Salinity gradient energy: Assessment of pressure retarded ...
Transcript of Salinity gradient energy: Assessment of pressure retarded ...
Engineering Conferences InternationalECI Digital Archives
Advanced Membrane Technology VII Proceedings
9-14-2016
Salinity gradient energy: Assessment of pressureretarded osmosis and osmotic heat engines forenergy generation from low-grade heat sourcesTzahi Y. CathColorado School of Mines, [email protected]
Johan VannesteColorado School of Mines
Michael B. HeeleyColorado School of Mines
Kerri L. HickenbottomHumboldt State University
Follow this and additional works at: http://dc.engconfintl.org/membrane_technology_vii
This Abstract and Presentation is brought to you for free and open access by the Proceedings at ECI Digital Archives. It has been accepted for inclusionin Advanced Membrane Technology VII by an authorized administrator of ECI Digital Archives. For more information, please [email protected].
Recommended CitationTzahi Y. Cath, Johan Vanneste, Michael B. Heeley, and Kerri L. Hickenbottom, "Salinity gradient energy: Assessment of pressureretarded osmosis and osmotic heat engines for energy generation from low-grade heat sources" in "Advanced Membrane TechnologyVII", Isabel C. Escobar, Professor, University of Kentucky, USA Jamie Hestekin, Associate Professor, University of Arkansas, USA Eds,ECI Symposium Series, (2016). http://dc.engconfintl.org/membrane_technology_vii/27
Salinity gradient energy: Assessment of pressure retarded osmosis andosmotic heat engines for energy generation from low-grade heat sources
Johan Vanneste, Kerri L. Hickenbottom, Tzahi Y. Cath
Colorado School of Mines
Dept. of Civil and Environmental Engineering
ECI, Advanced Membrane Technology VII
September 14th, 2016
Cork, Ireland
2
Structure of the Presentation
Energy
PressureRetardedOsmosis
Low-grade heat
Membrane Distillation
Output:
3
Structure of the Presentation
Energy
PressureRetardedOsmosis
Low-grade heat
Membrane Distillation
Output:
Part I: Pressure
retarded osmosis
(PRO): membranes
and spacers evaluation
Part II: Osmotic heat
engine (OHE): working
fluid selection
Part III: Techno-
economic assessment
of the OHE
4
Pressure Retarded Osmosis (PRO): Utilizing Energy from the Ocean
PX
TurbineDraw SolutionHigh Concentration
High Pressure
Feed SolutionLow Concentration
Low Pressure
Applied Pressure < Osmotic Pressure
Semi-permeable,
hydrophilic membrane
Power Density (W) = JwΔP (power output per membrane area)
Water Flux (Jw) = A(Δπ-ΔP) (simplistic!)
5
Open-Loop PRO
Applications
Seawater – River water
RO concentrate –freshwater
Dead Sea and Great Salt Lake
Limitations of open loop PRO
Extensive pretreatment needed
Potential irreversible membrane fouling
Solution chemistry and temperature are fixed
Source: Statkraft
6
LC HC
Jw
Wt
Wh,pro
Closed-Loop PRO: Osmotic heat engine (OHE) Controlled solution
temperature and chemistry
High osmotic pressures
yields high power densities
Lower reverse salt flux
Can utilize low-grade heat
Potential for energy
storage
Thermal
separation
Closed loop system =
No backwashing or
chemical cleaning!
HC = high concentration (DS)
LC = low concentration (DI water)
7
Membrane Distillation (MD): Utilizing LGH for Draw Solution and Feed Stream Regeneration
High Concentration
High Temperature
Low Concentration
Low TemperatureMicroporous, hydrophobic
membrane
Water Flux (Jw) = Aw(ΔPv*) (simplistic!)
Jw
Can utilize low-grade heat to simultaneously
separate and concentrate mixed streams
8
PRO MEMBRANE AND SPACER ASSESSMENT
9
PRO: An Energy Generating Membrane Process
PRO membranes are not
yet commercially
available and FO
membranes are used
PRO membranes must:
Exhibit high power density
& low reverse solute flux
Withstand high
operating pressures
PRO membrane spacers
must provide good mixing
and adequate support
LC HC
Active
Layer
Porous
Support
Js
Jw
CF,b
CF,m
CD,m
CD,b
ICP
C
EC
PD
HC = high concentration
LC = low-concentration
CD,b = bulk draw concentration
CD,m = membrane interface draw conc.
CF,b = bulk draw concentration
CF,m = membrane interface draw conc.
ECPdil = dilutive external CP
ICPC = concentrative external CP
10
Operating Conditions: PRO
Source water
Draw solution (DS):
1, 2, & 3 M NaCl
Feed: deionized water
Bench scale testing
SCADA system controls
temperatures, and collect data to
calculate water flux, batch recovery,
and salt rejection
Flat sheet FO membranes
Hydration Technologies Innovations
(HTI) thin-film composite (TFC)
HTI cellulose triacetate (CTA)
Oasys Water TFC
X company TFC
11
PRO Bench-Scale SystemMembrane module
PROzilla
Membrane spacers
Tricot (35-ch) Tricot (20-ch) Extruded mesh
12
PRO Membrane Evaluation:
0 100 200 300 400 500
0
1
2
3
4
5
6
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10
0 1 2 3 4
DS hydraulic pressure, psi
Po
wer
den
sit
y, W
/m2
DS hydraulic pressure, MPa
X HTI CTA Oasys HTI TFC
0 100 200 300 400 500
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
16
0 1 2 3 4
DS hydraulic pressure, psi
Sp
ec
ific
re
ve
rse
so
lute
flu
x,
g/L
DS hydraulic pressure, MPa
X HTI CTA Oasys HTI TFC
1 M NaCl draw solution (DS); 2x – 20 channel tricot on feed✓ High power densities
✓ Low specific reverse solute flux
✓ Good mechanical stability
13
Spacer Orientation
Parallel to flow
45° to flow1 M
2 M
3 M
0 200 400 600
0
5
10
15
20
25
0
5
10
15
20
25
0 1 2 3 4 5
DS hydraulic pressure, psi
Po
wer
den
sit
y, W
/m2
DS hydraulic pressure, MPa
HTI TFC 3 M HTI TFC 2 M HTI TFC 1 M
Feed spacer orientations
Corrugated heat
exchanger plate
Spacerless patterned membranes!
48 % increase!
14
Membrane Deformation
Virgin
membrane
Membrane
after 450 psi
Surface
Cross-
section
Membrane
after 600 psi
0
2
4
6
8
10
12
14
1 2 3 4 5 6 7W
ate
r fl
ux
, L
m-2
hr-
1S
pe
cif
ic r
eve
rse
so
lute
flu
x, g
L-1
Elapsed time, hr
Js/JwJw
Importance of long
term experiments!!!
700 psi
15
OHE WORKING FLUID SELECTION
16
𝐽𝑤 = 𝐴 𝜋𝐷,𝑏 exp −
𝐽𝑤
𝑘 − 𝜋𝐹,𝑏 exp 𝐽𝑤
𝑆
𝐷
1 +𝐵
𝐽𝑤 exp 𝐽𝑤
𝑆
𝐷 − exp(−
𝐽𝑤
𝑘)
− ∆𝑃
Reverse Solute Flux in PRO and Implications for OHE Process Performance
Non-idealities in PRO gives rise to
reverse solute flux (RSF, Js):
To sustain osmotic driving force, must
bleed a portion of the PRO feed
stream to the MD feed
Impacts of RSF on net power outputs
and generation costs are unknown!
HC LCLC HC
Jw
Jw
Js
Wt
MDPRO
Legend:
π = osmotic driving force
k = mass transfer coefficient (DS side)
S = membrane structural parameter
B = solute permeability coefficient
D = draw solution diffusivity
𝐽𝑠 = 𝐵 𝜋𝐷,𝑏 exp −
𝐽𝑤
𝑘 − 𝜋𝐹,𝑏 exp 𝐽𝑤
𝑆
𝐷
1 +𝐵
𝐽𝑤 exp 𝐽𝑤
𝑆
𝐷 − exp(−
𝐽𝑤
𝑘)
17
Eight salts met screening criteria (π > πNaCl & non-hazardous)
Tested at concentrations equivalent to 17.4 MPa osmotic pressure
(osmotic pressure of 3 M NaCl)
Draw solution Cost Concentration
$/kg M
CaCl2 83 1.6
HCOONa 14 4.1
KBr 42 3.2
LiBr 121 2.2
LiCl 74 2.6
MgCl2 14 1.5
Na(C2H5COO) 38 4.1
NaCl 6 3.00
4
8
12
16
20
24
0
100
200
300
400
500
600
LiC
lL
iBr
CaC
l2M
gC
l2N
a(C
2H
5C
OO
)H
CO
ON
aN
aC
lK
Br
(NH
4)2
SO
4K
Cl
C2H
3N
aO
2 N
a3C
6H
5O
7N
a2S
O4
Mg
(CH
3C
OO
)2M
gS
O4
NH
4H
CO
3K
HC
O3
NaH
CO
3
PV
P, k
Pa
Os
mo
tic
pre
ss
ure
, M
Pa
Osmotic pressure PVP (b)
Draw Solution Screening and Selection
18
Operating Conditions: PRO
Source water
Draw solution: varied, 30 ºC
Feed: deionized water, 30 ºC
Membrane: HTI TFC
Draw solution hydraulic
pressure kept constant at 2
MPa (300 psi)
Membrane
cell
Feed
Feed
loop
Draw
loop
Control
system
Draw
TT
P P
PP
CC
Dosing
~~
Flow meter
Back pressure valve
Gear Pump
Peristaltic pump
Plunger pump
Legend
Ultrasonic sensor
Conductivity probe
Pressure transducer
Temperature probe
Heat exchanger
~
P
T
C
19
Operating Conditions: MD
Source water
Feed solution: varied
Distillate: deionized water
Stream temperatures
Feed solution: 55 ºC
Distillate: 25 ºC
Flat sheet, hydrophobic,
microporous (0.2 μm)
membranes from 3M
Flow meter
Gear Pump
Peristaltic pump
Scale
Stir plate
Legend
Ultrasonic sensor
Conductivity probe
Pressure transducer
Temperature probe
Heat exchanger
~
P
T
C
0
.
0
0
0
0
0.0000
Membrane
cell
Distillate
Distillate
loop
Feed
loop
Control
system
Feed
TT
T T
PP
CC
Dosing
20
PRO Performance
Difference in power densities is because of the difference in
diffusivity and solute permeability coefficient (B)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
2
4
6
8
10
12
14
16
18
20
Dif
fusiv
ity,
(10
-9)
m2/s
Po
wer
den
sit
y, W
/m2
Power densityDiffusivity
21
Higher diffusivities lead to higher reverse salt fluxes
Salts with high reverse salt fluxes can be detrimental to
system costs
PRO Performance
0
1
2
3
4
0
20
40
60
80
100
120
140
160
B,
L/m
2-h
Dif
fusiv
ity,
(10
-9)
m2/s
Revers
e s
olu
te f
lux,
g/m
2-h
Solute flux
Diffusivity
B
22
Distillate conductivity decreased over time, indicating 100%
rejection of salts
Water flux decreases with decreased partial vapor pressure
difference
MD Performance
0
2
4
6
8
10
12
14
16
0
5
10
15
20
25
30
35
40
ΔP
VP, kP
a
Wate
r fl
ux,
L/m
2-h
FluxΔPVP
23
Preliminary Economic Evaluation of Draw Solutions
Co,h,pro
Qo,h,pro
HC LC LC HC
MD PRO
Jw Jw
Js
Co,h,md
Qo,h,md
Co,l,md
Qo,l,md
Ci,h,md
Qi,h,md
Ci,l,md
Qi,l,md
Ci,l,pro
Qi,l,pro
Co,l,pro
Qo,l,pro
Buffer TANK
Co,h,t
Qo,h,t
Ci,h,t
Qi,h,t
Ci,b
Qi,b
Co,b
Qo,b
Qr,l,pro
Legend Q = flow C = concentration
HC = high concentration LC = low concentration
Jw = water flux Js = salt flux
Subscripts i = inlet
o = outlet r = return l = low concentration
h = high concentration p = permeate b = bleed
t = tank T = turbine
Symbols
Heat exchanger
Gear pump
Plunger pump
Turbine-generator
Pressure exchanger
Wh,pro
Wl,pro
Wl,md
Wh,md
WT
Ci,h,pro
Qi,h,pro
MD
-LC
Ta
nk
PR
O-L
C
Ta
nk
Co,h,pro
Qo,h,pro
HC LC LC HC
MD PRO
Jw Jw
Js
Co,h,md
Qo,h,md
Co,l,md
Qo,l,md
Ci,h,md
Qi,h,md
Ci,l,md
Qi,l,md
Ci,l,pro
Qi,l,pro
Co,l,pro
Qo,l,pro
Buffer TANK
Co,h,t
Qo,h,t
Ci,h,t
Qi,h,t
Ci,b
Qi,b
Co,b
Qo,b
Qr,l,pro
Legend Q = flow C = concentration
HC = high concentration LC = low concentration
Jw = water flux Js = salt flux
Subscripts i = inlet
o = outlet r = return l = low concentration
h = high concentration p = permeate b = bleed
t = tank T = turbine
Symbols
Heat exchanger
Gear pump
Plunger pump
Turbine-generator
Pressure exchanger
Wh,pro
Wl,pro
Wl,md
Wh,md
WT
Ci,h,pro
Qi,h,pro
MD
-LC
Ta
nk
PR
O-L
C
Ta
nk
Co,h,pro
Qo,h,pro
HC LC LC HC
MD PRO
Jw Jw
Js
Co,h,md
Qo,h,md
Co,l,md
Qo,l,md
Ci,h,md
Qi,h,md
Ci,l,md
Qi,l,md
Ci,l,pro
Qi,l,pro
Co,l,pro
Qo,l,pro
Buffer TANK
Co,h,t
Qo,h,t
Ci,h,t
Qi,h,t
Ci,b
Qi,b
Co,b
Qo,b
Qr,l,pro
Legend Q = flow C = concentration
HC = high concentration LC = low concentration
Jw = water flux Js = salt flux
Subscripts i = inlet
o = outlet r = return l = low concentration
h = high concentration p = permeate b = bleed
t = tank T = turbine
Symbols
Heat exchanger
Gear pump
Plunger pump
Turbine-generator
Pressure exchanger
Wh,pro
Wl,pro
Wl,md
Wh,md
WT
Ci,h,pro
Qi,h,pro
MD
-LC
Ta
nk
PR
O-L
C
Ta
nk
Co,h,pro
Qo,h,pro
HC LC LC HC
MD PRO
Jw Jw
Js
Co,h,md
Qo,h,md
Co,l,md
Qo,l,md
Ci,h,md
Qi,h,md
Ci,l,md
Qi,l,md
Ci,l,pro
Qi,l,pro
Co,l,pro
Qo,l,pro
Buffer TANK
Co,h,t
Qo,h,t
Ci,h,t
Qi,h,t
Ci,b
Qi,b
Co,b
Qo,b
Qr,l,pro
Legend Q = flow C = concentration
HC = high concentration LC = low concentration
Jw = water flux Js = salt flux
Subscripts i = inlet
o = outlet r = return l = low concentration
h = high concentration p = permeate b = bleed
t = tank T = turbine
Symbols
Heat exchanger
Gear pump
Plunger pump
Turbine-generator
Pressure exchanger
Wh,pro
Wl,pro
Wl,md
Wh,md
WT
Ci,h,pro
Qi,h,pro
MD
-LC
Ta
nk
PR
O-L
C
Ta
nk
Experimental PRO and MD
results were used in the
model
Design constraint: PRO feed
concentration of 4 g/L
(DS specific)
Specific membrane and
module costs for PRO and
MD were referenced from
RO and MF literature,
respectively
24
Impact of Draw Solutions on System Costing
CaCl2 and MgCl2 were the salts that performed best in MD and had the
lowest specific reverse solute fluxes
Electricity generation costs are higher than expected because of low
PRO operating pressures (low power densities)
0
1
2
3
4
5
6
7
8
9
Ele
ctr
icit
y g
en
era
tio
n
co
st,
$ / k
Wh
25
MD membranes are the highest costs
Salts with high RSF result in more bleeding, subsequently
decreasing net energy and increasing MD membrane area
System Costing and Net Energy (1 MW gross)
0
100
200
300
400
500
600
0
2
4
6
8
10
12
Net
po
wer,
kW
Ca
pit
al
co
st,
$ m
illi
on
Salt costs
PRO membrane costs
MD membrane costs (bleed only)
MD membrane costs (no bleed)
Net power
26
With addition of MgCl2, high water flux of NaCl can be
maintained while RSF drops significantly
Mixed draw solutions: best of both worlds?
0
10
20
30
40
50
60
70
80
90
0
2
4
6
8
10
12
14
16
0 20 40 60 80 100
RS
F, m
mo
l/m
2-h
r
Wa
ter
flu
x, L
/m2-h
r
Percent MgCl2-OP
Water flux Solute flux
27
OHE TECHNO-ECONOMIC ASSESSMENT
28
OHE System Model and Cost Assumptions
Base case scenario
2.5 MW (net power) system
Draw solution: 3 M NaCl
PRO operating pressure:
3.4 MPa (~500 psi)
MD temperatures:
70 °C feed, 30 °C distillate
Assumptions
Low grade heat and cooling is free
Membrane costs referenced from
commercially available membranes
PRO costs referenced from RO
MD costs referenced from MF
Equipment efficienciesPump efficiency 70 %Turbine efficiency 90 %Pressure exchanger efficiency 95 %Generator efficiency 95 %Heat exchanger efficiency 60 %
Data and other assumptionsPlant life 20 yrPlant availability 90 %PRO membrane replacement 10 %/yr
MD membrane replacement 10 %/yrInterest (discount) rate, I 8 %Inflation rate, n 3 %Amortization factor 0.1Labor cost 0.03 $/m3
Specific membrane costs*
PRO membrane element cost 11 $/m2
MD membrane element cost 24 $/m2
PRO membrane housing cost 17 $/m2
MD membrane housing cost 14 $/m2
*Assumed 35% mark-down from
quoted distributor cost
29
OHE Power Generation
Power outputs
Gross Power: 4.9 MW
Net Power: 2.5 MW
About 20 % of capital costs
due to bleeding
Process efficiency
Theoretical efficiency1 4%
System efficiency 0.1%
System costs: $0.48 per kWh
Benchmark <$0.20 per kWh
1Lin et al., ES&T 2014
Hickenbottom et. al, in Prep.
Civil work36%
Generation - PRO13%
Regeneration - MD (total)26%
Hydraulic system
4%
Control system
21%
Capital Cost $57 million
MD
26 % PRO
13 %
Civil work
36 %
Control
system 21 %
Hydraulic
system 4 %
Cost of labor and
spare82%
PRO membrane replaceme
nt 4%
MD membrane replaceme
nt14%
O&M Cost $72 million
MD 14 %
PRO 4%Labor 82%
30
0
10
20
30
40
50
60
70
80
90
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0 2 4 6 8 10
Po
we
r d
en
sit
y, W
/m2
Ge
ne
rati
on
co
sts
, $
/kW
h
PRO hydraulic pressure, MPa (psi)
Cost Sensitivity: PRO Power Density and solute permeability(B)
(290) (580) (870) (1,160)
0
5
10
15
20
25
30
35
40
45
50
0
1
2
3
4
5
6
7
8
0 2 4 6 8
Po
we
r d
en
sit
y, W
/m2
Ge
ne
rati
on
co
sts
, $
/kW
h
B, L/m2-hr
Generation cost - w/bleeding
PRO power density
31
Assumptions and Model Outputs: Ideal Case OHE
25 MW (net power) system
(economy of scale)
Draw solution: 3 M NaCl
PRO operating pressure:
7.6 MPa (1,100 psi),
corresponding to a PRO
power density of 76 W/m2
MD temperatures:
85 °C feed, 15 °C
distillate
OHE Sensitivities
Base Ideal Unit
System size
(net power) 2.5 25 MW
Electricity generation cost 0.48 0.10 $/kWh
PRO operating pressures 4 7.6 MPa
PRO power density 45 76 W/m2
PRO recoveries 15 40 %
MD feed (LGH)
temperature 70 85 (95) °C
MD distillate (cooling)
temperature 30 15 (5) °C
MD recoveries 6 30 %
MD water flux 27 38 L/m2-h
Membrane replacement 10 5 %/yr
32
$0.48
$0.32
$0.20
$0.16 $0.13 $0.13 $0.12
$0.10 $0.10 $0.10 $0.10 $0.08 $0.09
$0.07
Ge
ne
rati
on
co
sts
, $
pe
r k
Wh
Costs of Competing Energy Generation Technologies
*Figure adapted from US EIA’s Annual Energy Outlook 2012;
and ElectraTherm 2013
Stationary
Engines
Organic
Rankine Cycle
(ORC)
OHE
Conventional coal-fired power ($ 0.03 per kWh)
33
Conclusions
Energy
PressureRetardedOsmosis
Low-grade heat
Membrane Distillation
Output:
Changing spacer
orientation from
parallel to 45°
increased PRO
power densities
with 48 %
HTI TFC membranes are the
best commercially available
PRO membranes,
withstanding operating
pressures up to 500 psi and
attaining power densities up
to 22 W/m2
Use of CaCl2 and MgCl2 as
working fluids, can
decrease OHE electricity
generation costs by > 46%.
Mixed draw solutions with
low RSF have the potential
to further decrease costs.
Future improvements
to PRO membranes
and power densities (>
76 W/m2) could make
the OHE a competitive
renewable energy and
energy storage
technology.
34
The future of osmotic power and the OHE?
0.0
0.5
1.0
1.5
2.0
2.5
1980 1990 2000 2010 2020
$/m
3
Year
Greenlee et al., Water Ressarce 43 (2009) 2317– 2348.
RO Desalination Costs
35
Industry partners
Hydration Technologies
3M
Academic collaborators
Yale University
Funding agencies
ARPA-E
EPA-STAR
Research Group
Ryan Holloway
Mike Veres
Tani Cath
Estefani Bustos-Dena
William Porter
Katie Shumacher
Curtis Weller
Acknowledgements
36
Thank you!