Optical performance of thin films produced by the pulsed laser deposition of SiAlON and Er targets

Optical performance of thin films produced by the pulsed laser deposition of SiAlON and Er targets

Accepted Manuscript Title: Optical performance of thin films produced by the pulsed laser deposition of SiAlON and Er targets Author: I. Camps J.M. Ra...

941KB Sizes 1 Downloads 42 Views

Accepted Manuscript Title: Optical performance of thin films produced by the pulsed laser deposition of SiAlON and Er targets Author: I. Camps J.M. Ram´ırez A. Mariscal R. Serna B. Garrido M. Per´alvarez J. Carreras N.P. Barradas L.C. Alves E. Alves PII: DOI: Reference:

S0169-4332(14)02701-9 http://dx.doi.org/doi:10.1016/j.apsusc.2014.12.013 APSUSC 29248

To appear in:

APSUSC

Received date: Revised date: Accepted date:

7-7-2014 3-12-2014 3-12-2014

Please cite this article as: I. Camps, J.M. Ram´irez, A. Mariscal, R. Serna, B. Garrido, M. Per´alvarez, J. Carreras, N.P. Barradas, L.C. Alves, E. Alves, Optical performance of thin films produced by the pulsed laser deposition of SiAlON and Er targets, Applied Surface Science (2014), http://dx.doi.org/10.1016/j.apsusc.2014.12.013 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Optical performance of thin films produced by the pulsed laser deposition of SiAlON and Er targets. I. Camps1, J.M. Ramírez2, A. Mariscal1, R. Serna1, B. Garrido2, M. Perálvarez3, J. Carreras3, N. P. Barradas4, L.C. Alves4, E. Alves5. 1

Laser Processing Group, Instituto de Óptica, CSIC, C/Serrano 121, 28006 Madrid, Spain. MIND-IN2UB, Departament d'Electrònica, Universitat de Barcelona, c/Martí i Franqués 1, 08028 Barcelona, Spain. 3 IREC, Fundació Privada Institut de Recerca en Energia de Catalunya. 4 2 C TN, Instituto Superior Técnico, Universidade de Lisboa, E.N.10, 2695-066 Bobadela, Portugal. 5 IPFN, Instituto Superior Técnico, Universidade de Lisboa, E.N. 10, 2695-066 Bobadela, Portugal.

cr

ip t

2

M

an

us

Abstract We report the preparation and optical performance of thin films produced by pulsed laser deposition in vacuum at room temperature, by focusing an ArF excimer laser onto two separate targets: a commercial ceramic SiAlON and a metallic Er target. As a result of the alternate deposition Er:SiAlON films were formed. The as grown films exhibited an Er-related emission peaking at 1532 nm. The role of the PLD energy density during deposition on the final matrix film was investigated, in order to achieve an optimized matrix composition with enhanced optical properties, and its effect on the light emission performance. Keywords: PLD, Pulsed laser deposition, SiAlON, silicon oxynitride, PL, photoluminescence, erbium, Er, ellipsometry, refractive index.

Ac ce

pt

ed

1. Introduction Silicon oxynitride (SiON) materials have been the subject of research since they offer a suitable technological platform for the development of integrated optoelectronic devices such as light emitting devices (LEDs) [1,2,3] or high speed silicon optical modulators [4,5]. This is due to their excellent properties that include a large refractive index that can be tuned from that of SiO2 (1.45) [6] to that of Si3N4 (2.01) [7], high transparency in the visible-near infrared range, good electrical properties and compatibility with the current Si-technology. More recently, in the search for efficient light emitting materials with superior performance and large integration capability has motivated the study of more complex matrices such as aluminum-doped silicon oxynitrides.[3,8,9] Among the several advantages provided by this quaternary host, the possibility of tuning the electrical and optical properties, the enhanced light emission from luminescent species and the decrease in the maximum phonon energy become major key points that stimulate the ongoing research of these materials. The aim of this work is to report the first preliminary results on the performance of SiAlON Erdoped thin films (Er:SiAlON) grown by PLD from a commercially available ceramic [10]. The SiAlON ceramic has been known for its outstanding wear, corrosion and thermal resistance offering cutting-edge solutions to many engineering and industrial problems. Although it has also been investigated as phosphors in powder form, it has rarely been studied in thin film form. Several advantages are foreseen when using SiAlON as a thin film, such as its excellent thermoelectric properties, the robustness of its composites, its suitability as a host matrix for rare earth ions, and 1

Page 1 of 13

cr

ip t

the fact that it is a highly available material in industry that covers a wide range of low cost applications [8,9,11]. Pulsed laser deposition (PLD) has proven to be excellent for the preparation of complex oxides, and production of high density films with good adhesion [12]. It is thus very attractive for the deposition of high quality films for optical applications. In our group, we have produced in the past rare-earth doped films by the alternate ablation of aluminum oxide ceramic host (Al2O3) and rare-earth targets (Er, Yb) in vacuum [13,14]. In this work we will show that good quality films of Er-doped SiAlON can be produced in a single step in vacuum when suitable deposition conditions are chosen and that they have potential for telecom optical applications at 1.5 µm.

Ac ce

pt

ed

M

an

us

2. Experimental The PLD system consisted of a UV laser ArF excimer (λ = 193 nm, 20 ns pulse duration) and a vacuum chamber equipped with a multi-target system that can accommodate up to four targets and alternate the target to be ablated. The laser beam was focused alternately onto the ceramic SiAlON target and the metallic Er targets at an angle of incidence of 45°. During the process of ablation the targets were rotating to prevent crater formation. For the deposition the so called onaxis configuration was used and thus the centre of the substrate holder coincides with the plasma expansion axis. The substrate was static at 43 mm from the target. All the experiments were performed at vacuum 1x10-4 Pa without any reactive atmosphere or substrate heating. The energy density values, or fluences, used to ablate the targets were chosen to be: 2.22, 2.83 and 4.10 J/cm2. In order to monitor the film deposition, in-situ reflectivity measurements of the films during growth were performed with a chopped diode laser (647 nm) using an incidence angle of 45° respect to the normal direction to the substrate surface. After the deposition of the films, ex-situ spectroscopic ellipsometry (SE) measurements were performed in the 300 nm to 1600 nm wavelength range at incidence angles of 60°, 65° and 70° using a VASE ellipsometer (J.A. Woollam Co., Inc.). The combination of the optical in-situ and ex-situ measurements was used first to obtain the deposition rates in the reference films and then to confirm the film thickness and optical linear properties for the prepared doped films. The deposits were achieved by the alternating ablation of SiAlON_101 (from Int. Syalons Co.) [10] and Er (99%) targets. All the films were designed to have the same total thickness and interlayered structure independent of the laser energy density used to grow them. On the SiAlON target 940, 620 and 480 pulses, for the 2.22, 2.83 and 4.10 J/cm2 energy densities respectively, were used to deposit on the substrate a 2.5 nm thick SiAlON layer. On the Er target only one pulse was used to generate the doping layer independent of the laser energy density. The procedure was repeated 60 times in order to form a film with a total thickness of about 150 nm. The composition of the samples was analyzed by Rutherford backscattering spectrometry (RBS) and X-ray photoelectron spectroscopy (XPS) measurements. The RBS spectra were obtained with 4 He at 2 MeV, at an incidence angle of 140°. The data were analysed with the standard code Ion Beam Analysis (IBA) Nuno’s DataFurnace (NDF) [15]. Transmission optical properties were obtained through a Varian Cary 5000 UV-Vis spectrophotometer. Silicon (100) and fused silica were used as substrates, to perform XPS/RBS and UV-Vis measurements, respectively. 2

Page 2 of 13

cr

ip t

Photoluminescence measurements were done under excitation at λ = 488 nm at a nominal power of 200 mW from an Ar+ ion laser (Spectra-Physics 2020-30). The light emitted by the sample was collected with a Czerny-Turner type Monochromator (Acton Spectra Pro 300i, with a diffraction grating of 300 g/mm for the IR range) and detected through a Hamamatsu H10330B-75 PMT. The signal was amplified with the standard lock-in technique and collected by a CPU. The lifetimes were obtained with the chopped signal at 20 Hz and a single exponential decay fitting. The produced samples were subjected to thermal annealing in air at atmospheric pressure from the as grown (25 °C) in steps of 100 °C for 1 hour beginning at 400 °C up to 700 °C and further steps of 50 °C for 1 hour up to 900 °C.

Ac ce

pt

ed

M

an

us

3. Results and discussion i) Compositional analysis. The results from the RBS spectra of the bulk ceramic SiAlON target and the films are presented in table 1. It should be noted that the SiAlON manufacturer did not provide any information on the target composition. It was found that the target in addition to the Si, Al, N and O, had some content (around 2%) of Ca and Y, which were probably used as sintering agents. The results from Table 1 show that the composition of the films is different from that of the original target, especially for the films deposited at the lower energy densities. This is a well know effect for the deposition of multi-component targets. The laser fluence has to be sufficiently high to induce ablation rather than pure evaporation from target, but a high fluence may lead to preferential self-sputtering and possibly implantation of the light atoms in the film [12,16]. In this work therefore we have chosen to start from a low energy density and increase it until a composition closer to that of the original target was found. Regarding to the RBS analysis of the films it should be noted first that the Si and Al contents are presented together, and second that their mass and Z (atomic number) are quite similar. Although for the bulk material it was possible to quantify the content, however for the films the quantification through the fitting of the RBS spectra was not able to yield reliable values for Al content below 3 at. % due to the overlapping of the Si and Al signals. However it was clear from the RBS analysis in a qualitative manner, due to the lack of a clear Al signal, that Al concentration of the films is lower than that of the target. In order to obtain quantitative information on the Al content, XPS analysis was performed on the film deposited at the higher energy density (4.10 J/cm2). The XPS yielded an Al content of 3 at. % and a Si content of 45 at. %. The sum of these two values agrees very well with the (Si+Al) value obtained from the RBS analysis that was 47.9 at. %. The RBS and XPS data are also in good agreement for the N content. Whereas, for the oxygen content determined by XPS is larger (12 at. %) than for RBS (8.6 at. %). The differences in these values are compatible within the error of the experimental techniques, so we can assume the oxygen content is around 10.0 at. %. Note that from XPS measurements the content of the heavy elements Er, Ca and Y could not be determined. The relative Si content in the films increased with respect to that in the target. This can be seen particularly in the films deposited at the higher fluence that shows a 47.9 at. % Si compared to 34.9 at. % in the target, an increase of over 10 %. This difference was even greater for the films deposited at lower fluences, but it was difficult to assess since it was not possible to differentiate the Si from the Al content in those films. The nitrogen content in the films is similar to that of the 3

Page 3 of 13

M

an

us

cr

ip t

target, especially for the film deposited at the highest fluence. In Figure 1 it has been plotted the compositional ratio of the elements in the films respect to that of the target as a function of the fluence. The compositional ratio was defined as [Nfilm - Ntarget]/[Ntarget], where Ntarget is the at. % of a given element on the target, and Nfilm the at. % of a given element in the film. From these results it can be seen that the films are deficient in nitrogen and oxygen compared to the original target composition. However as the fluence to ablate the target increases it can be seen that the nitrogen content increases only slightly, whereas there is a significant increase of the oxygen content. The Si+Al content in the films undergo a moderate decrease as the energy fluence increases. The graph shows that the films deposited at the higher fluences tend to have a composition closer to the one of the target. The increase of the oxygen ratio as a function of the fluence has been reported for multi-component oxides in the past [17]. Finally, in spite of the energy density increase of about a factor 2 (from 2.22 to 4.1 J/cm2) it should be noted that there is a negligible increase in the Er content in the films (from 0.06 to 0.07 %). The amount of ablated species does not increase linearly with the ablation fluence, thus probably we are close to the ablation threshold of the Er target [12]. From this analysis it is found that the film with a composition closer to that of the target is the one deposited at a higher fluence. From the RBS and XPS, the SiAlON composition can be described as Si4.45Al0.3N4O. This can be interpreted as it is composed basically by a mixture of silicon nitride (Si3N4) plus a silicon sub-oxide (Si1.45O) with a small amount of Al. However note that this small Al content is typical in many successful SiAlON based emitters, see for example ref.[18] where a green emitting SiAlON phosphor is reported with an Al content of only 1.334 %.

Ac ce

pt

ed

ii) Linear Optical Properties. Figure 2 (a) shows the transmission spectra in the visible and near infrared region measured both for the films which were grown at the highest and lower energy density values deposited on a fused silica substrate. As can be observed they are very similar. The spectra show the typical oscillations observed for a transparent dielectric layer due to interference effects. The transparency is high in the visible and infrared, with a band-edge in the UV. The complex refractive index (n=n+ik) was determined from the spectroscopic ellipsometry (SE) measurements for the films deposited on Si substrates. Figure 2 (b) shows the effective refractive index (n) and Figure 2 (c) displays the effective extinction coefficient (k). The SE parameters where fitted using a Cauchy dispersion law for the refractive index (n(λ) = A + B/λ2 + C/λ4), with A, B, and C being free parameters together with the total film thickness t. The obtained thickness values range from 155 to 164 ± 10 nm for the films deposited at 2.22 and 4.10 J/cm2 respectively, in good agreement with the originally designed thickness. The refractive index in the infrared range (at 1500 nm) is around 2.17 ± 0.02 for both films, which is significantly higher than that reported for silicon nitride films [7]. For longer wavelengths it is found that the refractive index of the film deposited at higher energies showed a higher refractive index. This result is most likely related to a densification effect, as the species arriving from material ablated at higher fluences have a higher kinetic energy [12]. The extinction coefficient was fitted with an exponential function. It was found that the samples show significant absorption for wavelengths shorter than 400 nm. The absorption edge was 4

Page 4 of 13

displaced toward longer wavelengths for the film deposited at the highest energy density. The reason for this is not understood at present and requires further investigation.

Ac ce

pt

ed

M

an

us

cr

ip t

iii) Er light emission. Photoluminescence (PL). Figure 3 shows the photoluminescence spectra obtained for the film deposited at the highest energy density as grown and after annealing at 850 °C. The Erbium (Er3+) ions embedded in the SiAlON matrix exhibited the characteristic 1.5 µm emission due to the radiative transition 4I13/2  4 I15/2 under λ = 488 nm excitation. The spectra shape is attributed to the Stark splitting of the degenerate 4f levels characteristic of Er3+ when embedded in a solid [19,20]. Note that the spectra show significant line broadening due to the amorphous nature of the matrix in which the Er-ions do not occupy well-defined sites [13,14], since the films were deposited at room temperature in vacuum, an amorphous structure was expected. It is remarkable that a measurable and distinct Er3+ emission was observed in the as grown sample prepared at room temperature, in spite of the fact that PLD film usually contain a large density of defects due to the high kinetic energy of the species arriving at the substrate. The defect density can be reduced significantly by annealing treatments, as has been shown in related works on Al2O3:Er [13,14] and silicon-rich oxynitrides doped with erbium [21]. In Figure 3 it can be observed that after the treatment at 850 °C the PL intensity increases by a factor 30, and the PL lifetime reached a value of 1.8 ± 0.2 ms. These values are similar to those reported for other Er-doped dielectric films such as Er-doped Al2O3 [13,14]. Figure 4 shows the evolution of the PL intensity as a function of the ablation energy density, both for the as-grown films and for those annealed at 850 °C. It can be seen that the PL intensity does not change as a function of the ablation fluence for the as-grown films. However, a significant increase of about a factor 2 is observed on the PL of the annealed samples. This result suggests that after annealing there is an Er-activation process that depends on the original composition of the coatings. When we look at the composition of the films as a function of the ablation energy density it is found that the increase of oxygen is the most significant. Indeed, the state of oxidation of the Er is an important factor to determine its activation as it has been demonstrated in the case of Erdoped silicon films [22,23,24]. 4. Conclusions Thin films of SiAlON doped with Er were successfully deposited and characterized. The SiAlON films with the best performance were those grown at the highest laser ablation fluence. These films show a high refractive index (n = 2.17 at 1500 nm) and, an oxygen and nitrogen content (8.6 and 41.1 at. %, respectively) close to that of the target (10.0 and 43.5 at. %, respectively). The composition can be described as Si4.45Al0.3N4O, showing a composition similar to that of the target except for the Al content that has been significantly reduced. The Er-doped SiAlON films exhibited NIR emission from the Er3+ ions centered in 1.5 µm, in a typical amorphous environment, even for the untreated as grown samples. These results might suggest that an in situ procedure could reduce post-annealing treatments, e.g., substrate bias and/or heating [25].

5

Page 5 of 13

ip t

5. Acknowledgments This work has been financially supported by the Spanish Ministry of Economy and Competitiveness through the project TEC2012_38901-C02-01 and -02, project TEC2012-38540-C02-02. I.C. acknowledges the financial support through JAE-Pre-2011_00578. A.M. acknowledges the financial support through BES-2013-062593. J.M.R. acknowledges the financial support of Secretariat for Universities and Research of Generalitat de Catalunya through the program FI-DGR 2013. 6. References

Ac ce

pt

ed

M

an

us

cr

[1] S. Cueff, C. Labbé, L. Khomenkova, O. Jambois, P. Pellegrino, B. Garrido, C. Frilay, R. Rizk, Mater. Sci. Eng. B 177 (10), (2012) 725-728. [2] Y. Berencen, Josep Carreras, O. Jambois, J. M. Ramírez, J. A. Rodríguez, C. Domínguez, Charles E. Hunt, B. Garrido, Opt. Express 19 (2011) A234. [3] R.-J. Xie and N. Hirosaki, Sci. Technol. Adv. Mater. 8 (2007) 588. [4] S. Feng, T. Lei, H. Chen, H. Cai, X. Luo, A.W. Poon, Laser Photon. Rev. 6 (2), (2012) 145-177. [5] G.T. Reed, G. Mashanovich, F. Y. Gardes, D. J. Thomson, Nature Photon. 4 (2010) 518-526. [6] L. Feng, Z. Liu, Mater. Sci. Eng. B 122 (1), (2005) 7-11. [7] V.M. Bermudez, F.K. Perkins, App. Surf. Sci. 235 (4), (2004) 406-419. [8] A. Rosenflanz, Curr. Opin. Solid State Mat. Sci. 4 (5), (1999) 453-459. [9] V.A. Izhevskiy, L.A. Genova, J.C. Bressiani, F. Aldinger, J. Eur. Ceram. Soc. 20 (13), (2000) 2275-2295. [10] International Syalons (Newcastle), “Syalon 101 – An Advanced Silicon Nitride Ceramic” (June 2014) http://www.syalons.com/materials/syalon101/ [11] G. Z. Cao, R. Metselaar, Chem. Mater. 3 (1991) 242-252. [12] C.N. Afonso, J. Gonzalo, R. Serna, J. Solís, Pulsed laser deposition for functional optical films, in “Laser Ablation and its Applications” Ed. by C. Phipps (Springer, USA, 2007) Ch. 13, pp. 315-338. [13] R. Serna, M. Jiménez de Castro, J. A. Chaos, C.N. Afonso, I. Vickridge, Appl. Phys. Lett. 75 (1999) 40734075. [14] A. Suárez-García, R. Serna, M. Jiménez de Castro, C.N. Afonso, I Vickridge, Appl. Phys. Lett. 84 (2004) 2151-2153. [15] N.P. Barradas, C. Jeynes, Nucl. Instrum. Methods Phys. Res. Sect. B 266 (2008) 1875. [16] J. Shou, Appl. Surf. Sci. 255 (2009) 5191. [17] J. Gonzalo, C. N. Afonso, J. Perriere, J. Appl. Phys. 79 (1996) 8042. [18] N. Hirosaki, R.-J. Xie, K. Kimoto, T. Sekiguchi, Y. Yamamoto, T. Suehiro, and M. Mitomo, Appl. Phys. Lett. 86 (2005) 211905. [19] A. Polman, J. Appl. Phys. 82 (1997) 1-39. [20] W.J. Miniscalco, Optical and electronic properties of rare-earth ions in glasses, in: RareEarth doped fiber lasers and amplifiers, Ed. by M.J.F. Digonnet (Dekker, New York, 1993). [21] Z. Lin, R. Huang, Y. Guo, C. Song, Z. Lin, Y. Zhang, X. Wang, J. Song, H. Li, X. Huang, Opt. Mater. Express 4 (2014) 816-822. [22] R. Serna, E. Snoeks, G.N. Van den Hoven, A. Polman, J. Appl. Phys. 75 (1994) 2644-2647. [23] R. Serna, J.H. Shin, M. Lohmeier, E. Vlieg, a. Polman, P.F. A. Alkemade, J. Appl. Phys. 79 (1996) 2658. [24] A. J. Kenyon, Prog. Quantum Electron. 26 (2002) 225-284. [25] M. Jiménez de Castro, A. Suarez-García, R. Serna, C. Afonso, J. García-López, Opt. Mater. 29 (2007) 539542.

6

Page 6 of 13

Figure Captions Table 1. Composition of bulk and the SiAlON produced with different ablation fluences. * Al at. % < 3 couldn’t be quantified by RBS measurements for the films.

ip t

Figure 1. Percentage element ratio for Si+Al, N and O in the thin films respect to the target, as a function of the fluence used in the ablation process.

cr

Figure 2. a) Transmission, b) refractive index, c) extinction coefficient and the inset d) zoom in on the region of the absorption edge, for samples grown with fluences of 2.22 and 4.10 J/cm2.

Ac ce

pt

ed

M

an

Figure 4. PL Intensity as a function of ablation laser fluence.

us

Figure 3. NIR emission due to the radiative 4I13/2  4I15/2transition, after the excitation with λ = 488 nm, pumped at a nominal power of 200 mW. As grown sample and after annealing at 850 °C, deposited at 4.10 J/cm2.

7

Page 7 of 13

*Highlights (for review)

Highlights -

cr

ip t

-

PLD production of Er-doped thin films from a low cost commercial SiAlON target The role of the ablation fluence on the composition, optical properties as well as on the light emission performance at 1.5 µm. The optimized performance is obtained for the samples deposited at the higher used ablation energy density. Further improvement was achieved through annealing.

us

Keywords

Ac ce p

te

d

M

an

PLD, Pulsed laser deposition, SiAlON, silicon oxynitride, PL, photoluminescence, erbium, Er, ellipsometry, refractive index.

Page 8 of 13

Figure 1

30

Si+Al

20 10 Target

-10

ip t

N

-20 -30

O

-40

cr

Element ratio film to target (%)

0

-50

-70 2.0

2.5

3.0

us

-60

3.5 2

4.0

4.5

an

Fluence (J/cm )

Ac ce p

te

d

M

Figure 1

Page 9 of 13

Figure 2

600

800

1000 1200 1400 1600

(a)

80 60 40

(b)

As Grown Samples

2.30

2.22 J/cm

2.25

4.10 J/cm

2 2

us

2.20 2.15 0.08

(c)

(d)

0.06

0.08

0.04

k

k

0.06

0.02

0.04

0.00

0.02

450

500

M

400

an

0.10

ip t

20 0 2.35

n

400

cr

Transmission (%)

200 100

550

600

650

Wavelength (nm)

0.00 200

400

600

800

1000 1200 1400 1600

Ac ce p

te

Figure 2

d

Wavelength (nm)

Page 10 of 13

Figure 3

3.0 2

F = 4.10 J/cm As Grown Ann. @ 850 ºC

2.0

x10

1.5 1.0 0.5 0.0 1450

1500

1550

1600

1650

Wavelength (nm)

Ac ce p

te

d

M

an

us

Figure 3

1700

cr

1400

ip t

PL_Intensity (arb. units)

2.5

Page 11 of 13

Figure 4

3.0

As Grown Ann. @ 850 ºC

2.5

1.5 0.3 0.0 2.0

2.5

3.0

3.5

4.0

2

Fluence (J/cm )

Ac ce p

te

d

M

an

us

Figure 4

4.5

ip t

2.0

cr

PL_Intensity (arb. units)

exc = 488 nm

Page 12 of 13

Table 1

ip t

Fluence Si + Al* N O Ca + Y Er (J/cm2) (at. %) (at. %) (at. %) (at. %) (at. %) Bulk Target 34.9 + 10.0 43.5 10.0 1.7 --RBS SiAlON_101 2.22 55.3* 38.9 3.4 2.4 0.06 2.83 54.2* 38.6 4.9 2.4 0.06 4.10 47.9* 41.1 8.6 2.1 0.07 Fluence Si + Al N O Ca + Y Er (J/cm2) (at. %) (at. %) (at. %) (at. %) (at. %) XPS 4.10 45 + 3 40 12 ----NOTE: Percentage errors in RBS on the at. % determination are 5% for Si+Al and N; 16 % for Ca, Y and Er; and finally 40% for O since is a light element with a smaller crosssection. The estimated error in the XPS values determination is ± 2 at. %.

Ac ce p

te

d

M

an

us

cr

Table 1

Page 13 of 13