Patent Application
20050058933
This invention relates to making semiconductors and, particularly, to photolithography.
In photolithographic processes, a photoresist is deposited. The photoresist is exposed to radiation via a mask. Some regions of the photoresist develop away and other regions remain.
In order for the photoresist to be most effective, especially in connection with advanced lithographic processes, it should be highly transparent. For example, in connection with 157 nanometer and extreme ultraviolet lithography, the photoresist may absorb too much of the incident radiation. While those photoresists were fully effective for prior generations of lithographic technology, with more modern lithographic techniques, these photoresists may be deficient with respect to their transparency. In particular, the existing photoresists may be too radiation absorptive, which degrades the patterning performance of the resist.
Conventionally, the photoresists have photoacid generators which, upon exposure to radiation, generate acids that implement the breakdown of the photoresist where exposed. Current photoacid generators are not optimized for changes in polymer resins that are implemented to improve transparency. As a result, phase separation may occur due to structural differences.
Thus, there is a need for alternate, transparent photoresists.
Photoacid generator compounds may exhibit improved transparency in comparison to materials currently being used in 157 nanometer and other lithographic technologies. These compounds may maintain sufficient quantum yield under irradiation for chemically amplified photoresists. The photoacid generators may also have improved compatibility with the resin matrix making up a photoresist, in some embodiments.
The photoacid generators may include anions and cations. The cations may have single or sigma bonds using orbital overlap, rather than conventional pi or double bonded species. These sigma-bonded species may enable absorbent moieties to trigger photochemical reaction mechanisms that produce acid upon irradiation of the catalyst precursor by vacuum ultraviolet and ionized radiation, such as, for example, electron beam, ion beam, and extreme ultraviolet lithography processes. The absorbent sigma-bonded moieties may be linked to catalyst precursors as triggers for catalyst formation. Absorption of 157 nanometer electromagnetic radiation, triggering catalyst formation, from a catalyst precursor with sigma-bonded moieties, may be less than with conventional technologies.
The use of absorbing moieties that have reduced absorption at longer wavelengths enable photogenerated catalyst design that may be optimized to 157 nanometer lithography with chemically amplified photoresist in some embodiments. Moieties of the form C—R and C—X, where R may be a hydrogen, an alkyl, a substituted alkyl, and X may be a halogen, a chalcogen or other heteroatoms may be used as catalyst precursors for chemically amplified photoresist for 157 nanometer radiation may be applied. While the fundamental deprotection mechanism common to chemically amplified photoresists may remain unchanged in some embodiments, the classical mechanism may occur without a decrease in overall resist transparency that is typically encountered by the present state of the art photoacid generators.
Currently, many of those skilled in the art have focused primarily on developing polymer resins with very low transparencies in order to facilitate resist patterning for 157 nanometer technologies. While this focus on developing low transparency resin has been a key enabler of the technology thus far, those skilled in the art are still primarily using photoacid generator materials optimized for longer wavelengths developed in earlier technologies, such as phenyl-based materials.
While phenyl-based materials do have good quantum efficiencies, these materials are highly absorbing at 157 nanometer and thus, are not fully optimized for applications at 157 nanometers. The sigma-bonded species are both highly transparent and have sufficient quantum yields at wavelengths of interest to further reduce the overall absorbance of the resist, thereby improving patterning performance. Additionally, due to structural similarities, the sigma-bonded photoacid generators also may be more compatible with polymer resin than conventional classes of photoacid generators.
In particular, the cation portion of the photoacid generator, which is the photon harvesting part, may be modified so that it is sigma-bonded. The highly conjugated phenyl groups typically involved in conventional techniques are replaced by moieties primarily comprising C—H and C—F single bonds. The reduced conjugation reduces the absorption, especially at 157 nanometers, and thereby improves the overall performance of the resist due to lower absorption. At the same time, photoacid generators that are sigma-bonded still exhibit sufficient quantum efficiency to be viable in photoresists through absorption of radiation coupled to bond scission.
Photoacid generators perform two coupled processes. First, radiation is absorbed by a competent moiety whose orbital energy is mated to photons. For longer wavelengths this has been accomplished through the application of pi-bonded species in conjugation. Such antenna moieties are efficient absorbers at longer wavelengths.
In the second process, photoacid generators fragment and form acid after radiation is absorbed. Thus, the energy from the radiation is coupled to bond breaking processes. At longer wavelengths, the efficiency of absorption is relatively low, necessitating multiple absorbing moieties on the photoacid generator. For example, triphenylsulfonium nonaflate has three radiation absorbing phenyl groups. The energy for bond breaking reactions is facilitated from efficient collection of radiation.
At shorter ultraviolet wavelengths of 248 to 193 to 157 nanometers, the efficiency of radiation absorption by molecular species improves, because the radiation is more energetic and capable of exciting more different kinds of chemical bonds. Indeed, few bonds do not absorb at 157 nanometers, complicating resist design.
At 157 nanometers, photoacid generator efficiency is not limited by absorbance, as the photoacid generator molecules are highly absorbing. The high absorption, while not materially improving quantum efficiency, does increase absorption significantly.
The conjugated moieties, such as phenyl groups, may be reduced or eliminated from the cation with the concomitant introduction of more transparent sigma-bonded moieties, such as substituted alkyl groups. The use of sigma-bonded moieties results in photoacid generators that still absorb in proximity to the site of bond scission, allowing coupling of absorption and fragmentation.
As an example, instead of using a conventional cation, the addition of a methyl iodide to a stirred solution of diisopropylsulfide in toluene, heated below reflux, may result in sigma-bond formation. Crystalline diisopropylmethylsulfonium iodide is isolated by evaporation and purified by recrystallization. Photoacid generator formation is accomplished by an ion exchange from the silver nonaflate. Ion exchange to make nonaflate sulfonium salt from sulfonium halide may result in the formation of a sulfonium halide and may follow procedures known in the art.
Referring to
Referring to
Referring to
While the use of single bonded systems, rather than double bonded or aryl systems is suggested in accordance with one embodiment of the present invention to ensure adequate quantum efficiency while decreasing the overall absorbance, it is also anticipated that modification of the anion portion of the photoacid generator may also be implemented. For example, the use of a more weakly coordinating, non-basic ion to form a stronger conjugate Bronsted acid is one example of a modified anion portion. In another example, the loading or concentration of the photoacid generator may be modified in the photoresist system to further improve quantum yields versus aryl systems. The anion modification and the concentration modification would not detrimentally affect resist performance from issues such as defects or solubility.
Additionally, while the focus of one embodiment of the present invention is not specifically on the anion portion of the photoacid generator, the new classic cations described herein is considered compatible, not only with perfluoroalkyl sulfonate (PFAS) or perfluorooctyl sulfonate (PFOS) anions, but also with new, more novel anions which are considered more environmentally friendly, such as imide and methide systems recently disclosed. See Lamanna et al., “New Ionic Photo-Acid Generators (PAGs) Incorporating Novel Purfluorinated Anions,” Proceedings of SPIE Vol. 4690 (2002).
While the present invention has been described with respect to a limited number of embodiments, those skilled in the art will appreciate numerous modifications and variations therefrom. It is intended that the appended claims cover all such modifications and variations as fall within the true spirit and scope of this present invention.
