MIT scientists explain when we'll have fusion power

Back in March, we posted about how this could be the year where the National Ignition Facility breaks even with laser fusion, reaching the point where as much power is generated as is input. This doesn't mean we've got a fusion power plant around the corner, though, and researchers have come clean about what the hold-up is.

Fusion power is what you get when you take two lightweight atomic nuclei and fuse them together into one heavier atomic nucleus, releasing energy in the process. It's far cleaner and far more efficient than fission power, and the only reason that we're not taking advantage of it right now is that it requires temperatures and pressures on the order of what you'd experience at the center of the sun to get it to work.

At MIT, they've been working on getting fusion to happen inside a Tokamak (called the Alcator C-Mod, pictured above), which is a piece of equipment that uses intense magnetic fields to confine and heat plasma to the point that fusion can be initiated, which is something on the order of tens to hundreds of million of degrees.

Slashdot readers had the chance to ask a group of MIT fusion power experts questions about the futuristic power source, and while a lot of the info was way, way, way beyond the comprehension of mere mortals, we went through and pulled out a bunch of the most interesting (and understandable) info.

The Q&A below is a condensed version of some of Slashdot readers' questions along with group answers from the MIT scientists, and if you're looking for more detail, in most cases you can find it in the full-length answers here.

When will fusion power my house (or vehicle)?

This is obviously an impossible question to answer, but we can give some thoughts about when it might happen, and why. First, the current official plan is that ITER will demonstrate net fusion gain (Q = 10, that is, ten times more fusion power out than heating power put in) in about 2028 or 2029. (Construction will be done by about 2022 but there's a six-year shakedown process of steadily increasing the power and learning how to run the machine before the full-power fusion shots.)

At that point, designs can begin for a "DEMO", which is the fusion community's term for a demonstration power plant. That would come online around 2040 (and would putt watts on the grid, although probably at an economic loss at first), and would be followed by (profitable, economic) commercial plants around 2050.

The talk is always about reaching break-even with fusion. What about capturing the power? Are we generating heat that will drive steam turbines? What schemes exist for capture and harnessing the power generated by fusion?

In a magnetic fusion reactor, each deuterium-tritium fusion produces a 3.5 MeV (mega- electronvolt) alpha particle (helium nucleus) which deposits its energy in the plasma (this self-heating is how you can have an 'ignited' plasma which doesn't require much or any external heating), and a 14.1 MeV neutron, which deposits its energy in a thick lithium blanket surrounding the toroidal reaction chamber.

It all comes out as heat, which is used to heat a working fluid, which turns a turbine, producing electricity. This is not expected to be a technological problem - the challenge is in getting a confined thermonuclear plasma to produce the fusion energy in the first place!

How do you explain the safety/benefits of fusion to a generation of people terrified of nuclear anything?

This is where fusion really shines. The two big problems (at least, perceived problems) of fission reactors are the risk of a meltdown, and what you do with the high-level radioactive waste. Fusion has neither of these issues!

Regarding the first, the reason why a worst-case accident in a fission reactor can be so devastating is because there is a lot of fuel in the reactor at any one time. In a fusion reactor, it's a completely different story. There will be less than a gram of fuel in a reactor at any one time--fresh deuterium-tritium fuel is continually added as it is burned--and so a runaway reaction is simply not possible.

As for the second benefit of fusion (waste), the reaction is completely different from that in a fission reactor. In fusion, the reaction is simple, deuterium + tritium = helium + neutron. So there is no "waste" from the unburned fuel - any tritium that isn't burned gets pumped out of the chamber and recirculated back in.

What happens when the magnetic fields that hold the 90,000,000 degrees Celsius plasma in place fail?

Holding a hot plasma stationary using magnetic fields without it ever touching material surfaces is very difficult - Richard Feynman once compared it to trying to "hold Jello with rubber bands." To understand what happens, you have to realize that the plasma is very, very light. In the Alcator C-Mod tokamak, it has a mass of only about 0.001 grams - about one- fiftieth as much as the smallest drop of water you can get from an eyedropper.

We do two things to make sure that the walls can survive these disruption events. The first is making them out of materials that can take a blast of heat, like tungsten, or else materials that ablate away rather than melting, like carbon fiber composites. The second is to develop "disruption mitigation" systems which can cause the plasma to radiate all its energy evenly over the entire wall surface, spreading the heat out and lessening the chance of causing localized melting.

But I want to stress again - disruptions are an operational problem, meaning they might cause a power plant to be offline for a while, but they're not a safety problem. There is no chance of a runaway reaction or meltdown in a fusion reactor.

Do you think a program the size of the Apollo program could kickstart fusion to general availability?

We think that we're roughly $80-billion away from a reactor. At current levels of funding (worldwide), that's about 40 years. Even given access to huge amounts of money, it's unlikely that a working reactor could be built in less than a decade - there are just too many facilities to build between current devices and a full-scale reactor in order to ensure success. But we could certainly do it faster than 40 years!

We can say this: an increase in funding would allow for different paths to be tried in parallel, like stellarators, tokamaks (ITER), spherical tokamaks, etc. Plus, we could build a facility in the United States to study the problem of plasma-wall interactions, which is a very important topic that has not been adequately studied up to this point (see our answer above about what steps are needed to get to a reactor).

Perhaps the most heartening thing about all this is the following:

We know exactly what we need to do. Not everything has a solution yet - that's why it's still a research project! But we generally know what the big challenges are to get to a working magnetic fusion reactor. The point is that it's not a money pit. There are unsolved challenges, but we know what they are, and with adequate support, these challenges will be overcome.

Fusion power is cheap, it's clean, it's safe, but most importantly, it's realistic. We just have to get there. It may take a while, or with some concerted effort, it may take less of a while, but once we have operational power plants (by 2050 or earlier) churning out unlimited amounts of electricity, it'll be the beginning of the end for fossil fuels and the beginning of the beginning (we can only hope) for climate recovery.

MIT Plasma Science and Fusion Center, via Slashdot

For the latest tech stories, follow DVICE on Twitter
at @dvice or find us on Facebook