D

o
c
u
m
e
n
t

n
u
m
b
e
r
A new source of water
Energy consumption of water capture
for gas-/coal-fired power plants
By: Ludwin Daal
2
Contents
Water capture by membranes
Purpose
Case definition
- Process considerations
- Model and settings
- Assumptions
Results Coal and Gas cases
- Consumption improvements on vacuum and
pressure drop on ID fan
- Savings
Conclusions
Working principle water vapour capture
Flue gas
P = 1atm
H
2
O
Hollow membrane fibre
Special Coating
Vacuum
NOx
H
2
O
SO2
CO2
H2O
NOx
SO2
CO2
H
2
O
Micro scale Macro scale
Vacuum &
Condenser
O2
N2
N2
O2
Flue gas
P = 1atm
H
2
O
Hollow membrane fibre
Special Coating
Vacuum
NOx
H
2
O
SO2
CO2
H2O
NOx
SO2
CO2
H
2
O
Micro scale Macro scale
Vacuum &
Condenser
O2
N2
N2
O2
3
Principle water capture
400 MW coal fired power
plant:
Emits 150 m
3
water per
hour to the atmosphere
Needs 30 m
3
/h water
20% capture
F
u
r
n
a
c
e
F
l
u
e

g
a
s

D
e
s
u
l
p
h
u
r
i
z
a
t
i
o
n

ESP
Membranes
F
u
r
n
a
c
e
F
l
u
e

g
a
s

D
e
s
u
l
p
h
u
r
i
z
a
t
i
o
n

ESP
Membranes
F
u
r
n
a
c
e
F
l
u
e

g
a
s

D
e
s
u
l
p
h
u
r
i
z
a
t
i
o
n

ESP
Membranes
Equiv. Gas Fired plant:
Emits 100 - 150 m
3
water per hour
Needs about 1 m
3
/h water
4
Field test Waste to Energy plant
5
12 years preliminary research
Background:
- Power companies expected doubling of water tariff in the 90s
- Surface water needed extensive water treatment steps
- Research by KEMA, University of Twente & Dutch Power industry
Overview of general technology development:
Field tests flue gas Relative humidity Duration Results
(1)
Year
Coal fired power plant
after reheat max. 60 C
95-99% 32 weeks 0.2 L/m
2
/h
500 1000 S/cm
2003
Coal fired power plant
after FGD 46 -48 C
100% >5000 hours 1.4 L/m
2
/h
20 S/cm
2006
Waste to Energy
max. 65 C
100% 1 year 3-4 L/m
2
/h
40 S/cm;
2007
Gas burner at 40-50 C 70% 20 hours ~1 L/m
2
/h 2009
Gas burner at 80-90 C 10% 100 hours ~0.03 L/m
2
/h 2010
>40% water capture, results warrent a follow up!
(1) water flux in litre liquid water per m2 membrane area per hour; water purity in specific
conductivity
Goal / ambition of EU project: Capture of
evaporated water - CapWa
produce a commercially available membrane modular
system suitable for industrial applications within 3-4
years. The produced demin water from this system
should be competitive with existing demin water
technologies. The starting point will be the water
vapour selective composite membranes that are
developed in the proof of principle project.
7
www.watercapture.eu
8
Purpose energy modelling: why, what for?
Confirm feasibility / perspective
What to focus on with respect to OPEX
- Most energy intensive aspects
ID fan (pressure drop)
Vacuum pump
Air cooled condensing
Not The Final Word
- Revision should & will follow
- Field testing to validate results
9
Driving force concept selection
10
Case definition: (1) application factors, ...
11
Coal Gas
Selected plant IEC Ruthenberg Gas Natural Aceca
Plant info 550 MW unit CC 380 MW, GE9FA turb.
CO
2
(v/v) 13.0% 4.0%
N
2
(v/v) 70.0% 74.0%
O
2
(v/v) 3.8% 14.0%
H
2
O (v/v) 13.2% 8.0%
Flue gas temperature 55 C 110 C
Flue gas flow 1900000 m
2
/h 24413 kg/h
Case definition: (1) application factors, ...
12
... (2) location factors, and...
CapWa technology for hot / arid areas, therefore:
- No water cooling
- Only Air Cooled Condensors (ACC)
- Two temperatures: 20 C & 50 C (exit flow)
13
... (3) membrane process factors
Recovery: plant self-sufficiency
Pressure drop: 10 mbar
Permeate compos.: 99% H
2
O, 1% CO
2
v/v
Driving force generation
- Liquid ring pump:
t
50%, cooled 80 C
- Steam jet pump:
th
14.8%, 3 bar steam, ...
Membrane area & separation implicit
- Hidden in pressure drop, composition
Assumption: fixing these variables is OK,
they should have no big influence on each other / results*
14
Case summary, model
Number of cases:
- 2 applications coal, gas
- 2 cooling temperatures 20 C and 50 C
- 2 vacuum methods steam jet, liquid ring
- 2 recoveries low and high
2 x 2 x 2 x 2 = 16 cases
Note: worst case scenarios
- Reference case: water cooling to 12 C
Model: KEMAs proprietary SPENCE

process modelling software package

15
Example SPENCE CapWa model scheme
Slide is not for reading every
number, but for the big picture
2. FG cooler (gas PP)
13. Membrane unit
4. ID fan
6. Condensor (+ACC)
9. Suction (+ACC)
10. Condensor
(for steam jet)
8. Liq. water pump
16
First results Coal cases Trend is important..
Reference
17
Results Gas cases similar trends as Coal
18
Further improvements in modeling work
Coal cases: reference & ACC (50
o
C); Gas case: ACC (50
o
C )
3-step recompression of vacuum system
determined p for fibres placed in row with 20% water recovery
19
Performance calculations by SPENCE

Proporionate energy consumption

20
Proportionate of energy consumption on OPEX
Membranes
26%
Piping
18%
Condensor
4%
Vacuum pump
2%
energy
50%
Division OPEX
DRY
Membranes
44%
Piping
30%
Condensor
7%
Vacuum pump
3%
energy
16%
Division OPEX
WET
ACC
Cooling water system
Pie diagrams for a 600 MWe coal plant with 20% capture rate 37 m
3
/hr
Proportionate CapEx and OPEX of a Gas unit (Dry region)
Difference between Coal fired units is amount of water
captured
21
membranes
17%
piping
28%
ID f an
32%
Condensor
19%
Vacuum pump
4%
Division CapEx
Membranes
36%
Depreciation
others
32%
Maintenance
others
5%
Energy
27%
Division OPEX
Pie diagrams for a 380 MWe gas plant 40% capture rate 1 m
3
/hr at ACC 35
o
C
22
Energy savings achievable for Coal units
Description
Saved
[kWe]
Capture
rate
Remarks Likelihood
No reheating of flue
gas by low pressure
steam
3300 >70%
Additional water loss
at FGD due to high
inlet temperatures
NO, hardly any cases
like this, increase use
of wet stack
Energy recovery
before FGD - 3rd
condensate preheater
6960 70%?
High CapEx and
OPEX for plastic /
ceramic heat
exchanger
NO, new power plants
are not built this way
due to poor payback
time
Condensate preheating 924 >12%
In wet cold areas
access to cooling
water, if accessible
by piping
YES, if piping can be
reached. Also savings
combined with the
savings described here
For a 600 MWe coal plant with x amount of capture
23
Analysis: trends and conclusions
Liquid ring pump better than steam jet
Relative order of cases same for gas & coal
After improvements: vacuum generation consumes most
energy (pump & ACC)
Water price in interesting range: <10 / m
3
(@ 0,05 / kWh)
- Coal 50% & Gas 30% OPEX determined by energy
High recovery better than low recovery *
Lower cooling T lower press. more vacuum energy *
- 12 20 50 C recompress 14 23 125 mbar
* However,...
24
Analysis: recommendations
...some assumptions not valid!
- Fixed recovery (pressure range too wide)
- Fixed membrane area
So: further modelling and field validation necessary:
- Recovery dependent on vacuum pressure
- Membrane area (hence pressure drop) variable effects
OPEX and CapEX
Separately: strive to reduce non-condensables in permeate
- Optimise selectivity
- CO
2
removal before recompression?
D
o
c
u
m
e
n
t

n
u
m
b
e
r
Thank you for your attention