QUANTIFICATION OF THE EXTENT OF REACTION IN A NEGATIVE, NOVOLAC-BASED, CHEMICALLY AMPLIFIED RESIST

Citation
Pm. Dentinger et Jw. Taylor, QUANTIFICATION OF THE EXTENT OF REACTION IN A NEGATIVE, NOVOLAC-BASED, CHEMICALLY AMPLIFIED RESIST, Journal of vacuum science & technology. B, Microelectronics and nanometer structures processing, measurement and phenomena, 15(6), 1997, pp. 2632-2638
Citations number
24
ISSN journal
10711023
Volume
15
Issue
6
Year of publication
1997
Pages
2632 - 2638
Database
ISI
SICI code
1071-1023(1997)15:6<2632:QOTEOR>2.0.ZU;2-J
Abstract
A technique for determining the extent of reaction and the molar absor ptivity of the Fourier transform infrared (FTIR) peak created during t he postexposure bake (PEB) reaction of the negative e-beam/x-ray Shipl ey resist, SAL 605, is described. Wafers were oven baked within a gas- tight bomb. The product of the linking reaction, methanol, was quantif ied from the gas in the bomb and within the resist film after the PEB. The ether peak, corresponding to the reaction between hexamethoxymeth ylmelamine (HMMM) and novolac was measured on the same wafer using FTI R. In addition, the absolute number of HMMM molecules reacted was meas ured by,eel permeation chromatography to yield the number of moles met hanol produced/HMMM reacted. The molar absorptivity for the ether peak , resulting from a reaction between an HMMM and a phenolic site on the novolac is 3.14 +/- 0.53 x 10(5) cm(2)/mole. Under conditions suffici ent for imaging in a 0.5 mu m thick film, the average number of reacti ons/HMMM molecule is 1.27 +/- 0.24. From previous work on the acid con centration in this resist [P. M. Dentinger, C. M. Nelson, S. J. Rhyner , J. W. Taylor, T. H. Fedynyshyn, and M. F. Cronin, J. Vac. Sci. Techn ol. B 14, 4239 (1996)] each acid moiety created was found to catalyze an average of 26 +/- 8 events during the PEB process. It appears that no more than 1.5%-2% of the available phenolic sites in the film need to be reacted for the required differential dissolution rate, and that the film does not crosslink. The ramifications of this work on modeli ng this type of resist will also be discussed. (C) 1997 American Vacuu m Society.