human traffic

i read a great article today in the january issue of physics today - on the subtleties of safety designs in cars and how light trucks are a serious danger to others on the road. i always like it when physics people use reason and analytical thought to shed light on subjects that are fraught with murky thinking and ill-conceived myths. so i thought i'd link to it! it's a great read.

also i went to a talk on quantum dots today by ranamurti shankar - not the musician, the guy who wrote that book on quantum mechanics that everyone uses - on the application of a statistical method, random matrix theory, to the analysis of quantum dots. it was a cool talk and he's an engaging speaker - i asked a question at one point, but it was pretty nerve-racking to speak up in front of a room full of smart people about stuff they know very well. i don't know why i'm so self-conscious but it sucks.

anyway, i have to trawl the web now and look for some funny to inject into my talk next week. hope i find some!

hee hee!

so, my research is about studying defects in photonic crystals. a photonic crystal (thank you, britney) is a synthetic material that is made of a periodic lattice of some kind of material. for example, one type of material that we use a lot is a lattice made of silica (glass) spheres, with air gaps between them, and arranged in a hexagonally close-packed or face-centered-cubic structure. this material is called a synthetic opal, because it has the same structure as precious opals found in nature.

so what makes it "photonic"? well, if you make the spheres approximately the same size as the wavelength of some light, then that light won't propagate through the lattice along certain crystal planes. this is called a photonic bandgap and is analogous to the electronic bandgap in semiconductors that makes them.. well, semiconductors. this optical bandgap can be used to confine light inside a structure, which can be very useful in the design of integrated optical circuits.

the point of all this is that the technology could have some seriously kick-ass uses, like optical computers that run at terahertz clock speeds. have you noticed that for the last few years, processor speeds have largely plateaued at around 3-3.5 GHz? for a while there, it seemed like every six months intel would release a new chip with a faster clock - processor manufacturers were racing to push their clock speeds faster and faster. not so much anymore - they are running up against some pretty fundamental limitations of semiconductor technology. most people think it's unlikely that anyone will be able to make an electronic processor that runs faster than 10 GHz.

one reason is that as you crank up the clock speed, the processor runs hotter due to resistive losses in the circuits. so intel and others have compensated by adding ever-larger heat sinks and fans, but those are only a stopgap solution. another problem is size - as the circuits get smaller, the current-carrying paths get closer together. so close, in fact, that the electrons which carry the bits actually tunnel across the gaps and into places they're not supposed to be. a processor that works on light pulses instead of electrical pulses doesn't face these technological limitations (though there are plenty of limitations imposed by optical circuitry!)

so, anyway. that's the why - here is the how. we are going to use a nanomanipulator - sort of like a MEMS device, with piezoelectric actuators, all crammed inside an electron microscope - to modify the structure of a photonic crystal. we should be able to make a waveguide - like a current-carrying path - in order to guide light through the crystal and around corners, or introduce point defects that will act as small optical cavities. if we can get the defects right, we'll have succeeded in not just slowing light down, but stopping it entirely and localizing it in one spot. this nanomanipulator is being built in the group of sun yu, who is one of my supervisors on the project.

so, i was reading one of the papers he sent me, which is about using a probe attached to the end of one of these nanomanipulators to measure position in an array of carbon nanotubes. here is a picture of the probe. when i saw it, i burst out laughing.. i have no idea who designed it or what possibly could have been on their mind, but who says science ain't sexy? i just might have to submit this to maxim's unintentional porn department, coz it's perfect. anyway, next time, i'll post some more detailed stuff about what i'm doing in the lab.

i booked a flight to france today - going to visit andy and adeline in poitiers from the 17th of february to the 27th. i haven't seen andy in like 5 years and i'll be meeting adeline, who is no longer just his girlfriend, but his wife. and their dog, marvin or chester or something. surreal, if you know andy.

ok, this post is now stupid-long.. so i'll stop. bye!

dog jerk

so tonight i went to second cup at north york centre and read part of an issue of the economist (i always seem to be 2 weeks behind in reading this damned magazine - it's too good to skip anything and too dense to just plow right through in one sitting - love it, love it love it!). anyway, at one point i went outside for a smoke, and i noticed this poor little dog who was tied with his leash to an iron railing next to the building. it is cold tonight in toronto; it's apparently -10F and 20 mph winds.

anyway this pooch is sitting there, shivering and whining, and the owner is nowhere to be found. so i started petting him and it was obvious he was freezing, since he starts pushing himself against my body to get some warmth and cover from the wind. i knelt there for a while with him until he seemed a little warmer, and then went inside and started asking people if the dog was theirs. while i was asking around, someone pointed out that the owner had shown up, so i went back out and gave the guy a hard time about leaving his dog out there to freeze. he didn't say anything and just walked away, but hopefully he felt bad about it and won't do it again.

it really bothers me when people do shit like that. this dog was some little beagle mutt - couldn't have weighed more than 15 pounds and only had a thin layer of fur, and was attached to the fence with a short enough leash that he could only move in a 3-foot space. he couldn't even run around to keep himself warm - just had to sit there and shiver. biologically this dog's no more robust than a two-year-old human child in a sweater, and nobody would think about leaving one of those out there in this cold for a minute, much less a half-hour.

people can be so irresponsible with their pets - we force them out of nature and into our lives, demand that they abide by our rules and our schedules, make them dependent on our care and then we do shit like this where they are prevented from even taking care of themselves in the most basic of ways. for all of the sanctimonious preaching about personal freedom that people carry on, we incarcerate these animals for the most frivolous of reasons - which i guess would be alright if they were well cared-for, since they'd live a safer and more-comfortable life than in nature - but most people fail to uphold their end of the bargain. there oughta be a test you have to take.

anyway, that's enough of a lecture. have to knuckle down - i'm supposed to give a talk at the group meeting on the 25th and most people usually present the results of their research. being a twit, i have no results to present, so i'm not sure what the hell to talk about. they say "an off topic will do" so i might talk about accelerators and try to relate the subject to materials science via synchrotron light. if i can get it to work by then, i'll maybe talk about the elastic modulus measurements i've been trying to do on kai landskron's PMO thin films with gilbert's AFM. i think my own research is so tenuously-defined right now that i don't really know what to say about it yet.

i'm gonna try to add more science content to this blog, because, hey, i'm nerdy like that!

asking for trouble

spring term starts next week - the department opens tomorrow. i'm thinking about taking this guy's course on quantum dots. lots of people in my group use quantum dots and i don't really understand what the big deal is, so i think it'll be useful if not directly applicable to my research.

so lately i have been spending a lot of time on yahoo! answers (my user name: natelipkowitz), which is a site like ask metafilter where users can post questions or answer others' questions. it's a fun place for know-it-alls like me to go and get that narcissistic warm and fuzzy feeling of having your answer chosen as the best - affirmation for those who need it, i guess. it's addictive, and i think the deeper (or maybe just better-sounding) reason is that it helps me gel me thinking on certain topics. i like to answer questions in the science and politics categories. usually when i answer a question i do some research to support my answer, and in forming the response i'm forced to synthesize what i know with what i've learned. it is definitely true what they say - the best way to learn something is to teach it.