Imagine a world where humanity masters  every planetary resource available to   it—our first step on the famous Kardeshev  scale of technological advancement.

How   distant is that step?

Will we even become a true  Type-1 civilization, and how can we get there?

In 1964  Soviet physicist Nikolai Kardashev  proposed a method for measuring the technological   advancement of a civilization.

The Kardashev scale  has become a go-to metric for what technologically   advanced civilization might be capable of.

It’s  based on amount of energy commanded —with a Type   I Civilisation controlling access to all available  energy on a planetary scale, a Type 2 accessing   an entire star’s energy output, and a Type 3  harnessing energy from a galactic scale.

Fun side   note: Kardashev originally came up with this scale  in a paper discussing the detectability of alien   civilizations, because the distance over which an  alien broadcast might be detectable depends on how   much energy is pumped into that transmission.

Despite this modest intended application,   the whole idea of the Kardashev scale really  captured the imagination of both scientists and   science fiction-ists.

Exponential advancement on a  cosmological timescale implies frighteningly vast   differences between between civilizations born at  different times, and Kardeshev captures that.

Type   2 and 3 civilizations—mastery of one or many stars  respectively—are the stuff of extreme sci-fi—we’re   talking Dyson spheres, black hole engines,  artificial worlds, galactic empires, and warping   the fabric of space for fun.

By comparison,  the Type 1 civilization is just getting   started.

Maybe it’s traveled to the nearest star  systems and terraformed a planet or two at best.

Humans would be Type 0.

Maybe 0.5 at best.

But if  we hope to one day become a super-civilisation;   roaming the galaxy, mastering the  fundamental constituents of reality,   and finally making contact with the aliens who  have been waiting for us to graduate technology   preschool… we need to start somewhere.

And that  means becoming a Type I civilisation.

So what’s   it going to take for us to get there.

There’s a lot more to becoming a stable   super-civilization than energy  consumption—including some messy   geopolitical stuff that I’m neither qualified nor  have time to cover in a short YouTube video—so   today we’re going to stick to Kardeshev’s  simple criterion and ask what will it take to   attain the defining energy resources of a Type I.

In that 1964 paper, Kardeshev actually says that   humanity is close to the technological level of  a Type 1 civilization and defines a Type 1 power   capacity at roughly the current level.

But he was  speaking quite loosely, and these days we tend   to think of Type I as already having access to  planetary-scale energy resources—which is a factor   of 10,000 times higher than what we currently use.

So that’s the thought experiment for today—what   does it take to master planet-scale energy.

Let’s attach a number to that statement.

How   much energy can we access from the safety of  our home planet?

We want a stable civilization,   so we’re just going to consider renewable  sources.

Besides, a proper Type 1 would burn   all of the energy locked up in Earth’s  fossil fuels and accessible fissionable   material like uranium in a couple of days.

Renewable sources like wind and waves contain   a lot of energy, but the amount is piddling  compared to the amount of solar energy that   hits the Earth.

And that makes sense—ultimately,  almost every source of energy in the atmosphere   or biosphere comes from the Sun.

So we actually  have a very easy way to estimate the energy use   of a Type-1 civilization—it’s the amount of solar  energy hitting the entire planet every day.

And   directly collecting solar energy may also be the  easiest way to access that power.

So a sun-powered   Type-1 will be our first consideration.

There is  another way we could become Type-1 that may yield   even more energy, but I’ll come back to that.

The average solar irradiance at the Earth is   1361 W/m^2, and our planet intercepts that  energy over a disk with the Earht’s radius,   giving us an upper limit of 1.7x10^17 Watts to  play with—let’s say 10^17 Watts.

To put this into   perspective, we currently use around 10^13 watts  globally—10,000 times shy of our Type I goal.

Collecting anywhere close to 10^17 Watts in solar  power is going to take a lot more than covering   the planet in modern solar panels—not that we’d  want to do that anyway.

The first program in   upgrading our Kardeshev scale is improving our  solar collection capabilities.

So let’s start   by learning a bit more about solar cells.

The most common solar cells are typically   made from silicon and are an application of  semiconductor technology that we discussed here.

To summarize, you have two layers of silicon,  each mixed or doped with a different element.

One of these doped elements grants an extra  free electron to its layer—this is a p-type   senmiconductor and on its own those free electrons  allow the flow of current.

The other layer doped element subtracts one electron, creating  an electron hole and an n-type semiconductor.

And those holes can also flow as a current.

Now if you splice those lakers together,   the free electrons diffuse from p-type to  n-type—across the p-n junction—to fill the holes,   at which point the system is non-conducting.

But  if you provide energy to those electrons they jump   back across the p-n junction and are able to flow  around any circuit that the device is attached to.

This is exactly how transistors work,  with the energy being provided by a   small electric potential.

In solar cells,  the energy is provided by incoming photons   in a process called photoexcitation.

In a  solar panel, a circuit provides a path for   those electrons to return to the n-type layer  and their energy can be extracted en route.

But there’s a fundamental limit to how  much energy a solar cell can extract   from light.

This is called the Shockley-Queisser  limit.

There are a few effects that go into   determining this limit, but for now let’s  focus on one of the big ones: spectrum losses.

The energy required to move an electron  between layers is a very specific number:   typically the energy of a near-infrared  photon.

Photons with less energy,   which means most infrared, do nothing  at all.

Photons with more energy—all   visible and ultraviolet light—will excite their  electron, but the excess energy gets lost as heat.

The photons in sunlight come with a very broad  range of energies, so some loss is inevitable,   with an efficiency limit of around 50%.

Other  effects drag the efficiency—Shockley–Queisser   limit—down to around 30%.

In practice we've only  just cracked 25% efficiency at the moment.

But   this is for traditional solar cells with  a single p-n junction.

Newer designs and   materials offer the potential to exceed this  limit.

For example, layered p-n junctions   tuned to different energy photons can capture  a wider spectral range without wasting energy.

Certain special materials like perovskites have  a similar advantage.

Quantum dots are another   exciting possibility are quantum dots thanks to  their tunable absorption spectrum.

In general,   advances in materials science and nano-fabrication  are likely to boost us over 50% efficiency.

Even more important than conversion  efficiency is where we put these cells.

Space-based solar power has a major energetic  advantages.

Our atmosphere blocks most UV light,   but in space those UV photons increase the  overall available energy for solar cells by a   factor of 10.

Also there’s no night-time in space,  which adds another factor of two.

And of course,   it’s nice that we don’t have to cover Earth in  the things, leaving more room for the humans or   our AI overlords to inhabit our Type I world.

A major challenge with space-based solar is   getting the energy back to Earth.

The most viable  candidate is microwave power transmission.

2.45   GHz photons pass easily through our atmosphere.

Pulses of these would be collected by receiver   antennae – called rectennas — which would  convert those photons back to electricity.

Or maybe we finally build that space elevator  to send people up and channel power down.

So we’re really looking at swarms of solar arrays  orbiting Earth or in co-orbit around the Sun or   at Earth’s Lagrange points.

With the James Webb  Space Telescope, we’ve demonstrated that we’re   able to send large, fragile things into space.

To  achieve our 10^17 Watt power source we’ll need an   array quite a bit larger than the flattened disk  of the Earth in order to make up for the losses,   which will never be zero.

That’s more than a  hundred trillion square kilometers a lot of   solar collectors, and if we scale down our silicon  solar cell thickness by a factor of 10 that’s   still a few trillion kilograms, not including  supporting structure.

Fortunately silicon is   the second most abundant element on Earth after  oxygen, so a few trillion kilograms is nothing.

Add a bit of asteroid mining for the supporting  infrastructure, this feels distantly doable.

And of course this will put our new Type I  civilization on track to Type II — just keep   building arrays until we wreath the entire  Sun and we have an effective Dyson sphere.

Solar is the most obvious path to Type-I-dom,  but it may not be the best.

Remember that the   Sun generates energy through nuclear fusion—by  smashing together hydrogen nuclei to make helium,   and converting a small fraction of the  mass into energy.

So rather than being   content to soak up the energy that  the Sun happens to radiate our way,   why not build our own mini-Suns on Earth?

Let’s talk about artificial fusion.

In the Sun, particles are smashed  together and fused into new,   heavier nuclei in a series of reactions known  as the proton-proton chain.

It starts with lone   protons—hydrogen nuclei—and ends in Helium-4  nuclei.

The most energy is obtained during a   particular step of that chain, when two types  of heavy hydrogen—deuterium and tritium—smash   together to form helium 4.

About 0.4 of  1% of the mass of the ingoing particles   is converted into energy, mostly in the  kinetic energy of an ejected neutron.

That doesn’t sound very efficient, but remember,  E=mc2 describes the conversion of mass to energy,   and the c2 does some heavy lifting here.

Deuterium-tritium fusion releases four   times as much energy as Uranium fission  gram for gram.

That means to reach our   Type-I energy goal of 10^17 Watts we’d need  to convert a few hundred kilograms of heavy   hydrogen into helium per second.

Fortunately  we have a massive supply of deuterium—I’ll   come back to the tritium.

Earth’s oceans contain  around 1.4 billion billion metric tons of water,   one eighth of which is hydrogen by mass, and 0.016  percent of that is hydrogen is deuterium.

If we   burned the oceans deuterium, we’d be able to power  our Type 1 civilization for a few million years   before running out.

And if we can’t make it to  Type 2 by then, well, maybe it wasn’t meant to be.

Having the fuel for fusion is one thing, but we  also have to have the reactors.

How long before   we have that?

Commercial fusion is famously 50  years away no matter when you ask.

That said,   there are major programs to bring this  to reality.

The most promising approaches   use powerful magnetic fields to  confine an extremely hot plasma,   trying to duplicate the crushing effect of  the intense gravity in the core of the Sun.

Perhaps the most famous is the International  Thermonuclear Experimental reactor—ITER—in France;   a type of reactor called a Tokomak.

There are  other magnetic confinement projects and designs,   and also non-magnetic and non-confinement options,  but that’s not for this video.

Currently magnetic   confinement seems most promising for commercial  reactors, and ITER is the furthest along.

There are still many challenges to overcome  before magnetic confinement fusion is   commercially viable—from dealing the the plasma  instabilities to safely extracting the energy   from the extreme-energy neutrons produced in  fusion.

The main issue is getting more energy   out that you put in to get the reaction started  and maintained.

And that seems to be a matter of   scale.

Larger reactors should have an easier time  doing that—and this is the great hope of ITER,   with its proposed 500 MegaWatt output  supposedly only needing 10% of that to   sustain the reaction.

This is scheduled  to come online in around 10 years or 50.

OK, back to the question.

Assuming we get fusion  technology, how many reactors do we need to power   our aspiring Type-I status?

Let’s say we build  many ITERs.

We only expect a fraction of its   output to be convertible to electricity—those  neutrons are hard to catch.

At 20% efficiency   we’d need 100,000 reactors to power our  super-civilization.

That is a lot, but it’s not …   utterly preposterous.

If we double efficiency and  scale reactor size by a factor of 10 then 5,000   reactors scattered across the globe will provide  every Watt we’ll need for millions of years.

But with great power comes great need for  storage and distribution of that power.

That means the development of a planetary-scale  supergrid.

Rather than stringing lossy copper   cables across the planet, we’re going  to want to parallel-track development   of high temperature superconducting  cables.

Underground networks of, say,   yttrium barium copper oxide can distribute  our new extreme current density with minimal   power dissipation, thanks to their zero  resistance when they become superconducting.

Challenges here include the fact that even  high temperature superconductors need to   be cooled quite a bit, which means both  energy and cumbersome piping.

The energy   part we seem to have sorted, and we’re going  to require those chunky pipes anyway due to the   typical brittleness of superconductors.

This  is stuff that companies are working on today,   so we can hope it’s sorted by the time we need it.

OK, so we have a couple of paths  to Type-1-civilization-hood.

But   now that we have a completely ridiculous amount  of energy, what do we do with it?

Why did we go   to all this trouble anyway?

Honestly, maybe  we didn’t need to.

Maybe we can settle down   and live in modest harmony with our planet  and chill on these expansionist tendencies.

Maybe it’s juvenile to obsess over some made  up metrics in a soviet paper from the 60s.

But seeing as we got this far, and we  have an imaginary Type-1 civilization,   we may as well do something with it.

Part of  being a Type I civilisation is the ability to   not just harness the energy on Earth, but to  be in control of its natural forces.

We now   have access to energy on the same scale as the  atmosphere and the ocean, and so can, presumably,   change them for the better.

So I guess we start  out by saving the planet, reverse climate change   and all that, now that we actually plan  to stick around as a species for a while.

They say if you have a hammer, every problem is  a nail.

But energy is the most versatile hammer   there is, and it’s the ultimate currency  of problem-solving.

Near limitless clean   energy can enable colossal scales of resource  gathering and manufacturing—from mass-scale   desalination and farming to asteroid mining,  advanced manufacturing, building perfect cities   for all peoples.

We could turn the Earth into a  paradise of abundance.

Assuming we also solve all   world conflicts, which energy can’t do directly  although surely eliminating scarcity helps.

Energy is information and power is computation.

Our supercomputers will run hot and grind out   solutions to all technological challenges and  unlock the secrets of the universe with sheer   brute force processing and unthinkably capable  AI.

We’ll duplicate our energy facilities to   power the terraforming of other worlds, creating  simulacra of our Earthly paradise.

And so humanity   will claw its way towards the next levels of  Kardashev’s scale.

Or we might just spend the   energy mining Kardeshev-coin and training our  AI successors.

But whether it’s us or them,   Earth will be on track to spawning  a civilization of the Type II,   soon mastering the energy of an entire  star, and then of many.

And then maybe   we look back at Nicholai Kardeshev’s little paper  and realise that we are the beacons he imagined,   and that we’ve announced our presence to new  Type-0s still emerging across galactic spacetime.