Chemically-amplified positive DUV photoresists are well known to exhibit small profile deviation at the resist substrate interface, commonly called footing, when processed on substrates like silicon oxynitride (SiON), titanium nitride (TIN), and boron phosphorous silicate glass (BPSG). Severe footing can cause etch problems resulting in critical dimension (CD) nonuniformity and degraded lithographic performance. The objective of this paper is to examine possible solutions to footing on SiON substrates by focusing on three main areas: photoresist formulation, photoresist processing and substrate manipulation. The differences in photoresist formulation lead to various footing results. Three Shipley(TM) DUV photoresists, APEX-E 2408(TM) POSITIVE DW PHOTORESIST, UV5(TM) POSITIVE DUV PHOTORESIST and UV6(TM) POSITIVE DUV PHOTORESIST are examined to benchmark the performance on a baseline 300 Angstrom SiON film deposited on silicon. The formulation study is designed to discover the main differences in the photoresist that cause different amounts of footing by examining four areas: photo-acid generator (PAG) type, FAG loading, additive loading and polymer type. Photoresist processing is studied to understand the main impact of several different processing steps on footing. The studies are performed with one or more of the following Shipley Positive DW Photoresists, APEX-E 2408, UV5 and UV6 with a 300 Angstrom SiON film. The main process steps examined are developer time, photoresist thickness, and softbake and post-exposure bake (PEB) temperature. Prime temperature and dehydration bake after prime are also examined. The develop studies are performed to determine if longer develop times or would erode the foot. Photoresist thickness is examined to determine if the foot size is affected by the position on the swing curve. Softbake and FEB are examined to determine if manipulating bake conditions reduces footing. The substrate priming and dehydration bake studies are performed based on success in minimizing footing on BPSG by using a dehydration bake after prime.(1) The properties of the substrate are closely examined to determine the effect on interfacial footing for APEX-E 2408 and UV6. Although SiON ARL layers have been shown to offer significant improvements in photoresist patterning and yield,(2) DUV resist performance can be further improved by reducing the amount of footing. Factors that can contribute to the amount of footing include substrate type, surface treatment, substrate thickness and substrate reflectivity. The surface treatments examined include an oxide cap, a 10-second Post-Ash and several nitrous oxide plasma treatments, which vary the gas flow, power, and time. The studies on SiON film thickness and reflectivity show the greatest effect on photoresist footing. By changing the SiON film thickness from the reflection minimum of 300 Angstrom to the reflection maximum of 550 Angstrom, the footing is reduced by more than half for APEX-E 2408 and nearly eliminated for UV6. The use of a nitrous oxide plasma treatment reduces the foot by more than a factor of three. The combination of a surface treatment with the reflection maximum of 550 Angstrom completely eliminates all traces of footing for APEX-E 2408 and UV6. This work demonstrates that there is an interaction between the surface induced foot and the effective optical dose (reflectivity). Clearly, the purpose of an ARL is to minimize CD swing by reducing reflectivity. A 300 Angstrom SiON baseline film has a swing ratio of 5.3% for UV6 which increases to 17.7% for a 550 Angstrom SION film. This study gives additional information to minimize the overall CD variation caused by both footing and CD swing.
