Interlaboratory validation of the Mehlich 3 method for extraction of plant-available phosphorus

• Published in: Journal of AOAC International Vol. 92; no. 1; pp. 91 - 102
• Main Authors: Zhang, Hailin, Kaiuki, Solomon, Schroder, Jackie L, Payton, Mark E, Focht, Charlie
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 01-01-2009
• Subjects: Calibration; Composition; Crop yields; Extraction (Chemistry); Hydrogen-Ion Concentration; Laboratories - standards; Materia medica, Vegetable; Methods; Phosphorus - isolation & purification; Phosphorus content; Phosphorus in the body; Plant Development; Plant extracts; Plant-soil relationships; Plants - chemistry; Properties; Reproducibility of Results; Sensitivity and Specificity; Soil - analysis; Soils; Spectrophotometry, Atomic - instrumentation; Spectrophotometry, Atomic - methods; United States
• ISSN: 1060-3271; 1944-7922
• DOI: 10.1093/jaoac/92.1.91

The Mehlich 3 (M3) method is widely used for extraction of plant-available phosphorus (P) from soil over a wide range of pH values. The method is also used by many laboratories to determine multiple plant-available nutrients simultaneously. However, this method has not been statistically validated within and among laboratories. The objective of this study was to determine the repeatability (within-laboratory performance) and reproducibility (among-laboratories performance) of the M3 method by using a wide variety of soils. An in-house homogeneity test was conducted for 10 soils. Three replicates of each of the 10 soils were sent to 26 domestic and international laboratories primarily for P analysis. Samples were scooped, weighed, or both scooped and weighed for extraction. The P in extracts was quantified by the participating laboratories by using inductively coupled plasma-atomic emission spectrometry (ICP-AES) or colorimetrically. For the scooped samples analyzed colorimetrically, the repeatability relative standard deviation (RSDr) ranged from 2.07 to 12.1%; the RSDr ranged from 2.2 to 21.4% for the scooped samples analyzed by ICP-AES. For the weighed samples analyzed colormetrically, the RSDr values were 1.09-9.34%, and for the weighed samples analyzed by ICP-AES, they were 1.70-5.76%. For the reproducibility data, the RSDR values ranged from 6.85 to 50.8% for the scooped-colorimetry category, from 6.95 to 73.9% for the scooped-ICP-AES category, from 7.19 to 42.6% for the weighed-colorimetry category, and from 5.29 to 35.9% for the weighed-ICP-AES category. The greatest RSD values were associated with the Susitna soil, which had the smallest concentration of extractable P. Because of the relatively small concentration of P in this soil, the laboratories were attempting to measure solution concentrations that were close to the detection limits. The Horwitz ratios (HorRat) were also used to evaluate the repeatability, HorRat(r), and reproducibility, HorRat(R). Overall, the M3 P method appears to be both repeatable and reproducible across the 4 categories, and the vast majority of the HorRat values for both repeatability and reproducibility were within the acceptable range. The results of this study indicate that the M3 P method for the determination of plant-available P in soil is both accurate and precise when standardized procedures are used. The method has been shown to be suitable for use as a reference method for testing soil materials for extractable P.