View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/C3RA46944A (Paper) RSC Adv., 2014, 4, 21108-21114

Bio-tribological TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H multilayer coatings development with a built-in mechanism of controlled wear

L. Major*a, J. M. Lacknerb and B. Majora
aInstitute of Metallurgy and Materials Science, Polish Academy of Sciences, 30-059 Cracow, Reymonta St.25, Cracow, Poland. E-mail: l.major@imim.pl; Fax: +48-12-2952804; Tel: +48-12-2952800
bJOANNEUM RESEARCH-Materials-Institute for Surface Technologies and Photonics, Leobner Strasse 94, 8712 Niklasdorf, Austria

Received 22nd November 2013 , Accepted 26th March 2014

First published on 26th March 2014

Abstract

Development of a new generation of multilayer coatings as well as a microstructure understanding of the mechanisms operating at the smallest length scale (nano- and atomic-scale) during wear, opens an avenue for the fabrication of future high-tech functional surfaces. Coatings for the presented work were fabricated by a pulsed laser deposition supported by magnetron sputtering. Microstructure characterization has been performed on as-deposited coatings as well as on coatings after mechanical wear test. Thin foils for detailed TEM microstructure observation were cut directly from the mechanically deformed area, using the FIB technique. Wear mechanisms operating at the small length scale of TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H multilayer coatings subjected to mechanical wear was studied by means of transmission electron microscopy (TEM). Cracking of the multilayer systems propagated layer by layer. The highest stress concentration during mechanical uploading was moved through the multilayer coating by breaking only one layer at the time.

1. Introduction

In recent time much attention in the coatings development has been paid to their multifunctionality.1 Multilayer coatings can lead to benefits in performance over comparable single-layers and can combine properties of different materials with one protective coating.2–4 Crack initiation and propagation are often responsible for wear and the removal of coatings. The introduction of a number of interfaces parallel to the substrate surface can act to deflect cracks or provide barriers to the dislocation motion, increasing the toughness and hardness of the coating.5–9 Despite much research on the development of multilayer coatings with superior mechanical and tribological properties, a question still remains about the mechanisms, operating at the smallest length scale, underlying the mechanical response.10 The wear resistance is often attributed to the high hardness as well as to good chemical stability. Achievement of high hardness ought to be linked with a large number of internal interfaces, which act as sites of energy dissipation and crack deflection. Coatings can be applied for a diversity of reasons, namely for surgical tools. Requirements for tribological, multilayer coatings for medical tools, particularly are very high. In this field they can be applied to increase hardness, improve wear and corrosion resistance. The goal of the presented paper was to study wear mechanisms operating at the small length scale in multilayer TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H coatings subjected to mechanical testing, where very thin titanium layers (∼7 nm) were deposited at each TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H interface.

2. Materials and methods

A hybrid PLD system (Pulsed Laser Deposition connected with magnetron sputtering) equipped with a high purity titanium (99.9% Ti) and carbon (graphite) targets were used for multilayer coating depositions. By the application of magnetron sputtering in PLD coating plants, higher deposition rates can be reached and a very good film adhesion is achieved even at room temperature. Coatings were produced by the sequential deposition of amorphous carbon (a-C[thin space (1/6-em)]:[thin space (1/6-em)]H) and titanium nitride (TiN). Pure Ti layers were deposited in an argon (non-reactive) atmosphere, while for the TiN deposition the atmosphere was gradually switched to nitrogen. Additionally, to increase the quality of the coating adhesion and to reduce the residual stress concentration, the thin metallic titanium (Ti) layers were deposited at each interface. Details of the deposition process have been described elsewhere.11 The set of multilayer coatings with a different number of layers and phase ratios at constant coating thicknesses (1.5 μm) were produced. Additionally a-C[thin space (1/6-em)]:[thin space (1/6-em)]H and TiN single layer coatings were deposited as references. The list of coatings is presented below:

- a-C[thin space (1/6-em)]:[thin space (1/6-em)]H single layer

- TiN single layer

- 2 × TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H (4 layers with TiN and a-C[thin space (1/6-em)]:[thin space (1/6-em)]H with phase ratio 1[thin space (1/6-em)]:[thin space (1/6-em)]1)

- 8 × TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H (16 layers with phase ratio 1[thin space (1/6-em)]:[thin space (1/6-em)]1)

- 32 × TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H (64 layers with phase ratio 1[thin space (1/6-em)]:[thin space (1/6-em)]1)

- 8 × TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H (16 layers with phase ratio 1[thin space (1/6-em)]:[thin space (1/6-em)]2)

- 8 × TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H (16 layers with phase ratio 1[thin space (1/6-em)]:[thin space (1/6-em)]4)

- 8 × TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H (16 layers with phase ratio 4[thin space (1/6-em)]:[thin space (1/6-em)]1)

Bio-medical tests were done using smooth muscle cells.12,13 Smooth muscle cells (SMCs) were purchased from Lonza. Each vial had a concentration of 500[thin space (1/6-em)]000 cells per mL. The cells were stored in liquid nitrogen until use. Before adding cells, the medium was warmed at 37 °C in a water bath. Cells were taken from the liquid nitrogen container and placed for 2–3 min in a 37 °C water bath. SMCs were deposited directly on the surfaces of the coatings. After three days the cells were fixed in 4% paraformaldehyde. Then the cells were permeabilised in a Triton X-100 0.2% detergent for 4 min. The actin cytoskeleton was marked with AlexaFluor Phalloidin, the nucleus with DAPI. Fluorescent dyes were excited with the appropriate laser wavelength, 488 nm for actin cytoskeleton visualization, 405 nm to visualize nucleus of the cell. The cell adhesion to the coating surfaces were observed by confocal microscopy (Carl Zeiss Exciter 5).

The mechanical properties of the coatings were investigated by means of a ball-on-disc (Al2O3 ball with 6 mm radius) mechanical test. Details of the test are given in the table (Table 1).

Table 1 Parameters of the wear (ball-on-disc) test
Type of the test Load (N) Number of cycles Friction radius (mm) Line speed (m s−1)
Low stress condition (ball-on-disc) 1 20[thin space (1/6-em)]000 5 0.06
High stress condition 5 5000 4 0.05

Residual stress distribution was modelled by the finite elements technique. The microstructure of the as-deposited coatings and after the mechanical tests, was studied using scanning (SEM) and transmission (TEM) (TECNAI G2 F20 FEG (200 kV)) electron microscopes, which allows the microstructure observation at the smallest scale, especially in the high resolution mode (high resolution transmission electron microscopy-HRTEM).22 In HRTEM columns of atoms are observed. The chemical composition was characterized by the energy dispersive spectroscopy technique (EDS). Thin foils for TEM analysis were prepared using the focused ion beam technique (FIB-gallium ions) (QUANTA 200 3D Dual Beam), together with the OmniProbe in-situ lift out system. It allowed us to prepare foils directly from places of interest (from mechanically deformed areas-wear tracks).

3. Results

The main goal of the presented paper was the development of bio-tribological multilayer TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H coatings and to describe the microstructure changes and wear mechanisms operating at the small length scale of the described multilayer coatings after mechanical uploading.

The microstructure characterization has been performed on as-deposited coatings, as well as on those after mechanical wear test to observe changes in the microstructure caused by mechanical uploading. The potential application of the presented coatings are cardio-surgical tools.

To improve the adhesion of the coating to the substrate, a titanium (Ti) metallic buffer layer was deposited as a first one. Subsequently an appropriate multilayer structure was produced, where carbon layers were placed sequentially with titanium nitride layers.

3.1. Smooth muscle adhesion test

Medical tools (like tweezers) by contacting with vessels have the first contact with smooth muscle cells. The bio-test with these cells was treated as a criteria for the best multilayer coating selection for microstructure characterisations. The best cell-material interaction adhesion was found for the 8 × TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H coating (Fig. 1).
image file: c3ra46944a-f1.tif
Fig. 1 Smooth muscle cell adhesion observed by confocal microscopy.

3.2. Micromechanical analysis

In the low stress condition of the ball-on-disc test, wear indexes for multilayer coatings were between indexes for single layer coatings. The lowest wear was found for the a-C[thin space (1/6-em)]:[thin space (1/6-em)]H single layer coating (Fig. 2a). In the high stress condition the role of the multilayer structure started to be visible. Wear indexes for multilayer coatings were on the same level or even lower than index for the a-C[thin space (1/6-em)]:[thin space (1/6-em)]H single layer coating (Fig. 2b).
image file: c3ra46944a-f2.tif
Fig. 2 Wear (ball-on-disc) test results: (a) low stress condition (1 N), (b) high stress condition (5 N).

3.3. Microstructure analysis of as-deposited coatings

Connecting the biological test results with the mechanical ones, a coating with the optimal bio-mechanical properties was selected for the detailed microstructure change analysis caused by mechanical uploading. The characterisation has been done by transmission electron microscopy. The initial microstructure of the as deposited coating is described in Fig. 3. The carbon phase, in the multilayer system was characterised as having an amorphous structure, which confirmed the electron diffraction pattern, comprising blurred rings around a main spot. TiN and Ti layers consisted of columnar nano-crystallites. The titanium phase was deposited both as a thick buffer layer (first from the substrate) and as very thin inter-layers at each TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H interface, which is easily visible at the higher magnification bright field image (Fig. 4). The small amount of the Ti phase between the TiN and a-C[thin space (1/6-em)]:[thin space (1/6-em)]H layers improves the adhesion of TiN to a-C[thin space (1/6-em)]:[thin space (1/6-em)]H, which otherwise would be very weak due to a high residual stress. The thickness of the individual titanium layers was approximately 7 nm. The presence of the metallic titanium phase was confirmed by the EDS qualitative chemical analysis and the HRTEM technique in the atomic scale (Fig. 5). The above observations correspond to the deposition process. The pure titanium layers were deposited in an argon atmosphere. As a result, a very thin titanium phase was formed at each TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H interface. The atmosphere was subsequently switched to a mixture of argon and nitrogen. The nitrogen flow was gradually opened while the argon flow was closed. After reaching the half thickness of the TiN layer, the nitrogen flow was gradually closed, and the argon flow was opened. Similar to before, deposition from the Ti target was finished in an argon atmosphere, allowing the formation of a small amount of Ti phase on the next TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H interface. Concerning only the TiN phase in the layer, the Ti to N ratio was changed with the distance from the interface. It happened because the gases flow changed with the distance from the interface. The flow was gradually changed.
image file: c3ra46944a-f3.tif
Fig. 3 TEM bright field image of an as-deposited 8 × TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H multilayer coating, together with selected area electron diffraction patterns from selected parts of the coating.

image file: c3ra46944a-f4.tif
Fig. 4 TEM bright field image of an as-deposited 8 × TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H multilayer coating (higher magnification than Fig. 3, revealing thin Ti layers presented at each interface).

image file: c3ra46944a-f5.tif
Fig. 5 HRTEM image of the one Ti/TiN/Ti fragments of the coating; (a) bright field image of the one Ti/TiN/Ti fragments of the coating, (b) EDS qualitative chemical analysis, (c) HRTEM image of the analyzed fragment.

3.4. Microstructure analysis of coatings after mechanical test

The second step of the microstructure characterization was the analysis of the cross-section of the TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H multilayer coatings after the mechanical wear test. The topography of the coating after the test is presented by the SEM technique (Fig. 6). The place from which the thin foil was prepared is marked on the image. A detailed analysis of the wear mechanisms have been done by the TEM method. One of the main advantages of the described coatings was that cracking has not been performed suddenly. It occurred layer by layer (Fig. 7). The highest, localized stress concentration was moved during the coating deformation from the bottom to the top part of the coating, which was confirmed by numerical modelling (Fig. 8). The layers reunion after movement of individual layers to the closest layer of the same phase was possible by the presence of metallic thin layers at each interface. Another example shows layers motion towards the closed neighbouring layer, however reunion was performed again by thin Ti layers (Fig. 9). Therefore, layer motions were obtained towards the closed neighbour of the same phase, as well as towards the closest neighbouring layer.
image file: c3ra46944a-f6.tif
Fig. 6 SEM image of the topography of multilayer an 8 × TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H coating after the mechanical wear test.

image file: c3ra46944a-f7.tif
Fig. 7 TEM bright field image of the cracking place in an 8 × TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H coating (cross section): (a) TEM bright field image of the cracked place, (b) image revealing presence of thin Ti metallic layers at each TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H interface, (c) the magnified area of one crack presenting the ‘layers motion mechanism’, (d) place of layers reunion after ‘layers motion’.

image file: c3ra46944a-f8.tif
Fig. 8 Finite elements modelling of residual stress distribution during the mechanical test.

image file: c3ra46944a-f9.tif
Fig. 9 TEM image presenting another example of layers motion and reunions mostly because of the presence of the thin Ti metallic layers.

4. Discussion

In general, the layered structure can be vulnerable to failure due to a heterogeneous stress field and reduced toughness at the interfaces. Failures in layered structures can be complicated due to an elastic and toughness anisotropy. Residual stress and elastic mismatch can lead to mixed-mode loading for cracks on and off the interface, as presented by Shaw, Rice, Fleck and Ritchie.14–17 Elastic/plastic deformations lead to microcracking, providing a mechanism for subsequent wear through surface delamination as described by Luo.6 For cracking of the conventional multilayer system to take place, all the individual layers would need to be broken at the same time. It is suggested that cracking can be controlled by a crack-healing ability or by a plastic deformation and strong interfaces.18–20 Development of a new generation of coatings, which have both passive mechanical characteristics originated from the matrix material and active response sensitive to change in the local environment, which allows the fabrication of the future high-tech functional surfaces.21

In the described experiment, to reduce the high residual stress, a metallic buffer layer of 7 nm was deposited at each TiN/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H interface. Such a composition caused the unusual behavior of the multilayer coating during the mechanical wear, in particular the cracking process. The cracking line, in the described case, because of the presence of very thin Ti layers at each interface, could move through the multilayer coating in the direction of the applied stress by breaking only one layer at a time. Individual layers broke, moved and formed new connection with its neighbour. The mechanism has never been described before in the literature. The layers reunion probably was carried out by moving one broken layer relative to the other along a common interface (in this case-cracking line), while applying a compressive force. The friction heating generated at the interface softens both components, particularly plastically deformed thin Ti layers presented at each interface. The relative motion was then stopped. The connections were formed in the solid state. Anyway, the ‘layers motion’ phenomenon is similar to the edge dislocation motion (Fig. 10).22,23


image file: c3ra46944a-f10.tif
Fig. 10 Attempt to compare the ‘layers-motion mechanism’ to the edge dislocation motion: (a) movement of edge dislocation through the crystal lattice,23 (b) TEM bright field image of the layers motion mechanism.

5. Conclusion

The following conclusion can be drawn from this particular study:

- Multilayer structure and metallic interlayers in the TiN/Ti/a-C[thin space (1/6-em)]:[thin space (1/6-em)]H coatings play major roles in controlling the deformation process.

- Propagation of the deformation was realised layer by layer.

- The cracking line, due to the presence of very thin Ti layers at each interface, could move through the multilayer coating in the direction of the applied stress by breaking only one layer at a time.

Acknowledgements

Research project was financed by the National Science Centre (Polish-Narodowe Centrum Nauki, abbr. NCN) no. 3066/B/T02/2011/40. Polish-Austrian exchange project (2012–2014) 023/2012/2013/2014; 8548/R 12/R 14.

References

  1. Y. Brechet and J. D. Embury, Architectured materials, Scr. Mater., 2013, 68, 1–3 CrossRef CAS PubMed.
  2. Q. Luo, S. Cai Wang, Z. Zhou and L. Chen, J. Mater. Chem., 2011, 21, 9746–9756 RSC.
  3. J. Morgiel, L. Major, B. Major, J. M. Lackner and L. Nistor, J. Microsc., 2006, 223, 237–239 CrossRef CAS PubMed.
  4. T. Hattori, Y. Kaneko and S. Hashimoto, J. Mater. Sci., 2008, 43, 3923–3930 CrossRef CAS PubMed.
  5. S. Kumar, D. Zhou, D. E. Wolfe, J. A. Eades and M. A. Haque, Scr. Mater., 2010, 63, 196–199 CrossRef CAS PubMed.
  6. Q. Luo, W. M. Reinforth and W.-D. Münz, Scr. Mater., 2001, 45, 399–404 CrossRef CAS.
  7. Q. Luo, W. M. Reinforth and W. D. Münz, Wear, 1999, 225-229, 74–82 CrossRef CAS.
  8. L. Major, W. Tirry and G. Van Tendeloo, Surf. Coat. Technol., 2008, 202, 6075–6080 CrossRef CAS PubMed.
  9. J. M. Lackner, L. Major and M. Kot, Bull. Pol. Acad. Sci.: Tech. Sci., 2011, 59, 343–355 CAS.
  10. P. Goodhew, Nano Today, 2006, 1, 40–43 CrossRef ISSN:1748 0132, Elsevier.
  11. J. M. Lackner, W. Waldhauser and R. Ebner, Surf. Coat. Technol., 2004, 188–189, 519–524 CrossRef CAS PubMed.
  12. R. Major, F. Bruckert, J. M. Lackner, W. Waldhauser, M. Pietrzyk and B. Major, Bull. Pol. Acad. Sci.: Tech. Sci., 2008, 56, 223–228 Search PubMed.
  13. R. Major, J. M. Lackner, K. Gorka, P. Wilczek and B. Major, RSC Adv., 2013, 3, 11283–11291 RSC.
  14. M. C. Shaw, Eng. Fract. Mech., 1998, 61, 49–74 CrossRef.
  15. J. R. Rice, J. Appl. Mech., 1988, 55, 98–103 CrossRef.
  16. N. A. Fleck, J. W. Hutchinson and S. Zhigang, Int. J. Solids Struct., 1991, 27, 1683–1703 CrossRef.
  17. R. O. Ritchie, R. M. Cannon, B. J. Dalgleish, R. H. Dauskardt and J. M. McNaney, Mater. Sci. Eng., A, 1993, 166, 221–235 CrossRef.
  18. G. M. Song, Y. T. Pei, W. G. Sloof, S. B. Li, J. Th, M. De Hosson and S. van der Zwaag, Scr. Mater., 2008, 58, 13–16 CrossRef CAS PubMed.
  19. N. J. M. Carvalho, J. Th and M. De Hosson, Acta Mater., 2006, 54, 1857–1862 CrossRef CAS PubMed.
  20. Z. H. Xie, M. Hoffman, P. Munroe, A. Bendavid and P. J. Martin, Acta Mater., 2008, 56, 852–861 CrossRef CAS PubMed.
  21. D. G. Shchukin and H. Möhwald, Small, 2007, 3, 926–943 CrossRef CAS PubMed.
  22. J. Weertman and J. R. Weertman, Elementary Dislocation Theory, Oxford University Press, ISBN 0-19-506900-5, 1992 Search PubMed.
  23. X. Songa, F. Hofmanna and A. M. Korsunskya, Philos. Mag., 2010, 90, 3999–4011 CrossRef.
This journal is © The Royal Society of Chemistry 2014
Click here to see how this site uses Cookies. View our privacy policy here.