Porous low-k dielectric films have replaced silicon oxide as interlayer dielectric materials in semiconductor processing mainly because of its potential to reduce power consumption and cross talk noise. Low dielectric constant value also reduces resistive-capacitive delay which results in faster clock speed. However, implementation of low-k dielectrics in manufacturing environment requires an understanding of these new materials' characteristics and the challenges that appear in the integration of the low-k films into semiconductor processes. There are several low-k materials being used in integrated circuit manufacturing; however there is no clear winner. None of the low-k materials available commercially can fulfill all the required properties and successfully replace silicon oxide. Reducing the k value of these films by increasing porosity and decreasing density of the materials by using bulky groups such as methyl has a drastic effect on their mechanical, thermal and chemical properties.

Since most all the low-k materials for present and future technologies are porous in nature brings up several challenges. These films are highly susceptible to Atmospheric Molecular Contamination (AMC), especially contaminants containing polar O-H groups such as moisture and isopropanol (IPA) because of its porous structure. Even a small amount of absorbed moisture can increase the effective k value of the low-k film since the k value of water is as high as 79. Moisture retained in the film might outgass in further processing steps resulting in delamination and other potential issues. Etching and ashing are typical processing steps, which these films are exposed to during manufacturing environment. Residual gases from patterning (etching and/or ashing) processes can reside in pores and can outgass in later processes resulting in resist poisoning.

The interaction of impurities with porous spin-on and CVD (Chemical Vapor Depostion) deposited low-k dielectric films are studied using Mass Spectrometer and IR. IR spectroscopy is a powerful technique because, in this approach the surface is probed directly and any changes in the thin film can be related to changes in the absorbance spectra. The wavelength of the light absorbed is characteristic of the chemical bond of species in the film and the strength of the absorption is proportional to the concentration of the species present in the film.

A unique set up was developed in this research for real time and in-situ characterization of porous low-k dielectric films. In order to study the changes happening on the surface real-time, it is important to have a good control of the surrounding medium and an effective purge process, otherwise the accumulation of gas phase impurity will affect the results. The novel design of the cell allows well controlled purging of the samples and can be used for dynamics study. Transmission mode has been used in this study. For the purpose of increasing the sensitivity, the cell is designed in such a way that several sample coupons can be placed parallely in the path of light so that light passes through more sample length. Mercury Cadmium Telluride (MCT) detetor is used to further enhance the sensitivity of absorbance. Ultra high pure (UHP) N2 carrier gas is used to reduce the effect of impurities in the background. This unique set-up allows to study dynamics of impurity outgassing, the results could be used to extract fundamental transport parameters such as diffusivity and solubility of impurities in low-k films. This new cell can be applied to study over a wide range of temperatures ranging from room temperature to 350 oC.

In this study number of low-k films, including porous methylsilsesquioxane (MSQ) and black diamond II (BD II), are studied. These films are presently being used as interlayer dielectric materials in IC manufacturing and are exposed to wide variety of conditions during fabrication processes. The effect of various etching and ashing process on the outgassing rate are compared with the blanket and cured only films. The two model contaminants used for this study are moisture and IPA.

A fundamental process model is developed to study the mechanism of contaminant sorption, transport of gases through the bulk matrix as well as pores of the films. The experimental methodology along with physical model could be used to study the dynamics of impurity absorption and outgassing in low-k films, the effect of basic film properties, such as, porosity and film thickness on the contaminants uptake and removal. It could be also used to develop methods to improve the effectiveness of purge process in order to minimize the chemical and energy usage (for ESH gain and cost reduction), as well as decrease in the effective k-value (for performance gain).