[Markitron]

Hi All,

I have a question about UV curing of Resins if someone wouldn't mind helping me out.

I am trying to find a decent alternative to 2-part epoxy resins, and I wanna try out a UV-cured coating. Finding the resins themselves seems to be easy enough, but I am having trouble finding a light source. It looks like anything more elaborate than a standard blacklight is exceedingly expensive. Would a blacklight output at a high enough intensity to cure a thin coating? The resin requires a wavelength of approx. 360 nm which is about the same as an average blacklight as far as I know.

Thanks in advance.

[Arkcon]

Back in the day, I always directed people to Edmund Scientific, for small, intense short-wavelength UV lamps.  But these days, it seems like small UV LEDs are everywhere.  But do check the wavelength to be sure they're short enough.

[Markitron]

Thanks for the info, I was under the impression that the LED's have a narrower wavelength range and that you can't get them as low as 360 nm.

I'll try the blacklight anyway and see if it works.

[pgk]

I am afraid, not.
You need a special UV lamb that emits bellow 350 nm, with maxima mainly at 200 nm and additional maxima at 300 nm and 170 nm.
Explanation: UV cured resins are hybrid methacrylate polymers that are cured by photoinduced free radical polymerization of the double bond. The so formed isobutyric free radical is stabilized by the carbonyl conjugation and by the proton radical transfer of the methyl group. The corresponding bond energies, are:
C=C        ≈ 150 kcal/mole
C=O        ≈ 180 kcal/mole
C-H         ≈ 100 kcal/mole
By introducing the above values, to the equation bellow (attention to the units):
E = hv = hc/λ
We get:
C=C         λ ≈ 200 ± 15 nm
C=O         λ ≈ 170 ± 15 nm
C-H          λ ≈ 300 ± 15 nm    
The difference ± 15 nm represents the material’s interference to the absorbed wavelength.

[Markitron]

Many thanks for the reply, I'm specifically looking at epoxy resins here which generally cure in the UVA wavelength range. I'm not 100% sure yet but I believe it may cure via cationic catalysis and not free radical catalysis, ie When the cationic photoinitiatior is exposed to UV light a strong acid is released which catalyses the reaction. The reason I think this is the case is that the resin I have is a 'dark cure' and doesn't require constant UV light (which is the case with free radical-induced polymerisation).

As I said I'm not 100% about all of this, but I know it's not based on methacrylate polymers. In these methacrylate systems, what kind of intensity is required to initiate the formation of the free radical?

[pgk]

Sorry for the misunderstanding but you didn’t clearly mentioned that you wonna try out a UV-cured “epoxy” resin. Anyway, in cationic photopolymerization, a strong aromatic acid (that might be a photoiniciator, too) absorbs electrons and transforms the initially formed free radical to a cation and thus, the polymerization continues under cationic catalysis. However, the free radical must be generated at the beginnings. Thus:
C-O     ≈ 85 kcal/mole which corresponds to:
λ ≈ 350 ± 15 nm.
It might work.
Good luck.
PS: Referring to the beam intensity for the initiation of methacrylates, you have to check the emission UV spectrum of the lamp, in order to verify the maxima at the given wavelengths (A little lower intensity than maximum, doesn’t matter). Otherwise, you have to calculate the radiation flux at a given wavelength, with respect to the surface of the substrate, the coating's mass and the distance from the lamp.

[Enthalpy]

Pgk, you should once for good have a look at some real absorption spectra and compare them with the bond energies. They do NOT fit - sorry, but NOT. Then you can forget these very wrong explanations.

[Enthalpy]

Led begin to emerge at 360nm, for instance at Ushio
http://www.ushio.co.jp/en/products/light_source/led/index.html
but check the prices... Usual GaN Led emit at 405nm, shorter waves need the newer AlGaN.

Easier: take a medium pressure mercury arc lamp. It produces much power around 360nm.

Cheap: just take a bunch of fluorescent lamps. Put several straight tubes side by side if needed. They gave me enough UV to polymerize MMA in few minutes.

Or use sunlight if your location permits it. Which also means: when using your resin, beware of sunlight and fluorescent lamps. To avoid premature polymerization, work in red light, or at most under filament bulbs if covering the resin.

[pgk]

Dear Enthalpy,
Please, do not confuse the bonding/antibonding orbital excitation energy, as plotted in the UV spectrum, with the bond dissociation energy (though, strongly related). UV spectrometry measures the π → π*, n → σ* and n → π* excitation energy in combination with the auxochromic/bathochromic effect of the adjacent and/or conjugated group, at ultra-short time exposure; but not the pure bond dissociation energy (neither the σ → σ* excitation). Therefore, UV spectrum is not the correct tool, in order to conclude whether the said concept is right or wrong. Contrary, appropriate methodologies and instrumentations that exist, accurately prove the said concept, together with the principles of photochemistry.
Photoiniciators that are activated around 400 nm (black light), are already proposed and they are suitable for the photopolymerization of alkenes. However, these photoiniciators might not be necessary for the photopolymerization of oxiranes (epoxies) that dissociate around 350 nm.
Fluorescent lamps also emit bellow 300 nm. But fortunately, their UVB energy content is very low and in practice, the corresponding power is negligible above a few cm of distance.
Sunlight also contains a significant amount of UVB and UVC radiation that has escaped from the ozone layer and the clouds scattering. Unfortunately, the sunlight content of UVB and UVC radiation is important and dangerous; especially, at low geographical longitudes and/or high sea level altitudes, during summertime.
Thus, using a dozen of fluorescent tubes, in touch or exposure under sunlight might probably work, but it cannot be considered as a scientific and reproducible methodology.
In case of doubt, Sir, do not hesitate to ask for further details and comments, before being so severe, in your criticism.
Sincerely yours
pgk

Question: Does UV spectrometry alter the sample?
Or, is UV spectrometry a destructive analytical method or not?

[Markitron]

Many thanks for the great replies all, my materials will arrive in the next few days so I'll see how it goes.

[pgk]

Good news.

[Corribus]

Question: Does UV spectrometry alter the sample?
Or, is UV spectrometry a destructive analytical method or not?

Not a question that can be answered generally. Usually, UV spectrometry is not a destructive technique. Then again, most experiments are done in very dilute solutions and with very short exposure times. Any analytical technique have the potential to interact destructively with a sample, though. I've had samples in the past that did not emerge from a UV-Vis experiment unscathed. It depends completely on the nature of the sample and the experimental conditions and your tolerance for what you'd call "destructive".

[pgk]

Exactly!! However, ultra-sensible samples (e.g. sensible photoiniciators or heteroannular-homoannular conjugated isomers) might be affected. Besides, very dilute solutions means an elevated ratio of radiation/mole.
           

[Enthalpy]

[...]before being so severe, in your criticism.[...]

Yes. Reading my text again, I formulated it less delicately than I wanted. Apologies.

What I wanted to point out is that the so widespread claim that UV absorption edges relate with bond energies is not confirmed by experiment. For instance methane is transparent >152nm: <4.5*10^-24cm² per molecule (90m absorption distance!), while the photon energy is 800kJ/mol, widely above 400kJ/mol for the C-H bond.

Or take ethylene: at 172nm, a photon would have enough energy to break the double bond - but it only achieves to excite one of the four electrons. And at >209nm ethylene only absorbs <1.5*10^-23cm², still dropping sharply, despite the photon energy would widely suffice to excite an electron.

I know many courses and books claim a relationship with the bonding energy, but this only proves that the authors didn't check what they wrote. Some even contain false curves to support the wrong claim.

Air is transparent at the desired 360nm, as sunburns tell. I did use successfully a bunch of fluorescent lamps as a UV source to polymerize MMA.

[pgk]

No problem.
The truth is that bonding/antibonding orbital excitation does not always coincide with bond dissociation energy. It may happens, if the antibonding orbital is a HOMO that is in touch with a LUMO one, but it also may happens that the electron has absorbed supplementary energy and completely escapes from the molecule and therefore, the molecule irreversibly changes, when radiation stops, e.g. stable free radical formation.
UV spectrometry cannot clearly measure the σ → σ* excitation and therefore, methane or epoxy photoexcitation is not visible by UV spectrometry.
Extinction coefficient (and flux variations), is a question of the ratio of absorbed/transmitted energy and not of the energy of the absorbed photon, alone.  Besides, extinction coefficient is amplified by a nearby conjugation (roughly: how extended the photon trajectory through the molecule, is). Therefore, extinction coefficient and variations, are not the correct tool to estimate the absorbed energy.
I agree that using fluorescent lamps is a smart and inexpensive solution that works but I am just wondering whether it is a general purpose solution or not.