1
Development of a Novel Biphasic CO2 Absorption Process
with Multiple Stages of Liquid–Liquid Phase Separation
for Post-Combustion Carbon Capture
(DOE/NETL Agreement No. DE-FE0026434)
Yongqi Lu
Illinois State Geological Survey, Prairie Research Institute
University of Illinois at Urbana-Champaign
2016 NETL CO2 Capture Technology Meeting
Pittsburgh PA • August 11, 2016
Project Overview
Project objectives
Developing new biphasic solvents
Demonstrate phase separation-coupled CO2 absorption
process
Generate and assess engineering and scale-up data
Project duration
10/1/15 – 9/30/18 (36 months for two BPs)
Funding profile
2
DOE funding 1,999,996
BP1 1,079,663
BP2 920,333
Cost share (Cash & In-kind) 501,052
BP1 269,920
BP2 231,132
Total 2,501,048
Project Participants
University of Illinois
Illinois State Geological Survey
• Solvent screening & development
• Solvent equilibria, kinetics & properties measurements
• Absorption and desorption column testing
Illinois Sustainable technology Center
• Evaluation of solvent stabilities and corrosion impact
Applied Research Institute
• Molecular dynamics simulation study for solvent screening
Trimeric Corporation
• Process feasibility and TEA analysis
3
Conventional Monophasic Absorption Approach
4
Conventional CO2 Absorption Process (e.g., MEA)
Flue gas
Cleaned gas
Absorber
(40-50 C)
CO2-rich solvent
Cross-heat
exchanger
Stripper
(90-140 C)
Aqueous CO2
lean solvent
Combined
CO2-lean solvent
CO2
compressor
Reboiler
Steam
Reflux
Cooler
5
Advantages of Biphasic CO2 Absorption Processes
Conceptual of Biphasic Absorption Processes by Other Developers
Reduced equipment size due to reduced mass of solvent to be regenerated in
stripper
Reduced energy use and compression cost due to increased CO2 loading
capacity (concentrated feed), reduced mass, and increased stripping pressure
Impacts on stripper:
CO2-rich
phase
CO2-lean
phaseFlue gas
Cleaned gas
Absorber
(40-50 C)
CO2-rich solvent
Cross-heat
exchanger
Stripper
(90-140 C)
Aqueous CO2
lean solvent
Combined
CO2-lean solvent
LLPS
Reboiler
Steam
Reflux
Cooler
CO2
compressor
Biphasic CO2 Absorption Process with
Multi-Stages of Liquid-Liquid Phase Separation (BiCAP)
6
LP
steam
Flue gas
Cleaned
flue gas
Absorption
column
High-
pressure
stripper
Reboiler
CO2 (1)
Cooler
Cross heat
exchanger
Lean
solvent
CO2-rich
phase
CO2-lean
phase
Condensate
to power
plant cooling
tower
Inter-stage
cooler
Rich
solventWater
separator
Rich
solvent
tank
Solvent
makeup
Solvent
tankCondenser
Phase
separator
LP
steam
Flash
CondenserCO2 (2) to
compressor
Condensate
to cooling
tower
Reduced viscosity with separation of rich, viscous phase improves mass
transfer rate
Lean phase to next packed bed improves kinetics
Reduced mass of solvent to next packed bed
Adds impacts on absorber to the impacts on stripper:
Advantages of Multi-Stage Phase Separation (LLPS)
during CO2 Absorption
Modified operating curve allows for a higher mass transfer driving
force and thus a faster absorption rate7
0 0.2 0.4 0.6 0.8 1 1.2
CO
2p
art
ial
pre
ssu
re,
kP
a
CO2 loading, mol/mol solvent alkalinity
Flue gas inlet
Flue gas outlet
Equilibirum
curve
D'
BB'
CC'
D
A
Operating
curve
12 kPa
1.2 kPa
A
B
B’
C
C’
D
LLPS
LLPS
LLPS
Illustration of operating & equilibrium curves with 3 stages of CO2 absorption-LLPS
(the equilibrium curve that may change after each LLPS is not displayed in this illustration)
BiCAP vs MEA and Other Biphasic Processes
8
Biphasic processes vs MEA
Biphasic solvents have
larger loading capacity for
CO2 stripping due to
absorbed CO2 concentrated
in one phase as feed
solution to the stripper
Reduced mass and
elevated P for CO2 stripping
Reduced heat duty (low
sensible heat and
stripping heat)
Reduced compression
work requirement
BiCAP vs other biphasic processes
Absorption process:
Multi-LLPS in BiCAP allows for low CO2
loading and low viscosity throughout the
absorber, resulting in a faster absorption
rate and reduced absorber size
Solvent:
Phase transition behavior of BiCAP
solvents are tunable, facilitated with the
use of a unique solubilizer(s), allowing
for a wide range of solvent component
selection
Desorption process:
Desorption with a flash step to obtain
high-pressure CO2 and reduce
compression requirements
Project On-Track and Initial Milestones Achieved
9
WBS Lead Description Start End Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12
1.0 Project management and planning 10/01/15 09/30/18
1.1 ISGS/All Project management and planning 10/01/15 09/30/18
1.2 ISGS/All Briefings and reports 10/01/15 09/30/18
2.0 Screening and characterization of biphasic solvents 10/01/15 06/30/16
2.1 ISGS Solvent screening tests on CO2 absorption and phase transition behavior 10/01/15 06/30/16
2.2 ISGS Solvent screening tests on CO2 desorption performance 10/01/15 06/30/16 a
2.3 ARI Molecular simulation study for solvent screening 10/01/15 06/30/16 b
3.0 Measuring phase equilibria, absorption kinetics, & solvent properties 01/01/16 09/30/16 A
3.1 ISGS Measurement of VLE data under absorption & desorption conditions 01/01/16 09/30/16
3.2 ISGS Measurement of CO2 absorption kinetics 04/01/16 09/30/16 c
3.3 ISGS Measurement of solvent properties 07/01/16 09/30/16
4.0 Determining thermal and oxidation stabilities of the selected solvents 04/01/16 12/31/16
4.1 ISTC Oxidation stability of solvents under simulated absorption conditions 04/01/16 12/31/16
4.2 ISTC Thermal stability of solvents under simulated desorption conditions 04/01/16 12/31/16 e
5.0 Testing CO2 absorption and phase separation in a packed-bed column 04/01/16 03/31/17 B
5.1 ISGS Modification of absorption column to incorporate multi-LLPS operation 04/01/16 09/30/16 d
5.2 ISGS Parametric testing of CO2 absorptionand LLPS in the packed-bed column 07/01/16 03/31/17 f
5.3 ISGS Rate-based modeling of CO2 absorption in the packed-bed column 10/01/17 03/31/17
6.0 Development of a process sheet and preliminary process analysis 04/01/16 03/31/17
6.1 Trimeric Development of a conceptual process flow sheet 04/01/16 12/31/16
6.2 Trimeric Preliminary process analysis 07/01/16 03/31/17 g
7.0 Testing CO2 desorption in a high-pressure flash and stripping column 04/01/17 03/31/18 C
7.1 ISGS Modification of an existing packed-bed column by incorporating a flash unit04/01/17 09/30/17 h
7.2 ISGS Parametric testing of high-pressure flash and stripping 07/01/17 03/31/18 j
7.3 ISGS Design modeling of CO2 desorption in the flash and stripping column 10/01/17 03/31/18
8.0 Assessing the impact of solvent corrosion on the equipment 04/01/17 12/31/17
8.1 ISTC Assessing the impact of solvent corrosion on the equipment 04/01/17 12/31/17 i
9.0 Technical and economic feasibility study 10/01/17 09/30/18 D
9.1 Trimeric Process simulation and mass & energy balance calculations 10/01/17 06/30/18
9.2 Trimeric Technical and economic feasibility study 01/01/18 09/30/18 k
START/ENDSOPO BREAKOUT SCHEDULE BUDGET PERIOD 1 BUDGET PERIOD 2
Aug 11, 2016
COMPLETE
Project Scope and Technical Approach
10
Biphasic solvent
screening tests
(~50 solvents)
Molecular dynamics
simulations for
solvent screening
Preliminary selection
of solvent
components
Downsize to 5-10
solvents
VLE, kinetics, and
properties
measurements
Downsize to 2-3
solvents
Assessing solvent
oxidation and thermal
degradation
Parametric testing of
CO2 absorption &
LLPS
Process flow sheet
and techno-economic
analysis
Parametric testing of
high-pressure flash
and stripping
Assessing solvent
corrosion impact on
equipment
Biphasic
solvent
screening
(Q3)
Biphasic
solvent
characterization
(Q5 & Q9)
Biphasic
solvent-enabled
process testing
(Q6 & Q10)
Process
analysis & TEA
(Q12)
Key Milestones and Success Criteria
11
BP1 (by Q6):
Identify 2-3 top-performing solvents
(based on phase transition & CO2 enrichment behavior, CO2 loading capacity,
absorption kinetics, and viscosity)
Complete lab testing of 2-3 solvents in an absorption column with multi-LLPS
(CO2 capacity and kinetics 5 M MEA; each LLPS stage ≤ 5 min; 80% CO2
enrichment in the rich liquid phase)
Demonstrates reliable operability of the multi-stage absorption & LLPS
configuration during lab-scale testing
BP2: (by Q12)
Complete lab testing of 2-3 solvents in a flash / stripping system
(≥5 bar stripping pressure; working capacity ≥2 times that of 5M MEA)
Initial techno-economic feasibility study shows significant progress toward
achievement of DOE performance goals
Task 2. Solvent Screening Experiments: Absorption Capacity
12
0.0
1.0
2.0
3.0
4.0
5.0
6.0s1 s2 s3 s4 s5 s6 s7 s8 s9
s10
s11
s12
s13
s14
s15
s16
s17
s18
s19
s20
s21
s22
s23
s24
s25
s26
s27
s28
s29
s30
s31
s32
s33
s34
s35
s36
s37
s38
s39
s40
s41
s42
s43
s44
s45
s46
s47
s48
s49
s50
s51
s52
Tota
l CO
2 lo
adin
g in
so
lven
t, m
ol/
L
Solvents
Total loading, mol/L
--- 5M MEA (benchmark)
Overall CO2 capacity tested in gas impingers for 60 min of absorption
under atmospheric CO2 at 40C:
Most solvents achieved a comparable or slightly higher CO2 loading
than 5M MEA for absorption
Phase Transition Behavior
Formation of dual phases and their volumes are tunable
CO2 loading is highly concentrated in rich phase (91-99% of total loading)
Loading capacity of CO2 desorption for most solvents is improved by 37-
234% over 5M MEA (as only rich phase solution is used for regeneration)13
M
EA s1
s2
s3
s4
s5
s6
s7
s8
s9
s1
0s
11
s1
2s
13
s1
4s
15
s1
6s
17
s1
8s
19
s2
0s
21
s2
2s
23
s2
4s
25
s2
6s
27
s2
8s
29
s3
0s
31
s3
2s
33
s3
4s
35
s3
6s
37
s3
8s
39
s4
0s
41
s4
2s
43
s4
4s
45
s4
6s
47
s4
8s
49
s5
0s
51
s5
20
2
4
6
8
10CO
2 loading in rich phase
% of rich phase CO2 in total CO
2 uptake
CO
2 lo
ad
ing
in
ric
h p
hase, m
ol/L
90
92
94
96
98
100
No
PS
No
PS
No
PS
No
PS
No
PS
% o
f ri
ch
ph
ase C
O2 in
to
tal u
pd
ate
No
PS
Desorption Pressure
PCO2 = 0.710 bar at 120 C at lean CO2 loadings of ~0.30.5 mol/mol
(determined to achieve 90% CO2 removal), which is much higher than
that of lean MEA (pressure at ~0.25 mol/mol)14
10
100
1000
0 0.2 0.4 0.6 0.8 1
CO
2 p
art
ial p
ressu
re, b
ar
CO2 loading, mol/mol of total amine in rich phase
120C, De-S1
120C, De-S2
120C, De-S3
120C, De-S4
120C, De-S5
120C, De-S6
120C, De-S7
120C, De-S8
120C, De-S9
120C, De-S10
120C, De-S11
120C, 5M MEA
MEA op range
BiCAP desorption op range
(lean conditions to be determined)
10
1
0.1
Task 2. Molecular Dynamics Modeling for Solvent
Screening: Methodology Flowchart
15
Predicted efficiency based
on Gibbs free energy of
carbamate formation
Efficiency of phase separation• Implemented steered molecular dynamics simulation to
characterize phase separation process
• Computed free energy change for phase separate
system for selected solvent compositions including MEA
• Developed a method to characterize chemical potential
differences for species in rich an lean phases
Efficiency of carbon capture• Calculated enthalpy of formation for carbamate,
zwitterions, and protonated amines
• Performed stability analysis for selected ion pairs via
association energy in order to construct initial
configurations for phase separation simulations
• Computed Gibbs free energy change of selected reactions
Characterization of reaction pathways and barriers• Identified multiple reaction schemes as a function of solvent composition
• Characterized efficiency of bicarbonate protonation
• Computed energy barriers between transition states and carbonic acid via DFT
• Validated reactive molecular dynamics force field for proton transfer
Experimental Input• NMR, MS-GS, water content in each phase, pH
Transport and mixing properties• Computed self-diffusion coefficient of reactants and
products
• Computed viscosity for selected amines
Characterized transport
including diffusivity of species
Predicted efficiency based
on free energy change for
phase separation
Speciation
completed
Molecular Dynamics Modeling for Efficiencies of
Carbon Capture and Phase Separation
Carbon capture efficiency screening is performed via thermodynamic
calculations using semi-empirical molecular orbital theory (18 reactions
considered). Below is one example for the carbamate formation:
16
• G < 0: spontaneous
reactions; H < 0:
exothermic reactions
• Approach is general to
screen any
stoichiometry &
reactions of interest.
𝐴𝑚 + 𝐶𝑂2 + 𝐴𝑚 → 𝐴𝑚𝐶𝑂2− + 𝐴𝑚𝐻+
System F111
after SMD run
Zwitterions and
carbamates
“steered” to
separate inside
the simulation
domain
System’s energy
change during SMD
Favors phase
separationPhase separation
not favored
• Work done by the
system or constraint is a
measure of the driving
force behind the phase
separation process
Phase separation efficiency screening is performed via steered molecular
dynamics simulation
Task 3. Absorption Rate
Rate tests using both a stirred tank reactor (exemplary results displayed
in plot above) and a wetted wall column reactor (work is ongoing)
Rates tunable by addition of a promoter or selection of a different
solubilizer
* S18 (S17+promoter) and S31 (solubilizer different from S17)17
0.0E+00
5.0E-04
1.0E-03
1.5E-03
2.0E-03
2.5E-03
3.0E-03
0.2 2
CO
2ab
so
rpti
on
rate
, m
ol/
m2.s
CO2 partial pressure, psia
MEA S17 S18
S23 S26 S31
S40 S28 S53*
0.22-0.25 mol CO2/
mol amine
~0.5 mol CO2/
mol amine
0.29-0.42
mol/mol
Task 3. Viscosity Measurement and Optimization
Lean phase viscosity < 9 cP (data not displayed)
Rich phase viscosity was successfully decreased from ~400 cP to ~30
cP by optimizing total solvent concentration and selecting suitable amine
structures
18
0
50
100
150
200
250
300
350
400
450
500
#1 #2 #3 #4 #5 #6 #7 #8 #9 #10#11#12#13#14#15#16#17
Ric
h p
ha
se
vis
co
sit
y, c
P
Rich phase solution
60-70wt% amines
50wt% amines
40wt% amines
Task 5. A Lab Absorption System with 3-Stages of Packed
Beds and LLPS Vessels Designed and Under Fabrication
3 stages (4-in ID, 7-ft packed-bed for each stage) arranged horizontally
to accommodate lab ceiling limit
3 stages in one vertical column envisioned for practical use 19
CO2
cylinder
Aircompressor
~ Elec steam generator
T4
T3
T2
T1
P1
P2
T8
T7
T6
T5
P3
P4
T12
T11
T10
T9
P5
P6
Other gases (Optional)
MFC
MFC
MFC
FM
T0
Water cooler
Water cooler
Water cooler
Elec heater ~
Gas exhaust
Gas 3 to analysis
Gas 4 to analysis
Gas 1 to analysis
Gas (CO2)analyzer
Gas stream drying
Gas 1 Gas 2 Gas 3 Gas 4
T, P, Flow ratedisplay/control
Ts Ps Fs
Pump
Pump
Pump
LLPS
Regenerator
Solvent tank
Liquid sampling
T13
T14
Rotameter
Gas 2 to analysis
Rotameter
Rotameter
Vent
Liquid samplingLiquid sampling
Pump Pump
Liquid sampling
LLPS LLPS
Lab Prototype Phase Separation Vessel Achieved Efficient
and Stable Separation
20
Phase separation vessel design
Based on density difference
(lean phase ~0.85 vs. rich
phase ~1.1 g/cm3)
Residence time ≤ 5 min
(preferred at <1-2 min)
Actual separation performance
Separation efficiency better
than the design
Able to maintain constant
levels of both G-L and L-L
interfaces
Both interface levels adjustable
by adjusting their weir heights
Very stable operation
(Liquid volume of 10 L, total volume
of 15 L, liquid flow rate of 2 L/min)
Task 6. Preliminary Process Flow Diagram
Developed for BiCAP
Work underway to improve process/unit configuration, identify opportunities to minimize
equipment items to reduce cost, and assess integration options into a power plant 21
Future Work Plan in this Project
VLE data under absorption &
desorption conditions
Absorption kinetics
Solvent properties (Hr; viscosity,
diffusivity, Cp, density)
Thermal & oxidation stabilities
Solvent corrosion tendency
For 2-3 selected solvents:
Testing CO2 absorption and LLPS
in a 3-stage packed-bed column
Testing CO2 desorption in a flash
+ stripping column
Preliminary process analysis and
TEA feasibility study
22
Biphasic
solvent
screening
(Q3)
Biphasic
solvent
characterization
(Q5 & Q9)
Biphasic
solvent-enabled
process testing
(Q6 & Q10)
Process
analysis & TEA
study
(Q12)
WWC for kinetics
Low-P VLE
High-P VLE
Oxidative
stability
Thermal stabilityRx calorimeter
GC-MS
NMR
Stripping system
to be modified
CO2
cylinder
Aircompressor
~ Elec steam generator
T4
T3
T2
T1
P1
P2
T8
T7
T6
T5
P3
P4
T12
T11
T10
T9
P5
P6
Other gases (Optional)
MFC
MFC
MFC
FM
T0
Water cooler
Water cooler
Water cooler
Elec heater ~
Gas exhaust
Gas 3 to analysis
Gas 4 to analysis
Gas 1 to analysis
Gas (CO2)analyzer
Gas stream drying
Gas 1 Gas 2 Gas 3 Gas 4
T, P, Flow ratedisplay/control
Ts Ps Fs
Pump
Pump
Pump
LLPS
Regenerator
Solvent tank
Liquid sampling
T13
T14
Rotameter
Gas 2 to analysis
Rotameter
Rotameter
Vent
Liquid samplingLiquid sampling
Pump Pump
Liquid sampling
LLPS LLPSAbsorption column
under constructionStripping system
to be modified
Next-Stage Technology Development
Current project is a laboratory development
If process and TEA feasibility proven in the current project,
next stage would be a close-loop bench or small pilot demo
with simulated or actual flue gas
Rigorous process design and optimization modeling to
enhance performance
Analysis of technical risks and mitigation for scale-up
Identify industrial partners (design, construction, and
testing)
23
Acknowledgements
DOE/NETL Project Manager: Andrew Jones
University of Illinois:
Kevin O’Brien (Co-PI; PhD, Director)
Hong Lu (PhD, Chemical/Environmental Engineer)
David Ruhter (MS, Lab Manager)
Yang Du (PhD, Chemical/Environmental Engineer)
Qing Ye (PhD Student)
Wei Zheng (PhD, Senior Chemist)
Brajendra K Sharma (PhD, Senior Chemical Engineer)
Viktoriya Gomilko (MS, Assistant Research Chemist)
Joe Pickowitz (Environmental Engineer)
Santanu Chaudhuri (PhD, Principal Research Scientist)
Naida Lacevic (PhD, Lead Simulation Specialist)
Trimeric Corporation:
Ray McKaskle (Subaward PI; P.E., Senior Chemical engineer)
Andrew Sexton (PhD, P.E., Senior Chemical Engineer)
Kevin Fisher (VP, P.E., Senior Chemical Engineer)
24
Top Related