MEASURING METAL IONS IN A SOLUTION
This invention relates to electrode devices and methods for measuring the quantity of metal ions in a solution, particularly (but not exclusively) the measurement of lead in blood.
BACKGROUND TO THE INVENTION
It has been well known for many years that lead is a toxic element and can have a number of serious health effects. Young children are especially at risk from lead poisoning. They can be exposed to lead from sources such as water (lead pipes), food and air (leaded petrol) . Old paint can contain a high concentration of lead and the most common cause of lead poisoning in young children is from eating paint chippings or dust from the walls or windows of old houses.
In the United States of America the Center for Disease Control (CDC) considers lead poisoning such a serious problem that it recommends all children in the country under 6 years of age be screened for lead. The amount of lead which a child has been exposed to is determined by measuring the concentration of lead in the child's blood . Over the years, as more has been learned about the adverse effect of lead on children, the blood lead concentration considered safe has steadily declined. In 1985 the CDC considered a lead concentration of less than 25 μg/dl to be acceptable. In 1991 the CDC lowered the safe lead level to 10 μg/dl.
Atomic absorption spectroscopy (AA) can be used to measure lead in blood very accurately at low concentrations but the method is not practicable for screening because the instrument is large, very expensive, and requires a highly trained operator.
The electrochemical method of anodic stripping voltammetry (ASV) is another way of measuring lead in blood. In ASV, an electrode in contact with a solution to be tested is held at a negative potential for a sufficient period of time to reduce metal ions in the solution and concentrate them at the electrode. The potential is then ramped or scanned in the positive direction and any metals present will be stripped from the electrode when the unique oxidation potential of the respective metal is reached. The current produced during the stripping of each metal is proportional to the concentration of that metal in the test solution.
The CDC is now actively encouraging development of a system which is small, portable, cheap and easy to use, but can measure low concentrations of lead in blood with good accuracy. An example of an ASV system intended to meet these criteria, described in WO 91/08474, uses disposable electrodes suitable for field testing.
One disadvantage of these disposable ASV electrodes is that they are less sensitive under acid conditions of pH<2.
Recently, Feldman et al, "Electrochemical determination of low blood lead concentrations with a disposable carbon microarray electrode" Clin Chem 1995, 41: 557 - 63 described use of disposable microelectrode arrays for measuring lead in blood. However, the testing protocol required the electrode to be pre-plated with mercury. The electrodes also responded differently to different bloods and so standard additions had to be used to determine lead concentration.
It is an object of the invention to alleviate some or all of the above disadvantages of the prior art, and to provide an improved electrode measuring device for use not only in lead determination, but which may also be used to measure a number of different metallic ions in solution.
SUMMARY OF THE INVENTION
The present invention provides an electrode device for measuring the quantity of ions of a predetermined metal which are present in a solution, the electrode device comprising a working electrode and a chemical species for chemically enhancing the sensitivity or specificity of the electrode device, the chemical species being disposed within a layer overlying or in close proximity to the surface of the working electrode or being incorporated into the working electrode.
The electrode device is preferably of the type described in WO 91/08474 and comprises a conductive (eg carbon) working electrode and a reference electrode (eg of silver/silver chloride)- which are made by printing tracks of conductive ink on to a glass or plastics base. The preferred configuration of working electrode device is a microelectrode array which gives a high signal to background ratio and is not affected by dissolved oxygen. Such an array can be made by printing an electrically insulating layer over the carbon working electrode device and then using laser photoablation to make small holes in the insulating layer and expose the underlying electrode device.
It is often desirable to determine one metal in the presence of another. Whilst specificity can generally be achieved when stripping voltammetry is used, on the basis that the potential required to strip each different metal from the electrode is unique, a problem arises if the stripping potentials are close together, or one metal is present in large excess over another as deconvolution of the signals from different species can be difficult.
For example, when detecting lead in domestic water supplies and human blood samples the presence of large quantities of copper which are common in such samples can seriously reduce the lead response. This is because copper is plated onto the surface of the electrode and the
efficiency of lead plating on the copper surface is greatly attenuated. For example, in the presence of 1 ppm copper, the signal from a 10 ppb lead solution is depressed by almost 70%.
It has been found that the efficiency of plating of the target analyte is improved if a second or third transition Group VIII metal, or silver, gold or mercury is plated onto the working electrode, either before or during the measurement step.
Accordingly, in preferred embodiments of the invention, the chemical species comprises one of said metals or a salt thereof disposed in a matrix or binder layer coated on the working electrode and the metal is plated onto the electrode during the operation of the device. For example, the chemical species may be mercury, a mercury salt such as mercuric chloride or a silver salt. Mercury or silver is co-plated onto the working electrode together with the target analyte. This alleviates the problem of overlapping peaks, for example copper/lead peak overlap, apparently because lead plates more easily onto mercury or silver than onto copper. The silver or mercury surface is continually refreshed so that lead deposition is always favoured.
Preferably, the matrix or binder layer comprises a polymeric material, and for example the electrode device may be fabricated using Hydroxy-ethyl-cellulose (HEC) as binder for a mercury salt from which metallic mercury is to be plated. However, using HEC the resulting mercury film can tend to be somewhat variable due to loss of mercury during the wetting step. Preferably, the matrix or binder layer comprises a substance that has low solubility under the conditions of intended use, which, for the measurement of heavy metal ions, are normally acidic. An example of such a material is carboxy-methyl cellulose (CMC) which is soluble around neutral pH, and so can be readily coated onto the electrode surface, but is insoluble under acidic conditions. This is advantageous for the purpose of
detecting heavy metals which should be analysed under acidic conditions to ensure that the metals are freed from complexing agents in solution. If a binder composition which includes CMC is used the resultant mercury film is more reproducible and consequently analyte signal precision is improved.
An alternative way of promoting stability is to include a mercury ligand or chelating agent, such as Hydroxy ethyl ethylendiaminetri-acetic acid (HEDTA) , in the matrix layer. The binding molecule promotes stability by maintaining the mercury in a stable complex during storage, yet readily releases the mercury under the acid conditions prevalent during a test.
A further problem which can arise when different metals are present in the test solution is the formation of intermetallic compounds. An example of such a problem is the determination of zinc in the presence of copper. A zinc/copper intermetallic forms, and is stripped from the electrode at a potential close to the copper stripping peak, resulting in depression of the zinc peak.
A known solution to the intermetallic problem is use of gallium which forms an intermetallic compound with copper in preference to zinc. The invention includes the incorporation of gallium into the matrix layer.
The problem of determining one metal in the presence of another can alternatively or additionally be addressed if the matrix layer includes a reagent which diminishes or removes overlapping stripping voltammetry peaks caused by interfering metallic species. One alternative for such a reagent is one which selectively precipitates the unwanted species. For example, ferricyanide selectively removes copper. Another alternative is a reagent which interacts with one or both of the species concerned with the effect of separating the peaks. For example, iodide causes peak separation by changing the relative oxidation potentials of
lead and copper.
As an alternative to co-plating the working electrode during measurement, the working electrode may be pre-plated with the chemical species (for example a noble metal such as silver) during manufacture of the device. This is a particularly preferred option when the metal to be detected is lead because the presence of silver appears to discriminate against the deposition of copper during the determination of lead. The amount of silver applied is critical, since if insufficient silver is deposited lead sensitivity is diminished, whilst if too much silver is deposited a large oxygen reduction background is observed. We have found that very good lead stripping peaks can be obtained by pre-plating the electrode at approximately -1.0V versus silver chloride for between 5 and 600 seconds using a solution containing between 1 and 100 ppm of silver ions. Ideally, silver is plated for 60 seconds using a lOppm silver solution. These conditions appear to lead to an ideal distribution and size of silver nuclei on which lead nucleation can initiate.
The noble metal may alternatively comprise one of the so-called "platinum metals" i.e. rhodium, palladium or platinum.
As an alternative to plating, silver may be incorporated during the printing of the carbon electrode by using an appropriate mixture of silver and carbon inks or by printing carbon ink made from carbon particles metallised with silver.
The invention also provides a method for measuring the quantity of ions of a predetermined metallic element which are present in a solution comprising the steps of contacting said solution with an electrode in the presence of a chemical species for chemically enhancing the sensitivity of the electrode and determining the number of said ions coming into contact with said electrode.
The chemical species may be added to the test solution and may comprise a reagent which diminishes or removes overlapping stripping voltammetry peaks caused by interfering metallic species. One alternative for such a reagent is one which selectively precipitates the unwanted species. For example, ferricyanide selectively removes copper. Another alternative is a reagent which interacts with one or both of the species concerned with the effect of separating the peaks. For example, iodide causes peak separation by changing the relative oxidation potentials of lead and copper.
In a method for measuring the quantity of lead ions, ferricyanide may be added to the solution (for example by subjecting to 1:10 dilution with a 5% potassium ferricyanide solution) prior to testing to prevent any copper ions in the solution being deposited at the electrode. Alternatively, iodide, for example potassium iodide, may be added to the solution prior to testing.
BRIEF DESCRIPTION OF THE DRAWINGS
Embodiments of the invention will now be further described, by way of example, with reference to the accompanying drawings, in which: -
Figure 1 is a schematic plan view of an electrode device according to the invention,-
Figure 2 is a schematic transverse section through the electrode device of Figure 1, taken on line II - II;
Figure 3 is a graph of lead concentrations -fc-n bloods measured by the electrode devices according to a first example of the invention against those measured by a reference method;
Figure 4 is a stripping voltammogram obtained using electrode devices according to a second example of the
invention,-
Figure 5 is a graph of anodic stripping peak height measured by electrode devices according to a third example of the invention against lead concentration,- and
Figures 6 and 7 are stripping voltammograms obtained using electrode devices according to sixth and seventh examples of the invention respectively.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
Figures 1 and 2 show a disposable ASV electrode device comprising an insulating substrate 1 onto which three carbon electrodes, namely a counter electrode 2a, a central electrode 2b and a working electrode 2c have been deposited, for example by screen printing. The central electrode 2b has a silver/silver chloride layer 3 printed over the carbon, thereby forming a reference electrode.
A layer of dielectric 4 has subsequently been printed, covering the electrodes but leaving a window 7 for the counter and reference electrodes. The dielectric layer then has holes 5 ablated therethrough by a laser. Finally a reagent layer 6 is microdosed over the photoablated holes 5.
The carbon microelectrode array revealed by the holes measures 2mm x 3mm and consists of 13 rows of 19 microelectrodes, each with a diameter of approximately 40 m. A reagent layer containing approximately 11 μg of mercury is ink jetted over the array using a Bio Dot Microdoser (Bio Dot Limited, Diddington, Cambridge UK) and in this example contains mercuric nitrate, as well as a chelating agent (for stability of the mercury) and film formers, namely CMC and HEC.
Example 1
The procedure for performing a blood lead test with a set of electrode devices as described above involved two steps. Electrochemical measurement was preceded by an initial acidification step in which 200μl of blood was added to 2.10ml of 0.9M HC1 in a 20ml polystyrene vial. The vial was then briefly mixed by hand and put on a bottle roller for approximately 5 minutes in order to ensure complete release of the bound lead before the blood was tested. Earlier experiments showed that one minute was sufficient for complete release but results were unchanged for up to 45 minutes.
The electrochemical testing was carried out using an Autolab (EcoChemie B.V, Utrecht, Netherlands) in the differential pulse mode. 75μl of acidified blood was pipetted on to a disposable electrode device and deposition was effected by polarising the electrode at -800mV for 165 seconds. The stripping sequence employed a pulse amplitude of 50mV, step 5mV, pulse width 3ms, and trough width 120ms. Lead stripping peak heights were measured using the Autolab software.
The linear response of the electrodes was tested using one blood spiked with Pb(N03)2 (Aldrich AA standard) to give a range of lead concentrations up to 1000μg/l. The exact lead concentration of each spiked blood was measured using atomic absorption. Each blood was tested using 5 disposable electrodes. A graph of peak current against lead concentration was linear up to about 600 μg/1 with a slope of 0.0070μA/μg/l and intercept of 0.212μA. The intercept may be due to lead in the mercury layer on the electrodes. This graph was used as a calibration curve for subsequent electrodes.
The accuracy of the electrodes was determined by a correlation study using 20 venous bloods whose lead concentrations had been measured by atomic absorption. The
bloods were from both children and adults and were between
4 and 11 days old before being tested. Each blood was tested with 2 electrodes. The lead concentrations were calculated from the lead stripping peaks using the spiked blood calibration. The results of the correlation study
(Figure 3) show an excellent agreement between the disposable electrodes and the reference method. All 40 measurements were within 20 μg/1 of the reference method.
Of the 22 measurements on bloods with lead concentrations below 200 μg/1, 21 of them were within 10 μg/1 of the reference method. The low intercept and slope close to l demonstrate that the spiked blood calibration works well.
The precision of the electrodes was tested using 23 venous bloods with lead concentrations of 50, 110 and 250 μg/1. Each blood was acidified once and the solution was tested using 10 electrodes. The observed coefficients of variation were 11.9, 7.2 and 2.7% respectively.
Precalibration of the electrodes removes the need to perform the standard additions described by Feldman et al.
Example 2
Carbon microarray electrodes were coated with a layer having the following formulation:-
0.25% HEC
50mM potassium chloride 2.4% mercuric chloride
It was intended that mercury be plated onto the electrodes during testing.
A blood sample containing 150μg/l lead was diluted tenfold with 0.9M HC1 and pipetted onto the electrodes. A typical stripping voltammogram is shown in Figure 4 in which the lead peak on the left hand side is distinguishable but partially overlaps a copper peak on the right hand side.
Using a blood sample with a lead concentration of 200 μg/1 these electrodes gave a response of 6.49 μA with a coefficient of variation of 10.5%.
Example 3
A further set of electrodes were made according to Example 2 with an additional coating of 1.5 μl of a 4% solution of CMC in water . The blood sample containing 200 μg/1 lead that had been applied to the electrodes of Example 2 was also applied to these electrodes. The lead response was increased to 8.24μA and the coefficient of variation was reduced to 5.3%
Figure 5, which is a graph of peak current against lead concentration for these electrodes, shows good linearity.
Example 4
Electrodes were manufactured having the following composition for the binder layer:-
3% CMC 0.1% HEC
0.037M mercuric nitrate 0.04M HEDTA
50mM potassium chloride
The HEDTA molecule promoted stability by maintaining the mercury in a stable complex during storage, yet readily released the mercury under the acid test conditions.
Exjunplq 5
The following composition was used as the binder layer for disposable electrodes: -
3% CMC 0.1% HEC
0.5% mercuric chloride
50mM potassium ferricyanide
50mM sodium chloride
The potassium ferricyanide selectively precipitated copper from a blood sample and removed the copper peak from the stripping voltammogram.
Example 6
Electrodes were coated with the following composition: -
0.23% HEC 45mM potassium chloride
2.2% mercuric chloride
0.5% potassium ferricyanide
Figure 6 shows the stripping voltammogram obtained with these electrodes using the same blood sample which produced the voltammogram of Figure 4 with the electrodes of Example 2. As can be clearly seen from Figure 5, incorporation of ferricyanide into the electrode has eliminated the copper peak and hence improved the baseline for lead determination.
Exam le 7
A blood sample was diluted 9:1 with 0.2M potassium iodide solution and was then pipetted onto a disposable carbon microarray electrode. The stripping voltammogram (Figure 7) shows that the lead and copper peaks are clearly separated.