An atomic absorption method is described which employs electrothermal heating of the sample in an enclosed chamber to produce the atomic sample vapor. The atomization takes place in a selected atmosphere which may be maintained at any desired pressure, usually in the range of 1-300 Torr. This system increases absorption sensitivity by reducing foreign gas effects such as the loss of ground state sample atoms through chemical reaction with ambient gases. The method is capable of detecting 10~12 gram of an element in a few microliters of solution. In addition to significant improvement in absolute sensitivity as compared to flame methods, it is simple to operate and more adaptable to a variety of special techniques.The sensitivity which is theoretically available in atomic absorption makes this technique potentially ideal for trace analysis. The greatest limitation of commercial atomic absorption instruments is their use of the inefficient flameatomizer system. Primarily because of physical limitations, the amount of the element which must be supplied to the

A flame emission spectrophotometric method has been developed for the determination of ultramicro amounts of sodium present in thin silicon oxide films and on cleaned silicon surfaces. Ultrapure 5% hydrofluoric acid is used as the solvent and the sodium measured directly in the solution by flame emission analysis. This method, which has a sodium detection limit of 0.2 ppb, is used in place of the more contamination‐prone neutron activation analysis to determine sodium distribution profiles in thin silicon oxide films as well as to measure the sodium contamination level of cleaned silicon surfaces. Sodium contamination on the surface of a single silicon wafer, having a surface area of approximately 16 cm2, can be detected down to 8×1011normalNa atoms/cm2 .

The extremely low sodium vapor pressures which are present in hot quartz furnace tube atmospheres have been measured by atomic absorption spectroscopy. In the system described, sodium vapor pressures are measured between 10 -6 and l0 -9 Torr; however, this range can be extended by simple changes in optics and/or absorption cell length. The data show that the sodium vapor pressure in a new quartz tube changes slowly during heating until the rate of sodium in-diffusing from the hot ceramic liner is balanced by the rate of sodium out-diffusing to the cooler portions of the system. If the design of the furnace tube system is such that the source of the sodium, the furnace liner, is separated by an air space from the inner quartz tube, then the equilibrium sodium vapor pressure inside this tube may be decreased at least tenfold. Foreign gas composition has little effect on the sodium vapor pressure, providing no chemical reactions occur in the system. However, the reaction of hydrogen with the quartz surface results in a marked increase in the sodium vapor pressure. The system described may be easily adapted for determination of other contaminant vapors which could be present in hot quartz furnace tube atmospheres~