- Email: [email protected]
Study of magnetic iron nitride thin films deposited by high power impulse magnetron sputtering
SCT-20260; No of Pages 6 Surface & Coatings Technology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Study of magnetic iron nitride thin films deposited by high power impulse magnetron sputtering Akhil Tayal a, Mukul Gupta a, Ajay Gupta b, V. Ganesan a, Layanta Behera a, Surendra Singh c, Saibal Basu c a b c
UGC-DAE Consortium for Scientific Research, University Campus, Khandwa Road, Indore 452 001, India Amity Center for Spintronic Materials, Amity University, Sector 125, Noida 201 303, India Solid State Physics Division, Bhabha Atomic Research Center, Mumbai, 400 085, India
a r t i c l e
i n f o
Article history: Received 14 January 2015 Revised 16 April 2015 Accepted in revised form 6 May 2015 Available online xxxx Keywords: Iron-nitride High power impulse magnetron sputtering
a b s t r a c t In this work, we studied phase formation, structural and magnetic properties of iron-nitride (Fe-N) thin films deposited using high power impulse magnetron sputtering (HiPIMS) and direct current magnetron sputtering (dc-MS) techniques. The nitrogen partial pressure during deposition was systematically varied both in HiPIMS and dc-MS processes. Resulting Fe-N films were characterized for their microstructure, magnetic properties and nitrogen concentration. We found that HiPIMS deposited Fe-N thin films show improved soft magnetic properties and likely to possess globular nanocrystalline microstructure. In addition, it was found that the nitrogen reactivity with Fe get suppressed in HiPIMS discharge as compared to that in dc-MS plasma. Obtained results can be understood in terms of distinct plasma properties of HiPIMS discharge. © 2015 Elsevier B.V. All rights reserved.
1. Introduction HiPIMS is a recently developed technique for the deposition of thin films. Unique plasma conditions associated with it, makes it a preferred choice over conventional deposition methods [1–4]. As compared to dcMS technique, the plasma density in HiPIMS discharge is of the order of 1019 m−3, about 2 orders of magnitude larger than that in dc-MS plasma [2]. In this technique a high power pulse is applied to a magnetron target at low duty cycle (between 0.1 and 1%) that produces a highly dense plasma. In such kind of situation it was observed that the number of sputtered ionized species exceeds neutrals. The volume fraction of ionized species depends on various process parameters such as pulse duration, gas pressure, and peak current [2,5]. These characteristic properties of HiPIMS plasma result in improving film qualities such as film density, hardness, surface roughness, better adhesion, and dense microstructure. Moreover, due to high metal ionization in HiPIMS process, it is expected that thin films deposited via reactive sputtering would display superior properties. As such HiPIMS technique has been frequently utilized for the deposition of metal nitrides such as Al-N [6], Cr-N [7–9], Ti-N [10], and Nb-N [11] and metal oxide thin films such as TiO2 [12,13], Al2O3 [14,15], ZnO [16], ZrO2 [17], and Fe2O3 [18, 19]. In these studies, it was observed that properties of films deposited using reactive HiPIMS process are superior. Konstantindis et al. found that the formation of rutile phase in TiO2 thin film is more favorable as compared to anatase phase when sputtered using HiPIMS technique. Moreover, HiPIMS deposited films show higher refractive index [13].
E-mail addresses: [email protected]/dr, [email protected] (M. Gupta).
Ehiasarian et al. observed that pretreatment using HiPIMS process has improved the adhesion and mechanical properties of CrN thin films [20,21]. Similarly, Reinhard et al. observed improvement in corrosion resistance properties of HiPIMS treated CrN/NbN superlattice structure [11]. Recently, Zhao et al. observed that the optical transmittance of Zirconia thin films deposited using HiPIMS process is more as compared to dc-MS process [17]. Looking at the vast capabilities of HiPIMS technique in depositing various kinds of thin films, it is surprising to note that HiPIMS process has not yet been applied for the deposition of magnetic thin films. Very recently, HiPIMS process has been employed to deposit Fe2O3 [18,19] and FeCuNbSiB thin films [22]. Still magnetic nitride films have not yet been studied with HiPIMS technique. It is well known that transition metal magnetic nitrides are an important class of materials for their usage in various technological applications [23,24]. Therefore, it will be immensely useful to study magnetic nitride films deposited using HiPIMS technique. It is well known that Fe-N compounds are interesting both from the basic and applied point of view. These compounds have a wide range of usage, such as in tribological coatings, magnetic read–write heads, and memory devices [25–27]. In the present work we deposited a series of Fe-N films using HiPIMS technique and compared them with films deposited using dc-MS technique. The structure (local and long range), growth and magnetic properties of the deposited thin films were investigated using various characterization techniques. We found that HiPIMS deposited Fe-N films show improved soft magnetic properties and likely to have a globular nanocrystalline microstructure. In addition, it was found that nitrogen reactivity with Fe get suppressed in HiPIMS process as compared to that in dc-MS process. The obtained
http://dx.doi.org/10.1016/j.surfcoat.2015.05.008 0257-8972/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
2
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
(a)
Table 1 Deposition parameters for samples prepared using HiPIMS and dc-MS techniques. Here Vav is average voltage density (Vcm−2), Iav is average current density (Acm−2), Pav is average power density (Wcm−2), Vp is peak voltage density (Vcm−2), Ip is peak current density (Acm−2), Pp is peak power density (Wcm−2), v is pulse frequency (Hz) and pulse duration is in (μs). Process
Vav
Iav
Pav
Vp
Ip
Pp
Pulse
v
HiPIMS dc-MS
– 7.2
– 0.007
6.8 2.3
15.9 –
1.1 –
750 –
150 –
60 –
(b)
results are presented and discussed in terms of plasma properties of HiPIMS discharge.
2. Experimental details Fe-N thin films were deposited using HiPIMS and dc-MS techniques on Si(100) and float glass substrates using a AJA Int. Inc. make ATC Orion-8 series sputtering system. Pure Fe (purity 99.995%) target (diameter 75 mm, thickness 1 mm) was sputtered using a mixture of Ar and N2 gases. Total gas flow was kept constant at 50 sccm and the relative partial pressure of nitrogen defined as RN2 ¼ PN2 100%= PN2 þ P Ar , (where PN2 and PAr is nitrogen and argon gas flow) was kept at 0, 2, 5, 10, 20, 30, 40 and 50. Before deposition a base pressure of 2 × 10−6 Pa was achieved. During the deposition pressure was kept constant at 0.4 Pa using a dynamic throttling valve and substrate temperature was kept at 423 K. A substrate to target distance of 15 cm was kept fixed during all deposition sequences. Deposition parameters used in dc-MS and HiPIMS processes are listed in Table 1. To get nearly similar deposition rate average power in HiPIMS process was kept higher as compared to dc-MS process. For dc-MS process AJA DCXS 1500 power supply was used, whereas for HiPIMS process a Hüttinger Electronic TruPlasma Highpulse 4002 generator was used. Typical thickness of deposited films was about 80–100 nm. It is well know that the sputtering of a magnetic material using a magnetron source is rather difficult due to shunting of magnetic field lines of magnetron magnets by magnetic target. Therefore, magnetron source used in this work was specially configured for sputtering of Fe targets as shown in Fig. 1(a). The magnetic field strength of magnets in outer ring was about 4 kG with that of central magnet was about 2 kG. Fig. 1(b) shows a sketch of magnetic field lines from the magnetron source configured in un-balanced mode for sputtering of magnetic targets [28]. Structural characterizations of the samples were carried out with Xray diffraction (XRD) using a standard X-ray diffractometer (Bruker D8
Fig. 2. X-ray diffraction patterns of Fe-N thin films prepared at varying RN2 using HiPIMS(a) and dc-MS(b) techniques.
Advance) equipped with Cu K-α X-rays source in θ–2θ geometry. Surface morphology was obtained using atomic force microscopy (AFM) using a Digital Instruments make Nanoscope E AFM system with a Si3N4 cantilever. Magnetic properties were studied using a Quantum Design make superconducting quantum interference device-vibrating sample magnetometer (S-VSM) and polarized neutron reflectivity (PNR). PNR measurements were performed at Dhruva Reactor of Bhaba Atomic Research Center, Mumbai [29]. A magnetic field of strength 2000 Oe was applied along the films plane to saturate samples magnetically. The local structure of nitrogen atoms was studied using soft X-ray absorption spectroscopy (SXAS) at BL-1 beamline of Indus-2 synchrotron radiation source at RRCAT, Indore [30]. SXAS measurements were performed in an UHV chamber in total electron yield (TEY) mode. For accurate measurements of nitrogen concentration, 14N depth profiles were measured using secondary ion mass spectroscopy (SIMS) technique (Hiden Analytical SIMS Workstation). For sputtering, O+ 2 primary ions were used with 5 keV energy and 400 nA beam current. SIMS measurements were performed in an UHV chamber with a base pressure of the order of 8 × 10−8 Pa and during measurements the chamber pressure was 8 × 10−6 Pa. In SIMS measurements quantification of N concentration was performed using well characterize reference samples: α-Fe(N) and ε-Fe2.23N with known nitrogen concentration as 11 at.% [24] and 30 at.% [31], respectively.
(a)
Cylindrical Magnet
(b)
Annular Magnet
Cu Block
Water channel
Fig. 1. A photograph of the magnetron source (AJA International Inc.) showing distribution of magnets (a) and a typical sketch of magnetic lines of force from a magnetron source configured for sputtering of magnetic targets (b).
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
Fig. 3. Variation of crystallite size and coercivity with increasing nitrogen partial pressure. Here solid and open symbols represent crystallite size and coercivity, respectively for HiPIMS and dc-MS samples.
3. Results and discussion Phase formation of Fe-N thin films has been extensively studied by several co-workers using different thin film deposition techniques such as magnetron sputtering [32,33], ion beam sputtering [34,35], e-beam evaporation [36], and pulsed laser deposition [37,38]. In general, as nitrogen partial pressure is increased in a physical vapor deposition method, different types of Fe-N phases are formed and they can be broadly classified as: nanocrystalline α-Fe-N → amorphous αFe-N →α″ − Fe16 N2 → γ′-Fe 4N → ε − Fe 3 − z N (0 ≤ z ≤ 1)→ζFe 2 N → γ‴ ‐ FeN → amorphous/nanocrystalline γ‴-FeN. Since the formation of Fe-N phases with HiPIMS process has not yet been studied, we deposited a series of Fe-N samples using HiPIMS technique and compared them with its sibling i.e. dc-MS technique. Fig. 2 shows XRD patterns of Fe-N thin films deposited using HiPIMS(a) and dc-MS(b) techniques at different nitrogen partial pressures. For RN2 ¼ 0 and 2%, the structure is bcc α-Fe, both in HiPIMS and dc-MS processes. As RN2 increases, predominantly α-Fe structure (along with faint peaks corresponding to γ′ phase) can be seen up to RN2 ¼ 10% in samples deposited using HiPIMS technique, whereas the amount of γ′ phase appears to be larger in samples deposited using dc-MS technique. At RN2 ¼ 20%, intensity of peaks corresponding to γ′ phase increases in HiPIMS deposited sample, but in dc-MS deposited sample along with α-Fe(N) and γ′ phases, peak corresponding to ε phase can also be seen. Further increase in RN2 from 30% to 50% leads to the formation of ε phase in both cases, but faint peaks corresponding to γ′ can only be seen even up to RN2 ¼ 50% in HiPIMS deposited samples. Observed phase formation with varying RN2 is similar for samples
3
deposited using dc-MS process as observed earlier in the literature, but it appears to be somewhat different for HiPIMS deposited samples. To see the variation of crystallite size with increasing RN2 , average crystallite size (d) was calculated using Scherrer formula (for the most intense peak): d = 0.9λ/bcosθ [39], and shown in Fig. 3 [with λ wavelength of X-rays, b an angular width in terms of 2θ and θ is Bragg angle]. As such crystallite sizes are similar in both cases except for RN2 ¼ 0 and 2%, where HiPIMS deposited samples have significantly larger crystallite sizes as compared with samples deposited using dc-MS technique. We did S-VSM and PNR measurements on selected samples to measure the coercivity (HC) and the saturation magnetization (MS), respectively. Fig. 4 shows normalized magnetization with applied magnetic field (M-H) loops of Fe-N samples deposited by varying the RN2 using HiPIMS[(a1)–(e1)] and dc-MS[(a2)–(e2)] techniques. It can be seen that the M-H loops of pure Fe films are almost identical for samples deposited using HiPIMS and dc-MS processes, with HC ~ 10 Oe. As nitrogen is introduced during the deposition, HC increases suddenly to a value of about 70 Oe for dc-MS samples and remains at this value up to RN2 ¼ 10%. However, this behavior is strikingly different for HiPIMS deposited samples; while HC appears to be negligible for RN2 ¼ 2 and 5% [inset of Fig. 4[(a1)–(d1)]] samples, it increases to about 10 Oe for RN2 = 10%. At RN2 = 20%, HC increases in both samples, its value is about 100 Oe and 170 Oe for samples deposited using HiPIMS and dc-MS processes, respectively. Obtained values of HC are plotted in Fig. 3. Typical error in measuring HC is about ±5 Oe. The observed variation in the HC in HiPIMS deposited samples can be understood in terms of random anisotropy model (RAM) [40,41]. If average grain size is below ferromagnetic exchange length (Lex ~ 35 nm for present case [32,42,43]), HC decreases with decreasing the grain size. This happens as exchange interaction between neighboring grains dominates resulting in reduced effective magnetic anisotropy leading to improved soft magnetic properties. However for dc-MS samples, HC increases with decreasing grain size and cannot be understood with RAM. It is known that HiPIMS plasma is highly ionized and its ion energy distribution is broader(~100 eV) as compared to dc-MS plasma (~20– 30 eV) [2,44,45]. Typically average ion energy in HiPIMS discharge is about 10 times larger than that in dc-MS discharge. Such conditions lead to denser films with the relatively smaller amount of defects and vacancies in samples deposited using HiPIMS discharge [2,4,46]. Relatively smaller density of defects could result in smaller stress in the deposited films. This effect leads to improved soft-magnetic properties of films deposited using HiPIMS process. The sudden increase in HC for RN2 = 20%, can be understood in terms of change in the crystal structure as discussed in our XRD results. It may be noted that M-H loops of
Fig. 4. Normalized M-H loops of Fe-N thin films deposited using HiPIMS[(a1)–(e1)] and dc-MS[(a2)–(e2)] at RN2 = 0%[(a1), (a2)], 2%[(b1),(b2)], 5%[(c1), (c2)], 10%[(d1), (d2)], and 20%[(e1), (e2)]. Inset of figure [(a1)–(d1)] shows blown up region of M-H loops near the coercive field.
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
4
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
Fig. 5. PNR patters of Fe-N thin films prepared using HiPIMS(a) and dc-MS(b) at varying nitrogen partial pressure. Inset of figure (b) shows variation of average magnetic moment with increasing nitrogen partial pressure.
samples deposited using HiPIMS technique are somewhat asymmetric, which could be due to the growth of γ′ phase along with α-Fe phase, that also inferred with the XRD data. To investigate the variation in average magnetic moment with increasing nitrogen partial pressure, PNR measurements were performed. It is known that PNR is a very precise tool to measure accurate average magnetic moment specially for magnetic thin films, as it is insensitive to thin film dimensions [47]. Fig. 5 shows PNR patterns of Fe-N thin films deposited using HiPIMS(a) and dcMS(b) techniques prepared at varying RN2 . The obtained PNR patterns were fitted using a computer program based on the Parratt formalism [48] and the average magnetic moment was calculated. Inset of Fig. 5(b) shows the variation of magnetic moment with increasing RN2 . It can be seen that up to RN2 = 10%, magnetic moment of samples does not show any variation (within experimental accuracy) both for dc-MS and HiPIMS processes. However, at RN2 = 20% there is a sudden decrease in the value of magnetic moment in both cases. Interestingly, the sample deposited using dc-MS technique at RN2 = 50% becomes non-magnetic, whereas, it remains magnetic at this nitrogen partial pressure when deposited using HiPIMS technique. It is known that the
(a1)
(b1)
formation of non-magnetic iron nitrides with increasing nitrogen concentration is only observed when N concentration N 30 at.% [37,49,50]. As non-magnetic Fe-N phase is only observed in the case of films prepared using dc-MS process, it indicates that nitrogen incorporation has increased in samples prepared using dc-MS technique as compared to that with HiPIMS technique. The surface morphology of samples was investigated using AFM. Fig. 6 shows AFM images of samples deposited using HiPIMS[(a1)–(d1)] and dc-MS[(a2)–(d2)] techniques at RN2 = 0%[(a1),(a2)], 2%[(b1),(b2)], 5%[(c1),(c2)], and 10%[(d1),(d2)]. The scan area for all measurements was kept constant at 2 μm × 2 μm. For HiPIMS deposited samples, the AFM images indicate that the particle size distribution is more uniform as compared to samples deposited using dc-MS process. Recently it was observed that in HiPIMS deposited films, due to high ionized flux and high adatom energy, surface mobility increases. This process results in repeated nucleation process [4], resulting in a transition from the columnar (as in dc-MS process) to the globular microstructure. In extensive studies performed on Ti-N films deposited using HiPIMS process, Ehiasarian et al. found that as peak current density increases from 0.1 to 0.7 Acm−2 [51], the microstructure becomes denser. Similar results were observed for Cr-N films, where a transition from columnar to globular type growth was observed with increasing peak current density in HiPIMS process [46,52]. Since in our case peak current density is significantly large at 1 Acm−2 (as compared to 0.007 Acm−2 in dc-MS), it is expected that films may have a globular type microstructure. Our surface morphology and magnetization measurements also indicate a denser microstructure formed in HiPIMS deposited samples as compared to that in dc-MS deposited samples. Soft X-ray absorption spectroscopy is a tool to investigate the local structure of nitrogen atoms. We did SXAS measurements near nitrogen K-edge. Obtained spectra are shown in Fig. 7 for selected samples deposited at RN2 = 5%, 20%, and 50% using HiPIMS(a) and dcMS(b) techniques. Generally, a SXAS spectrum of metal nitrides consists of five features as observed in our samples. These features are assigned as (i) I-transition from N 1s to the unoccupied hybridized state of Fe3d– t2g and N 2p (ii) II-transition from N 1s to hybridized Fe 3d–eg and N 2p state (iii) III, IV, V-transition from N 1s to hybridized N 2p and Fe 4sp state [53–56]. In order to compare the relative nitrogen concentration in samples deposited using HiPIMS and dc-MS processes, background before pre-edge and after post-edge was subtracted using a computer program IFEFFIT-Athena [57]. As can be seen from Fig. 7 that the intensity of feature I increases with increasing RN2 both for HiPIMS and dc-MS samples. However, the relative intensity of feature I (for RN2 = 50%), is
(c1)
(d1)
(c2)
(d2)
400nm
(a2)
(b2)
400nm Fig. 6. AFM images of Fe-N thin films prepared using HiPIMS[(a1)–(d1)] and dc-MS[(a2)–(d2)] at RN2 = 0% [(a1), (a2)], 2%[(b1), (b2)], 5% [(c1), (c2)], and 10%[(d1), (d2)].
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
5
systems such as CrN [60,61], TiN [60,62], and TiO2 [63,64] the amount of reactive gas get reduced in HiPIMS plasma, because of significantly higher temperature produced in HiPIMS process due to high power impulse (750 Wcm−2 for 150 μs) as compared to that in dc-MS process (2.3 Wcm−2). This leads to more expansion of process gas (Ar and N2) from the vicinity of a target in a HiPIMS process (than in dc-MS process), reducing the volume of available gas for reaction as well as low deposition rates. This phenomenon is generally known as ‘gasrarefaction’ [2,4,63,64]. Observed variation in N incorporation in Fe-N thin films can be understood with this phenomenon. Fig. 7. Nitrogen K-edge X-ray absorption spectrum of iron nitride thin films prepared using HiPIMS(a) and dc-MS(b) at varying RN2 .
significantly enhanced in dc-MS deposited sample as compared to that in HiPIMS sample. Since the edge intensity in the XAS pattern is proportional to the nitrogen concentration [58], the obtained results clearly indicate that nitrogen concentration is higher for samples prepared using dc-MS technique as compared to HiPIMS technique. From the above analysis using various characterization techniques it appears that the amount of nitrogen in HiPIMS deposited samples is less than that in dc-MS. In order to quantify N concentration, we did SIMS measurements. Form SIMS depth profiles, atomic concentration of an element A in the matrix of element B, can be calculated using a following expression: C A ¼ RSF A
IA IB
ð1Þ
where, CA is atomic concentration, RSFA is relative sensitive factor and IA or IB is observed intensity in a SIMS depth profile for element A or B. In general, exact calculation of RSF is a tedious process due to involved matrix effects, but in the simplest case of two elements RSF can be calculated by using a reference sample with known concentration [24, 59]. Using Eq. (1) we measured N at.% both in HiPIMS and dc-MS deposited samples. Fig. 8 shows the observed variation of nitrogen concentration with increasing RN2 . It can be seen that up to RN2 = 20%, N at.% increases almost linearly in both cases. Between RN2 20 to 40%, N at.% increases marginally but there is a sudden jump at RN2 = 50% in sample deposited using dc-MS process. In contrast, a linear increase in N at.% can be seen almost up to RN2 = 40%, and get saturated for RN2 = 50% in samples deposited using HiPIMS process. However, overall N at.% is always higher in samples deposited using dc-MS technique as compared to that with HiPIMS technique. Intuitively, it may appear that the nitrogen concentration should be more for HiPIMS deposited samples as compared to dc-MS due to highly ionized flux in HiPIMS discharge. However, as observed in other nitride
Fig. 8. Variation of atomic concentration of nitrogen with increasing nitrogen partial pressure deposited using HiPIMS and dc-MS techniques.
4. Conclusion The results obtained from this work provide some distinct properties of Fe-N films deposited using HiPIMS technique: (i) improved soft magnetic properties (ii) denser microstructure and (iii) reduced nitrogen reactivity. Highly ionized and energetic flux produced in HiPIMS (as compared to that in dc-MS) process give rise to high adatom mobility leading to denser microstructure resulting in improved soft magnetic properties. A reduction in N concentration in HiPIMS deposited films as compared to dc-MS deposited films (for a similar value of RN2 ) can be understood by enhanced ‘gas-rarefaction’ in HiPIMS process as compared to that in dc-MS process. Acknowledgments We acknowledge D. M. Phase, D. K. Shukla, R. Sah and S. Karwal for utilization of SXAS beamline. Help provided in AFM measurements by M. Gangrade, and in S-VSM measurements by R. J. Choudhary and P. Pandey is gratefully acknowledged. One of the authors (A. T.) wants to acknowledge CSIR, New Delhi for a research fellowship. References [1] J. Andersson, A. Anders, Phys. Rev. Lett. 102 (2009) 045003 (URL: http://link.aps. org/doi/10.1103/PhysRevLett.102.045003). [2] J.T. Gudmundsson, N. Brenning, D. Lundin, U. Helmersson, J. Vac. Sci. Technol. A 30 (2012) (URL: http://scitation.aip.org/content/avs/journal/jvsta/30/3/10.1116/1. 3691832). [3] A. Anders, J. Andersson, A. Ehiasarian, J. Appl. Phys. 102 (2007) (URL: http:// scitation.aip.org/content/aip/journal/jap/102/11/10.1063/1.2817812). [4] D. Lundin, K. Sarakinos, J. Mater. Res. 27 (2012) 780–792. [5] K. Sarakinos, J. Alami, S. Konstantinidis, Surf. Coat. Technol. 204 (2010) 1661–1684, http://dx.doi.org/10.1016/j.surfcoat.2009.11.013 (URL: http://www.sciencedirect. com/science/article/pii/S0257897209009426). [6] A. Guillaumot, F. Lapostolle, C. Dublanche-Tixier, J. Oliveira, A. Billard, C. Langlade, Vacuum 85 (2010) 120–125, http://dx.doi.org/10.1016/j.vacuum.2010.04.012 (URL: http://www.sciencedirect.com/science/article/pii/S0042207X10001612). [7] A.P. Ehiasarian, J.G. Wen, I. Petrov, J. Appl. Phys. 101 (2007) (URL: http://scitation. aip.org/content/aip/journal/jap/101/5/10.1063/1.2697052). [8] J. Lin, W.D. Sproul, J.J. Moore, S. Lee, S. Myers, Surf. Coat. Technol. 205 (2011) 3226–3234, http://dx.doi.org/10.1016/j.surfcoat.2010.11.039 (URL: http://www. sciencedirect.com/science/article/pii/S0257897210012132). [9] M. Hala, N. Viau, O. Zabeida, J.E. Klemberg-Sapieha, L. Martinu, J. Appl. Phys. 107 (2010) (URL: http://scitation.aip.org/content/aip/journal/jap/107/4/10.1063/1. 3305319). [10] M. Lattemann, U. Helmersson, J. Greene, Thin Solid Films 518 (2010) 5978–5980, http://dx.doi.org/10.1016/j.tsf.2010.05.064 (URL: http://www.sciencedirect. com/science/article/pii/S0040609010007340). [11] C. Reinhard, A. Ehiasarian, P. Hovsepian, Thin Solid Films 515 (2007) 3685–3692, http://dx.doi.org/10.1016/j.tsf.2006.11.014 (URL: http://www.sciencedirect. com/science/article/pii/S0040609006013332). [12] M. Aiempanakit, U. Helmersson, A. Aijaz, P. Larsson, R. Magnusson, J. Jensen, T. Kubart, Surf. Coat. Technol. 205 (2011) 4828–4831, http://dx.doi.org/10.1016/j. surfcoat.2011.04.071 (URL: http://www.sciencedirect.com/science/article/pii/ S0257897211004270). [13] S. Konstantinidis, J. Dauchot, M. Hecq, Thin Solid Films 515 (2006) 1182–1186, http://dx.doi.org/10.1016/j.tsf.2006.07.089 (URL: http://www.sciencedirect. com/science/article/pii/S0040609006009060, proceedings of the 33rd International Conference on Metallurgical Coatings and, Thin Films {ICMCTF} 2006). [14] E. Wallin, U. Helmersson, Thin Solid Films 516 (2008) 6398–6401, http://dx.doi.org/ 10.1016/j.tsf.2007.08.123 (URL: http://www.sciencedirect.com/science/article/pii/ S0040609007015374). [15] E. Wallin, T.I. Selinder, M. Elfwing, U. Helmersson, Europhys. Lett. 82 (2008) 36002 (URL: http://stacks.iop.org/0295-5075/82/i=3/a=36002). [16] S. Konstantinidis, A. Hemberg, J.P. Dauchot, M. Hecq, J. Vac. Sci. Technol. B 25 (2007).
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
6
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
[17] X. Zhao, J. Jin, J.-C. Cheng, J.-W. Lee, K.-H. Wu, K.-C. Lin, J.-R. Tsai, K.-C. Liu, Thin Solid Films (2014)http://dx.doi.org/10.1016/j.tsf.2014.05.060 (URL: http://www. sciencedirect.com/science/article/pii/S004060901400635X). [18] S. Kment, Z. Hubicka, J. Krysa, J. Olejnicek, M. Cada, I. Gregora, M. Zlamal, M. Brunclikova, Z. Remes, N. Liu, L. Wang, R. Kirchgeorg, C. Lee, P. Schmuki, Catal. Today 230 (2014) 8–14, http://dx.doi.org/10.1016/j.cattod.2013.11.045 (URL: http://www.sciencedirect.com/science/article/pii/S0920586113006500, selected contributions of the 4th International Conference on Semiconductor, Photochemistry (SP4)). [19] Z. Hubicka, S. Kment, J. Olejnícek, M. Cada, T. Kubart, M. Brunclíková, P. Kšírová, P. Adámek, Z. Remeš, Thin Solid Films 549 (2013) 184–191, http://dx.doi.org/10. 1016/j.tsf.2013.09.031 (URL: http://www.sciencedirect.com/science/article/pii/ S0040609013014995, the 40th International Conference on Metallurgical Coatings and Thin Films). [20] A. Ehiasarian, P. Hovsepian, L. Hultman, U. Helmersson, Thin Solid Films 457 (2004) 270–277, http://dx.doi.org/10.1016/j.tsf.2003.11.113 (URL: http://www.sciencedirect. com/science/article/pii/S0040609003017115). [21] M. Lattemann, A. Ehiasarian, J. Bohlmark, P. Persson, U. Helmersson, Surf. Coat. Technol. 200 (2006) 6495–6499, http://dx.doi.org/10.1016/j.surfcoat.2005.11.082 (URL: http://www.sciencedirect.com/science/article/pii/S0257897205012442, {EMRS} 2005 Symposium K Protective Coatings and Thin Films {EMRS} 2005 Symposium K on Protective Coatings and Thin Films). [22] I.-L. Velicu, V. Tiron, G. Popa, Surf. Coat. Technol. 250 (2014) 57–64, http://dx.doi. org/10.1016/j.surfcoat.2014.03.015 (URL: http://www.sciencedirect.com/science/ article/pii/S0257897214002242, 4th International Conference on Fundamentals and Applications of {HIPIMS} and Final Event of {COST} Action {MP0804} Highly Ionized Pulse Plasma Processes {HIPP} Processes). [23] J.M.D. Coey, P.A.I. Smith, J. Magn. Magn. Mater. 200 (1999) 405–424, http://dx.doi. org/10.1016/S0304-8853(99)00429-1. [24] A. Tayal, M. Gupta, N.P. Lalla, A. Gupta, M. Horisberger, J. Stahn, K. Schlage, H.-C. Wille, Phys. Rev. B 90 (2014) 144412, http://dx.doi.org/10.1103/PhysRevB.90. 144412 (URL: http://link.aps.org/doi/10.1103/PhysRevB.90.144412). [25] C. Navo, J. Alvarez, M.J. Capitan, F. Yndurain, R. Miranda, Phys. Rev. B 78 (2008) 155417, http://dx.doi.org/10.1103/PhysRevB.78.155417. [26] C. Navo, J. Alvarez, M.J. Capitan, J. Camarero, R. Miranda, Appl. Phys. Lett. 94 (2009) 263112, http://dx.doi.org/10.1063/1.3159630. [27] Y.-K. Liu, M.H. Kryder, Appl. Phys. Lett. 77 (2000) 426–428 (URL: http://dx.doi.org/ doi/10.1063/1.126998). [28] B. Window, N. Savvides, J. Vac. Sci. Technol. A 4 (1986). [29] S. Basu, S. Singh, J. Neutron Res. 14 (2006) 109–120 (URL: http://www. tandfonline.com/doi/abs/10.1080/10238160500475301, arXiv: http://www. tandfonline.com/doi/pdf/10.1080/10238160500475301). [30] D.M. Phase, M. Gupta, S. Potdar, L. Behera, R. Sah, A. Gupta, AIP Conf. Proc. 1591 (2014) 685–686 (URL: http://scitation.aip.org/content/aip/proceeding/aipcp/10. 1063/1.4872719). [31] M. Gupta, A. Tayal, A. Gupta, R. Gupta, J. Stahn, M. Horisberger, A. Wildes, J. Appl. Phys. 110 (2011) 123518, http://dx.doi.org/10.1063/1.3671532. [32] R. Gupta, M. Gupta, Phys. Rev. B 72 (2005) 024202, http://dx.doi.org/10.1103/ PhysRevB.72.024202. [33] E.B. Easton, T. Buhrmester, J. Dahn, Thin Solid Films 493 (2005) 60–66, http://dx.doi. org/10.1016/j.tsf.2005.06.063 (URL: http://www.sciencedirect.com/science/article/ pii/S0040609005007030). [34] D.M. Borsa, D.O. Boerma, Hyperfine Interact. 151–152 (2003) 31–48, http://dx.doi. org/10.1023/B:HYPE.0000020403.64670.02. [35] X. Zhao Ding, F. Min Zhang, J. Sheng Yan, H. Lie Shen, X. Wang, X. Huai Liu, D.-F. Shen, J. Appl. Phys. 82 (1997) 5154–5158, http://dx.doi.org/10.1063/1.366319. [36] M. Vergnat, P. Bauer, H. Chatbi, G. Marchal, Thin Solid Films 275 (1996) 251–253, http://dx.doi.org/10.1016/0040-6090(95)07055-9 (URL: http://www.sciencedirect. com/science/article/pii/0040609095070559, Papers presented at the European Materials Research Society 1995 Spring Conference, Symposium E: Structure and Properties of Metallic Thin Films and Multilayers). [37] P. Schaaf, Prog. Mater. Sci. 47 (2002) 1–161, http://dx.doi.org/10.1016/S00796425(00)00003-7.
[38] M. Gupta, A. Gupta, P. Bhattacharya, P. Misra, L. Kukreja, J. Alloys Compd. 326 (2001) 265–269, http://dx.doi.org/10.1016/S0925-8388(01)01316-0 (URL: http://www. sciencedirect.com/science/article/pii/S0925838801013160, Proceedings of the International Conference on Magnetic Materials (ICMM)). [39] B.D. Cullity, Elements of X-ray Diffraction, Addison-Wesley, MA, 1978. [40] G. Herzer, IEEE Trans. Magn. 25 (1989) 3327. [41] J.F. Löffler, H.-B. Braun, W. Wagner, Phys. Rev. Lett. 85 (2000) 1990–1993 (URL: http://link.aps.org/doi/10.1103/PhysRevLett.85.1990). [42] J.F. Loffler, J.P. Meier, B. Doudin, J.-P. Ansermet, W. Wagner, Phys. Rev. B 57 (1998) 2915. [43] X. Bao, R.M. Metzger, W.D. Doyle, J. Appl. Phys. 73 (1993). [44] D. Lundin, P. Larsson, E. Wallin, M. Lattemann, N. Brenning, U. Helmersson, Plasma Sources Sci. Technol. 17 (2008) 035021 (URL: http://stacks.iop.org/0963-0252/17/ i=3/a=035021). [45] A. Hecimovic, A.P. Ehiasarian, J. Appl. Phys. 108 (2010) (URL: http://scitation.aip. org/content/aip/journal/jap/108/6/10.1063/1.3486018). [46] E. Lecoq, J. Guillot, D. Duday, J.-B. Chemin, P. Choquet, J. Phys. D. Appl. Phys. 47 (2014) 195201 (URL: http://stacks.iop.org/0022-3727/47/i=19/a=195201). [47] S.J. Blundell, J.A.C. Bland, Phys. Rev. B 46 (1992) 3391–3400 (URL: http://link.aps. org/doi/10.1103/PhysRevB.46.3391). [48] L.G. Parratt, Phys. Rev. 95 (1954) 359. [49] G. Chen, N. Jaggi, J. Butt, E. Yeh, L. Schwartz, J. Phys. Chem. 87 (1983) 5326–5332. [50] A. Tayal, M. Gupta, A. Gupta, M. Horisberger, J. Stahn, Thin Solid Films 536 (2013) 39–49, http://dx.doi.org/10.1016/j.tsf.2013.03.026 (URL: http://www.sciencedirect. com/science/article/pii/S0040609013004574). [51] A.P. Ehiasarian, A. Vetushka, Y.A. Gonzalvo, G. Sfrn, L. Szkely, P.B. Barna, J. Appl. Phys. 109 (2011) (URL: http://scitation.aip.org/content/aip/journal/jap/109/10/10. 1063/1.3579443). [52] J. Alami, K. Sarakinos, F. Uslu, M. Wuttig, J. Phys. D. Appl. Phys. 42 (2009) 015304 (URL: http://stacks.iop.org/0022-3727/42/i=1/a=015304). [53] J. Pfluger, J. Fink, G. Crecelius, K. Bohnen, H. Winter, Solid State Commun. 44 (1982) 489–492, http://dx.doi.org/10.1016/0038-1098(82)90130-2 (URL: http://www. sciencedirect.com/science/article/pii/0038109882901302). [54] F. Esaka, H. Shimada, M. Imamura, N. Matsubayashi, T. Sato, A. Nishijima, A. Kawana, H. Ichimura, T. Kikuchi, K. Furuya, Thin Solid Films 281282 (1996) 314–317, http:// dx.doi.org/10.1016/S0040-6090(96)88640-8 (URL: http://www.sciencedirect. com/science/article/pii/S0040609096886408). [55] C. Mitterbauer, C. HAbert, G. Kothleitner, F. Hofer, P. Schattschneider, H. Zandbergen, Solid State Commun. 130 (2004) 209–213, http://dx.doi.org/10.1016/j.ssc.2004.01. 045 (URL: http://www.sciencedirect.com/science/article/pii/S0038109804000936). [56] I. Jouanny, P. Weisbecker, V. Demange, M. Grafouté, O. Peña, E. Bauer-Grosse, Thin Solid Films 518 (2010) 1883–1891, http://dx.doi.org/10.1016/j.tsf.2009.07.039. [57] B. Ravel, M. Newville, J. Synchrotron Radiat. 12 (2005) 537–541 (URL: http://dx.doi.org/ 10.1107/S0909049505012719). [58] Y. Kihn, C. Mirguet, L. Calmels, J. Electron Spectrosc. Relat. Phenom. 143 (2005) 117–127, http://dx.doi.org/10.1016/j.elspec.2004.02.170 (URL: http://www. sciencedirect.com/science/article/pii/S0368204804004098, electron Energy Loss Spectroscopy in the Electron Microscope Electron Energy Loss Spectroscopy in the Electron Microscope). [59] M. Py, J.P. Barnes, D. Lafond, J.M. Hartmann, Rapid Commun. Mass Spectrom. 25 (2011) 629–638 (URL: http://dx.doi.org/10.1002/rcm.4904). [60] A. Vetushka, A.P. Ehiasarian, J. Phys. D. Appl. Phys. 41 (2008) 015204 (URL: http:// stacks.iop.org/0022-3727/41/i=1/a=015204). [61] G. Greczynski, L. Hultman, Vacuum 84 (2010) 1159–1170, http://dx.doi.org/10. 1016/j.vacuum.2010.01.055 (URL: http://www.sciencedirect.com/science/article/ pii/S0042207X10000795). [62] F. Mitschker, M. Prenzel, J. Benedikt, C. Maszl, A. von Keudell, J. Phys. D. Appl. Phys. 46 (2013) 495201 (URL: http://stacks.iop.org/0022-3727/46/i=49/a=495201). [63] K. Sarakinos, J. Alami, M. Wuttig, J. Phys. D. Appl. Phys. 40 (2007) 2108 (URL: http:// stacks.iop.org/0022-3727/40/i=7/a=037). [64] C. Huo, M.A. Raadu, D. Lundin, J.T. Gudmundsson, A. Anders, N. Brenning, Plasma Sources Sci. Technol. 21 (2012) 045004 (URL: http://stacks.iop.org/0963-0252/21/ i=4/a=045004).
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Study of magnetic iron nitride thin films deposited by high power impulse magnetron sputtering Akhil Tayal a, Mukul Gupta a, Ajay Gupta b, V. Ganesan a, Layanta Behera a, Surendra Singh c, Saibal Basu c a b c
UGC-DAE Consortium for Scientific Research, University Campus, Khandwa Road, Indore 452 001, India Amity Center for Spintronic Materials, Amity University, Sector 125, Noida 201 303, India Solid State Physics Division, Bhabha Atomic Research Center, Mumbai, 400 085, India
a r t i c l e
i n f o
Article history: Received 14 January 2015 Revised 16 April 2015 Accepted in revised form 6 May 2015 Available online xxxx Keywords: Iron-nitride High power impulse magnetron sputtering
a b s t r a c t In this work, we studied phase formation, structural and magnetic properties of iron-nitride (Fe-N) thin films deposited using high power impulse magnetron sputtering (HiPIMS) and direct current magnetron sputtering (dc-MS) techniques. The nitrogen partial pressure during deposition was systematically varied both in HiPIMS and dc-MS processes. Resulting Fe-N films were characterized for their microstructure, magnetic properties and nitrogen concentration. We found that HiPIMS deposited Fe-N thin films show improved soft magnetic properties and likely to possess globular nanocrystalline microstructure. In addition, it was found that the nitrogen reactivity with Fe get suppressed in HiPIMS discharge as compared to that in dc-MS plasma. Obtained results can be understood in terms of distinct plasma properties of HiPIMS discharge. © 2015 Elsevier B.V. All rights reserved.
1. Introduction HiPIMS is a recently developed technique for the deposition of thin films. Unique plasma conditions associated with it, makes it a preferred choice over conventional deposition methods [1–4]. As compared to dcMS technique, the plasma density in HiPIMS discharge is of the order of 1019 m−3, about 2 orders of magnitude larger than that in dc-MS plasma [2]. In this technique a high power pulse is applied to a magnetron target at low duty cycle (between 0.1 and 1%) that produces a highly dense plasma. In such kind of situation it was observed that the number of sputtered ionized species exceeds neutrals. The volume fraction of ionized species depends on various process parameters such as pulse duration, gas pressure, and peak current [2,5]. These characteristic properties of HiPIMS plasma result in improving film qualities such as film density, hardness, surface roughness, better adhesion, and dense microstructure. Moreover, due to high metal ionization in HiPIMS process, it is expected that thin films deposited via reactive sputtering would display superior properties. As such HiPIMS technique has been frequently utilized for the deposition of metal nitrides such as Al-N [6], Cr-N [7–9], Ti-N [10], and Nb-N [11] and metal oxide thin films such as TiO2 [12,13], Al2O3 [14,15], ZnO [16], ZrO2 [17], and Fe2O3 [18, 19]. In these studies, it was observed that properties of films deposited using reactive HiPIMS process are superior. Konstantindis et al. found that the formation of rutile phase in TiO2 thin film is more favorable as compared to anatase phase when sputtered using HiPIMS technique. Moreover, HiPIMS deposited films show higher refractive index [13].
E-mail addresses: [email protected]/dr, [email protected] (M. Gupta).
Ehiasarian et al. observed that pretreatment using HiPIMS process has improved the adhesion and mechanical properties of CrN thin films [20,21]. Similarly, Reinhard et al. observed improvement in corrosion resistance properties of HiPIMS treated CrN/NbN superlattice structure [11]. Recently, Zhao et al. observed that the optical transmittance of Zirconia thin films deposited using HiPIMS process is more as compared to dc-MS process [17]. Looking at the vast capabilities of HiPIMS technique in depositing various kinds of thin films, it is surprising to note that HiPIMS process has not yet been applied for the deposition of magnetic thin films. Very recently, HiPIMS process has been employed to deposit Fe2O3 [18,19] and FeCuNbSiB thin films [22]. Still magnetic nitride films have not yet been studied with HiPIMS technique. It is well known that transition metal magnetic nitrides are an important class of materials for their usage in various technological applications [23,24]. Therefore, it will be immensely useful to study magnetic nitride films deposited using HiPIMS technique. It is well known that Fe-N compounds are interesting both from the basic and applied point of view. These compounds have a wide range of usage, such as in tribological coatings, magnetic read–write heads, and memory devices [25–27]. In the present work we deposited a series of Fe-N films using HiPIMS technique and compared them with films deposited using dc-MS technique. The structure (local and long range), growth and magnetic properties of the deposited thin films were investigated using various characterization techniques. We found that HiPIMS deposited Fe-N films show improved soft magnetic properties and likely to have a globular nanocrystalline microstructure. In addition, it was found that nitrogen reactivity with Fe get suppressed in HiPIMS process as compared to that in dc-MS process. The obtained
http://dx.doi.org/10.1016/j.surfcoat.2015.05.008 0257-8972/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
2
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
(a)
Table 1 Deposition parameters for samples prepared using HiPIMS and dc-MS techniques. Here Vav is average voltage density (Vcm−2), Iav is average current density (Acm−2), Pav is average power density (Wcm−2), Vp is peak voltage density (Vcm−2), Ip is peak current density (Acm−2), Pp is peak power density (Wcm−2), v is pulse frequency (Hz) and pulse duration is in (μs). Process
Vav
Iav
Pav
Vp
Ip
Pp
Pulse
v
HiPIMS dc-MS
– 7.2
– 0.007
6.8 2.3
15.9 –
1.1 –
750 –
150 –
60 –
(b)
results are presented and discussed in terms of plasma properties of HiPIMS discharge.
2. Experimental details Fe-N thin films were deposited using HiPIMS and dc-MS techniques on Si(100) and float glass substrates using a AJA Int. Inc. make ATC Orion-8 series sputtering system. Pure Fe (purity 99.995%) target (diameter 75 mm, thickness 1 mm) was sputtered using a mixture of Ar and N2 gases. Total gas flow was kept constant at 50 sccm and the relative partial pressure of nitrogen defined as RN2 ¼ PN2 100%= PN2 þ P Ar , (where PN2 and PAr is nitrogen and argon gas flow) was kept at 0, 2, 5, 10, 20, 30, 40 and 50. Before deposition a base pressure of 2 × 10−6 Pa was achieved. During the deposition pressure was kept constant at 0.4 Pa using a dynamic throttling valve and substrate temperature was kept at 423 K. A substrate to target distance of 15 cm was kept fixed during all deposition sequences. Deposition parameters used in dc-MS and HiPIMS processes are listed in Table 1. To get nearly similar deposition rate average power in HiPIMS process was kept higher as compared to dc-MS process. For dc-MS process AJA DCXS 1500 power supply was used, whereas for HiPIMS process a Hüttinger Electronic TruPlasma Highpulse 4002 generator was used. Typical thickness of deposited films was about 80–100 nm. It is well know that the sputtering of a magnetic material using a magnetron source is rather difficult due to shunting of magnetic field lines of magnetron magnets by magnetic target. Therefore, magnetron source used in this work was specially configured for sputtering of Fe targets as shown in Fig. 1(a). The magnetic field strength of magnets in outer ring was about 4 kG with that of central magnet was about 2 kG. Fig. 1(b) shows a sketch of magnetic field lines from the magnetron source configured in un-balanced mode for sputtering of magnetic targets [28]. Structural characterizations of the samples were carried out with Xray diffraction (XRD) using a standard X-ray diffractometer (Bruker D8
Fig. 2. X-ray diffraction patterns of Fe-N thin films prepared at varying RN2 using HiPIMS(a) and dc-MS(b) techniques.
Advance) equipped with Cu K-α X-rays source in θ–2θ geometry. Surface morphology was obtained using atomic force microscopy (AFM) using a Digital Instruments make Nanoscope E AFM system with a Si3N4 cantilever. Magnetic properties were studied using a Quantum Design make superconducting quantum interference device-vibrating sample magnetometer (S-VSM) and polarized neutron reflectivity (PNR). PNR measurements were performed at Dhruva Reactor of Bhaba Atomic Research Center, Mumbai [29]. A magnetic field of strength 2000 Oe was applied along the films plane to saturate samples magnetically. The local structure of nitrogen atoms was studied using soft X-ray absorption spectroscopy (SXAS) at BL-1 beamline of Indus-2 synchrotron radiation source at RRCAT, Indore [30]. SXAS measurements were performed in an UHV chamber in total electron yield (TEY) mode. For accurate measurements of nitrogen concentration, 14N depth profiles were measured using secondary ion mass spectroscopy (SIMS) technique (Hiden Analytical SIMS Workstation). For sputtering, O+ 2 primary ions were used with 5 keV energy and 400 nA beam current. SIMS measurements were performed in an UHV chamber with a base pressure of the order of 8 × 10−8 Pa and during measurements the chamber pressure was 8 × 10−6 Pa. In SIMS measurements quantification of N concentration was performed using well characterize reference samples: α-Fe(N) and ε-Fe2.23N with known nitrogen concentration as 11 at.% [24] and 30 at.% [31], respectively.
(a)
Cylindrical Magnet
(b)
Annular Magnet
Cu Block
Water channel
Fig. 1. A photograph of the magnetron source (AJA International Inc.) showing distribution of magnets (a) and a typical sketch of magnetic lines of force from a magnetron source configured for sputtering of magnetic targets (b).
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
Fig. 3. Variation of crystallite size and coercivity with increasing nitrogen partial pressure. Here solid and open symbols represent crystallite size and coercivity, respectively for HiPIMS and dc-MS samples.
3. Results and discussion Phase formation of Fe-N thin films has been extensively studied by several co-workers using different thin film deposition techniques such as magnetron sputtering [32,33], ion beam sputtering [34,35], e-beam evaporation [36], and pulsed laser deposition [37,38]. In general, as nitrogen partial pressure is increased in a physical vapor deposition method, different types of Fe-N phases are formed and they can be broadly classified as: nanocrystalline α-Fe-N → amorphous αFe-N →α″ − Fe16 N2 → γ′-Fe 4N → ε − Fe 3 − z N (0 ≤ z ≤ 1)→ζFe 2 N → γ‴ ‐ FeN → amorphous/nanocrystalline γ‴-FeN. Since the formation of Fe-N phases with HiPIMS process has not yet been studied, we deposited a series of Fe-N samples using HiPIMS technique and compared them with its sibling i.e. dc-MS technique. Fig. 2 shows XRD patterns of Fe-N thin films deposited using HiPIMS(a) and dc-MS(b) techniques at different nitrogen partial pressures. For RN2 ¼ 0 and 2%, the structure is bcc α-Fe, both in HiPIMS and dc-MS processes. As RN2 increases, predominantly α-Fe structure (along with faint peaks corresponding to γ′ phase) can be seen up to RN2 ¼ 10% in samples deposited using HiPIMS technique, whereas the amount of γ′ phase appears to be larger in samples deposited using dc-MS technique. At RN2 ¼ 20%, intensity of peaks corresponding to γ′ phase increases in HiPIMS deposited sample, but in dc-MS deposited sample along with α-Fe(N) and γ′ phases, peak corresponding to ε phase can also be seen. Further increase in RN2 from 30% to 50% leads to the formation of ε phase in both cases, but faint peaks corresponding to γ′ can only be seen even up to RN2 ¼ 50% in HiPIMS deposited samples. Observed phase formation with varying RN2 is similar for samples
3
deposited using dc-MS process as observed earlier in the literature, but it appears to be somewhat different for HiPIMS deposited samples. To see the variation of crystallite size with increasing RN2 , average crystallite size (d) was calculated using Scherrer formula (for the most intense peak): d = 0.9λ/bcosθ [39], and shown in Fig. 3 [with λ wavelength of X-rays, b an angular width in terms of 2θ and θ is Bragg angle]. As such crystallite sizes are similar in both cases except for RN2 ¼ 0 and 2%, where HiPIMS deposited samples have significantly larger crystallite sizes as compared with samples deposited using dc-MS technique. We did S-VSM and PNR measurements on selected samples to measure the coercivity (HC) and the saturation magnetization (MS), respectively. Fig. 4 shows normalized magnetization with applied magnetic field (M-H) loops of Fe-N samples deposited by varying the RN2 using HiPIMS[(a1)–(e1)] and dc-MS[(a2)–(e2)] techniques. It can be seen that the M-H loops of pure Fe films are almost identical for samples deposited using HiPIMS and dc-MS processes, with HC ~ 10 Oe. As nitrogen is introduced during the deposition, HC increases suddenly to a value of about 70 Oe for dc-MS samples and remains at this value up to RN2 ¼ 10%. However, this behavior is strikingly different for HiPIMS deposited samples; while HC appears to be negligible for RN2 ¼ 2 and 5% [inset of Fig. 4[(a1)–(d1)]] samples, it increases to about 10 Oe for RN2 = 10%. At RN2 = 20%, HC increases in both samples, its value is about 100 Oe and 170 Oe for samples deposited using HiPIMS and dc-MS processes, respectively. Obtained values of HC are plotted in Fig. 3. Typical error in measuring HC is about ±5 Oe. The observed variation in the HC in HiPIMS deposited samples can be understood in terms of random anisotropy model (RAM) [40,41]. If average grain size is below ferromagnetic exchange length (Lex ~ 35 nm for present case [32,42,43]), HC decreases with decreasing the grain size. This happens as exchange interaction between neighboring grains dominates resulting in reduced effective magnetic anisotropy leading to improved soft magnetic properties. However for dc-MS samples, HC increases with decreasing grain size and cannot be understood with RAM. It is known that HiPIMS plasma is highly ionized and its ion energy distribution is broader(~100 eV) as compared to dc-MS plasma (~20– 30 eV) [2,44,45]. Typically average ion energy in HiPIMS discharge is about 10 times larger than that in dc-MS discharge. Such conditions lead to denser films with the relatively smaller amount of defects and vacancies in samples deposited using HiPIMS discharge [2,4,46]. Relatively smaller density of defects could result in smaller stress in the deposited films. This effect leads to improved soft-magnetic properties of films deposited using HiPIMS process. The sudden increase in HC for RN2 = 20%, can be understood in terms of change in the crystal structure as discussed in our XRD results. It may be noted that M-H loops of
Fig. 4. Normalized M-H loops of Fe-N thin films deposited using HiPIMS[(a1)–(e1)] and dc-MS[(a2)–(e2)] at RN2 = 0%[(a1), (a2)], 2%[(b1),(b2)], 5%[(c1), (c2)], 10%[(d1), (d2)], and 20%[(e1), (e2)]. Inset of figure [(a1)–(d1)] shows blown up region of M-H loops near the coercive field.
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
4
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
Fig. 5. PNR patters of Fe-N thin films prepared using HiPIMS(a) and dc-MS(b) at varying nitrogen partial pressure. Inset of figure (b) shows variation of average magnetic moment with increasing nitrogen partial pressure.
samples deposited using HiPIMS technique are somewhat asymmetric, which could be due to the growth of γ′ phase along with α-Fe phase, that also inferred with the XRD data. To investigate the variation in average magnetic moment with increasing nitrogen partial pressure, PNR measurements were performed. It is known that PNR is a very precise tool to measure accurate average magnetic moment specially for magnetic thin films, as it is insensitive to thin film dimensions [47]. Fig. 5 shows PNR patterns of Fe-N thin films deposited using HiPIMS(a) and dcMS(b) techniques prepared at varying RN2 . The obtained PNR patterns were fitted using a computer program based on the Parratt formalism [48] and the average magnetic moment was calculated. Inset of Fig. 5(b) shows the variation of magnetic moment with increasing RN2 . It can be seen that up to RN2 = 10%, magnetic moment of samples does not show any variation (within experimental accuracy) both for dc-MS and HiPIMS processes. However, at RN2 = 20% there is a sudden decrease in the value of magnetic moment in both cases. Interestingly, the sample deposited using dc-MS technique at RN2 = 50% becomes non-magnetic, whereas, it remains magnetic at this nitrogen partial pressure when deposited using HiPIMS technique. It is known that the
(a1)
(b1)
formation of non-magnetic iron nitrides with increasing nitrogen concentration is only observed when N concentration N 30 at.% [37,49,50]. As non-magnetic Fe-N phase is only observed in the case of films prepared using dc-MS process, it indicates that nitrogen incorporation has increased in samples prepared using dc-MS technique as compared to that with HiPIMS technique. The surface morphology of samples was investigated using AFM. Fig. 6 shows AFM images of samples deposited using HiPIMS[(a1)–(d1)] and dc-MS[(a2)–(d2)] techniques at RN2 = 0%[(a1),(a2)], 2%[(b1),(b2)], 5%[(c1),(c2)], and 10%[(d1),(d2)]. The scan area for all measurements was kept constant at 2 μm × 2 μm. For HiPIMS deposited samples, the AFM images indicate that the particle size distribution is more uniform as compared to samples deposited using dc-MS process. Recently it was observed that in HiPIMS deposited films, due to high ionized flux and high adatom energy, surface mobility increases. This process results in repeated nucleation process [4], resulting in a transition from the columnar (as in dc-MS process) to the globular microstructure. In extensive studies performed on Ti-N films deposited using HiPIMS process, Ehiasarian et al. found that as peak current density increases from 0.1 to 0.7 Acm−2 [51], the microstructure becomes denser. Similar results were observed for Cr-N films, where a transition from columnar to globular type growth was observed with increasing peak current density in HiPIMS process [46,52]. Since in our case peak current density is significantly large at 1 Acm−2 (as compared to 0.007 Acm−2 in dc-MS), it is expected that films may have a globular type microstructure. Our surface morphology and magnetization measurements also indicate a denser microstructure formed in HiPIMS deposited samples as compared to that in dc-MS deposited samples. Soft X-ray absorption spectroscopy is a tool to investigate the local structure of nitrogen atoms. We did SXAS measurements near nitrogen K-edge. Obtained spectra are shown in Fig. 7 for selected samples deposited at RN2 = 5%, 20%, and 50% using HiPIMS(a) and dcMS(b) techniques. Generally, a SXAS spectrum of metal nitrides consists of five features as observed in our samples. These features are assigned as (i) I-transition from N 1s to the unoccupied hybridized state of Fe3d– t2g and N 2p (ii) II-transition from N 1s to hybridized Fe 3d–eg and N 2p state (iii) III, IV, V-transition from N 1s to hybridized N 2p and Fe 4sp state [53–56]. In order to compare the relative nitrogen concentration in samples deposited using HiPIMS and dc-MS processes, background before pre-edge and after post-edge was subtracted using a computer program IFEFFIT-Athena [57]. As can be seen from Fig. 7 that the intensity of feature I increases with increasing RN2 both for HiPIMS and dc-MS samples. However, the relative intensity of feature I (for RN2 = 50%), is
(c1)
(d1)
(c2)
(d2)
400nm
(a2)
(b2)
400nm Fig. 6. AFM images of Fe-N thin films prepared using HiPIMS[(a1)–(d1)] and dc-MS[(a2)–(d2)] at RN2 = 0% [(a1), (a2)], 2%[(b1), (b2)], 5% [(c1), (c2)], and 10%[(d1), (d2)].
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
5
systems such as CrN [60,61], TiN [60,62], and TiO2 [63,64] the amount of reactive gas get reduced in HiPIMS plasma, because of significantly higher temperature produced in HiPIMS process due to high power impulse (750 Wcm−2 for 150 μs) as compared to that in dc-MS process (2.3 Wcm−2). This leads to more expansion of process gas (Ar and N2) from the vicinity of a target in a HiPIMS process (than in dc-MS process), reducing the volume of available gas for reaction as well as low deposition rates. This phenomenon is generally known as ‘gasrarefaction’ [2,4,63,64]. Observed variation in N incorporation in Fe-N thin films can be understood with this phenomenon. Fig. 7. Nitrogen K-edge X-ray absorption spectrum of iron nitride thin films prepared using HiPIMS(a) and dc-MS(b) at varying RN2 .
significantly enhanced in dc-MS deposited sample as compared to that in HiPIMS sample. Since the edge intensity in the XAS pattern is proportional to the nitrogen concentration [58], the obtained results clearly indicate that nitrogen concentration is higher for samples prepared using dc-MS technique as compared to HiPIMS technique. From the above analysis using various characterization techniques it appears that the amount of nitrogen in HiPIMS deposited samples is less than that in dc-MS. In order to quantify N concentration, we did SIMS measurements. Form SIMS depth profiles, atomic concentration of an element A in the matrix of element B, can be calculated using a following expression: C A ¼ RSF A
IA IB
ð1Þ
where, CA is atomic concentration, RSFA is relative sensitive factor and IA or IB is observed intensity in a SIMS depth profile for element A or B. In general, exact calculation of RSF is a tedious process due to involved matrix effects, but in the simplest case of two elements RSF can be calculated by using a reference sample with known concentration [24, 59]. Using Eq. (1) we measured N at.% both in HiPIMS and dc-MS deposited samples. Fig. 8 shows the observed variation of nitrogen concentration with increasing RN2 . It can be seen that up to RN2 = 20%, N at.% increases almost linearly in both cases. Between RN2 20 to 40%, N at.% increases marginally but there is a sudden jump at RN2 = 50% in sample deposited using dc-MS process. In contrast, a linear increase in N at.% can be seen almost up to RN2 = 40%, and get saturated for RN2 = 50% in samples deposited using HiPIMS process. However, overall N at.% is always higher in samples deposited using dc-MS technique as compared to that with HiPIMS technique. Intuitively, it may appear that the nitrogen concentration should be more for HiPIMS deposited samples as compared to dc-MS due to highly ionized flux in HiPIMS discharge. However, as observed in other nitride
Fig. 8. Variation of atomic concentration of nitrogen with increasing nitrogen partial pressure deposited using HiPIMS and dc-MS techniques.
4. Conclusion The results obtained from this work provide some distinct properties of Fe-N films deposited using HiPIMS technique: (i) improved soft magnetic properties (ii) denser microstructure and (iii) reduced nitrogen reactivity. Highly ionized and energetic flux produced in HiPIMS (as compared to that in dc-MS) process give rise to high adatom mobility leading to denser microstructure resulting in improved soft magnetic properties. A reduction in N concentration in HiPIMS deposited films as compared to dc-MS deposited films (for a similar value of RN2 ) can be understood by enhanced ‘gas-rarefaction’ in HiPIMS process as compared to that in dc-MS process. Acknowledgments We acknowledge D. M. Phase, D. K. Shukla, R. Sah and S. Karwal for utilization of SXAS beamline. Help provided in AFM measurements by M. Gangrade, and in S-VSM measurements by R. J. Choudhary and P. Pandey is gratefully acknowledged. One of the authors (A. T.) wants to acknowledge CSIR, New Delhi for a research fellowship. References [1] J. Andersson, A. Anders, Phys. Rev. Lett. 102 (2009) 045003 (URL: http://link.aps. org/doi/10.1103/PhysRevLett.102.045003). [2] J.T. Gudmundsson, N. Brenning, D. Lundin, U. Helmersson, J. Vac. Sci. Technol. A 30 (2012) (URL: http://scitation.aip.org/content/avs/journal/jvsta/30/3/10.1116/1. 3691832). [3] A. Anders, J. Andersson, A. Ehiasarian, J. Appl. Phys. 102 (2007) (URL: http:// scitation.aip.org/content/aip/journal/jap/102/11/10.1063/1.2817812). [4] D. Lundin, K. Sarakinos, J. Mater. Res. 27 (2012) 780–792. [5] K. Sarakinos, J. Alami, S. Konstantinidis, Surf. Coat. Technol. 204 (2010) 1661–1684, http://dx.doi.org/10.1016/j.surfcoat.2009.11.013 (URL: http://www.sciencedirect. com/science/article/pii/S0257897209009426). [6] A. Guillaumot, F. Lapostolle, C. Dublanche-Tixier, J. Oliveira, A. Billard, C. Langlade, Vacuum 85 (2010) 120–125, http://dx.doi.org/10.1016/j.vacuum.2010.04.012 (URL: http://www.sciencedirect.com/science/article/pii/S0042207X10001612). [7] A.P. Ehiasarian, J.G. Wen, I. Petrov, J. Appl. Phys. 101 (2007) (URL: http://scitation. aip.org/content/aip/journal/jap/101/5/10.1063/1.2697052). [8] J. Lin, W.D. Sproul, J.J. Moore, S. Lee, S. Myers, Surf. Coat. Technol. 205 (2011) 3226–3234, http://dx.doi.org/10.1016/j.surfcoat.2010.11.039 (URL: http://www. sciencedirect.com/science/article/pii/S0257897210012132). [9] M. Hala, N. Viau, O. Zabeida, J.E. Klemberg-Sapieha, L. Martinu, J. Appl. Phys. 107 (2010) (URL: http://scitation.aip.org/content/aip/journal/jap/107/4/10.1063/1. 3305319). [10] M. Lattemann, U. Helmersson, J. Greene, Thin Solid Films 518 (2010) 5978–5980, http://dx.doi.org/10.1016/j.tsf.2010.05.064 (URL: http://www.sciencedirect. com/science/article/pii/S0040609010007340). [11] C. Reinhard, A. Ehiasarian, P. Hovsepian, Thin Solid Films 515 (2007) 3685–3692, http://dx.doi.org/10.1016/j.tsf.2006.11.014 (URL: http://www.sciencedirect. com/science/article/pii/S0040609006013332). [12] M. Aiempanakit, U. Helmersson, A. Aijaz, P. Larsson, R. Magnusson, J. Jensen, T. Kubart, Surf. Coat. Technol. 205 (2011) 4828–4831, http://dx.doi.org/10.1016/j. surfcoat.2011.04.071 (URL: http://www.sciencedirect.com/science/article/pii/ S0257897211004270). [13] S. Konstantinidis, J. Dauchot, M. Hecq, Thin Solid Films 515 (2006) 1182–1186, http://dx.doi.org/10.1016/j.tsf.2006.07.089 (URL: http://www.sciencedirect. com/science/article/pii/S0040609006009060, proceedings of the 33rd International Conference on Metallurgical Coatings and, Thin Films {ICMCTF} 2006). [14] E. Wallin, U. Helmersson, Thin Solid Films 516 (2008) 6398–6401, http://dx.doi.org/ 10.1016/j.tsf.2007.08.123 (URL: http://www.sciencedirect.com/science/article/pii/ S0040609007015374). [15] E. Wallin, T.I. Selinder, M. Elfwing, U. Helmersson, Europhys. Lett. 82 (2008) 36002 (URL: http://stacks.iop.org/0295-5075/82/i=3/a=36002). [16] S. Konstantinidis, A. Hemberg, J.P. Dauchot, M. Hecq, J. Vac. Sci. Technol. B 25 (2007).
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008
6
A. Tayal et al. / Surface & Coatings Technology xxx (2015) xxx–xxx
[17] X. Zhao, J. Jin, J.-C. Cheng, J.-W. Lee, K.-H. Wu, K.-C. Lin, J.-R. Tsai, K.-C. Liu, Thin Solid Films (2014)http://dx.doi.org/10.1016/j.tsf.2014.05.060 (URL: http://www. sciencedirect.com/science/article/pii/S004060901400635X). [18] S. Kment, Z. Hubicka, J. Krysa, J. Olejnicek, M. Cada, I. Gregora, M. Zlamal, M. Brunclikova, Z. Remes, N. Liu, L. Wang, R. Kirchgeorg, C. Lee, P. Schmuki, Catal. Today 230 (2014) 8–14, http://dx.doi.org/10.1016/j.cattod.2013.11.045 (URL: http://www.sciencedirect.com/science/article/pii/S0920586113006500, selected contributions of the 4th International Conference on Semiconductor, Photochemistry (SP4)). [19] Z. Hubicka, S. Kment, J. Olejnícek, M. Cada, T. Kubart, M. Brunclíková, P. Kšírová, P. Adámek, Z. Remeš, Thin Solid Films 549 (2013) 184–191, http://dx.doi.org/10. 1016/j.tsf.2013.09.031 (URL: http://www.sciencedirect.com/science/article/pii/ S0040609013014995, the 40th International Conference on Metallurgical Coatings and Thin Films). [20] A. Ehiasarian, P. Hovsepian, L. Hultman, U. Helmersson, Thin Solid Films 457 (2004) 270–277, http://dx.doi.org/10.1016/j.tsf.2003.11.113 (URL: http://www.sciencedirect. com/science/article/pii/S0040609003017115). [21] M. Lattemann, A. Ehiasarian, J. Bohlmark, P. Persson, U. Helmersson, Surf. Coat. Technol. 200 (2006) 6495–6499, http://dx.doi.org/10.1016/j.surfcoat.2005.11.082 (URL: http://www.sciencedirect.com/science/article/pii/S0257897205012442, {EMRS} 2005 Symposium K Protective Coatings and Thin Films {EMRS} 2005 Symposium K on Protective Coatings and Thin Films). [22] I.-L. Velicu, V. Tiron, G. Popa, Surf. Coat. Technol. 250 (2014) 57–64, http://dx.doi. org/10.1016/j.surfcoat.2014.03.015 (URL: http://www.sciencedirect.com/science/ article/pii/S0257897214002242, 4th International Conference on Fundamentals and Applications of {HIPIMS} and Final Event of {COST} Action {MP0804} Highly Ionized Pulse Plasma Processes {HIPP} Processes). [23] J.M.D. Coey, P.A.I. Smith, J. Magn. Magn. Mater. 200 (1999) 405–424, http://dx.doi. org/10.1016/S0304-8853(99)00429-1. [24] A. Tayal, M. Gupta, N.P. Lalla, A. Gupta, M. Horisberger, J. Stahn, K. Schlage, H.-C. Wille, Phys. Rev. B 90 (2014) 144412, http://dx.doi.org/10.1103/PhysRevB.90. 144412 (URL: http://link.aps.org/doi/10.1103/PhysRevB.90.144412). [25] C. Navo, J. Alvarez, M.J. Capitan, F. Yndurain, R. Miranda, Phys. Rev. B 78 (2008) 155417, http://dx.doi.org/10.1103/PhysRevB.78.155417. [26] C. Navo, J. Alvarez, M.J. Capitan, J. Camarero, R. Miranda, Appl. Phys. Lett. 94 (2009) 263112, http://dx.doi.org/10.1063/1.3159630. [27] Y.-K. Liu, M.H. Kryder, Appl. Phys. Lett. 77 (2000) 426–428 (URL: http://dx.doi.org/ doi/10.1063/1.126998). [28] B. Window, N. Savvides, J. Vac. Sci. Technol. A 4 (1986). [29] S. Basu, S. Singh, J. Neutron Res. 14 (2006) 109–120 (URL: http://www. tandfonline.com/doi/abs/10.1080/10238160500475301, arXiv: http://www. tandfonline.com/doi/pdf/10.1080/10238160500475301). [30] D.M. Phase, M. Gupta, S. Potdar, L. Behera, R. Sah, A. Gupta, AIP Conf. Proc. 1591 (2014) 685–686 (URL: http://scitation.aip.org/content/aip/proceeding/aipcp/10. 1063/1.4872719). [31] M. Gupta, A. Tayal, A. Gupta, R. Gupta, J. Stahn, M. Horisberger, A. Wildes, J. Appl. Phys. 110 (2011) 123518, http://dx.doi.org/10.1063/1.3671532. [32] R. Gupta, M. Gupta, Phys. Rev. B 72 (2005) 024202, http://dx.doi.org/10.1103/ PhysRevB.72.024202. [33] E.B. Easton, T. Buhrmester, J. Dahn, Thin Solid Films 493 (2005) 60–66, http://dx.doi. org/10.1016/j.tsf.2005.06.063 (URL: http://www.sciencedirect.com/science/article/ pii/S0040609005007030). [34] D.M. Borsa, D.O. Boerma, Hyperfine Interact. 151–152 (2003) 31–48, http://dx.doi. org/10.1023/B:HYPE.0000020403.64670.02. [35] X. Zhao Ding, F. Min Zhang, J. Sheng Yan, H. Lie Shen, X. Wang, X. Huai Liu, D.-F. Shen, J. Appl. Phys. 82 (1997) 5154–5158, http://dx.doi.org/10.1063/1.366319. [36] M. Vergnat, P. Bauer, H. Chatbi, G. Marchal, Thin Solid Films 275 (1996) 251–253, http://dx.doi.org/10.1016/0040-6090(95)07055-9 (URL: http://www.sciencedirect. com/science/article/pii/0040609095070559, Papers presented at the European Materials Research Society 1995 Spring Conference, Symposium E: Structure and Properties of Metallic Thin Films and Multilayers). [37] P. Schaaf, Prog. Mater. Sci. 47 (2002) 1–161, http://dx.doi.org/10.1016/S00796425(00)00003-7.
[38] M. Gupta, A. Gupta, P. Bhattacharya, P. Misra, L. Kukreja, J. Alloys Compd. 326 (2001) 265–269, http://dx.doi.org/10.1016/S0925-8388(01)01316-0 (URL: http://www. sciencedirect.com/science/article/pii/S0925838801013160, Proceedings of the International Conference on Magnetic Materials (ICMM)). [39] B.D. Cullity, Elements of X-ray Diffraction, Addison-Wesley, MA, 1978. [40] G. Herzer, IEEE Trans. Magn. 25 (1989) 3327. [41] J.F. Löffler, H.-B. Braun, W. Wagner, Phys. Rev. Lett. 85 (2000) 1990–1993 (URL: http://link.aps.org/doi/10.1103/PhysRevLett.85.1990). [42] J.F. Loffler, J.P. Meier, B. Doudin, J.-P. Ansermet, W. Wagner, Phys. Rev. B 57 (1998) 2915. [43] X. Bao, R.M. Metzger, W.D. Doyle, J. Appl. Phys. 73 (1993). [44] D. Lundin, P. Larsson, E. Wallin, M. Lattemann, N. Brenning, U. Helmersson, Plasma Sources Sci. Technol. 17 (2008) 035021 (URL: http://stacks.iop.org/0963-0252/17/ i=3/a=035021). [45] A. Hecimovic, A.P. Ehiasarian, J. Appl. Phys. 108 (2010) (URL: http://scitation.aip. org/content/aip/journal/jap/108/6/10.1063/1.3486018). [46] E. Lecoq, J. Guillot, D. Duday, J.-B. Chemin, P. Choquet, J. Phys. D. Appl. Phys. 47 (2014) 195201 (URL: http://stacks.iop.org/0022-3727/47/i=19/a=195201). [47] S.J. Blundell, J.A.C. Bland, Phys. Rev. B 46 (1992) 3391–3400 (URL: http://link.aps. org/doi/10.1103/PhysRevB.46.3391). [48] L.G. Parratt, Phys. Rev. 95 (1954) 359. [49] G. Chen, N. Jaggi, J. Butt, E. Yeh, L. Schwartz, J. Phys. Chem. 87 (1983) 5326–5332. [50] A. Tayal, M. Gupta, A. Gupta, M. Horisberger, J. Stahn, Thin Solid Films 536 (2013) 39–49, http://dx.doi.org/10.1016/j.tsf.2013.03.026 (URL: http://www.sciencedirect. com/science/article/pii/S0040609013004574). [51] A.P. Ehiasarian, A. Vetushka, Y.A. Gonzalvo, G. Sfrn, L. Szkely, P.B. Barna, J. Appl. Phys. 109 (2011) (URL: http://scitation.aip.org/content/aip/journal/jap/109/10/10. 1063/1.3579443). [52] J. Alami, K. Sarakinos, F. Uslu, M. Wuttig, J. Phys. D. Appl. Phys. 42 (2009) 015304 (URL: http://stacks.iop.org/0022-3727/42/i=1/a=015304). [53] J. Pfluger, J. Fink, G. Crecelius, K. Bohnen, H. Winter, Solid State Commun. 44 (1982) 489–492, http://dx.doi.org/10.1016/0038-1098(82)90130-2 (URL: http://www. sciencedirect.com/science/article/pii/0038109882901302). [54] F. Esaka, H. Shimada, M. Imamura, N. Matsubayashi, T. Sato, A. Nishijima, A. Kawana, H. Ichimura, T. Kikuchi, K. Furuya, Thin Solid Films 281282 (1996) 314–317, http:// dx.doi.org/10.1016/S0040-6090(96)88640-8 (URL: http://www.sciencedirect. com/science/article/pii/S0040609096886408). [55] C. Mitterbauer, C. HAbert, G. Kothleitner, F. Hofer, P. Schattschneider, H. Zandbergen, Solid State Commun. 130 (2004) 209–213, http://dx.doi.org/10.1016/j.ssc.2004.01. 045 (URL: http://www.sciencedirect.com/science/article/pii/S0038109804000936). [56] I. Jouanny, P. Weisbecker, V. Demange, M. Grafouté, O. Peña, E. Bauer-Grosse, Thin Solid Films 518 (2010) 1883–1891, http://dx.doi.org/10.1016/j.tsf.2009.07.039. [57] B. Ravel, M. Newville, J. Synchrotron Radiat. 12 (2005) 537–541 (URL: http://dx.doi.org/ 10.1107/S0909049505012719). [58] Y. Kihn, C. Mirguet, L. Calmels, J. Electron Spectrosc. Relat. Phenom. 143 (2005) 117–127, http://dx.doi.org/10.1016/j.elspec.2004.02.170 (URL: http://www. sciencedirect.com/science/article/pii/S0368204804004098, electron Energy Loss Spectroscopy in the Electron Microscope Electron Energy Loss Spectroscopy in the Electron Microscope). [59] M. Py, J.P. Barnes, D. Lafond, J.M. Hartmann, Rapid Commun. Mass Spectrom. 25 (2011) 629–638 (URL: http://dx.doi.org/10.1002/rcm.4904). [60] A. Vetushka, A.P. Ehiasarian, J. Phys. D. Appl. Phys. 41 (2008) 015204 (URL: http:// stacks.iop.org/0022-3727/41/i=1/a=015204). [61] G. Greczynski, L. Hultman, Vacuum 84 (2010) 1159–1170, http://dx.doi.org/10. 1016/j.vacuum.2010.01.055 (URL: http://www.sciencedirect.com/science/article/ pii/S0042207X10000795). [62] F. Mitschker, M. Prenzel, J. Benedikt, C. Maszl, A. von Keudell, J. Phys. D. Appl. Phys. 46 (2013) 495201 (URL: http://stacks.iop.org/0022-3727/46/i=49/a=495201). [63] K. Sarakinos, J. Alami, M. Wuttig, J. Phys. D. Appl. Phys. 40 (2007) 2108 (URL: http:// stacks.iop.org/0022-3727/40/i=7/a=037). [64] C. Huo, M.A. Raadu, D. Lundin, J.T. Gudmundsson, A. Anders, N. Brenning, Plasma Sources Sci. Technol. 21 (2012) 045004 (URL: http://stacks.iop.org/0963-0252/21/ i=4/a=045004).
Please cite this article as: A. Tayal, et al., Surf. Coat. Technol. (2015), http://dx.doi.org/10.1016/j.surfcoat.2015.05.008