Chinese researchers discover new salty cooling solution that can drop temperatures by more than 50 degrees Celsius in seconds — depressurizing saturated fluid triggers massive amounts of heat transfer

News
By published

Will this salt solve our cooling problems?

Data Center
(Image credit: Getty / MASTER)

Researchers from the Chinese Academy of Sciences have discovered that saturating water with ammonium thiocyanate salt and pressurizing it, and then suddenly releasing the pressure, would allow it to cool by 30 degrees C (54 degrees F) at room temperatures, even going as far as a 50-degree C (122 degrees F) drop in warmer environments. The research paper, published in Nature [PDF], presents this as a greener alternative to current refrigerants, with its theoretical efficiency hitting almost 80% — much higher than conventional refrigerants — making it a great solution for the thermal issues that data centers face.

The researchers compare the system to a wet sponge, with water acting as the sponge and salt acting as the absorbed fluid. When you apply pressure on the sponge (the water), the fluid (the ammonium thiocyanate salt) will be squeezed out and release heat. But when you reduce the pressure, it will quickly reabsorb the solution and reabsorb the heat from its environment, causing a quick temperature drop.

“Cooling systems currently account for nearly 40 percent of a data center’s total electricity consumption,” state media CCTV said. “This finding could provide a more efficient cooling solution for these high-energy-consumption facilities.” The development could reduce the power needed to cool AI chips, especially as we get chips that deliver more performance while requiring higher power. Aside from this, it can also be used in other applications, like HVAC systems.

Still, this has the potential to reduce data center electricity consumption — a big problem that many AI tech companies are grappling with right now. Even China, which has ample electricity supply for all the data centers it’s building, is considering this a breakthrough, allowing it to be more cost-efficient and competitive when in the AI race.

Google Preferred Source

Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Jowi Morales
Jowi Morales
Contributing Writer

Jowi Morales is a tech enthusiast with years of experience working in the industry. He’s been writing with several tech publications since 2021, where he’s been interested in tech hardware and consumer electronics.

10 Comments Comment from the forums
  • hotaru251
    i mean not really shocked this works. pressurizing and then releasing pressure is crazy heat transfer.
    Reply
  • Air2004
    hotaru251 said:
    i mean not really shocked this works. pressurizing and then releasing pressure is crazy heat transfer.
    I gotta run your statement by the interwebs before I can trust it.

    Ok, your statement checks out.
    Reply
  • Thunder64
    And salt helps too.
    Reply
  • Jame5
    Scientists: "Did you know that pressurizing/depressurizing coolant is an effective method of heat transfer?"

    Entire HVAC industry: ...
    Reply
  • Notton
    The article already compares the solution to how HVACs work, but the claim is the fluid is even better than the various refrigerants currently used.
    However, no numbers are given, so by exactly how much? IDK.

    I didn't look into it too much, but I'm guessing this new solution's main competitor is R-717 (ammonia)?
    Reply
  • spaceminions
    Jame5 said:
    Scientists: "Did you know that pressurizing/depressurizing coolant is an effective method of heat transfer?"

    Entire HVAC industry: ...
    The part you're missing is that normally you are compressing a *refrigerant* which *changes from a gas to a liquid* whereas this barocaloric effect occurs without a gas phase. Compressing gases can be inefficient and annoying; the compressors in refrigeration systems are nicer but think about how loud and hot an air compressor can be. Pumping around water with stuff in it is pretty simple and easy.
    Edit: So it looks like their particular salt absorbs a lot of heat when it dissolves, like instant cold packs, but it also comes back out of solution when pressurized, which releases the heat again of course. Pumping gases around is slower than pumping liquids on a mass basis, since liquids are denser, so the flow rate needed should be less on a volume basis, which would be good. Might also be nice that you don't have to prevent accidentally feeding liquid refrigerant to the pump inlet, so no superheat? Idk.
    Reply
  • edzieba
    Aside from that, it’s hygroscopic and absorbs moisture, potentially affecting the long-term stability of the solution.
    ... It's used as a saturated solution. In water. Regardless of how hygroscopic it is, it physically cannot absorb any more water.
    Reply
  • spaceminions
    edzieba said:
    ... It's used as a saturated solution. In water. Regardless of how hygroscopic it is, it physically cannot absorb any more water.
    Ah no, wrong way around. A saturated solution means if you had any less water you couldn't dissolve all of the salt. It can definitely keep absorbing more water, until it becomes dilute enough that it loses as much as it's gaining. But it should be a closed loop and not do that anyway, and maybe you could trap a bit of excess salt so that any water that somehow gets in won't change the concentration.
    Reply
  • bit_user
    spaceminions said:
    Edit: So it looks like their particular salt absorbs a lot of heat when it dissolves, like instant cold packs, but it also comes back out of solution when pressurized, which releases the heat again of course.
    This is what I was wondering. I figured there must be some nonlinearity they're taking advantage of. As I understand it, that's what makes 2-phase cooling so attractive.

    Thanks for your informative comments!
    : )
    Reply
  • edzieba
    spaceminions said:
    Ah no, wrong way around. A saturated solution means if you had any less water you couldn't dissolve all of the salt. It can definitely keep absorbing more water, until it becomes dilute enough that it loses as much as it's gaining.
    That's just regular old dilution though, with hygroscopy (or lack thereof) completely irrelevant to the addition of more water.
    Reply