m01 · Bioelectric sources: transmembrane current, dipoles and double layers
Education only. Not for diagnosis, treatment or any clinical decision.
In 60 seconds
Every electrogram starts with current crossing cell membranes. Inside a strand of heart muscle, current flows down the voltage gradient; wherever that current changes, the difference leaves or enters through the membrane. Across a wavefront this gives a source of outward current just ahead of the front and a sink just behind it. A source and a sink close together form a current dipole, whose potential falls off as one over distance squared, faster than a single source. Spread dipoles over a surface and you get a double layer, whose potential is simply proportional to the solid angle it subtends at the electrode. Put together, a propagating front is a moving line of dipoles of strength \(\beta\,\Delta V\), set by the voltage jump and not by how steep the front is. Seen from a height \(h\), a front of width \(w\) looks exactly like an infinitely thin front at height \(w + h\): that is the unipolar electrogram of m03. Uniformly resting or uniformly activated tissue makes no potential at all; only fronts do.
From the field
- Far-field signals are dipoles seen from afar. Unipolar recordings carry far-field signal that can hide small local signals in diseased tissue (Stevenson and Soejima 2005). A distant front contributes like a dipole, falling as one over distance squared (Equation 2), so it fades with distance but never vanishes.
- Signals come from fronts, not from tissue states. In this model, tissue that is uniformly at rest or uniformly activated drives no membrane current and so produces no potential (Equation 4; lab part F). Solid-angle theory reads the ECG this way (Holland and Arnsdorf 1977). A quiet electrogram means no front nearby, not necessarily no tissue.
- A steeper front is not a stronger source. The front’s dipole moment is \(\beta\,\Delta V\) whatever its width (Equation 5). What width changes is how far the source seems to be from the electrode (Equation 6).
- Tissue is not isotropic. In calf trabecular muscle, conduction was 3.0 times faster along fibers than across them, and the intracellular resistance across fibers was 9.4 times that along them, while the extracellular ratio was 2.7 (Clerc 1976). Because the two ratios differ, the uniform double layer used here is an approximation; m09 treats anisotropy properly.
Derivations
Symbols used throughout. \(\sigma\) conductivity of the medium around the tissue (S/m). \(I\) current of a point source (A). \(\mathbf p\) current dipole moment (A·m). \(\tau\) double-layer strength, dipole moment per unit area (A/m). \(\Omega\) solid angle (sr). \(V_m\) transmembrane voltage. \(\Delta V\) voltage jump across a front. \(\beta\) intracellular conductance of a strand or sheet, as in m03. \(I_i\) intracellular current. \(i_m\) membrane current, positive outward. \(w\) front width, \(h\) electrode height, \(s\) signed distance from the front (positive where it has not arrived), all in mm. Results are in relative units with \(\sigma = \beta = \Delta V = 1\) unless stated.
We treat the medium as infinite, homogeneous and isotropic, and the fields as quasi-static, the standard starting model (Plonsey and Barr 2007). m02 adds boundaries and different media.
Point sources
A point source injecting current \(I\) into an infinite conductor drives a current density \(I/(4\pi r^2)\) outward through every sphere around it. With \(\mathbf J = \sigma\mathbf E\) and \(\mathbf E = -\nabla\phi\),
\[ \phi(\mathbf r) = \frac{I}{4\pi\sigma\,\lvert\mathbf r - \mathbf r'\rvert} \tag{1}\]
It satisfies Laplace’s equation everywhere except at the source, and the current through any closed surface around it is \(I\) (lab part A). Potentials from several sources add.
Key takeaways
- A point source’s potential falls as one over distance.
- Superposition: any source distribution is a sum of point sources.
Dipoles
Put a source \(+I\) at \(\mathbf r_0 + \mathbf d/2\) and a sink \(-I\) at \(\mathbf r_0 - \mathbf d/2\). As \(d \to 0\) with \(\mathbf p = I\mathbf d\) fixed, the pair becomes a current dipole:
\[ \phi(\mathbf r) = \frac{\mathbf p\cdot(\mathbf r - \mathbf r_0)}{4\pi\sigma\,\lvert\mathbf r - \mathbf r_0\rvert^3} \tag{2}\]
Close to one pole the pair looks like a single source (slope −1 on log-log axes); far away, like a dipole (slope −2). The difference between the pair and the ideal dipole falls as the square of the separation; the next term is a quadrupole (lab part B). Since every membrane current has a matching return current, the far field of any patch of active tissue starts at the dipole term.
Key takeaways
- A dipole’s potential falls as one over distance squared and changes sign across the plane through its center.
- Far-field signals in electrograms are dominated by the dipole term.
Double layers and solid angles
Spread dipoles over a surface with moment \(\tau\) per unit area, all normal to it. A patch \(dA\) seen at distance \(r\) and angle \(\theta\) from its normal contributes \(\tau\,dA\cos\theta/(4\pi\sigma r^2)\), and \(dA\cos\theta/r^2\) is the solid angle it subtends. A uniform double layer therefore gives
\[ \phi = \frac{\tau\,\Omega}{4\pi\sigma} \tag{3}\]
with \(\Omega\) signed positive on the side the dipoles point to. On the axis of a disk of radius \(R\) at height \(z\), \(\Omega = 2\pi(1 - z/\sqrt{z^2 + R^2})\). Three consequences follow. Close to a large layer the potential barely changes, because \(\Omega\) stays near \(2\pi\); far away the disk looks like one dipole of moment \(\tau\pi R^2\). Crossing the layer, \(\Omega\) jumps by \(4\pi\) and the potential by \(\tau/\sigma\). A closed double layer subtends zero solid angle from outside, so it produces no potential there (lab part C). This is the basis of solid-angle theory of the electrocardiogram (Holland and Arnsdorf 1977).
Exact
Assumptions
- Model: Equation 1, Equation 2 and Equation 3 in an infinite homogeneous conductor, evaluated on the axis of each source. Tested against
epml.sources. - Parameters: relative units (unit current, unit dipole moment, unit layer strength, \(\sigma = 1\)); disk radius 0.5–10 mm.
- What it ignores: boundaries and changes in conductivity (m02), off-axis positions, and the finite size of real electrodes.
Key takeaways
- A uniform double layer’s potential is its strength times the solid angle it subtends, over \(4\pi\sigma\).
- Only the edge of a layer matters for a closed surface; uniformly activated regions make no potential.
The wavefront
Along a thin strand, intracellular current flows down the voltage gradient. Wherever that current changes, the difference crosses the membrane:
\[ I_i = -\beta\,\frac{\partial V_m}{\partial x}, \qquad i_m = -\frac{\partial I_i}{\partial x} = \beta\,\frac{\partial^2 V_m}{\partial x^2} \tag{4}\]
This assumes the extracellular potential gradient is small next to the intracellular one, as in m03 (eq-m03-cable). For the arctangent front of m03 (eq-m03-front), \(i_m\) is outward ahead of the front and inward behind it. Its integral is zero, and its first moment, the front’s dipole moment per unit length of front, is
\[ p = \int s\,i_m\,ds = \beta\,\Delta V \tag{5}\]
for every front width (integrate by parts; lab part D). Seen edge-on, the front is a line of these dipoles. The 2-D potential of a line source is \(-\ln r/(2\pi\sigma)\) per unit current, so a line of dipoles in the tissue plane, seen from height \(h\), gives
\[ \phi(s) = \frac{p\,s}{2\pi\sigma\,(s^2 + h^2)} \tag{6}\]
Summing the field of every membrane-current strip of the arctangent front gives exactly \(\beta\Delta V\,s/(2\pi\sigma(s^2 + (w + h)^2))\), the unipolar electrogram of m03 (eq-m03-unipolar-closed). So a front of width \(w\) at height \(h\) is indistinguishable from an infinitely thin front, Equation 6, at height \(w + h\) (lab part E).
Exact
Assumptions
- Model: Equation 4 for the arctangent front of m03 and Equation 6, in an infinite homogeneous conductor; planar front, thin sheet. Tested against
epml.sources. - Parameters: relative units (\(\beta = \Delta V = \sigma = 1\)); front width 0.05–3 mm and height 0.2–5 mm, the range used in m03.
- What it ignores: curved fronts, anisotropy (Clerc 1976), the ionic detail of the upstroke (m08) and boundaries (m02).
Key takeaways
- Ahead of a front, current leaves the cells (source); behind it, current enters (sink).
- The front’s dipole moment is \(\beta\,\Delta V\); its width acts like extra electrode height.
Clinical correlate
This section describes published findings only. It is not guidance for patient care, and it has not yet been reviewed by an EP physician.
Where this shows up in EP. Every unipolar and bipolar electrogram recorded by a mapping catheter is the potential of membrane-current sources near and far (m03). Reviews of recording technique describe the trade-off this module explains: unipolar recordings see local fronts and far-field activity, and the far field can hide small local signals in diseased tissue (Stevenson and Soejima 2005; Bakker 2019). Solid-angle theory has long been used to interpret the surface ECG in terms of activation fronts and their boundaries (Holland and Arnsdorf 1977).
What the evidence is. Reviews and theory papers (Stevenson and Soejima 2005; Bakker 2019; Holland and Arnsdorf 1977) and one experimental study of anisotropy in calf muscle (Clerc 1976). The physics in this module is standard; how far it explains a particular electrogram depends on geometry, anisotropy and boundaries that the simple model leaves out.
What a reader should ask:
- Where is the front relative to the electrode? Its distance sets the falloff (Equation 2) and the shape (Equation 6).
- Is the signal from a front at all? Uniform tissue makes no potential (Equation 4).
- What would anisotropy or a boundary change? The uniform-layer picture is an approximation (Clerc 1976).
Visual summary
Schematic
Assumptions
- Schematic: not to scale, no quantitative claim. Labels follow Equation 1 to Equation 6.
How experts think
When an electrogram looks surprising, ask:
- Where are the sources? Fronts make potentials; uniform tissue does not (Equation 4).
- How far away are they? Point-like near, dipole-like far (Equation 1, Equation 2).
- What solid angle does the front subtend at the electrode? A large front close up gives a large, slowly changing potential (Equation 3).
- What does width do? It acts like extra height and lowers the amplitude without changing the dipole moment (Equation 6).
- Which assumption breaks? Boundaries, blood and tissue (m02), anisotropy (Clerc 1976), curvature.
One sentence to keep: an electrogram is the solid-angle view of moving source–sink pairs, and only the fronts are visible.
Cheat sheet
| Quantity | Result |
|---|---|
| Point source (Equation 1) | \(\phi = I/(4\pi\sigma r)\); current through any surface around it is \(I\) |
| Dipole (Equation 2) | \(\phi = \mathbf p\cdot\hat{\mathbf r}/(4\pi\sigma r^2)\), \(\mathbf p = I\mathbf d\); falls as \(1/r^2\) |
| Double layer (Equation 3) | \(\phi = \tau\Omega/(4\pi\sigma)\); jump \(\tau/\sigma\) across; zero outside a closed layer |
| Disk on its axis | \(\Omega = 2\pi(1 - z/\sqrt{z^2 + R^2})\) |
| Membrane current (Equation 4) | \(I_i = -\beta\,\partial V_m/\partial x\), \(i_m = \beta\,\partial^2 V_m/\partial x^2\); net zero |
| Dipole moment of a front (Equation 5) | \(p = \int s\,i_m\,ds = \beta\Delta V\), for any front width |
| Front as dipoles (Equation 6) | \(\phi = p\,s/(2\pi\sigma(s^2 + h^2))\); a front of width \(w\) at height \(h\) ≡ thin front at \(w + h\) |
Values used in this module
| Value | Source |
|---|---|
| Conduction velocity 3.0 times faster along fibers than across them (calf trabecular muscle) | Clerc (1976) |
| Intracellular resistance across/along fibers 9.4; extracellular 2.7 | Clerc (1976) |
| All other values: relative units, \(\sigma = \beta = \Delta V = 1\) | our own code |
Interview questions
Try each question before you open its model answer. In a real interview, say which assumptions you are making.
Q1. Why does a point source’s potential fall as 1/r?
Model answer
The current \(I\) spreads over spheres of area \(4\pi r^2\), so the current density is \(I/(4\pi r^2)\) and the field \(I/(4\pi\sigma r^2)\); integrating the field gives \(I/(4\pi\sigma r)\) (Equation 1).
Q2. Why can’t heart tissue produce a monopole far field?
Model answer
Charge is conserved: every outward membrane current has a matching inward current nearby, so the net source is zero and the expansion starts at the dipole term (Equation 4, Equation 2).
Q3. How does a dipole’s potential depend on direction and distance?
Model answer
As \(\cos\theta/r^2\) (Equation 2): largest along the dipole, zero on the plane through its center perpendicular to it, and of opposite sign on either side.
Q4. What is a double layer, and what is its potential?
Model answer
A surface covered with dipoles normal to it, moment \(\tau\) per unit area. Its potential is \(\tau\Omega/(4\pi\sigma)\), where \(\Omega\) is the signed solid angle it subtends (Equation 3).
Q5. What happens to the potential as you cross a double layer?
Model answer
It jumps by \(\tau/\sigma\), because the solid angle changes from about \(+2\pi\) to \(-2\pi\) (Equation 3).
Q6. Why does uniformly activated tissue produce no potential?
Model answer
With \(V_m\) uniform, \(\partial^2 V_m/\partial x^2 = 0\), so no membrane current flows (Equation 4). In solid-angle terms, a closed double layer subtends zero solid angle from outside (Equation 3).
Q7. Derive the membrane current across a front.
Model answer
Intracellular current is \(I_i = -\beta\,\partial V_m/\partial x\); what it loses along the strand leaves through the membrane, \(i_m = -\partial I_i/\partial x = \beta\,\partial^2 V_m/\partial x^2\) (Equation 4). It is outward ahead of the front and inward behind it.
Q8. Does a steeper front make a bigger source?
Model answer
No. Its dipole moment is \(\int s\,i_m\,ds = \beta\,\Delta V\) for any width (Equation 5). A narrower front puts the source effectively closer to the electrode, which raises the peak potential (Equation 6).
Q9. Why does a front of width w at height h look like a thin front at height w + h?
Model answer
The front’s dipole density along \(s\) is \(-\beta\,\partial V_m/\partial s\), a Lorentzian of width \(w\) with total \(\beta\Delta V\). Smearing the line-dipole field at height \(h\) with a Lorentzian of width \(w\) gives the line-dipole field at height \(w + h\), so the widths add (Equation 6, eq-m03-unipolar-closed).
Q10. How does a large flat double layer look close up and far away?
Model answer
Close up it subtends nearly \(2\pi\), so its potential is almost constant; far away it is one dipole of moment \(\tau\) times its area, falling as \(1/r^2\) (Equation 3; demo m01-d1).
Q11. What does anisotropy do to this picture?
Model answer
Conductivities differ along and across fibers, and differently inside and outside cells: 9.4 versus 2.7 in one study (Clerc 1976). The equivalent sources are then no longer a uniform layer normal to the front, so the solid-angle picture is an approximation.
Q12. A unipolar signal shows a large slow deflection with no sharp downstroke. What might it be?
Model answer
A distant front seen as a dipole: far-field signal falls slowly and lacks the sharp local intrinsic deflection , Equation 2. Bipolar recording (m03) rejects much of it.
Code lab
The lab builds every source in this module in code and ends by reproducing the m03 unipolar electrogram from them.
| Part | What you do | Data |
|---|---|---|
| A | A point source: Laplace’s equation and current conservation (Equation 1) | synthetic |
| B | From a source–sink pair to a dipole; falloff slopes (Equation 2) | synthetic |
| C | A double layer and its solid angle; the jump across it (Equation 3) | synthetic |
| D | Membrane current across a wavefront; net zero and dipole moment (Equation 4) | synthetic |
| E | The wavefront as a line of dipoles; the m03 closed form (Equation 6) | synthetic |
| F | Why only wavefronts make potentials | synthetic |
- Python notebook: labs/python/m01_lab.ipynb. Run it from the course repository after
pip install -r requirements.txt, or open it in Google Colab. - GNU Octave or MATLAB: labs/octave/m01_lab.m covers parts A–E with the same checks. It needs no toolboxes.
- Data: none to download; everything is in relative units.
Each part ends with checks: lines that pass with the code as given. If you change something and a check fails, that is information, not an error to hide.
Self-check
Short checks you can do in your head. Open each answer after you have tried it.
S1. You double the distance from a point source. The potential?
Answer
Halves (Equation 1).
S2. You double the distance from a dipole. The potential?
Answer
Falls to a quarter (Equation 2).
S3. What solid angle does an infinite plane subtend from one side?
Answer
\(2\pi\), half the full sphere (Equation 3).
S4. Net membrane current across a complete front?
Answer
Zero (Equation 4).
S5. You halve the front width. The dipole moment?
Answer
Unchanged, \(\beta\,\Delta V\) (Equation 5).
S6. Front width 0.3 mm, electrode height 1 mm. An infinitely thin front would give the same signal at what height?
Answer
1.3 mm (Equation 6).
Literature figures
These figures are worth looking at alongside this module. We link to them instead of copying them: some licenses do not allow reuse here, and others we could not confirm. Each entry says what to look for and where this module explains it.
- Conduction along and across fibers. Clerc’s measurements of impulse spread in calf trabecular muscle, with longitudinal and transverse currents [Clerc (1976); free full text]. This module: the wavefront, and the limits of the uniform double layer.
Further reading
- Bioelectricity. Plonsey and Barr’s textbook covers sources, volume conductors and dipoles (Plonsey and Barr 2007).
- Solid-angle theory. Holland and Arnsdorf’s review of the ECG in terms of double layers (Holland and Arnsdorf 1977).
- Recording technique. Reviews of unipolar and bipolar recording in clinical EP (Stevenson and Soejima 2005; Bakker 2019).
- Related modules. m02 (volume conduction, boundaries and Poisson’s equation), m03 (electrograms as spatial filters), m08 (action potentials), m09 (propagation and anisotropy).