Dear Peter, I presume you are trying to use the phthalimide as reference for calibrating 15N shifts. However, if you are able to appropriately calibrate your 1H shifts, there is no more need for an addtional nitrogen reference, you may calculate the required reference frequency from the proton spectrum. This is the recommended procedure nowadays and you may calibrate you 15N shifts either to nitromethane or to ammonia, as you prefer, with the same procedure. See Pure & Applied Chemistry vol 80 (2008) pp. 59-84 or Magnetic Resonance in Chemistry vol 46 (2008) pp 582-598.

I never use an additional standard when measuring 15N.

Thanks for the advice!

I am not really using the phtalimide to calibrate 15N-chemical shifts, I am measuring multimerization kinetics of small proteins (15N-labelled) and the Phtalimide is used as a reference or "probe" to check on what's going on with the shim. That way, its possible to control whether the observed kinetic funtion is caused by the aggregation or just by the worsening of the shim.

There is just one N-H coupling in the molecule, as it tis symmetric, the molecule is just not soluble enough, to get a sufficient signal in the HSQC...

How fast is your aggregation? If the aggregation is slow enough, it may be better to measure aggregation directly through diffusion edited (DOSY) techniques or off-resonance experiments. In any case, a 1H chemical shift reference should work just as well for checking the shim.

Potassium phtalimide (CAS 1074-82-4) is more soluble than phtalimide in water, DMSO, etc.

Its alkylation is described here http://www.orgsyn.org/demo.aspx?prep=cv1p0119 .

Jeffrey, that sounds interesting. I normalized the kinetics to the phtalimide signal, in order to be able to compare between replicate measurements. Do I need this as well in DOSY and off-resonance experiments? I am rather new to the field of NMR (more or less only applying the method). Can you point me to some literature on DOSY and off-resonance oulse-sequences?

DOSY is by far the simplest as the other methods give a mixture of internal and exchange dynamics. DOSY measures the diffusion of molecules in a magnetic gradient. The farther the molecule travels when the gradient is on the more the signal decreases. A small molecule has a higher diffusion constant than a larger one, hence its signal is attenuated more. The method is very accurate in measuring effective hydrodynamic radii, much more so than dynamic light scattering for small molecules.

http://www.umich.edu/~chemnmr/docs/diffusion.pdf

A typical set up would be to take a 1D 1H, an HSQC, and then a DOSY spectrum and then repeat. Other loops of experiments are possible depending on your protein.

https://www.researchgate.net/publication/235748979_Resolution_of_Oligomeric_Species_during_the_Aggregation_of_A1-40_Using_19F_NMR?ev=prf_pub

is an example of a real time NMR experiment (not using DOSY). If you want to give me more details about your particular experiment off line I may be able to help more.

Article Resolution of Oligomeric Species during the Aggregation of A...

Hi Peter,

good solvents for amides are CF3CH2OH and (CF3)2CHOH (trifluoroethanol and hexafluoroisopropanol). pH=2 is not a problem for this alcohols.

Succinimide as analog of phtalimide seems to be more suitable because of higher solubility both in organic solvents and water.