Media Info
    News releases
    Fact sheets
    Interview us
    Myths & truth
    Gun dictionary    
    Scary words
    SEA Effects

Photos & Videos
News Coverage
General Info
Explosive Experiments
Bomb Help (NOT!)

Shirts, hats, stuff
Join News List
Privacy Policy

Reloading Supplies
Reloading Data

Second Amendment Foundation banner

Project Diesel

bulletFirst Boomer
bulletProject Diesel II
bulletProject Diesel II, April 11, 1999 Update
bulletProject Diesel II, June 3, 1999 Update

horizontal rule

On September 2, 1998, with Don Wood as backup, I tried a mixture of diesel and ammonium nitrate. The video isn't very interesting, but here is the audio of that failure (AwCrap.wav -- a 280 Kbyte WAV file). Don did the video taping and was the inspiration for the title of the files. The tin can bouncing and rolling sound you hear in the background was an empty pop can rolling off the hood of my pickup. The muzzle blast from my 30.06 knocked it off. We were using the hood for a benchrest for the rifle and video camera. We tried two pop cans full of this, with the same result -- disappointment.

Next I tried adding some potassium chlorate and constructed two targets. Watch and listen to this video for a change in our attitude. Again Don did the taping and was the inspiration for the title. Here is a video of the first can and click here for the second one.  

I went back out the range again with Don Nelson a while later to do some more tests.  Here is one of the videos obtained then.

And to answer a question from a couple different sources, here is what happens when you put it in a six ounce juice can (SmallCan.wmv 374 Kbyte WMV file). Of course there were a couple other variables changed too. The percentage of potassium chlorate was reduced, and the AN was not ground up like it was in the first case.

The recipe I ended up using for the first Boomershoot was:

2 cups prilled (NH4NO3)
1 cup KClO3
1/4 cup diesel

Both the ammonium nitrate and KClO3 were ground up (but NOT together). The diesel was mixed with the NH4NO3 then the KClO3 was added.

I still have some more work to do. I would like to make the targets cheaper and more sensitive.   They cost about $1.00/target currently and if the velocity of a .30 caliber bullet gets below about 2200 fps they sometimes don't go off.  I think it may be ambient temperature related too.  I suspect that what is actually going on is that at impact time the ammonium nitrate and potassium chlorate combine to produce ammonium chlorate, which explodes at a temperature of 102 C.  If true, then the optimum mixture is something a little richer in KClO3 than what I was using.  This might make things more sensitive.  Other options to try include using various catalysts for the reaction. I have some on order and hope to try them out soon. Another thing that I think will be a good idea is to use half-pint milk cartons instead of the pop cans.  This will be much easier to fill, less danger from shrapnel, and make a lost target more likely to decompose safely in a week or two rather than wait around for years for some kid to find it or end up as some sort of crude land mine,

For those that have the access to the chemicals and a legal location to do this sort of thing I suspect the best recipe for this, that I know for certain will work, is 55% KClO3, 36% NH4NO3 and 9% diesel/kerosene/fuel-oil/etc.  All the percentages are by weight.  I believe the safest method of mixing is to grind up the KClO3 separate from the NH4NO3, (both should be very fine) mix the diesel and NH4NO3 together, then add in the KClO3. Use all the normal precautions for working with explosives and chlorates in particular.  Small batches, face shields, body armor, heavy gloves, ear protection, clean work environment, no sparks or open flames, no wooden or iron/steel tools, be extremely careful about friction and impact, test for friction, impact, and spark sensitivities with very small samples, dispose of all left over finished product in a timely and safe manner, and probably a hundred other things that I just do without even thinking about it and you must learn in order to be safe too.  Some of the mixes mentioned below have not been stored for more than a couple days so I don't know but what it may spontaneously explode after 72 hours.  I have also read cautions about working with KClO3 in the presence of direct sunlight and have done all my work in the shade.  There are cautions in the literature about moisture with the ammonium salts and KClO3. Keep your chemicals dry and remember that NH4NO3 will absorb moisture from the air.  Probably in sufficient quantities to cause problems. Be sure to follow all the ATF rules on proper distances from occupied buildings, roads, etc.   See for a start. Also, if you are stupid enough to engage in recreational explosives and get yourself arrested, hurt, killed, or a combination thereof, the extent of my liability will be to nominate you for a Darwin Award.  And always remember:

There are old pyros,
and there are bold pyros,
but there are not very many
old, bold pryos.

If you really want to get involved I suggest you let someone else do it (I plan on having a shoot each spring) or read a bunch of books on the subject before you start.

horizontal rule

Project Diesel II

Three goals with this project, in the following order.  1) Reduce the risk, 2) Increase the sensitivity, 3) Reduce the cost per target.  I had been used aluminum soda cans, metal being "a bad thing" to have explode in your vicinity should an accidental detonation occur.  Also, a lost can would probably hang around for years without deteriorating and become a hazard if someone were to decide to recycle it or whatever.  The mixture isn't quite sensitive enough to be reliably detonated at great distances.  Perhaps even with .223 rounds on cold days at short ranges.  Not good.  The mixture I used in the October '98 event cost about $1.00 per target about $0.90 of that was KClO3.

I have switched to half-pint paper milk cartons for the containers.  These should be safer in the event of an accidental detonation and they should deteriorate within a few days (weeks?) in the sun and weather.  Once exposed to the rain the NH4NO3 should dissolve into the ground and  completely eliminate the hazard of an accidental explosion.  Also, the cartons open wide enough to make it easier to load than the soda cans.  Folding them into roughly a cube and sealing them with glue and tape should make for a relatively good target.

In an effort to reduce the cost of the targets and increase the sensitivity I obtained some MnO2 which is used as a catalyst with KClO3 (in the presence of MnO2, KClO3 breaks down to release it's oxygen at about 70 C rather than the usual 400 C). I was hoping it would also catalyze NH4NO3 (ammonium nitrate). On February 27, 1999 I mixed up some NH4NO3 with diesel (ANFO) in the usual proportions (diesel is about 5.5% by weight) and added about 1% MnO2. I took a very small quantity and rubbed (hard) it between the face of a hammer and a metal plate. Nothing. I light tapped it between the face of two hammers -- nothing. Tapping harder finally got a small snap occasionally. This is more than I can get to happen with ANFO by itself. I added about 1 % SiO2 (sand) to increase the friction. About the same. I burned a small amount of it. Nothing much, it didn't burn very well. I added about 2 % KClO3 to the ANFO, MnO2, SiO2 mixture. Rubbing didn't do anything, tapping lightly got snaps fairly frequently with occasional "good ones". Burning was about like a burning a match head for a similar size piece. Similar experiment using Fe2O3 as the catalyst. Similar results but probably just a bit less reactive.  These tests indicate that it probably is safe for mixing, static and fire hazards, and putting in the target containers.

Out to the range the next day (February 28, 1999). The temperature was about 40 F, which I suspect affects the results. Higher temperatures making it more likely to detonate. I forgot to bring the KClO3, but I was mostly interested in seeing if I could get things to detonate without the use of KClO3 anyway. I mixed some of the stuff by volume (no scales at the range), 1 cup ANFO, 1/2 teaspoon MnO2, 1/2 teaspoon SiO2 inside a half-pint milk carton. This should figure out to be about 1 % MnO2 and 1 % SiO2. I put two .22 LR rounds through it from 30 yards away. 9.5 feet from the muzzle they were 910.7 and 1116 fps (yeah, I know, crappy ammo). Nothing. One round of 7.62 x 39 at 2236 fps at the chronograph. Nothing. I switched 7.62 x 39 ammo getting 2366 fps (an estimated 2300 fps at the target). Nothing except the box was torn up so badly there wasn't much material left in it.

I ran out of time (and being a "cop magnet" I got a speeding ticket on the way home -- 57 in a 45 MPH zone...), but I now know that ANFO, MnO2, and SiO2 combined won't detonate with rifle fire. At least not at target velocities of <= 2300 fps at 45 F. Next time I'll try adding some small quantities of KClO3. Maybe I can get detonation with much smaller quantities of KClO3.  And I suspect that as I increase the amount of KClO3 the sensitivity will increase.  I'll increase it until a supersonic .22 LR will set it off, but a subsonic round will not.  At least that is my plan.

Other possible catalysts for the NH4NO3 include potassium dichromate, and ammonium dichromate.  But they are both rather poisonous.  I have some on order and if the benefit is worth the additional hazards I'll give it serious consideration and a full report here.

I did some very small scale experiments with the potassium dichromate and ammonium dichromate.  I mixed up about a teaspoon of the mixtures and did friction and impact tests with hammers and steel plates.  It didn't seem (I know -- very subjective) nearly as good using the MnO2 as a catalyst.

April 11, 1999 Update:  

I finally got out to the range with the MnO2 catalyzed mixture.  Fred and Mike hung around after the NRA Personal Protection Class I taught and helped with the video, shooting (Fred supplied his .22 pistol and ammo), and offered helpful suggestions.  And most importantly they were available to check and see if there were enough piece left of me to consider calling an ambulance should I have an accident.

The NH4NO3 was ground up in a blender into fine sand sized particles the night before and sealed from contact with air for combining with diesel and other stuff on the range.  The assumed specific density for the various components are:
NH4NO3 : 0.90
KClO3 : 0.93
MnO2 : 1.32
Diesel : 0.82
C10H8 : 0.54

Yes, I know, I should be using mass, not volume with the powders.  But it's tough to do out on the range.  None of the powders were packed, all were fluffy, like you would do with flour, in the kitchen, making bread. 

Mix 1:
I mixed up 600 mL of NH4NO3 with 40 mL of diesel and 22 mL of mothball flakes (naphthalene, C10H8).  In a separate bowl I mixed 100 mL KClO3 with 10 mL of MnO2 and 0.5 mL of baking soda.  Then the two mixtures were combined.  The hypothesis of using some C10H8 was that it would sublime in the milk carton creating a vapor that would be easily ignited with the free oxygen from the KClO3 and the NH4NO3 .  This reduced the amount of diesel required which I suspect reduces the amount of heat required to get the mixture up to temperature (assuming temperature is the triggering physical event).  This is very, very lean on the KClO3 compared to what I used last year, about one fourth as much.  The cost per target would be in the $0.50 range.  I called this my "lean mixture".  The baking soda was to reduce the chance that acids would form and cause spontaneous decomposition from the breaking down of the KClO3 .  This was suggested by one of the fireworks books I have been reading on working safely with the chlorates.

This mixture was put into three 1/2 pint milk cartons with 200 grit aluminum oxide sandpaper on the back inside surface.  A fourth carton without sandpaper was filled also.  The cartons were taped shut and put on the berm about 30 yards away.  At least one solid hit was made with a .22 LR from both a pistol and a rifle without anything more interesting than puffs of dust.  Out came the .223 with 62 grain FMJ ammo.  These are a little slow in this short barreled gun, about 2490 fps at the target.  They didn't do anything noteworthy either.  Next came the 50 grain ammo using Hornady VMAX bullets. These bullet clocked in at about 2860 fps at the target. This detonated everything.  Even the milk carton without the sandpaper.  Here is the video snippet from this first carton of the "lean mixture" we got to detonate.  Fred was just kidding of course... :-)

Mix 2:
The first mixture was rewarding, but not satisfactory.  This would cut the price of the targets in half and obviously it's fairly safe.  But the requirement of the higher speed bullets (also expanding rather than FMJ -- what was it that made the difference?) is unacceptable. Next I mixed up 500 mL of NH4NO3 with 60 mL of diesel.  In a separate bowl I mixed 200 mL KClO3 with 10 mL of MnO2.  Then the two mixtures were combined.  The difference from before is a higher percentage of KClO3, no C10H8, and no baking soda.  I really should have put in the C10H8 with less diesel, but I was in a hurry.  This mixture was also put in cartons with sandpaper.  Again the .22 LR ammo wouldn't do anything.  The 62 grain .223 ammo got some smoke out of one of the cartons and the 50 grain VMAX bullets detonated it.  No difference between sandpaper and not (the one that smoked was sandpaper, however another one that didn't smoke also had sandpaper).

Mix 3:
Next was 200 mL NH4NO3 with 25 mL of model racing fuel (45% nitromethane CH3NO2 plus methanol and lubricating oil).  In a separate bowl I mixed 33 mL KClO3 with 3 mL of MnO2.  Then the two mixtures were combined. I think I may have screwed up on the racing fuel.  I hadn't (and still haven't) done the proper chemical balance stuff to make sure I have the right percentages.  There was a little smoke with the 50 grain VMAX bullets, but no detonation.  One of my helpers believes he can get some pure nitromethane.  This would help with the chemical balance problem (who knows what the percentage of methanol and oil is and the type of oil?).  Also, it's the CH3NO2 that is of interest to me.  Supposedly, when hot, this will explode from impact without additional chemicals.  It is a bit expensive though.  See MSDS for more information on it..  No time to explore this before the Boomershoot later this month, but maybe for the next.

Mix 4:
Next I mixed up 200 mL of NH4NO3 with 32 mL of diesel.  In a separate bowl I mixed 150 mL KClO3 with 5 mL of MnO2.  Then the two mixtures were combined and put into 1/2 pint milk cartons without sandpaper. I called this my "rich mixture" because of the high concentration of KClO3 .  Approximately the same percentages as last year with the addition of MnO2 for a catalyst.  Again the .22 LR ammo would not get any results.  But the 62 grain FMJ (2490 fps target velocity) bullets reliably set everything off with a very satisfying thump to the chest.  Here is the video from the first of this mixture, my "rich mix".  I hadn't really realized it until this last test of the mixture that the previous ones were not giving as much "thump" as the first.  And the echo's from the surrounding hills that continued for a surprisingly long time also gave evidence that this was a much different result than previous mixtures.  Fred made a comment about it giving a quite a "wallop". Then Mike and I discussed it a bit more at length.  Both of us were quite sure that this last mixture resulted in by far the most satisfying explosion.

More tests are needed.  I still haven't reduce the cost, nor have I increased the sensitivity enough that a just supersonic .22 caliber bullet can detonate it.  I have confirmed that 1/2 pint milk cartons can be used which should increase the safety some.  I have also confirmed that adding the MnO2 does not make the mixture super sensitive to the point of an extreme safety hazard.

June 3, 1999 Update:

 Just so people know what I used (and I don't forget too!) for the April 25, 1999 event.  I would put 2 1/4 cups of NH4NO3 and 1/4 cup of mothball flakes in the blender and grind it up for about 45 seconds.  Then I would put that in a mixing bowl and add 1/4 cup of diesel.  Then I would put 2 1/4 cups of KClO3, 1/2 tsp MnO2 and 1/2 tsp baking soda in the blender and let it grind while I was mixing 1/4 cup diesel into the NH4NO3/C10H8 mixture.  After the KClO3 mixture was done I would stir it into the NH4NO3. Then the stuff was put into the 1/2 pint milk cartons (makes about 5) and closed up with tape.

horizontal rule

Email: Joe Huffman
Last updated: February 05, 2009