HI all--
some other items to add to the list....
1. Nothing in ash collector. When we've had the quartz wool "eaten
away" by the sample, we'll still see at least some of the tin oxide ash
in the ash collector. In those cases, the molten glass from the filters
seem to have been eaten away too. Rarely (maybe once), we've had the
quartz wool fall out of the ash collector--leaving the ash in the
reactor. With filters, especially large ones, the molten glass tends to
stick to the ash collector. When first playing with filters, we had the
glass ooze out the vent holes on the side--sort of a taffy pull when to
get it out (not good).
2. Erratic 13C and 15N issues
a. Incomplete or erratic combustion of the filters. The molten filters
can entrain some of the materials/gases being analyzed and "burp' them
out later. The upshot is you'll get some random peak coming through in
a later run--sometimes as a nice peak, sometimes as a wavey baseline,
which may or may not interfere with that run's sample or standard
peaks. It may even be released when the next sample is combusted. This
is more of an issue with large filters. We've put in a long wait time
after peak centering to let the system settle down between runs, in case
this is happening.
b. Tiny hairline crack in reactor. Nasty samples, or "fines" in the
packing reagents can eat a hole in the reactor--and a very small leak in
the reactor can be hard to pickup , though flushing Ar along side of
reactor will do it--but remember furnace and ceramic bushing are hot.
Unfortunately, it can seem to mimic other problems.
c. Reactor plugging up/drop times are off (related to nothing in ash
collector). You should notice some other changes too, such as shifts in
peak retention times.
d. Oh yeah, you mentioned filament gone bad. And a dirty source would
also cause problems
3. Chromatographic/peak picking problems. Yeah, we had that strike
again recently. In methods, CO2 detection, advanced mode (bottom
section), set integration start time to some point after the switch has
occurred on your chromatogram, but before the CO2 should start.
take care
gerry
Charlotte Lehmann wrote:
> Anne,
> Have you talked with John Cocca at Costech? When we had a problem
> along these lines a year ago, he had some suggestions for tracking
> down the problem (i.e., is it with the EA rather than the IRMS?).
> I checked the following:
> -- are the positions of the capillaries in the ConfloIII correct?
> -- when does the flash occur? (is there an obstruction in O2 line? in
> the He line?) It should be about 6 or 7 sec.
> -- can you delay the sample a little longer?
> -- checked for leaks?
> -- checked flow rates?
> -- checked the sample start time?
> -- if you have the EAS software, you can run the EA by itself to
> determine if the problem is with the EA or with the ConfloIII and
> IRMS. The N:C should be approx. 0.05.
> -- is the water trap oversaturated? how fresh is your Mg-perchlorate?
> It has a shelf life apparently. Keep it in a dessicator.
> -- can you set the jump earlier?
> -- have you baked out the EA's GC column? Put oven up to 110C for 4
> hours to bake out, with the He diluter in. Then set it back to 60C.
> Does this improve the separartion between the N & C peaks? Do this
> every 3-4 months.
> -- if not improvement in separation from baking out the column, set
> the oven T to 55C; increases run time by 25%.
> -- you can start your He diluter sooner (even before the jump occurs).
> For example if your switch gas is at 205 sec, you can have go at 204
> sec rather than 212 sec.
> -- are your BGD's for Mass 45 positive numbers or negative numbers? If
> the BGD is negative, there are problems with the integration by
> ISODAT. You can change the integration parameters so it starts the
> integration a little later (based on slope).
> -- you can set up a method to just run for CO2 and see if there are
> issues with the IRMS.
> Good luck.
> --
> Charlotte Lehmann
> Assistant in Instruction
> Bates College
> Department of Geology
> 206A Carnegie Science Building
> 44 Campus Avenue
> Lewiston, ME 04240
> Phone: 207-786-6485
> FAX: 207-786-8334
>
>
> Quoting Fname Lname <[log in to unmask]>:
>
>> This is very timely as I have run my first batch of freshwater POM
>> samples with very low N on our Costech ECS4010 (DeltaPlus XP) and found
>> that the first two inserts came out fine after about 100 burns (about
>> 8cm of ash build-up) but when I pulled the third insert out of the
>> combustion tube, there was nothing in it. I am assuming they melted
>> into the combustion tube? I was also finding my 13C values widely
>> erratic. Does this have anything to do with the sample meltdown?
>>
>> A few, maybe related or not questions:
>>
>> My filament readings were bouncing around the last couple of weeks and
>> finally burned out today. Would this have something to do with the
>> erratic 13C (and sometimes 15N)?
>>
>> And a simple chromatographic problem.......I cannot seem to get my peak
>> start of CO2 away from the end of the peak jump settling which I think
>> is sometimes contributing to my varied 13C numbers. My peaks are fairly
>> close, yet fully resolved so my peak jump timing is pretty critical. (I
>> am also going to possibly work on the flow rate too for better
>> separation but I'm already at a low of 90ml/min). I think it's an
>> integration parameter thing but I haven't found the right settings.
>>
>> Suggestions please.
>>
>> Thanks, Anne
>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> Anne M. Cotter
>> U.S. Environmental Protection Agency
>> Mid-Continent Ecology Division
>> 6201 Congdon Blvd
>> Duluth, MN 55804
>> 218-529-5183
>> fax 218-529-5003
>> [log in to unmask]
>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>
>>
>>
>> Shapna Mazumder
>> <[log in to unmask]
>> > To
>> Sent by: Stable [log in to unmask]
>> Isotope cc
>> Geochemistry
>> <ISOGEOCHEM@list Subject
>> .uvm.edu> Re: [ISOGEOCHEM] filter problems
>>
>>
>> 04/04/2008 11:45
>> AM
>>
>>
>> Please respond
>> to
>> Stable Isotope
>> Geochemistry
>> <ISOGEOCHEM@list
>> .uvm.edu>
>>
>>
>>
>>
>>
>>
>> Gerald,
>> We run freshwater POM filters quite regularly on our Costech-Delta V. I
>>
>> have found that if samples have been acidified with HCl, the quartz
>> insert
>> and column takes a beating. A plug of silver wool below the quartz wool
>> in
>> the insert helps. I just run ~25 samples and change the insert to avoid
>>
>> deterioration in the columns. Hope this helps.
>> Shapna
>>
>> Water and Aquatic Sciences Research Program
>> University of Victoria, Department of Biology
>> PO Box 3020, Station CSC,
>> Victoria, BC, Canada V8W 3N5
>> (250) 472-4789
>> [log in to unmask]
>> www.uvic.ca/water
>>
>>
>> ----- Original Message -----
>> From: "Gerard Olack" <[log in to unmask]>
>> To: <[log in to unmask]>
>> Sent: Friday, April 04, 2008 9:28 AM
>> Subject: [ISOGEOCHEM] filter problems
>>
>>
>>> Hi All--
>>>
>>> We've been running some filters samples on a Costech ECS4010--fresh
>> water
>>> phytoplankton. Supposedly these are fresh water, no preservatives,
>> and
>>> filtered through standard glass fiber filters. However we've had
>> problems
>>> with those samples eating holes in the quartz ash collector, the
>> quartz
>>> wool, and sometimes the quartz reactor tube too. In the short term,
>> extra
>>> quartz wool and an extra quartz tube in the ash collector minimizes
>>> damage.
>>>
>>> Has anyone else seen this, and if so, have you been able to trace the
>>> source of the problems? We've done other filter samples without
>> unusual
>>> problems.
>>> take care
>>>
>>> gerry
>>>
>>
>>
>
|
