"What is flame?" - a contest entry
Annotated to illustrate some content design ideas.

In March 2012, the Center for Communicating Science at SUNY Stony Brook, held a contest: explain "What is flame?", to an 11-year-old. The "Flame Challenge".

Here is my entry (slightly tweaked). I've annotated it to point out some content design ideas.

Contest entries were constrained to be either plain text, a single image, or an isolated video. So when reading below, every time you wish a word was a link, a number was a graphic, a description was a video, or something was an interactive app ... yeah. Oh well. CCS is coming from journalism.

So this isn't content, just a writing exercise. And there's been zero user testing. Big picture, my hope was the contest would stir up a lot of expertise and collaborative creativity, which I might afterwards distill into illustrative content. Not much surfaced, so we'll see. But this isn't that.

Some things to note:

  • Small stories, woven in tapestry.
  • Answer common questions, and explain familiar experiences.
  • Avoid glaring gaps in descriptions.
  • Accessibility comes from greater insight, not from sacrificing correctness.
  • These four require a very high level of expertise. And are interdisciplinary. Which is (part of) why they are so very rare.
  • Real, non-toy understanding, has unfamiliar requirements. Such as learning the boundaries of concepts, instead of regurgitating labels atop fuzzy clouds of uncertainty.
  • Media matters. Attempting all this without non-linear mixed media, and interactive apps, is hopeless. Witness how badly I've done it here.

My context entry

... and some notes

"What is flame?"

The word "flame" has various meanings, both in common use, and across fields and professions. Why use physics and not firefighting? For contest-entry brevity, I dropped a sketch of "flame"'s consistency and variance. The one-liner is: mostly consistent use in the sciences and engineering, some divergence in professions, and a themed spread in common use.

What is flame? And what isn't? With stories.

Why include stories? Because how can you understand something without stories?

Why include jargon? Google surfing!

Jargon - communication, not understanding.
What role for jargon? It's a hook for googling. And vocabulary for student scientific discussion. But a mere label, attached to a concept-inventory failure, isn't understanding. And an awareness of meaning variation is needed.

What is flame?

Flame is combusting gasses.

Pithy definitions aren't always possible. Happily, for flame, in most of science and engineering, one is.

Combustion? A chain reaction of hot chemistry. Heat helps things burn, and burning makes heat. Fuel and oxidizer mix, hotter molecules smash harder, breaking, trading electrons and atoms, making more heat.

Gasses? For oxidizer, usually oxygen from air. For fuel, either something already gas (example: methane), or something nearby boiled (wood, candle wax).

Boiling wood, being unfamiliar, needs a video.

Assume active learners, and motivate them.
Organizing content 'basics first', or just-in-time, assumes and creates lazy learners. Structuring for active learning both permits richer, more engaging and powerful content, and creates the skills desired of science education.

With ambient google, and given desired skill outcomes, even a just-in-time approach to content seems unnecessarily restrictive. But I don't know of any research testing this speculation.

Whence light? Hot stuff glows.

Which cries out for an interactive app showing intensity and color vs temperature.

Whence flicker? Eddies from hot gas rising past surrounding air; and uneven burning of gas.

And this cries out for video, 2D and 1D flame models.

Blowing out a flame? The heat is moved away from the fuel. Breath on birthday candles; dynamite on oil well fires.

Blowing a flame hotter? More oxygen. Breath; bellows; blast furnace; basic oxygen furnace.

Answer common questions.
Answering common student questions should be a minimum standard for content.
"I know better than that!"
Ambient misconceptions and errors have value - being able to recognize institutions and high-status individuals "getting it wrong" is empowering.

Eg, the web has claims of 'dynamite snuffs fires by using up the oxygen'. A nice counter-story is dynamite fishing, which can segue to reefs, habitat destruction, and fisheries management.

What isn't flame?

Sketch concept boundaries!

Not flame... Hot gas, but not from combustion. The Sun. Auroras. Lightning. Neon lights.

Not flame... Detonation, instead of deflagration (combustion). A different chain reaction. Molecules smash, break into faster pieces, more smashing. Like dominoes. A shockwave. No oxidizer needed.

Acetylene tanks provide a nice 'combustion vs detonation' story. Running a tank low, and then banging on it, is bad.

Adding content can have negative cost.
By pruning misconceptions you would have spent more time fighting, adding content can save time.

Not flame... Combustion of solids and liquids. Directly, without turning them into gas first. See smoldering.

Non-toy understanding requires recognizing applicability.
Ideal gas laws don't apply to solids. But when did you ever see a problem whose correct answer was "the approximation is inapplicable, providing only a lower bound of...".

Some bite-sized stories

Story... Fuel and oxidizer - whichever ingredient there isn't enough of, limits how much flame gets made. (Sounds like cooking? Yes, cooking is chemistry). With non-engineered flames, the bottleneck is more often the oxidizer. Until the fuel is used up. Candles are orange and sooty because they can't get enough air to completely burn (the soot is unburned carbon). Unless you spray an oxygen tank at them. Then they get really bright, hot, and burn fast. Hospitals have signs saying "Danger: Oxygen In Use: No Smoking or Open Flame". Apollo 1 used pure oxygen "air", intended to be low-pressure... but a test was done at sea level. Death.

Enable students to tell stories.
To create and exercise understanding. But this requires much better content.
Use stories to prune misconceptions.
Misunderstandings are robust, requiring identification and focused targeting. Stories, at least in a collaborative active-learning setting, may be a useful way to scrape off misconceptions.
Use failure-analysis case studies.
Case studies, accident and failure analyses, and project post-mortems, are perhaps the most widely enjoyed engineering and professional literature. They both provide insight, and demonstrate that it matters.

Science and engineering do matter. As with so much else, "getting it wrong" costs people time, quality of life, and life.

Story... A candle flame in space is round, blue, slow and cool. No gravity, so less dense (hot) gas doesn't rise. Thus round. No draft sucking in fresh air, so oxygen only slowly diffuses in, instead of being turbulently mixed. Thus slower and cooler. Soot isn't pulled away without having a chance to find oxygen and burn. Thus not much yellow glowing soot. That thin blue carbon glow at the bottom of an Earth candle is what's left. Thus blue.

Off-topic story... Astronauts sleep near fans to avoid CO2 headaches and suffocation. From exhaled CO2 accumulating around them in a ball. Instead of rising (hot breath), or sinking (heavy CO2) away.

Story... Smoldering is non-flame combustion. A porous solid or aerosol liquid (so there's lots of surface area for the oxidizer to reach) can burn without first becoming gas. They just glow. As with cigarettes. and glowing charcoal bricks. Smoldering is cooler and slower than flame. But a fire can switch from smoldering to flames, and back. So a flaming match ignites a smoldering cigarette, which ignites a smoldering couch. Which somewhere gets hot enough to make gas and ignite it, becoming a flaming couch. Which may later cool to smoldering ashes, until the fire gets cold, or runs out of embers, and dies.

Story... Combustion is one kind of redox reaction (electrons moved between atoms and-or molecules). REDuction-OXidation. Others are rust, and your mitochondrial redox.

Providing context.

Story... Mitochondria (critters living in your cells) use protein machines, electron transport chains, instead of combustion's smashing brute force. To combine carbon (from food) and oxygen (from air). Getting energy, and CO2 waste. Powerplant of the cell. It's cooler than combustion. But while lots of stuff combusts if you get it hot enough, steel for example, we can only burn carbohydrates and lipids (and protein if you're starving). So you're a cool fire, sort of. Or, since the energy freed was originally from sunlight, you are living sunlight.

Thanks to Penny Chisholm (MIT) for "you are living sunlight".

Story... The firefighting "fire tetrahedron" is: fuel, oxygen, heat, chemical chain reaction. Reduce (ha, pun) the oxygen or fuel or heat, and if it's the bottleneck, you reduce the fire. Or you can mess with the reaction. A fire stops if gets cold, or runs out of fuel or oxidizer. So you can kill a flame by making it cold (adding water; spreading out campfires; bulldozing firebreaks; ripping open roofs; or blowing with mouth, wind, and explosives). Or by denying it fuel ("turning off the gas"; sand buckets). Or oxygen (displaced by C02 or other inert gas; snuffing a candle with a cup). Or by interfering with the chemistry (Halon steals hydrogens). Dry chemical fire extinguishing agents do a variety of these.

Story... Before matches, there were fire pistons. Tube with a slider. Slam slider down. Compression makes air hot (SCUBA and SCBA tanks are filled sitting in cool water). Hot enough to start a small smolder.

Story... In a tree: sunlight + CO2 + fertilizer -> wood + O2. In a campfire: wood + O2 -> light'n'heat + CO2 + ash. You get back what you put in, but the light is weaker.

Off-topic story... Yes, tree: sunlight + CO2 -> O2. But! A tree is also O2 + sugar -> CO2 + energy, just like us. Rainforests are _net_ producers of O2 for only a few hours around noon. Not all the time.

Rather than facing the misconception directly, this would be better as stories. A graph of rainforest products with time of day. Or plants dying in inert atmospheres.

Story... Premixed (fuel and oxygen) flames burn underwater, in inert (nitrogen) atmospheres, and in space. Rockets carry both fuel and oxidizer (cars - only fuel). They can't get air fast enough on the ground, and there's no air in space.

Story... The usual flame: oxidizer everywhere, inject fuel. Inverted flame: fuel everywhere, inject oxidizer. Picture an imaginary alien civilization in a methane atmosphere. Like early Earth (the methane, not the aliens). Vehicles carry oxidizer tanks. They are "oxidizered up" at oxidizer stations. The aliens eat oxygen rich foods.

Thanks to wbeaty's reddit post for the aliens.

Story... Atoms are like superballs. With magnets, so they clump. They form shapes, like connect-sticks toys. And they blink when you smack them.

Story... Atoms are a rowdy mob of superballs in a nano mosh-pit of hell. Think of bouncing bullets. Temperature says how fast the molecules are moving. Oxygen (O2) in air averages 500 meters per second (1000 mph) at room temperature. And twice that in a candle.

Thanks to someone (talk at MIT) for "rowdy mob".

Story... So why do farts spread so slowly? Think mosh pit. Molecules barely move, before they're bounced. They zig-zag (gas), or just jiggle (liquid).

Story... If atoms were superballs, your room is the size of virus (100 nm). In a 100 nm wide mosh pit of air, there'd be 5000 superball pairs (4/5 N2, 1/5 O2) zipping around the room. At... faster than the speed of light. Sigh. You need to slow time waayyyy down (1 second lasting years) to see them bounce around. They hit each other about once per room crossing (mean free path is about 100 nm). It's all very violent. Nanometer stuff happening in nanoseconds.

Story... But molecules are so strongly unbreakable, they mostly just bounce. You need them heated 50% faster to get lots of them breaking apart and recombining.

Story... The superballs stick together, like with magnets. They clump. Often they resist being pushed together (like reversed magnets), until they get close, change their minds (unlike magnets), and snap together. The snap speeds them up. Combustion requires getting atoms moving fast enough to smash apart and snap together. The superballs blink different colors when they spin, snap, and smash. That's the light, colors, and heat you see.

My coverage of light is very incomplete - lack of expertise. Spinning is IR, but how much of IR is spinning? Blue is C2, but is that on smash, snap-n-resnap, or both? These are just the simple, obvious things you would report about a flashing superball. Clarifying insight... from better experts.

Misunderstandings are common. Visual soot is not blackbody (way too cool), merely a 'mess of lines' with blackbody-like shape.

Story... How long can a flame last? A leak from a natural gas deposit can burn for decades or centuries. A hydrogen balloon burns faster - "bang".

Boundaries, the edges of the envelope, are interesting places. And illuminate what's happening away from the edge.

Story... The upper limit on flame size is keeping it fed. And deciding what to call a single flame. Forest firestorms create their own wind, and can explosively boil pine trees into tree candle fireballs. The lower limit on size is keeping warm. If the reactants are already hot, the flame can be a thin sheet or a tiny dot. But being smaller gives more surface for each bit of volume. Millimeter-sized turbines would be great for long-lasting refuelable cell phones, but when a tiny flame brushes a wall, it gets cold and goes out. So it's not easy.

Story... Lower pressure, gives less oxygen. Campfires are difficult at 2500 m above sea level (Bogota). At 4500 m, cigarettes go out, and paper burns 20% slower. At 10 km, paper doesn't burn. You need an O2 tank to burn newspaper on top of Everest.

Partial pressure needs more stories.

Needs a graphic.

Story... At sea level, the flammability limits of hydrogen in air are 4% and 75%. Methane: 5% and 15%. Gasoline 1% and 7%. Too much or too little, and the fuel doesn't burn.

Number salad, because no graphics.

Story... The atmosphere is 21% oxygen. The minimum oxygen concentration to burn paper is 14%. For hydrogen, it's 4%. To snuff a fire, get it lower.

Pressure, %O2, pO2, and combustion rate, cry out for a fun interactive app with videos.

Story... Earth's atmosphere is currently 21% oxygen (by volume). It was 0%, then low (1-10%), then jiggled (13-35%). Before the Great Oxidation Event (2.45 billion years ago), it was near zero. For the last 450 million years, fossil charcoal shows it's (mostly) been between 13% and 35%. So? Under 13%, no wildfires. 13-16%, wildfires are rare, and only very dry plants burn. >25% fires become widespread, even in wet areas. >30% fires everywhere. >35% even wet plants burn.

More number salad.

Needs a "% by mass" link with story.
Needs photos of regional smoke plumes.
Needs a story of a high-oxygen period, describing continents and flora, making it visualizable.

Story... House fires are ventilation limited. A flaming room uses oxygen from connected rooms. With closed doors and windows, it runs out. Sleeping people die, and the flames quiet. Until a door or window is opened. Firefighters, guided by anecdote instead of science, ventilated too much. Making bigger, faster fires. Excess death. Two lessons... Close doors between you and a fire. Use science.

Source: Impact of Ventilation on Fire Behavior in Legacy and Contemporary Residential Construction

A related 'consequences of non-scientific analysis' story is arson investigators getting innocents jailed based on misconceptions about fire behavior.

"Close doors [...]" ties science back to real-world, and actionable, health and safety suggestions.

Everything is connected. And the world is full of wonderful stories.

And that's it.

Missing stories

Stories I particularly regret not including:

  • Details of the light.
  • Turning candle flames off/on with cooling/heating wire meshes.
  • Acetylene tank combustion and detonation.
  • The great velocity difference between diffusion and convection.
  • Partial pressure.
  • Oxydizers other than oxygen. And hypergolics.
  • Pyrolysis.

Some other absent stories:

  • Self-relighting candles.
  • EM tweaks. Heating flames with microwaves. Deflecting flames.
  • Flame oscillations.
  • Flame resistant garments.
  • Radiant heat. (required a rough quantitative context)
  • Beatrice the Biologist's anthropomorphized chemistry (video). "oxygens get crazy and steal stuff", etc.


  • I worry/hope that the flame-vs-smoldering distinction might break down at nanoscale, thus requiring unification.

Other thoughts:

  • Lacking a larger instructional context, to provide a sense of scale, and skill with rough quantitative reasoning, I here did only ad hoc superballs for scale, and almost nothing for RQR. Much better is possible.
  • Non-hypertext meant all stories had to be inline, making them harder to write, and less interesting.
  • Too many of the stories are passive, without thought-spuring reader interventions.

Thanks for your interest.

Page history

A page draft. Thanks to L for comments.
The contest.
Mark it up No JS?
Email me

Fb share