Miniature Dilution Refrigerator

Almost all of the emerging applications for quantum technologies require low temperatures. This is fundamental to maintaining quantum coherence so while technologies such as on-chip cooling may lead to less stringent requirements for device platforms, it is clear that the need for sub-Kelvin systems is going to keep growing. Two-stage cryocoolers are well developed and typically provide cooling power at both 40K and 4K and temperatures down to about 1K may be achieved by pumping on liquid Helium-4. However, the only really practical to provide continuous cooling at lower temperatures is to use Helium-3: either by pumping on its surface (to about 250mK) or by dilution refrigeration (to a few mK). The most compact system available today needs three-phase power, cooling water, a separate rack to house gas handling equipment, a dewar of liquid nitrogen to cool a charcoal trap and stands above head-height to allow demountability. Furthermore most machines use between 10 and 35 litres of Helium-3. Currently, Helium-3 is only produced as a by-product of the manufacture and purification of tritium for nuclear weapons so it is a rare and politically volatile resource. As the requirement for low-temperature devices increases, it is essential that they get smaller, lower power, easier to use and use less Helium-3. These improvements will lead to lower environmental impact, wider access to the technology and lower costs. Compact Cryogenics is a new venture to take advantage of this market opportunity to make a desktop size, low power dilution refrigerator with no external Helium-3 circulation and a total Helium 3 requirement of 2-3 litres. The key aspect of this proposal is the innovative circulation system for helium-3/helium-4 mixture. Most attempts to date to commercialise a dilution refrigerator without external pumps have attemped to use a cycle in which a pair of independent Helium-3 refrigerator stages take it in turns to provide cooling power at 300mK, and this 300mK cold stage is used to run a 'cryogenic -cycle' dilution stage, in which Helium-3 evaporating from the still is condensed at a 300mK plate and so returned to the mixing chamber. This cycle was first developed by researchers in the USSR in the 1980s. These refrigerators had the disadvantage of very low cooling power at the mixing chamber as the mixture circulation could only be very small. A two-sorption pump, circulating design with 1K pot was developed in 1974 and a prototype made in the 1990s. There appear to be no current patents relating to this. We propose a radically different design in which the helium mixture itself is continuously circulated by a cold pump consisting of four cryopump sections and is cooled to the point of condensation by a Joule-Thompson expansion. In this configuration, the mixture circulation rate (and so cooling power) is limited only by the available 4K cooling power of the cryocooler. This is made possible by our introduction of greatly improved cryogenic valves, which can be actively driven in order to open and close each cryopump section.

Compact Cryogenics is a startup company. Jonathan Warren has many years of experience in specifying, designing and selling low temperature systems into physical science research laboratories around the world. The initial product developed as a result of this project will be targeted primarily at customers in the field of low-temperature detectors (such as superconducting nanowire single-photon detectors - SNSPDs), where there is a clear need for lower temperatures in a small, low-power, often portable format. Such users, in need of small-format systems, are currently using small 4K cryocoolers, or systems with Helium-4 circulation to get to about 1K, but significant improvements in performance can be acheived by moving to lower temperatures.
This is a relatively small market but is seeing considerable growth due to interest in using single photon detectors for quantum key distribution among other applications. We would expect to sell 3-5 sub-Kelvin units for this application in year two of the company, consistent with our limited manufacturing capability. This would equate to a turnover of about £300k. During this period we would refine the product in order to develop it into a system of more general usefulness in the quantum device research and development sector. This broader commercialisation would require verified, long-term performance data, optimised cooling power and base temperature and user-friendly operating software which would be developed over this period. We would expect to make a full 50mK demonstration system for loan to interested laboratories as well as showing the product at international conferences in order to establish it in the market.
Over the longer term (years 3-5), we would expect to establish a dedicated assembly facility (although parts machining would continue to be outsourced) allowing turnover to increase to £800k- £1.2M by the end of year four.