Author’s Accepted Manuscript

Experimental study of the effect of coating

thickness and substrate roughness on tool wear
during turning

M. Bar-Hen, I. Etsion

[Link]/locate/jtri

PII: S0301-679X(16)30435-2
DOI: [Link]
Reference: JTRI4450
To appear in: Tribiology International
Received date: 28 September 2016
Revised date: 4 November 2016
Accepted date: 6 November 2016
Cite this article as: M. Bar-Hen and I. Etsion, Experimental study of the effect of
coating thickness and substrate roughness on tool wear during turning, Tribiology
International, [Link]
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 galley proof before it is published in its final citable 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.
 Experimental study of the effect of coating thickness and substrate
roughness on tool wear during turning

M. Bar-Hena, I .Etsionb*
a
R&D Tool Division, Turning Department, Iscar LTD, Tefen 2495900, Israel
b
Department of Mechanical Engineering, Technion, Haifa 32000, Israel

*Corresponding author: etsion@[Link]

Abstract

The effect of coating thickness and substrate roughness on the tool wear in turning is
studied experimentally. TiAlN coating of various thicknesses is applied on tungsten
carbide (WC) tool substrate having various surface roughnesses. The tool wear is
measured following a fixed cutting distance and speed tests. In general, when plotting
the wear vs. coating thickness alone or vs. substrate roughness alone a trend of
decreasing wear with either increasing coating thickness or substrate roughness is
observed. However, the scatter of the results is very large. On the other hand,
excellent fit is obtained when the wear results are plotted vs. the dimensionless ratio
t/Rsu where t is the coating thickness and Rsu is the average radius of curvature of the
rough substrate asperities. An interesting correlation is shown between the present
experimental results and some theoretical models for yield inception of a coated
spherical asperity.

Keywords: Wear; Cutting tools; Coating; Surface roughness

1
Nomenclature

D depth of cut
E young's modulus
F feed rate
H hardness
R radius of a spherical asperity substrate
Rsu, r average radius of curvature of the rough substrate asperities
V cutting speed
VB width of flank wear
Y yield stress
t coating thickness

β roughness parameter, β = Rsu ∙ ∙

η area density of the asperities

Poisson's ratio

standard deviation of asperities heights

Subscripts
co coating
su substrate

2
1. Introduction

In turning, a cutting tool (see Fig. 1(a)) with a single cutting edge is used to remove
material from a rotating workpiece to generate a cylindrical shape. The primary
motion is provided by rotating the workpiece and the feed motion is achieved by
moving the cutting tool slowly in a direction parallel (also called "longitudinal
turning") or perpendicular (also called "facing") to the axis of rotation of the
workpiece [1,2].

(a)

Cutting tool
Holder

(b)

Cutting
Measured edge
area

Fig. 1: (a) General Scheme of turning tool. (b) Cutting tool and area on which the
substrate roughness and coating thickness measurements were made

During cutting operation, one or more wear types may occur, the most relevant for
this work are the crater and flank wear, see Fig. 2. Crater wear appears on the rake

3
face in the area where the tool is in contact with the chip. High temperatures and
pressures are typical for this zone. During operation, the crater gradually increases
and the edge becomes sharper. This causes an increasing stress on the edge, which
eventually leads to edge breakage. Flank wear appears on the flank face (see Fig.
1(b)) where the tool is in sliding contact with the newly cut workpiece material. The
wear normally starts at the cutting edge and grows continually away from the edge.
Flank wear is the most common type of wear and the preferred wear type, as it offers
predictable and stable tool life [3–6]. This research is specifically directed to flank
wear type.

VB

Fig. 2: Example of crater (left) and flank (right) wear in a cutting tool

A useful way of presenting the amount of flank wear is by the parameter VB, which is
defined as the width of the wear scar measured from the cutting edge as shown in Fig.
2. More detailed explanations are provided in Ref. [7].

The most common type of cutting tools used in industrial applications today consists
of a substrate material with one or more outer layers of coating. The main purpose of
applying the coating is increasing wear resistance and performance during work. The
coating thicknesses range from microns to several millimeters [8].

In several studies on turning [9–17], which investigated the wear of coated cutting
tools compared to that of uncoated ones, it was found that the wear rate of the latter is
much higher. Some experimental studies [18–22] examined the effect of flank hard
coating thickness on wear resistance in cutting tools. Different coatings were used
deposited by two methods: physical vapor deposition (PVD) [18,21] and chemical
vapor deposition (CVD) [19,20,22]. The range of the coating thicknesses was between
1 to 29 µm, cutting speed was between 34.8 to 335 m/min, feed was between 0.15 to
0.25 mm/rev and depth of cut was between 0.5 to 2 mm. Workpiece material was
different at each study. The obtained results were ambiguous: in [18,22] the tool wear
resistance continually increased with increasing coating thickness. In [19] the tool life
was longer with thickness between 7.5 to 10.5 µm, in [20] the flank wear resistance
increased with increasing coating thickness up to 6 μm and decreased drastically at
coating thickness of 10 μm. In [21] an optimum coating thickness of 3.5 µm was

4
found that maximized the tool life (or wear resistance). In each of the above five
studies [18–22] only a limited number of 3 to 5 coating thickness values was tested
over the entire thicknesses range. Also, in these references no indication on test
repetition for reliable statistics is provided except for two repetitions in [22].

Some experimental studies [23,24] examined the effect of substrate surface roughness
and coating thickness on wear using a pin-on-disc tribometer. The pins were spheres
of alumina and 100C6 steel [23] or bearing AISI 52100 steel [24] loaded against discs
of ferritic 35CD4 steel or AISI 440C steel, respectively. The discs had surface
roughness in the range of 0.01-1 µm Ra and were coated with titanium nitride (TiN)
[23] or Molybdenum disulfide/Titanium (MoS2/Ti) [24] having the range of 0.1-5 µm
thickness. It should be noted here that Refs. [23,24] do not concern cutting tools or
turning in particular. The reason for citing these references is that the surface
roughness was determined by the single parameter Ra, which does not fully represent
surface roughness. According to McCool [25], surface roughness should be more
accurately described by the following three parameters: standard deviation of
asperities heights s , radius of curvature of their summits r, and the area density of the
asperities .

No theoretical models on the effect of coating thickness or surface roughness on wear

resistance could be found in the literature. The situation is somewhat better regarding
the resistance to plastic yielding see e.g., Refs. [26–29]. These studies deal with a
coated sphere that simulates a single spherical asperity of a coated rough surface and
show the effect of the ratio t/R on the resistance to plastic yielding, where t is the
coating thickness and R is the radius of the spherical substrate. The aim of these
studies is to develop models, which will enable scientific approach to optimization of
coating thickness, differently from the commonly used less accurate trial and error
approach. It was found in these models (see e.g., [26]) that for a given set of
mechanical properties such as the moduli of elasticity of the coating and substrate co ,
su and their corresponding yield strengths co , su the most influential parameter on
the resistance to plastic yielding is the ratio t/R.

In some cases of very thin hard coatings weakening effect was observed [26,27,30–
34] where the tribological performance of a coated case is worse compared to that of
an uncoated one. Ref. [34] is of particular interest, showing cases in turning where
coated cutting tools had lower wear resistance than uncoated ones. It was found in
[26] and [30] that the range of dimensionless thicknesses t/R corresponding to such
weakening effect depends on the dimensionless parameters co ⁄ su and su ⁄ su .

As can be seen from the above literature review coated cutting tools were mostly
found to have significant higher wear resistance than uncoated ones. However, no
conclusive results can be found as to the effect of the coating thickness on flank wear
resistance. In addition, no reference could be found in the literature to the effect of
substrate surface roughness on wear resistance in turning. Some theoretical studies
exist on the effect of coating thickness on the resistance to plastic yield of a single

5
asperity of a rough surface. In these studies it was found that the ratio of coating
thickness over the radius of curvature of the asperity substrate, and not the coating
thickness alone, is the most significant parameter. Hence, the goal of this paper is to
attempt, for the first time, an experimental study, which focuses on the combined
effect of both coating thickness and tool substrate surface roughness on the tool wear
in turning. This may help in optimization of cutting tool design by controlling the tool
substrate roughness prior to selecting a preferred coating thickness based on this
roughness.

2. Experimental details

2.1 Specimens preparation

Tool substrate specimens were prepared from a Tungsten carbide (WC) powder
93.81% with Cobalt (Co) binder 6% and Chromium carbide (Cr3C2) additive 0.19%,
by the common pressing and sintering process.

2 [mm]
3 4 5
6
2
2 [mm]

30°
1

Fig. 3: Scheme of the scans to measure surface roughness of tool flank substrate

After the sintering stage, the surface roughness of the tool flank surfaces was
measured using the method suggested by McCool [25]. The measurements were done
on an area of 2mm by 2mm where the flank wear is supposed to be developed (see the
indicated measured area in Fig. 1(b)), using optical profilometer (InfiniteFocusSL,
Austria) with X10 lens. Since the tools surface roughness is anisotropic, six linear
scans with angular interval of 30° were made (see Fig. 3) to obtain the standard
deviation of asperities heights, , the average radius of curvature of their summits, r,
and the area density of the asperities, (see also [35]).

In order to expand the range of the above roughness parameters, some of the tools
flank surfaces were polished peripherally using brushing-polishing machine (Gerber
BP-MX, USA) shown schematically in Fig. 4. The polishing brush A is made of

6
natural double drawn Tampico fibers of 160 mm length. The fibers are manually
smeared before any polishing operation using 50 karat diamond abrasive polishing
paste “Diastar” (Gerber, Switzerland) having grains size in the range of 5.5 to 8 m.
The tools are inserted into a dedicated accessory B designed to reveal only the area
intended to be polished and hide all other areas. The brush and the accessory rotate in
the same direction while the accessories holder C rotates in the opposite direction.
The polishing time for each flank surface was 10 minutes at a rotational brush speed
of 300 revolutions per minute. After the polishing operation the tools were washed
manually with tap water to remove remnant of the paste particles.

Fig. 4: General scheme of the polishing machine (A - brush, B – accessory which

hold the tools, C –accessories holder)

Table 1 presents the range of the measured roughness parameters and their
dimensionless product β= for the as received and polished specimens. This last
parameter β is extensively used in theoretical modeling of contacting rough surfaces,
see e.g. [36]. As can be seen in Table 1 the polished specimens have much lower
standard deviation of asperities heights and higher radius of curvature of asperities
summit, as would be expected for smoother surfaces.

Subsequently, coating of Titanium Aluminum Nitride (TiAlN) was deposited by PVD

using cathodic arc deposition at temperature of about 5000 C. The coating of the entire
tool samples to be tested was done in a single batch with the same coating parameters.
The range of coating thicknesses was obtained by placing the samples on several
racks with different spacing and in various different locations in the coating chamber.

The coating thickness was measured on the same area of 2mm by 2mm where the
flank wear is supposed to be developed (see Fig. 1(b)), using an X-ray machine
(CMI900, UK). The measurement method is based on bombarding the sample with
excitation photons using an X-ray tube, measuring the secondary or fluoresced X-rays

7
with a detector and determining the thickness and/or composition of the coating via
the X-ray’s intensity. The error in this coating thickness measurement technique is
±5%.

Adhesion of the coating to the substrate was qualitatively analyzed, on a randomly

selected sample of 5 tools, by a Vickers indentation test machine (AR90, FV-700,
Japan) with activation force of 200 N. This is the standard method used by the tool
producer for the coating to be tested. The indenter was a square-based diamond
pyramid with angle of 136°. The indentation trace was analyzed by optical
microscope and good adhesion (see illustration in Fig. 5) of the coating to the
substrate was found.

Table 1 The range of different roughness parameters for the as received and polished
specimens

Range of values for the Range of values for tools

Measured parameter
as received tools with peripheral polishing

Standard deviation of
0.037-0.4 0.0079-0.052
asperities heights, [µm]

Average radius of
curvature of asperities 26.88-79.41 94.84-145.83
summit, r [µm]

Area density of the

0.001-0.0031 0.0017-0.0028
asperities, [ ]

Roughness parameter, β 0.004-0.045 0.0018-0.012

The mechanical properties of the substrate and coating materials (obtained from the
data base of the tool producer) are presented in Table 2.

Table 2 Mechanical properties of the substrate and coating materials

E [Gpa] Y [GPa] H (GPa)

Substrate 566 5.816 17.11 0.197

Coating 550 10 29.42 0.3

8
 Fig. 5: General illustration of good adhesion

2.2 Turning tests

Turning tests using CNC lathe (Okuma LB, America) were performed in a wet cutting
environment with pressurized coolant of 29 MPa in order to diminish crater wear.

Each tool underwent 25 facing passes without being removed from the holder. Each
pass started at a workpiece diameter of 244 mm and ended at a workpiece diameter of
80 mm (see Fig. 6). The cutting velocity, feed rate and depth of cut were selected
according to the tool geometry (and data base of the tool producer) and were
maintained constant (see Table 3). Constant cutting velocity was maintained by
controlling the spindle RPM according to the variable diameter of the workpiece in
each pass. The turning time for one pass of each tool was 1.26 min. resulting in total
turning time 31.5 min. for each tool. After the conclusion of all 25 passes, the tool
was removed from the holder and its flank wear was measured. Removing each tool
from the holder only once after many passes ensures that wear development is not
affected by remounting the tool in non-identical positions.

244 [mm]

Velocity

Feed

80 [mm]

Depth
of cut

Fig. 6: Schematic description of a facing pass

9
Table 3 Turning parameters

Parameters Symbol Unit Value

Cutting velocity V m/min 220

Feed F mm/rev 0.15

Depth of cut D mm 2

The cylinder workpiece material for the present experiments was AISI 4340 steel of
244 mm in diameter and a length of 980 mm (It was hardened by heating to 830-
860oC and oil cooling. Then it was tempered by heating to about 550oC and air
cooling). Its chemical composition and mechanical properties are shown in Table 4.

Table 4 Chemical composition (%) and mechanical properties of the AISI 4340 steel
workpiece

C Si Mn P S Cr Mo Ni Fe
(Carbon) (Silicon) (Manganese) (Phosphorus) (Sulfur) (Chromium) (Molybdenum) (Nickel) (ferrum)

0.4 0.25 0.7 0.035 0.04 0.8 0.25 1.85 95.675

E [Gpa] Y [MPa] H (MPa)

205 710 960 0.29

Burr and oxide layers were removed from the outer surface of each workpiece prior to
the actual turning.

3. Results and discussion

Overall 50 tools were tested using one work corner of each tool. 17 tools of the 50
were peripherally polished and 33 tools remained as received. For all the 50 tested
tools the coating thickness t varied within a range of 1.87 to 5.59 μm. The average
radius of curvature of the substrate roughness varied from 26.88 to 145.83 μm (see
Table 1). Hence, the range of the parameter t/Rsu obtained for all the tested
combinations is from 0.015 to 0.157. The flank wear width of the cutting tools (see
Fig. 2) was measured by an optical microscope according to the explanation in [7] and
was found in the range from 0.131 to 0.496 mm.

10
Figures 7 and 8 present two extreme tested cases of high and low flank wear width,
respectively. Part (a) of each of these figures shows a photograph of the zone on the
flank containing the wear scar along with the flank surface just below that scar. Part
(b) in each figure presents a profilometer scan starting on the flank surface below the
wear scar and running upward perpendicular to the cutting edge, at the center of the
depth of cut of the flank. As can be seen, the depth of cut in the two figures was D=2
mm according to the preset test parameters. The measured coating thickness t of the
tools in Figs. 7 and 8 was 2.25 and 5.28 µm, respectively. The average radius of
curvature Rsu of the substrate asperities in Figs. 7 and 8 was 132.5 and 50.8 µm,
respectively. In Fig. 7(b) the profilometer scan started at about 1.8 mm below the
cutting edge. It shows for the wear scar a local width of about 0.5 mm and a
maximum depth of about 0.045 mm at the cutting edge. Also, some protrusions of the
order of 10 m are seen on the surface profile below the wear scar (see the bold
arrow). The reason for their presence may be chips or tool particles, which adhere to
the flank during the turning.

D = 2 mm

VB =
0.496 mm (a)

1 mm

(b)

Fig. 7: Tool with high flank wear (a) photograph of the zone on the flank containing
the wear scar and (b) profilometer scan perpendicular to the cutting edge.

In Fig. 8(b) the profilometer scan started at about 2.2 mm below the cutting edge
showing for the wear scar a local width of about 0.1 mm and a maximum depth of
about 0.009 mm at the cutting edge. Here again some protrusions, similar to these in

11
Fig. 7(b), are seen on the surface profile below the wear scar. Note that in the case of
Fig. 7 with the high wear the ratio t/Rsu is 0.017 while in the case of Fig. 8 with the
low wear this ratio is 0.1. In both cases the maximum depth of the wear scar indicates
that the coating was totally removed near the cutting edge.

D = 2 mm

VB =
0.13 mm
(a)

1 mm

(b)

Fig. 8: Tool with low flank wear (a) photograph of the zone on the flank containing
the wear scar and (b) profilometer scan perpendicular to the cutting edge.

Figures 9 and 10 present the results of the flank wear VB of all the 50 tested tools vs.
the coating thickness t and the radius of curvature Rsu, respectively. In each figure the
best fit of the 50 test points is also shown by the solid line and the goodness of fit R2
is indicated. In Fig. 9 the general trend of the results is decreasing wear with
increasing coating thickness, similar to the trend reported in Refs. [18,22]. However,
the scatter of the results is very large and the goodness of fit is only 57%, indicating a
somewhat poor correlation. The trend in Fig. 10 shows increasing flank wear VB with
increasing average radius of curvature of the substrate roughness Rsu (smoother
surface). Here the goodness of fit is somewhat better being 64%, which still indicates
a poor correlation.

12
 0.55

0.5
Flank wear VB [mm]
0.45

0.4

0.35
R² = 0.5731
0.3

0.25

0.2

0.15

0.1
1 2 3 4 5 6
Coating thickness t [μm]

Fig. 9: Flank wear VB vs. Coating thickness t

0.55

0.5
Flank wear VB [mm]

0.45

0.4

0.35
R² = 0.631
0.3

0.25

0.2

0.15

0.1
20 40 60 80 100 120 140
Substrate roughness Rsu [μm]

Fig. 10: Flank wear VB vs. Substrate roughness presented by Rsu

When the flank wear VB was plotted vs. the other roughness parameters (not shown
here) namely, standard deviation of asperities heights s, area density of the asperities
, and roughness parameter β, the goodness of fit R2 was 27%, 4% and 2-3%,

13
respectively. It seems that the correlation between flank wear and coating thickness
alone or substrate roughness parameters alone is poor. Hence, these parameters by
themselves are not good candidates for reliable prediction of the flank wear VB.

As indicated in the Introduction, no theoretical models on the effect of coating

thickness or surface roughness on wear resistance could be found in the literature.
Assuming that there may be some correlation between wear resistance and resistance
to plastic yield inception, which was recently modeled (see e.g., [26–29]), it was
attempted to plot the flank wear vs. the dimensionless coating thickness t/Rsu. This
parameter was found very useful in modeling the effect of coating thickness on yield
inception of a single coated spherical asperity [26–29]. The results are presented in
Fig. 11 and show a very clear trend of decreasing flank wear VB with increasing ratio
t/Rsu. As can be seen in Fig. 11 the goodness of fit is 94%, which is indeed much
higher than that in Figs. 9 and 10, and indicates good correlation.

0.55

0.5
Flank wear VB [mm]

0.45

0.4

0.35

0.3

0.25
R² = 0.9405
0.2

0.15

0.1
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18
Dimensionless thickness t/Rsu

Fig. 11: Flank wear VB vs. Dimensionless thickness t/Rsu

It can be seen in Fig. 11 that most of the sampling points are within the range of t/Rsu
from 0.02 to 0.1 and, in this range, show exponential decrease of VB with increasing
t/Rsu. The expression of the best fit curve shown in Fig 11 for the relation between
t/Rsu and the flank wear VB in this range has the form:

0.03 1∙( ⁄
-0.58
su (1)

14
It can also be seen in Fig. 11 that for t/Rsu above 0.1, the VB approaches
asymptotically a constant value of 0.131 mm.

It is of interest to examine the results shown in Fig. 11 in light of the findings in the
models for plastic yield inception of a single asperity of rough surface with hard
coating [26], [29] and [30]. In these models the effect of the dimensionless coating
thickness having the form t/R (where t is the coating thickness and R the radius of the
spherical asperity substrate) on the value of the critical load to cause yield inception
and on the location of this yield was investigated. For very small values of t/R the
yield inception occurs within the substrate and the resistance to such yield may be
very small. A weakening effect of reducing the resistance to yield inception compared
to an uncoated case was found in [30] below a transition value (t/R)T given by:

1.41
( ⁄ )T (2)
( su ⁄ su )0.958

The maximum weakening effect occurs at a dimensionless thickness (t/R)MW given by:

1.066
(⁄ ) 0.225
(3)
( su ⁄ su )( co ⁄ su )

Fig. 12: Typical locations of the onset of plastic yielding taken from [26]

15
An optimum dimensionless coating thickness (t/R)p, which provides maximum
resistance to plastic yielding was found in [26]. According to this model increasing
the dimensionless coating thickness above this value causes the yield inception to
occur within the hard coating and provides better protection against plastic yield in
the substrate [29]. An expression for (t/R)p, which depends on material properties
only, is provided in [29] in the form:

0.536 -1.608 0.594

( ⁄ )p 2.824 (
co
) ( co
) ( su ) (4)
su co su

Figure 12 summarizes the theoretical three locations of yield inception according to

the range of the dimensionless coating thickness (see Refs. [26] and [30]). When
t/R<(t/R)T the plastic yield inception in the models occurs within the substrate in
location 1, when (t/R)T<t/R<(t/R)p the location of yield inception is in location 2 at
the substrate side of the coating/substrate interface and when t/R>(t/R)p the yield
inception occurs in location 3 within the coating slightly below the contact area.

0.55
Location 1 Location 2 Location 3
0.5
Flank wear VB [mm]

0.45 0.0175 0.067

su T su p

0.4

0.35

0.3

0.25

0.2

0.15

0.1
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
0.011 Dimensionless thickness t/Rsu
su

Fig. 13: Flank wear VB vs. Dimensionless thickness t/Rsu with the range of values
corresponding to the locations of onset of plastic yielding shown in Fig. 12.

16
Using the material properties of the coating and substrate from Table 2 and replacing
R by Rsu in Eqs. (2) - (4), the following values are obtained for the tested tools in the
present experiment: ( ⁄ su = 0.011, ( ⁄ su T = 0.0175 and ( ⁄ su = 0.067. The
vertical dashed lines in Fig. 13 mark these three values upon the results of the flank
wear VB as a function t/Rsu from Fig. 11. Fig. 13 also demonstrates the range of t/Rsu
values corresponding to the theoretical three different locations of onset of plastic
yielding shown in Fig. 12. It can be seen that the tools with maximum flank wear
belong to the range of location 1 where the theoretical first plastic yielding occurs in
the substrate. The tools with the minimum flank wear belong to the range of location
3 where the first plastic yielding in the model occurs in the coating. Hence, a good
correlation seems to exist between the theoretical results of resistance to plastic yield
inception and the present experimental results of resistance to tool flank wear. In
order to achieve minimum flank wear the recommended value of t/Rsu should be about
50% larger than ( ⁄ su p .

Conclusion

The effect of coating thickness, t, and substrate roughness, represented by the average
radius of curvature of the rough substrate asperities, Rsu, on the tool wear in turning
was studied experimentally. Tool specimens were prepared from WC powder by
conventional industrial processes and coated with various thicknesses of TiAlN using
PVD. The surface roughness of the tool flank was measured prior to the coating
deposition and some of these tools were polished to obtain a larger range for the
dimensionless coating thickness t/Rsu. Flank wear VB was measured on 50 tools
following 31.5 min. of turning under fixed operating conditions for each tool. When
plotting the measured flank wear results vs. coating thickness alone or vs. substrate
roughness alone a general trend of decreasing wear with either increasing coating
thickness t or decreasing Rsu (increasing substrate roughness) was observed. However,
the scatter of the results was very large with a low goodness of fit being only 64% or
less, which indicates poor correlation. A much better correlation, with 94% goodness
of fit, was obtained when the wear results were plotted vs. the dimensionless ratio
t/Rsu. It was found that the flank wear VB decreases exponentially with increasing
t/Rsu up to a certain value of this dimensionless coating thickness above which VB
becomes a small constant. This finding may be very helpful in selecting a preferred
coating thickness based on the measured substrate roughness prior to coating.

An attempt was made to examine the present experimental results in light of recently
published theoretical models dealing with plastic yield inception of a single spherical
coated asperity. An interesting and surprising correlation was shown between the
effect of the dimensionless coating thickness on the present resistance to wear and on
the resistance to plastic yield inception in the theoretical models.

Finally, a word of caution is in place. The present experiment was limited to a certain
set of materials and to fixed operating conditions such as cutting speed, feed and
depth of cut. Other tool materials and operating conditions may significantly affect

17
the wear development see e.g., [36–39]. Hence, more research work is required in the
future to explore a larger range of these factors. It should also be clarified that the
surprising correlation presented here, between the effects of the dimensionless coating
thickness on the experimental flank wear resistance and on the theoretical resistance
to yield inception of a coated asperity, does not mean that the two physical
mechanisms are the same. Much more work, which is out of the scope of this paper, is
required to identify the exact failure mechanisms of the coated tools. Such work
should include issues such as adhesion and thermal effects, residual stresses etc.
Hopefully, the present results will trigger enough curiosity among researchers in the
field to further explore this interesting correlation.

Acknowledgment

The authors would like to graciously thank Iscar LTD for providing its facilities to
carry out the research work. This paper is part of IEA AMT IA technical activities.

References

[1] G.T. Smith, Turning and Chip-breaking Technology, in: Cutting Tool
Technology Industrial Handbook, Springer, London, 2008: pp. 34–61.
doi:10.1007/978-1-84800-205-0.

[2] F. Klocke, Fundamentals of Cutting, in: Manufacturing Processes, Springer,

Berlin Heidelberg, 2011: pp. 39–71.

[3] J. Hu, Y.K. Chou, Characterizations of cutting tool flank wear-land contact,
Wear. 263 (2007) 1454–1458. doi:10.1016/[Link].2007.01.080.

[4] S.K. Khrais, Y.J. Lin, Wear mechanisms and tool performance of TiAlN PVD
coated inserts during machining of AISI 4140 steel, Wear. 262 (2007) 64–69.
doi:10.1016/[Link].2006.03.052.

[5] A. Mkaddem, A. Ben Soussia, M. El Mansori, Wear resistance of CVD and

PVD multilayer coatings when dry cutting fiber reinforced polymers (FRP),
Wear. 302 (2013) 946–954. doi:10.1016/[Link].2013.03.017.

[6] S. Chinchanikar, S.K. Choudhury, Predictive modeling for flank wear

progression of coated carbide tool in turning hardened steel under practical
machining conditions, The International Journal of Advanced Manufacturing
Technology. 76 (2014) 1185–1201. doi:10.1007/s00170-014-6285-6.

[7] Tool-life testing with single-point turning tools, ISO 3685. (1993).

[8] S. Hogmark, S. Jacobson, M. Larsson, Design and evaluation of tribological

coatings, Wear. 246 (2000) 20–33. doi:10.1016/S0043-1648(00)00505-6.

18
[9] F.A. Soliman, O.A. Abu-Zeid, M. Merdan, On the improvement of the
performance of high speed steel turning tools by TiN coatings, Wear. 119
(1987) 199–204. doi:10.1016/0043-1648(87)90109-8.

[10] W. König, R. Fritsch, Physically vapor deposited coatings on cermets:

performance and wear phenomena in interrupted cutting, Surface and Coatings
Technology. 49 (1991) 316–324. doi:10.1016/0257-8972(94)90248-8.

[11] A.A. Minevich, Wear of cemented carbide cutting inserts with multilayer Ti-
based PVD coatings, Surface and Coatings Technology. 53 (1992) 161–170.
doi:10.1016/0257-8972(92)90118-T.

[12] S.C. Lim, C.Y.H. Lim, K.S. Lee, The effects of machining conditions on the
flank wear of TiN-coated high speed steel tool inserts, Wear. 181-183 (1995)
901–912. doi:10.1016/0043-1648(95)90214-7.

[13] C.Y.H. Lim, S.C. Lim, K.S. Lee, Wear of TiC-coated carbide tools in dry
turning, Wear. 225-229 (1999) 354–367. doi:10.1016/S0043-1648(98)00366-4.

[14] A. Devillez, F. Schneider, S. Dominiak, D. Dudzinski, D. Larrouquere, Cutting

forces and wear in dry machining of Inconel 718 with coated carbide tools,
Wear. 262 (2007) 931–942. doi:10.1016/[Link].2006.10.009.

[15] R.P. Martinho, F.J.G. Silva, A.P.M. Baptista, Wear behaviour of uncoated and
diamond coated Si3N4 tools under severe turning conditions, Wear. 263 (2007)
1417–1422. doi:10.1016/[Link].2007.01.048.

[16] Y.J. Lin, A. Agrawal, Y. Fang, Wear progressions and tool life enhancement
with AlCrN coated inserts in high-speed dry and wet steel lathing, Wear. 264
(2008) 226–234. doi:10.1016/[Link].2007.03.007.

[17] K. Aslantas, I. Ucun, A. Çicek, Tool life and wear mechanism of coated and
uncoated Al2O3/TiCN mixed ceramic tools in turning hardened alloy steel,
Wear. 274-275 (2012) 442–451. doi:10.1016/[Link].2011.11.010.

[18] E. Posti, I. Nieminen, Influence of coating thickness on the life of TiN-coated

high speed steel cutting tools, Wear. 129 (1989) 273–283.

[19] J.-D. Kim, Effect of Carbide and Oxide Coating Thickness on Cutting Tool
Wear in the Turning Process, Journal of Testing And Evaluation. 23 (1995)
311–314.

[20] H. Cheng, M. Hon, Influence of TiN coating thickness on the wear of Si3N4-
based cutting tools, Surface and Coatings Technology. 81 (1996) 256–261.

[21] K. Tuffy, G. Byrne, D. Dowling, Determination of the optimum TiN coating

thickness on WC inserts for machining carbon steels, Journal of Materials
Processing Technology. 155-156 (2004) 1861–1866.
doi:10.1016/[Link].2004.04.277.

19
[22] F. Qin, Y.K. Chou, D. Nolen, R.G. Thompson, Coating thickness effects on
diamond coated cutting tools, Surface and Coatings Technology. 204 (2009)
1056–1060. doi:10.1016/[Link].2009.06.011.

[23] J. Takadoum, H.H. Bennani, Influence of substrate roughness and coating

thickness on adhesion, friction and wear of TiN films, Surface and Coatings
Technology. 96 (1997) 272–282.

[24] J.H.W. Siu, L.K.Y. Li, An investigation of the effect of surface roughness and
coating thickness on the friction and wear behaviour of a commercial MoS2 –
metal coating on AISI 400C steel, Wear. 237 (2000) 283–287.

[25] J.I. McCool, Relating Profile Instrument Measurements to the Functional

Performence of Rough Surfaces, Journal of Tribology. 109 (1987) 264–270.

[26] R. Goltsberg, I. Etsion, G. Davidi, The onset of plastic yielding in a coated

sphere compressed by a rigid flat, Wear. 271 (2011) 2968–2977.
doi:10.1016/[Link].2011.06.025.

[27] W. Song, L. Li, A. Ovcharenko, D. Jia, I. Etsion, F.E. Talke, Plastic yield
inception of an indented coated flat and comparison with a flattened coated
sphere, Tribology International. 53 (2012) 61–67.
doi:10.1016/[Link].2012.04.022.

[28] W. Song, L. Li, I. Etsion, A. Ovcharenko, F.E. Talke, Yield inception of a soft
coating on a flat substrate indented by a rigid sphere, Surface and Coatings
Technology. 240 (2014) 444–449. doi:10.1016/[Link].2013.12.073.

[29] Z. Chen, R. Goltsberg, I. Etsion, Plasticity evolution in a coated sphere

compressed by a rigid flat, Tribology International. 98 (2016) 116–124.
doi:10.1016/[Link].2016.02.004.

[30] R. Goltsberg, I. Etsion, A model for the weakening effect of very thin hard
coatings, Wear. 308 (2013) 10–16. doi:10.1016/[Link].2013.09.019.

[31] X. Huang, H. Kasem, H.F. Shang, T.M. Shao, I. Etsion, Experimental study of
a potential weakening effect in spheres with thin hard coatings, Wear. 296
(2012) 590–597. doi:10.1016/[Link].2012.08.018.

[32] X. Huang, I. Etsion, T. Shao, Indentation pop-in as a potential characterization

of weakening effect in coating/substrate systems, Wear. 338-339 (2015) 325–
331. doi:10.1016/[Link].2015.07.009.

[33] X. Huang, I. Etsion, T. Shao, Effects of elastic modulus mismatch between

coating and substrate on the friction and wear properties of TiN and TiAlN
coating systems, Wear. 338-339 (2015) 54–61.
doi:10.1016/[Link].2015.05.016.

20
[34] R. ’Saoubi, .P. Johansson, J. . Andersson, Wear mechanisms of PVD-
coated PCBN cutting tools, Wear. 302 (2013) 1219–1229.
doi:10.1016/[Link].2013.01.074.

[35] I. Etsion, M. Amit, The Effect of Small Normal Loads on the Static Friction
Coefficient for Very Smooth Surfaces, Journal of Tribology. 115 (1993) 406–
410. doi:10.1115/1.2921651.

[36] L. Kogut, I. Etsion, A Finite Element Based Elastic-Plastic Model for the
Contact of Rough Surfaces, Tribology Transactions. 46 (2003) 383–390.
doi:10.1080/10402000308982641.

[37] S.R. Das, D. Dhupal, A. Kumar, Experimental investigation into machinability

of hardened AISI 4140 steel using TiN coated ceramic tool, Measurement. 62
(2015) 108–126. doi:10.1016/[Link].2014.11.008.

[38] Z. Hao, Y. Fan, J. Lin, Z. Yu, Wear characteristics and wear control method of
PVD-coated carbide tool in turning Inconel 718, The International Journal of
Advanced Manufacturing Technology. (2015) 1329–1336.
doi:10.1007/s00170-014-6752-0.

[39] B. Davoodi, B. Eskandari, Tool wear mechanisms and multi-response

optimization of tool life and volume of material removed in turning of N-155
iron–nickel-base superalloy using RSM, Measurement. 68 (2015) 286–294.
doi:10.1016/[Link].2015.03.006.

[40] R. Yigit, E. Celik, F. Findik, S. Koksal, Tool life performance of multilayer

hard coatings produced by HTCVD for machining of nodular cast iron,
International Journal of Refractory Metals and Hard Materials. 26 (2008) 514–
524. doi:10.1016/[Link].2007.12.003.

Highlights

 The effect of coating thickness and substrate roughness on the tool wear in
turning is studied experimentally.
 TiAlN coating of various thicknesses is applied on tungsten carbide (WC) tool
substrate having various surface roughnesses.
 The tool wear is measured following a fixed cutting distance and speed tests.
 It is found that tool wear depends not on coating thickness alone or surface
roughness alone but on a dimensionless combination of these two parameters.

21