Why Sink Conditions and Hydrodynamics Matter
A dissolution test measures how quickly drug leaves a dosage form and enters the surrounding medium. The measured rate is not determined by the formulation alone: it also depends on how close the medium is to saturation and on how fluid motion carries dissolved drug away from the dissolving surface. These factors matter when interpreting a profile. A slow measured rate may reflect formulation-controlled release, limited solubility, inadequate medium volume, weak mass transfer, or a combination of them.
The earlier material on formulation attributes that control drug release describes how dosage-form properties can affect release. Here the focus is on the surrounding conditions. Sink conditions limit the buildup of dissolved drug relative to its saturation solubility; hydrodynamics govern transport near the dissolving surface. Neither condition by itself guarantees that an in vitro rate represents release in vivo.
Deriving a Volume-Based Sink Criterion
Suppose a dose \(D\) is placed in a well-mixed medium of constant volume \(V\), and assume the entire dose could dissolve. Its resulting concentration would be \(D/V\). Under the one-third criterion, the volume must be large enough that this concentration does not exceed one-third of saturation solubility:
The quantity \(D/C_s\) is the volume that would contain the dose at saturation. The sink criterion therefore calls for at least three times that volume. For a more stringent target \(C_b \leq 0.1C_s\), the corresponding requirement is \(V \geq 10D/C_s\). This is a capacity calculation: it indicates whether the medium can accommodate the dissolved dose without approaching saturation. It does not, on its own, establish that a particular test remains below the target at every sampling time.
The calculation assumes that \(C_s\) is known for the relevant medium, temperature, and chemical conditions; that volume is effectively constant; and that the stated dose is the amount capable of dissolving into the medium. If the medium is replaced or sampled without replacement, or if solubility changes during the test, the simple volume calculation must be adjusted. A formulation may also release only part of the dose during the observation period, but a capacity calculation based on the full dose is a conservative check.
Worked Example: Minimum Volume for the One-Third Criterion
Hypothetical test. A tablet contains 100 mg of drug. The measured saturation solubility in the test medium is 0.50 mg/mL. Find the minimum volume for the one-third criterion and assess a proposed 900 mL test volume.
Using \(V \geq 3D/C_s\), with the dose in mg and solubility in mg/mL:
At 900 mL, the maximum concentration if all 100 mg dissolves is \(100/900 = 0.1111\) mg/mL. Relative to saturation solubility, this is \(0.1111/0.50 = 0.2222\), or 22.2% of \(C_s\). Since 22.2% is below one-third, the 900 mL volume meets the stated criterion, assuming the specified solubility and an unchanged volume. Under the stricter 10% criterion, the required volume would be \(10(100)/0.50 = 2000\) mL, so 900 mL would not meet that target.
How Hydrodynamics Enter the Dissolution Rate
A useful starting point is the diffusion-layer model, also called the Noyes–Whitney form. It treats drug transport from the solid surface through a thin fluid layer into the bulk medium. Let \(M\) be the mass dissolved, \(A\) the exposed surface area, \(D_m\) the molecular diffusion coefficient in the medium, and \(h\) the effective thickness of the diffusion layer. If the concentration at the solid surface is \(C_s\), the bulk concentration is \(C_b\), and transport through the layer is approximately steady and one-dimensional, Fick’s law gives the mass transfer rate:
This model assumes that the surface concentration is at saturation, the bulk is sufficiently mixed to have a representative \(C_b\), the exposed area and layer properties are defined for the period considered, and diffusion across the layer is the controlling transport process. Real dosage forms can violate these assumptions: area can change as particles erode or disintegrate, surface concentration may not reach \(C_s\), and convection or chemical reactions may also affect transport. The equation is therefore a mechanistic guide, not a universal description of every dosage form.
The concentration difference \(C_s-C_b\) is the driving force. Under a strong sink, \(C_b\) is small and the driving force is close to \(C_s\). As drug accumulates in the medium, the gradient and hence the predicted rate decline. Agitation can alter the effective layer thickness \(h\): stronger fluid motion commonly reduces the thickness or increases surface renewal, increasing mass transfer. The effect is not necessarily proportional to agitation speed, and it can also change the dosage form’s exposed area, wetting, or mechanical erosion.
Worked Example: Effect of Boundary-Layer Thickness
Hypothetical calculation. A dissolving surface has area \(A=2.0\ \mathrm{cm^2}\), the diffusion coefficient is \(D_m=5.0\times10^{-6}\ \mathrm{cm^2/s}\), and \(C_s=0.50\ \mathrm{mg/mL}=0.50\ \mathrm{mg/cm^3}\). First take \(h=0.010\ \mathrm{cm}\) and negligible bulk concentration. Then consider agitation that reduces the effective \(h\) to \(0.0050\ \mathrm{cm}\).
At negligible \(C_b\), the equation gives:
Halving \(h\) doubles the rate in this model:
If \(C_b=0.050\ \mathrm{mg/mL}\) instead, the gradient is \(0.50-0.050=0.45\ \mathrm{mg/cm^3}\), which is 90% of the sink-case gradient. At \(h=0.010\ \mathrm{cm}\), the predicted rate is therefore \(0.030(0.90)=0.027\ \mathrm{mg/min}\). These rates are model calculations under the stated assumptions, not measured values or claims about a real product.
Volume, Accumulation, and the Meaning of an Observed Rate
Volume can affect the time course even when the initial dissolution rate is unchanged. To see why, suppose the dissolved mass is \(M=VC_b\), the medium is well mixed, volume is constant, and the solid remains available in excess. If \(D_mA/h\) is constant, the rate equation becomes:
Dividing by \(V\) gives a first-order approach to saturation. With initial concentration \(C_b(0)=0\), integration yields:
This derivation shows two distinct effects. At \(t=0\), the mass dissolution rate is \((D_mA/h)C_s\), independent of \(V\) in this simplified model. Later, a smaller volume accumulates concentration more quickly, so its concentration gradient falls sooner. The equation does not account for exhaustion of the solid dose; it applies only while enough undissolved drug remains to support the assumed surface condition.
Worked Example: Comparing Two Medium Volumes
Hypothetical calculation. A 100 mg dose is tested in well-mixed media with \(C_s=0.50\ \mathrm{mg/mL}\). Let \(D_mA/h=30\ \mathrm{mL/min}\), and compare 300 mL with 900 mL after 5 minutes. Assume the dose remains in excess, volume is constant, and the model’s transport coefficient does not change.
For 900 mL, the exponent is \(-(30/900)(5)=-0.1667\). Thus:
For 300 mL, the exponent is \(-(30/300)(5)=-0.5000\):
The smaller volume has reached 39.3% of \(C_s\), compared with 15.4% for the larger volume. Both concentrations are below one-third in this particular calculation, but the smaller-volume result is closer to saturation and has a weaker instantaneous driving gradient. The initial rates are equal under the model; the later amounts differ because concentration accumulates more rapidly in the smaller volume.
Assumptions, Diagnostics, and Failure Modes
A sink calculation is only as reliable as its inputs. Solubility should correspond to the test medium and conditions, rather than being borrowed from a different pH or temperature. If \(C_s\) changes during the test, for example because the medium’s composition changes, a single fixed value may misclassify the sink status. The bulk concentration should also be assessed over time when possible; meeting a capacity criterion does not ensure that a local region near a dissolving surface remains dilute.
Hydrodynamic conditions also need to be sufficiently controlled for results to be interpretable. If agitation varies between runs or units, differences in boundary-layer transport can be mistaken for formulation differences. An unexpectedly strong dependence of the profile on agitation may indicate mass-transfer limitation, sensitivity to mixing, or a change in dosage-form behavior such as erosion. A profile that slows as bulk concentration rises may be consistent with loss of sink, but a similar shape can have other causes; concentration measurements and formulation observations help distinguish them.
The diffusion-layer interpretation becomes unreliable when its assumptions fail. Examples include a changing effective surface area, substantial local concentration gradients despite bulk mixing, a dosage form that swells or erodes, or an apparent rate controlled by another process. Replicate profiles, concentration measurements, and observations of the dosage form can reveal instability or unexplained variability. These diagnostics flag questions for investigation; they do not identify a unique mechanism by themselves.
Comparing Approaches and Interpreting In Vivo Relevance
| Approach | What it helps establish | Main limitation |
|---|---|---|
| Capacity calculation, \(V \geq 3D/C_s\) | Whether the chosen volume can accommodate the dose under a stated sink convention | Does not prove the measured bulk concentration stays below the target at all times |
| Time-varying concentration assessment | Whether accumulation during the test weakens the driving gradient | Requires appropriate concentration and volume information |
| Diffusion-layer model | How solubility, concentration gradient, area, and mass transfer can affect instantaneous rate | Relies on simplifying assumptions about the surface, mixing, and transport |
| Agitation sensitivity assessment | Whether the observed profile depends strongly on hydrodynamic conditions | Changing agitation may alter more than boundary-layer thickness |
A sink test is useful for separating a formulation’s release behavior from a pronounced reduction in concentration gradient as the medium approaches saturation. But a strongly sink-maintained test may not reproduce the fluid volumes, changing conditions, or absorptive processes present in the gastrointestinal tract. In vivo, dissolution and absorption can occur together, and removal of dissolved drug from the lumen can help maintain a gradient. Conversely, local fluid conditions and motility can differ from a well-mixed laboratory vessel. Therefore, an in vitro sink result supports a statement about the specified test conditions; it is not, by itself, a prediction of in vivo release or absorption.
A common mistake is to reason, “The volume is larger than three times the saturation volume, so every measured rate is an intrinsic property of the formulation and directly predicts in vivo release.” The correction is that the volume criterion addresses bulk capacity under a stated solubility assumption. Measured rates can still depend on hydrodynamics, surface changes, and other test conditions, and in vivo relevance requires evidence beyond the sink calculation.
Check Your Understanding
Answer each question, then compare with the solutions below.
- A hypothetical 80 mg dose has \(C_s=0.40\ \mathrm{mg/mL}\). What minimum volume meets the one-third criterion?
- In the diffusion-layer model, what happens to the instantaneous rate if \(C_b\) increases while all other terms stay fixed?
- Why can two test volumes have the same initial rate but different amounts dissolved after several minutes?
- Name two assumptions of the diffusion-layer model that may fail for a swelling or eroding dosage form.
Solutions. (1) \(V_{\min}=3(80\ \mathrm{mg})/(0.40\ \mathrm{mg/mL})=600\ \mathrm{mL}\). (2) The gradient \(C_s-C_b\) decreases, so the predicted instantaneous rate decreases. (3) The initial bulk concentration is the same in the simplified model, so the initial gradient and rate are the same; subsequently, the smaller volume accumulates concentration more rapidly and loses driving force sooner. (4) Exposed area may not remain constant, and the concentration at the surface may not equal a fixed saturation concentration; changing layer properties or incomplete bulk mixing are additional possible failures.