Absence of Fermi level pinning at metal-In<i>x</i>Ga1−<i>x</i>As(100) interfaces

Brillson, L. J.; Slade, M. L.; Viturro, R. E.; Kelly, M. K.; Tache, N.; Margaritondo, G.; Woodall, J. M.; Kirchner, P. D.; Pettit, G. D.; Wright, S. L.

doi:10.1063/1.97027

Cited by 32 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Having more than one substrate element reacting with the oxide at the interface with different formation and decomposition energetics will lead to reaction channels resulting in various desorption and segregation products. 2 To preserve a stoichiometric InGaAs surface, one of the reported solutions employs an As-capping layer that is desorbed at relatively low temperatures 3,4 prior to further deposition. However, some methods for epitaxial growth of InGaAs on GaAs such as metal-organic chemical vapor deposition are not feasible with As capping.…”

mentioning

confidence: 99%

Indium stability on InGaAs during atomic H surface cleaning

Aguirre-Tostado

Milojević

Hinkle

et al. 2008

Applied Physics Letters

View full text Add to dashboard Cite

Atomic H exposure of a GaAs surface at 390°C is a relatively simple method for removing the native oxides without altering the surface stoichiometry. In-situ reflection high energy electron diffraction and angle-resolved x-ray photoelectron spectroscopy have been used to show that this procedure applied to In 0.2 Ga 0.8 As effectively removes the native oxides resulting in an atomically clean surface. However, the bulk InGaAs stoichiometry is not preserved from this treatment. The In:Ga ratio from the substrate is found to decrease by 33%. The implications for high-mobility channel applications are discussed as the carrier mobility increases nearly linearly with the In content. © 2008 American Institute of Physics. ͓DOI: 10.1063/1.2919047͔One of the approaches being considered for future complementary metal-oxide-semiconductor devices is the substitution of silicon with high-mobility channels, where In x Ga 1−x As is one of the preferred candidates for metaloxide-semiconductor field-effect transistor ͑MOSFET͒ applications. 1 However, the interface thermal stability and interface quality of oxide insulators on InGaAs and other III-V semiconductors are far more complex than those on Si. Having more than one substrate element reacting with the oxide at the interface with different formation and decomposition energetics will lead to reaction channels resulting in various desorption and segregation products. 2 To preserve a stoichiometric InGaAs surface, one of the reported solutions employs an As-capping layer that is desorbed at relatively low temperatures 3,4 prior to further deposition. However, some methods for epitaxial growth of InGaAs on GaAs such as metal-organic chemical vapor deposition are not feasible with As capping. 5 An alternative to As capping includes passivation using the surface oxides which can be removed upon hydrogen plasma treatment ͑HPT͒ at a substrate temperature below 250°C, where it has been reported that no detectable surface decomposition occurs by x-ray photoelectron spectroscopy ͑XPS͒. 3,5 Atomic hydrogen treatment ͑AHT͒ from a H 2 thermal cracking source is an alternative technique for the removal of the oxide capping layer being advantageous over HPT as no H ions are supplied. This letter describes the reaction channels observed during the removal of air exposed grown native oxide on ͑13.5 nm͒ In 0.2 Ga 0.8 As/ GaAs͑001͒ upon AHT at 390°C and subsequent in-situ reflection high energy electron diffraction ͑RHEED͒ and XPS analysis. It is found that the AHT procedure results in an atomically clean ͑2 ϫ 4 reconstructed͒ In x Ga 1−x As surface. However, an indium depletion at the surface and subsurface regions is detected which may be expected to impact performance such as interface defect state densities and channel mobility in MOSFET devices. 6,7 The sample employed for this study was a commercial ͑13.5 nm͒ In 0.2 Ga 0.8 As layer grown by molecular beam epitaxy 8 on a semi-insulating GaAs͑001͒ wafer with a GaAs buffer layer of 535 nm thick. The In 0.2 Ga 0.8 As layer and GaAs buffer la...

show abstract

mentioning

confidence: 99%

Indium stability on InGaAs during atomic H surface cleaning

Aguirre-Tostado

Milojević

Hinkle

et al. 2008

Applied Physics Letters

View full text Add to dashboard Cite

show abstract

“…Table 2 reports the values of the contact potential difference obtained for the three different layers and for the two tip-sample distances (z 100 and 200 nm). The measured values for the same zone differ by less than AE 0.1 V. Measurements taken on Ga x In 1-x As (100) layers with clean, ordered surfaces by soft Xray photoemission spectroscopy show that the Fermi level is unpinned [14]. This gives the value of the electron af®nity: 4.77 eV.…”

Section: Measurements On the Au Surfacementioning

confidence: 87%

Contact Potential Difference of Au and GaInAs by Electrostatic Force Microscopy

Bresse

2000

Mikrochim Acta

View full text Add to dashboard Cite

For a metallic surface (Au) and highly doped (N) and (P) semiconductor surfaces (GaInAs) and for localised zones (2 Â 2 mm) we have measured using an electrostatic force microscope the variation of the gradient of the electrostatic force by the signal (phase of the oscillating movement of the metallised tip) as a function of the sample-tip potential difference (À 4 V to 4 V). In both cases the signal shows a quadratic variation with the sample-tip potential difference. The variation of the signal is of the order of magnitude of the theoretical predictions obtained by modelling the shape of the tip by a truncated cone a portion of a sphere.Using the parabolic curve that ®ts the experimental results, the value of the contact potential difference, corresponding to a zero value of the electrostatic force gradient, can be determined with an accuracy of 50 mV. The contact potential difference, measured between the metallised tip and the metal (Au), taken as a reference, allows the work function of the metal tip to be determined (5.25 eV). The values of the contact potential difference for the GaInAs (N) and (P) surfaces can be explained by the Fermi level pinning due to surface charges.The electrostatic force microscope (EFM) which is derived from the atomic force microscope (AFM) is sensitive to the electrostatic force between the metallised tip and the sample. This electrostatic force can be due to the presence of existing charges or to an applied bias between the tip and the sample.The interest of this new microscopy is its ability to image local voltages of working microelectronic structures [1, 2], charges on insulator surfaces [3] and to image the ferroelectric domains [4].The curve of the electrostatic force versus the potential difference between the tip and the sample allows measurement of the contact potential difference which is directly related to the work function difference. As for the Kelvin method [5], this also allows local potentiometry to be performed [6], or the dopant distribution to be mapped [7].The electrostatic force due to the effect of a potential difference between the tip and the surface of a metal or a highly doped semiconductor can be evaluated by an analytical model using a geometrical simulation of the tip shape and the cantilever [8,9].In this paper, we have used an EFM working in thè`t apping'' mode. The electrostatic force is separated from the van der Waals force by using a``lift'' method which allows ®rst the topography to be recorded using the van der Waals forces and then the surface topography to be followed at a distance constant equal or greater to 100 nm. At this distance, the tip is only sensitive to the electrostatic force.In the case of a metallic surface and of a highly doped semiconductor, we have calculated, using the geometrical shape of the tip, the electrostatic force as well as its gradient with the distance. We have done experiments by applying a bias to the Au contacts and the GaInAs (N) and (P) layers with the metallised tip grounded. For each sample, a 2 Â 2 mm ...

show abstract

“…However, they found considerably different values for ¢Bn,Au· They attributed this difference to surface preparation, i.e. air-exposed [KA73] vs. clean [BI86a,b], and suggested that air-exposure of the surfaces led to pinning. ... …”

Section: Conduction Behaviormentioning

confidence: 99%

“…This variation is unacceptable for the production of many devices, which may require control to within 0.02eV [BR86]. By comparison, it has been shown that the Fermi level of~ InxGa 1 _xAs does not pin for atomically clean surfaces, but does for air-exposed [BI86a,b].…”

Section: Experimental Evidencementioning

confidence: 99%

See 1 more Smart Citation

Thin film reactions on alloy semiconductor substrates

Olson¹

1990

View full text Add to dashboard Cite

Absence of Fermi level pinning at metal-InxGa1−xAs(100) interfaces

Cited by 32 publications

References 23 publications

Indium stability on InGaAs during atomic H surface cleaning

Indium stability on InGaAs during atomic H surface cleaning

Contact Potential Difference of Au and GaInAs by Electrostatic Force Microscopy

Thin film reactions on alloy semiconductor substrates

Contact Info

Product

Resources

About