| Author | Message | ||
     weyer |
I have been using a Foss/NIRSystems interactance probe (the circular one with the little feet on it) to examine a coating on aluminum sheeting. The probe is usually tilted a bit from 90 degrees. When there is no coating, I observe the spectrum of an organic matrix. It can be positive or negative probably depending on how I tilt the probe with respect to the sheeting. The peak heights are about 0.01 absorbance units, which is more than the actual coating. Can anyone tell me what I am looking at with the clean sheeting? Is there any organic substance between the window and the fibers which might come into the light path under these conditions? | ||
| David Hopkins (Hopkins) |
Lois, I can think of 3 possibilities, maybe the last is best. First, this is difficult to address without seeing the scan. I wonder if it could be due to mismatch of the sample and baseline optical paths. If the baseline is with a nice diffuse reflector, and the sample is more like a specular reflecting surface sending a lot of light back to the detector, you could be seeing features of the reference in your �sample� scans. This effect might be enhanced by a contaminated reference. I have read that Spectralon is very avid to pick organics. Is it possible that your reference has picked some contamination? Maybe you could see it by scanning a fresh (different) reference material. Perhaps the problem would go away if you tried a thick ceramic reference. You can buy ceramic disks directly from Coors. Their website is www.coorstek.com. I assume you know where you can obtain Spectralon disks. I have observed this effect with the probe, when there is a possibility that ambient (especially fluorescent) light can enter the optical path. A series of oscillations appears that look a lot like interference bands or a series of sharp bands that on first observation look like organic bands in the 1720 or 2300 nm region. There was some kind of artifact that we could eliminate by running scans with the environmental lights off and constructing a hood or tepee around the sample and probe, to block ambient light. We made measurements in a fume hood with the lights off, under a hood constructed of black plastic film cut like a teepee. I would be interested to hear whether any of the suggestions help you. Best Regards, Dave Hopkins | ||
| Art Springsteen |
RE:Dave Hopkins comments: The Spectralon (sintered PTFE) contamination option is possible but somewhat unlikely. Typically what you would see is the inverted features of the contaminant in teh Spectralon in your sample scan, since the 'contaminated' Spectralon was used to set the 100% line. That being said, and I do have a little expertise about sintered PTFE  , I've never seen a lot of features in the NIR due to contamination. Typically they show up in the UV, and then only as a decrease in reflectanec below 320 nm. If the sinteerd PTFE is showing contamination features, it's probably visibly contaminated and you wouldn't be using it as a standard anyway... I'm not sure option 1 of Dave's is a good choice either- I don't think you'd see features. Number 3 is a possibility. Lois didn't mention what type of features she saw- might they be the absorbances in sintered PTFE- they typical C-F overtones? | ||
| Art Springsteen |
I'm not sure I agree with Dave Hopkins on the 'contaminated Spectralon' issue. Yes, sintered PTFE can absorb contaminants from the environment, but typically they are at VERY low levels. You might see them in the UV but I never saw them in the mid-IR unless the sample was grossly contaminated- to the point where the reflectance everywhere was low. Dave's option #3 is possible, while his first option is again a possibility- what you may be seeing is the inverted absorbances of the C-F stretches in PTFE. Just a thought... | ||
| Lois Weyer (Weyer) |
Dave and Art, Thank you for your comments. I am using clean aluminum foil (stiff foil) as a reference, however, so no PTFE, ceramic, or any other reference material is involved. I have tried shutting off the lights and wrapping the probe end and sample in black velvet at Dave's suggestion, but the anomaly did not go away, or even decrease. I'm back to wondering about the optical path beyond the window, and have inquired of the vendor to see if they can give me any clues. | ||
| art springsteen |
Lois, The aluminum foil may have a coating on it. And black velevet is a notoriously bad actor in the NIR. The reflectance of a number of 'black' velvet's I've seen have been quite high- they must use a metal salt as a pigment. Still, it's probably something in the probe. | ||
| Lois Weyer |
I have a new puzzle I need help with. I'm trying to understand how a lamp change affects a spectrum. I have a Powerpoint file of 4 slides to illustrate the problem and will send it to whoever would like to see it. The slides show average spectra of the same, liquid, samples taken after a lamp change and a prealigned-lamp upgrade. The strong peaks show that the pre-aligned lamp modification caused a loss in linearity. There is also a small amount of moisture increase in the last series. However, neither of these effects are important to my analysis. The important changes appear to be a decrease in the intensity of the major bands at lower wavelengths and distortions above 1600 nm, especially in the 2000 nm region, with an isosbestic point at 2014. I need to measure a very small change near 2030 nm. Any ideas? | ||
| Art Springsteen |
Hi Lois, Sounds like your pre-aligned lamp wasn't very well pre-aligned. We see this on our Lambda-19's. The new lamp should give a spectrum that is the same or have slightly better intensity over the whole wavelength range (I'm talking the W/X source). But it doesn't- it will have distinct differences from the previous lamp if one looks at the raw energy scan. Now, in the double beam instrument, this all washes out in the background correction, but that may not be so for your instrument. Can't explain the moisture difference but the linearity problems may be due to shifting of pre-amp gains as the lamp intensity changes over the spectrum. Doesn't solve your problem but... I might try another lamp and see if that helps alleviate the problem. | ||
| Gabi Levin |
Hi Lois, About lamp change - I would like to see the spectral information. Another issue - is the spectrometer single channel or dual beam? If dual, how is the dual setup work? [email protected] Sorry I can offer help with the coating on aluminum. How thick is the coating? Thanks, Gabi |