Recalibration of Calibration Standards

posted on Thu, Jan 7 2010 9:20 PM

rated by 0 users

We calibrate the XYZ scan axes of our AFM using a traceable height and pitch standard made of thermally grown silicon dioxide on silicon. The specimen contains these patterns:

10 um pitch, 2-dimensional array of pits, nominally 200 nm deep

2 um pitch, 1-dimensional array of ridges, same depth as the pits.

The manufacturer's certificate of traceability does not discuss recalibration or indicate any need for it.

Based on the materials of construction, I believe that the pitch values and the pit depth are stable and that no recalibration is needed, provided that we are careful in handling and using the specimen.

For example, any nanodebris visible in the image is excluded from the measurement.

However, the quality consultant says that it is common to send such samples to an outside lab for recalibration. He added that we should either do recalibration or write a justification

 for why recalibration is not necessary.

I am curious what other people do about this. One can consider this in several different environments:

-the AFM manufacturer, who wants to provide good calibration for the instruments

-the AFM user who is supporting a production or QC application

-the AFM user who is mainly supporting R&D, but needs to show traceability in order to meet the requirements of a quality management system.

Here are a few practical questions to start the discussion.

If you have such a standard recalibrated, what is the time interval?

If you do not recalibrate the standard, what rationale do you give for why it is unnecessary?

Can you articulate the reasons so that it is clear they are based on good science and engineering data and practices?