# What are some general tests that would determine an unknown compound?

I know a little bit about unknown compounds but I am still new to the topic here. I know that you could use filtration, chromatography and distillation. You can also combine the unknown with others to form a compound. And you can measure properties and such and such to figure out what the compound is.

If I were to do this, what should I watch out for, and what would be the best way to proceed? I would appreciate it if you don't go "PhD" on me.

This question will be ambiguous for most people, but I am looking for a general idea and what you would normally do, instead of "what I would do in this situation".

• Normally you use some combination of NMR, UV-VIS and mass spectroscopy. – Jori May 14 '15 at 22:25
• If its an inorganic compound, one can use the now outdated method of Inorganic Qualitative Analysis. This is especially useful if you're not familiar with spectroscopy (which I am not). However keep in mind that this method has many limitations. – Binary Geek May 15 '15 at 4:51

As usual, it depends on what is available.

Mass spectrometry, $^{1}\ce{H}$ and $^{13}\ce{C}$ NMR, IR, UV/VIS spectroscopy, would be obvious choices in organic chemistry nowadays. XRD of a suitable crystal would be nice to have!

Back in the days, (undergrad) laboratories didn't have sophisticated analytical instruments and you were happy to find a melting point apparatus, a refractometer and an IR spectrometer.

You would have used IR spectroscopy to identify funcional groups, possibly supported your hypothesis by using staining reagents for TLC, like Dragendorff's reagent and then started to synthesize derivatives from which you knew that they show sharp, well-defined melting points.

You would have converted

• amines to sulfonamides
• carbohydrates to osazones
• ketones and aldehydes to 2,4-dinitrophenylhydrazones or semicarbazones

and compared the melting points of the derivatives with tabulated values. Those were the days…

• Your last comment reads almost like you would be sorry about that we have advanced O.o – Martin - マーチン May 15 '15 at 8:07
• @Martin-マーチン No, not at all. I don't miss the 60 MHz Varian NMR (CW, ink plotter). And remembering that one had to measure the refractive index of rather volatile acid chlorides almost brings back the tears - but those were not the tears of joy ;-) But it is still remarkable how much was achieved in natural products chemistry and structure elucidation in those early days. – Klaus-Dieter Warzecha May 15 '15 at 8:19

Is anything about the material known? And how much of it do you have? If you think the material might be inorganic, an adsorption, emission, or mass spectroscopy might be the first step, as that will tell you what elements are in it. If the sample is large enough, and can be formed into crystals, then x-ray diffraction will likely give a thorough idea of the structure. If it's organic, NMR is probably the most powerful technique, and only requires a minuscule amount of the sample. Mass-spec and IR-spec are also commonly used.