The surface of a metal, say Al is easily oxidized. If we establish such a metal contact (of thickness < 20 nm) on n-Si, in which the native oxide is always present prior to deposition of Al. If we take XPS spectra for analysis, then how to distinguish Al2O3 from the metal surface to that of present in the Al/SiOx/Si interface?
Thanks.
Dear Indudhar, try to vary the exit angle of the X-rays. For small exit angles, i.e. angle between the surface of your sample and the direction of your electron analyser, you should first observe a strong peak related to the oxide of the surface, and smaller metal contribution. If you increase the exit angle, then metal contributions should increase, and surface oxide will decrease. If you reach the penetration depth of the photoelectrons originating from the inner interphase, the Si signals will appear, and if the inner interphase is oxidized, also the oxide will increase again with respect to the metal.
You may simulate the behavior assuming a layered structure, check the literature e.g. from Haupt & Strehblow, and Hoppe & Strehblow from the late 1980ies and early 1990ies, or the papers from e.g. Peter Cumpson on this topic!
Good luck, Dirk
Dear Dirk,
In my case, I have noticed greater metal oxide contribution than elemental metal. I have also observed the signals of Si and its native oxide.
For Al lines the binding energy 2p is a little bit lower for the pure metal. But how is in your case? Some data:
Al 2p Al2O3 74.10
Al 2p Al2O3/Al 73.00
Al 2p Al2O3/Al 75.80
Al 2p Al/Al2O3 72.90
Al 2p Al/Al2O3 73.00
Al 2p Al/Al2O3 73.00
Al 2p Al/Al2O3 75.60
Al 2p Al/Al2O3 75.80
Al 2p Al/Al2O3 75.80
Al 2p Al/Al2O3 75.90
Al 2p Al/AlOx 73.00
Al 2p Al 72.50
Al 2p Al 72.70
Al 2p Al 72.80
Al 2p Al 72.80
Al 2p Al 72.90
Dear Indudhar, of course you have to take care for a proper electrical contact of your sample, so that no charging appears. If you observe the Si substrate peaks, then of course you are well within the escape depth of the photoelectrons, implying that your surface layers are very thin, i.e. about 5 nm or even less. What happens if you change the angle of the measurement?
Hi Svetlana,
I have mentioned that the area (and FWHM) of Al/AlxOx or Al/Al2O3 contributing to XPS spectra is greater than Al metal.
(FYI: I have got B.E. of 72.66 eV for Al metal and about 75.44 eV for Al/AlxOx or Al/Al2O3 in core XPS Al2p spectra. However, in the core XPS Si2p spectra, the area of elemental Si is greater than Si/SiOx for the same Al/Si interface)
Considering your last answer this implies that the thickness of the Si oxide is small (eventually only in the monolayer range), while that of the outer Al-air (vacuum) surface seems to be larger. but again, you have to do some measurements with angular variation and fit the data to a layered model system in order to get more quantitative results.
Hi,
The surface oxidation involves carbon contaminants, so look at your C1s spectra if you got it. You should see there signals of C-O, C=O, CO2 from the surface.
In addition, maybe looking at O1s spectra will allow you to distinguish between peaks of silicon oxides, aluminum oxides, and again - carbon oxides adsorbed to Al surface.
P.S.
~20 nm of Al should be way to thick to receive any signal from Si/Al interface if your layers are continues.
As Jane pointed out, 20 nm is too thick to see the bottom interface.
What question are you trying to answer using XPS of the Al-(SiO)-Si interface?
Hi Jane,
I definitely see SiO2 in O1s spectra. The thickness I'm talking about is the thickness of metal+its oxide+semiconductor oxide is which definitely less than
Indudhar, In any case what you might do is a sputter depth profiling experiment. In that case, you will see the enrichment of oxide at the surface and the interface.
I'm trying to answer about difference between AlxOx/Al/SiOx/Si and AlxOx/Al/(AlxOx/SiOx)/Si interfaces. I found it is very difficult answer in the latter case. Is that a experimental limitation of XPS?
Hi,
the only experimental limitation in XPS is that if the thickness of the Al layer is more than say 2-3 nm it is difficult to see the interface. The solutions are:
1- Use hard X-ray at the Synchrotron (check this paper and work made by Marcus Bär: M. Wimmer, et al. "Chemical interaction at the buried silicon/zinc oxide thin-film solar cell interface as revealed by hard x-ray photoelectron spectroscopy", J. Electron. Spectrosc. Rel. Phenom. 190, 309 (2013).
Once you are at the synchrotron you could use X-ray Absorption spectroscopy to have further information.
2- If you don't have the possibility of measuring at the synchrotron, you can use the lab XPS set-up, but you should do an experiment starting from the Si substrate with it's native oxide, deposit a small amount of Al (less than 1nm, for example) and measure, deposit more Al, measure etc. up to reach a complete Al film with no Si or SiO2 visible. From this "in-situ" type of experiments, one can get enough information on the interfaces.
Indudhar,
(A) Regarding your statement: "I definitely see SiO2 in O1s spectra. The thickness I'm talking about is the thickness of metal+its oxide+semiconductor oxide is which definitely less than
Hello Indudhar,
As other experts already suggested you to do depth profiling and or synchrotron based XPS/ proper interface experiments like layer by layer deposition to understand different interfaces. All these experiments need special facilities.
With careful measurements one could identify the O KLL auger peak for Al2O3 and SiO2. This may help you to understand and distinguish the Oxygen at different interfaces. You can have look to this article on Auger Peak of O KLL
http://pubs.acs.org/doi/abs/10.1021/ac50059a017
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Thin Films and Nanotechnology