- Email: [email protected]
Atomic force microscopy indentation of fluorocarbon thin films fabricated by plasma enhanced chemical deposition at low radio frequency power
Thin Solid Films 517 (2009) 3310–3314
Contents lists available at ScienceDirect
Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f
Atomic force microscopy indentation of fluorocarbon thin films fabricated by plasma enhanced chemical deposition at low radio frequency power L. Sirghi ⁎, A. Ruiz, P. Colpo, F. Rossi ⁎ European Commission, Institute for Health and Consumer Protection, TP-203, Via E. Fermi, 21020 Ispra (VA), Italy
a r t i c l e
i n f o
Article history: Received 17 April 2008 Received in revised form 16 January 2009 Accepted 16 January 2009 Available online 24 January 2009 PACS codes: 68.37.Ps 68.03.Cd 62.20.de 62.20.F62.23.St
a b s t r a c t Atomic force microscopy (AFM) indentation technique is used for characterization of mechanical properties of fluorocarbon (CFx) thin films obtained from C4F8 gas by plasma enhanced chemical vapour deposition at low r.f. power (5–30 W) and d.c. bias potential (10–80 V). This particular deposition method renders films with good hydrophobic property and high plastic compliance. Commercially available AFM probes with stiff cantilevers (10– 20 N/m) and silicon sharpened tips (tip radius b10 nm) are used for indentations and imaging of the resulted indentation imprints. Force depth curves and imprint characteristics are used for determination of film hardness, elasticity modulus and plasticity index. The measurements show that the decrease of the discharge power results in deposition of films with decreased hardness and stiffness and increased plasticity index. Nanolithography based on AFM indentation is demonstrated on thin films (thickness of 40 nm) with good plastic compliance. © 2009 Elsevier B.V. All rights reserved.
Keywords: Fluorocarbon Nanoindentation Elastic properties Plasticity index
1. Introduction Apart of being an instrument for imaging surfaces with high resolution, the atomic force microscope is also an important tool for assessment of mechanical properties of materials at nanoscale. Thus, the atomic force microscopy (AFM) indentation is widely used for determination of elasticity modulus and hardness of thin films and microscopic objects [1–3]. Indentation is also one of the most used AFMbased lithography techniques [4]. This technique has the advantage that the AFM imaging capability can be used for in situ characterization of the fabricated patterns or nanostructures. Indentation of polymer films is of great interest for nanolithography because of good plastic compliance of polymers [1] and their intensive use in micro and nano patterning. AFM indentations can be used to trigger solvent-assisted ejection of polymer at the indentation imprint in order to build array of pits [5]. Indentation of polymers with heated AFM tips can be an efficient technique to build pits on shape memory polymers [6]. The present paper reports the results of AFM indentation of fluorocarbon (CFx) thin films obtained by plasma enhanced chemical vapour deposition (PECVD) using a low-power r.f. discharge in C4F8 gas. Polymeric CFx films are known to rend a surface with very low energy and relatively good hardness and wearing resistance. These properties make these films ⁎ Corresponding authors. E-mail addresses: [email protected] (L. Sirghi), [email protected] (F. Rossi). 0040-6090/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2009.01.055
interesting for coatings that reduce adhesion of water, oily contaminants, bacteria, etc. Such coatings are desirable for magnetic recording media [7], electric insulators for electric power lines [8], and biomedicine applications [9,10]. Fluorocarbon films with good hydrophobic and hardness properties are usually synthesised by PECVD that employs discharge plasmas to dissociate precursor gases [11,12]. In order to obtain a good compromise between the good hydrophobic property of pure fluorocarbon films and the good hardness and wear resistance of pure hydrocarbon films (which are not hydrophobic) fluorinated carbon films are synthesised by PECVD in mixed hydrocarbon and fluorocarbon gases [11]. For these films, it was found that their hydrophobicity improves by the increase of the fluor content, while the hardness and wearing resistance decrease [12]. An alternative method to synthesise fluorocarbon film consists of sputtering deposition in an r.f. magnetron discharge using polytetrafluoroethilene (PTFE) target [13]. Systematic studies of dependence of hydrophobic property and mechanical properties of these fluorocarbon films on different deposition parameters (discharge power, mixture gas content, substrate biasing potential, gas pressure) are reported in [14]. Our study regards the fluorinated carbon films obtained by PECVD in pure fluorocarbon gas at low r.f. power and d.c. biasing potential, films that show especially good hydrophobicity and plasticity compliance. Details on properties and deposition method of these films are given in a previous published paper [16]. It was found by different authors [15,16] that the CFx films obtained by PECVD in pure fluorocarbon gases at low discharge power have a higher density of
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314 Table 1 Deposition parameters and surface properties of CFx films P (W)
Vsb (V)
t (nm/min)
θw (deg.)
Roughness (nm)
E (GPa)
Hn (GPa)
ψ
5 10 20 30
−10 −20 −60 −80
7 20 36 45
106 ± 1 106 ± 1 105 ± 1 106 ± 1
0.36 ± 0.05 0.52 ± 0.07 0.58 ± 0.07 0.59 ± 0.08
1.75 ± 0.3 2.1 ± 0.6 2.4 ± 0.6 3.2 ± 0.7
0.31 ± 0.03 0.52 ± 0.09 0.88 ± 0.19 1.57 ± 0.46
0.84 ± 0.05 0.77 ± 0.04 0.6 ± 0.04 0.55 ± 0.5
P is the power of r. f. discharge in C4F8 gas; Vsb, the self-biasing potential of the deposition substrate; t, the deposition rate; θw, the water contact angle of the film surface; E, the Young modulus of elasticity; Hn, the film hardness at nanoscale; and ψ, the plasticity index.
CFx bonds and a better hydrophobic property. However, there is no study reporting the mechanical properties of these films, and the effect of lowering of discharge power and biasing potential on these properties has not been investigated. Tang et al. found that hardness and stiffness of CFx films obtained by r.f. magnetron sputtering of PTFE increase by the decrease of the discharge power [13]. This was explained by the decreased fluor content of the films deposited at lower discharge power values. The AFM indentation experiments presented by this paper show that the reduction of the discharge power in the PECVD of CFx films results in decrease of the film hardness and stiffness and increase in plastic compliance. Moreover, it is shown that the CFx films deposited at low discharge power, while having good hydrophobic property, are suitable to AFM indentation-based nanolithography because of their increased plasticity index. Nano patterns consisting in arrays of pits made by AFM indentations are demonstrated. The pits (indentation imprints) are imaged by the same AFM tip to determine their depth and diameter. Force depth-sensing curves acquired during the indentations are processed to determine the film elasticity. While these curves did not fit well a power law, the Oliver and Pharr method [17] of determination of film elasticity was not applicable to these indentation experiments. Therefore, the film stiffness at the beginning of the unloading part of the indentation along with contact radius measurements were used to determine the sample elasticity under the Sneddon approximation of a flat punch [18]. The film hardness is determined by the measurements of the maximum indentation force and the diameter of the resulted imprints.
3311
The indentation experiments were performed by the same AFM apparatus and probe that were used for tapping mode AFM characterization of the film surface. The AFM probe (NSG 11 from NT-MDT Co.) had a stiff cantilever (spring constant around 16 N/m) with a sharpened conical tip (silicon) with cone angle of 22° and apex curvature radius below 10 nm. The AFM tip was kept oscillating in tapping mode at a distance from the film surface equal to the oscillating amplitude (30 nm). Then, indentations were performed by precise forward and backward extension of the AFM scanner. The imprints resulted from the indentation experiments were immediately imaged by tapping-mode AFM. The spring constant of the cantilever was determined by thermal noise method [20], while the shape and condition of the AFM tip was determined by scanning electron microscopy (image not shown) and by imaging the ultra sharp edges of a silicon calibrating grating (TGG1 from NT-MDT). Indentation experiments required the use of AFM probes with stiff cantilevers (15–20 N/m) and very sharp tips (curvature radius under 10 nm).
3. Results and discussion 3.1. Imaging and characterization of indentation imprints To determine the elasticity modulus, E, and hardness, Hn, of the deposited CFx films, sets of 49 indentations were performed on a 7 × 7 arrays of points distributed on an area of 500×500 nm2 on the surfaces of thick films (thickness around 200 nm). Similar indentations were performed on thin films (thickness around 40 nm) for nanolithography purpose. Fig. 1 shows the topography image obtained for an array of imprints on the surface of a CFx film with the thickness of 200 nm deposited at r.f. discharge power of 10 W. Each of the 49 indentations was performed in 1 s time with a maximum loading force of about 300 nN. Similar indentation arrays were performed for indentation time varying between 0.2 s and 2 s. For this time interval, there were no noticeable differences in the imprint parameters. This means that patterns consisting of pit array can be performed on the surface of these films with relatively high speed.
2. Experimental setup and fluorocarbon thin film deposition The fluorocarbon films were deposited on silicon substrate in a plasma reactor that used an asymmetrical r.f. (13.56 MHz) capacitive discharge at low discharge power (5–30 W) in pure C4F8 gas flow at the pressure of 6.5 Pa. The deposition substrate was mounted on the active discharge electrode, which was self biased at a negative potential ranged, depending on the discharge power, between −10 V and −80 V. The film thickness was measured by a stylus profilometer (Alpha-Step IQ from KLA Tencor Co.). Films with two values of thickness were deposited, i.e. thick films (thickness around 200 nm) used for hardness and elasticity measurements, and thin films (thickness around 40 nm) used for AFM indentation-based nanolithography. The static values of water contact angle were measured with a contact angle goniometer (Digidrop from GBX Instruments Co.) on sessile drops of water on the deposited films. The surface topography of the films was characterized by taping mode AFM measurements performed with a commercial AFM apparatus (Solver PRO from NTMDT Co.) with close-loop scanner system. Results of these investigations in terms of deposition rate, water contact angle, and surface roughness for the films deposited at different discharge parameter values are given in Table 1. Results of the nanoscale hardness, elasticity modulus, and plasticity index, which were measured by the AFM indentation methods described below, are also given in the Table 1. More details on deposition system and structure of the deposited CFx films can be found in previously published papers [16,19].
Fig. 1. Topography image of an array of 7 × 7 indentation imprints on a CFx film (thickness of 200 nm) deposited at 10 W. The insets show high resolution (90 nm × 90 nm) topography image of a single imprint. The average depth of imprints is 7 nm, while the average value of imprint diameter is 26 nm. The array pitch is 70 nm. The height scale is 10 nm.
3312
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314
The inset on the topography image in Fig. 1 shows the high resolution topography image of one imprint. The topography image shows axis symmetrical imprints with the diameter of about 26 nm and depth of about 7 nm. Effect of material pile up around the imprints is visible. The axial symmetry of the imprints is illustrated by the roughly identical height profiles along two perpendicular axes (depicted as X and Y on the detail topography image) passing through the imprint centre (see Fig. 2). Because the deformation of the film surface was not entirely plastic, the imprint shape differs from that of the AFM tip. The theoretical height profile of the AFM tip surface is also depicted in Fig. 2. The height profile of the AFM tip is approximated by the parabolic dependence z=
r2 ; 2R
ð1Þ
where z, is the tip surface height, r is the lateral distance measured from the tip apex, and R, the tip curvature radius, which for this particular experiment was estimated to 7 nm by analysis of AFM images of sharp edges of a grating sample (TGG1 from NT-MDT). Since the imprints have a larger curvature radius (estimated to 12 nm) than that of the AFM tip, the values of the imprint depth and diameter can be determined directly on the as-acquired AFM images of the imprints, without being necessary to process the images for removing of the tip-sample convolution effects [21]. As illustrated in Fig. 2, the imprint diameter, dimp, and depth, himp, are determined by the height level of the film surface before indentations (which is approximated to the height level of the film surface far from the indentation place). Indentations the CFx films deposited at low r.f. power discharge can be used for a routine check on the condition of the tips of tapping AFM probes. Indentations performed with AFM probes having blunt tips result in imprints that, by comparison with the imprints performed by sharp tip, have much larger diameter and smaller depth. Also, the force–displacement curves acquired during the indentations performed with a blunt tip show, due to the large contact area, a much higher stiffness of the tip-sample contact. In some cases we noticed that the AFM images taken by the AFM tips after the CFx films indentation experiments were of a better quality than images taken with the same AFM tips before the indentations. We attribute this improvement in the AFM image quality to cleaning through indentations of the AFM tip apex. The CFx films have water and oily repellent surfaces, so that airborne contaminants on the AFM tip apex [22] may be removed during indentations of these films. Indentations performed on thin films (thickness of 40 nm) left imprints with slightly larger depth, when compared to the imprints depth performed on thick films (thickness of 200 nm). This is explained by the fact that the hard substrate limited the elastic deformations of the film, thus enhancing the plastic deformations. We
Fig. 2. Plots of the X and Y height profiles of the imprint shown in detail in Fig. 1(a) along with the plot of height profile of the AFM tip (curvature radius of 7 nm). The origin of height values is taken at level of the CFx film surface before indentation.
Fig. 3. Inverse topography image (a) and height profile (b) of a set of six indentations performed on CFx film (thickness of 40 nm) deposited at 10 W of the r. f. discharge power. The values indicated on the height profile plot are for Fmax.
performed sets of imprints with increasing loading force on thin CFx films (thickness of 40 nm) in order to determine the dependence of the imprint depth on the indentation force. Fig. 3 shows a 3D image (a) and height profile (b) of six imprints that resulted by indentation of the thin CFx film deposited at 10 W with increasing maximum loading force. The height profiles of the resulted imprints were used to determine himp and dimp for different values of Fmax. Fig. 4 shows the dependence of himp on Fmax for thin CFx films deposited at different discharge power values. While the imprint depth remained smaller than the film thickness, it increased linearly by the increase of Fmax. Therefore, linear fits of the experimental data are also shown on Fig. 4. For these measurements, the values of dimp were proportional to square root of Fmax, which is an indication of constant hardness. 3.2. Measurements of plasticity index, elasticity modulus, and hardness To avoid the effect of the stiff substrate [23], the elasticity modulus and hardness were measured based on analysis of force–displacement data acquired on AFM indentation experiments performed on thick CFx films (thickness around 200 nm). The force–displacement data were acquired for indentations performed on array of 49 points on
Fig. 4. The dependence of the imprint depth (himp) on the maximum value of the loading force (Fmax) for the thin CFx films (thickness of 40 nm) deposited at the r.f. discharge power values of 5, 10, 20, and 30 W, respectively.
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314
Fig. 5. Typical force–displacement curves corresponding to indentations of the CFx films deposited at r.f. discharge power values of 10 W, 20 W and 30 W, respectively. The force– displacement curve corresponding to indentation of the film substrate (silicon) is also shown. The origin, O, of the AFM tip displacement is corresponding to the non deformed film surface. The arrows indicate the position of the film surface after the indentations.
surfaces of CFx films deposited at r.f. discharge power values of 5, 10, 20 and 30 W, respectively. Fig. 5 illustrates a comparison between force–displacement curves obtained in indentation experiments on the film substrate and on CFx films deposited at different r.f. discharge power values. The force versus indenter displacement curves are determined on the bases of the force versus scanner extension curves acquired during the indentations. The indenter displacement, h, was determined by subtracting the cantilever deflection, δ, from the scanner extension, z. h = z−δ
ð2Þ
The position of the sample surface before indentation has been chosen as the origin of the indenter (tip of the AFM probe) displacement. The force–displacement curve acquired on the substrate was used for calibration of the cantilever deflection. While the indenter was pushed toward the substrate the indentation force raised steeply without noticeable displacement of the indenter, which means that the substrate deformation was negligible small due to its large hardness and stiffness. Since the displacement of the silicon surface is negligible, h = 0 and δ = z. The force–displacement curves presented in Fig. 5 show that, at the same maximum indentation force, the films deposited at lower r.f. discharge power suffer larger plastic and elastic deformations (larger values of indenter displacement inside the bulk of the films). The plastic deformation occurs during the loading process (portion OA of the force displacement curve of the CFx film deposited at 10 W in Fig. 5) when the AFM tip is displaced until hmax, the loading force increasing until the maximum value, Fmax. The unloading process (portion AB of the force displacement curve of the CFx film deposited at 10 W in Fig. 5) is an elastic relaxation of the film surface when the loading force decreases continuously. Because of the adhesive force between the AFM tip and the CFx film surface, the force acquired during the unloading process reaches minimum at negative values and cancels out at the moment of detachment [24], which indicates the imprint depth, himp (shown by arrows in Fig. 5). The mechanical work performed during the loading, Wl, is partially recovered by the mechanical work of the elastic force of the sample surface during unloading, Wu (Wu b Wl). The area encompassed by the loading and unloading force curves, Wl–Wu, represents the energy loss in plastic deformation of the sample surface. Therefore, the plasticity index of the sample material is defined as [25] ψ=
Wl −Wu ; Wl
ð3Þ
3313
and represents the fraction of energy used for the plastic deformation from the total mechanical work performed during loading. The value of ψ ranges between 0 (fully elastic deformation) and 1(fully plastic deformation). According to the definition above, thin films with high values of ψ are suitable for indentation nanolithography. Films with low hardness and high elasticity modulus are suitable to the indentation lithography because indentation of these films results in large plastic deformations with relatively small elastic deformations (high plasticity index). The plasticity index was determined according the Eq. (3) by processing force displacement curves acquired in indentations of the CFx films deposited at different discharge power values. The obtained values of ψ, which are given in Table 1, shows that the films deposited at discharge power lower than 10 W are suitable to indentation lithography, indentation of these films being mainly plastic (ψ N 0.75). According to the Oliver and Pharr method [18], the dependence of the loading force on the indenter displacement during the unloading part of an indentation performed by an axis symmetrical indenter is a power law dependence. This dependence can be used to determine the contact geometry and the elastic modulus of the sample material. However, the force–displacement curves acquired in indentations of our CFx films did not fit well a power law dependence. This may happened because: 1) the peculiar geometry of the contact as result of particular geometry of the AFM tip and imprint, and 2) action of the indenter-sample adhesive force. Therefore, we used the Doerner and Nix [26] approximation of the flat indenter at the beginning of the unloading process. According to the Sneddon equation [19], the contact stiffness at the beginning of the unloading process is Smax =
dF dh
max
= E⁎ 2rc;max :
ð4Þ
where E⁎ is the reduced elastic modulus of the indenter-sample system, which is 1 1−m2i 1−m2s = + : Ei Es E⁎
ð5Þ
The parameters Ei and νi are the Young's modulus and Poisson ratio of the indenter, while Es and νs are the same parameters for the sample. Because CFx films are much softer than the silicon tips of the AFM probes (Es ≪ Ei) and E⁎ i
Es : 1−m2s
ð6Þ
The parameter rc,max in Eq. (4) is the indenter-sample contact radius at the beginning of the unloading process. Considering that the
Fig. 6. The dependence of CFx film hardness (Hn) and elasticity modulus (E) on the power of the r. f. discharge used in film deposition.
3314
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314
indentation imprint retains the information of the contact area at the maximum loading force, the contact radius at the maximum loading force can be evaluated as rc;max i
dimp : 2
ð7Þ
Then, according Eqs. (4), (6), and (7), the CFx film elastic modulus can be approximated by Smax : Es = 1−m2s dimp
ð8Þ
The information on the imprint diameter is also used to determine the contact area at the maximum loading force and the nanoscale hardness of the CFx films as: Hn =
Fmax 4 Fmax = : 2 π rc;max π d2imp
ð9Þ
The force–displacement data were acquired for indentations on array of 49 points on surfaces of CFx films deposited at 5, 10, 20 and 30 W, respectively. The acquired force–displacement curves were processed to determine Smax and Fmax, The AFM images of the resulted imprints were processed to determine dimp. The obtained information that was used to determine E and Hn according the Eqs. (8) and (9), respectively. Fig. 6 presents plots with the dependence of Hn and E on the power of the r.f. discharge used for PECVD deposition of the CFx films. Both parameters decrease with the decrease of the discharge power. The relatively low values of E and H found for these CFx films are attributed to the low values of r. f. discharge power and selfbiasing potential. Thus, the hardness of these films is typically under 1 GPa. This dependence of film mechanical properties on discharge parameters can be explain by the prevalence of CFx polymer structures in the films deposited at low discharge power and biasing potential [13]. It has been shown by different research groups that hardness of CFx films decreases with the increase of fluorine content of the films. Thus, Bottani et al. showed for the CFx films obtained by PECVD from C2H2 and CF4 gas mixture that film hardness decreases from 13.5 GPa to 2.5 GPa by the increase of the CF4 content of the source gas from 0% to 60%, respectively. The low values of hardness of our deposited films can be attributed to high density of CFx groups in the film structure [16]. Moreover, decrease of plasma discharge power from 30 W to 5 W in our CFx film depositions reduced the ion bombardment of the film surface and favoured formation of polymeric CFx structures with the effect of decrease of hardness from 1.57 GPa to 0.31 GPa. 4. Conclusion We used the AFM indentation technique to build arrays of pits in CFx thin films obtained by PECVD on silicon substrate. Commercially available AFM probes with relatively stiff cantilevers (spring constant of 15–20 N/m) and sharpened tips (curvature radius smaller than
10 nm) were used for performing indentations and imaging of the resulted indentation imprints in tapping-mode AFM. For thin CFx films (thickness of 40 nm) the depth of the indentation imprints increases linearly by the increase of the maximum loading force. The film elasticity modulus and hardness were determined by the measurements of the indenter-sample contact stiffness at the beginning of the unloading process, maximum indentation force, and indenter-sample contact radius at maximum loading force. The latter was approximated by the imprint radius. These measurements showed that the decrease of the discharge power used in PECVD of the CFx films decreased the film hardness and stiffness, while increased the film plasticity index. The film plasticity index was determined by the ratio of the mechanical work used for plastic deformation on the mechanical work of the loading force in the indentation experiments. Films deposited at low discharge power are suitable to nanolithography because of their large plasticity index. Indentation of soft CFx films and imaging imprints can be a routine method of cleaning and checking on the tip condition of commercial tapping AFM probes. References [1] M.R. VanLandingham, S.H. McKnight, G.R. Palmese, J.R. Elings, X. Huang, T.A. Bogetti, R.F. Eduljee, J.W. Gillespie Jr., J. Adhes. 64 (1997) 31. [2] C. Reynaud, F. Sommer, C. Quet, N. El Bounia, T.M. Duc, Surf. Interface Anal. 30 (2000) 185. [3] Y.-G. Jung, B.R. Lawn, M. Martyniuk, H. Huang, X.Z. Hu, J. Mater. Res. 19 (2004) 3076. [4] X.N. Xie, H.J. Chung, C.H. Sow, A.T.S. Wee, Mater. Sci. Eng., R 54 (2006) 1. [5] B. Cappella, E. Bonaccurso, Nanotechnology 18 (2007) 155307. [6] F. Yang, E. Wornyo, K. Gall, W.P. King, Nanotechnology 18 (2007) 285302. [7] T.E. Karis, G.W. Tyndall, D. Fenzel-Alexander, M.S. Crowder, J. Vac. Sci. Technol., A 15 (1997) 2382. [8] H. Ji, A. Côté, D. Koshel, B. Terreault, G. Abel, P. Ducharme, G. Ross, S. Savoie, M. Gagné, Thin Solid Films 405 (2002) 104. [9] A. Shirakura, M. Nakaya, Y. Koga, H. Kodama, T. Hasebe, T. Suzuki, Thin Solid Films 494 (2006) 84. [10] T. Hasebe, A. Shimada, T. Suzuki, Y. Matsuoka, T. Saito, S. Yohena, A. Kamijo, N. Shiraga, M. Higuchi, K. Kimura, H. Yoshimura, S. Kuribayashi, J. Biomed. Mater. Res. 76A (2005) 86. [11] C.E. Bottani, A. Lamperti, L. Nobili, P.M. Ossi, Thin Solid Films 433 (2003) 149. [12] P. Ayala, M.E.H. Maia da Costa, R. Prioli, F.L. Freire Jr., Surf. Coat. Technol. 182 (2004) 335. [13] G. Tang, X. Ma, M. Sun, X. Li, Carbon 43 (2005) 345. [14] D. Koshel, H. Ji, B. Terreault, A. Cote, G.G. Ross, G. Abel, M. Bolduc, Surf. Coat. Technol. 173 (2003) 161. [15] A. Milella, F. Palumbo, P. Favia, G. Cicala, R. d'Agostino, Plasma Process. Polym. 1 (2004) 164. [16] L.-M. Lacroix, M. Lejeune, L. Ceriotti, M. Kormunda, Tarik Meziani, P. Colpo, F. Rossi, Surf. Sci. 592 (2005) 182. [17] W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564. [18] I.N. Sneddon, Int. J. Eng. Sci. 3 (1965) 47. [19] M. Lejeune, L.-M. Lacroix, F. Brétagnol, A. Valsesia, P. Colpo, F. Rossi, Langmuir 22 (2006) 3057. [20] N.A. Burnham, X. Chen, C.S. Hodges, G.A. Matei, E.J. Thoreson, C.J. Roberts, M.C. Davies, S.J.B. Tendler, Nanotechnology 14 (2003) 1. [21] P. Markiewicz, M. Cynthia Goh, Rev. Sci. Instrum. 66 (1995) 3186. [22] L. Sirghi, O. Kilian, D. Gilliland, G. Ceccone, J. Phys. Chem., B 110 (2006) 25975. [23] C.A. Clifford, M.P. Seah, Nanotechnology 17 (2006) 5283. [24] L. Sirghi, F. Rossi, Appl. Phys. Lett. 89 (2006) 243118. [25] B.J. Briscoe, L. Fiori, E. Pelillo, J. Phys., D, Appl. Phys. 31 (1998) 2395. [26] M.F. Doerner, W.D. Nix, J. Mater. Res. 1 (1986) 601.
Contents lists available at ScienceDirect
Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f
Atomic force microscopy indentation of fluorocarbon thin films fabricated by plasma enhanced chemical deposition at low radio frequency power L. Sirghi ⁎, A. Ruiz, P. Colpo, F. Rossi ⁎ European Commission, Institute for Health and Consumer Protection, TP-203, Via E. Fermi, 21020 Ispra (VA), Italy
a r t i c l e
i n f o
Article history: Received 17 April 2008 Received in revised form 16 January 2009 Accepted 16 January 2009 Available online 24 January 2009 PACS codes: 68.37.Ps 68.03.Cd 62.20.de 62.20.F62.23.St
a b s t r a c t Atomic force microscopy (AFM) indentation technique is used for characterization of mechanical properties of fluorocarbon (CFx) thin films obtained from C4F8 gas by plasma enhanced chemical vapour deposition at low r.f. power (5–30 W) and d.c. bias potential (10–80 V). This particular deposition method renders films with good hydrophobic property and high plastic compliance. Commercially available AFM probes with stiff cantilevers (10– 20 N/m) and silicon sharpened tips (tip radius b10 nm) are used for indentations and imaging of the resulted indentation imprints. Force depth curves and imprint characteristics are used for determination of film hardness, elasticity modulus and plasticity index. The measurements show that the decrease of the discharge power results in deposition of films with decreased hardness and stiffness and increased plasticity index. Nanolithography based on AFM indentation is demonstrated on thin films (thickness of 40 nm) with good plastic compliance. © 2009 Elsevier B.V. All rights reserved.
Keywords: Fluorocarbon Nanoindentation Elastic properties Plasticity index
1. Introduction Apart of being an instrument for imaging surfaces with high resolution, the atomic force microscope is also an important tool for assessment of mechanical properties of materials at nanoscale. Thus, the atomic force microscopy (AFM) indentation is widely used for determination of elasticity modulus and hardness of thin films and microscopic objects [1–3]. Indentation is also one of the most used AFMbased lithography techniques [4]. This technique has the advantage that the AFM imaging capability can be used for in situ characterization of the fabricated patterns or nanostructures. Indentation of polymer films is of great interest for nanolithography because of good plastic compliance of polymers [1] and their intensive use in micro and nano patterning. AFM indentations can be used to trigger solvent-assisted ejection of polymer at the indentation imprint in order to build array of pits [5]. Indentation of polymers with heated AFM tips can be an efficient technique to build pits on shape memory polymers [6]. The present paper reports the results of AFM indentation of fluorocarbon (CFx) thin films obtained by plasma enhanced chemical vapour deposition (PECVD) using a low-power r.f. discharge in C4F8 gas. Polymeric CFx films are known to rend a surface with very low energy and relatively good hardness and wearing resistance. These properties make these films ⁎ Corresponding authors. E-mail addresses: [email protected] (L. Sirghi), [email protected] (F. Rossi). 0040-6090/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2009.01.055
interesting for coatings that reduce adhesion of water, oily contaminants, bacteria, etc. Such coatings are desirable for magnetic recording media [7], electric insulators for electric power lines [8], and biomedicine applications [9,10]. Fluorocarbon films with good hydrophobic and hardness properties are usually synthesised by PECVD that employs discharge plasmas to dissociate precursor gases [11,12]. In order to obtain a good compromise between the good hydrophobic property of pure fluorocarbon films and the good hardness and wear resistance of pure hydrocarbon films (which are not hydrophobic) fluorinated carbon films are synthesised by PECVD in mixed hydrocarbon and fluorocarbon gases [11]. For these films, it was found that their hydrophobicity improves by the increase of the fluor content, while the hardness and wearing resistance decrease [12]. An alternative method to synthesise fluorocarbon film consists of sputtering deposition in an r.f. magnetron discharge using polytetrafluoroethilene (PTFE) target [13]. Systematic studies of dependence of hydrophobic property and mechanical properties of these fluorocarbon films on different deposition parameters (discharge power, mixture gas content, substrate biasing potential, gas pressure) are reported in [14]. Our study regards the fluorinated carbon films obtained by PECVD in pure fluorocarbon gas at low r.f. power and d.c. biasing potential, films that show especially good hydrophobicity and plasticity compliance. Details on properties and deposition method of these films are given in a previous published paper [16]. It was found by different authors [15,16] that the CFx films obtained by PECVD in pure fluorocarbon gases at low discharge power have a higher density of
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314 Table 1 Deposition parameters and surface properties of CFx films P (W)
Vsb (V)
t (nm/min)
θw (deg.)
Roughness (nm)
E (GPa)
Hn (GPa)
ψ
5 10 20 30
−10 −20 −60 −80
7 20 36 45
106 ± 1 106 ± 1 105 ± 1 106 ± 1
0.36 ± 0.05 0.52 ± 0.07 0.58 ± 0.07 0.59 ± 0.08
1.75 ± 0.3 2.1 ± 0.6 2.4 ± 0.6 3.2 ± 0.7
0.31 ± 0.03 0.52 ± 0.09 0.88 ± 0.19 1.57 ± 0.46
0.84 ± 0.05 0.77 ± 0.04 0.6 ± 0.04 0.55 ± 0.5
P is the power of r. f. discharge in C4F8 gas; Vsb, the self-biasing potential of the deposition substrate; t, the deposition rate; θw, the water contact angle of the film surface; E, the Young modulus of elasticity; Hn, the film hardness at nanoscale; and ψ, the plasticity index.
CFx bonds and a better hydrophobic property. However, there is no study reporting the mechanical properties of these films, and the effect of lowering of discharge power and biasing potential on these properties has not been investigated. Tang et al. found that hardness and stiffness of CFx films obtained by r.f. magnetron sputtering of PTFE increase by the decrease of the discharge power [13]. This was explained by the decreased fluor content of the films deposited at lower discharge power values. The AFM indentation experiments presented by this paper show that the reduction of the discharge power in the PECVD of CFx films results in decrease of the film hardness and stiffness and increase in plastic compliance. Moreover, it is shown that the CFx films deposited at low discharge power, while having good hydrophobic property, are suitable to AFM indentation-based nanolithography because of their increased plasticity index. Nano patterns consisting in arrays of pits made by AFM indentations are demonstrated. The pits (indentation imprints) are imaged by the same AFM tip to determine their depth and diameter. Force depth-sensing curves acquired during the indentations are processed to determine the film elasticity. While these curves did not fit well a power law, the Oliver and Pharr method [17] of determination of film elasticity was not applicable to these indentation experiments. Therefore, the film stiffness at the beginning of the unloading part of the indentation along with contact radius measurements were used to determine the sample elasticity under the Sneddon approximation of a flat punch [18]. The film hardness is determined by the measurements of the maximum indentation force and the diameter of the resulted imprints.
3311
The indentation experiments were performed by the same AFM apparatus and probe that were used for tapping mode AFM characterization of the film surface. The AFM probe (NSG 11 from NT-MDT Co.) had a stiff cantilever (spring constant around 16 N/m) with a sharpened conical tip (silicon) with cone angle of 22° and apex curvature radius below 10 nm. The AFM tip was kept oscillating in tapping mode at a distance from the film surface equal to the oscillating amplitude (30 nm). Then, indentations were performed by precise forward and backward extension of the AFM scanner. The imprints resulted from the indentation experiments were immediately imaged by tapping-mode AFM. The spring constant of the cantilever was determined by thermal noise method [20], while the shape and condition of the AFM tip was determined by scanning electron microscopy (image not shown) and by imaging the ultra sharp edges of a silicon calibrating grating (TGG1 from NT-MDT). Indentation experiments required the use of AFM probes with stiff cantilevers (15–20 N/m) and very sharp tips (curvature radius under 10 nm).
3. Results and discussion 3.1. Imaging and characterization of indentation imprints To determine the elasticity modulus, E, and hardness, Hn, of the deposited CFx films, sets of 49 indentations were performed on a 7 × 7 arrays of points distributed on an area of 500×500 nm2 on the surfaces of thick films (thickness around 200 nm). Similar indentations were performed on thin films (thickness around 40 nm) for nanolithography purpose. Fig. 1 shows the topography image obtained for an array of imprints on the surface of a CFx film with the thickness of 200 nm deposited at r.f. discharge power of 10 W. Each of the 49 indentations was performed in 1 s time with a maximum loading force of about 300 nN. Similar indentation arrays were performed for indentation time varying between 0.2 s and 2 s. For this time interval, there were no noticeable differences in the imprint parameters. This means that patterns consisting of pit array can be performed on the surface of these films with relatively high speed.
2. Experimental setup and fluorocarbon thin film deposition The fluorocarbon films were deposited on silicon substrate in a plasma reactor that used an asymmetrical r.f. (13.56 MHz) capacitive discharge at low discharge power (5–30 W) in pure C4F8 gas flow at the pressure of 6.5 Pa. The deposition substrate was mounted on the active discharge electrode, which was self biased at a negative potential ranged, depending on the discharge power, between −10 V and −80 V. The film thickness was measured by a stylus profilometer (Alpha-Step IQ from KLA Tencor Co.). Films with two values of thickness were deposited, i.e. thick films (thickness around 200 nm) used for hardness and elasticity measurements, and thin films (thickness around 40 nm) used for AFM indentation-based nanolithography. The static values of water contact angle were measured with a contact angle goniometer (Digidrop from GBX Instruments Co.) on sessile drops of water on the deposited films. The surface topography of the films was characterized by taping mode AFM measurements performed with a commercial AFM apparatus (Solver PRO from NTMDT Co.) with close-loop scanner system. Results of these investigations in terms of deposition rate, water contact angle, and surface roughness for the films deposited at different discharge parameter values are given in Table 1. Results of the nanoscale hardness, elasticity modulus, and plasticity index, which were measured by the AFM indentation methods described below, are also given in the Table 1. More details on deposition system and structure of the deposited CFx films can be found in previously published papers [16,19].
Fig. 1. Topography image of an array of 7 × 7 indentation imprints on a CFx film (thickness of 200 nm) deposited at 10 W. The insets show high resolution (90 nm × 90 nm) topography image of a single imprint. The average depth of imprints is 7 nm, while the average value of imprint diameter is 26 nm. The array pitch is 70 nm. The height scale is 10 nm.
3312
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314
The inset on the topography image in Fig. 1 shows the high resolution topography image of one imprint. The topography image shows axis symmetrical imprints with the diameter of about 26 nm and depth of about 7 nm. Effect of material pile up around the imprints is visible. The axial symmetry of the imprints is illustrated by the roughly identical height profiles along two perpendicular axes (depicted as X and Y on the detail topography image) passing through the imprint centre (see Fig. 2). Because the deformation of the film surface was not entirely plastic, the imprint shape differs from that of the AFM tip. The theoretical height profile of the AFM tip surface is also depicted in Fig. 2. The height profile of the AFM tip is approximated by the parabolic dependence z=
r2 ; 2R
ð1Þ
where z, is the tip surface height, r is the lateral distance measured from the tip apex, and R, the tip curvature radius, which for this particular experiment was estimated to 7 nm by analysis of AFM images of sharp edges of a grating sample (TGG1 from NT-MDT). Since the imprints have a larger curvature radius (estimated to 12 nm) than that of the AFM tip, the values of the imprint depth and diameter can be determined directly on the as-acquired AFM images of the imprints, without being necessary to process the images for removing of the tip-sample convolution effects [21]. As illustrated in Fig. 2, the imprint diameter, dimp, and depth, himp, are determined by the height level of the film surface before indentations (which is approximated to the height level of the film surface far from the indentation place). Indentations the CFx films deposited at low r.f. power discharge can be used for a routine check on the condition of the tips of tapping AFM probes. Indentations performed with AFM probes having blunt tips result in imprints that, by comparison with the imprints performed by sharp tip, have much larger diameter and smaller depth. Also, the force–displacement curves acquired during the indentations performed with a blunt tip show, due to the large contact area, a much higher stiffness of the tip-sample contact. In some cases we noticed that the AFM images taken by the AFM tips after the CFx films indentation experiments were of a better quality than images taken with the same AFM tips before the indentations. We attribute this improvement in the AFM image quality to cleaning through indentations of the AFM tip apex. The CFx films have water and oily repellent surfaces, so that airborne contaminants on the AFM tip apex [22] may be removed during indentations of these films. Indentations performed on thin films (thickness of 40 nm) left imprints with slightly larger depth, when compared to the imprints depth performed on thick films (thickness of 200 nm). This is explained by the fact that the hard substrate limited the elastic deformations of the film, thus enhancing the plastic deformations. We
Fig. 2. Plots of the X and Y height profiles of the imprint shown in detail in Fig. 1(a) along with the plot of height profile of the AFM tip (curvature radius of 7 nm). The origin of height values is taken at level of the CFx film surface before indentation.
Fig. 3. Inverse topography image (a) and height profile (b) of a set of six indentations performed on CFx film (thickness of 40 nm) deposited at 10 W of the r. f. discharge power. The values indicated on the height profile plot are for Fmax.
performed sets of imprints with increasing loading force on thin CFx films (thickness of 40 nm) in order to determine the dependence of the imprint depth on the indentation force. Fig. 3 shows a 3D image (a) and height profile (b) of six imprints that resulted by indentation of the thin CFx film deposited at 10 W with increasing maximum loading force. The height profiles of the resulted imprints were used to determine himp and dimp for different values of Fmax. Fig. 4 shows the dependence of himp on Fmax for thin CFx films deposited at different discharge power values. While the imprint depth remained smaller than the film thickness, it increased linearly by the increase of Fmax. Therefore, linear fits of the experimental data are also shown on Fig. 4. For these measurements, the values of dimp were proportional to square root of Fmax, which is an indication of constant hardness. 3.2. Measurements of plasticity index, elasticity modulus, and hardness To avoid the effect of the stiff substrate [23], the elasticity modulus and hardness were measured based on analysis of force–displacement data acquired on AFM indentation experiments performed on thick CFx films (thickness around 200 nm). The force–displacement data were acquired for indentations performed on array of 49 points on
Fig. 4. The dependence of the imprint depth (himp) on the maximum value of the loading force (Fmax) for the thin CFx films (thickness of 40 nm) deposited at the r.f. discharge power values of 5, 10, 20, and 30 W, respectively.
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314
Fig. 5. Typical force–displacement curves corresponding to indentations of the CFx films deposited at r.f. discharge power values of 10 W, 20 W and 30 W, respectively. The force– displacement curve corresponding to indentation of the film substrate (silicon) is also shown. The origin, O, of the AFM tip displacement is corresponding to the non deformed film surface. The arrows indicate the position of the film surface after the indentations.
surfaces of CFx films deposited at r.f. discharge power values of 5, 10, 20 and 30 W, respectively. Fig. 5 illustrates a comparison between force–displacement curves obtained in indentation experiments on the film substrate and on CFx films deposited at different r.f. discharge power values. The force versus indenter displacement curves are determined on the bases of the force versus scanner extension curves acquired during the indentations. The indenter displacement, h, was determined by subtracting the cantilever deflection, δ, from the scanner extension, z. h = z−δ
ð2Þ
The position of the sample surface before indentation has been chosen as the origin of the indenter (tip of the AFM probe) displacement. The force–displacement curve acquired on the substrate was used for calibration of the cantilever deflection. While the indenter was pushed toward the substrate the indentation force raised steeply without noticeable displacement of the indenter, which means that the substrate deformation was negligible small due to its large hardness and stiffness. Since the displacement of the silicon surface is negligible, h = 0 and δ = z. The force–displacement curves presented in Fig. 5 show that, at the same maximum indentation force, the films deposited at lower r.f. discharge power suffer larger plastic and elastic deformations (larger values of indenter displacement inside the bulk of the films). The plastic deformation occurs during the loading process (portion OA of the force displacement curve of the CFx film deposited at 10 W in Fig. 5) when the AFM tip is displaced until hmax, the loading force increasing until the maximum value, Fmax. The unloading process (portion AB of the force displacement curve of the CFx film deposited at 10 W in Fig. 5) is an elastic relaxation of the film surface when the loading force decreases continuously. Because of the adhesive force between the AFM tip and the CFx film surface, the force acquired during the unloading process reaches minimum at negative values and cancels out at the moment of detachment [24], which indicates the imprint depth, himp (shown by arrows in Fig. 5). The mechanical work performed during the loading, Wl, is partially recovered by the mechanical work of the elastic force of the sample surface during unloading, Wu (Wu b Wl). The area encompassed by the loading and unloading force curves, Wl–Wu, represents the energy loss in plastic deformation of the sample surface. Therefore, the plasticity index of the sample material is defined as [25] ψ=
Wl −Wu ; Wl
ð3Þ
3313
and represents the fraction of energy used for the plastic deformation from the total mechanical work performed during loading. The value of ψ ranges between 0 (fully elastic deformation) and 1(fully plastic deformation). According to the definition above, thin films with high values of ψ are suitable for indentation nanolithography. Films with low hardness and high elasticity modulus are suitable to the indentation lithography because indentation of these films results in large plastic deformations with relatively small elastic deformations (high plasticity index). The plasticity index was determined according the Eq. (3) by processing force displacement curves acquired in indentations of the CFx films deposited at different discharge power values. The obtained values of ψ, which are given in Table 1, shows that the films deposited at discharge power lower than 10 W are suitable to indentation lithography, indentation of these films being mainly plastic (ψ N 0.75). According to the Oliver and Pharr method [18], the dependence of the loading force on the indenter displacement during the unloading part of an indentation performed by an axis symmetrical indenter is a power law dependence. This dependence can be used to determine the contact geometry and the elastic modulus of the sample material. However, the force–displacement curves acquired in indentations of our CFx films did not fit well a power law dependence. This may happened because: 1) the peculiar geometry of the contact as result of particular geometry of the AFM tip and imprint, and 2) action of the indenter-sample adhesive force. Therefore, we used the Doerner and Nix [26] approximation of the flat indenter at the beginning of the unloading process. According to the Sneddon equation [19], the contact stiffness at the beginning of the unloading process is Smax =
dF dh
max
= E⁎ 2rc;max :
ð4Þ
where E⁎ is the reduced elastic modulus of the indenter-sample system, which is 1 1−m2i 1−m2s = + : Ei Es E⁎
ð5Þ
The parameters Ei and νi are the Young's modulus and Poisson ratio of the indenter, while Es and νs are the same parameters for the sample. Because CFx films are much softer than the silicon tips of the AFM probes (Es ≪ Ei) and E⁎ i
Es : 1−m2s
ð6Þ
The parameter rc,max in Eq. (4) is the indenter-sample contact radius at the beginning of the unloading process. Considering that the
Fig. 6. The dependence of CFx film hardness (Hn) and elasticity modulus (E) on the power of the r. f. discharge used in film deposition.
3314
L. Sirghi et al. / Thin Solid Films 517 (2009) 3310–3314
indentation imprint retains the information of the contact area at the maximum loading force, the contact radius at the maximum loading force can be evaluated as rc;max i
dimp : 2
ð7Þ
Then, according Eqs. (4), (6), and (7), the CFx film elastic modulus can be approximated by Smax : Es = 1−m2s dimp
ð8Þ
The information on the imprint diameter is also used to determine the contact area at the maximum loading force and the nanoscale hardness of the CFx films as: Hn =
Fmax 4 Fmax = : 2 π rc;max π d2imp
ð9Þ
The force–displacement data were acquired for indentations on array of 49 points on surfaces of CFx films deposited at 5, 10, 20 and 30 W, respectively. The acquired force–displacement curves were processed to determine Smax and Fmax, The AFM images of the resulted imprints were processed to determine dimp. The obtained information that was used to determine E and Hn according the Eqs. (8) and (9), respectively. Fig. 6 presents plots with the dependence of Hn and E on the power of the r.f. discharge used for PECVD deposition of the CFx films. Both parameters decrease with the decrease of the discharge power. The relatively low values of E and H found for these CFx films are attributed to the low values of r. f. discharge power and selfbiasing potential. Thus, the hardness of these films is typically under 1 GPa. This dependence of film mechanical properties on discharge parameters can be explain by the prevalence of CFx polymer structures in the films deposited at low discharge power and biasing potential [13]. It has been shown by different research groups that hardness of CFx films decreases with the increase of fluorine content of the films. Thus, Bottani et al. showed for the CFx films obtained by PECVD from C2H2 and CF4 gas mixture that film hardness decreases from 13.5 GPa to 2.5 GPa by the increase of the CF4 content of the source gas from 0% to 60%, respectively. The low values of hardness of our deposited films can be attributed to high density of CFx groups in the film structure [16]. Moreover, decrease of plasma discharge power from 30 W to 5 W in our CFx film depositions reduced the ion bombardment of the film surface and favoured formation of polymeric CFx structures with the effect of decrease of hardness from 1.57 GPa to 0.31 GPa. 4. Conclusion We used the AFM indentation technique to build arrays of pits in CFx thin films obtained by PECVD on silicon substrate. Commercially available AFM probes with relatively stiff cantilevers (spring constant of 15–20 N/m) and sharpened tips (curvature radius smaller than
10 nm) were used for performing indentations and imaging of the resulted indentation imprints in tapping-mode AFM. For thin CFx films (thickness of 40 nm) the depth of the indentation imprints increases linearly by the increase of the maximum loading force. The film elasticity modulus and hardness were determined by the measurements of the indenter-sample contact stiffness at the beginning of the unloading process, maximum indentation force, and indenter-sample contact radius at maximum loading force. The latter was approximated by the imprint radius. These measurements showed that the decrease of the discharge power used in PECVD of the CFx films decreased the film hardness and stiffness, while increased the film plasticity index. The film plasticity index was determined by the ratio of the mechanical work used for plastic deformation on the mechanical work of the loading force in the indentation experiments. Films deposited at low discharge power are suitable to nanolithography because of their large plasticity index. Indentation of soft CFx films and imaging imprints can be a routine method of cleaning and checking on the tip condition of commercial tapping AFM probes. References [1] M.R. VanLandingham, S.H. McKnight, G.R. Palmese, J.R. Elings, X. Huang, T.A. Bogetti, R.F. Eduljee, J.W. Gillespie Jr., J. Adhes. 64 (1997) 31. [2] C. Reynaud, F. Sommer, C. Quet, N. El Bounia, T.M. Duc, Surf. Interface Anal. 30 (2000) 185. [3] Y.-G. Jung, B.R. Lawn, M. Martyniuk, H. Huang, X.Z. Hu, J. Mater. Res. 19 (2004) 3076. [4] X.N. Xie, H.J. Chung, C.H. Sow, A.T.S. Wee, Mater. Sci. Eng., R 54 (2006) 1. [5] B. Cappella, E. Bonaccurso, Nanotechnology 18 (2007) 155307. [6] F. Yang, E. Wornyo, K. Gall, W.P. King, Nanotechnology 18 (2007) 285302. [7] T.E. Karis, G.W. Tyndall, D. Fenzel-Alexander, M.S. Crowder, J. Vac. Sci. Technol., A 15 (1997) 2382. [8] H. Ji, A. Côté, D. Koshel, B. Terreault, G. Abel, P. Ducharme, G. Ross, S. Savoie, M. Gagné, Thin Solid Films 405 (2002) 104. [9] A. Shirakura, M. Nakaya, Y. Koga, H. Kodama, T. Hasebe, T. Suzuki, Thin Solid Films 494 (2006) 84. [10] T. Hasebe, A. Shimada, T. Suzuki, Y. Matsuoka, T. Saito, S. Yohena, A. Kamijo, N. Shiraga, M. Higuchi, K. Kimura, H. Yoshimura, S. Kuribayashi, J. Biomed. Mater. Res. 76A (2005) 86. [11] C.E. Bottani, A. Lamperti, L. Nobili, P.M. Ossi, Thin Solid Films 433 (2003) 149. [12] P. Ayala, M.E.H. Maia da Costa, R. Prioli, F.L. Freire Jr., Surf. Coat. Technol. 182 (2004) 335. [13] G. Tang, X. Ma, M. Sun, X. Li, Carbon 43 (2005) 345. [14] D. Koshel, H. Ji, B. Terreault, A. Cote, G.G. Ross, G. Abel, M. Bolduc, Surf. Coat. Technol. 173 (2003) 161. [15] A. Milella, F. Palumbo, P. Favia, G. Cicala, R. d'Agostino, Plasma Process. Polym. 1 (2004) 164. [16] L.-M. Lacroix, M. Lejeune, L. Ceriotti, M. Kormunda, Tarik Meziani, P. Colpo, F. Rossi, Surf. Sci. 592 (2005) 182. [17] W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564. [18] I.N. Sneddon, Int. J. Eng. Sci. 3 (1965) 47. [19] M. Lejeune, L.-M. Lacroix, F. Brétagnol, A. Valsesia, P. Colpo, F. Rossi, Langmuir 22 (2006) 3057. [20] N.A. Burnham, X. Chen, C.S. Hodges, G.A. Matei, E.J. Thoreson, C.J. Roberts, M.C. Davies, S.J.B. Tendler, Nanotechnology 14 (2003) 1. [21] P. Markiewicz, M. Cynthia Goh, Rev. Sci. Instrum. 66 (1995) 3186. [22] L. Sirghi, O. Kilian, D. Gilliland, G. Ceccone, J. Phys. Chem., B 110 (2006) 25975. [23] C.A. Clifford, M.P. Seah, Nanotechnology 17 (2006) 5283. [24] L. Sirghi, F. Rossi, Appl. Phys. Lett. 89 (2006) 243118. [25] B.J. Briscoe, L. Fiori, E. Pelillo, J. Phys., D, Appl. Phys. 31 (1998) 2395. [26] M.F. Doerner, W.D. Nix, J. Mater. Res. 1 (1986) 601.