In situ study of hydrogen silsesquioxane dissolution rate in salty and electrochemical developers
(a) Schematic diagram of the sample holder. This holder limited the area of the silicon wafer that was exposed to solution during electrochemical development. The wafer was sandwiched between two acrylic slides using fluorosilicone O-rings to prevent developer solution from contacting the slide on any place other than the 1/4 inch diameter central hole. (b) Schematic of experimental setup for immersion development of the sample with a reference and counter electrode.
Schematic of the QCM apparatus. To perform in situ development rate measurements, a thin layer of HSQ was spun directly onto the active side of the quartz crystal resonator. An oscillating electric field at the resonance frequency was applied to the gold electrodes. Its resonance frequency increased as mass was removed from the crystal.19 For electrochemical studies, a controlled electric potential was applied to the evaporated gold layer beneath the HSQ on the active side. A peristaltic pump was used to control the flow rate of developer solution into the QCM microfluidic module. The solution passed over a temperature sensor and controller before contacting the active side of the crystal. O-rings prevented the solution from contacting the backside of the crystal. In the electrochemistry module, a platinum counter electrode and Ag/AgCl reference electrode were used with a potentiostat to control the applied electric potential. One can access a picture of this apparatus on the manufacturer’s website (Ref. 22).
Top-down scanning electron micrograph of 11-nm pitch nested-L structures in HSQ resist fabricated using electron beam lithography and a standard salty developer process (1% NaOH and 4% NaCl). Note that the waviness in the horizontal direction on the features was due to vibrations during imaging.
Scanning electron micrographs comparing developed resist cross sections with and without a potential applied during development. A Raith 150 electron beam lithography tool was used to expose a 35 nm pitch (top) and 45 nm pitch (bottom) line pattern into the HSQ on highly p-doped silicon. (a) Features developed without an electric potential in 1% NaOH solution. (b) An increase in resist removed between the features was evident when a positive 5 V potential was applied to the substrate surface during development in 1% NaOH solution.
Comparison of the dependence of the development rate on salt concentration and applied potential. The development rate of the HSQ in salty developer and electrochemically-enhanced developer was determined with a QCM. (a) 96 nm of HSQ resist was spun onto a gold-plated quartz crystal. The crystal was mounted in a QCM flow module and developed with 1% NaOH mixed with varying concentrations of NaCl with no applied electric potential. The development rate increased as the concentration of NaCl increased. (b) Similarly, the development rate increased with the strength of the applied positive electric potential in 1% NaOH solution with no NaCl.
In situ observation of the dependence of the development rate on the applied voltage. A quartz crystal was coated with 96 nm of HSQ and then placed in an O2 plasma asher for 30 s at 150 W in order to partially cross-link the HSQ film. A 1% NaOH developer solution was passed over the HSQ in a microfluidic module. A positive electric potential of 2.5 V was applied to a gold electrode beneath the HSQ from 35 to 65 s, causing a sharp increase in the development rate. From 160 to 190 s, 4 V was applied to the crystal.
Resist thickness remaining vs dose. A dose matrix was exposed into a 170 nm thick layer of HSQ on highly p-doped silicon. (a) Samples were developed for 15 s in 0.3% NaOH, 1.3% NaCl salty developer solution, then the resist thickness remaining was measured using a surface profilometer. (b) The samples were developed for an additional 60 s (totaling 75 s) then measured. When an electric potential was applied to the sample, more resist was removed in the first 15 s than was removed with conventional development, indicating an increased development rate. However, after full development, no increase in contrast was observed. Similar results to (b) were observed in the 1% NaOH solution without NaCl, but are not shown, and the scenario described in (a) was not tested without NaCl.
Article metrics loading...
Full text loading...