Home      Publications      Teaching      Software      Hardware      Dissertation

Scientific Hardware

Liquid Nitrogen Immersion Cryostat

I designed this liquid nitrogen immersion cryostat for infrared spectroscopy of frozen samples, but it has also proven useful for visible spectroscopy as well.  Conventional (non-immersion) cryostats for spectroscopy position the sample in an evacuated region, which works fine for solid phase samples or for liquid samples that can be completely sealed.  However, for liquids that need to be rapidly frozen, or for sample holders that cannot be sealed (such as for Stark effect spectroscopy), an immersion cryostat has proven to be a better option.

cryostat picture

The cryostat is described in my paper in the Review of Scientific Instruments, which can be downloaded from my Publications web page, or in Chapter 2 of my dissertation which is on the Dissertation web page.  The list of parts, some notes, and original drawings are available in this pdf file: Drawings.pdf.

A few changes have been made added to my design since I had 2 cryostats made in 2001.  Tony Kanchanawong, in the Boxer group, designed a window assembly for 4 windows, rather than the 2 that I used.  I don't have the machine drawings for it and I don't know if it was ever built.  Also, Janis Technologies will now make an improved version of this cryostat by special order (contact Pete Shields) for about $10,800.  Here is a diagram of what they are selling: JanisPict.pdf.  As far as I know, only one of these has been sold.  It is probably somewhat cheaper to make a custom one, and especially if campus machine shops are subsidized.

While the cryostat is designed primarily for liquid nitrogen, it has also been successfully used with liquid oxygen, which is useful primarily for its different infrared absorption properties.  Following is a transmission spectrum through the cryostat with liquid nitrogen and liquid oxygen, and no sample.  Solid line: empty cryostat, long dashes: liquid nitrogen filled, short dashes: liquid oxygen filled.  The cryostat windows were broadband anti-reflection coated, for 5 microns, 2 mm thick Cleartran from Spectral Systems.


Note that it is hard to avoid absorption by water.  With effort, it is possible to reduce the size of the water peak, but it always increases every time the cap is opened, such as for sample changing.  I also designed a liquid helium insert for the cryostat for 4K and colder research although I never built it.  It's described the downloadable pdf of parts listed above.