Yes, Virginia, India’s H-bomb fizzled.
K Santhanam (who was director of test site preparations for India’s 1998 nuclear tests; pictured above, handing the firing keys to the range safety officer) has admitted what everyone else has known for a long time — that India’s 1998 test of a thermonuclear device was unsuccessful:
“Based upon the seismic measurements and expert opinion from world over, it is clear that the yield in the thermonuclear device test was much lower than what was claimed. I think it is well documented and that is why I assert that India should not rush into signing the CTBT,’‘ Santhanam told [the Times of India] on Wednesday.
Sources said that Santhanam had admitted that the test was a fizzle during a discussion on CTBT organised by IDSA. Karnad also participated in the seminar. He told TOI that no country has succeeded in achieving targets with only its first test of a thermonuclear device.
“Two things are clear; that India should not sign CTBT and that it needs more thermonuclear device tests,’‘ said Santhanam.
This is a subject we have covered in some detail here at Arms Control Wonk.com (see The Bomb, Dmitry. The Hydrogen Bomb, 10 April 2005.)
1. Yes, India’s thermonuclear device probably probably did fizzle, looking at the seismic data.
2. Some Indian scientists, including the former chairman of India’s Atomic Energy Commission PK Iyengar and now Santhanam, keep pointing out this somewhat embarrassing fact because it is part of an argument for India to resume nuclear testing.
3. India’s evident need to resume testing to complete development of a thermonuclear device is the principal reason that I opposed a “clean” NSG exemption for India (See: Will India Test Again?, 23 June 2008 ).
Here is my original post on the subject, reproduced because I am lazy and I recall the reading list was somewhat helpful:
Did India successfully test a two-stage thermonuclear device in May 1998?
There are substantial reasons for skepticism. India claimed that it detonated three devices on 11 May 1998 at Pokhran (right)—a 43-kiloton thermonuclear explosion, a 12-kiloton fission explosion and a 0.2-kiloton fission explosion. (India then claims to have conducted low yield tests on 13 May 1998.)
Seismic analyses (particularly Wallace et al) conclude the cumulative yield for the 11 May tests was only 12-kilotons. A yield that low is probably “too small to have been a full test of a thermonuclear weapon”—suggesting the test fizzled.
The US intelligence community reportedly shares this conclusion. Govenment officials told Mark Hibbs of Nucleonics Week that analysts from Livermore’s Z Division “have now concluded that the second stage of a two-stage Indian hydrogen bomb device failed to ignite as planned.” Subsequently, “senior U.S. expert” confirmed to Hibbs that this account was correct.
Indian scientists have been quick to dispute these estimates, arguing that Western scientists have made inaccurate assumptions about the geography of the Indian test site. This argument has always struck me as unconvincing, in part because of data that has been presented from the 1974 test.
A former chairman of India’s Atomic Energy Commission, PK Iyengar, has used calculations similar to those of Wallace et al to suggest that the second stage of the two-stage thermonuclear weapon failed to ignite—“the fusion core burnt only partially, perhaps less than 10 per cent.” Iyengar, however, has an axe to grind—he wants India to resume nuclear testing.
Such failures have plagued new nuclear designers before. China’s seventh nuclear test (CHICOM 7)—and second thermonuclear weapon—also fizzled, resulting in a yield estimated at the time between 15-25 KT.
Similarly, Livermore’s first attempt at “super” also failed—resulting from what Herb York called “a simple design flaw … engendered by the novelty of the technology and by our inexperience.”
Brian Barker et al, “Monitoring Nuclear Tests,” Science 281:5385 (25 September 25, 1998) 1967-68 (subscription).
Mark Hibbs,”India May Test Again Because H-Bomb Failed, U.S. Believes,” Nucleonics Week 39:48 (26 November 1998) 1.
Mark Hibbs, “Because H-Bomb Fuel Didn’t Burn, Iyengar Pleads For Second Test,” Nucleonics Week (1 June 2000) 6.
PK Iyengar, “Nuclear Nuances,” The Times of India (22 August 2000) (full text in the comments).
SK Sikka et al,”The recent Indian Nuclear Tests: A Seismic Overview,” Current Science 79:9 (10 November 2000) 1359-1366 (draft).
Gregory van der Vink et al, “False Accusations, Undetected Tests and Implications for the CTB Treaty,” Arms Control Today 28:4 (May 1998) 7-13).
Terry C. Wallace, “The May 1998 India and Pakistan Nuclear Tests,” Seismological Research Letters 69 (September 1998) 386-393 (preprint).]
Herbert York, Making Weapons, Talking Peace: A Physicist’s Odyssey from Hiroshima to Geneva (Basic Books, 1987) 78.
Credible Deterrent Through Testing
By P K IYENGAR, August 2000
(The author is former chairman of the Atomic Energy Commission)
AFTER a long period of indecision and ambivalence regarding its nuclear preparedness, India detonated five nuclear devices in May 1998. Consequently, it declared itself a nuclear weapons country. However, it is unlikely that we will be accepted as a weapons country under the Comprehensive Test Ban Treaty (CTBT), because there is no provision for threshold states maturing to become nuclear powers.
When Prime Minister Atal Bihari Vajpayee visits the US next month, the CTBT will certainly be one of the important issues raised by the Americans. There may be political arguments both for and against signing the CTBT, but if we have decided to follow a policy of nuclear deterrence, which will require weaponisation, then, scientifically, we have no option but to continue testing. It is the scientific case that I wish to make here.
If one goes by the numbers for the total nuclear yield put out by the Department of Atomic Energy, which I see no reason to dispute, the yield of the thermonuclear device detonated on May 11, 1998 was around 40 kilotons. This is a rather low yield. If the yield was deliberately kept low to restrict damage to the nearby villages, then surely it would have been more sensible to test the thermonuclear device separately, and not along with the 15 kt fission device. Now, the thermonuclear device itself consisted of two parts: the fission trigger and the fusion core.
The crucial question is not what the total yield of the device was, but what was the ratio of fission energy to fusion energy? Clearly, for a given total yield, the greater the fraction of the fusion energy, the more efficient is your thermonuclear device. In my opinion, that ratio musts have been around 1:1, and no one has so far, to my knowledge either publicly or privately, disputed that number. Therefore, by my estimate, the fusion yield could not have been more than 20 kt. Further, it seems likely that a fission `spark-plug’ was used at the centre of the fusion core, in which case the actual fusion yield would have been even less.
Sticking to the larger number of a 20 kt fusion yield, one can easily calculate that the amount of LiD fusion material needed would be only around 400 grams or around 500 cc. This is a very small size for the fusion core, and the actual core used must certainly have been much larger. This suggests that the fusion core burnt only partially, perhaps less than 10 per cent. This can easily be checked; if the burn was only partial, there should have been a lot of tritium produced, which should have been detected after the explosions.
In such a complex system as a two-stage thermonuclear device, getting any burn at all is a credit to the abilities of the scientists and engineers of the Bhabha Atomic Research Centre (BARC). However, a thermonuclear device that only burns partially is certainly inefficient. Logically and scientifically, the next step would be to improve the design of the device to achieve greater efficiency. This is particularly important from the point of view of a weaponisation programme.
The government has declared a policy of maintaining a minimum nuclear deterrent. Nuclear deterrence means that we have a demonstrable nuclear capability that deters a potential adversary from attacking us. For us to have a nuclear deterrent we must weaponise. For this, we must have fusion weapons, because these are smaller, lighter, and more efficient than fission weapons. But for that deterrent to be viable, we must master all aspects of thermonuclear weapons, and demonstrate that expertise not just in one, but many thermonuclear designs, particularly those of greater efficiency.
Whether that should include a neutron bomb or not, is not of the essence. In a neutron bomb, one establishes a thermonuclear burn by igniting only a part of the core, and making the burn propagate. This is the crux of the matter in designing an efficient thermonuclear device. One may not have a neutron bomb in one’s arsenal, but it would strengthen our abilities if we successfully tested one.
Some people argue that we have benchmarked our computer simulations using the data from the Pokhran tests, and, therefore, further weapons can be designed based on those computer simulations. We should note that we have conducted only one thermonuclear test, and that too of low yield. It is, as mentioned before, likely that this device burnt only partially. Devices that are more efficient will have to be built. In order to weaponise, we will need missile-mountable devices, which will have a different geometry. They will also have to be of higher yield. Then these will have to be made compact, and integrated with delivery as well as command and control systems. Can our nuclear deterrent be credible if we go through this long process of weaponisation without a single additional test? The bottom line is that we just cannot hand over to the army, or deter potential aggressors with, weapons based on computer simulations.
It is unscientific to embark on a long programme of weaponisation, and develop elaborate plans for maintaining a credible nuclear deterrent, all based on just one, low yield, thermonuclear test. When we do not do this for the Agni or Prithvi missiles, why would we want to take this risk for nuclear weapons? I am sure that the BARC scientists themselves, like their DRDO counterparts, would prefer to take a more conservative approach and test further to refine their designs and their capabilities. This is the scientific way. It would be wrong for the government to pressure the scientists to put a premature end to nuclear tests, for political expediency.
In principle, India accepts nuclear disarmament, and hopes its problems will be solved if all countries accept non-discriminatory, global, nuclear disarmament. Yet, this is unlikely to happen, from what we see around us as well as in the `N5’ (five nuclear weapons) countries. In spite of long and friendly discussions with the US, we haven’t come to any concrete decision relating to a new status under the CTBT. If we are to maintain our independence in today’s world, it is essential for us to have a credible nuclear deterrent, and this requires us to continue testing.
— Jeffrey Lewis · Aug 27, 09:08 AM ·
Mnhm-hmnh. As the prospect of ratifying the CTBT draws closer, we can expect to hear more of how everyone’s weapons don’t work well enough to give up testing.
That’s not to say that I don’t believe that India’s H-bomb didn’t work as planned. The evidence is there.
From the United States, we can expect to hear that there are hitherto undisclosed safety questions that can only be resolved by testing, although this ploy becomes less and less credible with repetition, along with the assurances that the RRW doesn’t need testing.
Or, come to think of it, maybe it does!
— Cheryl Rofer · Aug 27, 09:18 AM ·
Is this cool or what?
I remember what happened when I wrote that article in the fall of 1998 saying in the headline that the US had concluded that the Indian “H-Bomb failed.”
Almost overnight after the article was published I got a huge bundle of papers from BARC and DAE sent to me by diplomatic pouch from Mumbai informing me with all kinds of numbers that I was wrong.
I gave the papers to laboratory geoscientists at several European countries and the US. One main CTBTO monitoring scientist told me explicitly: “Nope. The stuff in these papers is shitty science. They haven’t shown that you are wrong.”
That having been said, please note however that, as PK Iyengar had made the case to me back a decade ago, once again this “news” is surfacing in India because their bomb makers want to keep testing. Some things in India are changing fast. Other things aren’t.
— mark hibbs · Aug 27, 09:57 AM ·
I got into a huge pissing match with the Indians on this issue as I was the principal author of Barker et. al. 1998 which had the yield estimates far below the Indian press releases. A number of Indian scientists tried to submit a comment to Science rebutting our analysis. We asked them to provide the in-country seismic data on which they based their analysis, but they refused. Luckily, in the end, their comment was rejected and never published.
On a related note, I saw the other day that wikipedia has a glowing description of the Indian 1998 tests, citing the inflated yields and saying the tests were a huge technical accomplishment. Seehttp://en.wikipedia.org/wiki/Pokhran-II
In the next day or so, I plan to submit a corrected analysis.
— charles meade · Aug 27, 10:48 AM ·
Charles, I recall one of your co-authors back then explained to me in nitty-gritty detail your frustration on this with these guys. Please do correct the record for posterity.
— mark hibbs · Aug 27, 11:14 AM ·
Their arguments at the time were quite remarkable. They said that our seismic data didn’t reflect the true yield because of a complex interference pattern caused by the simultaneous tests. Under these circumstances, they said that one could only obtain the correct yield from near field data. We said, “fine, show it to us”. They refused and that was the end of their paper.
— charles meade · Aug 27, 12:24 PM ·
The Indian argument:
For us to have a nuclear deterrent we must weaponise. For this, we must have fusion weapons, because these are smaller, lighter, and more efficient than fission weapons.
is a lot of hooey.
They claim to be building a deterrent force, not a war-fighting arsenal with a counter-force capability.
For the size and mass of their likely early-generation fusion designs, they can instead use basic fission bombs yielding in the multi-dekakiloton range – multiples of the hell weapons that incinerated Hiroshima and Nagasaki.
That should be sufficient to deter any rational adversary. And if they aren’t rational, then you have no deterrent.
A nice discussion of the actual yields of the Indian tests is from Carey Sublette back in 2001, here
— Yale Simkin · Aug 27, 12:40 PM ·