Reactivation of a fluorescence detector

[bunnahabhain]

Hi everybody
We have a Dionex RF-2000 detector that I am about to reactivate. First thing I tried is the Dionex PQ. It failed .

The first PQ test is the noise test: Excitation 350 nm -> emission 394 nm. Autozero by the detector is performed with 350 -> 450 nm, after one minute, the values are set to 350 -> 394. Quality criterium is: Emission range 40 - 80 mV at 350 -> 394. But: After the emission wavelength is changed by the program, I get a nevative reading. So I recorded an emission spectrum from excitation 350 nm, it looks like that:



I understand now why I get a negative reading from the program above, but I don't understand the spectrum. Any input from the experts?
Thanks a lot
Jörg

[M Farooq]

One should get a Raman peak at 394 nm by excitation at 350 nm for pure water. Can you establish wavelength accuracy of your monochromators first?

This is from the manual: http://www.dionex.com/en-us/webdocs/810 ... ct2008.pdf
The Fluorescence sequence is used to determine the wavelength accuracy of the emission spectrum. When testing the Dionex RF 2000 detector, make sure that the ZWAVE parameter is set to 1 on the instrument (see section 3.4.1). Water is pumped through the flow cell at a flow rate of 1ml/min. For an excitation wavelength of 350 nm, the emission wavelength changes in 1nm increments from 380 nm to 410 nm. The relative signal maximum is compared to the theoretical maximum.

[bunnahabhain]

Hi M Farooq

The ZWAVE parameter was set to 1. This is the result from the emission spectrum:

This does not really surprise me: The PQ program autozeroes at 350 -> 450 (in the middle of the peak I presented in my first post), then it scans in a range where there is only noise. I guess the instrument is completely out of alignment. I sthere a way for me to correct this, or will I have to call a technician?

Jörg

[danko]

Are you sure the water you're using is absolutely pure?

Also, check the Ex. and Em. with some fluorescent sustance to confirm the correct wavelengths.

Best Regards

[M Farooq]

Hello Jörg,

You are right, there is nothing but noise in the spectrum. The absence of Raman peak of water in the scanned region shows that the alignment of gratings/optics is out. We don't have exactly the same detector but Raman peak is a generic test for all spectrofluorometeric measurements and wavelength accuracy tests. Is it possible that the cell is dirty? The Raman peaks are not very intense though.
If cleaning the cell with solvents (by flushing) still shows the same level of noise, it is time to get technical assistance. Good luck.

[M Farooq]

Are you sure the water you're using is absolutely pure?

Also, check the Ex. and Em. with some fluorescent sustance to confirm the correct wavelengths.

Best Regards

Although it is highly desirable to use pure water, the Raman peak test for wavelength accuracy has nothing to do with the purity of water. Unlike fluorescence, scattering peak intensity is not affected by impurities. Even tap water should show a scattering (Raman) peak in the scanned region.

[bunnahabhain]

Thank you M Farooq

I flushed the cell with 0.1 M HNO3 / water / acetone and the spectrum I posted earlier remains unchanged. You said the Raman peak is not very intense, what I see is a very strong peak at 463, maybe the raman peak is just overlaid by that.

Here is the spectrum I obtained from phenanthrene (ex. 266 nm) in MeOH/H2O (10/90):


I guess also here we can see the 463 and 808 nn peaks overlaying the phenanthrene spectrum. I don't think I will get any further on my own. Thank you for your assistance!
Jörg

[danko]

Pure water is essential - believe me.

Also, try scanning phenol (Ex. 272/Em. 300 nm) to verify the calibration of the gratings.

Best Regards

[bunnahabhain]

Hi danko

I really trust you that pure water is essential, I use LC-MS grade water for the emission spectrum of water, it's the best I have.

Now have a look at Phenol in water, ex 272 nm:


And water ex 272 nm:


I guess this is not what we want to see

Joerg

[M Farooq]

Hi Joerg,

It is interesting to see the spectra which apparently make no sense whatsoever. The presence of peaks around 700-800 nm shows that there is something wrong with the calibration of the gratings, although the manual says that the detector is good till 600 nm. The peaks around 600-800 nm is not even a second order peak in your emission spectra (which appears in some fluorescence measurements as 2 x excitation wavelength).

Please share the solution or the fault once a technician finds it out. The emission spectrum of phenol is even stranger which shows that phenol glows red under UV

Regards,

Farooq

[bunnahabhain]

Update:
I have just installed a new lamp, see attached pictures:

Water (ex 250, em scan):


Water (ex 350, em scan):


ca. 1 mg/L Phenol in water (ex 272 scan):


With the new lamp, the first peak in the em spectrum matches the excitation wavelenght, the second matches 2x ex.
For phenol, we see a maximum at 300 nm, but again, overlaid by the "water".
What does all this tell me?

Jörg

[danko]

Hi Jörg,

As you might have figured it out, the peak at the excitation wavelength is Light scattering which is normal. The peak at the excitation wavelength x 2 is second order light scattering and its normal too.
The only thing not normal is the missing Raman Light Scattering (RLS)
I can think of one possibility and that is too much Stray Light in the detector which decrease the sensitivity in general. Keep in mind that RLS results much weaker signal than genuine fluorescence. So maybe your RLS is swamped by the Firs order Scattering and the background signal (The stray light). I believe the weak phenol emission signal supports that hypothesis.
I think you’ll need to have the detector checked by a relevant service engineer.

Best Regards

[bunnahabhain]

This is also what I thought. It is already packed in the box for repair. As soon as I have it back I will give an update here. Have a nice weekend,
Jörg

[danko]

A slight correction of what I wrote previously: RLS is often used for Rayleigh Light Scattering. This and Tyndall Scattering is not shifted - i.e. is at the same wavelength as the excitation ditto. Raman Scattering is slightly shifted and the name is seldom abbreviated. Most often it’s just called Raman.

Best Regards

And nice weekend to you too.

[bunnahabhain]

This took a while, but now I know more. The detector was sent to Shimadzu for repair. They said that the grating and some other parts in the housing are deformed, most likely by mechanical forces. I don't know how this can happen, but anyway it is a write-off. Sad to lose an instrument...

Cheers,
Jörg