Slashdot Mirror


High-Power Thermoelectric Generator Utilizes Thermal Difference of Only 5C (newelectronics.co.uk)

A silicon-nanowire thermoelectric generator has been developed by a team of researchers from Waseda University, Osaka University, and Shizuoka University. From a report: According to the Japanese researchers, this experimentally demonstrated a high-power density of 12 microwatts per 1cm2, enough to drive sensors or realise intermittent wireless communication, at a small thermal difference of only 5C. Silicon-based thermoelectric generators conventionally employed long, silicon nanowires of about 10-100nm, which were suspended on a cavity to cut off the bypass of the heat current and secure the temperature difference across the silicon nanowires. However, the cavity structure weakened the mechanical strength of the devices and increased the fabrication cost. The team says their generator has overcome this issue.

"Because our generator uses the same technology to manufacture semiconductor integrated circuits, its processing cost could be largely cut through mass production," says Professor Takanobu Watanabe of Waseda University. "Also, it could open up a pathway to various, autonomously-driven IoT devices utilising environmental and body heats. For instance, it may be possible to charge your smartwatch during your morning jog someday."

47 of 100 comments (clear)

  1. Forget 5C, how about higher differentials? by ZorinLynx · · Score: 3, Interesting

    Being able to efficiently convert heat into energy is a holy grail. Are these TEGs better than previous tech when you have large differentials?

    1. Re:Forget 5C, how about higher differentials? by michelcolman · · Score: 2

      This thing could already power watches as if by magic. My watch battery lasts a few years, and if I did the math correctly, that's well below 60 microwatts. Surface area is more than 5 cm2, and temperature differential is well over 5C, so it should be plenty of power.

      (Not talking about smartwatches, obviously).

    2. Re: Forget 5C, how about higher differentials? by SuricouRaven · · Score: 1

      Converting large differentials to useful energy is easy - on a large scale. It does not scale down well. High-performance TEGs (And ones that don't melt so easily) could make it practical to scavenge leftover heat energy from many industrial processes, improving efficiency.

    3. Re:Forget 5C, how about higher differentials? by nospam007 · · Score: 1

      "This thing could already power watches as if by magic. My watch battery lasts a few years, and if I did the math correctly, that's well below 60 microwatts. Surface area is more than 5 cm2, and temperature differential is well over 5C, so it should be plenty of power.

      (Not talking about smartwatches, obviously)."

      I see, you have a small wrist, But what if one could tolerate a cold ass to power their iWatch?

    4. Re:Forget 5C, how about higher differentials? by BlueStrat · · Score: 1

      This thing could already power watches as if by magic. My watch battery lasts a few years, and if I did the math correctly, that's well below 60 microwatts. Surface area is more than 5 cm2, and temperature differential is well over 5C, so it should be plenty of power.

      (Not talking about smartwatches, obviously).

      InB4 panties with heatsinks/convection-cooling radiator fins suddenly come into fashion among women. :)

      Strat

      --
      Progressivism (aka US 'Liberalism'): Ideas so good they need a police/surveillance-state to enforce.
    5. Re:Forget 5C, how about higher differentials? by Applehu+Akbar · · Score: 1

      I hear the "hum" of a new industry preparing to get it's...feet...wet.

      Followed by the flesh-cooking sizzle of users with wet feet running electric appliances.

  2. 12 uW/cm^2 isn't much by Nutria · · Score: 4, Insightful

    Unless my math is wrong, that's less than an eighth of a watt per square meter.

    --
    "I don't know, therefore Aliens" Wafflebox1
    1. Re:12 uW/cm^2 isn't much by iggymanz · · Score: 1

      useful wireless? no, I think not, for lowest power class 3 bluetooth for example that 1.2 microwatts of the square centimeter is off by factor of 1000.

    2. Re:12 uW/cm^2 isn't much by Zocalo · · Score: 4, Interesting

      Depends on what OP meant by "intermittent". There are plenty of telemetry sensors that only need to transmit a tiny amount of data at intervals in excess of an hour, or in the event of a given condition being met, for which 12 uW/cm^2 might be more than enough to charge up a battery or capacitor with enough juice to do so when the time comes.

      --
      UNIX? They're not even circumcised! Savages!
    3. Re:12 uW/cm^2 isn't much by 110010001000 · · Score: 2

      That makes sense. It doesn't matter if you have 1 centimeter or 1 meter or 1 kilometer of generator - you will be "off by a factor of 1000".

    4. Re:12 uW/cm^2 isn't much by 110010001000 · · Score: 2

      I was going to say that too, but I guess iggymanz is the expert, so I think we should notify the researchers.

    5. Re:12 uW/cm^2 isn't much by DrTJ · · Score: 4, Informative

      "experimentally demonstrated a high-power density of 12 microwatts per 1cm2"

      Somebody need to work on their adjectives. Solar panels are in the order of 10-20 mW/cm, i.e. 1000x more.

      If 12 uW/cm qualifies as "high-power density", then solar panels must be "super-power density".

      Maybe the researchers are from Krypton.

    6. Re:12 uW/cm^2 isn't much by iggymanz · · Score: 1

      I know, pesky math and physics can be such a downer. how long are you going to have to charge for 1kb payload packet of bluetooth class 3 power?

    7. Re:12 uW/cm^2 isn't much by 110010001000 · · Score: 1

      I'm not sure. I'll go let the researchers know that it isn't possible to do what they say might be possible. Be right back!

    8. Re:12 uW/cm^2 isn't much by bobbied · · Score: 1

      Unless my math is wrong, that's less than an eighth of a watt per square meter.

      Can you say "LOW, LOW duty Cycle"? You'd be better off with a small solar cell over this..

      --
      "File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
    9. Re:12 uW/cm^2 isn't much by angel'o'sphere · · Score: 1

      A 5C temperature gradient you can have nearly everywhere.
      A solar cell only works where you habe light, obviously.

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    10. Re:12 uW/cm^2 isn't much by Darinbob · · Score: 1

      I'm working on stuff that needs a decade or two battery life (no recharge) and is wireless. Give it a hundred square centimeters and you're in a good range to help with this. The snag though is that temperature doesn't change a lot, but if you can store the electricity in a capacitor then it helps extend the battery life.

    11. Re:12 uW/cm^2 isn't much by Darinbob · · Score: 1

      Don't think about this in 2D like it's a solar panel. Fold it up and you can get 100 square centimeters in a compact space. Now if you can keep it generating electricity then that's great, but more likely you'll need to store up the electricity in a supercap.

    12. Re: 12 uW/cm^2 isn't much by Darinbob · · Score: 1

      Some products only need to transmit once every 20 minutes or more. The snag is that you need to receive a lot more often and that ends up being a much large power drain than transmission.

      A millisecond transmission is a snag in itself. To squeeze more data in you need more bits per second, but that also decreases range. Most independent sensors (not wired to a reciever) would want better range than bluetooth class 3.

      Still, this technology could be a good start. Low power wireless computing chips are being introduced and getting more efficient every year, once the market for such things became popular.

    13. Re:12 uW/cm^2 isn't much by bobbied · · Score: 1

      A 5C temperature gradient you can have nearly everywhere. A solar cell only works where you habe light, obviously.

      9 Degrees Fahrenheit over say half a square meter is NOT that easy to come by and isn't going to net you much energy. But my point is a solar collector might be a better choice in a lot of cases because it's a lot smaller size for the same energy. Batteries and other storage devices can take up the gaps and you'd still have size to spare. Not to mention that most temperature differentials are caused by solar heating anyway, at least the ones you would want to be using.

      --
      "File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
    14. Re:12 uW/cm^2 isn't much by iggymanz · · Score: 1

      researchers often have to be like IPO marketers, make big claims or the funds dry up.

    15. Re:12 uW/cm^2 isn't much by AmiMoJo · · Score: 1

      Solar is the default option because it's cheap and pretty good, but often you want to put sensors in places where there isn't much light. Inside buildings, underground (there is a lot of infrastructure down there), underwater, in permanently shadowed areas etc.

      In those cases a small thermal gradient might now provide enough energy to do something useful. And 5C is only the lower limit, often there is a much larger gradient available if you have things like running water.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
  3. I already charge my smartwatch while jogging. by Anonymous Coward · · Score: 3, Funny

    I just leave it at home on the charger.

    1. Re:I already charge my smartwatch while jogging. by Darinbob · · Score: 1

      I do even better, I left mine at the store.

  4. Signals processing for detection of temp gradients by WillAffleckUW · · Score: 1

    This is a very useful signal detection implementation for low temperature gradients, useful in monitoring range fluctuations, and is obviously not useful for power generation beyond that needed to drive monitoring circuitry and software to communicate those gradient changes.

    More like for nuclear power plants or battery cells with temperature operation limitations, or for use in industrial and commercial processes, where you need to run a resin at a specific temperature range to cure it before you apply it.

    Kudos! Good job!

    --
    -- Tigger warning: This post may contain tiggers! --
  5. Anything USEFUL from this tech? by geekmux · · Score: 1, Interesting

    "For instance, it may be possible to charge your smartwatch during your morning jog someday."

    My automatic watch has been running without batteries for years, and all I have to do is wear it.

    Millennial, please sell me again why I need this solution that lacks a problem...

    1. Re:Anything USEFUL from this tech? by Anonymous Coward · · Score: 1

      They might say they use their watch for telling the time about as often as they use their cell phone for making phone calls.

    2. Re:Anything USEFUL from this tech? by Darinbob · · Score: 1

      Not connecting to Facebook and Twitter might be a feature that people are willing to pay less for!

  6. Re:Trump will die in Federal Prison by Rick+Schumann · · Score: 5, Insightful

    You don't want Trump to die. You want Trump to live a long, healthy life, completely and totally discredited and disgraced, and never forgotten as the liar, cheat, and villian that he is. You want him to be a constant reminder to everyone of all the evil selfish destructive things he's done, and be punished for it until his last breath is drawn, and then only when he's died of purely natural causes and is put in the ground (face down, preferably), forever immortalized in the history books as the worst president the United States has ever had, an example to all future candidates of what not to be when you're sitting in the Oval Office, and a warning to future generations of the GOP.

  7. Great a solid state Stirling engine alternative. by technosaurus · · Score: 1

    Way less efficient than a low temperature differential Stirling, but plenty of room for improvement. It could also use the coldness of space as the heat sink using some new tech... can't find the link at the moment, but there was a TED talk on it.

  8. Re:Trump will die in Federal Prison by SuricouRaven · · Score: 1

    If Trump dies, Pence gets in. His policy positions are almost identical, as both strictly follow the Republican platform. Pence would at least be more stable though, and less inclined to insult rivals and national leaders on twitter.

  9. Doesn't take much by foxalopex · · Score: 3, Interesting

    I have a solar powered Casio watch that does atomic clock synchronization every day. It's pretty cool in that it keeps very accurate time and has been running for years and years. All it needs is some ambient office lighting to keep it charged everyday. I've heard the rechargable lithium battery will eventually wear out however.

    1. Re:Doesn't take much by Walking+The+Walk · · Score: 1
      My Citizen WR 100 SolarTech will be 19 years old in September, and still works wonderfully. It doesn't have atomic time sync, but keeps time accurately enough that there's no drift during the 6 months between Daylight Saving settings.

      Another poster mentioned his uses a capacitor not a battery - I'm not an electrical engineer, but don't capacitors drain faster and have a voltage drop-off as they drain? My watch keeps time even without charging - I can leave it in a drawer for 2 months and it's still going smoothly with the correct time when I pull it out. I always assumed that meant it had a reasonably sized rechargeable battery?

      --
      A recursive sig
      Can impart wisdom and truth
      Call proc signature()
    2. Re:Doesn't take much by Walking+The+Walk · · Score: 1

      I decided to google it, apparently there's an entire Wikipedia article on the Eco-Drive watches (I forgot that was the name of the range my model belongs to.) According to the article and its references, it has a secondary battery that will live 20 to 40 years.

      --
      A recursive sig
      Can impart wisdom and truth
      Call proc signature()
  10. Re:Trump will die in Federal Prison by Rick+Schumann · · Score: 1

    Somewhat amazingly, Trump Derangement Syndrome seems to be a real thing. Seeing these types of posts basically proves it.

    You're completely and totally correct, and I agree with you 100%: Donald Trump is in fact deranged. There really needs to be psychological vetting of candidates for public office so we don't get crazies in positions of power.

  11. Re:Trump will die in Federal Prison by Rick+Schumann · · Score: 1

    Word has it actually that Pence is worse than Trump in significant ways since he's notably a Dominionist and would ruthlessly enforce Fundamentalist Christian agendas and not give a fuck about anyone else. Trump seems to want to drag us back socially to the 1940's; Pence would drag us all the way back to the pre-Renaissance era socially.

  12. Re:Trump will die in Federal Prison by SuricouRaven · · Score: 1

    Yet that's somehow still preferable to a president who is just one snappy decision away from starting WW3.

  13. High-Power Thermoelectric Guitar by Trogre · · Score: 1

    I gotta go to bed.

    --
    "Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
  14. Uh, it's already been done by an amateur by mark_reh · · Score: 1
  15. Re:Trump will die in Federal Prison by mark_reh · · Score: 1

    I expect to die chained to a wall in one of the secret dungeons at Mar-a-Lago, right between Bobby DeNiro and John Stewart.

  16. Re:Trump will die in Federal Prison by jwhyche · · Score: 2

    Plus that Pence is a political insider and doesn't have any of the baggage that Trump carries with him. He knows how to talk politics with out antagonizing his rivals. He is true political animal unlike Trump is.

    And that makes him 10 times more dangerous. All the mistakes we are seeing from Trump people with TDS want to attribute to Trump being mentally addled. Which isn't the case, they are simply the mistakes of a political novice. Pence will not make those mistakes.

    Since Pence's policy positions are the same as the Republican party, he has a much greater chance to pull them off. To make it worse Pence thinks he on a mission from God to put women back in the kitchen and gays back in the closet.

    Trump dying or being removed from office will be the worse thing that could happen.

    --
    I read at +2. If your post doesn't reach that level I will not see or respond to it.
  17. Re:Trump will die in Federal Prison by jwhyche · · Score: 2

    Yet that's somehow still preferable to a president who is just one snappy decision away from starting WW3

    Which won't happen. Contrary to popular myth there is no "Big Red" button that Trump can push to start WWIII. Outside of a retaliatory strike where nuclear weapons are first used against the US or her alias, the first use of nuclear weapons by the US government is a complex process that requires several levels of cooperation within the armed forces.

    It was done this way to prevent just such a scenario as people with TDS like to envision about Trump rolling out of bed and pressing the "big red button."

    --
    I read at +2. If your post doesn't reach that level I will not see or respond to it.
  18. Re:Trump will die in Federal Prison by SuricouRaven · · Score: 1

    He doesn't have to press a button. He is Commander in Chief. There are plenty of things he could do to start a war. It's easy for a small regional conflict to grow, pulling in allies on both sides. That's how WWI got started.

    The North Korea powderkeg, for one. His attitude has been unstable - he used to deliberately antagonise the leadership with twitter barbs like 'little rocketman,' then he moved on to proclaiming himself a dealmaker and negotiating the very rough outline of an agreement. If tensions flare up a bit more once again he could easily order an attack, which would result in NK retaliating against SK, and China joining in to defend their own interests in keeping a big buffer area between them and any American allies and inevitable escalation.

  19. Re:Trump will die in Federal Prison by jwhyche · · Score: 2

    Which probably won't happen. Unlike is predecessors, Trump has shown a reluctance to use military force as a first response. It is very doubtful that Trump would order any such attack against NK. More likely things would simply resort to the status quo.

    I know the TDS infect media has so many people quaking in their boots about Trump starting WWIII but you should just ignore such scare tactics. We are doing just fine.

    --
    I read at +2. If your post doesn't reach that level I will not see or respond to it.
  20. Re:Trump will die in Federal Prison by drinkypoo · · Score: 1

    Since Pence's policy positions are the same as the Republican party, he has a much greater chance to pull them off.

    Nonsense. Trump's policy positions are the same as the Republican party's wet dreams; you can tell this is true because they have consistently supported him. The reason Pence has a greater chance to pull off anything is that he seems more reasonable to many people than the commander of cheese. That means he'll be able to fool many moderates (aka fence-sitters) into believing that he is more reasonable than Trump.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  21. Re:Your deep in the tank for trump by jwhyche · · Score: 2

    He's the only President who casually makes nuclear threats against his enemies. (on twitter no less).

    Which directly lead to NK coming to the negotiation table, which lead to a end of the Korean War, which lead to a path to disarming NK of its nuclear weapons. Not bad for a few tweets.

    --
    I read at +2. If your post doesn't reach that level I will not see or respond to it.
  22. Re:Trump will die in Federal Prison by SuricouRaven · · Score: 1

    Which probably won't happen. Probably. Are you happy with a president so fickle that the best we can say is that he probably won't start a major war?