Solvent-to-Feed Ratio Optimization

Reference ID: MET-682F | Process Engineering Reference Sheets Calculation Guide

Introduction & Context

Solvent-to-feed ratio optimization is a critical procedure in Process Engineering, specifically within the design and operation of continuous countercurrent leaching systems. This process is widely utilized in the vegetable oil industry (e.g., palm kernel or soybean extraction) to separate valuable lipids from inert solid matrices using a solvent, typically hexane.

The objective is to determine the optimal solvent flow rate that balances high oil recovery targets against the escalating costs of solvent recovery and capital expenditure for extraction stages. Proper optimization ensures that the miscella concentration remains within manageable viscosity limits while minimizing the loss of oil in the spent solid underflow.

Methodology & Formulas

The calculation follows a mass balance approach across the leaching cascade. The system defines the feed rate F (total mass flow of solid feed), the inert solid mass flow rate D, and the oil mass fraction X_F. For a binary feed consisting only of oil and inert solids (dry basis), the inert solids flow rate is:

\[ D = F \cdot (1 - X_{F}) \]

The underflow ratio U represents the mass of solution retained per unit mass of inert solids. The mass flow of solution retained in the underflow is calculated as:

\[ R_{u} = U \cdot D \]

The target oil loss in the spent solids, based on the recovery efficiency η, is defined as:

\[ O_{\text{loss}} = (1 - \eta) \cdot (F \cdot X_{F}) \]

The concentration of oil in the underflow solution x_U is the ratio of lost oil to the total solution retained:

\[ x_{U} = \frac{O_{\text{loss}}}{R_{u}} \]

The solvent flow rate S is determined by the chosen solvent-to-feed ratio R:

\[ S = R \cdot F \]

The overflow solution flow V (the miscella) is derived from the overall mass balance around the leaching system. The total mass entering (feed + fresh solvent) must equal the total mass leaving (overflow solution + inert solids + underflow solution):

\[ F + S = V + D + R_{u} \]

Rearranging and substituting F = D + F·X_F for a binary feed yields the correct overflow solution flow rate:

\[ V = S + F \cdot X_{F} - U \cdot D \]

Operational Constraint: The overflow flow rate must be positive for feasible operation, requiring:

\[ S + F \cdot X_{F} > U \cdot D \quad \text{or equivalently} \quad R > \frac{U \cdot D}{F} - X_{F} \]

The resulting miscella concentration y_m (mass fraction of oil in the overflow solution) is calculated as:

\[ y_{m} = \frac{F \cdot X_{F} - O_{\text{loss}}}{V} = \frac{\eta \cdot F \cdot X_{F}}{S + F \cdot X_{F} - U \cdot D} \]

The economic feasibility is assessed using the concentration ratio, which compares the miscella concentration to the underflow concentration:

\[ \text{ConcRatio} = \frac{y_{m}}{x_{U}} = \frac{\eta \cdot U \cdot D}{(1 - \eta) \cdot (S + F \cdot X_{F} - U \cdot D)} \]

Parameter	Constraint/Regime	Threshold/Limit
Underflow Ratio (U)	Empirical Bounds	0.3 ≤ U ≤ 1.0 kg solution/kg inert
Miscella Concentration (y_m)	Viscosity/Boiling Point Limit	y_m ≤ 0.50 (mass fraction)
Solvent-to-Feed Ratio (R)	Operational Range	0.3 ≤ R ≤ 2.0 (subject to V > 0 constraint)
Recovery Target (η)	Commercial Viability	η ≥ 0.95

The number of ideal countercurrent stages N required to achieve the target recovery can be estimated using the Kremser equation for leaching with constant underflow and an equilibrium constant of unity (perfect mixing in each stage). The extraction factor is defined as:

\[ E = \frac{S}{U \cdot D} \]

For E ≠ 1, the number of ideal stages is:

\[ N = \frac{\ln\!\left(\frac{E - \eta}{1 - \eta}\right)}{\ln(E)} - 1 \]

For the special case E = 1, the number of stages is N = η/(1-η). In practice, actual stages are scaled by an efficiency factor (typically 0.6–0.8 for leaching).

The total annual cost is derived from the sum of operating costs (solvent and energy) and annualized capital costs (number of stages N):

\[ C_{\text{total}} = (S \cdot C_{\text{solvent}} + V \cdot C_{\text{energy}}) \cdot H_{\text{annual}} + (N \cdot C_{\text{stage}}) \]

Where H_annual is the annual operating hours (typically 8000 h/yr for continuous operation), C_solvent is the solvent cost per unit mass, C_energy is the energy cost per unit mass of miscella processed, and C_stage is the annualized capital cost per ideal stage.

To identify the optimal solvent-to-feed ratio, process engineers should follow these systematic steps using the mass balance methodology described above:

Map the feasible region: Calculate the minimum solvent ratio R_min = (U·D/F) − X_F required to maintain positive overflow (V > 0). This establishes the lower bound for R.
Perform a sensitivity analysis: Vary R across the operational range (from R_min to 2.0) and compute the corresponding miscella concentration y_m, concentration ratio, and required number of stages N using the Kremser equation.
Evaluate the trade-off: Plot total annual cost C_total versus R. Increasing R reduces the required number of stages (lower capital cost) but increases solvent and energy operating costs due to the larger miscella volume V.
Identify the economic optimum: The optimal R occurs at the minimum of the total cost curve, typically where the marginal savings in capital cost equal the marginal increase in operating cost.
Validate constraints: Ensure the selected R yields y_m ≤ 0.50 and η ≥ 0.95. Confirm through pilot-scale testing to account for non-ideal mixing and stage efficiencies.

Operating near or below the minimum feasible solvent-to-feed ratio (R approaching R_min) introduces several critical process risks:

Overflow collapse: When S + F·X_F ≤ U·D, the calculated overflow V becomes zero or negative — all solvent is entrained in the underflow, halting miscella production entirely.
Excessive miscella concentration: As V decreases, y_m = η·F·X_F / V increases sharply. Concentrations exceeding 0.50 cause prohibitive viscosity, reduced mass transfer rates, and potential boiling point elevation issues in the solvent recovery column.
Insufficient stage count: Low R reduces the extraction factor E = S/(U·D), requiring exponentially more stages to maintain the target recovery η. If existing equipment has insufficient stages, oil loss O_loss will exceed the design specification.
Phase inversion risk: Reduced solvent volume in the contactor may lead to emulsion formation or phase inversion, particularly if the feed contains surface-active impurities.
Fouling and precipitation: Localized high oil concentrations near the feed inlet can cause precipitation of phospholipids or waxes, leading to fouling of extraction internals.

The solvent-to-feed ratio directly governs the miscella flow rate V = S + F·X_F − U·D, which is the primary driver of utility costs in the solvent recovery unit. Key energy impacts include:

Reboiler duty: The distillation column must vaporize the solvent content of V. The solvent mass in the miscella is V − η·F·X_F = S − U·D + (η·F·X_F terms). At higher R, V increases nearly linearly, raising reboiler steam consumption proportionally.
Condenser cooling load: The latent heat of the recovered solvent must be rejected in the overhead condenser. Higher V increases cooling water or air-fin cooler demand.
Pumping energy: The solvent recycle loop handles a flow rate of approximately S (fresh solvent) plus internal recycle. Larger R increases pumping costs and pipe sizing requirements.
Optimum trade-off: While a higher R increases energy costs through the V·C_energy term, it reduces the required number of stages N (lower capital cost). The total cost curve C_total typically exhibits a distinct minimum, which represents the economic optimum for the facility.

Worked Example: Soybean Oil Extraction

A continuous countercurrent leaching system processes soybean meal to extract crude oil using hexane. Determine the miscella concentration, underflow concentration, and verify that all operational constraints are satisfied.

Input Parameters

Feed rate, F	2000 kg/h
Oil mass fraction, X_F	0.18 (18 wt%)
Inert solids flow, D = F·(1 − X_F)	2000 × 0.82 = 1640 kg/h
Underflow ratio, U	0.60 kg solution/kg inert
Recovery target, η	0.97 (97%)
Solvent-to-feed ratio, R	0.80
Annual operating hours, H_annual	8000 h/yr
Solvent cost, C_solvent	0.15 $/kg
Energy cost, C_energy	0.05 $/kg miscella
Annualized stage cost, C_stage	50,000 $/yr per ideal stage

Step-by-Step Calculation

Step 1 — Oil in Feed:

\[ F \cdot X_{F} = 2000 \times 0.18 = 360 \text{ kg/h} \]

Step 2 — Solution Retained in Underflow:

\[ R_{u} = U \cdot D = 0.60 \times 1640 = 984 \text{ kg/h} \]

Step 3 — Oil Loss in Spent Solids:

\[ O_{\text{loss}} = (1 - \eta) \cdot F \cdot X_{F} = (1 - 0.97) \times 360 = 10.8 \text{ kg/h} \]

Step 4 — Underflow Oil Concentration:

\[ x_{U} = \frac{O_{\text{loss}}}{R_{u}} = \frac{10.8}{984} = 0.01098 \approx 0.0110 \text{ (1.10 wt%)} \]

Step 5 — Solvent Flow Rate:

\[ S = R \cdot F = 0.80 \times 2000 = 1600 \text{ kg/h} \]

Step 6 — Overflow (Miscella) Flow Rate (corrected mass balance):

\[ V = S + F \cdot X_{F} - U \cdot D = 1600 + 360 - 984 = 976 \text{ kg/h} \]

Check: Total mass in = F + S = 2000 + 1600 = 3600 kg/h. Total mass out = V + D + R_u = 976 + 1640 + 984 = 3600 kg/h. ✓

Operational constraint: V = 976 kg/h > 0. The minimum solvent ratio is R_min = (U·D/F) − X_F = (984/2000) − 0.18 = 0.312. Since R = 0.80 > 0.312, operation is feasible. ✓

Step 7 — Miscella Concentration:

\[ y_{m} = \frac{F \cdot X_{F} - O_{\text{loss}}}{V} = \frac{360 - 10.8}{976} = \frac{349.2}{976} = 0.3578 \approx 0.358 \text{ (35.8 wt%)} \]

Step 8 — Concentration Ratio:

\[ \text{ConcRatio} = \frac{y_{m}}{x_{U}} = \frac{0.3578}{0.01098} = 32.6 \]

Step 9 — Number of Ideal Stages (Kremser Equation):

\[ E = \frac{S}{U \cdot D} = \frac{1600}{984} = 1.626 \]

\[ N = \frac{\ln\!\left(\frac{E - \eta}{1 - \eta}\right)}{\ln(E)} - 1 = \frac{\ln\!\left(\frac{1.626 - 0.97}{0.03}\right)}{\ln(1.626)} - 1 = \frac{\ln(21.87)}{0.4861} - 1 = \frac{3.085}{0.4861} - 1 = 5.35 \]

Rounding up, N ≈ 6 ideal stages. With a stage efficiency of 0.70, approximately 9 actual stages are required.

Step 10 — Annual Cost Estimate:

\[ C_{\text{total}} = (1600 \times 0.15 + 976 \times 0.05) \times 8000 + (6 \times 50{,}000) \]

\[ C_{\text{total}} = (240 + 48.8) \times 8000 + 300{,}000 = 288.8 \times 8000 + 300{,}000 \]

\[ C_{\text{total}} = 2{,}310{,}400 + 300{,}000 = 2{,}610{,}400 \text{ \$/yr} \]

Constraint Verification

Constraint	Value	Limit	Status
U	0.60	0.3–1.0 kg/kg	✓ Pass
y_m	0.358	≤ 0.50	✓ Pass
R	0.80	0.3–2.0	✓ Pass
η	0.97	≥ 0.95	✓ Pass
V > 0	976 kg/h	> 0	✓ Pass

Engineering Note: The miscella concentration of 35.8 wt% is within the viscosity limit. If optimization demands a lower solvent ratio, engineers must verify that y_m does not exceed 0.50 and that the minimum feasible ratio R_min = 0.312 is respected.

"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