It is rare to find any significant level of phosphate in a wall-wash sample because inorganic phosphate is actively scavenged by the micro-biota that live on even the most inhospitable, natural rock surface. However, sometimes it is good to be able to verify phosphate levels for those samples where the ionic balance of the usual electrolytes we measure fails to make sense.
The phosphomolybdate blue method has a long history of use as a sensitive assay for phosphate, and, consequently, there are great number of published reference methods. The main differences between these methods is the choice of reagent used to achieve quantitative reduction of phosphomolybdate to the molybdenum blue complex that comprises the basis of the actual measurement.
To this end, I have found that antinomy-catalysed ascorbic acid results in robust quantitative behavior, and thus have developed a method around its deployment.
Ammonium molybdate solution:
- 2.5g of ammonium molybdate tetra-hydrate
- 80ml of concentrate sulphuric acid
- make up to 250ml
- weigh-out 0.6g antimony metal (>99% pure)
- add 8ml of concentrated hydrochloric acid
- add 1ml of de-ionised water
- add 1ml 35% hydrogen peroxide
- when the reaction ceases, add a further 1ml of hydrogen peroxide
- repeat the above step as many as four times over 24hrs until the metal has completely dissolved
- weigh-out an amount of tartaric acid equimolar to the amount of antimony originally used (0.74g) and dissolve in 100ml of de-ionised water
- slowly add the tartaric acid to the dissolved antimony with continuous stirring. A dense white precipitate of the oxychloride is initially formed. This clears over a number of minutes to form the soluble tartarate salt.
- make the solution up to 500ml to give an approximate concentration with respect to antimony of 0.01M
- 1ml aliquot of sample
- 2ml of ammonium molybdate solution
- 1ml of antinomy tartarate solution
- 9ml of freshly-made ascorbic acid
- make up to exactly 25ml with de-ionised water
- measure the absorbance at 820nm after 3hours
The response is linear and two-point calibration with standard and blank is adequate. The typical working range is 0 to 1umole in a 1ml sample aliquot.