Ok, I've read a SHITLOAD on this idea and the problem(s) have always been that the two areas of interest - the 200-400nm range and the >1200nm
range, are both invisible to normal sensors.
That meant that no matter what I thought up, the whole problem was going to founder on the rocks of the need for a photodetector/photodiode/etc.
array.
Then I starting reading about "Paired Emitter Detector Diodes", the basic principle of which is that one diode of the pair emits at a lower wavelength
than the optimal wavelength of a second diode which is being run in reverse bias. Now, here is the tricky part, the reverse-diode actually does not
measure the light incident upon it, but the time taken for the LED to accept and convert enough radiation-energy to discharge.
That said (and I think I said it right - I am hellishly tired), the fact of the matter is that the authors on the subject state unequivocally that a
diode can be used as a sensor to measure light absorption from any wavelength under that for which it is rated.
If this is true, then a basic blue diode would be capable of accepting and effectively measuring the total light in the UV region incident upon it
(such as is given off by a halogen/tungsten/quartz tube).
In terms of sensors, there is obviously going to be a need to design a better mousetrap (although anyone looking at the literature via google for
PEDD/Paired-Emitter-Detector-Diode will soon get some ideas on design).
Now I know this is not perfect - but the measurement of the entirety of the incident radiation, allowing quantification thereof, while a motorized,
actuator stage is used to build a moving mirror.
Now, what could we possibly use as a beamsplitter for UV-VIS? As Quartz Cuvettes, so obviously glass/plastic are out. un0me2 - 4-9-2010 at 03:58
Another option is for amateur chemists, who want to take advantage of the newly discovered capacity to build UV-detectors (see attached papers) from
low-temperature ZnO nanowires and crossed nanowires, from low-cost Zn salts and fairly simple syntheses to examine organic compounds (outline[url], [url=http://en.wikipedia.org/wiki/Ultraviolet%E2%80%93visible_spectroscopy]Wiki-page and the Woodward-Fieser Rules). It is a long way from the almost-unequivocal results of FT-IR, but by the looks of what is coming out, it may be
achievable by amateur chemists, using simple solution chemistry and then a microscope for fabrication, or use one of the vertical array techniques,
based upon hydrothermal syntheses of ordered nanorod arrays.
The fact remains, these miniscule wires are UV Detectors of some interest (Here's why), imagine being able to use a low-melting metal (tiny amounts of gold should work) to cap the vertical arrays & then use that to transport the electrical impulses to a detector? Or horizontal arrays which form a circuit on ANY substrate?
There has to be a way for us to use this - the fact is we can synthesize the nanocrystals, maybe on an etched silicon surface, so as to be able to
utilize the photo-electric effect when they are contacted by UV light.DDTea - 4-9-2010 at 06:12
Alright, I've read this thread numerous times and tried to find what you are talking about online, but I still can't figure out what you're on about.
Would you mind clarifying a few things so that we're on the same page:
What exactly are you trying to accomplish? From what I can tell, you're trying to acquire spectra of organic molecules in the 200-400 nm region as
well as the >1200 nm range. You want to use a Michelson interferometer as a way of creating the desired wavelengths, a double beam instrument with
a reference cell and sample cell, and a detector that utilizes the photoelectric effect. The samples will be held in quartz cuvettes. Am I correct
so far?
Quote:
Ok, I've read a SHITLOAD on this idea and the problem(s) have always been that the two areas of interest - the 200-400nm range and the >1200nm
range, are both invisible to normal sensors.
What kind of "sensor" are you talking about? That's an *extremely* broad word that can have many different meanings depending on the context and who
is saying it. The detector (e.g, Photodiode array, PEDD, CCD, whatever) could simply be called a "sensor," but so could the whole device,
encompassing the sample source, incident radiation, monochromators, detectors, etc. (e.g., UV-Vis Spectrophotometer, FT-IR, AES, etc.)
Quote:
Now, what could we possibly use as a beamsplitter for UV-VIS? As Quartz Cuvettes, so obviously glass/plastic are out.
You could simplify the whole design by eliminating the beam splitter altogether. It's not necessary to run a reference cell simultaneously, although
it is certainly convenient because it can eliminate most instrumental drift. However, you can account for instrumental drift in a single-beam
instrument: there is something called Dynamic Measurement Correlation. In DMC, you measure your reference cell (typically a standard solution) before
you measure your unknown sample, then measure the reference cell again afterward. By tracking the change in absorbance of the reference cell over the
time period of the measurement, you can correct for drift.
As far as the Woodward-Fieser Rules are concerned, those are actually pretty good approximations for a lot of molecules. Anyone using UV-Vis
spectroscopy should take a look at them. However, it's also important to know the fundamental science too--what kind of electronic transitions are
occurring and solvatochromic effects, in particular. Familiarity with "model" compounds' spectra and absorption maxima are also useful.
I'm still not sure I understand what you're trying to do though.
[Edited on 9-4-10 by DDTea]un0me2 - 4-9-2010 at 06:49
Well, here is a paper, quite an interesting one, on building a Mg0.1Zn0.9O sensor, on
a glass plate, with conductive silver paste/paint and a gap, filled with nanorods/nanowires of the above compound (in toluene solution) and allowing
it to evaporate.
The absorption is about 100% (obviously going to electrical current) for that from ~200nm-390nm which would give some idea about the compound(s) in
the mix (or more appropriately, the disappearance thereof). In terms of instrumentation, it would be the very, very, basic starting point. But it
beats hell out of guessing. The sensor under discussion is shown in the linked paper. Using a basic tungsten/halide lamp with a filter to stop visible
light, should allow for the collection of the UV spectra, shouldn't it?IrC - 4-9-2010 at 08:59
Not sure if these links are useful for your work but lately I have been modifying digital cameras for UV, IR, or UV + IR photography and so far have
fairly good results which will be even better when I can afford improved optical materials such as getting a Baader U filter, real hard to get cheaply
in a 58 mm mount. I wonder if closeup UV images would be useful and how this compares to the detectors you mention.
Lately I have been wondering how useful this UV and IR imaging would be in chemistry experiments so this thread got my interest. DDTea - 4-9-2010 at 19:55
The following is taken from Skoog, Holler, and Crouch, Principles of INstrumental Analysis Sixth Edition, 2007. Thomoson Brooks/Cole. ISBN:
0-495-01201-7. Figure 7-2 and Figure 7-3, since I don't have a scanner, I will manually enter the data regarding choice of construction materials for
different optical components. As you see, the decision of what materials to use is dependent on which wavelengths are to be measured. The
wavelengths listed are the ranges over which the materials are usable. All values are approximate. Some of this might not be useful to you, but I
included the whole table for completeness.
Materials for cells, windows, lenses, and prisms
Lithium fluoride: ~125-5500nm
Fused silica or quartz: ~180-3300nm
Corex glass: ~300-2000nm
Silicate glass: ~390-2000nm
NaCl: 200-13,000nm
KBr: 150-25,000nm
TlBr or TlI (that's Thallium Bromide/Iodide, to be clear): 150-35,000nm
ZnSe: 650-18,000nm
Wavelength selectors: Continuum sources
Fluorite prism: 110-200 nm
Fused silica or quartz prism: 175-3300nm
Glass prism: 390-2700nm
NaCl prism: 2000-16,000nm
KBr prism: 10,000-30,000nm
Gratings: can cover entire spectrum from 100nm (Vacuum UV) with 3000 lines/mm to 40,000nm (Far IR) with 50 lines/mm.
Interference wedge: 400-17,000nm
Fused silica is what I use, cannot afford high end quartz optics right now. 92% transmission from uV-A and lower. Best I can do now but works
extremely well. When you take the IR block filter out you must measure the thickness, and find pre-made stuff with the correct I.R. and thickness to
put the focus back to perfection. Cheapest place I have found so far wants $540 for 10 windows using stock material they have, over a grand/window to
have them built to my exact specs. Problem is their stuff is 0.1 mm thicker than the window in the camera, wrecking the focus, especially for closeup
work such as photographing chemical experiments. As far as having a perfect window so focus is not altered made, 10's of thousands for one, a couple
hundred buck each for 5,000. It is that first one made you need Bill Gates money for. Much cheaper in quantity. So this limits me to not much
available after doing a tremendous amount of work when the best you can do is get close. Assume you are not sure the material of the OEM filter
therefore also unsure of it's index of refraction. Work to measure it to high accuracy, then calculate the new thickness needed when all you can find
pre-made and cheap of a different thickness or I.R. has to come out right to keep focus on the money. Not easy at all. So far the only low budget way
around this is to waste one camera tearing apart the assembly to find a perfect way to shift a lens in the assembly by a precisely controlled distance
to regain focus with just any cheap filter or window negating the need for custom optics. By cheap I mean high quality but surplus from something
else.
I assume only God could get a programmer out of Sony, let alone one with all the programming data and instructions. If I had this I could reprogram
the camera to not shift focus for the window (IR block filter), omit the window and just screw filters on up front having no need to use a window.
This would also make the camera more fun at pools. Sony in the late 90's added code to fix light levels, lock iris, and disable manual focus, manual
light balance, etc. for this reason making it much harder to modify and fully use it for the kinds of work we might be doing. I found taking pics and
clips of thunderstorms in only 720 nm and lower IR give fantastic and cool images. I am going to modify another one for uV-A + IR to add the
ultraviolet coming from discharges to the infrared background of the overall storm. I can imagine somewhere like Venus or Mars would look similar,
very cool scenery.
So you use a low pass filter to block all above IR, screw on a variable density filter (ND2 to ND400) to control the light, and set it on a tripod as
the shutter is open meaning the slightest jiggle blurs the image. You also shift from LCD big screen to the eye LCD window. Otherwise someone would
beat the crap out of you if they see their girlfriend in her transparent bikini on your screen as they are walking by. No I have never used it for
this but I am over the hill. If I was in my 20's who knows.
Anyhow a lesson on voyeur cams was not the point. Just explaining what you have to do to the camera for other pictures not TSA scanner related. Yes
they see all you have at the airport. What I am working with is crap compared to what tax dollars bought them. I just started working on converting
one to only ultraviolet but finding visible block filters are expensive. Very.
Anyway to the point of the thread I still do not know from anyone if cameras like this would be useful in the photomicrography of chemistry
experiments, in wavelengths well above or below visible. Image wise in closeup mode you can take a pic in raw data format of 15 mb for one image,
detail is incredible. It would be helpful to know if so, what frequency ranges would be best.
Edit to add: thanks for doing all the typing, very useful data. Much better than any tables I have found elsewhere.
[Edited on 9-5-2010 by IrC]aliced25 - 3-9-2011 at 05:39
Ok, I've been doing a bit more work on this as well.
What I've come up with is using a master grating and taking a polymer cast of that (essentially what they do with CD/DVD's). Then using Aluminium PVD
(getting a backing pump in the near future) deposit a thickish layer of Al on the polymer. Then to avoid the trouble caused by overlap/thickness
issues on the backside of the PVD layer, use a solvent to dissolve the polymer leaving a true aluminium grating for less cost than currently available
(essentially a grooved mirror).
If that were designed from the outset (as a master ought to be) to split the 200-1,000 nm range then a single grating should suffice. It would be a
solid mirror, so there is no need for silvering, etc.
The entire thing is placed in an evacuated chamber with a quartz window/slit. The light coming through the slit is diffracted off the grating onto a
Lumogen (PVD coated lumogen onto optical glass screen - also within the vacuum enclosure). The lumogen upshifts the 200-400nm light to visible
wavelengths which register on an ordinary CCD/CMOS sensor (uncoated).
PVD is the gamechanger here, if it works then UV-NIR Spectroscopy should be able to be made quite affordable, the closed (vacuum chamber) unit could
even be adapted to be used with various light sources, including the Capacitive Plasma Atomic Emission example mentioned elsewhere.
