Cleaning the X-lens assembly
First, there is a video that came with the instrument that shows how to disassemble, reassemble, and clean the lens assembly. That’s all fine, but the recommended cleaning procedure mostly involves fancy yellow and blue sandpaper. That was hard work and took a long time. In desperation, I contacted the PLASMACHEM-L@LISTSERV.SYR.EDU list server. Based on replies kindly sent by Javier Seravalli and Richard Pappas, I have abstracted this sequence of things to try:
- DI water and 2% Citranox or Barkeeper’s Friend in an ultrasonic cleaner.
- 1% nitric acid and 1% hydrochloric acid in an ultrasonic cleaner.
- Scrub with Barkeeper’s Friend, using a watery paste and Q-tips.
- If all else fails, use the fancy sandpaper like the video instructions say.
Always finish by ultrasonic cleaning in DI water, followed by careful drying and reassembly with gloves.
Rotary vacuum pump advice
This is from Dr Ashley Norris, Norris Scientific, PlasmaChem list server, March 2023:
“I’ve had a fair bit of experience running our 7700 on a range of old rotary pumps, and it works to some degree. Baseline pumping requirements for the entire range of Agilent mass specs are mostly the same (chambers are all roughly the same size, aside from the step change between the single and MS-MS instruments, interfaces are roughly the same, etc…) and at risk of being quite wrong, I would argue that the rotary pumps are mostly interchangeable across the line up – assuming the objective is to just get it to “work” and not hit any specific performance figures.
With regards to the comments from Dr Pappas about a pump being “insufficient”, there are two key observations that may be of use:
- Pretty much any rotary pump, provided the system is leak free, and provided it can pump a vacuum of < 5 Pa (a top-notch rotary pump should reach an ultimate vacuum of 0.1 Pa), will be able to back the turbo pump and allow the system to pump down. This can be advantageous if, for example, the system has to sit for some time (> 3 months, say) while you wait for a new pump, and you want to preserve the detector (which lasts longer if stored under vacuum) and keep the chamber clean.
- When you light the plasma and the valve to the interface opens, the rotary pump has to pump the space between the cones as well as back the turbo. This is where the pump rate becomes significant, and this is where you need a suitably sized pump to ensure it can suck on the cone orifice enough to prevent the pressure inside the chamber from climbing too high and shutting down the turbo pump.
What works to your advantage here is that the onboard vacuum system of any Agilent instrument does a really good job of making this process pretty much bullet proof. The only risk from “messing around” with the vacuum system would be additional wear and tear due to frequent restarts of the turbo pump, but really, a modern turbo should be good for it, just be aware that the risk is there. If your rotary pump is not up to scratch then the system will let you know. For example, if you use a pump that you have to hand and try to pump the system, if it can’t rough the chamber in a suitable amount of time the system will stop pumping and throw an error for you: you need a better pump. Likewise, if the system pumps down OK, but you then go to light the plasma and find the operation aborts every time (and you don’t have gas leaks in your sample introduction system) then your rotary pump is unable to pump hard enough on the system and you need a bigger pump.
If you only have smaller pumps around, then something to consider would be doubling up two smaller pumps. You can run the second pump directly from a wall outlet, start it at the same time as the main pump, and you join the pumps in parallel with a vacuum “T” fitting and suitable O-rings and joining clamps (these are normally all KF25, which is a fairly common size).
Something you should be aware of, but not in any rush to mess with, is that the MassHunter engineering mode allows the pumping timeout and pressure thresholds to be adjusted. This would (in theory) allow you to relax the performance values so that the vacuum system can tolerate a sub-standard pump. However, I do not recommend doing this since the risk of damaging the turbo pump is too high, plus there would be implications for your instrument warranty (if at all applicable), performance, and detector lifetime.
And finally, if you do run with a sub-standard pump, and the system pumps and runs “fine”, you should also be aware that there may be some analytical challenges that arise if the chamber pressure is much higher than you are used to. I’m assuming here that it won’t be a problem for you, but of course, as always in this game, it depends.
I wanted to share this information with you in the hope that it may allow you to get your system going while you wait for your replacement pump, or parts, to arrive. But if you are unsure about any of it, the safest option is to stick with the official pump recommended by Agilent and to follow the advice of their support team.”
Update from Brian Dreyer: “Contact provac.com. They sell rebuilt MS40+ and in all likelihood offer repair service. Cheers,” This follows several other PlasmaChem posts that indicate that Agilent has stopped support for the MS40+ pump, and rebuild kits are getting very hard to find. One suggested source is Ideal Vacuum Products. Various PlasmaChem people suggest getting an equivalent spec pump, and some especially favor oil-free, “no-maintenance” pumps. The main specs are a pump rate of 40 cubic meters per hour, an ultimate vacuum of 5×10-2 millibars, max power 1200 watts or less, operating power about 750 watts or less, 220 volts.
Removal of Sb and Sn from distilled HCl
From Malcom Reid, University of Otago, NZ: We find both Sn and Sb challenging to consistently purify to below 20ppt. We use the distillate from our quartz sub-boiling stills to feed Savillex stills and see the gradual improvement (with time) Savillex state in their literature. We haven’t felt the need to double (triple?) distil through the Savillex stills although this may help as Gregory has suggested.
Obviously measuring the reagent blanks can use a lot of your production – we dry down 20 and 40ml and hope to measure a volume consistent concentration on our quadrupole instrument. Our last batch tested gave 13 and 14 ppt Sb (volume corrected) for 20 and 40mL. Sn about 10x higher.
From Alessandro Bragagni, Università degli studi di Firenze, Italy: We also observed high Sn content in distilled HCl. We solve this issue by passing our HCl in anion exchange resin (Bio-Rad AG1X8). It is quite
easy if you have a large column with a funnel (e.g. those from Bio-Rad). Looking on KD for anion resin, this approach should work for Sb as well.
De Laeter J. R. and Jeffery P. M. (1965) The isotopic composition of terrestrial and meteoritic tin. /J. Geophys. Res./ *70*, 2895–2903.
Kirchenbaur M., Heuser A., Bragagni A. and Wombacher F. (2018) Determination of In and Sn Mass Fractions in Sixteen Geological Reference Materials by Isotope Dilution MC-ICP-MS. /Geostandards and
Geoanalytical Research/ *42*, 361–377.
