Glenn,

Bring this evidence of Kr to your supplier and see what they can do to 
reduce the variability.  Band-aids include using a correction equation 
and/or calibrating and running samples off the same Ar bottle. 

Duane Miller
Analytical Lab Manager
APL Engineered Materials, Inc.
Urbana, IL 61802
[log in to unmask]

"PLASMACHEM-L: Analytical Chem.(ICP's, DCP's, MIP's)." 
<[log in to unmask]> wrote on 09/29/2011 11:23:16 AM:

> Hi Duane.  I got these counts (0.1s integration) from two different 
> tanks from two different lots aspirating 18.3 megaOhm water.
> 
>    Tank1      Tank2
> 77   3.4      3.9
> 78   613      628
> 79   129      132
> 81   222      223
> 82   69      593
> 83   70      597
> 84   356      3036
> 85   9.2      18
> 86   113      954
> 87   12      14 
> 
> Based on what Ed mentioned and the counts at 84 and 86 it would seem
> that there is some Kr variability at a minimum.
> 
> Glenn
> 
> 
> Glenn Hyland
> Public Health-Madison & Dane County
> 210 MLK Jr Blvd, Room 507
> Madison, WI 53703
> 608-243-0348
> 
> 
> 
> 
> -----Original Message-----
> From: PLASMACHEM-L: Analytical Chem.(ICP's, DCP's, MIP's). [mailto:
> [log in to unmask]] On Behalf Of Duane Miller
> Sent: Thursday, September 29, 2011 9:57 AM
> To: [log in to unmask]
> Subject: Re: Mass 82
> 
> Glen,
> 
> A mass spectrum from 77 to 87 (skip 80) would be helpfull in identifying 

> the species at mass 82 in each of your cylinders.
> 
> Duane Miller
> Analytical Lab Manager
> APL Engineered Materials, Inc.
> Urbana, IL 61802
> [log in to unmask]
> 
> "PLASMACHEM-L: Analytical Chem.(ICP's, DCP's, MIP's)." 
> <[log in to unmask]> wrote on 09/29/2011 09:25:43 AM:
> 
> >   We are running an Agilent 7500ce and have recently had problems 
> > with Se at 82.  We run EPA 200.8 for most samples so we must use 
> > mass 82 and no collision/reaction gases.  We have a two cylinder 
> > automatic switchover manifold and use grade 5 Ar as we are unable to
> > get liquid Ar delivered to our facility.  The problem first 
> > presented itself as calibration blanks run every 10 samples being 
> > quite positive or negative out of range (-idl<concentration<+idl) 
> > during a long run.  I was able to determine this happened after a 
> > tank switch over.  To try and accommodate this I warmed up and tuned
> > on one tank and switched to a new tank just prior to calibration. 
> > This worked for some time.  Now I am experiencing a drift in the 82 
> > signal during a tank too.  Yesterday, I began to monitor counts at 
> > 82 during warm up.  I had 73 at mass 82 on the first tank.  I 
> > manually switched to another tank and the counts went up to 1400.  I
> > put a third tank on and manually switched to it and the counts were 
> > about 800.  The change in signal was instantaneous with the 
> > changeover each time.  All 3 of these tanks were from different lots.
> >   I have seen many discussions here about Kr being present and 
> > accumulating in dewars but I couldn't find any discussions about Kr 
> > in cylinders.  I talked with the supplier and they felt Kr wouldnt 
> > be an issue because it wouldn't be liquefied in the tank at that 
> > pressure so it wouldn't accumulate in the tanks nor would the tank 
> > be stratified either.  The supplier also felt since Kr is present at
> > low levels in the atmosphere that the proportion in their gases 
> > would end up being relatively constant across lots.  In looking 
> > around the only other instance of problems I have found at 82 was 
> > Br81H+.  I am not sure if Br is checked in grading the gas or not 
> > and haven't taken it up with the supplier yet.  Has anyone else 
> > experienced this sort of thing or have any suggestions?  Thanks in 
> advance.
> > 
> > Glenn
> > 
> > 
> > Glenn Hyland
> > Public Health-Madison & Dane County
> > 210 MLK Jr Blvd, Room 507
> > Madison, WI 53703
> > 608-243-0348