Cryogenic Sample Mounting Guide
Also check out our Sample-Mounting---9-Things-to-Know.pdf for a quick overview on sample mounting techniques in a cryostat.
To achieve the best base temperature on the sample, it is important to take several things into consideration. There are the obvious considerations such as the grease or adhesive used to adhere the sample to the sample mount, but also the conductance of all the components inside the sample, and the other heat loads brought into the sample such as wires or coaxial cables. Finally, the material of the sample determines the ability to absorb thermal energy and can impact the sample temperature.
MOUNTING THE SAMPLE
Since the most important variable in good thermal conduction between components is the effective contact area, the sample should be mounted to a cleaned surface free of debris that could act as an insulator. To achieve this clean surface, first, wipe away any remnant grease or adhesive with acetone or isopropyl alcohol. Also, the surface that the sample is mounted to should be smooth and polished, ideally it should be C101 copper, also known as OFHC or OFE copper, that is gold plated. The gold plating is necessary due to an emissivity dependence on the surface roughness. If copper oxidizes, this does not make a good surface for the sample to adhere to, as the emissivity increases. Gold is also malleable and thus readily deforms, which increases contact area upon sample mounting. Other materials such as aluminum or steel will not work as well and could lead to an increase in base temperature of the sample. Below is a simple list of methods used to increase thermal conductivity between a mounted sample and a cold platform.
N-grease is most often used to mount samples due to the ease of use. N-grease is not an adhesive, however, mutual surfaces containing N-grease can become tightly bonded at cryogenic temperatures. The conductivity of N-grease is 0.095W/m K (~300W/m K for good thermal conduction in copper), and as such, it is important to use a thin layer between the copper surface of the sample mount and the sample for good thermal conduction. In preparation for the application of N-grease, the surface of the metal should have a shiny appearance, and when the sample is pressed in place, grease should not press out the sides of the sample.
GE Varnish is semi-permanent and can be removed after adhered with the use of isopropyl alcohol or acetone. Since some samples cannot be subject to IPA or acetone, GE Varnish is not used often. If GE Varnish is going to be used, keep in mind the relatively low conductance (0.06W/m K). Because of the low conductance of varnish, to get good thermal conductance to the sample, the layer of GE Varnish should be kept quite thin. To achieve an optimally thin GE Varnish layer, a small amount of acetone can be added to thin the layer that is applied. After a very thin layer of varnish has been prepared on the surface, quickly (
Silver paint being very thermally conductive and with low electrical resistance can be occasionally used to mount samples. More often silver paint is used for non-permanent electrical connections or to quickly bond wires to the sample. It is important to note that silver paint can be messy and get on sample surfaces if not careful.
A user can also use other greases, adhesives, or even epoxies, but be sure that it can go into vacuum and will not outgas, and can also be used for low temperatures.
For optimal thermal conductance, and therefore the lowest base temperature possible, it is important to make sure that the sample is bonded well to the surface below it, but also all surfaces are bonded properly such that the ability to dissipate any added heat load is optimized. All surfaces have grooves in them to some degree, but ideally the surface would be as smooth as possible, and not rough. N-grease will fill the grooves that are naturally present in all materials. When available, it is best to bolt surfaces together. Although all samples cannot be tagged or bolted down, be sure that there is some sort of grease or adhesive between the metal and the sample.
In the Montana Instruments system we have set up the sample space for highest thermal conductance in using our HiCon material which has a conductance of 10000W/m K. At all junctions we apply N-grease, maximize cross sectional areas and apply pressure using screws with Belleville washers. Belleville washers keep contact as the temperature changes; for long joints it is especially helpful to use Belleville washers as the bolts contract at a different rate than the material around it, and the Belleville washers will keep thermal contact between all parts.
As informed by the conductance equation:
Where Q is heat flow, ΔT is the change in temperature, k is the thermal conductivity of the material which changes with temperature, A is the cross sectional area, and L is the path length. If you are designing a sample mount or sample chamber, keep these principles in mind.
In a closed cycle system, the user needs to be aware of the heat load imparted onto the sample, unlike bath cryostats where everything is cooled, in a closed cycle system the heat must be dissipated. In a closed cycle system, heat dissipation is accomplished by having good thermal conductance throughout, such that the heat can be pulled out of the sample. However, the user must also be aware of the sample emissivity and conductance. If a user has a sample that has high emissivity and low conductivity, then the sample will absorb any incoming radiation such as an incoming laser beam and will not be able to dissipate the heat load very efficiently. An example of this is the Cernox thermometer used in the Montana Instruments system; it is made of ceramic which has low conductivity and has high emissivity, so if the wire is not lagged properly, or if the thermometer is not shielded properly, it will read a falsely high temperature.
To get the lowest base temperature possible there are a few other considerations for mounting the sample. If the sample has electrical contacts running up to it, it is important that the wiring is done properly to not add a heat load to the sample. Please check the Wiring Guide for more information on properly wiring a sample.
All windows should be in place when making a measurement. With one window out, the sample is now exposed to radiation coming in from 30K which can raise the temperature of the platform and sample several degrees Kelvin. For optimal performance, keep all windows in place.
Attocube positioners are made of titanium and do not have a high conductance, therefore a thermal link or braid should be added to the stages to increase the thermal conductance through the positioners.
If the sample temperature is high, first ensure that it is not just the sample thermometer reading high. For instance, if the data indicate a higher temperature, then start to troubleshoot with added heat loads. Check that everything mentioned here has been considered:
- Smooth surfaces, and ideally soft like gold plating
- Bolted connections where possible – which provides force
- Low emissivity and high conductance materials throughout the system, including the sample material
- Some type of grease or adhesive to hold the sample in place
- No added heat loads from wiring, missing windows
If everything has been considered and the sample temperature is still high, perhaps replace the grease with something that has higher conductance, like silver paint, to see if this affects the base temperature.