FumCALF-20 for Efficient PVSA Biogas Upgrading

• FumCALF-20 exhibits a high CO2 working capacity (3.62 mol/kg) and low CH4 uptake (0.27 mol/kg), enabling >90% CH4 purity and recovery in PVSA.
• The study integrates multiscale methods including Monte Carlo simulations, dual-site Langmuir fitting, and process-level optimization to link molecular behavior with performance.
• Its optimized pore structure (PLD 3.4 Å, LCD 5.0 Å) and moderate adsorption enthalpy balance adsorption kinetics and energy efficiency for biogas upgrading.

FumCALF-20 is a member of the CALF-20 isoreticular series evaluated as an adsorbent for biogas upgrading by pressure vacuum swing adsorption (PVSA). In a multiscale assessment integrating structural characterization, atomistic grand canonical Monte Carlo simulations, dual-site Langmuir isotherm fitting, and process-level optimization, it was identified as the only material among CALF-20 and five isoreticular derivatives that could reach methane purity greater than 0.90 while maintaining methane recovery of at least 0.90 under the modeled PVSA conditions (Shin et al., 20 Jul 2025). Within that study, its reported combination of high CO$_2$ uptake, low CH$_4$ affinity, and moderate adsorption enthalpy distinguished it from the other members of the series for energy-efficient biogas upgrading.

1. Position within the CALF-20 isoreticular series

Cyclic swing adsorption processes, particularly PVSA, are described as a promising technology for upgrading biogas by separating carbon dioxide from methane. The same study emphasizes that rational design of adsorbent materials with tailored properties is important for deployment of high-performance PVSA technology. Metal-organic frameworks, and especially the CALF-20 isoreticular series, have attracted interest because of high CO$_2$ selectivity and thermal and water stability (Shin et al., 20 Jul 2025).

Within this context, FumCALF-20 was assessed alongside CALF-20 and five derivatives in a workflow that connected molecular-scale adsorption calculations to process-scale cycle optimization. The central result was not merely that FumCALF-20 adsorbs CO$_2$ effectively, but that it remains the only member of the six-material set to satisfy the specified simultaneous process targets for methane purity and methane recovery. A plausible implication is that, in this series, favorable equilibrium properties are necessary but not sufficient; process-level regenerability and methane slip are equally decisive.

2. Structural descriptors and equilibrium adsorption behavior

The reported pore characteristics of FumCALF-20 are a pore volume of $0.52\ \mathrm{cm^3\,g^{-1}}$, a pore limiting diameter of $3.4\ \text{\AA}$, and a largest cavity diameter of $5.0\ \text{\AA}$. Its isosteric heats of adsorption, obtained from Widom insertion in RASPA, are $-27.7\ \mathrm{kJ\,mol^{-1}}$ for CO$_2$ and $-18.9\ \mathrm{kJ\,mol^{-1}}$ for CH$_4$0 (Shin et al., 20 Jul 2025).

Single-component equilibrium loading was represented with a dual-site Langmuir model,

$_4$1

with

$_4$2

For FumCALF-20, the fitted parameters are as follows.

Quantity	CO$_4$3	CH$_4$4
$_4$5	$_4$6	$_4$7
$_4$8	$_4$9	$_2$0
$_2$1	$_2$2	$_2$3
$_2$4	$_2$5	$_2$6
$_2$7	$_2$8	$_2$9
$_2$0	$_2$1	$_2$2

Under the specified condition of $_2$3 and $_2$4, the reported CO$_2$5 working capacity is $_2$6, the CH$_2$7 working capacity is $_2$8, and the CO$_2$9/CH$0.52\ \mathrm{cm^3\,g^{-1}}$0 selectivity $0.52\ \mathrm{cm^3\,g^{-1}}$1 is approximately $0.52\ \mathrm{cm^3\,g^{-1}}$2 at both adsorption and desorption pressures. These values are the immediate equilibrium basis for the material’s process-level performance.

3. Molecular simulation framework and relation to CALF-20 theory

The molecular simulations for FumCALF-20 employed DREIDING for the framework and TraPPE for CO$0.52\ \mathrm{cm^3\,g^{-1}}$3 and CH$0.52\ \mathrm{cm^3\,g^{-1}}$4, with Coulomb plus $0.52\ \mathrm{cm^3\,g^{-1}}$5–$0.52\ \mathrm{cm^3\,g^{-1}}$6 Lennard-Jones interactions, a cutoff of $0.52\ \mathrm{cm^3\,g^{-1}}$7, and tail corrections. Partial charges on the framework were assigned with PACMAN DDEC06. Periodic boundary conditions were used, and simulation boxes were replicated until the minimum dimension exceeded $0.52\ \mathrm{cm^3\,g^{-1}}$8 the cutoff, namely $0.52\ \mathrm{cm^3\,g^{-1}}$9. CO$3.4\ \text{\AA}$0 was treated as a rigid three-site molecule and CH$3.4\ \text{\AA}$1 as a united-atom single site. Translation, rotation, and insertion/deletion moves were assigned equal probabilities. Temperatures were $3.4\ \text{\AA}$2, $3.4\ \text{\AA}$3, and $3.4\ \text{\AA}$4, pressures ranged from $3.4\ \text{\AA}$5 to $3.4\ \text{\AA}$6, equilibration and production each used $3.4\ \text{\AA}$7 cycles, and $3.4\ \text{\AA}$8 was obtained by Widom insertion over $3.4\ \text{\AA}$9 cycles in RASPA 2.0 with a rigid framework. Binary isotherms were predicted with an extended DSL model rather than direct mixture GCMC (Shin et al., 20 Jul 2025).

A separate study on CALF-20, rather than FumCALF-20, developed a statistical-mechanical multi-site Langmuir model derived from partition-function arguments and coupled it to transition-state-theory-based diffusion estimates. In that work, the CALF-20 CO$5.0\ \text{\AA}$0 isotherm was represented with two inequivalent sites, and the framework reproduced isotherms from $5.0\ \text{\AA}$1 to $5.0\ \text{\AA}$2, Henry’s constants, saturation loadings, and self-diffusion trends with reported quantitative agreement to molecular simulation benchmarks (Gonçalves et al., 10 Jul 2025). This suggests a broader methodological context for the CALF-20 family: equilibrium and transport can be analyzed within a common thermodynamic-kinetic formalism, even though the FumCALF-20 PVSA study itself relied on DSL/EDSL equilibrium models and process optimization rather than an explicit diffusion theory of that type.

4. PVSA cycle configuration and optimization problem

Process evaluation was carried out with a five-step modified Skarstrom PVSA cycle consisting of: pressurization from $5.0\ \text{\AA}$3 up to $5.0\ \text{\AA}$4, adsorption at $5.0\ \text{\AA}$5 with CH$5.0\ \text{\AA}$6-rich product at the top, heavy reflux using CH$5.0\ \text{\AA}$7-rich recycle to boost purity, counter-current depressurization to $5.0\ \text{\AA}$8 with CO$5.0\ \text{\AA}$9 desorption, and light reflux with a CH$-27.7\ \mathrm{kJ\,mol^{-1}}$0-rich purge. The governing assumptions were ideal gas behavior, a one-dimensional non-isothermal plug-flow model, linear driving force mass transfer, Ergun pressure drop, and no radial gradients. The dynamic model was discretized by finite volume methods with WENO reconstruction and solved in MATLAB with ode15s (Shin et al., 20 Jul 2025).

The optimization variables were the adsorption pressure $-27.7\ \mathrm{kJ\,mol^{-1}}$1 from $-27.7\ \mathrm{kJ\,mol^{-1}}$2 to $-27.7\ \mathrm{kJ\,mol^{-1}}$3, desorption pressure $-27.7\ \mathrm{kJ\,mol^{-1}}$4 from $-27.7\ \mathrm{kJ\,mol^{-1}}$5 to $-27.7\ \mathrm{kJ\,mol^{-1}}$6, adsorption time $-27.7\ \mathrm{kJ\,mol^{-1}}$7 from $-27.7\ \mathrm{kJ\,mol^{-1}}$8 to $-27.7\ \mathrm{kJ\,mol^{-1}}$9, superficial velocity $_2$0 from $_2$1 to $_2$2, and light and heavy reflux ratios $_2$3 and $_2$4. Multi-objective optimization used Thompson Sampling Efficient Multi-objective Optimization (TSEMO). The process objectives were to maximize CH$_2$5 purity and CH$_2$6 recovery subject to recovery at least $_2$7, while the economic objectives were to maximize productivity and minimize energy requirement subject to purity at least $_2$8 and recovery at least $_2$9.

The reported performance metrics were

$-18.9\ \mathrm{kJ\,mol^{-1}}$0

$-18.9\ \mathrm{kJ\,mol^{-1}}$1

$-18.9\ \mathrm{kJ\,mol^{-1}}$2

and

$-18.9\ \mathrm{kJ\,mol^{-1}}$3

reported as $-18.9\ \mathrm{kJ\,mol^{-1}}$4 per ton CH$-18.9\ \mathrm{kJ\,mol^{-1}}$5.

5. Process-level performance and comparison with other derivatives

On the purity-recovery Pareto front, FumCALF-20 is the only material reported to reach CH$-18.9\ \mathrm{kJ\,mol^{-1}}$6 purity at least $-18.9\ \mathrm{kJ\,mol^{-1}}$7 while maintaining recovery at least $-18.9\ \mathrm{kJ\,mol^{-1}}$8. All other CALF-20 derivatives top out below approximately $-18.9\ \mathrm{kJ\,mol^{-1}}$9 purity under the same recovery constraint. The optimal operating window for FumCALF-20 was reported as $_4$00–$_4$01, $_4$02–$_4$03, $_4$04–$_4$05, $_4$06–$_4$07, $_4$08–$_4$09, and $_4$10–$_4$11 (Shin et al., 20 Jul 2025).

Under the purity constraint of at least $_4$12, the economic Pareto front places FumCALF-20 at an energy consumption of approximately $_4$13–$_4$14 and a productivity of approximately $_4$15–$_4$16. The study’s summary describes these as competitive values in the modeled setting.

The comparative analysis is especially important because it shows that equilibrium CO$_4$17 working capacity alone does not determine process success. TtdcCALF-20 has a similar CO$_4$18 working capacity of $_4$19, but its CH$_4$20 working capacity is $_4$21, and this corresponds to only about $_4$22 CH$_4$23 purity in PVSA. Parent CALF-20, by contrast, is reported to suffer from poor regenerability, with CO$_4$24 remaining bound at $_4$25, and from low capacity. These comparisons directly support the conclusion that methane co-adsorption and low-pressure desorption behavior are decisive discriminants within the series.

6. Mechanistic interpretation, limitations, and projected development

The study attributes the performance of FumCALF-20 to three linked factors: a high CO$_4$26 working capacity of $_4$27, very low CH$_4$28 uptake of $_4$29, and a resulting CO$_4$30/CH$_4$31 selectivity of approximately $_4$32. In addition, the reported moderate $_4$33 is interpreted as providing sufficiently strong CO$_4$34 binding for uptake while still allowing facile desorption under vacuum, and the pore network, summarized by $_4$35 and $_4$36, is described as balancing transport and capacity (Shin et al., 20 Jul 2025).

The same source also identifies scale-up constraints. Vacuum pumping to about $_4$37 is reported to increase capital and energy costs, which motivates possible hybrid vacuum/pressure stages. Water co-adsorption, pelletization effects, and mass-transfer resistances are identified as issues requiring experimental validation. Thermal management and multi-bed arrangements are stated to be required for continuous operation.

Reported recommendations for further optimization include introducing tailored kinetic enhancers such as hierarchical porosity to reduce cycle times, exploring mixed-linker variants to fine-tune $_4$38 and selectivity, and integrating heat- and work-recovery loops while evaluating humidity resilience for real-world biogas feeds. These recommendations indicate that the current result is a molecular-to-process screening outcome rather than a final deployment study. Even so, within the single-column PVSA framework examined, FumCALF-20 is singled out as the only CALF-20 derivative capable of delivering at least $_4$39 CH$_4$40 purity and at least $_4$41 recovery, which is the defining result behind its current prominence in this research niche.