatings are tested on the Warm and Hot Upsetting-Sliding Test, a friction test dedicated to hot metal
forming. Tests consist in coated stainless steel tools sliding against steel specimens heated to 1100°C. The
tests are adjusted so that the sliding tool generates a plastic strain up to 2.6 in the vicinity of the specimen
CE

contact surface. Tests were performed with and without lubricants. Results show:
1- the tested sol-gel coated remain firmly bonded to the tool surface,
2- when compared to uncoated tools, the sol-gel coating decreases the friction coefficient between
AC

tools and workpieces,

3- the loading of the alumina sol-gel coating with h-BN particles significantly improves its friction
behaviour. The friction coefficient is lower, sticking phenomena are reduced and a better aspect of
the workpieces after hot forging is obtained,
4- The addition of graphite lubrication is beneficial whatever the nature of the coating.

These results are encouraging and demonstrate that sol-gel could be an adequate alternative to well-known
coatings for the production of tools used in hot forging processes. Next step could be to use commonly used
tool steels instead of stainless steels. For that, a protective pre-treatment has to be performed to prevent
corrosion of the tools during the sol-gel deposition process and so to enhance adhesion.

19/24
 ACCEPTED MANUSCRIPT

Acknowledgments

This work was carried out within the frameworks of the Carnot Arts, the Institute CNRS Research
Federation on Ground Transports and Mobility, the ELSAT2020 and the Pristimat-Interreg IV projects
supported by the European Community, the French Ministry of Higher Education and Research, the Hauts
de France Regional Council. The authors gratefully acknowledge the support of these institutions.

Bibliography:

[1] C. DEBRAS. Analyse multifactorielle de la dérive vers l'usure des outillages de frappe à froid.

PT
Mécanique des matériaux. Université de Valenciennes et du Hainaut-Cambresis, 2016. PhD
Dissertation, in French.
[2] M. HAWRYLUK, Review of selected methods of increasing the life of forging tools in hot die forging

RI
processes, Archives of Civil and Mechanical Engineering, Volume 16, Issue 4 (2016) 845-866
[3] S. LEROUX, C.BOHER, L.PENAZZI, C.DESSAIN, B.TAVERNIER, A methodology and new criteria to
quantify the adhesive and abrasive wear damage on a die radius using white light profilometry,

SC
Tribology International, 52 (2012) 40–49
[4] J. HARDELL, B. PRAKASH, High-temperature friction and wear behaviour of different tool steels during
sliding against Al–Si-coated high-strength steel. Tribology International 41 (2008) 663–671
NU
[5] J. KONDRATIUK, P. KUHN, Tribological investigation on friction and wear behaviour of coatings for hot
sheet metal forming, Wear 270 (2011) 839–849
[6] SALLIT, C. RICHARD, G. BÉRANGER, D. KIRCHER and H. MICHAUD, Experimental study of wear
behaviour of hot forging tool steels under dry conditions: 40CrMoV13 against C35E, Tribology Letters,
MA

Vol. 12, No. 3 ( 2002 ) 147–154

[7] J. HARDELL, B. PRAKASH, Tribological performance of surface engineered tool steel at elevated
temperatures, Int. Journal of Refractory Metals & Hard Materials, 28 (2010) 106–114
[8] O. BRUCELLE, G. BERNHART, Methodology for service life increase of hot forging tools, Journal of
D

Materials Processing Technology 87 (1999) 237–246

[9] D.H. KIM, B.M. KIM, C.G. KANG, Die life considering the deviation of the preheating billet temperature
E

in hot forging process, Finite Elements in Analysis and Design, 41 (2005) 1255–1269
PT

[10] E. DAOUBEN, A. DUBOIS, M. DUBAR, L. DUBAR, R. DELTOMBE, N.G TRUONG DINH, L. LAZZAROTTO,
Effects of lubricant and lubrication parameters on friction during hot steel forging, Int J Mater Form
(2008) 1223 –1226
CE

[11] D.J. JEONG, D.J. KIM, J.H.KIM, B.M. KIM, T.A.DEAN, Effect of surface treatements and lubricants for
warm forging die life, Journal of Materials Processing Technology 113 (2001) 544-550
[12] C. BOHER, S.LEROUX, L.PENAZZI, C.DESSAIN, Experimental investigation of the tribological behavior
and wear mechanisms of tool steel grades in hots tamping of a high strength boron steel, Wear 294–
AC

295 (2012) 286–295

[13] A. DUBOIS, M. DUBAR, L. DUBAR, Warm and Hot Upsetting Sliding Test: Tribology of Metal Processes
at High Temperature, Procedia Engineering (2014) 1964-1969
[14] M. DUBAR, A. DUBOIS, L. DUBAR, Wear analysis of tools in cold forging: PVD versus CVD TiN coatings
Wear 259 (2005) 1109–1116
[15] H.G. PRENGEL, W.R. PFOUTS, A.T. SANTHANAM, State of the art in hard coatings for carbide cutting
tools, Surface and Coatings Technology 102 (1998) 183-190
[16] T. WANG, J. PU, C. BO, L. JIAN, Sol–gel prepared Al2O3 coatings for the application as tritium
permeation barrier, Fusion Engineering and Design 85 (2010) 1068–1072
[17] H. PASCHKE, M. WEBER, P. KAESTNER, G. BRAEUER, Influence of different plasma nitriding treatments
on the wear and crack behavior of forging tools evaluated by Rockwell indentation and scratch tests,

20/24
 ACCEPTED MANUSCRIPT
Surface and coatings technology 205 (2010), No.5, pp.1465-1469
[18] M.H. STAIA, L. DUBAR, M. DUBAR, E.S. PUCHI-CABRERA, A. IOST, P. DEBAETS, A. DUBOIS, Mechanical
characterization of a prototype a-C:Cr,Si and its tribological behaviour at high temperature, Tribology
International, 100 (2016) 242-254
[19] M. HAWRYLUK, J. ZIEMBA, Possibilities of application measurement techniques in hot die forging
processes, Measurement, vol. 110 (2017) 284-295
[20] B. NAVINSEK, P. PANJAN, A. CVELBAR, Chraracterization of low temperature CrN ad TiN (PVD) hard
coatings, Surface and Coatings Technology, 74-75 (1995) 155-161
[21] J. LAIMER, M. FINK, C. MITTERER, H. STÖRI, Plasma CVD of alumina—Unsolved problems, Vacuum,
Volume 80, Issues 1–3 (2005) 141-145
[22] M. RAOUFI, SH. MIRDAMADI, F. MAHBOUBI, SH. AHANGARANI, M.S. MAHDIPOOR, H.ELMKHAH,
Effect of active screen plasma nitriding pretreatment on wear behavior of TiN coating deposited by

PT
PACVD technique, Applied Surface Science, Volume 258, Issue 20 (2012) 7820-7825
[23] A. CHAOUCHI, S. KENNOUR, S. D’ASTORG, M. RGUITI, C. COURTOIS, S. MARINEL, M. ALIOUAT,
Characterization of sol-gel synthesized lead-free (1 – x)Na0.5Bi0.5TiO3-xBaTiO3-based ceramics,
Journal of Alloys and Compounds, Volume 509, Issue 37 (2011) 9138-9143

RI
[24] J. RÖDEL, A.B.N. KOUNGA, M. WEISSENBERGER-EIBL, D. KOCH, A. BIERWISCH, W. ROSSNER, M.J.
HOFFMANN, R. DANZER, G. SCHNEIDER, Development of a roadmap for advanced ceramics: 2010–

SC
2025, Journal of the European Ceramic Society 29 (2009) 1549–1560
[25] J.G. CHĘCMANOWSKI, B. SZCZYGIEŁ, High temperature oxidation resistance of FeCrAl alloys covered
with ceramic SiO2–Al2O3 coatings deposited by sol–gel method, Corrosion Science, Volume 50, Issue
NU
12 (2008) 3581-3589
[26] M.A. DOMÍNGUEZ-CRESPO, A. GARCÍA-MURILLO, A.M. TORRES-HUERTA, F.J. CARRILLO-ROMO, E.
ONOFRE-BUSTAMANTE, C. YÁÑEZ-ZAMORA, Characterization of ceramic sol–gel coatings as an
alternative chemical conversion treatment on commercial carbon steel, Electrochimica Acta, Volume
MA

54, Issue 10 (2009) 2932-2940

[27] D.C.L. VASCONCELOS, J.A.N. CARVALHO, M. MANTEL, W.L. VASCONCELOS, Corrosion resistance of
stainless steel coated with sol-gel silica, Journal of Non-Crystalline Solids 273 (2000) 135-139
[28] M.L. ZHELUDKEVICH, I. MIRANDA SALVADO, M. G. S. FERREIRA, Sol–gel coatings for corrosion
D

protection of metals, Journal of Materials Chemistry, Volume 15 (2005) 5099-5111

[29] R. DI MAGGIO, L. FEDRIZZI, S. ROSSI, P. SCARDI, Dry and wet corrosion behaviour of AISI 304 stainless
E

steel coated by sol-gel ZrO2-CeO2 films, Thin Solid Films, 286 (1996) 127-135
[30] S.K. TIWARI, R.K. SAHU, A.K. PRAMANICK, R. SINGH, Development of conversion coating on mild steel
PT

prior to sol gel nanostructured, Al2O3 coating for enhancement of corrosion resistance, Surface &
Coatings Technology 205 (2011) 4960–4967
[31] D. WANG, G. P. BIERWAGEN, Sol–gel coatings on metals for corrosion protection, Progress in Organic
CE

Coatings 64 (2009) 327–338

[32] I. MEHNER, H.-W. ZOCH, W. DATCHARY, G. PONGS, H. KUNZMANN, Sol-Gel Coatings for High Precision
Optical Molds, CIRP Annals - Manufacturing Technology, Volume 55, Issue 1 (2006) 589-592
AC

[33] J. RAMS, A. UREÑA, M.D. LÓPEZ, A.J. LÓPEZ, Characterisation of multilayered sol–gel silica coatings on
aluminium–SiC composites, Surface and Coatings Technology, Volume 201, Issue 6 (2006) 3715-3722
[34] A.O. OLOFINJANA, J.M. BELL, A.K. JAMTING, Evaluation of the mechanical properties of sol–gel-
deposited titania films using ultra-micro-indentation method, Wear, Volume 241 (2000) 174–179
[35] A.R. PHANI, F.J. GAMMEL, T. HACK, Structural, mechanical and corrosion resistance properties of
Al2O3–CeO2 nanocomposites in silica matrix on Mg alloys by a sol–gel dip coating technique, Surface
& Coatings Technology, Volume 201 (2006) 3299–3306
[36] G. RUHI, O.P. MODI, I.B. SINGH, Pitting of AISI 304L stainless steel coated with nanostructured sol–gel
alumina coatings in chloride containing acidic environments, Corrosion Science, Volume 51, (2009),
3057–3063
[37] G. RUHI, O.P. MODI, A.S.K. SINHA, I.B. SINGH, Effect of sintering temperatures on corrosion and wear
properties of sol–gel alumina coatings on surface pre-treated mild steel, Corrosion Science, Volume

21/24
 ACCEPTED MANUSCRIPT
50 (2008) 639–649
[38] THOMAS HÜBERT, STEFAN SVOBODA, BERND OERTEL, Wear resistant alumina coatings produced by a
sol–gel process, Surface & Coatings Technology, Volume 201 (2006) 487–491
[39] THOMAS HÜBERT, JACK SCHWARZ, BERND OERTEL, Sol-gel alumina coatings on stainless steel for
wear protection, Journal of Sol-Gel Science and Technology, Volume 38 (2006) 179–184
[40] W. ZHANGA, W. LIU, C. WANG, Characterization and tribological investigation of sol-gel Al2O3 and
doped Al2O3 films, Journal of the European Ceramic Society, Volume 22 (2002) 2869–2876
[41] B. FELDE, A. MEHNER, J. KOHLSCHEEN, R. GLÄBE, F. HOFFMANN, P. MAYR, Deposition of alumina
coatings on monocrystalline diamonds by sol-gel techniques, Diamond and related Materials 10
(2001) 515-518
[42] A. DUBOIS, J. QUDIN, P. PICART, Elastoplastic finite element analyses of an upsetting-sliding test for
the determination of friction at medium and high contact pressure, Tribology international 29 (1996)

PT
603-613
[43] L. DUBAR, A. DUBOIS, M. DUBAR, Friction Analysis in Bulk Metal Forming, in 60 Excellent Inventions in
Metal Forming, A.E. Tekkaya, W. Homberg, A. Brosius, Eds., Springer Vieweg (2015) 23-30
[44] DUBOIS, D. PATALIER, P. PICART, J. OUDIN, Optimization of the upsetting-sliding test parameters for

RI
the determination of friction laws at medium and high contact pressures, Journal of materials
processing technology 62 (1996) 140-148

SC
[45] L. LAZZAROTTO, L. DUBAR, A. DUBOIS, P. RAVASSARD, J. OUDIN, Three selection criteria for the cold
metal forming lubricating oils containing extreme pressure agents, Journal of Materials Processing
Technology 80 (1998) 245-250
NU
[46] A. DUBOIS, Metal Forming and Lubrication, In: Saleem Hashmi (editor-in-chief). Reference Module in
Materials Science and Materials Engineering, Oxford: Elsevier (2016) 1-10
[47] L DUBAR, B HAUW, A DUBOIS, J OUDIN, Numerical identification of coating bulk behaviour in metal
forming, Computers ad Structures, volume 79, Issue 10 (2001) 1027-1037
MA

[48] R. LEVEQUE, Aciers à outils – Mise en œuvre, In: Les Techniques de l’Ingénieur, M4587 V2, 2013 (in
French)
[49] A. MARSAL, F. ANSART, V. TURQ, J.P. BONINO, J.M. SOBRINO, Y.M. CHEN, J. GARCIA, Mechanical
properties and tribological behavior of a silica or/and alumina coating prepared by sol-gel route on
D

stainless steel, Surface and Coatings Technology 237 (2013) 234-240

[50] L.C. LIM, M.O. LAI, J. MA, Tempering of AISI 403 stainless steel, Materials Science and Engineering,
E

A171 (1993) 13-19

[51] N. ÖZER, J.P. CRONIN, Y.J. YAO, A.P. TOMSIA, Optical properties of sol-gel deposited Al2O3 films, Solar
PT

Energy Materials & Solar Cells 59 (1999) 355-366

[52] I. PIWONSKI, K. SOLIWODA, The effect of ceramic nanoparticles on tribological properties of alumina
sol–gel thin coatings, Ceramics International 36 (2010) 47–54
CE

[53] J. EICHLER, C. LESNIAK, Boron nitride (BN) and BN composites for high-temperature applications,
Journal of the European Ceramic Society 28 (2008) 1105–1109
[54] Y. KIMURA, T. WAKABAYASHI, K. OKADA, T. WADA, H. NISHIKAWA, Boron nitride as a lubricant
AC

additive, Wear, 232 (1999) 199–206

[55] A. GRECO, K. MISTRY, V. SISTA, O. ERYILMAZ, A. ERDEMIR, Friction and wear behaviour of boron based
surface treatment and nano-particle lubricant additives for wind turbine gearbox applications, Wear,
271 (2011) 1754–1760
[56] Z. PAWLAK, T. KALDONSKI, R. PAI, E. BAYRAKTAR, A. OLOYEDEF, A comparative study on the
tribological behaviour of hexagonal boron nitride (h-BN) as lubricating micro-particles—An additive in
porous sliding bearings for a car clutch, Wear, 267 (2009) 1198–1202
[57] J.J. FUENTES-GALLEGO, E. BLANCO, L. ESQUIVIAS, Ellipsometric characterization of an AISI 304
stainless steel protective coating, Thin Solid Films, 301 (1997) 12–16
[58] E. NOURI, M. SHAHMIRI, H.R. REZAIE, F. TALAYIAN, A comparative study of heat treatment
temperature influence on the thickness of zirconia sol–gel thin films by three different techniques:
SWE, SEM and AFM, Surface & Coatings Technology, 206 (2012) 3809–3815

22/24
 ACCEPTED MANUSCRIPT
[59] E. DAOUBEN, Comprehension des mécanismes d’endommagement en surface des outils de forge a
chaud – effet des lubrifiants, Université de Valenciennes et du Hainaut-Cambresis, 2009. PhD
Dissertation, in French.
[60] ES PUCHI-CABRERA, MH STAIA, JD GUÉRIN, J LESAGE, M DUBAR, D CHICOT, An experimental analysis
and modeling of the work-softening transient due to dynamic recrystallization, International Journal
of Plasticity 54 (2014) 113-131
[61] T. TISON, A. HEUSSAFF, F. MASSA, I. TURPIN, R.F. NUNES, Improvement in the predictivity of squeal
simulations: Uncertainty and robustness, Journal of Sound and Vibration, Volume 333, Issue 15
(2014) 3394–3412

PT
RI
SC
NU
MA
E D
PT
CE
AC

23/24
 ACCEPTED MANUSCRIPT

Highlights

- alumina sol-gel coating doped with a-alumina and hBN particles are studied
- friction tests were performed à 1100°C
- both sol-gel coatings reduce adhesive wear
- sol-gel coating doped with hBN strongly reduces friction

PT
RI
SC
NU
MA
E D
PT
CE
AC

24/24