Full transcript
0:00One of the great unsolved mysteries of the universe surrounds us, but we
0:03can't see it. Dark matter. It's passing through your body right now unnoticed,
0:08and because it doesn't absorb or reflect light, it's very difficult for us to detect. But Nancy
0:13Grace Roman's engineering has been tuned to search for the clues of its existence. Aim a
0:18telescope at just the right point in the sky, and you might just see the clues of its existence.
0:23The Nancy Grace Roman Telescope is going to revolutionize this hunt for not just dark matter,
0:28but for exoplanets, too. The Roman telescope is equipped with tools that can cut through
0:33the glare of a sun 1 billion times brighter than the planets orbiting it to image them directly.
0:39And with clever use of physics, it will even be able to detect rogue planets, dim planets
0:44roaming space with no star of their own. This is what the Nancy Grace Roman telescope can do.
0:50and I'm going to show you how it's going to do it. This is how the Nancy Grace Roman Telescope works.
1:00The Roman Telescope fills a critical gap in our abilities to observe the universe. The
1:04James Webb telescope launched in 2021 and is a massive telescope in comparison to Roman.
1:10Its primary mirror is 6.5 m across, made from 18 hexagonal segments of burillium coated in a thin
1:17layer of gold. Gold because it reflects infrared light better than almost any other
1:22surface. Burillium because it's light, stiff, and does not warp when cooled to the cryogenic
1:27temperatures web needs to operate. Design's driven by James Webb's mission to take images
1:32of the very furthest reaches of the universe where time and expansion of the universe has stretched
1:37the visible light into the infrared spectrum, needing cryocooled sensors and massive collecting
1:42areas to image. Nancy Grace Roman's mission is different. Its primary mirror is 2.4 4 m across,
1:49the same diameter as Hubble's. This means the telescope can fit inside a launch fairing without
1:54any complicated folding mechanisms. NASA actually had little to say about the size of the telescope.
2:00Its structure and primary mirrors were donated to NASA by the National Reconnaissance Office. It's a
2:06cancelled spy satellite, and NASA was tasked with adapting the hardware for scientific purposes. On
2:11mirror size alone, WEB wins easily. But the mirror size is not what Roman is for. Web's
2:17field of view is tiny and it needs an incredibly long exposure time to gather enough light to see
2:22anything. It was designed to peer much further back into the history of the universe, looking
2:27for the stretched infrared light emitted hundreds of millions of years ago after the Big Bang. The
2:33Nancy Grace Roman telescope is looking closer to our cosmic neighborhood and has some amazing tools
2:38to make it perfect for the job of hunting down exoplanets. The first trick of its engineering is
2:43its ability to cast a much wider net than either the James Webb telescope or the Hubble telescope
2:49to catch the effects of dark matter and to spot exoplanets. We have a better shot not focusing
2:54on one particular area of the sky rather than a large swathed. This is the Hubble Deep Field
3:00taken in 1995. A tiny patch of sky about the size of a pin head held at arms length. Almost none of
3:07these dots are stars. They're some of the youngest and most distant galaxies. With modern sensors,
3:13this telescope can cast a net 100 times the area of Hubble at the same resolution, and it can take
3:19the same image just 15 minutes later. What takes Hubble months to survey, Roman can do in hours.
3:26In a single exposure, it captures a patch of sky about the size of two full moons side by side.
3:32Roman will be able to image the visible sky of its field of view in months. Something that would take
3:38web decades. It does this with an absolutely massive array of 18 infrared detectors. Each
3:44one 4,096 pixels by 4,096 pixels combining to form 300 megapixel images of the universe in a
3:54single exposure. This is going to be an enormous amount of data. Over the course of Hubble's
3:5930-year mission, it generated 172 terabytes of information. Web is projected to generate 1,000
4:06terabytes over its 5-year primary mission. Roman is going to obliterate those numbers with 20,000
4:12tab. These massive images are full of valuable data, all of which will be available to the
4:18public. You could scour through these images and potentially find a new planet yourself,
4:22but to do that, you're going to need to know how to deal with massive data sets. And today's
4:28sponsor, boot.dev, will teach you everything you need to know. Their core content is free,
4:33and they have a 30-day, no questions asked refund policy. So, hear me out here. When I was learning
4:38to code with Forran back in college, I got pretty bored listening to the teacher. I just
4:43started coding a textbased adventure game with the plot of Superbad at the back of the class.
4:48I learned way more not listening and just doing and actually had fun doing it. And
4:53boot.dev takes the same learning approach. In their Python introductory class, you will learn
4:58the fundamentals of building the code for a role- playinging game. And the platform rewards you just
5:03like an RPG would, too. You can learn everything you need to handle that NASA data and potentially
5:09discover planets right on boot.dev. They even have a dedicated data analyst path. You can't just open
5:1620,000 terabytes of space data in Excel. To find a new exoplanet, you will need to search NASA's
5:21databases with SQL, fetch the raw image files using APIs, write Python code to inspect the
5:27pixels, and manage it all inside a Linux terminal. And you can learn how to do all of that with their
5:32Python SQL APIs and Linux courses. And you will learn them with actual hands-on projects. Instead
5:39of just copying code off a screen, you're building real back-end projects directly in the terminal,
5:45mastering Python, Golang, Linux, SQL, and system design. In an era where AI can spit out simple
5:52scripts in seconds, surface level coders who rely on cheat codes are getting left behind. Companies
5:58don't need people who just copy and paste snippets. They need engineers who understand
6:02core architecture and can read complex code bases and actually know how to manage massive data sets.
6:09And you can do all of this without paying the absolutely astronomical fees of attending college.
6:14Go to boot.dev and use my code realengineer to get 25% off the entire first year on an annual
6:21plan. Or you can follow this QR code or the link in the description. Then you can start dealing
6:27with the 20,000 terabytes of data which will be sent back down to Earth by this huge 500 megabit
6:33pers downlink antenna. The optical path of light inside the telescope is incredibly complex. This
6:40is the path light takes to reach the detector. The first thing it hits is the primary mirror.
6:45Roman's primary mirror is made of the same ultra- low expansion glass as Hubble's primary mirror,
6:51engineered to barely deform as temperatures shift. However, manufacturing techniques have advanced so
6:57much since Hubble's time that despite being the exact same diameter, the engineers of the Roman
7:02telescope have reduced the weight of the mirror from 828 kg to just 186. The surface is coated
7:11with a layer of silver less than 400 nm thick, about 200 times thinner than a human hair. Web's
7:18mirror is gold because it reflects deep infrared better than almost anything else on the periodic
7:23table. But the Roman telescope needs to see the visible spectrum and near infrared. In the visible
7:29range, gold absorbs blue instead of reflecting it, giving it its gold color. Silver doesn't have that
7:36problem. Light continues from the primary mirror to the secondary mirror, a smaller convex mirror
7:42about 55 cm across, mounted at the forward end of the telescope on a six-legged actuated support
7:48structure called a hexopod. The hexopod can tilt, shift, and adjust the secondary mirror's
7:54position on orbit to maintain precise focus as the telescope expands and contracts thermally. A two
8:01mirror design can cancel out spherical aberration and coma, the two aberrations that blur images
8:07taken straight through a single mirror. But this design leaves a stigmatism in the image,
8:12which stretches stars into little ellipses the farther they sit from the center of the frame.
8:17That's why Hobble's two mirror design only allowed a field of view a few arc minutes wide. Roman
8:23fixes this by adding an additional powered mirror to eliminate a stigmatism and give it a field of
8:29view up to 100 times bigger than Hubble's. But there is one more optical problem. When
8:35light bounces off a curved mirror, the photons coming from the edges of the mirror have to
8:40travel a slightly different distance to their focus point than the rays hitting the center.
8:45That means the focal plane isn't straight. To deal with this, the 18 detectors on Roman's focal
8:51plane don't sit on a flat grid. They're arranged along a curve. Roman will use this massive field
8:57of view and spend a large portion of its mission photographing hundreds of millions of galaxies
9:03spread across billions of light years. But it's not just looking at the galaxies themselves, but
9:08rather their apparent shape and how they subtly get distorted by the gravitational pull of matter
9:13between them and us, including dark matter. Dark matter does not absorb, reflect, or release light,
9:21but its gravity does bend light. The clearest evidence we have of its existence is from the
9:26bullet cluster, where two galaxy clusters collided violently. The normal visible gas slamped together
9:33and overheated into a glowing cloud of X-rays. But the bulk of the matter, which we couldn't see,
9:38coasted straight through one another as if the impact had never happened. But if we couldn't
9:43see that mass, how did we know it was there? With gravitational lensing. A galaxy that should appear
9:50circular could arrive at Roman's detector looking slightly stretched, pulled along the axis of
9:55where the greatest concentration of dark matter sits. The distortion is only about 1% but catch
10:01it across enough galaxies and a map of where dark matter has clumped emerges. We have other evidence
10:07of dark matter's existence from the way galaxies spin. A spiral galaxy based on what we can see
10:13should spin like this. But this mysterious unseen matter can make them spin more uniformly like a
10:19vinyl. Between these collision dynamics and galaxy rotation dynamics, dark matter makes up roughly
10:2585% of all matter in the universe. Roman's goal is to turn these subtle gravitational distortions
10:32into a highresolution map of where this invisible mass is hiding. And Roman's massive field of view
10:39is going to help us search for these rare events and build that map faster than ever.
10:44Roman is capable of watching roughly 100 million stars at once, which is also going to help us find
10:50exoplanets and rogue planets through microlensing events. Moments when a planet or a free floating
10:57object drifts across the line of sight of a background star, and its gravity momentarily
11:03bends and brightens that star's light. The Nancy Grace Roman telescope is going to give us
11:07an insane amount of data on exoplanets, including rogue planets. To learn more about what they are,
11:14I spoke with my friend Kobe from the Astro Kobe channel. Can you tell me a little bit about how
11:19the Roman telescope is actually going to impact our search for exoplanets? Okay,
11:24so over the last 30 years of searching, with every telescope that's on and off the Earth,
11:29humanity's confirmed something like 6,000 exoplanets. And the key thing is almost all
11:34of them sit within this tiny little bubble around us. In the grand scheme of things, we've barely
11:39started exploring. And Roman is the telescope that fundamentally changes this. So by staring
11:45at 100 million stars, that's in the galactic core, they'll detect planets up to 26,000 lighty years
11:51away and find somewhere between something like 60 and 200,000 new worlds over just a 5-year survey.
11:58That's something like 55 new planets every single day. And then there's the rogue planets. These are
12:04just worlds without any star at all. Most of them were flung out of their solar systems just as they
12:09finished forming. And they've basically just been drifting through interstellar space all alone ever
12:14since. Some of them for like billions of years. Importantly, they're nearly completely invisible
12:20to us. There's no sun to light them up. So, we can only know that they are there using gravitational
12:25lensing when one drifts in front of some distant star. Its gravity bends that starlight and makes
12:31it flicker a little bit, giving it away. Do we have any idea of how many rogue planets there
12:36might actually be based off those flickers? When astronomers went out and actually counted how
12:41often we see those flickers, the answer genuinely shocked the whole community, the best estimate was
12:46something like 20 rogue planets for every single star in the galaxy. That translates to literally
12:52trillions of them. The most common type of planet in the Milky Way could be these worlds with no sun
12:58at all. Obviously, because these things are just drifting through the galaxy all alone without a
13:03sun, their surface can get pretty cold. I think they normally sit around 30 Kelvin. That's like
13:08-243° C. A completely frozen, lifeless world. Or so you'd think. If that planet just happened
13:16to hold onto a thick hydrogen atmosphere on its way out of the system, the radioactive decay in
13:22its core could keep that atmosphere from freezing solid, trapping enough heat to maintain a liquid
13:28ocean locked under kilometers of ice. If life ever did actually start down there, that rogue
13:34world would become kind of like an arc carrying it across the galaxy. Going from confirming 6,000
13:41exoplanets in our entire history to potentially over 200,000 in just 5 years is going to be a
13:48massive leap in our knowledge of the universe. And it gets even better. This huge detector array also
13:54has a secondary instrument alongside it that will allow us to image these planets directly. a very
14:00difficult task when the light of a star can be a billion times brighter than the planet itself. So,
14:06we need a way to block the starlight. This is what the coronagraph instrument is for. A traditional
14:12coronagraph uses a simple disc to cover the star. But the Roman telescope's engineering
14:18is much more complex. A flat mirror near the end of the path splits off a fraction of the
14:23beam sideways into the chronograph instrument. This instrument uses shape-shifting mirrors and
14:29a range of precision engineered masks to actively cut out unwanted light to allow us to peer through
14:35a stars glaring light. The first challenge is keeping the light steady. Manufacturing
14:40a perfect mirror is basically impossible. Every surface will have tiny bumps that scatter light
14:46unpredictably. And scattered light is just noise that will make detecting extremely faint light of
14:52distant planets impossible to pick out. But the engineers of the Roman telescope have created
14:58a way to measure it and cancel it out. There are two mirrors inside the chronograph's optical path
15:04that contain thousands of Poctric actuators. Po electric materials can change their shape
15:10when you apply a voltage. And by applying different voltages to different actuators,
15:15the telescope can push the mirror surface up and down at specific points. Roman will be measuring
15:21the residual light constantly and applying corrections 20 times a second to cancel out
15:26the scattering. The light now meets Roman's masks, which are all precision engineered to
15:31block a star's light. A simple disc can't block the light. Light will just spill around the edges,
15:37and the shape of the mask needed changes with what we are looking at. One mask uses precision cut
15:43layers of metallic and dialectric films to delay the phase of light passing through them by 180°
15:50before recombining them to cancel out the stars light, allowing us to directly see the faint light
15:56from the planet itself, enabling us to pinpoint its location. With its location known, we can
16:03switch to other masks that allow us to create dark zones in only the regions we need. This allows a
16:09broader range of wavelengths through which is needed for spectroscopy which will allow us to
16:14split the wavelengths and measure them which can give us valuable information on the composition
16:18of the planet's atmosphere. This technology is groundbreaking. Kobe explained to me just how
16:24limited our planet detection ability has been without Roman's advanced coronagraph. The thing
16:29is this coronagraph system really will change the entire field of exoplanet astronomy forever.
16:35The current detection methods for exoplanets are really good at finding these huge hot gas giant
16:40planets that orbit extremely close to their star. But it's nearly impossible for them to find mature
16:46planets in the cold outer edges of these systems. Roman's microl lensing survey will finally be able
16:52to uncover these planets by the thousands from their gravity alone. And then for the first time
16:57ever, it will be able to follow up using this coronagraph and directly image them. Literally
17:02every single planet that we've ever directly photographed before was still glowing from the
17:07heat of its own formation. And they're always orbiting extremely far away from their star.
17:12Something like 10 times farther out than Jupiter is from our sun or even more. A mature quiet world
17:18like our very own Jupiter shining in nothing but the reflected light of its star has simply been
17:24impossible for our technology to image. And all of that completely changes with Roman. This advanced
17:29coronagraph is the first instrument capable of actually imaging these worlds. I think officially
17:35it's not even being flown as a science instrument. It's actually a technology demonstration. This is
17:40the first flight test of this kind of starlight cancelling technology that NASA is going to need
17:45for its next great observatory, the habitable world's observatory that's planned for the 2040s.
17:51HWO will take everything that this chronograph can do and push it like 10 times further to photograph
17:57Earth-sized planets around sunlike stars. Literally searching their atmospheres for signs
18:03of alien life. But importantly, if they actually want to do all of this science, Roman can't be
18:09in orbit around the Earth like Hubble. These programs are designed specifically to fill in
18:14our gaps in knowledge and to do all this science. Roman cannot be in orbit around Earth like Hubble.
18:20The telescope will park itself in a special orbit, Lrangee point 2. L2 is a gravitational
18:26balance point a million and a half km from Earth where the pull of the sun and Earth combine to let
18:32a spacecraft orbit along with Earth as it circles the sun. So the two never drift out of alignment.
18:39It's a pretty unique orbit that has only been technically viable in the last couple of years.
18:44When Hubble launched, we didn't have the technical know-how nor the communication systems in place
18:49for L2 to be useful. It was stuck in low Earth orbit instead, cycling through blistering heat
18:55and freezing cold every 90 minutes with Earth itself blocking out half the sky at any given
19:01moment. Now, web and three other telescopes live in L2 with Roman becoming the fourth observatory
19:08in the orbit. That fixed geometry means a telescope parked there can keep the sun, earth,
19:14and moon permanently at its back. Sitting in a stable, deep, cold thermal environment year round,
19:20allowing it to see deeper into the infrared range. But the center of the galaxy, where we want to
19:26look for exoplanets, is only visible in seasons. During the summer months, the galactic center
19:32is in line with the sun. In winter, the problem flips. Now the sun sits almost directly behind the
19:38spacecraft. This means that the schedule for Roman is already determined for the next few years. When
19:44the center of the galaxy is visible, Roman will point to that. And when it's not, it will look
19:49at the distant galaxies in the higher latitudes of the sky. Getting Roman to L2 is a problem in
19:55its own right. The telescope is too heavy for a standard Falcon 9. So the Falcon Heavy will
20:01lift it into a parking orbit first. Then the upper stage fires again for a trans L2 injection burn,
20:07pushing Roman onto a trajectory that will carry it out 1.5 million kilometers over roughly 30 days.
20:14But L2 is not technically stable. An object placed exactly there will drift away over time because
20:21the equilibrium is a saddle point, not a bowl. So Roman does not sit at L2. It orbits around it,
20:28tracing a slow loop called a halo orbit. Roman's mission will be limited by how much fuel it can
20:33carry to stay at L2. Web, which shares the same L2 halo orbit regime, fires its thrusters roughly
20:40every 3 weeks for station keeping. And the James Web Space Telescope taught engineers something
20:46useful. The Aron 5 delivery was so precise that the midcourse corrections on the way out used a
20:52fraction of the expected fuel. And the mission now has enough propellant to potentially operate for
20:5820 years rather than the planned 10. Roman's team has that lesson in mind. Every kilogram of fuel
21:04saved on the journey to L2 is months of additional mission life. The question today on many of the
21:11scientist minds as the Nancy Grace Roman telescope launches from the Kennedy Space Center aboard
21:16its Falcon Heavy is whether they can repeat the Aran Fi's precise delivery. Thanks for watching.
21:22If you want to get started on learning how to code, check out our sponsor boot.dev. And
21:27if you use my code, realeng engineer, you'll get 25% off your first year on an annual plan.