TOC runs

[WISPHARM]

We just got our new TOC analyzer, and I am writing the SOP's. What is industry standard for system suitability and standards. From what I hear many folks are running system suits with every batch. If we were to do it once a week, could standards be used for the batches in between the system suits to confirm suitability. If so, just before the run, or before and after? Blanks?

[sphereman]

I think it depends on your particular instrument and the risks your company is willing to take.

We have a GE Sievers 900 series that uses UV/Persulate and Selective membrane conductivity that we calibrate annually per the manufacturer's instructions. We currently run system suitability once a month and after any maintenance. When we installed our instrument, we did run it every week in order to gather the data to justify decreasing the frequency.

A contract lab I used to do business with had a Shimadzu the used combustion and NDIR detection. They calibrated the instrument and ran system suitablity every day. I am not familiar with the instrument to know if it was necessary or if they were being over careful.

[WISPHARM]

We are also running a 900 Sievers. Do you run any control standards between your system suitability? Blanks?

[sphereman]

We do not run anything other than system suitability on a monthly basis. The only other standards we run are the calibration standards (KHP and sodium bicarb) annnually. Also, every six months we run the calibration verification standards (Sucrose and sodium bicarb). The only blanks we run are either part of system suitability or the ocassional blank to flush the system between known high TOC samples.

Most of our samples are WFI, Purified Water or Clean Steam from our water system so we are not putting a lot of dirty samples on the instrument.

[WISPHARM]

Thanks for the input sphereman, Can anyone else with a Sievers 900 comment on the use of standards and blanks between system suitability runs?

[bluejay]

We run an older Tekmar Phoenix (UV Persulfate) for process control. We calibrate infrequently (~every few months) but run a calibration verification of blank and a standard each day we run.

EPA method 415.3 has a lot of information about TOC analysis. You can download it from the following website:

https://www.nemi.gov/apex/f?p=237:48:11 ... 23167:::48

This method looks at TOC levels in drinking water which tend to be pretty low. If you're looking at higher levels like we are, a certain amount of their requirements are overkill but good to know about when things don't run as they should.

[oscarBAL]

We use Tekmar TOC; The SS test for this is done by runing 500ppb of Sucrose and 500 ppb bezoquinone and this comes form USP; recovery in both cases is betwen 90 and 110%. this is for pharmaceutical grade water.

[WISPHARM]

OscarBAL,

Are you running SS everyday? If not do you perform some sort of check standard in between, blanks?

[oscarBAL]

Hi Wishpharm; yes we run SS everytime but we not run everyday, and also we run Chck TSD and Blank always.

[WISPHARM]

If you could be so kind oscarBAL, can you get me an idea of what a batch of TOC runs looks like? i.e system suitability,followed by 3 samples, blank, check standard, repeat 3 samples, blank, check standard. I wonder if you could run system suitability, run all your samples then rerun system suitability? would having the ss vials reinjected be subject to increased TOC over time say a few hours?

[sphereman]

What types of water are you testing, is it USP grade WFI, Purified Water and Clean steam only or are you running samples expected to be greater than 500 ppb? If you are just running samples that are expected to be less than the limit response, then with the instrument you are using there should not be an issue with higher TOC in a re-run system suitability.

[WISPHARM]

Purified water for cleaning validation at OTC pharma manufacturer. We are typically seeing 50-200 ppb levels.

[oscarBAL]

Hi Wishpharm; sorry for the delay;
We usually run; Cleaning, Calibration Curve; blank, SS, Check Std, Samples ans Check std at the end (you have to define the bracketing of the Check STD).
This procedure is very stable you can re run SS at the end but this not so important as the Chek STD; I do not know how much it would have to increase your las SS being above the stistical variability.

[janzalone]

oscarBAL

Do you purchase commercially prepared standards or are you making them from stcok solutions? The GE 900 TOC is our second TOC in the lab and we are trying to run standards by making 100X dilutions from a 50 PPM commercially prepared stock solution. I am running a reagent blank first using my dilution water and labeling the standard as a standard in the software. So far I've tried three different stock standards and no matter what I've done I always get 110% to 120% recovery. My blank values are good with usually 20 -40 PPB TOC. We need to run bracketing standards for our cleaning verification samples. Has anyone tried diluting stock standards to make 500 PPB check standards? GE has been absolutely no help. Their only contribution is to let me know that they don't support diluting standards. What happens if you have to dilute a sample? Will that also be biased high? There 500 PPB standards gives good results so the instrument is properly calibrated. We are using all class A volumetric pipettes and flasks to make a simple 100X dilution. What is going on? I think it's too much of a risk to run standards once a month and then find out you have a failing result which cast a dark cloud over the previous months worth of cleaning data.

Thanks

John

[oscarBAL]

Janzalone; I started in a new job two months ago but let me see if I can help...
We used to run SS and Check Standard for any sample set. our check Standard (500ppb) was diluted by our self from a working solution got it from ERA. We were getting high recoveries for Check Standard then we realized there something in the calculation formula the way the software subtract the blank that always added aroun 13% higher result specially for lower ppb values.
I wasn't using GE but check formula could be a good Ideal put attention to how they substract the blank and what they use as y-intercept.