| Pages:
1
2
3 |
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
Remarkable reaction with adjustable delay
By accident I found a very interesting reaction, not because of its reaction products, but because of a really spectacular transition from seemingly
no reaction at all to near explosive violence. The experiment involves mixing two solutions and then for tens of seconds or even a few minutes nothing
seems to happen and after that, suddenly a very violent reaction starts, which completes in a fraction of a second
The experiment is very simple.
Take appr. 100 mg of NaBrO3 and dissolve this in just enough water. Add water drop by drop and try to dissolve this small amount of NaBrO3. Assure
that all solid is dissolved, no solid particles may be left behind. It is easiest to use a dry test tube, put the NaBrO3 in that and then add water
drop by drop, each time swirling until no more NaBrO3 dissolves and continue doing this until no solid is left.
Take appr. 100 mg of NH2OH.HCl (hydroxylammonium chloride) and dissolve this in just enough water in the same way as you did with the NaBrO3.
Dissolving the NH2OH.HCl requires less water than dissolving the NaBrO3.
Mix both solutions. Simply add them to each other in a test tube, swirl and then set the test tube aside and step back a meter or so.
Wait (this can take a minute, or even a few minutes) and then suddenly within a fraction of a second a near-explosive reaction occurs. The water boils
away and a small amount of brown vapor is produced. The entire reaction takes less time than a blink of the eye (actually, you can't follow it with
your eyes, it goes very fast, maybe a tenth of a second and then all is over).
You can play around a little with quantities. E.g. take 150 mg of NH2OH.HCl and 100 mg of NaBrO3 and you get NO/NO2 after the reaction. Take just 50
mg of NH2OH.HCl and 100 mg of NaBrO3 and you get bromine vapor after the reaction. In all cases, there is a delay, which can range from appr. 1 minute
to well over 5 minutes. If you hand-warm the liquids while dissolving the solids, then you need less water and the concentration of the solutions is
somewhat higher, and waiting time is shorter. If you dilute each solution by a factor of 1.5 before mixing, then you get a much longer waiting time
(between 5 and 10 minutes) before the reaction starts.
An interesting variation is to dissolve 100 to 200 mg of NH2OH.HCl in water and add some solid NaBrO3 (50 to 100 mg) to the solution. Again, for a
minute (or even a few minutes) nothing seems to happen, the NaBrO3 just sits there at the bottom, under the liquid. Then there is a POP sound, due to
a small explosion and a plume of vapor and brown gas escapes from the test tube.
--------------------------------------------------------
I think this is very remarkable, because of the long delay, relative to the duration of the reaction. This is a very special type of dynamics. Such
violent reactions are not that special (e.g. 65% HNO3 with isopropylalcohol can be equally violent when mixed), but it is the ratio
delay : (duration of reaction)
which surprises me. The delay is hundreds or even thousands times as long as the duration of the reaction.
With all other reactions of this violence, the delay is at the same time scale as the duration of the reaction.
--------------------------------------------------------------------
Finally, a warning: This reaction can be carried out safely with the indicated quantities, but do not scale up!!
[Edited on 7-11-13 by woelen]
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
In the meantime I googled "bromate hydroxylamine". A lot of links appear which talk about complex dynamics and one link talks about clock reactions:
http://www.readcube.com/articles/10.1002/kin.550260305
I also did the experiment with (NH2OH)2.H2SO4 instead of NH2OH.HCl and this shows exactly the same type of behavior. I also did experiments with KBrO3
instead of NaBrO3. This shows similar behavior, but this only works with solid KBrO3 (this is much less soluble than NaBrO3 and solutions of this
probably are too dilute).
I did not read complete articles (no access), but the search makes clear that there is something interesting about this reaction.
|
|
|
bbartlog
International Hazard
Posts: 1139
Registered: 27-8-2009
Location: Unmoored in time
Member Is Offline
Mood: No Mood
|
|
Any speculation about the nature of the delay (or indeed the overall net reaction)? Presumably there is some initial slow reaction that ultimately
creates conditions for the fast reaction to take place, but what is it? A gradual increase in the pH as the HCl is oxidized?
The less you bet, the more you lose when you win.
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
I did not yet speculate (nor read) anything on the nature of the reaction. I want to think that over tonight, but I did not yet find the time to do
that.
|
|
|
blogfast25
International Hazard
Posts: 10562
Registered: 3-2-2008
Location: Neverland
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by woelen  | | I did not yet speculate (nor read) anything on the nature of the reaction. I want to think that over tonight, but I did not yet find the time to do
that. |
Is it possible that this is just a plain old runaway? Reaction proceeds very slowly at RT but generates heat. Slow temperature rise slowly increases
reaction rate, thereby increasing heat output and further accelerating reaction rate and so on and so on until reaction rate become extremely high?
You could test that hypothesis by adding the second reagent to the first but while the first is hot (or warm) (and by some form of remote control!)
|
|
|
froot
Hazard to Others
Posts: 347
Registered: 23-10-2003
Location: South Africa
Member Is Offline
Mood: refluxed
|
|
Or maybe see if placing the test tube into a ice water bath after mixing affects the delay.
We salute the improvement of the human genome by honoring those who remove themselves from it.
Of necessity, this honor is generally bestowed posthumously. - www.darwinawards.com |
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
@froot: Heating certainly affects the delay. I tried that. I took the saturated solutions (which are around 15 C, the same as my lab ambient
temperature) and mixed them. I did the same experiment again, using samples from the same solutions, but then I warmed the liquids in my hands before
mixing them. This is a difference of 15 degrees or so. In the latter case, there still is a delay, and the characteristic of the reaction is the same,
but the delay is shorter. So, temperature affects the delay.
@blogfast25: An ordinary runaway is different. As you say already, in a runaway you have accelerating rate and it keeps on accelerating on and on. I
have seen runaways quite a few times (e.g. with nitric acid and organics, but also while making peroxochromates from H2O2 and chromates) and then you
see a slow reaction, which gradually becomes more violent, e.g. faster bubbling, until it runs out of control. In this reaction there is no visible
change at all and then suddenly there is production of a lot of gas (I think most of it is N2) and the liquid becomes hot. The rate of production of
the gas, however, is not increasing visibly, it just goes POP. During the entire delay, nothing visible happens at all! And then suddenly it does POP,
and then the reaction is over. I hope to be able to make a movie of this next weekend in the daytime with good light at 200 frames per second. Then I
get more info on the duration of the reaction. It certainly is much less than a second, faster than a blink of the eye.
|
|
|
deltaH
Dangerous source of unreferenced speculation
Posts: 1663
Registered: 30-9-2013
Location: South Africa
Member Is Offline
Mood: Heavily protonated
|
|
I believe that clock reactions were greeted with great skepticism when they first appeared because many scientists argued that such chemical kinetics
went against known chemical behaviour (of the type blogfast describes)... particularly the ones that 'tick' forwards and backwards...
was considered total pseudoscience lol
Clear cut case of 'nature not understanding the theory'
Now here's a real challenge for you woelen. Can you demonstrate a clock reaction coupled to chemiluminescence? Preferably a ticking
one would be much more spectacular over a off then on one, but even the latter would be wonderful nevertheless.
[Edited on 7-11-2013 by deltaH]
|
|
|
blogfast25
International Hazard
Posts: 10562
Registered: 3-2-2008
Location: Neverland
Member Is Offline
Mood: No Mood
|
|
Woelen: you're probably right but a very steep exponential style increase in reaction rate could still be confused visually with a hockey
stick type phenomenon... At the end of the day, even during the delay, something (invisible or poorly visible) has to be going on.
|
|
|
DubaiAmateurRocketry
National Hazard
Posts: 841
Registered: 10-5-2013
Location: LA, CA, USA
Member Is Offline
Mood: In research
|
|
Have you tried making Hydroxyl ammonium nitrate ?
|
|
|
bbartlog
International Hazard
Posts: 1139
Registered: 27-8-2009
Location: Unmoored in time
Member Is Offline
Mood: No Mood
|
|
| Quote: | | a very steep exponential style increase in reaction rate could still be confused visually with a hockey stick type phenomenon |
Except that the overall reaction in this case has a very steep inflection point that doesn't match any single reaction rate function. Anyway, the
obvious way to test this idea is to measure the temperature progression of the reaction, see what the temperature is just before it goes poof, and
then *start* the reagents at that temperature. If it's really just a single stage, then it should take almost no time at all; if there's some
unperceived priming reaction, then there will still be an appreciable initiation period. I assume that's why woelen warmed his reagents: to
demonstrate this test.
Ethylene glycol and bleach (6% aqueous sodium hypochlorite) also react in a delayed, initiation-and-then-reaction way. But it's quite boring compared
to this.
The less you bet, the more you lose when you win.
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
Let's try to stay on-topic. Discussing the synthesis of
hydroxylammonium nitrate is interesting, but not in this topic.
---------------------------------------------------------------------------
I tried whether this clock-type kinetics of this reaction is specific for bromate or not. And yes, it is.
1) When KIO3 or NaIO3 is added to a solution of NH2OH.HCl, then immediately a violent reaction occurs. A plume of iodine vapor escapes from the liquid
immediately.
2) When NaIO4 is added, the behavior is very similar to the case of adding NaIO3 or KIO3.
3) When NaClO2 is added, then you see a more classical runaway. First, there is gentle bubbling and foaming for a few tens of seconds, but slowly the
bubbling and foaming intensifies and at a certain point in time it becoms quite rapid and then in second or so, the reaction is over in a final
climax, being a puff of white fume and gas (water vapor and most likely N2).
Another direction of research is to try to find a set of differential equations, which shows clock-type behavior. This certainly will be a non-linear
set of equations. The set also must be simple. Only two reactants are involved, so the equations cannot have many states. The final property of these
equations is that the involved quantities must be non-negative, as they should represent concentrations of chemicals.
Unfortunately, my search on internet does not yield very much more info than I already have. Quite a few articles exist, which mention special
kinetics in the hydroxylamine/bromate system, but much more information is not available freely.
@bbartlog: I have been thinking about measuring temperature of the reaction mix, but I only dare perform the experiment on a very small scale and for
decent temperature measurements you need at least a few ml of solution. Next weekend I want to try the reaction with lower concentrations of chemicals
in somewhat larger volumes. I already did this with small amounts and then I had a waiting time of more than 7 minutes (!!) and after that it did
*whoosh* and the reaction was over. After the whoosh, the test tube was hot.
[Edited on 7-11-13 by woelen]
|
|
|
blogfast25
International Hazard
Posts: 10562
Registered: 3-2-2008
Location: Neverland
Member Is Offline
Mood: No Mood
|
|
bbart: nowhere did I imply a single reaction rate.
Woelen: do you have access to the full article?
What do you mean by "so the equations cannot have many states"?
Differential equations could be the way to go but with sets of simultaneous non-linear equations, computer iterative programs would almost certainly
be needed.
There are some really simple clocks too: KI/H2O2/Na2S2O3/starch is a classic. No non-linearities involved.
|
|
|
Metacelsus
International Hazard
Posts: 2539
Registered: 26-12-2012
Location: Boston, MA
Member Is Offline
Mood: Double, double, toil and trouble
|
|
I found an interesting article that may explain it:
| Quote: |
A kinetic study of oxidation of hydroxylamine by bromate ion in acid sulfate solution using spectrophotometric and potentiometric methods is reported.
Oxidation of hydroxylamine to nitrate is quantitative and followed competitive, consecutive, and auto catalytics steps characterized by induction
periods. In the slow rate limiting step, hydroxylamine on reaction with HOBr (kmath image) forms an intermediate I, which further reacts fast with
second molecule of HOBr (kmath image) giving nitrite. Nitrite reacts with HOBr (kmath image) yielding the final product nitrate. Nitric acts as an
autocatalyst also and its initial addition decreased the induction periods. In excess of hydrogen ion concentration all the reaction steps follow
second-order kinetics. All the second-order rate constants are reported and the reaction mechanism is proposed.
|
http://onlinelibrary.wiley.com/doi/10.1002/kin.550161011/abs...
Sadly, I don't have access to the full text.
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
I do not have access to the full article. I'll look further, but I have no real hope for that.
With 'many states' I mean many independent state variables, i.e. the dimension of the system.
I know of the simple clock, but that is not representative for this. The clock involves Na2S2O8, Na2S2O3, KI and starch and works like this:
Na2S2O8 is a strong but slow oxidizer. It oxidizes I(-) ion slowly to iodine, while not oxidizing S2O3(2-). As long as S2O3(2-) ion is present, the
iodine is reduced at once to I(-) again and the S2O3(2-) is oxidized to S4O6(2-). As soon as all S2O3(2-) is used up, the iodine remains in solution
and then gives a dark blue complex with starch. So, the liquid remains colorless for a while and then at once turns dark blue. The dynamics of this is
described by highly non-linear equations, but the behavior of these equations is simple. There is no sharp inflexion point in the reaction, everything
runs smoothly. It looks a sharp reaction, due to the sensitive nature of the starch indicator, but the underlying dynamics are tame and predictable.
The reaction I describe in this thread, however, must be of quite a different nature. There is a very sharp rise in reaction rate in an amazingly
short amount of time. The solution of the differential equation system hence switches from minute's time scale to millisecond's time scale and that
makes the system so special for me.
|
|
|
vulture
Forum Gatekeeper
Posts: 3330
Registered: 25-5-2002
Location: France
Member Is Offline
Mood: No Mood
|
|
Some thinking points:
Did you try the other reactions with the same quantities of water? Maybe the boil off of the water is required before the reaction really takes off?
Does bromate show anomalous solubility compared to the other oxidizers?
One shouldn't accept or resort to the mutilation of science to appease the mentally impaired.
|
|
|
PHILOU Zrealone
International Hazard
Posts: 2893
Registered: 20-5-2002
Location: Brussel
Member Is Offline
Mood: Bis-diazo-dinitro-hydroquinonic
|
|
@Woelen,
To keep things relative to each other you should have tested NaClO3 and not NaClO2.
I think you would also have a delayed reaction with chlorate.
The reaction is for sure due to the formation of HBrO3 and resultant NH2OH.HBrO3
Hydroxylamine bromate contains a strong reducer and a strong oxydant in the same molecule... just like NH4ClO3 it is on th edge of stability.
Probable cause of the delay of explosion is the slow building up of concentration of NBr3.
PH Z (PHILOU Zrealone)
"Physic is all what never works; Chemistry is all what stinks and explodes!"-"Life that deadly disease, sexually transmitted."(W.Allen)
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
I tried the NaClO3 as well, but this is not interesting at all. It does not react and this is exactly what I expected. My excuse for not mentioning
that. I considered it common knowledge that chlorate ion is sluggish in aqueous solution and only acts as serious aqueous oxidizer at low pH or at
elevated temperatures, which cannot be achieved in aqueous solution at normal pressure. I can imagine that finely powdered solid KClO3, mixed with
finely powdered solid NH2OH.HCl can make a very energetic mix, but this is far from the conditions of the special reaction which I observed.
I asked access to the paper, mentioned by Cheddite Cheese. It looks promising. I hope to be able to derive the set of differential equations if I have
access to the paper and have insight in the reaction mechanism. Simulating a set of differential equations is not a problem for me. If they are in the
form dx/dt = f(x), with x a vector of state variables then things are really simply (e.g. Runge-Kutta with variable step size detection to find the
sharp inflexion point), if they are implicit algebraic differential, i.e. of the form dx/dt = f(x,y); g(x,y) = 0 with algebraic state y, then I have
more of a challenge. This may require DASSL or a similar piece of software.
I do not think that stuff like NBr3 or NH2OH.HBrO3 is involved. There must be a much more intricate mechanism.
@vulture: I did quite a few different tests, with solid KBrO3 and solid NaBrO3 and solutions of that. All show clock-type behavior (except too dilute
solutions). The solubility of the bromates is not really anomalous compared with the other oxidizers. The sodium salt dissolves more easily than the
potassium salt.
[Edited on 7-11-13 by woelen]
|
|
|
blogfast25
International Hazard
Posts: 10562
Registered: 3-2-2008
Location: Neverland
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by woelen | I asked access to the paper, mentioned by Cheddite Cheese. It looks promising. I hope to be able to derive the set of differential equations if I have
access to the paper and have insight in the reaction mechanism. Simulating a set of differential equations is not a problem for me. If they are in the
form dx/dt = f(x), with x a vector of state variables then things are really simply (e.g. Runge-Kutta with variable step size detection to find the
sharp inflexion point), if they are implicit algebraic differential, i.e. of the form dx/dt = f(x,y); g(x,y) = 0 with algebraic state y, then I have
more of a challenge. This may require DASSL or a similar piece of software.
|
Well, I certainly look forward to that treatment!
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
A little bit of the mystery is resolved. The reaction is not a true clock reaction, but it is a so-called branching chain reaction (see link for a
very short, but clear qualitative description):
http://www.britannica.com/EBchecked/topic/77570/branching-ch...
A branching chain reaction is a chain reaction, where one of the products in the chain of reactions catalyses the conversion of one of the initial
reactants. It has aspects of an autocatalytic reaction, but from initial reactant(s) to catalysing product is not through a single step, but through a
chain of steps.
In the case of my reaction, the formation of bromide ions and H(+) ions catalyses the reaction.
The net reaction (when excess NH2OH.HCl is present, which is split into ions NH3OH(+) and Cl(-) in solution) is:
2BrO3(-) + 6NH3OH(+) ---> 2Br(-) + 3N2O + 9H2O + 6H(+) + heat
This net reaction occurs through many steps, this equation only gives the final result.
Initially, when there is no bromide and only a little amount of H(+), due to splitting of NH3OH(+) in NH2OH and H(+), bromate reacts with NH3OH(+)
very slowly, giving intermediate species HBrO2 and [NH2(OH)2](+). Both species react further very rapidly, giving final products Br(-), N2O, H(+),
water.
With the increase of the concentration of Br(-) and H(+) another reaction occurs, much more rapidly, but not so fast that it can be considered
momentaneous:
BrO3(-) + 5Br(-) + 6H(+) --> 3Br2 + 3H2O (simplified, through a chain of steps, involving HBrO2 and HOBr as transient species).
The Br2 in turns reacts rapidly with NH3OH(+), forming N2O, water, bromide and H(+). This reaction can be considered momentaneous (quasi-static,
non-differential).
Many steps go so fast, that they can be considered non-differential and can be considered quasi-static. A few steps have a perceptible range and these
cannot be described as quasi-static and lead to a differential equation.
The type of equations for a branching chain reaction lead to non-linear equations, exhibiting super-exponential behavior. Such systems can have
solutions which remain close to 0 for a long time and then suddenly the solution 'explodes'. In technical terms, such a solution is called a
non-thermal explosion.
I'll try to find a set of differential equations, based on real chemical reactions, which on simulation reproduces the effect which I observe.
---------------------------------------------------------------------
Another thing, which makes the system even more extreme is that in the short time of the non-thermal explosion the liquid heats up considerably (close
to the point of boiling) and this accelerates the reaction even more. The produced heat also works "autocatalytic" in some sense. In a normal runaway,
the heat is only one factor, which leads to the fast reaction and this can lead to exponential runaway. Here, in this reaction, the non-thermal
explosion, described above, combined with the strong heating of the reaction, gives the effect of the sudden explosion.
A nice article, describing some of the concepts is attached to this post.
If I find a good set of equations, which describe this reaction and reproduces my experimental results, then I certainly will make a web page about
this interesting phenomenon, with references, pictures and a Java-program which demonstrates the behavior.
Attachment: clock_reactions.pdf (69kB)
This file has been downloaded 558 times
[Edited on 8-11-13 by woelen]
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
I made a few movies of this reaction:
Excess bromate: http://www.homescience.net/chem/exps/hydroxylamine_bromate/e...
Excess hydroxylamine: http://www.homescience.net/chem/exps/hydroxylamine_bromate/e...
Slow motion: http://www.homescience.net/chem/exps/hydroxylamine_bromate/s...
|
|
|
bfesser
Resident Wikipedian
Posts: 2114
Registered: 29-1-2008
Member Is Offline
Mood: No Mood
|
|
<del>woelen, I've taken the liberty of uploading your videos to YouTube (unlisted), so that they can be embedded here. Some
users don't have native AVI support on their devices; this is a way around that. As I said, they're unlisted, so they're only really visible from
these links, but I can remove them if you like; just say the word.</del>
[edit] Amazing work, by the way! The slow motion video in particular shows some surprising and unexpected (for me) phenomena.
[edit] Videos removed.
[Edited on 9.11.13 by bfesser]
|
|
|
deltaH
Dangerous source of unreferenced speculation
Posts: 1663
Registered: 30-9-2013
Location: South Africa
Member Is Offline
Mood: Heavily protonated
|
|
Spectacular! I enjoyed the comparison of the excess one or the other. Lovely work, well done woelen!
[Edited on 9-11-2013 by deltaH]
|
|
|
kmno4
International Hazard
Posts: 1503
Registered: 1-6-2005
Location: Silly, stupid country
Member Is Offline
Mood: No Mood
|
|
I tried this on test tube-scale, ~50mg reactants + ~1cm3 water
Violent reaction , however without bromine, only colourless gases, but it is a matter of proportion I think. In my hands it took few seconds from
mixing reactants to the spectactular end.
I also tried N2H4•H2SO4 instead of NH2OH•HCl.
Reaction similar, but it speeds up much longer (with increasing amount of bubbles), and the end is not so violent. This time I got at the end orange
solution with strong Br2 smell.
When some solid KBrO3 was added to spent solution (with excess of hydroxylamine), no delay was observed but sudden reaction starts at once.
[Edited on 9-11-2013 by kmno4]
Слава Україні !
Героям слава !
|
|
|
woelen
Super Administrator
Posts: 8027
Registered: 20-8-2005
Location: Netherlands
Member Is Offline
Mood: interested
|
|
For this time it is OK to me to leave them on Youtube, but generally I am not charmed of it at all. The movies are screwed, all detail is lost and
video quality is crappy. I want people to watch the original movies and if they are on Youtube, then people do not watch the originals. Especially the
slow motion video loses a lot of detail.
Another issue with Youtube is that the material becomes owned by Youtube as soon as you upload. If somewhere in the future I see my movies (or frames
of it) used (e.g. in a commercial), then I cannot do anything against that. For that reason I have my own webspace. If people do not have AVI support
on their PC, I recommend them to install VLC media player, which is available for Windows and Linux (not sure about Apple OS-X) without any cost.
|
|
|
| Pages:
1
2
3 |
![[*]](images/xpblue/default_icon.gif){kind=icon}
