Index Home About Blog
Newsgroups: rec.pyrotechnics
From: tip*@ai.chem.ohiou.edu (Thomas I Perigrin)
Subject: Re: Now flashpowder dynamics....
Date: Tue, 26 May 1998 14:44:14 GMT

In article <1998052018260000.OAA17570@ladder03.news.aol.com>,
sheetla@aol.com (Sheetla) wrote:

> Here is a reference for you.  Bircumshaw and Phillips , JCS 1953, 703 -
> state that potassium perchlorate forms Kcl, O and Cl upon decomposition.
> They also state that the decomposition is "complicated and irreproducible
> in a vacuum".
>
> ----------------------------------------------------
> Bill, this does not make sense to me, Kclo4 has only 1 chlorine per
> potassium, where would the free chlorine come from? if that were to
> occur, free potassium would have to form (i dont think so!) or oxygen
> would have to be more reactive than Chlorine (ahhaha) and displace the
> KCl, and we know thats not possible.

See this in my hands?   Its a can o' worms.   I'm about to open it....

Lets start with a small can o' worms.  (gets out can opener)  *ratcheta
ratcheta ratcheta*

Let us discuss the reaction

R1          2H2 + O2 -> 2H2O.

This reaction doesn't involve 2 hydrogen molecules and 1 oxygen molecule
going into a "chemical dark closet", closing the door, and suddenly,
viola, by magic, water appears....  There is a mechanism.  All reactions
have a mechnism, which is the microscopic atom-by-atom and bonb-by-bond
description of how the reaction occurs.  It is often very complex, and
involves all sorts of funny intermediates, and can lead to odd products.

It could be thought that the formation of water goes like this:

Ra           H2  + energy ->  2 H*
Rb           O2  + energy ->  2 O**
Rc           H* + O** + H*  ->  HOH

However, it was found experimentally that this is not the case.   While it
COULD happen that way, it happens a different way.   First, let me
explicate what does happen, and then explain why...

There are three phases in this reaction:

1) initiation
2) propogation
3) termination

INITIATION

In the initiation step, a molecule of hydrogen is broken into two hydrogen
atoms (with a single electron each, this is called a radical, as opposed
to an ion).  This is done by a  flame, a spark, UV, whatever...  I
represent it as "energy". [I am going to use an asterisk to mark a
radical... this is unusual, most of the time we use asterisks for excited
states and dots for radicals, but I can't be sure everyone would see the
dot character.]

R2          H2 + energy -> 2 H*

PROPOGATION

These two hydrogen radicals then each go their own way.  Let us follow
one...  the other does the same thing.  Eventually one of them runs into
an oxygen, as shown in R3.

R3           H*  + O2  ->  HOO*

This is called the hydroperoxy radical.   The hydroperoxy radical then
roams around until it finds a hydrogen molecule, as shown in R4.

R4           HOO* + H2  ->  HOOH + H*

Oh look, we have generated another hydrogen radical!   We can see the
ramifications of that if we sum R3 and R4, and come up with R5.

R5           H* + O2 + H2  ->  HOOH + H*

Reaction R5 shows that the hydrogen atom is catalytic!   It catalyzes the
reaction of hydrogen and oxygen to give hydrogen peroxide, and is
regenerated to do it again and again and again.  The hydrogen peroxide is
unstable at elevated temperatures, and so it breaks apart as shown in R6.

R6           HOOH  ->  2 HO*

The hydroxy radicals formed in R6 then find hydrogen atoms as shown in R7
(only one is show, the other does the same).

R7           HO*  + H2  -> HOH + H*

Oh look  - another hydrogen radical! This means that we are not just
"turning them over", we are generating NEW ones.    Summing R5, R6 and R7
gives:

R8           H* + O2 + 3 H2 ->  2 H2O + 3 H*

This means that the reaction is BRANCHING...  as it proceeds, it proceeds
faster and faster.  This leads to an explosion.   That is one of the
reasions why a mixture of hydrogen and oxygen explodes when sparked,
rather than just burns nicely.

TERMINATION

Occasionally the radicals find each other and terminate the chain.  For
example, the two radicals formed in R2 could recombine as shown in R2'

R2'          2 H*  ->  H2

However, this is uncommon.  Radicals (unlike ions) do not have a charge
and don't attract each other.  Thus, when the radicals are formed they
just randomly bump into molecules by Brownian motion.   The concentrations
of H2 and O2 are very high compared to the concentration of H*, and so the
chances the two H*s will find each other before they run into an oxygen
molecule are very small.  And because the radicals are so reactive, once
they do bump into a suitable molecule they temd to react immediately, and
are thus consumed quickly.   They don't hang around long enough to find
each other.

Another temination reaction is shown in R9.

R9           HO*  +  H*  ->  H2O.

At first thought this might seem more reaonable than the complicated
mechanism shown above.  However, once again it has to do with
concentration and random motion.   The concentrations of hydrogen and
oxygen are much higher than the concentrations of hydroxy radical and
hydrogen radical until the reaction is nearly complete.  Thus, until the
hydrogen gas and oxygen gas are nearly consumed, the termination reaction
R9 is an inconsequential part of the reaction.


WHY NOT Ra, Rb and Rc?.

Thus, you can see the reaction of   2 H2 + O2 -> 2H2O is a complex process
that involves many steps.   Essentially none of the water is formed by Ra,
Rb, and Rc.

Ra           H2  + energy ->  2 H*
Rb           O2  + energy ->  2 O**
Rc           H* + O** + H*  ->  HOH

Why?   Once again it has to do with probability, and also with thermodynamics.

During the beginning phases of the reaction there is actually a lot more
molecules than heat...  thus the heat available per molecule for reaction
is low.  Thus, lower energy processes will dominate those that require the
input of a lot of heat.   The splitting of oxygen into two radicals (Rb)
requires 59 Kcals/mole, while the reaction of H* with O2 to give
hydroperoxy radical RELEASES 46 kcals/mole.  That means that if the
hydrogen radicals finds an oxygen, it is favorable for it to react.   Of
course, it did require an input of 51 kcals/mole to create the hydrogen
radical, but that is required in either pathway.  That penalty has to be
paid.

Of course, the spark (flame, etc...) that generated the hydrogen radicals
CAN produce some oxygen radicals.   What is their fate?  Well, by
probability, the oxygen radicals are MUCH more likely to run into hydrogen
molecules (H2) rather than hydrogen atoms (H*).   That occurs according to
R10.


R10          O** + H2 ->  HO*  + H*

Note that this plays back into the mechanism above.   The hydrogen atom
enters the mechanism at any step that requires H*,  and the hydroxy
radical enters at R7 regenerating more HYDROGEN radicals.  Thus, in a
manner of speaking, ocygen radicals are converted into hydrogen radicals.

SUMMARY

The reaction occurs through a complicated multi step branching chain
reaction.  This explains why a little initiation event, releasing even a
few reactive radicals, can quickly become an explosion.


-----------

SO, WHAT DOES THIS HAVE TO DO WITH KClO4 + Al?

While the reaction of KClO4 + Al is more complex, the same principles
hold.  The reaction does NOT proceed through the reactions:

Rx           KClO4  ->  KCl + 2O2
Rx (alternate)  KClO4 -> KCl + 4 O**
Ry           Al(crystal) ->  Al***(gas)
Rz           2 Al*** + 3 O**  -> Al2O3

Instead, the reaction is likely to proceed through a mutli step
mechanism.  The first step may very well involve some disproportionation
reaction to generate a reactive gaseous oxyhalo compound.  This can then
begin transfering oxygens to aluminum, generating heat and producing
radicals.  These can then help chew up the perchlorate, and transfer more
the oxygen to the aluminum.  At some point the heat will rise to a point
where the Al will begin to vaporize.  This may allow a secondary mechanism
to occur.   Eventually the reaction may deplete the metallic aluminum and
perchlorate, and secondary reactions between the partially oxidized
aluminum species and partially reduced oxychloro species will occur.
Asd the reaction proceeds, the major mechanism may shift again, giving
even more different products.

Anyway, it isn't clean.  And it certainly doesn't occur as one is led to
believe in early chem classes.

--
-------

To reply, remove the anti-spam * in my name


Newsgroups: rec.pyrotechnics
From: nelson@cs.rochester.edu (Randal Nelson)
Subject: Re: Now flashpowder dynamics....
Date: Wed, 27 May 1998 14:40:42 GMT

In article <tip*-2605981044140001@shimizu.chem.ohiou.edu>,
Thomas I Perigrin <tip*@ai.chem.ohiou.edu> wrote:

>See this in my hands?   Its a can o' worms.   I'm about to open it....

  ... Nice description of Hydorgen-Oxygen reaction path...

Can of worms indeed.  Even this is not the whole story.
Naively applied, the mechanism
described yields an exponentially growing reaction even for microscpic
initiation. If this were true, then a mix of hydrogen and oxygen would
explode as soon as a random cosmic ray produced a few hydrogen radicals.
This does not occur. H2 / O2 mixes as reasonably stable as such things go.

Generally, initiation of reactions like this is controlled by thermal
diffusion mechanisms: some active species are produced by a thermal process,
or at least their production rate is governed by temperature; if the
reaction zone is too small, then it cools off before enough of the active
intermediates can be produced to propagate the reaction.

In this case, it looks like the HOOH  ->  2 HO*  reaction is the
critical step, since as TP points out it needs elevated temperatures
to proceed, and thus heating of some (small) critical volume.

There are some other reactions (H2 / Cl2 comes to mind) that will
propagate from microsopic, or nearly microscopic initiation.
As far as pyrotechnics is concerned, all this relates to the notoriously
difficult issue of predicting or even quantifying the sensitivity
of various mixes...

RN


--
  Randal Nelson			 716-275-8488	University of Rochester
		      nelson@cs.rochester.edu	Computer Science Department
..!{allegra,decvax,rutgers}!rochester!nelson	Rochester, New York,  14627

Index Home About Blog