Print

Print


Hi Arnold,

I don't know if this is your problem (and we may have a different set up), but I just spent half the summer trying to figure out why our EA reduction tubes were getting used up so fast and thinking it was bad copper or poor packing. Turns out it had nothing to do with the reduction tube. It was the needle valve that controlled the O2 dosing. The O2 dosing was creeping upward over several months and I didn't even realize it for a while. Anyway, it was a bad needle valve causing the problem.

So...for whatever its worth. Hope you figure out your problem soon.

Regards,
Carol

On Thu, Sep 12, 2019 at 11:12 AM Arnold,Thomas E <[log in to unmask]> wrote:
All, 

We run our reduction reactor at 650C. The copper was just changed (it was the first thing we did after this problem developed), but it may not have been packed it properly. As Jason suggested (and other thereafter) perhaps I should try repacking it remove any preferential flow paths that may have developed form improper packing. 

Will try that again. 



On Sep 12, 2019, at 11:50 AM, Hayat Bennadji <[log in to unmask]> wrote:

EXTERNAL EMAIL: Exercise caution with links and attachments.



I agree with Loreta, it's time to change the reactor 
Hayat
Louisiana State University


From: Stable Isotope Geochemistry <[log in to unmask]> on behalf of Loreto Morales Rojas <[log in to unmask]>
Sent: Thursday, September 12, 2019 10:30 AM
To: [log in to unmask] <[log in to unmask]>
Subject: Re: [ISOGEOCHEM] Mass 30 instability on Delta V
 
TEA

How many analyzes have you done, copper serves me for 200 analyzes (standard and samples), when saturated a nitrogen peak begins to appear.

maybe it's copper



 

 


Loreto Morales Rojas




De: Stable Isotope Geochemistry <[log in to unmask]> en nombre de Arnold,Thomas E <[log in to unmask]>
Enviado: jueves, 12 de septiembre de 2019 11:23
Para: [log in to unmask] <[log in to unmask]>
Asunto: Re: [ISOGEOCHEM] Mass 30 instability on Delta V
 
As a follow up -

I baked out the EA oven, closed the valve to the IRMS and baked that out for 24hrs. We still have the same issue with mass 30 increasing within a single on/off run and across multiple on/off runs. Today I will switch out our He tank for a new one, and change the water trap (even though it show no sign on water consumption).

Not sure if there are any other suggestions beyond what has already been put forth. The standard deviations look fine after the 4th (or so) N2 on/off run, so maybe I am making too much of an issue out of this, but it is frustrating that such an issue would suddenly appear and be difficult to fix.

Thanks for the help,
-TEA

> On Sep 10, 2019, at 2:08 PM, Harlow, Benjamin <[log in to unmask]> wrote:
>
>
> EXTERNAL EMAIL: Exercise caution with links and attachments.
>
>
>
>
> Tom, it works nice to install a 3 way valve at the exit of the EA. That way you can quickly vent to the room (or a flow meter) instead of the conflo. You can also use this valve to block off all flow out of the EA for gas savings or a leak down test.
>
> I would turn off the HF on your conflo and vent the EA whenever you do maintenance such as chemical changes and GC bake out. Then restore the EA to work mode and let it get up to flow and temperature for an hour or so before switching the 3 way back to the conflo. This allows any contaminants introduced during the maintenance to bake off and flow somewhere other than the IRMS.
>
> Have your helium or O2 tanks been changed recently?
>
> Cheers, Ben
>
> -----Original Message-----
> From: Stable Isotope Geochemistry <[log in to unmask]> On Behalf Of Arnold,Thomas E
> Sent: Tuesday, September 10, 2019 10:33 AM
> To: [log in to unmask]
> Subject: Re: [ISOGEOCHEM] Mass 30 instability on Delta V
>
> Hi Ben,
>
> We have been using the same UHP tank of N2 for some time now (predating the issue at hand).
>
> I have not baked the GC oven in the EA before. When you say vent the EA away from the conflo, can I do this by simply turning off (in insodat) the high flow capillary that is going into the conflo? If I remove the line in front of our Costech EA that is carrying the gas from the machine to the conflo, the N2 backgrounds skyrocket (as expected). Thus, I’m not sure which is the best way to vent it away from the conflo.
>
> Thanks,
> TEA
>
>
>> On Sep 10, 2019, at 12:37 PM, Harlow, Benjamin <[log in to unmask]> wrote:
>>
>>
>> EXTERNAL EMAIL: Exercise caution with links and attachments.
>>
>>
>>
>>
>> Hi Tom,
>>
>> What are your backgrounds? You could have a bad tank as well. Have any been changed recently? If 30 is increasing throughout your EA sequences it could be that the water scrubbing chemicals are not good, or there is a pathway through the trap. When was the last time the GC was baked? (Do this with the EA vented away from conflo)
>>
>> What I meant about instability is that is usually a consequence of excessive water. If you have good N2 on-off stability then I would think the water in your system is manageable. Usually, when we see a strong stair step effect in m/z 30 N2 is not reproducible to 0.06 per mil or better. If N2 is continually flowing into the source you will see 30 rise. But with time it should return down.
>>
>> It is my understanding that if heaters are off, water is still likely hanging around the source but you may not see a large m/z 18. Compared to leaving heaters on all the time, the water background will be much less.
>>
>> If your system doesn't seem stable, it may be worth it to close the inlet, turn on all heaters and let the system pump for a day or two then try again.
>>
>> Ben
>>
>> -----Original Message-----
>> From: Stable Isotope Geochemistry <[log in to unmask]> On Behalf Of Arnold,Thomas E
>> Sent: Tuesday, September 10, 2019 6:52 AM
>> To: [log in to unmask]
>> Subject: Re: [ISOGEOCHEM] Mass 30 instability on Delta V
>>
>> Ben (and others),
>>
>> Thanks for the quick response, we also assumed water in the system. As a follow-up: why would the m/z 30 signal increase throughout a single run and across multiple runs? You mention instability as a reason. What do you specifically mean by this? That the EA is not properly sealed? It’s also strange because we see this issue when the EA/high flow is disconnected from the IRMS and the on/off runs are through the conflo.
>>
>> Thanks again,
>> TEA
>>
>>> On Sep 9, 2019, at 5:39 PM, Harlow, Benjamin <[log in to unmask]> wrote:
>>>
>>> Hi Tom,
>>>
>>> It is likely elevated water in your system. N from your N2 pulses combines with O from H2O to give you m/z 30. More time will help. We typically run with source and inlet valve heaters on all the time to keep water out. Sometimes a heat gun or hair dryer can be beneficial to drive water off other components. Monitor m/z 18 while heating various components. Watch for the signal to climb, eventually drop and stabilize.
>>>
>>> Every time you replace reagents or open up the EA more m/z 18 can get in your system. So venting the EA or allowing ample stability time is needed. If water is really bad then close the IRMS inlet and allow it to pump down independently.
>>>
>>> Usually if m/z 30 does the stair step on std on offs it is not a stable condition for us.
>>>
>>> Regards, Ben
>>>
>>> -----Original Message-----
>>> From: Stable Isotope Geochemistry <[log in to unmask]> On Behalf Of Arnold,Thomas E
>>> Sent: Monday, September 09, 2019 2:04 PM
>>> To: [log in to unmask]
>>> Subject: [ISOGEOCHEM] Mass 30 instability on Delta V
>>>
>>> To the great minds of the list serve,
>>>
>>> Our system: Delta V coupled to a costech EA via conflo IV
>>>
>>> Our issue: Mass 30 is drifting to higher and higher voltages across multiple N2 on/off runs — more than a 1V increase from our first to fourth on/off. I can send pictures to individual emails if needed.
>>>
>>> What we've tried thus far: 1)Replaced: the seals on the EA reactors, the copper wires in the reduction reactor, the water trap. 2) We turned on the heater to bake out the source and inlet. I believe this heater operates at a temperature of 50-80C. After the bake out, the amplitude of mass 30 was reduced, but the voltage drift was still there. 3) The problem presents itself in both EA mode and in the conflo only mode.
>>>
>>> Any suggestions on what to do? Should this even be considered a problem as our standard deviation looks fine after the 3rd series of on/off runs?
>>> Thanks,
>>> Elliott Arnold
>>> Pitt Geology
>>



--
Carol S. Loopstra
Soil, Water, and Climate Department
University of Minnesota - St. Paul