Torch meltdown cause

Hi,

I've just experienced a torch meltdown on a 5800 ICP-OES with an easy fit torch, while trying to heighten the plasma robustness using the Mg II/Mg I method.

To begin, colleagues of mine ran an intelliquant screening on unknown sample last friday that was very high in Boron and sodium levels and should have been dilluted. We haven't done any calibration for our intelliquant so the actual concentrations are unclear, but since even the weakest emission lines give a strong signal, i'm assuming it was still way too high.

When i got back on monday i noticed a white residue on the upper part of the torch. I tried cleaning the torch with 50% aqua regia for 2 hours. Since that did not work i tried 70% aqua regia, for 2-3 hours. Sadly this also did not remove the residue, i also tried a very diluted laboratory detergent with no result.The always helplful mentoined elsewhere on this forum that the residue might actually be devitrification of the quartz glass and i think this is actually the case. See the pictures below.

Since we did/do not have a spare torch on hand we settled on installing the torch again until our new torch would arrive. When trying to resume my plasma robustness tests earlier this week i noticed that the same exact settings resulted in about half of the Mg II/ Mg I ratio (from 6.47 to around 3.50, both far from robust). Could this be a result of the devitrification or should i look elsewhere to explain this? The settings in question were:

RF power: 1.5 kW
Neb flow: 0.5 L/min
Plasma flow: 10 L/min
Aux flow: 1 L/min

Today i continued with the plasma robustness and when i used the following setttings for about 30 seconds the meltdown occured:

RF power: 1.5 kW
Neb flow: 0.5 L/min
Plasma flow: 8 L/min
Aux flow: 2 L/min

I am trying to find out if the meltdow was caused either solely by the settings or the devitrification, or if it was probably a combination of both. Any help/suggestions are very welcome :)

Thanks in advance!

Parents
  • I understand that you are trying to measure plasma "robustness" as defined by Jean-Michel Mermet as referenced in the paper I will attach here. Plasma robustness is typically measured at RF Powers >1.3kW, nebulizer flows <0.8L/min and a wide diameter injector tube size (>2.0mm). You mentioned that you are using an easy-fit torch, which is standard and has an injector tube diameter of 1.8mm.  I believe the torch integrity you were using for the plasma robustness study had been compromised due to running high B and Na with it, causing devitrification of the quartz. Devitrification weakens the quartz structure. Running the plasma at the maximum power of 1.5kW, very low neb flow of 0.5L/min and plasma flow of 8L/min with an injector tube diameter of 1.8mm meant that the plasma was incredibly hot, and with a compromised torch structure, this all combined to result in a torch melt. You can run the Agilent system with lower plasma gas flows, as referenced in the 2nd paper I will attach here. Typical "robust" conditions can be achieved by using an RF power of 1.2-1.5kW, neb flow around 0.75L/min, plasma flow 12L/min and Aux flow 1L/min with the high solids torch which has an injector tube diameter of 2.4mm. 

  • Thank you very much Tina, you have truely helped me a great deal troughout my internship!

    We are indeed using the standard torch with 1.8 mm injector tube. So is the replacement we have ordered. The high B and Na sample was actually a random DIY project of a colleague and not part of the method i am trying to develop. The method in question has a concentration range of 0.08-10 ppm (when taking into account dillutions). So i think/hope the standard torch will suffice in terms of needing robust conditions 

    I will definitely study the papers once uploaded and resume development once the new torch arrives, with the newfound knowledge of how not to melt a torch :D

Reply
  • Thank you very much Tina, you have truely helped me a great deal troughout my internship!

    We are indeed using the standard torch with 1.8 mm injector tube. So is the replacement we have ordered. The high B and Na sample was actually a random DIY project of a colleague and not part of the method i am trying to develop. The method in question has a concentration range of 0.08-10 ppm (when taking into account dillutions). So i think/hope the standard torch will suffice in terms of needing robust conditions 

    I will definitely study the papers once uploaded and resume development once the new torch arrives, with the newfound knowledge of how not to melt a torch :D

Children
No Data
Was this helpful?