Understanding trade-offs in adsorption capacity, selectivity and kinetics for propylene/propane separation using composites of activated carbon and hypercrosslinked polymer

2021 
Abstract The separation of propane and propylene represents a key step of polypropylene synthesis. While this separation is currently performed via distillation, adsorption appears as a more sustainable alternative. Adsorption requires the use of adsorbents exhibiting fast kinetics, optimised uptake and selectivity for the process conditions. Yet, individual adsorbents for propylene/propane separation often exhibit trade-offs of uptake vs selectivity and kinetics vs selectivity. In this context, composite adsorbents may provide an answer to the challenge. To test this hypothesis, the properties of composites must be investigated, to identify whether they exhibit synergistic, additive or antagonist properties with respect to their individual components. Herein, we report a series of silver-exchanged hypercrosslinked polymer (HCP)/activated carbon (AC) composites for propylene/propane separation. The adsorbents a priori combine the high capacity and fast kinetics of AC and the high selectivity of silver-exchanged HCP. We synthesised and characterised the composites using analytical, spectroscopic and microscopy techniques. We then tested the materials using equilibrium and dynamic gas sorption measurements. The results showed that, at low pressure, the composites surpassed the trade-off between capacity and selectivity compared to the physical mixture of AC and HCP. At higher pressure, this synergistic behaviour was diminished and we only observed an additive behaviour. In any case, the composites retained the fast gas diffusion properties of AC. The findings point towards the possible use of such composites to tackle the challenge in propylene/propane separation for industrial application.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    46
    References
    1
    Citations
    NaN
    KQI
    []