[pcmcc]

Hi All,

I am currently trying to set up and qualify a small lab for pharmaceutical raw material identification. I am primarily looking at a PerkinsElmer Spectrum Two FTIR for identification of excipients, but my main concern is about raw material variability. Some of our raw materials are either going to vary some from lot to lot (eg Fetal Bovine Serum) or have quite a few ingredients (eg DMEM/F12 cell culture media). Since most of these materials will be nonUSP, my current thought is to build up a reference library in house and compare new lots to the previous. Does this sound like a viable identification method? Does anyone have any experience setting up an FTIR for this type of application? If so, any advice/thoughts you can share on selection, qualification, etc. would be greatly appreciated. Thanks!
 

[marquis]

Setting up your own library is viable and will work.  But, you need to limit the variables as much as you can.
Here's what I mean. 

Make sure the prep methods are always the same.  If one material is qualified using nujol mull, make sure it
is always prepared using nujol mull. Try to ensure the film thickness in the IR prep is always the same. And make
sure the same amount of raw material is used (for example, specify 0.1 g).  Otherwise, extraneous peaks will show up.
The extraneous peaks lead to  non conforming raw materials, CAPA, root cause analysis... You get the idea.

Even then, you will find out about additives not specifically listed.  Silica was one that showed up often for us.

Many materials don't show up in normal IR.  If you are using IR to qualify a raw material, make sure it absorbs
IR (this sounds obvious, but is often overlooked).