Natalya Bailey is a rocket propulsion engineer from MIT and now CTO of Accion Systems. Please support this podcast by checking out our sponsors:
- Munk Pack: https://munkpack.com and use code LEX to get 20% off
- Four Sigmatic: https://foursigmatic.com/lex and use code LexPod to get up to 60% off
- Blinkist: https://blinkist.com/lex and use code LEX to get 25% off premium
- Sun Basket: https://sunbasket.com/lex and use code LEX to get $35 off

EPISODE LINKS:
Natalya's Twitter: https://twitter.com/natalya926
Accion Systems: https://accion-systems.com/

PODCAST INFO:
Podcast website: https://lexfridman.com/podcast
Apple Podcasts: https://apple.co/2lwqZIr
Spotify: https://spoti.fi/2nEwCF8
RSS: https://lexfridman.com/feed/podcast/
Full episodes playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4
Clips playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41

OUTLINE:
0:00 - Introduction
1:46 - Intelligent life in the universe
5:47 - Life in our solar system
7:52 - Humans on Mars
11:26 - Robots vs human in space exploration
12:20 - AI in space
16:25 - How rocket engines work
21:42 - How ion engines work
26:05 - How colloid engines work
35:03 - Material science
37:52 - Nuclear powered rocket engines
42:51 - Electric propulsion out in space
46:18 - Satellites
51:12 - Photo of Earth from the Moon
52:50 - Humans on Mars
55:12 - Propulsion without fuel
1:03:07 - How to build a rocket company
1:10:05 - SpaceX and commercial spaceflight
1:14:38 - Advice to startup founders
1:21:13 - Book recommendations
1:29:31 - Meaning of life

CONNECT:
- Subscribe to this YouTube channel
- Twitter: https://twitter.com/lexfridman
- LinkedIn: https://www.linkedin.com/in/lexfridman
- Facebook: https://www.facebook.com/LexFridmanPage
- Instagram: https://www.instagram.com/lexfridman
- Medium: https://medium.com/@lexfridman
- Support on Patreon: https://www.patreon.com/lexfridman

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, MunkPak low carb snacks, 4 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.

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 1 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 at Lex Friedman.

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.

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, it's also- In our dimensions

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 Carl Sagan idea that they may be communicating at a time scale that's totally different.

Like their signals are in a totally different time scale or in a totally different kind of medium of communication.

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.

Like actually yeah, asking the question whether aliens exist might be the very medium by which they communicate.

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

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.

In a way that's like, if we broadcast the signal, and then it could somehow like percolate throughout the universe, like that signal having an impact.

That would have a signal, an effect on the most, the highest number of possible civilizations, what would that signal be?

It might not be like sending a few stupid little hello world messages.

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.

My train of thought has never 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.

Are you excited by the possibility that there's intelligent life out there.

Sort of you work on the engineering side of things.

It's this very kind of focused pursuit of moving things through space efficiently.

But if you zoom out, 1 of the cool things that this enables us to do is find, forget even an intelligent life, just life on Mars or on Europa or something like that.

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.

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.

So I'm not sure about this close to home.

Like do you think about the Drake equation much of like?

That was what, yeah, what got me into all of this, yeah.

Yeah, because 1 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 with 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 time scale 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.

Yeah, it's not in the monolith in the Utah desert, it's in the trees.

How do you think we'd 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 when and what it will take to get humans living there in a more permanent way, I have a glib answer, which is, when we can invent a time machine to go back to the early Cold War and instead of uniting around sending 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 of when you colonize with more than, I say, 5 people on Mars, you have to start thinking about the kind of rules you have on Mars.

And speaking of the Cold War, who gets to own the land?

You know, you start planting flags, you start to make decisions.

And like SpaceX says this night, it's probably a little bit trolly, but they have this nice paragraph in their contracts where it's like, 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 the founding fathers for this country, that's the kind of language.

It's interesting 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, like 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.

Like you should have a path to making money in like 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 like bacteria is what we're supposed to be doing in the first place.

And I'm probably an unusual person for thinking that in my industry because humans want to explore.

But I almost wonder, are we putting unnecessary obstacles?

We're very finicky biological things in the way of some more robotic or more silicon-based exploration?

And yeah, do we need to colonize and spread?

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, you know, more, this 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, I don't know if we're talking about AI in a truly, you know, artificial intelligence way or AI as we kind of use it today, but maybe sending a Petri dish or 2 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, maybe not even that petri dish.

So I see, I think what I'm saying is, 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 DNA in a petri dish.

And then human civilization destroys itself.

And then there'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 goal is, right?

Another way to look at it is we've preserved, it's like a little time capsule of knowledge, DNA, that will outlive us.

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

So even though you said you're unusual in 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.

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 we're talking vacations, would you like to 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 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.

Yes, I think by 2025 it's not going to be in use anymore.

But I think there are other, there are private companies that are going to be putting up stations and things.

So it's primarily like a research lab, essentially.

It's like the coolest possible research lab.

And now though my risk profile has changed a little bit.

I have 3 little ones and I won't be in the first thousand people to go to Mars, let's put it that way.

We have our troubles, but overall, it's pretty nice.

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 2 main types of rocket propulsion.

The 1 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 a fuel on an oxidizer and they react and combust and release stored chemical energy.

And that energy 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 between 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 batteries or solar panels.

And in this case, this stuff you're pushing out the back would be charged particles.

So instead of combustion and heat, 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 of stuff out the back and everything else goes forward.

What's the difference maybe in like 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 2 categories that distinguishes them, besides the chemical 1 being the big sexy flames and.

Chemical is very well understood, 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 1 of the last gaps there is probably what exactly are the products of combustion?

Are 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 1 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 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.

1 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, we only sort of understand how it works, but it works kind of.

And even an ion engine, which is probably 1 of the most straightforward because 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 electrical.

So like maybe, is there a categories of electric engines and then we can zoom in on ion propulsion?

There's the 2 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 2 categories.

So just kind of other plasma, like a VASIMR engine, which we could get into.

And then Those are probably the main 3 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.

Okay, so what is an ion propulsion ion engine?

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 1 of those electrons collide with 1 of those neutral atoms and turn it into an ion.

So kick off a secondary electron, and now you have-

And now you have, And now you have a charged 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 2 grids with holes in them.

And in terms of area, the same amount of those ions also make 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 2 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.

What temperature are we talking about here?

So in terms of like the chemical base engines, those are super hot.

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 gonna 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.

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 1 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 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.

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 0 right at the tip, and the electric field right at the tip of a sharp object would go to infinity, because it goes as 1 over the radius, and 1 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 that liquid.

So the field becomes strong enough there that you can pull ions out of the liquid.

We're talking about, or is it 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.

See all Lex Fridman transcripts on Youtube

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