Evaluation of on-line preconcentration and flow-injection amperometry for phosphate determination in fresh and marine waters

Abstract

Dissolved reactive phosphorus (DRP) was determined as orthophosphate (PO₄-P) in fresh and saline water samples by flow-injection (FI) amperometry, without and with in-valve column preconcentration. Detection is based on reduction of the product formed from the reaction of DRP with acidic molybdate at a glassy carbon working electrode (GCE) at 220 mV versus the Ag/AgCl reference electrode. A 0.1 M potassium chloride solution was used as both supporting electrolyte and eluent in the preconcentration system. For the FI configuration without preconcentration, a detection limit of 3.4 μg P l⁻¹ and sample throughput of 70 samples h⁻¹ were achieved. The relative standard deviations for 50 and 500 μg P l⁻¹ orthophosphate standards were 5.2 and 5.9%, respectively. By incorporating an ion exchange preconcentration column, a detection limit of 0.18 μg P l⁻¹ was obtained for a 2-min preconcentration time (R.S.D.s for 0.1 and 1 μg P l⁻¹ standards were 22 and 1.0%, respectively). Potential interference from silicate, sulfide, organic phosphates and sodium chloride were investigated. Both the systems were applied to the analysis of certified reference materials and water samples.

Introduction

Phosphorus is an essential plant and animal nutrient. However, due to human activities, inadvertent addition of phosphates to watersheds has caused eutrophication, which is commonly manifested by algal bloom formation [1]. Analytical methods for determination of phosphorus species, especially readily bioavailable dissolved reactive phosphorus (DRP) are essential for investigating the sources, cycling and fate of phosphorus (P) in aquatic ecosystems.

Spectrophotometric methods for determination of DRP are commonly based on the reduction of the phosphomolybdate complex (formed from reaction between reactive phosphate and acidic molybdate) to form the intensely coloured “molybdenum blue” product. Both batch [2] and flow-based [3], [4] procedures have been reported. Although the spectrophotometric method provides good sensitivity, it suffers both from interferences, e.g. silicate, turbidity [3] and refractive index (Schlieren) effects in estuarine and marine samples, which can cause large errors in quantitation [5]. Electroanalytical methods are more tolerant of these interferences, and do not suffer from the Schlieren effect. Voltammetric or amperometric determination of phosphate has been performed using a range of different electrochemical reactions including, (1) reduction of 12-molybdophosphate to molybdenum blue [6], [7], [8], [9], [10], [11], [12], (2) oxidation of molybdenum blue, which was electrochemically prereduced and adsorbed on the electrode [13], [14], (3) reduction of molybdovanadophosphate [15] and (4) oxidation of FePO₄ after reductive accumulation of Fe(II) in the presence of phosphate [16]. Flow-injection has also been applied for electrochemical determination of phosphate in order to improve performance of the technique. Some electrochemical methods involving the use of enzymes for determination of DRP have also been reported [17], [18]. An FI amperometric system incorporating inorganic pyrophosphatase and nucleoside phosphorylase–xanthine oxidase reactors was used for simultaneous determination of phosphate and pyrophosphate [17]. An FI amperometric system with a nucleoside phosphorylase–xanthine oxidase column was also used for determination of phosphate with a detection limit of 1.25 μM [18]. The analytical characteristics of these electroanalytical techniques are summarized in Table 1.

The three major objectives of this work, based on the reduction of 12-molybdophosphate on a glassy carbon working electrode (GCE), were to evaluate amperometric detection of phosphate in fresh and marine waters, investigate the use of an in-valve preconcentration column and develop a simple manifold suitable for application in field instruments.