thanks headstrong for hudlicky's book in it i found this reference for the dehalogenation of aromatic compounds y means of u.v. light and isopropanol and the reference is:
pinkkey d.t.,rigby r.d.g Tetrahedron letters 1969 1267 (36,68)
which lead to this finding

A long Google Books link

page 518 see scheme 8

uv light in the wavelength of 313 nm i would venture i'm not sure the wattage needed to accomplish this

essentially the aryl chloride is converted to an aryl radical and a Cl radical by (hv) simultaneously isopropanol is converted to a beta radical by the chlorine radical which give HX and pinacols as products typical yeilds are 77%

can anyone find the full reference for this???

another question: since the higher wattage high frequency uv lamps are typically 254 nm wavelength (used for germicidal applications) would a shorter, more energetic wavelength work just as well?

can anyone find the full reference for this???

The reference is here.

another question: since the higher wattage high frequency uv lamps are typically 254 nm wavelength (used for germicidal applications) would a shorter, more energetic wavelength work just as well?

Though I can't yet tell you which wavelengths are needed and why, CHEMISTRY LETTERS, pp. 1589-1592, 1987 used "a 30W low pressure mercury lamp located at the center of concentrically surrounded quarz vessel containing the acetonitrile solution of a polychlorobenzene." The original reference used a "a 125-W medium pressure mercury arc in a water-cooled quartz well." From some preliminary findings, it seems that the quartz is important. From this wiki page, "in low-pressure mercury-vapor lamps, only the lines at 184 nm and 253 nm are present, though only the light at 253 nm is usable unless synthetic quartz is used to manufacture the tube as the line is otherwise absorbed. In medium-pressure mercury-vapor lamps, the lines from 200-600 nm are present." Therefore, the lower wavelengths are necessary. I'll keep looking and update this post.

Photochemical Dehalogenation of Polyhalobenzenes. I. Survey of the Reactivity in Hexane Solution
Masahiro Nakada,Sachio Fukushi,Hideaki Nishiyama,Keiji Okubo,Kotaro Kume,Minoru Hirota and Tetsuo Ishii
Bull. Chem. Soc. Jpn., 56, 2447 - 2451 (1983)

depends on the peak absorption spectra of your substrate and what kind of reaction is taking place.
in the case of aryl chlorides the absorption spectra is 250 nm and less i don't have the exact maxima but it seems more prudent to use a UV-C lamps for a number of reasons.
1) it does'nt give off much heat energy so it could be sleeved in a quartz jacket and submersed in the reaction soup.
2) efficiency no wasted energy so a lower wattage could be used.

this is a good example of a uv-c sterilizing bulb with quartz sleeve and ballast.

http://www.americanaquariumproducts.com/QuartzSleeveBallast.html

enkidu you are such a wealth of knowledge i wanted to pose this question: since loperamide is negligbly soluble in isopropanol would adding methanol (which it's freely soluble in) interfere with the reaction scheme homolysis>radical formation>reduction?
i don't see any problems with it but, i just wanted to run it by you.

righto o was looking at that paper and the lamdba was >300nm rings absorb light in the 190-254nm range so i think it might not be even nessecary to use ketones because all they are doing is transfering photons to a system that does'nt absorb at that frequency/wavelength.
in other words it's a nice paper (publish or perish) but it is'nt needed for this system also they used a water cooling jacket that would'nt be needed for a sterilizing uv-c lamp because uv-a and uv-b produce heat and uv-c runs cool so that simplifies things a bit for us if you read the original document you'll see what i'm saying i think isopropanol is used because it forms a stable carbocation or radical being secondary, it would'nt hurt to throw a little co-solvent in the mix.
it would'nt even hurt to throw a little cyclohexanone in there too for good meausure.
that's another paper right there

i'd say you're right about most everything thus far. and so convenient! uv-c low pressure mercury bulbs emit 90% of their energy in the 250 nm range.
maybe 10% in the 190 nm range
the non-ozone producing ones are between 253nm and 340nm
the medium pressure bulbs that is.
chlorobenzene; peak absorption = around 240nm

one could probably get away with a teflon tube (fairly inert to uv light)
immerse the quart sleeved uv bulb into it and leave little clearance like you suggested.
some pvc's are photoresistant to uv light too.
convection should be sufficient to move the solvent around.

[insert by Enkidu from here].

trying to figure out how to setup that photoreaction and i think i got it.
UVC bulbs give off little heat so they can be submerged.
quartz sleeves and uvc bulbs are available everywhere for water purification and they are usually just setup in pvc.
so one could have a pve pipe capped at one end and the other capped with an opening for the qaurtz sleeve with a rubber taper seal for the opening.
the uvc bulb could be sealed up to protect it from thermal shock and if nessecary the pipe could be cooled by circulating water through it with a glass/plastic tube.
basically a 35-50 watt bulb will do nicely because of the high energy of the spectrum and the fact that aromatics absorb right in the 180-256 nm range.
it would be very easy.
of course uv light would break down pvc but there are different kinds of pvc that are resistant to uv attack.
if one wanted to be anal he could use polyfluorocarbon plastics.