Answer Liz’s blog questions here.
Answers
-
Hi Liz,
My two bobs worth!
1) Plate counts... if we use methanol the plates will be lower and that looks bad.
Qualification ...
a) System Test...as we only ~ 10% acetonitrile for RSD caffeine test its only a small amount so won't help much. (should use dial a mix instead of engineers premixing a large volume.)
b) Detector linearity and AQT stuff ...Could save quite a bit here however as UV cut-off is higher for methanol could alter outcome.... maybe use a shorter column or even guard col to minimise solvent usage.
Bottom line is we can still show the market that the total volume of solvent used for validation..Maybe we state that Waters can validate your system with only ... ml of acetonitrile. (calculation based on dial a mix with the acetonitrile for AQT and the use of MeOH and MeOH/H2O as the strong and weak needle wash) will be quite small per system.
My final opinion is as above: Dial a mix only for CH3CN and MeOH is used for needle washes.
Brian
0 -
Here are some links that may be of use "Acetonitrile Shortages: "Recommendations for Reporting Changes in Analytical Procedures".
(1) http://www.fda.gov/cder/ops/Acetonitrile_Shortages_FINAL_updated.pdf
(2)Access from: http://www.fda.gov/cder/ops/default.htm
Liz
0 -
I don't think Waters should stop using acetonitrile, and agree with Outback about marketing as being able to do the same with less. While the company I work for is converting some method from MeCN to MeOH, I think a better approach is trying to reduce the MeCN usage, not eliminate it. Economic 101 teaches the law of supply and demand. If everyone stops demanding MeCN in years to come, there will be a true shortage because no one will be supplying it, and the cost of MeOH will skyrocket do to the high demand.
As far as the Waters UPLC qualification, I was suprised to see that it has a 2 minute runtime for a single component solution. You market high-speed chromatography and then take 2 minutes per injection for the qualification. I can do that with my HPLC qualification. I think the runtime should be down around 30 sec per injecction.
0