The InnovationQ application will be updated on Sunday, May 31st from 10am-noon ET. You may experience brief service interruptions during that time.
Browse Prior Art Database

Ta2O5/Si3N4 Hard Mask/BARC for 248nm Lithography

IP.com Disclosure Number: IPCOM000016338D
Original Publication Date: 2002-Sep-19
Included in the Prior Art Database: 2003-Jun-21

Publishing Venue



Disclosed is a combination hard mask and bottom anti-reflection coating (HM/BARC) for use in lithography at 248 nm. This coating is composed of Ta2O5/Si3N4. Its intended use is in "image transfer" patterning. In the "image transfer" method of fabricating thin film structures, a "hard mask" is patterned using photo lithography and RIE (Reactive Ion Etching). Subsequently, this hard mask image is transferred to a base layer or other underlying materials (oxides, metals, cured polymers) with a different RIE chemistry. Prior art includes the use of Ta2O5 as a hard mask for the above process. Recent interest in patterning Ta2O5 with 248nm lithography requires an anti-reflection coating. Such an anti-reflection coating aids in the lithography process control by making the lithography process less sensitive to variations in resist, hard mask and base layer thickness. The currently disclosed structure is a bilayer Hard Mask that also acts as a bottom anti-reflection coating (BARC) at 248nm. The first layer is Ta2O5 and the second is Si3N4. These materials both act as a hard mask and can be sputtered in the thickness needed at low temperature. As a demonstration of its effectiveness experiments with this structure have been carried out and summarized in figure 1, below. The "Base Layer" is a cured photopolymer, and the HM/BARC combination is 1740 Å of sputtered Ta2O5 with 221 Å of Ion-Beam deposited Si3N4 (nominal composition). This Si3N4 has high refractive index and a k which is between those of the photoresist and HM layers making it an ideal film. The "swing curve" chart below shows the reflectance (in air) of a photoresist/BARC-HM/Baselayer stack. This is done for wafers both with and without the Si3N4 component. As can be seen, the reflectance swings are greatly suppressed by the addition of the BARC layer. In principle, an experimental "tuning" of the Si3N4 thickness should be able to reduce this reflectance swing from the current 2% to less than 0.5%.