Dear Pier,
Like I said, it was not easy to tell the whole story! I am sorry to use the
word “boiling” to confuse you. You are right, you can not see the acid boiling
at 120C. Since there is a magnet in the acid bath, it spins fast, make the
acid look like “boiling”. My point is because the reaction temperature is
high; the high acid vapor pressure may pose an issue. Let me bring you more
details: We use >100% acid, and it is cooked overnight at 120 degree C with
continuous Helium flow vent, so (I assume) there is no water left in the acid,
and we pump out between samples, the left over water  is supposed to be pumped
out too. I am not familiar with your question about CaO. Since our data come
out very well, repeatedly. The “CaO” thing may not pose an issue in this
system.

Zheng-Hua



>===== Original Message From Stable Isotope Geochemistry
<[log in to unmask]> =====
>Dear Zheng-Hua,
>
>Thanks for giving more detail on the changes in method. It is highly
>appreciated.
>However, from your statement that your acid boils at 120 degree C, I can
>conclude you do not use strong phosphoric acid of higher density (say the
>+100 percent acid), as actually originally used for the calibration of the
>system (there was discussion on this item years ago, in the early days of
>isogeochem list - see the archives!). If you use +98 percent acid the
>boiling point is 158 degree C (reference: Aldrich guide). What means that
>you are boiling of water if the acid boils (the acid does not really boil
>itself, but is just releasing boiling water as far as I understand - I do
>not think (hope) you will have any phosphoric acid vapour; fortunately not
>because then you were really in problems with risk of attack of your walls
>of the system or other nasty effects)... and what is the influence on the
>CO2 from your reaction? How is the control on exchange between this water
>and CO2, the O in the acid, the O in the CaO left over from reaction... to
>give some questions! This will introduce quite some more uncertainty in your
>method.
>No criticism to you - you do as many others do in applying this system.
>I just want to bring forward, as triggered by your posting, some of the
>worries I have... This kind of matters must have their effect on the
>results, although in general we see these very nice small precisions
>presented with the results. How comes...???
>
>Best regards,
>Pier.
>
>
>> Dear Pier and others,
>>
>> Well, it is a little bit hard to describe all details about the
modifications
>> in an email. It would be easier to tell the story with some diagrams.  Here
I
>> can tell you the main things I did with the original CarboFlo:
>>
>> 1) I spared a valve on the CarboFlo to control the Helium gas, so that I
can
>> pump out the whole system between samples (to eliminate the memory effect).
>> 2) Instead, I use a valve as a vent on DeltaS to replace the original vent.
>> 3) The original LN2 trap (for CO2) was replaced by a two-way stainless
steel
>> tube filled with silver wood to increase the CO2 trapped surface.
>> 4) The Carousel is sealed, so that all system can be pump out to the
>> background quickly.
>> 5) Since the original Carousel has some issues, such as, when it spans,
some
>> samples can be spread out of the silver tin cap,this may mess up other
>> samples. A single-ring cap was made to solve this problem.
>> 6) Since the hardware structure was modified, I had to change the software
for
>> the large sample and micro-volume measurement accordingly, such as, the
>> pumping out time, Slurry and LN2 trap time, pneumatic valve control, etc.
>>
>>  Since we use 120 degree C reaction temperature, this high temperature
>> causes the acid boiling. To avoid any acid get into the system, we always
keep
>> the Slurry on. I haven't found this is an issue for this modification so
far.
>>
>> Cheers,
>>
>> Zheng-Hua
>>
>>
>>
>>
>>
>>
>>
>>> ===== Original Message From Stable Isotope Geochemistry
>> <[log in to unmask]> =====
>>> Dear Zheng-Hua,
>>>
>>> What were those modifications in hardware and software that reduced your
>>> memory-effect? This way it does not tell us very much why and how you
>>> obtained this result. I am interested (and probably some others) to see a
>>> bit more details...
>>>
>>> Regards,
>>> Pier.
>>>
>>> ****************************************************************
>>> Delta Isotopes Consultancy
>>>
>>> Dr. Pier A. de Groot
>>> Pastoor Moorkensstraat 16
>>> 2400 Mol - Achterbos
>>> Belgium
>>> Tel. +32 (0)14 326 205
>>> e-mail: [log in to unmask]
>>>
>>> Visit my WEB-site about my ³Handbook of Stable Isotope Analytical
>>> Techniques², including pdf files of an order form and a flyer for
>>> downloading, at:
>>> http://users.pandora.be/handbook/index.html
>>> last update: October 1, 2004
>>> Volume I is now available. Volume II is expected to be available medio
2005.
>>> ****************************************************************
>>>
>>>
>>>
>>>
>>> Dear Bryan,
>>> You might be interested in knowing something else in addition to GasBench
>> II
>>> and GV's MulitFlow Geo.
>>> Our lab is equipped with a Finnigan CarboFlo device to run large carbonate
>>> samples (2-4 mg). As you probably know, this device has some memory-effect
>>> issues, particularly when running small samples. In order to minimize this
>>> effect, I modified this device a bit both for the hardware and therefore
>>> software. Now this device runs very well. Here are some numbers you might
>> want
>>> to know:
>>> ---------------------------------------------
>>> Auto carousel: 39 samples
>>> Standard Dev.: d13C=0.03 permil, d18O=0.1 permil
>>> Experiment time: ~30 hours
>>> Acid volume: ~10 ml
>>> Reaction temp: 120 degree C
>>> ----------------------------------------------
>>> To check the memory effect on the modified CarboFlo, I had used four
>> carbonate
>>> standard samples (NBS-18, CHCC, ANU P3, ANU M2), which d13C values range
>> from
>>> +2.24 to -10.68 permil, I found the memory-effect is tremendously reduced
>> to
>>> almost none at this d13C range, as you can see from d13C standard
deviation
>>> above.
>>>
>>> Cheers,
>>>
>>> Zheng-Hua
>>>
>>>
>>>
>>>> ===== Original Message From Stable Isotope Geochemistry
>>> <[log in to unmask]> =====
>>>> Dear All,
>>>>
>>>>       We currently have VG-micromass-GV "optima" IRMS at our lab which
>>>> we have been using for bulk carbon and nitrogen stable isotopes for the
>>>> past 11 years. For the most part, the instrument has operated
>>>> exceptionally well, with a few minor exceptions. One thing I don't like
>>>> is the operating software; I'm not a big lover of OS2, as I wish we had
>>>> something windows based, as it would certainly make data manipulation
>>>> easier.
>>>>
>>>> Currently we are looking into acquiring a new instrument and I have
>>>> researched both the DeltaPlus and GV's Isoprime. From those of you who
>>>> have responded via the ISOGEOCHEM site, there seems to be two camps out
>>>> there: those who prefer Finnigan instruments and those who prefer GV
>>>> instruments. It would seem preference may be a function of what
>>>> instruments people have worked on in the past and feel comfortable with,
>>>> or what was previously available at a particular lab. Certainly there
>>>> are features that are different between the two companies, for example
>>>> the operating software: Isodat NT vs MassLynx, but I don't know how
>>>> other hardware components(mass spectrometer, EA, GC, and respective
>>>> carbonate devices) compare.
>>>>
>>>> I am particularly interested in the ability to measure carbonates via
>>>> continuous flow and have heard that ThermoFinnigan's GasBench II
>>>> operates well for these measurements. I have not heard to much regarding
>>>> GV's MulitFlow Geo, and am wondering how it stacks up to the other
>>>> carbonate devices (i.e. Kiel). I know precision drops with the gas
>>>> benches compared to a Kiel system, but we are not doing paleoclimate
>>>> work and a drop in precision isn't critical, especially given our
>>>> research questions. I think GV and ThermoFinnigan both state that
>>>> precision is about .1permil for carbon and oxygen isotopes via CF. Have
>>>> any of you run carbonates using the MultiFlow Geo? Can large carbonate
>>>> samples(500ug - 1500ug) be run on the MultiFlow? I know ThermoFinnigan's
>>>> Gas Bench has an autodiluter, capable of reducing the signal to the mass
>>>> spec, but do not know if GV's MultiFlow has this capability?  I suspect
>>>> it does, but would be interested to hear any feedback on the
>>>> subject........
>>>>
>>>>                              Regards-
>>>>                              Bryan Taplin
>>>>
>>>> Bryan Taplin
>>>> Environmental Scientist
>>>> US Environmental Protection Agency
>>>> NHEERL- Atlantic Ecology Division
>>>> 27 Tarzwell Drive
>>>> Narragansett, RI 02882
>>>> Tel. (401) 782-9607
>>>> Fax (401) 782-3030
>>>
>>> Dr Zheng-Hua Li
>>> Senior Research Associate
>>> 306 G&G Building
>>> Dept of Earth and Planetary Sciences
>>> University of Tennessee
>>> Knoxville, TN 37996
>>> USA
>>> Tel: 1-865-974-9622
>>> Fax: 1-865-974-2368
>>> http://web.utk.edu/~zli
>>>
>>>
>>
>>