Thermally induced delamination of amorphous hydrogenated carbon coatings monitored by positron beam analysis

Surface and Coatings Technology 180–181 (2004)207–212 0257-8972/04/$ - see front matter 䊚2003 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2003.10.042 Thermally induced delamination of amorphous hydrogenated carbon coatings monitored by positron beam analysis R. Escobar Galindo *, A. van Veen , H. Schut , F. Rabbani , G.C.A.M. Janssen , a,aa b b J.Th.M. De Hosson c Defects in Materials, Interfaculty Reactor Institute, Delft University of Technology, Mekelweg 15, NL-2629 JB Delft, The Netherlands a Department of Materials Science and Technology, Delft University of Technology, Rotterdamseweg 137, NL-2628 AL Delft, The Netherlands b Materials Science Centre and NIMR, University of Groningen, Nijenborgh 4, NL-9747 AG Groningen, The Netherlands c Abstract In this work we have studied the adhesion of PVD amorphous hydrogenated carbon (a-C:H)thin coatings (-200 nm)on crystalline silicon by means of positron beam analysis (PBA). PBA allows the study of the a-C:H layer and the interface by monitoring changes on the Doppler broadening parameters Sand W. Depending on the deposition condition, using PBA some samples were found to be delaminated after deposition. Annealing experiments were performed on samples that did not show signs of delamination after deposition. Some of these samples were distinguished by the fact that nitrogen was used in the manufacturing process. Samples were annealed for 30 min in vacuum up to 600 8C in order to remove gases absorbed at the interface. PBA experiments were performed after each annealing step. For the sample without N there are no S–Wchanges 2 either in the coating or at the interface until 600 8C is reached. On the other hand, for the sample containing N , there is a 2 gradual change in these parameters with temperature. In this coating, after annealing to 500 8C, the increase in Wis related to release of H and consequent restructuring of the layer. The decrease in the value of Sbetween 500 and 600 8C indicates the 2 formation of graphitic crystallites. There is open volume at the coatingysubstrate interface since the Sparameter gradually increases while Wremains constant until 400 8C. This behaviour in S–Wis related to the removal of physisorbed hydrogen. At approximately 400 8C the H bonded to C at tetrahedral sites is also released which leads to an internal restructuring of the ‘lattice’, hence a decrease in the value of Sis observed. Finally, at 600 8C both coatings delaminate, as the interface parameters tend towards the same values as those of the sample which had delaminated after deposition. 䊚2003 Elsevier B.V. All rights reserved. Keywords: Positron beam analysis; Doppler broadening; Annealing; a-c:H; Delamination 1. Introduction Diamond-like carbon (DLC)coatings are of great interest in a wide range of industrial applications because of their low coefficient of friction, good wear rate, high hardness, optical transparency, low surface energy, chemical inertness and biocompatibility w1x. These films have been used as protective coatings in microelectronics devices w2–4x, in optics as commercially available scratch resistant materials for use in laser barcode scanners w5x, in magnetic and magneto-optic recording disks w6x, in tribological applications such as self lubricating bearings and metal forming tools and in biomedicine w7xw 8x. The diamond-like properties of the DLC *Corresponding author. Tel.: q31-15-278-1612; fax: q31-15-278- 6422. E-mail address: [email protected] (R. Escobar Galindo). films are mainly determined by the sp ysp ratio and 32 the hydrogen content. In a recent work Robertson w9x reviews in great detail the deposition and characterisation methods, electronic structure, defects, mechanical properties and applications of diamond-like-amorphous carbon films. The different forms of amorphous carbon are also described, including amorphous hydrogenated carbon (a-C:H)coatings, the subject of this study. One of the main concerns in the applications of DLC coatings is the high compressive stress developed by the films during manufacture. In many cases this results in the failure of the coating by delamination or buckling w10x. Nitrogen incorporation in a-C:H has proven to be an effective way of reducing the internal stresses of the films without noticeably changing the hardness w11– 14x. One of the reasons adduced for stress reduction is the increase in the density of voids in the films. 208 R. Escobar Galindo et al. / Surface and Coatings Technology 180–181 (2004) 207–212 Table 1 Description of the deposition conditions used to manufacture the amorphous carbon coatings and resultant film thickness Sample CH 4 Ar N 2 H 2 Bias Temperature Thickness (sccm)(sccm)(sccm)(sccm)(V)(8C)(nm) 1 50 0 0 50 550 270 135 2 50 0 20 50 550 270 90 3 100 80 20 100 600 ;300 200 4 50 90 10 50 400 ;300 125 Positron beam analysis (PBA)is a non-destructive technique particularly sensitive to the presence of open volume defects or voids w15x. This technique has been successfully applied to the study of different types of DLC films w11,16–18x. In this work we apply PBA to study the changes observed in the coatings and at the interface of a-C:H films as a result of annealing treatment. These results will be compared to Raman experiments performed on the same samples and presented in Ref. w19x. 2. Experimental Four different amorphous hydrogenated carbon coatings (a-C:H)were produced by a non-conventional process w19x. A Hauzer PVD machine was used to generate a reactive glow made from the decomposition of a range of gases with and without N , with CH as 24 the carbon containing gas. A description of the deposition conditions is given in Table 1. A DC-bias voltage of 550 V was applied to the substrate table and the gases were introduced at specific flow rates high enough to generate plasma. Two control samples (1 and 2 in Table 1)were made with temperature regulation i.e. the deposition was started at 270 8C and this temperature was maintained during the test by applying external heating not generated by the plasma. The deposition conditions and the gas mixture are identical for both samples except for the nitrogen content. The coatings were deposited onto Si N100Msingle crystal substrates, for a deposition time of 1 h. Prior to the coating procedure an etch step was included using Ar gas to remove the surface oxide layer on the Si substrate. After deposition the samples were annealed for 30 min under a vacuum of approximately 10 Pa in order y5 to remove gases absorbed at the interface. The annealing temperatures were: 150, 300, 400, 500 and 600 8C. PBA was used to analyse the coatings between each annealing experiment. The PBA experiments were performed with the Delft variable energy positron beam (VEP)w20x. The positrons were injected into the samples with energies tuned between 100 and 30 keV. The maximum implantation energy corresponds to a typical implantation depth of ;4mm for materials with a density of r;3 g cm . y3 All experiments were carried out at room temperature (RT)under a vacuum of approximately 10 Pa. PBA y6 results are described in terms of two parameters (Sand W)related to the Doppler broadening of the 511 keV annihilation photo-peak. The Sparameter indicates the fraction of positrons that annihilates with low momentum electrons (valence or conduction electrons). This parameter is related to the open volume defects present in the sample (e.g. pores). The Wparameter indicates the fraction of positrons that annihilates with high momentum electrons (core electrons). This parameter is related to the chemical environment where the annihilation takes place. Both parameters can be combined in S–Wmaps with a third variable (i.e. implantation energy, annealing temperature or strain)as a running parameter. The different positron annihilation sites (layers)can then be distinguished. The data was analysed with the VEPFIT w21xprogram. 3. Results and discussion 3.1. PBA on as deposited amorphous carbon coatings In Fig. 1 the PBA experimental results on the as received samples are presented. The samples were described in terms of 4 layers each of them ascribed to different positron annihilation sites: (1)surface, (2)a- C:H coating, (3)coating-substrate interface and (4) silicon substrate. These sites will hereafter be referred to as cluster points (CPs). The results summarised in Table 2 show that the control samples 1 and 2 have constant values of Sand Wfor the coating layer. The S parameter is similar for both samples but the Wparameter is higher for the coating with nitrogen (sample 2). This difference in Wis related to annihilation in the vicinity of nitrogen as reported to occur for thin Zirconia layers with argon w22x. The presence of the interface (CP3)is clearly observed for both samples in Fig. 1b from the turn of the S–Wvalues after the coating layer (CP2)towards the silicon substrate (CP4). The Svalue characterising the interface of sample 2 is higher than sample 1 without N indicating a more porous structure 2 at the interface of the first sample. Sample 3, although deposited under different conditions than the control samples 1 and 2, has similar positron results. On the other hand, sample 4 shows a completely different S– Wmap compared to the other three as deposited samples. 209R. Escobar Galindo et al. / Surface and Coatings Technology 180– 181 (2004) 207–212 Fig. 1. Doppler broadening results on amorphous carbon films. The Sparameter is plotted as a function of the positron implantation energy for (a)samples 1 and 2 and (c)samples 3 and 4 as deposited on silicon. The S–Wmaps are shown for (b)1 and 2 and (d)3 and 4 as deposited samples, respectively. The lines represent the VEPFIT analysis results. The numbered circles indicate the positron annihilation sites ascribed to 1—Surface; 2—Coating layer; 3—Interface and 4—Substrate. Table 2 VEPFIT results on amorphous carbon samples, studied at RT (before delamination)and after 600 8C annealing (after delamination) No. Before annealing After annealing S coat W coat S delam W delam S interf W interf S coat W coat S delam W delam 1 0.931 1.299 – – 0.946 1.429 0.919 1.536 0.907 1.666 2 0.932 1.398 – – 0.964 1.328 0.929 1.481 0.906 1.675 3 0.933 1.315 – – 0.989 1.121 0.918 1.572 0.906 1.637 4 0.930 1.308 0.908 1.658 – – – – – – The absolute errors in Sand Ware DS;0.002 and DW;0.03, respectively. In Fig. 1c, the Sparameter of sample 4 first increases to a value similar to the other samples but, instead of further increasing towards the Svalue of the silicon substrate, it drops to a value close to the surface. In the S–Wmap of Fig. 1d this change is clearly observed as the S–Wdata after reaching the coating cluster point (CP2)looped back in the direction of the surface to a cluster point (CP3)with a low Sand high W.Itis interesting to notice that after this CP the S–Wdata lie on a straight line towards the silicon substrate CP. We ascribed this different behaviour to delamination of the coating as we will explain it later in the next section. 3.2. PBA on annealed amorphous carbon coatings Annealing experiments were performed on samples 1, 2 and 3, coatings without signs of delamination after deposition. Samples were annealed for 30 min in vacuum up to a maximum temperature of 600 8C. PBA experiments were performed after each annealing step. In Fig. 2, we can observe the PBA results for the three annealed samples after the last annealing step as compared with the results for as deposited sample 4. The most remarkable observation is that the interface cluster point (CP3)for the three annealed samples is located at 210 R. Escobar Galindo et al. / Surface and Coatings Technology 180– 181 (2004) 207–212 Fig. 3. Schematic representing the different positron annihilation sites for a coatingysubstrate system with (a)a closed interface and (b) with an open interface (delamination).In(b)the positron parameters of site 3 are an average of the positron parameters of the two new surfaces (1* and 4*)created after delamination. The schematic is not to scale. Fig. 2. Doppler broadening results on amorphous carbon films. The Sparameter is plotted as a function of the positron implantation energy for (a)samples 1 and 2 after 600 8C annealing and (c)samples 3 after 600 8C annealing and 4 as deposited. The corresponding S–Wmaps are shown for (b)samples 1 and 2 and (d)samples 3 and 4, respectively. The lines represent the VEPFIT analysis results. The numbered circles indicate the positron annihilation sites ascribed to (1)surface; (2)coating layer; (3)interface and (4)substrate. the same position as for as deposited delaminated sample 4(see S–Wmaps in Fig. 2b and d). With increasing positron implantation energy the S–Wdata lie on a straight line towards the CP of the silicon substrate. The process can be schematically described in Fig. 3. Before delamination occurs we have a closed interface (3) between the coating and the substrate. This results in a typical S–Wmap as observed in Fig. 1b with a loop between CP2 and CP4 indicating the presence of the interface (CP3). Once delamination occurs there is no longer a close interface and positrons annihilate in the new internal surfaces created (1* and 4* in Fig. 3). CP3 is now ascribed to positron annihilation in these new surfaces. This is the reason the S–Wcurves for samples annealed to 600 8C, and as deposited sample 4, turn back towards surface values (Fig. 2b and d). After that the S–Wcurves map a straight line to the substrate, as a free-standing sample would, with only two positron annihilation sites: surface and bulk. The control samples 1 and 2 were manufactured to allow for the study of the role of nitrogen in the performance of the coatings during annealing. In Fig. 4, we follow the changes in S–Wvalues with temperature, inside the coating (Fig. 4a), and at the interface (Fig. 4b). In Fig. 5, the Sparameter for the interface is plotted as a function of annealing temperature. The sample without N (sample 1)shows no significant change in 2 the values of S–Weither in the coating or at the interface until 600 8C is reached. As mentioned before, at this temperature there is a dramatic change in the interface 211R. Escobar Galindo et al. / Surface and Coatings Technology 180– 181 (2004) 207–212 Fig. 4. S–Wmaps showing the CPs ascribed to positron annihilation as a function of annealing temperature, (a)inside the coating and (b) at the interface (the control samples 1—no N , is represented by open 2 circles and 2—with N , is represented by closed circles). Both maps 2 are plotted using the same scale in order to compare the changes observed. Fig. 5. The Sparameter corresponding to the interface of the control samples 1 (no N , open circles)and 2 (N , closed circles)is plotted 22 as a function of annealing temperature. The label RT denotes the samples prior to the annealing process. CP indicating delamination of the film. There is a more gradual change for the sample containing N (sample 2 2). Inside the coating the Wparameter increases after annealing at 500 8C indicating a structural change in the layer due to release of hydrogen and further graphitisation as deduced from Raman spectroscopy experiments w19x. The open porosity, already present at the interface of the as deposited sample, becomes larger as the Sparameter gradually increases with a constant W (no structural change)until 400 8C is reached (Fig. 5). This observation corresponds to the removal of physisorbed hydrogen from interfacial pores w23x.Soitis important to note that positrons distinguish a large open porosity (related to an increase in S)of a delaminated interface (related to changes in the value of Sand W towards surface values). However, according to Raman experiments w19x, at approximately 400 8C the gas released during annealing is not only the release of ‘unbound’ hydrogen, but also the H bonded at tetrahedral sites to C. The growth of the sp clusters, as seen 2 by Raman spectroscopy, was also monitored by positrons showing a decrease in the value of Sat 500 8C and an increase in the value of Wat the interface. Finally at 600 8C the same S–WCP is reached as for sample 1 and 3 at 600 8C, and delaminated sample 4 before the annealing process (RT stands for room temperature in Fig. 4). 4. Conclusions PBA has been performed on PVD hydrogenated amorphous carbon thin films synthesized under different deposition conditions. Delamination was observed after deposition in one of the samples studied. Annealing in vacuum up to 600 8C provokes the appearance of the first stages of delamination in the rest of the coatings. This was monitored by changes in the interfacial positron annihilation site. The role of nitrogen inclusion in the coating was followed through the study of two control samples deposited under the same conditions except for the inclusion of nitrogen in one case. The interface of the nitrogen-containing sample has a more open structure or pores. These pores are decorated with physisorbed hydrogen, whose release was monitored by positrons after annealing the sample up to 400 8C. Further annealing of the sample at 500 8C causes a restructuring both at the interface and in the coating due to release of bonded hydrogen and further graphitisation. The sample without nitrogen did not change noticeably during the annealing procedure until 500 8C. At 600 8C both samples showed the occurrence of delamination. Acknowledgments The authors thank the Foundation for Fundamental Research on Matter (FOM)and the Dutch Technology 212 R. Escobar Galindo et al. / Surface and Coatings Technology 180– 181 (2004) 207–212 Foundation STW (project number GTF.4901), for their financial and technical support. References w1xJ. Robertson, Surf. Coat. Technol. 50 (1992)185–203. w2xQ. Wei, J. Narayan, Int. Mater. Rev. 45 (2000)133–164. w3xQ. Wei, J. Sankar, J. Narayan, Surf. Coat. Technol. 146–147 (2001)250–257. w4xA.M. Baranov, A.E. Varfolomeev, A.A. Nefedov, et al., Diamond Relat. Mater. 9 (2000)649–653. w5xF.M. Kimock, B.J. Knapp, Surf. Coat. Technol. 56 (1993) 273–279. w6xJ. Robertson, Tribol. Int. 36 (2003)211–228. w7xV.M. Tiainen, Diamond Relat. Mater. 10 (2001)153–160. w8xM. Brizuela, L.A. Garcıa, J.L. Viviente, I. Braceras, J.I. Onate, ´ J. Mater. Sci. Mater. Med. 13 (2002)1129–1133. w9xJ. Robertson, Mater. Sci. Eng. R 37 (2002)129–281. w10xM. Wittmer, D. Ugolini, J. Eitle, P. Oelhafen, Appl. Phys. A 48 (1989)559–566. w11xY.H. Cheng, Y.P. Wu, J.G. Chen, X.L. Qiao, C.S. Xie, Diamond Relat. Mater. 8 (1999)1214–1219. w12xS.R.P. Silva, J. Robertson, G.A.J. Amaratunga, et al., J. Appl. Phys. 81 (1997)2626–2634. w13xP. Wood, T. Wydeven, O. Ysuji, Thin Solid Films 258 (1995) 151–158. w14xD.F. Franceschini, C.A. Machete, F.L. Freire Jr., Appl. Phys. Lett. 60 (1992)3229–3231. w15xA. van Veen, H. Schut, P.E. Mijnarends, in: P. Coleman (Ed.), Positron Beams and their Applications, World Scientific Publishing Co, 2000, Chapter 6. w16xD.T. Britton, M. Harting, M. Hempel, D. Gxawu, K. Uhlman, ¨ Appl. Surf. Sci. 149 (1999)130–134. w17xF. Rossi, B. Andre, A. van Veen, et al., J. Phys. III 5 (1995) ´ C1 179–C1 191. w18xL.G. Jacobsohn, G. Capote, M.E.H. Maia da Costa, D.F. Franceschini, F.L. Freire Jr., Diamond Relat. Mater. 11 (2002) 1946–1951. w19xF. Rabbani, R. Escobar Galindo, W.M. Arnoldbik, S. van der Zwaag, A. van Veen, H. Schut, Diamond Relat. Mater., submitted for publication. w20xA. van Veen, J. Trace Microprobe Tech. 8 (1990)1–29. w21xA. van Veen, H. Schut, J. de Vries, R.A. Hakvoort, M.R. Ijpma, in: P.J. Schultz, G.R. Massoumi, P.J. Simpson (Eds.), AIP 218, Positron Beams for Solids and surfaces, 1990, pp. 171–196. w22xJ.P. Holgado, R. Escobar Galindo, A. van Veen, H. Schut, J.Th.M. de Hosson, A.R. Gonzalez-Elipe, Nucl. Instrum. Meth. ´ B 194 (2002)333–345. w23xF. Rabbani, B.M. Vogelaar, Diamond Relat. Mater., accepted for publication.