69

After the recent Notre-Dame de Paris fire, there has been a heavily re-posted tweet going around in response to an earlier claim that a golden cross did not melt or deform - due to an act of God.

Screenshot of a tweet and response

Kaylee Crain: "After all the aftermath and destruction of the Notre Dame fire, the alter and cross remained untouched. Please explain to me how you don’t believe in God after seeing this."

Dan Broadbent: "Because the melting point of gold is 1064°C and a wood fire burns at around 600°C"

The melting point of gold varies based on purity, and can thus be lower than 1064°C.

However; do wooden buildings, such as the Notre Dame Cathedral, burn at temperatures above 600°C?

I'm obviously not interested in any debate over whether this was an "Act of God" or other unprovable matters.

In terms of personal research, what I have found is that while wood itself will not burn much hotter than 600°C, once it turns to charcoal - it can then reach over 1100°C. However I don't know enough about physics/chemistry or fires, to make a reasonable judgement on how that applies in a real-life fire.

  • 23
    The question about wood fire temps is valid, but the original Twitter statement with the pic is bogus. Look at many other pics of the damage (e.g. news.sky.com/story/…): the fire never reached that part of the church. Voting to close as not notable. – Jan Doggen Apr 17 at 15:01
  • 149
    Never mind the gold cross, why aren't any of those wax candles melted? – plasticinsect Apr 17 at 16:54
  • 94
    @plasticinsect you nailed it: the candles which are designed to burn are not burned, hence there was no fire here. That's consistent with the reports that only the roof burned. The damage below was from falling debris, not fire. – TemporalWolf Apr 17 at 17:47
  • 10
    The roof was burning. There was no significant fire in the nave. The vault collapsed in two places, allowing burning debris to fall through, but you can clearly see the debris did not fall on the cross and was quickly put out. People on the net vastly overestimate how much of the cathedral was actually in flames. The amount of fire below the vault was negligible compared to what was going on on the roof. Look at the candles, they aren't even melted. Lots of wooden structures are still intact as well. – Polygnome Apr 18 at 11:18
  • 20
    @R.Schmitz Yeah, "Wood fires can't melt gold" is a good analogy to "Jet fuel can't melt steel beams" -- both are wrong for the same primary reason of misunderstanding how fire works: in short, "The temperature a fuel can reach varies wildly with factors like container shape and airflow". And both have a secondary reason that invalidates the conclusion anyway: "the cross wasn't in the fire" / "metal doesn't have to 'melt' in order to soften and lose structural strength, see blacksmithing". – dgould Apr 18 at 18:44
81

Without acknowledging any of the conditions actually present in the church, wood fires can get much hotter than 600 °C.

The maximum temperatures measured within the pile were of the order of 800, 1000, and 1200 °C for piles composed of 1.27, 2.54, and 9.15 cm sticks respectively, although the maximum temperatures for a given size stick appeared, from all data obtained, to be somewhat dependent upon the structure of the pile. The prescribed temperature-time curve of a standard fire exposure test 1 is also shown in figure 4 from which a general agreement may be noted.
D Gross: "Experiments on the Burning of Cross Piles of Wood", Journal of Research of the National Bureau of Standards- C. Engineering and Instrumentation Vol. 66C, No.2, April-June 1962. (PDF)

A nice pile of wood with good ventilation can get apparently really hot:

Fire plume temperature data suggest a maximum turbulent flame temperature in fully developed compartment fires of about 1500 C for stoichiometric and adiabatic conditions. Experimental results for crib and pool fires are presented to support the trends indicated by the approximate analyses.

In general, from equations 12 and 13 for stoichiometric conditions, the temperature is given as

math formula

where Tf,ad is the stoichiometric adiabatic flame temperature. Recorded gas temperatures near the ceiling are reported as high as 1350 C [21], and mean temperatures over the peak burning period are 1000 to 1200 C for polyethylene fires [21] and approximately 900 to 1200 C for wood cribs [20]. For turbulent fire plumes, having a radiative loss fraction Xr, a similar formula applies to the combustion region. This turbulent flame (centerline) temperature is given as [18]

math formula

From the best available data [22–24], the turbulent mixing parameter, kT, is found to be approximately 0.5 for cp 1 kJ/kg K. As the fire diameter increases, the radiative fraction falls due to soot blockage [25]. Fig. 4 shows flame temperature data for turbulent plumes as a function of Xr. The extrapolated adiabatic temperature is approximately 1500 C. For a realistic adiabatic flame temperature of 2000 C, the actual turbulent mixing factor is approximately 0.75 or a turbulent dilution factor of 1.5. For a large fire in a compartment with large vents, the core maximum flame temperature should approach the turbulent adiabatic flame temperature.
James G. Quintiere: "Fire Behavior in Building Compartments", Proceedings of the Combustion Institute, Volume 29, 2002/pp. 181–193. DOI

But to be very sticklish, the claim is actually somewhat correct. Why?

Comparison between maximum theoretical and measured surface temperatures for redwood Graph of rate of heat release and heat of combustion for Douglas fir
M. J. Spearpoint And J. G. Quintiere: "Predicting the Burning of Wood Using an Integral Model", Combustion And Flame, 123:308–324 (2000). DOI

Or to put it simply:

A bonfire can reach temperatures as hot as 1,100 degrees Celsius (2,012 degrees Fahrenheit), which is hot enough to melt some metals.

Most types of wood will start combusting at about 300 degrees Celsius. The gases burn and increase the temperature of the wood to about 600 degrees Celsius (1,112 degrees Fahrenheit). When the wood has released all its gases, it leaves charcoal and ashes. Charcoal burns at temperatures exceeding 1,100 degrees Celsius (2,012 degrees Fahrenheit).
Gabriella Munoz: "How Hot Is a Bonfire?", Sciencing, April 26, 2018.

Wikipedia says

This is a rough guide to flame temperatures for various common substances (in 20 °C air at 1 atm. pressure):

Wood                    1,027 °C (1880.6 °F)
Methanol                1,200 °C (2192 °F)
Charcoal (forced draft) 1,390 °C (2534 °F)

and gives for adiabatic flame temperature maximum even:

Wood  Air 1980°C  3596°F

The French authorities and experts seem to have suggested that inside the church, around the altar, theoretically 800 °C might have been reached, with 'improper' firefighting methods, meaning that it stayed cooler in that area. Or that even temperatures of 2000—2500 °C occurred actually in the roof where the main part of fire burned:

Contrairement aux pompiers américains, les sapeurs-pompiers français s’attaquent aux incendies par l’intérieur et non de l’extérieur. Cette tactique est plus dangereuse pour les hommes mais plus efficace pour sauver le patrimoine, observe l’expert Serge Delhaye. Si l’on se concentre sur l’extérieur, on prend le risque de repousser les flammes et les gaz chauds, qui peuvent atteindre 800 degrés, vers l’intérieur et accroître les dégâts. »
"Six questions sur l’incendie de Notre-Dame de Paris", Le Parisien, Jean-Michel Décugis, Vincent Gautronneau et Jérémie Pham-Lê| 15 avril 2019, 23h40

Selon les premiers éléments de l’enquête, l’incendie aurait démarré dans les combles de la cathédrale. Le feu se serait propagé très vite sous l’effet du vent, dévorant l’une des plus anciennes charpentes de Paris constituée de centaines de poutres de chênes. La toiture de plomb de plusieurs centaines de tonnes qui reposait sur cette « forêt » est partie en fumée. La flèche haute de 93m qui surplombait la croisée du transept s’est effondrée en moins d’une heure. Sous la charpente, « les températures ont pu atteindre 2000 voire 2500°C, une température bien supérieure à la celle de la fusion du plomb », explique Guillaume Legros, enseignant-chercheur à l’Institut ∂’Alembert et ancien doctorant de José Torero, spécialiste mondial des grands incendies.
"L’incendie de Notre-Dame de Paris", Sorbonne Université, 19 April 2019.

Most sources seem to quote a temperature of 1000 °C for this incidence, but other sources even go up to 1400 °C:

Fires peak at 1,400°C, explains professor Guillermo Rein, the head of Imperial College London's fire-studying Hazelab.
Nicole Kobie: "The hot, dangerous physics of fighting the Notre Dame fire", Wire, Tuesday 16 April 2019

Interestingly, the place where the cross is located is far away from where most of the flames ravaged, beneath a stone vault that remained largely intact. But nevertheless the wood that did burn on and in the roof was well seasoned, old, and dry.

But even fresh wood that would otherwise make for quite a suboptimal fuel burns often hotter than 600 °C:

Q. At what temperatures do forest fires burn?
An average surface fire on the forest floor might have flames reaching 1 meter in height and can reach temperatures of 800°C (1,472° F) or more. Under extreme conditions a fire can give off 10,000 kilowatts or more per meter of fire front. This would mean flame heights of 50 meters or more and flame temperatures exceeding 1200°C (2,192° F).
(Natural History Museum of Utah: "Wildfires: Interesting Facts and F.A.Q." .PDF)

  • 3
    I was about to add the translation of the French bit, but I wonder if that's truly a relevant thing to do, considering it gets readable enough when you copy paste it into DeepL. – Clockwork Apr 18 at 6:00
  • 2
    Then again, what if a blind person stumbles on this answer... ? – Clockwork Apr 18 at 7:41
  • 1
    However, OP didn't ask about wood in general, but explcitly "wooden buildings, such as Notre Dame", where the wood is far from a perfect pile – Hobbamok Apr 18 at 11:22
  • 1
    @LangLangC I don't think I can do that. Technically, I can add some alt text that describes the formulas as e.g. "Uppercase T sub lowercase S lowercase T [...] lowercase H end sub minus uppercase T sub 0 end sub equals Δ H [...]", but that would be too cryptic for most people, let alone ones who do need screen-reading assistance. Any more reliable method (they do exist) would most probably be too much for a Stack Exchange post. – EKons Apr 18 at 19:45
  • 4
    Experimental data point - my in-laws house burned down, and their aluminum window frames were little pools of resolidified metal on the ground. Aluminum melts at >650C (yes, I know it is actually an alloy, but not intentionally a low-melting point eutectic alloy). – Jon Custer Apr 18 at 20:16
53

Probably yes: According to at least one expert, the temperature in the Notre Dame fire must have been extremely high, and probably exceeded 600°C.

Yesterday the Süddeutsche Zeitung, one of Germany's most reputable newspapers, published an interview with the director of the German Technisches Hilfswerk (the Federal Agency for Technical Relief) and former director of the Fire Departments of Berlin, Albrecht Broemme. The interview covers several aspects of the Notre Dame fire (for example, Broemme explains why using water bombers was out of the question). He also discusses why this was an extraordinarily difficult task for the fire fighters. One reason he mentions is the extreme heat of the flames:

Der Farbe der Flammen nach zu urteilen müssen die Temperaturen bei 800, 900 Grad gelegen haben.

(My translation: "Judging from the color of the flames, temperatures must have been 800 or 900°C.")

Of course, this interview is not a peer-reviewed publication on the temperatures that wood fires can reach. However, based on this expert statement, there is little reason to doubt that a fire such as the Notre Dame fire can be far hotter than the 600°C the Twitter comment mentions. Note of course that his statement does not answer whether the fire was really hot enough to melt the golden altar cross, or whether the choir was actually exposed to this extreme temperature.

25

Wood is a perfectly acceptable and common material used in metal forging even more so when it becomes partially combusted (charcoal). What really determines the heat though, is the amount of oxygen it can get. If there were medium-high winds blowing on the building it could have melted even steel beams.

Reference: https://youtu.be/x_wYozMBWNk

In that video you see a forge burning raw wood getting hot enough to make steel white which typically happens around 1200C (reference2: http://www.smex.net.au/reference/SteelColours02.php)

  • 17
    Q: "do wooden buildings, such as the Notre Dame Cathedral, burn at temperatures above 600°C?" My Answer: "It's certainly possible". How doesn't it target the question? – Jesse_b Apr 17 at 17:40
  • 3
    skeptics.meta.stackexchange.com/questions/1505/… Maybe this is more helpful. An answer always needs a reference. You can have a look at the answer of LangLangC or target the specific occasion, like answer from Schmuddi – Maxim Apr 17 at 17:49
  • 3
    References added – Jesse_b Apr 17 at 18:05
  • 9
    Exactly. Virtually every metal forge for thousands of years was wood-fueled. Given fuels have a minimum temperature at which they will combust at a given pressure, but that doesn't mean it's the only temperature at which they will combust. – reirab Apr 17 at 18:52
  • 9
    @Maxim How is this not an answer to the question? If anything, I think that it's actually the simplest explanation given so far of why the claim in question is incorrect. In particular, I think that the point about wood fires being used to melt metal all the time is actually really compelling. – EJoshuaS Apr 17 at 20:39
16

The entire type of assertion "if X is burning, and X is said to burn at Y temperature, then the fire cannot melt Z which melts at Z temperature" is fundamentally flawed at at least two levels.

  1. The burning point of a material is the usual minimum point where it starts to burn, but not the maximum temperature of a fire involving that material.

  2. The temperature that would apply to a melting point is the temperature of the air in the environment, and as Jesse_b very rightly answered, that temperature is more about air flow, and the overall situation of the space. The amount of burning fuel, amount of air in fire reactions, and the heat and air flow of the entire environment all contribute to how hot that environment gets.

References:

  • I've completed courses in chemistry, and I have a good memory. Fire is an exothermic reaction, meaning it creates heat. The amount of heat released in an entire large fire is a factor of the amount of fuel consumed. The type of fuel (e.g. wood) merely determines the rate and surface temperature. The temperature of something that might be melted is based not on the temperature of some other nearby object, but on the temperature of the object that might melt, which is determined by how much heat it receives from all nearby sources, both through radiation and through contact with heated air and other heated objects.

  • "Primary combustion begins at about 540° F, continues toward 900° F and results in the release of a large amount of energy. [...] Primary combustion also releases large amounts of unburned combustible gases, including methane and methanol as well as more acid, water vapor and carbon dioxides. These gases, called secondary gases, contain up to 60 percent of the potential heat in the wood. [...] The conditions needed to burn secondary gases are sufficient oxygen and temperatures of at least 1100° F. The air supply is critical. Too little air will not support combustion and too much will cool the temperature to a point where combustion cannot occur."

    (from "Stages of Wood Burning Combustion" by By Dwayne R Bennett - http://www.flameandcomfort.com/archives/blog/311)

  • "A bonfire should be treated with respect as it can reach temperatures as hot as 1,100 degrees Celsius (2,012 degrees Fahrenheit)."

    https://sciencing.com/hot-bonfire-8770.html

  • 3
    For whatever reason, many people seem to view a combustion point as similar to a boiling point, where an increase in temperature will increase the rate at which an endothermic process occurs, pushing the temperature downward. While mixtures (e.g. chafing fuels) can be formulated to behave that way, in most cases where fires get out of control, they do so because increasing temperatures increase the rate of exothermic reactions, resulting in thermal runaway. – supercat Apr 18 at 19:04
  • This seems to be addressing the question of whether it's possible for the fire to exceed 600 C, while the question asked seemed to be about whether it's possible for it to not exceed 600 C. – Acccumulation Apr 19 at 15:35
  • @Acccumulation Why do you think the question is asking that (a strange question, and even more irrelevant to the melting situation)? The question's summary (in bold at bottom) reads "do wooden buildings, such as the Notre Dame Cathedral, burn at temperatures above 600°C?" – Dronz Apr 19 at 18:44
  • As I commented to the main question, there is a conflict between the title and the main question body. The question body puts the question in the context of someone presenting the survival of metal object as being remarkable. That would be the case only if a fire being lower than 600 C is unusual. – Acccumulation Apr 19 at 19:10
  • @Acccumulation Was that in an earlier version of the question before edits? Where do you see words that lead you to think that is the question? I'm not seeing it. Even if that were the question, it would be seen to be inappropriately framed if the asker understood what I say in my answer - the melting point of a cross being reached is more about the resulting heat reaching the cross, which is not reducible to the ignition point of the wood. – Dronz Apr 20 at 5:33

You must log in to answer this question.

protected by Mad Scientist Apr 18 at 8:40

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged .