Deriving Membrane Flux From Permeability

From a Membrane Concentration Gradient to a Transfer Rate

When a drug crosses a cell membrane, the concentration difference between the two sides is not by itself a transfer rate. The rate also depends on how readily the drug crosses the barrier and how much membrane is available. A useful starting point is therefore to separate three quantities: flux through a unit of membrane area, the total membrane area, and the concentration difference that drives net transfer.

The relation derived here describes passive transfer under specified assumptions. It is a constitutive relation for membrane crossing, not a complete tissue model: it does not, on its own, describe how concentrations change over time in the surrounding compartments.

The sign convention carries information: if \(C_1>C_2\), net flux is from side 1 to side 2; if \(C_1<C_2\), it is from side 2 to side 1. If the two concentrations are equal under the model’s assumptions, the net flux is zero. This does not mean that individual molecules stop moving; it means opposing molecular movements balance.

Deriving the Linear Permeability Relation

Consider a flat membrane of thickness \(L\). Let \(x\) measure distance through the membrane, from side 1 at \(x=0\) to side 2 at \(x=L\). Fick’s first law states that diffusive flux within the membrane is proportional to the negative concentration gradient:

where \(C_m\) is drug concentration in the membrane and \(D_m\) is the drug’s diffusion coefficient there. The minus sign indicates movement down the concentration gradient. Suppose the membrane is homogeneous, \(D_m\) is constant, and the membrane concentration profile is at steady state. The gradient is then constant across the membrane, so:

Here \(C_{m,1}\) and \(C_{m,2}\) are the membrane concentrations at its two interfaces. If drug partitions between the aqueous solution and membrane with the same partition factor \(K_{m/w}\) at both interfaces, then \(C_{m,1}=K_{m/w}C_1\) and \(C_{m,2}=K_{m/w}C_2\). Substitution gives:

This has the form \(J=P(C_1-C_2)\), with effective permeability \(P=D_mK_{m/w}/L\). This expression shows why permeability is not simply a diffusion coefficient: it also reflects membrane partitioning and thickness. Permeability may be measured or estimated directly; the expression in terms of \(D_m\), \(K_{m/w}\), and \(L\) is a simplified physical derivation.

If the two sides have different partitioning into the membrane, the interface concentrations need not be \(K_{m/w}\) times the same factor on both sides. With interface factors \(K_1\) and \(K_2\), the derivation instead gives \(J=(D_m/L)(K_1C_1-K_2C_2)\). In that case, equal bulk concentrations do not necessarily imply zero net flux. More generally, the thermodynamic driving force is a difference in chemical potential; a concentration difference is a suitable linear driving force only when the concentration bases and partition conditions make it so.

Permeability, Flux, and Permeability-Surface-Area Clearance

Permeability \(P\) has units of length per time. For example, if \(J\) is measured in amount per area per time and concentration in amount per volume, then \(P=J/(C_1-C_2)\) has units of volume per area per time, which reduces to length per time. It describes transfer per unit area for a given concentration difference.

Multiplying permeability by area gives \(PA\), often called the permeability-surface-area product or permeability-surface-area clearance. Denote it here by \(CL_{PS}=PA\). Its units are volume per time, and the total net transfer rate is:

This is “clearance-like” because a volume-per-time coefficient multiplies a concentration to produce an amount-per-time rate. It is not automatically systemic drug clearance or elimination: it quantifies exchange across the specified membrane under the stated conditions. Nor should \(P\) and \(CL_{PS}\) be interchanged. The former is an area-normalized membrane property; the latter combines that property with the area available for transfer.

Worked Calculations

When the Relation Is Useful—and When It Can Mislead

The simple relation is useful when a membrane barrier is the rate-limiting resistance and the relevant concentrations are known on both sides. It provides a direct way to relate a measured or estimated permeability to net transfer and clarifies which parameters change flux: increasing \(P\), available area, or the driving concentration difference increases the rate in direct proportion, provided the model remains valid.

The earlier material on perfusion-limited tissue models treats a different approximation: blood delivery is sufficiently rapid that tissue exchange is not limited by passage across a membrane. A permeability-based description is appropriate when membrane crossing itself may limit equilibration. These are alternative limiting descriptions, not interchangeable labels for the same parameter. A real tissue may involve both blood delivery and membrane crossing; deciding which is limiting requires evidence about its concentration profiles and transport conditions.

Several observations should prompt scrutiny of the constant-\(P\) approximation. A measured rate that is not proportional to the concentration difference may indicate saturable transport, concentration-dependent binding, active transport, or changing membrane properties. An apparent permeability that varies with experimental mixing or stirring can signal an unstirred aqueous layer adding resistance. If membrane thickness or composition varies substantially across the surface, a single uniform \(P\) may conceal spatial heterogeneity. These effects do not prove that the membrane relation is useless, but they require a more suitable effective model or a better-defined experimental system.

A further diagnostic is the predicted equilibrium condition. Under the simple same-basis model, the net flux must be zero when \(C_1=C_2\), and its direction must reverse when the concentration difference reverses. If the system instead has unequal equilibrium concentrations because of partitioning, using \(C_1-C_2\) alone gives the wrong driving force; concentrations must be adjusted to a common thermodynamic reference or the partition terms must be represented explicitly.

Common Pitfall

A related practical error is multiplying \(P A\) by a concentration expressed in μmol/L while leaving \(PA\) in cm³/s. The units will not cancel correctly unless the concentration is converted to μmol/cm³ or the volume units are made consistent. A quick dimensional check should yield amount per time for a transfer rate.