Methanol Water Mixture as Secondary Heat Transfer Liquid

[StoneCold]

Greetings
We have a plant cooling system that has always given us some problems. The system runs a mixture of methanol and water (75 wt% methanol) as the secondary heat transfer fluid.

This fluid runs through the chiller and out to reactors, condensers, etc. in the plant. It is a closed loop system. Sometimes it appears that the methanol and water separate in "dead legs" and the fluid freezes.

We run the system near -20C, with the evaporator on the chiller running about -50C.

Have you ever experienced this problem?

Anyone else crazy enough to run methanol and water?

Lab tests have not shown any phase separation of the fluids at these temperatures.

Any suggestions?

Thanks
StoneCold

 

[MintJulep]

Possibly a particulate contaminate serving as a nucleous to either initiate crystallization, or locally alter the mixture?

 

[btrueblood]

Stone,

Dunno. I have seen water ice crystallize out of 50/50 glycol+water solutions, when a chiller bath got set to too low a temperature. The water ice doesn't re-melt very quickly until the temperature is above the melt point for pure water. Glycol and water reach a eutectic at 70% glycol by volume, and have a freezing point at that mixture of -70 F. Ethanol and water I have data for, and it shows a freeze point of about -60 F /-50 C at 72% by weight ethanol in water. Methanol I don't have data for; are you SURE that methanol/water will not form a slush at -50 C? If it does, then slush will form on the chiller coils, and can break off and flow downstream, possibly pooling or packing into dead zones.

Also, because methanol has a high vapor pressure, it tends to boil away if you aren't watching the mixture carefully, even at chilly temperatures. Are you reading the mixture on a daily basis, using a trusted and calibrated hydrometer?

Glycol/water mixture might work better for you, it has the ability to get well below your -50C chiller temperature with no possibility of slush forming (with glycol conc. above about 60% and less than 75%). Also, its vapor pressure is lower and thus less prone to loss that way.

Lastly, we see changes in apparent hydrometer glycol content readings in fluid that has picked up salts/corrosion byproducts over time. This is long-term stuff, though, and preventable with good corrosion protection systems.

 

[EdStainless]

The nice thing about alcoh/water is the low viscosity. Glycol is like pumping mud.
My only expr with meth/water is in small systems (cooling the CPU on a computer).
You need to check the SG of your mixture. There are a lot of decomposition products that can mess the system up.


= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[lilliput1]

Be careful! Methanol is flammable. I would recommend changing to food grade propylene glycol to minimize liability for accidents & environmental spills. Peace of mind & safety is worth the extra hp for pumping. Check available pump head & position of pump balancing valve, you may have enough spare to be able to handle the added viscousity.

 

[EdStainless]

There are other low viscosity additives available. I know that food plants use something called Enviro-Kool.

http://www.cwichem.com

It comes in different grades for different temperatures.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[chicopee]

w/ 75% methanol by weight,did you calculate the freezing point of the mixture? also w/ such high concentration of methanol have you experienced any degradation of elastomeric material such as seals in the system?

 

[StoneCold]

MintJulep,btrueblood,Edstainless,lilliput1, chicopee

Thanks for your responces.
We are dealing with the flamability issues and with the elasomers. Switching fluids now would require all new pumps as we don't have any head to spare.
What I am really wondering about is these strange freeze spots. Maybe I will try in the lab for extended periods and an experiment with some salt, or insoluble powder added to see what happens.

Thanks
StoneCold