Date of Award

Spring 1981

Project Type


Program or Major


Degree Name

Doctor of Philosophy


Because phosphorus is a common pollutant in natural waters, there is a need to develop a convienent, reliable technique to measure low phosphorus concentrations. Spectrophotometric procedures involving the molybdenum blue complex have so far proven to be the best technique for measuring phosphorus in the 0.01 and 10 mg/L range, however, these procedures are all time consuming and subject to interference.

This dissertation deals with flame photometry as a method for phosphorus analysis. In a hydrogen-rich flame phosphorus atoms undergo a chemiluminescent reaction to form an excited HPO molecule, which emits with maximum intensity at 526 nm. Normally this method is used to selectivity detect phosphorus compounds separated by gas chromatography, or to analyze gas phase phosphorus concentrations. When aqueous solutions are introduced to the flame as aerosols, sensitivity is reduced and metal ions interfere with the emission. An alternative method of sample introduction is evaluated in this dissertation. A Perkin-Elmer HGA 2100 graphite atomizer is coupled to a flame photometer. Samples containing 25 (mu)L of aqueous phosphate solutions are placed in the atomizer with a micropipet. The atomizer dries the sample at 100(DEGREES)C, then vaporizes the phosphorus compound at 2500(DEGREES)C. The phosphorus vapor is swept into the flame and the HPO emission is measured. Intensity is directly proportional to phosphorus concentration. This method of sample introduction has fewer metal ion interferences than the aerosol technique, however, common ions such as calcium, magnesium, and ion III, still interfere. Releasing agents such as EDTA and sodium silicate failed to eliminate the metal ion interferences. It was subsequently demonstrated that the metal ion interference takes place in the vapor phase after atomization.

It was found that silica reduced the degree of interference. A method of permeating the graphite tube with silica was developed. The procedure involves inpregnating the graphite tube with SiCl(,4) under reduced pressure, hydrolyzing SiCl(,4) to hydrated silica, drying and firing at a high temperature. The high temperature firing causes the dried silica to react with the graphite forming silicon carbide. This was confirmed by weighing tubes before and after treatment with silica and after firing.

The appearance temperature for phosphorus as phosphate in the presence of calcium is 1400(DEGREES)C for untreated graphite and 2000(DEGREES)C for the silicon carbide coated tube. At a higher temperature, the metal-phosphorus interaction causing the interference is less favorable, thus causing a reduction in the extent of the interference.

Eleven possible interfering ions were tested to determine the degree of interference on a 1 mg/L phosphorus solution. Among the 11 ions tested, none of them had a signal depressed more than about 10%, when the ion concentration was at 10 mg/L. When tested at the 100 mg/L ion level, six of the ions caused a deviation of more than 10% from the signal which a 1 mg/L phosphorus solution would give.

The relative standard deviation for five, 25 (mu)L samples of phosphorus is 3 to 6% at a concentration of 1 mg/L. The detection limit is 0.04 mg/L for a 25 (mu)L phosphorus solution.

By making a few minor modifications to the instrument, sulfur can be determined. Sulfur is also subject to metal ion interferences but this element was not studied as extensively as phosphorus.