[Beowulf] Re: Cooler room or cooler servers?

[Joshua Halpern]

The fluorinerts have huge global warming potentials.  3M is replacing 
them with a line called Novec engineered chemicals.  Both were designed 
for cleaning and cooling cutting applications

Among the amusing things you could do with fluorinerts was dip working 
electronics into them and they would still work.  I have not tried this 
with the Novecs

http://products3.3m.com/catalog/us/en001/oil_gas/specialty_materials/node_2CT5BPQWKBge/root_GST1T4S9TCgv/vroot_G1F6DNZDBVge/theme_us_oilgas_3_0/command_AbcPageHandler/output_html
or http://tinyurl.com/4uup6

Josh Halpern

David Mathog wrote:

> 
> On Mon, 11 Apr 2005 joelja at darkwing.uoregon.edu wrote:
> 
> 
>>On Mon, 11 Apr 2005 award at andorra.ad wrote:
>>
>>
>>>Some time ago, I was thinking of using an inert liquid such as 3M's
>>>Fluorinert  for refrigeration.
> 
> 
>>also you'd be helping keep the ozone hole large for the next 50 years...
>>
> 
> 
> According to the product information:
> 
> Fluorinert liquid FC-40 has zero ozone depletion potential. The
> material is exempt from the U.S. EPA and most State definitions
> of a volatile organic compound (VOC), and does not contribute to
> ground-level smog formation.
> 
> Fluorinert liquid FC-40, a perfluorocarbon (PFC), has a high global
> warming potential and a long atmospheric lifetime. As such, it
> should be carefully managed so as to minimize emissions.
> 
> 3M recommends that users of FC-40 liquid further limit emissions by
> employing good conservation practices, and by implementing
> recovery, recycling and/or proper disposal procedures. 3M offers
> a program for used fluid return.
> 
> The stuff is nonreactive but that doesn't mean it's completely
> safe.  I recall studies of freon type liquids as a liquid
> breathing medium for mice.  Once it got into
> their bodies it destroyed their livers.  The mechanism was
> interesting - since their bodies had absolutely no way to
> eliminate this teflon like material it just piled up inside the
> liver cells.  If memory serves massive cirrhosis was the result.
> Here'a another study with a similar liquid:
> 
> http://www.chestjournal.org/cgi/content/full/118/5/1436
> 
> In this one they let dogs breathe various fluids for a while and
> then followed them for years.  The most relevant part, to my mind, was:
> 
> ...However, in the whole lungs of four dogs killed at 5 years,
> vacuolated macrophages were still present in all lobes, but in greatly
> reduced numbers compared with earlier years. Concurrent analysis of the
> lung samples for Caroxin-F by gas chromatography revealed an exponential
> decrease in residual values over the years from 0.31 mg/100 g of tissue
> at 6 months, to 0.16 mg/100 g at 1 year, and approximately 0.11 mg/100 g
> at 5 years. Importantly, in no case was there any evidence of fibrosis
> or other permanent damage associated even with the heaviest macrophage
> burden. By chance, the two dogs that survived 10 years, although equally
> exposed to Caroxin-F and having the same postventilatory effects on
> pulmonary function and blood gas values as the others in their group,
> appeared to have the lightest fluorocarbon load as judged by the numbers
> of macrophages in their earlier biopsy specimens.
> 
> Translation - the body is not good at clearing this sort of material.
> Caroxin-F was the best material these researchers found and
> simpler Freon like substances should have much longer half lives
> in the body.  In this context it might be a good safety measure
> to use a fluid for cooling computers which has a boiling point
> sigificantly below human body temperature.  That should greatly
> reduce retention of the material, since these tend not to be
> very soluble in water.
> 
> Regards,
> 
> David Mathog
> mathog at caltech.edu
> Manager, Sequence Analysis Facility, Biology Division, Caltech
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
>