Steel burns on Rimworld?

[brcruchairman]

I think the main debate here is not about whether or not steel can burn on extreme circumstances, but rather what the circumstances are in which steel could burn. Steel does burn at high temperatures, to be sure; it's where all the flakes of slag and the like come from in metal processing. The oxidization of steel, even at high temperatures, is exothermic.¹ So okay, steel can theoretically burn. The question, to me, becomes: is it easy to burn?

The answer in the article cited above implies, reasonably, that it is not. For something like an open flame in an open environment, with a large hunk of steel at the edge, it's reasonable to expect not, as the rest of the steel will conduct the heat away from the point of contact. So a single spark lighting an entire steel wall on fire seems ridiculous, as I'm sure all here would agree.

What, though, if the steel wall is surrounded by fire? Soil is a reasonable insulator, so the ground likely couldn't siphon enough heat away through conduction to keep all energy out. So if a steel wall is surrounded by perpetual fire, sure, that could work. Mercifully, most fires in Rimworld burn themselves out. (If they didn't, I'd probably lose every colony I ever made to bad wiring and flash storms. ;_; ) So now we're looking at the following: could surrounding fire produce enough heat to ignite the steel wall before depleting its own fuel source?

Steel has a reasonable thermal capacity of about a half joule per gram-degree Celsius.² Let's assume, for the sake of example, a 1 m thick by 2 m tall by 10 m long steel wall surrounded on all sides by fire. Steel also has a density of about 8 g/cm^3. This can tell us the total energy required to raise the temperature of the wall to 600 *C, roughly the temperature of a match flame.

The volume of the wall is 1 m * 2 m * 10 m = 20 m^3. We want that in cm^3, so we multiple by (100 cm)^3 / (1 m)^3 = 20 000 000 or 2x10^7 cm^3 . Okay, cool. We know that would weigh 8 gm / cm^3, so we're looking at 1.6x10^8 g of steel. Let's say it's a nice 20 *C outside, so we need to go up 580 *C. With the specific heat capacity of 0.5 J/g *C, we're looking at 1.6x10^8 g * 580 *C * 0.5 J/g *C = 5.22x10^10 J, or 5 220 mega joules.

Okay, that's a lot of energy. Cool. How much wood would it take to yield that much energy? (We're going to ignore, for now, energy lost to the surroundings instead of transferred directly to the wall, though such losses would be significant.) An example wood, red oak, has combustible energy around 10.4 MJ/kg under less-than-ideal circumstances.³ That's a good start. Okay, so without heat loss to the environment, we're looking at a bit shy of 502 kg of red oak. Per the OneOak Project, oaks can weigh up to 14 000 kg.⁴

On the surface, it seems like an easy assumption to make that a burning oak next to a 1x2x10 m steel wall could easily set it ablaze. But here's the catch; how does that energy get from the oak to the wall? Air is a very good insulator, so we can reasonably ignore conduction. That leaves us with two methods; radiation and convection. Let's begin with radiation.

For radiation, heat is radiated out in all directions. That's neat, but we're only interested in heat heading towards the steel wall. Heat going away from the wall, up, or down into the ground are all ignored for this purpose; what we're left with is what I'll approximate to about an eighth. In short, for an arbitrary number of radiation vectors, I'll assert that only an eighth of them hit our wall. (To visualize, this is about a 90* arc wide and tall.) Unfortunately, here is where I hit the limit of my knowledge; I'm not a physicist by training, and though I could approximate and say, "About an eight of the total burnt energy gets to the wall" I could very easily be wrong. If anyone else has better knowledge of thermodynamics, I'd very much appreciate your input.

In the absence of an expert opinion, I'm going to take up the tried and true traditions of the internet and speculate wildly. I'm going to speculate that, given a bunch of burning trees surrounding a steel wall, they would heat the wall quickly enough and thoroughly enough to cause it to burn. This is only an opinion; I'm still hoping a physicist will come in to save the day with legitimate calculations, I stand stand eternally ready to admit I was totally wrong.

Anyway, that's enough of a wall of text; I hope someone found it interesting. I know I sure did; my first thought when coming into this was, "Phah, steel can't burn! Even slagging it takes lots of energy! Now way could steel have a self-sustained exothermic reaction!" I sure showed me.

EDIT: Also, Mossy Piglet, I think it was more about how steel rusts far more quickly in the presence of water than it does in only the presence of air.⁵ While steel should eventually rust unless it is, as you suggest, in an oxygen-free environment, the presence of water does seem to dramatically increase the rate at which rust accumulates.

¹: http://www.popsci.com/diy/article/2007-10/burning-metal
²: http://asm.matweb.com/search/SpecificMaterial.asp?bassnum=MQ304A
³: https://en.wikipedia.org/wiki/Wood_fuel#Energy_content
⁴: https://sylva.org.uk/oneoak/tree_facts.php
⁵: https://van.physics.illinois.edu/qa/listing.php?id=485

[newcadence]

mabe we should justg build out of bricaks

[Mutineer]

OK, for people who did not do chemistry..

Demonstrations - Combustion in Pure Oxygen - Burning Iron

https://www.google.co.nz/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&cad=rja&uact=8&ved=0ahUKEwjxl9y8hJzOAhUILpQKHUjGBrYQFgg2MAQ&url=https%3A%2F%2Fwww.angelo.edu%2Ffaculty%2Fkboudrea%2Fdemos%2Fburning_iron%2Fburning_iron.htm&usg=AFQjCNEEiliuTn-NWQJywbuPuDaowA7bRg&sig2=6pZV9tB48HidpXZ1uJzTbw&bvm=bv.128617741,d.dGo


We do not know concentration of oxygen on every planet., even terraformed. My chemistry teacher demonstrated burning of steel in oxigen in our class. It was 40 years ago when we still had education in schools...

[Mutineer]

On additional note, many materials you think as non burning can easy burn in very unseal conditions. Aluminium for example. The only reason it is not burning because it constantly create layer of oxide, which is extremely strong and has very high melting point. But if you prevent creation of this layer, it will self ignite in water. Adding a little mercury to aluminium can quickly burn it (do not do it in hope, mercury is extremely poisoned)

[brcruchairman]

Again, Mutineer, I don't think anyone is arguing that steel cannot possibly burn. I believe the argument is more that the conditions in which steel burns, particularly in terms of both total heat involved and the rate of heat transfer, are unlikely to occur in an uncontrolled environment. I happen to disagree, as per my previous post, but it's important to not misrepresent an opponent's argument; if we want to find some truth, and educate ourselves and/or others, we've got to understand what they're trying to say in order to decide if it has merit.

Also, I'll note that in a pure oxygen environment, sugar is a bomb waiting to go off, but I don't recall noting explosions from my food stockpiles. (True, there is no sugar there yet, but I hope you take my point that a high-oxygen environment is unlikely.) We can also gather that Rimworld is likely not a high-oxygen planet based on the lack of oxygen toxicity in the colonists. OSHA advises against a great than 21% oxygen atmospheres in a workplace.[sip]1[/sup] However, I don't like that number, as it seems low. Instead I'm going to go with 28%, which has been suggested to cause some pretty significant damage if used long-term.²  Given that we do not note hyperoxia in our pawns, I'm going to go ahead and say that we're not likely to have more than 28% oxygen on any Rimworld planets.

So yeah, Mutineer, your points are accurate, but I'm note sure they're relevant, given that we're looking at open flames in an open field interacting with unknown but presumably less than 10% carbon steel likewise in an open environment. Stuff like steel and aluminum CAN burn, yes, but the question we're asking is "is it LIKELY to burn?"

¹: https://www.osha.gov/SLTC/etools/shipyard/shiprepair/confinedspace/oxygendeficient.html
²:  New, A (1 February 2006). "Oxygen: kill or cure? Prehospital hyperoxia in the COPD patient". Emergency Medicine Journal 23 (2): 144–146.

[Havan_IronOak]

mabe we should just build out of bricks

Personally I'd prefer if you build out of straw or at worst sticks.

-- B.B. WOLF--

[Shurp]

So if we're going to assume that the Rimworld atmosphere is 90% oxygen meaning that plate steel will catch fire if a tree next to it is burning, how much more flammable will iron filings become?

Because I think it's time for steel fuel-air bombs.  Spray iron dust at your opponents and light it up!!!

[brcruchairman]

See, Shurp, that's the kind of creative and Dorf-like thinking that I've come to know and love from the Rimworld community. "This feature seems odd, and presents a huge danger. How can we weaponize it?"

[Goo Poni]

Besides, if steel didn't burn, wood and plasteel would be the only flammables. I think. Stone and uranium (??) don't burn. Not sure about silver, gold, or jade.

Is there a problem with that? Why is PlaSTEEL flammable? "Advanced spacer tech structural material." So something that you build a starship out of, but you're completely fucked if a fire starts because it won't be long before your ship turns into a fireball. Noice. Game design reason of "things should burn because burning" aside.

[Mutineer]

Not only sugar, grain for example. and not in pure oxygen, in usual condition.

<<Time is of the essence in salvaging wet feed and grain. Both will begin to heat and mold very quickly, leading to spoilage as well as the possibility of spontaneous combustion. >>

https://www.ag.ndsu.edu/flood/farm-ranch/salvaging-stored-wet-feed-and-grain

[BlackSmokeDMax]

What is the end game to this thread exactly? (to the OP)

1. Trying to get Tynan to make steel non-flammable in game?
2. Get him to change the name of steel to "flammable steel-like substance"?
3. Just have fun speculating about things?

[Goo Poni]

I would prefer the first of those three options.

[Nonmomentus Brain]

[work by the quoted author; too long to be quoted here]

While your post is an interesting read and certainly well-thought-out, there is a major problem with it.


You assume that at 600 °C, steel burns. This, however, is based solely on the temperature of a burning match, which might be an appropriate figure for a wood fire, but tells us nothing about steel. I have found a source indicating that the temperature of ignition of iron is somewhat higher, at 930 °C in an atmosphere of pure oxygen.¹ At a lower partial pressure of oxygen, the temperature of ignition is likely to be higher.

Under the assumption that the temperature of a match is equal to that of burning oak (according to your findings, 600 °C), it becomes clear that it is impossible to heat steel to the required temperature simply by burning wood, since the second law of thermodynamics prevents the flow of heat from a colder body to a hotter one on its own. However, the properties of oak trees and matches may be sufficiently dissimilar to allow for sufficient temperatures under certain conditions.


As an additional factor to be considered, in the oxidation of steel, iron oxide forms. Unlike the combustion products in more usual fires (mainly CO₂ and H₂O for wood), iron oxide is not gaseous at these temperatures. Therefore, it remains at the surface and blocks additional oxygen from reaching the unoxidized steel unless it is removed somehow.

According to Wolfram|Alpha², the melting points of various iron oxides appear to be too high in order for them to flow away from the point of oxidation at the given temperature. Not only would this prevent further combustion, but as a layer of iron oxide would form while it was being heated (since the greater heat would accelerate the rusting reaction, according to the Arrhenius equation), combustion might not start at all. On the other hand, in another study³, it was found that if exposed to oxygen during the heating process, iron ignites at slightly above 1400 K (or about 1150 °C). This indicates that oxides form during the heating process and inhibit ignition, but that reaching the melting points of the oxides as given by Wolfram|Alpha is not necessary for ignition. It is possible that this is because the melting point is lower, which can be caused by impurities (for example, different oxides mixed with each other, as is likely to be the case here).


In conclusion, though the previous findings indicate that it is possible to heat steel to a temperature of 600 °C, this temperature is insufficient for the ignition of iron, even under optimal conditions. However, it might be possible to reach sufficiently high temperatures anyway. Further examination is warranted.

Again, I would like to stress that I do not mean to attack your work, but merely add to it. I hope that my post was as interesting to read as yours.

¹Grosse, A.V. and Conway, J.B., 1958. Combustion of metals in oxygen. Industrial & Engineering Chemistry, 50(4), pp.663-672.
²http://www.wolframalpha.com/input/?i=melting+points+of+Fe_2O_3,+Fe_3_O_4,+FeO
³Nguyen, K. and Branch, M.C., 1987. Ignition Temperature of Bulk 6061 Aluminum, 302 Stainless Steel and 1018 Carbon Steel in Oxygen. Combustion Science and Technology, 53(4-6), pp.277-288.

[Andy_Dandy]

Um, steel rusting is just the iron in steel reacting with oxygen in the air to make iron oxide (rust), so there is no reason steel would not rust inside.

Unless you mean to say there is no oxygen inside...

And gameplay reasons, if you've heard of those.

[Shurp]

Concerning "plasteel", I assume this is steel reinforced carbon fiber.  And everyone who has watched auto racing other than NASCAR knows that carbon fiber burns nicely once it gets going.  All that's left of the car when it's done is a puddle and the steel engine block.  (Note that the engine block doesn't burn, even though it's in the middle of the fire...)

And the goal of the thread?  I dunno, it just struck me as odd and funny.  Like brcruchairman said, why complain about something if we can find a way to weaponize it instead?