E ECTRODE WITH PROTECTIVE COATING
The present invention is concerned with protective coatings for metallic electrodes and in particular protective coatings for metallic electrodes of the type employed in analysis (typically spectroscopical analysis) of biological material .
International (PCT) Patent Application WO 92/04630 describes a method of analysis of biological cell materials, particularly those comprising enzymes in membranes, and substrates for such cell materials and enzymes . The preferred method comprises causing an AC electrical potential of one or( more initial frequencies to fall across the biological material so as to produce a nonlinear dielectric spectrum and obtaining a detectable signal corresponding to the spectrum obtained by measuring the response of the material at at least one response frequency which does not overlap with the initial frequency or frequencies.
The apparatus used in the method described in the above patent application comprises a spectrometer having two electrochemical cells. The first cell is a test cell containing a sample of the biological system whose nonlinear dielectric properties are being studied. The second cell is a reference cell comprising the supernatant of the test suspension. The electrochemical cells are connected to a sinusoidal oscillator via respective electrode assemblies, each comprising four electrodes
connected together in pairs. A test waveform is applied to the (outer) current electrodes, and data is logged from the (inner) voltage electrodes at a sampling frequency for a time specified by the operator to produce for analysis a test spectrum and reference spectrum relating to the materials in the first and second electrochemical cells respectively.
International (PCT) Patent Application WO 93/18402 describes a method of analysing and/or monitoring biological material, particularly cellular biological material in which a response is obtained from enzymes retained within cell membranes. Once again, an AC electrical potential at one or more discrete frequencies is applied to cellular biological material via the (outer) current electrodes of a four-electrode assembly and data is logged from the (inner) voltage electrodes to produce a non-linear dielectric spectrum for analysis. In this case, however, the electrode assembly is preferably arranged to be positioned substantially flush with, for example, the skin of a patient so that the described apparatus can be used for non-invasive monitoring of physiological parameters of the patient.
Both of the above-described arrangements can be used to analyse or monitor many different types of living or viable cell material, any metabolic substrate such as oxygen, glucose, lactic acid or lactate, or the like, any inhibitor that affects metabolite concentrations, such as vanadates, any biological product.
Contamination (or fouling) of metallic components (such as electrodes) used in the analysis of biological material using the above-described and other methods has often been a problem. Generally, contamination arises due to the presence of foreign substances, such as foreign proteins or the like. In an attempt to alleviate such problems, metallic electrodes have been scrupulously cleaned; however satisfactory cleaning has often not been achieved. Typically, if such contamination has not been eradicated, the biologically relevant signal has been substantially unstable, distorted or concealed. Acceptable low-noise, repeatable results have generally only been obtained when substantially clean metallic electrodes have been employed.
It has been found that the surfaces of metallic electrodes are particularly susceptible to contamination by foreign substances, such as foreign proteins, sugars, lipids or the like. Contamination by foreign proteins has been found to be especially problematic. For example, adhering foreign proteins tend to insulate part of an electrode surface, thereby altering the electrode polarisation impedance and hence the characteristics of apparatus, such as a spectrometer, employed in the analysis of biological material. Such electrodes have, therefore, needed careful checking for surface interactions with any foreign substance .
Electrode cleaning to ensure repeatable analysis, in particular repeatable spectroscopy (such as non-linear dielectric spectroscopy) , has been a complex and empirical
task, due to the lack of knowledge of the exact form of the causative mechanisms operating at an electrode/electrolyte interface. No certain ways of obtaining a quiet and repeatable reference signal from an individual electrode surface have been found but simple cleaning by abrasion has generally been employed in conjunction with applying a slowly alternating voltage to oxidise surface contaminants and subsequently remove oxidation products by the reduction half-cycle. The electrode cleaning process has been deemed successful when a repeatable, artefact-free signature has been obtained from a well-known and reliable reference system, such as a resting yeast suspension or the like. When a clean, recognisable, repeatable, artefact-free harmonic signature has been obtained from a frequency sweep covering its frequency range, the electrodes have been considered ready for use.
A further problem that has been experienced with the use of electrodes in the spectroscopy of biological material is that the maximum amplitude of signal has also been limited.
This has been because visible electrolysis has occurred whenever the voltage much exceeded + 1.5V zero-to-peak, and at suc voltages the electrode surfaces were much more susceptible to contamination and consequently the results became unstable.
A still further problem that has been encountered is that electrodes cleaned using existing techniques often become unstable within a few minutes. Continual control readings have, therefore, been vital during any series of
experiments to be sure the electrode surface behaviour has not substantially altered in use, in which case the results have needed to be abandoned and the experiments repeated.
There is a need, therefore, for the provision of clean, substantially non-contaminated metallic electrodes, that alleviate the hitherto described fouling associated with prior art electrodes, and that can be reliably employed in the analysis of biological material, particularly in the spectroscopical analysis of such biological material, such as non-linear dielectric spectroscopy substantially as described in UK Patent Specification 2247530. The present invention meets the above need and alleviates the hitherto problems associated with the prior art techniques.
According to the present invention there is provided a metallic electrode having at least one surface thereof provided with a protective coating, which protective coating is substantially electrically transparent and comprises at least one copolymer comprising at least a first polymer moiety attaching at least a second polymer moiety to said surface, said at least second polymer moiety extending from said surface so as to inhibit contamination of said surface by at least one foreign substance.
The first polymer moiety can suitably comprise at least one hydrophobic oligomer to which the at least second polymer moiety is attached. Preferably, the hydrophobic oligomer may comprise polypropylene oxide or the like. Advantageously, the hydrophobic oligomer can attach to the
surface of the metallic electrode and thereby substantially anchor at least the second polymer moiety thereto.
Preferably, the second polymer moiety is effective in obviating surface contamination of the metallic electrode as a result of physicochemical forces, such as steric repulsion, so as to substantially preclude attraction between the foreign substance and the protective coating and/or the metallic electrode surface. Suitably, the second polymer moiety may comprise at least one hydrophilic oligomer, such as polyethylene oxide or the like, arranged to extend from the surface of the metallic electrode.
In a preferred embodiment, at least one surface of a metallic electrode according to the present invention is provided with a protective coating comprising at least a first polymer moiety comprising at least one hydrophobic oligomer directly attached to the electrode surface, and at least a second polymer moiety comprising at least one hydrophilic oligomer attached to the electrode surface by the first polymer moiety and arranged to extend from the electrode surface so as to obviate contamination substantially as herein before described. In use, the hydrophilic oligomer of the second polymer moiety is arranged to extend into an electrolytic test medium of a biological material being analysed. The test medium may typically comprise a suspension of the biological material being analysed, or a supernatant obtained therefrom. A further advantage associated with a coated metallic electrode according to the present invention is that such
a coated electrode can be used in the analysis of suspensions as described above. Prior art electrodes used in the analysis of biological material generally required provision of a supernatant before the analysis could be carried out. This provision of a supernatant often proved to be problematic and at times impossible.
An advantageous feature of the present invention is that the copolymer can be tailored so as to obviate contamination by specific foreign materials. In particular, properties of the polymer moieties can be correlated with properties of the foreign substance. For example, the properties of the second polymer moiety can be correlated with the properties of a foreign substance likely to effect contamination of the surface of the metallic electrode. Suitably, where the foreign substance comprises a small protein molecule or the like, chains of the second polymer moiety can be arranged in relatively high surface density on the electrode surface. Alternatively, where the foreign substance comprises a large protein molecule or the like, the chains of the second polymer moiety can be arranged so as to be in relatively low surface density on the electrode surface. In both cases, it is generally preferred that the spacing between chains of the second polymer moiety extending from the electrode surface should be selected to be substantially less than the largest dimension of the foreign substance .
Optimisation of the copolymer properties substantially as
described above can suitably be achieved by the use of a copolymer having a general structure as shown by Formula (I) below (although of course other suitable copolymers may be used)
(P2)x - Px - (P2)2
(I) wherein
P-L denotes the first polymer moiety;
P2 denotes the second polymer moiety;
x is an integer such as 1 or 2; and
z is an integer typically in the range of 0 to 2.
A preferred copolymer for use in the present invention can be represented by Formula (la) below
[(M2) X - (Mx)b - [(M2)c]z (la)
wherein
Mx is a monomer from which the first polymer moiety is derived;
M2 is a monomer from which the second polymer moiety is derived;
x and z are substantially as herein before described with reference to Formula (I) and a, b and c denote the number of monomer units in each polymer moiety, where a, b and c, which can be the same or different, are each an integer of not less than 10.
It may be preferred that only one copolymer is employed in the protective coating used in the present invention. Alternatively, more than one copolymer may be employed in the protective coating used in the present invention. For example, the protective coating may include first and second polyethylene oxide - polypropylene oxide polyethylene oxide triblock copolymers, wherein the number of monomer units in one, or more than one, of the polymer moieties (denoted by a, b and c substantially as herein before described) of the first triblock copolymer differ from the number of monomer units (again as denoted by a, b and c) of respective polymer moieties of the second triblock copolymer.
In the case where a high surface density of chains of the second polymer moiety is required on the electrode surface substantially as herein before described, then b is selected so as to be a relatively low integer typically in a range of 30 to 40. Alternatively in the case where a low surface density of the chains of the second polymer moiety is required substantially as herein before described, then b is selected so as to be a relatively high integer, typically in the range of 60 to 80. Furthermore, the length of the chains of the second polymer moiety can be
selected so as to optimise repulsion of foreign substances thereby.
As described herein, it is preferred that the first polymer moiety comprises at least one hydrophobic oligomer, such as polypropylene oxide, and the second polymer moiety comprises at least one hydrophilic oligomer, such as polyethylene oxide. A preferred copolymer represented by formula (1) thus comprises a polyethylene oxide polypropylene oxide - polyethylene oxide triblock copolymer and a particularly preferred embodiment of the present invention comprises a metallic electrode having at least one surface thereof provided with a protective coating, which protective coating is substantially electrically transparent and comprises at least one polyethylene oxide - polypropylene oxide - polyethylene oxide triblock copolymer, whereby polypropylene oxide attaches to the electrode surface and polyethylene oxide extends from the polypropylene oxide so as , to obviate contamination of the electrode surface substantially as herein before described.
The term "substantially electrically transparent" as used herein to describe a protective coating employed according to the present invention refers to the property of the coating to exert substantially no influence on the measuring ability of a metallic electrode according to the present invention. In particular, the term "substantially electrically transparent" as used herein denotes the property of the protective coating employed according to the present invention, whereby the use of the protective
coating has substantially no effect on electrical interface conditions between a metallic electrode and electrolytic biological material being analysed.
A metallic electrode according to the present invention is of the type suitable for use in the analysis of biological material and preferably comprises a metallic electrode suitable for use in the spectroscopical analysis of electrolytic biological material, preferably non-linear dielectric spectroscopical analysis substantially as herein before referred to.
A foreign substance substantially as herein before described typically denotes one or more foreign proteins, sugars, lipids or the like, often associated with biological material requiring analysis. It is particularly advantageous that the present invention can obviate contamination of a metallic electrode by one or more foreign proteins.
The present invention further provides, in combination, a metallic electrode substantially as herein before described and electrolytic biological material requiring analysis. Typically the analysis is arranged to investigate or monitor a selected metabolic parameter, such as glucose concentration or the like, or material present in cell suspensions .
There is further provided by the present invention use of a protective coating substantially as herein before
described in obviating contamination of at least one surface of a metallic electrode by at least one foreign substance, which protective coating is substantially electrically transparent and comprises at least one copolymer comprising at least a first polymer moiety for attaching at least a second polymer moiety to said surface, which second polymer moiety can be arranged to extend from said surface so as to obviate said contamination.
The invention further provides a method of substantially obviating surface contamination of at least one surface of a metallic electrode by at least one foreign substance, which method comprises providing at least one surface of said metallic electrode with a protective coating which is substantially electrically transparent and comprises at least one copolymer comprising at least a first polymer moiety for attaching at least a second polymer moiety to said surface, which second polymer moiety can be arranged to extend from said surface so as to obviate said contamination of said surface.
The present invention is still further concerned with apparatus for use in the electrolytical analysis of biological material, which apparatus comprises at least one metallic electrode provided with a protective coating substantially as herein before described, and preferably the apparatus further comprises a test chamber for receiving a sample of electrolytic biological material requiring analysis and wherein a metallic electrode substantially as herein before described is arrangeable in
contact with a solution of the electrolytic biological material . Preferably apparatus according to the present invention comprise a spectrometer, suitably a non-linear dielectric spectrometer substantially as herein before described.
The invention still further provides in combination apparatus substantially as herein before described and electrolytic biological material requiring analysis, typically in respect of a selected metabolic parameter substantially as herein before described.
The present invention also provides a method of analysing an electrolytic biological material (typically by measuring a selected metabolic parameter thereof) , which method comprises using at least one metallic electrode substantially as herein before described in the analysis of the biological material. In particular, the present invention provides a method of obtaining a non-linear dielectric spectrum of electrolytic biological material ,- which method uses at least one metallic electrode substantially as herein before described in carrying out the non-linear dielectric spectroscopy.
The present invention will now be further illustrated by the following Examples and accompanying Figures, which do not limit the scope of the invention in any way.
The following is a description of the Figures referred to in the Examples:
Figure 1 : Power spectrum produced by NLDS (nonlinear dielectric spectrometer) . The fundamental frequency was 13Hz.
Figure 2 : Coefficient of Variation of 3rd harmonic of electrode response as a function of voltage and frequency for (a) uncoated gold electrode and (b) coated electrode .
Figure 3: Identification of the presence of cells in a suspension by electrode.
Figure 4: PLS predictions of glucose levels between separate fermentations. The (non-averaged) results obtained using non-coated electrodes are shown. The rmsep was 53% at 3 factors.
Figure 5 : PLS prediction * between separate fermentations using uncoated electrodes and median averaged data. The rmsep was 41% at 3 factors.
Figure 6: PLS prediction of the set of 12 produced by coated electrodes employed according to the present invention using median averaged data. .The rmsep was 45% at 6 factors.
Figure 7: PLS prediction of a fermentation taken 2 months after the application of the coating to the electrodes according to the present invention. The rmsep was 38% at 2 factors.
Figure 8: A schematic cross-sectional representation of an electrode assembly comprising four electrodes each according to an exemplary embodiment of the present invention.
Referring to Figure 8 of the drawings, four metallic electrodes 10, connected in pairs, are provided with a protective coating 12. The protective coating 12 is substantially electrically transparent and preferably comprises at least one polyethylene oxide - polypropylene oxide - polyethylene oxide triblock copolymer, whereby polypropylene oxide attaches to the electrode surface and polyethylene oxide extends from the polypropylene oxide so as to obviate contamination of the electrode surface. Wires 14 connect the electrodes 10 to a non-linear dielectric spectrometer (not shown) .
Example 1
An electrode assembly consisting of four electrodes connected in pairs' was used (as shown schematically in Figure 8 of the drawings) . Electrical energy was applied via the outer two electrodes whilst the transformed signal was detected across the inner two electrodes. The electrodes were cleaned by known abrasion techniques. The chamber was filled with lOOmM KCl and 60 spectrograms taken to study the stability of the electrode before coating.
The following control file was used:
5 9 (a)
04.5 0.75 1.0 1.25 1.5 (b)
10 17 31 56 100 177 316 562 1000 (c) .
Line (a) signifies that all combinations of 5 voltages and 9 sinusoidal frequencies were generated in the sweep. Line (b) gives the chosen voltages in Volts. Line (c) gives the chosen frequencies in Hz.
At each voltage/frequency combination, a power spectrum as in Figure 1 was produced and the amplitudes of harmonics 1 to 5 of this spectrum recorded. This led to each sample (object) being composed of 225 x-variables and 1 reference y-variable.
Spectrograms were taken at the following time intervals in minutes : -
Spectrograms 1-20 in 1 minute intervals (1-20 minutes) 21-40 in 2 minute intervals (22-60 minutes)
41-46 in 5 minute intervals (65-90 minutes) 47-49 in 10 minute intervals (100-120 minutes) 50-55 in 15 minute intervals (135-210 minutes) 56-60 in 30 minute intervals (240-360 minutes) .
The electrode was then coated with a PEO-PPO-PEO (polyethylene oxide - polypropylene oxide - polyethylene oxide) self assembling monolayer by filling the electrode chamber with a 2% wt/vol PEO-PPO-PEO solution available under the trade mark Synperonic PE/L64 and a 2% wt/vol PEO-
PPO-PEO solution available under the trade mark Lutrol P127 and leaving to evaporate to dryness. The above time study was repeated and the results compared.
The harmonics were separated and the Coefficient of Variation (C) was calculated for each voltage/frequency combination. The resulting 3rd harmonic surface of C vs . frequency (f) vs. voltage (V) is given in Figure 2a for the uncoated electrode and in Figure 2b for the coated electrode as above.
There was a significant improvement in stability for the coated electrode as above. The similarity in the shape of the two plots respectively shown in Figures 2a and 2b confirm that the PEO-PPO-PEO coating was electrically transparent and exerted substantially no effect on the electrical interface conditions between electrode and solution. This similarity was preserved through all harmonics 1-5.
The mean level of the above referred to plots was taken to give a single figure as an index of stability, along with the equivalent for all the other harmonics for both the coated and uncoated electrode. The mean of C can then be compared in Table IA.
Table lA
Similar figures for a repeat experiment on a second electrode are given in Table IB.
Table IB
The above tables demonstrate the improvement in stability due to a coating employed according to the present invention.
The Similarity in form of the C/V/f plots was also found to be repeated for all harmonics of the second electrode.
Exa ple 2
This Example describes the detection of cell suspensions and confirms that the coating described in Example 1 is electrically transparent.
A standard suspension of Saccharomyces cerevisiae was prepared as follows. Freeze-dried yeast (Allinson's baking yeast) was rehydrated to 50mg dry wt.ml"1 in a solution of 1% yeast extract w/v in distilled water. This was allowed to stabilise overnight to allow the yeast to exhaust its endogenous energy stores and enter a resting state before experiments were carried out. No cell growth occurred under the conditions used.
Yeast suspension was placed in the electrode chamber and a spectrogram taken as above but using the following control file:-
3 10
1.0 1.25 1.5
5 10 15 20 25 30 35 40 45 50.
With the first 5 harmonics at each voltage/frequency combination being recorded, this led to each sample
(object) being composed of 150 x-variables. A spectrogram of the supernatant was obtained with the conductivity exactly matched to that of the suspension with distilled water to compensate for the (insulative) volume fraction of the cells removed.
The process was repeated 30 times to produce 30 pairs of suspension and supernatant samples.
The 60 samples were split into 3 data sets. A Partial Least Squares (PLS) model was trained on one subset (the training set) and the validity of this model checked by predicting another (the validation set) . The third unseen subset was then predicted. The unseen data set (the test set) was classified as yeast-present vs. yeast-absent as shown in Figure 3.
Example 3
This Example describes the quantification of glucose in a batch fermentation. Having confirmed in Example 2 that the electrode coating employed according to the present invention as described in Example 1 has not compromised the ability of the spectrometer to detect cell suspensions, the ability of the coating to assist the spectrometer in modelling quantitative, metabolically significant parameters was studied. The chosen parameter was glucose concentration in batch fermentation of baker's yeast.
Three batch fermentations were carried out on consecutive days with a coated electrode as described in Example 1 using standard experimental protocols and data logging programs, to log non-linear dielectric spectrograms against time. Reference glucose concentration readings were provided by a hand-held glucometer (of the type available under the trade mark Reflolux) designed for at-home testing
by diabetics. A repeat triplicate of fermentations was then carried out on another coated electrode.
The control file used was the same as for the detection of suspensions above, but differential spectrograms were also recorded. A spectrogram was taken using suspension in the electrode chamber, then an identical one taken using supernatant to give the spectrogram due to the electrode interface; and the latter of these subtracted from the former to produce a spectrogram of cellular signal deconvolved from the electrode response. The suspension, supernatant and difference spectra were recorded to separate files. This led to each sample (object) being composed of 150 x-variables in each of the 3 files.
Five spectrograms were recorded with no glucose present, then glucose was added to a concentration of 170mM and spectrograms were taken every 2 minutes until the glucose was used up. For this yeast strain, this typically took 3 hours from the addition of the glucose. (Cell counting procedures showed that cell growth did not occur in these experiments.) Five subsequent spectrograms were recorded after the glucose levels in the suspension had reached zero .
The reference glucose levels were measured every third spectrogram with the above referred to glucometer. The measurement precision was calculated to be ± 10%. The glucometer had a detection range of 0.5mM to 27mM, so the higher glucose concentration samples used in yeast work
were diluted before readings were taken.
Each batch fermentation produced a data set. Median averaging in each variable with respect to sample number was used on these data sets, relying on the fact that glucose-related phenomena changed slowly during fermentation, and any sudden changes were spuria.
PLS modelling was carried out for the three sets of data collected from one electrode using one data set for the training set, a second data set as the validation set and the last data set as the test set. Six different combinations of prediction were derived from the 3 data sets. This procedure was repeated on the data sets from the second electrode to produce 6 more predictions . The Root Mean Square Error of Prediction (rmsep) was. used in forming and evaluating these PLS models.
Previous work had produced a typical prediction from three data sets of deconvolved difference spectra produced exactly as above but with uncoated electrodes and is shown in Figure 4. The rmsep was 41% at three factors. This prediction took many months of electrode cleaning and retaking of fermentation data until the electrode stayed stable and with no significant fouling for the three day test period. This formed the base line which the coating improved upon.
In prediction figures referred to herein, the measured reference data are shown as a solid line and the
multivariate predictions are shown by individual points. The upper plot gives the relation of these points to the ideal 1:1 line of perfect prediction; and the lower plot shows the actual function being modelled along with the relation of the predicted points to this function.
PLS predictions of a coated electrode prepared according to Example 1 are shown in Figures 5 and 6 using median averaged data. The rmsep obtained from Figure 5 is 53% at three factors. The rmsep obtained from Figure 6 is 45% at 6 factors .
Example 4
This example describes the stability of a protective coating prepared according to Example 1.
The long term stability of the coating was tested over a period of 2 months with further fermentations and predictions. After this period, one of the electrodes showed minor deterioration, producing the prediction of Figure 7 to compare with that of Figure 6. The other electrode began to show slightly more deterioration, suggesting that two months is close to the maximum period over which the coating can be considered stable enough for measurement. It is recommended that, for reliable results, re-coating be carried out at least monthly.