This Monday saw a meeting to discuss the development of the Materials Property Validation Centre. Feedback to the survey carried out by the National Physical Laboratory formed the basis for a discussion on how to implement the recommendations of the Materials IGT Report most effectively. 101 organisations had taken part in the survey – and 80% had expressed their interest in taking part in this workshop. It must be because it was a Monday, but there were only about 20 of us who turned up!

There were three distinct types of materials property we talked about. The first is the familiar and most obvious one – straight mechanical and functional performance. There are already a large number of sources for such data, from the Materials Safety Data Sheets provided by suppliers to a (very large) number of databases that contain parts of the whole data range. They are usually provided by those who are trying to sell their materials (or classes of materials) so quote often present the attractive data and leave out the less than flattering results! This can mean that to get a complete data set on a particular material (or group of materials for comparison) the user must assemble the data set from different sources. Graham Sims of the NPL explained the problems with repeatability and reproducibility of data and the lengths organisations like the NPL go to to ensure that data is valid.

The second type of “property” we discussed was that concerned more with performance in use – here the “basic” data is combined with the structure and the environment to characterise performance both out-of-the-box and over time. Many end-users and final manufacturers use this data as part of their specifications and can regard the data as proprietary. It can also be very application specific – increase the temperature by a few degrees in use, or introduce a chemical into the environment and the fatigue resistance may well fall off for example.

The third type of “property” is concerned with environmental impact. It has aspects of provenance and fate – the source of the material, and its processing up to the stage it is being considered will all affect the “imbedded energy” or “imbedded carbon” values of a material and the need to recycle or dispose of a material after the current use will determine the end of life equivalents. This is increasingly important for those seeking to select a material for an application, who have no preconceived idea about what the “right” material is for the application and want to both start with a clean sheet and take account of the overall environmental impact of their choice.

My take on the discussion was that the first type of data was widely available but needed correlation between sources, the second type was too closely related to specific end use to be easy to generate and that the third type was horrible difficult to measure and/or calculate - but might have the biggest impact in the future. On this final point, about a third of those taking part in the survey had thought that there was no need for this type of data - but the rest, and many in the workshop, considered that it might be prudent to start building the databases before we have to use them!

The delivery mechanism also caused some discussion. Many thought that a web-accessible database was the answer, others thought that regional centres where materials could be characterised to provide “valid” data for their further use or recycling were the answer. I think the meeting resolved to use a hub-and-spoke structure which did both.

This all got me thinking about what our "goal" might be. I had a quick look at how various groups look at the question of validity. I must have led a sheltered life, because I have now discovered there are 4 types of validity and they build together to give us confidence!! And I thought we had problems with just the data!!!


David