Capacitor composition?

[PedroDaGr8]

I have been trying to find some in depth information on capacitor chemical composition and one thing that I can seem to find is are the insides of capacitors acidic or basic or neutral? THe reason I am wondering is that I have acess to a lot of expensive equipment for chemical analysis, and I wanted to try and do some comparisons between the good capactiros and the bad ones. My idea:

Extract the electrolyte using a solvent.
Run a few IR scans to compare them (not entirely helpful but could give some info)
Then run a GCMS which would split a sample into its individual parts and then analyse it. The idea I am wondering is could be use it to track changes in the caps over time (not on the same cap but as companys alter the chemical formulas etc.) to see if we can spot good and bad quickly via chemicals. For example say a little known cap comes into play, we could see if it compares to a nother well known bad cap etc. The problem is, highly acidic compounds will destroy a GCMS capillary collumn, also compounds containing a labile fluorine will kill the column. I am unsure if anyone in here could answer this question, but I feel like this would be a nice side project for me. The hardest part will be finding a solvent that extract all of the electrolyte (though most of the aqueous ones should be fine in water).

[PeteS in CA]

Re: Capacitor composition?

I believe Al-El electrolytes are neutral, though I'd check that with some one who knows more or with litmus paper. Some wet slug tantalums (mil M39006/22 or /26) are very acidic, as in clean your PCB within seconds or kiss it goodbye.

[PedroDaGr8]

Re: Capacitor composition?

yeah, I figured that I would test with pH paper anyways, but always figured that I would ask first. I am mainly gona stick with Al-El's next step will be to get a bunch of crap caps (better to refine the method on crappy ones than good ones).

[gonzo0815]

Re: Capacitor composition?

As far as i remeber (if i ever have read something about that... ), the solution is probably more alkaline (in order to generate the AL oxide layer). But the purity of the used Alumina on the foils are a major factor too. There is an link burried deep in this forum, where some pro`s died some testing on bad caps from Taiwan. They stated, that the Al from Japanese capacitors was nearly 97.xx% clean or may be soemthing near 99.xx% wheras that from badcaps are only 96 or lower purity. They further explained, that if the purity is to low, an excess of hydrogen would be generated, wich are responsible for the bulging and early elektrolyt breakdown show by the bad caps.

[Spacedye69]

Re: Capacitor composition?

The aluminum oxide layer is created with sulfuric acid (think anodizing aluminum) and AFAIK, all aqueous are sulfuric as the main ingredient. I'm sure the oxide growth is done under vacuum with sulfuric vapor to control the growth. The quality contol and the secret ingredient stabilizing the hydrolysis under charge retention must be the key to good caps.

[Spacedye69]

Re: Capacitor composition?

interesting reading

[gonzo0815]

Re: Capacitor composition?

The oxid layer is not build under vacuum. After the cap is ready made, the oxide layer will be build up by applying a specific voltage regime. It is similar or equal to the formation proces, wich is sometimes necesary, if an cap was to long without any voltage.
Regarding the elektrolyt formular, my knowledge is limited to the earlier GPO caps, i have no clue abouth the water based formula.

[willawake]

Re: Capacitor composition?

you would need an esr and capacitance meter though to categorise the samples you are taking. i think scientific research into this issue is very poor and there is opportunity to make a name for yourself. the Hillman et al spacey linked is sadly the only one i know. still like gonzo said the aluminium purity is an issue also and its a problem and expense to get high purity. there are other items like the etching of the foils and also the aging of the caps to build up the dialectric which are reduced in the bad caps. so whilst looking at the electrolyte is interesting its still not the whole picture. but similarily to the Hillman perhaps it will give an idea about other factors in the failure mode. it would be interesting to take into account high operating temperature as a factor and also excessive ripple on the input. how you would do that dont ask me lol.

[Spacedye69]

Re: Capacitor composition?

Another link Elelectrolyte fills in voids on the anode oxide film. The original oxide film is formed before the foil is welded to the leads and put in the cap. A couple links have shown that they use a normal sulfuric or oxalic acid wet anodizing process most of the time to grow the film. More linky

[PedroDaGr8]

Re: Capacitor composition?

Wow some excellent reading here. I just got around to reading this today because I had knee surgery (from an injury in june where I dislocated my knee 2.5") last friday and was sort of out of it. Though I just fell on the bad leg last night and have a check up later today to make sure I did no more damage.

The Hillman article is actually quite interesting.
Here is my summary on what I learned from the article

While it was stated that the impure Al could be the cause (Gonzo not trying to point ya out )

here is an link burried deep in this forum, where some pro`s died some testing on bad caps from Taiwan. They stated, that the Al from Japanese capacitors was nearly 97.xx% clean or may be something near 99.xx% whereas that from badcaps are only 96 or lower purity. They further explained, that if the purity is to low, an excess of hydrogen would be generated, which are responsible for the bulging and early elektrolyt breakdown show by the bad caps.

In the beginning of this article they state:

Ignoring the topical constituents of oxygen
and carbon, the purity of the cathodic aluminum foil
from the Japanese capacitor worked out to be
approximately 99.1 wt%, which was within the limit
set by Dapo. The purity of the cathodic aluminum foil
from the Taiwanese capacitor was approximately
97.5%, which was below the minimum value stated by
Dapo. The insufficient purity of the Taiwanese
aluminum foil could cause gaseous hydrogen
production that would not be impeded by a
depolarizer, but the galvanic couples were not thought
to be sufficient to account for the rapid production of
hydrogen gas that was necessary to cause the
relatively rapid bulging of the capacitor cans.

what this basically says is that while there is a stark difference in the qualities of the Al, and it COULD be have as people previously hypothesized. They feel the difference is not nearly enough to account for the rapid H2 (Hydrogen gas) evolution that is occuring.

So there in turn must be something else going on. I think the most notable thing of interest is that the bad capacitors in the hillman article have a concentration of solubilized aluminium in their electrolyte, while the good caps do not.

This implies that somewhere along the way either the foil or the dielectric is getting eaten up. It is in fact the dielectric, Al2O3, that is being eaten up. I knew that that was going to happen the first instant that they mentioned that the pH in the bad caps was midly basic.

This reaction could happen via one of two routes, in the first (less likely) the aluminum oxide steals a hydrogen from water Al2O3 + H2O ---> Al2O2OH +-OH. The -OH goes on to attack the Al2O2OH to form Al2O2(OH)2 which this then repeats with a cleavage of the inter aluminum oxygen bonds to finally form 2 Al(OH)3.

On the other hand, due to the basic nature a more likely situation is that since the solution is neutral to slightly basic, there should be a good amount of free -OH in the solution. Which would then attack the Al2O3, forming Al2O3OH- which then goes on to pluck a proton off of a water forming Al2O2(OH)2 or represented anotherway Al2O3.H2O. Simply a hydrate, but the difference is, it is not simply a water molecule bond due to hydrogen bonding type mechanisms, instead it is actually covalently bound in the molecule.

From here the system repeats itself, with another attack of an -OH. This results in Al2O2(OH)3, with the same following step, the capture of an H+ giving us Al2O(OH)4 or Al2O3.2H2O. The last step is the same as the previous two, with an attack of an -OH, the difference being when this occurs, the molecule is split into Al(OH)3 and AlO(OH)2- with the latter rapidly picking up a proton (H+) to get the final product of 2 molecules of 2Al(OH)3 which conveniently is Al2O3.3H2O.

Another important thing to note is that the non bulging Taiwanese capacitor contained a high level of phosphate, while the one that bulged did not. I am going to do some more research into why this is important, it could be that the phosphate prevents the Al Hydroxide formation, I am not sure (sort of sheepishly too because my main research area is in aluminum-ligand compounds, albeit for reactions with nerve gases instead of this type of chemistry but still).

I guess the good and the bad of this is that they really to be honest did an excellent job on this research. The good being that their in depth explanation was not the greatest. I think once I get some more information, I am going to try and write up a long explanation on why the capacitors fail in simple English.

[PedroDaGr8]

Re: Capacitor composition?

By the way spacedye69 thanks a billion for all of those EXCELLENT links

Also williwake, thanks for the formatting. I am still pretty drugged up(1 Pericoset 7.5 every three hours) and the first thing to go for me is proper punctuation.