Supersaturation Control via Cooling Profile

Reference ID: MET-854F | Process Engineering Reference Sheets Calculation Guide

Introduction & Context

In industrial crystallization, maintaining a constant level of supersaturation is critical for controlling crystal size distribution, preventing secondary nucleation, and influencing polymorphic outcomes. The Cooling Profile for Constant Supersaturation is a process engineering technique used to design a temperature‑time trajectory that keeps the relative supersaturation at a fixed, optimal value throughout a batch process, thereby supporting effective polymorph control in pharmaceutical crystallization.

By precisely controlling the cooling rate, engineers can ensure that the rate of solute deposition onto existing seed crystals matches the rate of solubility reduction, thereby avoiding the metastable zone limit where uncontrolled nucleation occurs. Understanding the metastable zone width determination helps define the safe operating region for consistent crystal growth.

Methodology & Formulas

The calculation relies on a mass balance between the solute removed from the liquid phase and the mass deposited onto the crystal surface. The following steps define the iterative physics used to derive the cooling profile.

The number of crystals N is assumed constant; this assumption can be verified through agglomeration diagnosis in crystallizers. The initial surface area A₀ is calculated based on the seed mass M_cry,0, crystal density ρ_c, shape factor k_v, and characteristic length L₀:

\[ N = \frac{M_{\text{cry},0}}{\rho_{c} \cdot k_{v} \cdot L_{0}^{3}} \] \[ A = N \cdot k_{a} \cdot L_{0}^{2} \]

2. Crystal Growth Kinetics
The mass deposition rate \(\dot{m}_{c}\) is governed by the growth rate constant k_g, the current surface area A, and the target relative supersaturation \(\sigma_{0}\) raised to the growth exponent g:

\[ \dot{m}_{c} = k_{g} \cdot A \cdot \sigma_{0}^{g} \]

3. Surface Area Scaling
As mass is deposited, the surface area increases. Assuming the number of crystals remains constant and the shape factor is preserved, the area scales with the crystal mass M_cry:

\[ A = \left( \frac{A_{0}}{M_{\text{cry},0}^{2/3}} \right) \cdot M_{\text{cry}}^{2/3} \]

4. Cooling Rate Derivation – To maintain constant supersaturation, the cooling rate must compensate for the mass removed from the solution. Given a linear solubility curve defined by the slope b, the required temperature change rate is calculated as follows. For an alternative strategy based on solvent removal, refer to controlling supersaturation through evaporation rate.

\[ \frac{dT}{dt} = - \frac{\dot{m}_{c}}{V \cdot b \cdot (1 + \sigma_{0})} \]

Parameter	Constraint/Regime
Relative Supersaturation (\(\sigma_{0}\))	Must be ≤ 0.2 to remain within the metastable zone.
Temperature Gradient	T_final must be strictly less than T_initial.
Time Step (\(\Delta t\))	Must be a positive value to ensure convergence.
Growth Exponent (g)	Typically 1.0 for diffusion-controlled or surface-integration-controlled growth.

To maintain the metastable zone width (MSZW) during crystallization, you must balance the cooling rate with the crystal growth kinetics. Follow these steps:

Calculate the solubility curve for your specific solute-solvent system.
Implement a non-linear cooling profile, such as a cubic cooling curve, to account for the increasing surface area of the crystal population.
Monitor the supersaturation level using in-situ process analytical technology (PAT) like focused beam reflectance measurement (FBRM).
Adjust the cooling rate to ensure the supersaturation remains below the primary nucleation threshold.

Exceeding the metastable limit triggers primary nucleation, which negatively impacts your process in the following ways:

Uncontrolled secondary nucleation leads to a broad crystal size distribution (CSD).
Excessive fines generation complicates downstream filtration and drying operations.
Rapid precipitation often results in the entrapment of mother liquor, leading to lower product purity.
Increased risk of oiling out if the system enters the unstable region of the phase diagram.

Introducing seed crystals significantly alters the supersaturation dynamics by providing surface area for growth. Key considerations include:

Seeding allows for a faster initial cooling rate because the surface area of the seeds consumes supersaturation effectively.
The seed loading (mass percentage) dictates the maximum allowable supersaturation before secondary nucleation occurs.
Proper seed conditioning is required to prevent dissolution upon addition, which would otherwise shift the system back into the undersaturated region.

Worked Example: Cooling Profile for Constant Supersaturation

Scenario: A batch cooling crystallizer with a volume of \(V = 1.0\) m³ is seeded with \(1.0\) kg of crystals. The objective is to design a temperature trajectory \(T(t)\) that maintains a constant relative supersaturation \(\sigma_0 = 0.10\) throughout the batch, ensuring uniform crystal growth without secondary nucleation.

Knowns

Solution Volume: \(V = 1.0\) m³
Initial Temperature: \(T_0 = 60.0\) °C
Final Target Temperature: \(T_{\text{final}} = 20.0\) °C
Target Relative Supersaturation: \(\sigma_0 = 0.10\)
Solubility Slope: \(b = 0.5\) kg/m³/°C
Seed Mass: \(M_{\text{cry},0} = 1.0\) kg
Seed Characteristic Length: \(L_0 = 0.0001\) m
Crystal Density: \(\rho_c = 1500.0\) kg/m³
Volume Shape Factor: \(k_v = 1.0\)
Area Shape Factor: \(k_a = 6.0\)
Growth Rate Constant: \(k_g = 0.001\) kg/m²/s
Growth Exponent: \(g = 1.0\)
Time Step: \(\Delta t = 10.0\) s

Step-by-Step Calculation

Initialize Crystal Population:
The number of seeds is calculated from the seed mass and volume: \[ N_0 = \frac{M_{\text{cry},0}}{\rho_c \cdot k_v \cdot L_0^3} = 666666666.667 \] The initial surface area available for growth is: \[ A_0 = N_0 \cdot k_a \cdot L_0^2 = 40.0 \, \text{m}^2 \] Because the number of crystals remains constant, the area scales with mass to the two-thirds power. The area coefficient for this scaling law is: \[ A_{\text{coeff}} = \frac{A_0}{M_{\text{cry},0}^{2/3}} = 40.0 \]
Compute Instantaneous Growth Rate:
The mass deposition rate depends on the current surface area and the target supersaturation: \[ \dot{m}_c = k_g \cdot A \cdot \sigma_0^g \] At the final step in the calculation, the surface area has grown significantly: \[ \dot{m}_c = 0.001 \cdot 325.518 \cdot 0.10 = 0.032 \, \text{kg/s} \]
Update Crystal Mass and Surface Area:
The mass deposited in a single time step is \(\Delta M = \dot{m}_c \cdot \Delta t\): \[ \Delta M = 0.032 \cdot 10.0 = 0.322 \, \text{kg} \] The total crystal mass accumulates to a final value of: \[ M_{\text{cry}} = 23.215 \, \text{kg} \] The surface area is updated using the geometric scaling law: \[ A = A_{\text{coeff}} \cdot M_{\text{cry}}^{2/3} = 40.0 \cdot (23.215)^{2/3} = 325.518 \, \text{m}^2 \]
Determine Required Cooling Rate:
To keep the supersaturation constant, the cooling must exactly compensate for the solute removed by crystal growth. The required cooling rate is derived from the mass balance: \[ \frac{dT}{dt} = -\frac{\dot{m}_c}{V \cdot b \cdot (1 + \sigma_0)} \] Substituting the values from the final time step: \[ \frac{dT}{dt} = -\frac{0.032}{1.0 \cdot 0.5 \cdot 1.10} = -0.059 \, ^\circ\text{C/s} \] This corresponds to a temperature drop of \(\Delta T = -0.586 \, ^\circ\text{C}\) over the final 10-second interval.
Iterate Over Batch Time:
Steps 2 through 4 are repeated numerically. The iterative integration of the cooling profile yields a batch time of \(t = 1400.0\) s, at which point the temperature has dropped to \(T = 19.609\) °C, crossing the lower bound threshold.

Final Answer

The designed cooling profile brings the crystallizer from \(60.0\) °C to a final temperature of \(19.609\) °C over a batch time of \(1400.0\) seconds. The final crystal mass is \(23.215\) kg, and the total crystal surface area is \(325.518\) m².

"Un projet n'est jamais trop grand s'il est bien conçu." — André Citroën

"La difficulté attire l'homme de caractère, car c'est en l'étreignant qu'il se réalise." — Charles de Gaulle