All the light we see from the sky, Moses pointed out, comes from fusion power burning hydrogen, the commonest element in the universe---3/4 of all mass. A byproduct of the cosmic fusion is the star-stuff that we and the Earth are made of.
On Earth, 4 billion years of life accumulated geological hydrocarbons, which civilization is now burning at a rate of 10 million years' worth per year. In 1900, 98% of the world's energy came from burning carbon. By 1970, that was down to 90%, but it has not decreased since. It has to decrease some time, because there is only so much coal, oil, and gas. During this century every single existing power plant (except some hydro) will age and have to be replaced, and world energy demand is expected to triple by 2100.
To head off climate change, fossil fuel combustion has to end by about 2050. The crucial period for conversion to something better is between 2030 and 2050. The ideal new power source would be: affordable; clean; non-geopolitical; using inexhaustible fuel and existing infrastructure; capable of rapid development and evolution. Moses' candidate is the "laser inertial fusion engine"---acronym LIFE---being developed at Lawrence Livermore.
The question, Moses said, is "Can we build a miniature Sun on Earth?" The recipe involves a peppercorn-size target of hydrogen isotopes deuterium and tritium heated to 200 million degrees Fahrenheit for a couple billionths of a second. To get that micro-blast of heat, the National Ignition Facility (NIF) uses lasers---coherent light---at a massive scale. Laser engineer Moses notes that photons are perfect for the job: "no mass, no charge, just energy."
Moses ran a dramatic video showing how a shot at the NIF works. 20-foot-long slugs of amplified coherent light (10 nanoseconds) travel 1,500 yards and converge simultaneously through 192 beams on the tiny target, compressing and heating it to fusion ignition, with a yield of energy 10 to 100 times of what goes into it. Successful early test shots suggest that the NIF will achieve the first ignition within the next few months, and that shot will be heard round the world.
To get a working prototype of a fusion power plant may take 10 years. It will require an engine that runs at about 600 rpm---like an idling car. Targets need to be fired at a rate of 10 per second into the laser flashes. The energy is collected by molten salt at 1,000 degrees Fahrenheit and then heats the usual steam-turbine tea kettle to generate electricity. The engine could operate at the scale of a standard 1-gigawatt coal or nuclear plant, or it could be scaled down to 250 megawatts or up to 3 gigawatts. The supply of several million targets a year can be manufactured for under 50 cents apiece with the volume and precision that Lego blocks currently are. Moses said that 1 liter of heavy water will yield the energy of 2 million gallons of gas.
Fusion power, like nuclear fission power, would cost less per kilowatt hour than wind (and far less than solar), yet would be less capital intensive than fission. For the constant baseload power no carbon is involved, no waste stream, no possibility of meltdown or weaponization, and there is no such thing as peak hydrogen.-- by Stewart Brand