Register to reply

Rapid chilling water with LN2 pros and cons question

by anthell
Tags: chilling, cons, pros, rapid, water
Share this thread:
anthell
#1
Nov10-12, 06:34 AM
P: 10
I have this need to be able to rapidly chill (not freeze) water without lowering its O or increasing CO level

I am currently thinking using LN2 due to its lower solubility in water than O and it being somewhat safe

My question is,

1. How safe it is exactly? I really need to chill it fast but i can't lowering its O stats
2. I thought of using liquid oxygen but i can't help but ponder upon the risk of having an explosion....
3. Any other suggestions?

The amount of water i need to chill is around 20 tonnes.... and for all purposes in mind, a rapid 15C drop might be needed on a moment call
Phys.Org News Partner Engineering news on Phys.org
Hoverbike drone project for air transport takes off
Student develops filter for clean water around the world
Developing the next evolution in underwater communication
0xDEADBEEF
#2
Nov10-12, 08:47 AM
P: 824
LN2 doesn't work well due to the Leidenfrost effect, but it is fairly save if you know how to vent all the gas and pressure. I wouldn't use liquid oxygen unless I was forced to do it. There are many ways to cool that water, but you have not given any of the information that is actually important. From what temperature to what other temperature do you need to cool? How fast does it need to be? How well does the end temperature need to be controlled? How much space and money can it cost? Do you need to cool it in place or can you pump it somewhere else? How often do you need to do it? I guess my choice for most cases would be some pre-cooled multi disk block of copper or cheaper aluminium, through which I would pump the water. But really the fact that you didn't give any information that is necessary makes me doubt that you can build something useful or save.
anthell
#3
Nov10-12, 10:03 PM
P: 10
1. I would need to chill it from 27-30 to around 20 (sea water)
2. I would need to be able to replenish around 20-50 tonnes of chilled water within 15 mins
3. I would be using a large water cooler to keep the temperature down afterward but i could also do nothing since i would need to replenish it around 5 times a day
4. Cost and space, the less the better (I was thinking of using LN2 due to the fact that if this experiment failed, i havent really spend much buying large industrial grade water chiller)
5. The idea is, i would use large pump to get seawater directly, and i would need it chilled form around 30 to 20 by the time it arrived at my aquarium. It should take about 15 mins for the water to travel, but i dont think i can actually chill it along the way. Instead, by the time its close to my plant, i need a very powerful chiller to chill the water within minutes

0xDEADBEEF
#4
Nov11-12, 06:09 AM
P: 824
Rapid chilling water with LN2 pros and cons question

I see some information to work with. A rough calculation I did put the cost of cooling with liquid nitrogen to about 3000$ per cooling for 50 metric tons of water, using 7 tons of liquid nitrogen and producing 4500 cubic meters of gas. But you would need a big liquid nitrogen tank and everything, so it would be fairly expensive. And you would need to ensure that there is no asphyxiation. If one wants to have a small cooler one could cool something else like a lot of aluminium pipes first using a slow cooler and then mixing it with the water. But these would take up almost the same amount of volume as the water. So if you need the space anyhow why can't you use a second cooling tank, to prepare the water during the four hours that you have between refills?
flatmaster
#5
Nov11-12, 08:52 AM
P: 504
What is the ambient air temperature? As it is pumped, maybe you can spray it into a mist before it falls into a tank. You would get both evaporative cooling and conductive cooling with the air.
anthell
#6
Nov13-12, 10:52 PM
P: 10
$3000 per 7 tons of LN2 seems to suggest around 50 cents per kg. I did quick search on "how much liquid nitrogen cost" and it suggests around 5-10 cents instead. Can you elaborate on this? Are those super low price only feasible on insanely large quantity? Isnt a few tons daily not large enough?

I could try to chill 50 tons of water using chiller. But even to accomplish a mere 10C in 4 hrs seems to suggest a very powerful chiller with insane electrical bill? Am i overthinking this?

The ambient temperature is around 29-32C during the day and 25-27C at night
0xDEADBEEF
#7
Nov14-12, 12:22 PM
P: 824
I am better with Physics than with engineering. I also just googled some price and I used 30 cents per litre, since I don't know where you live. Maybe you can get a better price. I also made some assumptions about the exit temperature of the nitrogen, which was just a wild guess. Well at least in theory it should be much more economical to cool water from 30C to 20C than to cool nitrogen to −196 C due to the Carnot efficiency, and since the price of liquid nitrogen is basically the cooling cost + some margin, I would assume that the electricity cost is lower than the nitrogen cost. Apparently there are water chillers for air conditioning that can handle your cooling needs and there you know at least that they deliver what they promise. With the LN2 system you have to make sure that the design works that nothing explodes and that no one suffocates. I have no idea how much a large liquid nitrogen tank costs, you need a lot of customization, then you probably need to talk to an engineering office to make a safe design and after two years the thing is finally operational...
Enthalpy
#8
Nov14-12, 03:15 PM
P: 661
A chiller costs electricity but I don't see how liquid nitrogen, whose heat was absorbed at 77K, could take less electricity than a chiller at 300K. Liquid nitrogen consumed to cool near ambient temperature is an incredible waste. They may get electricity cheaper, that's all.

Do NOT use liquid oxygen. That's seriously dangerous stuff.

Produce in advance tiny ice cubes, drop them in water when needed?
Flow the water through a heat exchanger with a paraffin that melts at 15C and was frozen in advance? More compact than aluminium.
Q_Goest
#9
Nov14-12, 09:06 PM
Sci Advisor
HW Helper
PF Gold
Q_Goest's Avatar
P: 2,903
Hi anthell. I work for an industrial gas company. The values for amount of LN2 OxDEADBEEF gave you are in the ballpark. I calculate that you'll need approximately 3400 kg of liquid nitrogen to cool 35 tons of water by 10 degrees C. To get those numbers, there are some assumptions that need to be made about how the liquid is stored (ie: what pressure and amount of subcooling can be expected). But it puts you in the ballpark. Note that the amount of LN2 you use is linearly proportional to the number of tons of water you need to cool and the total temperature. Doubling the water will double the amount of LN2. Cutting the temperature drop in half will cut the amount of LN2 required in half.

The cost I found was roughly 20 to 50 cents per 100 SCF or about 6 to 15 US cents per kilogram. That assumes you're in the US. I don't know how well it compares to Europe but I suspect it's still pretty close. The cost depends on how much product you take and how far you are from an air separation plant. The farther it needs to be trucked, the more it'll cost. You may have to add some cost for a monthly service charge which is basically a cryogenic tank rental.

So the cost to cool 35 tons of water by 10 C is $204 to $510. To do this however, you'd need to enter into a contract for a few years, so don't underestimate how much product you'll need and the monthly service charge. You could get a spot job done to avoid the contract but I don't have costs for that and they'll be a bit higher I'm sure. I could find out if you'd like.

To cool the water with nitrogen, I'd suggest some kind of heat exchanger such as a shell in tube type or a plate type heat exchanger. They're not so expensive and they prevent the nitrogen from getting into the water. They'll also allow you to vent the nitrogen safely. But you'll need a pump for the water at the least.

All that said, if you're looking for a less expensive way of doing this, I kinda like Enthalpy's ideas which is to store the 'refrigeration' in the form of ice or paraffin. You could also store it as cold water for that matter and just pump it through a heat exchanger when you're ready. I'm sure in the long run, that kind of method would be less expensive than using LN2 because LN2 requires a large consumption of energy to liquify the nitrogen when compared to the relatively small amount of energy required to remove the energy from the water you're talking about.
anthell
#10
Nov14-12, 09:27 PM
P: 10
Thank you you gusy for all the helps so far.

The concept of using ice to chill the water is very possible if there is a company that actually manufacture frozen sea water, which i doubt since it ruins their machine, and its harder to freeze salt water. I cannot use normal ice because it affects salinity and other stuffs the sea water comes with....

Let say i consider using contact freeze pipe system, can anyone give a rough idea how much electricity will i am looking it at? or in term of wattage, so i can just compare it to the price per watt. Also, how big is the machine going to be to chill that much water at that rate?
Q_Goest
#11
Nov15-12, 04:59 PM
Sci Advisor
HW Helper
PF Gold
Q_Goest's Avatar
P: 2,903
Quote Quote by anthell View Post
Thank you you gusy for all the helps so far.

The concept of using ice to chill the water is very possible if there is a company that actually manufacture frozen sea water, which i doubt since it ruins their machine, and its harder to freeze salt water. I cannot use normal ice because it affects salinity and other stuffs the sea water comes with....
How about having a second, fresh water pool you can dump ice in. Pump the cold water from the fresh water pool through a heat exchanger and back. The salt water pool would also pump water through this heat exchanger. The ice in the fresh water pool should provide quite a good heat sink. I calculate you'll need just under 5 tons of ice to cool 35 tons of water by 10 C.

I'd welcome others to do the calculation as I didn't check it very carefully. I found heat of fusion for ice here.
AlephZero
#12
Nov15-12, 09:59 PM
Engineering
Sci Advisor
HW Helper
Thanks
P: 6,932
I agree with QGoest's ice calculation, assuming the ice is stored at close to 0 degrees C.

To estimate the electrical power:
Specifc heat of water 4200 J / Kg.K
To cool 35 tonnes of water 10 degrees = 4200 x 35,000 x 10 J = 1500 MJ
1/5 day = 17,280 sec
Average power = 1,500,000 / 17,280 = 87 kW

Including an allowance for less than 100% efficiency, maybe 100 kW (i.e. 2400 kWh per day or about 900 MWh per year).
Enthalpy
#13
Nov17-12, 05:39 AM
P: 661
For drinks, there are small ice cubes enclosed in a spherical plastic shell of D~15mm, with a small air bubble. The contained water is frozen in advance, the ice cube dropped in the drink, and later reused. They achieve your 15min and 10K in normal use.

You could enclose the freezing liquid in such small shells that you drop in your Ocean water.
  • Looks faster than a heat exchanger, easier to clean.
  • You can stir the water more easily than in a heat exchanger.
  • No worry with a solid in tubes.
  • The freezing liquid can taw around +10C (paraffin?), saving energy as opposed to fresh or sea water. Freezing them above 0C also simplifies the operations, as it avoids frost.
The ones I saw were injected half-shells welded together. Maybe the manufacturer can change the contents for you - or keep its fresh water. Or let a contractor inject them; it costs you a mould.
0xDEADBEEF
#14
Nov17-12, 01:25 PM
P: 824
Freezing sea water seems to be somewhat troublesome, too. On some non authoritative websites I found that gas is expelled from the ice while if freezes (arguably this is why ice cubes from the freezer are not clear) as well as some salt producing a layer of unfrozen water. Therefore the oxygen content may change as well as the salinity when you cool this way. I doubt that you can rent pre frozen cocktail coolers by the ton (but who knows...), if you freeze then yourself you will not gain anything in terms of your chiller's size. A real option that you might want to consider is buying ice from an industrial supplier, and cool using some kind of heat exchanger or sealed bags.
Enthalpy
#15
Nov20-12, 06:47 AM
P: 661
Yes, gas (and salts) is what gathers at the center of ice cubes, because they freeze from the outside in. There I describe a faster method that also makes clear ice:
http://www.chemicalforums.com/index....5184#msg175184

And yes again, sea water segregates when freezing. It lets columns of very cold brine sink from the near-surface to the Ocean's floor and freeze everything there that isn't brine - impressive videos.

However, if freezing quickly, segregation should be reduced. That's where the ice cubes help over a more difficult heat exchanger.

The chiller for ice cubes is smaller than a heat exchanger because it can take its time.


Register to reply

Related Discussions
C++ Compiler Pros & Cons Programming & Computer Science 1
The pros and cons of Gedankenexperimente General Discussion 11
The pros and cons of global warming; i.e. GCC Current Events 10
Pros and cons of the science -physics General Discussion 22
Linux: Pros and Cons Computing & Technology 21