Home Made Supercaps

Home made supercaps 1 Year, 10 Months ago Hi everyone,

I decided to register on this forum because I noticed that there are a few people contemplating to build their own supercaps and I think that I may be able to help.

More than a year ago I became intersted in supercaps andI spent a lot of time researching the subject and making prototipes.

The good news is that it is possible to make excellentDIY supercaps with water based electrolytes.

Unfortunately I don't have the time to go into all the detail, but I am happy to answer any specific questions regarding my supercap work.

My initial prototypes contained about 1g of AC and yielded capacitances up to 25F with maximum voltage of 1V. My favorite electrolite is Potassium Hydroxide (KOH) at a concentration of 5M.

The first picture shows some leftover bits from my early experiments. Ont he top row there are two sets of current collectors. The first set is made from nickel plated steel (recycled alkaline battery can). It worked very vell.The second set is stainless steel and is ok for starters.

The round black patches are the carbon electrodes, about 1.2mm thick each.

Also there is a separator there, made from copy paper.

The second picture shows an assembled cell without the electrolyte. The spring clips are needed to maintain electrical contact between the collectors and the AC electrodes.

The carbon electrodes are made from a mixture AC, Teflon and Carbon Black.

OK, here is my electrode making procedure:

1. Put 8g of Activated Carbon granules (the size of rice grains) and 4g of Virgin Teflon into a coffee grinder and grind the mixture for about 10 minutes.2. Add 1g of Carbon Black (Lamp Black) to the mixture and grind for a couple of minutes more.3. Take out 4.5g of the mixture from the grinder and set aside.4. Add 4.5g of new AC granules to the mixture in the grinder and grind for 10 minutes more.5. Take out the 9g of mixture from the grinder and repeat point 4 with the other 4.5 grams of mixture.

At this stage you should have a total of 18g of carbon mixture ready to be pressed into electrodes.

I turned up a simple brass cylindrical press tool for compressing small amounts of the dry mixture.For the press I used a toolmakers vice with smooth parallel jaws. Not much force is needed.

Don’t make the electrodes any thicker than 1.5mm.

The Teflon breaks down into microscopic fibres that bind the carbon grains together. If you can get fibrillating Teflon powder then you can start with AC powder, rather than AC granules.The Carbon Black increases electrical conductivity between the carbon grains and it is NOT essential.

No need for heat. The fibrillating properties of Teflon is what binds the carbon together.But don't expect to get something very hard. The electrodes crumble or break quite easily.

My electrodes were (as seen on first picture) 22mm diameter and 1.2mm thick and this arrangement produced over 20F capacitance.With two such units in series I could light an LED for several minutes.

I made a 1.088V power supply. A complete improvisation, but it works very well.The impedance mismatch between the transformers limits the output to 140mA, which is perfect for charging small caps.I had to unwind some of the secondary of the second transformer until I had the right output voltage.The two diodes across the output act as a zener diode and I had to try many different types until I

had the right voltage drop.

The relevant literature gives AC a capacitance of more than 70F/g in aqueous electrolyte. Much less in organic electrolyte but still the organic wins when it comes to energy density due to the higher working voltage.My latest capacitor contains 6g of AC and has a capacitance of about 150F. The carbon pellets have a diameter of 50mm and weigh about 3g each.I work out (roughly) the capacitance by measuring the time it takes to discharge the capacitor into a resistive load and then plug the numbers into the formula that gives the capacitance.For example, my 150F capacitor takes 91 minutes to drop from 1V to 0.5V when connected to a resistor of 68 ohms.

Carbon supercaps don't use a physical dielectric for storing the charges, you must be refering to the separator.As I said before I use copy paper but I tried thin cardboard as well with good results.

I guess I still don't fully understand the basics of ultracaps. How does charge separation occur if you use a separator that is porus? Wouldn't paper soaked in a liquid electrolyte be conductive? I know in electrolytic capacitors a porus separator is used, but the dielectric is formed by running a current through the device till an oxide layer forms on one plate. So in an aluminum electrolytic the dielectric is a tough layer of aluminum oxide on one plate. Which is why they are polarized.Why do ultra caps have a polarity?

If you take two non-reactive metal plates and partially immerse them into an electrolyte you get yourself a ‘supercapacitor’ that can be charged to just below the breakdown voltage of the electrolyte. EACH metal plate forms an Electric Double Layer Capacitor. If charged, such a capacitor will instantly discharge if the two metal plates touch. Of course this ‘supercap’ would have a small capacitance due to the small surface area of the electrodes (metal plates). Now if we apply a layer of activated carbon to the metal plates we increase the surface area of the electrodes by something like a million times, but the condition of the electrodes not touching remains, hence the need for a separator. One gram of AC has a surface area of hundreds of square metre. Also remember that a unit supercap always consists of two EDLCs connectedin series by the electrolyte.I can think of two reasons for ultra caps being polarised. One being non-symmetrical construction and the other is probably ionic intercalation that occurs after the first charging.

Your capacitor is built by :Making two currency collectors made from nickle-coated steel.

1. attaching an electrode made from activated carbon, teflon, and carbon black to each collector.

Are the electrodes attached to the collectors in any way (glue)?

Why must the thickness of the electrodes be no more than 1.5 mm?

Putting the two collectors together with the electrodes facing each other and separated by a piece of copy paper.

2 Inserting the device into a small plastic bag.

Is there anything else that goes in the bag? Like a solution of some kind?

Clamping the entire thing with clips to maintain contact between the electrodes & the collectors.It seems very simple to me. Is there anything I'm missing?

Joe

Your understanding is correct. Indeed you need an electrolyte in the small plastic bag.My favourite one is Potassium Hydroxide at 6M concentration.It takes a while for the electrolyte to be fully absorbed and it also deteriorates by reacting with CO2 from the air.If you can, make a sealed cell.Also by "copy paper" I realy mean tracing paper.Any paper like product is ok if it is capable to absorb the elecrtolyte and it does not desintegrate. Normal paper is made of cellulose fibers and it falls apart in alkaline electrolyte.

i only just found out about these capacitors and i am quite interested in making one but i have quite a few questions,

1) In your step by step guide you cut the separator the same size as the electrodes, wouldnt electicity leak around this and cause electrolysis on the two plates.

2) Is there a substitute for the teflon because this is the most expensive bit, i thought about grinding the AC into a powder and then adding the electrolite to this mix to make a paste, then you could roll it like a normal capacitor, would this work.

3) Whats the best type of metal for the backing plates for the electrodes, you used nickel and stainless and that is good for starters and thats 18g mix wih about 20F. You said that youve got 25F with just 1g of AC what metal is those electrodes or what are you doing differently.

4) Does the concentration make much of a difference because i have some KOH flakes from an electrolysis project and i can make any concentrate, it would be great to know the best concentration.

5) Does the thickness of the electrodes make much of a difference and why no more than a thickness of 1.5mm.

Thanks for sharing this with us, it would also be great if you could show us some more pics of other caps youve made. I never thought that batteries might be replaced by a simpler and safer solution.

1) I cut the separator a bit larger that the electrodes to make sure there is no physical contact between the electrodes. The separator has to be ‘leaky’ for ionic conduction. Electrolysis starts above 1V and that is what limit’s the usable voltage range.

2) I only know of another usable binder and that one is even harder to get than Teflon. It is called PVDF.You can certainly experiment with Teflon powder but if you want it to act as a binder you have to fibrillate it mechanically.

3)For the collector plates you need metals that do not react chemically with the electrolyte that you use.I don’t know how you associate 18g of electrode material with 20F. I get about 25F per gram.

4) I don’t know what the best concentration is but from information I found on the internet 5M or 6Mis ok.

5) The thicker the electrode the higher the internal resistance of the capacitor and that is what determines how fast you can charge or discharge it.

I have posted some images on this site that have disappeared along with some of the explanations I provided. I think this site had some sort of a crash and lost some of the material.So here is again my 150F sealed super cap containing 6g of active material driving a super bright LED through a DC toDCconverter. It can do so only for about two and half minutes because the circuit stops working below about 0.85V and it is not very efficient either.

Stefan these caps of yours are great. Working on my own, will post results when I have'em. A couple more questions for you: Do you happen to know if copper reacts with KOH electrolyte?I'd love to know what kind of max amps you see across a short. I want to work toward multi-cell caps capable of a couple hundred amps at frequencies above roughly 2kHz, single pulse.

My first very quick'n dirty cap with water electrolyte, saran wrap separator, and copper plates does about 45

microamps at 0.5 volts. I just wanted to see some carbon store a charge, so I call it a success. Now I'll move to a little more rigorous manufacturing. The caps you've posted have been hugely helpful in providing a simple, easy to produce starter model.BTW, anyone looking for teflon of the diy variety might try your local hardware plumbing department. Look for teflon washers, fittings, etc. Maybe somebody can figure a way to effectively shred teflon tape.

Oh yeah, almost forgot. I used a very little bit of concrete mix to bind my carbon, no teflon. Resistance through copper plate and carbon layer, about 18 kOhm. Completely unoptimized, I'm sure this could be greatly improved. Will post as I learn. Copper does react with KOH electrolyte so it is not suitable for collector plate.To get an idea of the internal resistance I measure it across the collector plates with the (dry) elecrodes pressed into contact (NO separator) and WITHOUT electrolyte.Usually I get several ohms.Max current is a bit more complicated than that though.

Thanks for your response, Stefan. And your patience with someone who's forgotten all of high school

chemistry.

When I get to testing my own caps I think I'll measure peak amps with timed discharges through a single resistor, repeated with resistors of decreasing value until heat becomes a problem. Calculate current for each discharge cycle, plot, and extrapolate.

Have you tried charging one of your caps before adding electrolyte? With a suitable separator you could use higher charging voltage and see how capacity is effected. I will be trying this with my own caps as soon as a busy schedule allows. I realize you may have zero interest in trying this - but I

gotta ask to find out.

Another quick question, I know this was mentioned by a previous poster, I'm hoping for greater detail. I'm curious about teflon-bound electrodes, somehow I can't see them having any structural integrity at all. So I guess I actually have two questions on this: What kind of handling can the pressed carbon/teflon electrodes survive, and how thin could you make an electrode of this type before it's too weak to be manipulated in hand-assembly of a cap?

I'm finding that handling/controlling/integrating the AC is *the* first real hurdle in building these things...

I’m located down under in Melbourne, Australia (originally from Transylvania). I certainly hope to make units well over 4KF as my interest is in power storage for general use and I will need very large cells for that. Low ESR is not important for me at this stage as charge/discharge time would be measured in hours.

CJ,Indeed, a timed discharge into a resistive load is the way to go when determining capacitance at various discharge currents.

Without electrolyte you have a normal capacitor and I don’t expect the internal surface area of the AC to play a role - but don’t let me hold you back.

The AC electrodes with Teflon binder don’t have much structural integrity although they are not too bad if the binder is well fibrillated. When the Teflon is fibrillated it becomes invisible even at high magnification.What I don’t like about Teflon is that it is very hydrophobic and it makes electrolyte impregnation of the electrode rather difficult.I will try to make binderless electrodes by vacuum sintering of AC powder and hope to get a material with much less electical resistance and good structural integrity that will allow me to make thick electrodes.

Sintering, yes! Would be great to hear your ideas on that, or maybe you could point to a good info source? I've also been thinking that some kind of sintering process is probably the way to go.

Tried a primitive version, one of those 'what would happen if...' little experiments. I tried to mix AC powder with liquid (hot) solder. No joy, it looked like the oxide layer on the solder - fast oxidation at that temp - prevented the AC from absorbing any of the the liquid metal. But in a vacuum... perhaps the evacuated pores of the AC might soak up some of the solder. Other thoughts I have on this are higher temps or doping (borax is an option) to lower viscosity of the solder. Dunno.

Thanks I won't I can't completely agree with your statement. A 'normal' capacitor still does not have the surface area advantage provided by AC. A 'dry' supercap would function the same as a normal cap - non-polar and no electric double layer (assuming insulating separator). But I suspect that it would still have far greater capacity than a typical cap, due to the AC.

There is one major catch here, I think, that could prove you right. I now understand how ions are held in place within the AC, and I'm not sure the same forces apply to single electrons. Given the difference in size between a gigantic ionic molecule and an electron, it might be that the AC simply can't 'grab' those unstable little electrons. Anyway, I do intend to make and test various dry caps to find out what the real deal is. But first there's the electrode construction problem to sort out, and at the moment all I have time for is the occasional long-winded post here.

Vacuum sintering sounds very promising...

Try this site for info on sintered AC:http://reticlecarbon.com

...and their US patent 6350520

I'm building a high vacuum system for coating optics and it will be very handy for AC sintering experiments.

Thanks again, Stefan. The Reticle site only mentioned 'proprietary' processes without going into any tech detail. The good stuff is in their patents. Following is a quote that conveniently contains reference to various of their patents...

From: http://greenpatentblog.com/2009/04/15/reticles-carbon-consolidation-process-produces-high-capacitance-electrode-material/

This text:QUOTE:Reticle’s manufacturing process and resulting carbon material are protected by a family of four U.S. patents: U.S. Patent Nos. 6,350,520 (claims granular active carbon material made by a high temperature and pressure process), 6,511,645 (claims a process for producing carbon material by consolidating amorphous carbon using elevated temperature compression), 6,544,648 (claims a processed carbon material consolidated under elevated temperature and pressure) and 6,787,235 (claims a processed carbon material consolidated in a hot isostatic press under elevated temperature and pressure).Hi guys,I’ve finally decided to register and to get hot news about your progress.Stefan, did you have tried sintering? What are results?You can use polyaniline (PANI) as a binder. It is a conductive and pseudocapative material, but you have to use acidic electrolyte.

Thanks for the binder sugestion.At this stage I want to get away from binders tough. Besides I'm not a chemist.My vacuum plant will become operational in a couple of weeks, so I will start my sintering experiments after that.

Hey guys, what’s going on, no one is experimenting any more?

I abandoned the idea of vacuum sintering after a disastrous first experiment.Confining the AC powder in vacuum just seem too difficult. Fine particles escapethrough the venting holes of the compression mold and contaminate the oil in the vacuum pump.

So I went back to using binders and made some good progress in using epoxy resin as a binder.See attached pic for some examples. The thinner disc is made from fine AC powder and it contains 33% resin. The thicker ones are made of a mixture of grit sizes and contain 20% resin.

So far I only tested the material with the 33% resin content and found it to have about 1/3 of the specific capacitance of the material described earlier in this thread - that used Teflon as binder. Not

surprising considering the high percentage of the binder.

Cool!resin is a bit cheaper and mayby the plates i a bit more mecanicly stable? want to build a small 2 seat car electric car with a onbord generator (48V)

Now i realy want to build a big supercap"battery, that happens if I stack AC/resin plates with tracing paper in between each layer?breakdown voltage on ac/resin cap?

Hi guys,

Stefan,We have similar problems with vacuum and very fine particles materials.We are using a tube furnace for sintering in CO2 or nitrogen. So you can try sintering in the inert atmosphere, not in the vacuumWe have tried different kinds of binders (silver and carbon paste, conducting glues), but PTFE seems to be best for common experiments. There are some interesting results with conducting polymers, but I will have some reliable results in several months.I have one tip for you, AC and carbon blacks are very hydrophobic materials. Try to boil them in the water with alcohol or hydrogen peroxide. After that add binder etc.

your idea sounds great, but it will be very hard to build supercap with energy and power needed for you car.

For testing our materials, electrolytes and separators we have bought nice “toy” http://www.el-cell.com

Don’t get carried away, my epoxy bonded material is very low performance stuff.I just started testing a cell to see how a thick electrode made of this material behaves. Preliminary measurements indicate a rather disappointing 100F for 18g of AC content. Mechanically the material is very good being a real solid that can be bonded to the collector plate and no contact pressure is needed afterwards.

PetrDvorak,Thanks for your suggestions. I will certainly try the boiling to improve wetability.

My vacuum system is for optical coatings and I’m not going near it with AC materials ever again. What I don’t like about the PTFE-bound AC is that it seems to require external mechanical pressure for optimum performance and I don’t think it can be used in the form of thick electrodes.Please keep us up to date with your progress.

I like your new toy.

ok, then a would need around 45 kg AC in "my" cap

I have a hydralic press i thing I´l try to do a couple of plates myself and press out as mutch epoxy as possible and see what happens.how can i connect these in a series? If I have got everything right I need too isolate every "cell" from each other because it will not work if the electrolyte is in contact in meween each "layer" of "cell"Edit: why did I not think of boiling my self? I know for a fact that AC works a helluva lot better then you want to purify alcohol if you boil it first

Is it possible that the epoxy is affecting the performance of the AC because it's encapsulating at least some of the granules completely?I was discussing this with a friend of mine, and he had what I thought was a novel idea. Apparently you can get hardened epoxy to soften with a little heat. What about mixing hardened epoxy dust with AC and pressing it under heat? His specific suggestion, which I thought was really neat would be to use the leftover dust from carbon fibre cuttings - you get epoxy granules with a bonus of exposed conductive carbon fibre at each end.

Another one that just occurred to me when I was eating lunch - what about mixing the ground AC with something like gelatin and casting the plates? This would form a collagen matrix that's ion permeable, but I have no idea how it would stand up to the various electrolytes.Maybe not gelatin, but I know there are a lot of other types of naturally derived polymers (guar, agar, etc) that could be used. Any thoughts?

Or cornstrch, chiting, etc. But After doing a little reading, it seems that many of these may degrade over time and in the chemistry.

What about a silica gel to bind the AC, plus an acid electrolyte, the same as a gel-cell lead acid battery?

Thank jndale for your suggestions.There are many things that can be tried but the bottle neck is in the testing of the various materials.Take for example my thick electrode test cap with 20% epoxy binder. After two days of soaking in the electrolyte I measured it at 100F, after another two days of soaking it came in at 200F and today, after 6 days total soaking I measured it to be 230F.

Wow, very interesting stuff!

I wanted to ask if there is a standard way for measuring the internal resistance of an electrode?I mixed up some gel based electrodes based on purified water, AC, and gelatin. It seems to be holding up well to 5M KOH, which was surprising to me, but I cannot tell how good of a conductor it is.

I measured the resistance with my probes based 1mm apart into the gel, and it's far more conductive than pure gelatin with no AC, but not nearly as much as if I just stick the probes into pure AC powder. Is there a standard technique to measure the internal resistance of the electrodes so that I can communicate any findings in a way that's useful to others?

Yes,there is method to measure resistance - Impedance measurement but I'm not sure, if it is possible to measure it at home

For testing various dry state AC materials a Four-Point resistivity meter should be useful but I think that would be an expensive toy.Anyway my thick electrode capacitor is up to 380F and that is more than 60% of what I would expect from PTFE-bound electrodes of the same weight.Of course the ESR is quite high. I did the test with 11mA average discharge current. It took 3 hours and 42 minutes to drop from 1V to 0.5V.

I know this is not quite in the spirit of what I set out to do (use only common materials) but the result I got with my latest binder is so good that I might consider using it if I don’t come up with something equally good but less exotic.It is a thermopolymer powder called HALAR 8014 used for powder coating. I used it in a proportion of 28.5% and it produced a nice hard material that seem to retain much of the micro porosity of the AC.The small device in the attached picture contains 3g of the material and it came in at 90F after two days ofconditioning. That is better than what I got with my PTFE-bound material.

Lotsems, read the thread from the start.

I have to ask what inspired that brainstorm. I mean, how did you pick that particular one, instead of polyethylene, polypropylene, other etc? I mean, there are at least 32 different kinds of Halar alone, just from one distributor! Did you find out it is conductive or something? Was it just trial and error, or is there a method?

Similiarly, for both the epoxy and Halar 8014 mixtures, was the amount of binder just trial and error, or is there a way to calculate this? Or did you just start by mixing it until is "just looked right"? Also, are you using fine AC particles, or a mix of different sizes? I've found that the large granules make processing to be a pain if you need to cut or shape the electrode, but having only small particles seems like it would raise the internal resistance since there is less contact between the granules (based on the few that I've made so far).

Also (last question, I promise) when you formed these electrodes, were you applying pressure, or doing anything else to force the AC particles to be in better contact with each other?

No method to my madness I’m afraid.More than a year ago I was trying to get some PTFE micro powder through a powder coating business,unsuccessfully, and the man gave me some DYKOR 830 which is a PVDF powder and the HALAR 8014 because that is what he had in stock. I read somewhere that PVDF can be used as binder but I had no success with it so far.I don’t think there is anything special about the HALAR 8014. Simply it is the first thermopolymer binder that worked for me and I think there are many more that would work. I will try to get some FBA powder which is just dry epoxy. It probably will be just as good as the HALAR considering that my normal epoxy bonded test cap reached 440F today with a discharge current of 20mA.For making the electrodes I weighed 5g of AC granules and added 2g HALAR, then I ground them together for 60 seconds total grinding time in 10 sec intervals. The ratio was just a guess intended to be a starting point.I have a 50mm diameter compression mold that I made. It is a very simple thing that consists of a 10mm thick base plate, a short cylindrical ring with a 50mm ID and a piston with a flange. Up to 12 screws (M6) can be used to bolt the whole thing together.I used aluminium foil at the bottom of the mold, and between the piston and the powder to stop the material from sticking to everything. I used 6 screws to compress the powder and I heated the whole thing to 250 degrees Centigrade. Afterwards I had to use emery paper to grind off the aluminium foil from both sides of the disc.

Ok, I have a dumb question for Stefan, Petr, or anyone.I've been experimenting with different electrolyte solutions and metal collectors, but have not having much luck doing what I believed would be a simple test. I'm sure I'm making a basic mistake, but I can't figure it out.

To avoid difficulties with making binders, I've simply made small bags out of some stretchy polymer netting, and used a nylon zip-tie to hold the granules of AC inside the bags, and tight to the metal electrode (see the picture). In this example, there are 5g of AC granules being held tight to a nickel rod in each electrode, and they are separated with tracing paper. I used an ohmmeter to verify that there was electrical contact between the nickel rods and the AC on the furthest part of the bag (it was about 170 ohms at most).

I covered the electrodes with 0.1 molar KOH solution (100 ml distilled H20 and 5.62g KOH) and applied 1.0V to the electrodes.I know this is not the best setup, since the AC granules are fairly large - I just want to see some charge being stored. But I don't seem to be storing any charge at all! There does seem to be current flowing into the unit though. In all of the previous experiments I've done, that means that there is a chemical reaction taking place - usually the carbon is destroying whatever metal I've decided to use as a current collector.In fact, I was having such bad luck with the carbon reacting with the metals, that I decided to just so a simplified version of Stefan's example to make sure I could do a basic experiment - but it's not working!!Any thoughts? I'm sure it's something obvious, but I can't tell what!

Thanks in advance.

All I can suggest is that you increase the concentration of the electrolyte and also the charge time. It can take days for the electrolyte to properly penetrate the nanopores of the carbon. Until that happens self discharge can be very high. I don’t even bother taking meaningful measurements without at least one overnight charge.I made a new test cap yesterday using HALAR bonded carbon molded directly onto the nickel current collectors. The electrodes weigh 7g each and I expect it to achieve about 420F. This morning, after overnight charging, it delivered 20mA for about one and a half hours by the time it dropped to 0.5V. That means a capacitance of about 290F, well within expectation. By tomorrow it should be close to 400F. It can already deliver 250mA into a 2ohm load at a voltage of 0.75V, meaning that the ESR can’t be more than 1ohm.

I agree with Stefan, you have to use KOH with higher concentration (5 or 6 molar). Try to make thiner electrode. You don't need separator in this configuration, leave small gap between electrodes. Or, pull electrode and separator together.I just remembered - cover your supercaps. CO2 can be accumulated in the electrolyte and reduce conductivity.

Thanks for the thoughts guys, and congrats on your capacitor, Stefan! One day I'll be building nice ones like that.

I think I found the problem, or at least *a* problem. When I built the electrodes, I may not have rinsed the AC granules off well enough, and there was a little bit of carbon dust floating in the electrolyte. I don't think it was a problem until it started settling on the bottom and forming a

conductive path between the electrodes. A little filtering seems to have improved things greatly. I didn't show it in the picture, but the entire apparatus is inside of a plastic bag to prevent interaction with the CO2 from the air. Also, I had to use the tracing paper because it's a tight fit and the electrodes are pressed against each other (which is what I wanted).

Unfortunately, I only bought 30g of lab grade KOH because I wasn't expecting to use KOH chemistry for my real device, this is the only reason I was skimping on the moles. I just bought it to verify that I

could build a known working device. It seems to be working much better now I wouldn't think that too weak of a mixture would cause problems though - it just wouldn't be as capacitive if all the ions were pulled into the AC, right?

I don't recommend you to use organic electrolytes (non-aqueous) at home. It is little bit danger and expensive. It is also better to operate in dry and clean atmosphere. We are using the dry box with argon atmosphere...

usable solvents:propylene carbonate (PC), acetonitrile, ethylene carbonate ...salts:LiClO4, TEABF4 ...Cheapest way is 0.5 M LiCLO4 and PC.

Be careful with acetonitrile !!!you can use Ni or Al collector - same as in the KOH electrolyte.

my motivation using organic is the energy boost (14 times more then hydro), since we can use 3V instead of 1V.is there a hydro way to get 3V per layer?is there a cheaper and easy to manage electrolyte with more than 1V limit per layer?

I understand to your motivation. I'm using only organic electrolytes.There is no way how to get 3 V with aqueous electrolytes. In my opinion the highest usable voltage at electrode in aqueous electrolytes is in lead-acid batteries. You can get potential about 2 V per cell

at Pb electrodes. But who wants use Pb in supercap...KOH is a good electrolyte with high conductivity, in fact it is used in the commercial aqueous supercaps and alkaline batteries.

Not to change the topic, but how does an Aluminum collector work with KOH electrolyte, Petr? I have had good luck plating steel with Nickel, but I would prefer to use Al. How can this be done without the KOH allowing the water to destroy the aluminum?

Or are you referring to KOH in an organic electrolyte?

jndale, in the industry, of some companies, Al is the collector and it is possible with aqua-acid but with KOH i know that they need to squeeze the layer to overcome the potential barrier over the alumina coat that is created on the Al due to KOH. in that manor, you loose about 0.3V from the energy but have a cheep Ucap .

Peter, why exactly is organic more "voltable" then aqua?is methanol and such not good solvents for salts?If you have any insights into how aluminum collectors can be used in a supercap, I'd really love to get that information. I have seen alumina coated aluminum in electrolytic caps, but not with a KOH electrolyte, and never with carbon present.

I tried making several supercaps with chemistries based on the ones from electrolytic capacitors (there are lots of books available on this, unlike supercaps), but they all failed because of their aluminum electrode. I found that in this situation, carbon behaves like a very noble metal and 22corrodes the aluminum. Likewise, I have not found any aluminum material that will resist the effects of KOH over a significant length of time.

Perhaps there is a technique to do this with heavily anodised aluminum, but if the oxide is thick enough to resist the chemical attack, it may no longer be conductive enough to make a good current collector? The insulating alumina is ideal for an electrolytic cap, but seems to be the opposite of what you would want in a supercap.I would much prefer to use aluminum than the nickel coated steel sheets I made. If you can enlighten me as to what I might be missing, or what I did wrong, I'd appreciate it very much.

I can't see why shouldn't it work.... :\Maxwell is using Al foil. http://www.ansoft.com/firstpass/pdf/CarbonCarbon_Ultracapacitor_Equivalent_Circuit_Model.pdf

it works fine for me. as i said there may be a need to squeeze the layer. if you tried it you should see if that aluminum foil is truly aluminum. if it is, there might be a problem with the electrodes or electrolyte.I use a one time baking plate cut collector - more thick, which for robust liking people like me is better than Al foil. foil is used at the industry so no bad choice in that.BTW - there's no better than nickle except gold. if you don't want to coat a collector with nickle, just recruit a few nickle coated coins - like, how unsurprisingly, the American Nickle.another tip for you coating people - its uneconomical to coat a whole surface with nickle - the carbon is conductive enough at these dimensions. just coat a net (1 cm wide gap/mesh is good enough) and have it as a collector.still waiting for Peter's answers and insight

I can see how this may work in an organic electrolyte (like the ones I assume Maxwell is using), but in

an aqueous solution, my aluminum always ends up destroyed. What are you using for an electrolyte? I've tried KOH, sodium borate, boric acid, and sodium bicarbonate.

all that is not KOH has some disadvantage - i use only KOH.Maxwell uses organic but other companies that uses aqueous are using Al or conductive PVC.try the baking plate material i think it'll solve your problems.

Sorry guys,

Al in KOH isn't good idea But it is used in nonaqueos electrolytes because of price.Why are organic more "voltable" ?I don't know why, some organic solvents have higher decomposition voltage, you can find it in some tables.

Methanol - I don't know, I have never read about using it as a solvent in supercaps. It is also a little bit useless-you work on the air, which contain H2O and I'm not sure, if you can buy methanol without water. This water always decrease the potential. So, if don't operate in the dry atmosphere, don't think about other electrolytes.

Al in KOH isn't good idea

Oh, I think it's a great idea! I've filled large balloons with hydrogen this way (actually NaOH, but it works the same in this case). Tie a fuse to it, light it, and let it fly! Who says you don't learn anything

useful in college! I figured KOH + Al could be used in a non-aqueous situation, because the Al isn't really reacting with the OH-, it's just that the OH- allows the Al to react with the H2O by disrupting the formation of the protective oxide layer. The bigger problem is with galvanic corrosion from the carbon though. I think it rules out using aluminum completely in an aqueous situation, but my mind is open if anyone has any ideas!> I'm not sure, if you can buy methanol without waterYes, you can buy anhydrous methanol and ethanol, but it's very hygroscopic and won't stay anhydrous for long unless you can work with it outside of the air, like Petr says.

> Why are organic more "voltable" ?Water with ions begins to electrolyze at about 1.2V. This will be your upper bound unless you are ok with chemical reactions taking place, as Petr hinted about the lead acid battery. Materials other than water have different breakdown voltages.

decrees voltage maybe, but if its fairly above 1V...say 2V?if methanol ethanol and such were so absorbent of water vapor combustion engines would have been useless ... I was thinking of acetone but don't know what salt to use...what organically dissolving salt is safe to use?also, as a membrane, pergament paper worked well with KOH but i never knew how long it will hold...do anyone?is pergament paper good for organic?

jndale, correct my mistake please

vapor combustion - this is electrochemistry, I've heard about people, who used old frying oil in their

cars

I can see 3 problems with "your" organic solvents (don't discuss watter).1. Ionic conductivity (connected with salts)2. Evaporation at room temperature3. Potential window

I have googled this paper: http://www.google.cz/url?

sa=t&source=web&ct=res&cd=2&ved=0CBAQFjAB&url=http%3A%2F%2Fwww.cstl.nist.gov%2Facd%2F839.03%2Fj15wu.pdf&ei=z2bnSuL4EYnmnAP38YCwCA&usg=AFQjCNFlwCk0NEmJdyInPyTIWQMAaaMscwlook at Table 1.,the conductivity of methanol and ethanol is much lower than water. There aren't concentrations of electrolytes, but as an example it is good. In this paper is also discussed the effect of water moisture at conductivity...How long does it takes to evaporate 1 ml of acetone and water...In my opinion, the potential window can be bigger, but in connection with lower ionic conductivity it is useless.

but water aren't conductive ether unless salts are molten in them. why is water conductivity needed when they are but a solvent for the ionic material? ( i know water are bipolar but still...).if water is the weak link and the salt is the active one in the "game", why not use another less fragile solvent even if its tainted with water?about KOH + Al - is it possible that the material i use as a collector - the one time baking plate - isn't Al?also can you comment on my previous post on the membrane?

ndor,I can’t comment on the chemistry as I’m not qualified, but in answer to your membrane question I recommend non-woven polypropylene. Once you learn to recognize it you’ll find it everywhere. It is used for all sorts of things like shopping bags, CD protectors, protective clothing etc.It comes in various thicknesses and colours but it always has the characteristic look that you see on the attached picture. Of all the materials that I tried it was the best and it should last a very long

time in any type of electrolyte.

Yes, I would guess that the baking pan may be a steel of some kind.

I did a little bit of googling, and it seems that alcohols electrolyze even easier than water. It looks like depending on the ion concentration and pH levels, ethanol and methanol start electrolyzing between 0.6V and 1.1VI was not able to find any information about acetone though. Ndor, maybe you could find this and enlighten us! Unfortunately, I'm not sure about acetone as a solvent after I found this page: http://www.newton.dep.anl.gov/askasci/chem03/chem03689.htmBut maybe there are other polar solvents that would work?Maybe it's time to start some threads for electrolytes and collector materials.

damn right jndale!Thanks a bunch Peter! I've swapped membrane on one of my Ucap : charge and discharge time was lower - i believe les energy was lost to ion friction in the membrane.Has anyone cosidered using ground up wax as a binder in place of teflon? If the wax doesn't melt, I'd figure that it wouldn't coat the AC- but it might still hold the AC together if pressed. Any thoughts?I'm also wondering why water saturated with table salt isn't used as an electrolyte? I would figure that it would already be used if it was a good solution, so what's wrong with using it?

We have tryed several types (materials) of binders. PTFE and PVdF are the best.KOH electrolyte has better conductivity than NaCl.

I spent a while studying the Maxwell website- their Ultracapacitors come in at an energy density of up to 5 1/2 Wh/kg- substantially lower than stefan's claim of 150F per gram of AC- according to these findings Stefan finds a capacity of 41.6 Wh/kg (although the Maxwell caps take into account the weight of the steel can, etc.)

I'm thinking that Stefan, your number may be quite a bit high- but as I'm beginning to experiment myself, I truly hope that I am wrong and that those energy densities are correct. I mean, that's approaching the energy density of a Pb battery... we couldn't do much better than that. The only solution that could leave Stefan's numbers correct would be if the Maxwell Ultracaps don't use AC. So could someone please clear this up for me? This is the url for the Maxwell Ultracaps data sheet. http://www.maxwell.com/ultracapacitors/datasheets/DATASHEET_MC_SERIES_1009361.pdf

Hi,In my opinion, Maxwell use AC or some similar carbon based materials.

I have tried to calculate the energy density of Stefan's capacitors. It is between 3 and 4 Wh/kg.1st. electrode 150 F / 6 g2nd. electrode 380 F / 18 g

Thanks PetrD for clearing that up. Indeed if I take into account the weight of current collectors, electrolyte and packaging I’ll be lucky to achieve 1.5 to 2 Wh/Kg.

Shotslinlgman, my best result is 35F/g, nowhere near 150F/g.

I read that nice thread and got few questions. 1. AC? What AC. Is that one found in medicine or for watter filtering or what?2. Carbon black? I'd search the web and seems that is just purified soot?3. Why stay in 1V limit. Is the electrolysis only problem?

1. AC stands for Activated Carbon and, yes, it is the stuff you find in water filters.2. Yes, Carbon Black is purified soot and it also used to be called Lamp Black.3. Yes again, electrolyte decomposition is the only problem limiting the voltage.

In the beginning of the post there are photo and say that alkaline battery has nickel plated can. I search the web and found that alkaline batery has SS can. What's true. On the other hands I have a bunch of NiMh cells (death) and can be recycled. But I can't find what material can is. The term NiMh say's nickel but seems that this is not metalic nickel but oxides and is in + pole. Can is - pole and I don't know the material.I realy want to try that. Is anybody try to substituted KOH with NaOH? If I know correct then we can use cooper collector plate.

... yet another question...

I'm not god in chemistry and don't have clue how to make 5M KOH. I can get KOH pellets from http://tinyurl.com/ydcd8e9and don't know how much pellet for just right concentration? Is someone good in chemistry there I just think about binder. And Laser printer toner come into my mind. As I remember correct that stuff is conductive too. Of course need to be warmed up to about 200 degres celsius when electrodes are pressed.And second one idea.

Stefan say that capacitance increases over days. Now I think why. I found two reasons. One is "weetability" of AC. So not all pores are wet on start. I think that if (wet) cell is put in vaccum chamber and do few cycles of moderate vacuum (mean refrigerating compressor) the electrolyte will be forced to all cavity.Other reason can be that there maybe is some oxide make on AC after initial current flow. If I understood there are two capacitors back to back. So One electrode is AC and other is electrolyte. and another one is electrolyte and AC. So if I look there should bee some dielectric too. And that should bee betwen AC and electrolite! So what dielectric is? Is it some oxide created during 1st charge (and need time to create itself) or it is just some "magic" of ion interaction?

Hi,I recommend you 6M KOH1l 6M KOH: 336,66g KOH and 835 ml H2OI agree with your idea of wetting and pores. It is good to use alcohol during boiling of caron (don't

drink )Oxide - I don't thinkNaOH electrolyte has worse conductivity than KOH.

I could be wrong about the metal can. When I used stainless steel for current collectors I noticed that one of them, can’t remember which one (anode or cathode), survived with very little signs of corrosion. So it may be possible that alkaline batteries use a SS can.As far as electrolyte penetration goes vacuum is used in mass production of Supercaps. The dielectric is formed as you suspect by “some magic of ion interaction”

I put to job my old coffe grinder. As wish to duplicate Stefan recipies I start with 8g of AC and 4g of Teflon (I make flakes from PTFE rod with knife) Then I start the grinder and dust come out in all dirrections! What a mess... So I abandon the original recipies and put all AC inside grinder and seal

cup with duct tape. Now seems little better So is that mixing some important thing as I don't see anything in grinder as cover comes black in seconds. As the original grinding time is aprox 20 minutes and my grinder has restriction that 45 seconds can be on and 10 minutes resting I will end grinding tomorow. For collector plate I had prepared NiMh cell sheel. (I had a lot of death NiMh but none of alcaline cells.) So I have little progress.

Is there some other one realy make that cap?

I guess, NiMh is alkanile cellI can make supercap, but in our lab. Stefan's receipt is for "home-made" supercaps really good.

NiMh is alkaline cell. KOH is electrolyte.

Today I try to make my 1st electrode. I Grind 16g AC and 4g of PTFE as stayed. I make dye to make electrodes 30mm in diameter. So I put 3g of mixture in dye and press with mini lathe tailstock. (does not need much pressure?!?). But when I wish to move pressed electrode out of piston the electrode destroys. Is just to fraggile. OK I say myself I gonna to press little harder. So I put mixture in dye again and put thing in 6 tones DIY hydraulic press. Came out better but still to fragile to be able to move out of dye. The (mess) comes out of grinder is just very fine dust and PTFE is totaly invisible to me. I'm shure that I use PTFE! So I have trouble in beggining as I see..

Finaly I made my 1'st cap. So I plug it on and calculating capacitance. So I search for formula and test it with known design.Stefan say 91 minutes and 68 Ohms discharging from 1V to 0.5V to give 150F?!?My formula say 56F Who is right?

I calculated time constant for 0.7T soC=(5460/68)*.7=56F

Em I miss something?For my 1'st cap I just reuse stuff from previous post (AC*PTFE). I take 3.5 g of mix and add 0.6g of laserprinter toner. And after that I divide this 4g to 2 pieces and put it separatly into mold. I heat up

mold to 200 degres Celsius and still hot put into 6 tons press. Electrodes are solid as rock I think I can use even less toner.And here are 1'st resultsIn all cases I use 68 Ohm resistor and discharging from 1V to 0.5V

After 10 minutes of soaking discharge time is 1:20After 15 minutes of soaking discharge time is 2:30After 30 minutes of soaking discharge time is 4:08After 40 minutes of soaking discharge time is 5:36so I got aprox 3.4F

I never tried to fibrillate teflon shavings so I can't help you there. The only type of teflon I tried was the powder type called virgin teflon. It is used for making teflon stock for machining.You can make demo supercap electrodes by mixing some AC powder with a bit of epoxy resin and putting the mix in your 30mm press overnight. Don't use the 5 minute setting stuff though and dilute it with a bit of methylated spirits before adding the AC. The mixture should look almost dry when it goes into the press. Use aluminium foil to stop the mixture from sticking to the mold. Use sandpaper to grind off the foil from the discs after you get them out of the mold.About 20%, by weight, of epoxy should do it.

If I remember ok you (Stefan)say's that epoxi is bad stuf. Overnight I got 16F with my last (and 1'st one)

Congrats on your first super cap. Have patience, keep the cap on the charge overnight and you will be surprised by the improvement you get as long as you don’t go above 1.1V.I used a formula that only gives the correct capacitance with constant current discharge and it seems that I also made a mistake by using the upper voltage rather than the middle of the range for the approximation. I only wanted a ball park figure. It should be more lake 120F but not 56F as in your calculation.When I want a more accurate figure then I discharge down to 0.37V to get the Time Constant and then divide by the value of the load resistance to get the capacitance

Time constant is discharging to 37% of voltage

TimeConstant * 0.7 is discharging to 50% of voltageTimeConstant=R*C 'for 37% dischargeTimeconstant=(R*C)*0.7 'for 50% dischargeI'm pretty shure about that. So not bad idea to check your formula again.Overnight (just soaking not charging) and after 10 minutes charge I got 16F (27 minutes @ 68Ohms) to go from 1V to 0.5V

I Hate to wait so much so now I use 33Ohm resistor.Last measurment at 33Ohm load is11 min 35 sec -->> 14.7FI gonna to make vacuum chamber to speed up soaking.

Vacuum chamber done...

I put half discharged capacitor into chamber and observe voltage during vacuuming. It's slightly

dropped. The funniest is as I notice that electrolite start to bubbling. What the fu.... ... then I realise that I'm just do a little to much vacuum and electrolite start to boil (at room temperature). So I do few cycles with vacuum and atmosphere presure and start charging after that. Seems that I got 2F more in that way. And electrodes are already soaked near two day's. But other thing popup. Before vacuuming when I remove power supply the voltage drops very fast to aprox 0.85V. After vaccuming seems more better. I charge capacitor to 1.05V and when I removed supply It's droped to 1.048 after few minutes. I will check what self discharge rate is tomorow.

Do you think that changing the ambient air pressure changes the electrolyte's resistance at all?

NO. Just more AC come weet and average voltage comes lower