SciCafe: Hacking the Stars with Hakeem Oluseyi

HAKEEM OLUSEYI (FLORIDA INSTITUTE OF TECHNOLOGY): It's my first time here. It's also the first time that I showed up to speak, and instead of giving me a bottle of water they gave me a beer. So I'll be back.

So I do give talks on different topics. Sometimes I like to come in, blow your mind with cool science, or give the inspirational talk of my life story. They forced me to talk about my most boring talk, which is my research. So are you ready? Alright, so here we go. My research. So we're going to start off.

This is baby me, when I was a little baby. This guy right here is my PhD advisors, the late great Dr. Arthur Bertram Cuthbert Walker II. These are my other colleagues. But here we were flying this rocket. We were the ones who pioneered the technology that gives you images like this from the Sun every day from the Solar Dynamics Observatory, or the Soho Spacecraft.

And I became really fascinated with a particular topic. And I made an amazing discovery. And the discovery is a question. So, it's related to another question. So let me give you the first question that everybody knows the answer to. And that is, what is matter made of?  What do you say? Okay, you guys are too nerdy, all right. The average human would say atoms. And that's correct.

Okay, now the second question which was my discovery was the question no one had ever asked me. And I feel like everyone should be asked this question from the time they understand language. And that is, where does light come from? The Sun. That's the answer everyone gives. But there's no Sun in this room, but yet there's a lot of light. So I'll give you the simplest answer. Now this isn't a hundred percent, there are some special circumstances where other things could happen, but for the most part matter makes it.

Did you catch that? So what I direct my students to do is, whenever you see matter–excuse me–light being emitted from somewhere–the lights of the room, a flame, a lightning bug–ask yourself what is the matter doing to emit that light? But the amazing thing for us as humans and scientists is that what matter makes light, the signature of what the matter is and what the matter is doing, is encoded in the light. And when the light travels through space-time, if space-time is contracting or expanding, the dynamics of space are also encoded in the light. So I became really fascinated with this idea of interpreting light. But back to our experiment.

So we were getting the images like this, and we were going to say, here, here's what the matter is, here's what it's doing. But just to give you an idea, here's our rocket payload. I spent the summer building this thing. Sixteen to 22 telescopes inside there. And each of them, each circle you see here is a different telescope. And each one takes an image of the Sun at a different wavelength, and that allows us to take an image of the Sun at a different temperature. Enough of that.

So I went to graduate school in the ‘90s at Stanford, as they said. And right next door there was this place called Silicon Valley. And everybody was graduating, then going and making millions. And so I thought, well hell, I'll go to Silicon Valley. And I did. And after a year and a half I had a bunch of patents, I had a half-million dollars in stock options. Unfortunately, I started in 1999. And you know what happened in 2001. The bubble burst.

But one of the things that happened is, what we see here is a process chamber where they make computer chips. And there's a process called plasma enhanced chemical vapor deposition. And when chips are made, many of them use plasmas. Many of these processes use plasmas. And the idea that I had was that a lot of the light that's being emitted can tell us what's happening in that process chamber.

Now, normally the way you find out if what happened in the process chamber was correct is after you do the process you measure the wafers. That's really wasteful, so they were looking for a way to get rid of that waste. And so what I did is, I did this big experiment where I took a spectrum.

So here's a spectrum from a plasma process chamber. On the X-axis is time, on the Y-axis is wavelength, and the colors represent the intensity of the light. So if you take a slice at one time, you get a spectrum that looks something like this. And as things change in the chamber, these little peaks and valleys move around. And I was able to find–this is my table of how things change as I change these parameters up here at the top. And… See I thought you guys were super nerdy. If you were, you would just scream. Or maybe you were so excited that you just went silent. But when you get a straight line in science…you hit the nail. Boom, it's in the wood.

So anyway, I did that. But it was a direct application of what I was doing in astrophysics, I applied in semiconductor manufacturing. And things were going really good, but I didn't like the corporate life. They have these crazy rules. They have this thing in Silicon Valley, they have it in a lot of other places too, they call it a boss. They have that thing. And they want you to show up in the morning, stay all day. So I decided to come back to science.

So the first thing I did is I joined the Supernova Cosmology Project, which is, you know, led by Saul Perlmutter who won the Nobel Prize for discovering dark energy. And there we were using stars. So when I say hacking, hacking stars, what I mean is repurposing stars. So there are many of us who study stars, because stars themselves are interesting. So what I did, the first hacking I did, was to take the knowledge I learned from studying stars, and apply it to making technologies here on earth.

In the second instance, now we want to understand space-time and how it behaves. And so supernova cosmology, at the time, was one of two areas of what we call precision cosmology. And here's how that works.

A supernova explodes somewhere in the distant universe. We know how bright it intrinsically is. And so, by measuring how bright it appears, we can tell how far away it is. Because an object appears more dim the more distant it is. The second thing is we can break that light down into its spectrum, you know something like this, and particular peaks and valleys occur at specific wavelengths. But because that light is traveling through intergalactic space, the space between the galaxies is expanding, and so the light gets stretched by the exact same amount that space expanded while the light was traveling through it.

So if we receive light from a distant source–let's say this one right here that occurs at like 400–let's say for example we find it at 800. You'll see the same pattern only it'll be shifted and stretched out, then we know it's at twice the wavelength that it should be, that means that the universe today is twice the size it was when the light left that thing.

So you get a measure of the size of the universe. That measurement of distance that I mentioned earlier, based on the brightness, tells you how long the light has been traveling through us. It's a measurement of time. So now, by looking at the spectrum of these exploding stars, you get the size of the universe versus time. That's pretty amazing. So I did that stuff and then I became a professor.

You've heard surprising things last night, and you might think this is equally surprising. You heard these guys–Michael Jordan, LeBron James, yeah. Well I'm actually the greatest basketball player who's ever existed. And I ruptured a my lumbar disc, and so I left the field for a couple of years while I was recovering. Okay, and what happened was I decided I wanted to come back and get involved in an emerging field, and the emerging field that I joined is what is known as supernova cosmology. Sorry, survey science. I had a little thing right there, time glitch. And so, survey science, it really went big with the Sloan Digital Sky Survey.

So instead of taking your telescope and pointing it at an object that you're interested in, instead you take your telescope and every night you tile the sky and you take an image of every location on the sky with some cadence. And you do that night, after night, after night, year, after year, after year, and over time you will find everything that changes its brightness within a limit, and everything that moves.

And so, the thing about that science is that you can't go and look at these images. The data I'm going to show you now is from a survey that was done by MIT Lincoln Labs to find near-earth objects. And we took the data, and we had five billion observations of twenty four million objects. So you're not going to look at that with your naked eye, that data. So you have to do big data analytics. And we were interested in galaxy formation and evolution.

And so here's our dominant model of how galaxies form and evolve. They grow through a process called accretion. Now this is the galaxy, but remember there's a supermassive black hole in the center, and there's a big giant dark matter halo surrounding this thing, right? So we say a galaxy, there's a lot more going on than just that galaxy. But because they grow in this way through accretion, there should be all sorts of debris left over from how our galaxy grew.

And so this type of science is called galactic archaeology. And so there's a type of pulsating star called an RR Lyrae star that we know how bright it is intrinsically, and so by looking at how bright they appear, we can map out the distribution of these stars around our galaxy, and look for over-densities of them. And they can perhaps tell us where there are tidal streams. Now something to note: in the disk, it's very dense with matter. Not really dense, because if you think about it–you want to hear something mind-blowing? All right. Get your mind around this.

The stars are about ten million times their size distant from each other, all right? About ten million times their own size distant from each other. The nuclei that make up the majority of your mass, all of your mass virtually, they’re about a couple hundred thousand times their own size away from each other. You look pretty solid, but you're really empty space. We're just so far away from those little tiny things. So you heard stars ten million times their own size apart, you heard nuclei in your body a few hundred thousand times their own size apart, how far apart what you think galaxies are from each other relative to their own size? What would your guess be? Yeah. Try ten. Yeah. The universe is more solid than you are. You're just not looking at it from far enough away. But that’s an aside.

So the idea here was that we took this data set that I told you, 24 million objects, five billion observations, and we found these new over-densities of these things. And these big orange balls are globular clusters, which are spherical distributions of about a million stars in a hundred light years. And we found what looks like tidal tails of globular clusters. But you know what I've always thought would be so cool? Imagine if you lived on a star on the edge of a globular cluster above our Milky Way galaxy. Half the year you'd see the face of the Milky Way in your night sky, then the other half you'd be looking into the globular cluster. Anybody get with me on that? Yeah.

Anybody been to the southern hemisphere and see the dark sky? Southern hemisphere sky? The northern hemisphere sky is nothing in comparison. You guys got to go. In the northern hemisphere, the Milky Way core never gets really high above the horizon. In the southern hemisphere it’s directly overhead, and you can see the large and small clouds of Magellan really easily.

Okay so now more recently. So what happened is once I got in the survey science, I had two research groups: one research group was doing solar physics, plasma physics; the other research group was continuing to do the survey science. And so I directed my graduate students to do what I did, pay attention to other areas that are like our area, but not our area. And so we were looking at these two problems with solar physics. One is called the coronal heating problem. The material you see here is about two or three million degrees. The surface of the Sun doesn't shine in this x-ray image, it's dark. And so that's a problem. Because if I have a 6,000 degree surface, why would I have a million degree atmosphere above it? The second problem is how is the solar wind accelerated? Gravity at the surface of the Sun is 30 times stronger than gravity at the surface of Earth, and yet the sun's atmosphere streams away in order to form the solar wind. What's accelerating that plasma?

So that's what we were studying and I directed my students to pay close attention to processes that are scale-invariant. That means the physics is the same although the size scale is very different. And so we were looking at these jets that you see here at the Sun's pole, you see these jets flying off, and how they work. And they work via a process known as magnetic reconnection. So here's a model of them. There's a jet, and what you see here is you have this planar structure where these lines represent magnetic fields. And the reason why the plasma gets accelerated, the reason the plasma is heated, it has to do with the interaction of these magnetic fields with themselves and also with the plasma.

And so what you see here where there is a spine, is what we call a magnetic null. So we have a region where all the magnetic field lines either point into or away from this region where the magnetic field goes to zero. And under certain conditions, a magnetic reconnection event can occur, and the plasmas are accelerated to very high speeds. And when I say it's scale-invariant, it happens on the surface of the Sun or the scale of millions of miles, it happens in the earth and other planets’ magneto spheres on the scales of miles, and it happens in the cores of galaxies on the scale of lightyears.

So my graduate student came to me one day and he said, Dr. O–there's a joke there but I'm not going to tell it. Not that joke, the other joke. This process is scale-invariant and it accelerates particles to 3,000 kilometers per second. Our top ion propulsion technologies accelerate ions only 50 or 60 kilometers per second. If we're able to harness this and do it in a volume of a cubic meter, then this could be a great new ion propulsion technology. And so the key thing here is two things: we have to first create this weird magnetic field configuration that's difficult to create, this fan spine creation.

But then you have to make it reconnect somehow, you have to force it to reconnect. And so you need what we call a perturbation. And we began doing the problem. There's also some plasma physics to be done there. So here's how we create the magnetic field. It just so turned out that in Melbourne Florida, where Florida Institute of Technology is–there's a company called the Advanced Magnet Lab. We gave them our problem, and they gave us a solution. I know it's complex, but trust me. Don't trust anybody.

And how do we give the perturbation? Well it turns out that the people who do fusion research had invented this device decades ago called a dense plasma focus. So what you have here is a series of electrodes at one polarity, and a central electrode at the opposite polarity. And I mean plus or minus voltages. And so you connect this thing to a huge capacitor bank. At the back is a piece of glass, a dielectric. You put it in a dilute gas, and you set off those capacitors and suddenly a plasmoid forms. Which is basically a glowing plasma. It moves to the end, this is what the magnetic field looks like, exactly what we need in that experiment–it moves to the end and it squeezes down so severely then it creates nuclear fusion.

It actually works. You don't get more energy out than you put in, but we realize that this is the exact perturbation that we need. So we bottle the thing, and so here you see the plasmoid move it to the end, and yeah it worked. So that's my research. Thank you.