Natalya Bailey: Rocket Engines and Electric Spacecraft Propulsion | Lex Fridman Podcast #157

The following is a conversation with Natalia Bailey, a rocket scientist and spacecraft propulsion engineer, previously at MIT and now the founder and CTO of Axion Systems, specializing in efficient space propulsion engines for satellites and spacecraft. So these are not the engines that get us from the ground on Earth out to space, but rather the engines that move us around in space once we get out there. Quick mention of our sponsors Monk Pack Low Carb Snacks, four Sigmatic Mushroom Coffee, Blinkist, an app that summarizes books, and Sun Basket meal delivery service. So the choice is snacks, caffeine knowledge, or a delicious meal. Choose wisely, my friends, and if you wish, click the sponsor links below to get a discount and to support this podcast. As a side note, let me say something about Natalia's story. She has talked about how when she was young, she would often look up at the stars and dream of alien intelligences that one day we could communicate with. This moment of childlike cosmic curiosity is at the core of my own interest in space and extraterrestrial life, and in general in artificial intelligence, science and engineering. Amid the meetings and the papers and the career rat race and all the awards, let's not let ourselves lose that childlike wonder. Sadly, we're on Earth for only a very short time, so let's have fun solving some of the biggest puzzles in the universe while we're here. If you enjoy this thing, subscribe on YouTube, review it on Apple podcast, follow on Spotify, support it on Patreon, or connect with me on Twitter lexfriedman. And now, here's my conversation with Natalia Bailey. You said that you spent your whole life dreaming about space and also pondering the big existential question of whether there is or isn't intelligent life, intelligent alien civilizations out there. So what do you think? Do you think there's life out there? Intelligent life?

Intelligent life? That's trickier, I think. Looking at the likelihood of a self replicating organism, given how much time the universe has existed and how many stars with planets, I think it's likely that there's other life, intelligent life. I'm hopeful. You know, I'm a little discouraged that we haven't yet been in touch, allegedly. I mean, in our dimensions and so on.

Yeah, it's also possible that they have been in touch and we just haven't. We're too dumb to realize they're communicating with us in whichever. It's the. It's this Carl Sagan idea that they may be communicating at a time scale that's totally different. Like their signals are on a totally different Time scale or in a, like a totally different kind of medium of communication. It could be. It could be our own. It could be the birth of like, human beings like that, whatever the magic that makes us who we are, the collective intelligence thing, that could be aliens themselves. That could be the medium of communication. Like the nature of our consciousness and intelligence itself is the medium of communication.

And like, being able to ask the questions themselves. I've never thought of it that way.

Like actually. Yeah, asking the question whether aliens exist might be the very medium by which they communicate. And it's like they send questions.

So some. This, like collective emergent behavior is the signal.

Is the signal, yeah.

Interesting.

Yeah, because maybe that's how we would communicate with. If you think about it, if we were way, way, way smarter, like a thousand years from now, we somehow survive. Like, how would we actually communicate? In a way that's like if we broadcast the same signal, you know, and then it could somehow like, percolate throughout the universe. Like that signal having an impact.

Multiverse.

Multiverse. Of course, that would have a signal, an effect on the most possible, the most, the highest number of possible civilizations. What would that signal be? It might not be like sending a few like, stupid little hello world messages. It might be something more impactful. It's almost like impactful in a way where they don't have to have the capability to hear it. It like forces the message to have an impact.

Right. My train of thought has never gone. Gone there, but I like it. And also somewhere in there, I think it's implied that something travels faster than the speed of light, which I'm also really hopeful for.

Oh, you're hopeful. Are you excited by the possibility that there's intelligent life out there? Sort of. You work on. On the engineering side of things. It's this very kind of focused pursuit of moving things through space efficiently. But, you know, if you zoom out, one of the cool things that this enables us to do is find, get even intelligent life. Just life on Mars, on Europa or something like that. Does that excite you? Does that scare you?

Oh, it's very exciting. I mean, it's. The whole reason I went into the field I'm in is to contribute to building the body of knowledge that we have as a species. So very exciting.

Do you think there's life on Mars? Like no longer. Well, already living, but currently living, but also no longer living like that. We might be able to find life, as some people suspect. Basic microbial life I'm not so sure.

About in our own solar system. And I do think it might be hard to untangle if we somehow contaminated other things as well, such. So I'm not sure about this close to home.

That would be really exciting.

Yes.

Do you think about the Drake Equation?

Much of that was what got me into all of this. Yeah.

Yeah. Because one of the questions is how hard is it for life to start on a habitable planet? Like if you have a lot of the basic conditions, not exactly like Earth, but basic Earth like conditions, how hard is it for life to start? And if you find life on Mars or find life on Europa, that means it's way easier. That's a good thing to confirm that if you have a habitable planet, then there's going to be life. And that like immediately that would be super exciting because that means there's like trillions of planets, basic life out there.

Though of all the planets in our solar system, Earth is clearly the most habitable. So I would not be discouraged if we didn't find it on another planet in our solar system.

True. And again, that life could look very different. It's habitable for Earth, like life, but it could be totally different. I still think that trees are quite possibly more intelligent than humans, but their intelligence is carried out over a timescale that we're just not able to appreciate. Like they might be running the entirety of human civilization and we're just like too dumb to realize that they're the smart ones.

Maybe that's the alien message. It's in the trees.

It's in the trees. Yeah. It's not in the monolith in the Utah desert.

It's in the trees, right? Yeah.

So let's go to space exploration. How do you think it would get humans to Mars?

I think SpaceX and Elon Musk will be the ones that get the first human setting foot on Mars and probably not that long from now from us having this conversation. You know, maybe we'll inflate his timeline a little bit, but I tend to believe the goals he sets. So I think that will happen relatively soon. As far as, you know, when and what it will take to get humans living there in a more permanent way. You know, I have a glib answer, which is, you know, when we can invent a time machine to go back to the early Cold War, and instead of uniting around sending people, people to the Moon, we pick Mars as the destination. So really I say that because there's nothing truly scientifically or technologically impossible about doing that. Soon it's more politically and financially, and those are the Obstacles, I think to that.

Well, I wonder when you colonize with know more than I say five people on Mars, you have to start thinking about the kind of like rules you have on Mars. And speaking of the Cold War, who gets to own the land? You know, you start planting flags and you start to make decisions. And like SpaceX has this nice, it's probably a little bit trolly, but they have this nice paragraph in their contracts which, where it's like, it's, it talks about that like human governments on Earth or Earth governments have no jurisdiction on Mars. Like the rules. The Martians get to define their own rules. It sounds very much like, like the founding fathers for this country. That's the kind of language, it's interesting that that's, that that's in there and it makes you think perhaps that needs to be leveraged. Like you have to be very clever about leveraging that to create a little bit of a cold war feeling. It seems like we humans need a little bit of a competition. Do you think that's necessary to succeed and to get the necessary investment? Or can the pure pursuit of science be enough?

No, I think we're seeing right now the pure pursuit of science. I mean that results in pretty tiny budgets for exploration. There has to be some disaster, impending doom, to get us onto another planet in a permanent way. I don't know financially, I just don't know if the private sector can support that. But I don't wish that there is some catastrophe coming our way that spurs us to do that.

Yeah, I'm unsure what the business model is for colonizing Mars.

Yeah, exactly.

Yeah, there is for. We'll talk about satellites. There's probably a lot of business models around satellites, but there's not enough short term business. I guess that's how business works. You should have a path to making money in the next 10 years.

Well, and maybe even more broadly. And looping back to something we said earlier. I don't know that getting humans off this planet and spreading bacteria is what we're supposed to be doing in the first place. So maybe we can go, but should we? And I'm probably a, an unusual person for thinking that in my industry because humans want to explore. But I almost wonder, you know, are we putting unnecessary obstacles like we're very finicky biological things in the way of some more robotic or you know, more silicon based exploration and. Yeah. Do we need to colonize and spread? I'm not sure.

What do you think is the role of AI in space? Do you in your work Again, we'll talk about it. But do you see more and more of the space vehicles spacecraft being run by artificial intelligence systems? More than just like the flight control, but like the management?

Yeah, I don't have a lot of color to the dreams I have about way in the future and AI, But I do think that removing, you know, it's hard for humans to even make a trip to Mars, much less go anywhere farther than that. And I think we'll have more. Again, I'm probably unusual in having these thoughts, but perhaps be able to generate more knowledge and understand more if we stop trying to send humans and instead, you know, I don't know if we're talking about AI in a truly, you know, artificial intelligence way or AI as, as we kind of use it today, but maybe sending a petri dish or two of like stem cells and some robotic handlers instead if we still need to send our DNA, because we're really stuck on that. But if not, you know, maybe not even that petri dish. Um, so I see, I think what I'm saying is, you know, I see a much bigger role in the future of AI for space exploration.

It's kind of sad to think that, I mean, I'm sure we'll eventually send a spacecraft with efficient propulsion. Like some of the stuff you work on out that travels just really far with some robots on it and with some, with some DNA in a petri dish. And then human civilization destroys itself and then they'll just be this floating spacecraft that eventually gets somewhere or not. That's a sad thought. Like this lonely spacecraft just kind of traveling through space and humans are all dead.

Well, it depends on what the, what the goal is. Right. Another way to look at it is we've preserved. It's like a little time capsule of knowledge, DNA, you know, that we've. That will outlive us.

Well, that's beautiful.

Yeah, it's how I sleep at night.

So you also mentioned that you wanted to be an astronaut.

Yes.

So even though you said you're unusual and thinking like it's nice here on Earth and then we might want to be sending robots up there, you wanted to be a human that goes out there. Would you like to one day travel to Mars? You know, if it becomes sort of more open to civilian travel and that kind of thing, like, are you like vacation wise, like if you talk, if we're talking vacations, would you like the vacation on Earth or vacation on Mars?

I wish that I had a better answer, but no. I wanted to be an astronaut. Because I, first of all, I like working in labs and doing experiments, and I wanted to go to like, the coolest lab, the iss, and do some experiments there. That's being decommissioned, which is sad, but there will be others, I'm sure.

The ISS is being decommissioned, yes.

I think by 2025 it's not going to be in use anymore, but I think there are private companies that are going to be putting up stations and things.

So it's primarily like a research lab, essentially, yes. Research lab in space. That's a cool way to say it's like the coolest possible research lab.

That's where I wanted to go. And now, though, my risk profile has changed a little bit. I have three little ones, and I won't be in the first thousand people to go to Mars, let's put it that way.

Yeah, Earth is kind of nice. We have our troubles, but overall it's pretty nice. Again, it's the Netflix. Okay, let's talk rockets. How does a rocket engine work? Or any kind of engine that can get us to space or float around in space?

The basic principle is conservation of momentum. So you throw stuff out the back of the engine, and that pushes the rocket and the spacecraft in the other direction. So there are two main types of rocket propulsion. The one people are more familiar with is chemical, because it's loud and there's fire, and that's what's used for launch and is more televised. So in those types of systems, you usually have fuel and an oxidizer, and they react and combust and release stored chemical energy. And that energy heats the, heats the resultant gas, and that's funneled out the back through a nozzle, directed out the back, and then that momentum exchange pushes the spacecraft forward.

Is there an interesting difference in liquid and solid fuel?

In those contexts, they're both lumped in the same. So chemical just means that the release of energy from those bonds, essentially. So a solid fuel works the same way. And the other main category is electric propulsion. So instead of chemical energy, you're using electrical energy, usually from, you know, batteries or solar panels. And in this case, the stuff you're pushing out the back would be charged particles. So instead of combustion and heat, you, you end up with charged particles and you force them out the back of the spacecraft using either an electrostatic field or electromagnetic. But it's the same momentum exchange and same idea stuff out the back, and everything else goes forward.

Cool. So those are the big two categories. What's the difference, maybe in the challenges of each? The use cases of Each and how they're used today, the physics of each and where they're used, all that kind of stuff. Anything interesting about the two categories that distinguishes them, besides the chemical one being the big sexy flames and fire? Fire, yeah.

Chemical is very well understood. You know, in its simplest form, it's like a firework. So it's been around since 400 BC or something like that, so that even the big engines are quite well understood, I think. You know, one of the. One of the last gaps there is probably what exactly are the products of combustion. Our modeling abilities kind of fall apart there because it's hot and gases are moving. And you end up kind of, you know, having to venture into lots of different interdisciplinary fields of science to try to solve that. And that's quite complex. But we have pretty good models for some of the more like emergent behaviors of that system anyways. But that's, I think, one of the last unsolved pieces. And really the kind of what people care about there is making it more fuel efficient. So the chemical stuff, you can get a lot of instantaneous thrust, but it's not very fuel efficient. It's much more fuel efficient to go with the electric type of propulsion. So that's where people spend a lot of their time, is trying to make that more efficient in terms of thrust per unit of fuel. And then there's always considerations like heating and cooling. It's very hot, which is good if it heats the gases, but, you know, bad if it melts the rocket and things like that. So there's always a lot of work on heating and cooling and the engine cycles and things like that. And then on electric propulsion, I find it like, much more refreshingly poorly understood.

Lots more mysteries.

Yeah, I think so. One of the classes I took in college, we spent 90% of the class on chemical propulsion and then the last 10% on electric. And the professor said, like, we only sort of understand how it works, but it works kind of. And it's like. That's interesting. Yeah. And you know, even an ion engine, which is probably one of the most straightforward because it has just an. It's just an electrostatic engine, but it has this really awesome combination of quantum mechanics and material science and fluid dynamics and electrostatics. And it's just very intriguing to me.

First of all, can you actually zoom out even more because you mentioned ion propulsion engine is a subset of electric. So is there a categories of electric engines? And then we can zoom in on ion propulsion?

Yes. So sure. There's the two most kind of Conventional types that have been around since the 60s are ion engines and hall thrusters. And ion engines are a little bit simpler because they don't use a magnetic field for generating thrust. And then there are also some other types of plasma engines, but that don't fit into those two categories. So just kind of other plasma, like a Vasimir engine, which we could get into. And then those are probably the main three categories that would be fun to talk about. Oh, and then of course, the category of engine that I work on, which has a lot of similarities to an ion engine, but could be considered its own class, called a colloid thruster.

Colloid, cool. Okay, so what is an ion propulsion? I imagine.

Okay, so in an ion engine, you have an ionization chamber, and you inject the propellant into that chamber. And this is usually a neutral gas like xenon or argon. So you inject that into the chamber, and you also inject a stream of really hot high energy electrons. And everything's just moving around very randomly in there. And the whole goal is to have one of those electrons collide with one of those neutral atoms and turn it into an ion. So kick off a secondary electron. And now you have plasma. Yes.

Okay.

And now you have. And now you have a charged, you know, xenon or argon ion and more electrons and so on. And then some fraction of those ions will happen to make it to this downstream electric field that we set up between two grids with holes in them. And, you know, in terms of area, the same amount of those ions also makes runs into the walls and lose their charge. And that's where some of the inefficiencies come in. But the very lucky few make it to those holes in that grid. And there are two grids, actually, and you apply a voltage differential between them, and that sets up an electric field. And a charged particle in an electric field creates a force. And so those ions are accelerated out the back of the engine, and the reaction force is what pushes the spacecraft forward. If you're following along and tallying these charges. Now, we've just sent a positive beam of ions out the back of the spacecraft. And for our purposes here, the spacecraft is neutral. So eventually those ions will come back and hit the spacecraft because it's a positive beam. So you also have to have an external cathode producer of electrons outside the engine that pumps electrons into that beam and neutralizes that. So now it's net neutral everywhere, and it won't come back to the spacecraft. So that's. That's an ion engine.

What temperature are we talking about here? So in terms of like the chemical based engines, those are super hot. You mentioned plasma here. How hot does this thing get? I mean, is that an interesting thing to talk about in a sense that. Is that an interesting distinction or is heat. I mean, it's all going to be hot?

No. So it's important, especially for some of these smaller satellites people are into launching these days. So it's important because you have the plasma, but also those high energy electrons are hot. And if you have a lot of those that are going into the walls, you do have to care about the temperature. So I'm having trouble remembering off the top of my head. I think they're at like 100 electron volts in terms of the electron energy. And then I'd have to remember how to convert that into Kelvin.

Can you stick your hand in it?

Not room temperature, not recommended. Yeah.

Okay, so what's a colloid engine?

So the same rocket people that came up with these ideas for electric propulsion probably in the middle of last century, also realized that there's one more place to get charged particles from if you're going to be using electric propulsion. So you can take a gas and you can ionize it. But there are also some liquids, particularly ionic liquids, which is what we use, that you also can use as a source of ions. And if you have ions and you put them in a field, you generate a force. So they recognize that. But part of being able to leverage that technique is being able to kind of manipulate those liquids on a scale of nanometers or, you know, very few microns, so, you know, the diameter of a human hair or something like that. And in the 50s, there was no way to do that. So they wrote about it in some books and then it kind of died for a little bit. And then with silicon MEMS computer processors, and when foundry started becoming more ubiquitous and my advisor started at mit, kind of put those ideas back together and was like, hey, actually there's now a way to build this and bring this other technique to life. And so the way that, the way that you actually get the ions out of those liquids is you put the liquid in again, a strong electric field. And the electric field stresses the liquid and you keep increasing the field and eventually the liquid will assume a conical shape. It's when the electric field pressure that's pulling on it exactly balances the liquid's own restoring force, which is its surface tension. So you have this balance and the liquid assumes a cone when it's Perfectly balanced like that. And at the tip of a cone, the radius of curvature goes to zero right at the tip. And the radius, sorry, the electric field right at the tip of a sharp object would go to infinity because it goes as one over the radius and one over the radius squared. And instead of the electric field going to infinity and maybe like generating a wormhole or something, a jet of ions instead starts issuing from the tip of the. Of that liquid. So the field becomes strong enough there that you can pull ions out of the liquid.

What is the liquid we're talking about? Or is that. There's a bunch of different ones.

You can do it with different types of liquids. It depends on how easily you can free ions from their neighbors and if it has enough surface tension so that you can build up a high enough electric field. But what we use are called ionic liquids, and they're really just positive. They're very similar to salts, but they happen to be liquid over a really wide range of temperatures.

This sounds like really cool. Okay, so how big is the cone? Are we talking what's the size of this cone that generates the ions?

If you have a cone that's emitting pure ions, the. I can't remember if it's the radius or diameter, but that emission is happening from of that cone is something like 20 nanometers.

Oh, I was imagining something slightly bigger. But so, like, this is. So this is tiny. Tiny, yes. Hence the only being able to do it recently.

Yeah, that's right.

So this is all controlled by a computer, I guess. Or like, how do you control. How do you create a cone that generates ions at a scale of nanometers? Exactly.

So the kind of main trick to making this work is that physically we manufacture hundreds or thousands of sharp structures and then supply the liquid to the tips. So that does a few things. It makes sure that we know where the ion beams are forming, so we can put holes in the grid above them to let them actually leave instead of hitting cool. But it also reduces the actual field we have to. The voltage we have to apply to create that field, because the field will be much stronger if we can already give the liquid a tip to form on. And those tips we form have radii of curvature on the order of probably like single microns. So we are working at a little bit larger scale. But once we create that support and the electric field can be focused at that tip, then the tiny little cone can form.

So, wait, so there's something in the. There's an already, like a hard material that, like Gives you the base for the cone and you're pouring like liquid over it, whatever, from the bottom.

Yeah, it's porous. So we actually supply it from the back of the chip and then.

And then liquid forms on top on that structure.

Yeah.

And then you somehow make it like super sharp, the liquid, so the ions can.

And then we've applied that field to get those ions, and that same field then accelerates them.

That's awesome. And there's like a bunch of these.

Yeah, I should have. I should have brought something. So we.

You could just pretend that you have some nanometer cones on.

So actually, you know, kind of about this scale we build, we call them thruster chips. And it's just a convenient form factor. And it's a square centimeter. And on each square centimeter today, we have about 500 of the actual physical. We call them emitters, those physical cones. And we're working on increasing that by a factor of four in the coming months.

In size or in the density.

In the density. The number of emitters within the same square centimeter chip.

So that thing. Cause I think I've seen pictures of you with like a tiny thing in your hand. That must be the thing. Okay, so that's an engine.

So that is kind of the ionization chamber and thrust producing part of it. What's not shown, you know, in that picture is the propellant tank. So we can keep supplying more and more of the liquid to those emission sites. And then we also provide a power electronic system that talks to the spacecraft and turns our device on and off.

So that's the colloid engine. Yeah, that's the core of the colloid.

It's the way I've been talking about it, it's more of ion electrospray. Colloidal tends to mean like liquid droplets coming off of the jet. But if you make smaller and smaller cones, you get pure ions. So we're kind of like a subset of colloid. Yes.

What aspects of this. You said that it's been full of mystery from the physics perspective. What aspects of this are understood and what are still full of mystery?

Yeah. Recently we've been understanding the kind of instabilities and stable regimes of how much liquid do you supply and what field do you apply and why is it flickering on and off or why does it have these weird behaviors? So that's. In the past, just couple years, that's become much more understood. I think the two areas that come to mind as far as not as well understood are the boundary between, you know, you have we actually use kind of big molecular ions. And if you're looking at the molecular scale, you have, you know, some ions that you've extracted, and they're in this electric field. One ion, you know, it's a big molecule. It's getting energy from the electric field, and some of that energy is going into the bonds and making it vibrate and doing weird things to it. Sometimes it breaks them apart and then zooming out to the whole beam. The beam has some behaviors as this beam of ions, and there's a big gap between what are those? What, how do you connect those, and how do we understand that better so that we can understand the beam performance of the engine?

Is that a theory question or is it an engineering question?

Theory, definitely. Axion is a startup, and we're more in the business of building and testing and observing and characterizing, and we're not really diving much into that theory right now.

Okay, zooming out a little bit on the physics. Apologize for the way too big of a question. But to you, from either us, you mentioned Axion as more of sort of an engineering endeavor. Right. But from a perspective of physics in general, science in general, or the side of engineering, what do you think is the most, to you, like, beautiful and captivating and inspiring idea in this space?

In this space. And then I'm going to zoom out a little bit more. But in this space, I keep butting up against material science questions. So over the past 10 years, I feel like every problem or interesting thing I want to work on. If you dig deep enough, you end up in material science land, which I find kind of exciting, and it makes me want to dig in more there. And I was just, you know, even for our technology, when we have to move the propellant from the tank to the tip of the emitters, we rely a lot on capillary action. And you're getting into wetting and surface.

Energies at a scale of like, nano.

Yeah, I mean, you're. It's. If you look further, it's quantum, too, but it all is, you know.

Wait, capillary action at the quantum level.

Yeah. So I would. I think it comes back to me to, you know, material science. There's so much we don't understand at these sizes, and I. I find that inspiring and exciting. And then more broadly, you know, I remember when I learned that the same equation that describes flow over an airfoil is used to price options, the Black Scholes equation. And I was like. And it's just a partial differential equation, but that kind of connectedness of the universe. I don't want to use options pricing and the universe in the same. But you know what I mean. This connectedness I find really magical.

Yeah. The patterns that mathematics reveals seems to echo in a bunch of different places.

Yes.

Yeah. There's just weirdness. It's like it really makes you think. I think you're definitely living in a simulation. Like whoever programmed.

I like that that's your conclusion is using.

Is using like shortcuts to program it. Like they didn't they just copy and pasted some code to the different parts.

Yeah. Think of something new or just paste from over there.

They won't notice.

My conclusion from that was I'm going to go interview for finance jobs. So I had like a little detour.

That's the back. That's the backup option. So in terms of using colored engines, what's an interesting difference between a propulsion of a rocket from Earth when you're standing on the ground to orbit and then the kind of propulsion necessary for once you get out to orbit or to like deep space to move around?

Yes. The reason you can't use an engine like mine to get off the ground is, you know, the thrust it generates is instantaneous thrust is very small. But if you have the time and can accumulate that acceleration, you can still reach speeds that are very interesting for exploration and even for missions with humans on them. An interesting direction I think we need to go as humans exploring space is the power supplies for electric propulsion are limiting us in that solar panels are really inefficient and bulky and batteries. I don't know when anybody's ever going to improve battery technology. I know a lot of people that work on that. And nuclear power. We could have a lot more powerful electric propulsion systems. So they would be extremely fuel efficient but more instantaneous thrust to do more interesting missions. If we could start launching more nuclear systems.

But so like something that's powered. Nuclear powered. That's the right way to say it.

Yeah.

But is in a small enough container that could be launched.

Yeah. So I mean as a world we do launch spacecraft with nuclear power systems on board. But size is one consideration. It hasn't been a big focus. So the reactors and the heaters and everything are bulky and so they're really only suitable for some of the much bigger interplanetary stuff. So that's one issue. But then it's a whole rat's nest of political stuff as well.

I heard I think Elon describe or somebody. But I think. I think it was Elon that described the Evtol like electrical vertical takeoff and landing vehicles. So basically saying rockets. Obviously Elon is interested in electric vehicles. Right. But he said that rockets can't. In the. In the near term, it doesn't make sense for them to be electrical. What do you see? A world where the rockets that we use to get into orbit are also electric based.

It's possible you can produce the thrust levels you need, but you need this a much bigger power supply. And I think that would be nuclear. And the only way people have been able to launch them at all is that they're in a 100 times redundancy safe mode while they're being launched and they're not turned on until they're farther off. So if you were to actually try to use it on launch, I think a lot of people would still have an issue with that, but someday.

It's an interesting concept. Nuclear. It seems like people like everybody that works on nuclear power has shown how safe it is as a source of energy. And yet we seem to be based on the history based on excellent hbos series. I'm Russian with the Chernobyl. It seems like we have. Our risk estimation about this particular power source is drastically inaccurate. But that's, that's a fascinating idea that we would use nuclear as a source for our vehicles and not just in outer space. That's cool. I'm going to have to look into that. That's super interesting.

Well, just last year, Trump eased up a little bit on the regulations and NASA and hopefully others are starting to pick up on the development. So now is a good time to look into it because there's actually some movement.

Is that a hope for you to explore different energy sources that the entirety of the vehicle uses? Something like. Like the entirety of the propulsion systems for all aspects of the vehicle's life travel is the same or electric. Is it possible for it to be the same like the colloid engine being used for everything you could.

And you would have to do it in the same way we do different stages of rockets now, where once you've used up an engine or a stage, you let it go because there's really no point in holding onto it. So I wouldn't necessarily want to use the same engine for the whole thing, but the same technology I think would be interesting.

Okay, so it's possible. All right. But yeah, comes down to the power source. The power source, that's really interesting. But for the current power sources and the. Its current use cases, what's the use case for electric, like the corollary engine can you talk about where they're used today?

Sure. So chemical engines are still used quite a bit once you're in orbit, but that's also where you might choose instead to use an electric system and what people do with them. And this includes, you know, the ion engines and hall thrusters. And our engine is basically any maneuvering you need to do once you're dropped off. There's. Even if your only goal was to just stay in your orbit and not move for the life of your mission, you need propulsion to accomplish that. Because the Earth's gravity field changes as you go around in orbit and pulls you out of your little box. There are other perturbations that. That can throw you off a bit. And then most people want to do things a little bit more interesting, like maneuver to avoid being hit by space debris, or perhaps lower their orbit to take a higher resolution image of something and then return. At the end of your mission, you're supposed to responsibly get rid of your satellite, whether that's burning it up. But if you're in geo, you want to push it higher into graveyard orbit. What's geo, what's so low Earth orbit and then geosynchronous orbit or geostationary orbit.

And there's a graveyard.

Yeah. So those satellites are at like 40,000 kilometers. So if they were to try to push their satellites back down to burn up in the atmosphere, they would need, you know, even more propulsion than they've had for the whole lifetime of their mission. So instead they push them higher where it'll take, you know, a million years for it to naturally deorbit. So we're also cluttering that higher bit up as well. But it's not as pressing as leo, which is low Earth orbit, where more of these commercial missions are going now.

So how hard is the collision avoidance problem there? You said some debris and stuff. So how much propulsion is needed? How much is the life of a satellite is just like a crap trying to avoid little things here and there.

I think one of the recent, you know, rules of thumb I heard was per year, some of these small satellites are doing like three collision avoidance maneuvers. So that's not too bad. Yeah, but it's not zero. And it takes a lot of planning and people on the ground and, you know, none of that really, I don't think right now is autonomous.

Oh, that's not good.

Yeah. And then we have a lot of folks taking advantage of, you know, Moore's Lawn cheaper spacecraft. So they're launching them up without the ability to maneuver themselves and they're like, well, I don't know, just don't hit me.

And three times a year, that could become affordable if it's like, if it gets hit, maybe it won't be damaged kind of thing. That kind of logic.

Affordable in that Instead of launching one satellite, they'll launch 20 small ones. Yeah. So if one gets taken out, that's okay. But the problem is that, you know, one good sized satellite getting hit, that's like a ballistic event that turns into 10,000 pieces of debris that then are the things that go and hit the other satellites. Yeah.

So do you see a world where, like, in your sense, in your own work and just in the space industry in general, do you see that people are moving towards bigger satellites or smaller satellites? Is there going to be a mix? Like, what's. And what are we talking? What does it mean for satellite to be big and small?

What size so big? The space industry prior to, I don't know, 1990, you know, I guess the bulk of the majority of satellites were the size of a school bus and cost a couple billion dollars. And now, you know, our first launches were on satellites the size of shoeboxes that were built by high school students. So that's very different. To give you the two ends of the spectrum, big satellites will. I think they're here to stay at least as far as I can see into the future for things like broadcasting, you want to be able to broadcast to as many people as possible. You also can't just go to small satellites and say Moore's Law for things like optics. So if you have an aperture on your satellite that doesn't follow Moore's Law, that's different. So it's always going to be the size it will be unless there's some new physics that comes out that I'm not aware of. But if you need a resolution and you're at an altitude that kind of sets the size of your telescope. But because of Moore's Law, we are able to do a lot more with smaller packages. And with that comes more affordability and opening up access to space to more and more people.

Well, what's the smallest satellite you've seen go up there? Like, what are the smallest kind? You said shoe boxes.

Yeah, so I think, you know, the smallest, smallest common form factor can fit a softball inside. So that's 10 centimeters on each side. But then there are some companies working on, you know, fractions of that even, and they're doing things like IoT type application. So it's Very low, you know, bandwidth type things. But they're finding some niches for those.

You mean like there's a business, there's a thing to do with them?

Yes.

What do you do with a small satellite like that?

You can, you know, track a ship going across the ocean as like if you need to, if you just pinging something, you know, you can handle that, that amount of data and latencies.

And so you have to have propulsion on that. You have to have a little engine. Yeah, but what, so what kind of satellites would you equip a colloid engine on?

Anything that's bigger than probably about 20 kg, anything that needs to stay up for more than a year or anything somebody spent more than like 100k to build are kind of the ways I would think about it.

That's a lot of use cases. What's a small sat? Like what category?

It's actually very big. I think it's like 700kg or keep hitting my microphone, maybe 1000kg down to 200kg or. People have their own kind of definitions of how they break them up. But smallsat is still quite large. And then it's kind of also applied as a blanket term for anything that's not a school bus size satellite. We need to get our jargon straight. Industry.

So do you see a possible future where there's a few thousand satellites up there now, a couple thousand of them functioning? Do you see a future where there's like millions of satellites up in orbit? Or forget millions, tens of thousands, which just seems like where the natural trajectory of the way things are going now.

Is going tens of thousands. Yes, the two buckets of applications, one is imaging and the other is communication. So imaging, I think that will plateau because one satellite or one constellation can take an image or a video and sell it to infinity customers. But if you're providing communications, like broadband Internet or satellite cell or something like that, satellite phone, you know, you're, you're limited by your transponders and so on. So to serve more people, you actually need more satellites. And perhaps at the rate, you know, our data consumption and things are going these days, yeah, I can see tens of thousands of satellites.

Can I ask you a ridiculous question?

Yeah.

So I've recently watched this documentary on Netflix about flat Earthers, the people that believe in a flat Earth. As somebody who develops propulsion systems for satellites and for spacecraft, what's to use the most convincing evidence that the Earth is round?

Probably some of the photos taken from the moon.

Photos from the moon. Okay, so it's not from the satellite space.

Yeah. I think seeing that perspective, maybe I'm just. I'm answering too personally because I really love those photos because they're beautiful.

Yeah.

I really like the ones that show the Moon and the lunar lander and they're taken a little bit farther back. So you see Earth, and first you're like, wow, that's tiny. And we're insignificant and that's kind of sad. But then you see this really cool thing that we landed on another planetary body and you're like, oh, okay.

Can you actually see her? I don't know.

Yeah, I'll send you. I'll send you that picture because I.

Love the pictures or videos of just Earth from more. From orbit and so on. Those. That's really beautiful. That's like a perspective shifter. That's the pale blue dot. Right. It probably appears tiny.

Yeah. And just that, you know, juxtaposition of the insignificance.

But you're in another.

We built this really cool thing. Take the picture. Yeah.

Oh, that'd be cool. I can't. I personally love the idea of humans stepping on Mars. I'm such a sucker for the romantic notion of that and being able to take pictures from Mars.

So you would go, I.

I would be. What did you say? You said you wouldn't be.

Not in the first thousand.

Which. It's funny because to me that's. That's brave to be in the first million. I think when the Declaration of Independence was signed in the United States, that was like 2 million people. So I would like to show up when they're signing those documents. Okay, so maybe the 2 million.

Oh, that's an interesting way to think.

Of it because, like, then we're like, participating in citizenry and defining the direction.

So it's not the technical risk. You just don't want to show up somewhere that's like America before.

Yeah. Because from a psychological perspective, it's just going to be a stressful mess. As people have studied. Right. It's like. It's people. Most likely the process of colonization, like, looks like basically a prison. Like, you're in a very tight enclosed space with people. And it's just a really stressful environment. You know, how do you select the kind of people that will go. And then there will be drama. There's always drama. And I just want to show up when there's some rules. But I mean, you know, it depends. So I'm not worried about the health and the technical difficulties. I'm more worried about the psychological difficulties and also just not being able to tweet, like, what are you gonna. How are you? How can you. There's no Netflix. So. Yeah, maybe not in the first million, but the first hundred thousand. It's exciting to define the direction of a new. Like, how often do we not just have a revolution to redefine our government, as, you know, smaller countries are still doing to this day, but literally start over from scratch. There's just our financial system. It could be, like, based on cryptocurrency. You could think about, like, how democracy, you know, we have. We have now the technology that can enable pure democracy, for example, if we choose to do that.

Yeah.

As opposed to representative democracy. All those kinds of things. So we talked about two different forms of propulsion which are super exciting. So the chemical base that's doing pretty well, and then the electric base is. Are there types of propulsion that might sound like science fiction right now, but are actually within the reach of science in the next 10, 20, 30, 50 years that you kind of think about, or maybe even within the space of even just like. Like, even ion engines, is there, like, breakthroughs that might 10x the thing, like, really improve it?

So, you know, the real game changer would be propellantless propulsion. And so every couple years you see a new. Now a startup or a researcher comes up with some contraption for producing thrust that didn't require, you know, we've been talking about conservation of momentum, mass times velocity out the back. Mass.

There's usually a mass forward.

Yes, exactly. And you have to, you know, carry that up with you or find it on an asteroid or harvest it from somewhere if you didn't bring it with you. So not having to do that would be, you know, one of the ultimate game changers. And I, you know, unless there are new types of physics, I don't know how we do it, but it comes up often. So it's something I do think about. And the one, I think it's called the Casimir effect. If you have two plates and the space between them is on the order of the wavelength of these ephemeral vacuum particles that pop into and out of existence or something. I may be confusing multiple types of propellantless forces, but that could be real and could be something that we use eventually.

What would be the power source?

Yeah, the most recent engine like this that was just debunked this year, I think, in March or something was called the EmDrive. And supposedly you used a power source so, you know, batteries or solar panels to generate microwaves into this resonant cavity. And people claimed it produced thrust. So they went straight from this really loose concept to building a device and testing it. And they said, we've measured thrust. And sure, on their thrust balance, they saw thrust. And different researchers built it and tested it and got the same measurements. And so it was looking actually pretty good. No one could explain how it worked. But what they said was that this inside the cavity, the microwaves themselves didn't change, but the speed of light changed inside the cavity. So relative to that, you know, there momentum was conserved and I don't, you know, I. Whatever. But finally.