NTB: Some people confuse cryogenics with cryobiology or cryonics, but they’re very different fields, aren’t they?
Shirron: [Laughs] Uh, yeah. Cryobiology is legitimately a subset of cryogenics, and it’s an interesting one because it exploits what we normally consider to be a hazard, and that’s the destruction of tissue when it’s cryogenically frozen. It’s done to achieve useful results like eliminating tissue that’s harmful to a person. It’s an exciting and, as I said, legitimate field. Cryonics is, perhaps, exciting, but it’s not legitimate.
NTB: I believe one of the directors of your group put a blurb on the Web site that says, “We do not freeze people”?
Shirron: [Laughs] Well, it’s true! We get enquiries every now and then from people who confuse the two. They’re interested in preserving their own bodies, or just the general techniques one employs to freeze bodies or parts of bodies. And again, there’s no utility in doing that, so it has to be considered a fringe “science”.
NTB: So what type of projects does the Cryogenics and Fluids Group typically get involved with?
Shirron: We have a broad range of interests and projects that we work on. I’d say our main lines of business are to develop cooling systems for space instruments, and these would fall broadly into two categories. There are stored cryogen systems, where we send up a tank full of liquid helium - which is the coldest liquid - or a solid cryogen, for example, hydrogen, neon, argon, even dry ice, at about 200 Kelvin, but that pushes the limits of what we consider to be cryogenics. Those solids and liquids have been used to cool various instruments over the years.
The other broad category, which is becoming prominent now, is cryo-coolers. These are mechanical refrigerators, not too unlike house refrigerators, but these are systems that can cool down to as low as about 4 Kelvin, and there are even some kinds that can go down to about 2 Kelvin
So, we not only develop those two types of systems – stored cryogens and cryo-coolers – but also work out all of the issues associated with their use. Integrating instruments into them, in the case of the stored cryogens, there are a lot of challenges associated with just keeping the cryogen in the tank in space. For liquid helium to work, you’ve got liquid in a tank and you need to confine the liquid, but you need to let boil-off gas go as heat gets into that liquid and boils it. So we work on things like the phase separators that retain the liquid inside the tank, but let the gas go.
NTB: One of the groundbreaking technologies you helped develop at NASA is a continuous duty multi-stage adiabatic demagnetization refrigerator (ADR). What is an adiabatic demagnetization refrigerator and how does it work?
Shirron: That’s a very good question. It’s a solid state refrigerator. It’s one of the few refrigeration techniques that’s based on solid refrigerants rather than gas or liquid refrigerants. The way it works is, the refrigerant is generally a crystalline structure that’s got a magnetic atom in each molecule. That magnetic atom has the ability to store and release heat. It does that through changes in the angular momentum of the outer most electrons in the shell of that atom. We can manipulate that magnetic state through an external magnetic field. When we apply a strong magnetic field to the refrigerant, we cause the magnetic spins to align with the external magnetic field.
To explain how this produces cooling, I’ll make use of a popular notion of entropy. People think, legitimately, of entropy as a measure of “disorder,” and in this magnetic spin system if you apply a magnetic field, all of the magnetic spins align with that field, so there is little disorder. It’s a very ordered state, so the entropy is low there. You can actually suppress it quite close to zero. But as you reduce the magnetic field, the magnetic spins no longer have to fully align with the field and there can be some misalignment and, therefore, some disorder. That means higher entropy. An important connection is that entropy is related to heat flow. A change in entropy is a change in heat divided by the temperature you happen to be at.
So, in our case, by changing the external magnetic field on a magnetic material, we can cause the material to give off heat or be able to absorb heat. And if the material is isolated, the heat is transferred to or from the lattice structure - that is, the non-magnetic part of the refrigerant – and the refrigerant warms up or cools down. Those four processes – warming and cooling and giving off and absorbing heat – are exactly the four processes that go on in any refrigerator, but they’re done very efficiently in an ADR and they are very pronounced effects at low temperature. So, it’s a very powerful technique for producing temperatures that are very close to absolute zero. But the thing you have to appreciate is, our starting point in temperature is usually no more than about 4 Kelvin. So we start cold, but we go much colder.
NTB: What are these types of refrigerators typically used for?
Shirron: They’re mostly used when you want to achieve temperatures below about 2 Kelvin. And they’re mostly used for space instruments. Let me justify that. There are other refrigeration techniques that are, in some ways, more capable than ADRs. One of the most common ones is a helium 3 refrigerator. It’s a liquid-based system that uses the evaporation of liquid helium 3 to produce cooling. It is more compact and can absorb larger heat loads than an ADR, but it can’t cool below about 0.3 Kelvin. For lower temperature, there are limited options: dilution refrigerators and ADRs.
Dilution refrigerators can cool to milliKelvin temperatures, and they have relatively large cooling power. The problem is they are not suitable for space. They’re a fluid-based system that relies on gravity to phase separate the refrigerant – a mixture of helium 3 and helium 4. Despite some effort, this technique has not been adapted for space. So, in a case where you need low temperature and you need to go into space, ADRs are really the only option.
At present, the only instruments that need such low temperature are telescopes that use ultra-sensitive detector arrays to detect x-ray and infrared photons. So they’re mostly used to support the astronomy community. We’re hoping there are some Earth observing uses, because that’s a higher priority for NASA right now, but mostly we’re limited to astronomy.
NTB: What advantages do adiabatic demagnetization refrigerators have over liquid cryogens?
Shirron: Well, they can produce lower temperatures than you can get with any liquid cryogen. The lowest temperature that we can produce in space with a liquid cryogen is about 1 Kelvin. I suppose you could say that a helium-3 refrigerator uses a liquid cryogen, and as I’ve mentioned that kind of system can produce temperatures as low as 0.3 K. But ADRs still have the advantages of lower temperature, very high efficiency, and the easy ability to operate in space.
NTB: What would you say was the biggest technical challenge you faced in designing an ADR that could cool down to 10 mK?
Shirron: There’s no doubt here. It was heat switches. Heat switches are used to regulate the inflow and outflow of heat between various components. The ADR we were designing was a multistage system that had four stages that would operate over the temperature range from as low as 10 mK up to about 5 Kelvin. The stages are connected in series, and heat is cascaded from the low temperature end up to the warm end. The heat switches are devices that can be very thermally conductive and very thermally insulating, and these are used to control the flow of heat through the refrigerator.
Because we’re covering the range from 5K to 10 mK, we needed heat switches that could operate at any temperature across that range. Existing heat switch techniques worked pretty well above 1 Kelvin, and pretty well below 0.2 Kelvin, but there were none that really worked well in that middle region, and that region was critical for us. Our solution, which was the development of a passively operated gas-gap heat switch, not only solved the problem of filling in that gap, but it resulted in switches that had better properties and better capabilities across the entire temperature range. A need for us turned into something that actually provided more capability than we expected.
NTB: You were recently part of the team that worked on NASA’s highly-touted Soft X-ray Spectrometer (SXS). Tell us about that project, what it’s designed to accomplish, and what role you played in it.
Shirron: The SXS – Soft X-ray Spectrometer – will be the most sensitive x-ray sensing instrument that’s been put into space. It’ll have the ability to measure the energies of individual x-rays striking the detector array. I like to think of this as literally taking color pictures in the x-ray spectrum. You’re used to seeing dental x-rays as just a gray, dull image, so imagine the dentist being able to show you a color picture of the x-ray spectrum. There can obviously be a lot more information in that image, so it will greatly expand scientific capability. The SXS will use a 64-pixel array of these very sensitive detectors. Each one of them is a small chip of silicon with a thermometer attached to it. To determine the energy of an x-ray, you absorb the x-ray in that chip of silicon, and you measure the temperature rise using the thermometer, and you directly relate the temperature rise to the energy of the x-ray. This kind of detection technique can resolve energies to about one-part-in-a-thousand, which I’m guessing is about as good as color pictures are at resolving visible photons.
One example of how these detectors will expand our ability to probe the universe concerns binary star systems. Some of these are oriented in such a way that at any given moment one of the stars is coming toward us and the other star is moving away from us. The energy of x-rays coming from these stars will be Doppler shifted, so x-rays from the star that’s coming toward us are going to be slightly higher energy than the x-rays coming off of the star moving away from us. The SXS detectors will be able to see the difference and be able to measure the velocity difference between those two stars. There are a lot of astronomy questions that such measurement sensitivity will be able to address, such as the formation and evolution of galaxies; the behavior of matter in strong gravity environments near black holes; and even help us answer questions like how much mass there is in the universe, so that we can understand this infamous dark matter that nobody knows how much of it is out there and how it interacts with visible matter. These detectors will be able to help us answer some of those questions.
The key here is that this measurement technique, the ability to precisely measure the energy of a single x-ray, is only possible if that chip of silicon and its thermometer are at a very low temperature. This is because at very low temperature materials have very low heat capacities. Heat capacity is the amount of heat needed to produce a change in temperature. The small heat capacity of the detectors allows the energy of individual x-rays to produce a substantial change in temperature. Again, that’s possible only if we’re operating in the 50 mK temperature range.
My role is to develop the ADR that will be used on this mission - the Astro-H Mission - to cool the detector array. Because of one of the peculiarities of the cryogenic system that our ADR will attach to, we have to have greater capabilities than, say, the ADR that was flown on the Astro E-2 mission with the XRS instrument. So, for this we’re going to use a two-stage ADR, and this will be the first-ever multi-stage ADR flown in space. It won’t be a continuous one, like we’ve been developing in our lab, but its design – and particularly the design of a lot of its components – is similar to, or identical to, the ones we use for the continuous ADR. This will provide us with an opportunity to test a lot of the components and heat transfer processes that are critical to the continuous ADR’s operation.
NTB: You’ve been working with Goddard’s Innovative Partnerships Program to help spin-off and commercialize the research you’ve been doing. Tell us about that program and some of the potential commercial applications you envision for ADR cooling technology.
Shirron: Well, I’m no expert on all that the IPP office does. What I do know is they work with researchers like me at Goddard to identify commercial uses and users who can benefit from NASA-developed technology. They also act as a liaison between technology developers and the patent office to orchestrate the filing for patents and the issuing of patents. Also, they act as a liaison with commercial companies who want to license the technology. They perform a very valuable function for us in all that negotiation. That takes a load off of us because, frankly, I don’t want to start thinking about all the legal aspects of patent licensing. That’s a great role they perform for us.
The IPP Office also provides funding to technology developers, to enhance both its utility to NASA and its commercial potential. Over the past few years, with NASA’s shift in emphasis to space exploration, funding for technology development for space science has dropped off considerably, and the IPP Office’s support has been critical to keeping the CADR development effort alive. This, fortunately, positioned us to propose the two-stage ADR for SXS and puts us in a good position for future x-ray and infrared missions.
NTB: But from a technical aspect, where do you see the potential in commercial applications for this technology?
Shirron: The main uses right now for ADRs are for detector development in labs and universities, in support of space missions. As I mentioned, for ground-based research there are other cooling techniques that have advantages over ADR. However, for detectors and instruments that will go into space, it is important to identify compatibility issues between the detectors and other components, so there is an inherent benefit to using ADR for detector development.
Beyond that, there are applications that could grow out of the easy availability of low temperature cooling that you can get from our continuous ADR, such as the monitoring of silicon chips coming off an assembly line. You might use x-ray imaging to screen them and certainly you’d want to limit x-ray doses, so with very sensitive x-ray detectors cooled by a continuous ADR – because you’ve got a continuous stream of chips coming off – you can do that evaluation in a potentially less destructive way.
Another one that’s farther out is something called quantum computing. This involves the use of quantum effects in very small circuits to perform the same functions that are now done at room temperature in silicon. Gates and memory storage units, ands and ors, flip-flops are being developed using these circuits, which carry the potential for extremely high processor speed and storage density. This work is currently being done using dilution refrigerators to do the cooling. But I can imagine that in the future, for real practical quantum computing on your desktop, you’re going to need a small, portable cooling system, and it would have to operate continuously, it would have to get you down to very low temperatures, it would have to be very robust, and that’s the niche that we see our technology fitting into. It’s very portable; it’s very reliable; it’s cooling at the push of a button. There’s virtually nothing to break down, so it has a lot of advantages over larger, fluid-based systems.
Those are just a couple of the possibilities I see for the future.
To download this interview as a podcast, click here