PDA

View Full Version : UV LED's



N2CHX
04-13-2011, 08:07 AM
Hey guys,

Any of you know a good source for UV LED's? Somewhere in the 250-270 nm range would be excellent.

KC2UGV
04-13-2011, 08:14 AM
I think Rudy might have some :lol:

WØTKX
04-13-2011, 09:11 AM
Google Fu has few answers, but that wavelength is not in the LED (380 nm) sweet spot,
and therefore would be expensive.

Mercury vapor lights are considered the best solution.
Those shorter wavelengths are more dangerous biologically.

Useful for sterilization, so maybe industrial medical? Maybe industrial epoxy curing?

rot
04-13-2011, 10:27 AM
Here ya go...make your own!
http://lib.semi.ac.cn:8080/tsh/dzzy/wsqk/selected%20papers/Journal%20of%20Physical%20Chemistry%20B/110-14685.pdf
OK...so I'm the Island smartazzficktard de jour.
oh well good luck!
:-D
Just had to do it.
Take Care
rot

NQ6U
04-13-2011, 10:29 AM
Here ya go...make your own!
http://lib.semi.ac.cn:8080/tsh/dzzy/wsqk/selected%20papers/Journal%20of%20Physical%20Chemistry%20B/110-14685.pdf

Trying to understand that document made my head hurt...

N2CHX
04-13-2011, 10:54 AM
LOL nice. I have access to all sorts of things like deuterium lamps and such, which I could use but I'm trying to keep it simple. I want to build a calibration standard for UV spectrometry. What we've been doing until now is running caffeine in Methanol, which produces a big peak at 208 nm and a smaller one at 272 nm. It's a slow, painful way of calibrating. I have an old diffraction grating I want to use, wherein I will shine a wideband (>10 nm) UV LED light source at it, and the output will be a slit in the chamber where the LED and grating will be fixed, at the correct angle for a known wavelength. This will give me a really accurate and stable narrowband UV light source, which I can then put in an RID or similar LC analyzer to calibrate the wavelength setting.

They sell such things, but they're thousands of dollars and I pretty much have all the parts to build one myself, and known working equipment to calibrate it against, so per usual I wanna roll my own....

WØTKX
04-13-2011, 11:02 AM
The skinny zinc oxide wires are interesting. Wonder if you could build that?

By Tesla's Ghost! I thought you were making a death-ray. :mrgreen:

Avatar Material. :lol:

http://rlv.zcache.com/nikola_tesla_death_ray_co_stickers-p217004307298107127q0ou_400.jpg

rot
04-13-2011, 11:16 AM
I don't understand. are you calibrating a spectrophotometer for analytical use?
the reason i ask is the RID and LC mean Refractive Index detector and Liquid Chromatography to me...not sure if this is what you meant..I do this all the time.
I do not know why this is slow and painful...unless you are doing something completly different.
You can get certified films to do this and most of what you pay for is the cert.

If you want to use Caffieine/Meoh which is fine to make sure of the methanol purity...It will have a uv cutoff (usually 190nm).
Just wondering.
rot

ps i just checked most methanol has a uv cutoff of 205...a bit to close for 208 work,

N2CHX
04-13-2011, 11:24 AM
I don't understand. are you calibrating a spectrophotometer for analytical use?
the reason i ask is the RID and LC mean Refractive Index detector and Liquid Chromatography to me...not sure if this is what you meant..I do this all the time.
I do not know why this is slow and painful...unless you are doing something completly different.
You can get certified films to do this and most of what you pay for is the cert.

If you want to use Caffieine/Meoh which is fine to make sure of the methanol purity...It will have a uv cutoff (usually 190nm).
Just wondering.
rot

ps i just checked most methanol has a uv cutoff of 205...a bit to close for 208 work,

Yep, that's exactly what I'm talking about. And yes, it's painful because we seem to get a lot of false peaks. We look for the peak at 272. The 208 peak is oxygen, as I'm sure you know way better than I. I'm just learning this stuff.

rot
04-13-2011, 11:49 AM
Well I'm a bit confused again..an RID has nothing to do with UV. RI is a light bending property of whatever you flowing through the cell. UV is a different detector type all together. Now for your false peaks if you are using an RI, then it is your column is not separating what you are looking for? Can you give me analyte, column, eluant, and detector briefs...I'll be glad to help if I can.
rot

WØTKX
04-13-2011, 12:47 PM
Cool stuff you mad scientists. I wish my job was more "techie" than it is.

Had some lulzy discussions about our combo IR/visible light cameras at work as they lose focus on IR only when the lenses get coated with clear ice. Management did not (and my boss) and bitched that the cameras were defective. Took phone calls from the engineers who designed it to calm them down.

Arguing that science is defective. Flat earthers are everywhere!

N2CHX
04-13-2011, 01:29 PM
Well I'm a bit confused again..an RID has nothing to do with UV. RI is a light bending property of whatever you flowing through the cell. UV is a different detector type all together. Now for your false peaks if you are using an RI, then it is your column is not separating what you are looking for? Can you give me analyte, column, eluant, and detector briefs...I'll be glad to help if I can.
rot

Well, I've only been doing this a month so my terminology is probably incorrect. Specifically, a variable wavelength detector like a TSP UV-1000/2000, PE 785A and similar detectors. As for the eluant, it's always MeoH or IPA. We don't do any analyzing or real chemistry here, we just fix the intruments. I'm an electronics person and this chemistry stuff is all new to me. I got hired to do component level board repair and the next thing I know, I'm fixing pumps, flowcells, and all sorts of other LC stuff. I haven't dived into the GC side much aside from the autosampler trays, towers and controllers, which is so boring it makes me fall asleep.

WØTKX
04-13-2011, 02:03 PM
Sounds like a very cool project, Kelli.

rot
04-13-2011, 02:15 PM
Well, I've only been doing this a month so my terminology is probably incorrect. Specifically, a variable wavelength detector like a TSP UV-1000/2000, PE 785A and similar detectors. As for the eluant, it's always MeoH or IPA. We don't do any analyzing or real chemistry here, we just fix the intruments. I'm an electronics person and this chemistry stuff is all new to me. I got hired to do component level board repair and the next thing I know, I'm fixing pumps, flowcells, and all sorts of other LC stuff. I haven't dived into the GC side much aside from the autosampler trays, towers and controllers, which is so boring it makes me fall asleep.

well effin cool. i have worked with chromatography stuff for ca 30 years. love it. i used to fix pumps check values autosamplers and stuff. can't do it anymore. i call the vendor. Anyway for a uv/vis vwd if you need a source...buy one. if you need to confirm frequency you don't need a source you need a certified material to confirm the freq/absorption. you can buy these as well. One that has known properties. you really don't have to be right on freq. what you want is the column/ detector to respond with one peak at the same time each injection. if you are using meoh eluant and a meoh/caffiene soution...you should have a stable a stable baseline and one resoved peak at some time. if you are getting a bunch of crap...go bitch out the chemist. you can do nothing until all the shit is off the column and the baseline runs at zero au, clean, and at a stable psi. there's a bunch of variables and you can waste huge amounts of time chasing this around.
i could go on forever, but if you want send me a pm. ill give a phone number and we can chat. i'm just had knee surgery so the turnaround maybe slow. im inna fuckin chair immobilized right now and typing like a dick. anyway i ain't to busy for the next months so it ain't no problem. just a bit slow on this mode. in the future reccomend Agilent stuff..I'm ussually involved in 5 or 6 system and they are effin works horses. we have 2 running 24/7.
rot

N2CHX
04-13-2011, 02:15 PM
Sounds like a very cool project, Kelli.

I think so, but if there's an easier way, I'm all for it. I bow to rot's experience and knowledge. I never thought I'd be working on stuff like this. I'm way out of my normal realm, as far as this HPLC/GC stuff. As I said, I got hired for component level board repair but they needed some major help with the LC stuff and quickly realized I had a knack for the mechanical aspects of it too, so I swing between doing that and my low-lever repair, as needed.

N2CHX
04-13-2011, 02:25 PM
well effin cool. i have worked with chromatography stuff for ca 30 years. love it. i used to fix pumps check values autosamplers and stuff. can't do it anymore. i call the vendor. Anyway for a uv/vis vwd if you need a source...buy one. if you need to confirm frequency you don't need a source you need a certified material to confirm the freq/absorption. you can buy these as well. One that has known properties. you really don't have to be right on freq. what you want is the column/ detector to respond with one peak at the same time each injection. if you are using meoh eluant and a meoh/caffiene soution...you should have a stable a stable baseline and one resoved peak at some time. if you are getting a bunch of crap...go bitch out the chemist. you can do nothing until all the shit is off the column and the baseline runs at zero au, clean, and at a stable psi. there's a bunch of variables and you can waste huge amounts of time chasing this around.
i could go on forever, but if you want send me a pm. ill give a phone number and we can chat. i'm just had knee surgery so the turnaround maybe slow. im inna fuckin chair immobilized right now and typing like a dick. anyway i ain't to busy for the next months so it ain't no problem. just a bit slow on this mode. in the future reccomend Agilent stuff..I'm ussually involved in 5 or 6 system and they are effin works horses. we have 2 running 24/7.
rot

I definitely bow to your experience, and yeah I'd love to pick your brain. I'm loving the learning new stuff again. I became really stagnant in radio, because I stopped learning and got sick and tired of fixing the same crappy shit all the time, usually at 3 am.

Agilent, yes.... Right now I'm surrounded by no less than three Agilent 1100 stacks. But we deal a lot with PE, Jasco, Waters, Shimadzu and a few others because there's demand for it. Last week I worked on a few Jasco detectors and I was really impressed by the stability and sensitivity, even compared to the Agilent stuff.

Basically I rip the instruments apart, clean everything, replace seals, polish the seal surfaces and pistons (on pumps), and replace anything else worn out or about to fail. You never know what has been run through some of these things. I just finished up a pair of Waters 515 pumps that were so corroded and filled with salts it was gross. On top of that, one was leaking oil through the stepper motor. When it came in it didn't work and looked ready for the trash bin. When it left my bench it looked almost brand new and passed testing no problem.

rot
04-13-2011, 02:53 PM
well cool. i sure can't fix 'em anymore. refurbs are excellent and have used a many. labs will abuse the shit out of these things.
i sit back and advise these days. we have bright new chemists that with just a little help do real well on their own. New apps deserve new equiptment, at least for me. I'm the old lab relic who pisses and moans about peak integration software, cause back in my day we had to manually triangulate.
anyway great endeavor...chromatography has always been my favorite.
im sure you know more about the electronic shit than i do..i had to solder a zeroing pot on an old Biorad pump about 20 years ago and that's about the extent of what i can do.
rot

KK4AMI
04-13-2011, 04:33 PM
I definitely bow to your experience, and yeah I'd love to pick your brain. I'm loving the learning new stuff again. I became really stagnant in radio, because I stopped learning and got sick and tired of fixing the same crappy shit all the time, usually at 3 am.

Agilent, yes.... Right now I'm surrounded by no less than three Agilent 1100 stacks. But we deal a lot with PE, Jasco, Waters, Shimadzu and a few others because there's demand for it. Last week I worked on a few Jasco detectors and I was really impressed by the stability and sensitivity, even compared to the Agilent stuff.

Basically I rip the instruments apart, clean everything, replace seals, polish the seal surfaces and pistons (on pumps), and replace anything else worn out or about to fail. You never know what has been run through some of these things. I just finished up a pair of Waters 515 pumps that were so corroded and filled with salts it was gross. On top of that, one was leaking oil through the stepper motor. When it came in it didn't work and looked ready for the trash bin. When it left my bench it looked almost brand new and passed testing no problem.

Deep UV LEDs (240nm-280nm) are kind of a rare breed. In other words the Chinese haven't started mass producing them yet. Neat wavelength to work in especially for food processing. Gotta get rid of those mercury vapor lamps. Here is one, but good luck finding cheap sales yet.
http://www.micronkk.com/1_imaging/set/SET_UVTOP_Catalog.pdf

ab1ga
04-13-2011, 06:50 PM
LOL nice. I have access to all sorts of things like deuterium lamps and such, which I could use but I'm trying to keep it simple. I want to build a calibration standard for UV spectrometry. What we've been doing until now is running caffeine in Methanol, which produces a big peak at 208 nm and a smaller one at 272 nm. It's a slow, painful way of calibrating. I have an old diffraction grating I want to use, wherein I will shine a wideband (>10 nm) UV LED light source at it, and the output will be a slit in the chamber where the LED and grating will be fixed, at the correct angle for a known wavelength. This will give me a really accurate and stable narrowband UV light source, which I can then put in an RID or similar LC analyzer to calibrate the wavelength setting.

They sell such things, but they're thousands of dollars and I pretty much have all the parts to build one myself, and known working equipment to calibrate it against, so per usual I wanna roll my own....

Be careful not to practice false economy here, you may well be able to buy a cheaper source for your purposes than you could ever build yourself.

StellarNet produces a number of sources, including a combination mercury/argon source which emits multiple lines between 253 and 1014 nm. It can run off of batteries or a wall wart, takes 15 seconds to warm up, and costs less than $300. It has a fiber optic output connection so it's a lot easier than aligning optics.

There may be downsides to the gas tubes as light sources, but the line emissions are very narrow, and their wavelength very well known, all with no moving parts. LED sources usually require sophisticated driver circuitry and often temperature regulation to maintain useful stability, and diffraction gratings have a resolution limit dependent on the number of grooves per mm cut into them, so you'll need to check that before proceeding.

If you really want only one output wavelength, then you can do that by adding an interference bandpass filter on the output, which won't hurt the width of the primary line but will get rid of those that are nearby. You may not need this though, because two lines in the detector passband is much more useful for calibration than just a single line.

Good luck with everything,