| Author | Message | ||
     Gabi Levin (gabiruth) Advanced Member Username: gabiruth Post Number: 25 Registered: 5-2009 |
Hi Howard - I am not sure right now and I am away on business - but I can write directly to you - as I am not sure how much will be still consdidered non commercial posting. Please send me your mail address - mine is [email protected] | ||
| David Adamson (davida) Junior Member Username: davida Post Number: 6 Registered: 5-2009 |
What I have found is that if using a thin cling wrap film there is a definite risk of getting these interference patterns and sometimes they are so subtle as to be nearly unnoticable. This is a problem as it upsets the results with quite a few false negative identifications. One has to be very, very careful to have the film thoroughly pressed onto the surface of the probe with not even the slightest wrinkle in it and this is too time consuming. So our practice in the testing of pharmaceutical materials is not to cover the probe when testing harmless excipient materials but scan through the plastic liner when testing toxic active materials. As the liners are much thicker material we do not get the fringing, ever. The two spectral regions where the liner produces peaks are easily eliminated from the chemometric assessment without affecting the overall result. If we measured through the liners for the excipients there would be a problem as often the distinguishing peaks are in the same regions as the plastic liner peaks. Cheers | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 278 Registered: 9-2001 |
Gabi - that sounds interesting. Is it something you can talk about, or is there a patent describing how you do that, that you can refer me to? I suppose that in my previous posting I should have also mentioned that the thicker the film is, the more effective the averaging of the different pathlengths becomes, so that the fringes become less and less prominent. In fact, by the time the "film" has become so thick that it's better called "sheet" or even "plate", it becomes almost impossible to create fringes even if you want to, and even if you're not measuring using tranflectance. This is because the natural divergence of illuminating light will cause sufficient path length differences to create the averaging effect. Tony - the main point I was trying to make in my previous post was that interference fringes are really very delicate constructs and many conditions have to be met in order to create them. Therefore they are correspondingly easier to destroy. The mid-infrared crowd is very familiar with this, since because of the longer wavelengths they work at, the more severe their fringing problems are. Interference is generated when two different light rays coincide. This has to be done on a scale that is smaller than the wavelength of the light involved. What we call "fringes" are generated when the interference varys between constructive and destructive, at different wavelengths. Incidentally, this is NOT what optical engineers and physicists call "fringes", although they are related. But let's not go into that. Because of the small scale at which the light has to interact coherently, the interference effect is easy to disrupt. For example, when you said "transflectance", I assumed there was a diffusing reflector behind the sample. Anything diffuse about the light will disrupt the interference phenomena and prevent fringes from forming, so the diffusing reflector would be one way to do that. Therefore, for example, another way to prevent fringes would be to roughen one or the other (or both) surfaces of the film. Still another way would be to use a diffuse source. Even without diffusion, the interference could be prevented by not having the two surfaces be plane and parallel. A way to prevent fringes from a window, for example, would be to make the window slightly wedge-shaped. Still another way would be to use illuminating optics with a low f-number, so that the light will hit the sample, or the film, over a wide range of angles, simulating the effect of diffusion. If the sample is a liquid, then matching of the index of refraction to the film would reduce or eliminate the reflection from one or the other surface of the film, thereby eliminating the interference by eliminating one of the interfering rays. \o/ /_\ | ||
| Tony Davies (td) Moderator Username: td Post Number: 204 Registered: 1-2001 |
Hello Howard, Two questions about your post. 1)"So it's not that reflection cancels transmission". ?? What did you mean to say? 2) I was talking about transflection with just clear polymers. There is not much diffuse reflection occuring within the sample so I would not expect the returning rays to be "returning at so many different angles". I see the returning rays as giving an equal but opposite fringe effect so that in summation (not averaging) the fringes disappear. Just to go a little further with you rays at different angles. If we assume that the polymer thickness is constant then rays at more or less than 90 degrees will travel further than rays at 90 degrees. If the wavelength of the illumination gives fringes with 90 degree illumination it will not do so for rays at different angles. For the divergent rays to give rise to fringes the wavelength needs to change and thus in a scanning spectrometer this will be observed as a broadening of the fringes. As the variation in the ray angle increases the broadening will increase until the fringes are no longer observed. Best wishes, Tony | ||
| Gabi Levin (gabiruth) Advanced Member Username: gabiruth Post Number: 24 Registered: 5-2009 |
Very good explanation Howard. The interfrence fringes which are significant problem in all thin film, including transparent coatings on metals, production of thin films for food packaging like saran wrap, etc. This is why we at Brimrose designed the thin film analyzer - where we illuminate at a special angle and collect at a special angle. The light that goes in goes at an angle that avoids the reflection from the surface where index of refraction change occur, and it goes out from the film at an angle that avoids the possibility of constructive and destructive interference. | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 277 Registered: 9-2001 |
Good point, Gabi. I think Tony doesn't quite explain the mechanism for the loss of the fringes in transflectance correctly, though. All the interference fringes are caused by the interference of rays reflected from the two faces of the film. When the film's thickness is exactly a multiple of 1/2 the wavelength then you get constructive interference, and when 1/4 wavelength more, destructive interference. Across the spectrum, you get alternating effects because of the changes in wavelength. So it's not that reflection cancels transmission. It's that the nature of the fringe, whether it's on an interference maximum (consturctive) or a minimum (destructive) depends very delicately on the pathlength of the light ray through the film (relative to the film thickness). Rays at different angle will traverse different pathlengths, therefore some will constructively interfere and some will interfere destructively. In transflectance (or diffuse reflectance), the rays will be returning at so many different angles that returned rays that have undergone constructive intereference will be averaged with rays that have undergone desctructive interference, and all will average out to the same intensity. Also, if the sample is liquid, then refractive index-matching on one side of the film will reduce or eliminate the fringes, because it will eliminate the reflection from one side of the film. \o/ /_\ | ||
| Gabi Levin (gabiruth) Advanced Member Username: gabiruth Post Number: 23 Registered: 5-2009 |
Hi Tony, I have seen fringes in transflectance as well - when the backing of the film is like mirror - I did not have an opportunity to look at powders or slurries through a thin film. In the evnets that we run into "film" the PE is thick - bags of pharma raw materials etc. and then there is no propblem. The other issue with using a film is that it becomes part of the spectrum - and needs to be corrected for. Gabi | ||
| Tony Davies (td) Moderator Username: td Post Number: 203 Registered: 1-2001 |
Hi Gabi et al., I did some work on identifying and quantitating components of plastic laminates [Analyst, 110,643 (1985)] and did not have any problems with interference fringes. I later discovered that this was because we were using transflectance - the only mode available with our Neotec 6350 Mark 1 spectrometer. When transmission measurements became available the fringes messed-up the spectra! Luck was on our side in 1984! I assumed that the fringes in the reflected energy cancelled the fringes seen in transmission. I think this worked because the absorptions by the film were very low and thus there was little difference in the energy levels of the input and reflected (apart from the fringes). The point of this message is that if you put a film over a stainless steel probe you are going to measure NIR in transflectance and I would not expect any problem with fringes. Best wishes, Tony | ||
| Gabi Levin (gabiruth) Advanced Member Username: gabiruth Post Number: 22 Registered: 5-2009 |
Hi guys, Teflon tape should be tested - truly absorption is negligible - but it does reflect - I don't kow why. Possibly if very thin - could be used. need to test. About other films - all C-H containing films when thin enough - below 40 micron or so will exhibit the phenomenon known as inteference fringes. This will be extremely problematic for any application. Interference fringes stem from the fact that when light goes from air into the film, some of it is reflected due to the difference in index of refraction, some goes into the film, some of that is reflected back from the other side of the film, and due to that, at certain wavelengths you have cosntructive interference and at some you have destructive interference - thus the overall signal becomes like a sine wave - and there is no useful information in the signal. Gabi Levin Brimrose | ||
| rahimi (aarahimi) New member Username: aarahimi Post Number: 2 Registered: 2-2008 |
Dear everybody It would be appreciated if some one let me know that is it possible to detect bacteria in a water solution or blood serum using NIR spectroscopy . Regards Rahimi | ||
| David Adamson (davida) New member Username: davida Post Number: 5 Registered: 5-2009 |
Hi Jerry. What is it that you want to cover? If you are using typical NIR fibre optic probe, these are completely water proof. I can only guess that you want to prevent contamination of the probe or sensing element from a water based solution. Any kind of plastic cling wrap material might work so long as you can tell the chemometrics to ignore regions where the small plastic peaks show up. Trouble is in my experience, is that these materials are not particularly pure and can have a wide spread of small peaks which can be a problem as I have found personally. In my case we use pharmaceutical grade Polyethylene which is pure and consistent and only has peaks in a few places. However, it is too thick to wrap on the probe but can be measured through as a barrier between the probe and the test material. | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 275 Registered: 9-2001 |
Jerry - Art makes some good points. Polyvinylidene fluoride is not completely fluorinated, so it still contains C-H bands. In fact, I have a spectrum of that, and there are some bands in the 2200-2400 nm region, as well as in the 1600-1800 nm region and the 1100-1400 nm region. This has been published, along with spectra of other polymers, in Spectroscopy; 9(1); p.27-32 (1994). But if the plastic is thin enough, the absorbance bands, especially in the NIR, should be very weak, and might be weak enough to not matter to you. In this case, any "ordinary" plastic might work, even Saran wrap itself. I'd try it out, if I were you. Also, I forgot to make another point: PTFE (Poly-TetraFluoroEthylene - capitalization intended to show the origin of the abbreviation) is the "ordinary" white teflon that we think of. But there are other formulations for polymers made from completely fluorinated monomers. One of them that I know about, and was able to get a sample of, is FEP (Fluorinated Ethylene-Propylene) teflon; as the name indicates, it is a co-polymer. But since this material is also completely fluorinated, it also lacks any of the C-H bonds that give us most of our absorbance bands in the NIR. It is also MUCH less scattering than PTFE teflon; whereas PTFE appears white in the visible region, FEP appears only slightly milky, and this amount of residual scattering is also reduced in the NIR. The FEP material is also somewhat softer than the PTFE, and less readily available. You'll probably have to get some from a manufacturer. Also, getting the thin film you want might be problematic. \o/ /_\ | ||
| Art Springsteen (artspring) Member Username: artspring Post Number: 14 Registered: 2-2003 |
Hello, Howard is right, up to a point. PTFE film will have some slight absorbances in the 2150 and 2400 nm range and a very weak one around 1360 nm. But it IS translucent, even down to about 50 microns thick. What about something like poly(vinylidene fluoride)? Commercially available, and not all that expensive. Could could also try something like Saran� wrap, which used to be poly (vinylidene chloride). I would think that if they are VERY thin, the absorbances would be very weak. | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 274 Registered: 9-2001 |
Jerry - good luck! There are not many materials that are non-absorbing in the NIR. Your best bet is Teflon. Of course, it's also translucent rather than transparent, but it should be even better in the NIR than in the visible region. The same stuff is also sold by other companies under other names ("Teflon" is the trade name for Dupont's material so nobody else is allowed to use it) Then your problem will be to get it in a thin enough layer that it will do what you want. The thinnest commercially supplied form that I'm aware of is 1/64 inch - slightly less than 0.5 mm., maybe that will be thin enough for you, both optically and mechanically. But at least it's relatively easily available. I usually get it from Small Parts, Inc (www.smallparts.com); you can buy as little as 1 square foot from them, for only a few dollars. There's also another form that I know about, which is even thinner. This is the material in a very thin tape form, sold by plumbing supply places under names like "Tape Dope". It's normally used to seal joints in pipes; it looks like Teflon and feels like teflon, but I don't know if it's really pure Teflon (under any name). But this is also pretty inexpensive, so it may be worth a try if you don't need too big a piece, since the tape is only about 1/2 inch wide. If you try this out, let us know how it worked out, especially how absorbance-band-free it is. There's also a couple of other plastics that have no absorbance bands. One is a polyether, so that has no hydrogen because all the bonds are C-O-C and they're clear like glass. I forget their names, but you probably can't get them, anyway. I tried a few years ago, and after I filled out and signed a complicated form to get a license to buy the stuff, I found that I could only get it in powder form at a price of about $1,000/gram. I think either the military or the DHS is restricting the sales of it. Probably better to not get involved. \o/ /_\ | ||
| Jerry Jin (jcg2000) Intermediate Member Username: jcg2000 Post Number: 19 Registered: 1-2009 |
Dear All, I am looking for a plastic-wrap like film that is transparent to NIR light and not soluble in water. I would appreciate it if you could provide info where to buy it. Thanks. Jerry Jin |