E. M. Malykhin

The mechanism of low-k SiOCH film modification by oxygen atoms

The interaction of oxygen atoms with three types of plasma enhanced chemical vapor deposition low-k SiOCH films is studied. The samples were treated by O atoms in the far plasma afterglow conditions in a special experimental system designed for this study. The experimental system allowed avoiding the effect of ions and vacuum ultraviolet (VUV) photons on surface reactions and controlling the oxygen atom concentration over the samples. Fourier-transform infrared spectroscopy, x-ray fluorescence, and atomic force microscopy techniques were used to analyze the changes occurring in low-k films. Monte Carlo model for O atom interaction with low-k material that includes penetration, recombination, and reactions with methyl groups was developed. It is shown that the surface recombination on the pore wall surface determines the profile and penetration depth of O atoms into the films. The reaction of O atoms with methyl groups has lower probability and therefore proceeds in the background mode.

Recombination of O and H Atoms on the Surface of Nanoporous Dielectrics

The interaction of O and H atoms with SiOCH nanoporous low-dielectric-constant (low-k) films is studied in the far plasma afterglow in the absence of ion and photon fluxes on the surface. The loss probabilities of O and H atoms are directly measured by plasma-induced actinometry. Modification of low-k films during the experimental scans was studied by the Fourier transform infrared spectroscopy technique. The model of O- and H-atom recombination in nanoporous materials was developed to analyze the experimental data. It is shown that the main mechanism of the O and H loss is their surface recombination. The consumption of these atoms in the reactions with the carbon-containing hydrophobic groups has a minimal contribution. Thus, the surface recombination defines a damage depth in low-k films. It was shown that the oxygen atoms lead to the noticeable removal of CH3 groups. On the contrary, hydrogen atoms do not break Si-CH3 bonds, allowing the avoidance of plasma damage in the case of the hydrogen-plasma-based resist strip in appropriate conditions.

Surface recombination of oxygen atoms in O 2 plasma at increased pressure: I. The recombination probability and phenomenological model of surface processes

This work deals with the study of oxygen atom loss on a quartz surface in a glow discharge plasma in pure O2 at increased pressures (5?50?Torr). O atom loss probabilities are obtained from the radial distributions of oxygen dissociation degree measured by the actinometry method. It is shown that the applicability of the actinometry method at high pressures requires the knowledge of the spatial distribution of a reduced electric field for the correct calculation of the electronic excitation rates of oxygen and actinometer atoms. The analysis of the obtained data within the framework of a simple phenomenological model of the surface processes revealed that O atom surface recombination with physisorbed oxygen atoms and molecules (producing O2 and O3, respectively) is the main loss channel for oxygen atoms in O2 plasmas at increased pressures. The oxygen atom loss probability can noticeably grow in comparison with the case of low pressure due to the essential increase in the surface occupation degree by physisorbed atoms and molecules.

Surface recombination of oxygen atoms in O 2 plasma at increased pressure: II. Vibrational temperature and surface production of ozone

Ozone production in an oxygen glow discharge in a quartz tube was studied in the pressure range of 10?50?Torr. The O3 density distribution along the tube diameter was measured by UV absorption spectroscopy, and ozone vibrational temperature TV was found comparing the calculated ab initio absorption spectra with the experimental ones. It has been shown that the O3 production mainly occurs on a tube surface whereas ozone is lost in the tube centre where in contrast the electron and oxygen atom densities are maximal. Two models were used to analyse the obtained results. The first one is a kinetic 1D model for the processes occurring near the tube walls with the participation of the main particles: O(3P), O2, O2(1?g) and O3 molecules in different vibrational states. The agreement of O3 and O(3P) density profiles and TV calculated in the model with observed ones was reached by varying the single model parameter?ozone production probability on the quartz tube surface on the assumption that O3 production occurs mainly in the surface recombination of physisorbed O(3P) and O2. The phenomenological model of the surface processes with the participation of oxygen atoms and molecules including singlet oxygen molecules was also considered to analyse data obtained in the kinetic model. A good agreement between the experimental data and the data of both models?the kinetic 1D model and the phenomenological surface model?was obtained in the full range of the studied conditions that allowed consideration of the ozone surface production mechanism in more detail. The important role of singlet oxygen in ozone surface production was shown. The O3 surface production rate directly depends on the density of physisorbed oxygen atoms and molecules and can be high with increasing pressure and energy inputted into plasma while simultaneously keeping the surface temperature low enough. Using the special discharge cell design, such an approach opens up the possibility to develop compact ozonizers having high ozone yield at the low energy cost of O ? O3 conversion.

Removal of amorphous C and Sn on Mo:Si multilayer mirror surface in Hydrogen plasma and afterglow

Removal of amorphous carbon and tin films from a Mo:Si multilayer mirror surface in a hydrogen plasma and its afterglow is investigated. In the afterglow, the mechanism of Sn and C films removal is solely driven by hydrogen atoms (radicals). Probabilities of Sn and C atoms removal by H atoms were measured. It was shown that the radical mechanism is also dominant for Sn atoms removal in the hydrogen plasma because of the low ion energy and flux. Unlike for Sn, the removal mechanism for C atoms in the plasma is ion-stimulated and provides a much higher removal rate.

The effect of He plasma treatment on properties of organosilicate glass low-k films

The effect of low-pressure He plasma on properties of nanoporous organosilicate glasses low-k films with 24% and 33% open porosity is studied. The influence of ions, VUV radiation, and metastable atoms are extracted separately using a special experimental system designed for this purpose. The low-k films treated in He plasma were exposed to O or H atoms in the downstream of high-pressure O2 or H2 rf discharge. The changes in chemical composition and structure occurring in low-k films were measured before and after all treatments. The loss probabilities of oxygen and hydrogen atoms on the low-k film surface were measured for both treated and pristine films. It is shown that the film pretreatment in He plasma leads to the noticeable densification of the top surface layer up to complete sealing all the films studied. The sealing layer prevents O atoms from deep penetration to the film bulk and carbon extraction. The sealing mechanism related to the joint impact of low-energy ions and VUV photons with metasta...

Loss of Hydrogen Atoms in H 2 Plasma on the Surfaces of Materials Used in EUV Lithography

Low-pressure hydrogen is an important component of the working medium in extreme ultraviolet (EUV) projection lithography. Under the action of EUV photons and fast secondary electrons on the gas medium, plasma and atomic hydrogen, actively interacting with the surface, are produced. This interaction is very important, because it largely determines the lifetime of the multilayered EUV optics. In this study, the loss of atomic hydrogen under the conditions of a low pressure (<10 Torr) RF plasma discharge on the surfaces of materials used in EUV lithography is investigated. The surface loss probabilities of H atoms on these materials are measured. It is shown that surface recombination of atomic hydrogen goes according to the Eley-Rideal mechanism via direct recombination of H atoms from the gas phase with chemically and physically adsorbed atoms. In this case, the surface recombination probability is mainly determined by the density of chemical adsorption sites. The density of adsorption sites and the desorption energy of H atoms are estimated. The desorption energy of physically adsorbed H atoms on pure metal surfaces (or surfaces exposed to plasma) is about 0.5 eV, and the density of sorption sites is close to the surface density of atoms. This results in a high loss probability of H atoms on metals (∼0.1). Therefore, to provide efficient transportation of hydrogen atoms, it is necessary to use materials with the lowest loss probability of H atoms, i.e., dielectrics.

The structure of thin carbon films deposited at 13.5 nm EUV irradiation

In this work, the structure and chemistry of thin nm-thick carbon films deposited on a substrate using strong 13.5 nm EUV irradiation under a strong vacuum were studied. The film structure was studied by Raman spectroscopy in comparison with the Raman spectra of well-known carbon phases: diamond, single-wall nanotubes, nano- and micro-crystalline graphite and amorphous carbon. As well, FTIR spectroscopy was used to study possible IR-active chemical bonds, primarily, hydrogen bonds. It was shown that films deposited on a surface under EUV irradiation consists of amorphous sp2-carbon. The mechanisms of deposition are discussed briefly. Knowledge about the structure and chemistry of such carbon films is very important for EUV lithography.

Plasma cleaning of multilayer mirrors in EUV lithography from amorphous carbon contaminations

The feasibility of plasma cleaning of the multilayer mirrors used in 13.5-nm EUV lithography from carbon contaminations is studied. Experiments conducted in electrodeless plasma of the surface-wave low-pressure discharge in helium and hydrogen demonstrated the high rate, efficiency, and selectivity of this cleaning without any damage of the mirror’s upper protection layer, even at the atomic level. The optimal working parameters of plasma cleaning are determined and its possible mechanism is discussed.