Air Pressure, Fire "Consuming" Oxygen

[D H]

A temperature-based explanation simply cannot explain why no bubbling is observed while the candle is burning. Burning increases the molarity of the gas. The only explanations for the observed fact of no outgassing occurs (in fact, the water rises) while the candle is burning are that the temperature is decreasing even though heat is being added or that gas is being absorbed faster than it is being created. How can the temperature decrease while heat is being added? Add to this the fact that gases are very poor conductors of heat. The heat transfer rate to the glass and the water will be slow.

[Klaynos]

But the candle flame is reducing in size... so the temp is dropping...

 

Incidently I'm only taking this position because someones got to... and I'd really like to know what is happening....

 

In the coloured liquid one you can see that the liquid is sucked (har) up from the dish...

[Cap'n Refsmmat]

The temperature of the flame is dropping, but not necessarily the temperature of the air in the glass.

 

If someone has apparatus to run the experiment with pressure and temperature sensors inside the glass, that would be perfect.

[hermanntrude]

this isn't combustion in oxygen, either. it's more complicated than that. By closing the system, you're forcing partial combustion to occur, resulting in the formation of CO, carbon monoxide. I don't think that explains the reduction in pressure, though. I think it's probably due to condensation of wax vapours. Some of the wax is volatilised during the combustion (particularly during the partial combustion), becoming gaseous, and then when the flame goes out (and also before, as the temperature starts to fall) the wax condenses out resulting in a large reduction in gaseous molecules, resulting in a drop in pressure. There way also be some water in the gaseous phase due to the heat of the candle.

[Cap'n Refsmmat]

So if we continue to apply heat to the glass, the water level should not rise?

[mooeypoo]

I'm actually going to try that.

 

I'm going to try a few things, I just finished talking to a chem professor at my uni.. he suggested a few tests I could do to check what exactly operates there.

 

So... Expect a follow up vid with these tests

[D H]

You can look for wax residue on the glass and floating on the water to confirm or reject the vaporized wax hypothesis. Look for signs of water on the glass to confirm or reject the glass condensation hypothesis.

 

Some tests:

1. Test of temperature hypothesis. Put the glass in the water bath, no candle. Heat the glass with a hair dryer for some time. This will raise the gas temperature, so you should see some bubbling from the glass. Let the glass cool. My prediction: The water will rise as the system cools, but not nearly so dramatically as with the candle.
    
    
   
2. Temperature sensitivity test. Conduct the candle test multiple times with the water bath at different temperatures. CO₂ solubility varies by a factor of 5.7 or so between 0C and 60C while H₂O vapor pressure varies by a factor of 33 or so over the same temperature range.

 

The carbon monoxide hypothesis: That's a good one. Since CO has negligible solubility in water, and increases the molarity of the gas, it does not explain the water rising. It certainly does make things more complex.

[Mr Skeptic]

Or find a source of flame that burns at a higher or lower temperature.

 

Good point. A big resistor would be just perfect.

[hermanntrude]

weigh everything (glass, water, candle) before and afterwards. I think that will give you the best chance of finding out what happened.

 

CO isn't a hypothesis. A theory is much closer, or even a well-educatd prediction based on very well-known chemistry. Any textbook will tell you that in the absence of excess oxygen, combustion gives rise to carbon monoxide as well as carbon dioxide. But you are correct that it doesn't give any reason for the drop in pressure.

 

I'm 99.9% certain that it's to do with a gaseous substance consodensing. Whether that substance is wax or something else (most like water, but i dont quite know how), i'm not so sure. I think wax is the prime candidate.

[D H]

CO isn't a hypothesis.

You raised it as a complicating factor. For that to be the case, the amount of CO production would have to be significant. Carbon monoxide will be produced in some quantity, but if that quantity is only a tiny fraction of the combustion products this incomplete combustion can be ignored.

 

I'm 99.9% certain that it's to do with a gaseous substance consodensing. Whether that substance is wax or something else (most like water, but i dont quite know how), i'm not so sure. I think wax is the prime candidate.

Some agent is responsible for the rising water level. One thing is certain: this agent cannot be condensation of candle wax. Think about it. The candle wax starts as a solid. Any candle wax that vaporizes, fails to combust, and fails to condense represents a net increase in the molarity of the gas. This does not support the rise in the water level; it is in fact counterindicative. Any candle wax the vaporizes, fails to combust, and later condenses back as a solid represents a zero net change. This is neither supportive or counterindicative; condensed wax cannot explain the phenomenon.

 

======================

 

 

In this link, http://dc19.4shared.com/download/12762082/14547575/Incomplete_Combustion.pdf the authors (unknown authors, unfortunately) report on an experiment very relevant to the topic at hand. They burned a candle in an enclosed space and measured temperature and the quantities of O2, CO2 and CO. A summary:

• The candle went out when the oxygen level fell to 16.5%. The candle does not burn all of the oxygen in the glass. It goes out much, much sooner than that.
   
   
  
• Incomplete combustion occurs, but minimally. 1/1080 of the carbon combustion products are CO. The rest are CO2. Incomplete combustion does occur, but it is negligibly small.
   
   
  
• The temperature rises by a paltry 2 degrees C. Moreover, subsequent cooling is very slow. The small temperature rise and the long cooling period means this cannot be a temperature-related phenomenon.

 

The last observation indicates that temperature is not the dominant factor in the rise in the water level. Without detailed measurements it appears in Mooey's video that the water rises about 1 up of the glass and the glass appears to be about 10 cm tall. That rise corresponds to a pressure difference of about 98 pascals. Using the ideal gas law, the temperature would have had to risen by 30 degrees C to account for the rise solely with temperature. The temperature rose by 2 degrees in the cited experiment. Even if gas does escapes from the glass during the burning (no bubble is observed in Mooey's video) and even if cooling is quick (which it isn't), a paltry 2 degree change cannot account for the drastic rise in water level.

 

My hypothesis has been from the onset that the phenomenon results from a reduction in the molarity of the gas in the glass caused by condensation of combustion products in the water bath. A simple and safe test to falsify the condensation of combustion products in the water is to use a bath of very hot water instead of cold or room temperature water. The hot water has a much reduced ability to dissolve CO2 and to absorb water vapor. If the water level still rises significantly the condensation hypothesis is falsified.

 

One final observation: Googling this phenomenon leaves me very saddened by the state of science education. Some teach the phenomenon results from to burning of oxygen. Mooey made this mistake herself, but later corrected it. While there is absolutely nothing wrong with a student making such an error, there is a lot wrong with a teacher making this error. Some attribute it to temperature, but don't do the simple analysis to show what kind of temperature change is needed and don't do the simple observation to see if gas bubbles out of the glass while the candle is burning. Nobody attributes it to a change in molarity.

[swansont]

 

[*]The temperature rises by a paltry 2 degrees C. Moreover, subsequent cooling is very slow. The small temperature rise and the long cooling period means this cannot be a temperature-related phenomenon.

 

Something is very wrong here. If we had a 1L bottle, and our oxygen concentration drops 3.5% of that volume (20% to 16.5%), that's 1.56 millimoles of O2 burned. If the paraffin is C₂₅H₅₂, we need 37.5 moles of O2 burned per mole of paraffin, so that's 51.6 micromoles, or a mass of 14.6 milligrams of paraffin lost to combustion. The heat of combustion is ~42 kJ/g, so this releases 615 Joules of energy. Even ignoring factors of two for approximations, that's way off. It only takes a couple of Joules to raise air temperature 2 ºC for a volume of 1L. (C_p and C_v are different, but in the vicinity of 22.4 J/mol-K) The heat capacity of paraffin is around 2 J/g-K. How massive was the candle, or at least the part that heats up. — tens of grams? That's a few tens of Joules of energy.

 

The value also doesn't jibe with the caution about using a metal plate as a heat sink in the case where you use a plastic container. You won't melt plastic with a 2 degree rise in temperature.

 

So, where was the temperature probe? In the water, perhaps? If it was in the air we're missing several hundred Joules of energy, or I've made a math mistake I can't find. (or someone has embezzled it). But the water makes sense, because C_p is 4.18 J/g-K, so less than 100 ml of water will take that energy right up and only rise two degrees. The air temperature could have easily risen 30 ºC and the water near the probe only 2 ºC.

[D H]

Something is very wrong here.

I agree. Using 42 kJ/mole and ignoring details such as heat absorbed by the candle, the glass, and the water leads to a temperature rise of more than 300 degrees. Memo to self: Never trust a random web site, and think before you post.

 

I tried the experiment myself with a water bath at different temperatures, ice cold, tap temperature, hot, and scalding hot. (Hey, it's Saturday.) I observed miniscule quantities amounts of condensed wax on the glass. I observed some outgassing in the scalding hot water, but not in any of the others. The water level starts to rise immediately, but the pace picks up as the candle burns down. The amount the water rose differed significantly depending on the water bath. Ice cold water: ~2 cm; tap water: ~1 cm; hot and scalding hot water: ~0.5 cm.

[ydoaPs]

Another experiment to do would be to measure the effect of the heating of the milk by the candle. Do it again in a graduated cylinder with no glass and measure how much the milk expands. This could be part of why the level increases prior to the flame extinguishing. I'm pretty sure the video relies on multiple effects, however.