Unlike cryostats based on the boiling of liquid helium, cryocoolers are based on the cyclic compression and expansion of helium gas (closed-cycle cryostat). They are interchangeably referred to as:
- Closed-circuit refrigerators
- Cryogen-free (dry) dilution refrigerators
The single-stage systems of the early days achieved temperatures of 45 K, while the double-stage systems reached 9-10 K. They had enormous appeal because they were so easy to start up. All you had to do was plug them in and you were ready to go. For scientists, they marked the end of having to perform laborious liquid nitrogen and liquid helium filling operations, all too often at unearthly hours (reliable helium level and helium consumption measurements were a later development).
The cryocoolers therefore had a bright future ahead of them and as early as in the 1970s Dominique Brochier, head of cyrogenics, thought that this technology was going to sweep aside conventional cryogenics just as the electric engine had swept away the steam locomotive. In actual fact, things didn’t quite work out that way. The cooling capacity of the cryocoolers was relatively limited and the Orange cryostats and the associated infrastructure (helium recovery, automatic filling, etc.) gradually cancelled out the vast majority of their advantages.
The first cryocooler on D9
(a Displex CS1003) was probably a "split-pair Stirling" refrigerator (see below)
How to turn cryocoolers every which way
No doubt because of their relatively modest performance levels, the first cryocoolers were snubbed by ILL instruments, with the exception of the 4-circle single-crystal neutron diffractometers. On this type of diffractometer, the crystal in fact rotates in all directions and for a long time this was incompatible with liquid helium cryostats.
By 1972, this difficulty had been overcome by the helium-flow cryostat on D10, but the cryocoolers offered a much cheaper and much simpler solution. In those days, it was easier to use cryocoolers with neutrons than with X-rays because X-ray diffractometers were generally too small and fragile to accommodate these heavy and bulky pieces of equipment.
In 1976, at the initiative of M.S. Lehmann and Russ Simms, D9 was the first 4-circle single-crystal diffractometer to be equipped with a single-stage cryocooler (Displex CS1003 from Air Products) reaching temperatures down to 50 K [1,2]. This was followed shortly after by D15, which, at the suggestion of J.M. Reynal, was fitted with a double-stage cryocooler (Displex CSA 202 from Air Products) capable of cooling to 10 K . This was possible thanks to the sturdiness of D15’s offset phi-circle Eulerian cradle (Mark VI from Grubb-Parsons). In 1981, D9 was equipped with a similar system but one which allowed x, y and z positioning of the sample and temperatures down to 14 K . In 2016, D9 is still using a similar device although now capable of cooling to 2 K.
However, to achieve even lower temperatures, there are only two options: either go back to liquid helium with D10’s gas flow dilution cryostat (50 mK) or change diffractometer geometry, i.e. switch from 4-circle geometry to "normal beam" geometry.
Although cryocooler technology has come a long way, in particular with the advent, at the beginning of the 2000s, of "pulse tubes" capable of reaching 2-3 K, the Orange cryostats, which have been constantly improved, continue to have the upper hand in terms of cooling capacity, temperature range and sample change time.
Mogens (Mons) Lehmann
©2014, Dorte Lehmann
1976: The cold head of the Displex cryocooler
mounted in the Eulerian cradle of the diffractometer D9 (Tmin= 50 K)
1981: General view of D9
On the left-hand side, we can see the cryocooler’s compressor
1981: Single-stage cryocooler
mounted in D9’s Eulerian cradle
1986: Displex DE 201 (Air Products)
mounted on D9
2001: The cryocooler
of the upgraded diffractometer D9 (Tmin= 2 K)
©2001, ILL and Studio de la Revirée
1981: The double-stage cryocooler of D15 (Tmin= 10 K)
Jean-Michel Reynal is keeping a close eye on his masterpiece!
- "Electron Density in bis (Dicarbonyl p-cyclopentadienyl iron) at liquid Nitrogen Temperature by X-ray and Neutron Diffraction", A. Mitschler, B. Rees and M.S. Lehmann, J.A.C.S. 100 (1978) 3390. This article only contains one reference to the D9 cryocooler. The assembly was really only described in article .
- "Use of a single-stage, closed-cycle cryorefrigerator for routine measurements at moderately low temperatures on a four-circle neutron diffractometer", Allibon J.R., Filhol A., Lehmann M.S., Mason S.A. and Simms P., J. Appl. Crys. 14, 326-328 (1981).
- "Cooling to 10 K on a four-circle diffractometer by means of a double-stage cryorefrigerator", Filhol A., Reynal J.M., Savariault J.M., Simms P., Thomas M. (1980) J. Appl. Cryst. 13, 343-345.
- "A Simple, Adjustable Mount for a Two-Stage Cryo-Refrigerator on a Eulerian Cradle." J.M. Archer and M.S. Lehmann, J. Appl. Cryst. 19 (1986) 256.