Search

Criteria for the selection of adsorbents, for layer preparation, for loading, and for the choice of developing solvents are discussed. From a knowledge of the absorption/scattering (i.e., L/S) ratios of common adsorbents in the UV, of the emission spectrum of the source, and of the absorption spectrum of the test substance in solution, a rational choice of scanning wavelength is possible. A favourable signal-to-noise ratio when scanning below 280 nm depends upon the adjustment of optical factors so as to maximise the light flux, in this case the source, condenser geometry, and a reduction of layer thickness. It is well established that the most significant noise arises from the particulate nature of thin layers and the irregularities associated with them. An amplifier offering a variable signal-averaging period has some advantage. Recordings should indicate variation in layer thickness so that a correction of the Kubelka-Munk form may be applied to the results.

An alternative circuit for the output from the author's flying spot scanner has been developed so that transmission in the two dimensions may be presented continuously as a negative logarithm on a flat-bed recorder. Thus, a rapid visual survey of all the detectable information present in as many as six chromatograms on one 75 × 75 mm plate can now be made. The proportion of each resolved zone can be estimated approximately from manual calculations based on the sum of peak heights. In a binary mixture scanned at 330 nm the coefficient of variation found was about 4%. The chromatography, recording and manual calculation of, for example, twelve impurities in terms of two standards occupy about 2 h. When scanning at wavelengths below 280 nm, a reduction in layer thickness to 0.14 mm is made to reduce the increase in light absorption of silica gel G. In the examples shown, −log T recordings in the UV are more sensitive than visual or visualisation methods for testing the resolution of mixtures. The effect of tailing of standards in the estimation of impurities in a manufactured intermediate product is also discussed.

The construction and use of an instrument designed to facilitate research into analysis by quantitation of chromatographed zones is described. The instrument determines and records on computer tape the UV transmittance at 0.5 × 0.5 mm intervals in the two dimensions of the thin layer. The application of the general K ubelka -M unk transmission equation to the interpretation of the data is described. The flexibility of the method lies with the analyst who may write computer programmes suited to his analytical problems.

An end-point half-cell suitable for use in chloride titrations has been prepared by dipping a silver rod into an aqueous suspension of a cation-exchange resin, partly converted to the silver form. Such a half-cell has a considerable reserve capacity for maintaining silver ion concentration at the required level, and therefore has a long useful life in routine operation.

A method for the in situ quantitative analysis of thin layer chromatographic zones on microscope slides is described. It comprises the quantitative application of 1 μl of solution to the plate, two-dimensional scanning of the zone with a point of monochromatic illumination, then off-line data processing by computer so that a simplified form of the general K ubelka -M unk equation for transmission can be applied. The standard deviation observed on independently analysed 0.510 μg loads of acetone dinitrophenylhydrazone on silica gel thin layers using the experimental prototype apparatus is 2.7%.

The optical properties of some common thin-layer chromatographic adsorbent layers are compared. Silica Gel G is much more transparent above 280 nm than are similar thicknesses of other adsorbents. Below 280 nm, the attenuations of these adsorbents rise more or less steeply. Improvements to the author's apparatus for densitometry by transmission in the UV region are described, followed by off-line data processing on an I.B.M. 360 computer. Examples shown are: preliminary tests of the precision of a steroid determination at 238 nm where the background “absorbance” is about 2.8; separation and determination of minor impurity zones at 340 nm on glass support plates; and at 298 nm the active agent remaining after the decomposition of a 0.2% aqueous solution.

A simple theoretical relationship applicable to in situ quantitative analysis by light absorption is proposed. The equation is based upon the Kubelka—Munk theory of radiative transfer. The conditions necessary for the application of the theory to quantitative analysis have been discussed. The simple theory is composed of two separable effects, “hyperchromatically” and “curvature of response”. These two effects have been evaluated experimentally on silica gel layers and correspond closely to that predicted. The validity of the theory is thus confirmed.