Why Electrodes Matter: Electrode-Electrolyte Interface
An electrode system can be divided into two main components – the Electrode and the Electrolyte. Where the electrolyte meets the electrode surface, an interface known as the Electrode-Electrolyte Interface is formed. The characteristics of this interface are very important in the recording of a high quality electrocardiogram (ECG).
The current generated from the heartbeat flows through this interface, allowing us to record the ECG. Within the metal electrode, this current is carried by electrons, however in the electrolyte and skin it is carried by ions. As you can imagine, the transition at this interface between one type of carrier to the other can inflict an impedance on this current. When designing an electrode, our aim is to reduce the impedance and potential across this site as much as possible. To enable us to do so, an understanding of the electrochemical interaction between the electrolyte and electrode is required.
Electrode Potential
When choosing a material for the ECG electrode, it may be perceived that the more conductive a material the better the performance. However, the half-cell potential of the chosen material is arguably just as important. If you choose a material with a high conductivity, but also with a high, unstable half-cell potential; baseline wander, attenuation and interference are inevitable.
At the electrode to electrolyte surface, there is a constant flow of electrons in and out of the electrode. When the current that flows in one direction, is equal to and cancels out the current flowing in the opposite direction; this is known as the equilibrium. The potential at this equilibrium can be known as the Equilibrium, Reversible or Half-cell potential. This potential depends on the composition and ionic concentration of the electrolyte and how this interacts with the electrode metal.
All electrodes have a half-cell potential. This potential is measured in reference to the half-cell potential of the Standard Hydrogen Electrode (SHE), which is said to have a potential of 0mV [1]. A list of varies electrode materials and their half-cell potentials can be seen below.
Electrodes can be separated into two categories depending on their half-cell potential; Polarisable and Non-polarisable. Electrodes with a half-cell potential closer to zero are known as Non-polarisable, and those which have relatively high half-cell potentials are known as Polarisable. It is impossible to produce a perfectly polarisable or non-polarisable electrode.
The term polarisation is used in reference to the change in potential of an electrode from its half-cell potential upon application of dc current [1]. The potential difference between these states can be referred to as the over-potential. The total potential of an electrode at any given time is therefore equal to the half-cell potential of the electrode material plus the over-potential produced.
Non-polarisable electrodes produce minimal over-potential when DC current is applied. In comparison to polarisable, which produce a very large over-potential with DC current application.
The size of the over-potential would not be an issue if the potential produced across both electrodes were identical, as these would simply cancel each other out. However, due to impurities within the electrode metal and slight variation in concentration of the electrolyte between electrodes this is not possible. Any potential difference produced between electrodes, will then get amplified along with the ECG. This will cause baseline wander(drift) and distortion in the ECG.
Although both non-polarisable and polarisable electrodes produce over-potential, non-polarisable electrodes produce a lower, more stable potential.
When designing an electrode, we should aim for a material with as low of a half-cell potential as possible.
The conducting electrolyte should also be taken into consideration; an electrolyte with a high ionic concentration will further reduce the half-cell potential. Using silver/silver-chloride electrodes as an example; the half-cell potential is usually estimated to be about 0.22 volts. If the electrolyte ionic concentration is saturated, the half-cell potential can reduce to as low as 0.20 volts. However, if the ionic concentration is lower, the half-cell potential can increase to approximately 0.27 volts [1]. Other methods to reduce the half-cell potential include increasing the electrode surface area which can be done by increasing the actual size of the electrode and/or surface roughening.
Electrode - Electrolyte Interface Equivalent circuit
As discussed above, the half-cell potential of the electrode has a direct effect on the ECG. The half-cell potential can be modelled in an equivalent circuit along with the multiple other components present at this interface.
A large or mismatched impedance can cause signal attenuation, signal distortion and interference within the ECG. Impedance at the electrode-electrolyte interface is a combined measure of the opposition to current through the electrode interface (resistance) and the ability to store charge at the interface (capacitive reactance) [2]. As current can be capacitively coupled as well as resistively transmitted through the electrode-electrolyte interface, we can model this in an equivalent circuit model as a resistor in parallel with a capacitor, as seen below.
The circuit model comprises of a double-layer capacitance (Cdl) in parallel with the charge transfer resistance(RCT). Both are in series with the half-cell potential of the electrode (Erev), and the relatively small resistance (Rs) due to the sum of the lead and electrolyte resistances.
As previously stated, current is transferred via ions in the electrolyte, whereas it is transferred via electrons in the electrode. This creates a resistance at this interface, which is known as the Charge Transfer Resistance (RCT). Ideally, we want the signal to flow through the interface unimpeded (RCT = 0). Rct is directly linked to the half-cell potential; the lower the half-cell potential, the lower this resistance.
The electrode-electrolyte interface possesses not only resistive properties (RCT), but also capacitive properties. The capacitance component within the equivalent circuit is known as the Double Layer Capacitance (Cdl). This double layer occurs at the electrode-electrolyte interface when a voltage is applied.
One layer is within the solid electrode; this layer is usually negative as metal is typically more negatively charged compared to the electrolyte. As opposite charges attract, solvated cations within the electrolyte move towards the negatively charged electrode. This forms the outer Helmholtz plane (OHP). These two layers are separated by a monolayer consisting of solvent molecules and anions that have been adsorbed onto the electrode surface. This is known as the inner Helmholtz plane (IHP) and forms our dielectric between the two oppositely charged layers. Current is diffused over this interface electrostatically, similarly to a regular capacitor. [3]
The double layer capacitance behaves similar to capacitors in that it is frequency dependent. As the frequency increases, the reactance within the capacitor decreases exponentially (known as the capacitors complex impedance).
Due to the higher impedance present within the capacitor at lower frequencies, the current passes resistively. At higher frequencies the impedance within the capacitor significantly decreases, and the current therefore passes capacitively. This explains the distortion that can be seen at lower frequency waveforms of the signal; such as the P, S and T waves, and modification of the S-T segment. [3]
Testing Standards
In order for an electrode to be commercially regulated for patient monitoring, they must comply with ANSI/AAMI EC 12 (2000) standard [3]. This is a standard produced by the Association for the Advancement of Medical Instrumentation (AAMI) which requires electrode manufacturers to analyse the characteristics of the electrode-electrolyte interface, ensuring they meet the minimum safety and performance requirements for clinical use. These tests are currently the only widely accepted electrode standard tests in use for ECG Electrodes [3].
All of these tests are conducted using a pair of electrodes placed gel-to-gel. The justification for all of these tests being conducted solely between electrodes, is due to the correlation found by UBTL [4] between the gel-to-gel and skin-to-gel testing. They state that impedance as measured on abraded skin, correlates well with the impedance measured with the electrodes connected gel-to -gel (correlation of 99%), however the impedance measured on clean, non-abraded skin correlates very poorly (correlation of 47%). This comparison was also carried out for DC offset, which UTBL found a correlation of a factor 2.5 between the offset measured on abraded skin, versus that measured with electrodes connected gel-to-gel [4]. These correlation values are used to give a skin-electrode theoretical value and therefore results in missing the more important real properties of the electrolyte–skin interface. Since the creation of this standard in 1979, skin impedance spectroscopy work has shown that the electrode-skin interface is independent to that of the electrode-electrolyte and the correlation found by UBTL does not provide us with an accurate value for this. Because of this, we can sometimes find electrodes that perform better on the bench tests do not always perform as well when used on the skin. Regardless, these are still very useful standards in analysing the electrode-electrolyte interface.
The range of requirements within this standard are:
AC Impedance
DC Offset Voltage
Combined Offset instability and internal noise
Bias current tolerance
Defibrillation overload recovery
1. AC Impedance
A high contact impedance within the electrode system can cause considerable signal attenuation, distortion and increase the 50/60HZ interference pickup (The effect of these on the ECG is displayed in the ECG Artefacts Blog post). Not only can high contact impedance produce a corrupted ECG, it can also cause the electrodes to heat up, running the risk of causing burns to the patient’s skin if used in conjunction with electrosurgery or defibrillator discharges [4].
The AC impedance bench test calculates the contact impedance present within the electrode-electrolyte interface by applying a sinusoidal current of known amplitude and observing the amplitude of the resulting voltage across electrodes. The magnitude of this impedance is then given by the ratio of the amplitude of output voltage to that of the current [4].
For this test, the AAMI have identified a maximum impressed current of 0.1mA, and a set frequency of 10Hz. All bench tests carried out were set to a frequency of 10Hz. The UBTL carried out testing at frequencies of 1Hz, 10Hz and 60Hz and observed a similar electrode behaviour at each frequency. It was therefore concluded that tests carried out at 10Hz would be valid at both 1 and 60Hz. If testing at least 12 pairs of electrodes, a maximum average value of 2kΩ across all of the electrodes must be met . In addition to this, a single pair of electrodes should not exceed 3KΩ [4].
The impedance at the electrode-electrolyte interface is minimal in comparison to the impedance present within the outer layer of the skin. Conductive gel, skin preparation and select positioning of the electrodes are therefore very important in producing a low impedance. Although this test is not an accurate representation of the overall impedance within the ECG, this standard ensures the electrode-electrolyte interface is not contributing significant impedance in addition to that of the skin. Details on this will be presented in the Skin-Electrolyte Interface blog.
Although not required by the standard, it is beneficial for the manufacturer to carry this test out at various frequencies (instead of just at 10Hz as done within this standard). Electrochemical Impedance Spectroscopy (EIS) is an electrical impedance measurement method which estimates the contact impedance and its phase angle over a range of frequency points [5]. By adding these extra frequency points it provides a more in depth analysis of the impedance within our electrode system.
2. DC Offset Voltage & 3.Combined Instability and internal noise
The DC Offset Voltage measures the voltage across a gel-to-gel electrode pair as a result of the difference in their electrode half-cell potentials. Ideally, the potentials of both electrodes should be identical and therefore cancel each other out. However, the small impurities within the metal electrode and slight ionic concentration variations prevalent in the gels results in differences between these potentials. These differences are then amplified along with our ECG causing DC offset. By using non-polarising electrodes within our ECG we can minimize this difference.
As well as potential difference between the electrodes, there is also a potential difference between various skin sites on the body; depending on the thickness of the skin, and the abundance of hair follicles and sweat glands. The difference of the skin tends to be much larger than what we see within the electrodes. When under excessive DC voltage, the DC offset will increase and cause the amplifier within the circuit to saturate, producing an unreadable ECG. A reasonable limit on the offset voltage is therefore required to prevent the electrodes from significantly contributing to the overall DC offset voltage [4].
To measure the DC offset voltage, two electrodes placed gel-to-gel form a circuit with a DC voltmeter; this must have a minimum input of 10MΩ and a resolution of 1 mV or better [4]. The measuring instrument will apply a maximum of 10 nA bias current to the electrodes under test, and begins after a 1-min stabilization period but before 1.5 minutes have elapsed. The electrodes should not exhibit an offset voltage greater than 100mV.
As the maximum allowable dc offset should be less than 300mV, the Committee decided that the limit for gel-to-gel dc offset should therefore be less than 300/2.5mV(i.e. 100mV). in correspondence to the correlation to a factor of 2.5 that was found as discussed above.
Combined offset instability and internal noise is a bench top test evaluating the baseline wander(also known as offset drift) prevalent in the ECG electrode. This standard is carried out using the same measurement technique as the DC Offset Voltage, however it is carried out over a period of 5 minutes (after the initial 1 minute stabilization) to provide us with a rate at which the offset drifts.
The limit for the level of offset drift is giving by a rating system produced by the American College of Cardiology’s Task Force on the Quality of Electrocardiographic Records. The highest rating is reserved for a drift of less than 0.1mV/sec, and baselines ranging from 0.1-0.4mV/sec were said to be less desirable but acceptable [4]. Using the outer limit of this, 0.4mV/sec, and dividing it by the factor of 2.5 again, we get our offset drift limit of 0.150mV/sec. The pair of electrodes should therefore not generate a voltage greater than 150mV peak-to-peak in the passband of the 0.01– 1000 Hz.
4. Electrode Tolerance to Bias Currents
When a DC current passes through the electrode-electrolyte interface, it is no longer in equilibrium. Upon DC current application the reactants for the chemical reactions that occur here, may become depleted and cause significant deviation from the electrode half-cell potential [4], otherwise known as polarisation (described above in ‘Electrode Potential’ section). The potential difference we witness at this interface, is also prevalent within the skin; albeit much bigger. This can be caused by excessive motion, loss of contact between the skin and the electrode, and disconnected leads. Many cardiac monitoring systems have a limit set so that the system will notify clinicians when experiencing large offset voltages that are indicative of disconnected leads or loss of contact between electrode and skin. If the potential difference within the electrode-electrolyte interface is drastic it will also set of these alarms.
The electrode tolerance to bias currents is typically much better in non-polarised than in polarised, due to the much slower polarisation that takes place within the non-polarised electrode.
To analyse this, we apply a set input bias current and continuously monitor the offset voltages. It is recommended that the change in electrode potential be less than 100mV when subjected to a continuous bias current of 200nA. The test must be carried out for the period of time the manufacturer recommends the electrodes be worn without needing replacement. This must be a minimum of 8 hours [4].
5. Defibrillation overload recovery
The requirement for ECG electrodes to withstand defibrillator overload has been incorporated in the FDA draft standards for electrocardiographic and cardiac monitors. The cardiac monitor has added circuit components in place that allow the overload current of the defibrillator impulse to harmlessly bypass the input stage of the monitor. However when this current passes over the electrodes it can cause polarisation of the electrode. As the electrode becomes more polarised at each impulse, the potential difference between electrodes increase and the DC offset voltage increases. This can quickly render an ECG unreadable.
It is crucial that a clinician, in the event of defibrillating or cardioverting, be able to see an uncorrupted ECG within 5–10s in order to judge the efficacy of the delivered impulse and to decide if another is required. The ECG must not only be visible on the monitor but also be recognizable and clinically useful [3]. The offset voltage across the electrode-skin interfaces, which drastically increases as a result of the defibrillation impulse, must therefore return to below 300mV within 5s following the discharge. The offset voltage also should not drift with time by more than 1mV/s [4].
To measure the defibrillation overload recovery, the electrode pair are connected to the circuit as seen below. During animal testing, it was found the electrodes absorb an average of 1.5millicoulombs and rarely more than 2 millicoulombs of unidirectional charge when the defibrillation pads are touching the electrodes. The following simulation was created to represent the worst case scenario, with the product of the voltage multiplied by the capacitance of the discharge capacitor equal to 2 millicoloumbs.
With SW1 closed, and SW2 & 3 open, the capacitor is allowed to charge. It must be charged for at least 10 seconds until it is fully charged to 200V, and then SW1 can be opened. The capacitor is then discharged by closing SW2 for a maximum of 2 seconds, which releases a surcharge of voltage into the electrodes. SW2 is then opened and SW3 closed immediately, connecting the electrodes to the offset measuring system. The electrode offset is recorded to the nearest 1mV 5 seconds after the closure of switch SW3 and every 10 seconds thereafter for the next 30 seconds. The overload and measurement are repeated three times. The 5-sec offset voltage after each of the four discharges of the capacitor shall not exceed 100 mV, and any difference in the following 10 second interval should not exceed ± 11 mV (± 1 mV/sec) [4].
In the latest standard for cardiac monitors and electrocardiographs specifies a ± 300mV offset. This would however only stand if the offset voltage had a very small rate of change below 1mV/sec, which wasn’t the case with the electrodes that produced an offset voltage of over 100mV during the testing. The standard therefore set the limit to 100mV, which presents no problem to most of the silver-silver chloride electrodes available today [4]. All of the electrodes which did not satisfy this limit, where polarisable electrodes.
Overview
With the understanding of the electrode-electrolyte interface, it should now be easy to see why the silver/silver-chloride electrodes are our gold standard in ECGs today. Not only are these very cheap and easy to manufacture, due to there low half-cell potential it also makes them a very stable electrode for recording the ECG. This low half-cell potential provides minimal over-potential with DC current application within the ECG. As well as this, the silver-chloride coating on the silver helps reduce the charge transfer resistance (Rct). This is due to the combination of both metal ions and chloride ions within the surface coating, allowing the current to pass easily between the chloride ions present in the electrolyte and the metal ions in the electrode. All of these contribute to a very stable, reliable and cheap electrode.
In the next blog we will go into detail on the Skin-Electrolyte interface and artefacts which arise from problems we see here.
Sources :
[1] Macy A. Chapter 4: Electrodes – Alan Macy [Internet]. Alan Macy. 2017 [cited 2020 Sep 2]. Available from: https://alanmacy.com/books/the-handbook-of-human-physiological-recording/chapter-4-electrodes/
[2] Myllymaa S, Pirinen S, Myllymaa K, Suvanto M, Pakkanen TA, Pakkanen TT, et al. Improving electrochemical performance of flexible thin film electrodes with micropillar array structures. Measurement Science and Technology. 2012 Oct 30;23(12):125701.
[3] McAdams E, Webster JG. Medical Devices and Instrumentation / Bioelectrodes [Internet]. 2nd ed. New York: Wiley-Interscience; 1988. Available from: https://onlinelibrary.wiley.com/doi/abs/10.1002/0471732877.emd013
[4] IEEE Recommended Practice for Measurements and Computations of Electric, Magnetic, and Electromagnetic Fields with Respect to Human Exposure to Such Fields, 0 Hz to 100 kHz. New York, Ny, Usa Ieee; 2010.
[5] Ghoshal D, Bera T. Comparison of two and four electrode methods for studying the impedance variation during cucumber storage using Electrical Impedance Spectroscopy (EIS. 2017; Available from: https://www.researchgate.net/publication/314195670_Comparison_of_two_and_four_electrode_methods_for_studying_the_impedance_variation_during_cucumber_storage_using_Electrical_Impedance_Spectroscopy_EIS
