Hard X-ray Photoelectron Spectroscopy (HAXPES) characterisation of electrochemical passivation oxide layers on Al–Cr–Fe complex metallic alloys (CMAs)

    Hard X –Ray Photoelectron Spectroscopy (HAXPES) characterisation of electrochemical passivation oxide layers on Al-Cr-Fe Complex Metallic Alloys (CMA) Alessandra Beni, No´emie Ott, Magdalena Pawelkiewicz, Micheline Ward´e, Kirsty Young, Birgitta Bauer, Parasmani Rajput, Blanka Detlefs, J¨org Zegenhagen, Ronan McGrath, Marie-Genevi`eve Barth´es-Labrousse, Lars P.H. Jeurgens, Patrik Schmutz PII: DOI: Reference:

S1388-2481(14)00156-8 doi: 10.1016/j.elecom.2014.05.024 ELECOM 5169

To appear in:

Electrochemistry Communications

Received date: Revised date: Accepted date:

12 May 2014 16 May 2014 20 May 2014

Please cite this article as: Alessandra Beni, No´emie Ott, Magdalena Pawelkiewicz, Micheline Ward´e, Kirsty Young, Birgitta Bauer, Parasmani Rajput, Blanka Detlefs, J¨ org Zegenhagen, Ronan McGrath, Marie-Genevi`eve Barth´es-Labrousse, Lars P.H. Jeurgens, Patrik Schmutz, Hard X –Ray Photoelectron Spectroscopy (HAXPES) characterisation of electrochemical passivation oxide layers on Al-Cr-Fe Complex Metallic Alloys (CMA), Electrochemistry Communications (2014), doi: 10.1016/j.elecom.2014.05.024

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.

ACCEPTED MANUSCRIPT Hard X –Ray Photoelectron Spectroscopy (HAXPES) characterisation of electrochemical passivation oxide layers on Al-Cr-Fe Complex Metallic Alloys

RI

PT

(CMA)

SC

Alessandra Beni1*, Noémie Ott1, Magdalena Pawelkiewicz1, Micheline Wardé2, Kirsty Young3,

NU

Birgitta Bauer4, Parasmani Rajput5, Blanka Detlefs5, Jörg Zegenhagen5, Ronan McGrath3, Marie-

1

MA

Geneviève Barthés- Labrousse2, Lars P. H. Jeurgens1, Patrik Schmutz1

Empa, Swiss Federal Laboratories for Materials Science & Technology, Ueberlandstrasse 129,

8600, Duebendorf, Switzerland.

ICMMO, CNRS UMR 8182 Université Paris Sud, 91404, Orsay Cedex, France.

3

Department of Physics, University of Liverpool, Liverpool L69 3BX, United Kingdom.

4

Department of Earth and Environmental Sciences, Ludwig Maximilians University,

TE

D

2

5

AC CE P

Theresienstrasse 41, 80333 Munich, Germany. European Synchrotron Radiation Facility, ESRF, 6 rue Jules Horowitz, 38043 Grenoble Cedex,

France.

*Corresponding author

E-mail addresses: [email protected] (A. Beni), [email protected] (P. Schmutz) Telephone: +41 (0)58 765 4845.

Abstract

A Hard X-Ray Photoelectron Spectroscopy (HAXPES) characterisation of the passivation layers formed by electrochemical polarisation of Al-Cr-Fe complex metallic alloys is presented. By employing X-ray excitation energies from 2.3 to 10.0 keV, the depth distributions of Al- and Cr1

ACCEPTED MANUSCRIPT oxide and hydroxide species in the (Al,Cr)-passive layers could be determined. Simultaneous analyses of the shallow Al 2s and deep Al 1s core level lines (respectively, more bulk- and

PT

surface-sensitive) provided complementary information to effectively determine the depth-

RI

resolved contributions of hydroxide and oxide species within the passivation layer. A Cr

NU

SC

threshold concentration of 18 (at %) was found for effective passivation at pH 1.

MA

Keywords

Electrochemical passivation, Hard X-Ray Photoelectron Spectroscopy (HAXPES), Al alloys,

AC CE P

TE

D

Complex Metallic Alloys, composition depth profiling, oxide/hydroxide.

1. Introduction

Al-Cr-Fe complex metallic alloys (CMA) are intermetallic phases occurring as crystalline approximants of quasicrystals [1, 2] and present peculiar bulk [3] and surface properties, such as low friction, hydrophobicity and high corrosion resistance in the 0-14 pH range [4, 5], being good candidates as multifunctional corrosion protecting coatings. Their corrosion resistance in very aggressive environments, such as acidic sulphuric or chloride containing electrolytes, has been proven by electrochemical techniques [4, 5]. The formation of a 6-8 nm thin protective passive layer composed of several Al and Cr oxyhydroxide stacks was evidenced by X-ray Photoelectron Spectroscopy (XPS) [4, 6] and complementary XPS and Auger sputtering depth profile analyses, which are typically used to determine the depth-resolved chemical constitution of layers with a 2

ACCEPTED MANUSCRIPT thickness exceeding the XPS information depth. However, the main disadvantages of the latter techniques are the possible preferential sputtering effects and the need for an accurate calibration

PT

of the sputtering rate [7]. Investigation of the oxidation mechanism not only requires detailed

RI

information on the oxide constitution, but also on the oxidation-induced changes in the alloy adjacent to the alloy/oxide interface [8]. Unfortunately, the information depth obtained by angle-

SC

resolved XPS analysis using a conventional Al- or Mg- X-ray source is limited up to about 6 nm

NU

and therefore cannot probe the alloy/oxide interface for ‘thicker’ passive oxide layers, as formed by e.g. electrochemical polarisation [10]. Furthermore, the energy resolution of standard

MA

laboratory set-ups is typically insufficient to unequivocally resolve the oxide and hydroxide components in measured core-level spectra of oxidised alloys [9].

D

The present study demonstrates that Hard X-ray XPS (HAXPES) is a very powerful

TE

technique to characterise relatively thick, multi-element oxyhydroxide layers on Al-Cr-Fe CMAs

AC CE P

as formed by potentiostatic anodic polarisation. By tuning the energy of the incident X-rays up to 10 keV [10, 11], HAXPES can effectively probe the chemical constitution as a function of depth over the entire passive layer and across the reacting alloy/oxide interface. In addition, the use of a synchrotron source with monochromatised X-ray optics provides an enhanced energy resolution as compared to conventional set-ups, thus allowing simultaneous analysis of shallow and deep core-level photoelectron lines (like Al 2s and Al 1s), thereby opening the field to new physics. As demonstrated in this work, use of high-energy resolution at various excitation energies is essential for identifying the chemical environments in the multi-element oxyhydroxide layers formed by electrochemical treatments for which a direct comparison with known reference oxide and/or hydroxide bulk compounds is not possible.

3

ACCEPTED MANUSCRIPT 2. Experimental Details 2.1 Materials and electrochemical treatment

PT

Four Al-based CMA intermetallics; the orthorhombic Al78.8Cr15.5Fe5.7 and Al79.1Cr17.8Fe3.1, the

RI

hexagonal Al78.7Cr20.6Fe0.7 and the cubic Al64.2Cr27.2Fe8.1 (at %) phases were investigated.

SC

Corresponding single crystals were prepared at LMU (Munich) [12]. The polycrystalline cubic phase was prepared at LSGMM (Nancy).

NU

All the samples were polished to mirror finish using diamond paste and ethanol and

MA

electrochemical anodically potentiostatic polarised in the low passive domain at 0VSCE for 24 h in

2.2 HAXPES measurements

TE

D

sulphuric acid electrolytes at pH 0 or 1, using a conventional three electrode cell.

AC CE P

The measurements were conducted at the ID32 beamline of the ESRF. The excitation energy was varied from 2.3 to 10.0 keV. A double-crystal Si(111) monochromator and a PHOIBOS 225 HV hemispherical electron analyser [13] were used to record HAXPES spectra at various photoelectron take-off angles. High-resolution measurements were performed using additional Si(333) and Si(555) post-monochromators. The binding energy (BE) was calibrated using gold. CasaXPS (2009 Casa Software Ltd) was used for data analysis, applying a Shirley-type background and asymmetric lineshapes for the metallic peaks and symmetric Gaussian/Lorentzian lineshapes for the oxidic ones (See the Appendix for the details). The lineshapes of the metallic peak components were determined from the spectra of clean alloys. Due to the variation of the energy resolution with the excitation energy (different source/analyser resolution, use of post-monochromators), the lineshape parameters can be different for different excitation energies. 4

ACCEPTED MANUSCRIPT 3.

Results and Discussion

Fig. 1 displays exemplary Al 2s, Al 1s and Cr 2p3/2 spectra recorded at 6.0 and 10.0 keV for the

PT

Al78.7Cr20.6Fe0.7 alloy after two different surface treatments, but identical electrochemical potentiostatic polarisations in sulphuric acid at pH 1: see caption for details. For all specimens

RI

studied, no oxidised state of Fe could be detected in the Fe 2p3/2 core-level spectra, indicating that

SC

Fe does not contribute to the formation of the passivation layer.

NU

At least two oxidic components were identified in the Al 2s spectra positioned at shifts of +1.7 and +2.7 (± 0.1) eV with respect to the metallic peak, as attributed to Al-oxide-like and Al-

MA

hydroxide-like environments, respectively. Such components were also identified in the Al 1s spectra at relative chemical shifts of +1.6 and +2.5 (± 0.1) eV. Figs. 1a,b shows that the Al 1s

D

core-level, as positioned at much higher BE, provides an enhanced surface sensitivity as

TE

compared to the Al 2s core level. Due to the enhanced degree of hydroxylation near the surface,

AC CE P

the Al 1s spectra thus show a more pronounced contribution of the Al-hydroxide component. The Cr 2p3/2 oxidic spectra (Fig. 1c,d) can also be described by at least one oxide-like and one hydroxide-like component with a respective chemical shift of +2.5 and +3.6 (± 0.1) eV, in accordance with Ref. [14].

In summary, the HAXPES analysis evidences different local environments (i.e. oxide and hydroxide) for the Al and Cr cations in the defective oxyhydroxide layers formed on the Al-Cr-Fe complex metallic alloys after electrochemical treatment in sulphuric acid. The high degree of local disorder of the oxyhydroxide layers modifying the local coordination spheres around Al and Cr cations in the oxide- and hydroxide-like environments (e.g. bond lengths and angles, as well as ligand-type) , results in relative large intrinsic widths and chemical shifts of the fitted oxide and hydroxide components.

5

ACCEPTED MANUSCRIPT

80 60 40

20

20

0 124

120

118

116

114

Al 1s 6 keV

0 124

1000

NU

600 400

x10

3

122

RI

40

CPS

CPS

60

Al 2s 10 keV

PT

b)

Al 2s 80 6 keV

SC

a)

122

120

118

116

114

Al 1s 10 keV

500

MA

200 0

0 1566 1564 1562 1560 1558 1556

1566 1564 1562 1560 1558 1556

CPS

200

100

0

AC CE P

Cr 2p3/2 6 keV

580

578

576

574

Binding energy (eV)

Binding energy (eV)

d)

Cr 2p3/2 10 keV 30

CPS

c)

TE

D

Binding energy (eV)

20 10 0

572

580

578

576

574

572

Binding energy (eV)

Fig. 1. Reconstructed high-resolution Al 2s, Al 1s and Cr 2p3/2 spectra for the Al78.7Cr20.6Fe0.7 alloy after metallographic polishing without (a,c) and with (b,d) sputter/annealing and subsequent electrochemical polarisation in sulphuric acid (pH 1) for 24 hours at 0VSCE, as recorded at 6 and 10 keV, respectively.

6

ACCEPTED MANUSCRIPT b)

122

120

118

116

0 582

114

Binding energy (eV)

3

5

0

580

578

e)

Cr 2p3/2 3.8 keV 1000

200

500

0 116

580

578

576

574

50 60 70 take off angle (deg)

530

528

O 1s 3.8 keV

1000

536

CrOx/Al2sOx (2.3 keV) CrOx/Al1sOx (2.3 keV)

Ratio

40

AC CE P

0.6

30

2000

i)

2.0

532

0

h)

D

(2.3 keV) (10.0 keV) (2.3 keV) (10.0 keV)

534

Binding energy (eV)

572

1.5

0.2

536

Binding energy (eV)

0.8

0.4

f

IOx/IMet

1.0

582

114

Binding energy (eV)

CrOx/Al1sOx CrOx/Al1sOx CrOx/Al2sOx CrOx/Al2sOx

1.2

Ratio

118

TE

g)

120

MA

100 0

572

NU

CPS

CPS

300

122

574

Binding energy (eV)

Al 2s 3.8 keV

124

576

SC

124

O 1s 2.3 keV

10

RI

0

400

x10

4000

2000

1000

d)

15

CPS

2000

CPS

CPS

CPS

CPS

3000

c)

Cr 2p3/2 2.3 keV

6000

PT

Al 2s 2.3 keV

CPS

a)

40

534

532

530

Binding energy (eV)

528

Al 1s 2.3 keV Al 2s 2.3 keV Al 1s 6.0 keV Al 2s 6.0 keV Al 1s 10.0 keV Al 2s 10.0 keV

35

6 4

1.0

2 0.5

0 10

20

30 40 50 60 70 take off angle (deg)

18

20 27 Cr content (at%)

Fig. 2. a)-f) Reconstructed Al 2s, Cr 2p3/2 and O 1s spectra for the electrochemically polarised Al64.2Cr27.2Fe8.1 alloy in sulphuric acid (pH 0) after 24 hours at 0VSCE , as recorded at 45º take-off angle. g-i ) Corresponding intensity ratios at different excitation energies and electron take-off angles for the alloys after polarisation in sulphuric acid for 24 hours at 0VSCE: g) Cr20% alloy (pH 1), h) Cr27% alloy (pH 0) and i) Al intensity ratios as a function of the alloy Cr content (pH 1). The estimated error can amount to ±10%.

Evaluation of the HAXPES spectra as a function of the excitation energy allowed a nondestructive, depth-resolved determination of the chemical constitution of the passivation layers. 7

ACCEPTED MANUSCRIPT The Al 2s and Cr 2p3/2 spectra (Fig. 2, a-f) recorded at 2.3 keV and 3.8 keV for the Al64.2Cr27.2Fe8.1 alloy confirm the incorporation of both Al and Cr in the grown oxyhydroxide

PT

passivation layers with an enhanced degree of hydroxylation towards the outer surface. For low

RI

excitation energies (i.e. more surface-sensitive), an additional minor component appears in the Al 2s and Cr 2p spectra, which is attributed to the residues of sulphate species from the electrolyte

SC

solution interacting with the passive layer outer surface. The O 1s core-level could be described

NU

by three components at 534.5 eV, 532.7 and 531.5 eV, as attributed to Al-hydroxide-like, overlapping Al-oxide/Cr-hydroxide-like and Cr-oxide-like local chemical environments,

MA

respectively [15]. Unfortunately, due to the large intrinsic width of the O 1s line, an increased energy resolution could not resolve the individual oxide and hydroxide components in the O 1s

D

envelope in presence of so different Al and Cr cations.

TE

Passivation of the Al-Cr-Fe alloys is ensured by the formation of the (Al,Cr)-

AC CE P

oxyhydroxide layer, as outlined above. The Croxide/Aloxide intensity ratios (summing the oxide and hydroxide components) evidence that the oxyhydroxide layers are richer in Cr cations towards their outer surface (Fig. 2g,h). The average thicknesses of the passivation layers for the electrochemically treated Al-Cr-Fe phases can then be estimated by assuming a simplified bilayer structure [16], consisting of an Al-oxide bottom layer and a Cr-oxide top layer on the alloy substrate. Table 1 reports thickness estimations using the intensity attenuations (using atomic densities of 51, 47 and 41 at/nm3 for the alloy, Al2O3 and Cr2O3, respectively [9]) of the HAXPES Cr 2p3/2 and Al 2s signals recorded at 6 keV where the metallic and oxidic compounds are well defined. As shown in Fig. 2i), the lower the Cr content, the thicker the oxyhydroxide layer and hence the higher the ionic mobility (i.e. the poorer the corrosion resistance). The Cr 17.8 and 20.6 at% phases present similar oxide thicknesses, confirming that the Cr threshold for the formation of a barrier-type passivation layer in sulphuric acid (pH 1) is correlating with a 8

ACCEPTED MANUSCRIPT minimum Cr content > 18 at% in the alloy, in accordance with previously published

Al oxide

(at%)

calculated

calculated

thickness (Å)

thickness (Å)

Al78.8Cr15.5Fe5.7

21

110

Al79.1Cr17.8Fe3.1

10

44

56

MA

30

D

9

AC CE P

Calculated IMFP (Å) in Al2O3

2.3 keV

78

TE

Al78.7Cr20.6Fe0.7

Angle (deg)

SC

Cr oxide

NU

Composition

RI

PT

electrochemical potentiostatic current transient measurements [5].

10 keV

Al 1s

9.7

Al 1s

69.0

Al 2s

22.5

Al 2s

79.0

Cr 2p 18.6

Cr 2p 75.8

Al 1s

18.9

Al 1s 134.0

Al 2s

43.6

Al 2s 153.2

Cr 2p 36.1

Cr 2p 147.1

Table 1. Oxide thickness estimation using equations in [16] and IMFP in Al2O3 at various excitation energies and angles according to [17].

4.

Conclusions

HAXPES was used to perform a non-destructive depth profile analysis of the passivation layers formed by electrochemical polarisation of Al-Cr-Fe complex metallic alloys in very acidic electrolytes (sulphuric acid pH 1 or 0). Using Hard X-rays, deep core-level photoelectron lines which cannot be excited with conventional laboratory sources were collected (i.e. Al 1s). The 9

ACCEPTED MANUSCRIPT shallow and deep core levels provided complementary information: i.e. the Al 1s signal allowed an enhanced surface sensitivity. Analysis and cross correlation of Al 1s, Al 2s, Cr 2p and O 1s

PT

spectra recorded at progressively increasing X-Ray excitation energies (from 2.3 to 3.8 keV) and

RI

enhanced energy resolution revealed the distributions of Al and Cr cations in hydroxide-like and oxide-like local chemical environments as a function of depth below the surface and indicate a

SC

relative enrichment of Cr cations towards the outer passive layer surface.

NU

Higher excitation energies (6-10 keV) were utilised to probe the interface (i.e. to record a reliable signal from the alloy substrate), thus allowing an estimation of the thickness of the passivation

MA

layers. As a final result, a Cr threshold of 18 at % was identified, necessary for the formation of a

AC CE P

TE

D

thin protecting barrier type passive oxyhydroxide layer on the Al-Cr-Fe complex metallic alloys.

10

ACCEPTED MANUSCRIPT Appendix Designation

BE

FWHM

(± 0.1 eV)

(± 0.1 eV)

84.0

0.6

Al 2s

Metal

118.1

1.1

Al 2s

Al oxide

119.8

1.6

Al 2s

Al hydroxide

120.8

1.6

Al 2s

Al sulphate

122.4

Al 1s

Metal

1559.8

Al 1s

Oxide

1561.4

Al 1s

Hydroxide

1562.4 1564.5

Cr 2p3/2

Cr2O3

Cr 2p3/2

Cr(OH)3

Cr 2p3/2

A(0.001,1,10)SGL(80) SGL(30)

SGL(30)

1.5

SGL(100)

1.1

SGL(50)

1.7

SGL(30)

1.7

SGL(30)

SC

Metal

1.8

SGL(50)

574.0

1.2

A(0.4,0.5,0)*SGL(20)

576.5

2.6

SGL(30)

577.6

2.6

GL(30)

CrO3 or Cr sulphate

579.5

2.0

SGL(30)

Fe 2p3/2

metal

706.9

0.6

A(0.5,1,10)SGL(40)

O 1s

Cr oxide

531.5

1.5

GL(30)

NU

Al sulphate

MA

Al 1s Cr 2p3/2

A(0.1,0,400)SGL(100)

RI

Metal

Au4f7/2

Lineshape CasaXPS

PT

Core Level

hydroxide

532.7

2.2

SGL(20)

O 1s

TE

D

Al oxide and Cr O 1s

534.5

1.5

SGL(30)

(overlapping)

AC CE P

Al hydroxide

Table A1: Chemical assignment of the fitted components in the passivation layers on the Al64.2Cr27.2Fe8.1

alloy after electrochemical treatment in sulphuric acid at pH0 (excitation energy 2.3 keV, no postmonochromators used).

11

ACCEPTED MANUSCRIPT Acknowledgements

PT

The work was developed within the framework of the European integrated Centre for the

RI

Development of new Metallic Alloys and Compounds ‘C-MAC’ (http://www.eucmac.eu). The authors are grateful to Prof Gille (LMU Munich) and M. de Weerd (LSGMM Nancy) for

SC

sample preparation and thank Dr Maehl and Dr Thiessen (SPECS GmbH) for providing the

AC CE P

TE

D

MA

NU

simulated relative transmission factors of the PHOIBOS 225 analyser.

12

ACCEPTED MANUSCRIPT References

AC CE P

TE

D

MA

NU

SC

RI

PT

[1] J.M. Dubois, Useful Quasicrystals, World Scientific, Singapore, 2005. [2] R. Gaspari, R. Erni, Y. Arroyo, M. Parlinska-Wojtan, J. Dshemuchadse, C.A. Pignedoli, D. Passerone, P. Schmutz, A. Beni, Journal of Applied Crystallography, 47 (2014), doi:10.1107/S1600576714008656. [3] J. Dolinsek, P. Jeglic, M. Komelj, S. Vrtnik, A. Smontara, I. Smiljanic, A. Bilusic, J. Ivkov, D. Stanic, E.S. Zijlstra, B. Bauer, P. Gille, Physical Review B, 76 (2007) 174207-174213. [4] E. Ura-Binczyk, N. Homazava, A. Ulrich, R. Hauert, M. Lewandowska, K.J. Kurzydlowski, P. Schmutz, Corrosion Science, 53 (2011) 1825-1837. [5] A. Beni, N. Ott, E. Ura-Binczyk, M. Rasinski, B. Bauer, P. Gille, A. Ulrich, P. Schmutz, Electrochimica Acta, 56 (2011) 10524-10532. [6] N. Ott, A. Beni, A. Ulrich, C. Ludwig, P. Schmutz, Talanta, 120 (2014) 230. [7] S. Hofmann, Auger- and X-Ray Photoelectron Spectroscopy in Materials Science, 2013. [8] E. Panda, L.P.H. Jeurgens, E.J. Mittemeijer, Corrosion Science, 52 (2010) 2556-2564. [9] J.K. Parle, A. Beni, V.R. Dhanak, J.A. Smerdon, P. Schmutz, M. Wardé, M.G. Barthés-Labrousse, B. Bauer, P. Gille, H.R. Sharma, R. McGrath, Applied Surface Science, 283 (2013) 276-282. [10] J. Zegenhagen, C. Kunz, Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 547 (2005) V. [11] J. Rubio-Zuazo, G.R. Castro, Surface and Interface Analysis, 40 (2008) 1438-1443. [12] B. Bauer, P. Gille, Zeitschrift Fur Anorganische Und Allgemeine Chemie, 637 (2011) 2052-2058. [13] J. Zegenhagen, B. Detlefs, T.-L. Lee, S. Thiess, H. Isern, L. Petit, L. André, J. Roy, Y. Mi, I. Joumard, Journal of Electron Spectroscopy and Related Phenomena, 178–179 (2010) 258-267. [14] W. Fredriksson, S. Malmgren, T. Gustafsson, M. Gorgoi, K. Edström, Applied Surface Science, 258 (2012) 5790-5797. [15] J. van den Brand, W.G. Sloof, H. Terryn, J.H.W. de Wit, Surface and Interface Analysis, 36 (2004) 8188. [16] M. Liu, S. Zanna, H. Ardelean, I. Frateur, P. Schmutz, G.L. Song, A. Atrens, P. Marcus, Corrosion Science, 51 (2009) 1115-1127. [17] S. Tanuma, C.J. Powell, D.R. Penn, Surface and Interface Analysis, 43 (2011) 689-713.

13

ACCEPTED MANUSCRIPT

PT

Highlights

SC

RI

• HAXPES characterisation of electrochemical passivation layers for Al-Cr-Fe CMA’s.

NU

• Oxide and hydroxide signals identified for Al 2s/Cr 2p photoelectron core levels.

MA

• Shallow and deep core levels analysed to achieve diverse surface sensitivity.

TE

D

• Cr 18% at concentration threshold necessary for stable passivation at pH 1 H2SO4.

AC CE P

• X-ray energy variation for cations and hydroxide/oxide distribution identification.

14