Browse Prior Art Database

A HIGH CONTRAST POSITIVE E-BEAM RESIST PROCESS

IP.com Disclosure Number: IPCOM000006184D
Original Publication Date: 1991-Jul-01
Included in the Prior Art Database: 2001-Dec-12
Document File: 4 page(s) / 154K

Publishing Venue

Motorola

Related People

C.M. Brown: AUTHOR [+2]

Abstract

The application of direct-write E-beam lithogra- phy using an AEBLE 150 depends heavily upon resist sensitivity. If the resist is not sensitve, large layer to layer misalignments can result due to the space charge regions set up in the resist patterning layer from the large deposited charge. This charge induced misalign- ment effect is especially prevalent when positive resists are used to define contacts or via cuts in moder- ate to thick dielectric layers.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 45% of the total text.

Page 1 of 4

0 M

MOTOROLA INC. Technical Developments Volume 13 July 1991

A HIGH CONTRAST POSITIVE

E-BEAM RESIST PROCESS ~

by C.M. Brown and J.N. Helbert

  The application of direct-write E-beam lithogra- phy using an AEBLE 150 depends heavily upon resist sensitivity. If the resist is not sensitve, large layer to layer misalignments can result due to the space charge regions set up in the resist patterning layer from the large deposited charge. This charge induced misalign- ment effect is especially prevalent when positive resists are used to define contacts or via cuts in moder- ate to thick dielectric layers.

  Positive E-beam resists are notoriously insensitive (see Table I), especially the higher contrast (or resolu- tion) systems that are dry-process compatible. Of course, sensitive E-beam resists that are not dry-pro- cess compatible are not usable for making integrated circuits. The bi-component DNQ resists of Table I also function as classical photoresists, hence, they generally require larger E-beam exposures or deposited radiation charge to function.

  In 1990, Dammel et al(l) reported a tri-component resist for E-beam and X-ray applications which func- tions via a chemical amplification dissolution inhibitor reduction mechanism. For this system, the dissolution inhibitor component is chemically decomposed by the acid produced via radiation induced acid formation from the photoacid generating component of the resist. Since the acid produced is not consumed in the disso- lution inhibitor reaction, it can carry out much more solubility inhibitor reduction in the exposed area per unit dose than that which occurs for the classical sys- tems. Hence, this system, Ray-PF, is very sensitive and overcomes the sensitivity problems of the systems of Table 1. The sensitivity of Ray-PF depends on the substrate and the feature size, but typical exposures range from 0.5-S pC/cm', at least an order of magni- tude better than for the corresponding conditions for the most sensitive system of Table I. Ray-PF contrast values are so high they are hard to quantify, but values

>25 are typical; this level of contrast value far exceeds those for the systems of Table I.

  In 1989, an early coimnercial sample of Ray-PF was received for testing bn the Motorola AEBLE 150 system in ATC. The p:ocess recommended by the vendor was not a high performance or contrast pro- cess, and could not be employed to do high resolution E-beam lithography without extensive process opti- mization. Over approxiinately the next year, a high performance process was developed for AZ Ray-PF by employing statistic$lly designed experiments. They were carried out toiprovide a contrast optimized and a high contrast/resolutmn process for this system. After running several response surface experiments, central composite designs with 2-4 variables, a we11 optimized high contrast p;ocess was developed as ver- ified by the high resolutidn images of Figure 1 and the stable and optimized response surface...