I am performing a determination of riboflavin with fluorescence spectrophotometry. I've dissolved my riboflavin in a dilute GAA solution, created a calibration curve and measured my unknown's fluorescence.

For these types of analyses I usually perform triplicate measurements, however I've noticed that spectrofluorometric measurements always decrease with replicate number. I assume this is due to 'photobleaching' of my samples. Since the phenomenon occurs with replicate rather than over time, and since my sample is in a dark environment, I would imagine this can only be the result of the excitation light emitted by the instrument itself. I find it surprising that this causes such a dramatic change in fluorescence from measurement to measurement. I also notice that with enough replicate measurements, eventually the intensity decline seems to bottom out. Can anyone explain the specifics of this phenomenon to me?

  • Am I right in assuming that this is photobleaching (deactivation of the chromophore) rather than photodegredation (breaking the molecule), or are these the same?
  • How exactly is the fluorophore deactivated?
  • Can the degree of fluorescence intensity decline be predicted quantitatively?
  • Would my calibration be more accurate if I used single measurements instead of taking the mean of three replicates?
  • $\begingroup$ While I can't answer all of these questions, I doubt you're looking at photobleaching. Photobleaching happens when a significant portion of the electrons have left the ground state, and unless riboflavin has an exceptionally long fluorescence lifetime, you really shouldn't see this effect. Photobleaching is what you see when you wail on a few thousand molecules with hundreds of watts of power for extended periods of time. My best guess is you're looking at some sort of photocatalyzed reaction. Does the fluorescence return if you leave the sample in the dark for ~5min? $\endgroup$ – chipbuster Oct 11 '16 at 5:21
  • $\begingroup$ I may also not have understood your details about "replicates rather than time" comments correctly--please let me know if that seems to be the case. $\endgroup$ – chipbuster Oct 11 '16 at 5:23
  • $\begingroup$ I think the sample's intensity is going down as a function of the number of times I hit it with the excitation wavelength (to take a measurement), rather than just over time. $\endgroup$ – gannex Oct 11 '16 at 5:42
  • $\begingroup$ My gut says "photodegredation," (which is not the same thing as photobleaching) but I have no idea what the nature is. It could be that the excited riboflavin forms a complex with the GAA, which could either be permanent or slowly degrade over time. If it's the latter, you could potentially restore the quantum yield by keeping the sample in the dark for a while (or heating it up to break up the complex). $\endgroup$ – chipbuster Oct 11 '16 at 5:45
  • 1
    $\begingroup$ these are good ideas @chipbuster. I initially thought photodegredation, but I was skeptical because I thought it would only happen to small, easily ionizable molecules, so I figured with a large organic like riboflavin it must be some sort of small-scale degredation of the chromophore that leads to deactivation of the fluorescence. Separate samples would certainly be best. I just figured since spectrofluorophotometry is such a common technique, more people would know this problem. $\endgroup$ – gannex Oct 11 '16 at 5:59
  • That looks like a bleaching to me. The notion that bleaching only occurs with excesive light intesity on small populations of fluorophores is news to me. Might also that in my field we just use the bleaching term loosely. Best way to test is to take a long exposure of your sample and monitor the fluorescence intensity over time.
  • With our chemistry we usually encounter intensity loss that can be fitted with exponential decay function which would explain the bottoming out that you are describing. We often take multiple such "bleaching curves" and fit them to correct for intesity loss in our experiments that can take multiple hours.

  • In case you are only exposing your sample for short periods i would just make sure you measure each sample the same number of times that should rule out any error between samples. Thus reducing to only one measurment will not make it more correct.

  • Bleaching is poorly understood thus far. There is not a way to predict what excatly happens with a specific molecule.

Also how big is the effect for example (Start value)/(3rd measurment value) ?

  • $\begingroup$ The difference between the 1st and 3rd measurement at 0.2-1.0 ppm ranges from 0.04 to 0.2 emission untis. It's a pretty small difference, but I did try some with many repeated measurements and it declines for quite a while but eventually bottomed out. I didn't save those data though. I agree with you that it's bleaching. Making a bleaching curve is a good idea. $\endgroup$ – gannex Oct 28 '16 at 22:08
  • $\begingroup$ the (start value)/(3rd value) ranges from 1.008 to 1.012 for the standards of 0.2-1.0 ppm. For only three measurements, it's not a big change. $\endgroup$ – gannex Oct 28 '16 at 22:12
  • $\begingroup$ Could you also address the other subquestions so that this can be a complete answer? $\endgroup$ – jonsca Oct 28 '16 at 23:22
  • $\begingroup$ so you are looking at and 0.8 to 1.2 % change in intensity which is imho an okay value depending on your setup. To rule out any other processes one could also alter excitation intensity to check if the relative intesity loss over multiple measurments dependes on the excitation intensity. I updated the answer to adress every aspect you brought up. But I am at a loss with regards to the mechanism of bleaching effects. $\endgroup$ – simlan Oct 29 '16 at 10:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.