{"id":1180,"date":"2014-11-14T01:29:57","date_gmt":"2014-11-13T23:29:57","guid":{"rendered":"http:\/\/antoine.wojdyla.fr\/blog\/?p=1180"},"modified":"2015-02-01T01:52:20","modified_gmt":"2015-01-31T23:52:20","slug":"seeing-things","status":"publish","type":"post","link":"http:\/\/antoine.wojdyla.fr\/blog\/2014\/11\/14\/seeing-things\/","title":{"rendered":"Seeing small"},"content":{"rendered":"

I don’t know how to do this on a small scale in a practical way, but I do know that computing machines are very large; they fill rooms. Why can’t we make them very small, make them of little wires, little elements \u2013 and by little, I mean\u00a0little. For instance, the wires should be 10 or 100 atoms in diameter, and the circuits should be a few thousand angstroms across.
\n\u2013 Richard Feynman (1959)<\/p><\/blockquote>

Being\u00a0 a researcher, all my friends and family have very little idea of what I’m work on. The trouble is that when I try to explain, I need to resort to notions such as wavelength or transistor, only to discover that most people have no clue of what this is, expect to joke about being on the same… wavelength. But very small wavelength.<\/p>To put it straight : I look at small<\/span> things.<\/strong><\/p>Yet, I was just awarded a prize for best research at the EUV lithography symposium in Washington, D.C., and it’s only fair of me to explain what I’m doing for a living!<\/p>\n

\"sharp_photo\"<\/a><\/p>\n

The SHARP EUV microscope I’m working with<\/p>\n<\/div>

 <\/p><\/p>\n

I am working at the Center for X-Ray Optics, in a team specialized in EUV lithography. For a starter, let’s explain what this is !<\/p>\n

EUV Lithography<\/h2>

(Photo-)lithography is the art of making microchips. As complicated as they might be (they now feature billions of transistor), fabricating microchips is not from photography.<\/p>It all starts with a silicon wafer, which is somehow just a piece of highly purified and melt send that has the cool property of being a semi-conductor. The wafer is then coated with what is called a photoresist, which is a sort of resin that gets harder when lit with ultra-violet light. Using this cool property, one can “print” any kind of\u00a0 pattern on the wafer. This is done using a photo-lithography tool (which is a fancy inverted microscope) and a photo-mask (the equivalent for vinyls would be a goldmaster<\/em>, though in this case it is pressed against the product), whose image will be replicated on the photoresist. <\/em>The resist is then chemically processed, in order to leave polymer where it was exposed, and nothing everywhere else.<\/p>Then the wafer can be bombarded with ions to alter its electrical properties-this is called doping-, or some metal can be deposited, the photo-resist left being here to locally prevent these changes, forming some kind of an electrical circuit (well, it’s way more complicated than that, but you get the idea).<\/p>Repeating this process with different patterns and doping, one can fabricate\u00a0 very complicated electrical functions and finally get a working microchip.<\/p>\n

*
\n*\u00a0\u00a0\u00a0 *<\/p>As a rule of thumb, one can print features that are about the size of the wavelength, due to a physical limit called diffraction.<\/p>\"lens<\/a><\/p>As you try to fabricate smaller and smaller feature to satisfy the Moore’s law<\/a>, you need to have smaller and smaller wavelength. Current photo-lithography tool are using deep ultra-violet light to expose the wafers, with a wavelength of 193 nm. However, they are now printing features that are about 50 nm in dimension. The industry was able to keep using the same wavelength for a while by using immersion lithography<\/a> and multi-patterning<\/a>, but the options are becoming more and more limited.<\/p>That’s where my work comes into play. To keep up with the pace of progress, the principal solution is to change the wavelength to 13.5nm, in the extreme ultra-violet range (EUV).
\nMany new issues arises : dimensions are so small that everything becomes a disturbance : temperature, vibration, or even the granularity of matter.\u00a0 (I’ve made a presentation for public outreach, available here<\/a>)<\/p>My work is focused on the inspection of EUV photo-mask, which is one of the main issues for EUV lithography : a single defect on a mask end up printed on every wafer going out of the fab. That’s why we need new techniques to spot them (they’re as small as a few microns) to be able to repair them.<\/p>\n

The SHARP EUV Microscope<\/h2>

I am working in the SHARP team<\/a>. SHARP stands for
\n“SEMATECH High-NA Actinic review project” where :<\/p>\n