E-Beam Lithography, Part 1

hi I'm Scott brasswell with the nanotech user facility at the University of Washington and today we're going to do a training video on ebeam lithography lithography in general is a way of transferring a pattern um we use two different kinds of lithography to transfer patterns of very small size photolithography and ebam lithography or ebl for short photolithography uses light to expose a photosensitive resist similar to exposing silver Emulsion in the photographic process light shes through a photo mask like a transparency and the pattern of the Mask is reduced and transferred to the resist after a short bath and developer you can see the pattern in the resist layer with this method we can generate patterns many millimeters down to single pattern elements of about 2 Micron the pattern transferred by photo lithography will have the same thickness as the resist layer had that's useful because we can backfill the lines and holes in the pattern with metals or oxides or nitrides to grow a pattern with the chemical and physical properties that we want alternately we can etch the pattern into the Silicon wafer itself and create deep troughs and holes where the resist was exposed in order to create pattern elements with nanoscale Dimensions we need the resolution of an electron beam in ebam lithography we use the Electron Beam of an sem to copy our pattern onto a wafer cated with resist it's a Serial process like tracing a picture out of a book and so it takes much longer than a parallel process like photolithography the first step in ebl training is to learn the software uh generate a design CAD file for the pattern that you want and then a run file for the exposure conditions that you want so that can be done on your own computer once you've installed the software the second step is to come in and prepare the substrate once the substrate is ready uh you can book some sem time uh come in and actually do the writing on the machine the first step of the ebl process is preparing the substrate so I start with a a clean silicon wafer and I'm going to rinse it first with acetone uh to remove any impurities then isopropanol and uh finally a distilled deionized water rinse and then we'll dry it so the first step is the acetone rinse before before the acetone dries you rinse it with isopropanol to make sure there's no acetone residue and then with distilled deionized water and before the water dries I use dry nitrogen to blow all the water droplets off of the surface both the back and the front once this wafer is clean I can spin coat it with the ebeam sensitive resist and make sure that the bonding to the surface is good the thickness of the resist layer on the wafer can be controlled by playing with the viscosity of the resist that we're applying and also the speed that it's spin spinning at so to apply the resist layer I'm going to use this spin coating machine here um that has a vacuum Chuck on top I just set the clean wafer in the center and then I'll test to see that the vacuum is holding by starting the machine if the wafer starts to spin then the vacuum is good so I'm going to interrupt that process and now I'm going to apply the resist in this case we're using a 3% polymethyl methacrylate resist the resulting layer for spinning this at 4,000 rpm should be about 300 nanm I'm going to apply an excess of the resist and and the extra volume will all spin off into the bowl each pipet full is about 1 ml so for an entire wafer I'm going to use 4 ml as the thickness of the coating changes you'll see the color of the wafer also changes until we reach our final thickness and the resulting W for a 300 nmet stick should be a green color full spin time is 45 seconds now I'm going to transfer that wafer to a hot plate at 180° for 1 minute that'll drive off the excess solvent and solidify the resist after 1 minute I'm going to pull the wafer from the hot plate let it cool and then before we pattern the surface we'll break this into smaller pieces the chamber size on our system is limited to about 5 cm uh in the X and the Y Direction ction uh so we can't write on this entire wafer so the next thing I'm going to do is break it into smaller pieces uh so that it fits into our system this wafer is single Crystal silicon and uh it'll break it's so pure it'll break along a a perfect cleavage line in the crystal plane if I can just get a fracture started it'll propagate all the way through and so I can cut a perfectly square piece from this the ideal size for our chamber is about 1 cm squared the surface of this resist is featureless so in order to have something to focus on so that I know that the resist is at the right height I'm going to carve numbers into the corners on this chip they don't need to be deep enough that they score the Silicon they just need to be through the resist layer so I'm numbering the corners clockwise 1 through 4 that'll give me a reference point when the chip is in the system in addition to loading the chip I need two standard samples first I'll load gold nanoparticle substrate because it has very good contrast and we can use it to optimize the beam before writing it's kind of like sharpening your pencil before you draw a picture and the second sample that we need is called the Faraday cup and it's basically a 75 Micron aperture that collects the entire beam so that we can measure the amount of current that's coming through from that current we can calculate the dosage that we need to expose the resist fully and one last step before I load this into the chamber I want to blow the debris from the surface of the chip uh that might have landed there from the describing the numbers into the corners

Channel: cmditr

Views: 89,704

Keywords: E-Beam Lithography, University of Washington, Scott Braswell, Photonics

Id: bvgITKqYpuY

Length: 9min 30sec (570 seconds)

Published: Tue Apr 06 2010