| MadSci Network: Physics |
Hello Daniel This Mad Scientist must admit to being a little bit mystified about the prior history of your query as noted in the message section of your submission. I'm guessing you are interested in plotting percent of water molecules broken down vs. temperature. I have not much specific data on the absolute amounts decomposed at given temperatures, but I can take a shot at giving you a reasonable temperature range over which the thermal decomposition of water will occur. In any thermal system, the particles (be they molecules, atoms, or ions) will have a distribution of energies. Those in the "hot tail" can initiate reactions that become more prevalent later. That is, the amount of freed H2 O2, OH, O, and H running around will not go immediately from zero to 100% decomposition but will rise with temperature. One quick way to estimate the temperature about which this reaction will occur is to compare kT (Boltzman's constant times absolute temperature in Kelvin) against the bond energy of the bonds (O-H in this case) to be broken. Doing this gives me a crude estimate of 2300K. I did a web search (www.google.com, using the keyword 'water' and the keyphrase "thermal decomposition") and got several thousand hits. Most of these were rather useless (thermal decomposition of other compounds in which one of the products was water) but I found one discussing a solar concentrator setup that was getting about 2% of the hydrogen in the feed water coming off as H2 at 1800K. So the reaction is certainly happening at that temperature. Switching the web search to keyword 'water' and the keyphrase "stellar atmosphere" got a few hundred hits. I found two papers that discussed the presense of water (as evidenced by its absorbtion bands in the stellar spectra) in red dwarf (i.e. low mass, fairly low surface temperatures--for a star, anyway) stars. One paper (in Mon. Not. R. Astrom. Soc., v. 277, pp 767-776) showed H2O bands in such a star with a temperature of 2800K. The second paper (in Publ. of the Astrom. Soc. of the Pacific, v. 110, 1007- 1011) noted water in other, similar red dwarf stars. From these we can conclude that some water must still exist even as high as 2800K. If you can find some data on the amount of OH noticed in red dwarf stars vs. the amount of H2O (this level of detail is beyond the expertise of your Mad Scientist, but you might contact the authors of the papers), then perhaps an estimate of the percent decomposition could be made. However, the chemical equilibrium can easily be skewed in a star because there is obviously quite an excess of hydrogen around and we're no longer dealing with a pure H2O feed system...not to mention competition from reactions with metals for the available oxygen, e.g. TiO, ZrO, etc. So, it seems that the thermal decomposition is beginning to occur about 1800K, and is can not be complete at 2800K. All that said, you must be very careful when setting up your experiment to be wary of catalytic effects by other materials in the system. For instance, it can be readily demonstrated that boiling water (373K) on silicon will result in some hydrogen being absorbed into the silicon and inactivating boron dopant by the formation of B-H pairs in the silicon lattice. Small amounts of contaminants, especially anything with carbon or hydrocarbons, will also lead to hydrogen evolution by water-gas-shift kinds of reactions (e.g. H2O + C --> H2 + CO ...methane is also a possible product gas). To make a purely water-only system is harder than you think, because of the difficulty of confining high temperature (>1800K) steam and its extremely aggressive chemical nature.
Try the links in the MadSci Library for more information on Physics.