unome
Hazard to Others
Posts: 134
Registered: 17-10-2009
Member Is Offline
Mood: No Mood
|
|
LED Arrays - Can we get the full spectrum?
I mean, can we get from the mid-UV all the way to the [url=http://en.wikipedia.org/wiki/Infrared]FIR?
I know UV LED's are available, visible LED's are available and NIR LEDS are available, but what about those going into the MWIR, LWIR/FIR ranges?
I've searched around a bit, all I can seem to find is MWIR LED arrays for pretending to be high-temperature targets (military app's obviously).
Has anyone seen much in the way of LED's that could theoretically enable us to build a better light-source for FTIR?
Just thinking, cos I have an idea of how to improve the RT sensors...
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
http://www.roithner-laser.com/
Laser Diodes 395 nm - 14 µm
LEDs 245 nm - 7.0 µm
High Power LEDs 350 - 1550 nm
LEDs are not good sources for wide band spectroscopy as they are not continuous sources like thermal emitters.
You could have a large number of LEDs, each producing a different band, but there may be regions where no commercial devices emit. It would be
difficult to arrange them so as to provide all the bands coming from a single spot, not doing so complicates the rest of the optics.
When the bandwidth demands are less the devices like quantum cascaded lasers or synchrotrons can be used. These are generally used for applications
where you know what class of compounds you have or want to monitor, so you only look at a small section of the IR spectrum. They aren't so useful in
a general purpose machine, and they are much more expensive than more conventional thermal sources.
The detectors are more of a problem, FTIR needs fairly quick response and wide bandwidth, as does dispersive IR. VNIR to UV can use one and two
dimensional detectors, and do FT without moving parts; only at their very widest range, sat 200 to 1200 nm in a single instrument, do the match the
bandwidth of a convensional research IR spectrometer.
Conventional thermal sources such as globars or resistance heating elements can be operated in open air, and so avoid issues with the transparency uf
ant envelope. They're pretty simple to operate and not that expensive. With some fine KANTHAL wire wound on a mullite or alumina form you've got your
IR light source.
|
|
|
bquirky
Hazard to Others
Posts: 316
Registered: 22-10-2008
Location: Perth Western Australia
Member Is Offline
Mood: No Mood
|
|
I agree. semiconductors dont make good broad band sources (unless they over heat) its hard to go past good old incandecent light sources for
bandwidth (unless you need a very high intensity source)
dont forget that if it is just bandwidth you seek incandecant lamps become efficant as all that 'waste IR & UV light become usefull energy'
but if you want more Umph
Consider a Xenon arc lamp for higher intensitys they can be had quite cheap from ebay as aftermarket headlights for cars. prehaps a hunded bucks or so
If you want still higher intensity. a broard band light source I saw at a trade show used a Highly Foucused IR laser amed at a presurised Xenon
Envelope. this resulted in an Ionised spheare of plasma 200 micron diamiter puting out 10 watts (about $9k)
if you want brighter than that you will nead a 'super continuem laser' witch can give you about 4 watts in a single mode in the 400-1500nm range
($100k)
if thats not enough.. then you might have to find a cyclatron 
some instruments actualy use a gas flame
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
I'd had similar thoughts, and then discovered that automotive Xenon lights are more properly metal arc lights. They have prominent bands in their
specture, mercury and others; unlike pure xenon arc lamps the are approximately flat from the deep red into the near UV. :-(
|
|
|
bquirky
Hazard to Others
Posts: 316
Registered: 22-10-2008
Location: Perth Western Australia
Member Is Offline
Mood: No Mood
|
|
prehaps use the ballast on a xenon flash tube a bit of a hack but maby posible
|
|
|
un0me2
aliced25 sock puppet
Posts: 205
Registered: 3-2-2010
Member Is Offline
Mood: No Mood
|
|
I'm thinking more along the lines of the mid-UV - mid-IR (0.2-4.5cm-1) region, which if possible would give us all the information that we need...
The reason I was thinking of using LED arrays is to allow for a rapid on-off cycle (in order to avoid using a mechanical chopper/blocker), such as is
required with RT Thermo-Pneumatic detectors (Golay-type). They have highly accurate, rapid response (~25-30ms), but require a relaxation period (ie.
they must be isolated from the source of radiation) for about the same period (~25-30ms), so say 50-60ms/scan (so a stepper motor which can move the
stage the required distance in that same period would be good - and probably not that difficult to build given that Scientific American provided plans
for building a Michelson Interferometer at home, long before the advent of highly accurate servomotor driven, highly accurate, computer controlled
devices). The rapid on-off, flashing cycle would be trivial to design with LED's (fuck, flashing LED circuits can be bought in kits), but fatal to
Tungsten Halide lamps
On the plus side, Arsenic Trisulfide glass is invisible to Mid-IR radiation and can be made into lenses, mirrors, and fibers (also windows) and is
apparently not going to turn into a puddle with atmospheric water, just thinking...
[Edited on 1-4-2010 by un0me2]
|
|
|
bquirky
Hazard to Others
Posts: 316
Registered: 22-10-2008
Location: Perth Western Australia
Member Is Offline
Mood: No Mood
|
|
Choppers might not be too bad if your sensor be synced to the chopper insted of the othe way aound
allthough syncing xeon flashtubes has been perfected by the photography crew. puting a resitor in serise with the tube will extend its duration.
just an idle thought.
|
|
|
unome
Hazard to Others
Posts: 134
Registered: 17-10-2009
Member Is Offline
Mood: No Mood
|
|
I'm actually wondering - if we could work out what would work as a decent lens for the light going to the sensor, then with two or 3 modified
Golay-type sensors (depending upon how long they take to cool/reset), we could direct the light from one sensor to another, then back to the first and
have a constant scanning job with high precision & NO NOISE.
EDIT
Actually, as the Golay-type sensors equilibrate (ie. the gas loses its heat and reduced in volume) rapidly once they are no longer exposed to the
electromagnetic radiation (in fact if various sources are to be believed, as rapidly as they respond to the radiation in the first place), then
running two, side-by side, with a moving block in front of them, that allows the radiation to fall #1 & not #2 and then vice versa, then this
could be run as a scanning device - high precision, no noise, high-scan rate.
[Edited on 5-4-2010 by unome]
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
Golay detectors are not noise free. While they don't have dark current or shot noise, they do respond to mechanical vibration, acoustical noise, and
ambiant thermal fluctations.
Part of their low-noise performance comes about from the band pass filtering of their chopped detection system; similar techniques are used elsewhere
- chopping or AC stimulation with synchronous detection are common, sometimes pseudorandom sequences are used to reduce the impact of noise near a
fixed chopping/excitation frequence (at the cost of slower response times).
Getting a good wide band black coating, covering UV to mid-IR, can be tricky. It's within the reach of amateurs, but you need to reach the subject
first. And you'll need to do so if you plan to make your own Golay detectors.
Pyroelectric detectors are only a bit less sensitive than Golays, and are considerably smaller and cheaper.
|
|
|
aliced25
Hazard to Others
Posts: 262
Registered: 31-7-2010
Member Is Offline
Mood: No Mood
|
|
That aside, I'm seriously looking at the use of LED's - the spectrum of warm white and blue-white, when overlaid, give a pretty good fit up to nearly
800nm (from just under 400). Two LED's use a shitload less power than the alternatives and are easily fiber coupled.
On the note of phosphors, I know there are several that emit in the UV, I note for instance, Calcium Tungstate - which appears (by reference to the
fact that it shifts incident light a certain distance, or appears to) to Stoke/Red-Shift Light. Is it a Stokes Shift? If so, could one reasonably
presume that a particularly simply acquired noble gas, that gives off emissions in a continuum of 100-150nm in a dielectric discharge plasma, could be
mated with a phosphor (or two, like any normal plasma>phosphor>light source), to change the output to give 200-400/500nm excitation?
[Edited on 15-12-2010 by aliced25]
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
Phosphor shifts come from the same underlying effects as Raman shifts (and thus Stokes) In crystalline materials the bands are generally fairly
narrow, particularly emission wavelengths; for liquids including glasses the bands usually are fairly wide.
However you have to remember that Raman type shifts are a fixed wave number for a particular absorber-emitter. As the wavenumber goes up with
decreasing wavelength, the shifts in nm goes down; the common shifts with 800 nm excitation give a spread of nearly 200 nm, while the same shifts at
100 nm yield barely a 20 nm spread. You need to be kicking electrons into more widely separated bands, wider bandgaps limit your choices of materials
and too some extent the wavelengths.
Note that deep UV tends to be hard on many materials, below 200 nm means 6 ev or more per photon which is enough to kick electrons out of many
molecules (12 ev will ionise most molecules).
Again, there's a reason you don't see a method that seems to be better than the current ones in commercial use, in general because the alternatives
don't work better than the old standbys (in a commercial sense - ease of fabrication, reliability, reproducibility, and so on)
|
|
|
aliced25
Hazard to Others
Posts: 262
Registered: 31-7-2010
Member Is Offline
Mood: No Mood
|
|
I'm looking at calcium tungstate, given that we have numerous accounts that it is possible to either X-ray or XUV from the noble gasses in atmospheric
plasma discharges. Calcium tungstate seems to have an extremely wide emission band (200-500nm according to some sources) from X-ray excitation. I'm
trying to find the wavenumbers of the shift, but it is hard to get at, presumably there is a search term I'm not aware of.
PS I'm also going to ask, does anyone here know of a compact MID-IR, Broadband (4000–400 cm<sup>−1</sup>/2.5-25 µm) light source?
The Chalcogenide fibers are out "NOW", the moving stage will be an issue, but if I'm going to fuck around trying to look at that, the sensing of the
UV-NIR just got a whole lot simpler, I'll FT that in the same way (possibly in the same instrument, in which case Xenon Short-Arc's would be the best
bet).
PPS I've been looking HARD, quite frankly a lot of the "old standbys" are still in use because (1) Scientists are by their nature, ultra-conservative
(2) They are exotic enough for the Companies making them to keep selling them at incredibly inflated prices; and (3) make no use (except for some
minor exceptions) of the improvements in technology in the last 20 years (which actually, on inspection, appear to surpass that of the previous 20
years).
Quite frankly, if Ocean Optics hadn't come out with a Fiber Optic Spectrometer 20 years ago, they still wouldn't exist. It is a bizarre market in
fact, there seems to be some tacit "price-fixing" anti-competitive behavior going on. Unfortunately, the benefits of technology would lower the entry
cost to the market and open it up to those who need these machines and that is being resisted by the existing players... Why? The only thing that
comes to mind is that they want to hang onto their market share, which is bizarre. If they lower the entry cost (and cost of ownership), the market
would grow many times over, maybe they worry about actually having to compete on a level playing field?
It's funny actually, once we get through the Far-Infrared, we encounter the MW Spectrum, and the MW Spectrum can be used to generate X-rays... Running
round in rapidly decreasing circles
EDIT
Found these, small TO-5-like packages, wire-wound - go from the NIR to the end of the Mid-IR :
If the output was split, then the two equal beams were put into fiber (Chalcogenide fiber) this could be used very easily in a fiber-interferometer
(the beam splitter is before the fiber-coupling - so 2 fibers come out).
Anyone got any suggested reading on FT-UV-NIR Spectroscopy? I'm thinking if I have to use a moving stage for one, it would be an awful waste of a
moving stage not to try and get both at the same time (In fact, I think someone is actually already offering that).
In terms of the microbolometers, I'm wondering if I could use a rotating array of them and just have one placed in front of the output beam (lock the
rotation to the stage movement). However long they take to come back to nil reading, is going to determine how many bolometers are needed. I'm happy
enough to mount a peltier, with forced air cooling, but it isn't going to cool 'em down as quick as all that I'd have thought.
[Edited on 16-12-2010 by aliced25]
|
|
|
aliced25
Hazard to Others
Posts: 262
Registered: 31-7-2010
Member Is Offline
Mood: No Mood
|
|
IR Bolometer Array Cameras for Automotive Use could really fuck things up as far as sensors go... Imagine how fucking cheap they'll get quickly
once they're needed in the automotive (one of the major drivers of low cost electronic components) industry?
Shit, a cheap sensor, a fucking quick one too, a diode-type blackbody light source and a hovering (ie. no friction) moving stage (which would not have
to stop given the speed of the camera) and IR optical fibers... Prediction, amateur FT-IR Ownership is going to become a reality this decade.
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
"once they are needed..." And get into volume production, a research paper can be 2 to 10 years away from full commercial production. Cost? maybe
around 50 Euros, at least to start with. Small in terms of the vehicle price; consider the cost of airbags.
Note that these are intended for longwave IR, thermal sources, are may be tuned for a particular band as opposed to the 3-16 um for GP IR. Also note
they use a Si window, a good choice for the operating environment as Si is pretty hard and inert, however see the Si transmission curve at http://rmico.com/technical-notes/bk7-quartz-ge-si
Maglev stages are not new, magnet arrays just allow them to not have big bulks of iron in the fixed stage. They can be annoying to get to work well
for small steps, try building one simple for technology development first, use a red laser diode or low power HeNe for measurements., before going off
and building instrumentation around one.
As for | Quote: | (1) Scientists are by their nature, ultra-conservative (2) They are exotic enough for the Companies making them to keep selling them at incredibly
inflated prices; and (3) make no use (except for some minor exceptions) of the improvements in technology in the last 20 years (which actually, on
inspection, appear to surpass that of the previous 20 years).
|
If the first were true then Ocean Optics would not be anywhere nearly as successful as they are. A researcher isn't going to give a tinker's dam as
to how the light source works, so long as it is dependable in performance and reasonably easy to set up and use. Companies are often looking for ways
to reduce production costs, if a particular field has been footdragging then startups tend to spring into existence. Factors you are leaving out
include the availability of a technology and its relative performance, and patents which can look out late-comers to a new technology. Right now I'd
say you've little exposure to the manufacturing world, which also can be conservative but at the edges is pretty radical and cutting edge, and the
constraints that affect it.
Hmm - a rotating array of low output device is going to be interesting to build, noise pickup will be a real problem; it'll be a bit more complex of
an assemble than you might first think.
|
|
|
aonomus
Hazard to Others
Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline
Mood: Refluxing
|
|
Such LEDs do exist, but they cost alot of money.
See:
http://thorlabs.com/NewGroupPage9.cfm?ObjectGroup_ID=2814
http://thorlabs.com/navigation.cfm?Guide_ID=2102
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
One more line commercial microspectrometers
http://sales.hamamatsu.com/en/products/solid-state-division/...
http://www.eureca.de/neu-english/optoelectronic/microspectro...
|
|
|
franklyn
International Hazard
Posts: 3026
Registered: 30-5-2006
Location: Da Big Apple
Member Is Offline
Mood: No Mood
|
|
The region from long wave infrared to millimeter microwave remains undeveloped.
In the microwave frequency range devices are not light emitting diodes. The Gunn diode
is the semiconductor device used for oscillators that output at sub millimeter wavelength.
The emitter itself will be an antenna or waveguide, as in the signal generator of cell phones.
Devices that can image in this spectrum operate at cryogenic temperatures or else the
blackbody background noise will swamp the signal. Why it remains a technical challenge.
http://en.wikipedia.org/wiki/Gunn_diode
|
|
|
aliced25
Hazard to Others
Posts: 262
Registered: 31-7-2010
Member Is Offline
Mood: No Mood
|
|
Sorry, been busy (fucking families & Xmas, anyone want to guess why it is called boxing day here? )
I was just looking at this paper again, it intrigued me from the very beginning - now one problem I am having is finding a decent (read sufficiently
cheap) source for imaging the UV-NIR in one instrument.
Now, the author's in this paper suggest that with the use of fluoride based prisms, the range could be extended down into the Vacuum UV (the absolute
limits of current fiber technology, Sapphire/Doped Silica at that and a LOOOONG way from cheap), and also right up beyond the capacity of either
standard CCD/CMOS imagers... Now there are patents, granted...
They aren't terribly well written and are a combination of being so vague as to be utterly incomprehensible or so literal, exact and specific as to be
avoided rather easily (the more words you have the bigger the loophole, you should know this already, most are speculative at best - trouble being,
they make claims they cannot substantiate using the systems described, probably not a fantastic move given that credibility is all in such
disagreements.
@not_important...
Look at the prices, I'm not saying the technology isn't there, I'm saying it is hell overpriced. Not one of these instruments is as complicated as the
average DVD Recorder, the reason they are so much dearer? Gee, wouldn't be anything to do with a reluctance to dip beyond a certain point? Maybe the
fact the market is stagnant? Go look up the price of a bare bones laser meter........ But look at what the effect of the mobile phone & automotive
markets is on electronics, continual advances make some rather juicy equipment, obsolete for the original purpose, available at rock bottom prices
soon after it is released.
@frankyln,
I'm onto a source of Xenon Short-Arc's for 1/10 the market price (75-350W), provided I can work out a power source, I'll go with the same light that
Perkin-Elmer prefers...
Attachment: Luet.Boher.Leroux.Imaging.Polarization.Interferometer.for.Flat.Panel.Display.Characterization.pdf (316kB)
This file has been downloaded 780 times
Attachment: Padgett.etal.SinglePulse.FourierTransform.Spectrometer.having.no.Moving.Parts.pdf (623kB)
This file has been downloaded 964 times
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by aliced25  | ...
@not_important...
Look at the prices, I'm not saying the technology isn't there, I'm saying it is hell overpriced. Not one of these instruments is as complicated as the
average DVD Recorder, the reason they are so much dearer? Gee, wouldn't be anything to do with a reluctance to dip beyond a certain point? Maybe the
fact the market is stagnant? Go look up the price of a bare bones laser meter........ But look at what the effect of the mobile phone & automotive
markets is on electronics, continual advances make some rather juicy equipment, obsolete for the original purpose, available at rock bottom prices
soon after it is released.
...
|
Those are instruments, not consumer goods. Just saying "less complex than..." overloooks many aspects. Low volume is on aspect of the price
difference, and there is certainly less price pressure for the scientific app gear, however there are other reasons.
The sensors are significantly better than those in consumer goods, read the specs carefully. Thinned-back CCDs are not cheap, but give much better UV
response than standard ones. Even the less fancy CCDs are lower noise and leakage than consumer goods; both important to low light levels. And I
doubt there's are many consumer cameras with TECs built in. The gratings are much better than those in DVD players, where all they do is work on a
narrow set of wavelengths.
You need to spend a few years in the electronics industry to get a better grasp of physical reality. Consumer goods only need to be good enough to
fool the ear and eye; research instrumentation often needs to record as close to reality as possible. As an example, for decades telephony used 8-bit
companded PCM at 8 Kbaud, and it sounded OK to most people. But that compressed 13-bit range could not properly transmit a pair of tones where one
was much lower in amplitude than the other. Cheap and effective, but often not useful for instrumentation applications - I've had to fix designs
where low cost highly integrated telephony A/D & D/A converters were used for something other than telephony.
Even gear for process control often is specialised and of limited range, as it is not uncommonly working with known composition samples where one
aspect of the sample is being monitored. This was the case for much of the VNIR stuff, and for many of the small and open coil NMR stuff too; they
weren't for general structural decoding but rather for tracking a particular type of something-H in a sample, distinguishing the relative ratios of
C-O-H vs -C-NH, or monitoring moisture content.
I suggest reading the following PDF http://sales.hamamatsu.com/assets/applications/SSD/fft_ccd_k...
|
|
|
aliced25
Hazard to Others
Posts: 262
Registered: 31-7-2010
Member Is Offline
Mood: No Mood
|
|
Yes, it is true that the "size" of the market dictates the "low volume" aspect - not something I intend to launch into here, suffice to say, I'd
seriously and quite simply state that the size of the market is dictated by the cost of the equipment. There are a lot of Customer's out there who'd
prefer to use instruments, but simply cannot afford to do so.
As to the cost of specially backthinned CCD/CMOS Sensors, etc. there are numerous startups working on 8" blanks who are willing and able to build
pre-thinned sensors for a whole lot less than the majors are asking. The same would go for the majority of the electronic, electrooptic, light-based,
optical, etc. components. Even purchased singly there is often anything up to a 10,000% markup over what it was purchased for by the original
middlemen, I've seen quotes for sharp cut-on filters made from AR Coated Schott Glass for <$0.50/100 units, given the sales price per unit is ~$50
(if you look hard - there is better, but not by much), the sheer effrontery of the industry leaders is astounding. Put simply buy 100 units and still
be 99% less out of pocket than if you'd bought 1 from the majors.
This is why it is possible to look at building something that costs a whole lot less that has the equivalent (if not conquerable) specifications as
the products coming from the majors, not because of any short-cuts on quality/accuracy. As for the majority of the improvements that have come
through, they come from recently declassified US Army/Airforce/Navy documents. There are some interesting articles indeed on the use of windows for
the mid-UV-mid-IR which would be subjected to heat, describing the incredible structural/engineering changes in Sapphire when heated (and the inherent
magnification of flaws & structural weakness found therein). Most of this research is still languishing unused, there is a hell of a lot of it
online (search for various terms with "site=dtic.mil" amongst the terms - although please don't dismiss NASA either), finding it doesn't take any
special tools except a google search engine and a modicum of brains.
The Russian Journal Articles that are on Springer for nix aren't such a bad place to look either, especially for luminophors / dosimeters, that
coupled with the XUV-XRAY output of hollow cathode lamps in noble gasses may offer a particularly low-current UV-A/B/C light source (I mean using
phosphors with plasma isn't precisely patentable is it?).
As to the use of Xenon Flash-Tubes, they are about as inexpensive as it gets, the use of the slowest of them with a moderately fast camera will enable
the binning of most frames (which have nothing but fluorescence information - longer lived than the absorption - or maybe using some of it for
calibration of the adsorption output) will give the data that costs so fucking much at the moment. As for UV Phosphors, there really are quite a lot
of them, working out how to use them on a thin film before the camera might be a worthwhile activity.
@franklyn,
As to imaging in the IR - it hasn't been missed, it has been classified up until relatively recently. Go have a look, uncooled microbolometer and
microthermopile arrays do exist, just the price is still quite high, because up until now the only lawful customer was the US Military, which will
gladly pay $5,000 for a bloody hammer.
[Edited on 26-12-2010 by aliced25]
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
That comment on microbolometers - I think that it is nonsense, they're mentioned as sensors for lab IR gear going back decades. Also see
"Microbolometers for infrared detection by TL Hwang - 1979 " Maybe so for some modifications, including arrays of them for imaging, and indeed
imaging is useful if you want to do the no moving parts FT-IR (but check the resolution needed). There are new designs using newer fab techniques,
but I suspect it was as much the technology as the military governing the availability of those - lot of stuff the military classifies overlaps with
research published elsewhere - like the cover of a journal showing a simulation of laser-driven implosion fusion that was classified in the US with
the cover being torn off of copies out in non-restricted areas.
I'd like to see a price list or quote of the thinned back illuminated CCDs, ones that come near matching the performance of those in that PDF. Ditto
for the specs on the cheap glass filters - ones that you've talked about in the past both block low shift Raman lines and lacked sufficient OD for the
laser line. Easy to say "there are cheap equivalents" but proof is showing them as actual products or hard quotes.
| Quote: | | I mean using phosphors with plasma isn't precisely patentable is it? |
I've earned money doing patent searches - you'd be surprised at what can get patented. Specific excitation sources, phosphors, wavelengths produced,
exact structure of the device, and much more can be used to take an old concept and turn out a patentable new product.
|
|
|
aliced25
Hazard to Others
Posts: 262
Registered: 31-7-2010
Member Is Offline
Mood: No Mood
|
|
The use of uncooled microbolometer arrays is extremely sensitive, I'm talking about the larger, imaging arrays... Actually have a look.Vanadyl Acetylacetonate degrades thermally to the necessary VO<sub>2</sub> needed for the major imaging uses - they aren't that old,
NASA is still working on them and IIRC they still haven't got past the couple of K pixels. There are also carbon ones based upon the thermolysis of parylene (Liger, Matthieu (2006) Uncooled carbon microbolometer imager. Dissertation (Ph.D.), California Institute of Technology). There are also negative
photoresists that also give pyrolytic carbon on thermolysis (SU-8 type), which may make a great deal more sense than CVD (given the photoresist is
already in use in the process) Microbolometers are most assuredly not new, the uncooled, imaging microbolometer array is not something that is all
that old. For low-cost instruments, the 4*4 pixel array setup would be great, a digital 8-bit output from each pixel could potentially carry a
shitload more information than a single cooled bolometer (the process is essentially the same as that of CMOS sensors apparently, they need a bias
current, but as a result, can be read out a lot more easily than a micropile array).
As for patent searches, the difficulty with patents is that most people try very hard to include as much as they possibly can, in order not to limit
the fucking things. The less words, the less restriction, every sentence has a hole in it (a rule to live by). A hollow cathode X-Ray-XUV-VUV
light-source (Xenon or Argon - one is 100 times more expensive than the other, which to choose) exciting a broad output UV-Emitting phosphor, with essentially nothing more in the way of a power supply than a
diode? That is huge, working out which phosphor? That is the trouble. As for patenting it, be kind of dumb not to, no? Ethical question, should I
patent what I cobble together based upon research other people carried out? I'm trained in law not ethics & patents are legal.
As for the resolution question, there are some fucking interesting papers out that show that a flip-chip type design could be used to build the IR sensor on top of an already fabricated CMOS Sensor - which would
open up the 2-5K pixel width device gate, which if it were square would also allow for "pushbrooming" (NASA's term, not mine, there are some nice
signal-noise algorithm's but, thanks be to god). A gold coated holographic grating for the IR and a no moving parts, no fucking around, instant IR
Absorption Spectrometer? Something to dream about at least...
As the Golay Cell is used in a lot of older/high end systems, and it works by heating carbon (essentially, it actually uses the heating of the carbon
to heat the xenon gas, which expands, yada yada yada - yeah that bloody Golay again) using incident IR Radiation, then I suspect the Carbon-film
model would be better for Instruments (it images the relevant part of the EM Spectrum).
[Edited on 30-12-2010 by aliced25]
|
|
|
aliced25
Hazard to Others
Posts: 262
Registered: 31-7-2010
Member Is Offline
Mood: No Mood
|
|
I recall posting the Schott Glass Transmission data on the colored glass filters, the transmission is not perfect and it is a longer-slope than is
really going to be useful at the lower wavelengths/cm<sup>-1</sup>. That said, up at the red end of the spectrum, not very much will be
lost at all, the only issue will be correcting for less than perfect transmission over around 5-10nm.* That said, the price is significantly lower
than the multiply coated fast-cut-on filters, which may be useful for some purposes (particularly down the other end of the visible-spectrum), but
provided allowance is made for the variable transmission in the colored-glass, it'll do (plenty of people have used it before - I rather think that
might be why Schott make it?).
The uncooled microbolometers are a long way from a "normal" bolometer, the reason I'm picking up on the carbon-film is that the Golay Pneumatic Cell
(Golay, Marcel, 'Pneumatic radiation detector' (1968) US Patent 3,384,749) which has been used in what, if I recall correctly, are some of
the most sensitive units for the relevant region due to the fact that carbon has an excellent response to carbon as a radiation absorbent in the
short-mid IR (whereas the majority of sensor designs using uncooled arrays use VO<sub>2</sub>, which is great in the Long-IR, but not real
good at the others). He used soot, but as a lot of soot producers are now being shown to actually grow layers of graphene on the surface (including
some really basic ones like camphor) and soot-impregnated fiber to make the patent claim.
* As the relevant wavelengths would be well and truly known, the software would be designed to convert all of the incident light to the theoretical
100% (taking into account also the fall in relative/perceived intensity over wavelength). In terms of phosphors, simple ones like the vanadate-type
(which do emit in the 200-400nm region when hit with extreme UV/Soft X-ray excitation) look like the simplest route.
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
