Transport of Supercritical CO2 3

[Armen75]

We are having a big debate here on the properties of CO2 at high pressures, and a lot of the confusion is related to reading the phase diagram (Pressure/Enthalpy) for CO2.

If we start at 250 bar / 100oC (supercritical) and transport in a pipeline far away with a destination pressure of 150 bar and 40oC. Is the destination a liquid or is it still supercritical.

What if the destination conditions are 40bar and -20oC.

Which comes back to saying is there such a thing as sub-cooled liquid CO2.

 

[zdas04]

Terminology helps some here. At your first point you are in the Dense Phase region (the term "super critical" doesn't really mean anything in CO2, and probably not very much anywhere else). If you assume a straight line reduction in pressure and temperature (kind of risky since that line is really close to the Critical point and dropping into the Vapor region can make strange things act really weird) then you end up well to the right of the Melting Line and above the Boiling Line (i.e. as a subcooled liquid).

Continue to drop the pressure and temperature then you'll get to the melting line somewhere in the neighborhood of -56C and above 5 bar and turn into a solid.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

"It is always a poor idea to ask your Bridge Club for medical advice or a collection of geek engineers for legal advice"

 

[Armen75]

Sorry, you are right, Dense Phase.

To paraphrase, even if the pressure remains above the critical point, but the temperature is low enough, the fluid goes from dense phase to a liquid and eventually to a solid.

The melting line is shown.
Which brings up the question of WHEN does it stop being dense phase and start being a liquid? Is it somewhat like a phase transition

 

[zdas04]

You can see it on both a P-T diagram and a P-h diagram. The thing that is confusing on a P-h diagram is that the entropy lines are smooth through Dense Phase line (which is horizontal adjacent to the Gas region and curved and mostly vertical next to the liquid region). If you look really closely, you can see that the density lines are disconnected across the gas/dense phase line.

On the P-T diagram the liquid/dense phase line is vertical so anything to the left is liquid and anything to the right is dense phase--it is a real phase transition. A big difference between the two is compressibility. You pump liquid, but you compress dense phase.

David

 

[MortenA]

If you havnt already visited union.dk for co2 info - then go there. (im not with union - its just a very good handbook).

The co2 phase diagram shows that at 40 barg/-20ºC you will be in the liquid phase

The term "sub-cooled" mean that you have a liquid that actually should have frozen - you can observe this with water if you are a little careful.

The next phase transition is at around -50ºC (little lower i think). Be avare: A very common process simulator such as HYSYS will NOT find the solid phase for co2!

Best regards

Morten

 

[zdas04]

Morton,
"Sub cooled" has nothing to do with freezing. It means that you are off the saturation curve into the liquid region. It sounds like you are talking about a Super Cooled Liquid in your post. Similarly, "super heated" means that you have a gas that is away from the saturation line.

David

 

[Armen75]

Thank you both for your help, very informative.

To clarify, "at what point does the fluid stop being in dense phase and is considered a liquid"
On the P-H graph, is it :
a) Anthing to the left of the critical point and above the bell (557 kj/kg)?
b) Anthing colder than the critical point (304K) ?

Question 2
and, should there be a concern with this phase change if it will occur in a continuous flowing pipe intended for injection?

 

[zdas04]

Armen,
First, I recently realized that I shouldn't have jumped on you about the term "super critical". On my Mollier diagram the dense phase region is labeled "SCF" for "Super Critical Fluid". It is just that the term "critical" (and by implication "super critical") is used so often in fluid mechanics for so many unrelated things that I start feeling that the use of the term should be discouraged. Just my own bias, sorry.

Also, all of my references have the critical point at 87.76F (360.9K) and 1070 psia (73.8 bara). You might want to check to verify the 304K you referenced.

For your first question, I think I have liquid CO2 anywhere less than the critical temperature at a pressure greater than the Boiling Line as long as temperature is greater than the Melting line.

Any time you are dealing with pressurized CO2 you should be concerned. A long-time poster on these boards named Montemayor once posted an example of a leak in a liquid CO2 vessel. Pressure dropped isothermally to the saturation line, then pressure and temperature followed the boiling line down to the sublimation line to atmospheric pressure at around -120F (-84C). Most piping has problems at those sorts of temperatures.

With CO2 Sequestration being such a hot topic this week, a lot of people are evaluating the pluses and minuses of gas transport vs. liquid transport and gas injection vs. liquid injection. The issues are complex and there is no one-size-fits all answer. You can compress dense-phase CO2, but it is one of the best surfactants on the planet--it will "clean up" all of the lubrication in a recip compressor and special materials are required. Starting with gaseous CO2 and liquefying it is very energy intensive (for a project I looked at last year, the total cost including capital and operating cost for this option was 3 times gaseous transport and twice dense phase transport).

Phase change happens with many fluid processes and it doesn't have to be a catastrophe, but unplanned phase change is rarely a good thing. For example, in your original example if your ending point had been a little bit lower pressure then the pressure traverse would have taken you from Dense Phase to Gas to Liquid--as a gas, a constant mass flow rate as required by the continuity equation would have yielded extremely high velocities that could easily be a noise problem or an erosion problem. If the CO2 should still be gas (or a liquid very close to saturation) at the pump suction then cavitation is a serious risk.

David

 

[dcasto]

I'll disagree with the Zadas/mortemayor issue on depressuring supercriticle fluids. most pipings have no issues with dropping the pressure on fluids well below the piping "min temp" I've depresssured supercriticle ethylene hudreds of times, not a single issue (-121 on api 5xl grade x52)

If you look at the density of the liquid when you drop below the criticle point, it won't be much different than when it was a criticle "gas" fluid, the issue comes only when you want to try to measure the stream an it's two phase or you want to pump it, pumps do not like phase changes on the suction.

I've ran a centrifugal pump and a PD compressor in parrallel on a supercritical fluid, they both worked just fine.

I've seen systems two phase and it's really weird. the top of a line is several degrees warmer than the bottom. Separators can't seperate because the gas and liquid phases have the same density, the heat in/out causes ice to form on minute and to melt off 15 seconds later as the fluid goes from gas to liquid and back.

Injecting CO2 into a formation is also fun. it may be a gas at the surface, but the CO2 will always be a liquid when it enters a formation. I had a petroleum resivoir engineer tell me I couldn't inject the CO2 into a formation because the gas would have too high of a volume. When I told him to drop the college books on gas resivoir injection and use the saltwater injection books he would see that I , a chemE was right. I was right when the well went on line.

 

[zdas04]

Well in any case, the thread I was talking about was thread135-145412. I haven't worked with the phase diagram of ethylene and don't know if it has similar phase behaviour to CO2 or not. In my limited experience, CO2 has been different from anything I've worked with, but I haven't worked with many different gases. Maybe your experience with ethylene is more relevant than Art Montemayor's comments about his 50 year career with CO2 and other industrial gases.

David

 

[dcasto]

ethylene looks very much like CO2 In the US C2= lines operate between 800 psi and 2200 psi. the criticle pressure is near 840 psi and 50 F. Its boiling point is -155.

It doesn't have a triple point to mess things up, more.

 

[davefitz]

I think that if it is a long pipeline then heat transfer over the length of the pipe becomes important in the analysis.

You would need to break the pipeline into perhaps 200 short segments, and calculate the heat loss and pressure drop across each segment. The program would need to be able to access thermophysical properties of CO2 and recognize change of phase accordingly.

 

[Armen75]

Yes, that is what we did: breaking the 33km pipeline into 30m segments, we found that in the winter scenario we just barely dip into the liquid (or dense liquid) part of the graph (somewhere around the 25th km)

As mentionned by Dcasto, the density should be close enough between the 2 fluids, we are assuming minimal impact from density.
I can see how measurement can be a huge problem if you do not know the phase, but we are not measuring it at the outlet.
The bigger concern is with the viscosity. What happens to the flow if the viscosity starts jumping around with the fluid alternating between phases.
Thinking this through, we assume that vibration due to cavitation shouldn't be an issue since there is no implosions due to volume changes.

Ultimately, we can stay away from the "change" by starting off with slightly warmer CO2 and therefore ending up with dense phase conditions at delivery.
wondering if the worry is worth the trouble...

 

[dcasto]

armen, what pressure drop equations are you using? on our 160km legs we broke it down into 15km sections, then used Hazen williams with an EOS for density of the ethylene. Unless you are getting to a point where you are going back and forth and the density is swinging wildly, then calculating every 30 meters isn't gonna change much over 300 meters. It doesn't hurt with free compture power.

A major company used Panhandle to calculate the pressure drop for their client we were working with (actually a distant arm in the same company) and I told the client that the engineering firm was wrong. In any case, when the line started up, yep, I missed it by 2% and the engineering firm was off 15%.

 

[Armen75]

We are using a Hysis model. Initially based on a Lee-Kessler EOS, and then added extra checks using advanced equations-of-state (as suggested by Aspen). Even the guys at Aspen told me that the phase indicated by the software in the label is incorrect but they reassured me that that the properties generated are ok.
(results with Lee-Kessler and AEOS were very similar)

BTW, in the continuing quest of knowledge, a new term was introduced to me yesterday: Dense Liquid; which related back to my original question.... it is a liquid. (It is a long enough pipe to ensure the cooling past the critical temperature to liquid)

Yes, raw computer power is great, as there is no real advantage of breaking it down to such small seegments, definitely overkill.... just a "might as well". (I think Aspen uses Darcy as a default, will see about Hazen)

Again, I might be creating an issue when there isn't one; Hysis didn't identify any issues. It is the Team that is worried there may be one due to the change from Dense gas to Liquid.

As it stands, I am totally on-board the 3 issues that were mentionned :
1- the concern of instrumentation, measurement and control
2- temperature and metalurgy issues during depressuring
3- Lubrication due to liquid CO2

Finally, as we are going to inject the CO2 through a valve, we are suggesting the option to start off slightly hotter and giving a sufficient margin to remain in the densa gas area without transitonning into the liquid.

Thank you all for your input

 

[MortenA]

Armen - be avare: HYSYS (as its called) WILL NOT calculate the correct transition to solid CO2 - CO2 does NOT have a solid phase in HYSYS.

Use tables/diagras instead - its much more precise - at least as a check! If you dont digress into the solid zone - then you have no problem of course.

For good reference take a look here:

http://www.union.dk/media/CO2-GB.pdf

This is original data all the way back to Plank & Kuprianoff - that has withstood the test of time

Best regards

Morten

 

[dcasto]

we always heat the ethylene before meaureing it to make sure we were 10 to 20 degrees above the criticle temp for measurement. We kept a back pressure valve on meters to assure we were where we wanted to be.

There is no issure about metalurgy upon blowdown. The hoop stress level is not high enough. This doesn't relieve you off doing some pipe stress due to contraction on above ground piping. Get on kinder morgans CO2 pipeline pages and I think they pictures of some of their sites. Better yet, go to the four corners area and look at their sites.

valve grease will be an issue, all the valve manufacturers will tell what they use for this service.