North Korea’s Nuclear Weapons: The Great Miniaturization Debate

The great debate over North Korea's miniaturization capabilities continues. On December 20, 2014, the South Korean Ministry of National Defense (MND) released a white paper that contained a surprising statement about North Korea’s nuclear program.[1] “North Korea seems to have made significant progress in miniaturizing its nuclear weapons.”

The MND Minister had made a similar statement in October, but for some reason, this time his statement sparked a flurry of stories in South Korean press, such as the Chosun Ilbo and JoongAng Ilbo, as well as in US publications like the Wall Street Journal and the New York Times.

This chatter forced the South Korean government to clarify the statement. “Seoul and Washington have reached consensus that the North already reached a significant level of technology to miniaturize nuclear weapons through three nuclear tests,” an MND official told the Chosun Ilbo. “But there is no intelligence report that the North has already succeeded in miniaturizing nuclear weapons.”

Well that clears it up.

This is now the third time something like this has happened in the past few years—a statement that North Korea has developed a nuclear weapon small enough to arm a ballistic missile of one sort or another, followed by oddly parsed statements suggesting that maybe they haven’t.

In Spring 2013, for example, a US Defense Intelligence Agency (DIA) threat assessment was mistakenly marked unclassified stating that North Korea might be able to arm ballistic missiles with nuclear weapons, prompting the Department of Defense and Director of National Intelligence to release clarifications of their own. And, in October of that year, the Commander of US Forces Korea stated his personal opinion that North Korea probably could do so, prompting a statement by the ROK Minister of National Defense.

At some level, this debate strikes me as a bit bizarre. The North Koreans have conducted three nuclear weapons tests since 2006, including one they openly declared to have been of a “miniaturized” device; they have also created a Strategic Rocket Force and published a picture of a map showing their nuclear targeting plan against the United States. I realize that North Korean propaganda is often balderdash, but the idea that North Korea might be developing nuclear-armed ballistic missiles isn’t really  in the same category as claims that Kim Jong Un doesn’t poop.

Whether North Korea can arm a ballistic missile with a nuclear warhead, particularly a ballistic missile that can reach the United States, depends on the answer to three questions:

  1. Can North Korea make a nuclear weapon small enough?
  2. Can North Korea’s compact nuclear weapon survive the shock, vibration and temperature change associated with ballistic missile flight?
  3. Can North Korea construct a “reentry vehicle” that can survive the extreme heat of reentry, a problem that gets worse with range?

I think the answer to each of these questions is, “yeah, probably.” While I understand the caution in crediting the North Koreans for capabilities that are only under development, there is ample open source information to support such a judgment. Reasonable people may still disagree, but no one should be surprised by the prospect of nuclear-armed North Korean missiles.

Can North Korea Make a Nuclear Weapon Small Enough?

The simplest question is whether North Korea can build a nuclear weapon small enough—both in terms of mass and compactness—to fit atop a ballistic missile. The United States intelligence community has a term of art—simple fission device—to describe first generation nuclear weapons, like the aptly named “Fat Man,” that are much too large to place on a ballistic missile.

As a general technical matter, however, the US intelligence community has always stated that a country could skip right toward building much smaller devices on the order of 1,000 kg—although such weapons would be unreliable without nuclear testing. This device would look something like the US Mark 7, which weighed about 750 kg. Some of my colleagues have pointed out that North Korea could probably do much better, trying out something like the Mark 12 which weighted on 450 kg. (See chart.)

Select Early US Nuclear Warhead Designs for Comparison
Mass (kg) Diameter (cm) Yield (kt) Deployed
Mark III (“Fat Man”) 4,700 150 20 1945, 1947-1950
Mark 5 1400 110 120 1952-1963
Mark 7 800 80 70 1952-1967
Mark 12 500 60 20 1954-1962
Source: Chuck Hansen, Swords of Armageddon, electronic edition.

Not surprisingly, as early as 1999, DIA was arguing that North Korea might try to build a 650-750 kg device, even if others in the US intelligence community were skeptical. DIA just assumed that North Korea would go straight to a Mark 7-like design.

There is plenty of reason to think that North Korea tried to do precisely that. During the 2000s, there were many reports of North Korean conducting extensive testing of high explosives. A nuclear weapon is mostly a conventional explosive. Making the bomb more compact largely involves design innovations that require fewer explosives to achieve a given level of compression (such as levitated pits and better electronics). One explanation for all the testing of conventional explosives is that North Korea was trying to develop a device small enough to be delivered by missile. In 2005, a North Korean defector stated that North Korea had done precisely that, build a 1,000 kilogram device that was—just as the US intelligence community would have predicted—not reliable. (The defector also said the next device would be smaller.)

When North Korea’s first test in 2006 produced a very disappointing yield, many of us took the small yield to be confirmation of this general hypothesis—North Korea had tried to skip directly to a compact device and it did not work. At one point, a reporter told me this was also the working hypothesis within the US intelligence community. Since then, North Korea has conducted two more nuclear tests that produced far higher yields—a few kilotons in 2009, followed by several kilotons in 2013. Following that lest test, the North Koreans announced they had “miniaturized” their nuclear devices.

It seems very plausible to me that, after three tests, the North Koreans have a nuclear weapons design somewhere in the Mark 12 to March 7 range—450-750 kg in mass with a diameter between 60-90 cm. Lots of states have moved quickly to develop relatively smaller devices. (See chart.) The Chinese provided a uranium-based design to Pakistan that was 500 kg and 90 cm in diameter, which the Pakistanis miniaturized and passed on to Libya and lord knows who else.

Other Early Generation Compact Nuclear Devices
Mass (kg) Diameter (cm) Yield (kt) Vintage
CHIC-4 (China/Pakistan/Libya) 500-1200 90 10 1960s, 1980s
Iraq (Al Qa’qaa molds) 900 80 20 1980s
Pakistan Miniaturized 200 60 10 1980s
R265/R288 58
Sweden* 400-500 35 20 1950s
*The Swedish design made use of oblate high explosives configuration.
Source: Author estimates.

Such a warhead is certainly small enough to arm a Nodong and might just fit on a notional DPRK inter-continental ballistic missile (ICBM). The problem here is how to estimate the capabilities of a DPRK ICBM that does not exist–based on Unha technology or the KN-08 mockups? If North Korea can’t make a warhead compact enough for its ICBMs, it is more likely to be because the ICBM doesn’t have enough payload space.

Can North Korea’s Compact Nuclear Weapon Survive the Shock, Vibration and Temperature Change Associated with Ballistic Missile Flight?

This is a more interesting problem. It’s all well and good to design a much smaller nuclear weapon using fancy electronics and so on, but the design must be rugged enough to survive the shock, vibration and temperature extremes of taking a ride on a ballistic missile. “The difference has to do with the confidence level in the actual ability of the North Koreans to make a weapon that will work in a missile,” James Clapper, Director of National Intelligence, explained in 2013, “And neither we nor the North Koreans know whether that’ll actually—whether they have that—such capability, if they have it, will actually work.”

This was a real problem for the Chinese in the 1960s, too. The Chinese developed a missile-delivered warhead for their DF-2 ballistic missile—the same design that China provided to Pakistan—and originally planned to simulate the abuse suffered during a real launch, followed by an underground test of the roughed-up bomb.

The Chinese, however, decided that it was too hard to simulate the extreme conditions of flight. After a fair amount of back-and-forth between the weaponeers and the central leadership, Zhou Enlai authorized a very unusual live test of a real nuclear weapon on a real ballistic missile. China fired a nuclear-armed DF-2 in October 1966. It worked. The Chinese weren’t alone. We had the same debate in the United States a few years before. Like China, we also settled for a one-off demonstration called Operation Frigate Bird, in which a US submarine fired a nuclear-armed Polaris missile at a nuclear test site in the South Pacific. It worked too, although it later turned out that the warhead in question was judged unreliable.

We might lack confidence in North Korea’s ability to manufacture a reliable miniaturized nuclear weapon. I wonder, though, how much that matters. Do the North Koreans lack confidence in their warheads? What if we underestimate them? What if they are drunk off Juche? What if, like Operation Frigate Bird, the unreliable weapon just happens to work when it’s fired? There is an interesting discussion to be had about reliability, confidence and deterrence, but I wonder whether it adds much to our assessment of North Korea.

Can North Korea Construct a “Reentry Vehicle” that Can Survive the Extreme Heat of Reentry?

Finally, no matter how rugged one makes a nuclear warhead, it has to be packaged in a reentry vehicle that can survive the heat created as it reenters the earth’s atmosphere. The North Koreans could certainly package a warhead in a blunt reentry body that would be inaccurate, very heavy and potentially vulnerable to theater missile defense systems—but it would still survive reentry.

The North Koreans, however, have paraded missiles with so-called “triconic” reentry vehicles that are sort of a compromise between blunt reentry bodies and the slender cones that arm missiles in the US and other advanced nuclear powers. A triconic reentry body must deal with heat through ablation—in other words, the reentry body must be made of material that burns off, taking the heat with it.

This can be a significant challenge for an ICBM, where reentry speeds can reach 7 km/s. China, for example, struggled in the 1970s with developing a reentry vehicle for the DF-5 ICBM that could handle such temperatures. China Today, a series of publications on the technical history of China’s defense industries, describes the problem as “a technical difficulty” which is about as colorful as China Today gets. Ultimately, though, the Chinese solved that problem. In fact, I can’t think of a single state that has been able to build an ICBM, but not able to put a passable reentry vehicle on top of it.

It is common to say North Korea would require a program of testing to overcome these problems. That’s understandable. In the 1960s, reentry vehicle designers probably struggled to model reentry environments and had a limited choice in materials. But today? After more than fifty years of space flight? With a large body of open source information, better computer simulation capabilities and fancy new materials? Maybe a little help from their friends? And maybe a little overconfidence?

And, let’s be clear about the problem here. The warhead probably won’t burn up. Even the North Koreans don’t suck that badly. When designers talk about how hard it is to design an ablative reentry vehicle, what they really mean is designing one where the ablation occurs evenly around a spinning reentry vehicle. The Chinese were as worried about “the stability of the warhead in flight” as they were about protecting the bomb package inside. An unstable reentry body might fail completely, but it is more likely to just wildly miss the intended target—say landing in San Jose when it was aimed at San Francisco. That’s a problem, of course, but Kim Jong Un might be content with such an outcome.


It is not surprising that some people in the US or ROK government think that, yes, North Korea might be able to do these things. Nor is it surprising that others would counsel caution, suggesting that North Korea hasn’t put all together in a single test. North Korea’s missile and nuclear “developments have been accompanied with extremely belligerent, aggressive public rhetoric toward the United States and South Korea,” Clapper testified in 2013. “North Korea has not, however, fully developed, tested or demonstrated the full range of capabilities necessary for a nuclear- armed missile.” In other words, prove it.

But is that really what we want? Looking at the Chinese example, do we really want to insist that North Korea arm a missile with a live warhead and conduct a demonstration? A much better solution is trying to negotiate limits on North Korea’s nuclear and missile programs. Such limits would not eliminate the threat these programs pose, but they might keep them unreliable. That would be an achievement.

I’ll be the first person to say that we should not exaggerate the capabilities of North Korea’s nuclear forces, but underestimating them is every bit as bad. The North Koreans are developing military capabilities that we will, sooner or later, have to deal with. I just happen to think that negotiations, as frustrating as they may be, are the best of a series of unappealing options.

Moreover, underestimating the North Koreans often means that, when they surprise us, our political system over-compensates, passing from denial straight into panic. Consider the case of the August 1998 Taepodong launch. The US intelligence community had assessed, in 1995, that “No country, other than the major declared nuclear powers, will develop or otherwise acquire a ballistic missile in the next 15 years that could threaten the contiguous 48 states and Canada.” As it turns out, more than fifteen years later, they were right. (And the fine print on North Korea and the Taepodong program was pretty decent, as well.) So, when North Korea launched a Taepodong in 1998 with an unexpected third stage that failed, the intelligence community got a great big “congratulations” for a job well done. Oh, wait, no it didn’t. The intel was right, but that didn’t matter in part because the technical assessment didn’t convey North Korea’s ambitions to develop a capability that outstripped its abilities.

Just imagine if North Korea were to conduct a live demonstration of a nuclear weapon on a Nodong out to sea. Even if it didn’t work, Washington, Seoul and Tokyo would go bonkers. That’s worth keeping in mind. Yes, the North Koreans probably stink at making compact warheads and accurate reentry vehicles. But that’s not quite the same thing as saying they aren’t trying, that they don’t have some confidence in these capabilities or that we shouldn’t keep trying to find ways to discourage them from testing these systems.

Jeffrey Lewis is Director of the East Asia Nonproliferation Program at the James Martin Center for Nonproliferation Studies (CNS), Monterey Institute of International Studies, and a frequent contributor to 38 North.


[1]The statement, in Korean, is “북한의 핵무기 소형화 능력도 상당한 수준에 이른 것으로 보인다.”

Stay informed about our latest
news, publications, & uploads:
I'm interested in...
38 North: News and Analysis on North Korea