Thin Solid Films 688 (2019) 137265

Contents lists available at ScienceDirect

Thin Solid Films

journal homepage: www.elsevier.com/locate/tsf

Commentary on using H/E and H3/E2 as proxies for fracture toughness of T

hard coatings
Xinjie Chen, Yao Du, Yip-Wah Chung
⁎

Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA

ABSTRACT

While positive correlation between fracture toughness and H/E and H3/E2 has been reported for some ceramic coatings, we outline an argument that such correlation
cannot be true in general, especially for nanolayer and nanocomposite coatings that present multiple interfaces to impede crack growth. Literature data are presented
to support this argument.

Hard coatings, mostly derived from ceramic materials, are widely higher toughness where there is no plastic deformation.
used for wear protection applications. High hardness, while desirable The proxy H3/E2 very likely originated from the sphere-on-flat
under conditions of abrasive wear, is not the only important physical Hertzian contact analysis [9]. Given the contact between a sphere of
property for these applications. Given the brittle nature of most ceramic radius R, elastic modulus E and an infinitely rigid flat surface with a
materials, fracture toughness is another important property to consider. normal load N, the maximum Hertzian contact stress Pmax is given by:
Many papers have been written about measuring fracture toughness of 1 1
coatings [1–7]. In general, accurate determination of fracture toughness 1 6NE 2 3 NE 2 3
Pmax = 0.578
of coatings without substrate effects requires rather involved mea- R2 R2
surements and/or sample preparation. Therefore, it is desirable to find No significant plastic deformation occurs when Pmax < σy, which
an easier method to determine or rank fracture toughness of coatings. can be assumed to scale with the coating hardness H, i.e.,
Over the past 15 years or so, two ratios, viz., H/E and H3/E2 (H being
1
the hardness and E the elastic modulus) have been proposed as proxies NE 2 3
0.578 < = H
for fracture toughness of coatings. In some publications, E* = E/ R2
y

(1 − ν2) is used instead, ν being the Poisson ratio. For simplicity, we

will use E as the notation in this discussion. We are not sure how this where α is some constant. Therefore, one can write:
idea first came about. One possible reason for using H/E as a proxy for 3R2 H
3
N < 5.168
fracture toughness is that since H is proportional to the yield strength E2
σy, H/E is a measure of the elastic strain limit, which may have been
The above equation indicates that for a given external load and test
interpreted to be a measure of toughness. Another reason may have
geometry, a coating with larger H3/E2 is less likely to be plastically
arisen from the work of Greenwood and Williamson [8], in which they
deformed and should therefore have higher toughness.
defined the plasticity index φ as follows:
Musil and coworkers reported and discussed such positive correla-
tions between H/E (and H3/E2) and fracture toughness [10–17]. In their
1
E 2
= experiments, fracture toughness was determined qualitatively by
H
bending the coating through a certain deflection and observing the
where σ = surface roughness and β = asperity radius. Contacts with occurrence of surface cracks. Conducting such bending tests on several
small plasticity indices require large contact stresses to induce the onset coating systems, viz., Al-Cu-O [10], Al-O-N [11], Zr-Al-O [12], Ti-Ni-N
of significant plastic deformation. At constant surface roughness and [13], and Zr-Si-O [14], they found that coatings with H/E > 0.1 do not
asperity radius, H/E is inversely proportional to the plasticity index. form cracks, while those with H/E < 0.1 do. Examples of results from
Therefore, an argument can be made that a coating with larger H/E is Ti-Ni-N and Al-Cu-O coatings are shown in Fig. 1. They noted that
less likely to deform plastically at a given stress and hence should have elastic recovery, microstructure, and macro-stress generated during the

⁎
Corresponding author.
E-mail address: ywchung@northwestern.edu (Y.-W. Chung).

https://doi.org/10.1016/j.tsf.2019.04.040
Received 4 April 2019; Received in revised form 22 April 2019; Accepted 24 April 2019
Available online 25 April 2019
0040-6090/ © 2019 Elsevier B.V. All rights reserved.
X. Chen, et al. Thin Solid Films 688 (2019) 137265

Fig. 2. Correlation between fracture toughness and H/E and H3/E2 for a series
of aluminosilicate glasses with varied network modifier ions (data from Ref.
Fig. 1. Correlation between fracture toughness and H/E and H3/E2 in Ti-Ni-N
[27]).
and Al-Cu-O coatings (data from Ref. [13] and [16]).

correlation only at light impact forces, but not at higher forces. A clear
growth of coatings are all important factors contributing to cracking
effect of plasticity was shown in Shi et al.'s work [29] on multilayer W/
resistance [17].
VC coatings, in which they demonstrated either an inverse or zero
More recent work by Hahn et al. [18] on TiN/CrN superlattice
correlation between toughness and H/E and H3/E2 (toughness being
coatings shows a slightly positive correlation between H/E and fracture
measured by a modified Vickers method), as shown in Fig. 3. In their
toughness, the latter being measured by a microcantilever bending
experiments, the W thickness fraction within the four multilayer coat-
method. The significance of H/E in wear control has been discussed by
ings ranges between 60 and 90%. Daniel et al. [30] synthesized a series
Leyland and Matthews [19] and Lackner et al. [20].
of monolithic TiN coatings with a chevron-like microstructure and
It is attractive to use H/E and H3/E2 as proxies for fracture tough-
found no correlation between fracture toughness and H/E and H3/E2,
ness of coatings because H and E can be readily measured by na-
with toughness measured by the microcantilever bending method. In
noindentation. In spite of initial reports on such positive correlations,
the investigation on WC-Co coatings, Lima et al [31] found an inverse or
we explore in this commentary if such correlation is applicable in
no correlation between toughness and H/E, with coating toughness
general. Since H/E represents elastic strain to failure and H3/E2 re-
measured by the Vickers method. These two datasets are shown in
presents the resistance to plastic deformation, neither ratios address the
Fig. 4.
contribution of plasticity to fracture toughness. Moreover, the quantity
More examples of either inverse or no correlation can be found in
H3/E2 was derived from surfaces in contact while the property of
diverse material systems reported in the literature, e.g., Fe/VC multi-
fracture toughness has no intrinsic dependence on contact. For fracture
layer coatings [32], TiB2-TiC-Al2O3 composite coatings [33], CrN/AlN
events dominated by ceramic materials, plastic deformation is negli-
multilayers [34], and CrN/Cr and TiN/SiOx multilayers [35].
gible. Therefore, in coating systems where elasticity dominates fracture
In summary, H/E and H3/E2 may have been good proxies for frac-
toughness, H/E and H3/E2 could be good approximations to rank the
ture toughness of coatings if coating fracture involves no plasticity and
fracture toughness of coatings. However, in cases where plastic de-
leads to instant failure. However, the latter conditions may not be sa-
formation plays a role and cracks do not propagate to instant failure,
tisfied in most nanolayer and nanocomposite coatings, in which pro-
the positive correlation discussed in the preceding paragraphs might
pagating cracks are arrested frequently at interfaces. As such, use of H/
not hold.
E or H3/E2 as a measure of coating toughness is not generally applic-
There have been discussions about the use of micro/nanostructure
able.
control to improve fracture toughness, e.g., nanolayers, nanocompo-
sites, or matrix-precipitate lattice mismatch [21–25]. Such micro/na-
nostructure could toughen the coating through crack deflection at in-
terfaces, interface delamination, ductile interlayer ligament bridging,
and crack tip blunting because of nanoscale plasticity [26]. For ex-
ample, Wang et al. [21] synthesized Cu/Zr metallic glass nanolaminates
and achieved tensile ductility > 10% (Zr metallic glass has almost zero
ductility). Such toughening mechanisms may not have a proportional
effect on H/E and H3/E2. Under these conditions, H/E and H3/E2 will
not be good proxies for toughness – this statement applies to coatings as
well as bulk materials.
To illustrate, let us examine the correlation between fracture
toughness (as measured by the Vickers method) and H/E and H3/E2 for
a series of (bulk) aluminosilicate glasses with varied network modifier
ions [27]. As shown in Fig. 2, fracture toughness does not correlate with
either H/E or H3/E2. From publications that report H, E, and some
measures of toughness on coatings, we have found many examples of
either no correlation or inverse correlation. For example, Beake [28]
synthesized TiFeN and TiFeMoN films and assessed their fracture
Fig. 3. Correlation between fracture toughness and H/E and H3/E2 for VC, W,
toughness from repeated nano-impact tests. He found positive
and four W/VC multilayer coatings (shaded) from Ref. [29].

2
X. Chen, et al. Thin Solid Films 688 (2019) 137265

addition of Al in ZrO2 thin film on its resistance to cracking, Surf. Coat. Technol.
207 (2012) 355–360.
[13] J. Musil, R. Jílek, R. Čerstvý, Flexible Ti-Ni-N thin films prepared by magnetron
sputtering, J. Mater. Sci. Eng. 4 (2014) 27–33.
[14] J. Musil, J. Sklenka, J. Prochazka, Protective over-layer coating preventing cracking
of thin films deposited on flexible substrates, Surf. Coat. Technol. 240 (2014)
275–280.
[15] J. Musil, F. Kunc, H. Zeman, H. Polakova, Relationships between hardness, Young's
modulus and elastic recovery in hard nanocomposite coatings, Surf. Coat. Technol.
154 (2002) 304–313.
[16] J. Musil, Hard nanocomposite coatings: thermal stability, oxidation resistance and
toughness, Surf. Coat. Technol. 207 (2012) 50–65.
[17] J. Musil, Flexible hard nanocomposite coatings, RSC Adv. 5 (2015) 60482–60495.
[18] R. Hahn, M. Bartosik, R. Soler, C. Kirchlechner, G. Dehm, P.H. Mayrhofer,
Superlattice effect for enhanced fracture toughness of hard coatings, Scr. Mater. 124
(2016) 67–70.
[19] A. Leyland, A. Matthews, On the significance of the H/E ratio in wear control: a
nanocomposite coating approach to optimised tribological behaviour, Wear 246
(2000) 1–11.
[20] J.M. Lackner, L. Major, M. Kot, Microscale interpretation of tribological phenomena
in Ti/TiN soft-hard multilayer coatings on soft austenite steel substrates, Bull.
Polish Acad. Sci. Tech. Sci. 59 (2011) 342–355.
Fig. 4. Correlation between fracture toughness and H/E and H3/E2 for TiN [21] Y. Wang, J. Li, A.V. Hamza, T.W. Barbee, Ductile crystalline–amorphous nanola-
(data from Ref. [30]) and WC-Co coatings (data from Ref. [31]). minates, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 11155–11160.
[22] S. Zhang, D. Sun, Y. Fu, H. Du, Toughening of hard nanostructural thin films : a
critical review, Surf. Coat. Technol. 198 (2005) 2–8.
Acknowledgments [23] Y. Wang, S. Zhang, Toward hard yet tough ceramic coatings, Surf. Coat. Technol.
258 (2014) 1–16.
[24] A.N. Ranade, L.R. Krishna, Z. Li, J. Wang, C.S. Korach, Y.W. Chung, Relationship
This work was supported by the US National Science Foundation between hardness and fracture toughness in Ti–TiB 2 nanocomposite coatings, Surf.
(Grant No. CMMI-1462850). We wish to thank Prof. Joe Greene for Coat. Technol. 213 (2012) 26–32.
[25] C. Wang, K. Shi, C. Gross, J. Miranda, M. De Mesquita, Y.W. Chung, Toughness
inspiring us in the tackling of such a difficult subject. enhancement of nanostructured hard coatings : design strategies and toughness
measurement techniques, Surf. Coat. Technol. 257 (2014) 206–212.
References [26] H. Holleck, V. Schier, Multilayer PVD coatings for wear protection, Surf. Coat.
Technol. 76 (1995) 328–336.
[27] M. Tiegel, R. Hosseinabadi, S. Kuhn, A. Herrmann, C. Rüssel, Young' s modulus,
[1] L.M. Keer, C.H. Kuo, Cracking in a loaded, brittle elastic half-space, Int. J. Solids Vickers hardness and indentation fracture toughness of alumino silicate glasses,
Struct. 29 (1992) 1819–1826. Ceram. Int. 41 (2015) 7267–7275.
[2] K. Holmberg, A. Laukkanen, H. Ronkainen, K. Wallin, S. Varjus, A model for [28] B.D. Beake, V.M. Vishnyakov, J.S. Colligon, Nano-impact testing of TiFeN and
stresses, crack generation and fracture toughness calculation in scratched TiN- TiFeMoN films for dynamic toughness, J. Phys. D. Appl. Phys. 44 (2011) 085301.
coated steel surfaces, Wear 254 (2003) 278–291. [29] K. Shi, C. Wang, C. Gross, Y.W. Chung, Reversing the inverse hardness–toughness
[3] A.A. Volinsky, J.B. Vella, W.W. Gerberich, Fracture toughness, adhesion and me- trend using W/VC multilayer coatings, Surf. Coat. Technol. 284 (2015) 80–84.
chanical properties of low-K dielectric thin films measured by nanoindentation, [30] R. Daniel, M. Meindlhumer, W. Baumegger, J. Zalesak, B. Sartory, M. Burghammer,
Thin Solid Films 429 (2003) 201–210. C. Mitterer, J. Keckes, Grain boundary design of thin films: using tilted brittle in-
[4] Z.H. Xia, W.A. Curtin, B.W. Sheldon, A new method to evaluate the fracture terfaces for multiple crack deflection toughening, Acta Mater. 122 (2017) 130–137.
toughness of thin films, Acta Mater. 52 (2004) 3507–3517. [31] M.M. Lima, C. Godoy, P.J. Modenesi, J.C. Avelar-batista, A. Davison, A. Matthews,
[5] S. Zhang, D. Sun, Y. Fu, H. Du, Toughness measurement of thin films: a critical Coating fracture toughness determined by Vickers indentation: an important
review, Surf. Coat. Technol. 198 (2005) 74–84. parameter in cavitation erosion resistance of WC–Co thermally sprayed coatings,
[6] S. Zhang, X. Zhang, Toughness evaluation of hard coatings and thin films, Thin Surf. Coat. Technol. 178 (2004) 489–496.
Solid Films 520 (2012) 2375–2389. [32] C. Wang, J. Miranda, Y. Yang, Y.W. Chung, Investigation of hardness and fracture
[7] A. Riedl, R. Daniel, M. Stefenelli, T. Schöberl, O. Kolednik, C. Mitterer, J. Keckes, A toughness properties of Fe/VC multilayer coatings with coherent interfaces, Surf.
novel approach for determining fracture toughness of hard coatings on the micro- Coat. Technol. 288 (2016) 179–184.
meter scale, Scr. Mater. 67 (2012) 708–711. [33] M. Masanta, S.M. Shariff, A.R. Choudhury, Evaluation of modulus of elasticity,
[8] J.A. Greenwood, J.P. Williamson, Contact of nominally flat surfaces, Proceedings of nano-hardness and fracture toughness of TiB2–TiC–Al2O3 composite coating de-
the royal society of London. Series A. 295 (1966) 300–319. veloped by SHS and laser cladding, Mater. Sci. Eng. A 528 (2011) 5327–5335.
[9] K.L. Johnson, Contact Mechanics, Cambridge University Press, London, UK, 1985. [34] M. Schlögl, C. Kirchlechner, J. Paulitsch, J. Keckes, P.H. Mayrhofer, Effects of
[10] J. Blažek, J. Musil, P. Stupka, R. Čerstvý, J. Houška, Properties of nanocrystalline structure and interfaces on fracture toughness of CrN/AlN multilayer coatings, Scr.
Al–cu–O films reactively sputtered by DC pulse dual magnetron, Appl. Surf. Sci. 258 Mater. 68 (2013) 917–920.
(2011) 1762–1767. [35] R. Daniel, M. Meindlhumer, J. Zalesak, B. Sartory, A. Zeilinger, C. Mitterer,
[11] J. Musil, R. Jílek, M. Meissner, T. Tölg, R. Čerstvý, Two-phase single layer Al-O-N J. Keckes, Fracture toughness enhancement of brittle nanostructured materials by
nanocomposite films with enhanced resistance to cracking, Surf. Coat. Technol. 206 spatial heterogeneity: a micromechanical proof for CrN/Cr and TiN/SiOx multi-
(2012) 4230–4234. layers, Mater. Des. 104 (2016) 227–234.
[12] J. Musil, J. Sklenka, R. Čerstvý, T. Suzuki, T. Mori, M. Takahashi, The effect of