A theoretical simulation method for analyzing the influence of cyclodextrin on nanofiltration membrane structure and separation performance
By combining molecular dynamics simulations and quantum chemical calculations, the microscopic effects of cyclodextrin on nanofiltration membranes were analyzed, solving the problem of low selectivity of traditional nanofiltration membranes in separating 1,3-propanediol and 1,2-propanediol, and achieving high-selectivity separation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XIAMEN UNIV
- Filing Date
- 2026-04-16
- Publication Date
- 2026-07-10
AI Technical Summary
Existing technologies struggle to efficiently separate and purify 1,3-propanediol from fermentation broths containing complex polyol impurities. Traditional nanofiltration membranes exhibit low selectivity when separating isomers with similar physicochemical properties, and existing characterization methods are insufficient to resolve the microscopic behavior of cyclodextrins at the water/oil interface.
A nanofiltration membrane system was constructed using a combination of molecular dynamics simulations and quantum chemical wave function calculations. The effects of cyclodextrin on the nanofiltration membrane structure and separation performance were analyzed. By using molecular dynamics simulation trajectory data, radial distribution function, and geometric structure optimization, the hindrance effect of cyclodextrin on monomer diffusion behavior and the geometric matching degree between target molecules and cavities were quantified, and a selective separation mechanism was established.
The microscopic dynamic mechanism of nanofiltration membrane structure was revealed, the sieving mechanism of molecular geometry matching was elucidated, and a theoretical prediction model of microstructure and macroscopic separation performance was established, providing a theoretical basis for the design and optimization of modified nanofiltration membranes. The separation of 1,3-propanediol and 1,2-propanediol with high selectivity was achieved.
Smart Images
Figure CN122369737A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of membrane separation performance research and computational chemistry, and in particular to a theoretical simulation method for analyzing the effects of cyclodextrin on the structure and separation performance of nanofiltration membranes. Background Technology
[0002] With the widespread application of bio-fermentation in the production of 1,3-propanediol, how to efficiently separate and purify 1,3-propanediol from fermentation broth containing complex polyol impurities (such as 1,2-propanediol) is a pressing technical challenge in the field of membrane separation. Traditional commercial nanofiltration membranes often exhibit low selectivity when separating isomers with extremely similar physicochemical properties.
[0003] By introducing cyclodextrin (CD) as a modifier in the aqueous phase or intermediate layer during interfacial polymerization, the sieving performance of nanofiltration membranes can be significantly improved by utilizing its unique rigid conical structure with a hydrophobic inner cavity and a hydrophilic outer edge. However, existing characterization methods are mostly limited to macroscopic performance testing (such as flux and rejection rate) and surface morphology observation (such as SEM and FTIR), making it difficult to deeply analyze the microscopic behavior of cyclodextrin at the water / oil interface at the molecular and atomic scale, and also unable to intuitively quantify the geometric matching and force mechanism between the target molecules (1,3-propanediol and 1,2-propanediol) and the cyclodextrin cavity during mass transfer. This leads to an unclear structure-activity relationship of the modified nanofiltration membrane, limiting the further targeted development of high-performance selective separation membranes. Therefore, it is urgent to establish a theoretical simulation method to systematically elucidate the influence mechanism of cyclodextrin on membrane structure and separation performance from the perspective of microdynamics and molecular spatial configuration. Summary of the Invention
[0004] The purpose of this invention is to solve the above-mentioned problems in the prior art and provide a theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes. This method combines molecular dynamics simulation and quantum chemical wave function calculation to explain the microscopic process of cyclodextrin-regulated interfacial polymerization and its selective sieving mechanism for specific molecular pairs.
[0005] To achieve the above objectives, the present invention adopts the following technical solution:
[0006] A theoretical simulation method for analyzing the effects of cyclodextrin on the structure and separation performance of nanofiltration membranes includes the following steps:
[0007] 1) Construct a two-phase molecular dynamics simulation system including aqueous monomer, oil monomer, cyclodextrin and solvent system, and obtain molecular dynamics simulation trajectory data of monomer diffusion and distribution behavior of the system under interfacial polymerization reaction conditions;
[0008] 2) Based on the trajectory data, extract the number density distribution of monomers along the interface normal direction, and calculate the radial distribution function between cyclodextrin and monomers to quantify the blocking effect of cyclodextrin on monomer diffusion behavior.
[0009] 3) Density functional theory was used to optimize the geometry of the target isolate and cyclodextrin, and the maximum cross-sectional size of the target isolate was calculated based on the electron density isosurface. At the same time, the effective cavity size parameters of cyclodextrin were extracted.
[0010] 4) Analyze the change in crosslinking density of the membrane layer based on the described retardation effect, analyze the geometric matching degree between the target analyte and the cyclodextrin cavity based on the described size parameters, establish the synergistic influence mechanism of the change in crosslinking density and geometric matching degree on the separation selectivity, and evaluate the selective separation capability of the cyclodextrin modified nanofiltration membrane for the target analyte.
[0011] In step 1), the aqueous phase monomer is piperazine (PIP), the oil phase monomer is trimesoyl chloride (TMC), the solvent is n-hexane, and the cyclodextrin is α-cyclodextrin, β-cyclodextrin, or γ-cyclodextrin.
[0012] In step 1), the intermolecular interactions are described using the OPLS-AA all-atomic force field, and molecular dynamics simulations are performed under the NPT ensemble, with the temperature controlled at 290~310 K, the pressure controlled at 0.9~1.1 bar, and the simulation time at 100~200 ns.
[0013] In step 2), the method for extracting the number density distribution is as follows: the simulation system is divided into several thickness layers along the interface normal direction, and the evolution curve of the number of monomer molecules in each layer over time is statistically analyzed.
[0014] In step 2), the calculation of the radial distribution function includes: analyzing the number of hydrogen bonds between the hydroxyl groups on the cyclodextrin surface and the amino groups of the piperazine molecule, so as to quantitatively characterize the retardation effect of cyclodextrin on the piperazine monomer.
[0015] In step 3), the geometric structure is optimized by using the B3LYP functional combined with the 6-311G(d,p) basis set, and the maximum cross-sectional size of the target separation material is calculated based on the electron density isosurface method.
[0016] In step 3), the effective cavity size parameters of the cyclodextrin include the cavity inner diameter and cavity volume, which are calculated using an electron density isosurface analysis tool.
[0017] In step 4), the establishment of the synergistic influence mechanism includes: when the size of the target separated product is smaller than the inner diameter of the cyclodextrin cavity, it is determined that it can form a preferential mass transfer channel through the cyclodextrin cavity; when the size of the target separated product is larger than the inner diameter of the cyclodextrin cavity, it is determined that it is blocked by steric hindrance; at the same time, combined with the decrease in membrane crosslinking density caused by the hindrance effect, the increase in membrane flux is predicted.
[0018] In step 4), the target analytes include 1,3-propanediol (1,3-PDO) and 1,2-propanediol (1,2-PDO); the separation selectivity of the modified nanofiltration membrane for the 1,3-propanediol / 1,2-propanediol mixed system is predicted by comparing the differences in geometric matching between the two isomers and the cyclodextrin cavity.
[0019] Step 4) may also include: establishing the structure-activity relationship between the amount of cyclodextrin added and the crosslinking density of the membrane layer, and between the crosslinking density of the membrane layer and the separation selectivity.
[0020] Compared with the prior art, the beneficial effects achieved by the technical solution of this invention are:
[0021] 1. Revealing the microscopic dynamic mechanism of membrane structure evolution: Through molecular dynamics simulations, this invention can quantitatively assess the hindering effect of cyclodextrin on the diffusion behavior of aqueous monomers through hydrogen bonding networks. This method reflects the dynamic evolution of monomer distribution and reaction rate during interfacial polymerization at the molecular scale, providing a microscopic theoretical explanation for macroscopic phenomena such as reduced crosslinking degree and thinning of membrane layers in modified nanofiltration membranes.
[0022] 2. Elucidating the sieving mechanism based on molecular geometric matching: Utilizing quantum chemistry and visualization calculation methods, this invention constructs a three-dimensional geometric model of the target isomers (such as 1,3-propanediol and 1,2-propanediol) and the cyclodextrin cavity. This method objectively explains, from the perspective of spatial configuration and steric hindrance, the microscopic selective separation process in which linear molecules utilize the cavity as a preferential mass transfer channel for permeation, and branched molecules are repelled and retained due to exceeding the cavity's inner diameter.
[0023] 3. Establishing a theoretical prediction model for microstructure and macroscopic separation performance: This invention integrates microscopic dynamics analysis with molecular spatial configuration calculations to construct a theoretical analysis closed loop encompassing modifier regulation, membrane microstructure, and macroscopic separation performance. Compared to traditional macroscopic characterization methods, this invention provides a method for theoretically predicting the separation mechanism of specific mixture systems at the molecular level, offering a reliable theoretical basis for the design and structural optimization of selective nanofiltration membranes. Attached Figure Description
[0024] Figure 1 The above is a visualization simulation of PIP and TMC using Multiwfn; where (a) is PIP and (b) is TMC.
[0025] Figure 2 These are snapshots of the stability of the MD simulation system for the interface aggregation process; (a) at 20 ns and (b) at 160 ns.
[0026] Figure 3 This is a snapshot of the system stability at the end of the MD simulation of the interface aggregation process.
[0027] Figure 4 Let be the number density of α-CD in the system; where (a) is the simulation at 20 ns and (b) is the simulation at 160 ns.
[0028] Figure 5 Let be the number density of PIP in the system; where (a) is the simulation at 20 ns and (b) is the simulation at 160 ns.
[0029] Figure 6 The above are visualization simulations of PDO molecules using Multiwfn; where (a) is 1,2-PDO and (b) is 1,3-PDO.
[0030] Figure 7 The results are visualization simulations of cyclodextrins using Multiwfn; where (a1) and (a2) are molecular binding cavity diagrams and independent cavity shape diagrams for α-CD, respectively; (b1) and (b2) are molecular binding cavity diagrams and independent cavity shape diagrams for β-CD, respectively; and (c1) and (c2) are molecular binding cavity diagrams and independent cavity shape diagrams for γ-CD, respectively. Detailed Implementation
[0031] To make the technical problems, technical solutions and beneficial effects of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.
[0032] Example 1
[0033] A method for investigating the effects of α-cyclodextrin on the structure and separation performance of polyamide nanofiltration membranes based on theoretical simulation is described below:
[0034] 1. Construction of molecular dynamics simulation of interfacial polymerization process
[0035] Initial Structure Preparation: In a computing workstation, the geometric structures of PIP, TMC, and α-cyclodextrin (α-CD) were optimized using Gaussian software at the B3LYP / 6-311G(d,p) theoretical level, and charge distribution was extracted. The visualization results of the optimized aqueous monomer PIP and oil monomer TMC microstructure and surface electrostatic potential distribution are shown below. Figure 1 As shown in (a) and (b) in the figure.
[0036] Simulation box construction: A simulation box with dimensions of 5 nm × 5 nm × 28 nm was constructed using GROMACS software. The bottom layer was filled with an aqueous phase containing PIP and α-CD, and the top layer was filled with a hexane phase containing TMC. The force field was uniformly constructed using OPLS-AA all-atom confinement.
[0037] Dynamics simulation: Energy minimization was performed on the initial system to eliminate local overlap and stress. Subsequently, a 160 ns dynamics simulation was run at 300 K and 1 bar pressure (NPT ensemble) with a step size of 2 fs. Long-range electrostatic interactions were calculated using the PME method, and trajectory data were collected. Figure 2 As shown, these are stable snapshots of the MD simulation system for the interface polymerization process at 20 ns (a) and 160 ns (b), demonstrating the dynamic evolution of the reaction system over time; further as... Figure 3 As shown, this is a snapshot of the system stability at the endpoint of the MD simulation of the interfacial polymerization process, reflecting the final spatial distribution of TMC, PIP, α-CD, and water molecules in the Z-axis direction after the reaction reaches equilibrium.
[0038] 2. Analysis of Microscopic Diffusion and Membrane Structure Regulation Mechanisms
[0039] Number density analysis: The evolution curve of the number density of PIP monomers along the Z-axis during the period of 0–160 ns was extracted using the built-in analysis module of GROMACS. Combined with... Figure 4 The number density distributions of α-CD in (a) and (b) at 20 ns and 160 ns, and Figure 5 Analysis of the number density evolution curves of PIP at corresponding times in (a) and (b) shows that the diffusion rate of PIP to the interface decreases significantly after the addition of α-CD.
[0040] Interaction analysis: The radial distribution function and the number of hydrogen bonds between PIP and α-CD were calculated. It was confirmed that the abundant hydroxyl groups on the surface of α-CD formed a dense hydrogen bond network with the secondary amine groups of PIP. This network exerts a drag force (retardation effect) on PIP, reducing the concentration of monomers participating in the interfacial polycondensation reaction per unit time. This theoretically confirms the phenomena of reduced crosslinking degree, loose structure, and thinning of the polyamide layer observed in macroscopic experiments.
[0041] 3. Visualization of molecular cavity size and calculation of configuration
[0042] Target molecular size: The optimized molecular models of 1,3-propanediol and 1,2-propanediol were imported into Multiwfn software, such as... Figure 6As shown in the figure. Calculations show that 1,3-propanediol exhibits a linear distribution with a relatively small maximum cross-section; while 1,2-propanediol, due to the presence of side-chain methyl groups, has a significantly increased maximum cross-sectional size.
[0043] α-CD Cavity Simulation: Import α-CD crystal coordinates, call the Promolecular Electron Density (POD) module in Multiwfn, set the electron density threshold to define the cavity boundary, accurately calculate the effective inner diameter and cavity volume of the α-CD, and output a 3D rendered isosurface map, such as... Figure 7 As shown in (a1) and (a2), this result provides an intuitive three-dimensional geometric basis for further investigation into the cavity matching mechanism.
[0044] 4. Establishment of structure-activity relationship and explanation of separation mechanism
[0045] A comparative analysis was performed based on the three-dimensional dimensional data obtained in step 3. It was found that the molecular size of 1,3-propanediol is highly compatible with the inner diameter of the α-CD. During permeation, the α-CD inner cavity forms a low-resistance "preferred hydration mass transfer channel," allowing 1,3-propanediol to rapidly penetrate the membrane. Conversely, the branched structure of the 1,2-propanediol molecule leads to strong steric repulsion (steric hindrance) with the α-CD inner cavity, effectively blocking it from the outside of the channel. Combined with the loose membrane network formed in step 2, macroscopically, this nanofiltration membrane not only exhibits high water flux but also displays an extremely high selective separation coefficient for 1,3-propanediol. This simulation completely closes the loop of structure-property verification from material structure to separation performance.
[0046] The advantage of this invention lies in its ability to accurately reconstruct the microscopic dynamics of cyclodextrin-regulated interfacial polymerization while successfully quantifying the three-dimensional dimensional fit between the target molecule and the cyclodextrin cavity by introducing a theoretical simulation method combining molecular dynamics and quantum chemical wave functions. Based on this simulation method, the following conclusions can be drawn: linear 1,3-propanediol molecules can highly adapt to the inner diameter of α-cyclodextrin and achieve rapid mass transfer; while 1,2-propanediol with a branched structure is trapped due to strong steric hindrance because its maximum cross-sectional size exceeds the inner diameter of the cavity. These simulation results systematically reveal the fundamental mechanism by which modified nanofiltration membranes achieve highly selective separation of 1,3-propanediol at the microscopic level.
[0047] The above description is only a preferred embodiment of the present invention. For those skilled in the art, several improvements and refinements can be made without departing from the technical principles of the present invention, and these improvements and refinements should also be considered within the scope of protection of the present invention.
Claims
1. A theoretical simulation method for analyzing the effects of cyclodextrin on the structure and separation performance of nanofiltration membranes, characterized in that, Includes the following steps: 1) Construct a two-phase molecular dynamics simulation system containing aqueous monomers, oil monomers, cyclodextrins and solvents, and obtain molecular dynamics simulation trajectory data of monomer diffusion and distribution behavior under interfacial polymerization reaction conditions; 2) Based on the trajectory data, extract the number density distribution of monomers along the interface normal direction, and calculate the radial distribution function between cyclodextrin and monomers to quantify the blocking effect of cyclodextrin on monomer diffusion behavior. 3) Density functional theory was used to optimize the geometry of the target isolate and cyclodextrin, and the maximum cross-sectional size of the target isolate was calculated based on the electron density isosurface. At the same time, the effective cavity size parameters of cyclodextrin were extracted. 4) Analyze the change in crosslinking density of the membrane layer based on the described retardation effect, analyze the geometric matching degree between the target analyte and the cyclodextrin cavity based on the described size parameters, establish the synergistic influence mechanism of the change in crosslinking density and geometric matching degree on the separation selectivity, and evaluate the selective separation capability of the cyclodextrin modified nanofiltration membrane for the target analyte.
2. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that: In step 1), the aqueous phase monomer is piperazine, the oil phase monomer is pyromellitic trimethylol chloride, the solvent is n-hexane, and the cyclodextrin is α-cyclodextrin, β-cyclodextrin, or γ-cyclodextrin.
3. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that: In step 1), the intermolecular interactions are described using the OPLS-AA all-atomic force field, and molecular dynamics simulations are performed under the NPT ensemble, with the temperature controlled at 290~310 K, the pressure controlled at 0.9~1.1 bar, and the simulation time at 100~200 ns.
4. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that, In step 2), the method for extracting the number density distribution is as follows: the simulation system is divided into several thickness layers along the interface normal direction, and the evolution curve of the number of monomer molecules in each layer over time is statistically analyzed.
5. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that, In step 2), the calculation of the radial distribution function includes: analyzing the number of hydrogen bonds between the hydroxyl groups on the cyclodextrin surface and the amino groups of the piperazine molecule, so as to quantitatively characterize the retardation effect of cyclodextrin on the piperazine monomer.
6. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that: In step 3), the geometric structure is optimized by using the B3LYP functional combined with the 6-311G(d,p) basis set, and the maximum cross-sectional size of the target separation material is calculated based on the electron density isosurface method.
7. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that: In step 3), the effective cavity size parameters of the cyclodextrin include the cavity inner diameter and cavity volume, which are calculated using an electron density isosurface analysis tool.
8. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that, In step 4), the establishment of the synergistic influence mechanism includes: when the size of the target separated product is smaller than the inner diameter of the cyclodextrin cavity, it is determined that it can form a preferential mass transfer channel through the cyclodextrin cavity; when the size of the target separated product is larger than the inner diameter of the cyclodextrin cavity, it is determined that it is blocked by steric hindrance; at the same time, combined with the decrease in membrane crosslinking density caused by the aforementioned hindrance effect, the increase in membrane flux is predicted.
9. The theoretical simulation method for analyzing the influence of cyclodextrin on the structure and separation performance of nanofiltration membranes as described in claim 1, characterized in that: In step 4), the target analytes include 1,3-propanediol and 1,2-propanediol; by comparing the differences in geometric matching between the two isomers and the cyclodextrin cavity, the separation selectivity of the modified nanofiltration membrane for the 1,3-propanediol / 1,2-propanediol mixed system is predicted.