NMR: As a rule of thumb, a 5 % impurity shouldn't interfere in NMR analysis. However, it all depends on the nature of your impurity/impurities. Is it another compound/a biproduct? Another isomer of your target compound? Residual solvent peak(s) ? Contaminants in your deuterated solvent? and so on.

The article below by Fulmer et al. comes in handy when dealing with impurities from reagents and solvents:

Fulmer, G.R; Miller, A.J.M.; Sherden, N.H.; Gottlieb, H.E.; Nudelman, A.; Stoltz, B.M.; Bercaw, J.E.; Goldberg, K.I. NMR Chemical Shifts of Trace Impurities: Common Laboratory Solvents, Organics, and Gases in Deuterated Solvents Relevant to the Organometallic Chemist, Organometallics, 2010, 29, pp. 2176–2179

IR: 95 %, but preferably above that. "Simple" mixtures of compounds could be identified with the aid of software that subtracts the spectrum of a pure compound from that of the mixture.

MS: as Martin stated above, this CAN tell you if you have the target molecule. We routinely run GC-MS or LC-MS on reaction mixtures and final products, but if you need to give your sample to someone else for MS analysis, I'd recommend you to aim for "as pure as possible".

You will, however, get a reliable mass spectrum as long as the right instrument/ionization method/detection mode is used and as long as no other compound is overlapping with your target compound = You don't need a completely pure compound if you just require the mass spectrum and fragmentation pattern.

For HRMS (e.g., for a publication), I'd make sure to have clean NMR spectra and all signals unambiguously assigned first.

Regards,

JR

MS informs whether you stuff is there or not. Even if you have 100% pure compound and you end up running on an inappropriate Mass Spectrometer, you may have nothing (e.g. +ve mode MALDI-TOF for negatively charged molecules doesn't give you anything)

IR and NMR: I'd prefer atleast 99% pure compound.

I think, 90-95% purity of your sample is good enough to get a reliable NMR, IR etc.

97% purity is reliable for NMR and even little less pure can do for IR. MS comfirms the presence of your compound so 70% purity should give a

clear signal.

NMR: As a rule of thumb, a 5 % impurity shouldn't interfere in NMR analysis. However, it all depends on the nature of your impurity/impurities. Is it another compound/a biproduct? Another isomer of your target compound? Residual solvent peak(s) ? Contaminants in your deuterated solvent? and so on.

The article below by Fulmer et al. comes in handy when dealing with impurities from reagents and solvents:

Fulmer, G.R; Miller, A.J.M.; Sherden, N.H.; Gottlieb, H.E.; Nudelman, A.; Stoltz, B.M.; Bercaw, J.E.; Goldberg, K.I. NMR Chemical Shifts of Trace Impurities: Common Laboratory Solvents, Organics, and Gases in Deuterated Solvents Relevant to the Organometallic Chemist, Organometallics, 2010, 29, pp. 2176–2179

IR: 95 %, but preferably above that. "Simple" mixtures of compounds could be identified with the aid of software that subtracts the spectrum of a pure compound from that of the mixture.

MS: as Martin stated above, this CAN tell you if you have the target molecule. We routinely run GC-MS or LC-MS on reaction mixtures and final products, but if you need to give your sample to someone else for MS analysis, I'd recommend you to aim for "as pure as possible".

You will, however, get a reliable mass spectrum as long as the right instrument/ionization method/detection mode is used and as long as no other compound is overlapping with your target compound = You don't need a completely pure compound if you just require the mass spectrum and fragmentation pattern.

For HRMS (e.g., for a publication), I'd make sure to have clean NMR spectra and all signals unambiguously assigned first.

Regards,

JR

Using low resolution NMR may not show the impurity signals/peaks in the spectrum. But using a higher resolution or cryoprobe will show even a little amount of impurity. If the integration is lower than the sample signals, impurities can easily be identified and should not interfere in structure elucidation. In short. 95% purity will give a good spectrum.

About MS: If you have 99% pure compound and not easily ionizable and 1% impurity which is highly ionizable, you will see the peak for impurity predominat over the pure compound. I would say, MS is tricky & not a quantitative method.

NMR: as mentioned above, you can quantify the aount of impurity (as long as it is >3%)..

IR: functional group dependent and not quantitative.

The best reliable way to know the formation of your desire product is NMR spectra. It should be minimum 90-95% pure.

Minimum purity for getting reliable IR & NMR/MASS is TLC

To get reliable IR & Raman the purity of the starting material is grater than 95% is more enough. The solvent impurity will reflect in NMR spectrum. But no such impurity patterns are present in the NMR & FTIR spectrum.

You can use a simple instrument called

CHROMATOTRON for accurate pure sample

According to me: The Spectroscopic techniques are sensitive & costly too. I Would like to suggest to all the researchers that, kindly confirm the purity of the synthesized/ isolated compound by some primary investigations like TLC (Rf), Melting/Boiling points (comparing with literature value). Afterwards will proceed for the Spectral techniques such as IR, GC/HPLC, MASS, NMR.