Pique the Geek 20090705: The Art and Science of Fireworks

( – promoted by buhdydharma )

Since yesterday was Independence Day and many of us saw a fireworks display, shot off our own fireworks, or both, I thought it would be timely to describe how fireworks actually work.  All fireworks have one thing in common:  they give off heat from some sort of chemical reaction.

Most fireworks also give off light, and most also emit some sort of sound.  Notable exceptions to this are sparklers which have little sound, smoke “bombs”, and the large set pieces that display shapes like an American flag or some such.  We shall focus on the ones that give off light and sound.

We shall divide the way that fireworks operate in the art and the science of them, and look at the science first.  Let us look at the chemistry.

All pyrotechnics, of which fireworks are a subset, contain, at a minimum, at least one fuel and at least one oxidizer.  The fuel provides the energy content and the oxidizer provides the oxygen necessary to burn the fuel at a rate depending on the purpose of the particular pyrotechnic mix.  For example, the oldest pyrotechnic mixture, blackpowder, contains charcoal and sulfur as the fuels and potassium nitrate (saltpeter) as the oxidizer.  Blackpowder is manufactured in such a way as to be a very fast-burning pyrotechnic, and so is useful as a propellant and as a bursting agent.

As a matter of fact, almost every firework that moves has blackpowder or a close derivative as the source of motive power.  From the humble bottle rocket to the very large professional mortars, blackpowder is the lifting charge.  There is really not a suitable substitute for blackpowder in fireworks.

Until quite recently, potassium nitrate was the only oxidizer available for fireworks.  The problem is that potassium nitrate begins to give up its oxygen only at relatively high temperatures, thus making colors other than white and yellow impossible.  Alas, the Founders of our Nation had only white and yellow for their celebrations.  (“The Rockets’ Red Glare” is a product of heat, not a true firework color in the modern sense.  Scott Key’s red glare was more like the element in an electric oven, not what we expect in modern fireworks).

The introduction of potassium chlorate in the 19th century revolutionized the manufacture of colored firework effects.  Potassium chlorate begins to give up its oxygen at a much lower temperature than nitrate, and so does not “wash out” colors.  Potassium chlorate is trickier to handle than potassium nitrate, so there are a number of safety precautions involved in working with it.  However, it allows the huge range of colors now routinely observed in modern fireworks.

Now for a little of the physics of fireworks.  The colors (other than white) that we see are produced in a couple of ways.  One way is by burning a material directly, like granular charcoal or iron filings to produce orange colors.  White is usually produced by aluminum powder because it burns at such a high temperature, the iron and charcoal at lower temperatures.  But these are not the brilliant colors of which we normally think.

The brilliant colors are produced by atomic emission, where an electron is promoted from a lower electronic level to a higher one by heat, in the case of fireworks.  As that electron falls back to the lower electronic level, light it emitted.  If everything is just right, this light is emitted in the visible.  That is why nitrate is a poor oxidizer for fireworks, because the high temperature promotes the electrons to a level too high to produce visible electromagnetic radiation when the electrons falls back.  However, potassium chlorate is just right for promoting electrons in many materials to just the right level for visible emission.

The easiest colors to make are green, red, yellow, orange, and gold (in addition to white).  Green almost always is produced by barium salts added to the pyrotechnic mixture (more on these mixtures in a bit).  Red is most often produced by strontium compounds, and the red produced by those materials is deep and intense.  Lithium salts can also produce a red, and it is lighter in shade than strontium red.  At large public displays sometimes you can see the two different shades, but the strontium one is much more common.  Yellow is produced by sodium (the reason that 18th century fireworks had a yellowish tint was that the potassium nitrate used was contaminated by sodium, and sodium is very forgiving when it comes to high temperatures).  Orange is produced by calcium salts.  Gold, as we said earlier, is either charcoal granules or iron filings.  The iron filings also give a sparkle effect.

By far the most difficult color to produce reliably is blue.  There are a couple of reasons for that.  For starters, the human eye is relatively insensitive in that wavelength, so a very intense light source is required to see it very well.  The other reason is that no element has a sufficiently intense blue emission line to be more than dimly visible.  Note that without exception, all of the colors mentioned above are from elements, which are fairly temperature insensitive.  The only good source of blue of sufficient intensity is the copper/chlorine molecular ion CuCl+.  This is an extremely fragile species, easily destroyed by high temperatures.  To produce and excite it to emit blue light, a copper salt and a chlorine donor are mixed together.  The chlorine donor is often powdered polyvinyl chloride plastic, which has the advantage that it absorbs some heat because it melts before donating the chlorine, cooling the mix.  Other coolants are also used to control the temperature.  If done correctly, a very bright, true blue is produced.  Not too many years ago blue effects were extremely rare.

Purple is also hard to make, but generally a strontium salt is mixed with the blue mix to add some red to the blue, making purple.  The temperature considerations are the same as for blue.  Similar to blue, purple effects are comparatively recent.

Now let us look at some of the engineering considerations, and this has to do mainly with lifting our firework to a height suitable for display and for most of the burning debris to go out before hitting the ground.  This is a fairly straightforward problem.  You know how heavy your payload is and the energy output from the blackpowder lifting charge is known.  It is fairly straightforward to calculate the amount of blackpowder to lift the projectile or rocket to the desired height.

Another engineering consideration is timing the bursting of the rocket or mortar.  This is often done with quickmatch, basically blackpowder impregnated cotton twine, manufactured to very close tolerances.  For cheap consumer rockets, the delay is just the time that it takes the column of blackpowder lifting charge to burn to the bursting charge, but for mortars that is not possible.  In most professional displays, a length of quickmatch is inserted into the bursting charge in the center of the mortar, and it is of the right length to ignite the bursting charge at the apex of the mortar’s lift phase.  Then it blows to bits.

Another engineering aspect is the coordination of firing of the different mortars in professional displays.  These displays are set up the morning of, or the day before, the display and everything is loaded.  No one is allowed near the fireworks once the display starts for safety reasons.  No mortar tube is ever reused for a given show due to the possibility of hot embers remaining in the tube, thus inviting disaster.

All modern professional displays are fired electrically.  For a typical mortar round, an electric match (also called a squib) with electrical leads is placed in the bottom of the tube.  Then the lifting charge of however much blackpowder is added, then the actual projectile.  All of the leads are numbered and connected to a master control panel, and each device is ignited by a switch in a predetermined sequence.  The largest ones are actually computer controlled, with the sequence programmed into a laptop which in turn controls the master control panel.  This makes coordination with music much easier than by manual firing.  No one is out amongst the tubes with a punk, lighting fuzes.

Now let us consider the art of fireworks.  Firecrackers are easy to make, but are not very artful.  Just stick a cheap fuze into a paper cylinder packed with flash powder and you are done.  Noisy, but not very elegant.

There are lots of different types of fireworks, so let us imagine a mortar with a diameter of say, six inches (that is a big one) that will burst to produce a spherical dispersion of stars (the trade term for the actual submunitions in the device that you see glowing).  Let us further say that we want two spheres, one inside the other, and for the outer one to start out green and change to red before fading away, and the inner one to be first blue then white.  This is not uncommon, but the technical details are daunting.

First, two different types of stars are required, and each type have to be almost exactly identical to each other so that the transition from one color to the other is uniform.  In addition, two different bursting charges have to be used to create the two spheres.  First we make the stars.

Stars are made of a fuel, often charcoal (aluminum for white) or other neutral material.  An oxidizer, potassium chlorate, is added to the fuel and then a binder is added.  The binder serves to hold everything together during the considerable pressure developed by the bursting charge.  Traditionally, shellac has been used as a binder, but synthetics are becoming more widely used.  Too make stars, it is essential to produce a plastic mass that can be controlled carefully in size and shape, usually spherical.  Part of the mix is moistened with whatever solvent us used (alcohol for shellac binders) and the color salt is added, in this case strontium (we want red as the second color, so it has to be on the INSIDE).  The mix is then blended (almost always in a separate structure, by remote control for safety considerations).  When blending is complete, the mix is taken to a “star room” where a mold is used to produce the proper sized stars.  This is pretty much done by hand, most often under the protection of a high-speed deluge system, which will flood the entire room with hundreds of gallons of water a minute if the sensors detect a fire of any kind.  Once molded, the stars art taken to a dryer (often forced warm air) to dry.  Now we have to make the outer layer, using barium in our mix since we want it green first.  The stars are placed in molds and the outside mix is applied to them in a uniform layer.  Then back to the dryer.

The stars for the inner sphere are made in the same way, but the inner portion has aluminum powder since we want it to turn from blue to white, and the outer shell is the copper/chlorine donor mix.  All of the stars may get a surface treatment with a very easy to ignite mix to assure that the blackpowder bursting charge ignites them, but this does not contribute to the color.

Now to assemble the round.  Traditionally, paper and glue were used to form the containers for each layer, but this practice is being replaced by premolded plastic components, which are more uniform and easier to use.  Imagine, again, what we want:  an outer sphere that starts out green and then changes to red, and in inner sphere that starts out blue and then becomes white.  Thus, we need two layers of stars of the different types, and two bursting charges.  In addition, we want the outer sphere to be larger than the inner one, and for it to be propelled just an instant before the inner one.  We have to build this from the inside out, so here are the steps:

We take a small burster container (all of these containers are two hemispheres that are glued together as they are filled) and fill it with the appropriate amount of blackpowder and add the appropriate length of quickmatch to get the desired delay (a fraction longer than the other delay).  Then we take the container for the inner sphere of stars and fill it with the appropriate volume of stars and glue it.  Next, we take the outer burster and place the appropriate amount of blackpowder in it, add the quickmatch, and glue it.  Finally we take the outer star container, fill it with the different stars, and glue it.  The round is now complete.

Basically, we have essentially made a bomb.  As a matter of fact, yesterday morning one technician lost his life and three or four others were seriously injured unloading a truck of fireworks for a professional display in Virginia, as memory serves.  Imagine a 40 minute display taking four seconds to complete.  That is what witnesses said happened there.

I am well aware of the hazards handling pyrotechnics.  Many of you do not know that I am an Army-certified Ammunition Handler, a certification that I had to get when I ran a developmental facility for new Army pyrotechnics.  We mainly did white and colored smoke formulation research, but we also worked with more energetic mixes like those used for flash bang and rubber pellet filled less than lethal (riot control) devices.  As a matter of fact, I (along with a couple of other fellows) hold a patent for a high order flash grenade mix that is safe to handle, unlike the old flash powder that goes high order just under its own weight.

I hope that everyone had a wonderful Independence Day observance.  I was afraid that the professional display would not take place here since it rained most of the day yesterday, but at 10:00 PM it started, as the heavy booms of the first mortars told me.  I drove up to a clear area and watched from the comfort of my car (too many trees from my back deck) until it was over, that that made me think to write this piece to give you a flavor for how this is done.  Any specific questions about other pyrotechnics are welcome.

Well, you have done it again.  You have wasted a perfectly good batch of electrons reading this poor post.  And even though Gov. Sanford finds another soulmate every time he reads me say this, I sincerely mean it when I state that I learn much more than I could ever hope to teach by writing this series.  So please keep those questions, comments, corrections, and suggestions coming.

For the life of me, I can not get the “L” on Learning in the tags to capitalize.  Every time I try to change it, it goes back to lower case.  Any suggestions?

Warmest regards,


Crossposted at dailykos.com

Is the Pony/Pie/Hide rating system too cutsie?

View Results

Loading ... Loading ...


Skip to comment form

  1. come from pyrotechnics?  (This is a very subtle reference to the series that I am finishing up soon, and also a very popular American band from the 1960s and 1970s.  A pony to anyone who can guess both).

    Warmest regards,


  2. Two twelve packs of bottle rockets in each bottle.  Plumbers torch ignites fuses of bottle rockets.

    Keep face down and low to ground and please do not do in 1/4 acre suburban lot.  Need like really open space away from flammable stuff.

    “I always get “hit” when I do this stuff.”

    Wife says “You never told me that”

    “Ya, but was it fun, did you like my fireworks display?”

  3. buhdydharma.  I very much appreciate it.  It is nice to have a place to write where my thoughts are appreciated, and to think that I might be opening other minds.

    Warmest regards,


Comments have been disabled.