Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Why does CO2 keep keep heat within the atmosphere? (on the molecular level)

  1. Jun 16, 2010 #1
    I know I know this isn't really related to molecular biology, but I figured since Biologists have a strong background in ecology in most cases, I'd be able to get more help here than the Earth forums (especially because biologists know a lot of chemistry).

    Basically I understand that CO2 keeps heat in the atmosphere. But how? I'm talking on the molecular and sub-atomic level here (i.e the molecule itself, it's electrons). Thanks!

    EDIT: I understand what greenhouse gasses do, why they're important, the fact that there are more in the atmosphere than should be, the reason there are more in the atmosphere than should be (carbon emissions from fossil fuels mainly), and almost every eco-related aspect of carbon dioxide. I'm talking chemistry now. Why do these molecules act this way? (think chem and physics)...maybe this should be on another forum page haha.
    Last edited: Jun 16, 2010
  2. jcsd
  3. Jun 16, 2010 #2

    D H

    User Avatar
    Staff Emeritus
    Science Advisor

    Greenhouse gases such as CO2 have one thing in common: They are fairly transparent in the visible range but are not so transparent to infrared light (the more potent greenhouse gases are downright opaque in the infrared). Most of the energy in sunlight is in the visible range. Greenhouse gases, being more-or-less transparent in the visible range, let sunlight pass through to the Earth's surface with only a slight reduction in intensity.

    The Earth's surface radiates that absorbed energy in the infrared. Greenhouse gases absorb that ground-generated infrared energy. Those greenhouse gases re-reradiate that absorbed energy but do so according to their temperature, and this temperature depends mostly on altitude. This absorption / re-radiation process continues up the atmosphere, so by the top of the atmosphere the energy radiated by the Earth as a whole is somewhat attenuated.

    We need that greenhouse effect to some extent. If the atmosphere (including water vapor and clouds) was transparent in the infrared the average temperature of the Earth's surface would be a chilly -18 C or so. The problem then isn't that greenhouse gases exist; we need them to exist. The problem is that too much of a greenhouse effect will raise the Earth's surface temperature upward.
  4. Jun 16, 2010 #3
    Okay that was kind of what I was looking for. The "transparency" of carbon dioxide molecules with respect to infrared was a good analogy.

    But WHY? What chemical or electrochemical properties allow for carbon dioxide to be less "transparent" to infrared, and absorb it? Is it its specific heat and the bonds between the atoms? Is it the states of the electrons? Thanks.
  5. Jun 16, 2010 #4


    User Avatar
    Science Advisor
    2017 Award

    It's related to both the bonds between the atoms and the states of the electrons. Here's a quick explanation.

    When molecules absorb electromagnetic energy, the energy from the photon will kick the molecule into a higher energy state. There are two types of transitions that molecules can undergo: molecular transitions, which alter the molecule's shape or movement (for example, by inducing vibrations or rotation of the molecule), and electronic transitions, which excite electrons to higher energy levels. Because of quantum mechanics, the energies of both vibrational states and electronic states are discrete (quantized); thus, only photons that correspond to the energy of that a specific transition can be absorbed.

    As is probably apparent, it is much easier to cause molecules to vibrate than it is to kick an electron to a higher energy level. It turns out that the energies required for molecular transitions (specifically transitions between different vibrational states) fall in the IR region of the electromagnetic specturm (i.e. IR photons contain energies on the order of the energies required for these vibrational transitions) and electronic transitions fall in the UV region of the electromagnetic spectrum (for special molecules with special properties, electronic transitions can be moved down in energy into the visible region; this is how dyes work).

    Carbon dioxide has a particularly strong absorbance in the IR spectrum due to a specific vibrational mode (the asymmetric stretching mode) which excites the carbon to bounce around between the oxygens. Carbon dioxide is also a very stable molecule, electronically, so it does not absorb well in the visible or UV region. These properties of carbon dioxide contribute to its properties as a greenhouse gas.
  6. Jun 16, 2010 #5
    Now that was VERY helpful. Thanks a lot drasil. Could I draw a parallel between infrared and carbon dioxide, and Water and Microwaves? I understand microwaves use the polar properties of water to turn them, which results in more kinetic energy and more heat. Infrared causes carbon dioxide to vibrate, but not in that way I assume.

    Thank you :D
  7. Jun 16, 2010 #6


    User Avatar
    Science Advisor
    2017 Award

    Yes. Rotational transitions correspond to the microwave region of the EM spectrum. So, what happens in a microwave is analogous to what CO2 does in the atmosphere. Also note that water vapor is a good greenhouse gas as well. It is transparent in the UV/vis range but strongly absorbs in the IR. In fact, it is a much better greenhouse gas than carbon dioxide.
  8. Jun 16, 2010 #7
    I read that, thought "hmm", searched to see whether CO2 really does have such a narrow spectrum, and found:
    Is there a point at which further increasing the CO2 concentration in a planet's atmosphere will cease to further insulate the planet from the background dark of space?

    (To avoid giving anyone the wrong idea, I should also ask whether there is any point from which increasing atmospheric CO2 concentration ceases affecting ocean chemistry.)
    Last edited: Jun 16, 2010
  9. Jun 16, 2010 #8


    User Avatar
    Science Advisor

    Be careful with that statement. That's true for a single molecule or a gas at the zero-pressure limit. Otherwise you have line-broadening. Look at the spectrum of any gas at 1 atm and you're not going to see many sharp lines in the IR. CO2 does not have a narrow spectrum in the atmosphere.

    Cesiumfrog: That quote is from a climate-change-denier site, and it's simply based on bad science.

    Seriously, the role of CO2 and other greenhouse gases in global warming has been studied and debated for over a century since Arrhenius first came up with the idea. A naïve model based on the Lambert-Beer law isn't going to change this very well-established fact.

    Anyone who doesn't believe it is welcome to go stand in front of a CO2 laser.
  10. Jun 16, 2010 #9
    alxm, that's why I only quoted the actual argument without linking to the source. But regardless (and without specific reference to Earth's climate), I still ask the question I did above. Is the mistake treating an atmosphere purely as an absorber, rather than as layers that are both absorber and emitter?
  11. Jun 16, 2010 #10
    IR-absorbing gas molecules such as H2O, CO2, CH4 or SF6 all have a common property at the chemical level in that they are all molecules with covalent bonds between two different atoms with two different masses. Note that molecules with a covalent bond between two identical atoms (eg. N2 or O2) are not significant IR absorbers and therefore not greenhouse gases.

    When there are two different masses on either end of the covalent bond, that bond can vibrate or resonate quantum mechanically, and the energy levels of that system correspond to the energies of photons in the IR spectrum, and therefore these molecules can absorb (or emit) IR light.
  12. Jun 17, 2010 #11


    User Avatar
    Science Advisor

    Yes, that's one of them. You also have scattering, pressure broadening, the influence of other molecules, etc.

    As for your question about oceans, I guess you could put a 'maximum' for oceanic acidification at pH = 6.35, since that's the first pKa for carbonic acid. But the ocean being such an ionic mixture, and with all the possible buffering effects, it's really hard to say.
  13. Jun 17, 2010 #12
    I actually only mentioned oceans to make it clear (to anyone that stumbles upon my post out of context) that, regardless of how plausible that argument against greenhouse theory may seem, fossil carbon would still be devastating to ecosystems chains that we're accustomed to depending on. I would have thought that as long as the partial pressure of CO2 keeps increasing then (all other factors being equal at least) the amount of carbonic acid would also continue increasing (making it ever more taxing for organisms to maintain exoskeleton growth). Is this wrong, or are you just saying that 6.35 is the unbuffered pH if the atmosphere were 100% CO2?

    Do you really think that "scattering, pressure broadening, the influence of other molecules, etc" make many orders of magnitude increase to the length at which absorption saturates? It seems to me that those are factors that may indeed broaden the spectra beyond what quantum mechanics would predict in isolation, but would not separate (decrease the overlap of) the spectra of the CO2 at different altitudes. It would still remain the case that 99% of the original rays that the greenhouses gases in the atmosphere are ever going to absorb would all be absorbed in a distance very much less then the height of the atmosphere. If the wavelength is close to the resonance then I hear no reason to doubt the photon is very likely to be absorbed in the first few metres of the atmosphere, and if it is far from resonance then it would easily transmit past the entire atmospheric column unattenuated. Changing the greenhouse gas concentration by a factor of 2 would alter the distance in which a ray is absorbed (if it were already going to be absorbed anyway), but would not practically alter the overall likelihood for a specific ray to transmit through the entire height of the atmosphere (nor alter the portion of energy that will be absorbed from a beam of thermal radiation).

    Hence, I think the true problem is a complete misconception of the greenhouse effect.

    If the atmosphere merely blocked some of the surface's heat from radiating directly to space (and trusting that the absorption saturation or "extinction" length for atmospheric concentrations of greenhouse gases really is several orders of magnitudes smaller than the height of the atmosphere) then even an order of magnitude change in greenhouse gas concentration would negligibly alter the total quantity of "blocked heat radiation". This might lead one to jump to the conclusion that greenhouse gas concentrations are no longer affecting temperature.

    I think the critical concept is that half of this intercepted energy is radiated back again. The top layer of the atmosphere radiates a unit of energy back to the layer below, elevating the temperature of everything contained below (to have achieved radiative equilibrium). The second from top layer further elevates the temperature of everything below it, and so on for every further layer (without having checked I suspect this asymptotes to the solar temperature). Halving the extinction length effectively doubles the number of layers, which would validly lead one to conclude that greenhouses gas concentrations affect temperature. But this is a more intricate process, do you think that denier was presuming a single-layer effect?
  14. Jun 17, 2010 #13

    This is really interesting, cesium. I enjoyed reading this. I believe I understand most of what you are saying, but could you clarify absorption saturation and "extinction"?

    I do not know all that much about the chemistry and physics behind the contributions carbon dioxide makes to the greenhouse effect, but I would be surprised if ecologists, chemists, and Environmental scientists/engineers have neglected the idea of thinking about the levels of Carbon Dioxide in "layers". What is the standard model that scientists use? Surely they don't believe that carbon dioxide is simply one layer. Thank you for your answers, they've been very helpful.
  15. Jun 17, 2010 #14
    Thanks zewpals. An example of extinction length is that, since water absorbs/scatters sunlight, it gets really dark if you go 1km down. (The illumination is extinguished.) It wouldn't seem much darker if you went 5 or 10km deep, since practically all of the sunlight which the ocean was ever capable of absorbing has already been absorbed in the first 1km. (Since adding more depth will not significantly increase the total absorption, we can describe it as saturated after 1km.)

    I agree that the standard model used by climate/atmosphere experts is more nuanced and well supported. But I've stumbled upon at least one website advocating a contrary conclusion, this one founded on the assumed fact that absorption of IR by air similarly saturates in a rather short distance. Rather than blindly ignore this dissenting view (nor blindly accepting alxm's dismissal of it) I sought to understand for myself whether the views were rational.
    Last edited: Jun 17, 2010
  16. Jun 18, 2010 #15

    Alright, I certainly agree that both arguments are worth exploration. Light is not readmitted in every direction I believed, though. Once it hits the water molecule it is either reflected or absorbed, right? The absorbed energy is transformed into another type of energy, such as heat or used to excite electrons.

    I understand what extinction entails now, but when molecules absorb heat, they release it in every direction; when molecules come into contact with light, they either absorb the light and do not release that energy as light, or reflect it. This is what I believe, but I'm not sure.

    This also raises another quick question. As some molecules are transparent to light, are some molecules transparent to infrared? Infrared does have a larger wavelength, so I assume it could be possible. Then again, I've never heard of anything that could not absorb heat. Why is infrared considered to be "heat" anyway? I thought heat came from the vibrations of molecules. Finally, are some molecules truly transparent to light? Or are they just less opaque than others (does light actually pass through them? and what chemical properties allow that?)? Thanks!

    Wow that was not a quick question haha. If you're too frustrated to answer all those, don't bother haha I understand. I miss the days in school where I was actually interested in what I was learning and asked soooo many questions. College classes suck.
  17. Jun 18, 2010 #16


    User Avatar
    Science Advisor

    I'm saying we've got more than sufficient chalk around that I can't imagine the CO2 levels ever pushing it past that buffer point.

    No, I was just pointing out some of the reasons why the Beer-Lambert law isn't applicable.

    Anyway, the main issue with greenhouse gases isn't the incoming radiation, it's the outgoing radiation. Note that the Sun is 6000 K and the Earth is about 270 K or so. They both emit roughly as blackbodies, so if you look at the Planck curves for these two temperatures you'll see that there's very little overlap. The emitted radiation of the earth is way down in the infrared.

    The issue is that greenhouse gases absorb more strongly at these very low infrared frequencies than O2 and N2. It's not that the incoming radiation is absorbed more efficiently, it's that the outgoing radiation is being stopped from leaving.
  18. Jun 18, 2010 #17


    User Avatar
    Science Advisor

    No, it's either absorbed or not (individual molecules do not 'reflect'; that's a macroscopic property of substances). It can be re-emitted as light, or transmitted 'non-radiatively' to other molecules, i.e. they bump into each other and transmit some energy. (actually the 'bumping' is electromagnetic force, which is mediated by photons, so technically you could look at it as radiative emissions only).


    No, but I'm sure you've noticed different substances conduct heat at vastly different rates.

    Yes, heat is the vibrations of molecules, but vibrational energy can be emitted as photons (see above) so you have an equilibrium between vibrational energy and radiation. Infrared isn't really heat. It's just an area of the spectrum, which is energetically in the same area as thermal energy at 'human' temperature ranges.
  19. Jun 24, 2010 #18
    Cesiumfrog, the significance of increasing CO2 is that it effectively thickens the greenhouse gas layer. The primary greenhouse gas on planet earth is water vapour, which is very abundant near ground level, but declines extremely rapidly with rising altitude. The reason for this is twofold - firstly, the temperature declines with altitude at around 6-10K per kilometer, and secondly, the saturation vapour pressure declines with temperature - around 7% for each 1K. As you go up towards the stratosphere, the WV content becomes less significant and CO2 becomes the dominant GHG. The effect is that CO2 "sits" on top of the WV layer and absorbs LWR emerging from the WV layer.
    It is the thickness of the greenhouse gas layer, or put another way, the altitude at which LWR escapes into space, that determines the greenhouse effect.

  20. Jun 24, 2010 #19
    T, would you explain how "the altitude at which LWR escapes into space" is the thing which "determines the greenhouse effect"?
  21. Jun 24, 2010 #20
    Molecules do reflect light. The sky wouldn't be blue if atmospheric molecules weren't around to scatter photons (elastic-Rayleigh scattering). And if molecules didn't inelastically scatter photons, then we couldn't probe their vibrational modes with Raman spectroscopy.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook