Flourescence settings

[miro2009]

Hello chromatographers,

I'm new to using flourescence detectors in our HPLC methods, we basically started implementing it as a detection system for it's higher sensitivity. As for the settings in our model Shimadzu RF 20Axs, there are 2 settings I'd like to ask about: Sensitivity (low-medium-high) and Gain (x1, x4 and x16). I understand they are settings for amplifying the signals obtained and could be used in combination for highest magnification, but are there any tricks or drawbacks that I should consider when using this option (other than possible slight noise increase)? would it affect signal linearity or accuracy for instance? I'm happy to observe significantly high signals with each setting while I was suffering with minute signals with UV detector.
Also, why isn't it common to find also UV detectors with such signal amplification option? In other words, how this "maginifaction" work, does it need special types of signals that could be possible to magnify (like flourescence)?

Another general question, how different is it to develop and work with a method using RF when compared to UV? are there any precautions/disadvanages of RF vs. UV that could be overlooked by a common user?


Newbie in the RF world!

[tom jupille]

UV absorbance and fluorescence involve fundamentally different principles.

For "high sensitivity" work in UV, we are looking at a small difference between two large quantities (UV detectors actually measure transmittance; they *calculate* absorbance as the negative log of transmittance). That means that there is plenty of light intensity and so no advantage to boosting the sensitivity of the photodetector; quite the contrary, since that can increase noise.

In fluorescence, we are looking at the presence or absence of emitted light at a particular wavelength. That means that the intensity of the light is very low, and boosting the sensitivity of the photodetector can help -- up to a point where the noise gets excessive.

Overall, fluorescence detection is fairly trouble free. Obviously much more selective than UV. Watch out for quenching of fluorescence in some molecules by dissolved oxygen as well as pH sensitivity and, of course, the usual suspects like bubbles.

[HW Mueller]

You just have to experiment with these parameters to get the optimum results. My fluorescence detector is actually a spectrometer which has only the slit widths as variables ahead of the PC connection, if I remember correctly.
The main difference between UV and Fluo that comes to mind is that fluorescence is even more susceptible to environment than UV. You better not have changes in polarity of the mobile phase (from one run to another), nor heavy metal ion fluctuatios, organic quencher fluctuations . . . .

[Alex Buske]

Hi,

I don't know the RF20, but with the RF-10 the peak height increases with low-medium-high sensitivity as does the noise, the same for the gain. AFAIR sensitivity is an internal (digital) amplification and gain is the amplification for the analog output only.
On method development: try to get FL spectra (ex and em) in the elution solvent. Gradient runs work perfectly with FL and spectra are quite similar in ACN and MeOH. Stable room temperature is a plus as well as degassed solvents.

[miro2009]

Thanks all for the very helpful and informative replies..
But back to the signal amplification issue during method development, so I just need to settle on a factor ( for eg. x16, x1000, x... my detector manual mentions its capable of up to x20000 when combining "high" sensitivity with x16 gain) and then I should specify this in my method protocol/SOP? Do they do that in publications and literature with such methods?
and if I settle on such high amplification where I find the noise still acceptable, should I expect problems regarding signal linearity, accuracy... basically during method validation?

Also, does this amplification cause any extra drain of the detector's life?

[tom jupille]

You are correct that you are balancing sensitivity with noise. My choice would be the lowest gain, etc. that give me adequate sensitivity. My experience with fluorescence spectrospcopy is several decades old, and was with photomultiplier tubes, so solid-state photodetectors may not have the same issues, but the photomultiplier tube life was better at lower voltages (i.e. a lower sensitivity setting).

The gain and sensitivity setting are likely to be specific to your model of instrument. I would certainly document them in the method or publication, but in an SOP I would leave future users some wiggle room by including a statement like "or equivalent", and include a system suitability requirement so they can know what "equivalent" means (perhaps signal/noise for a test standard ??).