1-s2.0-S1876610212010971-main
-
Upload
rizky-anugrah -
Category
Documents
-
view
222 -
download
0
Transcript of 1-s2.0-S1876610212010971-main
-
8/12/2019 1-s2.0-S1876610212010971-main
1/12
Energy Procedia 23 (2012) 3 14
1876-6102 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of SINTEF Energi AS
doi:10.1016/j.egypro.2012.06.057
Trondheim CCS Conference - 6
Comparison of current and advanced post-combustionCO2capture technologies for power plant applications
Miguel A. Gonzalez-Salazar1a*, Robert J. Perryb, Ravi-Kumar Vipperlac,Alvaro Hernandez-Nogalesa, Lars O. Norda, Vittorio Michelassia,
Roger Shislerb, Vitali Lissianskiba General Electric Global Research, 85748 Garching b. Munich, Germany
b General Electric Global Research, 1 Research Circle, Niskayuna, NY 12309, UScGE Energy, 300 Garlington Road, Greenville, SC 29615, US
Abstract
Most energy scenarios suggest carbon capture and storage (CCS) from power generation might contributeto reduce the carbon emissions necessary to stabilize the long-term global average atmospherictemperature. GE is actively investigating and developing novel technologies for both capturing and
compressing CO2from power plants with potential lower energy requirements and environmental impactthan state-of-the-art processes. One technology that is currently the focus of significant research effort is
phase-changing absorbents for post-combustion capture applications. This investigation compared theperformance of phase-changing absorbents to state-of-the-art monoethanolamine (MEA) capture for threedifferent flue gas conditions with CO2 concentrations ranging from 4 mole% to 13 mole%. Resultsindicate that depending on the flue gas conditions, the specific equivalent work necessary for operating
phase-changing absorbents is expected to be up to 40% lower than for MEA capture. However, as thelevel of maturity of phase-changing absorbents is certainly lower than MEA capture, higher uncertainty in
performance is expected. Besides lower energy requirements, a reduction of up to 6% in specific watercooling load is expected from the phase-changing absorbents compared to MEA capture, in particular forcases with high CO2concentrations in the flue gas.
2011 Published by Elsevier Ltd.Keywords: CCS; Carbon capture; Post-combustion; Phase-changing absorbents; CO2compression
* Corresponding author. Tel.: +49 (0) 89 55283-549; fax: +49 (0) 89 55283-180.E-mail address: [email protected]
Available online at www.sciencedirect.com
2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of SINTEF Energi AS
-
8/12/2019 1-s2.0-S1876610212010971-main
2/12
4 Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14
1.IntroductionMost energy scenarios suggest carbon capture and storage (CCS) from power generation might
contribute to reduce the carbon emissions necessary to stabilize the long-term global average atmospherictemperature. While renewables would likely keep growing worldwide in the future, CCS from power
plants would still be required to respond to an increasing energy demand while meeting emission targets.
CCS technologies mainly address coal-fired power generation, partly because it offers the potential toreduce over 40% of the energy-related anthropogenic greenhouse gas emissions. In addition, applyingCCS to other power plants combusting carbon containing fuels might offer even further potential toreduce emissions.
GE is actively investigating and developing novel technologies for both capturing and compressingCO2from power plants with potential lower energy requirements and environmental impact than state-of-the-art processes. One technology that is currently the focus of significant research effort is phase-changing absorbents for post-combustion applications.
This investigation compared the performance of phase-changing absorbents to state-of-the-art
monoethanolamine (MEA) capture for three different flue gas conditions with CO2 concentrationsranging from 4 mole% to 13 mole%. Evaluated applications included retrofit and greenfield power plants.While MEA is considered a mature and near commercial technology that might be employed in retrofitand greenfield applications, phase-changing absorbent is considered a next generation capture technologyand its performance was evaluated only for greenfield applications. With regard to CO2compression, anintegrally geared compression train with supercritical pumping was evaluated, as this solution proved to
be the least energy intensive for a wide operational range. Aspen Plus and Thermoflex were used tosimulate the performance of both technologies for the different study cases. As the energy requirementsfor the two capture technologies varied qualitatively, the concept of specific equivalent work (MJ/kg-CO2) was used for comparing the performance of the capture technologies. Finally, the specific watercooling load (MJ/kg-CO2) was also estimated.
2.ApproachMost studies in literature comparing the performance of CO2capture technologies for power plants
applications used two different methodologies. On one hand, some studies included very detailed modelsof the power plant and its interaction with the capture unit [1]-[3]. On the other hand, some other studiesdid not include any detail of the power plant and focused only on the capture unit [4]-[6]. In this study,
priority was given to understand the performance of the capture and compression technologies for genericflue gas conditions, rather than the performance of specific power plants with CCS. Thus, the
performance of both phase-changing absorbents and MEA was estimated at 90% capture for threedifferent flue gas conditions with CO2concentrations ranging from 4 mole% to 13 mole% (see Table 1).
These selected flue gas conditions are representative for large scale power plants fuelled with fuelsranging from natural gas to coal.
-
8/12/2019 1-s2.0-S1876610212010971-main
3/12
Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14 5
T
p
3
t
v
r
l
able 1.Flue gas
To estimat
ethodology
1. Identif
case th
2. Identifperfor
3. Define
4. Design
depend
5. Run th
6. Build t
7. Quanti
A more de
rocess is pres
.Phase-chaThe phase-
tramethyldis
iscosity (4 c
eadily conver
The amino
aterial is m
wer freezing
conditions for st
CO2c
H2O c
N2conO2con
Tempe
Pressu
Flow r
the mean a
as followed.
key perfor
e specific equ
the key desiance indicat
mean values
experiment t
ent variables.
e experiment
ransfer functi
fy the mean a
tailed descrip
ented in next
ging absorb
hanging abs
iloxane) [GA
@ 25oC) liq
ted back into
silicone offe
re thermally
point than
dy cases.
ncentration (
ncentration (
centration (mcentration (m
rature (C)
e (bar)
ate (kg/s)
d variability
The method
ance indicat
ivalent work
gn parameterrs.
and standard
o evaluate th
sing softwar
ons that defin
d standard d
tion of phas
sections.
nt
rbent is base
P-0] as show
uid that readi
the starting a
s a number
stable, has a
EA. Additio
ole%)
ole%)
ole%)ole%)
in performa
logy can be s
rs (depende
and the speci
(independen
deviation for
effect of the
e process sim
e the depende
eviation of ea
-changing ab
d on an amin
in Equatio
y forms a sol
inosilicone
f advantage
lower vapor
ally, the am
ase 1 Ca
4
9
7512
57
1,014 1,
650 6
nce for both
ummarized i
t variables)
ic water cool
t variables) f
the identified
independent
ulation.
nt variables i
ch of the per
sorption and
osilicone mat
1. This sub
id on contact
n thermal tre
over the tr
ressure, a lo
nosilicone s
se 2 Cas
8 13
0 17
8 674 2
7 57
14 1,01
50 65
capture tech
the followin
o compare b
ing load are u
r both techn
independent
variables and
terms of the
ormance indi
MEA captu
erial (1,3-bis
trate is a hi
with CO2. Th
atment.
ditional aqu
wer heat cap
rbent is used
3
4
ologies a Si
steps:
oth technolo
sed.
logies that in
ariables.
their interact
independent
ators.
e as well as
3-aminoprop
h boiling (2
is solid carba
ous MEA s
city and a s
in a non-aqu
x Sigma
ies, in this
fluence the
ions on the
variables.
the design
yl)-1,1,3,3-
5oC), low
mate salt is
stem. This
bstantially
eous based
(1
-
8/12/2019 1-s2.0-S1876610212010971-main
4/12
6 Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14
p
t
g
p
b
i
3
a
o
c
c
t
t
f
t
t
s
a
t
1
rocess which
e MEA syst
as to generat
AP-0 expos
reliminary p
alances and
odel accounput needed f
.1.Process dThe genera
nd formation
f lean GAP-
arbamate par
yclone-type s
e high press
osimetric Putemperatur
inal step is th
the desorbe
e desorber c
eparator whe
nd then deliv
e desorber.
O2. Dissocia
O2for transp
Posimetric is a r
significantly
m. In this n
a solid. Whi
d to wet CO
ocess model
he system p
s for captureor sensible he
scription
l process envi
of the solid.
0 sorbent ar
ticles that ar
eparator. The
ure desorber
p1
. The ris of 100-115
ermal desorpt
to provide h
onsisting of
e the vapor a
ered to the C
he lean solv
ion of the C
ortation and s
Fig. 1 CO2Isl
egistered tradem
reduces the
vel procedur
le some wate
still maintai
was develop
rformance.
of CO2by Gating of the s
sioned consis
his may occ
e sprayed in
formed are
third action
hich may b
h sorbent froC before bei
ion of the C
eat, which re
O2is cooled
nd entrained
2compresso
ent that is re
2at elevated
torage.
nd process flow
ark of the Gener
mount of en
, the neat G
r will be pres
s its friable
d for the CO
he process
P-0 sorbentrbent.
ts of four uni
r in a spray t
o the CO2-r
isolated and
is transport o
e between 5-
m the absorbng fed to the
2from the so
leases CO2fr
in a heat ex
liquid are sep
. The liquid
ormed is ret
pressures wil
diagram
l Electric Comp
ergy required
P-0 liquid r
ent in the flu
solid charact
2separation
odel was ca
, heat input n
operations (
wer configu
ich flue gas
collected in
f the solid fr
0 bar. This
er is fed to thdesorber for
rbent at ~125
om the rich s
hanger utiliz
arated. The
rom the bott
rned to the a
l reduce com
ny.
to heat and
adily reacts
gas, experi
r with no lo
nit to calcul
librated with
eeded to des
ee Fig. 1). T
ation (absorb
stream at a
a second op
m a low pres
ay be acco
e rich-lean hseparation o
oC under pr
orbent. The h
ng water. Th
O2gas is re
m of the sep
bsorber unit
pression cost
ondense wat
ith the CO2
ents have in
s in capture
te the mass
experimenta
rb the CO2a
e first is CO2
er) wherein fi
proximately
ration which
sure (~ 1 bar
plished by t
at exchangerthe absorbe
ssure. Steam
ot vapor fro
e stream then
oved from th
rator is retur
for further re
of making s
er found in
in the flue
icated that
apacity. A
and energy
l data. The
nd the heat
absorption
ne droplets
50oC. The
may be a
) regime to
e use of a
and heatedCO2. The
is supplied
the top of
flows to a
e separator
ed back to
action with
upercritical
-
8/12/2019 1-s2.0-S1876610212010971-main
5/12
Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14 7
3.2.Key system assumptionsThe energy needed by the CO2separator is provided by extraction of steam from the power plant. The
steam will pass through the desorber reboiler, and must have a condensation temperature as high as thetemperature in the desorber. The system has four process variables that dominate the performance:absorber temperature, desorber temperature, desorber pressure, and rich/lean heat exchanger approachtemperature. The system model accounts for the major energy penalties for CO2 separation, and theyinclude the energy required:
1. For vaporization of water.2. For desorbing the carbon dioxide (i.e. reaction energy).3. For sensible heating of the sorbent.The model also accounts for CO2compression energy and auxiliary loads. The sorbent rich loading is
defined as the weight percentage of CO2in the rich sorbent leaving the absorber column. The sorbent leanloading is defined as the weight percentage of CO2in the lean sorbent leaving the desorber column. Thesorbent net loading is defined as the difference between the rich loading and the lean loading and is
obtained from lab-scale experiments. The lab-scale isotherm data indicate that sorbent net loading of 8%is achievable with GAP-0. The key assumptions for the CO2separation unit utilizing the GAP-0 sorbentare listed in Table 2.
Table 2.Parameters used in the baseline (GAP-0).Parameter Value
Temperature of flue gas after direct contact cooler (oC) 32
Absorber temperature (oC) 49
Absorber pressure (bar) 1,03Desorber temperature (oC) 127
Desorber pressure (bar) 13,8Rich-lean heat exchanger temperature approach (oC) 5,5
The GAP-0 sorbent utilizes less energy than the MEA sorbent due to lower water in the sorbent mixtureand a low specific heat of the sorbent.
Low water in the sorbent mixtureThe model accounts for absorption of water in the flue gas by the MEA sorbent and the vaporization of
water in the desorber column. The baseline MEA sorbent concentrations are limited to 20-30% and theremaining is water due to viscosity and corrosion issues. The water in the sorbent necessitates significantamount of energy due to sensible heat as well as vaporization of the water.
Low specific heat of the sorbentThe specific heat of GAP-0 is 2,3 kJ/kg-C while the specific heat of MEA is 3,73 kJ/kg-C. The lower
specific heat for GAP-0 improves the energy efficiency of the process.
-
8/12/2019 1-s2.0-S1876610212010971-main
6/12
8 Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14
3
v
T
4
a
c
a
a
e
0
s
c
.3.Design ofThe param
alues are giv
able 3. Design o
Desor
Net l
Heat
Rich l
.MonoethaThe investi
state-of-the
haracterized
bsorber colu
bsorber inter
specially sig
,28 mol CO2/
tripped from
ooled down
ateSep (ra
Fig.
Experiment (
ters that we
n below (see
f Experiment (D
ption pressur
ading (%)
f reaction (k
ean heat exc
olamine (M
ated captureart and de
by high ene
n the CO2c
cooler was c
ificant at hi
mol MEA). I
the solution.
nd sent bac
e-based distil
2. MEA capture
DOE)
e varied for
Table 3 ).
OE) for GAP-0.
(bar)
/kg)
anger temper
A) capture
configuratioonstrated te
gy requirem
ntained in th
osen to imp
h CO2conce
the desorbe
inally, purif
to the abso
lation). Fig.
process
the design o
ature approac
consists of achnology. T
nts, thermal
e flue gas rea
rove the exo
ntrations (hi
, the reactio
ed CO2is se
ber. The pla
shows the ca
experiment
M
15
7
1
h (C)
n absorber we selected
degradation
cts with the a
thermic abso
her than 8%
is reversed a
t for compre
t was model
pture process
and the mea
ean Stan
,85
,3
68
1
th intercoolesolvent is 3
above 125
queous MEA
ption reactio
and high le
nd the absorb
ssion while t
ed with Asp
flow diagra
and standar
ard deviatio
3,45
2
186
5,5
and a stripp0%wt MEA,
and corrosi
. The configu
n. This impr
n loadings (
ed CO2is he
e regenerate
n Plus 7.1
.
deviation
n
r, which iswhich is
ion. In the
ration with
ovement is
igher than
ted up and
solvent is
and Aspen
-
8/12/2019 1-s2.0-S1876610212010971-main
7/12
Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14 9
s
a
d
a
a
t
f
r
i
c
r
s
a
T
F
One simple
udies that s
pproach of F
he flue gas b
uty is higher
lower outlet
nd the abso
utomatically
e same quan
om the stripp
The strippe
boiler temp
creases, the
ompression
boiler, whic
percritical c
t 45C.Fig. 3
able 4.Paramete
ig. 3. CO2captur
train is use
ggest the u
luor that prop
lower is set u
but the absor
pressure is a
ber. Table 4
aried to ach
tity of CO2t
er outlet is th
pressure is
rature varies
driving for
ork. Howev
h increases
nditions (15
shows the C
rs used in the ba
apture ratio
tripper press
tripper heig
olumns dia
emperature
tripper cond
acking type
e process flow di
to process a
e of two ab
oses the con
stream of th
er of flue ga
tomatically
shows the
eve 90% cap
at was captu
e same as the
et to 1.9 bar
from 114C
es to strip
r, a higher s
he energy p
bar) by usi
2compressi
eline (MEA).
Par
ure (bar)
t (m)
eter conditio
f flue gas an
nser tempera
agram
ll the exhaus
orber trains
truction of b
flash cooler.
s inlet tempe
designed to
fixed parame
ure in the ab
ed in the abs
absorber sol
o avoid possi
and 120C,
the CO2 are
ripper worki
enalty. After
g a 6-stage i
n process flo
meter
solvent at a
ture (C)
t gases from
and one stri
gger absorbe
. Compared t
ature is redu
vercome the
ters used in
sorber. The s
rption colu
ent inlet.
ble MEA de
epending on
enhanced,
ng pressure
the capture
tegrally gea
w diagram.
sorber inlet
the power pl
per [7], [8]
r diameters t
the downstr
ed, enhancin
pressure dro
the simulati
ripping proc
n. Therefore,
radation (125
the loading.
reducing the
eeds better q
process, the
ed compress
V
80%
4
4
FLEXI
ant. In spite
this study
reduce capit
am position,
the absorpti
s in the flue
n. The solv
ss is designe
the solvent l
C). At this p
As the stripp
reboiler dut
uality steam
CO2 is co
r train with i
alue
0%
1,9
10
looding
0C
0C
AC 1Y
f previous
ollows the
al cost [9].
the blower
on process.
gas cooler
ent rate is
d to desorb
an loading
ressure the
er pressure
y and the
to feed the
pressed to
tercooling
-
8/12/2019 1-s2.0-S1876610212010971-main
8/12
10 Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14
4.1.Design of Experiment (DOE)In order to evaluate the behavior of the capture plant under different flue gas conditions, three
variables were selected: lean loading, absorber height and heat exchanger temperature approach. The leanloading and the heat exchanger temperature approach affect the energy requirements in the stripper, in
particular the sensible and the latent heat. The absorber height affects significantly the absorption capacityand water cooling load in the absorber and just slightly the energy requirement in the stripper. Table 5shows the selected parameters for the design of experiments.For the sake of brevity not all steps of theSix Sigma methodology are shown.
Lean loadingThe lean loading is defined as the molar ratio of CO2to MEA in the absorber inlet solvent stream. A
low lean loading means a high capacity of the solvent to absorb CO2, but also a lower CO2 partialpressure at the bottom of the stripper which means a higher amount of energy to desorb CO2. Althoughother studies consider lean loading levels higher than 0,3 mol CO2/mol MEA [4], [10], the market preferslower loading levels to reduce the absorber capital cost. Based on previous experience and data found inthe literature [7], [8], the selected most likely values for the lean loading are 0,25, 0,27 and 0,29 for Case
1, 2 and 3 respectively, see Table 5. It is important to note that while these are most likely values,optimizing the lean loading for each case was not in the scope of this work.
Absorber heightThe absorber height was varied for the three flue gas conditions. As the CO2concentration of the flue
gas increases, higher solvent rate is needed to achieve 90% capture rate. While the diameter of thecolumns is automatically designed to achieve 80% flooding, it is still necessary to adapt the height to theincreasing solvent rate for the different flue gas conditions. Based on previous experience and data foundin the literature [7], [8], the most likely absorber heights are 15, 20 and 25 m for Case 1, 2 and 3respectively.
Heat exchanger temperature approachThe cold side temperature approach of the heat exchanger considerably affects the sensible heatrequirements in the reboiler duty. Recent papers [10] show the possibility of using 5C instead of 10C toimprove the performance of the plant. This reduction leads to a strong increase in the capital cost. Thesuitability of using a smaller or higher temperature approach will be determined by the business plan. Inthis investigation the selected most likely value for the heat exchanger temperature approach is 9C andthat agrees with another study [11].
Table 5.Design of Experiment (DOE) for MEAMean Standard
deviationCase 1 Case 2 Case 3
Absorber height (m) 15 20 25 1Lean loading (mol CO2 / mol MEA) 0,25 0,27 0,29 0,005
Heat exchanger approach (C) 9 9 9 0,8
-
8/12/2019 1-s2.0-S1876610212010971-main
9/12
Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14 11
5.Equivalent workAs the energy requirements for the phase-changing absorbent and MEA varied qualitatively, the
concept of specific equivalent work was used for comparing the performance of the capture technologies.The specific equivalent work has been used in the literature to compare the overall energy requirements(heating, electricity) of different process configurations, capture technologies or solvents [4], [6], [12].Rochelle et al. define the specific equivalent work as the sum of the electric power consumed in the
process (CO2compressor, pumps, flue gas blower, others) and the work that otherwise could be generatedwith the steam condensing in the reboiler, assuming a 75% Carnot efficiency (see Equation 2).
Weq=
(2)
While the specific equivalent work as defined above might be useful to compare capture technologieswithout the need for specifying details of the power plant, it does not fully describe the overall energy
penalty. In particular the first term of the equation, defined as the work that could be generated with thesteam condensing in the reboiler (THeating), assumes that the steam extracted from the power plant issaturated and that the heating process is isothermal. However, extraction steam at the specific pressurerequired in the reboiler (~3 bar) rarely occurs in most of todays steam power plants and when it occurs isin superheated condition. This means that the actual extraction temperature is much higher than thesaturation temperature required in the reboiler (max. 125C for MEA to avoid solvent degradation) andtherefore the extracted steam should be desuperheated. This desuperheating effect is though not describedin Equation 2.
An alternative to account for the desuperheating effect in the specific equivalent work is suggestedhere (see Fig. 4). The approach of converting the heating requirements of the capture plant into specific
equivalent work is accomplished in two steps. In the first step, the needed steam flow is calculated inThermoflex based on the heat requirements of the desorption process (Q) and the conditions of theextraction steam. It is assumed that the reboiler has a pinch temperature of 10C and therefore therequired steam temperature should be 10C higher than the reboiler temperature. As such, the conditionsof the steam required for the desorption process with MEA are 2,7 bar/130C, 2,6 bar/128,8C and 2,5
bar/127,6C for cases 1, 2 and 3 respectively. For the phase-changing absorbent the conditions of therequired steam are 2,47 bar/127C. Regarding retrofit and greenfield applications, it is assumed that theextraction steam conditions in both cases are different. For retrofit applications extraction steamconditions are assumed to be those of state-of-the-art supercritical steam power plants, i.e. 5 bar / 291C(Case 11 from DOE/NETL report [1]). For greenfield applications it is assumed that future steam power
plants will be designed to have steam extraction close to the conditions required for the desorptionprocess, i.e. 3,1 bar/135C. Note that as the pressure and temperature of the available steam are higherthan required, a throttle valve and a desuperheater are used to ensure the right conditions. Throttling anddesuperheating the extraction steam have been commonly used in the CCS literature [1]-[3].
Once the amount of extraction steam is estimated, the second step is calculating the equivalent powerthat could be otherwise generated. For this purpose a simplified process layout including a low-pressuresteam turbine, a condenser and water pumps was built in Thermoflex. A condenser pressure of 0,069
bar (1 psia) and a dry step efficiency of 90% for the steam turbine are assumed. Although the described
-
8/12/2019 1-s2.0-S1876610212010971-main
10/12
12 Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14
t
a
p
a
6
a
s
d
4
s
f
o-steps app
ccount for th
ressure drop
Fig.
The modifi
Weq=
Where WEqddition to th
ater cooling
WCD =
.Results anResults for
nd MEA are
pecific water
efined in the
.1. Results in
pecific equiv
or MEA capt
oach accoun
e effect of r
t the inlet an
. Two-steps app
d expression
.Turbine is thespecific eq
load is define
discussion
specific equ
shown in Fig.
cooling load
Design of Ex
dicate that fo
lent work ne
re.
ts for the de
nning the L
d reduced effi
oach for calculat
for the speci
equivalent pivalent work,
d as follows:
valent work
5. The figur
within one st
eriment (DO
all flue gas
cessary to op
uperheating
turbine in o
ciency).
ing the specific e
ic equivalent
wer calculat the specific
and specific
shows the v
ndard deviati
E) for each c
onditions stu
rate phase-ch
effect in the
ff-design for
quivalent work.
work is show
d in the twwater coolin
water coolin
ariability in t
on (1 sigma
pture techno
died in both r
anging absor
specific equi
retrofit cases
n in followin
-steps appro load is also
load for ph
e specific eq
) based on va
ogy and expl
etrofit and gr
ents is expe
alent work,
(effects incl
equation:
ch describeestimated. T
se-changing
ivalent wor
riation in the
ained in secti
enfield appli
ted to be low
it does not
de mainly
above. Inhe specific
absorbents
and in the
parameters
ons 3.3 and
cations, the
er than that
(3)
(4)
-
8/12/2019 1-s2.0-S1876610212010971-main
11/12
Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14 13
o
f
a
a
o
a
c
c
a
d
c
s
4
cs
s
c
7
s
s
r
Reduced e
f three main
or non-absor
bsorption an
nd 3) lower
ccurs at a
bsorbents is
oncentration
EA retrofit,
ases the spe
pplications,
esorption pro
With regar
oncentration
pecific water
mole% CO
ooling load tpecific water
EA capture
tandard devia
hanging abso
Fig. 5. (a) Comp
.ConclusionGE is activ
O2 from po
tate-of-the-ar
tate-of-the-ar
etrofit applica
ergy require
actors: 1) les
ing co-solve
desorption a
O2compress
ressure high
expected to
in flue gas t
at 13 mole%
cific equival
s it is assu
cess in green
to the speci
for both pha
cooling load
concentrati
han MEA recooling load,
and therefor
tion of the sp
rbent was ab
arison of specifi
s
ely investiga
er plants w
processes. T
monoethan
tions.
ents are exp
heat is requi
ts that shoul
oids the ther
ion power is
er than ME
e more pron
e specific e
CO2the redu
ent work fo
ed that the
ield cases tha
fic water co
se-changing
is 8% lower
ns phase-ch
pectively. Inthe level of
higher unce
ecific equival
ut 0,05 MJ/k
equivalent wor
ting and dev
th potential
his investiga
lamine (ME
ected for pha
red in the des
be heated (
mal separatio
equired as th
(~16 bar).
ounced at hi
uivalent wor
ction could b
r MEA in
conditions
n in retrofit c
ling load, th
absorbents a
or MEA than
nging absor
spite of offaturity of th
rtainty in pe
ent work in
-CO2.
; (b) Compariso
loping novel
ower energy
ion compare
) capture f
se-changing
orption proce
ater in the c
n and distillat
e desorption
This impro
gher CO2 co
k of phase-c
e as high as 4
reenfield ap
f extraction
ases.
lowest obse
d MEA cap
for phase-ch
ent presente
ring potentiae phase-chan
formance is
EA was abo
of specific wat
technologies
requirements
the perform
r different f
bsorbent co
ss as a pure a
ase of MEA),
ion processes
rocess in the
ed perform
centrations.
anging abso
2%. It is imp
plications is
steam are
rvable value
ture. At this
nging absor
d 5% and 6
l lower speciing absorbe
expected. Fo
t 0,010 MJ/
r cooling load.
for both ca
and environ
ance of phas
ue gas cond
pared to M
sorbent avoi
2) phase cha
needed in M
phase-changi
nce of phas
hile at 4
bent is 25%
rtant to note
lower than
ore appropri
occurs at 8
CO2 concen
ent. Howeve
lower spe
ic equivalent is certainly
instance the
g-CO2, while
turing and c
ental impa
-changing ab
tions in gre
A because
ds the need
nge during
EA capture
ng concept
e-changing
ole% CO2
lower than
that for all
in retrofit
ate for the
ole% CO2
ration, the
, at 13 and
cific water
work andlower than
calculated
for phase-
mpressing
t less than
sorbents to
nfield and
-
8/12/2019 1-s2.0-S1876610212010971-main
12/12
14 Miguel A. Gonzalez-Salazar et al. / Energy Procedia 23 (2012) 3 14
Results indicate that depending on the flue gas conditions, the specific equivalent work necessary foroperating phase-changing absorbents is expected to be up to 40% lower than for MEA capture. Besideslower energy requirements, a potential reduction of up to 6% in specific water cooling load might beexpected for phase-changing absorbent over MEA, for the cases of 4 and 13 mole% CO2 concentrations.However, as the level of maturity of the alternative capture technology is certainly lower than MEAcapture, higher uncertainty in performance is expected.
Acknowledgements
The information, data, or work presented herein was funded in part by the Advanced ResearchProjects Agency - Energy (ARPA-E), U.S. Department of Energy under the following contract: Award
Number DE-AR0000084 in collaboration with University of Pittsburgh.
Disclaimer: "The information, data, or work presented herein was funded in part by an agency of theUnited States Government. Neither the United States Government nor any agency thereof, nor any oftheir employees, makes any warranty, express or implied, or assumes any information, apparatus, product,or process disclosed, or represents that its use would not infringe on privately owned rights. Reference
herein any specific commercial product, process, or service by trade name, trademark, manufacturer, orotherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by theUnited States Government or any agency thereof. The views and opinions of authors expressed herein donot necessarily state or reflect those of the United States Government of any agency thereof."
References
[1] DOE/NETL. Cost and performance baseline for fossil energy plants, Volume 1: bituminous coal and natural gas to electricity.Revision 2; 2010.
[2] Galindo Cifre P, Brechtel K, Hoch S, Garcia H, Asprion N, Hasse H, Scheffknecht G. Integration of a chemical process modelin a power plant modeling tool for the simulation of an amine based CO 2scrubber.Fuel 2009; 88:2481-2488.
[3] Korkmaz , Oeljeklaus G, Grner K. Analysis of retrofitting coal-fired power plants with carbon dioxide capture. EnergyProcedia2009; 1: 1289-1295
[4] Plaza J M, Van Wagener D, Rochelle G T. Modeling CO2capture with aqueous monoethanolamine. International Journal ofGreenhouse Gas Control 2010; 4: 161-166
[5] Kothandaraman A, Nord L, Bolland O, Herzog HJ, McRae GJ. Comparison of solvents for post-combustion capture of CO2bychemical absorption.Energy Procedia 2009; 1: 1373-1380
[6] Oyenekan B A, Rochelle G T. Stripper models for CO2capture by aqueous solvents. Greenhouse Gas Control Technologies2005; 7: 1861-1864
[7] Berstad D, Arasto A, Jordal K, Haugen G. Parametric study and benchmarking of NGCC, coal and biomass power cyclesintegrated with MEA-based post-combustion CO2 capture.Energy Procedia2011; 4: 1737-1744
[8] Lawal A, Wang M, Stephenson P, Obi O. Demonstrating full-scale post-combustion CO2 capture for coal-fired power plantsthrough dynamic modelling and simulation.Fuel2010. Article in Press.
[9] Reddy S, Johnson D, Gilmartin J. Fluors Econamine FG PlusSM technology for CO2 capture at coal-fired power plants. Airand Waste Management Association - 7th Power Plant Air Pollutant Control 'Mega' Symposium 20081, pp. 63-79
[10] Van Wagener D H, Rochelle G T. Stripper configurations for CO2 capture by aqueous monoethanolamine and piperazine. Energy Procedia2011; 4: 1323-1330
[11] Schach M O, Schneider R, Schramm H, Repke J U. Exergoeconomic Analysis of Post-Combustion CO2 Capture Processes.Computer Aided Chemical Engineering2010; 28: 997-1002
[12] Rochelle G T, et Al. CO2capture by absorption with potassium carbonate.DOE Report, DE-FC26-02NT41440; 2007