Aneis - 12-11-2010 at 16:32
First of all I would like to say this is my first post, so don't be hatin'.
I am currently writing my extended essay for the IB course in chemistry about the emission spectra’s of various fluorescent molecules. I have done
several experiments regarding Luminol with different catalysts such as Cobalt Nitrate and Potassium hexaferricyanate and am about to move on to the
diphenyl oxalates.
When checking Wikipedia's article about glow sticks (DPO/CPPO is the photon emitter there) I found something quite interesting and I quote:
"After activation, the glow sticks gradually shift their emission spectral distribution somewhat towards red."
Following up the source of the statement I ended up at an American army report for potential uses of glowsticks in cockpits when using night vision as
its spectrum did not interfere with them as much as normal diodes.
The link can be found here http://www.dtic.mil/cgi-bin/GetTRDoc?Location=U2&doc=Get...
On page 4 Figure 3 we can see the emission of the glowstick after 50 minutes and furthermore after 92 minutes has shifted somewhat towards infrared at
least in parts of the regions of the spectra. As I am not an advanced organic chemists I can find no exact explanation for this but my best theory is
that the fluorescence resonance energy transfer (FRET) between the diphenyl oxalate ester and the fluorescent dye over time becomes less energetic as
the molecules are further and further apart (the concentration goes down as the reagents are used up). Interesting to note is that seemingly only the
right sight of the spectrum shifts while the maximum still is at a fairly constant 525 nm as well as the “towards UV” part of the graph, also the
minimum point at aprox. 640 nm doesn’t seem to have changed towards red, this might be due to the emission range of the fluorescent dye is at its
max here.
Please note that these are just speculations and I would be more than delighted to here some more opinions on the matter whether you are experienced
organic chemists or not.
PS.
(1)
Please note that I have performed similar emission studies for glowsticks myself and can confirm that what can be seen on figure 3 is what happens for
almost all (all 8-12h I’ve tried) glowsticks.
(2)
If any moderator or administer consider this thread better in the organic chemistry I don't mind moving it though the focus of this topic is mainly
luminescence and not organic chemistry.
DS.
[Edited on 13-11-2010 by Aneis]
DDTea - 13-11-2010 at 02:05
I don't think organic chemistry is where you'll find answers to this question. This has more to do with physical chemistry/quantum mechanics or
Molecular Orbital Theory.
According to Fig. 3 in the paper you referenced, the maximum emission wavelength doesn't appear to shift appreciably at all. In fact, the entire
range of wavelengths appears to remain constant. Some longer wavelengths definitely become more prominent, though.
With that in mind, I would say the primary electronic transition remains the same. The slight shift in the emission maxima could correspond to
rotational/vibrational relaxation of the excited state prior to emission of the photon. That could also explain the increased prominence of longer
wavelengths with the progression of time. It would be really interesting to see a higher resolution spectrum where the vibrational structure is more
apparent.
Aneis - 13-11-2010 at 05:28
Oh, so by referring to different vibrational relaxation states you mean that the photon can be at different levels of excitement in the molecules,
would these levels be equivalent to the different electron levels in an atom?
Would this occur due the molecule colliding with other molecules and thus loses energy that way, because then one would assume to see more "towards
red" emission in the beginning of an reaction rather than in the end (if it does not collide with the by-products formed from the reaction), or due to
other factors?
Unfortunately I don't have any emission spectra on this computer at the moment but instead at an university where I perform the experiments. I will go
there this Tuesday so hopefully I will be able to attach them here then.
DDTea - 13-11-2010 at 05:50
If you can, I'd toy around with the bandwidth of the monochromators on the instrument. They should be controllable through the software. A narrower
bandwidth gives higher resolution. However, in the liquid phase, there are a lot of sources of band broadening that can reduce the resolution of the
fine rotational and vibrational structure. Really, for that sort of resolution, it's best to do it in the gas phase--unfortunately, that isn't always
feasible!
But yes, you pretty much got it: on top of every electronic level, there are vibrational energy levels. On top of every vibrational energy level
are a series of rotational energy levels. To some extent, they almost form a continuum--almost. In any case, higher vibrational energy levels can
relax by transferring energy through collisions to the matrix (that is, everything else in solution!); that is, they heat their surroundings. If a
photon is emitted prior to this vibrational relaxation, it will be of slightly higher energy (and thus, lower wavelength) than if it emitted a photon
after relaxing.
I am just speculating now, but one explanation why the "redder" transitions are more prominent later in the reaction could be due to intersystem
crossing (i.e., phosphorescence--look up Jablonski diagrams for an illustration of this). Electrons would undergo a spin forbidden transition with a
longer lifetime and subsequently relax from a lower-energy excited state. This process has a significantly longer lifetime than fluorescence. As
such, they may contribute to the overall emission spectrum as reaction time increases.
I hope that makes sense because I'm incredibly tired.
Ozone - 13-11-2010 at 07:35
Phosphorescence is extremely unlikely in the liquid state--the fluor molecule is *much* more prone to de-excite via collisional relaxation resulting
in loss as heat. For this reason, many fluors (fluorescein, napthalene, etc.) are also phosphors when frozen (e.g. 77 K) or otherwise dispersed in an
amorphous (glassy) solid state (particularly an ionic one, e.g. boric acid glass, see Lewis et al, 1941).
Hayer (2002) presents a nice overview of the phenomenon while Shulman and Walling (1973) point out phosphorescence in gas-phase (the collisional
frequency of the molecules is comparatively small due to low density) and solid substrates, e.g. TLC plates and paper.
Delayed fluorescence has been observed in liquid state (Parker and Hatcher, 1962), but the phenomena is rare, the time constants are small and
emission intensity weak. In any case, any radiative transfer of energy (that might be observed) from the triplet state (in both liquid and
solid-state) will likely be shunted to delayed fluorescence by thermal excitation. This gives the same emission wavelength as the S1 fluorescent
deexcitation (which is at odds with the red-shift) albeit with a loss of intensity to heat with time constants identical (or very nearly so) to the
classic triplet deexcitation (phosphorescence).
This is not to say that the electrons may not make it into the triplet manifold. From there, they are quite reactive and photochemistry is likely to
take place. Be it one-shot photobleaching or the formation of dye dimers (for example) which may emit at longer wavelengths, the accumulation of such
products will become more likely as the time-of-reaction becomes long and may account for some of observed bias to longer emission wavelengths.
Cheers,
O3
Attachment: Hayer et al 2002 Delayed Fluorescence and Phosphorescence.pdf (199kB)
This file has been downloaded 1080 times
Attachment: Schulman and Cheves ! Walling 1973 phosp adsorbed ionic organic molecules.pdf (479kB)
This file has been downloaded 667 times
Attachment: Parker and Hathard 1972 T1-S1 emission in fluisd.pdf (742kB)
This file has been downloaded 699 times
Attachment: Lewis et al 1941 JACS photochem proc in rigid media.pdf (1.4MB)
This file has been downloaded 574 times
Aneis - 13-11-2010 at 07:39
I don't have the equipment available to perform the emission analysis in gaseous state. This doesn't really matter though as it is a comparison
between different colours and intensities of glowsticks that I am looking at and not exact values, and I guess the organic solvent (which is the same
will interfere equal in each experiment). It is possible to change the bandwidth of the monochromator, however even without, using the "zoom" function
on the program one can see a clear shift in this part of the spectrum.
Just one question about the vibrational and rotational energy levels; if a photon of UV light at say 350nm is emitted by the decomposed diphenyl
oxalate ester and then absorbed by the fluorescent dye, will the light be emitted by it be random (of course inside its range) or at a certain
wavelength.
Taking for example 9,10-diphenylanthracene which emits blue light between 400nm - 550nm (with a peak at aprox. 440nm) would a photon absorbed at say
350nm give of light at higher wavelength than one of 330nm i.e is there a relationship between the light absorbed and the light emitted?
Phosphorescence might be the key to the shift in spectrum. Looking at a Jablonski diagram we can see that are some delayed fluorescence in the form of
phosphorescence, the question however is how much of this occurs and if it even occurs in chemiluminescence at all or just in normal fluorescence (a
quick search on google for "phosphorescence in glowstick/chemiluminescence etc.” didn’t give any results).Despite this it actually makes quite a
lot of sense compared to my theory!
Edit: seeing Ozone's post makes this again less likley. A quite large amount has to be emitted trough phosporesence to be able to affect the emission
spectrum in such an extent as done in Fig. 3
Regarding photobleaching it is important to note that it is not a laser used to excite the fluorescent dyes but rather the absorption of higher energy
photons. That the light emitted by the solution itself would decompose some of the fluorescent dyes seems to me rather unlikely (though this of course
depends on the stability of the compound).I might have misunderstood what you meant here, so please correct me if I am wrong.
I see also a problem in the formation of dimers. First of all only very small changes of the structure of molecule is needed to drastically change the
emission (for example, many flourescent dyes are anthraces with just some minor things such as the position of chlorides that differs). Adding a
complete (though same) new molecule would drastically change the emission spectrum and would be much more noticeable on the spectrum. Also one would
see a shift at the right minimum point of the spectrum at aprox. 640nm if this was to occur as the range of flourescence would become larger (if not
the dimer formed would emit light at lower wavelength in which case the we would see a shift at the left minimum point).
[Edited on 13-11-2010 by Aneis]
Aneis - 23-11-2010 at 03:22
Does anyone else have any ideas/opinions on why the spectrum shifts?
Aneis - 27-11-2010 at 08:38
Actually never mind, I think I got my answer looking trough the Parker and Hathard article regarding delayed fluroescence. Thank you very much both
DDTea and Ozone for your help!
Also, sorry for triple post 