Hottest Part Of Oxy Fuel Flame (LP Gases)

[Ahmdo Molah]

Searching online for 'hottest part of a flame', it is easy to encounter info about oxyacetylene flames and that the hottest part is around the tip of the inner cone.

But for alternate fuels(LP gases) like propane, butane, etc., simple graphic info is very limited (so it seems). I did find some (below) indicating that for propane, the hottest part is farther away from the torch tip and the inner cone - somewhere around 2/2.5 times the length of the inner cone.

[ Wilhelmsen.com ]
https://www.wilhelmsen.com/ships-ser...ne-vs-propane/

"Propane releases only a small proportion of heat in the inner flame cone (less than 10%), so most of the heat in the flame is located in the outer cone.
Acetylene releases almost 40% of its heat in the inner flame cone."

Hottest Part Of Oxy-Fuel Flame (Wilhelmsen).jpg


[ Miller/Smith ]
* From the Smith Little Torch Manual (p.21) 1-3.Flame And Heat Placement BTU Output Per Cubic Ft Of Fuel Burned With Oxygen

The left column of the chart below indicates the approximate BTU at the tip of the inner cone and the longer tail of the flame.

Acetylene: 500 vs 970 ( 500 / 1470 = 34% at tip of cone)

Propane: 250 vs 2250 ( 250 / 2500 = 10% at tip of cone)

Flame & Heat Placement BTU Output... (Smith).jpg


[ DelphiGlass.com ]
https://www.delphiglass.com/blog/how...f-and-my-torch
from "Me, Myself, and My Torch" by Cere Ceddon

This is from a glass artisan's perspective.

Flame Zones (C.Seddon).jpg

"A - hottest part of the flame - used most of the time
B - used for metal fuming - gold or silver is vaporized at the tip of the inner cones
C - inner cones - composed of unburned gasses - cool and useless for heating
D - lower edge - used for precise, controlled heating - used for adding detail or stretching thin controlled tapers from rod
E - diffused heat area - used for pre-heating and for some tube work
F - outer reaches - used for preheating and for final heating of tubing just before blowing or stretching"


If anyone has more information on this matter, especially on flame temperature measurement methods used for determination, I'd be grateful.

[Mark Kelly]

If anyone has more information on this matter, especially on flame temperature measurement methods used for determination, I'd be grateful.

My first job out of Uni was in a research group and involved runnng an atomic absorbtion spectrophotometer which used an oxy acetylene atomiser*. That was a long time ago but from memory the flame temperature is determined by using a dual line absorbtion technique where the flame temperature is proportional to the ratio of the absorbance of the two spectral lines (I think it's a log relationship as it derives from the Boltzmann equation).

The reason for the flame temperature distribution is fairly simple: it depends on the relationship between gas flow rate and reaction rate. With acetylene, the reaction rate dominates so temperature reduces as you move out in the flame. With other gasses the reaction rate is slower so there is more combustion further out.


*if you think industrial grade acetylene is pricy, try high purity instrument grade.

[Ahmdo Molah]

Atomic Absorption Spectroscopy(AAS) is definitely interesting, but a bit more than I can chew at the moment . OTOH, I do get that LP fuels and their slower reaction rates result in the hottest zone being further out than the tip of an acetylene flame's inner cone.

What I'm wondering is how the sources I've mentioned above or their (unmentioned) 'sources' came to the conclusion that... "oh, the hottest point is... there!"

There was a time (not too long ago) when I thought a thermocouple, or an IR(infrared) camera would do the trick. I found out that thermocouples, pyrometers, and IR cameras can all be pretty inaccurate and/or inconsistent due to their inherent limitations.

[Mark Kelly]

The technique I mentioned allows measurement of the flame temperature in different zones.

See fig 3 in this paper for example (the two elements technique being discussed wasn't around when I was using an AA).

FWIW IR cameras are useful when the object is emitting in the infra red, below around 700 oC. Thermocouples can be used to directly measure flame temperature but you need special high temperature miniature ones and they disturb the flame.

[Ahmdo Molah]

I SEE!

After reading the entire research paper (like a cow...) I looked up "atomic absorption spectroscopy principle" and fortunately the very first site I encountered demystified the process bit by bit (without the fancy equations) in plain talk.
https://www.technologynetworks.com/a...cations-356829

So... if I'm not too off track... light from a light source with a filament of a specific material(s) (and therefore clearly defined spectral lines) is passed through a flame of a specified fuel (in an 'atomized state'), then passed through a filter (that blocks out unnecessary spectral bands), and a sensor(CCD, charge coupled device) detects the resulting spectral line intensity (after the partial absorption of source light by the flame).

It seems that in this method, the highest temperature location would be determined by the location where the absorption of source light is greatest (where the flame's reaction intensity and therefore the amount of relevant atoms are most concentrated).

Thank you!

[Mark Kelly]

Not quite.

The flame measurement is a bit subtler than just measuring absorbance. What you are looking for is the Boltzmann distribution, by measuring at two different wavelengths you can find the slope of that distribution which is proportional to temperature.

You cannot use a filament as source, if you did you would get a continuous spectrum rather than spectral lines (radiation from a solid is governed the Stefan Boltzmann law). To get spectral lines the material must be in gas form (eg each atom has six degrees of freedom), this is achieved using a hollow cathode tube which sputters atoms off the cathode material using inert gas ions accelerated from the anode.

Yes Boltzmann's distribution and Stefan Boltzmann Law are from the same guy: he invented stastitical mechanics which is why one of the seven foundational quantities of the universe is named after him (by Max Planck).

[Ahmdo Molah]

Not quite. ...

Thanks for your corrections to my over simplifications. In my excitement at even a VERY general/partial understanding of the overall process of AAS, I wanted to express my gratitude.

I will certainly take plenty of time to check out the fascinating details you've mentioned. Time for some more COFFEE. Cheers!

[Ahmdo Molah]

... What you are looking for is the Boltzmann distribution, by measuring at two different wavelengths you can find the slope of that distribution which is proportional to temperature. ...

You done it... (got me to get my NERD ON!)
I FINALLY understand (at least) the significance of the Boltzmann Distribution and related equations which interrelate BOTH variables for absorption ratios(a1/a2)* AND the temperature(T) of the (acetylene) flame that atomizes the test samples.
* Of two different metals (as in the article you linked, Cu/Fe or a1/a2).

Regarding your comment about "slope of that distribution";
1. Distribution of what variable(s) are you implying?
2. What is the numerator/denominator that form the slope? Do you mean 'a1/a2'?

I ask because I've searched for 'dual-line atomic absorption xxxxx' and could not find accessible material online other than the research article you linked unfortunately. I'd greatly appreciate your explanation or advice on where I can get some info that explains the conceptual or mathematical process that is already stated in equation form in the article.

The following YouTube video was helpful only to a limited extent in understanding this topic.
How Temperature Affects Atomic Spectroscopy

...You cannot use a filament as source, if you did you would get a continuous spectrum rather than spectral lines (radiation from a solid is governed the Stefan Boltzmann law). To get spectral lines the material must be in gas form (eg each atom has six degrees of freedom), this is achieved using a hollow cathode tube which sputters atoms off the cathode material using inert gas ions accelerated from the anode. ...

Upon checking out some more info on actual AAS machines, I now understand that the metal(s) to be utilized for test need to be the same in the hollow cathode tube(HCT) and the atomized sample, and that the flame itself is for atomizing the test sample (i.e. the energy from the sputtered atoms of the HCT are absorbed by the atomized atoms in the flame).

For those that are curious, but lack the urge to read about AAS...

[ AAS Introduction ]
Atomic Absorption Spectroscopy (AAS) Explained - PART 1


AAS PART 2: Calibration Curves Explained



[ Actual AAS Setup ]
Atomic Absorption Spectroscopy Part 1


Atomic Absorption Spectroscopy Part 2



[ Atomic Absorption Spectrometer Teardown (& Roasting A Hotdog) ]

[Mark Kelly]

1. Distribution of what variable(s) are you implying?
2. What is the numerator/denominator that form the slope? Do you mean 'a1/a2'?

The molecules in the gas will either absorb radiation or emit it depending on whether they are in the ground state or an excited state*. The ratio of absorbances at different frequencies thus corresponds to the relative population of atoms in the ground vs excited state at different energy levels (since energy corresponds to frequency according to Planck's Law). Boltzmann's distribution describes the way that this relative population of atoms at a given energy level varies with temperature, so if you can measure the variation you can impute the temperature.


*Ground state means the electrons are in the lowest energy orbitals thus minimising the energy in the electron cloud. Excited state means one or more electrons have been boosted to a higher orbital. When the electron reverts from a higher orbital to a lower orbital it emits a photon of an energy corresponding to the difference in energy of the orbitals. Conversely, if radiation of an energy corresponding to a difference in orbital levels is passed though a cloud of atoms in the ground state, some of them will absorb the energy in the photons and boost the electron to a higher orbital. The fact that this only happens at defined energy levels means there is a minimum threshold, called a quantum. This is the basis of quantum theory.

If you understand quantum theory, you are doing better than 99.99% of us. I certainly don't so there may be holes in my explanation above (that was a quantum theoretical joke, which means it is only theoretically funny).

[Ahmdo Molah]

According to Mr. Garro's opinion...

Welcom To Flagstaff.jpg

[Mark Kelly]

The fact that this only happens at defined energy levels means there is a minimum threshold, called a quantum. This is the basis of quantum theory.

If you understand quantum theory, you are doing better than 99.99% of us. I certainly don't

This is a poorly worded, it should say something like "This only happens at defined energy levels so each transition corresponds to an amount or quantum of energy. This is the basis of quantum theory." The next bit is still true.

[Ahmdo Molah]

... "This only happens at defined energy levels so each transition corresponds to an amount or quantum of energy. This is the basis of quantum theory." ...

Relating this back to 'imputing' the temperature of a flame, my general conceptual understanding of how it is done with AAS is;

1. The light energy source or Hollow Cathode Tube(HCT) generates an EMR(Electro Magnetic Radiation) - the "quantum of energy" that causes electrons of metal atoms in the HCT to 'jump' to the higher orbit or 'excited' energy state, and when these atoms return back to the unexcited 'ground' state(the excited electron that had jumped to higher orbit returns to its unexcited initial lower orbit), release the energy in the form of photons/EMR(Electro Magnetic Radiation).

2. The released photons/EMR from the HCT are focused in a beam through an atomizing flame of acetylene fuel for example, where a metal of the same type as in the HCT is atomized from a solution (in which the metal is dissolved).

3. Some of the photons/EMR from the HCT beam are absorbed by the atomized atoms that are in unexcited 'ground' state as the beam passes through the flame.

4. The monochromator filters out 'noise' from the beam and passes only specific bands of the metal in the HCT and test sample.

5. The sensor receives the beam after absorption(or redundant addition from an excited atomized sample) of engery has affected it.

6. The temperature of the flame can be imputed/extrapolated by an equation* which interrelates among various factors the absorption ratios of the metal samples and the temperature, based on the Boltzmann Distribution.
* The image below taken from the research paper you linked above, shows the two metals Cu & Fe, their frequency/absorption bands a1 & a2, and the equation for T (temperature) which includes the absorption ratio.

Temperature Equation Based On Boltzmann Distribution.jpg

where
E = excitation energy
k is the Boltzmann constant
a1, a2 = absorption by Cu & Fe
A = is the Einstein transition probability
g, g' = the statistical weights corresponding to the lower and upper level of the spectral line
𝜆 = the absorption wavelength
M = the standard atomic mass

Air+Acetylene Flame Temperature By Height.jpg

As the research paper was intended for professionals in the field and therefore lacks step by step explanations of the process of derivation towards the key equations and any example calculations, I am seeking the other source research papers (regarding dual-line atomic absorption method) mentioned in the paper you link.

In the meantime my NERD is very elated even with the bits understood so far thanks to you sir! Cheers!

[Mark Kelly]

1. Close enough. The excitation of the HCT is actually inert gas ions accelerated from the anode which collide with the material on the cathode and knock individual metal atoms off, creating a cloud of these metal atoms. Most of the metal atoms are in an excited state because there is so much energy in the ion collisions.

2 -6. Yep, that's pretty much how it works.

[Ahmdo Molah]

I appreciate your clarifications of my simplifications!

I just finished reading the wiki on 'Boltzmann constant'(actually taking the time to understand the equations which were a bit more accessible to me)and found that it helps enormously in understanding (conceptually) how the Boltzmann Distribution is so meaningful.

https://en.wikipedia.org/wiki/Boltzmann_constant

Thanks again!

Now onto using this knowledge to advantage for practicing fillet brazing!

[Mark Kelly]

To get spectral lines the material must be in gas form (eg each atom has six degrees of freedom), this is achieved using a hollow cathode tube which sputters atoms off the cathode material using inert gas ions accelerated from the anode.

I made a mistake here, sorry. I forgot that a vaporised metal forms a monoatomic gas, so each atom only has three degrees of freedom that are thermodynamically significant.

[Ahmdo Molah]

As I'm currently using butane as my fuel, I wondered how its reaction time(flame speed) compares with propane(or other fuels). Apparently there is relatively limited research that has been done on butane compared with other fuels such as propane and methane(natural gas).

But I did find a research paper that digs into this matter indirectly - a study on natural gas and the effects of other component gases such as methane, ethane, propane, and butane.

The HAL Open Science research paper and download link:
'Measurements of Laminar Flame Velocity for Components of Natural Gas'
https://www.google.com/url?sa=t&rct=...zuwPxc3fvnzxsy

On p.5 the authors present the following results which compares previous research data by other researchers and the authors' test results for laminar flame velocity of methane, ethane, propane, and butane with air.

*Note: The equivalence ratio Φ is defined as the ratio of the fuel mass flow rate to the air mass flow rate divided by the same ratio at the stoichiometry of the reaction considered**.
** In other words...
Φ = 1 : reaction according to stoichiometry,
Φ less than 1 : less fuel (than air/O2),
Φ greater than 1 : more fuel (than air/O2)

Flame Speeds to Equivalence Ratios of Hydrocarbons/Air Mixture
Flame Speeds of Hydrocarbons.jpg

According to the authors' results, the maximum flame velocities of propane and butane (in air mixture) were 40cm/s and 39cm/s respectively, so the difference is relatively small (considering acetylene/air flame velocity is roughly 3 times higher).

Acetylene/Air Flame Speed to Equivalence Ratios
Acetylene & Air Flame Speed.gif

Source: Laminar Flame Speeds
http://ignis.usc.edu/Mechanisms/USC-...me%20speed.htm

The HAL research paper also presents a flame velocities graph for methane/air mixtures with varying concentrations of O2 - 21%(normal air) up to 35%. Note the proportional increase in flame velocities with increase in O2 concentrations. Obviously with an O2 percentage in the 90s range the flame velocities would be significantly higher (well beyond the range of the graph shown below).

Also of interest, at lower O2 concentrations the peak flame speed is at an equivalence ratio greater than 1, whereas with increases in O2 concentration the peak flame speed approaches equivalence ratio of 1.

Flame Speeds With Varied O2 Enrichment
Flame Speeds With Varied O2 Enrichment.jpg

[Scheisserad61]

From a Harris brochure I found online :Screenshot (2669).jpg