Acid Diffusion Length Research Articles

Improving resist resolution and sensitivity via electric-field enhanced postexposure baking

This article explores a methodology for enhancing the resist resolution and improving resist profile based on confining the photoacid drift/diffusion by an external electric field. A properly offset alternating electric field applied to the resist film during postexposure bake (PEB) can enhance the photoacid drift in the vertical direction, drive acid to the desired direction, and thereby confine the lateral acid diffusion. The experiments were conducted on UVIIHS resist with JEOL electron-beam exposure tool and UVII-10 resist on ASML KrF stepper, respectively. The scanning electron microscopy pictures show that electric-field enhanced PEB can control the resist profiles and increase the verticality of resist sidewalls. Electric-field-enhanced PEB also significantly improves the tolerance of over and underexposure and provides better critical dimension control. It is estimated that it reduces the lateral acid diffusion length by about 70–90 nm in KrF lithography.

Mask Error Factor in Proximity X-Ray Lithography

We discuss the mask error factor (MEF) in proximity X-ray lithography. The MEFs for dense-line, isolated-line, and isolated-hole patterns were estimated by using aerial image simulations and exposure experiments. When the minimum feature size was 100 nm, MEFs in the patterns were kept within unity by setting the proximity gap to 15 µm. For feature sizes of 70 nm, the proximity gap should be reduced to as small as 10 µm to maintain the MEF at around unity for dense- and isolated-line patterns. For isolated-hole patterns, a high-resolution resist with a shorter acid-diffusion length (<20 nm) should be used to keep the MEF at around unity. The MEF changes depending on geometric conditions, such as the pattern size, proximity gap, X-ray wavelength, and absorber thickness. The MEF is determined mainly by the interference between the diffracted X-rays, which have different phases. Thus, the MEF can be controlled by choosing the appropriate exposure conditions. In most cases, the MEF in proximity X-ray lithography is much smaller than that in optical lithography, which is advantageous for the proximity lithography because it allows a larger margin in the critical dimension control for 1X X-ray masks.

Enhancement of resist resolution and sensitivity via applied electric field

This article presents a methodology for enhancing the resist sensitivity and resolution based on confining the photoacid drift/diffusion by external electric field. An alternating electric field applied to the resist film during postexposure bake can enhance the photoacid drift in the vertical direction, reduce the bake time, and thereby confine the lateral acid diffusion. A mathematical model is presented and a rigorous solution is obtained in the case of Fickean diffusion and constant electric field. The experiments were conducted on UVIIHS resist with a JEOL electron-beam exposure tool. The scanning electron microscope pictures show that electric-field enhanced postexposure bake (PEB) can reduce the PEB time requirement by 30%, and at the same time, improve the sharpness of two-dimensional corners and increase the verticality of resist sidewalls. Electric-field-enhanced PEB also significantly improves the tolerance of overexposure and provides better critical dimension control. It is estimated that it reduces the lateral acid diffusion length by about 70 nm, or 50%.

Aqueous base development and acid diffusion length optimization in negative epoxy resist for electron beam lithography

A new aqueous base developable, chemically amplified negative resist based on epoxy chemistry is evaluated for high-resolution, high-speed e-beam lithography. This resist is formulated using partially hydrogenated poly(hydroxy styrene) and epoxy novolac polymers. Degree of hydrogenation controls the aqueous base solubility and microphase separation phenomena. Reduction of edge roughness compared to the pure epoxy systems is observed whereas the absence of swelling phenomena allows lithography up to 100 nm regime and a sensitivity of 4–8 μC/cm2 at 50 keV. The diffusion coefficient has been evaluated both from high-resolution line and dot exposures and it is found to be 5×10−14 cm2/s for the optimal thermal processing conditions selected.

Analysis of Quencher Diffusion in Chemically Amplified Resists and Its Effect on Imaging Characteristics.

Diffusion of quencher contained in most chemically amplified resists today can cause proximity effects such as dependence of pattern feature size on pattern density or iso-dense bias. Diffusion coefficients were measured for several quencher compounds, and significantly large diffusion coefficients (e.g. 0.003μm2/sec at 100°C) were observed compared with those observed for acid. To evaluate the influence of quencher diffusion on macroscopic imaging behavior of resists, a new resist model for mutual diffusion of acid and quencher was introduced and was implemented to our in-house lithography simulator. Simulation results based on the model explained various proximity effects observed on real wafers, assuming quencher diffusion length as large as 0.2μm. This is significantly long compared with acid diffusion lengths reported and agrees with the results of direct measurement. Our results suggest that very small amount of quencher can have a strong influence on imaging characteristics of chemically amplified resists.

Novel Diffusion Analysis in Advanced Chemically Amplified DUV Resists Using Photometric Methods.

The demand for smaller device dimensions in microlithography drives the need to understand and control diffusion during photoresist processing.In advanced chemically amplified systems the lithographic performance is strongly influenced by diffusion of acid and base additives. Photoacid generation efficiencies and diffusion parameters were quantitatively evaluated using an established in situ photometric method employing a pH-sensitive organic dye. [1-3] A kinetic model for the post-exposure bake (PEB) has been proposed, transferred to molecular reaction dynamics and extanded to transport properties. The experimental data for this model have been obtained from UV/VIS spectroscopy measurements. The UV/VIS data show that a photoacid loss reaction during post-exposure bake has to be taken into account. Rough estimations of the acid diffusion lengths are given by the rate of the diffusion-controlled neutralization reaction for blanket exposures. The acid diffusion range was minimized, as the temperature of pre baking was raised and the PEB temperature was reduced. Chemical kinetics around the glass transition temperature of the resist are discussed.Furthermore, a face-to-face contact experiment (wafer-sandwich) involving the dye was used to determine acid loss during PEB. Comprehensively, the acid diffusion within the matrix and acid evaporation out of the matrix could be quantified.Results are used to improve the lithographic performance of the dual-wavelength CARL® [4, 5] resist system presently used at Infineon Technologies.

A study of dissolution characteristics and acid diffusion in chemically amplified DUV resist

The effects of dissolution characteristics and acid diffusion behavior on lithographic performance were evaluated in tert-butoxycarbonyl (t-BOC)-protected chemically amplified positive deep-ultraviolet (DUV) resists. The resists consisted of t-BOC-protected polyhydroxystyrene as a base resin and 2,4-dimethylbenzenesulfonic acid derivative as a photoacid generator (PAG). In particular, the line width difference between an isolated line and a dense line (iso/dense bias) was investigated by changing the post-exposure bake (PEB) temperature. As a result, clear relationships among dissolution characteristics, acid diffusion length, and iso/dense bias were obtained. Moreover, suitable dissolution characteristics and acid diffusion length for reducing iso/dense bias were clarified.

Polymer structure effect on dissolution characteristics and acid diffusion in chemically amplified deep ultraviolet resists

The role of molecular weight dispersion (Mw/Mn) of base phenolic resin was investigated in chemically amplified positive KrF resists. The resists consisted of tert-butoxycarbonyl (t-BOC) blocked polystyrene as base resin and 2,4-dimethylbenzenesulfonic acid derivative as photoacid generator, and the Mw/Mn value was changed as 1.2, 4.0, and 9.0. Not only dissolution rate contrast, but also the slope m of log(dissolution rate)-log(exposure dose) plots increased with decreasing Mw/Mn. In the case of low Mw/Mn, the molecular structure becomes more homogeneous, so resist film could be dissolved uniformly via constant developer penetration into film. This uniform dissolution characteristics may be attributable to the origin of dispersion effects mentioned above. Furthermore, higher acid diffusion property and smooth pattern side wall were obtained without side wall roughness in lower Mw/Mn. This fact indicates that acid diffusion length is also uniform within resist film in the case of low Mw/Mn. Based on the experimental analysis, the clear relationship among Mw/Mn, dissolution characteristics and acid diffusion behavior in resist film was obtained. Moreover, it was found that lower Mw/Mn has advantage for improving inherent resist performance.

Acid and base diffusion in chemically amplified DUV resists

In order to clarify the photogenerated acid diffusion in resist film, the diffusion behavior of acid, as well as the role of additional base component was investigated in tert-butoxycarbonyl (t-BOC) protected type chemically amplified positive deep ultraviolet (DUV) resists. The resists consisted of t-BOC protected polystyrene as a base resin, 2,4-dimethylbenzene sulfonic acid derivative as a photoacid generator (PAG) and N-methyl-pyrrolidone (NMP) as an additional base component. Acid diffusion coefficient was suppressed by the addition of base component. Moreover, the change of base concentration corresponded directly to the lithographic performance, such as sensitivity, resolution capability and resist profile, especially T-topping formation. Based on the experimental analysis, the clear relationship between acid diffusion length and additional base was obtained.

Relationship between Remaining Solvent and Acid Diffusion in Chemically Amplified Deep Ultraviolet Resists

For clarification of the diffusion path of photogenerated acid in the resist film, the prebake temperature dependence and tert-butoxycarbonyl ( t-BOC) blocking level dependence on acid diffusion were investigated for chemically amplified deep ultraviolet (DUV) positive resists. The resists consisted of a t-BOC protected polystyrene base resin and a 2,4-dimethylbenzenesulfonic acid derivative as a photoacid generator (PAG). The concentration of remaining solvent in the resist film decreased as the prebake temperature or t-BOC blocking level increased. The acid diffusion coefficient was almost independent of exposure dose; however, it decreased with increasing prebake temperature or blocking level. Therefore, it was considered that the concentration of remaining solvent in the resist film corresponds to one of the acid diffusion paths. Moreover, hydrophilic OH sites, whose concentration was the reciprocal of the blocking level, were also considered to correspond to a diffusion path in the polymer matrix, based on the fact that the acid diffusion length increased gradually with decreasing t-BOC blocking level, even when the solvent concentration remained constant. Based on the experimental analysis results, the existence of two diffusion paths, as well as of explicit relationships among the remaining solvent, t-BOC blocking level and acid diffusion length was confirmed.

Theoretical Study of Latent Image Formation in Chemically Amplified Resists

Theoretical analysis of chemically amplified resists indicates that a higher reaction order of acid-catalyzed reaction and lower diffusion length of acid enhance image quality. We identified the relation between diffusion length and the number of catalytic acids appropriate for inducing a given amount of acid-catalyzed reactions as a function of the difference between the activation energies of diffusion (E ad) and acid-catalyzed reaction (E ar), and the post-exposure baking (PEB) temperature. When E ad=E ar, the diffusion length required for a certain amount of acid-catalyzed reaction is inversely proportional to the square root of the amount of generated acids, regardless of PEB temperature. When E ad≠E ar, the required diffusion length depends on PEB temperature. When E ad<E ar, the diffusion length is smaller at higher PEB temperatures, and this will be preferable for enhancing the resist resolution.