Development of Wet Resist Strip Strategies based on the Identification of Structural Changes in the Resist after Arsenic Ion Implantation (Ontwikkeling van natte fotolak verwijderstrategieën gebaseerd op de identificatie van strukturele veranderingen in de fotolak na arseen implantatie)
Development of Wet Resist Strip Strategies based on the Identification of Structural Changes in the Resist after Arsenic Ion Implantation
Tsvetanova, Diana; S0186224;
In the processing of integrated circuits, the source and drain of a p-type and n-type complementary metal-oxide-semiconductor field-effect transistor (CMOS) are defined by implantation of donor and acceptor ions respectively. During the ion implantation, a photoresist (PR) is used as a masking material. The removal of high dose (≥ 1E15 at/cm2) and low energy ion implanted photoresist (II-PR) after implantation of ultra shallow extension and halo regions (ultra shallow junctions (USJ)) is considered as one of the most challenging front-end-of-line processing steps for 32 nm and beyond technology nodes of logic devices. This is due to the difficulties of removing the modified layer (crust) formed on the top and sidewalls of the PR during the ion implantation in combination with the compatibility towards ultra shallow implanted novel (Si)Ge, III/V substrates, metal gate and high k-materials. Commonly used resist strip processes such as fluorine-based dry plasma ash and hot sulfuric/peroxide mixtures induce unacceptable levels of oxidation and material loss. Alternative cleans need to be developed for removal of II-PR.This thesis concentrates on an in-depth understanding of the physical and chemical processes induced by arsenic ion implantation in 248 nm DUV PR. Based on this, wet approaches using UV irradiation and organic solvents are proposed and evaluated for PR stripping after USJ implantation for sub-32 nm CMOS technology nodes.In order to gain insight into the physical processes induced by ion implantation in the PR Stopping and Range of Ions in Matter (SRIM) calculations have been used for evaluation the magnitude of the ion stopping mechanisms as a function of the implantation conditions. SRIM cannot predict the main degradation behavior of the PR but it allows evaluating the physical parameters (Linear Energy Transfer (LET) and dose) that modulate the chemical reactions. It has been demonstrated that the variation of the electronic (ionization) and nuclear (vacancy) stopping can be used as a guideline for the degree of chemical modification, and consequently the stripping resistance, of PR as a function of the implant energy, dose, tilt angle and tentatively ion species.In order to build up understanding for the chemical processes induced by arsenic ion implantation in PR various analytical techniques have been combined for PR characterization as a function of the implant energy and dose. A novel knowledge for the PR chemical degradation has been established. A radical mechanism of crust formation has been proposed, which involves cross-linking and chain scission reactions of the resist. The PR cross-linking is the dominant reaction especially for high doses and energies. It is induced by ß-cleavages in the PR which lead to loss of hydrogen and formation of carbon macroradicals that recombine to form C-C cross-linked crust. Moreover, formation of αß-unsaturated ketonic and/or quinonoid structures by cross-linking reactions has been suggested. The dopant atoms can additionally provide rigid points (As-C bonds) in the crust. For higher doses and energies further dehydrogenation occurs leading to formation of triple bonds in the crust. Different π-conjugated structures are formed as a result of cross-linking and dehydrogenation. No experimental evidence for the presence of an amorphous carbon in the crust has been revealed. In addition, presence of substrate atoms in the form of SiO2 has been found on the II-PR side walls indicating substrate sputtering during ion implantation.Polar organic solvent chemistries have been recognized as the most promising for PR removal after USJ implantation for future CMOS technology nodes. Physical force is needed to break the crusted top and side walls of the PR and enable dissolution of the underlying bulk PR. After a solvent strip with physical force some small amount of crust side wall footing residues remains that cannot be removed by increasing the total physical force applied. The chemical analysis of these residues has revealed that they are carbon rich containing C-C/C-H, C-Si and dopant As-C and As-O bonds. The use of a UV treatment in air or nitrogen ambient before and/or after the solvent strip with physical force has been proposed and evaluated in order to enable a complete removal of the II-PR.The chemical degradation of the II-PR by UV irradiation and oxidizing species has been studied in detail using the same approach as after ion implantation. It has been found that the UV induces scission in the crust resulting in an increase of its solubility in organic solvents. Moreover, oxidizing species present in the ambient during the UV treatment additionally can induce scission and oxidation of the crust. In parallel with the modification of the crust, cross-linking, scission and oxidation of the underlying bulk PR occur. The dominant degradation reaction of the bulk PR is cross-linking induced by the UV which leads to a decrease of its solubility in organic solvents. Thus, the effect of the UV pretreatment on the II-PR removal by organic solvent will depend on the balance between the scission of the crust and cross-linking of the bulk PR.In order to define optimum conditions for the UV pretreatment of the II-PR the effect of the UV parameters on the properties of the II-PR and its removal has been investigated. The scission of the crust and cross-linking of the bulk PR depend on the wavelength and dose of the UV radiation. Short wavelengths (< 200 nm) and low exposure doses have been found to be the most optimum for UV pretreatment since the high photon energy and the high absorption of the crust versus bulk resist favor scission of the top crust and limit the cross-linking of the underlying bulk PR due to limited UV penetration depth. A post-treatment with short UV wavelengths is recommended after the solvent strip for removal of residual organic contamination of bulk resist, solvent or priming layer as well as crust footing residues if still present on the substrate. Finally, the use of oxidizing ambient during the UV treatment facilitates the effect of the UV pre- and post-treatments.