A Combined Computational and Experimental Approach to Ultra-High - PowerPoint PPT Presentation

A Combined Computational and Experimental Approach to Ultra-High Permeability Mixed Matrix Membranes for Post-Combustion CO 2 Capture Dave Hopkinson, Surendar Venna, Ali Sekizkardes, Sameh Elsaidi, Samir Budhathoki, and Jan Steckel Month 31, 2016 Solutions for Today | Options for Tomorrow

Membranes need very high performance to be used in CO 2 capture from fossil energy Robeson Upper Bound Challenge: Need to process large amount of gases with low available driving force Mixed Matrix Membrane Potential Mixed Matrix Membranes CO 2 /N 2 Selectivity Inorganic Polymer filler CO 2 Permeability (Barrer) Lloyd M.Robeson, Journal of Membrane Science, 320, 2008, 390-400 2 Performance vs cost plot, Courtesy: William Koros

For a 10% reduction in COE over reference plant, CO 2 permeance of 4000 GPU and CO 2 /N 2 selectivity of 25 is needed 20 15 0.1 COE reduction (%) 10 0.25 0.5 α CO2/N2 5 1 25 0 50 -5 50 (advanced) -10 100 (advanced) -15 -20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 CO 2 Permeance (1000 GPU) Recycle Gas Air Booster Secondary Air to Boiler ID-Fan FG Booster Fan Vent Gas Air Sweep Membrane Flue Gas FGD Membrane Module Array Secondary Air Flue Gas Membrane Vent Gas Module Array Two stage membrane CPU Exhaust Gas Retentate Gas Retentate CW CW To Stack process with air sweep Raw CO 2 Booster Fan CO 2 Product Gas CPU Vacuum Pump Water Water Keairns et al, A cost and performance analysis of polymeric membrane-based post- 3 combustion carbon capture, In review

MMMs can increase membrane performance beyond the Robeson Upper Bound Robeson upper bound 100 MMM performance CO 2 /N 2 selectivity 10 Matrimid-UiO-66 NETL Polymer 1 polyphosphazene-SIFSIX NETL Polymer 2 PIM-BILP NETL Polymer 3 IXPE-Silica gel 1 1 10 100 1000 10000 100000 CO 2 Permeability (Barrer) Assumptions of Robeson UB: pure polymers; 35 ⁰C; pure gas; solution -diffusion 4

How do we choose the best pair of polymer and filler particle? By chemical intuition? SIFSIX POP UiO-66 Silica Microporous Polyimide Ionic XL Polyethers Polyphosphazenes Polymers n Polyphosphazene polymer development for mixed matrix membranes using SIFSIX-Cu-2i as performance enhancement filler particles, Journal of Membrane Science, 535 (2017) 103-112. 5 Incorporation of benzimidazole linked polymers into Matrimid to yield mixed matrix membranes with enhanced CO 2 /N 2 selectivity, Journal of Membrane Science, 554 (2018). Carbon Dioxide Separation from Flue Gas by Mixed Matrix Membranes with Dual Phase Microporous Polymeric Constituents, Chemical Communications, 52 (2016) 11768-11771.

According to the Maxwell Model, properties of the polymer and filler must be complementary R polymer Matrimid CO 2 Permeability = 10 Barrer Interface Matrimid CO 2 /N 2 selectivity = 30 45 40 35 CO 2 /N 2 Selectivity R seive 30 25 CO 2 Permeability 20 15 10 𝑄 𝑒 + 2𝑄 𝑑 − 2∅ 𝑒 𝑄 𝑑 − 𝑄 𝑒 𝑄 𝑓𝑔𝑔 = 𝑄 𝑑 5 𝑄 𝑒 + 2𝑄 𝑑 + ∅ 𝑒 𝑄 𝑑 − 𝑄 𝑒 Matrimid with 23% filler particle 0 1E+00 1E+01 1E+02 1E+03 1E+04 1E+05 1E+06 1E+07 Assumptions of Maxwell Model: Filler Particle Permeability (Barrer) • Resistors in series MMM CO2 Permeability (Barrer) MMM CO2/N2 Selectivity • No particle agglomeration • For optimum selectivity, permeability of particle • Low particle loading, spherical should be < 100X greater than polymer • Ideal interface • MMM permeability improvement has limitations 6 Journal of Molecular Structure 739 (2005) 87 – 98

Computational modeling is used to predict MOF and MMM properties 7 Budhathoki et al, Energy Environ. Sci. 2019, 12, 1255

Permeability of MOFs is calculated based on pore geometry Pore Limiting Diameter Diffusivity Solubility Grand Canonical Monte Carlo simulations are used Molecular dynamics simulations are used to to calculate CO 2 and N 2 solubility for rigid MOFs calculate CO 2 and N 2 diffusivity MOF Permeability = Solubility X Diffusivity Mixed Matrix Membrane Permeability is from the Maxwell Model 8

Predictions of MMM permeability are in good agreement with literature data Blue markers = CO 2 permeability; Green markers = N 2 permeability 9

CO 2 permeability and CO 2 /N 2 selectivity is calculated for MMMs with hypothetical MOFs CO 2 /N 2 Selectivity CO 2 Permeability (Barrer) 10

Compared to pure polymer, MMMs can dramatically reduce the cost of capture MMM CO 2 removal system: Two stage membrane NETL Polymer with air sweep • Cost Reduction from ~$61 to ~$46 per tonne CO 2 • Reduction of ~24% 11

There are many practical considerations for a high performance membrane High performance Ultra-thin, defect- polymer free selective layer 1 3 2 4 Support with Nano-size MOF with optimum pore size matched properties and density 12

PIM-1/MEEP-Polyphosphazene polymers combine the best properties of each PIM-1: High Permeability Low Selectivity Brittle films Physical aging reduces permeability MEEP: Low Permeability High Selectivity Gummy films 13 J. Mat. Chem. A 2018 , 6 , 22472

Thin film PIM-1/MEEP has reduced aging compared with neat PIM-1 PIM-1/MEEP: 150 nm Gutter layer: 250 nm 500 nm MEEP PIM-1 PIM-MEEP suffers less aging than PIM-1 due to (1) chain-chain entanglement (2) MEEP chain/PIM-1 pore intercalations 14

A hollow fiber support needs to be optimized for flux, pore size, and pore density Our current hollow fiber membrane supports: • N 2 permeance >100,000 GPU • CO 2 /N 2 selectivity ~ 0.8 (Knudsen diffusion) • Surface pore size ~ 20 nm • Resistant to mild solvents The support should have at least an order of magnitude higher gas flux compared to selective layer 15

MOF A can now be synthesized in a variety of particle sizes with the same structure a b c d e f TEM Images (scale bars = 200 nm) Diameter 43 ± 9 67 ± 11 82 ± 12 104 ± 16 151 ± 24 248 ± 34 (nm) Surface area 1158 ± 2 1353 ± 3 1205 ± 2 1393 ± 3 1409 ± 4 1410 ± 4 (m 2 /g, N 2 77 K) 16

NETL MMMs are above the Robeson Upper Bound with high CO 2 permeability Robeson Upper Bound Experimental MMMs Neat PIM-1/MEEP Simulations Other reported MMMs 17

Increasing MOF concentration improves P CO2 with little effect on a CO2/N2 18

MMMs show stable performance when tested in actual flue gas with contaminants NCCC, Alabama NETL’s membrane flue gas test unit at the MMM with MOF A National Carbon Capture Center 19 Kusuma et al.,Journal of Membrane Science, 533, 2017, 28 – 37

Summary: NETL has taken a multifaceted approach to MMM development for low cost CO 2 capture MMM NETL Polymer • Using high throughput computational techniques, properties of polymer/MOF can be matched to make better MMMs • • For an NETL polymer, the cost of capture can MMMs developed at NETL are above the be reduced from $61 to $46/tonne CO 2 Robeson Upper Bound • High permeance hollow fiber supports have been fabricated • • MMMs have been tested at NCCC with real Techniques for thin film coatings of MMMs are flue gas and show stable performance being developed 20

Thanks to our team! MOF development: Membrane fabrication Past team members: Sameh Elsaidi and testing: Surendar Venna Jeff Culp Victor Kusuma Anne Marti Nathaniel Rosi Fangming Xiang Jie Feng Patrick Muldoon Shouliang Yi Ganpat Dahe Lingxiang Zhu Dave Luebke Polymer development: Zi Tong Hunaid Nulwala Ali Sekizkardes Erik Albenze James Baker Team leads: Alex Spore Dave Hopkinson Hyuk Taek Kwon Simulations and Kevin Resnik Megan Macala economic analysis: Olukayode Ajayi Program management: Samir Budhathoki Lynn Brickett Jan Steckel John Litynski Wei Shi Christopher Wilmer Acknowledgement: This project was funded by the Department of Energy, National Energy Technology Laboratory, an agency of the United States Government, under the Carbon Capture Field Work Proposal and in part through a support contract with AECOM (DE-FE0004000). Neither the United States Government nor any agency thereof, nor any of their employees, nor AECOM, nor any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. 21