Artificial neural networks for instantaneous analysis of real-time Rutherford backscattering spectra

Demeulemeester, Jelle; Smeets, Dries; Barradas, N.P.; Vieira, Armando; Comrie, C.M.; Temst, Kristiaan; Vantomme, André

doi:10.1016/j.nimb.2010.02.127

Cited by 32 publications

(22 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…17 This approach is based on pattern recognition and allows one to automatically link acquired RBS spectra to the quantitative information of interest, e.g., thickness of the growing or shrinking phases, stoichiometry, roughness, etc. 18 With this approach, huge RBS data sets can be analyzed quasi instantaneously, without deteriorating the quantitative accuracy.…”

Section: Experimental Details a Sample Preparation And Data Acqumentioning

confidence: 99%

On the growth kinetics of Ni(Pt) silicide thin films

Demeulemeester

Smeets

Comrie

et al. 2013

Journal of Applied Physics

Self Cite

View full text Add to dashboard Cite

We report on the effect of Pt on the growth kinetics of d-Ni 2 Si and Ni 1Àx Pt x Si thin films formed by solid phase reaction of a Ni(Pt) alloyed thin film on Si(100). The study was performed by real-time Rutherford backscattering spectrometry examining the silicide growth rates for initial Pt concentrations of 0, 1, 3, 7, and 10 at. % relative to the Ni content. Pt was found to exert a drastic effect on the growth kinetics of both phases. d-Ni 2 Si growth is slowed down tremendously, which results in the simultaneous growth of this phase with Ni 1Àx Pt x Si. Activation energies extracted for the Ni 1Àx Pt x Si growth process exhibit an increase from E a ¼ 1.35 6 0.06 eV for binary NiSi to E a ¼ 2.7 6 0.2 eV for Ni 1Àx Pt x Si with an initial Pt concentration of 3 at. %. Further increasing the Pt content to 10 at. % merely increases the activation energy for Ni 1Àx Pt x Si growth to E a ¼ 3.1 6 0.5 eV. V C 2013 AIP Publishing LLC [http://dx

show abstract

Section: Experimental Details a Sample Preparation And Data Acqumentioning

confidence: 99%

On the growth kinetics of Ni(Pt) silicide thin films

Demeulemeester

Smeets

Comrie

et al. 2013

Journal of Applied Physics

Self Cite

View full text Add to dashboard Cite

show abstract

“…artificial neural networks (ANNs), has been applied to analyze these data sets. It has been shown that ANNs yield a reliable and instantaneous fully quantitative analysis of real-time RBS spectra [13]. This approach is based on pattern recognition [14] and allows to automatically link acquired RBS spectra to the quantitative information of interest [15].…”

Section: Methodsmentioning

confidence: 99%

In situ study of the growth properties of Ni-rare earth silicides for interlayer and alloy systems on Si(100)

Demeulemeester¹,

Knaepen²,

Smeets³

et al. 2012

Journal of Applied Physics

Self Cite

View full text Add to dashboard Cite

We report on the solid-phase reaction of thin Ni-rare earth films on a Si(100) substrate, for a variety of rare earth (RE) elements (Y, Gd, Dy and Er). Both interlayer (Ni/RE/) and alloy (Ni-RE/) configurations were studied. The phase sequence during reaction was revealed using real-time x-ray diffraction whereas the elemental diffusion and growth kinetics were examined by real-time Rutherford backscattering spectrometry. All RE elements studied exert a similar influence on the solid phase reaction. Independent of the RE element or its initial distribution a ternary Ni 2 Si 2 RE phase forms, which ends up at the surface after NiSi growth. With respect to growth kinetics, the RE metal addition hampers the Ni diffusion process even for low concentrations of 2.5 at. %, resulting in the simultaneous growth of Ni 2 Si and NiSi. Moreover, the formation of Ni 2 Si 2 RE during NiSi growth alters the Ni diffusion mechanism in the interlayer causing a sudden acceleration of the Ni silicide growth. *

show abstract

“…Artificial neural networks (ANNs) can be constructed capable of effectively analysing classes of IBA data (see [6]) and are now being used to handle real-time data obtained to determine the detailed annealing kinetics of various systems [95]. Much intervening work has shown that ANNs can be trained to handle multiple spectra, or multiple techniques; and it is clear that any sort of IBA can be implemented in an ANN for which a valid training set can be defined.…”

Section: Iba For Large Datasetsmentioning

confidence: 99%