Arte SheetsLearn. Try. Understand.

Food Dehydration & Pharmaceutical Stability, GAB Moisture Sorption Isotherm

GAB Food Moisture Sorption Isotherm (M0, C, K) Formula

Calculate Effective Multilayer Activity K*aw, GAB Equilibrium Moisture Content (% dry basis), Uncorrected BET Moisture Prediction (%), and High-aw BET Divergence Error (%).

Why classic 2-parameter BET breaks down above aw = 0.45 in dehydrated foods—and how GAB’s third parameter K extends accuracy to aw = 0.90

Freeze-dried foods, milk powders, pet kibble, and pharmaceutical tablets store shelf-life stability on their Moisture Sorption Isotherm relating Dry-Basis Equilibrium Moisture M (% db, g water / 100 g dry solids) to Water Activity a_w (0 < a_w < 1). Classic 2-parameter BET theory (M0 monolayer moisture, C Guggenheim energy constant) assumes multilayer water molecules above the monolayer have the exact same heat of liquefaction as pure bulk water—blowing up hyperbolically toward infinity above a_w ≈ 0.45. The Guggenheim-Anderson-de Boer (GAB) Isotherm (adopted by European COST 90 & AOAC food engineering standards) introduces a third thermodynamic parameter K (0.70 <= K < 1.00) accounting for structured multilayer sorption enthalpy, scaling a_w to **K * a_w: MGAB = (M_0 * C * K * a_w) / ((1 - K * a_w) * (1 - K * a_w + C * K * a_w))** (% db).

Verify on Spray-Dried Skim Milk Powder at 25 °C—Storage_Water_Activity a_w = 0.65 (B2), GAB_Monolayer_Moisture M_0 = 5.80% db (B3), Guggenheim_Energy_Constant C = 14.5 (B4), and Multilayer_Factor K = 0.84 (B5): Scaled Activity K * a_w equals 0.546 (E2), True GAB Equilibrium Moisture equals 12.08% db (E3), whereas Uncorrected BET (K = 1.0) predicts 15.98% db (E4)—a +3.90% db overprediction (E5).

GAB vs BET Sorption Isotherm KPI (M0=5.8% db, C=14.5, K=0.84, aw=0.65 vs 0.82)FormulaVerified Output
Effective Multilayer Activity K*awMultilayer_K * Water_Activity_aw0.546 (0.689 @ humid aw=0.82)
GAB Equilibrium Moisture (% db)(M0 * C * K * aw) / ((1-K*aw)*(1-K*aw+C*K*aw))12.08% db (18.07% db @ humid aw=0.82)
Uncorrected BET Moisture (% db)(M0 * C * aw) / ((1 - aw) * (1 - aw + C * aw))15.98% db (31.74% db blow-up @ aw=0.82)
BET Overprediction Error (% db)Uncorrected_BET_Pct - GAB_Moisture_Pct+3.90% db (+13.67% db error @ aw=0.82)

3-Parameter GAB Multilayer Sorption Isotherm vs 2-Parameter BET Limit Formula

Validate Water_Activity a_w in (0.02, 0.95], Monolayer M_0 > 0.2%, Guggenheim C > 0.5, and Multilayer K in [0.50, 1.00].

=LET(aw_f, B2, mz_p, B3, cg_k, B4, km_f, B5, ka_w, km_f * aw_f, mg_p, (mz_p * cg_k * ka_w) / ((1 - ka_w) * (1 - ka_w + cg_k * ka_w)), mb_p, (mz_p * cg_k * aw_f) / ((1 - aw_f) * (1 - aw_f + cg_k * aw_f)), IF(OR(aw_f <= 0.02, aw_f > 0.95, mz_p <= 0.2, cg_k <= 0.5, km_f < 0.5, km_f > 1), "Invalid Isotherm Input", ROUND(mb_p - mg_p, 2)))

Tropical Packaging Permeation (aw = 0.82) BET Asymptote Divergence Check (Set Water Activity B2 = 0.82)

Expose the powder pouch to 82% Equilibrium Relative Humidity (a_w B2 = 0.82): GAB Equilibrium Moisture E3 rises realistically to 18.07% db (K * a_w = 0.689 in E2), while 1 / (1 - a_w) drives Uncorrected BET E4 to an unphysical 31.74% db (+13.67% db error in E5).

=IF(ISTEXT(E5), E5, LET(ka_w, B5 * B2, ROUND((B3 * B4 * ka_w) / ((1 - ka_w) * (1 - ka_w + B4 * ka_w)), 2)))

Try the example

Use the original values or change them to test your understanding.

Further reading