Development of diamond-like carbon (DLC) coatings with alternate soft and hard multilayer

architecture for enhancing wear performance at high contact stress

ABSTRACT
 We proposed that multilayer DLC coatings may have potential for tribological applications at
high stresses.
 In this study, DLC coatings consisting of alternate soft-layer and hard-layer were deposited
onto M2 steel substrates using unbalanced magnetron sputtering technique.
 The tribological performance at high contact stresses (2.5 GPa, 3.2 GPa, and 4 GPa) was
evaluated by a ball-on-disc tribometer.
 The residual stress, the hardness and reduced modulus are increased with percentage of hard-
layer thickness.
 Multilayer coatings with the soft-top layer improve wear performance compared with single-
layer DLC coatings. But, multilayer coatings without the soft-top layer show poor wear
performance under high stresses.
 The soft-top layer reduces the wear volume during run-in period by forming a transfer layer
so that decreases the total wear.
 The 50% hard multilayer coating with the soft-top layer provides the best wear performance
at an extreme stress of 4 GPa among all coatings in this study because it combines the low
magnitude of the residual stress with the acceptable hardness value.

INTRODUCTION
 In tribological applications, DLC coatings with high hardness usually provide high wear
resistance [2].
 However, it has limited tribological applications where high contact stresses are required
[3].
 In previous studies, the tribological performance of DLC multilayer coatings mostly
focuses on low loads wear onto silicon substrates.

EXPERIMENTS
COATING DEPOSITION
 DLC coatings were deposited using a unbalanced closed field magnetron sputtering
system (Teer UDP-650). Sputtering technique is commonly used in commercial coatings
[12]. Teer UDP-650 has a computer-controlled system to precisely monitor and control
deposition process.
 A chromium based transition layer was deposited first onto a M2 steel substrate to
improve adhesion strength.
 All carbon coatings were started with a soft-layer deposited at bias of −40 V onto the pre
deposited transition layer, followed by the hard-layer deposited at bias of −120 V and
then alternate soft-layer and hard-layer.
 The multilayer structure has ten bilayers and each bilayer includes a soft-layer and hard-
layer. The thickness of each bilayer is ~60 nm.
 The thickness ratio of a hard-layer to a soft-layer is varied inside a bilayer while the
bilayer thickness is fixed. It is defined as the percentage of hard-layer thickness in this
work.
 Besides multilayer coatings, a single-layer soft DLC was deposited at bias of -40 V; a
single layer hard DLC was deposited at bias of -120 V. Total five DLC coatings were
prepared with different percentages of hard-layer thickness (0%, 30%, 50%, 70% and
100%). The layer thickness was controlled by deposition time.
 In this series of coatings, except for the single-layer soft DLC, others all had the hard-
layer as a top layer. In another series, a 50 nm soft-layer was deposited onto previously
described coatings as a top layer.
 Coatings were deposited onto M2 steel substrates to determine hardness, reduced
modulus, adhesion strength and tribological properties.
 Coatings were also deposited onto silicon substrates to measure thickness and residual
stresses.

COATING CHARACTERIZATION
 The cross-sectional morphology was observed through a high resolution transmission
electron microscope (HRTEM; JEOL 2100F).
 The silicon substrate thickness is 400±10 μm with diameter of 50.8±0.3 mm.
 The thickness ratio of DLC coating to silicon wafer substrate is ~0.001.
 Thus, σ can be calculated from the curvature by the well-known Stoney’s equation [15]
 The film thickness and curvature radii were measured by Form Talysurf PGI (Taylor
Hobson).
 Nanomechanical test instrument (Hysitron TI700) was used to determine hardness and
reduced modulus of DLC coatings by well-known Oliver and Pharr method [17]. A cube
corner indenter was used due to higher aspect ratio and being sharper than Berkovich
indenter and experimental error 10%.
 The adhesion strength is determined using a scratch tester (Teer ST3001) based on ASTM
standard C1624-05 [20]. In the test, a diamond indenter was drawn across the coatings at
a constant speed and a progressively increasing normal force. The failure is observed by
optical microscope.
 A ball-on-disc tribometer (Teer POD-2) was used to evaluate the tribological behavior of
DLC coatings under dry sliding conditions. A 5 mm diameter WC-6% Co ball was used
as a counterpart. The normal loads are 20 N, 40 N and 80 N which correspond to the
initial Hertzian contact stresses of 2.5 GPa, 3.2 GPa and 4 GPa respectively. The sliding
speed is 55 mm/s (320 RPM). The diameter of the wear track was fixed at 3.3 mm. The
wear volume was calculated by measuring the cross section of wear tracks using optical
surface profile (Veeco Wyko NT9300).
 A scanning electron microscope (SEM; FEI Quanta 450) was used to observe the wear
track morphology of DLC coatings.

RESULTS & DISCUSSION

MICROSTRUCTURE
 Fig. 1a shows the TEM bright field image of the cross section of the 50% hard multilayer
coating on silicon substrate. Alternate dark layer and bright layer with clear interface are
observed.
 The bright layer is the soft-layer deposited at bias of -40 V, while the dark layer
corresponds to the hardlayer deposited at −120 V.
 The contrast difference results from different density between a hardlayer and a soft-layer
[22].
 The thickness of a hard-layer or a soft-layer is ~30 nm, so the total thickness is ~ 600 nm
measured from the TEM cross-sectional image.
 Fig. 1b is a cross-sectional image at a higher magnification of a bilayer inside the
multilayer coating, which indicates that hardlayer and soft-layer show the amorphous
structure.
MECHANICAL PROPERTIES
From Table I, in single-layer coatings, the magnitude of the residual compressive stress of the
soft DLC coating is ~0.5 GPa; the magnitude of the residual compressive stress of the hard
DLC is ~5.0 GPa.
The compressive stress stems from energetic particles bombardment during the deposition,
which induces a compact structure of coating.
As for multilayer coatings, the coating with higher percentage of hard-layer thickness has the
larger magnitude of the residual stress.
The hard DLC possesses the highest hardness and reduced modulus. The soft DLC has the
lowest hardness and reduced modulus. The multilayer coatings have hardness values between
the soft DLC and hard DLC.
For DLC coatings, hardness is mainly governed by the fraction of sp3 bond [23].
Thus, the hardness and reduced modulus are increased with percentage of hard-layer
thickness.
The adhesion results are shown in Table I. All coatings have a critical load over 60 N. There
is no evidence of cracking or delamination of the coatings, which indicates good adhesion.

TRIBOLOGICAL PERFORMANCE
 Fig. 2a shows The single-layer soft DLC outperforms other coatings and has the lowest
wear volume after 21333 cycles.
 It demonstrates that multilayer DLC coatings without the soft-top layer do not improve
the wear performance of coatings.
 It should be noticed that though the run-in period only accounts for 5% duration of the
entire wear test.
 Multilayer and hard DLC coatings, it reaches around 30%-40% of the total wear. In
contrast, for the soft DLC the wear volume of the run-in period is only ~9% of the total
wear.
 It indicates that for DLC coatings without the soft-top layer, most of the wear has
occurred during the run-in period.
 The different run-in behavior between soft DLC and others might be caused by the top
layer.

 Fig. 2b shows wear volume of these DLC coatings with the soft-top layer at 1067 and
21333 cycle under the same applied load of 20 N.
 Though only 50 nm thick soft DLC is added as a top layer, the wear performance has
been altered.
 the wear volume of 30%, 50%, 70% hard multilayer and hard DLC coatings with the soft-
top layer is reduced by 33%, 59%, 73% and 69% respectively.
 The wear volume of the run-in period is decreased by an average of ~70% for multilayer
and hard DLC coatings with the soft-top layer.
 It suggests that the soft DLC as a top layer can effectively reduce wear volume of the run-
in period so that decrease the wear volume of the entire wear test.
 Fig. 2c show 50%, 70% hard multilayer and hard DLC coatings with the soft top always
show lower wear volume than the soft DLC coating.
 The 30% hard multilayer shows worse wear performance than the soft DLC.
 Hard DLC coating with the soft-top layer survives and shows the lowest wear volume.

 Fig 2d shows wear volumes of DLC coatings at an extreme applied load of 80 N.

 Only 50% and 70% hard multilayer coatings and soft single-layer DLC can survive.
 The 30% and 100% hard coatings with the soft top failed after certain cycles.
 The 50% hard multilayer provides the lowest wear volume, only half of the wear volume
of the single-layer soft DLC coating.

 Fig. 3a The coefficient of friction (CoF) versus cycle curves for DLC coatings without the
soft-top layer at a load of 20 N.
 Except Soft DLC, The CoF is quite high in the beginning it reaches to 0.5 at 20 cycles,
takes 110 more cycles to fall to 0.15, and arrives at the steady state at 2000 cycles with
CoF 0.1.
 The high CoF is associated with high wear volume during the run-in stage as indicated in
Fig. 2a.

 Fig. 3b, on the contrary, the maximum CoF is reduced to ~0.15 for all coatings with the
soft-top layer.
 The low CoF corresponds to the low wear volume during the run-in period as indicated in
Fig. 2b.
 Since all coatings were capped with the soft-top layer, the CoF has similar behavior.
 The CoF is initially around 0.15 and the fall rapidly to 0.1 during 100 cycles then reduced
gradually to steady-state values of 0.08 to 0.1 by 2000 cycles, then remains constant. In a
word, the wear volume of the run-in period is related to the CoF.
 Fig. 4 shows optical images of wear surfaces of WC balls after 1067 cycles and after
21333 cycles at a load of 20 N.
 After 1067 cycles (run-in period), a discontinuous transfer layer has been formed onto
WC counterface against soft DLC (Fig. 4a) and 50% hard multilayer (Fig. 4b) coatings
with the soft-top layer.
 After 21333 cycles (entire wear test), the area of transfer layer is increased and it still well
covers the contact area (Fig. 4d and e).
 In contrast, WC counterface against hard DLC does not show a large area transfer layer
after 1067 and 21333 cycles (Fig. 4c and f)). Only wear debris (the black area indicated
by an arrow in Fig. 4c and f) was produced around the contact area.

 Though multilayer coatings indeed reduce residual stresses as expected, multilayer

coatings without the soft top do not provide good wear performance at high contact
stresses.
 Based on above analysis, we find that good wear performance depends on whether the
DLC coating has a soft top layer or not.

 For hard DLC coating, due to high residual stresses and brittleness the crack was induced
at load of 20 N during the run-in stage as shown in Fig. 5a and b, and hence abrasive
debris was produced as shown in Fig. 4c.
 This debris could not be expelled from the wear track in the beginning, so that entrapped
debris between two opposing surface scratches both the counterface and hard DLC
coating, which cause “three-body wear”[28].
 This abrasive wear leads to high CoF and wear volume during the run-in period.

 For multilayer coatings capped with the soft DLC or the single-layer soft DLC coating,
during the run-in stage a transfer layer begins to build up as shown in Fig. 4a and b.
 Such low CoF makes the coating survive run-in stage and have access to the stable wear
even under high load.
 Compared with the hard DLC, no crack is found inside the wear track for soft DLC as
shown in Fig. 5c
 All multilayer coatings offer lower magnitudes of residual stresses than hard DLC coating
and higher hardness values than soft DLC coating, which provides the possibility of good
wear performance.
 The 50% hard multilayer coating combines the acceptable hardness with the low
magnitude of residual stresses, which provides the best wear performance at high loads.
 Thus, for extreme load 80 N, multilayer coatings can be the best choice for tribological
applications.

CONCLUSIONS
(1) The residual stress, hardness, and reduced modulus are increased with percentage of hard
layer thickness. The residual stress and reduced modulus of multilayer coatings are found to
follow the mixing rule.
(2) Multilayer coatings with the soft-top layer improve wear performance compared with
singlelayer DLC coatings. In contrast, multilayer coatings without the soft-top layer show
poor wear performance under high stresses. Coatings with the soft-top layer can build up a
transfer layer during run-in period to provide low CoF and wear volume so that decrease the
total wear at high loads.
(3) The 50% hard multilayer coating with the soft-top layer provides the best wear
performance at an extreme stress of 4 GPa among all coatings in this study because it
combines the acceptable hardness with the low magnitude of residual stress. Also, the soft top
layer plays a key role in improving wear performance of 50% hard multilayer at high loads.
Effect of Diamond-Like Carbon Thin Film on the Fatigue Strength of AISI 4340 Steel
(Sujitno, T., et al, 2019)

ABSTRACT
 AISI 4340 is known as super strength steel and widely applied in military equipment,
automotive components because of its excellent behavior in wear, corrosion, fatigue,
high temperature, and high-speed operating conditions. However, due to continuing
work which reduced the performance of that component surface during their service
life, an effort to improve the surface properties for a longer service life should be
carried out.
 This paper presents the research result of the influence of diamond-like carbon
coating deposited using home-made DC Chemical Vapor Deposition (DC-CVD) on
the surface of AISI 4340 steel.
 As a carbon source, a mixture of argon (Ar) and methane (CH4) with a ratio of 24%:
76% was used in this experiment.
 The conditions of the experiment were 400°C of temperature at various gas pressures
(1.2 mbar, 1.4 mbar, 1.6 mbar, 1.8 mbar, and 2.0 mbar) for 5 hours of coating time.
 Investigated surface properties are hardness, fatigue strength, and surface
morphology.
 It was found that the optimum conditions in enhancing fatigue strength at 1.4 mbar of
pressure.
 At these conditions, the fatigue strength increase from 401 MPa to 514 MPa, the
microhardness increase from 327 VHN to 625 VHN.
 Based on surface morphology observation of the fracture surfaces, it shows that for
raw material, an initiation crack starting from the surface.
 However, after being coated for 1.2 mbar, 1.4 mbar, and 1.6 mbar, the initial crack
begins from the inside.
 The high hardness layer hinders the fatigue crack initiation.
METHODOLOGY
 In this research, it is reported the effect of diamond-like carbon coating on the fatigue
strength of AISI 4340 steel.
 Deposited by home-made DC Chemical Vapor Deposition (DC-CVD) on the surface
of AISI 4340 steel.
 For the purpose, as a carbon source, we use a mixture of argon (Ar) and methane
(CH4) with a ratio of 24%:76%.
 The conditions of the experiment were 400°C of temperature at various pressures (1.2
mbar, 1.4 mbar, 1.6 mbar, 1.8 mbar, and 2 mbar) for 5 hours of coating time.
 Fatigue test was conducted using a rotating bending machine. The specimens were
machined into the shape & size according to JIS Z 2274 standard as shown Figure 1.
RESULTS
HARDNESS TEST
 The influence of pressure variation at 400°C of temperature and 5 hours coating time
on the surface hardness of coated AISI 4340 steel is shown in Figure 3.
 It showed that for the higher pressure of the coating process the higher hardness and
reached the optimum hardness around 625 VHN.
 This optimum hardness was achieved at 1.4 mbar of pressure.
 However, after 1.4 mbar, the hardness tends to decrease.
 Correlation between pressure and formed plasma during the coating process is the
more pressure, the more formed ions/molecules.
 Formed ions during the coating process are C+, Ar+, and CxHy+; these ions
contribute to the formation of diamond-like carbon coating [9].
 As the flow rate of the gas increases, the pressure of the plasma inside the chamber
increases, this in turn increases the rate of generation of atomic hydrogen and methyl
radicals; as a result, the growth rate of the film increases.
 However, increasing of Ar concentration above a certain level, the CH4 concentration
becomes so low thus the etching rate due to Ar becomes more pronounced than the
deposition rate from CH4.
 As a result, the growth rate drops.

 Bombardment of Ar and H ions on the surface of the DLC film produces more
dangling bonds due to sputtering of hydrocarbons and abstraction of H from the film,
thus the incoming ad-atoms bonds with nucleated regions giving rise to larger grain
size.
 Raising in the grain size will cause the material to become softer.
 Reducing the grain size will cause the material to become stronger.
 Grain size reduction is also mean to increase the toughness of metal.
FATIGUE TEST
 The fatigue test to failure was carried out in the air up to N = 107 cycles.
 Figure 4 shows the S–N curves of the uncoated and coated samples treated for various
of pressure at 400°C of temperature and for 5 hours of deposition time.
 Table 3 shows the effect of pressure on the fatigue strength of AISI 4340 coated at
400°C of temperature and 5 hours of duration time.

 It shows that the fatigue strength of as-received specimens is 401 Mpa.

 After being nitrided for various of pressure, showed that for the higher pressure of the
coating process, the higher fatigue strength and reached the optimum fatigue strength
around 514 MPa, and this condition was achieved at 1.4 mbar of pressure.

 However, after 1.4 mbar of pressure, the fatigue strength tends to decrease.
 Increased fatigue strength is caused by the increase of surface hardness and the
formation of compressive residual stresses in the surface during the coating process.

 The high hardness layer hinders the fatigue crack initiation.

 The compressive residual stress is superimposed to the external load and leads to a
reduction of the effective stress in tension.
 Since only tensile stress produces fatigue cracks and contributes to crack propagation,
a reduction of the effective stress in tension increases the fatigue strength.
FATIGUE FRACTURE SURFACE ANALYSIS
 The fracture surface was observed using Scanning Electron Microscopy (SEM).
 Figure 5 shows the typical SEM images surfaces for as–received and DLC coated
specimens.
 Based on SEM observation of the fracture surfaces shown that initiation crack starts
from the surface.
 However, after being coated for 1.2 mbar, 1.4 mbar, and 1.6 mbar, the initial crack
start from the inside.
 The high hardness layer hinders the fatigue crack initiation.
 The compressive residual stress is superimposed to the external load and leads to a
reduction of the effective stress in tension.
 Since only tensile stress produces fatigue cracks and contributes to crack propagation,
a reduction of the effective stress in tension increases the fatigue strength.

CONCLUSIONS
 The hardness of the as-received materials is 327 VHN, after being coated for various
pressure and time; generally, the hardness increased and reached optimum in order of
625 VHN.
 This condition was achieved at 1.4 mbar of pressure and 5 hours of deposition time.
 In the rotary bending test, it was found that, for as-received materials, the fatigue
strength is 400 MPa.
 After being coated for various pressure and time, the fatigue strength increased and
reached optimum in order of 514 MPa.
 This condition was achieved at 1.4 mbar of pressure and 5 hours of deposition time.
 A surface morphology observation of the fracture surfaces shown that for raw
material, initiation crack starts from the surface.
 However, after being coated for 1.2 mbar, 1.4 mbar, and 1.6 mbar, the initial crack
starts from the inside.
 The high hardness layer hinders the fatigue crack initiation.
The effect of hydrogen on the tribological behavior of diamond like carbon (DLC)
coatings sliding against Al2O3 in water environment (Zhang, T., F., 2016)

ABSTRACT
 The tribological behavior of hydrogenated and hydrogen free diamond like carbon
(DLC) coatings sliding against Al2O3 in air and pure water was investigated by ball-
on-disc reciprocation wear test, and the effect of hydrogen was discussed.
 The results showed that in air environment, the Al2O3 slide against DLC under solid
lubrication condition, and the wear of DLC coatings was mainly affected by coating
mechnical properties.
 The hydrogen lowered the coating hardness and brought more DLC wear in air
environment.
 In water environment, the solid lubrication effect is weak, the hydrogen began to play
an important role in DLC wear.
 The hydrogenated DLC coating with saturated -CH bonds showed better wear
resistance even though its hardness is lower.
 The hydrogenated DLC coating with unsaturated -CH bonds was easy to graphitize
and generate massive wear no matter in water or in air environment.

INTRODUCTION
 In vacuum or inert atmosphere, hydrogen free DLC coating shows relatively high
friction and heavy wear because of strong covalent bond interactions between its free
σ-bonds and atoms in counterpart ball [5].
 Oppositely, hydrogenated DLC coating shows low friction and slight wear due to the
elimination of free σ-bonds by hydrogen.

 In humid air, the friction coefficient of the hydrogenated DLC coating increases with
increasing humidity due to the interruption of lubricating layer and friction induced
coating oxidation [6].
 However, hydrogen free DLC coating can keep a low friction coefficient in humid
environment because the water molecular will terminate the free σ-bonds on the
coating surface [7].

 In water environment, water molecules will react with hydrogenated and hydrogen
free DLC coating forming oxygen containing hydrophilic group at the sliding surface
[8].
 These hydrophilic groups will produce a surface layer rich in water, providing
lubrication for the counter surfaces.
 As a result, both the hydrogenated and hydrogen free DLC coatings show lower
friction coefficient in water than that in air [9,10].
 In water environment, hydrogenated DLC coating generally shows a higher specific
wear rate than in air environment [9,11].
 Hydrogen free DLC coating is reported to perform well in water environment, but its
high hardness may cause wear of the counterface material [13].
  For hydrogenated and hydrogen free DLC coatings, there is still no final conclusion
about which one is more suitable for the application in water environment.
METHODOLOGY
2.1 Coating preparation and characterization
 Biomedical CoCrMo alloy wafers with 14 mm diameter and 1.5 mm thickness were
used as the substrates.
 Before the DLC deposition, the substrate was grinded and polished to a roughness of
2~10 nm.
 Then hydrogenated and hydrogen free DLC coatings were prepared on the CoCrMo
substrates by plasma immersion ion deposition (PIID, labeled as HD1) [14], electron
cyclotron resonance plasma enhanced chemical vapor deposition (ECR-PECVD,
labeled as HD2) [15] or filtered cathode vacuum arc (FCVA, labeled as D3) [16],
respectively.

 The coating thickness and roughness were measured by a step profiler (XP2,
AMBIOS, USA) and atomic force microscope (CSPM 5000, BenYuan, China).
 The hydrogen content of HD1 and HD2 was measured by elastic recoil detection
analysis (ERDA, NEC/9SDH-2, China).
 The indentation hardness and modulus of the DLC coating were evaluated by
indentation test using a dynamic ultra micro hardness tester (DUH-211S, Japan).
 Load-unload curves were measured to calculate the indentation hardness and
modulus, as described in Ref. [17].
 The properties of DLC films mentioned above are shown in Table.1.
 The structure of the DLC coatings was characterized by Raman spectroscopy (λ=514
nm, Renishaw Invia, UK).
 The C-H bonds in the DLC coatings were analyzed by Fourier transform infrared
spectroscopy (FTIR, Nicolet 5700, US).

2.2 Tribological properties of the DLC coatings

 The tribological properties of the DLC coatings were tested by a ball-on-disc
tribometer (CSEM, Switzerland).
 Al2O3 balls were used as the friction pairs because Al2O3 is commonly used to make
femoral heads in COM type artificial joints.
 Reciprocation friction mode was performed to evaluate the friction and wear of the
DLC coating.
  The Al2O3 balls slid on the DLC coatings under 2 N with the sliding speed of 2.5
cm/s for 50,000 cycles.
 The sliding took place in air and pure water environment respectively and the wear
track length was 6 mm.
 After the wear test, the bulk wear loss was calculated based on the cross-section
profile of the wear track measured by the XP2 step profiler (AMBIOS, USA).
 The wear rate was normalized with respect to the applied load and sliding distance.

RESULTS
 The structure of DLC coatings was evaluated by Raman spectroscopy, as shown in
Fig.1.
 The Raman spectra of DLC coating consists of a D peak at about 1350 cm-1 and a G
peak at about 1560 cm-1.

 The integrated intensity ratio (ID/IG) was often used to reflect the content of sp3
clusters [18].
 Fig.1 shows that hydrogenated HD1 and HD2 samples have lower ID/IG ratios than D3
sample, which indicates the hydrogenated DLC coatings (HD1 and HD2) have more
sp3 bonds than the hydrogen free D3 sample.

 For the hydrogenated DLC coatings, the sp3 bonds consist of sp3-CC bonds and sp3-
CH bonds.
 The sp3-CH bonds make no contribution to coating hardness, so the hardness of
hydrogenated HD1 and HD2 is lower than that of hydrogen free D3 (Table.1).
Fig.1 Raman spectrum and ID/IG ratio of HD1, HD2 and D3 samples
 The FTIR spectrum from 2500 cm-1 to 3500 cm-1 was used to analyze the –CH bonds
in hydrogenated DLC coatings, as shown in Fig.2.
 For HD1 sample, a large peak centering at about 2900 cm-1 reveals a strong absorption
of saturated sp3-CH bonds vibration [19].
 And a mild broad peak can be observed at about 3300 cm-1, which may result from
the vibration of the unsaturated sp1-CH bonds (C≡CH) [20].

 For HD2 sample, small absorption peaks at 2860 cm-1, 2920 cm-1 and 2955 cm-1 can
be observed indicating the saturated sp3-CH structure (-CH3) in HD2.
 A strong broad peak between 3000 cm-1 and 3300 cm-1 also can be observed, which
indicates the HD2 sample contains -CH bonds with unsaturated sp2 (C=CH2) and sp1
structure [21].

 Fig.2 also shows that the HD1 sample has stronger peak around 2900 cm-1 than HD2
sample, which implies HD1 sample contains more saturated sp3-CH bonds than HD2.
 The HD2 sample shows stronger absorption peak (3000~3300 cm-1) of unsaturated
structure indicating HD2 has more unsaturated sp2 & sp1-CH bonds compared with
HD1.
 For hydrogen free D3 sample, no -CH absorption peaks can be observed because there
is no hydrogen in the coating.

Fig.2 FTIR spectrum of HD1, HD2 and D3 samples

3.2 Friction of the DLC coatings

 The tribological properties of the DLC coatings were tested against Al2O3 balls using
reciprocation friction mode by a ball-on-disc tribometer in pure water and air,
respectively.
 After ~3,000 laps’ friction, the friction coefficient became relatively stable.
 During steady wear stage, the friction coefficient at ~10,000, ~25,000 and ~40,000
laps was recorded to calculate the average friction coefficient.
 The average friction coefficients of the HD1, HD2 and D3 in air and water are shown
in Fig.3.
  In air environment, HD1 shows lower friction coefficient than HD2 and D3. D3
shows the highest friction coefficient.
 In water environment, HD1 still exhibits the lowest friction coefficient, and D3
exhibits the highest.
 All the three DLC samples show lower friction coefficients in water compared to
those in air environment, which may result from the water lubrication effect for the
sliding surfaces [22].

Fig.3 Average friction coefficients of HD1, HD2 and D3 tested in air and water environment

3.3 Wear of the DLC coatings

 After the wear test, the cross-section profile of the wear track was measured at three
different positions of the wear track.
 The average bulk wear rate was calculated according to the 3 measurements.
 The average wear rates of the DLC coatings in air and in pure water are shown in
Fig.4.

 In air environment, hydrogen free D3 sample, which has the highest hardness in three
samples, shows the lowest wear rate. And hydrogenated HD2 sample, which has the
lowest hardness, exhibits the highest wear rate.

 In water environment, the hydrogenated HD1 sample shows a lower wear rate than
hydrogen free D3 sample although the hardness of HD1 is lower.
 However, the hydrogenated HD2 sample exhibits a higher wear rate than the
hydrogen free D3 sample.

Fig.4 Bulk wear rates of HD1, HD2 and D3 during the wear test in air and water environment
3.5 Analyses of the friction counterfaces
 The optical morphologies of the Al2O3 counterfaces are shown in Fig.6.
 In air environment, transfer layer is formed on the Al2O3 counterface when the Al2O3
ball
 slid against HD1 and HD2.
 For D3 sample, wear debris can be observed on the Al2O3 counterface as well.
 These wear products generated in air may provide solid lubrication for the sliding
surfaces.

Fig.6 OM images of the Al2O3 counterfaces in air and water environment

 Fig.7 shows the Raman spectra of the wear products forming in air environment.
 All the wear products show a higher ID/IG ratio than the as deposited ones (Fig.1),
which reveals a graphitized structure of the wear products.

 In water environment, wear products were invisible on all the Al2O3 counterfaces
probably because water may restrain the formation of the transfer layer [25].
 The solid lubrication in water environment may be insufficient (tidak mencukupi) and
the sliding surface may be exposed directly to the water environment

Fig.7 Raman spectra of the wear products on Al2O3 counterfaces forming in air environment
DISCUSSION
Additionally, oxygen and vapor will take part in the tribochemical reaction and form oxygen
containing groups on the surfaces of wear track and wear products.

For hydrogenated DLC films, hydrogen can terminate the carbon dangling bond and restrain
the formation of oxygen containing groups.
So compared with the hydrogen free DLC (D3), hydrogenated DLC (HD1 and HD2) show
lower friction coefficient in air.

Compared to HD1, HD2 contains more unsaturated -CH bonds, the unsaturated -CH bonds is
unstable and easy to oxidate generating oxygen containing groups on the sliding surface, so
the friction coefficient of HD2 is higher than HD1.

In this paper, the hydrogenated HD1 and HD2 with lower hardness show larger wear rates,
and the hydrogen free D3 with higher hardness shows lower wear rate.
It reveals film hardness has important influence on the DLC wear in air environment.

In Water environment, for hydrogenated and hydrogen free DLC coatings, the lubrication
film may consist of H2O molecules formed by physical absorption and CH/COH groups
generated by tribochemical reaction [26].

For hydrogenated HD1 sample, hydrogen may terminate the dangling carbon bond, restrain
the tribochemical reaction and reduce the formation of oxygen containing groups.
So the friction and wear of HD1 are relatively low in water environment.

Hydrogenated HD2 sample also shows low friction coefficient in water environment,
however, the low hardness and structure softening during the sliding induce a large wear rate.

For D3 sample, without the termination of hydrogen, more oxygen containing groups will be
produced on the sliding surface as lubrication film.
The oxygen containing groups may induce hydrogen bond effect with water molecule which
is strong [27].
So the friction and wear of D3 in water environment is relatively high.

CONCLUSIONS
Hydrogenated and hydrogen free DLC coatings were prepared on the CoCrMo alloy.
In air environment, DLC coatings slide against Al2O3 balls under solid lubrication condition,
the wear rate is mainly related to coating hardness.
The hydrogen lowers the coating hardness and brings more DLC wear.
In water environment, DLC coatings slide against Al2O3 balls under water lubrication
condition.
In that case, the hydrogen in hydrogenated DLC coating become vital to influence the wear.
The hydrogenated DLC coating with saturated sp3-CH bonds produces the least friction and
wear, even though its hardness is lower.
The hydrogen free DLC coating may generate relatively high friction and wear without H
termination in water environment.
The hydrogenated DLC coating with unsaturated sp2 & sp1-CH bond reveals an unstable
structure, which is easily transformed to a softer graphite phase and produce a high wear rate
whatever in air or in water environment.
Tribological performance of an H-DLC coating prepared by PECVD (Solis, J., 2016)

ABSTRACT
 This paper describes the tribological performance under dry conditions of duplex layered
Hydrogenated Diamond-Like Carbon (H-DLC) coating sequentially deposited by
microwave excited plasma enhanced chemical vapour deposition (MW-PECVD) on AISI
52100 steel.
 The architecture of the coating comprised Cr, WC, and DLC (a-C:H) with a total
thickness of 2.8µm and compressive residual stress very close to 1 GPa.
 Surface hardness was approximately 22GPa and its reduced elastic modulus around
180GPa.
 Scratch tests indicated a well adhered coating achieving a critical load of 80 N.
 The effect of normal load on the friction and wear behaviours were investigated with steel
pins sliding against the actual coating under dry conditions at room temperature (20 ± 2
◦C) and 35–50% RH.
 The results show that coefficient of friction of the coating decreased from 0.21 to 0.13
values with the increase in the applied loads (10–50 N).
 Specific wear rates of the surface coating also decrease with the increase in the same
range of applied loads.
 Maximum and minimum values were 14 × 10−8 and 5.5 × 10−8mm−3/N m, respectively.
 Through Raman spectroscopy and electron microscopy it was confirmed the carbon-
carbon contact, due to the tribolayer formation on the wear scars of the coating and pin.
 In order to further corroborate the experimental observations regarding the graphitisation
behaviour, the existing mathematical relationships to determine the graphitisation
temperature of the coating/steel contact as well as the flash temperature were used.

METHODOLOGY
2.1. Materials and deposition
 A WC/a-c:H coating was deposited on AISI 52100 steel plates of dimensions 7 × 7 × 3
mm3 using an industrial scale Microwave excited Plasma Enhanced Chemical Vapour
Deposition (MW-PECVD) Flexicoat 850 system.
 The continuous deposition procedure includes deposition of an adherent Cr interlayer (by
DC magnetron sputtering) followed by an intermediate hard tungsten carbide (WC) layer
(by DC magnetron sputtering and gradual introduction of acetylene gas to chamber).
 The surface topography of H-DLC coating surface is shown in Fig. 1.
 This industrial scale PVD/CVD system incorporates two 1000 Watt microwave plasma
sources (2.45 GHz).
 The negative bias voltage used was 420 V.
 The substrate was maintained at less than 200 ◦C.
 During deposition, control of the substrates temperature is very important for heat
sensitive metallic materials such as the steel used in this study.
 The substrates were cleaned by Ar+ plasma using pulsed DC bias prior to deposition of
the adhesion layer.
 The thickness of the coating was determined by means of the abrasion ball cratering
technique utilizing a Calotester apparatus (tribotechnic, France).
 The total thickness of the a-C:H layer was 2.8 ± 0.2µm.
 The compressive residual stress of 0.9 ± 0.05 GPa was determined through substrate
deflection before and after H-DLC film deposition on a pure Si wafer.
2.2. Coating characterisation
 A surface roughness of 18 ± 5 nm for this film was evaluated using two-dimensional
contacting profilometry (Talysurf5, Taylor-Hobson, UK).
 Surface roughness data of 8 mm trace was analysed to the least square line, with Gaussian
filter, 0.25 mm upper cutoff and bandwidth 100 ÷ 1.
 The hardness and the reduced elastic modulus were evaluated by depth-sensing Nano-
indentation using a Nano Test (MicroMaterials, Ltd. Wrexham, U.K.), an enclosed box
platform with regulated temperature, software suite and micro capture camera.
 The diamond indenter was a Berkovich tip.
 The load was incremental with depth from 1 to 50 mN and a matrix of 50 indents was
used.
 The adhesive strength of the coating was tested using a commercial apparatus
(millennium 200, TRIBOtechinque, France) fitted with a Rockwell spherical diamond
indenter (tip radius of 500 µm).
 Scratch-tests were performed using progressive loads from 0.1 to 80 N with a load rate of
100 N/min and for a transverse scratch length of 8 mm in dry condition.
 The scratch tester is equipped with an acoustic emission monitoring sensor.
 The local chemical bonding structure present at the coated surface, as well as in the wear
tracks, was determined using Raman spectroscopy (Renisaw Invia Raman microscope
system), and a wave length of the incident laser beam of λ= 488 nm.
 All measurements were carried out in air at room temperature (20 ±2 ◦C).

2.4.2. Optical microscope, scanning electron microscope (SEM), atomic force microscope
(AFM) and raman spectroscopy
 A Leica optical microscope DM6000 was employed in this study to analyse the worn
regions for both tribo systems, i.e. Steel/Steel and H-DLC/Steel.
 The microstructure, morphology and thickness of the films as well as the worn surfaces
were studied using a Zeiss EVO MA15 Variable Pressure and field emission SEM in
direct mode.
 An integrated Oxford Instruments Energy Dispersive X-ray (EDX) analysis system was
used to qualitatively and quantitatively evaluate the presence of C and Cr/W
outside/inside the wear scar and also to determine the occurrence of material transfer.
 In depth chemical profiles were obtained by glow discharge optical emission
spectroscopy (GDOES).
 Surface topography, in turn, was observed with AFM (Bruker, ICON dimension with
scan asyst) for the H-DLC and steel surfaces outside and inside the wear scar.
 AFM scan images were obtained using a silicon tip (cantilever stiffness ∼0.4 N/m and tip
radius of ∼10 nm) in contact mode and a scan area of 5 µm × 5 µm.
 The carbon coating structure with respect to sp2/sp3 ratio was studied by Raman
Spectroscopy.
 A 488 nm excitation wave length laser source with 2 mW power was used.
 The ratio ID/IG was considered as an indicator of the carbon sp2/sp3 structure and also
transfer layer structures.

RESULTS
 Thus, the adhesion of the coating to the substrate is widely measured by means of the
scratch test, which provides information regarding the level of adherence in terms of the
occurrence of interfacial fractures.
 Scratch tests indicated that the H-DLC coating was well adhered to the AISI 52100 steel
substrate with very little cracking and no evidence of adhesive failure and therefore, no
critical failure as shown in Fig. 4.
 The arrow A points out an early on set of periodic low amplitude acoustic emission (AE)
peaks indicating no partial adhesive failure but only microscopic angular cracks at the
edge of the groove (64 N).
 The high amplitude AE peaks pointed out by arrow B corresponds to the total penetration
depth (8 mm). There is evidence of transverse semi-circular cracks caused by the higher
load (75 N) at the rear of the contact end.
 It can then be considered a critical load >80 N for the present coating, which is a more
than acceptable value for industrial applications [30].
 it can be suggested that the chromium deposited as the adhesion layer between the DLC
layer and the substrate, allowed for the relaxation, to a certain extent, of the compressive
residual stresses while the ductility to resist the interfacial fracture improves.

 The variation of the coefficient of friction (!) as a function of sliding distance in dry
conditions and under several applied normal loads is displayed in Fig. 7.
 At the beginning of the test in dry condition, the CoF for 10 and 20 N loads increased
progressively with the increasing of the sliding distance until it reached an average upper
value of 0.22–0.23, respectively, and then, a slightly decreasing trend is visible until it
levelled off with average values of 0.20–0.21.
 This initial rise in the CoF is attributed to the initial high roughness (Fig. 8a) and reduced
contact area between the interfaces.
 Due to sliding, the rough surface features smoothen as a result of the elastic and plastic
effects of the frictional forces acting on the coating, which in turn increase the contact
area that is strongly associated to the CoF, specifically its appearance and surface
hardness corresponding to the deformation resistance of the contact area.
 In Fig. 10, the specific wear rates of the H-DLC coatings are depicted.
 Under the same sliding velocity condition (0.1 m/s), with the increase in normal load, the
average specific wear rates decrease.
 The maximum and minimum specific wear rates are 14 × 10−8 and 5.5 × 10−8mm−3/N
nm, respectively.
 The former under a normal load of 10 N and the latter under a normal load of 50 N for 6 h
or 2160 m dry sliding.
 The reduction in specific wear rate seems to be abetted by the formation of a transfer
layer on the counter-layer, because of structural transformations triggered by the normal
load applied during the sliding.
 Among these structural effects, the re-crystallisation process becomes very active at the
contact interface in such a way that, due to the reciprocating sliding, a re-orientation of
the re crystallised layers brings the basal planes parallel to the top surface of the coating
with the consequent reduction in the wearing rate [41].
 Specifically, the sliding-induced localised annealing at the contact asperities, likely causes
a gradual destabilization of the carbon-hydrogen bond in the sp3 tetrahedral structure and
as a result, a transformation of this sp3 structure into a graphite-like sp2 structure takes
place [11].

CONCLUSIONS
 A hydrogenated Diamond-like Carbon has been characterised both prior to, and following
tribological testing.
 Friction and wear behaviour under dry unlubricated conditions of a [WC/a-C:H]/steel
contact has been assessed for different normal loads in ambient atmosphere.
 The adhesive strength of the coating was found to be more than acceptable for industrial
applications with a scratch resistance above 80 N.
 According to the outcomes from the present research, the H-DLC coating could be useful
for wide range of industrial applications such as automotive, aircraft and machine
components.
Effects of TiCN interlayer on bonding characteristics and mechanical properties of
DLC coated Ti-6Al-4V ELI alloy (Kang, S., 2015)

ABSTRACT
 The delamination of diamond-like carbon (DLC) films on titanium alloys can be
prevented by using an interlayer coating.
 In this study, a DLC film was deposited using a filtered vacuum arc (FVA), and a TiCN
interlayer was applied between the Ti-6Al-4V ELI alloy and the DLC film.
 The interface boundary, carbon bonding, and mechanical properties of the DLC-coated
Ti-6Al-4V were investigated, and the results indicate that DLC thin films with a TiCN
interlayer are suitable coating materials for Ti-6Al-4V.
 The coated DLC films had a crystal structure that consisted of an amorphous phase while
the TiCN interlayer exhibited a crystalline structure.
 The TiCN interlayer was necessary in order to improve the hardness, elastic modulus, and
interface bonding.
 The TiCN interlayer on the DLC film was observed to increase the ID/IG ratio as a result
of the sp3 bonding of the carbon atoms.
 The resulting friction coefficient of the DLC-coated Ti-6Al-4V with the TiCN interlayer
had an extremely low value of 0.03.

METHODOLOGY
 Ti-6Al-4V ELI (ASTMGrade 23) specimens had a diameter of 15mm and a thickness of
3mm.
 The specimens were ultrasonically cleaned in acetone and ethanol, and the TiCN
interlayer filmswere deposited using an ion-plating system with the purity of the Ti target
exceeding 99.9%.
 The initial pressure of the chamber was less than 7.5 × 10−7 Pa, and the TiCN interlayer
was deposited onto the substrate in a mixture of C2H2 and N2 gases for 80 min.
 The target and substrate were precleaned using an ion etching process, and the DLC films
were then deposited using an FVA system with a carbon target with a purity exceeding
99.8% and a working pressure of 1.0 × 10−5 Pa during the DLC coating process.
 A negative bias of−100 V DC and a substrate temperature of 100 °C were applied, and
the DLC layer was then deposited onto the substrate in an argon atmosphere for 60 min.
 The thickness of the film was observed using a field emission scanning electron
microscope (FE-SEM, S-4700, Hitachi, Japan); the interfacial microstructure of the TiCN
interlayer was observed using a high-resolution transmission electron microscope (HR-
TEM, TECANAI F20, Philips, Netherlands); and the binding structures of the DLC films
were analyzed using a laser Raman spectrometer (SPEX 1403, USA) with a 514.5 nm Ar
laser beam as the excitation source.
 The hardness and elastic modulus of the DLC-coated Ti-6Al-4V ELI specimen with and
without the TiCN interlayer were measured on the top surface of the specimens by using
a nanoindenter (Nanoindenter, MTS systems, USA) where the nanoindentation was
performed with a Berkovich indenter at a strain rate of 0.05 s−1. A 500-mN maximum
load with an indentation depth of 1000 nm was applied to the specimen with a loading
rate of 79.12 nm/s, and the variations in the results obtained from the data were addressed
by repeating the indentations for over 20 tests.
 The wear properties of the DLC-coated specimens were measured with a wear tester
(Wear tester, RB 102 PD, R&B, Korea) using the pin-on-disk method with a load of 20 N
and a rotation speed of 150 rpm, respectively.
RESULTS
 Fig. 1(a) shows FE-SEM morphologies of the DLC films coated on the Ti-6Al-4V alloy
with a thickness of 0.5 μm.
 The sequence for the Ti-6Al- 4V/TiCN/DLC coating layer is also shown in Fig. 1(b).
 The TiCN interlayer had a thickness of 0.9 μm.
 The interface boundary between the DLC film and the TiCN interlayerwas clearly
observed, and no cracks or pin holes were found in the films.

 The hardness and elastic modulus were measured for the DLC films using the
nanoindenter and are shown in Fig. 4.
 The hardness and elastic modulus of the DLC films without the TiCN interlayer were
measured at 66.2 GPa and 561.1 GPa, respectively, while those with the TiCN interlayer
were 78.7 GPa and 755.9 GPa, respectively.
 The presence of the TiCN interlayer effectively increases the mechanical properties of
DLC films, and this can be attributed to the increase in sp3 fraction content in the DLC
films.

 In addition, the TiCN interlayer could affect the mechanical properties of the DLC layer
due to the thin top DLC layer during indentation.
 Fig. 5 shows the friction coefficient of the Ti-6Al-4V, Ti-6Al-4V/DLC, and Ti-6Al-
4V/TiCN/DLC samples.
 The friction coefficient of the noncoated Ti-6Al-4V alloy was 0.35 while that of the
DLC-coated Ti-6Al-4V alloy was less than 0.1.
 The rapid increase in the curve at 2300 cycles was shown in a latter friction condition, and
the delamination of the DLC film resulted from the weak bonding of the DLC-coated
specimen without the TiCN interlayer.
 The friction coefficient of the DLC-coated Ti-6Al-4V with a TiCN interlayer was
extremely low, with a value of 0.03.
 The constant friction coefficient of the specimen was maintained, even after 4500 cycles,
and the DLC-coated specimens with a TiCN interlayer appeared to be less worn after the
wear test.

 The HR-TEM images and the selected area diffraction patterns (SADPs) of the Ti-6Al-
4V/TiCN/DLC are shown in Fig. 2.
 In the upper right hand corner of Fig. 2(a), the SADP indicates a DLC coating in an
amorphous phase with a halo pattern.
 This phase is referred to as a tetrahedral amorphous carbon (ta-C) phase [13].
 The lower right hand corner of the SADP in Fig. 2(a) is shown in the polycrystalline
phases with multi-ring patterns.
 The coated DLC films had a crystal structure that consisted of an amorphous phase while
the TiCN interlayer exhibited a crystalline structure.

CONCLUSIONS
 DLC thin filmswere successfully coated on Ti-6Al-4Vwith a TiCN interlayer by using an
FVA.
 The coated DLC films exhibited a crystal structure that consisted of an amorphous phase
while the TiCN interlayer had a crystalline structure.
 The TiCN interlayer significantly contributed to improving the hardness, elastic modulus,
and interface bonding of the thin film.
 The increase in the sp3 bonding of the carbon atomwas influenced by the TiCN interlayer
on the DLC film with an increased ID/IG ratio.
 The DLC-coated Ti-6Al-4V with the TiCN interlayer had an extremely low friction
coefficient of 0.03.

Hollow-cathode chemical vapor deposition of thick, low-stress diamond-like

carbon films (Miller, J., 2020)

ABSTRACT
 A radio-frequency (RF), hollow-cathode plasma source with confining magnetic field is
described for the chemical vapor deposition of thick ( > 10 μm), amorphous diamond-like
carbon ablator films for inertial confinement fusion applications.
 Plasma is characterized by optical emission spectroscopy, while properties of the resultant
films are measured by a combination of profilometry, Rutherford backscattering
spectrometry, elastic recoil detection analysis, X-ray diffraction, Raman spectroscopy,
and atomic force microscopy.
 The dependence of the deposition rate, film density, elemental composition, self-bias and
residual stress is reported as a function of RF power.
 Higher density films were found when using Ar plasma, than N2 or H2 plasma.
 The coatings produced are x-ray amorphous, exhibit low compressive stress ( ~ 100
MPa), high density ( < 1.7 g/cm3), hydrogen content of ~ 30 at.%, and a low average
roughness of 0.75 nm.
 Applications of these films as tunabledensity ablators for inertial confinement fusion
experiments are discussed.

METHODOLOGY
 All coatings were performed in an in-house designed RF-driven hollow cathode system
schematically illustrated in Fig. 1(a) and (b).
 The basic principle of the hollow cathode discharge is the overlap of the negative glow
zones of opposite cathode surfaces, leading to increased plasma densities.
 Inside the hollow cathode discharge, electrons are electrostatically trapped and oscillate
inside the cavity, expending all their energy for excitation and ionization of atoms,
increasing the ionization efficiency as well as the energy of the generated ions as a
function of power.

 The hollow cathode was made of a stainless steel tube with an inner diameter of 1.3 cm
and a length of 7.6 cm.
 Propylene (C3H6, 99.9% purity) was utilized as the hydrocarbon precursor.
 The carrier gas was either H2 (99.9%), N2 (99.9%), or Ar (99.999%).
 During deposition, Si (100) substrates diced into 1.9 × 1.3 cm2 chips were clamped onto
a water-cooled sample holder located 8cm from the end of the hollow cathode tube.
 The substrate holder becomes negatively self-biased by the plasma.
 The cathode plasma anode system is electrically floating and the plasma establishes itself
at a positive potential with respect to ground.
 Prior to coating, the chamber was pumped down to (4–5) × 10 5 Pa with a turbo
molecular pump.
 A mix of C3H6 and carrier gases were flowed through the hollow cathode toward the
substrate surface.
 Proportions of C3H6 in the gas feed mix were in the range of 5–100%, with typical total,
continuous gas flow rates of 0.5–20 sccm and operating pressures of 0.4–5.3 Pa.
 An RF power supply was used to ionize the gas inside the hollow cathode and the RF
power was varied between 25 and 150 W in a chamber volume of approximately 2000
cm3.
 A current-carrying magnetic coil, external to the vacuum chamber, was placed around the
hollow cathode to provide higher ion energy and stabilize the ionized gases at lower
operating pressures.
 Film thicknesses and deposition rates were measured via stylus profilometry (KLA
Tencor, Inc) on a masked section of the substrate.

RESULTS
 Table 2 shows representative stress values for coatings deposited at different RF power
levels and to different thicknesses.
 It reveals that all these films have very low compressive residual stresses of ≲ 100 MPa.
 This is at least an order of magnitude lower compared to residual stresses in typical DLC
films produced by other methods such as cathodic arc that exhibit stresses in the GPa
range [6,11,12].
 Such low residual stress levels in our films enable the deposition of ultrathick films
without delamination.

CONCLUSIONS
 A magnetically-confined RF hollow cathode plasma system has been developed,
demonstrating the growth of ultra-thick (45 μm), low-stress ( ≲ 100 MPa), ultra-smooth
(0.75nm average roughness) amorphous DLC films with densities as high as 1.7 g/cm3,
making these films promising for ICF applications.
 However, there is still a significant amount of work required to achieve larger film
densities and optimize growth parameters resulting in lower impurity levels and higher
deposition rates.