No abstract

Publication informationJournal of Hospital Infection, Article in Press :Publisher Elsevier

Measurements of the electron temperature, plasma density, and floating and plasma potentials with Langmuir probes in radio-frequency discharges often represent a challenge due to rf oscillations of the plasma potential. These oscillations distort the probe characteristic, resulting in wrong estimates of the plasma parameters. Both active and passive rf compensation methods have previously been used to eliminate rf fluctuation effects on the electron current drawn by an electrostatic probe. These effects on an uncompensated probe have been theoretically and experimentally studied by Garscadden and Emeleus [Proc. Phys. Soc. London 79, 535 (1962)], Boschi and Magistrelli [Nuovo Cimento 29, 487 (1963)], and Crawford [J. Appl. Phys. 34, 1897 (1963)]. They have shown theoretically that, assuming a Maxwellian distribution and sinusoidal plasma-potential oscillation, the electron temperature can be deduced directly from an uncompensated Langmuir probe trace, by taking the natural logarithm of the electron current. It is the purpose of this paper to bring back the attention onto this result, which shows that under certain discharge conditions it is not necessary to build any rf compensation in a Langmuir probe system. Here we present and reference experimental data found on the literature which support this result. Also computational data are presented.

Many of today's continuous-wave plasma etch processes of semiconductor devices utilize chlorine or chlorine-based gases which require stable power delivery conditions. When the plasma process is intentionally (pulsed) or unintentionally driven into an unstable mode, the population of electronegative species become time-dependent. This paper examines the instability coupling between the external circuit and a chlorine plasma within an inductively coupled plasma tool. Standing-wave ratio measurements are used to map instabilities in a 2 mTorr pure chlorine discharge, as a function of power and the parallel capacitor (C P ) of the T network (matching box and antenna). At low power (<180 W) no instabilities are observed whether the system is matched or mismatched. Instabilities are only observed at input power levels from 180 to 280 W when the system is matched. At larger powers (>280 W) instabilities are observed when the system is mismatched with respect to the system input impedance (50 ) and the mismatch makes the load reactance negative. The asymmetry of the match is explained using an equivalent electrical model and power balance arguments. The methodology developed allows the identification of four different types of instability and indicates operating regimes which will be free of instabilities.

Thermoplastic substrates made of cyclo olefin polymer (COP) are treated with oxygen plasma discharges to introduce polar groups at the surface. This is the first step in the process of surface functionalization of COP substrates used in biosensor devices. A molecular dynamics model of basic COP structure is implemented using the second-generation reactive empirical bond order (REBO) potentials for hydrocarbon–oxygen interactions. The model includes covalent bond and Van der Waals interactions. The bombardment of a COP surface with mono-energetic atomic oxygen ions, energy in the range 1-35 eV, is simulated and reported here. The dynamics of the substrate modification reveals that the substrate top layer is de-hydrogenated and subsequently builds up an oxygen–carbon matrix layer, ∼10 Å thick. Analysis of the modified substrates indicates that surface yield is predominantly peroxide groups.