Silicon Supercapacitor Promises Built-in Energy Storage For Electronic Devices 95
Science_afficionado writes "A news release from Vanderbilt University begins, 'Solar cells that produce electricity 24/7, not just when the sun is shining. Mobile phones with built-in power cells that recharge in seconds and work for weeks between charges. These are just two of the possibilities raised by a novel supercapacitor design invented by material scientists ... that is described in a paper published in the Oct. 22 issue of the journal Scientific Reports. It is the first supercapacitor that is made out of silicon so it can be built into a silicon chip along with the microelectronic circuitry that it powers. In fact, it should be possible to construct these power cells out of the excess silicon that exists in the current generation of solar cells, sensors, mobile phones and a variety of other electromechanical devices, providing a considerable cost savings. ... Instead of storing energy in chemical reactions the way batteries do, “supercaps” store electricity by assembling ions on the surface of a porous material. As a result, they tend to charge and discharge in minutes, instead of hours, and operate for a few million cycles, instead of a few thousand cycles like batteries.' The full academic paper is available online."
Meh (Score:5, Insightful)
Re:Meh (Score:5, Funny)
You didn't leave us your number, you insensitive clod!
Re: (Score:3)
And he knows wherever we are, we'll come running
Re: (Score:2)
Thanks for clearing that up, much appreciated.
Re: (Score:2)
Re: (Score:3, Interesting)
Um, I'd think this would be a pretty big deal for computers. I'm not sure if you've looked lately, but the boards are covered in caps made of all kinds of materials, some rather rare. Direct integration with SSDs is the first major use I can see off the top of my head.
Re: (Score:3)
Re: (Score:3)
Not the same thing at all.
The process here is not to replace capacitors, the process it to replace the battery.
Capacitors in a circuit have an entirely different reason for being there than the battery. They are for very short term storage of potential or smoothing of power switching.
If this works, and if it acquires any further attention and attracts funding, it might replace batteries in phones, but it won't replace capacitors in circuits.
Re: (Score:2)
Why not?
Re: (Score:3)
Why not?
Well, I suppose it could, but that is not the focus of this research. What we use now in circuits is outrageously cheap, and fulfilling the need.
What we have now in battery technology is very expensive, and barely keeping up with demand.
Re: (Score:2)
Why not?
Why not indeed. To effectively smooth power fluctuations in high frequency circuits, the capacitor should be as close as possible to the power consuming circuits. What could be closer than integrating the capacitor directly into the silicon on the same die? This could also eliminate the need for external capacitors, reducing part counts, board real estate, and cost.
Re:Meh (Score:5, Insightful)
Capacitors used on your motherboard are mostly there as part of filter circuits and therefore chosen for their unique transient response (i.e. exactly how they behave over time once a voltage is applied). In other words, the discharge rate matters. Can't be too fast, can't be too slow.
Designing silicon based super caps for long term energy storage with slow discharge does not automatically mean that the same tech will replace regular electrolytic caps. I'm not saying they won't, haven't even read TFA, but the design goals are certainly distinct.
Re: (Score:2)
Can't be too fast, can't be too slow
For what it's worth, when designing a switching power supply, you want the equivalent series resistance (or ESR) of the filter capacitors to be as low as possible. This is why you'll often see several capacitors in parallel with each other rather than just one big one: paralleling them lowers the effective ESR.
Re: (Score:2)
The article is pretty light on details, but it's entirely possible that this capacitor type will have a too high equivalent series resistance (ESR) to work well as a decoupling capacitor.
Wrong (Score:2)
To be frank it would be a lot harder for to replace the batteries than the capacitors in circuits but I suspect the journalist wanted to put things in terms that looked simpler - most people know about batteries.
Re: (Score:2)
Posting to undo accidental mod.
Re: (Score:2)
You know that you don't need much skill or money to replace capacitors right? They come off pretty easily.
Re: (Score:2)
I just junked a computer because the capacitors were all leaky and it wouldn't run stable any more. If the chips (CPU, RAM, etc) didn't need capacitors any more because they had the necessary capacitance built right in, and it was solid-state, I think that would be great.
It would be great, if the new supercaps are much more reliable than the current aerogel supercaps.
I've repaired a few MoBo's, (and some other consumer equipment as well), simply by replacing the bad electrolytic capacitors. But if the caps are 'in the chips', you can probably forget about repairing the equipment, unless you can find a replacement chip and are very good at fine-pitch close-quarter surface-mount rework soldering.
Re: (Score:2)
Re: (Score:2)
Like icebike said, this is about batteries. Your caps are leaking and exploding because they're cheap bottom of the barrel shit. Voltages are the same in today's computers as 1990's, but you didn't have the capacitors exploding because they simply used better components.
AND, the capacitors on your motherboard are for circuits that tie different chips together; the chips themselves have caps on silicone (and resistors and other components).
The subject we're discussing isn't more reliable capacitors; that's n
Re: (Score:3)
They've a few other advantages. Their power density is huge, which is great for things that need brief but very high-power surges.
I used a few of them in hobbyist high-power engineering to power the solonoid on an experiment. Five hundred amps DC at 12V with ease. Any attempt to do that with batteries or a high-current PSU would have taken up half the living room.
The experiment was to determine the effect a high magnetic field would have on an arc. It was quite the success: The arc behaved very differently
Re:Meh (Score:5, Informative)
I used a few of them in hobbyist high-power engineering to power the solonoid on an experiment. Five hundred amps DC at 12V with ease.
Not a "supercap "you didn't
They tend to have several ohms ESR. Low ESR supercaps are in the realm of 100m ohms. 500A at 12V requires a total system impedance of 0.024R.
Re: (Score:2)
It helps if you do your math on a napkin and ignore source impedence.
Re: (Score:2)
I did.
12/500=0.024
When I said "total system impedance" that includes the source. Hence having 0.1 ohms ESR in the capacitor supplying the power, its impossible to discharge 500A into anything with only 12V.
Re: (Score:2)
Re: (Score:2)
Welders are AC.
Re: (Score:1)
No, welders come in AC and DC variants.
There are plenty of switchmode DC arc welders around these days. I bought one the other day for $200 and it does 120A easy.
Re: (Score:2)
Find me one that can:
1. Manage 500A DC at 12V
2. Sustain that for the duration (About thirty seconds) of an experiment.
3. Still be affordable to a lowly helpdesk monkey.
Re: (Score:2)
"You can easily get 500 A from a car battery, it is about what is required to start the engine"
"You can easily get 500 A from a car battery, it is about what is required to start the engine"
6 Kilowatts needed to start a car engine ?
Surely you must be thinking of a truck
l can remember cars that you could start with a crank handle (yes the car was made before I was born )
Re: (Score:3)
Even if you manged to get 0.1 farad, it would provide 12 volts for only a few milliseconds at 500 amps.
Your average Lead - Acid car battery will put out 1000 amps into a dead short. Don't try this though, things tend to explode.
Super capacitors tend to have very high ESR, no good for power filtering.
Capacitors and inductors are the two things that don't reduce down to chip levels very well. One of the main reasons your cell phone isn't just one chip. Making these on chip is kind of a holly grael. If this we
Re: (Score:2)
2.5V, 2600F. The big ones. Strapped six in series to get something that could take the voltage.
Re: (Score:2, Informative)
Call me when a supercap has anything like the energy density - by any measure of cubic or weight - as a battery. Till then, they have only niche uses. I've seen various supercap articles that were about tech that was "About to change the world" for how many decades now? OK, sooner or later, they might...I'm still waiting, and I ain't gonna live for as many more decades as I've already been waiting. Till then, I'll drive my Volt.
DCFusor, you forgot one thing to be informative.
The article states their power density around 13wh/kg in one of their diagrams.
While l-ion batteries are up to 1500 wh/kg (common ones are however much less often around 500 wh/kg)
Re: (Score:2)
1500 Wh/kg is more than double what's been demonstrated for any battery chemistry even in a lab setting, and 500 Wh/kg is higher than any shipping lithium ion battery that I've ever heard of, so I'm going to say that your numbers are complete BS.
Re:Meh (Score:4, Informative)
500 Wh/kg is higher than any shipping lithium ion battery that I've ever heard of, so I'm going to say that your numbers are complete BS.
500 Wh/kg is available from lithium batteries, but not from rechargeable ones. Rechargeable ones are less than half of that.
Re: (Score:2)
Except he didn't claim 500 from lithium, but from lithium ion.
Re: (Score:3)
Good thing you didn't supply any "correct" values. Otherwise someone might accuse you of being helpful.
Re: (Score:3, Informative)
Geeze - just hit Wikipedia and (assuming the sock puppets haven't been at it), look at the specific energy:
100–265 Wh/kg
http://en.wikipedia.org/wiki/Lithium-ion_battery
Nothing like trolling by a factor of 10 in the morning.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
That's an energy density, not a power density. (And watts are represented by a capital "W", not a lower case one.)
The article does not contain the phrase "power density", except as a label of the graph. Likewise the paper on Nature. But the limit of that axis on that graph is 10 Wh/kg, and all points are to the left of that limit.
Is there anything else apart from making up units, and making up numbers that you'd like to confess to, to get it off your
Re: (Score:2)
No meh here (Score:5, Interesting)
The thing is, there are many applications where space and weight aren't an issue, but lifetime and power sourcing are. For instance, I have lots of room -- going ten X on the space involved isn't a problem for me in any way, but it'd be awesome to have a reliable, high-power capable storage system to replace the batteries I'm using now, which (a) aren't going to last very long and (b) are severely limited by comparison in terms of the maximum current that can be drawn from them.
The real problem is just an engineering one: we need some standard systems to give us usable energy in standard ranges (12vdc and/or 120/240vac) from ultracap stacks. There's nothing hard about that, it's a market and demand issue, no more. Given the demand, designing the hardware is a doddle.
And of course it's worth noting that UC size is going down while power is going up. Most likely, at some point they will cross the battery line, and that's the time to buy stock in whatever UC company pulls it off.
Plus, instead of poisoning the environment with a dead battery, you can will your UCs to your kids. :)
Re: (Score:2)
.
If you look at energy in a capacitor - Joules = 1/2 ce^2, most of the energy is at the top of charge, and you'd have to be a real swithing supply guru (which I happen to be) to get most of it out, even, at some usable range of output voltage - those parts would
Re: (Score:2)
(noticed what you actually pay for groceries vs the official inflation rate for example?)
Amen. The CPI has risen 30% here in the past 15 years, while the price of milk has gone up 70%. Gasoline costs have more than doubled, electricity costs have almost tripled, housing costs have more than tripled. Go figure.
Re: (Score:2)
I'll just note that some uni students, funded by NASA (IIRC about $5000), successfully fired a 3D printed one-piece titanium rocket engine. This engine cost less than 1/10 as much as the original fabricated engine it is destined to replace, and took a few weeks instead of over a year to manufacture. Another group are building entire housing structures by 3D printing, essentially, dirt. 3D printing is now recognized as critical to long term space habitation, development, exploration, etc.
Of course propone
Re: (Score:2)
Putting this together with the story about the oceans dying [http://www.usatoday.com/story/news/world/2013/10/21/newser-ocean-pollution-overfishing/3143007/], perhaps we could build harvester/m
Re: (Score:3)
I think I recall reading that one of the people who first discovered this problem, (a researcher at U Washington?) has been running research ships out there to see if that can be done. There are two big problems. First, the stuff isn't all that thick - while it sounds like the ocean looks like a venue after a rock festival, the stuff is not very thick in the open ocean. Second, it actually degrades - the sun breaks it down to smaller and smaller bits as the plasticizers get broken down until it's basical
Re: (Score:2)
Re: (Score:3)
Call me when a supercap has anything like the energy density - by any measure of cubic or weight - as a battery.
Too true. The article says these new supercaps are "significantly better than commercial supercapacitors". How much better is "significantly"? Unless it's an order of magnitude or more, it's probably not that big a deal. Why? Because unlike batteries, whose voltage remains nearly constant for a large portion of their usable charge, capacitor voltage starts decreasing as soon as discharge begins. So for optimum usefulness you need to charge a supercap to higher than the system voltage, then regulate it down
What's a luddite doing in science or engineering (Score:2)
keep on livin' the dream! (Score:2)
pic of lithium-based battery power densities (Score:1)
Looks like 200 Wh/kg is industry leading for widely used technology.
http://bioage.typepad.com/.a/6a00d8341c4fbe53ef019b0033dc98970d-800wi [typepad.com]
Re: (Score:2)
Tesla's currently using cells that are around 250 Wh/kg.
Re: Meh (Score:5, Funny)
Pretty high and mighty for a dude that owns a volt.
Tell me when you have a columb.
This will be a GREAT way (Score:5, Insightful)
to provide electricity for my flying car, and my holographic storage disks too!
Re: (Score:2)
to provide electricity for my flying car
I know you need 1.21GW peak power, but what's the total energy requirement?
Re: (Score:2)
I know you need 1.21GW peak power,
It's actually 1.21 jigawatts. The conversion factor from jigawatts into gigawatts is unknown.
but what's the total energy requirement?
Extrapolating from the documentary, where time travel takes about 0.5s, I'd say at least 0.6 jigabrowns, give or take. ;)
Lifted from "Day of the Tentacle" (Score:2)
Dr. Edison's "Super Battery" plans have been stolen!
Re: (Score:1)
and they said imitation diamond wasn't good enough!
Re: (Score:2)
Re: (Score:2)
You might be waiting awhile. I heard he went nuts and married a pigeon.
Re: (Score:3)
The purpose is science, so we know how things work and what we can do. Technology you can personally leverage comes later. ALWAYS. You think the transistors on the ICs, and the ICs themselves, sprung into being in the first microprocessor systems? No, they were lab critters and no more than that, well prior to the 4004 and successors. Crude, hacky looking things of no direct use to anyone. But now look at them.
I agree it's tantalizing to see and hear
Re: (Score:2)
Indeede. For another example, I just saw on History Detectives, the first transistor pocket radio (the Regency TR-1 [wikipedia.org]) came out in 1954, at a price of $49.95. That would be over $500 today. This was apparently the first consumer use of transistors. What-became-Sony's first effort came out in 1957 and didn't quite fit in your pocket.
Re: (Score:2)
The first transistor ever was a crude thing on a lab bench -- it wasn't used in equipment at all. But nice try.
In fact, here it is: http://en.wikipedia.org/wiki/File:Replica-of-first-transistor.jpg [wikipedia.org]
Re: (Score:3)
If I read correctly... a battery on the chip die? (Score:3)
If I read the article correctly, this would allow supercap batteries to be placed on the chip die. This doesn't sound like much, but it would be useful in keeping DRAM refreshed if there is a power outage for a brief bit, or enough juice to dump the DRAM to permanent storage (a small SSD.) If the processor state can be saved as well, this would allow a computer to start right back up almost exactly where it was before.
Of course, this wouldn't be enough power to keep a modern day CPU like a POWER7 running at full tilt for any significant length of time, but it might be enough to get the machine's components to save its state and shut down cleanly.
Then, there are the obvious uses for supercap batteries. A buffer for solar cells that can charge the regular batteries at exactly the power they need is one example, especially if combined with a MPPT controller. If the supercap cells are good enough with energy density, they could even be the primary batteries, although there was a patent application with working prototypes I read mentioned a bit ago [1] about high temperature batteries with a large energy density, and these would be a great candidate as primaries, while the supercaps would be additional storage, a buffer for optimal charging, and giving the ability to continue charging for a little bit of time once the solar panels stop receiving usable light.
[1]: http://arxiv.org/ftp/arxiv/papers/1307/1307.1305.pdf [arxiv.org]
Re: (Score:1)
If I read the article correctly, this would allow supercap batteries to be placed on the chip die. This doesn't sound like much, but it would be useful in keeping DRAM refreshed if there is a power outage for a brief bit, or enough juice to dump the DRAM to permanent storage (a small SSD.) If the processor state can be saved as well, this would allow a computer to start right back up almost exactly where it was before.
Of course, this wouldn't be enough power to keep a modern day CPU like a POWER7 running at full tilt for any significant length of time, but it might be enough to get the machine's components to save its state and shut down cleanly.
Then, there are the obvious uses for supercap batteries. A buffer for solar cells that can charge the regular batteries at exactly the power they need is one example, especially if combined with a MPPT controller. If the supercap cells are good enough with energy density, they could even be the primary batteries, although there was a patent application with working prototypes I read mentioned a bit ago [1] about high temperature batteries with a large energy density, and these would be a great candidate as primaries, while the supercaps would be additional storage, a buffer for optimal charging, and giving the ability to continue charging for a little bit of time once the solar panels stop receiving usable light.
[1]: http://arxiv.org/ftp/arxiv/papers/1307/1307.1305.pdf [arxiv.org]
The paper doesn't suggest you could fit these on the same slab as an IC; I'm not sure we'd want to either -- there are times when it's useful to be able to pull the plug; letting chips have control over their own power sounds like the beginning of the singularity.
Shipstone anyone? (Score:1)
Probably not on-chip (Score:2)
The paper doesn't suggest that putting this device on the same chip as an IC with other functions is possible. But it does indicate a promising material.
This is still at the level of "cool effect seen at microscopic level". It's not yet at "experimental device built, cycled for many cycles, here are the results", let alone "prototype demonstrated".
EEStor and other snake oil salesmen (Score:2)
There was a company, I think it was EEStor, that claimed they were building a Supercap that could be used to electric cars, I believe some Canadian carmaker (zep?) even licensed the tech, but the company never once displayed a working product, claiming that their process wasn't yet patented or some other kind of hand-waving. All very shady, and their press releases seem to be geared towards getting venture capital and producing nothing.
So the question: Is this a real breakthrough that will actually result i
Re: (Score:2)
> There was a company, I think it was EEStor
Yeah. Long and short, they couldn't make it work.
I suspect, but have no way of knowing, that they got great results out of small samples, and believed scale-up was all that was needed.
Similar problems infected the room-temperature superconductor field for a while. People would see sudden drops in resistance and interpret that as small portions of a larger mass going super. Then someone noticed that copper does the same thing not far below zero, yet no floating
Supercapacitor + wax heat sink = super-ultra-turbo (Score:2)
A few months back, I read about a wax heat sink [geek.com] that could allow processors to turbo to very high speeds for very short periods. But...
Unfortunately, dealing with the heat created by sprinting isn’t the only issue that needs to be resolved. Even if wax is up to the task, there needs to be improvements in battery technology before such a system would work in a portable device.
Intel engineer Steve Gunther told Wired, “if I can’t get the current out of the battery it doesn’t help. You need balanced solutions.”
Well, here's the technology that can help that.
Self-storing solar energy? (Score:3)
In fact, it should be possible to construct these power cells out of the excess silicon that exists in the current generation of solar cells, sensors, mobile phones and a variety of other electromechanical devices..
Not sure if they're making a comparison here or proposing an application, but wouldn't it be pretty spiffy if you had photovoltaic cells that stored the energy they collect and convert from sunlight, so it's there to use when you need it? Not sure what the leakage factor for a supercap of this type would be compared to current technology supercaps, you'd still have some energy stored for an hour or two at least, and I think that would be a game-changer for solar power.
Total rubbish (Score:2)
Click to the article, and from there to the original paper. Look for Figure 4. Extract energy density of 5 Wh/kg. Wiki up "energy density" and extract super capacitors at 0.018 MJ/kg. Ask Google to convert 0.018 MJ to Wh. Google returns 5, so 50 Wh/kg.
So this new super device has exactly the same performance as existing super caps.
Power my phone for a week my ass.