John Girdner Atwood

The laser illuminated absorptivity spectrophone: a method for measurement of weak absorptivity in gases at laser wavelengths.

A spectrophone measures absorptivity by sensing thermal expansion in a confined sample gas. Laser source excitation provides sufficient radiation to measure precisely very weak absorptivities at laser wavelengths. This paper describes the theoretical capability, design considerations, and experimental testing of a pulsed ruby laser absorptivity spectrophone and a cw CO(2) laser absorptivity spectrophone. A spectrum of the water vapor line at 6943.8 A was obtained. The peak absorptivity was 3 x 10(-6)cm(-1). In the vicinity of 9.6 micro, absorptivities of CO(2)-N(2) mixtures were measured down to 1.2 x 10(-7)cm(-1).

Dispersion of peaks by short straight open tubes in liquid chromatography systems

Abstract The theory of spreading of a sample peak in a long straight open tube is known. However, when the tube becomes shorter than 30 theoretical plates, the eluted peak becomes non-gaussian and the theory for long tubes does not apply. This is the case for connecting tubes, injection loops and detector flow cells in liquid chromatographic (LC) systems. In earlier work, we studied the theory of this case using a computer model combining Poiseuille flow with diffusion, obtaining unexpected results about how samples wash out of short tubes. This work extends that study to obtain the peak shapes and bandwidths eluted from straight open tubes ranging from 0.01 to 30 plates in length. An empirical expression was found for peak width which fits the results of the computer model to within 4% over this entire range. Below 3 plates, the normalized peak width is approximated by a constant times the inverse fourth root of normalized tube length in plates. It becomes as small as a quarter of the value predicted by the long tube theory for 0.01 plates. Experimental measurements on short tubes agree approximately with the computer model when diffusion is the only cause of radial mixing. An expression was derived which determines whether a tube is sufficiently straight so that secondary flow is unimportant compared with diffusion as a cause of radial mixing. Measurements on curved tubes are consistent with the expression. The conditions under which measurements of peak spreading in an open tube can be used to obtain the diffusivity of an LC sample are discussed. A consequence of these results is that LC systems can be designed with substantially less bandwidth contribution from extra-column components than would be predicted using the long tube theory. Another consequence is that variances contributed by consecutive segments of short open tubes are not additive unless there is complete radial mixing at the connections between segments.

Improvements in liquid chromatography column life and method flexibility by saturating the mobile phase with silica

Abstract We describe a technique which may expand the opportunities for liquid chromatography (LC) methods development. It has not been feasible to use high pH water-based mobile phases with silica packings, even those coated with octadecylsilane. A principal failure mode of such packings with both high pH and high temperature mobile phase is the dissolution of silica. We show that this can be controlled by equilibrating the mobile phase with silica using an appropriate column mouted in the oven ahead of the analytical column. We describe some early results of this technique for determining tricyclic antidepressant drugs at a mobile phase pH of 10.7 on uncoated 5-μm silica particles. Using an inorganic instead of an organic base also permits UV detection closer to 200 nm where most organic bases have become opaque. We measured the silicon dissolved in the column effluent by atomic absorption (AA). We used an injection procedure which permits the AA burner to take up solution at its optimum rate while the LC is used at any lower mobile phase flow-rate.

Early phases of the dispersion of a sample injected in poiseuille flow

Abstract A mathematical simulation of the dispersion of a sample injected in a viscosity-controlled (Poiseuille flow) stream has been made to determine the behavior of the sample from injection time until the sample distribution has become substantially gaussian. The results show that for velocities much greater than the chromatographic optimu, and for flow of the order one to ten theoretical plate heights, an unusually unsymmetrical distribution develops. It gives rise to doubly peaked chromatograms, in agreement with curves actually observed in corresponding experiments.

Very-high-speed liquid chromatography : II. Some instrumental factors influencing performance

Abstract Instrumental factors influencing very-high-speed liquid chromatography including injection volume, detector flowcell volume, detector response time and total instrumental bandwidth are discussed. Very-high-speed analyses performed using relatively short columns (100mm) of conventional internal diameter (4.6 mm) packed with small particles (3 μm) result in very high performance as determined by the ability to generate over 400 theoretical plates/sec. Minimizing the total instrumental bandwidth by reducing volumes of the injector, connecting tubes and detector flowcell is necessary along with a very rapid detector response time in order to attain this performance. Limitations in performance on peaks having low capacity factors are apparent due to extra-column effects. The relatively high flow-rates which are utilized result in unexpectedly improved chromatographic performance since extra-column effects due to the flow path are not as great at higher flow-rates as originally expected. The largest contribution to extra-column effects appears to arise from the detector flowcell volume and the detector response time.

Optical Properties and Laser Thresholds of Thirty-nine Ruby Laser Crystals

Measurements were made of optical path uniformity, chromium concentration, fluorescence linewidth, large-angle scatter, and laser threshold on 39 flame fusion rubies. Results show that index of refraction variation is explainable by variation of chrome concentration and wander of the optic axis. Residual stress plays little role in optical defects, but may broaden the 77°K linewidth. Optical quality of lasers cut from regions of boules selected by interferometry remains constant after cutting, but warping destroys their parallelism. Threshold increases with chromium concentration, indicating that the rods are optically thick at the pumping bands. Threshold is not correlated with other optical properties, however.