[DIYbio] Re: is it possible to isolate a unknown compound through HPLC

More specifics would help. When you say 'crude extract' do you mean 'crushed or lysed plant cells in water' or do you mean 'salt precipitation and ethanol extraction of plant material that's been pushed through a .22 um filter'? Crude extract meaning the extraction was poor/simple or meaning that it was a good/complex extraction just done by hand. Also, when you say 'each compounds' do you have something specific you're looking for or are you expecting an HPLC-detector system to point you in a direction. Because, depending on the detector system, you're going to get pointed in about 100 different directions. As for actually doing the deed, you can easily overload an HPLC column, you can equally easily overload a detector. Given that you haven't overloaded the HPLC or the detector, then the answers spit out are always relative, especially for unknown complex mixtures. As a rule of thumb, 1000:1 relative concentrations are 'saturated'. Using a mass detector with a 1mL sample with 100 mg/mL of unkown solute, a competently designed modern system should be able to detect e.g. 1 mg of protein in 99 mg of "stuff", etc. (~99:1). The well designed systems should be able to see 100-10 ug in a 100 mg sample (~1000:1). It's not too difficult to see low nano- to  high pico- levels of compounds in a background of low micro-level contaminants. There are rather easy and even simple techniques for getting well beyond the 1000:1 level. Most LC-NMR and LC/LC-MS/MS/MS systems can be tailored to do this, bead based affinity systems can be done DIY. However, without knowing exactly what you're looking for, what Nathan says about getting into $100K+ instruments and Ph.D.-candidate level work is fairly cogent.