Figure 1: The faceted topography of magnetron sputtered aluminium on glass
RMS roughness = 6 nm.
|
The performance of coated and bonded aluminium is dependent, at least in part, on its surface chemistry. There have been numerous publications in which XPS has used to determine the chemistry and thickness of the oxide film at the aluminium surface. It is apparent that an agreement exists on the method to calculate the oxide thickness from the clearly resolved Al2p metal and oxide components.1 However, there is no consensus on the approach for fitting the oxide and hydroxide components of the O1s, indeed some suggest that it is not advisable.2 Recently, a method based on defining the separation between the O1s and Al2p components has been developed using an oxyhydroxide standard.3
Here, we apply this curve fitting method to determination of the hydroxyl concentration of the film formed at the surface of magnetron sputtered aluminium (99.999%). Importantly, this sample allows us to reliably define the age of the surface film and allows control of the alloying elements. Future work will use this material to assess the effect of plasma etching upon the surface chemistry of the aluminium surface, hence an understanding of the effect of ambient storage on the surface chemistry is important.
|
 |
-
- Aluminium was deposited in an argon plasma at a power of 50 W from a magnetron sputter target for 10 minutes to give an aluminium deposit of approximately 30 nm thickness. No substrate (glass) was observed in any of the spectra. A base pressure of 2.5x10-7mbar was obtained prior to the introduction of argon (flow = 10sccm and pressure= 5.2x10-3mbar). To minimise the through film hydration, the deposition chamber was vented to oxygen prior to opening to the ambient atmosphere. This was defined as zero time. The substrate was not observed to increase above the ambient temperature during
deposition.
Each XPS analysis was carried out on a separate sample using a Kratos AXIS Ultra, which employs a magnetic immersion lens to increase the solid angle of photoelectron collection at small analysis areas and to minimise the aberrations of the electron optics. All data were acquired using monochromated Al ka X-rays and processed with Vision 2 software.
|
Figure 2: Al2p and O1s core levels from a magnetron sputtered aluminium exposed to the atmosphere for 23 min.
Figure 3: Proportion of O1s of component peaks against atmospheric exposure time.
|
Aluminium was deposited onto glass cover slides producing a mirror finish. The aluminium surface topography was determined by AFM to be dominated by crystal facets of ca. 100 nm lateral size (Figure 1). This morphology was unchanged after ambient storage over the periods considered in this work.
The Al2p core level acquired from these samples comprised a metallic component, within which the spin splitting of the core level may be resolved, and a broader oxide component to higher binding energy (Figure 2). To provide a satisfactory fit to the envelope, a third component was required. This component is tentatively assigned to aluminium atoms at the oxide-metal interface.
Using the methodology developed from analysis of an oxyhydroxide standard, the O1s peak may be fitted to quantify the oxide (O-1), hydroxide (O-2) and water/carbonaceous oxygen environments (O-3).3 The lack of distinct features in the O1s spectra requires that to fit these data, the difference in binding energy between the O1s components, O-1, O-2 and O-3 and the oxide component of the Al2p peak be constrained. The values used were 456.6 ±0.1 eV, 458.0 ± 0.1 eV and 460.2 ±0.2 eV respectively. This is discussed in detail in ref. 3.
Using this curve fitting approach, the O1s core level of samples after a range of atmospheric exposure times revealed the gradual increase of the proportion of hydroxide ions in the surface film (Figure 3). It is apparent that a high proportion of hydroxide is incorporated in the film at all times considered, t>23 minutes. This suggests a very high initial hydration of the surface film. Beyond this time the film clearly continues to hydrate, the last analysis (35 days) still failing to identify a plateau in the hydroxide proportion (O-2).
|
Figure 4: Increase in film thickness calculated from the Al2p metal and oxide/hydroxide ratios assuming that the volume density ratio of aluminium in metal to that in oxide is 1.5 and the inelastic mean free path of the photoelectrons in the film was 2.75nm
Figure 5: Carbon concentration against atmospheric exposure time.
Figure 6: Water contact angle against atmospheric exposure time.
The error bars are the standard deviation calculated from a
number of measurements on different areas of one sample.
|
The knowledge that the oxide overlays the metal allows the oxide to metal ratio area ratio, Io /Im , to be converted into a film thickness, d(nm) using equation 1:
 .......Equation 1
where the inelastic mean free path (IMFP) in the oxide,  o = 2.8nm for Al Ka generated Al2p photoelectrons emitted normal to the surface (q=90°) of a  Al2O3 overlayer on aluminium, the ratio of the volume densities of aluminium atoms in metal to oxide, Nm /No=1.5 and the ratio of the IMFP of these electrons in oxide to metal, m /o = 2.6/2.8.1
The increase of film thickness through reaction with a combination ofatmospheric water and oxygen, is illustrated in Figure 4. It is clear from these data that the film thickness is still increasing, if at a far reduced rate, after 35 days ambient storage. These data are well fitted by a logarithmic relationship. The adsorption of atmospheric hydrocarbon is apparent in Figure 5. Thus, despite the introduction of the polar hydroxyl groups into the surface film with time, the hydrophobic nature of the surface increased as measured by the water contact angle in Figure 6. It is interesting that while the contact angle appears to still be increasing after one month of ambient exposure, the majority of the increase, from about 10° to 80°, occurred during the first 24 hours.
|
Figure 7: Curve fitted Al2p and O1s and C1s core levels from a magnetron sputtered aluminium sample exposed to the atmosphere for one month at normal and shallow take off angles.
|
Variable angle XPS spectra indicated that at shallower analysis depths the hydroxyl component becomes more prominent (Figure 7). This suggests that the proportion of hydroxyl ions at the surface of the oxide/hydroxide film is greater than in the film. The additional information that can be provided by application of regularisation algorithms to these data is currently under investigation. 4
|
The effect of ambient storage of magnetron sputtered aluminium has been characterised. Logarithmic increases with ambient storage time of the hydroxyl concentration, film thickness, carbon concentration and water contact angle have been identified. The adsorption of atmospheric carbon was found to be the dominant factor in influencing the wettability of the surface, which became increasingly hydrophobic with time.
Initial angle-resolved XPS measurements suggest that the proportion of oxygen ions in hydroxyl environments is greater at the film surface.
The procedure determined elsewhere for curve fitting the O1s core level has been successfully applied to the spectra obtained herein. Satisfactory fits of the O1s envelopes were obtained using the O1s-Al2p separations previously determined.
Qualitatively, sensible trends in the hydroxyl concentration are produced, however, a quantitative verification of this method has yet to be obtained.
|
[1] B. Strohmeier, Surface and Interface Analysis, 15, 51 (1990).
[2] P.M. Sherwood, Surface Science Spectra, 5 (1), 1 (1998).
[3] M.R. Alexander, G.E.Thompson and G.Beamson, "Characterisation of the oxide/hydroxide surface of aluminium using X-ray Photoelectron Spectroscopy",
submitted to Surface and Interface Analysis.
[4] P. J. Cumpson “Guide for constructing concentration-depth-profiles from angle-resolved XPS measurements”,
NPL report CMMT(D)178, December (1998).
This poster was presented at ASA-6
"Loughborough Adhesion" Conference, Loughborough University, UK, 18th - 20th April 2000
|
![[top]](../buttons/TOP.GIF)
Back to the text
