Premium
Progress in deep‐UV resists using CARL technology
Author(s) -
Leuschner R.,
Borndörfer H.,
Kühn E.,
Sebald M.,
Sezi R.,
Byer M.,
Nölscher Ch.
Publication year - 1992
Publication title -
polymer engineering and science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.503
H-Index - 111
eISSN - 1548-2634
pISSN - 0032-3888
DOI - 10.1002/pen.760322103
Subject(s) - resist , photoresist , materials science , laser linewidth , lithography , stepper , photolithography , optics , nanotechnology , optoelectronics , analytical chemistry (journal) , laser , chemistry , organic chemistry , physics , layer (electronics)
Three positive working Si‐CARL resists for bilayer applications with oxygen‐RIE pattern transfer were investigated, and their lithographic performance at deep‐UV exposure was compared. With all three we obtained good focus latitudes for 0.35 μm lines and spaces ranging from 1.6 to 2.2 μm exposed with an 0.37 NA KrF‐excimer laser stepper. The zero bias exposure dose required for resist R1, of the diazodiketone type, is relatively high (70 mJ/cm 2 ), but R1 has the advantage of not suffering from linewidth fluctuations caused by post‐exposure delay time effects. Processing for resists R2 and R3, which are based on acid catalyzed deprotection of t‐BOC‐imide and t‐butylester, respectively, had to be optimized to avoid bridging of isolated spaces after Chemical Amplification of Resist Lines (CARL). This bridging is caused indirectly by evaporation of triflate acid during PEB. Resist R2 needs only 8.5 mJ/cm 2 for exposure but has a poor linewidth increase after CARL, which seems to be the reason for eroded 0.25 μm patterns after oxygen‐RIE. Resist R3 shows the steepest resist slopes and the best overall performance. The ultimate resolution for resists R1 and R3 is 0.25 μm, which, according to the Rayleigh equation for resolution, corresponds to a k‐factor of 0.37.