Home | Survey | Topics | Index | References | Dictionary | Contributing | Gallery | Community
| Measuring attenuation of light: General comments on experimental arrangements | Prev topic | Next topic |
A number of experimental arrangements for measuring the attenuation of light by a medium is possible, each with a distinctive set of advantages and disadvantages. Two such arrangements which rely on the measurement of transmission of light by the medium are briefly discussed in this thread. Details of experimental arrangements for measuring attenuation of light in natural waters are discussed, for example, by Barth H et al (1997), Voss KJ and Austin (1993), Austin RW and Petzold (1977b), Tyler JE et al (1974). Arrangements for and problems with measuring attenuation of light by the atmospheric air are discussed, for example, by Malm WC et al (1986) and Pritchard BS and Elliot (1960).
For small pathlengths, i.e. for substantially attenuating media, the medium is typically enclosed in a cell (cuvette, sample container) each with two plane-parallel windows. It is important to mention that the optical quality of the inner surface of the cell, as well as the inner diameter of the cell, are of essence in the case of long-aspect (diameter << length) cells when the attenuation of light by significantly-scattering media is measured. This is especially important for cells with transparent walls if the refractive index of the sample medium is greater than that of the medium surrounding the sample cells (presumably air). Indeed, light scattered by the medium may contribute to the transmitted light power following reflection at the internal surface of the cell. In fact, if this reflection is deliberately enhanced, for example, by mirror-coating the inner surface of the cell, and if the acceptance angle of the detector is made sufficiently large, the measurement of light attenuation for a light scattering and absorbing medium is reduced to the measurement of light absorption (for example, Pegau et al 1997, see also end of note Acceptance angle of the detector).
For large pathlengths, for example, in the measurement of light transmission by the weakly-attenuating atmospheric dispersions, enclosing the dispersion (medium) in a cell might be difficult. Hence, in this case, it is the distance between self-contained light source and detection systems of a transmissometer which typically determines the thickness of the medium layer (for example, Malm WC et al 1986 - atmosphere, Brander H and Blackinton 1984 - seawater). Note that multipass cells are also used to extend the effective pathlength without undue lengthening of the cell itself (for example, Carleer M et al.).
The detector of transmitted light must have a linear response with a high dynamic range, i.e. the ratio of the maximum to the minimum signal, because the power range may include a low power transmitted by a sample cell and a much higher power transmitted by a reference. At the detector system level, the dynamic range (DR) can be extended by modifying the detector system gain (for example, Lee ML and Lewis 2003, a photomultiplier detector system, DR ~ 1012) and/or the signal integration time (for example, Eppeldauer G and Hardis 1991, a photodiode detector system, DR ~ 1014).
The dynamic range can also be increased with the use of neutral density filters. However this increases the measurement error, particularly if filters are stacked together and moved in and out of the beam path, changing the stack depth as needed to keep the transmitted light power within the detector's linear range. Stacking filters is generally a bad idea because errors may be introduced by multiple reflections at the filters' surfaces. The effect of these reflections can be reduced by tilting the filters slightly in the beam path to cause the reflected light to deviate out of the beam path. This in turn introduces a slightly longer beam path through each filter which again increases the error because the manufacturers state transmission of a filter at normal incidence. In fact, in any experimental arrangement it is generally best not to move items in and out of the beam path because it is virtually impossible to replace them in exactly the same place and orientation, which inevitably introduces errors in the measurements. This, incidentally, endorses the beam-splitter/light-chopper arangement (for example, Two sample cells with different pathlengths) which avoids the need to replace the cells with each sample as required with a single-cell arrangement for the transmission measurement.
| CITATION: Swanson N. L., Jonasz M. 2007. Measuring attenuation of light: General comments on experimental arrangements (www.tpdsci.com/Tpc/AtnCfMsSetGen.php). In: Top. Part. Disp. Sci. (www.tpdsci.com). |
HISTORY: Published: 21-Nov-2007 Modified: 01-Jan-2008 Peer-reviewed: 22-Dec-2007 |
| Copyright 2005-2008 MJC Optical Technology. All rights reserved. | Terms of use | Menu |