"The Simpsons, Radioactive Potato Salad,
And Running Out Of Electricity"
by John Wilder
"You know, Oppenheimer probably didn’t realize that his little gadget would one day power cat videos on YouTube®. But yet, here we are, preparing to stare down the barrel of an energy crisis that makes the 1970s oil embargo look like a minor hiccup at the gas pump.
America’s tech overlords are building A.I. data centers faster than a caffeinated beaver on gas station Chinese boner pills. These behemoths suck down electricity like it’s free beer at an open bar to toss electrons so we can make A.I. cat videos because there weren’t enough cats in real life. The scale is enormous: gigawatts upon gigawatts, enough to finally get Marty all the way back to 1985. But that begs this question: Where’s all that juice coming from?
Coal? Ha! That’s so 19th century, and the eco-warriors have pretty much chained themselves to the last coal plant, screaming about carbon footprints. Natural gas? Did everyone forget demand peaks in winter when everyone is cranking up the heat and prices spike like Nvidia® stock? Are we going to have to keep our homes at 40°F (3.14 millipedes) just so ChatGPT® can make GloboLeftist women on the East Coast even more neurotic?
We need power, so, naturally, the bright sparks in Silicon Valley and D.C. turn to the holy grail: The Simpsons. Sure, Homer® looks incompetent, but he hasn’t melted Springfield down. Yet. When The Simpsons started, they were mocking nuclear power in the typical GloboLeft drive to get it shut down. Deep down, though, nuclear really always has been the only viable transition plan into the future. Oil really will run out at some point, abiotic or not. But nuclear? If done right, it really can be clean, reliable, and if we don’t let Soviets do it, pretty safe. So, problem solved.
Not. We’re facing an immediate energy cliff. In 2025, nuclear isn’t a parachute, it’s really more like a bedsheet and some twine. With a little help from Constant Reader Ricky, who sent me an email. I’ll quote him directly because, well, he nails it better than I could.
Ricky writes: “Existing commercial power reactors in the US have two key characteristics – their uranium is enriched from the natural 0.7% U-235 assay to a level of 3%, and they are cooled with pressurized water as the heat transfer fluid to run the turbines. The reactors were INITIALLY fueled via uranium enrichment done long ago in monstrous factories that are now closed. An effectively experimental centrifuge enrichment operation in Piketon, Ohio shut down in 2016 without ever producing a pound of reactor fuel (we bombed a similar setup recently in Iran).
Believe it or not, the US CURRENTLY fuels its commercial nuclear power reactors for the past ten years with Russian 3% enriched uranium, even through the Ukrainian war. The Russians basically dilute some of their bomb grade 93% enriched uranium stockpile down into 3% reactor fuel as an export profit center.”
Key point courtesy of Ricky: “The current American commercial nuclear power program is 100% dependent on the Russians and has been for the last decade.” He adds, “But we want that because that every kilogram of Russian uranium that goes IN a New York City power reactor is one less kilogram of Russian uranium that can go into an incoming nuclear bomb OVER New York City.” He’s right. I want the Russians to hit the Somilsotans first. And then New York City twice. It’s the only way to be sure.
It’s like we’re in a bad spy novel, relying on our geopolitical rivals for the fuel that keeps our lights on. We can stamp our feet as much as we want to, but as long as Mom and Dad are paying the power bills, they call the shots.
With AI data centers projected to gobble up an extra 200-300 gigawatts by 2050 (that’s tripling our nuclear capacity), we’re supposed to ramp up nuclear like it’s no big deal. It’s like the steady high school girlfriend you’ve been dating off and on for a year who you can always call for a date at the last minute.
Nope. Building that kind of capacity? Recent estimates peg adding just 63 GW at $354 billion. We’re talking trillions when you factor in overruns. The Vogtle plant in Georgia – two reactors, “just” 2.2 GW, clocked in at $35 billion after fifteen years of delays. Nuclear power makes NASA look prompt and frugal.
Okay, we’ll just do micro-reactors. Except these micro wonders ditch the “obsolete” 3% enriched uranium for something hotter: 20% enriched stuff, packaged in pellets like, I don’t know, energy kibble. Supposedly, they’re meltdown-proof, corrosion-resistant, great with kids, fun at parties, and perfect for high-temperature gas or molten salt reactors. And they’re much smaller than kibble, like poppy seed sized, but kibble is a funnier word and I really don’t want to think how stupid it is to build highly radioactive balls that you could put into someone’s potato salad at the neighborhood picnic?
Cool, so where do we get this 20% enriched uranium for our nuclear kibble? We downblend our surplus bomb-grade stuff from the Cold War. The US has 480 metric tons total, but half is reserved for nuking India (it’s the only way to be sure), and 100 tons reserved for Navy reactors. Bringing those numbers up to date and turning it into nuclear kibble leaves 86 metric tons up for grabs.
So, we have a safe plan. What’s stopping us? Adding 250 GW of new nuclear by 2050 (a Department of Energy guess) requires 5,350 metric tons (it’s like a ton, but it has a French accent) of enriched uranium kibble. Do the math: 86 tons available vs. 5,350 needed? It’s like trying to fill an Olympic®-sized pool by spitting into it.
Okay, let’s restart a program that used to make the stuff. Great! The Piketon, Ohio centrifuge plant we mentioned above, let’s use that. They’re planning on delivering 900 kilograms (a ton for those of us from countries that have put people on the Moon) by 2026. So, we need over 5,000 tons. We’ve made one. Oh, scratch that, not even one yet. Want to take odds on that bet?
Even if we magically create tons of usable uranium, Harry Potter-style®, there’s no supply chain for turning it into nuclear kibble. Right now, it’s a prototype lab in New Mexico fiddling with demos. We’d need a whole new industry. And we’d need to have started on this (checks watch) twenty years ago. That’s the bitch of exponential growth. We could play with 2030 numbers (“only” 50 GW), but since no concrete has been poured for this new capacity and there is no path to creating this fuel, it’s more realistic to discuss if Superman© could beat The Witcher®. It’s a non-starter.
We’re dependent on foreign fuel, short on domestic capacity, and staring at timelines measured in decades, not quarters. Maybe it’s time to rethink the whole “AI will save us” stock market hype or at least stock up on candles and spears. And hey, if that microreactor ends up in my yard, Homer© and I will host a barbecue, BYOGC. (Bring your own Geiger counters, you know, potato salad). Thank heavens we let The Simpsons create our energy policy."
No comments:
Post a Comment