- Email: [email protected]
TiN multilayers grown by pulsed laser deposition
Applied Surface Science 257 (2011) 5332–5336
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Characteristics of ZrC/ZrN and ZrC/TiN multilayers grown by pulsed laser deposition D. Craciun a , G. Bourne b , G. Socol a , N. Stefan a , G. Dorcioman a , E. Lambers b , V. Craciun a,b,∗ a b
Laser Department, National Institute for Laser, Plasma, and Radiation Physics, Bucharest, Romania Major Analytical Instrumentation Center, Materials Science and Engineering, University of Florida, Gainesville, FL 32611, USA
a r t i c l e
i n f o
Article history: Received 27 May 2010 Received in revised form 16 November 2010 Accepted 16 November 2010 Available online 24 November 2010 Keywords: ZrC ZrN TiN Pulsed laser deposition Thin films Multilayers
a b s t r a c t ZrC/ZrN and ZrC/TiN multilayers were grown on (1 0 0) Si substrates at 300 ◦ C by the pulsed laser deposition (PLD) technique using a KrF excimer laser. X-ray diffraction investigations showed that films were crystalline, the strain and grain size depending on the nature and pressure of the gas used during deposition. The elemental composition, analyzed by Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS), showed that films contained a low level of oxygen contamination. Simulations of the X-ray reflectivity (XRR) curves acquired from films indicated a smooth surface morphology, with roughness below 1 nm (rms) and densities very close to bulk values. Nanoindentation results showed that the ZrC/ZrN and ZrC/TiN multilayer samples exhibited hardness values between 30 and 33 GPa, slightly higher than the values of 28–30 GPa measured for pure ZrC, TiN and ZrN films. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The excellent mechanical properties and thermochemical stability of refractory metal carbides and nitrides recommend them for important applications as hard and protective coatings [1–3]. It is generally rather difficult to deposit high quality ZrC films because of their high melting temperature, low sputtering rate, and high reactivity with oxygen or water vapors [4]. Driven by stringent requirements of special applications, such as coatings for field emission tips, nuclear fuel or outer space thermal radiators [5–7], significant progress has been made in the last decade and good quality ZrC films were obtained by using either physical vapor deposition (PVD) or chemical vapor deposition (CVD) based techniques [8–12]. As one of the most versatile research technique, pulsed laser deposition (PLD) allows for relatively low to moderate deposition temperatures of refractory carbides and nitrides without sacrificing their crystalline quality [13–15]. Some of the best mechanical properties of ZrC yet reported were measured on films deposited using the PLD technique [16,17]. Furthermore, it has been showed that by using a higher repetition rate laser for ablation, the substrate temperature could be reduced to only 300 ◦ C while
∗ Corresponding author at: Laser Department, National Institute for Laser, Plasma, and Radiation Physics, Bucharest, Romania. E-mail address: [email protected]fl.edu (V. Craciun). 0169-4332/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2010.11.106
the growth rate is increased and the crystalline quality maintained [18]. The deposition of multilayers with different crystalline lattices that could block at their interfaces the propagation of dislocations from one material to another was shown to be a way of increasing the hardness of such thin film structure [19–22]. Since it is known that generally carbides are more brittle than nitrides and that the refractory metal nitrides exhibit very similar properties to carbides, being also used as hard and protective coatings [1,3,16,23–25], we have investigated ways to further improve the quality of films by depositing multilayers of ZrC and a refractory metal nitride such as ZrN or TiN [26] and present new results here. 2. Experiment The depositions were performed in a PLD system using a KrF excimer laser ( = 248 nm, pulse duration = 25 ns, 8.0 J/cm2 fluence, 40 Hz repetition rate) that has been already described in detail elsewhere [18,26]. The films were deposited for tens of minutes from polycrystalline ZrC, ZrN, and TiN targets (Plasmaterials, Inc.) on p++ (1 0 0) Si substrates (MEMC Electronic Materials, Inc.) that were cleaned in acetone, then ethanol, rinsed in deionized water, and finally blown dry with high purity nitrogen before being loaded into the deposition chamber. The nominal substrate temperature was set at 300 ◦ C. Depositions were performed under a high purity atmosphere of CH4 or Ar (2 × 10−3 to 10−2 Pa). After
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
Fig. 1. XRR curves recorded from pure ZrC, TiN, and ZrN deposited films.
deposition, the films were slowly cooled to room temperature under the same atmosphere as that using the deposition, but at a much higher pressure of around 104 Pa. The films mass density, thickness, and surface roughness were investigated by X-ray reflectometry (XRR, Panalytical X’Pert MRD) using Cu K˛ radiation. The same instrument was used for structural characterization in symmetric and grazing incidence X-ray diffraction (XRD and GIXD) using a parallel beam configuration. The grain size was evaluated following the Scherrer equation [27] and, where possible, by the model of Williamson and Hall [28]. The chemical composition of the films was investigated by Auger electron spectroscopy (AES) in a Perkin-Elmer PHI 660 system (5 kV, 30◦ take off angle) and by X-ray photoelectron spectroscopy (XPS) in a Perkin-Elmer PHI 5100 ESCA system (300 W, Mg K˛ ). To obtain elemental depth profiles, measurements were collected after various time cycles of Ar ion sputtering (4 kV, 1–3 A/cm2 ; for XPS measurements the Ar ion beam was rastered over a larger area of 10 × 7 mm2 ). Cross section samples were prepared on a focused ion beam (Strata DB 235) for transmission electron microscopy (TEM) analysis on a JEOL 2010F instrument. The mechanical properties of the films were measured with a nanoindentation device (Hysitron inc.) equipped with a cube-corner diamond tip and set to run 100 indents per sample. The nanoindentation experiments were performed in displacement control with a contact depth of up to 30 nm. Hardness and reduced modulus were determined from the loaddisplacement data following the model of Oliver and Pharr [29], after excluding the results obtained for penetration depths smaller than 5 nm [30]. 3. Results and discussion XRR curves recorded from deposited pure ZrC, TiN and ZrN films are presented in Fig. 1. One could note different values for the critical angle of each material, a consequence of different density values. From simulations of the acquired curves using the Panalytical WinGixaTM software, which is based on Parrat’s formalism [31], we obtained values for the films density and surface roughness that are shown in Table 1, together with the deposition conditions and hardness values. The Kiessig fringes [32] were only present in the XRR curve of ZrC sample, which was much thinner than the other
5333
Fig. 2. XRD patterns (G, grazing incidence; T, symmetrical geometry) collected from ZrC/ZrN multilayers; the vertical bars are the reference diffraction peak positions of reference ZrC (full line, Card 35-0784) and ZrN (dotted line, Card 35-0753) powders [33].
Fig. 3. XRD patterns (G, grazing incidence; T, symmetrical geometry) collected from the ZrC/TiN multilayers; the vertical bars are the reference diffraction peak positions of ZrC (full line) and TiN (dotted line, Card 32-1489) powders [33].
two films. According to the simulations results, the films oxidized when exposed to the ambient, the oxide layers being estimated to be around 2–3 nm thick. The XRR curves acquired from multilayers showed evidence of interference peaks due to the presence of many interfaces, but we could not fit them satisfactorily with a model, most likely due to variations of thickness from layer to layer. However, the critical angle values of the top ZrC layer were consistent with densities very similar to those estimated for pure ZrC films. XRD and GIXD investigations showed that the deposited films were crystalline, as one can see in Figs. 2 and 3, where diffraction patterns from several representative samples acquired in symmetrical and grazing incidence geometry are displayed. The deposition conditions employed for each multilayer structure are displayed in Table 2. To facilitate a comparison, we also introduced in Table 2 and Figs. 2 and 3 some results from our previous study [26]. Since the lattice parameter of ZrC and ZrN are very similar, the recorded Bragg diffraction lines from ZrC/ZrN multilayers appeared as a convolution of the two lines, without being resolved. Analyzing the
Table 1 Deposition parameters and mechanical properties of grown pure films. Sample ZrC ZrN TiN
Deposition atmosphere (Pa) −3
2 × 10 CH4 2 × 10−3 CH4 2 × 10−3 CH4
Density (g/cm3 )
Roughness (nm)
Hardness (GPa)
Reduced Modulus (GPa)
6.45 7.09 5.21
0.5 0.6 0.5
30.1 28.5 29.9
226.0–228.3 214.5 242.3
5334
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
Table 2 Deposition parameters and mechanical properties of grown multilayers. Sample
Deposition atmosphere (Pa) −3
2 × 10 2 × 10−3 1 × 10−2 5 × 10−3 5 × 10−3 5 × 10−3 7 × 10−3 2 × 10−3
ZCN1 ZCTN1 ZCN2 ZCTN2 ZCN3 ZCTN3 ZCN4 ZCN5
CH4 CH4 Ar Ar CH4 CH4 CH4 CH4
Number of pulses and multilayers
Hardness (GPa)
Reduced modulus (GPa)
(720 + 720) × 20 (720 + 720) × 20 (1500 + 1500) × 20 (1500 + 1500) × 20 (1500 + 1500) × 20 (1500 + 1500) × 20 (1500 + 1500) × 8 (1500 + 1500) × 8
24.7–28.5 27.0–28.2 29.8–32.4 32.5–33.2
218.7–235.9 225.5–230. 1 235.7–251.7 264.5–270.3
25.7 26.5–27.3 30.8–30.9
248.7 195.3–208.1 211.7–222.6
Table 3 Structural parameters and mechanical properties of grown films. Sample
Grain size (nm)
ZrC ZrN TiN ZCN1 ZCN2 ZCN3 ZCN4 ZCN5 ZCTN2
Strain % (W–H)
W–H
Scherrer
4.9–7.5 16.7–14.2 90.9–33.3 5.7 13.0 19.3–20.0 21.7–17.9 20.1–20.5 8.8 15.9
7.7 6.5 7.9 7.4 6.4 6.9 6.1 7.4 5.5
0.5–0.6 2.1–1.9 3.1–2.6 4.2 1.6 2.6–2.3 2.8–2.9 2.4–2.2 3.7
5.8
was rather small and further research is needed to understand if the smaller grain size and change of texture played a role in its increase. As a general observation, it is worth mentioning that all films exhibited very high hardness values, amongst the best yet reported [14,16–18]. To test the lateral uniformity of the deposited multilayers structures we performed nanoindentation measurements on a 1-cm line across the surface of the samples ZCTN2 and
1.0
diffraction patterns recorded using the symmetrical geometry one could notice that the multilayers grown under CH4 atmosphere exhibited a (1 1 1) texture, while the use of an Ar atmosphere resulted in more equiaxed films, with a lower grain size and level of strain. This is in contrast with pure ZrC films, which exhibited a strong (2 0 0) texture [18]. The change of the texture was probably caused by the deposition of the ZrN layer first. The deposited ZrC/TiN multilayers exhibited distinct diffraction lines corresponding to ZrC and TiN layers. The 2 positions of diffraction peaks acquired from ZCTN1 sample are shifted from the reference positions [33] towards those of ZrCN and TiCN compounds [23], indicative of a certain intermixing during the deposition. However, films deposited under Ar atmosphere or with thicker bilayer periods exhibited more separated and narrower diffraction peaks. It is also apparent, that the TiN diffraction peaks are narrower than the ZrC peaks for the same sample, indicating a larger grain size within the TiN layer (see Table 3). Nanoindentation results, displayed in Table 2, showed that the ZCTN2 and ZCN5 multilayered films exhibited nanohardness values higher than those recorded from pure films. However, the increase
Fig. 5. Cross-section TEM image of sample ZCTN2.
ZCN5
Max: 13463 Zr3
dN(E)
Concentration (a.u.)
Min: -12779
Zr1 66.1 %
50
105
160
270
325
C1s O1s
O1 4.1 %
C1 14.5 %
215
N1s
Ti1 15.3 % 380 435
Ti2p
0 490
545
600
20
40
60
80
100
120
Sputtering time (min)
Kinetic Energy (eV) Fig. 4. AES survey scan acquired from the bulk region of the ZCN1 multilayer sample.
Fig. 6. XPS depth profile of elemental composition of ZCN5 multilayer sample. Ti is present as an impurity, more so in the ZrC target than ZrN one.
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
5335
ZCN5
90 min
ZrC
20 min
10 min
Intensity (a.u.)
Intensity (a.u.)
50 min
30 min
10 min
ar ar
187 186 185 184 183 182 181 180 179 178 177
187 186 185 184 183 182 181 180 179 178 177
Binding energy (eV)
Binding energy (eV)
Fig. 7. XPS high resolution scans of Zr 3d region acquired after various sputtering times for samples ZrC and ZNC5.
ZCN2. The results showed that the structures were laterally uniform, exhibiting just small deviations in the values of the hardness. The ZCTN2 sample exhibited higher hardness values than the ZCN2 sample. The dimension of the crystalline grains and the strain present, obtained from the Scherrer equation and, when possible, from a Williamson–Hall plot for the (1 1 1) family of planes or all diffraction lines, are displayed in Table 3. Firstly, the acquired diffraction patterns were fitted using the Profit software from Panalytical and then the full width at half maximum (FWHM) values corresponding to CuK␣1 radiation were corrected for the instrumental broadening. It appears that the Williamson–Hall method is not suitable to calculate the grain size for ZCN samples because of the convolution between the ZrC and ZrN diffraction lines. Fig. 4 displays an AES survey spectrum corresponding to the bulk region of the ZCN1 multilayer sample. A minimum of oxygen concentration of 4.1% was measured, a little bit higher than the best value of 3.0% recorded for a pure ZrC film [18]. It appears that some oxidation occurred during deposition process because of the longer time required to grow the 20 multilayer structure as compared with the deposition of a pure ZrC film. An elemental AES depth profile for the same sample showed that oscillations of the C and N signals were visible only for the first 3 deposited layers; after that, the composition variations were no longer visible, the sample exhibiting either a mixed composition, or the layers thickness was too small to be resolved [26]. TEM investigations, as those shown in Fig. 5, revealed that the layers thickness decreased for longer deposition time, due to the formation of a track on the target surface and the coating of the chamber entrance window. One can also note that the grain size is around 10 nm in the first two layers and then decreases significantly, more so for the ZrC layers. This change also explains the rather large variation in the grain size values estimated from XRD patterns and the difficulty to model the acquired XRR curves. Improvements in the residual vacuum and an increase of the bilayer thickness resulted in multilayers samples with better separated layers. When such a sample was analyzed by XPS depth profiling, it was possible to distinguish the variation of the composition of even the first layers near the surface, as shown in Fig. 6. Taking into account that we measured a sputtering rate of 0.15 nm/min, one can infer the presence of an oxidized layer on the surface of around 2–3 nm, corroborating the XRR simulation results. Moreover, when comparing the high resolution scans of the Zr3d region acquired from the pure ZrC film and the ZCN5 sample it is evident that the surface oxide layer is now thinner for the multilayer sample, since the peak corresponding to Zr atoms bonded to C (located at around 179 eV) is clearly visible even in the as-received XPS spectrum, as one can see in Fig. 7. The as-received scan acquired from a ZrC film showed only the presence of Zr atoms bound to O (or
OH) atoms. The XPS depth profile investigation also revealed lower oxygen content in the bulk of the ZCTN2 sample than the values recorded for the pure ZrC film, due to the lower residual vacuum level achieved when growing this sample. 4. Conclusions The properties of ZrC/ZrN and ZrC/TiN multilayer structures were investigated. The use of an Ar atmosphere during deposition resulted in a ZrC/TiN sample with more equiaxed grains and less texture that exhibited hardness values up to 33 GPa, higher than the values measured for pure TiN, ZrC or ZrN thin films. For films deposited under CH4 atmosphere the best results were obtained when using a low pressure of 2 × 10−3 Pa. The oxygen concentration is rather high and further improvements should concentrate on ways to reducing it. It also appears that larger thicknesses of the bilayer could have a beneficial effect upon hardness properties, but more investigations are required to clarify this aspect. Acknowledgements We would like to acknowledge the Major Analytical Instrumentation Center for help with samples characterization. This work was funded by CNCSIS Ideas Project Code 1408. References [1] L.E. Toth, Transition Metal Carbides and Nitrides, Academic, New York, 1971. [2] E.K. Storms, The Refractory Carbides, Academic, New York, 1967. [3] W. Lengauer, Transition metal carbides, nitrides and carbonitrides, in: R. Riedel (Ed.), Handbook of Ceramic Hard Materials, vol. 1, Wiley-VCH, Weinheim, 2000. [4] V. Craciun, D. Craciun, J.M. Howard, J. Woo, Thin Solid Films 515 (2007) 4636–4639. [5] F.M. Charbonnier, W.A. Mackie, R.L. Hartman, T. Xie, J. Vac. Sci. Technol. B19 (2001) 1064. [6] G. Vasudevamurthy, T.W. Knight, E. Roberts, T.M. Adams, J. Nucl. Mater. 374 (2008) 241–247. [7] B.V. Cockeram, J.L. Hollenbeck, Surf. Coat. Technol. 157 (2002) 274–281. [8] Y.S. Won, Y.S. Kim, V.G. Varanasi, O. Kryliouk, T.J. Anderson, C.T. Sirimanne, L. McElwee-White, J. Cryst. Growth 304 (2007) 324–332. [9] T. Noguchi, T. Shimada, A. Hanzawa, T. Hasegawa, Thin Solid Films 518 (2009) 778–780. [10] M. Braic, M. Balaceanu, A. Vladescu, A. Kiss, V. Braic, A. Purice, G. Dinescu, N. Scarisoreanu, F. Stokker-Cheregi, A. Moldovan, R. Birjega, M. Dinescu, Surf. Coat. Technol. 200 (2006) 6505–6510. [11] J.E. Krzanowski, J. Wormwood, Surf. Coat. Technol. 201 (2006) 2942–2952. [12] H. Li, M. Ding, J. Feng, X. Li, G. Bai, F. Zhang, J. Vac. Sci. Technol. B24 (2006) 1436–1439. [13] V. Craciun, D. Craciun, I.W. Boyd, Mater. Sci. Eng. B 18 (1993) 178–180. [14] P.R. Willmott, H. Spillmann, Appl. Surf. Sci. 197/198 (2002) 432–437. [15] L. D’Alessio, A. Santagata, R. Teghil, M. Zaccagnino, I. Zacardo, V. Marotta, D. Ferro, G. DeMaria, Appl. Surf. Sci. 168 (2000) 284. [16] M. Morstein, P.R. Willmott, H. Spillmann, M. Dobeli, Appl. Phys. A 75 (2002) 647–654.
5336
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
[17] J. Woo, G. Bourne, V. Craciun, D. Craciun, R.K. Singh, J. Optoelectron. Adv. Mater. 8 (2006) 20. [18] D. Craciun, G. Socol, N. Stefan, I.N. Mihailescu, G. Bourne, V. Craciun, Surf. Coat. Technol. 203 (2009) 1055–1058. [19] W.D. Sproul, Science 273 (1996) 889. [20] D.J. Li, J. Yang, X.H. Zhang, M. Cao, J. Vac. Sci. Technol. B 25 (2007) L11–L13. [21] L.H. Yu, S.R. Dong, J.H. Xu, I. Kojima, Thin Solid Films 516 (2008) 1864–1870. [22] G. Abadias, S. Dub, R. Shmegera, Surf. Coat. Technol. 200 (2006) 6538–6543. [23] M.M. Larijani, M.B. Zanjanbar, A. Majdabadi, J. Alloys Compd. 492 (2010) 735–738. [24] L. Milosev, H.-H. Strehblow, B. Navisec, Thin Solid Films 303 (1997) 246–252.
[25] H.C. Barshilia, N. Selvakumar, K.S. Rajam, K. Gopinadhan, S. Chaudhary, J. Phys. D: Appl. Phys. 41 (2008) 205409. [26] D. Craciun, G. Socol, G. Dorcioman, N. Stefan, G. Bourne, V. Craciun, J. Optoelectron. Adv. Mater. 12 (2010) 461. [27] A.D. Krawitz, Introduction to Diffraction in Materials Science and Engineering, Willey, 2001. [28] G.K. Williamson, W.H. Hall, Acta Metall. 1l (1953) 22–31. [29] W.C. Oliver, G.M. Pharr, J. Mater. Res. 47 (1992) 1564. [30] Y.R. Jeng, C.M. Tan, Appl. Phys. Lett. 89 (2006) 251901. [31] L.G. Parratt, Phys. Rev. 95 (1954) 359. [32] H. Kiessig, Ann. Phys. 10 (1931) 769. [33] JCPDS-International Center for Diffraction Data Copyright© JCPDS-ICDD 1996, Card 35-0784, 35-0753, and Card 32-1489.r.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Characteristics of ZrC/ZrN and ZrC/TiN multilayers grown by pulsed laser deposition D. Craciun a , G. Bourne b , G. Socol a , N. Stefan a , G. Dorcioman a , E. Lambers b , V. Craciun a,b,∗ a b
Laser Department, National Institute for Laser, Plasma, and Radiation Physics, Bucharest, Romania Major Analytical Instrumentation Center, Materials Science and Engineering, University of Florida, Gainesville, FL 32611, USA
a r t i c l e
i n f o
Article history: Received 27 May 2010 Received in revised form 16 November 2010 Accepted 16 November 2010 Available online 24 November 2010 Keywords: ZrC ZrN TiN Pulsed laser deposition Thin films Multilayers
a b s t r a c t ZrC/ZrN and ZrC/TiN multilayers were grown on (1 0 0) Si substrates at 300 ◦ C by the pulsed laser deposition (PLD) technique using a KrF excimer laser. X-ray diffraction investigations showed that films were crystalline, the strain and grain size depending on the nature and pressure of the gas used during deposition. The elemental composition, analyzed by Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS), showed that films contained a low level of oxygen contamination. Simulations of the X-ray reflectivity (XRR) curves acquired from films indicated a smooth surface morphology, with roughness below 1 nm (rms) and densities very close to bulk values. Nanoindentation results showed that the ZrC/ZrN and ZrC/TiN multilayer samples exhibited hardness values between 30 and 33 GPa, slightly higher than the values of 28–30 GPa measured for pure ZrC, TiN and ZrN films. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The excellent mechanical properties and thermochemical stability of refractory metal carbides and nitrides recommend them for important applications as hard and protective coatings [1–3]. It is generally rather difficult to deposit high quality ZrC films because of their high melting temperature, low sputtering rate, and high reactivity with oxygen or water vapors [4]. Driven by stringent requirements of special applications, such as coatings for field emission tips, nuclear fuel or outer space thermal radiators [5–7], significant progress has been made in the last decade and good quality ZrC films were obtained by using either physical vapor deposition (PVD) or chemical vapor deposition (CVD) based techniques [8–12]. As one of the most versatile research technique, pulsed laser deposition (PLD) allows for relatively low to moderate deposition temperatures of refractory carbides and nitrides without sacrificing their crystalline quality [13–15]. Some of the best mechanical properties of ZrC yet reported were measured on films deposited using the PLD technique [16,17]. Furthermore, it has been showed that by using a higher repetition rate laser for ablation, the substrate temperature could be reduced to only 300 ◦ C while
∗ Corresponding author at: Laser Department, National Institute for Laser, Plasma, and Radiation Physics, Bucharest, Romania. E-mail address: [email protected]fl.edu (V. Craciun). 0169-4332/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2010.11.106
the growth rate is increased and the crystalline quality maintained [18]. The deposition of multilayers with different crystalline lattices that could block at their interfaces the propagation of dislocations from one material to another was shown to be a way of increasing the hardness of such thin film structure [19–22]. Since it is known that generally carbides are more brittle than nitrides and that the refractory metal nitrides exhibit very similar properties to carbides, being also used as hard and protective coatings [1,3,16,23–25], we have investigated ways to further improve the quality of films by depositing multilayers of ZrC and a refractory metal nitride such as ZrN or TiN [26] and present new results here. 2. Experiment The depositions were performed in a PLD system using a KrF excimer laser ( = 248 nm, pulse duration = 25 ns, 8.0 J/cm2 fluence, 40 Hz repetition rate) that has been already described in detail elsewhere [18,26]. The films were deposited for tens of minutes from polycrystalline ZrC, ZrN, and TiN targets (Plasmaterials, Inc.) on p++ (1 0 0) Si substrates (MEMC Electronic Materials, Inc.) that were cleaned in acetone, then ethanol, rinsed in deionized water, and finally blown dry with high purity nitrogen before being loaded into the deposition chamber. The nominal substrate temperature was set at 300 ◦ C. Depositions were performed under a high purity atmosphere of CH4 or Ar (2 × 10−3 to 10−2 Pa). After
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
Fig. 1. XRR curves recorded from pure ZrC, TiN, and ZrN deposited films.
deposition, the films were slowly cooled to room temperature under the same atmosphere as that using the deposition, but at a much higher pressure of around 104 Pa. The films mass density, thickness, and surface roughness were investigated by X-ray reflectometry (XRR, Panalytical X’Pert MRD) using Cu K˛ radiation. The same instrument was used for structural characterization in symmetric and grazing incidence X-ray diffraction (XRD and GIXD) using a parallel beam configuration. The grain size was evaluated following the Scherrer equation [27] and, where possible, by the model of Williamson and Hall [28]. The chemical composition of the films was investigated by Auger electron spectroscopy (AES) in a Perkin-Elmer PHI 660 system (5 kV, 30◦ take off angle) and by X-ray photoelectron spectroscopy (XPS) in a Perkin-Elmer PHI 5100 ESCA system (300 W, Mg K˛ ). To obtain elemental depth profiles, measurements were collected after various time cycles of Ar ion sputtering (4 kV, 1–3 A/cm2 ; for XPS measurements the Ar ion beam was rastered over a larger area of 10 × 7 mm2 ). Cross section samples were prepared on a focused ion beam (Strata DB 235) for transmission electron microscopy (TEM) analysis on a JEOL 2010F instrument. The mechanical properties of the films were measured with a nanoindentation device (Hysitron inc.) equipped with a cube-corner diamond tip and set to run 100 indents per sample. The nanoindentation experiments were performed in displacement control with a contact depth of up to 30 nm. Hardness and reduced modulus were determined from the loaddisplacement data following the model of Oliver and Pharr [29], after excluding the results obtained for penetration depths smaller than 5 nm [30]. 3. Results and discussion XRR curves recorded from deposited pure ZrC, TiN and ZrN films are presented in Fig. 1. One could note different values for the critical angle of each material, a consequence of different density values. From simulations of the acquired curves using the Panalytical WinGixaTM software, which is based on Parrat’s formalism [31], we obtained values for the films density and surface roughness that are shown in Table 1, together with the deposition conditions and hardness values. The Kiessig fringes [32] were only present in the XRR curve of ZrC sample, which was much thinner than the other
5333
Fig. 2. XRD patterns (G, grazing incidence; T, symmetrical geometry) collected from ZrC/ZrN multilayers; the vertical bars are the reference diffraction peak positions of reference ZrC (full line, Card 35-0784) and ZrN (dotted line, Card 35-0753) powders [33].
Fig. 3. XRD patterns (G, grazing incidence; T, symmetrical geometry) collected from the ZrC/TiN multilayers; the vertical bars are the reference diffraction peak positions of ZrC (full line) and TiN (dotted line, Card 32-1489) powders [33].
two films. According to the simulations results, the films oxidized when exposed to the ambient, the oxide layers being estimated to be around 2–3 nm thick. The XRR curves acquired from multilayers showed evidence of interference peaks due to the presence of many interfaces, but we could not fit them satisfactorily with a model, most likely due to variations of thickness from layer to layer. However, the critical angle values of the top ZrC layer were consistent with densities very similar to those estimated for pure ZrC films. XRD and GIXD investigations showed that the deposited films were crystalline, as one can see in Figs. 2 and 3, where diffraction patterns from several representative samples acquired in symmetrical and grazing incidence geometry are displayed. The deposition conditions employed for each multilayer structure are displayed in Table 2. To facilitate a comparison, we also introduced in Table 2 and Figs. 2 and 3 some results from our previous study [26]. Since the lattice parameter of ZrC and ZrN are very similar, the recorded Bragg diffraction lines from ZrC/ZrN multilayers appeared as a convolution of the two lines, without being resolved. Analyzing the
Table 1 Deposition parameters and mechanical properties of grown pure films. Sample ZrC ZrN TiN
Deposition atmosphere (Pa) −3
2 × 10 CH4 2 × 10−3 CH4 2 × 10−3 CH4
Density (g/cm3 )
Roughness (nm)
Hardness (GPa)
Reduced Modulus (GPa)
6.45 7.09 5.21
0.5 0.6 0.5
30.1 28.5 29.9
226.0–228.3 214.5 242.3
5334
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
Table 2 Deposition parameters and mechanical properties of grown multilayers. Sample
Deposition atmosphere (Pa) −3
2 × 10 2 × 10−3 1 × 10−2 5 × 10−3 5 × 10−3 5 × 10−3 7 × 10−3 2 × 10−3
ZCN1 ZCTN1 ZCN2 ZCTN2 ZCN3 ZCTN3 ZCN4 ZCN5
CH4 CH4 Ar Ar CH4 CH4 CH4 CH4
Number of pulses and multilayers
Hardness (GPa)
Reduced modulus (GPa)
(720 + 720) × 20 (720 + 720) × 20 (1500 + 1500) × 20 (1500 + 1500) × 20 (1500 + 1500) × 20 (1500 + 1500) × 20 (1500 + 1500) × 8 (1500 + 1500) × 8
24.7–28.5 27.0–28.2 29.8–32.4 32.5–33.2
218.7–235.9 225.5–230. 1 235.7–251.7 264.5–270.3
25.7 26.5–27.3 30.8–30.9
248.7 195.3–208.1 211.7–222.6
Table 3 Structural parameters and mechanical properties of grown films. Sample
Grain size (nm)
ZrC ZrN TiN ZCN1 ZCN2 ZCN3 ZCN4 ZCN5 ZCTN2
Strain % (W–H)
W–H
Scherrer
4.9–7.5 16.7–14.2 90.9–33.3 5.7 13.0 19.3–20.0 21.7–17.9 20.1–20.5 8.8 15.9
7.7 6.5 7.9 7.4 6.4 6.9 6.1 7.4 5.5
0.5–0.6 2.1–1.9 3.1–2.6 4.2 1.6 2.6–2.3 2.8–2.9 2.4–2.2 3.7
5.8
was rather small and further research is needed to understand if the smaller grain size and change of texture played a role in its increase. As a general observation, it is worth mentioning that all films exhibited very high hardness values, amongst the best yet reported [14,16–18]. To test the lateral uniformity of the deposited multilayers structures we performed nanoindentation measurements on a 1-cm line across the surface of the samples ZCTN2 and
1.0
diffraction patterns recorded using the symmetrical geometry one could notice that the multilayers grown under CH4 atmosphere exhibited a (1 1 1) texture, while the use of an Ar atmosphere resulted in more equiaxed films, with a lower grain size and level of strain. This is in contrast with pure ZrC films, which exhibited a strong (2 0 0) texture [18]. The change of the texture was probably caused by the deposition of the ZrN layer first. The deposited ZrC/TiN multilayers exhibited distinct diffraction lines corresponding to ZrC and TiN layers. The 2 positions of diffraction peaks acquired from ZCTN1 sample are shifted from the reference positions [33] towards those of ZrCN and TiCN compounds [23], indicative of a certain intermixing during the deposition. However, films deposited under Ar atmosphere or with thicker bilayer periods exhibited more separated and narrower diffraction peaks. It is also apparent, that the TiN diffraction peaks are narrower than the ZrC peaks for the same sample, indicating a larger grain size within the TiN layer (see Table 3). Nanoindentation results, displayed in Table 2, showed that the ZCTN2 and ZCN5 multilayered films exhibited nanohardness values higher than those recorded from pure films. However, the increase
Fig. 5. Cross-section TEM image of sample ZCTN2.
ZCN5
Max: 13463 Zr3
dN(E)
Concentration (a.u.)
Min: -12779
Zr1 66.1 %
50
105
160
270
325
C1s O1s
O1 4.1 %
C1 14.5 %
215
N1s
Ti1 15.3 % 380 435
Ti2p
0 490
545
600
20
40
60
80
100
120
Sputtering time (min)
Kinetic Energy (eV) Fig. 4. AES survey scan acquired from the bulk region of the ZCN1 multilayer sample.
Fig. 6. XPS depth profile of elemental composition of ZCN5 multilayer sample. Ti is present as an impurity, more so in the ZrC target than ZrN one.
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
5335
ZCN5
90 min
ZrC
20 min
10 min
Intensity (a.u.)
Intensity (a.u.)
50 min
30 min
10 min
ar ar
187 186 185 184 183 182 181 180 179 178 177
187 186 185 184 183 182 181 180 179 178 177
Binding energy (eV)
Binding energy (eV)
Fig. 7. XPS high resolution scans of Zr 3d region acquired after various sputtering times for samples ZrC and ZNC5.
ZCN2. The results showed that the structures were laterally uniform, exhibiting just small deviations in the values of the hardness. The ZCTN2 sample exhibited higher hardness values than the ZCN2 sample. The dimension of the crystalline grains and the strain present, obtained from the Scherrer equation and, when possible, from a Williamson–Hall plot for the (1 1 1) family of planes or all diffraction lines, are displayed in Table 3. Firstly, the acquired diffraction patterns were fitted using the Profit software from Panalytical and then the full width at half maximum (FWHM) values corresponding to CuK␣1 radiation were corrected for the instrumental broadening. It appears that the Williamson–Hall method is not suitable to calculate the grain size for ZCN samples because of the convolution between the ZrC and ZrN diffraction lines. Fig. 4 displays an AES survey spectrum corresponding to the bulk region of the ZCN1 multilayer sample. A minimum of oxygen concentration of 4.1% was measured, a little bit higher than the best value of 3.0% recorded for a pure ZrC film [18]. It appears that some oxidation occurred during deposition process because of the longer time required to grow the 20 multilayer structure as compared with the deposition of a pure ZrC film. An elemental AES depth profile for the same sample showed that oscillations of the C and N signals were visible only for the first 3 deposited layers; after that, the composition variations were no longer visible, the sample exhibiting either a mixed composition, or the layers thickness was too small to be resolved [26]. TEM investigations, as those shown in Fig. 5, revealed that the layers thickness decreased for longer deposition time, due to the formation of a track on the target surface and the coating of the chamber entrance window. One can also note that the grain size is around 10 nm in the first two layers and then decreases significantly, more so for the ZrC layers. This change also explains the rather large variation in the grain size values estimated from XRD patterns and the difficulty to model the acquired XRR curves. Improvements in the residual vacuum and an increase of the bilayer thickness resulted in multilayers samples with better separated layers. When such a sample was analyzed by XPS depth profiling, it was possible to distinguish the variation of the composition of even the first layers near the surface, as shown in Fig. 6. Taking into account that we measured a sputtering rate of 0.15 nm/min, one can infer the presence of an oxidized layer on the surface of around 2–3 nm, corroborating the XRR simulation results. Moreover, when comparing the high resolution scans of the Zr3d region acquired from the pure ZrC film and the ZCN5 sample it is evident that the surface oxide layer is now thinner for the multilayer sample, since the peak corresponding to Zr atoms bonded to C (located at around 179 eV) is clearly visible even in the as-received XPS spectrum, as one can see in Fig. 7. The as-received scan acquired from a ZrC film showed only the presence of Zr atoms bound to O (or
OH) atoms. The XPS depth profile investigation also revealed lower oxygen content in the bulk of the ZCTN2 sample than the values recorded for the pure ZrC film, due to the lower residual vacuum level achieved when growing this sample. 4. Conclusions The properties of ZrC/ZrN and ZrC/TiN multilayer structures were investigated. The use of an Ar atmosphere during deposition resulted in a ZrC/TiN sample with more equiaxed grains and less texture that exhibited hardness values up to 33 GPa, higher than the values measured for pure TiN, ZrC or ZrN thin films. For films deposited under CH4 atmosphere the best results were obtained when using a low pressure of 2 × 10−3 Pa. The oxygen concentration is rather high and further improvements should concentrate on ways to reducing it. It also appears that larger thicknesses of the bilayer could have a beneficial effect upon hardness properties, but more investigations are required to clarify this aspect. Acknowledgements We would like to acknowledge the Major Analytical Instrumentation Center for help with samples characterization. This work was funded by CNCSIS Ideas Project Code 1408. References [1] L.E. Toth, Transition Metal Carbides and Nitrides, Academic, New York, 1971. [2] E.K. Storms, The Refractory Carbides, Academic, New York, 1967. [3] W. Lengauer, Transition metal carbides, nitrides and carbonitrides, in: R. Riedel (Ed.), Handbook of Ceramic Hard Materials, vol. 1, Wiley-VCH, Weinheim, 2000. [4] V. Craciun, D. Craciun, J.M. Howard, J. Woo, Thin Solid Films 515 (2007) 4636–4639. [5] F.M. Charbonnier, W.A. Mackie, R.L. Hartman, T. Xie, J. Vac. Sci. Technol. B19 (2001) 1064. [6] G. Vasudevamurthy, T.W. Knight, E. Roberts, T.M. Adams, J. Nucl. Mater. 374 (2008) 241–247. [7] B.V. Cockeram, J.L. Hollenbeck, Surf. Coat. Technol. 157 (2002) 274–281. [8] Y.S. Won, Y.S. Kim, V.G. Varanasi, O. Kryliouk, T.J. Anderson, C.T. Sirimanne, L. McElwee-White, J. Cryst. Growth 304 (2007) 324–332. [9] T. Noguchi, T. Shimada, A. Hanzawa, T. Hasegawa, Thin Solid Films 518 (2009) 778–780. [10] M. Braic, M. Balaceanu, A. Vladescu, A. Kiss, V. Braic, A. Purice, G. Dinescu, N. Scarisoreanu, F. Stokker-Cheregi, A. Moldovan, R. Birjega, M. Dinescu, Surf. Coat. Technol. 200 (2006) 6505–6510. [11] J.E. Krzanowski, J. Wormwood, Surf. Coat. Technol. 201 (2006) 2942–2952. [12] H. Li, M. Ding, J. Feng, X. Li, G. Bai, F. Zhang, J. Vac. Sci. Technol. B24 (2006) 1436–1439. [13] V. Craciun, D. Craciun, I.W. Boyd, Mater. Sci. Eng. B 18 (1993) 178–180. [14] P.R. Willmott, H. Spillmann, Appl. Surf. Sci. 197/198 (2002) 432–437. [15] L. D’Alessio, A. Santagata, R. Teghil, M. Zaccagnino, I. Zacardo, V. Marotta, D. Ferro, G. DeMaria, Appl. Surf. Sci. 168 (2000) 284. [16] M. Morstein, P.R. Willmott, H. Spillmann, M. Dobeli, Appl. Phys. A 75 (2002) 647–654.
5336
D. Craciun et al. / Applied Surface Science 257 (2011) 5332–5336
[17] J. Woo, G. Bourne, V. Craciun, D. Craciun, R.K. Singh, J. Optoelectron. Adv. Mater. 8 (2006) 20. [18] D. Craciun, G. Socol, N. Stefan, I.N. Mihailescu, G. Bourne, V. Craciun, Surf. Coat. Technol. 203 (2009) 1055–1058. [19] W.D. Sproul, Science 273 (1996) 889. [20] D.J. Li, J. Yang, X.H. Zhang, M. Cao, J. Vac. Sci. Technol. B 25 (2007) L11–L13. [21] L.H. Yu, S.R. Dong, J.H. Xu, I. Kojima, Thin Solid Films 516 (2008) 1864–1870. [22] G. Abadias, S. Dub, R. Shmegera, Surf. Coat. Technol. 200 (2006) 6538–6543. [23] M.M. Larijani, M.B. Zanjanbar, A. Majdabadi, J. Alloys Compd. 492 (2010) 735–738. [24] L. Milosev, H.-H. Strehblow, B. Navisec, Thin Solid Films 303 (1997) 246–252.
[25] H.C. Barshilia, N. Selvakumar, K.S. Rajam, K. Gopinadhan, S. Chaudhary, J. Phys. D: Appl. Phys. 41 (2008) 205409. [26] D. Craciun, G. Socol, G. Dorcioman, N. Stefan, G. Bourne, V. Craciun, J. Optoelectron. Adv. Mater. 12 (2010) 461. [27] A.D. Krawitz, Introduction to Diffraction in Materials Science and Engineering, Willey, 2001. [28] G.K. Williamson, W.H. Hall, Acta Metall. 1l (1953) 22–31. [29] W.C. Oliver, G.M. Pharr, J. Mater. Res. 47 (1992) 1564. [30] Y.R. Jeng, C.M. Tan, Appl. Phys. Lett. 89 (2006) 251901. [31] L.G. Parratt, Phys. Rev. 95 (1954) 359. [32] H. Kiessig, Ann. Phys. 10 (1931) 769. [33] JCPDS-International Center for Diffraction Data Copyright© JCPDS-ICDD 1996, Card 35-0784, 35-0753, and Card 32-1489.r.