measuring and modeling of mixed adsorption isotherms for
play

Measuring and Modeling of Mixed Adsorption Isotherms for - PowerPoint PPT Presentation

Measuring and Modeling of Mixed Adsorption Isotherms for Supercritical Fluid Chromatography Overview Objectives Experimental Modeling Empirical Thermodynamic Conclusions Lazo, Giese, Lbbert Objectives Model adsorption data measured


  1. Measuring and Modeling of Mixed Adsorption Isotherms for Supercritical Fluid Chromatography Overview Objectives Experimental Modeling Empirical Thermodynamic Conclusions Lazo, Giese, Lübbert

  2. Objectives • Model adsorption data measured at supercritical conditions in a systematic way • Gain a better understanding of adsorption under supercritical conditions • Highlight particular characteristics and problems of adsorption from supercritical fluids Lazo, Giese, Lübbert 6/21/00 2

  3. Experiment Description I System Components Experimental Set-up Mobile Phase SCF CO 2 Modifier Isopropanol Feed Phytol Fixed Phase Silica Gel Phytol Molecule C 20 H 40 O 1 Gas Supply 7 Oven 2 High Pressure Pump 8 Mixing Loop 3 Pressure Control Unit 9 Analytical Column 4 Manometer 10 Detector CH 2 OH 5 Modifier 11 PC 6 Feed 12 Chromatograms Lazo, Giese, Lübbert 6/21/00 3

  4. Experiment Description II Experimental Conditions Elution Experiments Characteristic Band Profile for a Sigmoidal Isotherm T = 313.15 K 4 Dimensionless Concentration Modifier 3.5 P [bar] [mL/min] 3 120 0.153 SHARP FRONT DIFFUSE REAR 150 0.153 2.5 210 0.153 Single 2 Isotherms 240 0.153 1.5 SHARP REAR 210 0.100 DIFFUSE FRONT 1 210 0.237 0.5 120 0.153 Binary 210 0.153 0 Isotherms 0 1 2 3 4 5 6 210 0.237 Dimensionless Time ! Isotherm with point of inflection Lazo, Giese, Lübbert 6/21/00 4

  5. Perturbation Method ! The Perturbation Method is based on Equilibrium Theory 2 Perturbation Desorption 1.5         − − − − ε ε ε ε _ UV Signal   1 dq       = = = = + + + + i t ( C ) t 1         R , i o ε ε ε ε dC _         i C 1 Adsorption Analysis 0.5 t Trans t Cis 0 0 5 10 15 20 25 30 Dimensionless Time Lazo, Giese, Lübbert 6/21/00 5

  6. Binary Quadratic Isotherm trans-phytol data c i s - phy t ol da t a quadratic mixture isotherm qua dr a t i c mi x t ur e i s ot her m 4.5 5.6 4.4 Time [min] 5.4 4.3 5.2 4.2 P IN = 210 bar P IN = 210 bar IPA FLOW =0.153 mL/min IPA FLOW = 0.153 mL/min 5.0 4.1 0 1 2 3 4 5 6 7 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Concentration [mg/mL] Concentration [mg/mL] + + + + 2 + + + + q b C b C C 2 b C Binary Quadratic Isotherms: Θ Θ Θ Θ = = = = 1 = = = = 1 1 3 1 2 4 1 1 + + + + + + + + + + + + + + + + 2 + + + + 2 q 1 b C b C b C C b C b C s 1 1 2 2 3 1 2 4 1 5 2 ! Five Parameters were fitted + + + + + + + + 2 b C b C C 2 b C q Θ Θ = = = = Θ Θ = = 2 = = 2 2 3 1 2 5 2 ! 21 Parameters in total 2 + + + + + + + + + + + + + + + + 2 + + + + 2 q 1 b C b C b C C b C b C s 1 1 2 2 3 1 2 4 1 5 2 Lazo, Giese, Lübbert 6/21/00 6

  7. Thermodynamic Model Equation of State Literature, Experiments Critical Constants ! PVT Interaction Parameters ! VLE ! Solubility Gibbs Thermodynamic SCF Solute Modifier Isotherm Model Literature, Experiments ! Gravimetric Adsorption Data ! Volumetric ! Chromatographic Adsorbed Phase Model Lazo, Giese, Lübbert 6/21/00 7

  8. Vapor Pressure of Phytol Temperature [K] 300 350 400 450 500 550 600 650 700 750 800 1E7 1000 Critical Point Tc=664.04 K Pressure [Pa] Pc=8.7685 bar 0.1 Acentric Factor w =2.24590 w w w 1E-5 Operating Pv PRSV EOS k k 1 =2.46767 k k Conditions 1E-9 Pv Experimental Points T=313.15 K Pv 1-Eicosanol P=8.5e-10 Pa 1E-13 Lazo, Giese, Lübbert 6/21/00 8

  9. Phytol Solubility T = 313.15 K 0 10 Molar Fraction of Phytol x 2 [-] Decrease in solubility -5 10 ( ( ) ) ( ( ) )         P T V P 1 = = = = v m x exp         2 ∞ ∞ ∞ ∞ φ φ         φ φ P RT 2 -10 10 Phytol Sol. PRSV EOS x IPA = 0 Phytol Sol. PRSV EOS x IPA = 0.0314 Phytol Experimental Solubility Experimental range, upper limit -15 10 0 100 200 300 400 500 600 Pressure [bar] ! The experimental data are inside the theoretical solubility region Lazo, Giese, Lübbert 6/21/00 9

  10. Single Adsorption Isotherms Loadings versus Fugacities Loadings versus Concentrations 45 120 350 350 150 40 210 300 300 cis-phytol loading [mg/mL] 240 35 210- 250 250 210+ 30 200 200 25 150 150 20 100 100 dq/dc [-] 15 50 50 10 0 0 0.0 0.2 0.4 0.6 0.8 0 2 4 6 8 10 C [mg/mL] Fugacity [nPa] Lazo, Giese, Lübbert 6/21/00 10

  11. Single Adsorption Data Fitting 450 trans-phytol loading [mg/mL] 400 ! 10 parameters were fitted T = 313.15 K 350 300 ! Expansion till third virial 250 coefficient 200 ! Loadings of CO 2 and IPA 150 are assumed to be proportional to their 100 fugacities o Pseudo-experimental data 50 — Virial EOS model 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 trans-phytol fugacity [nPa]         ∑ ∑ ∑ ∑ ∑ ∑ ∑∑ ∑ ∑ ∑ ∑ ( ( ( ( ) ) ) ) f 2 3         i = = = = − − − − + + + + + + + + + + + + ! ln ln K A n B n n C         Virial Isotherm: i j ij j k ijk 2     n     A 2 A i j j k Lazo, Giese, Lübbert 6/21/00 11

  12. 12 Single Adsorption Isotherm at T = 313.15 K Lazo, Giese, Lübbert 6/21/00

  13. Binary Adsortion Data Fitting trans-phytol mixture isotherm at T = 313.15 K 450 ! Single Parameters remain 400 o Data trans-phytol loading [mg/mL] — Correlation 350 ! 5 additional Parameters — Prediction — Single ads. 300 ! 25 Parameters in total 250 200 150 100 50 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 trans-phytol fugacity [nPa] Lazo, Giese, Lübbert 6/21/00 13

  14. Linear Equilibrium Constants C IPA = 0.1 mL/min C IPA = 0.153 mL/min C IPA = 0.237 mL/min Experimental Points 120 120 110 110 Linear Equilibrium Constant K [-] 100 100 trans-phytol cis-phytol 90 90 80 80 70 70 C IPA increase C IPA increase 60 60 50 50 40 40 30 30 20 20 10 10 0 0 120 160 200 240 280 120 160 200 240 280 Pressure [bar] Pressure [bar] Lazo, Giese, Lübbert 6/21/00 14

  15. Conclusions • Enhanced solubility of phytol at higher fluid density and IPA concentration. • There is competition for the adsorbent actives sites among CO 2 , IPA, and phytol isomers. • Decreased desorption tendency at very high pressures: repulsive forces and adsorbent saturation. • The model can correlate the data very well but has poor predictive capabilities. The adsorbed phase description should be improved Lazo, Giese, Lübbert 6/21/00 15

Recommend


More recommend