HI All--
A couple of other items of note:
With the Zero-Blank autosampler (no isolation valve), we don't use the
"purge" line on Costech's ECS4010, so we put a pigtail on it (very low
flow) instead of just capping it. That eliminated an air leak.
Also, the o-ring between the carousel and the connector to the reactor
needs to be cleaned/replaced periodically.
take care
gerry
p.s. That said, we occasionally have samples that eat away at the
quartz--putting a hole in the reactor. We may not see high Ar (leak too
large), but we'll see delays in the gas elution (N2 and CO2), but the O2
pulse hits the reactor almost immediately--and He tanks will empty quickly.
Stephen Taylor wrote:
> Hi Bruce,
>
> Interested to hear someone else has spent a 'heap-o-time' argon leak
> checking a Costech ...
>
> We have a similar set up (Costech EA + Conflo III and Delta+XL) and
> observed similar leak problems immediately after installation. I then
> spent the better part of a week searching for a leak (and slowly
> losing what little sanity I have left ...) because of the same high N2
> backgrounds that you describe.
>
> Fortunately I have two peripherals (TCEA and EA) on the same mass spec
> which I can switch between using a Valco 4 port valve and which both
> use the same UHP helium supply. With the TCEA in line mass (40) bkgrd
> was/is ~7mV. Therefore, I discounted a bad tank of helium (as Ben
> pointed out can occur), and I started argon leak testing in earnest.
>
> Initially the mass (40) background of the Costech was >150mV. After
> bypassing everything ... the GC column, autosampler and reactors, h2o
> trap and TCD, the mass 40 was still >150mV. I then argon leaked
> checked every fitting inside the EA and eventually found two leaks: 1)
> the TCD itself and 2) a tee fitting between solenoids # 6 and #7. Both
> of these were replaced/repaired by the Costech rep here in Canada
> under warrantee. This reduced the argon background down to ~45mV. At
> this point I started putting components back in line. With everything
> back in line mass 40 remains ~45mV and mass (28) ~12mv, mass (29) ~9mV
> and mass (30) ~50mV.
>
> Regarding the O2 supply. The oxygen portion of the Costech is isolated
> from the rest of the system except when the O2 pulse is requested, so
> it can't be source of the elevated backgrounds. It can though affect
> your blanks if your O2 tanks is either contaminated or the
> pressure/flow isn't high enough. Check that you've got at least 1 bar
> of O2 pressure and a flow of 25 ml/min at the O2 vent. If the O2
> pressure is too low room air can back diffuse into your loops.
> For-what-its-worth I use 4.3 purity O2.
>
> So at the end of the day I am running with the backgrounds noted
> above, which are in agreement with what Ben, Gerard and Ali report. I
> know of one other very small leak inside the EA but the fitting is
> very hard to get to and I've decided to let it be.
>
> At these background levels a N2 blank correction is the only way to
> achieve acceptable data and I make sure to run several empty tin cups
> before samples to get an accurate blank size. FYI: a 3.2 x 4mm tin cup
> gives me a ~0.6 [Vs] peak area.
>
> Cheers,
> Steve
>
>
>
>
>
> At 12:21 PM 12/5/2008, you wrote:
>> We have a Costech EA, a Conflo III, and a Thermo Delta V Advantage
>> I.R.M.S. Currently using continuous flow for C13 and N15
>> determination on
>> sediment core samples. C13 analysis is very good. N2 is the issue,
>> our N2
>> background is typically in the 26mV 28mV range (on N28, cup 2), we
>> have
>> tried various grades of O2 combustion gas, from grade 4.5, 5.0 (< 5
>> ppm N2,
>> batch tested), and now grade 6.0 (supposedly < 0.2ppm N2, also batch
>> tested), with no change at all in the N2 background. Note: this
>> background
>> is present in blanks, (blank = no sample whatsoever, O2 injected,
>> with auto
>> sampler inhibited) as well as “empty” tin capsules, and of course
>> samples and
>> standards.
>>
>> We have (we believe) thoroughly leak checked the system with argon leak
>> check techniques. This identified a small leak at the water trap in
>> the EA,
>> which we promptly corrected. No other leaks were identified, typical
>> Ar40 on
>> cup 3 reads < 70 mV (it has been lower than 60 mV). All gas fittings,
>> regulators, and lines from all gas cylinders were also checked, again
>> no leaks
>> were identified. We plumbed He into the O2 line to see if N2 background
>> would change with the He in the combustion line. It did, N28
>> background was
>> 2mV 3mV on cup 2 with He in place of O2.
>> Question 1: Is our N2 background (~28mV) typical? If not, what is?
>>
>> Question 2: What grade O2 should we use (and is grade 6.0 the
>> “best”?)?
>> Manual says grade 5.0 or better, we have seen no difference in N2
>> background between 4.5, 5.0, and 6.0. Should I request from my gas
>> supplier
>> a specific O2 cylinder which has been tested and certified to <
>> 0.2ppm N2,
>> not just batch tested? Or, should I try a different supplier? We have
>> had no
>> other issues with our current supplier. I realize that cert’s can
>> be wrong, and
>> our O2 cylinder grades might not really be what they are supposed to
>> be, but
>> are we looking in the right direction, or barking up the wrong tree
>> (as Mom
>> used to say!).
>> Question 3: What else could give us our N2 background?
>>
>> Any help or direction would be greatly appreciated.
>>
>> Thank you in advance.
>>
>> Bruce Wegter
>> Sciences Instrumentation Technician
>> Geosciences Department
>> Hamilton College Clinton, New York 13323
>> e-mail: [log in to unmask]
>
> Stephen Taylor
> Lab Manager, Isotope Science Lab,
> University of Calgary, Physics & Astronomy,
> 2500 University Dr. N.W.
> Calgary, Alberta,
> Canada, T2N 1N4
> office: 403 220-8268, Labs: 403 220-8273 or 220-6813
> (f) 403 220 7773
> [log in to unmask]
> www.phas.ucalgary.ca/isl
|