A negative capacitance effect has been observed in metal-semiconductor contacts. This phenomenon is explained by considering the loss of interface charge at occupied states below Fermi level due to impact ionization. A modified Shockley–Read treatment is proposed to interpret the experimental observations. In particular, a two-energy-level simplified model is presented to simulate the capacitance spectrum. The results are in good agreement with the experimental data.
A capacitance-spectroscopy technique based on accurate phase detection has been developed to measure the unoccupied states at silicide-silicon contacts. For Pd and Ni silicides, a dispersed group of states was found to exist in the Si band gap with its peak at a level 0.63-0.65 eV above the valence-band edge. Siiicide formation alters their density and distribution to reflect the changes in the structural perfection and barrier height. Observations on the epitaxial NiSi 2 -Si(lll) interfaces reveal that the characteristics of these states are controlled by the degree of structural perfection of the interface instead of the specific epitaxy. This seems to be the first correlation of the structural and electronic properties of a silicide-silicon interface.
Using electron-spin resonance (ESR), we demonstrate that several E′ variant precursors exist in a variety of technologically significant thermally grown thin SiO2 films on Si. The E′ variants include two varieties with the ubiquitous Eγ′ line shape (zero-crossing g=2.0005, O3≡Si⋅) and a second very narrow line shape (zero-crossing g=2.0019, structure unknown). We tentatively label the g=2.0019 defect EP for provisional E′ and distinguish the Eγ′ variants Eγn′ (neutral) and Eγp′ (positive). We combine ESR, capacitance versus voltage electrical measurements, and charge injection sequences to compare the electronic properties of the defects. We find that paramagnetic EP defects are positively charged while paramagnetic Eγ′ centers can be either positively charged or, under some circumstances, neutral. We find that EP precursors have a very large capture cross section for holes (σ=10−13 cm2) and that paramagnetic EP defects have an even larger capture cross section for electrons (σ=10−12 cm2). Both EP capture cross sections are an order of magnitude greater than those of the Eγp′ defects. We find that EP centers are distributed much more broadly throughout the oxide than either the Eγp′ or Eγn′ defects. We also find a two order of magnitude variation in EP density dependent upon processing variations. In addition, EP centers, unlike the Eγ′ variations, are not stable at room temperature. With their large capture cross section for holes and even larger capture cross section for electrons, EP defects may be relevant to device reliability and charge trapping under conditions of a low, relatively pure hole fluence such as in hot hole injection in short n-channel metal-oxide-semiconductor field-effect transistors.
We demonstrate that at least two varieties of E′ defect precursors exist in a wide variety of conventionally processed thermal SiO2 thin films. We provisionally label the defects EP and E′γp. We find that EP defect capture cross sections exceed the corresponding E′γp values by an order of magnitude, that EP centers are distributed far more broadly throughout the oxides than are the E′γp defects, and that the EP resonance, unlike the E′γp resonance is not stable at room temperature.
A study has been conducted of the effects of deposition conditions on the radiation hardness of borophosphosilicate glass (BPSG). Films deposited by two common deposition techniques were evaluated using gamma cell testing, electron spin resonance (ESR), and capacitance voltage (CV) measurements. The results indicate that two stoichiometrically similar films can differ greatly in radiation tolerance depending on the deposition conditions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.