NHS should be stable in DMF. No decomposition to be expected. Last time I used IR is 15 years ago or so, so I cannot tell what the peaks of the CO-N-CO function of NHS are and how intense they are. What if you measure just NHS itself in the same concentration as you applied in your reaction?

You didn't rinse or something after casting? I mean, forget nhs, where did your dmf go!? You should see traces of that as well.Why gold if i may ask?

You can find a spectrum of nhs here http://webbook.nist.gov/cgi/inchi?ID=C6066826&Mask=80#IR-Spec

Maybe try to find some of the other peaks, in case of overlaps?

Hi Steven, thank you for your input. The NHS IR spectrum is available at SDBS (http://sdbs.db.aist.go.jp/sdbs/cgi-bin/direct_frame_top.cgi), the C=O vibrations are around 1780 cm-1 for KBr disc. I have two peaks in C=O area for activated ester and then no peaks after the reaction. I am attaching an image of this area so you can see what I am talking about.

I have purchased an activated ester as is, which means I currently don't have NHS in the lab. It wouldn't be impossible to get it, but I wanted to check first within the community if degradation is that type of a "common knowledge" which never gets written down in literature, because it is assumed as common knowledge :)

This is a bit puzzling. According to the reported IR spectrum of NHS, I would expect a big peak at 1700 from the C=O of NHS, but you don't see it in your product spectrum. On the other hand: in your starting material (black line), you only see the C=O of the activated ester, and no C=O of the succinimide, which I find a bit strange. Wouldn't you expect a peak there around 1700 as well?

My starting material is porphyrin ester which could influence the position of bands. I found some NHS and performed drop deposition from DMF solution. The thin film spectrum shows two bands at 1730 and 1781 cm-1 and a shoulder at around 1700 cm-1, which actually resembles the same area in porphyin ester when you look at the image provided above.

There is a small peak at 1715 cm-1 in the product spectrum when you look closely, but even if it comes from NHS, why would its intensity be so much weaker?

This just adds to the confusion, but I have since repeated the same experiment in DMSO, and got the same results.