I have never seen a plastic with a co-efficient of expansion lower than most metals. The plastics are normally higher by a factor of about 10, are you sure you have not lost a decimal place somewhere.
Also the plastic will almost certainly go through a second order transition in the range from ambient to -184C, which will cause it to contract somewhat more than the co-efficent of expansion would indicate as it passes through that phase change.
I think sealguy gave you your answer with the injection mouldablr fluorocarbons. Why don't you ask their manufacturers for details.
The figur's you need are co-efficient of expansion and elongation at break, for both properties over the temperature range involved.
Problem is I expect no one has tested for this.
Can you get test plaques or pices of barstock of the canditate materials, drill a hole the appropriate size to allow for mould shrinkage in one of them, heat the plaque, and super cool an insert, place the insert in the hole, allow the whole deal to normalise overnight, then cool the part to -190C
Regards
pat
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.