Supercapacitor-On-a-Chip Now One Step Closer (ieee.org)

← Back to Stories (view on slashdot.org)

Supercapacitor-On-a-Chip Now One Step Closer (ieee.org)

Posted by timothy on Monday February 15, 2016 @02:41AM from the but-I-just-bought-a-phone dept.

schwit1 writes: In 2010 Spectrum reported a new approach for creating chip-scale supercapacitors on silicon wafers, proposed by researchers at Drexel University in Philadelphia and the Universite Paul Sabatier in Toulouse, France. In an article published in Science, the researchers described how to make supercapacitor electrodes from porous carbon that could stick to the surface of silicon wafers so that they could be micromachined into electrodes for on-chip supercapacitors. Now the same team has finally succeeded in doing just that. In a paper published in this week's Science, researchers from the two initial teams report creating efficient porous carbon electrodes that really stick to the surface of a silicon wafer. They made layers of porous carbide derived carbon (CDC) that are completely compatible with all treatments used in the semiconductor industry, says Patrice Simon, a researcher at Universite Paul Sabatier who has researched porous CDC electrodes over the last ten years and co-authored both the 2010 and this week's paper in Science.

53 comments

Min score:

Reason:

Sort:

And? by mariox19 · 2016-02-15 02:47 · Score: 2

Is it too much to ask for a "And why this is a big deal" in the summary, or do I have to turn in my nerd card?

--
quiquid id est, timeo puellas et oscula dantes.
1. Re:And? by U2xhc2hkb3QgU3Vja3M · 2016-02-15 02:49 · Score: 2, Funny
  
  1 .Do you know what a chip is?
  2. Do you know what a supercapacitor is?
  3. Do you know where your towel is?
2. Re:And? by Cassini2 · 2016-02-15 03:03 · Score: 3, Interesting
  
  I'm not clear on this either. Why would someone want to build a large-area device, like a super-capacitor, on top of an IC? given that the cost per unit surface area of a modern IC is astronomical?
  The question is particularly complex, as we are talking about super-capacitors. Super-capacitors usually have terrible AC characteristics. If we were talking about high-performance ceramic capacitors, then the answer would be to improve the AC performance of the power distribution net, or for RF communications.
  Alternatively, the answer could be to power a portable device. However, a typical capacitor has such a large surface area that it is a roll (or a stack) of many layers. To power the portable device, the surface area of the chip would have to be very large relative to its power consumption. This would be an unusual combination, as power consumption is often a function of die surface-area.
  In concept, any device will have an application. However, I'm not clear on what the application for this device would be ...
3. Re:And? by Anonymous Coward · 2016-02-15 03:04 · Score: 1
  
  Is it too much to ask for a "And why this is a big deal" in the summary, or do I have to turn in my nerd card?
  Very few things are a big deal at all. Look through the last year of Slashdot articles. I doubt you will find much that actually had an immediate and significant impact on your life.
  What the summary implies is that the supercapacitors could be integrated on the same wafer as other components.
  If this is true then it could lead to a bunch of interesting component improvements.
  The thing here isn't that you suddenly can create consumer applications that weren't possible before, but rather that they can be made smaller.
  It could also make it quicker to design new devices since there isn't a need to select an external capacitor and design a layout if the capacitor is integrated in the IC that needs/uses it.
4. Re:And? by CheeseyDJ · 2016-02-15 03:05 · Score: 1
  
  Basically, a supercapacitor on a chip could be the next generation "battery" everyone is seeking for mobile phones etc. They would have the advantage of charging in a couple of minutes (or even quicker), and they could store enough charge to run the phone for days or even weeks.
5. Re:And? by blueshift_1 · 2016-02-15 03:09 · Score: 1
  
  I believe they are saying they can make Super caps smaller / denser than current implementations. This could possibly be a step towards using capacitors rather than batteries, since caps are far better at handling recharge cycles, delivering / receiving electricity. For a long time, there has been this promise of the tiny super capacitor that could replace all batteries - and we are still waiting.
6. Re:And? by Anonymous Coward · 2016-02-15 03:12 · Score: 0
  
  Most likely, it will look good in a large font in a diagonal across the corner of the box - "Now with SUPERcapacitors!"
7. Re:And? by Anonymous Coward · 2016-02-15 03:15 · Score: 0, Informative
  
  A supercapacitor is like a battery that can store 10 to 100 times more energy per unit volume, can discharge faster, can recharge faster, and can recharge more frequently with less degradation. Until recently, supercapacitors were about 10 times larger than batteries with the same capacity, but researchers have just figured out how to fabricate better supercapacitors. And now, apparently, they have a new approach that can do it on regular chips.
  
  Some short-term implications include: the biggest, heaviest, and least robust component in your phone is no longer needed; electronics will no longer depend on child slaves in Asia who solder batteries to chips; RFID chips could use this to continue computing long after the signal is lost while remaining microscopic; the heaviest component in flying drones no longer needs be heavy.
  
  Longer-term implications include: implantable nano-robots will no longer require large amounts of poisonous metals to store a charge; our dependency on the power-grid may be wiped out by the new ability to print super-cheap energy storage devices, creating a new decentralized energy economy; maybe we will soon be able to print super-tiny weapons that can intelligently recognize their targets and deliver lethal doses of energy.
8. Re:And? by Anonymous Coward · 2016-02-15 03:16 · Score: 0
  
  >there has been this promise of the tiny super capacitor that could replace all batteries
  Bullshit "promise" probably came from a wishful-thinking tech enthusiast crowd, or science "journalists" filling their articles up with words.
  In other words, it may be time to lose hope and treat articles with skepticism.
9. Re:And? by peragrin · 2016-02-15 03:20 · Score: 4, Informative
  
  This article Is about installing super capacitors on the board replacing regular capacitors. Not about replacing the battery
  
  --
  i thought once I was found, but it was only a dream.
10. Re:And? by sunderland56 · 2016-02-15 03:38 · Score: 1
  
  Basically, a supercapacitor on a chip could be the next generation "battery" everyone is seeking for mobile phones etc.
  If this were true, manufacturers would be using existing discrete supercapacitors in phones; but they aren't, are they?
  Integrating components on chip merely drives your costs down; it does not bring added functionality.
11. Re:And? by CheeseyDJ · 2016-02-15 03:45 · Score: 1
  
  Basically, a supercapacitor on a chip could be the next generation "battery" everyone is seeking for mobile phones etc.
  If this were true, manufacturers would be using existing discrete supercapacitors in phones; but they aren't, are they?
  Any existing supercap which stores enough energy to power a phone for a day or more is much too bulky to fit inside a phone, so miniaturising supercaps would bring new applications.
12. Re:And? by transami · 2016-02-15 04:00 · Score: 1
  
  Off the top of my head it would allow watch battery type applications. Where as small batteries are either disposable or rechargeable but with limited lifetime and very slow recharge times (hours), these would have very long lifetimes and be rechargeable in a matter of seconds.
  
  --
  :T:R:A:N:S:
13. Re:And? by Antique+Geekmeister · 2016-02-15 04:07 · Score: 4, Interesting
  
  I can think of several good reasons. Decoupling the "droops" on the local power lines from local circuits drawing or providing excess current for signal lizes comes to mind immediately. It's easy to put in a large local capacitor to decouple many devices, but harder to find the board space to put a small, high frequency capable capacitor _right next to_ the power leads that connect each chip to the power bus or to the power plane.
  I've seen a number of complex board designs ruined when a new engineer, or a middle manager, insisted on replacing a set of small capacitors with one large one. I've even myself had to wire in small capacitors, manually, on top of soldered in chips to provide the necessary decoupling. I'll also admit that that was decades ago: I don't have the eyes and hands for that kind of work anymore.
14. Re:And? by ShanghaiBill · 2016-02-15 04:18 · Score: 1
  
  they could store enough charge to run the phone for days or even weeks.
  Not true. Even the best supercaps have an energy density far lower than batteries.
15. Re:And? by fnj · 2016-02-15 04:23 · Score: 1
  
  What would make you think that miniaturizing supercapacitors would in any way improve their energy density?
16. Re:And? by fnj · 2016-02-15 04:25 · Score: 1
  
  I believe they are saying they can make Super caps smaller / denser than current implementations.
  Smaller? Obviously. Denser? Not obvious AT ALL.
17. Re:And? by CheeseyDJ · 2016-02-15 04:25 · Score: 1
  
  they could store enough charge to run the phone for days or even weeks.
  Not true. Even the best supercaps have an energy density far lower than batteries.
  Not true right now, but the article is all about miniaturising supercaps. Miniaturising supercaps to chip-scale could, in theory, massively increase the energy density such that it exceeds that of Lithium batteries. That still looks like it's a long way off though.
18. Re:And? by CheeseyDJ · 2016-02-15 04:36 · Score: 1
  
  What would make you think that miniaturizing supercapacitors would in any way improve their energy density?
  My understanding is that decreasing size and increasing energy density are linked because creating better ways to build better nanostructures underpins both aspects.
19. Re:And? by fnj · 2016-02-15 04:37 · Score: 4, Informative
  
  A supercapacitor is like a battery that can store 10 to 100 times more energy per unit volume
  Utter and complete bullshit. No supercapacitor comes anywhere near the volumetric energy density of even a fair to middling battery.
  You could have found this out in about 5 seconds, even if you are too ignorant to already know it.
  Supercapacitor: 0.06-0.05 MJ/l
  Lead acid battery: 0.56 MJ/l
  Lithium-ion battery: 0.9–2.63 MJ/l
20. Re:And? by Anonymous Coward · 2016-02-15 05:03 · Score: 0
  
  Because you're dealing with three dimensions, think of the surface as x-y with depth being z, due to the wonders of micromachining you can get incredible amounts of surface area in the x-z and y-z planes with minimal use of the x-y plane.
21. Re:And? by Khyber · 2016-02-15 05:14 · Score: 1
  
  Doesn't matter when the recharge time is a matter of a couple of minutes instead of hours!
  
  --
  Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
22. Re:And? by ShanghaiBill · 2016-02-15 05:26 · Score: 1
  
  the article is all about miniaturising supercaps.
  No it isn't. It is about making them stick to silicon.
  
  Miniaturising supercaps to chip-scale could, in theory, massively increase the energy density
  No. Energy density is an intrinsic property. It doesn't change with scale.
23. Re:And? by Bengie · 2016-02-15 05:40 · Score: 2
  
  Cost of an IC has no bearing on this. All they need to do is integrate them into the same package to reduce traces.
24. Re:And? by Bengie · 2016-02-15 05:41 · Score: 2
  
  That's also ignoring that as a cap loses power, the voltage goes down. Batteries have stable voltage output.
25. Re:And? by Anonymous Coward · 2016-02-15 06:05 · Score: 0
  
  Actually, the parent did say "supercapacitors were about 10 times larger than batteries with the same capacity" which agrees with your figure. Of course, they contradicted themself: "10 to 100 times more energy per unit volume"
26. Re:And? by WoOS · 2016-02-15 07:40 · Score: 1
  
  Decoupling the "droops" on the local power lines from local circuits drawing or providing excess current for signal lizes comes to mind immediately. It's easy to put in a large local capacitor to decouple many devices, but harder to find the board space to put a small, high frequency capable capacitor _right next to_ the power leads that connect each chip to the power bus or to the power plane.
  
  To my knowledge it is even worse. You need to decouple the power grid of the individual parts of the IC against each other as well as the power supply (often also on chip for microcontrollers). From what I learned in a previous job most of the empty space(*) on current microcontroller dies is converted into capacitance to buffer the effects of power draw by switching transistors. And discussions sounded as if that was already becoming a limiting factor, so more capacitance (by attaching super capacitors on top of the die) could help.
  But I am not an analog guy so I can't comment on whether the bad AC characteristics mentioned in another post will prevent such usage.
  (*)= There is quite a bit of empty space on silicon dies due to the fact that a) the layout programs are not perfect (or rather the problem of routing the signal lines on the existing metal layers is not solvable without leaving some silicon empty) and b) design rules requiring certain areas to be empty (not sure they are allowed to be filled with capacitance then).
27. Re:And? by mikael · 2016-02-15 08:52 · Score: 1
  
  https://en.wikipedia.org/wiki/...
  Supercapacitors are used in applications requiring many rapid charge/discharge cycles rather than long term compact energy storage: within cars, buses, trains, cranes and elevators, where they are used for regenerative braking, short-term energy storage or burst-mode power delivery. Smaller units are used as memory backup for static random-access memory (SRAM).
  If you start designed computers as self-contained modular blocks, you would want the CPU and memory together with some battery power - that's where the supercapacitors come in. You might want a digital camera with flash memory.
  
  --
  Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
28. Re: And? by MarkH · 2016-02-15 09:36 · Score: 1
  
  Thanks for link I didn't realise with all innovation that lithium is still only 5 times better than poor old lead acid by weight
29. Re:And? by Antique+Geekmeister · 2016-02-15 13:16 · Score: 1
  
  I think you're confused, but perhaps because I didn't go into enough detail for people who've not looked closely at it. We're also looking back into physics and electronics training required long ago in my career, when learning computers involved building them from bare components and involved learning the limitations of transistors themselves.
  A typical digital circuit is tied to ground and power, and tied to input and output circuitry. When current is drawn from ground, and power, to change output signal states, you get power droop and ground bounce because the power lines and ground lines or power planes and ground planes, themselves have capicatance, inductance, resistance, and impedance. The larger the number of output signals being driven and the larger the number of output signals switching state, the more current is being demanded from ground and from power lines. Providing some local reservoir of change at power line voltages, coupled much more closely to the power lines than the power supply itself or the net capacitance between the entire local ground plane and the entire local power plane, helps "decouple" this highly localized current drain from the overall power supplies. This is why they're called "decoupling" capacitors. Also, with modern FET based systems, the amount of current drawn to switch the state of a lot of FET's driven by a modern digital circuit in a high frequency circuit changing state many times per second can be quite large. Without that local reservor of charge provided by a local capacitor which sits directly on the power leads, next to the relevant circuit, the current will have to be drawn from the signal lines all the way back to the main power supply.
  This way lies feedback loops and unexpected digital signal failures as a set of registers all draw current at the same time and drop the local power supply lines before the avlid volatages needed for the digital circuits to perform as expected, but only when a lot of signals switch *at the same time* to a different state. This can be _nightmarish_ to debug: it doesn't show up in simulation unless your simulations are _very_ sophisticated, and it doesn't show up at lower frequency testing. It shows up at high frequency, at the worst possible moments, in high frequency operations when you don't have _time_ to error check every output in a low latency signal processing schema, and it starts showing up erratically dependent on the actual number of signal lines changing at any moment.
  I used to encounter it in low-cost electronics, and I can still remember encountering it as circuits switched from discrete components to surface mount, and designers were overruled in their demands for local decoupling capacitors in the circuit board designs. And yes, I can remember correcting such issues by finding the designer, comparing old and later board designs, and realizing that sets of local small ceramic capacitors had been been replaced with single, larger, polar capacitors which simply could not do the job.
30. Re: And? by mhotchin · 2016-02-15 16:37 · Score: 1
  
  By *volume*. By weight is going to be much, much worse.
31. Re:And? by Anonymous Coward · 2016-02-16 13:30 · Score: 0
  
  Thank you.
These supercapacitors by l0n3s0m3phr34k · 2016-02-15 02:53 · Score: 1

sound perfect for a rail gun.
Applications by Anonymous Coward · 2016-02-15 02:53 · Score: 0

I wonder what these might be used for.
Maybe integrated decoupling capacitors? Or as general purpose surface mount, chip scale supercaps?
1. Re:Applications by Anonymous Coward · 2016-02-15 03:21 · Score: 0
  
  Can they be used to power a Tesla Model S?
No, the missing phrase is... by xxxJonBoyxxx · 2016-02-15 02:53 · Score: 0

No, the missing flamebait phrase is "percentage of women doing X is woeful". (e.g., http://tech.slashdot.org/story...)
One step closer? by Anonymous Coward · 2016-02-15 02:55 · Score: 1

Not very meaningful without a context. If it has taken them years to take this step, and they are still one hundred steps from mass deployment, there are not many reasons to feel all that excited about it.
1. Re: One step closer? by Anonymous Coward · 2016-02-15 08:27 · Score: 0
  
  In other news, every day we are one day closer to everything that is ever going to happen! (Modulo relativistic effects.)
Same Ol' Same Ol' by Anonymous Coward · 2016-02-15 03:18 · Score: 0

It's just another "breakthrough" that will culminate in an unending series of "twenty years from now" prognostications.
Until it's available at Home Depot, it's just hype.
1. Re:Same Ol' Same Ol' by U2xhc2hkb3QgU3Vja3M · 2016-02-15 07:16 · Score: 1
  
  With the Dremel 3D printer available at Home Depot, it means 3D printing isn't hype anymore.
I have no interest by Anonymous Coward · 2016-02-15 03:25 · Score: 0

...until it appears in a bundle on NewEgg
Not an expert but... by Anonymous Coward · 2016-02-15 03:27 · Score: 0

Why can't we store electrons in a container like gas? I know they like to repel each other, but so do gas molecules when they are compressed.
1. Re: Not an expert but... by Anonymous Coward · 2016-02-15 07:56 · Score: 0
  
  This is the principle of static electricity. It is notoriously escape prone and electrifies anything in the vicinity, yet contains very little energy.
  A capacitor holds them in place with an equal and opposite positive charge, which is so much easier.
Higher capacitance per unit area by XXongo · 2016-02-15 03:29 · Score: 4, Informative

I'm not clear on this either. Why would someone want to build a large-area device, like a super-capacitor, on top of an IC? given that the cost per unit surface area of a modern IC is astronomical?
Because a supercapacitor is not defined by being a large area. A supercapacitor can be any area.
Its defining characteristic is a higher capacitance per unit area than conventional capacitors. So, a supercapacitor is actually smaller area than the same capacitance in a conventional cap.

The question is particularly complex, as we are talking about super-capacitors. Super-capacitors usually have terrible AC characteristics.
Some, but not all, applications of capacitors require good AC characteristics.
1. Re:Higher capacitance per unit area by disposable60 · 2016-02-15 06:40 · Score: 1
  
  I can't think of an on-die application that would not need _really_ good AC characteristics to be worthwhile. You're adding [super]capacitance there as the lowest ESR/ESL charge reserve you can make. If you can't beat an MLCC on top of the leadframe, you're adding expense to be 'cool', but not adding performance.
  
  --
  You're looking for quotes? See my journal.
too many links by Gravis+Zero · 2016-02-15 04:00 · Score: 2

there are too many links in the summary to bullshit that doesn't matter. seven links is to many. keep it down to one or two.

--
Anons need not reply. Questions end with a question mark.
If I had to guess why this is a good thing... by mykepredko · 2016-02-15 04:11 · Score: 4, Interesting

At first thought, putting capacitors on the chip means that EVERYTHING for an application could be put on a piece of silicon and not require any interconnections. This could be very valuable for (very) high frequency RFID tags where the chip contains the logic, radio, antenna and power supply good for a few milliseconds of operation without any external components. This could easily halve (or more) the cost of an RFID tag and reduce it to just dropping a chip into the tag's (or even product's) plastic mold - it's been a number of years since I saw the state of the at on RFID tags, but they were to cost $0.15 to $0.25 each in quantity. Without any external parts, this cost could drop to a few pennies.
The other application I can think of are chips which need a defined power down sequence or else be damaged/lose data. The obvious example for this would be in a Flash chip with a write buffer - if power was lost, the contents of the write buffer would be saved to non-volatile storage before it was lost.
Others? I think the RFID tag is probably the application where this technology would be most valuable.

--
Mimetics Inc. Twitter
1. Re:If I had to guess why this is a good thing... by bluefoxlucid · 2016-02-15 08:16 · Score: 1
  
  I had a weird dream the other day involving a high-temperature superconducting film of carbon nanotubes with a tungsten atom inside. The film was woven back and forth on a nanoscopic level, sort of in that compressed zig zag manner like corrugate; that, taken as a linear unit, was then drawn in a trace inductor scheme (a spiral of wire), making a trace of superconducting air-gapped tungsten impregnated carbon nanotubes.
  Apparently my brain was trying to work out implanting a microscopic nano-super-indacitor into a contact lens, providing field power reception and storage in one unit (capacitor AND inductor) to draw overlays directly onto OLED display positioned over the iris.
  Waiting for them to move from supercaps on chip to supercaps on silicone hydrogel.
  
  --
  Support my political activism on Patreon.
Seems counter-intuitive on many levels. by spads · 2016-02-15 05:52 · Score: 1

Capacitors are for storing up change, and ICs are supposed to be very low voltage/current requirements, so what do you need that for? Also, I thought surges such as a capacitor might produce are generally damaging to the micro-circuitry of an IC.

--
Bukowski said it. I believe it. That settles it.
1. Re: Seems counter-intuitive on many levels. by Anonymous Coward · 2016-02-15 08:07 · Score: 0
  
  The faster the chip runs, the more power, especially surge power, it needs.
2. Re:Seems counter-intuitive on many levels. by Qzukk · 2016-02-15 10:56 · Score: 1
  
  I thought surges such as a capacitor might produce are generally damaging to the micro-circuitry of an IC
  There's an application for this then: tamper-proof chips.
  
  --
  If I have been able to see further than others, it is because I bought a pair of binoculars.