Despite their terrible effects on the environment and limited supply, for now non-renewable do a really good job of meeting our energy needs.

In 2017, 80% of the power used in the United States was supplied using fossil fuels. And the need for energy doesn't appear to be shrinking any time soon.

Another 9% was delivered from nuclear fission, the process of splitting atoms, which releases far less CO2. Unfortunately, fission produces radioactive waste and also relies on non-renewable fuel sources, such as uranium and plutonium.

All of these methods operate on broadly the same principle, essentially operating as a heat engine. A working fluid, often water, is heated by the fuel to expand and do work, turning the blades of a turbine. The turbine is connected to an electrical generator that converts the rotational motion of the blades into electrical power, which is then fed into the grid.

So what about the remaining 11% of power? That came from renewable energy sources, the kind that are generated about as fast as we use them.

Some of the major renewable energy sources come from processes that are naturally occurring on earth: wind power; solar power; hydropower, which is based on flowing water; and geothermal power, which uses the heat of the earth deep underground. None of these sources are things we'll run out of. We have a good few billion years left of sunlight, for example. And what's more, renewable energy tends to release fewer harmful by-products, like carbon dioxide, into the environment.

Take hydropower, for example, which converts the kinetic energy from the motion of running water into electrical power. In a fast-flowing river, a run-of-river power plant diverts part of the river's flow, sometimes through a tunnel, to turn the turbines of a generator.

That works well in some places, but the problem with this approach is that it's tricky to control the generation of energy to meet demand. You don't want to put lots of power into the grid when it won't get used, and you want to be able to ramp up the supply when the demand suddenly spikes.

Like during the halftime break when the English football team -- that's soccer to you Americans -- played Colombia in the 2018 World Cup. A huge number of people in the UK opened their refrigerators to grab a drink or a snack, causing the compressors inside them to turn on. Then there were the people who had already had a bunch of drinks. All of those people simultaneously flushing their toilets during the break created an increased demand for power on the local pumping stations that maintain pressure in the water system.

The total increase in demand was measured to be 1200 megawatts. That's an extra demand for power equivalent to several power plants. With fossil fuels, you can control the amount of fuel being burned and therefore the amount of power being produced.

Run-of-river power plants struggle with this because the amount of power they generate depends on the flow of the river, which in turn depends on things like the rainfall during the time period and even the temperature -- both things we can't control. To get around this, the more common form of hydropower is a hydroelectric dam. In this case, you can install a dam that floods an area and creates a huge reservoir of water.

The water then flows through the generator's turbines at the bottom of the dam, which turn the water's kinetic energy into power. If you install an intake valve that opens or shuts to control the water flow through it, you can even manage the production of energy to meet the changing demands of the electrical grid.

Unfortunately, flooding an area with water isn't consequence-free. Changing the environment so suddenly and preventing the natural flow of water downstream can have devastating consequences for the local ecology. There's also the risk of the dam breaking if it was built improperly. Despite those challenges, hydropower has been enormously helpful. In recent years, it's produced as much as 16% of the world's energy and up to 70% of all the world's renewable energy.

The other renewable energy source that works in a similar way to hydropower is wind power, which also uses turbines. The main difference is that the fluid doing work on the wind turbines is air, instead of water.

One of the biggest engineering challenges here is designing the turbine blades to efficiently extract energy from that air. As we saw in fluid mechanics, predicting the flow of a fluid around an object can get seriously tricky. Blades have to be engineered to withstand the stress they're subjected to while also allowing the wind to efficiently rotate them to power the generator.

It's as complicated as designing an airplane wing. Once again, you run into the problem of demand. You can't control the strength of the wind to increase or decrease power generation as you need it.

Even if that were possible, you'd still have to transport it from the sparsely-populated open plains where the wind blows more easily to dense urban centers with low amounts of wind but high demand for power. Transporting that power becomes even trickier over long distances because you lose some energy as the electricity travels through the wires. For that reason and others, engineering considerations often play a big role in deciding where wind farms, as a collection of turbines is known, should be built.

So wind power has only generated 4% of the world's total power supply in recent years.

Location also plays an important role in another renewable energy source: geothermal power. Like conventional power plants, geothermal power relies on steam as the working fluid on the turbines connected to the generator, but in this as, you don't need fuel to generate the steam.

You can drill into underground deposits of hot volcanic rock, normally near the earth's tectonic plate boundaries, to use them as a heat source for a power plant. Then all you need is to pump water to that location and create another channel for steam to rise through to do work on the turbines.

The biggest problem comes with setting up a geothermal power plant in the first place. It can be expensive to drill and explore for underground conditions that are exactly right, and it is only really possible in certain parts of the world, like Iceland and Italy.

But there is one source of renewable energy that's so abundant and easily accessible, you only have to step outside on a bright sunny day to see it: solar energy. In fact, the amount of sunlight the Earth receives in just a single year is twice the total amount of energy that will ever be extracted from fossil fuels and the uranium used in nuclear fission, combined.
The challenge is finding efficient ways to harness that energy because turning sunlight into electricity isn't simple. The most promising technology we have is called the photovoltaic, or simply PV, cell. Most people know them by the name given to many cells arranged together, solar panels.

Unlike everything else we looked at, there's no trace of a turbine here. Instead, we're looking at semiconductors. Solar panels use two different semiconducting pieces to set up an electrical field that biases the movement of free electrons inside the material in a particular direction.

In short, the materials encourage an electrical current to flow when they receive energy, which then travels through the circuit, delivering power to whatever's connected to the PV cell. That means solar panels can deliver power directly to the grid. Between that and the abundance of sunlight, it seems like there shouldn't be an energy shortage problem at all.

But as we've seen with the other energy sources, cost, fluctuating demand, location, and transmission all factor in here. For starters, solar cells aren't all that efficient. The very best solar cells can convert 40% of the energy they absorb into electrical power.

But they're expensive to produce because of the high quality of silicon needed in manufacturing, among other reasons. On average, industrial PV cells are about 17% efficient, and once you factor in the cost of making the cells and energy storage, solar power ends up being anywhere between 3 and 6 times as expensive to produce as that from fossil fuels. Increasing solar panels' efficiency would bring this down dramatically.

Another big challenge for solar power is that like with the hydroelectric dam, you need a way to store energy to control the production in line with power demand. You won't generate much solar power on a cloudy day, whereas you might have a surplus on sunny days, but you can't store sunlight directly.

Instead, engineers are working on ways to temporarily store that extra solar power. These include solutions like batteries or even pumping water up a column to later give up its energy as hydropower during periods of high demand. Once again though, efficiency plays a big role on making both these methods a suitable form of energy storage.

Despite the efficiency and storage problems, there is one major advantage to solar panels: they can be deployed pretty much anywhere. Rather than having to transmit power across long distances, solar panels can simply be installed on smaller scales close to areas of demand, even on the roof of an individual home.

But manufacturing panels themselves brings its own set of issues. One of the raw materials used currently to make solar panels is quartz, which has to be processed to produce the high-quality silicon needed for making PV cells. This itself is an energy-intensive process which offsets some of the total energy production of solar panels across their lifetime of usage. Even worse, processing quartz can often produce toxic by-products like tetrachloride, which can end up spilling into the environment and causing damage to soil. 

It all sounds a little bleak, but the most difficult challenges in engineering are often the most important ones. In fact, the National Academy of Engineering in the US has identified making solar energy more economical as one of the grand challenges that engineers in the 21st century need to solve.

Future engineers have lots of ways to contribute towards making solar more feasible. Currently, researchers are looking at new storage systems, such as using solar power to drive hydrogen fuel production, which can be burned later on with no carbon dioxide emissions. More on that next time.

Engineers are also introducing new materials into the production of solar panels and improving the ways in which PV cells themselves are linked and arranged on the panels. There are even experimental methods being developed that use new structures on a molecular level, called nanocrystals. These increase the amount of energy given to the electrons in the material when light is absorbed instead of losing the energy as heat. That could drive the efficiency high enough to make it economically competitive with current power sources and increase the adoption of solar worldwide.

So there are lots of challenges ahead in bringing renewable energy sources to the forefront of electrical power production. But that's all the more space for future engineers to have an impact and create new solutions to the world's energy needs. 

In this episode, we looked at renewable energy sources and why we need them. We looked at how hydropower, wind, geothermal, and solar power are used to produce electricity; some of the challenges faced in doing so; and the areas engineers are working on to make their use more widespread. 

In our next episode, we'll see how engineers have moved beyond natural processes to invent entirely new ways of generating power. 

Crash course engineering is produced in association with PBS Digital Studios, which also produces "Eons," a series that journeys through the history of life on earth. With paleontology and natural history, "Eons" takes you from the dawn of life through the so-called "Age of Dinosaurs" and right up to the end of the most recent ice age. 

Crash Course is a Complexly production and this episode is filmed in the Dr. Cheryl C. Kinney Studio with the help of these wonderful people, and our amazing graphics team is Though Cafe.