We report on the solubility characteristics of poly‐(methyl methacrylate) electron‐resist for developer combinations of methyl isobutyl ketone (MIBK) and isopropyl alcohol (IPA). The solubility rate is determined in terms of the fragmented molecular weight which is a function of original molecular weight and the energy absorbed by the polymer in degrading to a lower molecular weight. An empirical formula for the solubility rate is R=false[R0+β/Mnormalfαfalse]e−Enormala/normalkT , where Mnormalf is the fragmented molecular weight, R0 , β, and α are constants characteristic of a given developer. The temperature dependence is characterized by an activation energy, Enormala , for a given developer. The solubility kinetics are discussed showing a rate limited process for most developer conditions. From the solubility rate and the characteristic energy absorption in the resist, the development time and contrast are defined and calculated for a variety of exposure parameters including beam energy, resist thickness, developer, developer temperature, and original molecular weight. Compared to the standard developer, 1:3 MIBK:IPA, an improved sensitivity of 10–50 times is obtained using MIBK at the expense of increased development time. For a given contrast level the development time is greatly reduced by increasing the developer temperature. A lower original molecular weight also reduces the required development time.

No abstract

No abstract

2o 20 -9 -88 MOLE% BCES/12 MOLE %COC -z~ -75 MOLE %BCES/25 MOLE %COC RLM THICKNESS-45Onto , , , i I , , , , l , 0 5 I0 ABSORBED ENERG'Y (milliJoules/cm 2 ) Fig. 4. Cholesteric-isotropic transition temperature vs. absorbed u.v. energy.pliers. 2 Purification was deemed unnecessary because it was initially established that the isotropic transition temperatures were in good agreement with literature values, which is an indication of high purity. Further evidence for sample purity is the high resistivity of the stilbenes, which was of the order of 1011 ohm-cm. The effect is tentatively attributed to a trans->cis conversion of the stilbenes in which the u:v.-generated cis-stilbene acts as an impurity which depresses the liquid crystal-isotropic transition temperature. The role of impurities in the depression of liquid crystalisotropic transition points is well known (6) and evidence to support the assumption that a trans-*cis conversion takes place is given in Fig. 5, which shows a comparison of the u.v. absorption spectrum of BCES before and after irradiation with the known spectra of nonmesomorphic transand cis-=-chloro-4-methylstilbenes (7).

The time evolution of exposure contours in poly-(methyl methacrylate) (PMMA) electron resist is calculated for several time-dependent developers comprised of various proportions of methyl isobutyl ketone (MIBK) and isopropyl alcohol (IPA). Contours are calculated for a line of charge incident on 4000 Å of resist coated on a low-atomic-number substrate. Parameters such as beam energy, incident charge per unit length, original molecular weight, and developer temperature are considered. The use of a time-dependent developer greatly influences the shape of the developed contours and the development time. The more active the developer (expressed in terms of the PMMA solubility rate), the greater the influence of developer time on the contours. Use of higher beam energies results in developed contours confined closer to the beam axis and less sensitive to development time. The incident charge per unit length is very important. For the MIBK developer and at low doses the characteristic undercut phenomenon is not obtained except for long development times, resulting in an increase in linewidth and a reduction in original resist thickness. The principal result of using lower original molecular weight and/or higher developer temperature is to reduce the development time for a given contour.