Talk:Monochromator

AU/OD

Who calls it AU!? It's always OD!

Yes, it seems odd ( OD ? ;) ) to me too !!! --Totophe64 13:25, 10 April 2006 (UTC)

Excellent article

I just want to say that this is, in my opinion, a truly excellent article that exemplifies the best of what Wikipedia can be. When I came here, I had a vague notion that a monochromator was a device for producing monochromatic light and probably involved prisms in some fashion. When I had finished reading the article, I felt I understood monochromators quite intimately - their origin, construction, use, and significance. The process was painless, which means in effect that the Wikipedia article was serving as an optimally impedance-matched source of knowledge for my (educated, but non-specialist) mind. Congratulations to the authors of this fine article. 198.202.68.51 (talk) 18:10, 24 October 2008 (UTC)

wrong about monochromator collimators

I am not going to take the trouble to find a reference right now, but, believe me, the collimators for the instruments I am intimately familiar with are spherical. These are the Applied Physics, later Varian-Cary, double prism and prism grating, .4 meter focal length instruments, which are considered high precision for chemical/biological work. When Aviv started to manufacture their own monochromators in the 1990's, paraboloids were under consideration because their cost had dropped, but I do not know if they were actually incorporated in the final product. There are a lot of tradeoffs, spheres are generally good enough. The real trick in the Cary design is the kinematic mirror mounts that make alignment shock resistant. The challenge for performance is stray light, not resolution. --AJim (talk) 05:03, 27 October 2008 (UTC)

I can only give a counterexample of a spectrograph that I have worked with: Oriel MS260i spectrograph: "... uses toroidal mirrors. instead of the more common but poorly imaging spherical mirrors" [1]. The difference is of course that this is an imaging spectrograph that has proper imaging of points along the slit, which requires differing vertical and horizontal radii of curvature of the mirrors. You can use a spherical mirror to have good horizontal imaging (one wavelength ending up as a narrow vertical line in the image), but it will be blurred vertically in that case if there were several light source along the entrance slit. We might specify the pros and cons of toroidal/spherical optics in this article, although I'm not sure where to find references for that. BTW I find the distinction between this article and Spectrometer a bit artificial. Han-Kwang (t) 09:09, 27 October 2008 (UTC)

(1) You are right that the image is curved. Cary solved that problem in the 50's by using curved slits, not by using torroidal mirrors. I had a long talk with Roland Hawes about it in the 80's. He said the correct curvature would have been parabolic, which is hard to make, but that a machinist showed them how to (2) easily cut eliptically-curved slits on a lathe that were good enough. The design issue is energy, how much light is admitted. Curved slits can be higher for the same resolution. They published a resolution analysis in the Cary 14 manual and included a slit-height reduction shutter for those (rare) situations when the full slit would not have enough resolution. They split the difference in the curvature, the entrance and exit slits curve in opposite directions, and the intermediate slit, between the monochromators, is straight. The quick performance test for these instruments was a UV benzene vapor spectrum. There was never a problem with resolution; the relative heights of the peaks showed the stray light performance. I want to stress that low stray light is the main challenge, and stability a very important practical consideration. These instruments were very good at that. I read the engineering report that evaluated holographic gratings when they became available against the blazed gratings they had been using. Cary switched to holographic gratings because they had much lower UV stray light. I built new instruments using the 14 monochromator and optics in the 80's and 90's. The only design change to the 14 optics we made was to add a chopper before the entrance slit on some instruments to get lower stray light in the near IR. As for the related articles, let me just say that there is always room for improvement. I think that the monochromator concept is important enough, though, as a scientific tool, that it deserves an article on its own. There is a great deal more to say about monochrmator design, and there are a number of variant designs, as you know. There is a great deal more to say about spectrometry, spectrometers, and spectrophotometers, too. (3) By far the most common spectrophotometers are relatively cheap designs with hand-tuned single monochromators, and I think they must use spherical collimators, although I have never asked. I revised the Czery-Turner description because the type of collimator is irrelevant there and broke out slits as a section and addressed the issue there. I think I am on the right track. Do you agree?

--AJim (talk) 19:45, 29 October 2008 (UTC)

That's a long paragraph where you mention several issues. (1) Curved image plane with a spherical collimator. I think we are talking about different problems here. If we assume that the slits are vertical, then the top of the slit has a larger distance to the center of the mirror than the center of the slit, and hence will get focused closer by. This is probably what you are referring to. However, this is all about focusing in the horizontal plane. I was talking about vertical focusing. With a spherical mirror, even with a curved slit, a point source on the entrance slit will be imaged as a vertical line segment on the monochromator. A vertical focus (horizontal line segment) will appear closer to the mirror. This is an entirily different issue from that of a curved image plane. It is because if the mirror is not oriented on-axis (as is the case in the Czerny-Turner design), different horizontal and vertical curvatures are required. (Think of the local curvature of an ellipsoid that images a point into a point). In a normal monochromator it is not a problem if there is a vertical line focus, since one is not interested in the intensity distribution along the slit. But in an imaging spectrograph, e.g. with several fiber sources on the entrance slit and a CCD array in the image plane, this is relevant and requires non-spherical optics.

(2) Interesting, I'd think at first that a lathe is not particularily suitable for making an ellipse, but I guess it's about cutting a tube at an angle to get an elliptical cut.

(3) Depends on your field. Compact, cheap CCD spectrometers are quite popular for some applications. [2] [3], although they don't give you a 10^6 resolution.

Han-Kwang (t) 10:21, 30 October 2008 (UTC)

About (1), when I asked Roland about things like that he very nicely made it clear that he knew the instrument was not perfect, but that it did not matter, it was good enough.

About (2), what I remember is that they made a stack of brass plates and cut the slit edge (either positive or negative curvature) into a bunch of them at once. The cut was at an angle and so also produced the knife edge.

About (3), I agree that detector arrays have made inroads. However, I am thinking of the Spec 20 type instrument, as diagrammed here, which has been made in large numbers for a long time, is very cheap to make, and is good enough for many routine analyses.

In fact, I wonder if the encyclopedia should not have more information about making simple instruments of this sort, because even a Spec 20 is out of financial reach in a lot of places.

--AJim (talk) 21:03, 24 December 2008 (UTC)

Errors from monochromators

Could a section be added that discusses instrumental errors that can be caused by monochromators? I'm thinking particularly of transmission as a function of wavelength and accurate wavelength calibration. For wavelength calibration I think it is worth noting here the use of the atomic emission lines of mercury and the absorption spectra of Holmium Oxide (Standard Reference Material 2034) detailed in NBS 260-102 as a wavelength calibration. We could also discuss effects due to spectral bandwidth such as wavelength inaccuracies and the effects on spectra like broadening of peaks. Nwagers (talk) 02:48, 13 June 2009 (UTC)

You mention two topics, wavelength calibration and the effects of the spectral transfer function.

As far as wavelength calibration goes, the article is necessarily very general because it covers so many possibilities, but I do not see why you could not add a section. I think that in the past most monochromators were first tested with a mercury pencil lamp placed in the entrance slit and were adjusted to the 546 nm green line. Especially for a grating, a single wavelength calibration is pretty good because most of the remaining error sources will be mechanical. I think that, for high resolution monochromators, a great deal of design effort is expended to produce a linear drive mechanism, for instance. I think that some manufacturers now use lasers instead of mercury lamps for alignment (and initial calibration). I think what you are aiming at though is to identify ways that a monochromator user could check the wavelength accuracy of an instrument. You mention some standards. I think you should give it a try. But I think in many cases you may be talking about a particular instrument instead of about a monochromator itself. I think another performance testing tool often used is vapor absorption lines, which give information about wavelength accuracy and bandwidth. Line absorption ratios can also be used to get information about stray light.

The second topic is less clearly relevant here, but generally quite relevant. There is a start, for instance, the triangular transfer function is described and stray light is mentioned. A picture would be nice. Some of this is related to particular uses. Ultraviolet-visible_spectroscopy#Spectral_bandwidth touches on this, for instance. The theoretical ideas, for instance that the measured transmission spectrum is the convolution of the monochromator transfer function with the sample transfer function is not developed. Does it belong to an article about the theory of spectroscopy? --AJim (talk) 21:57, 13 June 2009 (UTC)

The illustration labelled as a "Fastie-Ebert" monochromator is in fact of a Littrow mounting. The Original Ebert mounting, re-discovered by Fastie has the entry and exit beams disposed on either side of the grating. — Preceding unsigned comment added by 217.43.64.181 (talk) 22:27, 12 June 2012 (UTC)

External links modified (January 2018)

Hello fellow Wikipedians,

I have just modified one external link on Monochromator. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20080926072418/http://gratings.newport.com/library/handbook/cover.asp to http://gratings.newport.com/library/handbook/cover.asp

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 15:59, 22 January 2018 (UTC)