Estimating India’s nuclear weapons-producing capacity

Postage Stamp, Apsara Research Atomic Reactor in Trombay, Mumbai, circa 1965Indian postage stamp, Apsara Research Atomic Reactor in Trombay, Mumbai, circa 1965. Credit: Shutterstock

How many nuclear weapons India can make is an important question in South Asia, determining the foreign and military policies of a number of countries in the greater region, and a concern for the world at large.

But coming up with a reliable answer is tricky.

Most estimates, such as those of the Carnegie Endowment for International Peace, get the raw material for their reports on a given country—such as the endowment’s study, “A Normal Nuclear Pakistan”—by relying on a relative handful of sources, such as the yearbooks of the Nuclear Threat Initiative, the International Panel on Fissile Materials, and the Stockholm International Peace Research Institute, to name some common examples. Bearing in mind the relative paucity of original source material, a 2015 estimate by the Institute for Science and International Security concluded that India’s stockpile of fissile material was only sufficient to make approximately 75-to-125 nuclear weapons.

But two recent analyses have produced much larger estimates. In 2016, Syed Muhammad Ali estimated in his book Indian Unsafeguarded Nuclear Program: An Assessment that India has sufficient material and technical capacity to produce between 356 and 492 plutonium-based nuclear weapons. A year later, in a discussion paper for Harvard University’s Belfer Center, Mansoor Ahmed concluded that India has the capacity to produce up to 2,686 nuclear weapons.

What accounts for most of these dramatic differences, from the low double-digits to a high in the thousands? It is the possibility, all-too-easily discounted, that India could use civilian reactor-grade fissile materials in its nuclear weapons, and how efficient such an approach could be.

Experts claim there are good reasons to assume it is extremely unlikely for India to draw upon its civilian reactors to make fissile materials for its nuclear weapons. For one thing, it is difficult and expensive to reprocess the plutonium in the spent fuel from a civilian nuclear reactor for use as the vitals of a nuclear weapon. And “India in particular has historically had trouble achieving consistent operations in its reprocessing facilities,” as a Bulletin article titled “Fuzzy math on Indian nuclear weapons” noted in 2016. (The International Panel on Fissile Materials came to much the same conclusion in 2015, citing the long-term poor performance of India’s reprocessing plants at Kalpakkam and Tarapur.) And don’t forget that these are unsafeguarded reactors, which by definition means that IAEA inspections are not allowed—so the outside world cannot rule out the possibility that these reactors could be used for military purposes.

And reactor-grade plutonium, with its lower proportion of the desirable isotope of plutonium for nuclear weapons (plutonium 239), is much more likely to produce a much smaller explosive yield than wanted, or “fizzle,” thus requiring a larger volume of fissile material to reach critical mass. All in all, while it is possible to make a nuclear warhead from reactor-grade plutonium, these researchers think it is a much more complicated (and risky) approach than making a warhead from weapon-grade plutonium.

Yet, while the odds may be low, they are not zero.

And India does have an extraordinarily large stockpile of reactor-grade plutonium. (Ahmed estimates India has about five metric tons of reactor-grade plutonium that has already been separated, and another 10 metric tons in spent fuel awaiting separation.) Admittedly, this stockpile is most likely intended as fuel for India’s Prototype Fast Breeder Reactor, and not for nuclear weapons.

But what if, for whatever reasons, at some time in the future, under conditions we cannot foresee right now, that stockpile were to be repurposed? Purely as a thought experiment, how many nuclear weapons is it possible for India to produce from this supposedly second-rate collection of fissile material?

Using a different set of theoretical assumptions, and what we consider to be very credible mathematical formulas, we found that contrary to previous estimates, India has the capacity to produce many more nuclear weapons than generally assumed. We estimate that India could produce 1,044 nuclear weapons (914 plutonium-based and 130 uranium-based nuclear weapons), if one includes reactor grade materials from non-military programs in India, as well as that from the country’s weapon-grade nuclear material production program.

A new paradigm. To support these findings, we carefully examined every aspect of Indian unsafeguarded nuclear reactors and included important factors such as each reactor’s history, size, and amount of material produced over its working lifetime. We have considered the Reactor-Grade Plutonium (RGp), Weapon-Grade Plutonium (WGp), and Highly Enriched Uranium (HEU) produced by these unsafeguarded facilities.

In addition to the calculations contained here, we have attached a link to an executable file based upon the Python computer programming language. (Python helps us to give instructions to computers in a language it understands.) This will allow readers to do their own calculations of the nuclear weapons-generating capacity of India—or any other nuclear state. All the reader needs to do is enter data such as how much reactor-grade plutonium is generated per year by a given nuclear reactor, the reactor’s power in megawatt electric, plutonium conversion factors, the length of time that the reactor has been in operation, and so forth—much of which is accessible online, through publicly available sources. As we demonstrate in the rest of this article, by entering this data, one can get a rough approximation of a nation’s capacity to make nuclear weapons using reactor-grade material. But it is important to note that to run this software the reader needs to download and install Python 3.6.0, and then enter the scripts from the Word files found elsewhere in this article, before plugging in the numbers.

No debate: On existential threats, Biden, Trump, and CNN all largely failed

Before we begin, however, we need to deal with the elephant in the room.

Can Reactor Grade Plutonium be used in nuclear weapons? There is a commonplace conception that Reactor Grade Plutonium (RGPu) cannot be used in nuclear weapons.

This argument is incorrect.

Reactor-grade plutonium can, in fact, be used in nuclear weapons, according to a 1997 report published by the US Energy Department, which said at the bottom of page 39 that “In short, reactor-grade plutonium is usable, whether by unsophisticated proliferators or by advanced nuclear weapon states.” (The report also said that “The possibility that either a state or a sub-national group would choose to use reactor-grade plutonium, should sufficient stocks of weapon-grade plutonium not be readily available, cannot be discounted.”) In addition, declassified US government documents such as the Energy Department’s FAQ titled “Additional Information Concerning Underground Nuclear Weapon Test of Reactor-Grade Plutonium” highlight that the United States itself had conducted a nuclear weapon test using reactor-grade plutonium back in the early 1960s. And an Indian nuclear scientist acknowledged to the magazine Economic and Political Weekly that India has conducted at least one nuclear test using reactor-grade plutonium.

And there is one other argument for factoring in India’s reactor-grade plutonium when trying to estimate the country’s capacity to build nuclear weapons: The historical record shows that India has long considered the potential of using reactor-grade plutonium in nuclear weapons. A 2015 study by David Albright of the Institute of Science and International Security said at the bottom of page 11: “Although generally India is not believed to use reactor-grade plutonium in nuclear weapons, Indian nuclear experts are reported to have evaluated this plutonium’s use in nuclear weapons and India may have decided to create a reserve stock of reactor-grade plutonium for possible use in nuclear weapons.”

And India has a large supply of this reactor-grade plutonium to draw upon. The biographer Ganeshan Venkataraman, an Indian condensed matter physicist himself, recounts some of India’s history regarding this stockpile in his book Bhaba, and His Magnificent Obsessions. In November 1954, Homi Jehangir Bhabha, widely considered the father of India’s nuclear program, presented his vision of the country’s nuclear fuel cycle. It would consist of three phases: In the first stage, India would build Canadian-designed reactors, which run on natural uranium, to produce plutonium. In the second stage, India would build reactors which could run on a combination of the plutonium produced from the first stage and locally available thorium to produce uranium 233. In the third stage, India would build fast breeder reactors that could run on the resulting uranium 233.

As events turned out, however, India has not been able to master this complex nuclear fuel cycle. But it has produced a large amount of reactor grade plutonium, which can be used in nuclear weapons when necessary.

As for how much reactor-grade plutonium would be necessary to provide the fissile material for a nuclear bomb, that depends in part on the quality of the spent fuel from the reactor that it is scavenged from. The percentage of plutonium in a reactor’s spent fuel depends on a number of factors—including reactor design, fuel type, burnup, and the amount of time that plutonium is exposed to fusion within the reactors—so this percentage is not fixed. (Burnup is a measure of how much energy is extracted from a nuclear fuel, and also a measure of fuel depletion.)

But as the IAEA’s Department of Safeguards has noted, “even highly burned reactor-grade plutonium can be used for the manufacture of nuclear weapons capable of very substantial explosive yields.”

With these considerations in mind, let us take another look at that previously mentioned estimate by Syed Muhammad Ali—the one that said that India has the capacity to produce about 356-to-492 plutonium-based (including reactor grade plutonium) nuclear weapons—and recalculate.

First, we start with how much reactor-grade plutonium India is likely to have. In a 2005 article in Foreign Policy in Focus, Zia Mian and MV Ramana estimated that India may have produced a total of about 8,000 kilograms of reactor-grade plutonium up to that date, from power reactors not under safeguards.

[Click this Word file, Python Program for RGp and WGp, to download the scripts necessary to run the Python Coding programs for Reactor Grade Plutonium (RGp) and Weapon Grade Plutonium (WGp).]

Indian reactor data.

Next, let us assume that in converting from fertile to fissile material, we get a conversion factor of about 0.3—meaning loosely in lay terms that just under one-third of reactor-grade plutonium is converted into weapon-grade plutonium suitable for a nuclear bomb. (According to an FAQ on the website of the nonprofit organization—built entirely by a group of nuclear engineers—a light water reactor has an average conversion factor of 0.5. But to make our calculations more credible, we have gone with the more conservative 0.3 conversion factor.) This means that a stockpile of the size of India’s can produce approximately 2,400 kilograms of weapon-grade plutonium (8,000 kilograms x 0.3 conversion factor = 2,400 kilograms).

Next, let us assume that 4 kilograms of weapon-grade plutonium are needed as the fissile material for one nuclear bomb. (According to the Union of Concerned Scientists FAQ “How much is needed to build a bomb?” a modern, sophisticated implosion nuclear weapon design would require 2-to-4 kilograms of plutonium—which the previously cited paper from Harvard University’s Belfer Center and other sources indicate is a technological feat well within India’s capacity. To err on the side of caution, we will do the calculations with 4 kilograms of plutonium per weapon.) This means that about 600 nuclear weapons can be produced by India, using just its stockpile of reactor-grade plutonium alone, and excluding all other sources. (2,400 kilograms divided by 4 kilograms per bomb = 600 nuclear bombs.)

Texas wildfires force major nuclear weapons facility to briefly pause operations

As can be seen, this figure is substantially more than the commonly accepted maximum number.

And even this 600 nuclear bomb figure is low. After all, it was calculated using the best estimate—8,000 kilograms—of India’s reactor-grade plutonium as of the year 2005.

It is now a full 13 years later, and India has obviously had time to produce much more of this material. So, we need to recalculate, from scratch.

The latest estimate of Indian reactor and reactor-grade plutonium is given below, assuming 0.7 capacity factor, 0.3 kilograms per metric ton plutonium conversion factor, 310 days per year reactor availability factor, and that 4 kilograms of plutonium are required for each nuclear weapon. The formula and calculation areas follows:

Total Plutonium in Kilograms = Reactor Power in Megawatts electric x Capacity Factor x Days Under Operation x Plutonium Conversion Factor x 0.001

Indian reactor-grade plutonium, in kilograms.

As can be seen, this means that India has enough reactor-grade plutonium on hand at this time to make 839 nuclear bombs from this material alone. But we are interested in India’s total nuclear weapons capacity, so we need to also include its capacity to produce weapon-grade plutonium.

Weapon-grade plutonium calculations. According to official government documents, India has two reactors producing weapon-grade plutonium: the 100 megawatt (thermal) CIRUS reactor and the 40 megawatt (thermal) Dhruva reactor. However, to estimate India’s weapon-grade plutonium, we need to convert Megawatt Thermal to Megawatt Electric. (Jordan Hanania and his team at the University of Calgary’s Energy Education department note that “Megawatts electric or MWe is one of the two values assigned to a power reactor, the other being megawatts thermal or MWt. Megawatts electric refers to the electricity output capability of the plant, and megawatts thermal refers to the input energy required.”) Given that 3,000 MWt is equal to 1,000 MWe, we need to convert those figures, as seen below.

[Click this link, Python Program for Calculating Nuke Weapon, to download the Word file containing the scripts necessary to run the Python Coding program for Calculating Nuclear Weapons-making Capacity from Weapon-Grade Uranium.]

India’s total weapon-grade plutonium-based capacity.

The calculation is as follows: Total Plutonium (kg) = Reactor Power in MWt x Capacity Factor x Days Under Operation x 0.001

From these results, we see that means that another 75 nuclear weapons can be added to India’s total weapon-producing capacity, so we need to add that to our previous figure.

Total Plutonium-based Nuclear Weapon-Making Capacity: 839 + 75 = 914 Nuclear Weapons

But while we now have an idea of India’s nuclear capacity using both reactor-grade and weapon-grade plutonium, there is still the matter of the country’s uranium-based nuclear weapons.

Highly Enriched Uranium (HEU) calculations. India is also pursuing a uranium mining and leaching program using its unsafeguarded reactors. However, estimating Highly Enriched Uranium production is quite challenging. Unlike making the calculations for reactor-grade plutonium, HEU requires more classified data, which is unavailable in public domain—so analysts must guess the actual number of centrifuges in a particular plant, the enrichment capacity of each centrifuge (which is dependent on the diameter and height of each machine), the quality of the feed material, and whether the rotor is made of aluminum, maraging steel, or carbon fiber. None of these factors can be accurately estimated without intrusive inspections. Therefore, it is hard to know for sure exactly how much HEU it has acquired.

Consequently, we will not make our own assessment but instead use reports—from organizations such as ISIS, the International Panel on Fissile Materials, and South Asian Voices—that can help give an idea of the total, as they have already calculated the amount of HEU each individual source has produced. The average calculations and ranges for these figures are as follows:

Average calculations and ranges for numbers of India’s uranium-based nuclear weapons, to date.

That gives us an average total of 130 uranium-based nuclear bombs that have been produced by India up to today’s date.

When we add that to the previous figure of 914 for all weapons that India has the capacity to produce by using reactor-grade and weapon-grade plutonium, we come up with a grand total of 1,044 nuclear weapons.

What does it all mean? There are many estimates that have attempted to calculate India’s nuclear weapon-making capacity, but this study explores an often-overlooked aspect, using mathematical formulas and publicly available information about reactor-grade material. Using these parameters, our study finds that India has the capacity to produce many more nuclear weapons than commonly thought: 1,044 nuclear weapons (914 plutonium-based and 130 uranium-based nuclear weapons), obtained from different reactors.

Of course, there is a big difference between the possible theoretical capacity to do a thing—especially if it is difficult, expensive, labor-intensive, time-consuming, risky, and unorthodox—and actually doing it.

But as former White House Coordinator for Arms Control and Weapons of Mass Destruction Gary Samore once told Foreign Policy magazine: “I believe that India intends to build thermonuclear weapons as part of its strategic deterrent against China.” According to the article: “It is unclear, he continued, when India will realize this goal of a larger and more powerful arsenal, but ‘they will.’ ”

The path outlined here is one way this could happen, without the world even being aware.

Together, we make the world safer.

The Bulletin elevates expert voices above the noise. But as an independent nonprofit organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.

Get alerts about this thread
Notify of

Newest Most Voted
Inline Feedbacks
View all comments
5 years ago

Excellent paper. Thanks for the python program, it works greatly. The fact that India has the capacity to produce enough fissile material from its unsafeguarded reactors is dangerous, as they are not under IAEA safeguards.

5 years ago

Hi Syed! Great work.

5 years ago

These estimates are based on India doing something that no possessor of nuclear weapons has ever done: using reactor-grade products for an operational stockpile. Quite higher estimates can be derived from using breeder products for an operational stockpile — something that also has not been done, at least as far as I can tell.
These actions can’t be ruled out, of course, but these are worst case estimates. We all know where this gets us.

Sameer Ali Khan
Sameer Ali Khan
5 years ago
Reply to  krepon

There have been tests of RG Pu based weapons

Also, Indian scientists claim that they did so.
I am not aware of any means to independently verify these claims by Indian scientists.

5 years ago

Sameer, you may find more evidences in the book “Reactor Grade Plutonium and Nuclear Weapons,” which can be accessed at
Not only India, but also, Britain, US and other countries have done it. I hope it may be useful.

5 years ago

Please do not forget the extremely useful book “Reactor-grade Plutonium and Nuclear Weapons written by Greg Jones. The book is available from his website for download.


Receive Email