The nuclear industry has spent the last few years trying to make reactors smaller. The whole small modular reactor pitch, the one filling conference slides and startup press kits, comes down to shrinking a power plant until it fits on a truck, building it in a factory, and skipping the decade-long construction saga that kills projects before they pour concrete. Ship the box, plug it in, repeat.<\/p>\n
In Seversk, a closed city near Tomsk in western Siberia, Russia is building the exact opposite of that.<\/p>\n
It is called BREST-OD-300, and it is a reactor so large it cannot be delivered in one piece. The parts show up as steel shells as tall as a five-story building, and the machine gets assembled inside its own building because there is no other way to get it there. Instead of water, it is cooled by molten lead. And this is the year its vessel finally closes, even though the reactor itself will not start up until 2028 or 2029.<\/p>\n
Most reactor vessels are steel cans. You forge the rings or weld them together, and the finished vessel ships to the site in one heavy piece. In Britain, a forge recently closed up a complete small-reactor vessel with four welds in under a day<\/a>. BREST’s vessel does not work like that at all.<\/p>\n Rosatom calls the layout “integral.” The vessel is not an all-metal can but a metal-concrete structure: welded steel cavities with concrete poured into the gaps as the build climbs. It is also enormous, which is the practical problem. It can only be delivered in parts, and final assembly is only possible on the Siberian Chemical Combine site itself.<\/p>\n The pieces are the kind of thing you plan a highway around. Six of them were manufactured weighing more than 1,000 metric tons each, split between the Atommash plant in Volgodonsk and the Izhora works near St Petersburg. The central cavity shell, the one that will hold the core basket and the fuel, weighs 143 metric tons and stands over 14 meters tall. Crews craned it into place in late September 2025.<\/p>\n Then in December they installed the last of four peripheral cavity shells, each more than 15 meters high, the pieces that will house the two steam generators and the main coolant pumps. Moving them meant lifting power lines and pulling down road signs so the convoy could squeeze through.<\/p>\n Once the concrete is counted, Rosatom says the whole installation will weigh about 16,000 metric tons<\/a>. The transport packaging alone came to 700 metric tons. This is less a machine you install than one you pour in place, like a cathedral that happens to make electricity.<\/p>\n At the start of 2026 that vessel was roughly 70% assembled. The job for this year is to close the loop: connect the peripheral cavities to the central one, finish the concrete, and drop in the internals. World Nuclear News<\/a> puts the target plainly, quoting the site: “Completion of the reactor vessel assembly is scheduled for the end of 2026.”<\/p>\n Nearly every power reactor running today boils water. BREST does not. It runs molten lead through the core at around 500\u00b0C, at close to atmospheric pressure, and that is a big part of the safety pitch. No pressurized water means no steam explosion to design around, and lead is heavy and boils so high that losing the coolant is genuinely hard to do. Rosatom leans on a phrase, “natural safety,” and says the design let it drop the core-catcher melt trap entirely.<\/p>\n The catch is corrosion. Hot lead dissolves ordinary steel, slowly eating the very parts meant to contain it. Engineers had to develop special steels rated to survive up to 600\u00b0C in that environment. Nuclear Engineering International<\/a> flags liquid-metal corrosion as the main threat to the whole thing.<\/p>\n None of this is a brand-new idea, which is the interesting wrinkle. Back in the mid-1970s the Soviet Union built lead-bismuth-cooled reactors and used them to power seven submarines, so the concept has decades of hardware behind it. But pure lead is a harder animal than the lead-bismuth alloy. The alloy melts around 180\u00b0C; pure lead does not go liquid until north of 300\u00b0C, so the plumbing has to stay hot enough to keep the coolant from freezing solid inside the machine.<\/p>\n Other outfits are circling the same coolant. In Italy, Newcleo is assembling a full-scale lead demonstrator that spins a turbine with no fuel inside it at all<\/a>, a way to prove the metal before anything ever goes critical. BREST is the version with actual uranium in it, which is exactly why the corrosion question carries so much more weight.<\/p>\n The reactor is only a third of the point. BREST sits inside a complex Rosatom calls the Pilot Demonstration Energy Complex, and the whole idea is to put three things on one site: a plant that makes the fuel, the reactor that burns it, and a plant that reprocesses the spent fuel back into fresh fuel. A closed loop, so that ideally nothing radioactive ever has to leave the perimeter.<\/p>\n The fuel plant is already running. It went into operation at the end of 2024 with four production lines, built around a dense mixed uranium-plutonium nitride fuel that packs in more material than the older MOX fuel used in Russia’s sodium-cooled fast reactors. So far the line has pressed pilot pellets using depleted uranium; full uranium-plutonium production is waiting on regulatory sign-off to handle plutonium on site. The first reactor load alone needs more than 200 fuel assemblies.<\/p>\n A fast reactor like this can breed more fuel than it burns and chew through some of the long-lived waste that ordinary reactors leave behind, which is the entire reason for wrapping it in a closed cycle in the first place.<\/p>\n The third piece, the reprocessing module, is the one still largely ahead of them. Rosatom’s plan puts its construction across 2025 and 2026, with commissioning around 2030. Until that part exists and works, the loop is a diagram, not a closed cycle.<\/p>\n For years the headline number on BREST was 2026. When construction started in 2021, Rosatom said the reactor should enter operation in 2026, and that date got printed a lot.<\/p>\n It is gone. Alexey Ferapontov, deputy chairman of Russia’s nuclear regulator Rostechnadzor, told Nuclear Engineering International that start-up is now set for 2029, “or maybe 2028.” Rosatom’s own director general, Alexey Likhachev, has given the same 2028-2029 window for firing up the reactor and the complex together.<\/p>\n So the honest read is the split from the top of this piece. The vessel gets buttoned up in 2026. The reactor comes to life a couple of years after that. Both are true, and only one of them is a 2026 story.<\/p>\n It is also fair to say these are timelines from Russian state sources on a first-of-its-kind machine, and first-of-its-kind machines slip. This one already has, and 2028-2029 is not the first date it has been handed.<\/p>\n The 300-megawatt unit is not really the goal. It is the proof. If it works, Rosatom wants to scale the design up to the BR-1200, a 1,200-megawatt version of the same lead-cooled concept.<\/p>\n On paper the ambition is large. Russia’s power plans out to 2042 pencil in eight BR-1200 units across four sites, two apiece at stations in the Sverdlovsk, Chelyabinsk, Seversk, and Irkutsk regions. Around 2035, Rosatom expects to have an export version of the whole complex ready to sell abroad.<\/p>\n That is a fleet plus a product, and all of it rides on one demonstrator in Seversk proving that molten lead and a reactor you pour in place actually hold up.<\/p>\n The rest of the industry is betting on small. Vessels welded shut in a day, reactors that fit on two soccer fields, boxes you truck to a site and bolt down, like the 300-megawatt machine America just filed to build next to the old Oak Ridge labs<\/a>. Russia went the other way and built a reactor into a building it can never move.<\/p>\n If the lead behaves and the corrosion stays in check, that building becomes the first working power reactor of its kind. If it does not, it becomes the most expensive argument yet for why everyone else is trying to make these things smaller.<\/p>\n","protected":false},"excerpt":{"rendered":" The nuclear industry has spent the last few years trying to make reactors smaller. The whole small modular reactor pitch, … <\/p>\nThe molten lead is the hard part<\/h2>\n
The whole fuel cycle behind one fence<\/h2>\n
The 2026 that quietly disappeared<\/h2>\n
Eight more, and much bigger<\/h2>\n