The identification of degradation products requires normally the application of costly instrumentation (GC/MS/ LC/MS/MS, etc). An idea for an easy identification of new photolyzed products is the application of the "Spectrum" UV-analysis over a broad range of wavelengths (e.g. 190-500nm) in your spectrophotometer, and compare the different spectra you get for the samples retrieved with photolysis time. By this way, new peaks in different waveleghts can give a first insight over the new products formed, as well as their concentration increase or decay with time. You could also buy some reference chemicals of photolytic by-products of the aformentioned herbicides and compare their UV spectra with those measured previously (samples).

Foxboro has a potable IR analyzer and you can buy the mentioned herbicide standards to calibrate the instrument. You will be able to detect the changes by peak areas and wavelength. As Vijayan indicated the health hazards could be assayed only by toxicology testing

You say UV analysis - would that be HPLC with UV detector? Is your detector single wavelength of is it a Diode Array?

The first step would be to determine what UV degradation by-products you would expect to get. Do a literature search on each compound to determine degredation pathways of your parent compounds. That way you can specifically look for those compounds and cite previous work to validate that you are looking for the correct degredants.

Fro instance from-

http://www.cdpr.ca.gov/docs/emon/pubs/fatememo/metolachlor.pdf

Hydroxylation, dehalogenation, oxoquinoline formation, and demethylation

are the main processes observed during the photolysis of metolachlor, and

the major degradation product formed is 4-(2-ethyl-6-methylphenyl)-5-

methyl-3-morpholine (Mathew and Khan, 1996).

They should cite optimun wavelengths for each degredant. Don't be suprised to find out that some of them dont have a chromaphore and may need a different analysis to determine whether they are present.

you can make use of Gas Chromatography, coupled with the appropriate detector to identify the by- products of your reactions. From the knowledge of the by- products you can then predicts the toxicity.

To be sure of the degradation products at different UV fluence, LC-MS/MS is the best technique to be used and you can rely on its results while interpreting you results. Otherwise, GC-MS is an alternative in case of your molecules have low solubility in water.

Cytotoxicity tests are the most common techniques to be used in toxicity studies. Try making search in scopus about related work, I believe there are a lot.

You can use different types of bioassays. In general, we use COMET, SOS/UMU, MN et al for genotoxicity, and YEAST assay for EDCs, luminacent inhibition assay for cytotoxicity et al. You can define your targets of so-called "health effect" first, followed by the selection of the proper bioassays. This is quite similar the case as people go to hospital for "health issues".

Thanks for the input. I am trying to stay away from bioassays and anything that requires addition of other chemicals into the mixture. I am thinking about using RAMAN spectroscopy in conjunction with GC/MS for a 2-pronged analysis. That way I can get the typical MS results as well as the raman and between the two I hope I can get a good idea of what I'm looking at. I have read some studies that claim Raman makes water-based analysis a lot cleaner than typical IR. I still have to look at the "expected products" and determine if they will be responsive to Raman though. I like the GC because it uses tiny sample volumes and the process is mostly automatic. I'll just have to work out the details like carrier gas and flow rate and all that. I'm not super keen on LC simply because it produces a rather large volume of waste. The UV analysis I am performing is a single-wavelength study just to look at the reduction in concentration of the initial chemical. I am not looking at "natural" samples with a lot of junk in the matrix; I am simply dissolving the desired analytes in water. So I will know the initial concentrations from the mathematical formulation of my solutions. The device will probably have a diode array sensor anyway... but I still am not sure exactly what equipment I'll have access to. That's one of the joys of doing a science thesis in a department that has none of its own professors or equipment: I have to beg or borrow from other departments. (Fortunately this campus has a pretty big bio/ag division.) But after UV treatment, I have to assume that I know nothing about the reaction products. I don't want to say "I'm looking for the spectroscopic signature of compounds A, B, and C which are likely reaction products" but just answer the question "After photolysis, what chemical(s) are present?" Obviously I will have some idea of what I expect to find; that's just good science. But rather than go looking for specific products I just want a generic "unknown" analysis.

after UV treatment you can perform TLC to get initial idea of how many photo-degradation product are formed, separate them and then run for UV detection or Raman and try to get molecular mass either on GC coupled with MS or LCMS, this will help you, you can further elaborate your study by taking NMR data.......

After a lot of research, I've decided to go with FTIR and try GC-MS as well. The other methods just require too much prep time with the separations and extractions, and Raman needs me to boost the concentrations too high. I have over 500 samples to run and only 2 months to finish it all so speed is an issue. I'm using a Shimadzu 2450 UV instrument for tracking the concentration over time during the photolysis, but only for the initial compounds. The FTIR should make it quick and easy to see all the products' functional groups, and the MS report will give me some more info to decide which FTIR peak corresponds to a given product. So that should work. Luckily I'm doing everything in reagent-grade solvents and not environmental samples, so there won't be any need to remove the solvent or reduce the matrix before I go to the analysis.

Thanks all for the advice.