Hacker News new | past | comments | ask | show | jobs | submit login
The tragic physics of the deadly explosion in Beirut (wired.com)
264 points by ChuckMcM 5 months ago | hide | past | favorite | 175 comments

For all the horror, at least this happened at the quayside and half the explosive force radiated away almost harmlessly over the sea. That basically halved the actual destruction at the outset. If it had been in the middle of the city, there would have been twice the damage. This much is fairly obvious.

What isn’t quite so obvious is that roughly half the energy was also wasted upwards, and that a startlingly worse case would have been if the explosives had been slightly interred, coupling the blast more closely to the ground and enhancing amplitude of the p-waves and providing plenty of really substantial debris raining down after basically ‘earth-quaking’ the buildings.

Roughly combining the two 50% ‘wastages’ leaves you with about only one quarter of the energy was actually really channeled into destruction, which I’m going to be charitable and call a third. In ‘ideal’ conditions you might’ve had three times as much devastation, and that’s a pretty shocking thing to consider, given the actual carnage.

NOTE: I had appended these considerations in a more inchoate form to the tail-end of another comment, but I figured I’d make them an item of their own and tidy them up a bit. Apologies for the duplication.

The grain store probably shielded a lot of other infrastructure and people too.

However it seems very irresponsible to keep 85% of the country's grain in a single spot. It's scary how so many single points of failure exist.

Yeah I’ve been thinking about that too... harsh as it may sound, big concrete cylinders fillded with loose grain is pretty much the best pillow to cushion the blow you could think of.

EDIT: As for the grain silos being “single points of failure”, that depends on the scope of the system you’re considering. Nationally, yeah... but internationally, there’s plenty and it’s being shipped in. So there’s actually more redundancy than the narrow-scope analysis would suggest. The 85% figure is just a figment of where one draws the borders, and national borders probably don’t have much valence in the given context (fortunately). Now, if we were speaking about 85% of the world’s ability to ship grain... now that would be a catastrophe of unimaginable proportions.

I did consider that the dust present in a grain silo could make made a 'flour bomb' and compounded the explosion. Maybe not

To be perfectly honest, that was my very first reaction when I first watched the video: it looked like the fuming aftermath of Something Really Bad, and then BOOM! Something Really Bad unfolded on-screen. I figured that it must’ve been some kind of aerosol/powder combustion triggered by the fireworks’ conflagration, but then more learned analyses of cloud hue and knowledge of what was actually stored in situ (correctly) overrode my initial reaction and the fertilizer was identified as the main culprit.

There probably won't be much shipping of any kind in that port for a while... Hopefully they can make other arrangements.

Lebanon is a tiny country, the size of Connecticut. There are a lot of things that were irresponsible in this story in Beirut, but I don't think the grain silo is one of them

There have been some works to expand the Tripoli port to avoid single point of failure, but apparently (from the podcast I heard) they've been going slow for political reasons (including regular people of Beirut profiting from the near-monopoly).

I wonder how much of that was by design, if any.

An old friend used to work in a chemical factory in, must have been Minnesota in the 60s or 70s.

She would talk about how they had a row of sheds with three walls for mixing certain dangerous chemicals. Each shed had an open or really thin wall facing the river.

If one went up, well, that person was probably gone, but at least it wouldn't start a chain reaction. It would explode in exactly one direction -- away from the plant, away from all the other dangerous chemical sheds.

Someone said that a small cracker on your open hand would hurt and burn, but if held closed in your fist, you'd have a hole in it.

I have actually had this happen when I was young. We have legal fireworks called a 'Kanonslag' (loosely translates to cannon blast). Those contain about 0.45g of black powder. I ignited one, threw it, and it didn't explode. Now, my dad had it drilled in that you never pick up a previously ignited firework, but I could still see the fuse sticking out, only smoldering, so in my mind nothing could go wrong. When I picked it up and clutched it the fuse slid into the firecracker and it blew up. I felt my hand forced open and it felt like each of my fingers had been slammed by a sledgehammer. Now, I didn't suffer any permanent damage, but my hand rung for a solid hour or so.

Then there are also illegal fireworks like 'Bex nitraat' (Bex is the brand name, and nitraat doesn't require translation) from Poland. Those will maul the entire hand, and are a guaranteed loss of fingers.

There is other illegal stuff like a 'Vlinderbom' (Butterfly bomb, known as a Spanish cracker in other countries). These have the explosive power of a small hand grenade and the aftermath of using one incorrectly doesn't need elaboration.

I used to play around with m80's and m100's as a kid, those put the fear of death in me cause the wick literally lasted 3 seconds before ignition. I think they were some unregulated hand made fireworks my neighborhood friend got from Mexico. The sound was louder than a gun shot.

Also dry ice in bottles was fun and spud cannon got me interested in physics. Good times.

What's the point of drilling someone about only one specific danger of an extremely dangerous object?

Not the worst idea to specifically point out scenarios with a high ratio of actual-vs-perceived danger

better than nothing

Plenty of footage and images on the internets what happens when you hold fireworks in your hand. Not so much a hole as your fingers get shredded / blown off.

The commenter above was talking about a literal wheat cracker, not a firecracker. They were mentioning how much energy is actually stored in grain(it's a lot).

No, I don't believe so. I believe they were indeed talking about a small explosive load, like a firecracker.

On second thought, I think you're right. A cracker in a closed hand would get snuffed out.

well 1) lack of vocabulary on my side, I meant small stick like explosive for celebrations (1x6cm)

2) you might be right I didn't pursue video evidences of shredded fingers and stayed with the dude saying

Haha, no, he wasn't. He was talking ... well, this: https://www.youtube.com/watch?v=qk9MK5smzVE&t=3m44s

Armageddon (1998, movie)

You are assuming few things here, first there were victims on the water front,a cruise ship sank with people on it and other, one survived "Beirut Port Worker Thrown Into Sea by Explosion Found ALIVE 30 Hours Later" the explosion left a 43-meter (141 foot) deep crater.NYTIMES:" U.S. Contractor Knew of Explosive Material in Beirut Since at Least 2016

An American contractor spotted and reported the potential danger at least four years ago, but U.S. officials denied they were aware of the findings until last week, after the blast."

Yes I’m assuming a huge number of things, but that’s the nature of ballpark figures and estimation. We’re well into “spherical cow” territory here.

> half the energy

No, half of the remaining energy. You can't discard two halves and expect to have something left.

I’m not subtracting, I’m multiplying fractions.

Half stayed on-shore; of that half, half went into the sky. That’s half of a half which is a quarter, which I’m rounding up to a third because later I take the reciprocal so the higher proportion is a more conservative estimate.

> Just after 9 am on December 6, 1917, in Halifax, mustachioed 45-year-old train dispatcher Vincent Coleman knew the lazy plume of smoke coming from the explosive-laden vessel in Halifax harbor was a terrifying portent of worse to come. But as the other dispatchers ran for their lives, Coleman realized the number of train passengers set to arrive at Halifax any moment, and so he stayed long enough to send one final telegram: “Hold up the train. Munitions ship on fire and making for Pier 6 ... Goodbye boys.” Coleman died in the blast at 9:05, but his final message saved thousands, not just the passengers on the trains that were able to stop before entering the zone of destruction, but also the citizens already in Halifax: The telegram signal reached every operator in the surrounding region, and because of Coleman’s quick thinking, every doctor who felt the earth rumble, up to 160 km away, had almost immediate access to news about what had happened. They rushed in to help.

This particular tit-bit at the end of the article is a gem. Times like these tend to also bring the best in some of us.

I had trouble understanding exactly how this scenario played out. Interestingly, Wikipedia gave some more color on the timing and distance involved [1]:

  The death toll could have been worse had it not been for the self-sacrifice of an Intercolonial Railway dispatcher, Patrick Vincent (Vince) Coleman, operating at the railyard about 750 feet (230 m) from Pier 6, where the explosion occurred. He and his co-worker, William Lovett, learned of the dangerous cargo aboard the burning Mont-Blanc from a sailor and began to flee. Coleman remembered that an incoming passenger train from Saint John, New Brunswick, was due to arrive at the railyard within minutes. He returned to his post alone and continued to send out urgent telegraph messages to stop the train. Several variations of the message have been reported, among them this from the Maritime Museum of the Atlantic: "Hold up the train. Ammunition ship afire in harbor making for Pier 6 and will explode. Guess this will be my last message. Good-bye boys." Coleman's message was responsible for bringing all incoming trains around Halifax to a halt. It was heard by other stations all along the Intercolonial Railway, helping railway officials to respond immediately.
Also the Wikipedia about Vince Coleman [2] is missing the details from [1]. I guess this is part of the Wikipedia Wormhole people go through...

[1] https://en.wikipedia.org/wiki/Halifax_Explosion [2] https://en.m.wikipedia.org/wiki/Vince_Coleman_(train_dispatc...

Indeed, thanks, I had missed that after skimming through the filler in that piece.


> Within that radius, the injuries from such a massive blast in a downtown location can be as varied as the victims who experience them, but a number of them are likely from glass and other flying projectiles. Flat, delicate, frangible, and installed in large sheets, glass is the perfect target for a blast wave of even minuscule magnitude; it shatters and flies easier than any other substance.

The glass damage must be horrible. Your entire home covered in tiny pieces of glass.

I hate when glass breaks at home because no matter how much you clean, it gets everywhere and months later you discover small shards in some distant place.

Once I tripped on my way home after grocery shopping, almost at the destination. It wasn't even a big fall, but some glass bottles broke. They were next to plastic bags filled with oats. In the "recovery operations" I conducted afterwards, I threw away the remainders of the bottles and cleaned the oats bags under flowing water, to get rid of the glass bottle's contains on the bags. As I did this, I noticed the oats themselves getting wet, below the transparent plastic. I wondered, how can this happen, it's supposed to be water-proof. Then I found a tiny, almost invisible cut where a glass shard has entered the plastic bag. It was a big shard and I don't know what had happened had someone not noticed it and ate it.

Horrified, I threw away the entire bag of oats, the risk of more, possibly smaller shards remaining was too big. Other bags which passed the "water test" I kept and I ended up eating their contents.

I hope the people of Beirut are doing similar checks for their stored food, for their own safety. I read that many folks already have glass shards in their bodies due to the explosion but that's past damage that can only be managed. The future damage is preventable.

> The glass damage must be horrible. Your entire home covered in tiny pieces of glass.

Watch this video at .25x to really see how glass reacts in the pressure wave. You can see the look on the bride's face as the initial vibrations/sound reaches her. Then, you can actually see the window pane bow like a wind sail.


RT/RUPTLY is well known for stealing other peoples' video and putting their huge gaudy watermark on it. Here is the same video where at least the media logo is on the bottom right corner.


Wow, you can really see in the glass reflections, the arrival of the ground compression (?) waves fractions of a second before the air pressure wave hits. And the wife notices it before the husband does.

Is the sound faster than the shockwave or same speed? I think same speed. Might have been a previous explosion that triggered the look before the main one.

I think you're hearing/seeing the difference between the speed of sound through the ground, and through the air.

Sound moves 4-times faster in water than in air, I assume even faster in ground surface, so let’s just assume it arrives instantly via ground. This would mean a delay like in lightning/thunder, which is what one can see in the video (2–3 seconds ~ 0.5–1 km). (This is the restaurant https://goo.gl/maps/td5K1vxUDBN4i1CA8)

But I could equally imagine it’s some lesser explosion before the main blast.

Object moves -> air goes back and forth -> your eardrum moves -> pressure pattern over time registers as sound as a 20-20kHz ish spectrum captured at 0.1s ish timesteps

Supersonic airplanes move past the speed of sound(280-340m/s in air) and generate wavefront at its interfaces with air but the resulting disjoint tsunami wave expands at the speed of sound and heard as single ‘bang’ when it reaches ears or microphones

So there’s no way one could hear the bang through air before visual bang hits

However the speed of sound is about 10x faster on solid mediums such as ground or seawater so could be those

It's actually two "bangs" for a plane. Tum-dum. Source: grew up about 100km away from a military air base in Russia at which at least a couple planes went supersonic almost every day for years.

You can hear several sonic booms in this video, although they're from low-flying planes: https://www.youtube.com/watch?v=jmhU7SEo4gg. The higher the plane, the more distinct and "bassy" the "booms". Here's another example, a bit higher, a bit more distinct: https://www.youtube.com/watch?v=M_KbZON5sW0

> Is the sound faster than the shockwave or same speed? I think same speed.

Shockwaves propagate faster than sound by definition.

The speed of sound depends on the medium, as a few replies have mentioned.

TFA says shock-waves travel much faster than sound's speed in air (343 m/s) whilst pressure-waves (as in the explosion in Beirut at 200+ m/s) travel relatively slower.

I was around during the Enschede, Netherlands fireworks factory explosion [0], just outside the radius where all glass shattered. Saw the glass ripple effect along a line of appartment buildings with full-height windows. A really astounding sight I'll never forget. With the amount of curvature caused by the shockwave it seemed these windows must surely break. But they did not.

[0] https://en.wikipedia.org/wiki/Enschede_fireworks_disaster

Why does the camera lens not shatter in this case?

And in the same vein, what happens to peoples’ eyeglasses?

Does anyone know?

The ratio of the thickness of the glass compared to the unsupported span of material makes them relatively much more rigid than a big, floppy pane of glass. A glass window is not only much less rigid, but also maximizes the surface area that pressure from the blast wave can push against.

This and the fact that eyeglasses are usually not firmly attached to a huge chunk of reinforced concrete.


Also behind a big glass pane is an enclosed room. Air can rush past the glasses.

Eyeglass lenses are usually made from plastic or polyurethane variants these days, as they can be much lighter, thinner, and tougher than glass with similar-enough optical properties to cover the needs of most eye conditions.

I consider PC a wonder material yet I went back to 1.7 AR coated glass (higher than that it's a bit overkill unless you really need it IMO). It's crisper, actually cheaper, heat resistant and virtually scratch proof if you clean it properly. It's heavier but I'm not going back to finding pinpricks after bouts of wind and all the extra care. The thing about glass is that it can shatter upon impact, plastic lenses won't but basically get ruined all the same in my experience.

I’ve not seem glass as an option recently. I miss having glass glasses for the exact reasons you describe. Worn plastic is scratched and give nasty artifacts.

For camera lenses, they are probably smaller, thicker relative to frontal area, and better protected against shocks and vibrations

Eyeglasses are actually almost always made of dense plastic material these days for better refractive indexes so they shatter differently, also I wonder if physics people consider it to be practically suspended in thin air at the scales of these explosions

Do plastic eyeglasses even shatter? Ever since lenses stopped being made out of glass, the amount of abuse they've been able to take has been through the roof.

I can't imagine a situation where my glasses might shatter before my face would.

They're quite small and thick: https://www.personal-view.com/talks/uploads/FileUpload/77/0c...

Windows break because they're thin and extend over large area, which make them flex a lot.

Similarly, you can see in the video how windows of the surrounding cars are largely unaffected.

Have a look at the vehicle glass in the video.

Oh no. I really hope the bride and groom are OK!

They're fine. And she's an MD who spent the better part of the rest of that day doing what she could for the people around her.


That's actually a _different_ wedding photo, happening at exactly the same time: https://www.youtube.com/watch?v=GQssU3dTKUs

A different couple

So why don’t we laminate or whatever our glass so it does what car glass does? Saving money?

Yes. The reason is we don't expect to have as many similar explosions that would make laminating glass necessary everywhere.

If we'd expect that, then we'd have bigger problems in the first place.

As a Lebanese who lived in Lebanon for 5 years (far away now, happily) - I always found it amazing that glass facades were being built on the newer buildings. For a country/city that's known for the occasional explosion it definitely doesn't seem like a good idea (as this event has shown).

That said, even regular windows were launched into people's houses (along with their frames) so I guess it wouldn't necessarily help here.

Laminating the glass would probably help for your regular/occasional explosion, but not for this, as you'd just have a giant pane smack you hard i guess.

Car glass is pebbled so it crumbles on impact to minimize damage.

Quite hard to laminate a surface that isn't flat-ish. Bottles etc are just blown. And they're disposable anyway, no point in making them several times more expensive for a rare situation.

Seems like using tempered glass for windows in addition to things like shower doors where it's required (in the US) now could help as well. Will still shatter, but won't splinter.

I guess that process is also expensive?

All architectural glass should be tempered, by safety codes. This is to avoid the problem of non-tempered (annealed) glass only breaking into a few large pieces, creating heavy sheets that will give someone deep lacerations. When you see a window instantaneously shatter into many tiny pieces, that is tempered glass. The energetic breaking of tempered glass still creates a lot of splinters though.

What is "architectural glass"? Only the bathroom windows in my house are tempered glass.

> The energetic breaking of tempered glass still creates a lot of splinters though.

Interesting. I guess my only real experience with tempered glass was dropping a shower door while installing it. It absolutely shattered into a million pieces right away, but they weren't what I'd describe as splinters. More like little cubes without particularly sharp.

Unless there were also splinters everywhere, and I just didn't notice in the mess of my bathroom.

It has to do with the size, but I don't know the limit. Smaller windows aren't going to matter, because pieces won't be large enough to be dangerous. A sliding glass exterior door is always going to be tempered glass, as well as glass tables and shelves. Tempering also makes the glass more impact resistant.

I broke a shower door too, while cleaning it. There were definitely splinters mixed in with all those cubes (and some in the hair on my arms). I'd imagine the physics is similar to when you snap a piece of uncooked spaghetti while holding it close to its ends - sometimes you end up with a small energetic middle piece that flies away.

it would be (very) expensive to make laminate glass bottles.

here is an amazing view of the explosion. https://youtu.be/hp-n-ghagok

The same photographer posted a very slow motion version of the same video: https://youtu.be/0tQ80Sj3QUs?t=122

Also note that you can hit "." on a paused video on youtube to go frame-by-frame.

Some observations:

* The camera vibrates long before the pressure wave reaches it. I assume this is due to energy traveling through the ground (i.e. an earthquake).

* On the third building from the bottom left there is some light reflected from another building's windows. Once the camera stops moving, these seem to still be moving.

* As the camera falls, two frames show a Nikon camera strap and a separate camera. It appears as if the photographer is taking stills. I haven't been able to find that these photographs were published, perhaps none were taken of the blast.

I live about 9km away from the site of the blast(as the crow flies). I can definitely confirm that the order of events was:

1 - Earthquake

2 - Sound (sounded like rolling thunder or a jet breaking the sound barrier)

3 - Pressure wave (felt as a sudden, short-lived, strong wind at that distance)

I am curious about 2 and 3. Since sounds is also just a pressure wave, those two should have been the same speed as well. I guess the sounds must have travelled through the ground as well.

This is explained in the article. It describes the pressure wave as moving slower than the speed of sound. It was not a shockwave which would be generated by high-explosives which would travel at the speed of sound and be much more deadly.

The speed of sound in pretty much any solid (or liquid) body is much faster than in the air. That's why the earthquake came first, and it was probably the earthquake that caused the 2. noises.

> * The camera vibrates long before the pressure wave reaches it. I assume this is due to energy traveling through the ground (i.e. an earthquake).

Had to check it out, earthquake p-waves travel really fast!

"Typical values for P-wave velocity in earthquakes are in the range 5 to 8 km/s."


I calculated ~400m/s for the pressure wave (judging by destruction, not the white condensation wave).

I guess it makes it a proper shock wave, rather than just sound.

(Explosion at 7:30)

That's an amazing view. Terrifying. CGI people will be basing Hollywood explosions on this for years to come.

I'd imagine this is approximately what Hiroshima and Nagasaki would have looked like, but the direction of the pressure wave would be different as they were both air bursts.

This is absolutely nothing like a nuclear detonation. The energy released was less than 10% of Hiroshima, but besides that, the detonation velocity of ammonium nitrate can’t even hold a candle to the near instantaneous energy release of a nuclear weapon. It’s the difference between a pile of gunpowder and a block of TNT, and even that comparison doesn’t do it justice.

Thanks for writing this.

There’s been some strange tendency of late to conflate the weapons effects of nuclear or thermonuclear explosions with those of “really big amounts of conventional high explosive”. While that may even hold true at a sufficiently great radius, the proximate effects due to the near-instantaneous radiated energy is something that has no comparable analogue in the conventional world.

These effects are not really militarily useful, which is why even in the realm of small tactical nuclear weapons there has been a tendency towards minimising or not emphasising them, but...

The power of a nuclear weapon, intended properly as energy released per unit time, is in a league of it’s own.

Aside: for all the horror, at least this happened at the quayside and half the explosive force radiated away almost harmlessly over the sea. That basically halved the actual destruction at the outset. If it had been in the middle of the city, there would have been twice the damage.

EDIT: A startlingly worse case would have been if the explosives had been slightly interred, coupling the blast more closely to the ground and enhancing amplitude of the p-waves and providing plenty of really substantial debris raining down after basically ‘earth-quaking’ the buildings.

NOTE: Took the last paragraphs and placed them into their own comment, with logical conclusion. Apologies for the duplication.

> There’s been some strange tendency of late to conflate the weapons effects of nuclear or thermonuclear explosions with those of “really big amounts of conventional high explosive”.

Haven't people been using kilotons and megatons of TNT equivalent as a measure of nuclear bomb energy since the 1950s?

Yes, but that doesn’t mean that X tonnes of a random substance blowing up will be as destructive as the instantaneous release of equivalent energy that a nuke delivers

Yes, for the overall energy. But a nuclear bomb releases the energy faster than TNT, which makes a huge difference.

You wouldn't have seen them...if you were looking in the direction you'd be blinded instantly.

But really this isn't especially similar.

There are plenty of videos of actual nuclear explosions.. e.g. https://www.youtube.com/watch?v=mP3bcPvgIG8

> You wouldn't have seen them...if you were looking in the direction you'd be blinded instantly.

Feynman disagrees: https://www.realclearscience.com/blog/2016/04/what_its_like_...

I don't think his statement "bright light can never hurt your eyes" is correct, and I think "I got behind a truck windshield, because the ultraviolet can't go through glass, so that would be safe" is pretty questionable: glass filters ultraviolet, but I don't think it blocks it completely.

But fact is: Feynman looked at an atomic explosion and he wasn't blinded.

Nukes are different class of beast, being heavy metals nuclear materials don’t generate huge amount of gas mass that deliver destructive force in case with chemical explosives

Atomic/nuclear survivors account often starts with a blinding flash then move to a mushroom cloud and scorched world, also consider that the wreck of the building directly under Hiroshima nuke still stands to this day

I'm not a physicist but as I understand it, the shockwave of a nuclear blast can account for 50% or more of a nuclear blast's energy. This shockwave, as well as the fireball and thermal emission are all linked insofar as they're all caused by a massive amount of X-rays vaporizing the bomb and rapidly superheating the atmosphere near the bomb. These X-rays predominantly come from the energetic collisions of fission products and the fission itself.

AFAIK that building survived not because nuclear blasts produce weak shockwaves, but rather because it was a very strong earthquake resistant building in the first place, and because it handled being loaded from above well (concrete and brick generally handle compression well.) If it had been a bit further to the side, I think it might not have remained standing.

This blast was reminiscent of what happened in Halifax long ago:


To this date Halifax has a medical specialty: eye damage. People were looking at the fire on board the boat and ended up with plate glass shards in their eyes.

Thanks for posting this - I wasn't previously aware of the delineation between a shock wave and a pressure wave that the author explained clearly.

I'm curious, is there something specific about "high explosives" vs "low explosives", beyond just the magnitude of energy released, that explains why some generate shock waves and others pressure waves?

In a nutshell, high explosives detonate, low explosives combust. The two processes are distinctly different in how they operate at the molecular level. Basically, detonation is when a shock wave goes through the material, causing it to "blow up" on a molecular level as it passes (the molecules split as the shock passes through). Combustion happens much slower and is usually a chemical reaction involving some sort of fuel and oxidizer that is mixed and burned. I'm not an expert, but you can see the difference in a gasoline engine. As you crank up the compression ratio the air/fuel mix in the cylinder will eventually undergo a change from combusting to detonating, producing engine "knock" and rapidly destroying your engine if you let it continue. On a gas motor, around 13:1 is a good spot to see this happen. Someone will have more for you, but I think this is the basic concept seperating the two.

So gasoline mixture is both a high and low explosive? Your example seems to contradict your definition. This is also what seems confusing in the article too. Ammonium nitrate might be a low explosive but in this case it seemed to detonate. At that point is there any qualitative difference in the reaction between low and high explosives?

I am not an expert and do not understand it fully, but I know that fuel/air mix does make a phase transition of sorts and becomes a high explosive when the compression ratio is high enough. At low ratios it burns, at high enough ones it really does detonate with a shock wave running through it and causing the whole charge to blow up almost at once. Ammonium nitrate does the same if there is a big enough kick to get a strong detonation wave going through it, though I think there is much greater chance of this if there is some sort of solvent (gas, diesel, etc) mixed in with it.

> I think there is much greater chance of this if there is some sort of solvent (gas, diesel, etc) mixed in with it.

The fuel is what burns turning into combustion byproducts. Larger volume -> high pressure -> physics magic I don't feel like explaining -> shock wave. Burning pure AN is a waste, you're basically burning super lean. Add fuel and you get closer to stoic. More efficient reaction -> more output volume per input volume -> more bang.

It's like the difference between making a bunch of gas fuel (like propane or something) go up as a woosh vs a bang but with solid inputs instead of gasses.

I'm also no expert, but I thought that "detonation" in a gas engine just meant that there was too much fuel being injected, when the engine is at too low revs, so the piston can't move quickly enough to absorb the expansion and convert it to spinning the crankshaft - very similar to if the timing of the spark was too early.

Nothing to do with a transition to high explosive, although it will destroy an engine just the same.

Am I right in my understanding then that diesel engines work entirely on the principle of detonation, and that's why they're heavier as everything is designed to cope with what would make a petrol engine ping?

Does that make any sense?

Or is diesel injected in to the cylinder in a way that more or less produces a controlled combustion?

Diesel engines have a very high compression ratio, like 20:1 or more, but diesel fuel has a very high octane rating in comparison with gasoline, so it still combusts, not detonates, at these high ratios. What makes a diesel different than a gas engine is that the heat generated from the compression of the air/fuel mix is what actually causes the mix to start burning in the cylinder. So, for this reason you don't see any ignition system on a diesel usually, it doesn't need it (though maybe glow plugs or a block heater in cold areas, but not an actual ignition system). But the mixture still combusts in a diesel. Detonation is incredibly destructive to any engine, even a big beefy diesel. It can destroy any engine very quickly.

Diesel has a low octane number. Though octane isn't used for diesel fuel, it can be determined and is around 15-25.

Diesel engines avoid pre-ignition knocking by compressing only air. There is no compression of air/fuel mix. The fuel is injected into the hot, compressed air at the top of the compression stroke (at something like 3000+ PSI, IIRC).

The fuel does not have to be particularly volatile because of the high compression and the injector's atomization.

In theory, diesel engines can run on "anything". Diesel engines have been run on crude oil taken directly from the ground (of good purity), and on vegetable oil.

You can do that with old diesel engines but new ones are too sensitive and destroy themselves if you try to run them on anything but diesel.

It mostly because the high pressure pump use the Diesel for lubrication so if you use something else the pump destroy itself sending tiny metal shards down the pipe and into the injectors.

That's why accidentally running a diesel car on gas turn expensive fast. The contrary is much cheaper as just have to flush the fuel lines and you're good to go.

Modern diesel injectors:

30,000 to 40,000 PSI


>but diesel fuel has a very high octane rating in comparison with gasoline,

"octane" is the ability of a fuel to "resist combustion" from compression. Diesel fuel is not rated in "octane" but the opposite: the ability for it to combust from compression. For some reason in the "old days", it was commonly thought by some that by adding some diesel to gasoline that it would help "lubricate" the motor (as diesel does have an oily feel). However mixing diesel with gasoline only lowers its octane rating.

Yeah, I am not sure if octane is the right word here, you are probably right in what you say. I just know that diesel fuel is much more resistant to detonation at high compression ratios, given how a diesel works. In a gas motor, octane gives a measure of this, so I used the word there, but as you say it is probably not the right word.

Thanks all for clearing that up.

Detonation in an engine means something different. Detonation is when the fuel starts burning from a source other than the spark plug and the two flame fronts collide.

No, Diesel engines are not based on detonation but also on a controlled combustion. The combustion is triggered differently that is all.

And if you're "inside" an explosion that's too slow to generate a shock wave, you may not even be seriously injured.[0] At least, if the pressure spike is too low to rupture your body. And as long as objects don't get accelerated toward you.

As a child, a friend and I were electrolyzing water, in an enclosed space. And so the hydrogen flashed. But the concentration was so low that the air just turned faint blue. There was no damage, to us or the building.

0) https://www.stuff.co.nz/national/114372936/how-did-anyone-su...

The technical term for a subsonic combustion wave is deflagration.[0] A normal internal combustion engine deflagrates. If it starts pinging, that is detonation, and damages the pistons, etc.

[0] https://en.wikipedia.org/wiki/Deflagration

I've been led to believe that 'high' explosives are explosives that explode/combust faster than the speed of sound. Others (like black powder) burn slower.

More than just burn, 'high' explosives detonate [0], where the pressure wave itself provides the activation energy required to sustain the reaction.

[0] https://en.wikipedia.org/wiki/Detonation

> In an explosion, however, the devilish little instigator that is oxygen shoves the process into overdrive.

It's a great article, and I know that I'm being pedantic, but chemical explosions need not involve oxygen. For example, consider acetylene and silver acetylide, with a carbon-carbon triple bond. Or lead azide, with a nitrogen-nitrogen triple bond.

> chemical explosions need not involve oxygen

They do need an oxidiser though. Which may or may not involve oxygen. You’re right of course, but redox itself has some poorly chosen jargon.

> They do need an oxidiser though.

When lead azide detonates, you get elemental lead and diatomic nitrogen. So yes, the Pb++ oxidizes the azide.

When acetylene detonates, you get a mixture of organic compounds. But I don't believe that there's any redox involved. You just get various hydrocarbons with single and double carbon-carbon bonds. And just to be clear, this is when there's no oxidizing agent present.

In chemistry, an oxidising agent (oxidant, oxidizer) is a substance that has the ability to oxidize other substances — in other words to accept their electrons. Common oxidizing agents are oxygen, hydrogen peroxide and the halogens.

The jargon being that an oxidiser is an substance that can accept electrons, and doesn't have to by oxygen.


Just take oxidizer to me an “something that accepts electrons, like oxygen would”. Of course, there’s some oxidizers that can accept electrons from things that oxygen can’t (think of fluorine, for example) but... it’s just a term.

There’s also an enormous continuum of timescales, all the way from the detonation of high explosives, to deflagration of low explosives, combustion of fuels, to... sedate rusting, which is just oxidation of iron or steel. YMMV.

I agree. I mean, there are loads of organic chemicals with plenty of oxygen in them, and they aren't explosive. What makes ammonium nitrate explosive isn't the oxygen in it, it's the nitrogen in it. The manufacturing process for ammonium nitrate involves using quite considerable amounts of energy to persuade nitrogen to get itself hitched to hydrogen and oxygen, but it would much rather be free as diatomic nitrogen, and when that happens you get that energy back.

As I understand it, decomposition of ammonium nitrate (NH4NO3) alone produces a mixture of ammonia (NH3), nitric acid (HNO3), nitrogen dioxide (NO2), nitrous oxide (N2O), and water (H2O). But if there's a reducing agent around, such as charcoal, coal or fuel oil, you'll get lots of diatomic nitrogen.

Isn't the white cloud a de-pressurization front? The actual pressure wave is ahead of it and his speed estimate is probably low.

Correct. Adiabatic cooling due to depressurization causes temporary condensation. Though arguably it's more of a "back" than a "front", since it's in the wake of the pressurization front.

Makes sense.

And my understanding of the "condensation wave" is that it's formation and dissipation is highly dependent on local conditions like humidity, temperate, etc.

She claims that measuring the speed of that "condensation" wave proves it's not a shockwave, the actual shockwave can move faster than the "condensation wave" (because it's formation and dissipation is highly variable).

Well it's basically if you normally just raise pressure and lower it then it will propagate at the speed of sound (more or less - the extreme temperature shifts in the wave will cause some variation in the actual speed of sound).

High explosives cause air to be blown apart such that you're no longer just talking about a pressure wave, but talking about air flying with momentum faster than the speed of sound in a shockwave (at least until the shockwave loses energy and becomes a pressure wave).

I would expect the condensation wave (if one exists, which as you said is more likely during conditions of high relative humidity) to travel in sync with either a pressure wave or a shockwave though.

Does that make it like the vapor cone that appear around aircraft and launch vehicles sometimes?


Yup, same cause (substantial temperature drop due to adiabatic expansion).

Yes,you can see it quite clearly in the frame-by-frame in this video at between 2:15 and 2:25.

The cloud wave is clearly expanding but it is behind the pressure wave that blows out windows on both the front and back (relative to the wave) of the highrise buildings.

Spectacular and horrifying physics.


I'm starting to wondering how accurate this article is...

"High explosives produce shock waves; low explosives, like ammonium nitrate, produce pressure waves, which have a bit of slope to their shape, a period of time over which the pressure increases more gradually."

But a shock wave is an area of high pressure, so it's a "pressure wave" too.

And at least according to this Los Alamos article [1], "low explosives" versus "high explosives" is basically the difference between deflagration (rapid burning, e.g. black powder) versus detonation (chemical conversion happens faster than the speed of sound through the material, e.g. TNT).

By that definition, ammonium nitrate, is a high explosive. That's why they use it as the main blasting agent in mining operations.


as the article explains the difference between a pressure wave and shock wave is in their speed of propagation,

pressure waves travel below speed of sound, shock waves travel faster than the speed of sound.

How can a shock wave travel faster than the speed of sound? I thought the maximum speed a disturbance could travel in a material was exactly the speed of sound in that material, since that's what sound is.

The shock wave travels faster than the speed of sound in the medium at rest, but not faster than the speed of sound in the moving medium around the wave. Longer and better explanation:


I see, thank you. Both your explanation and the link were illuminating.

Sound is a more or less coherent propagating oscillation in the medium. A shock wave consists of a very thin front of high entropy propagating through the medium, which is not coherent and essentially disrupts the structure of the medium.

If I use a machine to mechanically fling air at, say, 500 m/s, then it will move faster than the speed of sound relative to ambient air, until it gets slowed down and turns into a normal pressure wave.

interestingly even light (in a medium) can travel faster than the speed of light in that medium

now both speeds will be slower than the speed of light in vaccum (as that is the highest speed that can be attained)

as the light "slows" down from faster speed to its normal speed radiation is emitted


OK, that was my own ignorance of the fact that "pressure wave" is standard nomenclature.

But her comment that ammonium nitrate is a "low explosive" doesn't line up with what I've read.

Low and high explosives are a very specific thing relating to how they behave in a compressed spaces.

If you've ever done model rocketry you're restricted to low explosives for both motors and ejection charges, at least in the US.

Is there any source for these definitions? I tried googling low and high explosives on reading this article but I basically couldn’t find anything. One of the only results was this article.


"Explosive materials may be categorized by the speed at which they expand. Materials that detonate (the front of the chemical reaction moves faster through the material than the speed of sound) are said to be "high explosives" and materials that deflagrate are said to be "low explosives"."

Note that the relevant speed of sound here is the speed of sound in the material, not the speed of sound in air at STP. Once the reaction rate goes supersonic, the pressure rise becomes very sharp, like a shock wave in front of a supersonic jet. The very sharp pressure rise means that there's very little force pushing/scattering the explosive before there's enough pressure to crush the explosive sufficiently to cause it to react, so momentum is sufficient to contain the explosive long enough to get most of it to react, without the need for any container to hold it in place.

Without detonation, burning ammonium nitrate would have had a hard time building up pressure higher than the bursting pressure of the windows on the warehouse. Sure, the bursting pressure of the widows of the warehouse is high enough to break nearby windows and do damage, but orders of magnitude less damage than seen in this case. In the case of a deflagration, the generated overpressure is usually limited by the strength of the container (in this case, the warehouse). In a detonation, the reaction rate is such that the solid/liquid becomes a very hot gas essentially in-place (most secondary high explosives have very few solid reaction products).

There's basically a hierarchy of tendency to transition from normal burning (deflagration) to detonation (supersonic flame front): primary high explosives, secondary high explosives, insensitive high explosives, low explosives. Low explosives are difficult or impossible to get to detonate. Primary high explosives are typically the most dangerous, because they easily detonate once they start burning. (Also, the sort of instability for easy deflagration-to-detonation transition also correlates with instability when struck/dropped/crushed. The most common commercial primary explosives are lead azid and mercury fulminate, which obviously also throw around toxic heavy metal compounds.)

Insensitive high explosives are used in things like weapons that are designed to travel through brick walls or steel armor and explode on the other side. In such a case, (in the pre-electronic age), you'd have something like a momentum-based striker that strikes a percussion cap when the weapon makes impact, igniting a low explosive delay element (timing train) made out of something like solid pressed black powder, which would burn for a predictable period of time before igniting a small amount of primary explosive (e.g. lead azide). The primary explosive would start out deflagrating, but it easily undergoes deflagration-to-detonation transition, and detonates. The detonation shock wave from the primary high explosive sets of a small amount of secondary high explosive (say, PETN), which gives enough kick to set of the main charge of insensitive high explosive (such as ammonium picrate).

The compression of nuclear weapon fission primaries is usually done via insensitive high explosives that are also resistant to radiation degradation. Even though accidental detonation in a crash/drop would unlikely to cause a nuclear explosion, spreading toxic radioactive plutonium everywhere is still Very Bad (TM). I'm not a chemist, but I would guess that chemical stability against radiation degradation also correlates with pressure/temperature stability of insensitive high explosives.

During the Vietnam war, supposedly soldiers would sometimes burn C4 plastic explosives to heat food. C4 is basically RDX mixed with just enough polymer to make it into a pliable putty. RDX is very unlikely to undergo DTD transition, so this was relatively safe-ish, and RDX is chemically similar to hexamine solid fuel tablets used in some camping stoves. Dynamite is also a secondary high explosive, but nitroglycerine-containing dynamites are significantly less safe to burn than RDX.

Without detonation, burning ammonium nitrate would have had a hard time building up pressure higher than the bursting pressure of the windows on the warehouse.

That was my thought as well. I was skeptical of the article saying it wasn't a detonation of ammonium nitrate and that a shockwave wasn't generated.

Yea, I hope the author was just saying that as it expaded, the shock wave slowed to a subsonic pressure wave with a resulting less sharp pressure rise, and the majority of the blast damage was caused by a subsonic pressure wave.

According to the link above, it's the difference between deflagration and detonation.

Yes, pure AN is a secondary high explosive with a detonation velocity of 2,700 m//s. The author may be using some shorthand for the detonation shock wave slowing below the speed of sound (and it's pressure front becoming less sharp) as it expands.

Wikipedia defines 'high explosive' as having a detonation velocity of 3-9 km/s, so by that standard pure AN is not a high explosive, but ANFO is.

I'm not sure about AN, but ANFO is a tertiary explosive. But primary/secondary/tertiary is a matter of sensitivity, not detonation velocity.

This may be a case of "to explain research to media, researcher dumbs down findings until they no longer reflect outcome of study"

I found this to be a well researched and very relatable piece of journalism with respect to the explosion in Beirut.

Seconds before the final explosion, you can see some flashes in the base of the smoke, like firecrackers going off. Anyone knows what these are?

> it looks like firecrackers going off

It’s literally firecrackers going off. It’s one of those cases where if it quacks like a duck, waddles like a duck, and swims like a duck, it’s not a camel disguised as a duck.

I think it was reported of fireworks explosion in the vicinity that lead to this event.

The whole "let's store fireworks next to several tons of explosives -- oh wait there's a fire" thing would sound incredibly contrived if it was in a story. It's almost like something out of a Road Runner cartoon, not real life.

The past year in Lebanon has been like that. I'm sure if someone wrote a script about this for a movie they'd be told to just "tone it down on the grimdark shit okay?" or something.

1 - Oct 2019: financial meltdown (still ongoing, local currency has lost 80% of value vs USD since then with no end in sight. Banks aren't giving out USD deposits either, you get to take them out in local currency and lose the difference in exchange rate)

2 - Protests, riots, etc... (since the meltdown and still ongoing)

3 - Pandemic, lockdown (Mar 2020, still ongoing with rising cases/day since measures were relaxed)

4 - Huge explosion takes out half the capital and what remains of the economy (The Beirut port runs about ~70% of all the sea trade volume of Lebanon).

The whole thing is literally stranger than fiction. Someone described the latest disaster like so: "The most shocking thing about the Beirut is explosion is how none of negligence and incompetence that led there is shocking".

Supposedly someone may have been carrying out welding on the building too, which makes the whole thing even more unbelievable.

According to this story, the welding was specifically as a result of the ammonium nitrate:

> In January 2020, a judge launched an official investigation after it was discovered that Hangar 12 was unguarded, had a hole in its southern wall and one of its doors dislodged, meaning the hazardous material was at risk of being stolen.

> In his final report following the investigation, Prosecutor General Oweidat “gave orders immediately” to ensure hangar doors and holes were repaired and security provided, a second high-ranking security official who also requested anonymity said.

> ...During the work, sparks from welding took hold and fire started to spread, the official said.

It's really mind-boggling.


It's a port. Everything is stored there.

Most likely volatile nuggets of the fertilizer that were finally able to ignite.

The contents of the warehouse were not fertilizer (it was AN pellets intended for use as an oxidizer in industrial explosives). "Nuggets" are not a form factor in which AN is made, transported or used.

From a scientific point of view it was unfortunate that no Rapatronic camera was installed in Beirut:



Contrary to the article, I think amonium nitrate is a _high_ explosive. That means it can explode via a supersonic shock or detonation wave. https://en.wikipedia.org/wiki/Table_of_explosive_detonation_... says it's detonation velocity is 2.7 km/sec.

Actually it is often not even considered an explosive, it is an oxydizer, the type actually used in mining is called ANFO (Ammonium Nitrate and Fuel Oil) and the oil (mormal diesel fuel) is added on-site:


and has a much faster speed, around 4,200 m/s and only used in mining because it is extremely cheap and relatively more secure when compared to other (more powerful) explosives.

Separate from contemplating the physics, and hopefully not macabre rubbernecking of heartbreaking images of human suffering... what is being done for humanitarian aid for Beirut, and how can HN help?

Loads of countries have sent search & rescue teams (the Dutch one returned after two days already because there was plenty of help on-site), hundreds of millions in relief funds have been gathered by countries across the world (https://www.al-monitor.com/pulse/originals/2020/08/lebanon-a...), charities all over are collecting, celebrities, companies and nations are donating tons (https://www.popsugar.co.uk/celebrity/george-and-amal-clooney...), and local charities and donation numbers are being set up worldwide.

HN can help by getting their super wealthy excessively profitable FAANG & co employers to donate money. Apple alone has $193 billion in cash on hand, more than twice as much as Lebanon's total GDP (https://www.cnbc.com/2020/04/30/apple-q2-2020-cash-hoard-her... they could rebuild Beirut three times over and still have money left. Amazon has in the range of $70 billion cash on hand.

Don't feel like you have to donate when it's corporations that hold (and hoard) all the wealth. Call them out on it if they offer a pittance.

Apple / Tim Cook has indicated they will be donating, but no mention of how much: https://twitter.com/tim_cook/status/1291746412779069440 Couldn't find anything from the other big companies.

from what i've heard (informally) google is matching donations.

edit: dug up the fb post a friend had about it, it appears to be an internal programme

Google is matching donations as usual, but they sent an extra email out specifically to ask for donations for helping Beirut.

Probably by supporting the Red Cross, Medicins sans Frontiers, etc. Where I live, you can donate to all collaborating charities with a simple bank transaction. They state that €25 provides clean water for 10 families, €50 gets mattresses and blankets for two families, etc.

How hard is it to not put nearly 3,000 tons of explosives in a city?

Well, not putting explosives in a city is not hard; not putting substances that have some industry uses and happen to be explosive (eg fertilizer) near industry is hard.

After all, it needs to be shipped, stored and used.

If there's shipping, there will be some port city not far.

And if there is industrial use, there will be workers, and some city as well to house the workers.


- Halifax explosion (how hard is it to not make explosive-carrying ships go to ports?)

- AZF explosion (how hard is it to not locate factories far away from cities?)

- Texas City Refinery explosion (how hard is it to not let refinery workers inside the refinery?)

- Tianjin explosions (how hard is it to not locate a port city near its port?)

And so forth

When the government doesn't function, such a thing can be extraordinarily difficult.


Why was so much NH4 stored in one place? Did a shipment of that size just arrive or has it been building up a stockpile? It is end of summer here in America, not sure what season is approaching outdoor Lebanese fertilizer schedules. Abundance of summertime high energy (summer solstice increased photovoltaic potential) could be the reason so much was purchased cheaply (infinite shelf life?)

The road to last week’s tragedy began seven years ago, when the Rhosus, a Russian-chartered, Moldovan-flagged vessel carrying ammonium nitrate from Georgia to Mozambique, docked in Beirut to try to take on extra cargo to cover the fees for passage through the Suez Canal, according to the ship’s captain.

Port authorities impounded the Rhosus in December 2013 by judicial order 2013/1031 due to outstanding debts owed to two companies that filed claims in Beirut courts, the state security report showed.

In May 2014, the ship was deemed unseaworthy and its cargo was unloaded in October 2014 and warehoused in what was known as Hangar 12.

In February 2015, Nadim Zwain, a judge from the Summary Affairs Court, which deals with urgent issues, appointed an expert to inspect the cargo, according to the security report.

The report said the expert concluded that the material was hazardous and, through the port authorities, requested it be transferred to the army. Reuters could not independently confirm the expert’s account.

Lebanese army command rejected the request and recommended the chemicals be transferred or sold to the privately owned Lebanese Explosives Company, the state security report said.

The report did not say why the army had refused to accept the cargo. A security official told Reuters it was because they didn’t need it. The army declined to comment.

The explosives company’s management told Reuters it had not been interested in purchasing confiscated material and the firm had its own suppliers and government import licences.

From then on, customs and security officials wrote to judges roughly every six months asking for the removal of the material, according to the requests seen by Reuters.


It's not clear to me why, if they knew it was dangerous and had been asking for it to be removed for years, they decided to store fireworks in the same building.

Probably because "they" is a group of people/teams and knowledge is not evenly distributed, nor is responsibility concentrated in the right places.

This is a classic example of "normal failure" and I am sure will join other similar incidents in the literature.

Storing the ammonium nitrate in a single location was bad enough.

Storing fireworks in the same warehouse complex was - I honestly can't think of a word that fits.

> Storing fireworks in the same warehouse complex was - I honestly can't think of a word that fits.

And on top of that, according to the article, the fire was started by welders who were trying to repair doors to make the ammonium nitrate secure. It's like the plot of some dumb bloke movie: so absurd it would be comic if it weren't so terrible.

It's pretty easy - the fireworks were probably a pretty small shipment, and were loaded along with a bunch of other shipments into an anjoining warehouse.

I realise this seems weird to people living in a country which has all sorts of extremely restrictive and impressive safety protocols, but in poorer countries, this kind of mistake happens __all the time__.

It just doesn't normally result in much more than a small fire.

The (more local) news I'm reading, it was apparently known to be dangerous immediately, and regularly reported (~twice a year), just never actioned to remove it. To me that's where the criminal responsibility lies. The beating-around-the-bush has begun though.

Port authorities (who were the ones reporting the danger) were arrested.

skimmed, since I'm on the clock, but holds the key info: https://www.mtv.com.lb/en/News/Local/1088201/Lebanons-leader...

From what I've read, the supply had been in this warehouse for quite a while, having been seized from a Russian freighter that owed port fees or something. Certain people kept prodding the powers that be to take care of the situation before something disastrous happened, and they were not heeded.

It had been sitting in that warehouse since 2014, seized from a cargo ship.

I wonder if "sitting" because infinite loop bureaucracy ("seize" means potentially murky waters of international maritime law). From an economics standpoint - inefficient to leave energy dense "fuel" idle without being put to use for such a long period.

I didn’t read all the way through but quickly got the impression that the author got the physics wrong. Combustion and detonation are very different processes. It’s actually not easy to make ammonium nitrate detonate on purpose, and just heating it is certainly not a reliable way of making it do so.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact