A Langmuir probe was used to measure various electron plasma parameters in O 2 capacitively coupled plasma. It was shown that the variation in these plasma parameters was due to changes in the electron heating mechanisms as the discharge conditions varied. The so called 'α-γ' mode transition in O 2 plasma (100 mTorr) was identified from the power evolution (30-600 W) of the electron energy probability function (EEPF), electron density (n e ) and effective electron temperature (T eff ). The EEPF evolved from Druyvesteyn to bi-Maxwellian with increasing applied power which resulted in a rapid decrease and an abrupt increase in T eff and n e respectively. Comparisons were made to the same mode transition for similar conditions in Ar plasma. The EEPFs were Druyvesteyn in the α mode and evolved into a Maxwellian like EEPF in the γ mode of an Ar plasma. Two distinct trends of n e versus power was observed, it was shown that the measured rf current and rf voltage had a similar behavior. The pressure evolution of the EEPF, n e , and T eff was also investigated in O 2 plasma operated at both 30 and 200 W. At 30 W the number of high energy electrons decreased and flattening of the low energy portion of the EEPF occurred with increasing gas pressure (10-100 mTorr) which indicates a collisionless to collisional heating transition. However, at 200 W the right combination of rf voltage and pressure was met for the discharge to evolve into the γ mode as the pressure increased. This was evident from significant narrowing of the EEPF as the pressure increased.

Actinometry is a non-invasive optical technique that can be used to quantitatively monitor atomic oxygen number densities [O] in gas discharges under certain operating conditions. However, careless application of the technique can lead to erroneous conclusions regarding the behaviour of atomic oxygen in plasma. One limitation on this technique is an accurate knowledge of the various rate constants required, which in turn is hampered by an insufficiently precise knowledge of the Electron Energy Distribution Function (EEDF) in the plasma. In this work Particle in Cell (PIC) simulations have been used to generate theoretical EEDFs. To validate a simulation the electron density n e produced by the PIC code is compared to experimental n e values measured using a hairpin probe. The PIC input parameters are adjusted to optimise agreement between the PIC and experimental n e results. This approach should in principle yield an EEDF that more accurately reflects the true EEDF in the plasma. The PIC EEDF is then used to generate rate constants for the actinometry model which should improve the accuracy of the quantitative [O] result for that particular set of plasma conditions. The actinometry [O] results are then compared to [O] results obtained using Two-photon Absorption Laser Induced Fluorescence (TALIF) to validate the approach.

Two-photon absorption laser-induced fluorescence (TALIF) spectroscopy was used for detection of absolute atomic oxygen density in a low-pressure capacitively coupled plasma source. We investigated the variation of atomic oxygen density for various mixtures of O 2 /SF 6 and report a significant five-fold increase of [O] when oxygen plasma was diluted with SF 6 by only 5%. We attribute this increase in [O] to a combination of a change in surface conditions caused by constituents of SF 6 plasma reacting with the reactor walls and also due to an increase in the electron temperature. Atomic oxygen production rates were determined using electron-energy distribution functions obtained with a cylindrical Langmuir probe. It was found that the effective electron temperature dramatically increased from approximately 1-8 eV as the SF 6 content varied from 0% to 60% which consequently resulted in a three-fold increase in the atomic oxygen production rate. TALIF was also used to investigate the variation of [O] due to fluorination of the reactor walls and also after etching resist-coated wafers. It was found that [O] increased by over a factor of three after fluorinating the walls with SF 6 plasma; on the other hand a coating formed on the reactor walls after a resist etch process resulted in a reduction of [O] by only 20%.

Detection of fluorine atoms in a capacitively coupled SF 6 /O 2 /Ar discharge has been achieved using appearance potential mass spectrometry (APMS). Atomic fluorine density, [F], variation with power in the range of 50-300 W was found to saturate. The dissociation fraction for a fixed gas pressure of 40 mTorr and a feedstock mix of SF 6 /O 2 /Ar (85/10/5%) ranged from 3.0 × 10 −4 -2.4 × 10 −3 . The addition of O 2 to SF 6 plasma was shown to have a significant effect on [F]. The dissociation fraction in SF 6 /Ar (95/5%) plasma at 20 mTorr was over a factor of 22 lower than SF 6 /O 2 /Ar (85/10/5%) plasma. For a fixed gas pressure and power of 60 mTorr and 500 W the F + signal showed a variation of up to a factor of 60 depending if O 2 was in the feedstock gas which indicates a marked enhancement of [F] with O 2 addition. Fluorine actinometry was also used to monitor relative fluorine atom concentrations. A comparison of actinometry with APMS showed that a classical actinometric approach for measuring fluorine atoms was unreliable for the conditions examined in this work. Over the power range investigated the absolute fluorine concentration increased by a factor of 10 whereas the actinometric signal only showed a two fold increase.

A Langmuir probe has been used to investigate electron heating mechanisms in a capacitively coupled oxygen discharge over a wide pressure range (50-800 mTorr) at a fixed applied power (200 W). Evidence presented here from experimentally obtained electron energy distribution functions (EEDFs) illustrates discharge transition from a collisionless (stochastic) to a collisional (Ohmic) dominant regime with increasing oxygen pressure. The discharge exhibited a bi-Maxwellian EEDF in the collisionless regime, dominated by stochastic heating whereas Druyvesteyn-like EEDFs were exhibited in the collisional dominant regime. Moreover, in the transition between these two regimes, parameters such as electron density, effective electron temperature and electron-neutral collision frequency exhibited significant variations.