Fogbank: That time the US forgot how to make Nukes

The U.S nuclear weapons program is extremely secretive, yet oddly we can still figure some things out about it from the information that is out there. The U.S department of Energy, the government department concerned with the safety and handling of nuclear material, is in charge of one of the largest nuclear stockpiles in the world, and those weapons need to be constantly upgraded and refurbished with newer technologies. Occasionally those technologies get hinted at by politicians when asking for more money, or defending delays, and sometimes its clear that processes to create technologies have been forgotten with engineers retiring. The secret material, publicly refereed to as Fogbank is a great example of this.

Aerial photo of the Y-12 facility
Aerial photograph of the Y-12 facility at Oak Ridge taken around 2007 (Credit: NNSA)

Fogbank first became well known around 2007, when it came to light that it was the root problem behind technical delays in the life extension project for the W76 warhead. This particular warhead is used on the Trident II, submarine launched ballistic missiles. They are known as D5s, and are used by both the U.S Navy, and and the Royal Navy of the UK. It took until 2018 before the W76-1 warheads were finally delivered.

A Trident II (D-5) launch from a submerged Royal Navy submarine
A Trident II (D-5) launch from a submerged Royal Navy submarine. Image from Lockheed Martin, but is a U.S. Department of Defence photo.

There is a material that we currently use and it’s in a facility that we built … at Y-12 … It’s a very complicated material that – call it the Fogbank. That’s not classified, but it’s a material that’s very important to, you know, our [W76] life extension activity.

The then NNSA director Thomas D’Agostino, talking to members of the House of Representatives in 2007. He is referring to the Y-12 National Security Complex, located near Oak Ridge.
Y-12 Security Complex
The Entrance building of the Y-12 security complex, a Department of Energy facility located near Oak Ridge (Credit: NNSA).

Late in 2007, D’Agostino described Fogbank as an “interstage material”, which heavily implies that it sits somewhere between the primary and secondary stages of the the two stage thermonuclear weapon. The W76 is a two stage warhead, so would need some sort of material to trigger the second stage. In a thermonuclear bomb, also known as a H-bomb, the first stage creates the easily produced fission reaction (what happens in a nuclear reactor), which creates a superheated plasma at the interstage (the expected role of Fogbank), which then triggers the fusion reaction in the second stage. Fusion is the process that happens in the Sun, and may be the way we power future nuclear reactors.

There’s another material in the [W76] – it’s called interstage material, also known as Fogbank, but the chemical details, of course, are classified,

Thomas D’Agostino speaking to senators late in 2007
NNSA director Thomas D’Agostino in 2009 (Credit: NNSA)

This has obviously lead to many engineers and scientists to comment on possible materials that Fogbank could be. The most common reccuring theme is that it is likely to be an Aerogel, a type of material known as an ultra light gel. Aerogels are extremely light while being surprisingly strong. Invented in the 1930’s, the concept of aerogels have been around for a long time, but it was NASA’s Glen Research Centre that introduced modern methods of manufacture, and even used them on a few space missions such as Stardust. There have been suggestions, such as the one from Jeffery Lewis, an expert on missiles and nuclear weapons, that the code name Fogbank could be a reference to the other nicknames for Aerogels. Names such as “frozen smoke” and “San Francisco fog” have all been used in reference to the light and almost see through solid.

The Stardust dust collector with aerogel blocks. (NASA)
The dust collector with aerogel blocks on NASA’s Stardust spacecraft. (Credit: NASA)

There are a few other things that imply that Fogbank could be an aerogel. In 2007, D’Agostino told legislators that Fogbank’s production process required the material to be purified using “a cleaning agent that is extremely flammable.”. Then in that same year he talked at the Widrow Wilson centre about “another material that requires a special solvent to be cleaned”, potentially the same material, and that the solvent used was identified as “ACN”, the common abbreviation of acetonitrile, a solvent commonly used in aerogel production. He describes the solvent by saying “That solvent is very volatile, it’s very dangerous. It’s explosive.” which describes acetonitrile well.

A 2.5 kg brick is supported by a piece of aerogel with a mass of 2 g.
A 2.5 kg brick is supported by a piece of aerogel with a mass of 2 g. (Credit: NASA/JPL)

More evidence comes from a 2007 briefing slide on a program known as the Reliable Replacement Warhead (RRW), which aimed to develop a new design of warhead to replace a selection of the existing versions. The slide points at an interstage material that is an expensive, “specialty” material that if replaced would eliminate the need for unique facilities. The RRW effort was de-funded in 2008 and cancelled by president Obama the following year.

NNSA slide from a 2007 briefing
The NNSA slide from a 2007 briefing about the RRW program that details the properties of the current interstage material. (Credit: NNSA)

Up until 1989 the Y-12 facility in Oak Ridge had a specialised site known as Facility 9404-11, which was apparently used to create Fogbank. 1989 was the year that the final W76 warhead was finished, so the facility was closed down and a new “purification facility” took its place. The original building was eventually torn down in 2004, and replaced with a new facility, renamed to Facility 9225-3. According to Denis Ruddy, who served as the president of the division of the Babcock and Wilcox Company (BWXT Y-12) that ran the facility between 2000 and 2014, the purification facility reprocessed a material that they were taking out of weapons so that it can be used in refurbished weapons. The material is apparently classified, and the use in the weapon is classified, and so is the process that they follow, unsurprisingly.

It is also public knowledge that this site uses ACN as a part of at least one of the processes going on in it. On three seperate occasions in March 2006, workers had to evacuate the facility after alarms were triggered. According to the Department of Energy, the facility is alarmed to monitor for ACN levels, but it is not confirmed whether the alarms that went off were from ACN. There was also an ACN spill that forced an evacuation of the facility in December 2014, and although there were no injuries, it took months for the facility to get back up and running.

The Purification Facility, also known as Facility 9225-3, at Y-12.
The Purification Facility, also known as Facility 9225-3, at Y-12.

In 2009, in an issue of the Nuclear Weapons Journal, an official publication of the Los Alamos National Laboratory, an article was published that disclosed that there was a decision to restart the manufacture of Fogbank in 2000, and confirmed that it was linked to the W76-1 warhead project. It also revealed that in the years between the projects that the NNSA had lost almost all of its institutional knowledge base regarding the manufacture and development of Fogbank. The article said “Most personnel involved with the original production process were no longer available.” So the newer personnel had to reconstruct the production process from the records. It also mentioned that “a new facility had to be constructed, one that met modern health and safety requirements.” The facility would make sense to be the 9225-3 facility as it stands on the site of the old one. The best part of it all, the new facility, with the new staff using old records to manufacture, created a higher purity final product than the initial W76 warheads used. The problem with this “improved” material was that it was actually too pure,the impurity was actually essential for the final product to work as intended. The process was concluded in 2008, almost a decade after first deciding to restart production, and the W76-1 warheads began that year.

The improved production process
The extremely vauge diagrams shown in the article describing the production process the NNSA initially developed to produce Fogbank in the 2000’s and the improved versions. (Credit: NNSA)

This is one example of how organisations such as the NNSA lose technical institutional knowledge over relatively short periods of time when the technology isn’t being used. They had similar problems with designing high explosive capabilities recently. In March 2020, The director of the Natural Resources and Environment team at the Government Accountability Office, Allison B. Bawden, highlighted that the NNSA had not produced a particular type of high explosives at scale since 1993, and highlighted the issues with Fogbank production as an example. This highlights one issue of governmental technology not being taken on by commercial business to continue. Either way, Fogbank is a prime example of the complexity, and secrecy of nuclear warhead production, and how in that secrecy, technology is lost.

The Difference Between Bolts vs. Screws

It is almost an age old question, and by many out there these two words are almost interchangeable. Ask one engineer and they will give you an answer, ask another you will likely get a slightly different answer. Over the last few hundred years engineers have developed fasteners, sometimes for very specific applications, so the line can sometimes become blurred. Some fasteners can only be defined when it has been put into an assembly, being dependent on the design. There are a few standards out there that make an effort to define this, with varying degrees of vagueness, but I am going to try and make sense of it in this post and give you a few examples to make it more obvious. So lets put a few definitions into the mix. The definition I think is the best I can find is from the Specification for Identification of Bolts and Screws, ANSI-ASME B18.2.1 1981. This document has been superseded a few times but the changes have been to add many more extra definitions rather than change this one. Their definition is:


A bolt is an externally threaded fastener designed for insertion through the holes in assembled parts, and is normally intended to be tightened or released by torquing a nut.


A screw is an externally threaded fastener capable of being inserted into holes in assembled parts, of mating with a preformed internal thread or forming its own thread, and of being tightened or released by torquing the head.

I like these definitions as they put it in super simple terms that you can get your head around. Think of how you can tighten the fastener, if you have to use the head then it much be a screw, if you have a nut on the other end it mush be a bolt. From there it gets a bit more complicated, as you can also often tighten a bolt by the head as well, but the point is that you can use either, whereas a screw can only be tightened or loosened by turning the head. The other key point is that a bolt should not be tapping its own thread in the part it fastens to. A screw doesn’t have to form its own thread in the material, but if it does it can only be a screw. Basically if it is pointy it is likely to be a screw, if it is flat ended it is more likely a bolt (but not always true as we will see. There is also one other way to loosely define a bolt and that is the way it drives. A screw drives from the center (like an Alan head or flat head screw) but a bolt tends to need to be fastened with a wrench, so away from the center. Some companies like Accu group use this as a definition but it is not defined in most standards I have read, but still a good rule of thumb. Now lets look at a few examples to get a better idea:

Bolts That Cannot be Unfastened by the Head

A definite subset of bolts, the round head, oval head and plow bolt have not way to be undone via the head. The round head and oval head both protrude above the surface, but are completely smooth and rounded on the edges, so there is no surface for a wrench to lever against, so they have to be fasted by a nut on the other end. These bolts usually have a non circular area near the head to stop it from turning when the nut is being attached.

Externally threaded fasteners with a head that cannot be used to fasten it in place is a bolt. Credit (1)

Screws That Cannot Use a Nut

The classic screw is something that we are all familiar with, with it tapering to a point, often with a straight thread with multiple pitch length, and cannot use a nut. The tapering prohibits the use of adding a nut, this describes a classic wood screw. Other screws such as tapping and grub screws with points or shanks are also definitely screws by the fact they often make their own thread and have a point in the end to make some sort of non screwed connection with another part.

An externally threaded fastener that which has a thread that cannot be used with a nut is a screw. Credit: (1)

Bolts That Need a Nut to Function

Some bolts such as a hex structural bolt that have a shaft the same diameter as the thread (no shoulder) and therefore go through a part and needs to be attached into a nut on the other side to be fastened. The bolt has a smooth shaft near the head which cannot be fastened on its own, therefore needing a nut. By the fact it needs a nut it has to be a bolt. Most classic bolts often need a nut to work in an assembly, and it is the best way to recognise a bolt over a screw.

A hex Structural bolt is a great example of a bolt that needs a nut to function as the lack of thread near the head cannot be used to fasten. Credit: (1)

Screws That Look Like Bolts

This is where things can get a bit iffy, fasteners that on the face of it look like bolts but act a bit more like screws. Set screws for instance are a screw as they never use a nut, and they are usually used to secure an object within or against another object. Things like attaching a gear or pulley to a shaft is a common example. The other is a shoulder screw which looks much like a normal bolt but is different by the fact that the non-threaded shaft is bigger than the threads, hence the shoulder. The threaded part does not tend to be screwed into a nut, leading to the definition of it being a screw rather than a bolt. They tend to be used as a shaft for rotating things like pulleys or gears.


  1. Distinguishing Bolts from Screws – U.S. Customs and Border Protection – July 2012
  2. I you can get access – Specification for Identification of Bolts and Screws, ANSI – ASME B18.2.1 1981
  3. If you can get access – Square, Hex, Heavy Hex, and Askew Head Bolts and Hex, Heavy Hex, Hex Flange, Lobed Head, and Lag Screws (Inch Series), ANSI – ASME B18.2.1 2012
  4. For interesting reading about a court case about this: Rocknel Fastener, Inc. v. United States, 24 C.I.T. 900, 118 F.Supp. 2d 1238 (Ct. Int’l. Trade 2000)

The Fallow Deer

So today I was out with a group of friends in a local estate near Plympton. The owner was there and while we were talking, he told us to just stop and listen for a second. So listen we did. Behind the bushes and the trees, we heard grunting. This grunting was the sound of Fallow deer. Apparently there are around 100 Fallow deer in the area, and for this 2 week period, its mating season. The sound we heard was a rare one, it was the sound of the buck’s grunting and thrashing around in the ferns. Why do they do this? Apparently, they try and make as much noise as possible, so they sound big and scary, so mates will be attracted to them. The louder they are, the more likely that they will find a mate. I found this quite interesting, I thought you guys might too. I wouldn’t try and go up against a Fallow deer at this time of year, you are likely to lose. Lots and lots of testosterone flowing.

As an interesting side note about Fallow deer, they are the only British deer to have palmate antlers, so they have that bit in between the horns, like frogs feet.