What is the boiling point of Steel?

[mecheng]

I'have searched a lot but could not find boiling point for steel. Can you tell me what's the boiling point of steel?

Also any link for a table which provides both melting and boiling points for commonly used metals would be greatly appreciated.

Here is what I found so far:

1. Melting Point: The melting point of most steel alloys typically falls in the range of 2,600 to 2,800 degrees Fahrenheit (1,427 to 1,538 degrees Celsius). However, this can vary depending on the specific composition of the steel.
2. Boiling Point: Steel does not have a well-defined boiling point because it doesn't transition directly from a solid to a gas at a specific temperature. Instead, it undergoes various physical and chemical changes when heated. At extremely high temperatures (well above its melting point), steel may start to vaporize, but this isn't a common or practical occurrence.

 

[NateTG]

I'have searched a lot but could not find boiling point for steel. Can you tell me what's the boiling point of steel?

Also any link for a table which provides both melting and boiling points for commonly used metals would be greatly appreciated.

Very few metals are used in pure, or even relatively pure, forms. Steel, for example is the name for a whole family of iron alloys. The melting point of steel can vary drastically depending on the composition, and if someone told me that it also depended on how the steel had been tempered, I'd be inclined to believe them. If pressed, I would suggest that the boiling point of iron (not steel) is 2750 C, so the boiling point of steel is likely to be close for most steels.

 

[brewnog]

http://www.ktf-split.hr/periodni/en/ is a periodic table, click any metal you like to find out all sorts of physical data, including boiling and melting points.

As NateTG already pointed out, this data is for elements in their pure form. Be aware that in engineering applications, metals are frequently alloyed which can vastly affect physical properties.

Finally, a quick look at a few metal suppliers' websites showed the boiling point of a medium carbon steel to be around 3000K, which:

a) Is probably about as accurate as you'll find,
b) Fits in nicely with NateTG's suggestion, and
c) Shows that you didn't look that hard, after all!

 

[FredGarvin]

I guess the more important question is why one is interested in the boiling point? I can't say that in my professional career I have ever had to look at steel's boiling point. Then again, I don't work in a foundry...

 

[Integral]

Alloys can and do have melting properties much different form any of the constituent metals.

I well remember the demonstration in my Thermodynamic class where the prof showed us 2 solid materials (I believe one was Indium) when he but them in the palm of his hand and kneaded them together, they melted into a liquid pool. The melting point of alloy was below that of the constituents.

But you are talking boiling. This is a much more complex question, it may be that it is simply not possible to "boil" steel, that is each of the lighter constituents element may be sublimated out as the material heats up. I do not have this a personal knowledge, it is just a guess.

 

[Astronuc]

NateTG and brewnog have pretty much answered the question.

Steel is a multi-element alloy, principally of Fe, with additions of C, Cr, Ni and other metals. The carbon is mostly in the form of carbides of the alloy metals Fe, Cr, and Ni. The carbides will have higher melting/boiling temperatures than the metal matrix.

The boiling points I found are:

Fe: 3134 K (2861 °C)
Cr: 2944 K (2671 °C)
Ni: 3186 K (2913 °C)

I would go with Fe, especially if the steel is structural steel as opposed to high alloy stainless steel.

I looked in a host of materials and steel reference books, but I could not find a 'boiling temperature' for steel, or any other high temperature material.

I suspect a book on foundry technology would have such a book. I may have another source that I can check tomorrow.

 

[Gokul43201]

We're not done spoiling the broth are we ?

I'd go with about 2860 °C. This should be very close for most low-alloy and low/medium carbon steels, and reasonably close for other steels (even a 2% C-steel will not likely boil at over 2900 °C).

If you need a higher accuracy, it is possible to find the boiling point of a liquid from raoult's law and the clausius-clapeyron equation. This however requires the knowledge of such data as the liquid composition and the enthalpy (latent heat) of vaporization of liquid Fe.

The liquid composition will be non-trivial. There will be liquid Fe along with other molten metals (alloying elements), all of which should probably be considered miscible, volatile liquids (so that the total vapor pressure is weighted by their mole fractions). Then there will be insoluble (non-volatile) solutes which will primarily be carbides and oxides (concentration depending on the conditions) of the metals (primarily of Fe), but fortunately these do not affect the total vapor pressure. Finally, there will be some soluble (non-volatile) solutes like free carbon which will lower the vapor pressure and hence raise the boiling point.

Unless you have a PC steel or no more than one major alloying element, the analysis will be much too hard to perform.

 

[Astronuc]

I like Gokul's number (2900°C) as a nice round number, or even 3000°C, and even nicer rounder number.

I found two other sites through Google, but the links were bad. Nevertheless, the Google summary of one site stated:

At the normal boiling temperature of iron, Tb =3330 K (3057 °C), the rate of change
of the vapor pressure of liquid iron with temperature is 3.72×10-3 atm/K.

While the other site stated Tb =3310 K (3037 °C).

These last two seem a bit too high.

 

[mecheng]

Thanks everyone for this wonerful response.

I'll assume that steel would boil around 2900C.

 

[ohwilleke]

I guess the more important question is why one is interested in the boiling point? I can't say that in my professional career I have ever had to look at steel's boiling point. Then again, I don't work in a foundry...

Another use would be safety considerations. For example, if you wanted to estimate on an order of magnitude basis whether a fire of a particular type with a known heat that will produce a certain temperatuure for a certain duration would structurally compromise a steel structure.

 

[Gokul43201]

Another use would be safety considerations. For example, if you wanted to estimate on an order of magnitude basis whether a fire of a particular type with a known heat that will produce a certain temperatuure for a certain duration would structurally compromise a steel structure.

For that, you might want to know the melting point, not the boiling point.

 

[NateTG]

For that, you might want to know the melting point, not the boiling point.

Actually (from the metallurgy is a black art department) the crystal structure of steel changes due to heat long before they hit their melting point. If you're at all worried about the melting point, the steel may already be gone.

 

[FredGarvin]

Nate is right. Steels transformation temperature is well below melting point. Creep issues come in below that as well.

I still am curious about the whole boiling point question. If the OP could maybe throw me a bone I'd appreciate it. It's only curiosity, no flaming intended.

 

[Astronuc]

I still am curious about the whole boiling point question.

Me too!

Adding to what Gokul, NateTG, and FredGarvin wrote, a rule of thumb for 'structural' materials, i.e. 'solid' and load-bearing structures, is - "limit the temperature to about 0.3 or 0.35 of melting temperature." Of course, that depends on the stress level. The issue is creep.

In general, 0.3 of T_melt is considered high temperature. For Fe or Ni based alloys, that would be 870°C (~1600°F), and Ni-based Superalloys (e.g. Inconels) are usually employed.

This is one area in which I am particularly interested, and going a step further, in a neutron (fission) or charged particle (fusion) environment. Materials for high temperature radiation environments and their behavior are areas that still require considerable R&D.

I am looking forward to interesting discussions in the future.

 

[Gokul43201]

Actually (from the metallurgy is a black art department) the crystal structure of steel changes due to heat long before they hit their melting point. If you're at all worried about the melting point, the steel may already be gone.

I'm aware of this, but typically, the recrystallization temperature (for instance) lies within 0.3 to 0.6 Tm (it can be pinned down to a smaller range in the absence of strain hardening), so a knowledge of Tm is useful for estimating safe working temperatures (as Astonuc did above).

 

[ohwilleke]

For that, you might want to know the melting point, not the boiling point.

Good point. Now that I think about it, boiling point would be more relevant for stuff like an industrial process that would "spray paint" steel onto some surface so as the get a thin, uniform layer, although once it is at gas, it is hard to still call "steel".

 

[Astronuc]

I found one site that discussed spraying 'liquid' metal, not 'vapor'.

Even with fusion welding, the objective is to melt a metal, which then resolidifies, hopefully as rapidly as possible.

In multi-element/multi-phase system, like many steels and superalloys, one has to be concerned with element/phase segregation and volatility of elements of low melting temperatures. Significant differences between the weld alloy and substrate (base metal), i.e. significant differences in local chemistry and microstructure, may produce significant differences in mechanical and corrosion behavior.

 

[brewnog]

Good point. Now that I think about it, boiling point would be more relevant for stuff like an industrial process that would "spray paint" steel onto some surface so as the get a thin, uniform layer, although once it is at gas, it is hard to still call "steel".

I'm struggling to think of an industrial application of this, if you're trying to 'spray paint' with metal, you do it in a liquid form (as with GMAW/MIG welding), although it's easy to see why one might think of this as vapour.

 

[Astronuc]

Melting point of steel is about - 1450-1520°C (from Corus Group), essentially half of the boiling point, ca. 2900°. It makes sense to spray with liquid temperatures around 1500-1600°C, rather than 2900°C. Much less energy required - and economics in industry is an important/critical (and non-technical) constraint.

 

[Gokul43201]

Are you folks talking about thermal spray coatings (like HVOF) ? I can't say I ever heard of steel spray coatings (actually I've only heard or CrO2 and WC based sprays). And the conditions in a spray are both (i) highly non-equilibrium, and (ii) rarely at 1 atmosphere. I can't see that the boiling point itself will be of very much use here.

The only applications that I can imagine where the knowledge of the boiling point would be useful is in vapor depositions or thermal evaporations onto a ceramic. But then again, these are usually done under high vacuum and I've never heard of either of these involving steels.

 

[FredGarvin]

I'll have to talk to my flame spray gurus at work to see if they know anything in this realm.

 

[Astronuc]

Application - Restoration of Dimensions in printing rolls, undersize bearings with Carbon Steel, Stainless Steel

from http://www.flamesprayinc.com/applications.htm