The W62 and W70 warheads were developed by Lawrence Livermore in the mid to late 1960s. The W62 warhead — housed in the Mark 12 reentry vehicle — was the MIRV warhead for Minuteman III and saw service from 1970 to 2010. The W70 was the nuclear warhead for the Lance tactical ballistic missile which saw service from 1972 to 1992 and was part of Nato nuclear weapons sharing during the Cold War.

The XW63 warhead was Lawrence Livermore’s initial proposal for the Lance warhead and competed against the Los Alamos XW64 for the role. The XW63 won the design competition, and was then cancelled in favour of the W70. The XW65 warhead was Lawrence Livermore’s proposal for the Sprint terminal anti-ballistic missile. The weapon was cancelled in favour of Los Alamos’ W66 warhead.

History

I won’t cover the background of the W62 in much detail as the partially declassified official history “MIRV: A Brief History of Minuteman and Multiple Reentry Vehicles” does a reasonably good job of it,[1] but as a short summary, the Mark 12 RV was part of a series of hardened, high-β (that is to say, high mass and low cross sectional area to reduce drag and help retain velocity during reentry) RVs proposed in 1961. In addition to the Mark 12, the Marks 13 and 14 for Titan II and Titan I/Atlas respectively were also proposed, but only the Mark 12 progressed past this point.[1, p. 15]

This led to a “system” of two RVs: the Mark 12(L) for light and Mark 12(H) for heavy[1, p. 47] and later further broadening of the system requirements lead to an even lighter RV design.[1, p. 49] The Mark 12 heavy was then spun off as the Mark 17 and was assigned the W67 warhead.[1, pp. 51–52] The W67 warhead reached Phase 3 in June 1966 and was cancelled in December 1967.[2] Further details can be found in my previous post on the topic Some Details on the W67/Mark 17.

Concurrently, the US Navy was involved in the program, looking to put six Mark 12 warheads on the B3 Polaris missile (the B3 Polaris later became C3 Poseidon).[3, pp. 117–118] However, the navy had much higher warhead hardening requirements than the air force did[1, p. 50] as they had a shorter deployment spacing, causing friction between the navy and air force.[1, p. 111] The navy ultimately dropped out of the program as the longer range requirement of the air force and the navy’s own hardening requirements lead to a heavier RV that did not fill either role very well.[1, pp. 23–24]

Though not clear through the redactions, I believe the very light Mark 12 proposal became the Mark 18, which is related to the Halberd program, and Halberd is definitely related to the Mark 3/W68 Poseidon warhead. The deployed Mark 12 was probably one of the three proposed Mark 12 light designs.

The warhead required numerous nuclear tests between 1963 and 1968, along with warhead vulnerability tests (Marshmallow, Gumdrop and Plaid are some listed)[1, pp. 15, 16, 112]. The initial W62 secondary proposed was of the “conventional design”, but later Lawrence Livermore proposed a different secondary design which would enable more forward placement of the warhead in the RV.[1, p. 112] To speculate, this probably meant a cylindrical secondary replaced by a spherical secondary.

The history of the W70 warhead is more complicated and less well documented. It began life as the XW63 warhead, competing against the XW64 from Los Alamos, with design engineering beginning in July 1964. In September 1965, the XW64 was cancelled in favour of the XW63, and then in November 1966 the XW63 was cancelled in favour of the W70. The XW63 and XW64 were both enhanced radiation devices (“neutron” devices), while the originally deployed W70 was not.[4, p. 321]

Lawrence Livermore at this point were working on the Arrow device[5, p. 24] (possibly called “Arrow II”) which was part of their proposal for the XW65 Sprint missile warhead, which was also an enhanced radiation device.[6, p. 23] It’s reasonable to believe that this secondary was also used in the XW63, or that the weapons were quite similar to each other in general. For example, a 1965 production sizing study states that the Sprint warhead is going to use a detonator system like MC2059 used in the XW63.[7, p. 11] Los Alamos’ W66 warhead, which the XW65 was cancelled in favour of, had a yield of 2 kt.[8, p. 54] The XW63 and XW65 probably had yields close to this.

It’s not clear to me why the XW63 was cancelled in favour of the W70. Hansen has nothing to say on the topic. The closest I have found is some comments from the 1969 Tactical Nuclear Weapons Symposium, in which Cecil Hudson from Lawrence Livermore talks about suppressed neutron clean devices as well as enhanced radiation devices. He mentions that the XW63 was an enhanced radiation weapon, but also that there are no military requirements for ERWs outside antimissile systems.[9, p. 342] I have to wonder if the military were not yet sold on the value of ERWs leading to the more conventional W70, but it seems ridiculous to think that the AEC would devote two whole weapons programs and many years of research to the effort of ERWs for Lance without a military requirement for such a weapon.

Regardless of the reasons, the W70 was selected in favour of the XW63. Three initial mods were produced but it’s not clear if they were produced concurrently (suggesting different yield ranges like the B61) or sequentially, but the Mod 1 apparently had more yield options, so I suspect the W70-0 was first and then soon upgraded to the Mod 1. The Mod 2 may have been some other kind of upgrade (safety?).[4, pp. 321–322]

In April 1976, the idea of an enhanced radiation Lance warhead emerged again. The program was cancelled in September 1977 before being resumed in November 1978. Production of the W70 Mod 3 began in May 1981. Concurrently, the W70 Mod 4 was considered as a dual-capable conventional/ERW design but this never entered production.[4, p. 322] To speculate, the W70-3 was mostly or entirely based on the W63. Why they produced a mod of the W70 and not a whole new weapon is probably politically based; an “upgrade” of an existing warhead is easier to slip into the defence budget than a whole new warhead.

W62 and W70 Relationship

Hansen alleges that the W62 and W70 shared at least one major component.[10, p. 76] I would go further and allege that W70 was basically a ruggedised, low fission fraction (“clean”) version of the W62. At 170 kt nominal yield, a clean version of the W62 could be expected to get about half that yield, and the W70 is commonly quoted as having a maximum yield of 100 kt, which would fit.

Another point towards the idea is the W70’s shape: it’s highly conical. It should be noted that it doesn’t need to be that conical. For example, the W70-3 is much wider at the front end that earlier W70 mods while still fitting in Lance, suggesting that it was adapted from another program.

If one looks at a diagram of the Mark 12 RV, you can see what sort of shape the warhead needs to be to fit:

While the Mark 12A/W78 RV is probably slightly different from the Mark 12/W62 RV in some ways, its features are similar to what would be expected interfacing with a warhead shaped like the W70: There is an aft ring with an inner diameter that appears to be the same as the straight-walled side supports, and a conical section leading into foam that supports the front end of the warhead (US warheads are normally only bolted aft with soft forward supports, for reasons of thermal expansion).

Secondaries

Other than the aforementioned speculated replacement of the original secondary with a spherical device, there are few details about the W62’s secondary.

Some nuclear tests in the correct era with the correct yields by Lawrence Livermore include Flintlock Corduroy (December 1965, 120 kt) Flintlock Dumont (May 1966, 200 kt) and Flintlock Kankakee (June 1966, 200 kt). In each case, the official yield is classified and has been estimated from seismic data and the like.

One possibility is that they took the secondary from an existing design, such as the W58. The W58’s total yield is known to have been 200 kt, and losing 30 kt because of various weight saving measures in the final warhead is not unreasonable. Further, if they added weight back into the W70 because of less stringent weight and size margins, the W70’s 100 kt figure more closely fits the roughly half clean weapon approximation.

This may be supported by a letter written by John Immele and Paul Brown of Lawrence Livermore to the editors of the Journal of International Studies in 1988. In it they describe how a corrosion problem in the W58 (described by Hansen as having been discovered in 1975[4, p. 458]) was assessed using data from a series of heavily instrumented tests of a similar design that was developed after the W58 was deployed.[11] The number of devices that fit the requirements of similar design and tested several times in highly instrumented test around 1975 or earlier is quite limited and the only device I can think of is the W62.

Details on the W65’s secondary, and therefore the W70-3’s secondary (assuming you find the relationship credible) are more available. A goldmine document is a record of the minutes of a meeting between the Y-12 Plant and the Kansas City Plant discussing production allocation of various W65 and W70 parts. The document discusses several components, listing their names and the security classification of their associated drawings. In several instances they describe materials used in the parts.[12]

Notable to the discussion of secondaries is the listing of a “Jar, Aft” and “Jar, Fwd”, both drawings of which are classified secret.[12, p. 1] The name is not revealing in itself, bur further on it is described that these “jars” are made from lead-antimony.[12, p. 4] From this and the fact drawings of the part are classified secret, we can likely assume that the “jar” is a codeword for some part of the secondary, possibly the secondary tamper.

Unfortunately, the W70 parts assigned to Kansas City in this meeting is quite limited and therefore they do not discuss them in great detail.

Radiation Case

Details on the W62’s radiation case design are sparse. In a nuclear sense, the only material requirement for a radiation case is that it be high-Z. In a practical engineering sense, it needs to be something that can actually be constructed, and should also be corrosion resistant (either through choice of material or through thickness) or at least can be sealed in a way that protects it. Also, for a missile warhead, the radiation case needs to be light.

The best material is probably gold.

In their 1981 budget request, the US DoE was seeking to borrow approximately 400 to 700 kg of gold each year between 1980 and 1985 from the US treasury for use in nuclear weapons, rather than purchasing the gold.[13, p. 810] Carey Sublette spoke on the matter and suggested that it was for making radiation cases in some weapons.

This does make sense as gold is very ductile allowing very thin layers to be used, and its high corrosion resistance means that these thin layers are not going to corrode away easily as would very thin layers of other metals lead or uranium. But when I say thin layers, I really do mean thin. So thin that the gold is probably not used by itself, rather it is probably plated or electro-vapour deposited onto a backing material. This backing material could be plastic or metal. Plastic for example has been suggested for the W76, but given the era, I believe aluminium or titanium was used.

Though the question of if gold was used in this era should also be asked. For example, a 1961 report on forming thorium metal describes it as ductile and being suitable for weapon case components,[14, p. 5] and a 2008 declassification decision declassified the fact that thorium is present in the radiation case of the W71[15] (though there has been some recent speculation that this may be because MagThor alloy was used as a backing material, something I agree with).

The W62 dates from eight years after the thorium forming report, so it may have used thorium, or it may have been the first of the missile warheads using gold deposited onto a lightweight backing. At the very least, it probably used aluminium as a backing material given the Mark 12A RV for the W78 is made from aluminium with the headshield bonded to it,[16, p. 1] and it’s reasonable to assume the W62 is made from materials similar to that of the Mark 12.

The W70 lacks specific details beyond some casing parts being made from aluminium, but the XW65 has more details revealed in the Kansas City/Y-12 meeting minutes document: it was aluminium with a 0.036” (0.9144 mm) plated lead layer. The document further asks why plating is being used in the design instead of a more conventional bonded lead liner, suggesting the bonding of lead to a shell is the standard method at the time[12, p. 4] This also proves that very thin radiation cases are feasible for weapons.

Primary

There is a surprising amount of details on the primary stages of these weapons.

The W62 used the MC1991 detonator assembly.[7, p. 9][17, p. 1] It is specifically described as a detonator assembly and not as a detonator.[18, p. 3] MC1991 is also used in the W58 warhead.[7, p. 9] I believe that phrase “detonator assembly” is used for something more than a simple detonator and is a major component of the “driver detonator”, which I believe is the system that initiates the main HE charge (be it air lenses or an MPI system).

One document describes the build schedule of several weapons systems. In it from July 1966 to June 1968, 8250 MC1991 for the W62 were to be produced, along with 4700 W62 driver detonators.[19, p. 30] These figures probably did not eventuate due to delays in the W62 program, and as a caution, the numbers themselves can not be used to calculate the number of detonators in each warhead as a large, potentially uneven number would be expended in test firings and another large number would go into surveillance, which is to say storage so that they can be test fired later to check for any aging issues.

However, I don’t believe that the use of MC1991 in both weapons means a straight copy of Kinglet is used in the W62. For example, the detonator system in the W58 is MC1991-1, while in the W62 it is MC1991-2.[7, p. 9] Further, in March 1965, Lawrence Livermore changed something about the primary that shaved 4.6-6.8 kg (10-15 lb) off the total warhead weight.[1, p. 111] As we know that devices can have subvariants (such as Cougar A and Cougar B), it’s probable that the W62 uses another version of Kinglet, which might be “Kinglet B” or even “Kinglet C”.

We are however able to make some estimates for the HE weight in the W62 and W70’s primary, thanks to a report on detonation velocity variations in LX-04, in which they provide a few small details on the matter. The main two are that the W70’s HE weighed about twice that of the W62’s,[20, p. 14] and three diagrams of the pressed hemispheres.[20, p. 12]

These diagrams show few dimensions, but the fact that W62 and W70 diagrams have clearly different inner radii suggests that these are not dummy diagrams with some lines and numbers slapped on, but rather are actual diagrams from some manual or specification, and using the numbers provided and some simple algebra, we can calculate some dimensions.

Starting with the W70 (figure C4), I measured the X+2.7cm line as being 1.143 times longer than the X line, which produces the equation:

X+2.7=X*1.143

Which allows us to calculate X as being:

X=18.88 cm or 188.8 mm

From this I measured the pit radius as 109 mm or a diameter of 218 mm.

Moving onto the W62 hydrostatic (figure C2), we have three lines to examine instead of one. Rather than repeat the basic algebra, I will skip to the result as the procedure is the same: for X+1.0 cm I calculated X=119 mm, and for X+2.9 cm I calculated X=121 mm, which is a very good agreement for measuring off a diagram. From this I measured the pit radius as 87 mm or a diameter of 174 mm.

Unfortunately, I did not get any clear answer from diagram C3. I suspect that the curve is just a rough approximation of whatever the specifications say as drawing it by hand for the diagram would have been hard. I’m not even sure what the curve is: some sort of half ovoid, quadratic or perhaps an offset half-circle?

Regardless, out next question is as to what diameter they would turn the hemispheres down to. Hansen describes Kinglet’s outer diameter as 11.2 to 11.57” (284 to 293 mm),[21, p. 398] however, the LX-04 paper focuses on the importance of PBX density on detonation velocity, stating that the calculated 21 ns breakout spread is small enough to take advantage of “the most sophisticated lighting systems” (lighting system appears in several sources and is probably another name for a implosion system),[20, p. 1] and therefore the density measurement locations may represent the outside radius of the parts.

Following this line of reasoning, the outer radius for the W62 HE would be 149 mm or a diameter of 298 mm assuming the X+2.9 cm line is the radius. Using the given density and the calculated pit radius, the total HE mass is 21.16 kg. Interestingly, Hansen gives the total HE weight of Kinglet (in the W55) as 22.2 lb (10.06 kg).[10, p. 540] I have to wonder if the figure was actually 22.2 kg, particularly as he gives the total weight of Kinglet as 58-63 lb (26.3-28.57 kg). Alternatively, the 22.2 lb figure might be correct for a single hemisphere, for a total mass of 20.12 kg.

Running the same reasoning on the W70 hemispheres however does not work, as the X+2.7 cm line gives an outer radius of 215 mm or a diameter of 430 mm, and an HE mass of 67.3 kg, which is three times that of the W62. But what if the X radius is the final radius? If we use a radius of 188.8 mm or a diameter of 376 mm, the HE mass comes out to 41.6 kg, which is almost spot on for the W70 being twice that of the W62.

1981 Lawrence Livermore explosives handbook lists (among other things) skid tests for various explosives. Besides standard tests, they also describe non-standard tests. Most of these non-standard tests are due to the use of non-standard HE masses, and for LX-04, one of those tests is for a 22.7 kg mass.[22, Secs 9–32]

Unusually, there is a single, partial image of the W70’s primary out there, in part one of Sandia’s documentary Always, Never: The Quest for Safety, Control, and Survivability. In it, a sectioned W70 warhead is visible in the background. Though only the back side of the warhead is visible, two small holes are cut into the back side, allowing a small fraction of a white sphere to be visible.

Though it’s hard to make any definitive conclusions from this image, given the W70’s diameter of 18” (460 mm), the device does appear to be approximately the correct size for an HE sphere of 376 mm. At the very least, it does not seem to be a very small primary like that used in some weapons of the time.

You may ask why they would go for a larger primary, but this would have advantages in reduced fissile material requirements as more HE means more energy to compress the pit. As Lance was designed to also carry an HE warhead, the slightly heavier weight caused by this likely was insignificant compared to the cost savings from less plutonium.

Other Details

The XW63 was planned to have three tritium gas boosting bottles, the largest of which was like that used on then present weapons, while the smallest of which had a maximum decay heat of approximately 100 mW.[24] From this and knowing that tritium has a decay heat of 0.3240 W/g,[25] we can calculate the tritium content of the smallest bottle to be about 0.309 g.

To speculate, the three bottles are for yield control. Because D-T gas fuses more easily than Li6D (which requires a neutron to fission the Li6 first for easy fusion), the XW63 probably controlled yield through the addition of D-T gas to the secondary as a fusion spark plug. The XW65 having a “jar” (i.e. something hollow) supports this notion.

It seems unlikely that it got all of its secondary yield from D-T gas as D-T fusion only produces 80.4 kt/kg. If we assume a fusion burn of 50%, that means 25 g of gas per kiloton yield, of which 66% of it is tritium or 16.5 g. An “average” boost amount in 60s era weapons seems to be about 5 grams of tritium, so even if the XW63 only produced 1 kt from fusion, the three bottles combined (5 g + <5 g + 0.3 g) are too small to supply the necessary D-T gas, therefore some fusion yield must come from Li6D.

Speculatively, it may have been possible to choose where the boost gas goes. For example, the device might have gotten 0.5 kt from fission and 1 kt from fusion when all of the gas sent was to the secondary, but it may have been possible to direct all that gas to the primary and get ~5 to 10 kt fission yield if desired.

Summary

The W62 and W70 warheads were strongly related, using a similar secondary design and possibly a similar primary design. The XW63 and XW65 warheads were strongly related to each other, and probably used the same primary as the W62. The enhanced radiation W70-3 was likely similar in design to the XW63.

The W62 secondary was probably shared with that of the W58, having similar yields and with the W58 having been evaluated following corrosion concerns using test data from a weapons test that was probably of the W62. The W70 probably used a low fission fraction version of this secondary. This secondary was spherical in shape.

The XW65 used a lead tamper in its secondary. The radiation case was made by plating lead onto aluminium. This lead case was only 0.91 mm thick and proves that previously speculated very thin radiation cases are possible in thermonuclear weapons. Tampers in some weapons may be called “jars”.

References

[1]          Daniel Ruchonnet, ‘MIRV: A Brief History of Minuteman and Multiple Reentry Vehicles’, Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States), COVD-1571, Feb. 1976. [Online]. Available: https://commons.wikimedia.org/wiki/File:1976-MIRV-A-Brief-History-Minuteman.pdf

[2]          T. M. Josserand, ‘R&A for UUR_Weapon_History_Phases_20170206.’, Sandia National Lab. (SNL-NM), Albuquerque, NM (United States), SAND2017-2818C, Mar. 2017. Accessed: Jul. 25, 2021. [Online]. Available: https://www.osti.gov/biblio/1429158-uur_weapon_history_phases_20170206

[3]          ‘Proceedings of the Special Projects Office, Task II – Monitor and Sponsor the Fleet Ballistic Missile Development Program, 43rd Meeting, 27, 28 May 1964’, May 1964. Accessed: Jun. 15, 2022. [Online]. Available: https://nsarchive.files.wordpress.com/2015/02/proceedings-of-the-special-projects-office-steering-task-group-task-ii-monitor-the-fleet-ballistic-missile-development-program-43rd-meeting-27-28-may-1964.pdf

[4]          Chuck Hansen, Swords of Armageddon, vol. VI, 7 vols. 2007.

[5]          ‘LETTER TO J M CASE, SUBJECT: OFFICIAL NAME CHANGE – BALLISTIC MISSILE DEFENSE SYSTEM ( ENCL: MISC LETTERS AND MEMOS RE CLASSIFICATION INDEX, CODE WORD, ETC. DTD 1959 – 1969 )’, NV0726671, 1960s. Accessed: Jun. 11, 2021. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=16359299

[6]          Robert McNamara, ‘Production and Deployment of the Nike-X’, Dec. 1966. Accessed: Jun. 24, 2022. [Online]. Available: https://nsarchive2.gwu.edu/nukevault/ebb281/4A.pdf

[7]          ‘Weapons Production System Sizing Study, June 30, 1965.’, Mound Laboratory, MLM-CF-66-6–290, Jul. 1965. Accessed: Jun. 26, 2021. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=16137953

[8]          ‘Minutes of National Security Council Meeting’, Washington DC, May 1969. [Online]. Available: https://history.state.gov/historicaldocuments/frus1969-76v34/d16

[9]          ‘Proceedings of the Tactical Nuclear Weapons Symposium’, Los Alamos National Lab. (LANL), Los Alamos, NM (United States), LA-4350-MS, Sep. 1969. Accessed: May 22, 2021. [Online]. Available: https://www.osti.gov/opennet/servlets/purl/1018707.pdf

[10]        Chuck Hansen, Swords of Armageddon, vol. V, 7 vols. 2007.

[11]        J. D. Immele, P. S. Brown, and S. Fetter, ‘An Exchange on Stockpile Confidence’, International Security, vol. 13, no. 1, p. 196, 1988, doi: 10.2307/2538901.

[12]        ‘MEMO TO D C GRAY, ET AL, SUBJECT: MINUTES OF MEETING HELD JULY 11, 1967 ON PARTS FOR THE XW-65 AND XW-70 Y-12 CONSIDERS TO BE APPROPRIATE FOR TRANSFER TO BKC ( AS24854 )’, NV0970202, Jul. 1967. Accessed: Aug. 07, 2022. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=16131995

[13]        U. S. C. S. C. on A. S. on E. and W. Development, Energy and Water Development Appropriations for Fiscal Year 1981: Hearings Before a Subcommittee of the Committee on Appropriations, United States Senate, Ninety-sixth Congress, Second Session. U.S. Government Printing Office, 1980.

[14]        A. E. Calabra, ‘Rolling and Forming of Thorium Metal’, Rocky Flats Environmental Technology Site (RFP), Golden, CO (United States), RFP-232, May 1961. Accessed: Aug. 10, 2022. [Online]. Available: https://www.osti.gov/biblio/1165441

[15]        Andrew P. Weston-Dawkes, ‘CLASSIFICATION BULLETIN – WNP-118 Declassification of a Material in a Specified Weapon’, Mar. 2008. Accessed: Aug. 10, 2022. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=1052069

[16]        W. Garcia, J. Hertz, J. Prunty, and H. McCutchen, ‘Composite Material Application to the MK12A RV Midbay Substructure.’, GENERAL DYNAMICS SAN DIEGO CA CONVAIR DIV, Sep. 1979. Accessed: Jul. 04, 2022. [Online]. Available: https://apps.dtic.mil/sti/citations/ADA076485

[17]        Schaefer, R D, ‘LRL-MD Coordination Meeting Minutes, October 6, 1966’, Mound Laboratory, MLM-CF-66-10–398; 66-10–398, Oct. 1966. Accessed: Jul. 02, 2021. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=935982

[18]        Foster, K W, ‘SANL 409, Blanket Work Order for FY 1967 University of California, Lawrence Radiation Laboratory to Mound Laboratory.’, Mound Laboratory, 66-7–209; MLM-CF-66-7–209, Jul. 1966. Accessed: Jun. 27, 2021. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=16137989

[19]        D L Scott, ‘PLANT ACQUISITION AND CONSTRUCTION DEVELOPMENT AND STANDARDS BUILDING ADDITION (DS)’, Mound Laboratory, 65-5-14(T); MLM-CF-65-5-14(T); 65-5–14; MLM-CF-65-5–14, Apr. 1965. Accessed: Jun. 26, 2021. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=16138185

[20]        K. J. Scribner and E. G. Zaslawsky, ‘Variations in the detonation velocity of the high explosive LX-04-1’, California Univ., Livermore (USA). Lawrence Livermore Lab., UCRL-52460, Apr. 1978. Accessed: Jun. 30, 2021. [Online]. Available: https://www.osti.gov/biblio/7028925-variations-detonation-velocity-high-explosive-lx

[21]        Chuck Hansen, Swords of Armageddon, vol. I, 7 vols. 2007.

[22]        B. M. Dobratz, ‘LLNL explosives handbook: properties of chemical explosives and explosives and explosive simulants’, Lawrence Livermore National Lab., CA (USA), UCRL-52997, Mar. 1981. Accessed: Jul. 18, 2022. [Online]. Available: https://www.osti.gov/biblio/6530310-llnl-explosives-handbook-properties-chemical-explosives-explosives-explosive-simulants

[23]        Always/Never: The Quest for Safety, Control, and Survivability, (Jun. 16, 2015). Accessed: Oct. 29, 2021. [Online Video]. Available: https://www.youtube.com/watch?v=0a1exo_vU_k

[24]        Bertram C Blanke, ‘To G R Grove: Discussions on Feasibility of Calorimetering Components of the XW63’, Mound Laboratory, 66-3–284; MLM-CF-66-3–284, Mar. 1966. Accessed: Jun. 25, 2021. [Online]. Available: https://www.osti.gov/opennet/detail?osti-id=16138209

[25]        W. L. Pillinger, J. J. Hentges, and J. A. Blair, ‘Tritium Decay Energy’, Phys. Rev., vol. 121, no. 1, pp. 232–233, Jan. 1961, doi: 10.1103/PhysRev.121.232.