lmh wrote:
If this is a research project with no regulatory implications, then there are all sorts of lesser options you can consider:
Of course Peter is right, bulk compounds are cheaper than deliberately prepared standards, but for many compounds you'll struggle to get them cheaply in any form. If you can find a closely related compound that is available more cheaply, you can use this as a standard, and express your results on that basis; it will, at least, give some level of standardisation from experiment to experiment, and reduce variation caused by instrument efficiency. In some fields there is no choice but to use a related, non-identical standard: for example, the plant kingdom contains well over 100 different glucosinolates, but so far as I know only one is available as a commercial compound, so there are an awful lot of publications that estimate all their glucosinolates based on sinigrin. This isn't evil, provided you make it clear what you are doing. It's slightly annoying having to do it, because most chromatography software will, with justification, dislike cooperating! You would need to run your standard curve using cheap-X, then apply it to a set of peak areas at different retention times, which quite probably means doing the area-to-amount calibration curve bit outside the chromatography software - always an unhappy thing to do. You can also consider measuring the relative response factor of the cheap-X standard compared to the expensive standards of compounds A, B, C... and doing a conversion by using this "fudge factor" each time. Again, this isn't evil, provided you state what you're doing, and consider all possible ways in which it could go wrong.
Big thing: if your calibration curves are always linear, then you don't need to prepare a whole dilution series each time. Consider one-point calibration, and don't replicate your curve more than you can afford.
Consider for each experiment whether you are interested in absolute amounts, or looking for changes. If you're looking to see if a variety of hops is different, or if you're screening loads of lines of hops hoping to find one that's different, you don't need to estimate the absolute amount, and raw peak areas are quite adequate. Only perform the calibration when there is a real need for calibrated, absolute results.
I work with glucosinolate analysis and I can confirm they are expensive for standards, but there are several that are available just be prepared to spend something like $500 for 10mg or more of each one. We look for Glucoraphanin on a regular basis and if you want accurate numbers you do have to use it to calibrate because Sinigrin is not 1 to 1 on response versus concentration.
For the terpenes analysis, you could try what we do in the environmental lab, where you calibrate once, then check the calibration each day with one level of standard, if it passes with say +/-10% or 15% then call your calibration good and use it that day for analysis.
How much working stock volume are you making and why does it only last three days? I would think it would last longer that that without going bad. If you recap the vials and store in a cooler it should last quite a while unless they break down really fast.