Periodic Reporting for period 1 - NODRY (Designing green and energy-efficient anti-dewetting protocols for liquid chromatography)
Periodo di rendicontazione: 2022-05-01 al 2023-10-31
Reverse Phase High Performance Liquid Chromatography (RP-HPLC) is a leading separation technique which relies on hydrophobic stationary phases. When highly aqueous solvents are used, dewetting may occur in the nanopores, leading to retention losses (see Fig. 1). This prevents the use of RP-HPLC for separating highly polar compounds, calling for the vast adoption of dangerous organic solvents. NODRY set up a joint simulation and experimental approach based on intrusion/extrusion protocols to test the wetting properties of RP-HPLC columns, with the aim of assessing materials and protocols to avoid dewetting. In particular, NODRY has explored the role of dissolved gasses on dewetting. Overall, these advancements may lead to more green and efficient RP-HPLC columns and protocols. The most important outcome of the project is the development of a simulation-experimental platform that allows rapid and low-cost development of new materials, conditions and protocols for HPLC. In particular, proposed experiments rely on an intrusion-extrusion cycle that, in combination with simulations, allows the optimization of HPLC columns and protocols using small quantities of stationary phase (50 mg per test) and solvent and rapid compression-decompression cycles instead of full-scale HPLC experiments.
Experiments:
The experiments were realized using commercial WC8 purchased from WATERS, consisting of a mesoporous silica gel grafted with the octylsilanes C8 with a density of 2.1 groups/nm^2 [Grosu et al., ACS Appl. Mater. Interfaces, 2017]. In addition, deionized water was used as non-wetting liquid and several gasses were introduced in the system.
Water intrusion-extrusion experiments were performed with an Auto Pore IV 9500 porosimeter (Micromeritics Instrument Corporation, Norcross, USA). From this experiment a PV-isotherm is obtained (Fig. 2) in which a step of volume reduction due to intrusion of water into the pores is evident at the intrusion pressure (Pint) and another drastic change in slope is observed after a pressure reduction, corresponding to the extrusion of the water from the pores (Pext). In RP-HPLC, Pext is a measure of the conditions under which dewetting is expected to occur. Ideally, materials and protocols used in RP-HPLC columns should result in no extrusion (Pext
As an example, we considered the case of gases dissolved in the solvent. Since the extrusion process of water is related with nucleation of bubbles [Tinti et al., PNAS, 2017], the presence of dissolved gasses is expected to affect this process and affect the extrusion pressure [Camisasca et al., JPCL 2021]. In RP-HPLC terms, this means that gasses might facilitate dewetting. Five configurations were used: one with outgassed system and four with different gases: Synthetic air, Ar, He, and Ethylene. For all gases, Pint doesn’t differ significantly between all the experiments. However, in the case of Pext there are considerable differences between all the experiments. The lowest Pext was registered for the outgassed experiment. Then, for the gassed samples, the maximum Pext is registered for Ethylene, being more than 1 MPa higher than the outgassed one. The highest effect of Ethylene is expected due to its highest hydrophobicity as compared to the other three gasses. Thus, we have confirmed that the presence of dissolved gas with high hydrophobicity have an impact on the water intrusion/extrusion performance.
Simulations:
We aimed to create a sufficiently detailed computational model of a grafted nanopore, capturing the salient features of RP-HPLC stationary phases and yet with a limited computational cost to enable the simulation of experimentally relevant time and length scales. With this aim, the coarse-grained (CG) force-field (MARTINI 3 [Souza, P. CT, et al. Nature methods (2021)]) was used which reduces the degrees of freedom while maintaining sufficient chemical and spatial resolution. Our model is based on MCM-41 functionalized with hydrophobic silane grafting [Lefevre, B., et al. J. Chem. Phys. (2004)], which has a reproducible cylindrical shape. This choice allows for considerable geometric simplification and for useful comparison with theory, while capturing important general properties of silica gels of interest for RP-HPLC. We simulated three chain lengths (C8, C12, and C18) and several grafting densities ranging from 0.6 to 2.5 molecules/nm^2 (See Fig. 3 for simulation schematics). Water intrusion/extrusion cycles were performed in-silico. First, we could characterize the local geometric and wetting characteristics of the pores. Secondly, we could related such features with the intrusion and extrusion properties, revealing that local heterogeneities play a crucial role which cannot be explained by the average properties. These findings are useful in RP-HPLC as 1) they suggest that controlling the quality of the surface functionalization is key and 2) provide the computational tools to screen new materials and procedures.
As a demonstration of this platform, we tested intrusion and extrusion properties of the system in the presence of dissolved gasses. Chloroform and Ethylene gases at a concentration of 0.006 molecules/nm^3 were introduced to assess their impact. The presence of gases slightly affected intrusion and heavily affected extrusion, being more pronounced for the C18 grafted pore. The trend is in accordance with the experimental results, where the effects are higher for Ethylene, and particularly in extrusion.
The experiments were realized using commercial WC8 purchased from WATERS, consisting of a mesoporous silica gel grafted with the octylsilanes C8 with a density of 2.1 groups/nm^2 [Grosu et al., ACS Appl. Mater. Interfaces, 2017]. In addition, deionized water was used as non-wetting liquid and several gasses were introduced in the system.
Water intrusion-extrusion experiments were performed with an Auto Pore IV 9500 porosimeter (Micromeritics Instrument Corporation, Norcross, USA). From this experiment a PV-isotherm is obtained (Fig. 2) in which a step of volume reduction due to intrusion of water into the pores is evident at the intrusion pressure (Pint) and another drastic change in slope is observed after a pressure reduction, corresponding to the extrusion of the water from the pores (Pext). In RP-HPLC, Pext is a measure of the conditions under which dewetting is expected to occur. Ideally, materials and protocols used in RP-HPLC columns should result in no extrusion (Pext
As an example, we considered the case of gases dissolved in the solvent. Since the extrusion process of water is related with nucleation of bubbles [Tinti et al., PNAS, 2017], the presence of dissolved gasses is expected to affect this process and affect the extrusion pressure [Camisasca et al., JPCL 2021]. In RP-HPLC terms, this means that gasses might facilitate dewetting. Five configurations were used: one with outgassed system and four with different gases: Synthetic air, Ar, He, and Ethylene. For all gases, Pint doesn’t differ significantly between all the experiments. However, in the case of Pext there are considerable differences between all the experiments. The lowest Pext was registered for the outgassed experiment. Then, for the gassed samples, the maximum Pext is registered for Ethylene, being more than 1 MPa higher than the outgassed one. The highest effect of Ethylene is expected due to its highest hydrophobicity as compared to the other three gasses. Thus, we have confirmed that the presence of dissolved gas with high hydrophobicity have an impact on the water intrusion/extrusion performance.
Simulations:
We aimed to create a sufficiently detailed computational model of a grafted nanopore, capturing the salient features of RP-HPLC stationary phases and yet with a limited computational cost to enable the simulation of experimentally relevant time and length scales. With this aim, the coarse-grained (CG) force-field (MARTINI 3 [Souza, P. CT, et al. Nature methods (2021)]) was used which reduces the degrees of freedom while maintaining sufficient chemical and spatial resolution. Our model is based on MCM-41 functionalized with hydrophobic silane grafting [Lefevre, B., et al. J. Chem. Phys. (2004)], which has a reproducible cylindrical shape. This choice allows for considerable geometric simplification and for useful comparison with theory, while capturing important general properties of silica gels of interest for RP-HPLC. We simulated three chain lengths (C8, C12, and C18) and several grafting densities ranging from 0.6 to 2.5 molecules/nm^2 (See Fig. 3 for simulation schematics). Water intrusion/extrusion cycles were performed in-silico. First, we could characterize the local geometric and wetting characteristics of the pores. Secondly, we could related such features with the intrusion and extrusion properties, revealing that local heterogeneities play a crucial role which cannot be explained by the average properties. These findings are useful in RP-HPLC as 1) they suggest that controlling the quality of the surface functionalization is key and 2) provide the computational tools to screen new materials and procedures.
As a demonstration of this platform, we tested intrusion and extrusion properties of the system in the presence of dissolved gasses. Chloroform and Ethylene gases at a concentration of 0.006 molecules/nm^3 were introduced to assess their impact. The presence of gases slightly affected intrusion and heavily affected extrusion, being more pronounced for the C18 grafted pore. The trend is in accordance with the experimental results, where the effects are higher for Ethylene, and particularly in extrusion.
NODRY demonstrated that simulation and experiments of intrusion and extrusion in hydrophobic nanoporous materials can provide important information for RP-HPLC. In particular, they provide an informative and inexpensive means of testing new materials, surface functionalizations, and operative conditions (temperature and gas concentration) for RP-HPLC columns, while characterizing the phenomenon of dewetting. As a specific case, we assessed how the presence of specific gases in water may influence dewetting, suggesting that a thorough degassing could be beneficial for RP-HPLC with highly aqueous solvents. Further developing this joint simulation and experimental platform could lead to a valuable design and test tool for HPLC industry and HPLC users alike.