chemiluminescence using 1100 fluorescence detector

[priz]

Hi,

I need to use my Agilent 1100 series fluorescence detector as a chemiluminescence detector and I have Chem Stations Rev A. 10.02 (1757).

Has anyone done this before? I'm running with "enable analysis when lamp is off", but I don't think it's actually working because my baseline is really flat when I should be picking up some background chemiluminescence and I haven't been able to detect any of my standards.

I think I need to use some different settings on the Chem Stations software, but I'm not sure how to do it. I'm waiting to hear back from a few different people at Agilent. Please let me know if you have any experience in doing this or if you have any suggestions.

Thanks

[Mark Tracy]

I don't know about the detector settings, but do check the physical setup. I assume that you are using post-column addition of the chemiluminescence reagent. The mixing tee needs to be as close to the detector cell as possible; the half-life of the reaction is pretty short.

[priz]

Hi Mark,

We are using post-column addition of the chemiluminescence reagent. Our mixing tee is pretty close to the detector, but we might be able to bring it a little closer. Our reaction is isoluminol, microperoxidase and lipid hydroperoxides.

thanks,
teresa

[danko]

Hi priz,

If you’re not sure, the detector actually is able to detect any signal with the lamp tuned off, you can just set the excitation wavelength to a value that won’t excite any of the compounds you have in the system while the emission wavelength is the “correct oneâ€

[priz]

Dear Danko,

Thanks so much for your suggestion. I've actually been trying what you suggested.

I've set my excitation at 700nm and my emission at zero-order to detect emission at all wavelengths. I find if I do this, I do detect chemiluminescence peaks. I've also been experimenting with physically blocking the excitation path at the flow cell. The effect of blocking light gets a little tricky to interpret with the calculations that Chem Stations does with the excitation and reference diode.

Right now, I am able to detect my chemiluminescence peak, but my working standard curve is 8000 - 30000 pmoles when my reference paper is detecting 10-600 pmoles. My background baseline is high (500 luminescence units) even when only my mobile phase pump (Methanol:tert-butanol 9:1) is on and not my chemiluminescence reagent pump. I'm concerned about the high background baseline. I guess I'm worried about swamping the detector (I come from a GC-FID background) if I'm starting with such a high baseline.

The noise is very high, but I have my photomultiplier tube (PMT) gain set high, 15, on a scale of 0-18 where 10-15 is used for "normal" applications according to Agilent. I think I need a high PMT gain for better signal to noise. If I reduce my PMT, I lose my peak. I'm going to work on finding a better balance. In hopes of reducing noise, I'm going to treat my solvents with a chelex resin or molecular sieves. I also plan to increase the response time to lower noise. Today I will try increasing the response time and adjusting the PMT.

If you have any other suggestions or ideas or know of anyone else who is doing this, please continue to share with me. I'm still not sure if my system is actually going to work for my application.

Danko, thanks again.

Priz

[danko]

Hi Priz,

Glad to hear you get a signal. I thought of suggesting zero-order emission, but I reached to the conclusion that you might end up with a noisy baseline. That’s why my suggestion was to identify the wavelength max of the emitted signal (you can easily do it now by extracting it from the generated data) and set it on the emission side. You see, when you use the zero-order option, the instrument amplifies all wavelengths and thus adds unreasonable amount of noise. If you choose a single wavelength, that is what the amplifier will “concentrateâ€