It depends on which solvents you are dealing with. For lower boiling solvents like ethyl acetate, ether, hexanes and acetone, etc. using a roto-vap followed by several hours on a high-vac should be sufficient to remove these solvents.
For higher boiling solvents, especially ones that have a higher boiling point than water, you will likely need to azeotrope it with several rounds of an appropriate, lower boiling solvent, and then remove the low boiling solvent as mentioned above.
You can find lists of azeotropes on wikipedia and elsewhere.
It depends on which solvents you are dealing with. For lower boiling solvents like ethyl acetate, ether, hexanes and acetone, etc. using a roto-vap followed by several hours on a high-vac should be sufficient to remove these solvents.
For higher boiling solvents, especially ones that have a higher boiling point than water, you will likely need to azeotrope it with several rounds of an appropriate, lower boiling solvent, and then remove the low boiling solvent as mentioned above.
You can find lists of azeotropes on wikipedia and elsewhere.
Its depends on what kind of pulse sequence you used. You can suppress the solvent peak by solvent suppression pulse sequence(water suppression)and other some more advance in Bruker NMR spectrometer.
Jacob's method is what I often used. Especially useful with proton NMR, where you can use a proton-free solvent like CCl4 or tetrachloroethylene to "chase" off volatile solvents by evaporation.
Using high vacuum pump.. You can remove your solvent. If water is there dissolve your compound in dcm add sodium sulphate and again concentrate, or other polar solvent is there then do water workup , or you can do short column..
Normally, it is impossible to remove all solvent peaks from your spectra because the deuterization of NMR solvent cannot reach to 100 per cent. If your sample is not completely dry, it will come up in the spectrum. Hence, my recommendation is that you should dry the sample completely (not only by rotavapor, but by freeze drying overnight or by speed vacuum machine)
Second problem must be water peak. Since water presents in every solvent (even very non-polar ones like chloroform-d), it is recommended to use some salt to absorb the majority of water. You can see the water peak appears strongly in methanol-d4 or DMSO-d6 due to its polarity. My recommendation is to keep solvent in room temperature and use parafilm to cover whenever you finish the sample preparation.
Some spectral technique to remove the peaks are not typical in routine NMR measurements.
For 1H NMR spectra, the solvent residual signals arise from the proton of isotopomers containing one less deuterium atom than the perdeuterated solvent: e.g., CDHCl2 in CD2Cl2. For 13C NMR spectra,the solvent signals arise from the 13C atoms at natural abundance in the perdeuterated solvent.