LIGO Document E0900480-v5
- FTIR Testing to Qualify Parts for LIGO UHV Service
The ultimate requirement of = .02 micrograms/cm^2 corresponds to ~ 0.1 monolayers of oil or polymer (assuming a density ~1 g/cm^3 and a molecular size of ~20 angstroms).
- V5: Added Hexanes as potential solvent for use, refer to E0900479 (v2 onward) for details on sampling with Hexanes.
V4: For the reasons cited below, we are relaxing the requirements on acceptable NVR level for sampled holes to the following:
For threaded holes the maximum acceptable NVR level is = 1.2 microgram/hole for through holes and = 2.0 microgram/hole for blind holes, both prior to the air bake.
If an FTIR is conducted after an air bake, the acceptance level is = 0.4 microgram/hole for through holes and = 0.7 microgram/hole for blind holes.
In general is is not necessary to perform the FTIR sampling both before and after an air bake. However, for the next few large plates we will perform FTIR sampling for JPL analysis at AstroPak, upon receiving at the Observatory and after air baking the part.
Reasons for the change:
Our pass rate with the previous criteria was not very high. We had to repeat cleaning on a number of parts. While this is good if we are catching holes which were missed during the cleaning, or other problems such as compromised packaging during shipment, there is evidence that our criteria is just at or below the level achievable with our cleaning processes, both in-house and at AstroPak (our principal contracted precision cleaner). Most significantly, AstroPak took considerable care and spent a great deal of effort in cleaning a recent large SEI plate and yet still failed to achieve the criteria in v3 of this document (when sampled at AstroPak and so not subjected to the shipping environment). See for more detail. For this recent part, the JPL FTIR results were 1.7 micrograms/hole for blind holes and 1.04 micrograms/hole for through holes. Without researching a change to the basic cleaning approach and spending considerably more time and money, it is unlikely that we can achieve better FTIR results.
DCC Version 3.4.0, contact