Much of the discussion about how we are going to rapidly change over our energy infrastructure to low carbon energy involves existing technology, or at most incremental advancements. The problem is, of course, that we are up against the clock and the best solutions are ones that we can implement immediately. Even next generation fission reactors are controversial because they are not a tried-and-true technology, even though fission technology itself is. It certainly would not be prudent to count on an entirely new technology as our solution. If some game-changing technology emerges, great, but until then we will make due with what we know works.

The ultimate game-changing energy technology is, I think, fusion. Fusion technology replicates the processes that power stars, mostly fusing hydrogen into other forms of hydrogen and ultimately into helium. Massive enough stars can then fuse helium into heavier elements, with more massive stars fusing heavier elements until we get to iron which cannot be fused to produce net energy. But even fusing the lightest elements takes a tremendous amount of heat and pressure, which has proved technologically difficult to achieve on Earth. We have been inching closer to this goal, however, and recently the National Ignition Facility at the Lawrence Livermore National Laboratory in California has inched over a significant milestone – ignition.

I wrote just last year about the NIF achieving another milestone, burning plasma. The pace of advancement seemed pretty brisk, and I speculated about how long it would be to achieve the next milestone, ignition. Well, here we are. You can read that article for background, but quickly, the NIF uses a fusion method called inertial confinement – an array of 192 powerful lasers to produce inward pressure sufficient to cause a vessel to implode, with the implosion causing sufficient heat and pressure to produce fusion. The NIF was built in 2009, but it took significant upgrades before it was powerful enough to achieve fusion in 2021. Some of the energy from fusion contributed to further fusion, a process called burning plasma. But in that experiment fusion contributed only 70% of the energy necessary to sustain fusion. That means that the fusion process was still a net energy loss. (Those powerful lasers require a lot of energy.)

The obvious goal is for fusion to create more than 100% of the energy necessary to sustain fusion. But even that is not quite enough. The fusion has to create substantially more than 100% of the energy necessary to maintain fusion – enough to payback all the energy necessary to get fusion started, and enough so that waste heat from the process can be used to boil water into steam to turn a turbine to generate electricity. This needs to create enough electricity to justify the expense of building a massive fusion reactor.

Details on the NIF milestone are still skimpy, because the official announcement will not be made until tomorrow (Dec. 13). The information was apparently leaked early, and so we only have sketchy details so far. But that seems to be the main thrust of the announcement, the NIF (as expected) has increased their energy output further, crossing over the line to ignition – producing enough energy to sustain fusion. What I don’t know is how far over the line they are.

This is exciting news (if not totally unexpected). But the ultimate question remains – how long before there is a working commercial fusion power plant contributing net energy to an electrical grid? I am in a long term debate with my brother Bob about this on the SGU. He thinks we will be there sometime in the 2030s. I think it will be sometime after 2050. I sincerely hope he is right, and he will likely argue that the latest NIF milestone is a powerful argument in his favor. Perhaps, perhaps not. I will likely write an addendum after the announcement tomorrow. A lot will depend on how far they have gone, and their estimates of how much farther they go with their current incremental advancements. Do they think they will get to commercial-level net energy? Will they get there in a year, or 10 years, or 20 years? Even if they get there within, say, a few years, they will then need to design and build a commercial fusion reactor. I suspect that will take at least 20 years to complete. Right there we are getting close to 2050.

The bottom line is that I still think, optimistically, we are talking in the 2050s for a working fusion reactor. And of course, in order to have a significant impact on global warming, we need lots of them, not just one. That’s why I say fusion power is a technology for the second half of the 21st century. The anticipated announcement from NIF just keeps us on track for that. Of course, we could put billions of dollars into fusion (and the Inflation Reduction Act is a good start) and fast track the development, permits, and all the red tape. If we “moon shot” the hell out of inertial confinement fusion power, building on the work at the NIF, very optimistically we might move up the time table to the 2040s. I just don’t think that is what’s going to happen.

Rather, I think we will build a test reactor, designed to produce net energy, but still a proof of concept. It will be late and over budget, but hopefully will work. And based on that success we will start building larger commercial reactors, slated to go online in the 2050s and 2060s at the earliest. Even that is optimistic, but it is plausible – all the more so now that the NIF has achieved ignition. This further means that we need a plan for our energy infrastructure to take us to about 2060, and which time, perhaps, we can replace retiring power plants with fusion reactors. (I have explained at length previously why I think we will need 20-30% baseload power even if we try to maximize renewable energy.)

I also think, however, that once we do achieve workable fusion power, that energy source will be a critical technology for humanity for the foreseeable future, for thousands of years. Even just incremental advances will make fusion power safer, more efficient, smaller, and more reliable. Fusion power has the advantage of potentially producing massive amount of energy for the necessary land footprint (unlike wind and solar). It can work on the Moon or Mars, it can work far from the sun, it can power ships, and its fuel is ultimately just hydrogen, the most abundant element in the universe.

This does bring up one technological limiting factor – the NIF uses deuterium and tritium fuel capsules, and these isotopes of hydrogen are hard and slow to make. This technology needs to develop also, to provide an ample supply of fuel for all those fusion reactors we would like to have. We may ultimately need to go with other fusion technology that can use, for example, H3, which can be mined from the surface of the Moon (also a controversial prospect, but plausible).

The fuel problem just highlights the fact that we are not there yet. It’s easy to get excited about reaching such a technological milestone as achieving ignition in a fusion experiment. This does give us a glimpse into the (hopefully) near future. But new technologies like this are challenging to develop and deploy. You will probably read a lot of stories and headlines this week about how fusion is now right around the corner, but I still think we are talking 30 years conservatively. This is not to downplay the significance of this milestone – it proves that fusion power is plausible. We can make more energy than it takes to initiate fusion. But we do need to keep this advance in context. We are still a lot of work and funding away from commercial fusion reactors.