.. AIM Documentation documentation master file, created by sphinx-quickstart on Fri May 16 14:38:34 2025. You can adapt this file completely to your liking, but it should at least contain the root `toctree` directive. MixPred =============================== **MixPred** is the module for mixture isotherm prediction for the given pressure and composition. **MixPred** GUI is shown below: .. image:: images/MixPred.png :width: 1000 :alt: MixPred :align: center ------- The isotherm fitting results from **IsoFit** and **HeatFit** can be directly loaded into **MixPred** module, and the user can specify the desired pressure range and mixture composition to calculate the mixture loadings. Watch how to use **MixPred** :ref:`here ` **MixPred** has two models for mixture adsorption loading prediction which are as follows: * Extended Dual-site Langmuir Model (EDSL) * Ideal Adsorbed Solution Theory (IAST) Extended Dual-site Langmuir Model (EDSL) --------------------------------------------- The extended dual-site Langmuir (EDSL) model is a generalization of the Langmuir model for multi-component mixture adsorption. The model is expressed as, .. math:: q_{mix,i}^{*} = \frac{q_{sat,1,i}b_{1,i}P_{i}}{1+\sum_{1}^{N}b_{1,j}P_{j}} + \frac{q_{sat,2,i}b_{2,i}P_{i}}{1+\sum_{1}^{N}b_{2,j}P_{j}} where :math:`N` is the number of components involved; :math:`q_{sat,1,i}`, :math:`q_{sat,2,i}`, :math:`b_{1,i}`, :math:`b_{2,i}`, and :math:`P_{i}` are the single or pure component Langmuir isotherm parameters and partial pressure of component :math:`i`, respectively. The pure component Langmuir isotherm parameters for individual components can be obtained by fitting Langmuir isotherm model to the pure component isotherm data for the given component using either **IsoFit** or **HeatFit**. Note that the EDSL model is thermodynamically consistent only when the saturation capacities for each component are equal. .. math:: q_{sat, 1, 1}=q_{sat, 1, 2}=q_{sat, 1, 3}= ... q_{sat, 1, N} \\ q_{sat, 2, 1}=q_{sat, 2, 2}=q_{sat, 2, 3}= ... q_{sat, 2, N} .. note:: The EDSL model in **MixPred** is applicable only when the isotherm of each individual component is described using the Langmuir isotherm model. Ideal Adsorbed Solution Theory (IAST) --------------------------------------------- IAST is a thermodynamic framework to calculate the mixture isotherms using pure component isotherm. The IAST is based on three fundamental assumptions: * The surface area of the adsorbent is equally accessible to all adsorbates. * The adsorbed phase is an ideal solution. * The adsorbent is homogeneous. The solution of the IAST involves solving non-linear equations consisting of a rediced grand potential. For the component :math:`i`, the reduced grand potential :math:`\psi_{i}^{*}` is given as, .. math:: \psi_{i}^{*} = \int_{0}^{P_{i}^{*}} \frac{q_{i}^{*}(P)}{P} \,dP where :math:`P_{i}^{*}` is the fictitious pressure and :math:`q_{i}^{*}` is the equilibrium loading of component :math:`i` as obtained from pure component isotherm model. The fictitious pressure :math:`P_{i}^{*}` is the pressure for component :math:`i` at which it exerts the same reduced grand potential as the other components in the mixture. :math:`P_{i}^{*}` is related to partial pressure :math:`P_{i}` of component :math:`i` given as, .. math:: P_{i} = x_{i}\,P_{i}^{*} \qquad \mathrm{for}\, i=1,2,3,\ldots, N. where :math:`N` is the number of adsorbing components in the mixture and :math:`x_{i}` is the adsorbed mole fraction of component :math:`i`. Note that, .. math:: \sum_{i}\,x_{i} = 1 IAST states that adsorption equilibrium is achieved when the reduced grand potential of each component is same, .. math:: \psi_{1}^{*} = \psi_{2}^{*} = \psi_{3}^{*} = ... \psi_{N}^{*} In **MixPred** the IAST equations above are solved for :math:`2N` unknows, :math:`P_{i}^{*}` and :math:`x_{i}`. The mixture adsorption loading for each component are then calculated using .. math:: q_{tot} = \frac{1}{\sum_{i}^{N}\,\left(\frac{x_{i}}{q_{i}^{*}(P_{i}^{*})}\right)} \\ q_{mix, i}^{*} = x_{i}\,q_{tot} .. note:: The IAST equations are highly non-linear and often extremely sensitive to the initial guess used. The solution can often be computationally expensive and time-consuming.