Sciencemadness Discussion Board - hydrogen susceptibility to sunlight question

hydrogen susceptibility to sunlight question

math - 24-8-2012 at 03:43

Hello,

wikipedia states that

Hydrogen gas forms explosive mixtures with air if it is 4–74% concentrated. The mixtures spontaneously explode by spark, heat or sunlight.

So if I was about to fill a balloon with hydrogen gas made by the reaction of NaOH with aluminium in a glass bottle, by directly attaching the balloon at the neck of the bottle, does it mean that inside there will be stored an air-hydrogen mixture and that exposing said balloon to sunlight it'll explode instead of just flying up?

What'd be a simple fix to this issue, if present? Should I let some of the reaction going, and then attach the balloon to the bottle, so that the air has been purged out?

Thank you

woelen - 24-8-2012 at 05:37

I have had hydrogen/oxygen mixes in sunlight many many times and never had an explosion. I think that this is one of Wikipedia's bad errors. I know of H2/Cl2 mixes which can explode in sunlight, but I've never heard of this property for H2/O2 mixes.

vmelkon - 25-8-2012 at 08:10

I have made hydrogen and air mixtures. I had to apply a spark to cause it to explode. Besides, I think if you do it right, you can have your balloon with more than 95% hydrogen.

ChemistryGhost - 25-8-2012 at 14:04

Maybe Wikipedia confused hydrogen molecules with hydrogen ions.

AJKOER - 27-8-2012 at 07:14

My research suggests that Hydrogen is most easily ignited with 29% H2 in air (H2/O2 ratio 2 to 1), but explosion occurs at 41% H2 in air (H2/O2 at 3.5). The lower explosion limit in pure O2 is 10%, but ignition is, in general, more difficult at LEL (and, in my opinion, most likely lower than 9/1 in sunlight). Source: http://arhab.org/pdfs/h2_safety_fsheet.pdf ). In fact, another web thread, one contributor suggests in sunlight (no documentation) an 8 to 1 ratio (Hydrogen to pure oxygen).

Perhaps, before calling Wiki in error on this point, someone should place a calculated amount of NaOCl, NaOH, Aluminum and H2O2 (added last) in several partially crushed one liter perfectly clear plastic bottles (to allow various amounts of H2/O2 mixtures) in direct (and/or diffused) sunlight. Allow several days for a reaction/explosion to take place.

Intended preparatory reactions:

NaOCl + H2O2 --> NaCl + H2O + O2 (g)

2 Al + 2 NaOH + 2 H2O → 2 NaAlO2 + 3 H2 (g)

[Edited on 27-8-2012 by AJKOER]

unionised - 27-8-2012 at 08:21

Neither hydrogen not oxygen has any absorptions that are accessible to visible light or even to the UV components of sunlight.
Light that's not absorbed doesn't have an effect.

http://en.wikipedia.org/wiki/Photoelectrochemical_processes#...
So, wiki says that wiki is wrong.

Incidentally, if I were planning to demonstrate this I'd avoid using hypochlorite because adding chlorine (whic is known to be photochemically active) would mess up the results.

AJKOER - 27-8-2012 at 09:18

To avoid the possible argument relating to small amounts of photochemically active Cl2O or Cl2 from the Bleach initiating any explosion, I agree it would be better to use MnO2 and H2O2 as the oxygen generator, else one would have to repeat the experiment in the event of a positive result without the use of NaOCl.

In my opinion, the main question is whether a single hydrogen bond can be loosened by sunlight to start a chain reaction with O2 resulting in an explosion.
__________________________________

{EDIT] I just find a reference, exploring the explosive chain reaction of H2 and O2. The reference states: "Book considers the example in which the initiation reaction is photo initiated: H2 --> 2 H rate=v=intensity of radiation=I...."

See page 11 at link: https://docs.google.com/viewer?a=v&q=cache:13Gw2yMjahIJ:www.southalabama.edu/chemistry/barletta/notes_ch26.ppt+Hydrogen+Oxygen+explosion+ratio+'ch ain+reaction'&hl=en&gl=us&pid=bl&srcid=ADGEESjn-2m4R3wlmjkJnXNLr0J5FtTE0AnodggTbJCKd7uuIfAS0_tNHZ4wsSOiqXb1qKFUdlBhOkgUAc42rT5D11D4AWd 6JpoGJy6USWujq3JRfxfL9prx4qRmwKnzOkVY-g0oNRom&sig=AHIEtbRtaIfoCdd7YgPYPgyYbZ28zmBHaQ

However, in my opinion, this summary homework reference to "Chapter 26: Chain Reactions" of the unspecified book does not provide a definite yes answer as the Termination steps (in the H2/O2 chain reaction) may be significant.

[Edited on 27-8-2012 by AJKOER]

unionised - 27-8-2012 at 09:39

Well, I agree with this bit
" the main question is whether a single hydrogen bond can be loosened by sunlight to start a chain reaction with O2 resulting in an explosion."
and I don't see how the answer can be yes.

People use deuterium lamps as a source of UV these days because they are brighter, but they used to use hydrogen lamps.
The UV (at least down as far as 185nm) escaped from the lamp.
So the hydrogen couldn't be absorbing it.
So we know that hydrogen doesn't absorb UV with a wavelength more than 185 nm (it could be smaller).

Oxygen is a slightly better candidate.It does have some absorptions at longer wavelength but they lead to a triplet state rather than to dissociation and they are very weak (for most cases you can ignore them- oxygen isn't green.)
The UV cut off for dissociation is about 195nm.

Now lets think about sunlight.
It comes from the sun with plenty of hard UV.
But - here's the point- any wavelengths short enough to be absorbed by O2 will be absorbed by O2 in the atmosphere before they get to the ground.

So, no light that's absorbed by oxygen reaches us (because the atmosphere would have absorbed it already)
And no light that could be absorbed by hydrogen could get through, because it's even more readily absorbed than the light that would split oxygen.

So, the light that we see isn't absorbed by either gas.
How can it have an effect?

Morgan - 27-8-2012 at 09:45

Static occurrence
http://www.h2incidents.org/incident.asp?inc=281

[Edited on 27-8-2012 by Morgan]

AJKOER - 27-8-2012 at 09:46

Unionised:

I had to amend my comment based on a recent find.

I am now more in the Yes camp, but not completely.

unionised - 27-8-2012 at 09:55

Morgan,
Good point.
Whatever the pros and cons of the sunlight might be, static is a serious threat.

AJKOER - 27-8-2012 at 11:10

OK, I found the reference I alluded to previously. If correct, it does shed light on the H2/O2 photo initiated chain reaction. To quote Jon Richfield:

"If you put nice, cold, dry, clean O2+H2 in a clean glass vessel in the dark under low pressure, with no catalysts and radiation anywhere, you might have to wait a few lifetimes for much H2O or H2O2 to show up, let alone a major reaction, flame or explosion.
However, the more you increase the temperature and pressure and radiation, the more spontaneous reaction proceeds, and the greater the prospect of ignition. Sprinkle a bit of Pd powder, or even put in a strip of Pd metal, and you can be pretty sure of reaction, in fact you might be asking for trouble.
Just plain dry O2 + H2 in bright sunlight will react slowly (H2+Cl2 would explode) and O2 + H2 at 100C to 200C also react slowly, but significantly.
Get the picture? A spark or an accident could set it off, but there is slow reaction under a wide range of circumstances. If you are talking about storing the mix together for geological timespans, you might as well label your stores as H2O.
You will be right for almost the whole time."

Link: http://www.last-word.com/content_handling/show_tree/tree_id/...

There is also a reference (see http://jcp.aip.org/resource/1/jcpsa6/v7/i8/p710_s1?isAuthori... ) that supports the non-explosive composition reaction. To quote: "The kinetics of the upper explosion limit and the nonexplosive thermal reaction of H2 and O2 uniquely demonstrates the formation of HO2 in collisions H+O2+M, its destruction on surfaces and its ability to propagate chains by gas phase reactions with H2. Below 500° HO2 gradually ceases to be a chain carrier in the gas phase; assuming its further reactions to be heterogeneous, involving self‐neutralization and reaction with H2, the kinetics of the mercury‐photosensitized H2☒O2 reaction are well interpreted." See also: http://pubs.rsc.org/en/content/articlelanding/1958/tf/tf9585...

From Jon Richfield, my take is that a photo-initiated chain reaction may proceed slowly forming H2O and usually not explosively. However, the reaction mix may be more sensitive especially with increased pressure and heat to, for example, static electric denotation and the presence of metals.

Personally, I am comfortable with this scenario and how it integrates known accidents with comments made. The answer, in my opinion, is a slightly cautious no with respect to the possibility of explosion.

[Edited on 27-8-2012 by AJKOER]

unionised - 27-8-2012 at 12:49

My take on it is that Jon Richfield is clearly talking through his hat if he thinks ordinary sunlight, here at ground level will do that.

In the presence of mercury (or chlorine) the reaction is plausible in sunlight; but oxygen+ hydrogen don't absorb daylight so it can't affect the rate of reaction. It's slow at room temperature- possibly so slow that initiation by cosmic rays is faster than the thermal pathway but the idea that light which isn't absorbed, somehow supplies initiation energy is, lets be clear about this,
A breach of the conservation of energy.

Of course, if you are talking about some putative reaction during the creation of the earth - and, in particular, in the higher reaches of the atmosphere where there's lots of hard UV it's a different story. But that's got little or nothing to do with the OP

[Edited on 27-8-12 by unionised]

watson.fawkes - 27-8-2012 at 14:33

There's a lot of "inconceivable" in this thread.

Quote: Originally posted by unionised

Neither hydrogen not oxygen has any absorptions that are accessible to visible light or even to the UV components of sunlight.

That's not true. The transition between triplet (ground state) and singlet oxygen is in the infrared. While this isn't the best example (forbidden transition in isolation, i.e. in gas phase), it's illustrative of the transition energies between molecular orbitals. These lower energies, however, do not to bond dissociation nor even ionization, but rather to excited molecular states. Almost all of these excited states are radicals, with lone electrons in some previously-unoccupied molecular orbitals. The transition of H₂ to its first excited state, with one electron in the σ^* state, is at 109 nm, in the UVC range. I couldn't quickly locate the data for the analogous energy for the excitation in O₂ for its transition π_2p^* &rightarrow; σ_2p^*, but it's a longer wavelength than that.

Furthermore, radiation absorption isn't the only way to get excited states. There's a certainly steady-state population of excited states from the distribution of velocities and occasional molecular collisions at high energies. There's more of these at higher temperatures, so it's unwise to assume that the kinetic chain that could initiate an explosion is the same one that happens at a shock front.

Quote: Originally posted by unionised

So, no light that's absorbed by oxygen reaches us (because the atmosphere would have absorbed it already)
And no light that could be absorbed by hydrogen could get through, because it's even more readily absorbed than the light that would split oxygen.

So, the light that we see isn't absorbed by either gas.

This argument is utterly bogus, because every absorption line is also an emission line. In the atmosphere, molecules do absorb some of the radiation, whereupon they go into an excited state. They then quickly emit radiation, returning to the ground state.

arsphenamine - 28-8-2012 at 15:49

Direct transition from triplet to singlet is up against a large energy barrier:
singlet ¹O₂ is ~95 kJ/mol higher in energy than ³O₂.

The explanation is quantum chemical, involving selection rules for spin, parity, symmetry, etc.

¹O₂ generation is usually catalyzed by a metal ion or a dye.
The catalyst may be biogenic/endogenous.
Yeah, you over there with the blue eyes; I'm tokkin ta you.

watson.fawkes - 29-8-2012 at 10:36

Quote: Originally posted by arsphenamine

Direct transition from triplet to singlet is up against a large energy barrier

Yeah, I hinted at that with the phrase "forbidden transition". It's the reason that singlet oxygen has such a long lifetime in the gas phase. For such a transition to occur, it (less or more) has to do an internal-double transition, which is a very much less likely occurrence. The lifetime is around six or more orders of magnitude larger when it's an isolated molecule in the gas phase than when it's interacting with other molecules of other species.

My point, however, was to estimate the transition energies of O₂ excited states, and the singlet state transition gives a lower bound. Since it's in the infrared, it fails to provide a lower bound that's above the visible band. Indeed, since the first H₂ excited state operates at lower principal quantum numbers, and thus at higher transition energies, it's a pretty good guess that all or most of the O₂ electronic transition energies are somewhere in the bands between infrared and UVC, that is, right where solar output is good.

If I had proper MO data, I would have preferred to have just cited it.

AndersHoveland - 29-8-2012 at 11:54

I have never read anything about sunlight being able to ignite mixtures of hydrogen and oxygen.

But shorter frequency UV light does cause a photochemical reaction that can lead to ignition if there is enough intensity of UV light.

Quote:

A sharp wavelength dependence on the amount of incident laser energy necessary to ignite a premixed flow of H2/O2 at atmospheric pressure is reported. This wavelength dependence exhibits a spectral profile similar to the two-photon fluorescence excitation curve for flame oxygen atoms, and the respective peaks correspond to exactly the same wavelength near 225.6 nm. This similarity clearly indicates that oxygen-atom production and subsequent excitation is an important step in the efficient (∼0.5-mJ) laser ignition of H2/O2 flows in this wavelength region. In addition, the dependence of the incident laser energy on the equivalence ratio reveals that the most efficient ignition occurs far into the fuel-lean region

Platinum gauze can also catalyze a reaction between hydrogen and oxygen at room temperature. During the reaction, the platinum can reach an incandescent heat, at which point it can ignite the gas mixture. It takes a long time for the reaction to proceed to the point of ignition, so for demonstration purposes the platinum is always heated before being placed into the stream of hydrogen. As the catalytic action becomes much faster when the platinum is heated, this greatly reduces the necessary time it takes for the platinum to reach a glowing heat.

[Edited on 29-8-2012 by AndersHoveland]

math - 29-8-2012 at 16:33

Thank you all for the explanations

Thinking about the reaction between aq. NaOH and Al, I was thinking if adding a large excess of water (possibly as heat moderator in the exothermic reaction) would impair the amount of H2 produced.

I think it won't, given H2 extremely low solubility in H2O, but my main concern is that the reaction would proceed so slow that the amount produced is lost through the balloon rubber at the same rate.

Maybe it's overthinking, but I'd like to use another refrigeration method other than ice.

Thank you

arsphenamine - 30-8-2012 at 06:01

Quote: Originally posted by watson.fawkes

My point, however, was to estimate the transition energies of O₂ excited states, and the singlet state transition gives a lower bound.
...
If I had proper MO data, I would have preferred to have just cited it.

What is proper MO data for you?

This is interesting stuff.

(edit)
good gloss on ³O₂ at Chemogenesis web book

[Edited on 30-8-2012 by arsphenamine]

watson.fawkes - 30-8-2012 at 09:38

Quote: Originally posted by arsphenamine

What is proper MO data for you?

Well, to start with, one with all the numbers in place, principally energies for each molecular state. When I was looking this up, I saw plenty of MO energy diagrams that were qualitative in nature, intended to teach concepts, particularly the difference between bonding and anti-bonding orbitals. (This includes the link you cited, which I saw in my earlier search.) A proper diagram would have an upper bar at the top that represents the ionization potential, conventionally set at zero, yielding the convention that the energy of any particular orbital state is a negative number. In addition, there would be a number of orbital states that are unoccupied in the ground state but are present for excited states, usually at least two or three such. In addition, atomic spectroscopy diagrams tend to group the angular momentum quantum number in columns, so that it's easier to see some the selection rules.

Now I did find this page of physical data from NIST, but I couldn't make much sense about it in a few minutes. There's lots of literature citations, but I don't have immediate access to the literature, and I haven't spent the time working through it. This one shows some promise about being a decent summary of the observed data: Krupenie, P.H., The spectrum of molecular oxygen, J. Phys. Chem. Ref. Data, 1972, 1, 423.

unionised - 30-8-2012 at 12:27

It seems a little knowledge is a dangerous thing.
"That's not true. The transition between triplet (ground state) and singlet oxygen is in the infrared. "
Kind of. There are two such transitions. About 759 to 765 nm and 687 to 692nm.
Something with a significant absorption at those wavelengths would be green.
You would think someone might have mentioned that.
Oh!, hang on. I did.
Then there's the other thing about them. As Watson helpfully points out.
"These lower energies, however, do not to bond dissociation nor even ionization"
So, once you fill in the missing word, that wraps it up for them being a contributor to the initiation of a free radical reaction. (Unless you are talking about two photon events, but those are so rare that the generation of radicals from cosmic rays would outpace them.)

So far so average.
"These lower energies, however, do not to bond dissociation nor even ionization, but rather to excited molecular states. Almost all of these excited states are radicals,"
Not exactly.
The ground state for oxygen is a di radical.
The excited states (the singlet ones) are not radicals.
To be fair, you would have been right for most diatomic molecules.

Then there's "the transition of H2 to its first excited state, with one electron in the σ* state, is at 109 nm, in the UVC range. "
Well, OK I agree with that- hydrogen doesn't do a lot photochemically until you get to 109nm.
I had pointed out that the first absorption must be below 185nm but I hadn't cared how much further down the wavelength scale it was because there's no UV that short in sunlight.

"Furthermore, radiation absorption isn't the only way to get excited states. "
Quite right!
"There's a certainly steady-state population of excited states from the distribution of velocities and occasional molecular collisions at high energies. "
Indeed!
Yes, you can get significant populations of higher excited states by heating the gas.
But that's relevant to the question of "can I set off Hydrogen and oxygen with a lit match?"
If there were enough thermally generated free radicals then the reaction wouldn't wait for the sunlight- the materials would ignite on mixing.
That's what happens with fluorine and hydrogen. The F F bond is weaker so there's a much better chance of thermal energy breaking the bond and starting the reaction (there's also a bigger energy release to help propagate the reaction)

There's another point about thermal energy. It makes molecules bump into one another.
So that brings us to "...every absorption line is also an emission line. In the atmosphere, molecules do absorb some of the radiation, whereupon they go into an excited state. They then quickly emit radiation, returning to the ground state."

Close, but no cigar. There's another fate for excited states.
Collisional deactivation is the process where two molecules (one in an excited state) collide and bounce of each other but they leave faster than they came in, and the energy of the excited state is transfered to kinetic energy (and vibrational+ rotational).
That's quite common.
The laws of thermodynamics mean that the sun (which is quite hot) wants to warm up the (rather cooler) atmosphere, and so it does.

It means that if there's a transition that can absorb light it does so overall.
That's why we are screened from UV by the ozone layer. It's why hard UV is sometimes called vacuum UV and it's pretty much the same reason that people talk about the greenhouse effect. It's even why those wavelengths in the visible/ near IR are demonstrably missing from sunlight (ask Mr Fraunhofer). They are not strong enough transitions to make a difference in a balloon full of gas, but if you have an entire atmosphere of them you can see them.

What it isn't, is "Bogus".

AJKOER - 30-8-2012 at 17:09

Quote: Originally posted by unionised

My take on it is that Jon Richfield is clearly talking through his hat if he thinks ordinary sunlight, here at ground level will do that.

For the record, Jon Richfield is a teacher with a specialty in Entomology.

I did like his ability to translate some of the literature, obviously outside his realm of specialization, an attribute of a good teacher at that.

watson.fawkes - 30-8-2012 at 18:19

Quote: Originally posted by unionised

The excited states (the singlet ones) are not radicals.

Apparently you don't know what a radical is when you get away from the standard examples. All it means is unpaired electrons. The excited state of O₂ where the two highest-energy electrons occupy (1) one of the two π^* MO's and (2) the σ^* MO is, indeed, a radical state. Both electrons are unpaired in their respective MO states. It's also a diradical, since there are two unpaired electrons. Unlike the diradical of the ground state of triplet oxygen, this diradical state does not have the same angular momentum degeneracy. It's a reactive state, much more so that ground-state O₂.

What you were claiming before is that light not absorbed by the atmosphere can't be chemically reactive. That claim is bogus.

Quote: Originally posted by unionised

The state of oxygen I mentioned has transition line to the ground state somewhere in the visible or near UV. (I wish we knew the figure; it would clarify a few issues.) That transition line does not necessarily lead to a strong atmospheric absorption band, however. While there are other ways of discharging an excited state to ground state, such as collisions, it does not mean that every such excited state has a fast enough alternative to emission that would lead to an atmospheric absorption band. A bulk absorption band requires that there by some other pathway, and that pathway must be fast enough to outcompete the natural emission rate (see Fermi's golden rule) enough to see some bulk absorption instead of mere re-emission.

One thing that's should be clear at this point is that if sunlight can ignite H₂-O₂ mixtures, it's not going to be mediated by an excited-H₂ state. The transition line is in the upper end of the UVC band, and there's no significant UVC flux (at sea level) that would lead to direct excitation. The excited states of O₂ are low enough energy that they aren't going to excite H₂ by a collision process. (To be sure, the rate isn't exactly zero, but if you want to compute the full partition function for this system and provide an upper bound on the rate, be my guest.) So I would have to say that excitation of O₂ to the state I named above seems like the best candidate at this point for the initiation point of a plausible mechanism.

Just to be complete, I would hardly guess that sunlight is capable of igniting mixtures reliably, nor at every concentration of the explosive range. In addition, I would have to guess that details of weather, latitude and altitude, all of which bear on atmospheric attenuation, are also at play. Nor do I omit the possibility that the reports have omitted the use of solar concentrators, in which case it could simply be a temperature effect as reported.

arsphenamine - 31-8-2012 at 08:24

General notes on O2 --

The ¹O₂ emission spectrum is typically reported as 1270 nm or 7875 cm^-1, putting it in the near infra red region, typically where we see only overtones of organic compounds (if at all).

The ³O₂ emission spectrum is reported as 680 nm or deep red and visible.

Oxygen pressure measurement using singlet oxygen emission

The best single reference I can find is
Physical Mechanisms of Generation and Deactivation of Singlet Oxygen
Schweitzer, Claude; Schmidt, Reinhard
Chemical Reviews, 2003, 103 (5), pp 1685–1758
DOI: 10.1021/cr010371d

The DOI is incorrectly reported by Wiley&Sons, and you should look on the ACS site first.

In more practical terms, you need a photosensitizer to generate singlet O2. Porphyrins generally work, especially those with a bound platinum in the center, but iron, magnesium, or copper will do. Fe,Mg,Cu - porphyrins correspond to heme, chlorophyl, and octopus bloods.

[Edited on 31-8-2012 by arsphenamine]

unionised - 31-8-2012 at 09:56

The ground state of oxygen is a radical, and, at ordinary temperatures it doesn't react with hydrogen.
Absorption of visible light can produce the singlet state but that doesn't practically cause a reaction for two reasons.
Firstly the transition is forbidden. If I shone a beam of light at the right frequency through a balloon full of oxygen practically all of it would go straight through.
A very small fraction of it would be absorbed.
Oxygen still isn't green.

Secondly, the singlet state also doesn't react with hydrogen. It would , in due course lose that energy- probably by collision.

Now the transition is very weak but it does happen.
If you have a broad band source- like the sun and you look at the spectrum through a whole atmosphere's worth of oxygen, the absorption does matter. There's not much left of the relevant wavelengths.
So the wavelengths of light that might get absorbed (to some small extent) but which don't cause the reaction (because the singlet state is also inert to hydrogen) have already been largely absorbed and are not there (as witnessed by Mr Fraunfofer).

OK I forgot that the singlet state is also a radical- but the point remains it's a poor absorber on a balloon sized scale and a much better one on an ocean of air sized one so there's not much chance of it triggering a balloon full of hydrogen and oxygen because it's not there.
Even if it was there it doesn't produce a reaction.
I know that the singlet state has some more accessible absorption bands, but then you are talking about a two photon process and, as I pointed out, those are rare. So rare as to be unimportant compared to, for example cosmic ray interactions.

As for "What you were claiming before is that light not absorbed by the atmosphere can't be chemically reactive. That claim is bogus"
Light that isn't absorbed can't cause a reaction.
http://en.wikipedia.org/wiki/Stark%E2%80%93Einstein_law#Star...

Why do they call hard UV "vacuum ultraviolet"?
Why is that radiation not present in sunlight?

[Edited on 31-8-12 by unionised]

math - 1-9-2012 at 16:48

Quote: Originally posted by math

Thank you all for the explanations

Maybe a better questions would be:

how can I calculate the quantity of water at X temperature to be used to keep the whole reaction at, say, 50°C maximum?

Thank you

watson.fawkes - 2-9-2012 at 08:09

Quote: Originally posted by unionised

The ground state of oxygen is a radical, and, at ordinary temperatures it doesn't react with hydrogen.
Absorption of visible light can produce the singlet state but that doesn't practically cause a reaction for two reasons. [...] Secondly, the singlet state also doesn't react with hydrogen. It would , in due course lose that energy- probably by collision.
[...]
OK I forgot that the singlet state is also a radical

Ignorance confirmed.

There are lots of states of oxygen, not just "triplet" and "singlet". Since you have persisted in your lack of self-knowledge of your ignorance, I went and found data. For a rather complete list, see Krupenie, P.H., The spectrum of molecular oxygen, J. Phys. Chem. Ref. Data, 1972, 1, 423. There's a copy on the NIST site and a reference threadlet in References here. Table 2 of that paper lists nineteen (19) of them. That table also lists O₂⁺ states, which are relevant to O₂ since they correspond to "Rydberg" states, where one electron is highly excited but not quite ionized.

Furthermore, even restricting attention to the lowest states of oxygen, there are two singlet states, not just one. Quite apparently you have not bothered to read the Wikipedia page on singlet oxygen, which I referenced before, so there's the link—read it this time. You wouldn't persist with saying "the" singlet state consistently if you had done so earlier. That page shows both of these singlet states quite well, along with the (reduced) spectroscopic symbols. O₂ has a pair of degenerate 1π_g orbitals. There are three occupancy modes of two electrons in these states; one is the ground state triplet, the other two are the singlets. And these three all have the same occupancy signature in the MO when you ignore degeneracy. That signature is (3σ_g)²(1π_u)⁴(1π_g)².

The excited states I was referring to upthread have occupancy signature (3σ_g)²(1π_u)⁴(1π_g)¹(3σ_u)¹; there are two such states. It turns out, though, that there are two other signatures: (3σ_g)²(1π_u)³(1π_g)³, with six states, and (3σ_g)¹(1π_u)⁴(1π_g)²(3σ_u)¹, with eight states. I'm still plowing through that reference paper. It's 120 pages long, filled with results, using notation I'm not facile with. Nevertheless, there are lots and lots of transitions in this molecule in the visible and near UV. Many are forbidden transitions with small cross sections, albeit, but apparently not all are. There are fourteen subsections on the various spectroscopic systems (roughly, related transition in some band), and hundreds to thousands of individual lines. I would like to think that I have enough humility that I won't call a potential effect "inconceivable" before I have even a basic understanding of the system I'm thinking about.

Quote: Originally posted by unionised

Light that isn't absorbed can't cause a reaction.
http://en.wikipedia.org/wiki/Stark%E2%80%93Einstein_law#Star...

You are, yet again, confusing atomic absorption with bulk gas absorption, and also confusing necessary conditions with sufficient ones. Atomic absorption is a necessary condition for bulk gas absorption. It is not always sufficient. If, for example, the natural decay lifetime of a state is significantly shorter than the transit time across the mean free path of the molecule, then you can have an atomic absorption line without a significant bulk gas absorption line (they'll pretty much always be detectable with sensitive instruments). I do not know whether this pertains to molecular O₂ or not, but I am quite sure that conflating these two concepts is bogus.

Quote: Originally posted by unionised

It seems a little knowledge is a dangerous thing.

Yes ... yes it is.

unionised - 2-9-2012 at 13:22

A few weeks ago I was at the Kennedy space centre.
They have lots of stuff there for tourists to gawp at and helpful tour guides to answer questions.

He looked bored so I pointed at the rocket engine that they had on display and asked if he could point out the spark plug.
He couldn't- but he agreed that there must be one somewhere because mixtures of hydrogen and oxygen don't spontaneously ignite.

He did point out that there are flares lit in the vicinity of the rockets to light any leaks before they build up and become a bigger problem (most things in the vicinity of a space rocket are built to stand up to a flash fire)

It's often sunny in Florida.

NASA don't think that sunlight ignites air/ hydrogen mixtures.

For the record, I'm fully aware that I don't know the details of the transitions.
I don't think they matter.
Do you, for example, accept that oxygen, as we breathe it, does not have any significant absorptions in the visible region of the spectrum?
I know that there is the pair of transitions to triplet states You may remember that I gave the wavelengths corresponding to both of them which makes this assertion "Furthermore, even restricting attention to the lowest states of oxygen, there are two singlet states, not just one. " a bit redundant. I know there are. I gave their excitation energies earlier when I was pointing out that they are not in the infra red. They are, in fact, in the visible range . That was my point, if they actually happened to any significant extent, then oxygen would be green. It isn't.

When I referred to the singlet state I was referring to the whole bunch of excited states- I didn't bother to specify rotational or vibrational levels either.
I grant you it's sloppy language but when it comes down to it, they don't matter.

Do you think that there are any excited states of oxygen which can be reached from the ground state by a single photon transition involving EM radiation that is present in sunlight at ground level and which react with hydrogen?

BTW, re "I'm still plowing through that reference paper."
Why?
"You are, yet again, confusing atomic absorption"
Bollocks. I never stated what was absorbing: the law works perfectly well with molecules or atoms since it's a restatement of the conservation of energy. I just pointed out that if a photon goes in and out then it can't do anything. Photons don't have a big repertoire: they get absorbed or they don't. If they don't get absorbed they don't have any effect.

"Atomic absorption is a necessary condition for bulk gas absorption."
That must be a real pig for the IR and µwave spectroscopists since they never deal with atoms- only molecules.

"Nevertheless, there are lots and lots of transitions in this molecule in the visible and near UV."
There are indeed, but not many of them start from the ground state which is (fairly nearly) the only occupied state at room temperature.
Specifically there are the two I mentioned (split into many sub levels by vibrational and rotational excitations).
And the interesting things about them are that they are still weak as hell (oxygen still isn't green)
and the radiation needed to induce them has already been seriously attenuated by the passage through a big thick cloud of exactly the right stuff to absorb them- to whit , oxygen.

Those excited states don't react with hydrogen.

BTW, feel free to actually answer the questions I asked earlier.

Here they are again

Why do they call hard UV "vacuum ultraviolet"?
Why is that radiation not present in sunlight?
(I guess I should have specified sunlight near ground level.)

watson.fawkes - 2-9-2012 at 18:19

Quote: Originally posted by unionised

BTW, feel free to actually answer the questions I asked earlier.

Won't bother. You don't want to engage what I've said and repeat yourself. No sense wasting time with you.

Poppy - 2-9-2012 at 18:28

What if the baloon is hit by the stream jet of a very high energty cosmic ray particle/ radiation, that would focus and cause the baloon to explode, maybe?
This is the way I use to get cancer...

arsphenamine - 2-9-2012 at 19:38

Quote: Originally posted by watson.fawkes

Ignorance confirmed.

ditto, speaking only for self.

Quote:

There are lots of states of oxygen, not just "triplet" and "singlet".

So, there I was, thinking I'd casually model singlet vs. triplet total energies at a moderate basis set and theory level.
Typically, without exactly specifying the valence spin states, you get a Δ E that's 10-15 kcal too high.
The less costly correlation methods want to select the highest energy spin orbits or a weighted average of all spin combinations.

After you've specified your spin state, every ab initio QC package out there has its own obtuse nomenclatural conventions.
Yeah, you can do closed-shell computations on an open shell triplet without a peep of complaint from the application.

Quote:

For a rather complete list, see Krupenie, P.H., The spectrum of molecular oxygen, J. Phys. Chem. Ref. Data, 1972, 1, 423. There's a copy on the NIST site and a reference threadlet in References here.

Thanks.

Several NB's ...

The Jablonski Diagram, an agreeable shorthand for spectroscopically-observable transitions between different spin states,
by its very existence confirms that absorption and emission spectra need not be the same or even complementary, viz.,
an absorption spectrum is not an emission spectrum.

It has been silently amended in practice to include extreme rotation, translation, bend+stretch+torsion changes from cavitation states,
given the industrial advent of sonochemical engineering and hydrodynamic cavitation reactions.

Last N.B. -- Some lab reactions produce ¹O₂ with peroxide and hypochlorite, O₂^-2 and OCl^-.
Both species are produced in the upper atmosphere. The chloride comes from freons.

I will accept that, in a pure sense, ambient UV radiation will not initiate O2+H2 combustion.
To my mind, the likelihood of 'adventitious' mediating substances (metal ions, chromophoric triplet sensitizers,etc) re-opens this line of inquiry.

unionised - 2-9-2012 at 22:02

Quote: Originally posted by watson.fawkes

Quote: Originally posted by unionised

BTW, feel free to actually answer the questions I asked earlier.

Won't bother. You don't want to engage what I've said and repeat yourself. No sense wasting time with you.

Simply not true. I quoted you about half a dozen times in my last post alone.