Energy Efficient GO-PEEK Hybrid Membrane Process for Post-combustion CO2 Capture - NETL
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Energy Efficient GO-PEEK Hybrid Membrane
Process for Post-combustion CO2 Capture
DOE Contract No. DE-FE0026383
Shiguang Li, Naomi Klinghoffer, Travis Pyrzynski, S. James Zhou, Howard Meyer
Gas Technology Institute (GTI)
Huynh Ngoc Tien, Fanglei Zhou, Jarvis Chen, Miao Yu
University of South Carolina (USC)
Yong Ding and Ben Bikson
Air Liquide Advanced Separations (ALaS)
NETL CO2 Capture Technology Meeting
August 8-12, 2016
Project overview
Performance period: Oct. 1, 2015 – Sep. 30, 2018
Funding: $1,999,995 from DOE; $500,000 cost share
Objectives: Develop a hybrid membrane process combining a
graphene oxide (GO) gas separation membrane configuration unit and a
PEEK hollow fiber membrane contactor (HFMC) unit to capture ≥90% of
the CO2 from flue gases with 95% CO2 purity at a cost of electricity 30%
less than the baseline CO2 capture approach
Team:
Member Roles
• Project management and planning
• Quality control and CO2 capture performance tests
• GO membrane development
ALaS • PEEK membrane development
High-level technical & economic feasibility study
2
Process description
Filter GO membrane Treated flue gas to stack
(conventional process) (1.3 (vol.)% CO2, 0.5 psi)
1 2 3
7
Flue Flash
gas 13.2 (vol.)% regenerator
Blower 4 CO2 enriched 8
CO2
permeate
Vacuum Surge
pump 5 tank
PEEK Plant’s
membrane low
absorber pressure
Condenser steam
CO2/SO2 selectivity through GO 6 9
membrane will be tested in the 11
H2O
current program: 10
Case 1: SO2 permeates
>90% purity ~99.5%
through the membrane, a 13
of CO2 purity of CO2 12
caustic scrubber is needed
before the GO membrane CO2 captured (95%
H2O to surge tank
Case 2: SO2 stays in the purity of CO2)
retentate, scrubber unneeded;
HFMC can handle 150 ppmv
SO2 (DE-FE-0004787) 3
GO membrane technology based on our pioneering
work published in Science (2013, 342 (6154) 95)
A
Contribution of the paper:
Structure defect-free GO membrane is
impermeable to gas
1.5 Structural defects on GO flakes can be
B 500 nm
h, nm
controlled as transport pathway for
1.0 selective gas separations
0.5 0 250 500 750
x, nm
Single-layered GO flake
prepared as thin as 1 nm 4
What is a membrane contactor?
High surface area membrane device that facilitates mass transfer
Gas on one side, liquid on other side
Membrane does not wet out in contact with liquid
Separation mechanism: CO2 permeates through membrane, reacts
with the solvent; N2 does not react and has low solubility in solvent
5
Singular PEEK HFMC technology currently at small
pilot scale development stage (DE-FE0012829)
Membrane modules
Surge tank
Heat
Solvent Flash tanks exchangers
storage
tank
12 m (L) x 7.5 m (W) x 3.5 m (H)
Commercial-sized modules 3D model of the 0.5 MWe plant
Pilot plant construction to To be tested at NCCC
6 in 2017
be completed by 9/30/16
GO-PEEK technical goals
CO2 permeance ≥
GO 1,000 GPU Separate and
membranes CO2/N2 selectivity ≥ capture ≥ 90% CO2
GO-PEEK 90 from a simulated
hybrid flue gas with 95%
process Intrinsic CO2 CO2 purity
PEEK permeance ≥ 3,000
membranes GPU
CO2 mass transfer
coefficient ≥ 3 (sec)-1
GPU= Gas Permeation Unit 7
Technical challenges of applying GO-PEEK
process to existing coal-fired plants
GO membrane performance – Needs significant improvement
1000 Li et al., ACS Appl. Mater.
Interfaces. 7, 5528 (2015)
CO2/N2 selectivity
100 Kim et al., Science 342, 91 (2013)
Our target
Our membrane when being
proposed
Shen et al., ACS Sustainable
10 Chem. Eng. 3, 1819 (2015)
Shen et al., Angew. Chem.
Int. Ed., 54 , 578 (2015)
1
0 500 1000
CO2 permeance (GPU)
Durability – Long-term stability of both GO and PEEK membranes
Scale-up and cost reduction – Both membranes in hollow8 fiber format
Progress on GO
Membranes
GO: single-atomic
layered, oxidized
graphene
9
Procedure developed for coating GO on hollow
fiber (HF) support
Valve A Valve B
Coating procedure:
1. Valves A and B are open, GO dispersion flows continuously in hollow fiber
2. Vacuum filtration is conducted for a controlled time; and
3. Close valves A and B, leave the coated fiber under vacuum for a controlled
10 timeGO membrane (thickness: ~40 nm) supported on
polyethersulfone (PES) hollow fiber
A B
PES Uncoated
fiber fiber
surface
1 cm
C D
Coated Coated
fiber fiber cross
surface section
20 µm
300 nm 11Coated fiber sealed in a mini-module for
gas permeation testing
Permeation testing unit
Mini-module
Water
bubbler
and
knockout
vessel 12Approaches to improve membrane performance
Create more structural defects on GO flake by HNO3
etching
Reduce GO flake lateral size by ultra-sonication
W/O ultra-sonication W/ ultra-sonication
13Optimized membranes showed superior performance
to GO-based membranes reported in the literature
1000
Li et al., ACS Appl. Mater.
Interfaces. 7, 5528 (2015)
CO2/N2 selectivity
Our
Kim et al., Science 342, 91 (2013)
membrane
100 PCO2: 540 GPU
Shen et al., ACS Sustainable αCO2/N2: 860
Chem. Eng. 3, 1819 (2015)
Shen et al., Angew. Chem.
Int. Ed., 54 , 578 (2015)
10
1
0 200 400 600
CO2 permeance (GPU) 14Comparison to other CO2/N2 separation membranes
Our GO membranes
PCO2: 540 GPU, αCO2/N2: 860
1000
1OSU composite membranes
Our GO membrane
at proposal stage
100 2MTR Polaris advanced
α (CO2/N2)
(lab scale) membranes
1. Ho, et al., 2015 NETL
CO2 Capture Technology
10 Meeting, June 23-26,
2015, Pittsburgh.
4GE 2. Merkel, et al., Ibid.
composite hollow 3. Venna, Ibid.
3NETL MOF mixed
fiber membranes 4. Bhandari, 2014 NETL
matrix membranes CO2 Capture Technology
Meeting, July 29-August
1, 2014, Pittsburgh.
1
0.1 10 103 105
CO2 permeance (GPU)
Robeson, J. Membrane Sci. 2008, Vol. 320, p390 15
Note: Polymer data points (red): 100 nm membrane thickness assumedProgress on PEEK
Membranes
A B
16Under the current program, we are developing PEEK
fibers with intrinsic CO2 permeance of 3,000 GPU
3rd Generation Current
PEEK fibers under project
development target
CO2 permeance (GPU)
3000
2000
2nd Generation
PEEK fibers developed
for pilot-scale testing
1000 1st Generation
PEEK fibers developed
for bench-scale testing
Beginning of bench-scale program
0
2009 2011 2013 2015 2017
Year
1 GPU = 1 x 106 cm3 (STP)/cm2 • s • cmHg 17To date, intrinsic CO2 permeance of 2,500
GPU obtained for a new 2-inch module
Feed inlet Retentate Permeate CO2
Test Temperature
pressure pressure pressure permeance
number (°C)
(psig) ( psig) (psig) (GPU)
1 22 5.35 2.7 0.27 2,500
2 22 4.31 2.4 0.18 2,400
18This module showed mass transfer coefficient > 3.0 (sec)-1
in capturing CO2 from low CO2-concentration feeds
Membrane contactor solvent: aMDEA
2 4
CO2 concentration for treated gas
Mass transfer coefficient, (sec)-1
1.6
3
1.2
(vol.%)
2
0.8
1
0.4
0 0
0 2 4 6 8 10
19
CO2 feed concentration (vol.%)Plans for future development in this project
Further development for GO and PEEK membranes
GO PEEK
CO2 permeance (GPU)
1000 12/31/17 Target
Target 3rd Generation
CO2/N2 selectivity
To date 3000
100 To date
2000
Proposal 6/30/16 Proposal stage
stage Target 2nd Generation
10 3/31/17
Target 1000
1 0
0 500 1000 7/15/2015 1/31/2016 8/18/2016
CO2 permeance (GPU) Date
Integrated GO-PEEK process tests to achieve technical goal
20After the current project, steps can be
taken to further reduce capture cost
Increase CO2 permeance for both GO and PEEK membranes
Improve manufacture process to lower membrane costs
Use advanced solvents instead of aMDEA
Use novel process for solvent regeneration
e.g. gas pressurized stripping reported by Carbon Capture Scientific1
e.g. advanced flash regeneration by UT 2
1: Scott Chen et al., Ibid
2. Gary Rochelle, 2016 NETL CO2 Capture Technology
21
Meeting, August 8-12, 2016, Pittsburgh.Summary
We are developing a novel CO2 capture process combining
a conventional gas membrane unit and a HFMC unit
GO membrane developed to date
CO2 permeance of 540 GPU and αCO2/N2 of 860 obtained at 80°C for
a humidified CO2/N2 mixture
Superior performance to GO-based membranes reported in the
literature
Selectivity higher than other CO2/N2 separation membranes
PEEK membrane developed to date
Intrinsic CO2 permeance of 2,500 GPU obtained
Mass transfer coefficient > 3.0 (sec)-1 in capturing CO2 from low CO2-
concentration feeds with aMDEA solvent
22Acknowledgements
Financial support
ALaS
DOE NETL José Figueroa
23You can also read
