David S. Fryer

Extraordinary elevation of the glass transition temperature of thin polymer films grafted to silicon oxide substrates

We used local thermal analysis and ellipsometry to measure the glass transition temperatures (Tg) of supported thin films of poly(4-hydroxystyrene) (PHS) and hydroxy terminated polystyrene (PS-OH). The films were spuncast from solution onto silicon oxide substrates and annealed under vacuum at elevated temperatures to graft the polymer to the substrate. Grafting was verified and characterized in terms of the thickness of and the advancing contact angle of water on the residual layer after solvent extraction. For PHS, each segment of the polymer chain was capable of grafting to the substrate. The thickness of the residual layer increased with increasing annealing temperature. For this polymer the critical thickness below which the Tg of the film deviated from the bulk value was nearly 200 nm after annealing at the highest temperature (190 °C); the Tg of films 100 nm thick or less were elevated by more than 50 °C above the bulk value. For PS-OH films the polymer was only capable of grafting at one chain end...

Scaling of Tg and reaction rate with film thickness in photoresist: A thermal probe study

A thermal probe technique, local thermal analysis, was used to measure the glass transition temperature (Tg) and reaction rate as a function of film thickness in chemically amplified photoresists. Using this technique, heat loss into a resist film was monitored as the temperature of the probe was ramped from ambient to temperatures as high as 200 °C. The thermal events, glass transition temperature or heat evolved during reaction, were recorded as a function of the probe temperature. The Tg of the photoresists UVN 30, UV6, UV3, KRS, and KRS-XE was measured for thick films and for ultrathin films approximately 50 nm thick. The measured Tg in ultrathin resist films was 4–22 °C higher relative to that measured in thick films. We also investigated the behavior of polyhydroxystyrene films, and found that crosslinking to the substrate can increase Tg by a large amount. The photoresist films were then exposed with x-ray radiation at the same dose (950 mJ/cm2) for both thick and ultrathin films to ensure a consta...

Study of the fundamental contributions to line edge roughness in a 193 nm, top surface imaging system

Top surface imaging systems based on vapor phase silylation have been investigated for use at a variety of wavelengths. This approach to generating high aspect ratio, high resolution images held great promise particularly for 193 nm and EUV lithography applications. Several 193 nm top surface imaging (TSI) systems have been described that produce very high resolution (low k factor) images with wide process latitude. However, because of the line edge roughness associated with the final images, TSI systems have fallen from favor. In fact, TSI does not appear in the strategy or plan for any imaging technology at this time. Most of the 193 nm TSI systems that have been studied are based on poly(p-hydroxystyrene) resins. These polymers have an unfortunate combination of properties that limit their utility in this application. These limiting properties include (1) high optical density, (2) poor silylation contrast, and (3) low glass transition temperature of the silylated material. These shortcomings are relate...

Study of acid diffusion in resist near the glass transition temperature

The diffusion of p-toluene sulfonic acid (PTSA), a photoacid, in novolac was investigated at temperatures below and just above the glass transition temperature (Tg=91 °C) using a technique based on ellipsometry. Bilayer samples consisting of a blended film of novolac and PTSA and a film of pure novolac were held at a constant temperature for a known period of time and then quenched. The diffusion of PTSA from the supply layer into the initially pure novolac transforms the initial step function profile in refractive index to a more diffuse gradient in refractive index. Ellipsometry data from the samples could be analyzed to obtain diffusion coefficients using a Fickian model of diffusion and the experimentally determined dependence of the index of refraction as a function of PTSA concentration. The values of the diffusion coefficients were 10−14 to 10−12 cm2/s for temperatures between 50 and 95 °C. The activation energy for diffusion below the glass transition temperature was 67 kJ/mol. At 95 °C, just abov...

Latent image formation: Nanoscale topography and calorimetric measurements in chemically amplified resists

The characterization of photoresists during the image formation process has traditionally relied on bulk methods during or after development. This article shows that it is not necessary to develop the photoresist in order to obtain significant information about the image formation process in x‐ray, electron beam, and UV lithography. The characterization of the image formation process in chemically amplified photoresists prior to development is difficult due to their sensitivity to electrons used in scanning electron microscopy and the weak changes in dielectric constant needed for optical microscopy. The advent and development of the atomic force microscope (AFM) have allowed local surface measurements of exposure induced changed in photoresists with sub‐μm scale resolution and negligible modification of the sample. A series of chemically amplified resists, positive and negative, have been studied with a variety of exposure radiation (electron beam, x‐ray, and UV). The results discussed in this article fo...

Photoresist and the photoresist/wafer interface with a local thermal probe

This paper reports the development of local differential thermal analysis and ellipsometry to measure the glass transition temperature (Tg) of thin films of photoresist. We apply the techniques to measure the glass transition temperature as a function of the film thickness of polystyrene and poly(methyl methacrylate) (PMMA). We also study the effect of the duration of post apply bake and the extent of reaction on the Tg of films of a negative photoresist, SAL605. Measurements of the Tg as a function of film thickness of polystyrene cast on native silicon oxide substrates primed with hexamethyldisiloxane show that the Tg is depressed incrementally as thickness decreases to a maximum of Tg - 20 degree(s)C for film thicknesses below 80 nm. Films of PMMA cast on native silicon oxide show a similar depression in the Tg of 10 degree(s)C for film thicknesses less than 70 nm. Our study of SAL605 photoresist finds that the Tg is not influenced by the condensation reaction between the crosslinker and resin. We find that there is a strong plasticizing effect by the residual solvent in SAL605 over short PAB times that can change the Tg by as much as 15 degree(s)C. The Tg reaches a steady value after 30 seconds of baking at 90 degree(s)C.

Using alicyclic polymers in top surface imaging systems to reduce line-edge roughness

Top surface imaging (TSI) systems based on vapor phase silylation have been investigated for use at a variety of wavelengths. This approach to generating high aspect ratio, high resolution images held great promise particularly for 193 nm and EUV lithography applications. Several 193 nm TSI systems have been described that produce very high resolution (low k factor) images with wide process latitude. However, because of the line edge roughness associated with the final images, TSI systems have fallen from favor. In fact, top surface imaging and line edge roughness have become synonymous in the minds of most. Most of the 193 nm TSI systems are based on poly(p-hydroxystyrene) resins. These polymers have an unfortunate combination of properties that limit their utility in this application. These limiting properties include (1) High optical density (2) Poor silylation contrast (3) Low glass transition temperature of the silylated material. These shortcomings are related to inherent polymer characteristics and are responsible for the pronounced line edge roughness in the poly(p-hydroxystyrene) systems. We have synthesized certain alicyclic polymers that have higher transparency and higher glass transition temperatures. Using these polymers, we have demonstrated the ability to print high resolution features with very smooth sidewalls. This paper will describe the synthesis and characterization of the polymers and their application to top surface imaging at 193 nm. Additionally, it will describe the analysis that was used to tailor the processing and the polymers physical properties to achieve optimum imaging.

Local thermal analysis of reaction and the glass transition in exposed resist

We report the results of an in situ study of reaction and the glass transition in exposed films of positive chemically amplified resists. Local thermal analysis was performed on exposed patterns in thin films of APEX-E. We measure the activation energy of reaction int he exposed resist to be 132 +/- 5 kJ/mol. Post-exposure delay effects were shown to reduce the reaction rate in the pattern by as much as 14 percent. Finally, we used local thermal analysis to determine the effect of reaction on the glass transition temperature in exposed APEX-E. We find that the glass transition temperature increased linearly by as much as 20 degrees C after complete de-protection matrix. We present a detailed examination of the use of differential thermal analysis to study reaction and the glass transition with a local thermal probe.

Remarkably efficient acid generation in chemically amplified resist from quantum chemistry modeling

The authors apply methods of quantum chemistry to obtain information on the equilibrium geometrical structure, IR and UV spectra of the photoacid generator (PAG) di(phenylsulphonyl) diazomethane (DPSD) and products of its photoinduced decomposition, and UV spectra of polyhydroxystyrene (PHOST) and poly(4-t-butoxycarbonyloxystyrene) (PTBOCST), and to characterize a viable alternative path for photoinduced decomposition of DPSD in PHOST-like resists. In contrast with previously published sulphone-based chemistry, the newly suggested path for acid generation results in formation of up to two acid molecules per single PAG molecule. The theoretical results are quantitatively compared with available experimental data.

Application of scanning thermal microscopy to the study of thermophysical properties of ultrathin photoresist films

Monte-Carlo simulation methods are used to investigate the glass transition temperature measurements from a scanning thermal microscopy study of thin photoresist films. We find that, consistent with our own experimental observations, film thickness has a profound effect on the glass transition temperature. Depending on whether the film is confined or not, we observe an increase or a decrease of the glass transition temperature. These findings are explained in terms of structural changes occurring at the molecular level.