[labnetwork] Issue with diborane in nitrogen mixture

Fri Nov 16 22:50:39 EST 2012

All,
Very interesting.
I appreciate Lief's recall of Bob Hamilton's (UC Berkeley)
experience with polymerized diborane and Steve Paolimi's (Harvard)
experience with polymerized diborane (decaborane) formation possibly
being caused by Joule-Thomson cooling.

New to me is Steve's information that diborane  mixtures with hydrogen are
not subject to this phenomena. At question then is the following:
1. Does dilution in H2 avoid the problem because hydrogen is one of the
few gases that does not experience Joule-Thomsen cooling; or,
2. Does dilution in H2 avoid the problem because hydrogen prevents
polymerization by constantly saturating any potential cross link sites;
i.e., it always drives the equilibrium toward monomer not polymer.

If number1 is correct, then dilution in helium should also solve the problem
- and I suspect it does not.
I expect number 2 is more important and Joule-Thomsen cooling is only
partially (if at all) responsible for polymerization. I expect 
polymerization
of stored diborane, enhanced by the cylinder experiencing temperature
extremes (> 40 C) in an outside gas bunker is the issue. (Note, residues
are common on the regulator high pressure side also - not just the low 
pressure side.)
I expect cylinders are cooled to slow the polymerization reaction, not to
decrease Joule-Thomsen cooling. Cooling cylinders buys time, but that's all.
Some information on diborane decomposition is attached.
This gas phase analysis article does not even address the non volatile 
decomposition
products (polymers like decaborane) but the temp/decomposition 
relationship is clear.

Suggestion:
What if you use 3% B2H6  / 4% H2  / balance N2
The hydrogen pushes the equilibrium to diborane but
the concentration does not raise any H2 handling concerns.

Bill Flounders
UC Berkeley

Paolini, Steven wrote:
>
> I have had extensive grief with this same scenario when I was in 
> industry. To make a long story short, Diborane in the absence of 
> Hydrogen will form solid particles because of the cooling effect 
> (Joule-Thomson) of the regulator  reducing the pressure. The solids 
> are what is known as Decaborane (B10H6). Unfortunately, they form 
> mostly on the seat of the bonnet and allow the regulator to "creep" 
> resulting in the high pressure leaking past the regulator and into the 
> low pressure side. Some engineers have tried to chill the bottle to 
> reduce the Joule Thomson effect but that just resulted in the material 
> forming elsewhere in the circuit.
>
> Your regulator is certainly suited for handling Diborane but you must 
> get used to the fact that they will degrade over time as this material 
> is formed. My suggestions are as follows:
>
> 1)Replace the regulator with a two stage unit, this will extend the 
> failure time.
>
> 2)Specify a large CV on all new regulators, this will extend the life 
> because the effect will be on a larger surface area.
>
> 3)"Stack" regulators in series, The less that the regulator has to 
> reduce pressure, the less Joule Thomson effect will happen. This is 
> expensive and clumsy but a colleague of mine tells me that he has run 
> this set up for three years without incident.
>
> 4)Best method for an absolute cure is to (if you can) switch your 
> mixture to H2 instead of N2, you may not be able to do this depending 
> on your process but if you can, it will eliminate the problem.
>
> I used to run 5000 PPM B2H6 in N2 as a dopant to a furnace, H2 was out 
> of the question because the process was close to the auto ignition 
> point of H2. I used Tescom 1/2" bonnet regulators and swapped them out 
> every three months. If we didn't change them regularly, they would 
> begin to creep and ultimately fail.  It is a phenomenon that you have 
> to deal with in an expensive manner.
>
> Steve Paolini
>
> Harvard University Center for Nanoscale Systems
>
> *From:*labnetwork-bounces at mtl.mit.edu 
> [mailto:labnetwork-bounces at mtl.mit.edu] *On Behalf Of *Leif Johansen
> *Sent:* Friday, November 16, 2012 7:25 AM
> *To:* Lab Network (labnetwork at mtl.mit.edu)
> *Subject:* [labnetwork] Issue with diborane in nitrogen mixture
>
> Dear all,
>
> Here at DTU Danchip we are facing some difficulties with diborane in 
> nitrogen mixtures.
>
> I the recent months we have made the following observations with our 
> 3% diborane in nitrogen mixture:
>
> ·Approximately half year ago, the safety valve opened up due to a too 
> large overpressure on the secondary side of the regulator. Usually, 
> the secondary pressure is set to 1-2 Bars overpressure. The safety 
> valve will open up at pressures exceeding ca. 7 Bars.
>
> ·At that time we believed that the incident was caused by a faulty gas 
> panel regulator. So we changed the regulator and continued to use the 
> gas.
>
> ·However, in the past month or so we have seen that the pressure rises 
> from the setpoint of 1.5 Bars to around 3 Bars. If we then consume gas 
> from the line, the pressure has been observed to drop in discrete 
> steps rather than in a smooth way.
>
> ·This week the pressure suddenly increased dramatically again, leading 
> to a new safety valve release. We have now shut down the gas line.
>
> My questions are :
>
> 1.The bottle is from March 2010. I have once read that over time 
> diborane decomposes into higher order borohydrides, which, unlike 
> diborane, are solid and can clog orifices and nozzles. This phenomenon 
> has already been addressed by Bob Hamilton in this forum. Is our 
> diborane bottle simply too old and emits "ear wax" into the gas line?
>
> 2.The bottle is located in a gas bunker outside, where the temperature 
> can vary quite a bit over the four seasons. The specified temperature 
> range is 0°C to 40°C (32°F to 104°F ). What happens if the temperature 
> exceeds this range?
>
> 3.The gas panel regulator is a "Rotatrex SI 240 V" gas regulator (240 
> Bar primary side, 7 Bar secondary side). I have read that certain 
> materials should be avoided inside valves and regulators used for 
> diborane gas. As far as I can read on the data sheet, the  body is 
> made of 316 L steel, the diaphragm is made of Hasetlloy and the valve 
> seat is made of Vespel or PVDF. Can anyone comment on whether our 
> present regulator is suited at all to handle diborane? If not, can 
> anybody recommend a better alternative?
>
> These issues are causing considerable down-times of our diborane 
> consuming equipment, so I would very much appreciate any information 
> you could share which could help us solve the problem.
>
> *Leif S. Johansen *
>
> Head of Operations
>
> DTU Danchip
>
> *Technical University of Denmark*
>
> 	
>
> http://www.dtu.dk/images/DTU_email_logo_01.gif
>
> Danchip
>
> Ørsteds Plads, Byg. 347
>
> 2800 Lyngby
>
> Direct +45 45255713
>
> Mobile +45 25348992
>
> lej at danchip.dtu.dk <mailto:lesjo at danchip.dtu.dk>
>
> www.danchip.dtu.dk/ <http://www.danchip.dtu.dk/>
>
>
>
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://www-mtl.mit.edu/mailman/listinfo.cgi/labnetwork

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20121116/383f30b0/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 1055 bytes
Desc: not available
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20121116/383f30b0/attachment.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Diborane_decomposition_2005.pdf
Type: application/x-pdf
Size: 544592 bytes
Desc: not available
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20121116/383f30b0/attachment.bin>