The Untold Story of China’s Forgotten Underground Nuclear Reactor

How social media and a little sleuthing turned up a Mao-era nuclear program.


Go to a conference about China’s nuclear weapons and you will hear, over and over again, that China is not very transparent when it comes to its nuclear program. That’s still true at a governmental level, but it is an increasingly outdated assessment of other aspects of Chinese society, especially in the age of social media. Western analysts have more access to information on these topics than they have ever had access to before, even if much of it is in Chinese. That has led to some startling discoveries.

For example, despite official secrecy about China’s production of plutonium for nuclear weapons, my colleague Catherine Dill and I discovered an underground nuclear reactor that China attempted to construct near Yichang in Hubei province during the 1960s and 1970s.

The Yichang reactor is different from the never-finished underground nuclear reactor near Fuling, in Sichuan province, which the government opened to tourists a few years ago. The reactor at Fuling was a surprise when Chinese authorities publicized it, but it was still only an unfinished copy of one of China’s above-ground nuclear reactors. The reactor at Yichang, on the other hand, is a totally different design. As far we can tell, the existence of the Yichang reactor has never been written about in English. We can find no evidence in declassified documents that the U.S. intelligence community knew, or knows, about the existence of this site.

We didn’t start out looking for a secret underground nuclear reactor. I recently finished writing an Adelphi book for the International Institute for Strategic Studies on China’s nuclear weapons program. My book makes extensive use of open sources. As part of the research, I was looking into part of China’s nuclear industry called the "827 Plant." China has lots of factories associated with the nuclear industry, so an unidentified plant wasn’t necessarily anything interesting.

I still wanted to know what role the 827 Plant played in China’s early nuclear weapons program, even if it was boring. When we started looking into the 827 Plant, I hoped we might find something exciting, like an unknown fuel-cladding plant. (That’s sarcasm, by the way. Not even nuclear-policy wonks think fuel-cladding plants are exciting.)

There is an enormous amount of open-source information available about China’s nuclear programs. Of course, that information is encrypted in a kind of tonal and ideogrammatic code referred to as Chinese. But it is out there. I have to admit that my Chinese skills are pretty limited. I once made the mistake of joking with a Chinese friend that I can only order a beer and start a fistfight in Chinese, which pretty much makes me proficient in the language. He laughed and has now spent the past decade introducing me as a "proficient" speaker of Chinese, largely to make me tell that joke again and again. Fortunately, the Monterey Institute, where I work, has a large number of amazing students and researchers, such as Catherine Dill, with both language skills and expertise in proliferation who can compensate for my deficiencies.

Searching for any reference to the 827 Plant, Catherine and I found a lot of references on resumes of Chinese nuclear engineers online. Plenty of Homer Simpsons worked there through the early 1980s, before finding employment in the civilian nuclear power program.

Then the resumes gave way to an amazing memoir we found online. Cui Zhaohui is a retired professor of nuclear engineering at Tsinghua University. He has written a very interesting memoir recounting his career, which started at China’s original plutonium production reactor near Jiuquan. Jiuquan was the original source of China’s military plutonium. While Cui never worked at the 827 Plant, he mentions in passing a colleague who was reassigned there. Almost as an afterthought, Cui mentions that the 827 Plant was "originally planned [as a] heavy-water nuclear reactor and reprocessing plant."

I had to read that again. Did he just tell me the 827 Plant was a secret nuclear reactor that I’d never heard about? Is this guy for real? Is he just pulling my leg? How does he know his friend wasn’t pulling his leg?

Suddenly, I was very interested in the 827 Plant. The Chinese plutonium-production reactors we know about were based on a totally different Soviet design that used graphite and plain old "light" water. Did China try to build a second underground nuclear reactor with a completely different design at the same time? We kept finding more and more evidence to suggest that Cui was right. One academic paper about a nuclear research facility in Beijing, for example, explained that it was a sort of prototype for a much larger, never completed heavy water reactor called the 827 Plant.

Eventually, we decided to locate the 827 Plant site. Lots of sources said the 827 Plant was near Chinese city of Yichang, in Hubei Province. Since Yichang is just downriver from the Three Gorges Dam, we had a temporary panic that the remains of the reactor were now underwater. 

Thousands of people lived and worked at the site before it was abandoned sometime in the 1980s. The children of those people are now adults. And like most grown-ups, they are nostalgic. They have reunions and even a blog. They have posted many pictures of the abandoned site online. The local municipality is attempting to redevelop the abandoned site as an artist’s community.

With so many pictures, it was easy enough to find the site in satellite images. The 827 Plant is located at 30°50’56"N, 111° 8’46"E, between Yichang and the Three Gorges Dam.

Then we noticed something amazing. If you look at the site in Google Earth with the Panoramio layer, there are a number of ground-truth images posted by the same person who maintains the 827 families blog. The title of one image (in Chinese) is "827 pump house." It appears to be the ruined pump house that brings river water for the secondary cooling loop for the reactor, which seems to have been constructed underground. Holy cow! Like the other reactor near Fuling, the paranoia of Maoist China had driven them to place the reactor underground to protect it from an attack that never came.

In the early 1960s, China had one plant to make highly enriched uranium near Lanzhou and was completing one nuclear reactor to produce plutonium at Jiuquan. In 1964, China began the "Third Line" effort — a massive construction effort to relocate all of China’s heavy industries, nuclear and otherwise, in the interior of the country. Often these factories, including things as mundane as steel mills, were placed underground to protect them from Soviet or American attack. As you might expect, the disruption of attempting to relocate the country’s heavy industries to underground caverns in the rural interior was a complete and total cluster… well, you know. Wikipedia calls the Third Line "an economic fiasco," which seems to me to be an example of the wisdom of crowds.

As part of the Third Line effort, China’s nuclear engineers were supposed to build a copy of the first reactor — the one where Cui worked — in an underground cavern being dug near Fuling. But placing a nuclear reactor under a mountain is about as slow and arduous as you might expect. At some point in 1969, with relations between Moscow and Beijing collapsing, Beijing decided it could not wait for the engineers to finish Fuling. The first proposal suggested physically picking up and moving the reactor near Jiuquan somewhere else. Eventually the technical personnel convinced the Chinese leadership this was total madness. So, instead, China started building a temporary replacement above ground, near a place called Guangyuan in Sichuan.

I always wondered how, in the middle of the paranoia associated with the Cultural Revolution, Chinese leaders came to their senses and ditched the underground reactor at Fuling in favor of the above-ground copy at Guangyuan. It turns out they didn’t. Instead of replacing one crazy project with a more sensible one, the Chinese doubled-down on crazy — continuing the reactor project at Fuling, starting a new one at Guangyuan and, we now know, starting the underground reactor at Yichang. That, come to think of it, sounds more like Mao’s China during the Cultural Revolution.

Not many leaders respond to resource bottlenecks by tripling reactor construction, but Mao Zedong wasn’t most leaders. As best we can tell, China never finished the heavy water at Yichang, just as it never finished Fuling or any other number of wildly implausible Third Line projects. Construction at Yichang lumbered on through the 1970s, before being shut down around 1980 or so. At this point, the Chinese government took a number of steps to transition its nuclear industry to civilian power generation, converting and eventually decommissioning the reactor near Jiuquan, as well as giving up on Fuling and Yichang. China would not build a heavy water reactor until it bought CANDU heavy water reactors from Canada, one in 2002 and another in 2003. (CANDU is a portmanteau of "Canada" and "deuterium oxide," better known as heavy water.) Yichang is just a footnote. A crazy, implausible footnote.

The failed effort to build an underground reactor near Yichang helps explain some mysteries about China’s plutonium production. China didn’t produce a ton of plutonium for a nuclear weapons state. (Well, it produced a ton — like one, maybe two, but almost certainly not more than four.) That means China’s nuclear arsenal cannot grow beyond a several hundred warheads unless it builds new reactors to make plutonium. One question has been whether there are, or were, other sites in China churning out plutonium in secret. That now seems unlikely. We now know that China, during the madness of the Cultural Revolution, tried to build several underground nuclear reactors in the 1970s and failed. During the 1980s, China ended these projects, converting and closing all of these facilities. Of the two plutonium production reactors that China finished, Jiuquan closed in the late 1980s and Guangyuan closed sometime in the 1990s. China never finished the reactors at Fuling or Yichang. China’s surprisingly small stockpile of plutonium isn’t so surprising once we know this historical context. They tried to make more. They just couldn’t.

Yichang also helps explain why China has been reluctant to negotiate to a treaty banning the production of fissile material for nuclear weapons purposes and cagey about the history of its plutonium production efforts. Secret projects like the underground nuclear reactor near Yichang may be one reason why. Publicizing this history, on the other hand, may provide an opportunity to engage the Chinese government on the need, even inevitability, of becoming more transparent.

It is time for Beijing to understand that we now live in an era in which social media and Google Earth can reveal the existence of a long-secret underground nuclear reactor. This is a different world than in the past, one that will be far more transparent than most governments realize at the moment. And we are only at the beginning of this era. There will be more disclosures. Like yet another unfinished secret underground nuclear reactor that I haven’t mentioned. That’s right, there is a third underground nuclear reactor project that we’ve found. But I will keep that one a secret. For now.

Jeffrey Lewis is director of the East Asia Nonproliferation Program for the James Martin Center for Nonproliferation Studies at the Middlebury Institute of International Studies at Monterey.

Trending Now Sponsored Links by Taboola

By Taboola

More from Foreign Policy

By Taboola