For the short answer, I reply “I shoot [tag]lasers[/tag] at stuff”. But that doesn’t really say anything at all, does it?

For people with a little more technical knowledge, I might talk about how I make wire grids that are a tenth the width of your hair in the University of Alberta’s [tag]Nanofab[/tag] and then shoot them with lasers. But that’s still not really saying anything.

I could go deeper and say that I’ve made grids of gold wires that range from 5 mm long wires to arrays of 10 micron by 10 micron squares (and various lengths in between), and that I’m trying to use the laser to see how electricity flows in them. But I still wouldn’t be at the meat of it yet.

If I wanted to be a really elitist bastard I could give a jargony talk about how I’m making [tag]subwavelength[/tag] [tag]metamaterials[/tag] for probing using terahertz-time domain spectroscopy in order to extract the complex [tag]conductivity[/tag] of the devices. I could also talk about how I might be observing negative differential conductivity and various resonances. But then I might just be hiding whether or not I know what’s going on (hint electrons are going backwards).

I could also talk about the laser system, which actually is four lasers creating a 1 kHz, 500 ?J, 100 fs (10^-15 s long) pulse, which can ionize air due to it’s large instantaneous intensities. And how we use that pulse and a second-order non-linearity in a crystal of (110) Zinc-Telluride to generate terahertz radiation by the process known as optical rectification. And that detection of this radiation is completed using electro-optic sampling. But all of this would take a senior level physics course to really grasp (as I’m just finally getting it all, and I’ve been working here for 3 years).

So the big problem here, is that I have a highly specialized region of physics – that is [tag]ultrafast[/tag] spectroscopy in condensed matter physics, doing terahertz conductivity measurements of metamaterials – which is quite far removed from the basic freshman and especially high-school physics, and that when I visit with old friends or family, or basically anyone outside my lab, I have to find some way to convey that there is a point to what I’m doing. I can’t say I’m entirely sure what the point is, but the research is cool (to me), and potentially could be useful for some future application, or at the very least for understanding some weird systems that are out there.

Finally, I could also say that I come in, spend 15 minutes starting the laser, wait 15 minutes for it to warm up, spend 10-30 minutes aligning a setup and then spend the rest of the day pushing a big “SCAN” button once every fifteen minutes after moving a couple knobs to adjust my sample around. In the downtime while stuff scans I play on the internet for long periods of time, periodically checking to see if the data is reasonable. And then most days that I don’t have the laser (which is 60-80% of them) I either play with spreadsheets to make data look reasonable, or just plain screw around on the internet some more. But if I put it that way, it sounds less cool, and more monotonous.

So when asked, what do I do, I usually stick with “lasers”.

[tags]science, carnivals, terahertz[/tags]

I’ve heard that the atom is something like 99% empty space, is that right? And also, would that be like our solar system, because I’ve heard it’s a lot of empty space too?

After I answered yes to these two questions (since I was a little taken aback by questions like this in the morning), he said thanks and hung up. My conclusion is he is just very interested in science, and I have to commend that.

But coming back to it, I want to go a little deeper into these questions.

First on the atom. We (generally) all know that atoms are small particles which make up all forms of matter. Atoms are made of electrons, neutrons and protons. The neutrons and protons form the nucleus, or core, of the atom, and the electrons orbit the core (kind of).

Let’s try to get a rough idea for the size of an atom. The volume is known, and is roughly around an Angstrom (or 1/10th of a nanometre depending on the atom). If we assume an atom like Helium, which has two protons, two neutrons, and two electrons, we can start getting some numbers.

The radius of a proton is roughly a femtometre (10^-15 m), and since the mass of a neutron is the same, it is also around a femtometre.

The electron is a bit trickier. Scientists originally guessed the radius of the electron by measuring its electrostatic force with other particles. Unfortunately, the electron is really small and very electrified. Therefore they were likely off, and many modern physicists consider the electron to be an infinitesimally small point particle (or as near as we can achieve). But to get a rough calculation of the “empty space” of an atom, let’s use the classical electron radius of 3 fm (notice that this is larger than the protons and neutrons, which are orders of magnitude heavier, a clue that the electron is likely smaller).

Now, the volume of a sphere is equal to 4/3 times pi times the radius cubed. However, we want the ratio of the stuff inside to the atom, so the 4/3 times pi cancels. Our result is then (for Helium):

(2*((1e15)³+(1e15)³+(3e15)³) )/(1e-10)= 0.0001=0.01%

This means that the helium atom is over (because I did some rounding up) 99.99% empty space!

To further blow your mind, remember that in solids (and especially in liquids and gases) atoms are not like plastic balls touching – they are in fact separated by some finite amount of space.

So when you look around, you’re seeing light (photons) interacting with less than 0.01% of what you see.

Now, how does this compare to the solar system?

First, I’m going to treat the solar system slightly differently. Because all of the planets fall along the same plane I’ll look first at the area of the plane that’s empty, and then at the entire “sphere” of the solar system.

I’m also going to ignore the asteroid belt. I had considered assuming it to be a solid belt, but considering the number of satellites that have passed through without harm, and the fact Ceres, the largest asteroid is under 1000 km in diameter, I feel justified in this. As well the moons of the various planets will be ignored, there are 166 observed, but I feel this will be justifiable by the size comparison. This calculation is meant to be fairly rough and to just give an estimate to the ratio.

I should also say that we do not really have an exact definition for the “radius” of the solar system. It’s speculated that the sun’s gravitational field is experienced 2 light years away in the nearby systems, but this distance would trump any calculations I attempt. I will use the heliopause as a rough outer shell, which is estimated to be 95 au’s away (an au is an astronomical unit, or the distance from the Earth to the sun, 150 million km).

So now we need the radii of the various planets and the sun (all rounded):

First thing we notice is that the sun is more than 99.99% (the ratio is about 1×10^-12) of the area of the stuff in “disk” of the solar system (my planar solar system), the numbers even larger for the “sphere”. So really, to get any idea of how much empty space there is, we could neglect everything except the sun and still have a pretty good estimate!

But to continue, the ratio of the “areas” of the sun to the entire solar system (since we can safely neglect everything else) is around 10^-9. And the ratios of the volumes is around 10^-13. These numbers are unfathomably small. If the solar system were a dartboard and you threw a dart at it, you would have a slightly better than 1 in a billion chance of hitting something.

So just try to keep in mind that almost everything in the entire universe, from the computer (or paper) you read this on to the farthest reaches of space is almost completely nothing.

In this email, Marshall defends Big Bang Cosmology (which is nice, compared to some YECs).  He starts with the story of the discovery of the cosmic microwave background, one of the greatest discoveries of the past century.

Unfortunately, Marshall continues speaking.  He grabs a couple interviews with Robert Wilson (co-discoverer of the CMB):

Robert Wilson was asked by journalist Fred Heeren if the Big Bang indicated a creator.

Wilson said, “Certainly there was something that set it all off.  Certainly, if you are religious, I can’t think of a better theory of the origin of the universe to match with Genesis.”

Which is likely the good PR thing to say when cornered by an interviewer.  Scientists are known for being a bit sheepish with discussing religion, mainly since people who grant funding might look less favourably on someone who actively attacked the religious (especially forty years ago).

I do have to disagree strongly with Wilson though, a theory of origins that matches with Genesis was well accepted through most of the Roman empire, the following dark ages, and right up until the eighteenth or nineteenth century, and that was simply that the six-day, 6000 years ago story of Genesis was literally true, not a scientific theory based on the evidence pointing to a 13.73 billion year old universe.  It’s also worth pointing out the mutually contradictory creation myths of Genesis chapters 1 and 2.

Finally, I’ll touch on the other interview Marshall mined for:

In an interview, Penzias was asked why there was so much resistance to the Big Bang theory.

He said, “Most physicists would rather attempt to describe the universe in ways which require no explanation. And since science can’t *explain* anything – it can only *describe* things – that’s perfectly sensible.  If you have a universe which has always been there, you don’t explain it, right?

“Somebody asks you, ‘How come all the secretaries in your company are women?’ You can say, ‘Well, it’s always been that way.’  That’s a way of not having to explain it.  So in the same way, theories which don’t require explanation tend to be the ones accepted by science, which is perfectly acceptable and the best way to make science work.”

Except science does explain things.  The majority view of philosophy of science is that science does explain things, the theory of evolution is an explanation, just as much as the big bang theory.  Do new theories pose new questions? Absolutely, but that’s part of the fun of science.  Marshall seems to be attempting to hint at an anti-supernatural bias in science, however, in my opinion the evidence just hasn’t lead there (despite times it could).  Science isn’t inherently naturalistic in its assumptions, merely natural explanations are much more plausible than supernatural ones (due to the apparent overabundance of nature, versus the apparent absense of supernatural stuff).

More posts to come on the fun musings of Perry Marshall.

First, by conductivity, we are refering to the ability for current to pass through a material.  Current is moving charges, and the charges that are moving are either electrons (negative charges, which move in a direction opposite to the current – this is just the convention) or holes (positive charges, basically a place in the material where an electron no longer is).  Electrons and holes are known as charge carriers.

In metals the holes are basically stationary, and it is free electrons in the material that move around.  Now the ions (the atoms that are missing electrons) in the metal are very heavy (relative to the electrons), so they make up a background sludge while the electrons move essentially freely, like a gas through the material.

So if we treat the electrons as a free gas of particles against a background drag (caused by the electric attraction between the electrons and the ions), we have the Drude model of conductivity (invented in 1900, all equations will be ommited here but are available on the Wikipedia page).  Applying the classical laws of physics (Newton’s), we can find that the velocity of the electrons in a metal should be proportional to the mobility times the electric field magnitude.

The mobility is a measure of the charge over mass times the mean free time of the charge carrier.  This free time represents the time each electron is free before getting trapped by another ion – but charge will continue to flow, as long as an electric field is applied, since more electrons are liberated.  The free time is equivalent to the mass of the electron divided by the drag coefficient that is felt by the electrons.  This means that the mobility is just the charge of the electron over the drag coefficient.

The physical result of this model is that when you turn a light switch on, it takes at least that mean free time before the current begins flowing between the switch and lightbulb in your room.  However, average times for this are well under a picosecond (for gold, its on the order of femtoseconds).

The Drude model can be extended to AC signals, and fairly accurately represents what happens when pulsed laser light is incident on a sample (either metal or semiconductor).

So where doesn’t it work?  Our lab has found that for thin films of gold (under 30nms), the model fails to account for the conductivity found under terahertz spectroscopy probes.¹ What accounts for the discrepancy then?

An extrapolation of the Drude model, called the Drude-Smith model, attempts to add the effect of the free electrons scattering off of various sources, be they other electrons, atoms, ions, or defects.²  This results in an extra “fudge” factor and is not completely backed theoretically, but fits the data seen from experiments.

The idea with my current experiment, is that by limiting the path length that the electrons are able to travel we can force the demonstration of non-Drude behaviour and can either back evidence behind the Drude-Smith approach, or attempt to discover other processes that are involved.

In today’s experiment I am probing wire grid polarizers.  Essentially wires of 200nm thick gold on a glass (fused SiO₂) substrate that are 10 microns wide. The wires are separated by 10 microns gaps and vary in length from 5 millimetres to 10 microns long.

Since the wavelengths of the pulses I’m sending through are centred around 300 microns, this means at the extreme end of my devices I have subwavelength devices.

By doing terahertz time-domain spectroscopy on these samples, we should be able to figure out the limits for convential polarizers and determine whether (and where) the Drude model of conductivity breaks down.

The unfortunate part of today is that since the final sample mount isn’t finished for the vacuum chamber, which will attach to a cryostat, I have to break vacuum every time I want to move my sample and probe a different length of wire. The lucky part is vacuum is achieved in a mere 5 minutes.

So why use a vacuum for this experiment? Typically terahertz experiments are done in a nitrogen purged environment. The reason for purging (or vacuum) is that water is the sworn enemy of terahertz radiation. More specifically water molecules have several rotational modes that correspond to frequencies in the terahertz range. These modes absorb much of the incident radiation. Essentially terahertz radiation gets damped by damp air. By filling a chamber completely with nitrogen you eliminate most of the water vapour. Using a vacuum chamber accomplishes the same thing but is much cooler (and allows for a cryostat to be used inside as well).

This material is the work that I’ll be presenting at this year’s EngPhys/IEEE/Physics Undergraduate Research Symposium, which I worked to launch last year. After that I’ll hopefully take the work to CUPC.

8:05 am So it’s a national holiday and I volunteered to work today. At least it’s the first time I’ll get to exclusively use the laser in a week, and probably the last time until next week maybe.

I skateboarded to campus today, which at 7:30 am was pretty quick because Whyte Ave was deserted.

Most of campus is abandoned but a few people are around right now.

A speaker on the computer I’m using (nearest to my experiment) is missing so I’m listening to Sonic through a built in tinny speaker.

My goal for today: align the modelocked terahertz spectroscopy setup so I can see a signal. I’ve tried this a couple times before, mostly to no avail. This makes my outlook for the next 6 hours to be pretty dim. At 3:00 pm I’m going with Sonia to the legislature to see the chambers on the one day democracy is open to the public.

8:15 am Laser started and modelocked, output power of 295 mW at somewhere around 30-50 fs pulses (1fs = 10^-15 s). This power is the best it’s been at in a week, and one of the highest I’ve been able to get, maybe the rest of today can go this well.

8:46 am The goal of this experiment is to split the source laser pulse into two, one beam going to the source and generating a terahertz pump beam, the other going through a delay and acting as the probe. Both pulses (the THz pump and optical – 800nm – probe) recombine at the Zinc telluride detector crystal and then go through some splitting optics into a photodetector which I measure through a lock-in amplifier with the computer I’m on that scans the delay (allowing me to measure in real-time the electric field of the terahertz pulse).

As of now I have the probe path almost re-aligned. It seems that every time I go back to this setup it’s completely out of alignment… This doesn’t bode well for the next time I attempt to do this experiment. At the very least, if I’m lucky, today I’ll be able to find the position on the delay stage that corresponds to the overlapping of the two pulses in the detector crystal, the point we call t-zero (t₀).

9:01 am Both beams seem to be aligned now. The probe beam is entering the detector and seems to give a nearly balanced signal. The pump beam is hitting the THz source (a pair of transmission lines with 50 V applied), so the gap reads <5 k ohms, when with no illumination gives a few M ohms. So now I’m scanning the majority of the length of the stage and seeing if any signal jumps.

So while I wait (which is what most of any research in any field is) I can explain a bit of the theory behind this experiment.

The optical pump beam is focussed onto a piece of gallium arsenide, a semiconductor that when illuminated by light below a certain wavelength (816 nm here) it becomes conducting. So with 50 V applied across a small area that is normally insulating, when the optical pulse hits it it becomes conducting. This creates a flow of electrons from one electrode to the other. Accelerating charges create electromagnetic radiation, and in this case it is emitted in the range of terahertz radiation.

This radiation is in between the microwave and infrared regimes, and has several unique properties, such as the ability to penetrate cloth and plastic, while being strongly absorbed by water and reflected by metals.

The set up I’m working on is based around using this radiation for spectroscopic probing of materials. This means we look at the pulse after it travels through free space, and compare it with the pulse after traveling through our material. We then compare the Fourier Transforms (or the frequency spectras) of the pulses and can determine the index of refraction, absorption and conductivity of the sample.

9:21 am Two-thirds through the scan and no signal, I’m going to have to screw around with something in a bit.

One idea I had, but may be hard to implement with such a weak probe beam is to use a different method to source the terahertz radiation. If I put an electro-optic crystal at the source I should be able to generate the same radiation, although it may be too weak to detect.

9:25 am No signal, trying scan for the last bits of the delay.

The delay stage I’m using is 100 mm long, I initially scanned from 15 mm to 85 mm with 20 um steps, which gives a resolution of 133 fs. The pulse I’m looking for is around a picosecond long (or 1000 fs). The delay stage changes the path length of the probe beam, and hopefully (if the optics are positioned right) will cause the beams to pass the same lengths at some point in its scan.

9:40 am Still no signal.

One thing of note, the crystal with the transmission lines has a silicon lens that is pressed against the back of it, unfortunately, on close inspection, it seems the lens has pushed the emitter a bit off of its holder, but luckily not cracking it (GaAs is extremely fragile – and expensive). I also found that my alignment had decided to wander a bit off from where it was on the pump beam, also not a good sign.

I think at this point I’m going to abandon this emitter (for now), as I’ve never seen any good come from it. I should mention that I know this setup can work, since I got a signal with a 2,000 euro emitter that we had on loan to characterize.

I’m going to try a different technique for now, and I will report a bit more on it once I get a scan going.

10:30 am Attempting a new scan now. I’ve changed the source from the transmission lines, which should give a strong signal (8 mV) to a [110] oriented crystal of GaAs, which should produce terahertz by a process called optical rectification.

What happens here is the optical pulse causes the crystal to become polarized, which “looks” like an accelerating current, creating the terahertz radiation.

To set all of this up, I had to first cut the crystal. Unfortunately the crystal supplier gave us square crystals, with the crystal axis not perpendicular to the edges. If you’ve ever cleaved a crystal you’ll realize how much this sucks. Basically, cleaving requires me to scratch one edge with a diamond tip blade, hopefully scoring it along a crystal axis, the crystal then cracks along an axis and splits nicely (think of when you chop wood and it goes right along the grain). Now with this crystal the crack went at about 45o to the edge.

After this I found a washer I could mount my triangular shard to, and drilled a larger aperture. Then I mounted the crystal to the washer with rubber cement (we’re about quality here), and put it in an optics mount.

Finally I realigned the setup, ensuring that the pump and probe beam would overlap on the detector crystal, and then I put the GaAs in place of the source.

I’m now running my scan from 15 mm to 85 mm again. I don’t know if this will really show anything, since the laser I’m using might not have the power to generate the terahertz pulse, but I may be able to detect it using some HgCdTe (mercury cadmium telluride) detectors, but those require liquid helium or liquid nitrogen, which I don’t have access to today (those detectors are also limited to the mid-infrared range, and likely wouldn’t show the main portion of the pulse around 1 THz).

Well back to check my email, Facebook, etc. while I wait for this scan (another major part of research).

10:51 am No signal, no surprise. Mild hunger and frustration begin to set in.

11:09 am I couldn’t find a signal with the prior set up so I’ve moved the generation crystal to the spot where the sample usually goes. This means the THz beam will travel less in air, and therefore be absorbed less by water vapour (which can take 20-30% of the intensity). No guarantees that this will work though. I may have to double check that my delay stage is in the right general position by physically measuring each beam path. I might also eat my sandwiches while I wait too. Work is so exciting sometimes.

11:30 am Still no signals. Sandwiches are a little stale around the edges. And the sudoku I’m working on is taking a while.

11:50 am The pump path length is 67″, the probe path length can be set to 67″ +/- ~5″. It should work.

12:04 pm I returned to using the transmission line, and fine tuned the alignment. It may or may not give anything at all now. Not sure what I plan to do if this fails.

12:49 pm No signals seen at all. Feeling too distracted to get anything meaningful accomplished. I may shut it down now and go read something that will get my blood pressure up.

12:55 pm This ends my liveblogging session.  The lasers off, nothing new is accomplished (aside from mounting a crystal).  Basically this was just another average day in the lab.  I may try this again if I care to.

Note: This is technically not peer-reviewed research yet, it has only been submitted online to a free online paper-publishing site, so holes may yet be found, or not.

What tops my list is hearing the talk from Dr. Lene Hau who used Bose-Einstein Condensates to slow, stop, and even transport light!  This has gotten me interested in the entire field of BEC and atom trapping and I’m now in contact with Dr. Kirk Madison at UBC about a potential summer job in his lab next year (and maybe grad school after).

I must complain that the majority of the people I met from the University of Calgary were tremendous douchebags – and I don’t say that out of a school rivalry.  We listened to one of their talks (which wasn’t even that good), and midway through he insulted another university (over a plagiarism scandal that was likely unrelated to the paper he was referencing) and after he muttered that the audience was “boring” when they didn’t ask him any questions.  Another guy from their school was running through the hall on our floor and some chick yelled “you’re a nice guy, I just don’t want to fuck you” (or something to that extent), who also is rumoured to have gotten his funding cut for getting a tattoo when he was supposed to be presenting his poster (which was pretty abhorrent – ie. didn’t present anything expect what sounded like a dreamt up unresearched paradox, handdrawn, with links to pictures).  And even in talking with them they were just not nice.  We did meet one nice UofC Physics guy, so I don’t want to shame them all, just the majority of them.

As for the rest of the conference though, it was fantastic.  We saw TRIUMF and SFU Physics labs.  I presented and was later told by someone that he heard my talk was good – as in someone passed along good words about my talk.  Overall it was a great experience.

This part of the spectrum is interesting because it lies in what’s known as the “THz gap” between traditional photonics and microwave technologies. Traditional electronics cannot switch fast enough to generate these pulses, and commercial lasers cannot generate that low of frequency of waves.

So how do we generate these waves? This is where ultrafast optics comes in. Using a technique called “modelocking” a standing wave is setup in a laser cavity and waves constructively interfere to create pulses with lengths on the order of femtoseconds (10^-15 s). These pulses can then be amplified in a multi-pass system and set into a specific repetition rate (1080 Hz in my lab). Each ~100 fs pulse contains approximately 1/2 mJ. These pulses are then of high enough intensity to cause nonlinear effects in most materials. We now enter nonlinear optics land. When this light is incident on specific nonlinear crystals (in our case Zinc Telluride), a second order nonlinear effect called optical rectification occurs. This effect essentially creates a DC bias inside the crystal and current essentially flows through the crystal itself. This accelerating currents creates electromagnetic radiation in the THz regime.

THz pulses are interesting to study due to their absorption properties. Water is the primary absorber of these wavelengths, while plastics, papers, and fabrics have much lower absorption amounts. This allows for the application of techniques for THz imaging. For instance a THz wave can scan through clothes but gets absorbed by water or reflected by metal. This allows for the potential use of security scanning where police can essentially ‘see’ through a person’s clothes and determine if there are any weapons on them. The resolution is better than 1 mm, which is good enough to see most weapons. The following image shows a THz imaging scan:

There are several other ways to generate THz radiation (which I am presently working on a few), as well as many more applications (inlcuding THz-TDS). It is an exciting field that I am enjoying as I learn more about it.