High pressure CO2 chamber

[the_anomaly]

Hello all,

I've seen a few videos on Youtube of liquid CO2 in high pressure clear chambers https://www.youtube.com/watch?v=-gCTKteN5Y4.  I'm interested in performing some experiments on liquid/supercritical CO2 and decided to build one myself out of acrylic:



It's in a box with a half inch polycarbonate door.  Just in case it ruptured, I wanted to be protected.   

Later on, I learned supercritical CO2 might have a detrimental impact on acrylic.  So now I'm thinking of building one using borosilicate glass as the chamber material.  Since the thickest wall I was able to find is 4mm, I'd like to fit the glass tube into a larger acrylic tube.  The acrylic tube provides wall thickness to keep the glass from rupturing.  The glass/SS electrodes isolate the CO2.  A simple model is shown here:




My question is, do you think it would be better to have the glass tube (barely) press fit into the acrylic or should I bore the acrylic out to maybe 0.030 - 0.050" larger than the glass tube diameter and fill that interstitial space with a clear epoxy?  My concern is I can only slightly press fit the glass tube into the acrylic - I know it won't tolerate much of an interference fit.  With epoxy, I can make sure there is solid material between acrylic and glass but it would be a lot more effort to fabricate.  I also have not looked at epoxies yet to see if there are any clear ones with high compressive strengths. 

 

[davekni]

Either way, the acrylic will be there only to protect you if the borosilicate glass breaks.  The glass is much harder/stiffer than acrylic.  Glass will not expand sufficiently under pressure to transfer any significant fraction of load to acrylic.  Glass will either handle the pressure itself or break.

[Twospoons]

Do you really need to see inside?  If you are more interested in just using supercritical CO2 for extractions then you might as well just use steel.

[klugesmith]

I agree with Dave: clear plastic jacket won't significantly reduce the stress in glass, even if the air gap is filled with epoxy. Look up the Young's modulus values.
So the plastic shell is a safety shield, offering possibly better visibility than a metal screen.  The gap filling material could be oil or gel. If its index of refraction matches the plastic, then the plastic surface doesn't need to be super smooth and polished.

If the plastic is strong enough, and chemically compatible with CO2, then you can skip the glass. The view would be less distorted if outside surfaces are flat instead of cylindrical.
Acrylic is relatively brittle and not very strong compared to clear polycarbonate plastic. 
For a 20 bar clear sight glass I procured some Lexan tubing with OD of 1 inch and ID of 3/4 inch. Have lost track of strength analysis.   Would guess the tube would be weakened by long exposure to very high pressure CO2, but that's a thing maybe findable on Internet.

I stand with Twospoons in questioning the utility of 360 degree visibility, unless your goal is to make cool videos for youtube.
If you want _some_ visibility, a small flat window might be more practical and safe than a clear cylinder.
 

[the_anomaly]

Klugesmith those are some good suggestions I have been turning them over in my head.  I should have given some more details in the original post.  I found these papers https://wp.icmm.csic.es/wp-content/uploads/sites/32/2015/06/JPCL2010.pdf and https://pubs.aip.org/aip/apl/article/96/14/143110/986587/Nanopatterning-of-carbonaceous-structures-by-field.  They describe breaking down CO2 into carbon and oxygen using electricity at the nanometer scale.  I didn't quite follow why they chose an atomic force microscope to deliver the field; I think these microscopes might have very fine control of distance from probe tip to surface?

Anyway it got me thinking about what is the dielectric strength of CO2 in gaseous, liquid, and supercritical forms.  I've seen some values given for standard room temp and pressure but what about at 100, 200, 500 psi?  From looking at Paschen curves, I believe at high pressures it will continue to increase.  In the papers, they achieve dissociation at 40V / nm.  I won't achieve that spacing with my setup but reaching 25 um (0.001") or less is not unreasonable.  My plan is to apply high voltages to different electrode surfaces, flat, pointy, half sphere and measure the breakdown voltage and see if I do get any dissociation into C + O2 or possibly CO + O.  I was even thinking of adding some catalysts.

So my needs are:

1.  A chamber that can withstand pressures up to 1200 psi.  More is better but this seemed like a good starting point.
2.  A Way to insulate the electrodes.  At 0.0001" or 1/10 of a thou I need just over 101 kV to meet the 40V / nm.
3.  See through, at least a window. 

For part 3, I fill my chamber from a 5lb tank.  I use a heat exchanger (SS tube through ICE) to liquify the high pressure gas.  Being able to see the liquid has been helpful.   

I have not found high voltage feed throughs capable of these voltages or maybe they exist but are out of my budget. 

For the picture of my existing chamber, the ID is 1.250" and each plate of acrylic is 1" thick.  I measured the deflection at 740 psi and I got 0.004".  My concern has been what davekni first said, the borosilicate would shatter before it deflected that far.  I suppose there could be two chambers, one the glass tube is in that I pressurize to keep the differential pressure between inside of tube and outside to a minimum.  This would be a more complex design.

I found Schott makes a heavy wall glass product capable of 600 psi https://www.schott.com/en-us/products/tubular-gage-glass-p1000375/product-variants?tab=heavy-wall.  And I was wondering if I made a mold and melted glass into it, if I could make a very thick wall glass chamber.  No

My interest in using glass is I came across a source saying that supercritical CO2 will diffuse through acrylic and possibly chemically damage it.  So my measurements of breakdown may not be accurate.  I have not found any information about borosilicate.  I know borosilicate is typically very chemically resistant so I was hoping it would hold up here.

If this gets too complex I might stop but its been on my mind for a while and I wanted to give it a shot. 

[Twospoons]

If you can find suitable high-voltage feedthroughs (probably the hardest part) then you could use the same sort of setup AppliedScience used : Al ring with the HV feeds, then thick glass flats front and back held by an external ring clamp.

Those HV feeds are going to be tough - high pressure and high voltage.