Treatment of flue gases for carbon capture and storage: analysis of a process based on carbonic anhydrase Russo M.E., Olivieri G., Napoli F., Marzocchella A., Salatino P. Chemical Engineering Department - Università degli Studi di Napoli Federico II - ITALY 1. Introduction Sequestration of CO 2 by absorption in water and conversion into stable bi/carbonates has been receiving attention as an effective Carbon Capture and Storage (CCS) technology among post-combustion treatments. The most attractive perspective is the safe and stable sequestration of the captured carbon into solid carbonates that could be exploited as building materials. The research has been directed towards the optimization of the main steps of the process: i) the absorption of CO 2 into the aqueous phase; ii) the precipitation of carbonate minerals from the enriched aqueous stream supplied with a proper metal ion source. The first step is strongly limited as regards both the absorption capacity of the liquid phase and the rate of CO 2 conversion in carbonic acid. Regarding the latter issue, several authors proposed a bio- mimetic approach: the Carbonic Anhydrase (CA) enzyme catalysis for the hydration of the dissolved CO 2 . The enzyme is ubiquitous in nature and it is able to rapidly convert CO 2 into bicarbonate ion as well as to catalyse the inverse reaction (turnover close to 10 6 s -1 ). The key features of the CA-assisted CO 2 capture processes are: i) the water where the CO 2 dissolves and converts; ii) CO 2 conversion catalysed by the enzyme; iii) carbon distribution among CO 2 , HCO 3 - , CO 3 -- as a function of the pH; iv) the metal ion source to sequester carbonate. The enzyme can be made available as dissolved or confined. The latter solution is pursued for process intensification because it allows to increase the enzyme load of the reactor - by immobilization on solid carriers or on membranes – then to increase the specific potentiality of the absorption unit. The selection criterion of the water stream must take into account the following issues: a) the buffering capacity of the liquid phase; b) the metal source to form carbonates; c) the required mass flow rate. The pH of the absorbing liquid phase has to be high enough to increase the CO 2 absorption capacity of the aqueous stream [1]. Bond et al. [2] assessed the activity of CA in a synthetic seawater. They pointed out that the effects of ionic species, high salinity, sulphates and nitrates – transported from the flue gas – on enzymes activity was negligible. Mirjafari et al. [3] reported a study adopting synthetic brines in lab-scale batch devices. They showed that the carbonate precipitation rate in the presence of the enzymes increased provided constant pH that guarantees high carbonate ion fraction in the solution. Favre et al. [4] highlighted the synergistic effects of the CA and the buffer system: the enhancement of hydration rate due to the adoption of the enzyme must be properly offset by the action of the buffer, if this is not the case exceeding catalysts activity leads to the reduction of solid carbonate formation rate. A rough assessment of the mass flow rate of the metal ion bearing water was reported by Bond et al. [2]. Given a typical flue gas stream from coal fired power plant (300 MW(e)), the stoichiometric calcium request is satisfied by 18 10 6 ton seawater /day (one order of magnitude larger than the cooling water of the same plant). 1
6
Embed
Combustion Institute - Treatment of flue gases for carbon ......Treatment of flue gases for carbon capture and storage: analysis of a process based on carbonic anhydrase Russo M.E.,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Treatment of flue gases for carbon capture and storage: analysis
of a process based on carbonic anhydrase
Russo M.E., Olivieri G., Napoli F., Marzocchella A., Salatino P.
Chemical Engineering Department - Università degli Studi di Napoli Federico II - ITALY
1. Introduction
Sequestration of CO2 by absorption in water and conversion into stable bi/carbonates has been
receiving attention as an effective Carbon Capture and Storage (CCS) technology among
post-combustion treatments. The most attractive perspective is the safe and stable
sequestration of the captured carbon into solid carbonates that could be exploited as building
materials. The research has been directed towards the optimization of the main steps of the
process: i) the absorption of CO2 into the aqueous phase; ii) the precipitation of carbonate
minerals from the enriched aqueous stream supplied with a proper metal ion source. The first
step is strongly limited as regards both the absorption capacity of the liquid phase and the rate
of CO2 conversion in carbonic acid. Regarding the latter issue, several authors proposed a bio-
mimetic approach: the Carbonic Anhydrase (CA) enzyme catalysis for the hydration of the
dissolved CO2. The enzyme is ubiquitous in nature and it is able to rapidly convert CO2 into
bicarbonate ion as well as to catalyse the inverse reaction (turnover close to 106 s
-1).
The key features of the CA-assisted CO2 capture processes are: i) the water where the CO2
dissolves and converts; ii) CO2 conversion catalysed by the enzyme; iii) carbon distribution
among CO2, HCO3-, CO3
-- as a function of the pH; iv) the metal ion source to sequester
carbonate.
The enzyme can be made available as dissolved or confined. The latter solution is pursued for
process intensification because it allows to increase the enzyme load of the reactor - by
immobilization on solid carriers or on membranes – then to increase the specific potentiality of
the absorption unit.
The selection criterion of the water stream must take into account the following issues: a) the
buffering capacity of the liquid phase; b) the metal source to form carbonates; c) the required
mass flow rate.
The pH of the absorbing liquid phase has to be high enough to increase the CO2 absorption
capacity of the aqueous stream [1]. Bond et al. [2] assessed the activity of CA in a synthetic
seawater. They pointed out that the effects of ionic species, high salinity, sulphates and
nitrates – transported from the flue gas – on enzymes activity was negligible. Mirjafari et al.
[3] reported a study adopting synthetic brines in lab-scale batch devices. They showed that the
carbonate precipitation rate in the presence of the enzymes increased provided constant pH that
guarantees high carbonate ion fraction in the solution. Favre et al. [4] highlighted the synergistic
effects of the CA and the buffer system: the enhancement of hydration rate due to the adoption
of the enzyme must be properly offset by the action of the buffer, if this is not the case
exceeding catalysts activity leads to the reduction of solid carbonate formation rate.
A rough assessment of the mass flow rate of the metal ion bearing water was reported by
Bond et al. [2]. Given a typical flue gas stream from coal fired power plant (300 MW(e)), the
stoichiometric calcium request is satisfied by 18 106 tonseawater/day (one order of magnitude
larger than the cooling water of the same plant).
1
ragucci
Font monospazio
ISBN 978–88–88104–11-9 / doi : 10.4405/ptse2010.VIII4
PTSE 2010
Liu et al. [5] reported investigation on feasibility of CA-based carbon capture and mineral
sequestration adopting produced waters or natural brines from deep reservoirs. Dissolved CO2
was continuously converted in a fixed bed reactor loaded with immobilised CA, solid
carbonate formed in a separate unit where the bi/carbonate bearing stream was mixed with the
brine.
The present contribute reports a part of a research program aimed at studying a CA-assisted
CO2 absorption process based on two step operation. The first step is an enzymatic absorption
unit, in particular a three-phase system: the solid phase is the carrier with immobilised CA;
the liquid phase, the aqueous stream bearing Ca++
; the gas phase, the exhausted flue gas. The
second step is a carbonate recovery unit where the liquid phase is processed. This study
focuses on a preliminary analysis of the process. The main phenomena occurring during the
CA-assisted CO2 absorption and carbonate precipitation in the presence of seawater as metal ion
source have been considered. As a first attempt, the absorption/precipitation process occurs in a
mixed device. The performances of seawater as metal ion source and the extent of the beneficial
effect of the CA enzyme on the process have been assessed.
2. Model assumption and equations
Figure 1 shows a sketch of the CA-assisted CO2 absorption and carbonate precipitation process
occurring in a single apparatus.
NCO2
Seawater
Seawater
Bi/carbonate
CaCO3 (s)
E0
Flue gas
gas phase liquid phase
Gas
NCO2
Seawater
Seawater
Bi/carbonate
CaCO3 (s)
E0
Flue gas
gas phase liquid phase
Gas
Fig. 1 Sketch of the absorption/precipitation unit. E0: carbon anhydrase activity. NCO2: CO2
gas-liquid flux.
Main assumptions of the model are hereby reported.
A) The unit is well-mixed with respect to liquid and is operated under continuous conditions.
The time-space referred to the liquid phase volume is
B) The unit is well mixed and differential with respect to the gas phase. Accordingly, the CO2
concentration in the flue gas stream was constant and equal to the flue gas content
(PCO2=0.15 atm).
C) Seawater at the equilibrium with atmospheric air has been adopted as aqueous phase. The
composition is [6]:
-5 - -3
2 3
-- -4 + -9
3
- -6 ++
[CO ] = 10 mol/kg [HCO ] =1.77 10 mol/kg
[CO ] = 2.6 10 mol/kg [H ] = 6.3 10 mol/kg
[OH ] =9.6 10 mol/kg [Ca ]=0.01028 mol/kg
(I)
2
Italian Section of the Combustion Institute
The contribution of magnesium to the precipitation of carbonates has been neglected.
D) The CO2 flux between the gas and the bulk liquid phase (NCO2, mol/s kg) is estimated as
2
2
CO l l 2 2
2 s.w. CO
N =K a ([CO ]*-[CO ])
[CO ]*=H P (II)
Klal is the product between the mass transfer rate and the specific interfacial area between
liquid and gas phases, Hsw (=0.0028 mol/kg atm) the Henry constant for CO2 dissolution in
seawater at 25°C [6].
E) The reactions occurring in the liquid phase are [6]:
- +
2 2 3
- -
2 3
-- + -
3 3
- - --
3 3 2
++ --
3 3( )
1) CO +H O HCO +H
2) CO +OH HCO
3) CO +H HCO
4) HCO +OH CO H O
5) Ca +CO CaCOs
(III)
F) The kinetic expressions of the reactions 1) - 4) in (III) are [6]:
-1 - + 4
d1 d1 2 d1 i1 i1 3 i1
- 3 - 4 1
d2 d2 2 d2 i2 i2 3 i2
d3 d3
r =k [CO ] k = 0.037 s r =k [HCO ][H ] k = 2.66 10 kg/ mol s
r =k [CO ][OH ] k 4.05 10 kg / mol s r =k [HCO ] k 1.76 10 s
r =k [
-- + 10 - 1
3 d3 i3 i3 3 i3
- - 9 -- 5 1
d4 d4 3 d4 i4 i4 3 i4
CO ][H ] k 5 10 kg / mol s r =k [HCO ] k 59.4 s
r =k [HCO ][OH ] k 6 10 kg / mol s r =k [CO ] k 3.06 10 s
(IV)
G) The CO2 hydration reaction catalysed by the CA
- +
2 2 3CO +H O HCO +H
E
(V)
has been described by means of the Michaelis and Menten kinetic model
cat 0 2
E
m 2
k E [CO ]r =
K +[CO ] (VI)
where Km has been set at 0.0174 mol/kg, in agreement with results reported by Mirjafari et
al. [4]. The CA activity E0 has been set at 0.001mol/kg. The value reported in the literature
regarding kcat ranges over a quite large interval. Accordingly, a sensitivity analysis has
been carried out changing kcat between 102 and 10
6 s
-1.
The reversible nature of the reaction (V) has been modelled by multiplying rE by *
1 1(1-K K ) , where *
1 3 2K [HCO ][H ] [CO ] and K1=kd1/ki1.
H) The growth kinetic of the calcite crystals has been expressed as the product between the
specific surface area Ac of the crystal and the net specific growth rate. The latter being the
difference between the growth and the dissolution rates of the crystals. Accordingly, the
growth and dissolution rates are:
n 7 2
growth growth
6 2
diss diss diss
R =k Ω-1 k =4.6 10 mol / s m n 2.22
R =k (1 ) k 4.6 10 mol / s m
growth
(VII)
3
PTSE 2010
where the saturation ratio is defined as Ca++
][CO3--]/Kps [7, 8]. The value of the
solubility product Kps for calcite in seawater was set at 4.27·107(mol/kg)
2 [6].
The growth surface area (A) has been estimated as the product between Ac seeding
concentration (ms). Assuming spherical particles characterised by diameter 10-3
m and
ms=0.1 kg/kgseawater, the A results 0.22m2/kgseawater. The precipitated calcite has been
assumed negligible with respect to ms. Accordingly, the A is constant in this study.
I) The buffer system adopted is able to accept protons produced by the dissociation of the
carbonic acid. Accordingly, the pH is constant and equal to 8.2, typical of seawater. The
ion product of water is Kw=6.06 10-14
(mol/kg)2 [6].
The model is based on the mass balance equations on the dissolved species extended to the