solubility of potassium perchlorate

[dmitri]

This thread was originally divided into three single ones. Merging them might have affected their order somewhat, so take that into account when reading.

Rhadon

++++++++++++++++

Can somebody help me with advise?

I am going to use potassium perchlorate as an oxidizer which has to be introduced
in nanopores (typical size is 100-200 A) of porous silicon (fuel).
Since its surface is organofillic it assures complete filling of the pores by organic liquid.
The problem is that solubility of KClO4 in ordinary liquids like methanol or ethanol is extremely low,
while pore filling factor is crucial for explosive interaction. This binary system can be used as an efficient microexplosive. We have realized combination of porous silicon with other solid oxidizers, it works very well
but all of them are hydrofillic.
Do you have any ideas about solubility of KClO4 in other organic liquids?
Or probably one can suggest another procedure (chemical, electrochemical)to fill the pores by KClO4?
Dmitri

[Anthony]

I think this thread would be more suited to the Chemistry Related forum.

You have broken one of our golden rules by creating a new topic on your first post, however it is an interesting question and your application is unique.

Care to give us some background information on the application?

[vulture]

This seems to be quite interesting. Silicon has a very high burnenthalpy, in combination with KClO4 it should form an insanely hot pyrotechnic mixture.

How about solvents like DMSO or DMF? Acetic acid? Trifluoroacetic acid/anhydride? Trichloroacetic acid/anhydride? Or will the last 2 attack silicon?

[dmitri]

A few comments concerning this system:
Porous silicon can be produced by electrochemical etching from ordinary bulk
silicon wafers. It consists from nanometer-sized silicon nanocrystals whose surface is terminated by one monolayer of hydrogen or oxygen.
Total internal surface is 100-1000 m^2 per 1 cm^3 and almost each 4th Si atom lies on the surface.
Porosity is easily tuned from 10 to 90% and stoichiometric mixture with oxidizer can be achieved.
Another crucial difference with gun powder is that system is selfconfined: nanocrystals are mechanically interconnected and porous silicon (fuel) matrix is solid and stable. In some sence system is similar to high chemical explosive: fuel atoms (silicon) and oxidizer molecules are in atomic contact along huge internal surface.
Therefore a chemical reaction (oxidation) proceeds in time scale of 100´s of nanoseconds (measured).
System is highly explosive, energy gain we achieved so far (measured) is ~ 8 kJ/g and for optimized system can be as high as 10 kJ/g.
The interesting point that it can be exploded very efficiently in negligible amounts, say in micro- milligramm range. Therefore on one bulk Si wafer a big amount of single explosive elements can be produced which can replace ordinary pyrotechnical elements used in industry, say airbag igniters.
According to my experience it works very efficiently already with few oxidizers but all of them are hydrophilic.
KClO4 probably can be the best one but its solubility in ordinary liquids is low.
Thank you for advise concerning solvents, I will try them.
Dmitri

[vulture]

Those energy outputs are normal for pyrotechnic mixtures, but for a HE it is indeed very high! :eek:

BTW, would you be on the same research group that had this "accident" with porous silicon, the very cold silicon that exploded when the protecting hydrogen layer was disrupted by oxygen?

[dmitri]

Indeed I am leader of this group.
Energy yield is similar to ordinary pyrotechnics but this number is not crucial for explosive interaction.
Burning fuel gives more. As I mentioned above this system is mechanically self-confined in nanometer size range and no additional geometrical confinement like for gun powder or other pyrotechnic mixtures is required.
Therefore pressure and temperature rise (and rate of further oxidation) is very fast.
Having some experience with this system I would not classify it as an ordinary pirotechnics. It is rather high microexplosive. You can look on our institute webpage: http://www.e16.physik.tu-muenchen.de/. Look on NEW, there you can see explosion of only 6 milligramm of porous silicon. You can compare this explosion with ordinary pyrotechnics.
Dmitri

[rikkitikkitavi]

A solvent for KClO4 with fairly high solubility I can not think of. Since the pores are extremely small it is impossible to create a suspension of fine (nm sized) KClO4 crystals and let these attach themselfes. The crystals probably also need to grow on the Si-surface for adherence?


However , have you considered NaClO4 or NH4ClO4?

/rickard

[zeocrash]

would an alcohol solution work. the chlorate would disolve in the water, that would mix with the alcohol, then the alcohol would be absorbed into the silicon.
i'm probably talking out of my ass but it's worth a shot
also how about using an oil, water detergent mix as a solvent same principle again, the silicon attracts the organic molecule, that in turn attracts the detergent molecule that in tern attracts the chlorate solution

[Crazy Swede]

I'm quite convinced that ethanol will not work! It might be so that the solubility is higher in methanol, especially if heated. But, I guess you will never reach an efficient concentration of the oxidizer.

My advice is to check out supercritical fluids since they are known to have great solving powers. Maybe supercritical carbon dioxide would work?

Good luck, and please inform us of any success! :)

[Snipie]

Maybe you can form the KClO4 inside the porous silicon, by fractional crystallisation.
NaClO4 + KNO3 are both very soluble.
First you add the (hot) NaClO4 solution to the porous silicon, and after that you add the (hot) KNO3 solution, and then you cool it.

[dmitri]

Thanks a lot for advises. I am almost sure that ordinary solvents will not work, solubility of KClO4 is too low. Fractional crystallization and supercritical fluids, especially supercritical carbon dioxide sound very promising. Within one month we are going to try supercritical drying procedure with CO2, this idea also came to our mind. Unfortunately the price of the supercritical drying machine is quite high (40K) but probably this is only the way. Do you know any data on solubility of salts, especially KClO4 in liquid CO2? I can not find references. Another problem is that it is not clear whether KClO4 will stay in the pores. According to our experience only a few salts can be fixed in the pores, most of them slowly migrate to the surface of porous silicon layer.
Dmitri

[Haggis]

This did not require a new topic. You should have just made a reply to your existing topic on this subject. Do this in the future.

[DBSP]

Locking this thread, dmitri this could have been posted in the already existing topic...

If any senior mod could merge this thread with the existing one on the subject it would be good.

Or should it simply be deleted?? up to you to decide.

[Marvin]

Anhydrous lithium perchlorate is very soluable in Acetone, slightly less in Ether and slightly less in Ethyl Acetate. Roughly of the order 1 part salt by mass dissolves in 1 part solvent.

You could either try these solvents, or lithium perchlorate itself, which would probably perform very well as an oxidiser.

I'm interested in nanosilicon, what are the conditions to electrolytically etch the wafers?
Current density, makeup of the eletrolyte?

Have you thought about using a convection type crystal growing method to grow a single crystal of oxidising salt around and through the nanosilicon wafer? It would be interesting to see the results. How are you applying the salts currently, are there voids in the material?

If the reaction takes place within 100ns, does that mean the material is supporting a detonation wave?

[Crazy Swede]

Dmitri wrote:
-------------------------------------------
Thanks a lot for advises. I am almost sure that ordinary solvents will not work, solubility of KClO4 is too low. Fractional crystallization and supercritical fluids, especially supercritical carbon dioxide sound very promising. Within one month we are going to try supercritical drying procedure with CO2, this idea also came to our mind. Unfortunately the price of the supercritical drying machine is quite high (40K) but probably this is only the way. Do you know any data on solubility of salts, especially KClO4 in liquid CO2? I can not find references. Another problem is that it is not clear whether KClO4 will stay in the pores. According to our experience only a few salts can be fixed in the pores, most of them slowly migrate to the surface of porous silicon layer.
Dmitri
-------------------------------------------

I'm sorry, I have no data for the solubillity of potassium perchlorate in supercritical carbon dioxide. But, there must be sources at any modern university or technical library!

[dmitri]

Unfortunately lithium perchlorate according to our experience does not stay in the pores for a long time and
is hydroscopic.
Micro-(2-5 nm size of silicon nanocrystals) or meso-PS layers (10-20 nm) can be prepared from boron doped Si substrates, with a typical resistivity of 1-10 Ohm cm and 1-10 mOhm cm respectivily. The electrochemical etching is done in a Teflon cell containing ethanoic hydrofluoric solution, with a Pt wire as a cathode. The etching solution is a 1:1 by volume mixture of hydrofluoric acid (49 wt.% in water) and ethanol. The etch current density is 30-100 mA/cm^2. The anodization time can be tuned from a few tens of seconds up to a few hours in order to obtain different layer thicknesses ranging from 1~micrometer to 500 micrometer.
Concerning supporting a detonation wave: geometrical configuration of this explosive is rather unusual.
It can be considered as two-dimensional system: thickness of the layer (say 100 micrometers) is much smaller than lateral sizes (centimeters).
Therefore ordinary detonation wave propagation concept should be reconsidered. Under local ignition of stripe (say by laser pulse) the lateral velocity of reaction propagation is at least 2 km/s. What it will be in ordinary bulk conditions(say cube or sphere configuration)I can not say, to my opinion it should be much faster.

[zeocrash]

dude, you should use the new reply button, not the new thread button to post your replies. it angers the mods when you keep creating new topics. anyway, i took the liberty of pasiting what you had to say into the current thread

[zeocrash]

since he seems to keep posting his replies as seperate topics, :S i thought i'd relay his words to you all in the main thread
Originally posted by dmitri
Unfortunately lithium perchlorate according to our experience does not stay in the pores for a long time and
is hydroscopic.
Micro-(2-5 nm size of silicon nanocrystals) or meso-PS layers (10-20 nm) can be prepared from boron doped Si substrates, with a typical resistivity of 1-10 Ohm cm and 1-10 mOhm cm respectivily. The electrochemical etching is done in a Teflon cell containing ethanoic hydrofluoric solution, with a Pt wire as a cathode. The etching solution is a 1:1 by volume mixture of hydrofluoric acid (49 wt.% in water) and ethanol. The etch current density is 30-100 mA/cm^2. The anodization time can be tuned from a few tens of seconds up to a few hours in order to obtain different layer thicknesses ranging from 1~micrometer to 500 micrometer.
Concerning supporting a detonation wave: geometrical configuration of this explosive is rather unusual.
It can be considered as two-dimensional system: thickness of the layer (say 100 micrometers) is much smaller than lateral sizes (centimeters).
Therefore ordinary detonation wave propagation concept should be reconsidered. Under local ignition of stripe (say by laser pulse) the lateral velocity of reaction propagation is at least 2 km/s. What it will be in ordinary bulk conditions(say cube or sphere configuration)I can not say, to my opinion it should be much faster.

[Rhadon]

Dmitri, please be so kind and put your replies into existing topics using the New Reply button (not New Thread!). This saves us a lot of work and confusion.

[dmitri]

Sorry,
it was my mistake, in future I will use NEW REPLY option.
Dmitri:(

[Chade]

And here I thought solid state physics would never yield anything fun. I was just reading the 'sister' thread 'superpowerful explosive' and from what Dmiti is saying, I really have to start to wonder. Megalomanias big list of explosives and weapons patents includes a patent relating directly to this that has a very interesting idea. US 2003148569 A1, and I'm guessing Dmitri knows about that, as I strongly suspect it's his patent. (does that mean we have to bung him a few quid if we want to try it?)

The porous silicon mixed with oxidiser is used as a primary explosive, but can still be integrated on a regular semiconductor chip. Any chip which needs extreme protection against copying can include the ultimate self destruct mechanism.

What I started to wonder about is, how far along could a dedicated home experimenter get with the manufacture of this stuff? I think it'd be fascinating if you could create your own sample, however poor, of such a new explosive. Cutting edge technology. (I might not be saving up for one of those 40K supercritical fluid machines though.)

I know you can make polysilicon* at home from glass if you can get hydroflouric acid, and I think you could simply etch that to form the porous silicon, rather than worry about the extremely difficult step of forming a silicon crystal. I don't imagine proper researchers would use polysilicon, as the crystalline defects would just make the results less predictable, as far as investigating this new explosive goes, especially if silicon crystals could be bought from industry. The HF could then also be used for the etching, as described.

Then it's a matter of applying the oxidiser as described in both this thread and the patent. The more I think about it, the more possible it seems to be. The alternate method in the patent of starting from colloidal silicon also seems possible in a home lab.

* polysilicon: lump of silicon with no regular structure over any significant distance. Polycrystalline silicon.

[vulture]

I know you can make polysilicon* at home from glass if you can get hydroflouric acid, and I think you could simply etch that to form the porous silicon,

You're not going to get silicon when etching glass with HF:

SiO2 + 4HF ----> SiF4 + 2H2O

You would then have to reduce the SiF4 with a reactive metal like magnesium. However, this requires severe heating under inert atmosphere and SiF4 is a toxic gas, so you'll get serious problems from pressure buildup.

[Chade]

I'm sorry, that was my fault for being vague. I should have phrased it as follows: I know a method for getting elemental silicon in a home lab, and the most difficult aspect to this method is acquiring hydroflouric acid. That's why I said you could do it if you could get hydroflouric acid, but looking at it, my method also requires elemental sodium, which is also hard to get.

Actually, I'm not sure if it's ever been discussed here before, despite the fact that elemental silicon does burn fairly well as a powder.
OK, I'll give the full method I have here.

Hydroflouric acid, to start with, can be made from flourspar (a common mineral, and I've mentioned it in another thread) mixed with sulphuric acid and heated using appropriate containers. HF gas is then passed into water to make a saturated solution.

HF can then, as described by vulture, make SiF4. In a resistant container (described in the hydroflouric acid synthesis thread) of the same type as used to make the HF solution, you add the acid and a ground silicate powder. Sand is ideal, but any silicate works. You can grind down glass, or even quartz or flint. To grind down something like flint, heat it to a white heat in a furnace, then throw it in a bucket of water to shatter it. Then use an iron mortar & pestle to powder it.

The HF/SiO2 mix forms SiF4 as vulture says in a gaseous form. This is in the body of the still used earlier for HF formation. It is placed where the H2SO4/flourspar mix was, and the SiF4 fumes bubble through the water. This gives:
3(SiF4) + 2(H2O) -> SiO2 + 2(H2SiF6)

Ideally, you should bubble it through a thin layer of mercury into the water to stop the tube being clogged with forming silica, but it's not vital.
A white precipitate forms, which can then be heated to form pure silica. This is the SiO2 bonding to water. Not sure of the exact details of this, but the water is driven off by heating.

Anyway, we're not interested in the pure silica, although you could use it and start the process again to ensure a very pure sample of silicon. What we're interested in is the H2SiF6 solution, which we seperate from the precipitate. Then add Potassium Nitrate (Hmmm, where could I get that I wonder...) and observe the solution very closely. After a time, a near transparent precipitate is formed. It is this that we filter off, and now appears as a very white powder. My reference* doesn't give a formula for this compound, but I think it should be K2SiF6, with the potassium directly substituting the Hydrogen.

Then this white powder is mixed with the sodium, cut into small pieces, and heated in a covered crucible at high tempreature for around an hour**. You will end up with a brown powder, insoluble in water. This is Silicon. You can burn it already in air, and even better in pure oxygen, to again form pure silica as a white powder.

To make polysilicon, heat the silicon in a covered crucible with Zinc at a high enough temperature to melt the zinc but not boil it for, again, an hour. This is then covered with Hydrochloric acid which removes the zinc, but leaves the iron grey silicon crystals behind.

Like I said, not the easiest method for the amateur.
I think it should be possible to use any salt of a substance with a higher reactivity than hydrogen to precipitate the H2SiF6. I'm also keeping my fingers crossed that Magnesium powder might work instead of sodium to produce the elemental Si. I can get magnesium from firelighters. At an extortionate price, mind you, but if I can produce my own silicon, it'd be worth it for that alone. Lithium would be more likely to work, but is even pricier.

Once you have your own silicon crystals, you can try Dmitri's amazing exploding silicon. Or perhaps you could take the Si powder and try to make it as small as possible and approach it that way? It may even be easier. Can ball milling get powders down to colloidal (nanoscale) sizes?
Also, no-one thinks of explosives when they think of silicon, so if you say you're approaching it to study the electronic properties of Silicon, you will find it much easier to get hold of the neccesary chemicals.

* As I said in the other thread, my reference here is a turn of the century (C19) book called 'Science for all'. I'm working on scanning the useful articles, but I've got a lot on at the moment. When they're done, I'll host them on some free website for y'all.
** Unfortunately, no detail is given for the temperature required for this step. I'm guessing it would suffice to leave it above a bunsen burner or camping gas stove on full for an hour.

[Crazy Swede]

Chade, wouldn't the use of magnesium metal, instead of metallic sodium, produce magnesium silicide?

[Chade]

Crazy Swede: Quite possibly. My chemistry knowledge isn't that great. I suppose I should try to predict whether it would work before just trying it, and wasting a lot of time, effort and materials.
Bit of a shame if group II metals don't work, as I've got some Calcium as well.

[Edit: Did some research and added following]

It seems Mg2Si can form. As can Na4Si, but that doesn't happen when sodium is used, I presume as they use approximately stochiometric amounts. I think this is one of those cases where the reduction of the compound occurs first, then, if there's any excess reactive metal, it will form a silicide. So I think magnesium should work. (You can't see, but I'm crossing my fingers here)

[dmitri]

Si powder in combination with oxidizer will work like ordinary gun powder. I do not think that
ball milling would allow to get sizes significantly smaller that 1 micrometer. For this size internal area is about 1m^2/cm^3 while for porous silicon it is in the range of 100-1000 m^2/cm^3. Therefore for Si powder
burning rate will be very small, like for ordinary pyrotechnical powders.
Second important point is that as I wrote porous silicon is selfconfined sponge-like structure.
For powders additional geometrical confinement to increase burning rate due to rising pressure is required.