UNIVERSITY OF PATRAS DEPARTMENT OF CHEMICAL ENGINEERING LABORATORY OF HETEROGENEOUS CATALYSIS
___________________________________________
Chemical Reaction Engineering & Catalysis in Future Distributed Power Generation Systems
P. Panagiotopoulou, Panagiotopoulou, D. Kondarides and X. E. Verykios Department of Chemical Engineering University of Patras
New Frontiers in Chemical & Biochemical Engineering In Honor of Professors Anastasios Karabelas and Stavros Nychas Thessaloniki, November 26-27, 2009
Why Distributed Power Generation? Currently: Centralized power generation (coal or gas power plants, nuclear power plants, hydropower plants) Advantages: Economy of scale Disadvantages: Long distance transmission of electricity (cost, loss of power) Health and safety issues Environmental problems
Future: Distributed power generation – Power production on site and on demand Advantages: District heating – high efficiency Lower maintenance costs Less capital intensive
Reduced pollution – Early adaptation of fuel cells & hydrogen
1
Why Hydrogen & Fuel Cells? Acid Rain
¾ The environment, of course!
Land erosion
ΝΟx, SOx, HC, CO, CH4, CO2
Atmospheric pollution
Greenhouse Effect
¾ Depletion of fossil fuels
2
Global Fuel Mix
Where does hydrogen fuel come from? • •
Renewable electricity (photovoltaic, wind) via electrolysis No GHG emissions
• • • •
Biomass – derived fuels (bioethanol, biogas, etc.) Very low GHG emissions > high capital cost > technically challenging integration
• •
Fossil fuels - Natural gas GHG emissions reduction by 50%
NEED FOR FUEL REFORMATION & PROCESSING
3
HYDROGEN PRODUCTION VIA FUEL PROCESSING
Heat Air or Steam Reforming Fuel
Heat
Heat
Heat
High Temperature WGS Η2 CO CO2 350-450ºC
Low Temperature WGS Η2 CO CO2 200-260ºC
Heat
750-900ºC
Η2 CO CO2
Electricity
Heat CO Elimination
Fuel Cell
Η2 CO2 60-120ºC Air
Η2Ο, CO2
Heat
Heat
Air or Steam
Η2 Reforming
Fuel
750-900ºC
Η2 CO CO2
High Temperature WGS
350-450ºC
PSA Η2 CO CO2
CO2 20-40ºC
Heat
Catalytic Reactions Taking Place Steam Reforming: CH4 + H2O = CO + 3H2
Partial Oxidation: CH4+O2= CO+2H2
CmHn + mH2O = mCO +(m+1/2 n H2 CH3OH + H2O = CO2 + 3H2
CmHn+1/2mO2=mCO+1/2nH2 CH3OH + ½ O2=CO2+2H2
Autothermal Reforming: CH4 + ½ H2O + 1/2 O2 = CO +5/2 H2 CmHn + ½ m H2O +1/4 m O2 = mCO + ( ½ m + ½ n) H2 CH3OH + ½ H2O + ¼ O2 = CO2 + 2.5 H2
Carbon Formation: CH4 = C + 2H2 CmHn = xC + Cm-xHn-2x + xH2 2CO = C + CO2 CO + H2 = C + H2O
Water - gas shift: CO + H2O = CO2 + H2 CO oxidation: CO+O2 =CO2 H2 + ½ O2 = H2O
OR
CO methanation : CO + 3 H2 = CH4 + H2O CO2 + 4 H2 = CH4 + 2 H2O
4
Catalytic Steam Reforming vs Autothermal Reforming
Advantages of Steam Reforming ¾ higher hydrogen yield ¾ higher hydrogen concentration
⇒ higher F.C. efficiency ¾ lower volumes
Disdvantages of Steam Reforming ¾ need for heat exchanging surfaces
⇒ complicated reformers ¾ slower start-up ¾ carbon deposition issues
¾ less hazardous
ENERGY PRODUCTION VIA H2 AND FUEL CELL 5 KW(ELECTRIC) NET POWER PRODUCTION Process flow diagram and mass and energy balances draw‐up
5
ENERGY PRODUCTION VIA H2 AND FUEL CELL 5 KW(ELECTRIC) NET POWER PRODUCTION Pipe and Instrumentation Diagram (P&ID)
Fuel processors for distributed power generation DESIRABLE CHARACTERISTICS •
Highly compact
•
High efficiency
•
Produce no atmospheric pollutants
•
Produce hydrogen suitable to be fed to FCs
•
Integrated with FCs in process and in control
•
Short start –up time
6
Main Issues / Problems in Fuel Processing
Steam reforming : Rapid heat transport to the reformation zone (Heat-exchanger type reactors) Heat transfer within catalyst pellets Water – gas shift reaction: Very active catalyst so as to approach equilibrium rapidly Methanation reaction: Selective catalyst for methanation of CO and NOT of CO2
Advanced Reactors for Steam ReformingReforming- HIWAR
Flue gas
combustion catalyst film reforming catalyst film
Combustibles
Reforming Feed
Reformate
heat exchange zone
reaction zone
heat exchange zone
Heat-Integrated Wall Reactor (HIWAR)
> Heat Exchanger type reactor reduces dead volumes and heat transfer resistances > Heat is produced very close to area of demand >Very rapid heat transfer through the metallic tube wall (low resistance) > Required amount of reforming catalyst is significant lower than in a typical fixed bed operation
7
Plate type HIWAR Reformer plate preparation
Plate coated w/ Alumina by plasma spraying
Raw Plate
Plate coated w/catalyst
Plate reactor for steam reforming Plate reformer 450W
Combustion inlet
Reactants inlet
Reactants outlet
13cm
Combustion outlet
4cm
13cm
8
Advanced Reactors for Steam Reforming Effect of reforming flow 800
Combustion
Reforming
100
700
X, S (%)
T (oC)
2
80
600 500 Pt Ni
A Β Γ Δ Ε
400 300 200 100
XEtOH SH SCO SCO SCH SCHCHO SC H SC H
0
2
60
4 3
2 4
40
2 6
20 0
10
20
30
40
50
Reactor's Length (cm)
60
B(440)
Γ(700)
Δ(1000) Ε(1240) -1
Reforming flow(cc.min )
Combustion feed: 5.5 % EtOH, 7.5 % H2O, 87% Air, FT=2300 cc.min-1 Reforming feed: (A) 300 cc.min-1 He, (B-E) EtOH/H2O = 1/3
Advanced Reactors for Steam Reforming Change of the catalysts position 900
XEtOH SH2 SCO SCO SCH SCHCHO SC H SC H
Combustion
80
600
X, S (%)
T (oC)
750
100
Pt
Reforming
Ni
450 A B Γ Δ Ε
300 150 0
2
60
4 3
2 4
40
2 6
20
10
20
30
40
50
Reactor's Length (cm)
60
0
Β(400)
Γ(700)
Δ(1000) (Ε(1240) -1
Reforming flowrates (cc.min )
Combustion feed: 5.5 % EtOH, 7.5 % H2O, 87% Air, FT=2300 cc.min-1 Reforming feed: (A) 300 cc.min-1 He, (B-E) EtOH/H2O = 1/3
9
Water-gas Shift reaction CΟ + H2O
CO2 + H2
ΔH= - 41.1 kJ/mol
Effect of the nature of the dispersed metallic phase 3%CO, 10%H2O
0.5%M/TiO2 80
-1
TOF (s )
Conversion of CO (%)
100
Pt
60
Pd
Ru
40 20
Ru Pd
0,1 Pt
Rh
0,01
0 200
1
300
400
o
Temperature ( C)
500
Effect of reaction temperature on the conversion of CO over Pt, Rh, Ru and Pd catalysts supported on TiO2.
1,4
Rh
1,6
1,8
2,0
2,2
2,4
-1
1000/T (K )
Arrhenius plot of TOFs obtained over Pt, Rh, Ru and Pd dispersed on TiO2
Catalytic activity depends on the nature of the dispersed metallic phase, following the order: Pt > Rh > Ru > Pd with Pt being about 20 times more active than Pd.
10
Effect of the nature of the support - Pt catalysts 3%CO, 10%H2O
100
0.5% Pt
MnO
TiO2 La2O3 CeO2 YSZ MnO Al2O3 MgO SiO2
60 40
TiO2
1 SiO2
-1
80
TOF(s )
Conversion of CO (%)
0.5%Pt/MOx
MgO
0,1 La2O3
20
Al2O3
0,01 0 100
200
300
400 o
500
CeO2
YSZ
1,4
Temperature ( C)
1,6
1,8
2,0
2,2
-1
1000/T(K )
Catalytic performance of Pt supported on commercial oxides
Arrhenius plot of turnover frequencies (TOF) of Pt catalysts dispersed on the indicated metal oxides
The turnover frequency (TOF) is 1-2 orders of magnitude higher when Pt is supported on “reducible” (e.g. TiO2, CeO2 etc.) rather than on “irreducible” (e.g. Al2O3, MgO, SiO2) metal oxides.
Effect of metal loading and crystallite size – Pt/TiO2 3%CO, 10%H2O
x%Pt/TiO2
-1
80
rCO (mol.s .gcat )
60
1E-5
-1
Conversion of CO (%)
100
Pt loading (wt. %) 0.0 0.1 0.5 2.0 5.0
40 20 0 200
300
400
500
o
Temperature ( C)
Effect of metal loading on the catalytic performance of Pt/TiO2 catalysts.
5.0% 2.0%
1E-6
0.5% 0.1%Pt
1E-7 1,4
1,6
1,8
2,0
2,2
2,4
-1
1000/T (K )
Arrhenius plots of rate of CO conversion obtained over x%Pt/TiO2 catalysts.
¾ Conversion of CO at a given temperature increases significantly with increasing Pt loading in the range of 0.1 – 5.0%. ¾ The activation energy of the reaction does not practically change.
11
Effect of metal loading and crystallite size x%Pt/TiO2
x%Ru/TiO2 1.0 < dRu < 4.5 (nm)
1.2 < dPt < 16.2 (nm)
0,01
-1
TOF (s )
0,1
Ru loading (wt. %) 0.1 0.5 1.0 2.0 5.0
1
Pt loading (wt. %) 0.1 0.5 2.0 5.0 o 2.0 (600 C, 2h) o 5.0 (600 C, 2h) o 5.0 (650 C, 4h) o 5.0 (700 C, 4h)
-1
TOF(s )
1
0,1
1,4
1,6
1,8
2,0
2,2
2,4
1,4
1,6
1,8
2,0
2,2
2,4
-1
-1
1000/T (K )
1000/T (K )
x%Pt/CeO2
0.9 < dPt < 1.7 (nm)
Pt loading (wt. %) 0.1 0.5 1.0 5.0
Pt loading (wt. %) 0.5 2.0 5.0
1 -1
-1
Pt/Al2O3
TOF(s )
1
The rate is proportional to the metal surface area
x%Pt/Al2O3
2.0 < dPt < 9.1 (nm)
TOF (s )
TOF does not depend on metal loading, dispersion and mean crystallite size but only on the amount of exposed surface metal atoms.
0,01
0,1
0,1
0,01
0,01 1,4
1,6
1,8
2,0
2,2
1,4
2,4
1,6
-1
1,8
2,0
2,2
2,4
-1
1000/T (K )
1000/T(K )
Effect of the morphology of the support – TiO2
0.5%Pt/TiO2 16 nm
-1
TOF (s )
25 1
500 ppm PC
18
H2 consumption
10
UV
35
PC
0,1
P25
AT 0,01 1,6
1,8
2,0
2,2
UV
P25
2,4
-1
1000/T (K )
Effect of the nature of TiO2 support on the catalytic activity of Pt. The catalytic activity of Pt is improved significantly when supported on TiO2 with low primary crystallite size (high surface area). The WGS activity of Pt/TiO2 catalysts is enhanced with increasing the “reducibility” of the support.
AT 100 200 300 400 500 600 o
Temperature ( C)
Consumption of H2 obtained during TPR of preoxidized Pt/TiO2 catalysts with 0.5%H2/He The “reducibility” of the support increases with decreasing the primary crystallite size (dTiO2).
12
Effect of the addition of alkali/alkaline earth promoters
100 1
80 -1
TOF (s )
Conversion of CO (%)
3%CO, 10%H2O 0.5%Pt/(TiO2-alkali/alkaline earth)
60 Promoter none 0.72% Ca 1.80% Ba 0.06% Na 0.34% Cs
40 20 0
0,1
Promoter none 1.80% Ba 0.72% Ca 0.06% Na 0.34% Cs
1,8
o
Temperature ( C)
2,0
2,2
2,4
-1
1000/T(K )
Effect of the addition of alkali/alkaline earth on the catalytic performance and reaction rate of 0.5%/TiO2 catalysts Addition of small amounts of alkalis/alkaline earths results in a significant improvement of catalytic activity
Effect of the addition of alkali/alkaline earth - TOF
x% Cs TOF at 250 C (s )
1,6
x% CaΟ
1,6
0,8 0,4 0,0
0,00
0,05
0,10
0,15
Na content (wt.%)
0,20
-1
TOF at 250 C (s )
-1
1,2
1,5
o
1,2
o
o
-1
TOF at 250 C (s )
x% Na
0,8 0,4 0,0
1,0
0,5
0,0
0,0
0,2
0,4
0,6
Cs content (wt.%)
0
1
2
3
4
CaO content (wt.%)
Effect of alkali/alkaline earth-promotion on TOFs at 250oC of Pt/TiO2 catalysts The specific reaction rate goes through a maximum for alkali:Pt =1 and Ca:Pt=14
13
Effect of the addition of alkali/alkaline earth - H2-TPD
x% Na LT: Hydrogen chemisorbed on the surface of dispersed metal crystallites. crystallites
Na content (wt.%) (e) 0.20
HT: Spillover hydrogen associated with the support.
Responce of H2
100 ppm
(d) 0.12
MT: Hydrogen adsorbed at sites located at the metal/support interface.
(c) 0.06
(b) 0.017
HT LT
MT (a) 0.00
100
200
300
400 o
500
Temperature ( C)
Temperature programmed desorption of H2 obtained from Na doped Pt/TiO2 catalysts
Addition of alkali/alkaline earth results in weakening of hydrogen adsorption on sites located at the metal/support interface, which is reflected to a significant shift of the corresponding TPD peak toward lower temperatures.
Effect of the addition of alkali/alkaline earth
1,5
o
-1
TOF250 C (s )
0.5%Pt/(TiO2-alkali/alkaline earth)
1,0
Promoter none Li Na K Cs Ca
The specific reaction rate is determined to a large extent by the chemisorptive properties of sites at the metal-support interface.
0,5 200
220
240
260
280
o
Tmax( C) Volcano-type dependence of TOF at 250oC on the desorption temperature (Tmax) of the medium temperature (MT) peak observed in H2-TPD profiles of Pt/X-TiO2 catalysts
14
Effect of space velocity under realistic reaction conditions
0.5%Pt/(1%CaO-TiO2) LTS: 1.6%CO, 29.9%H2O, 16.3%CO2, 52.2%H2
100
Conversion of CO (%)
Conversion of CO (%)
HTS: 9.7%CO, 38.7%H2O, 6.8%CO2, 44.8%H2
80 60 40 -1
20 0
Space velocity (h ) 4044 7380 9800
100
-1
Space velocity (h ) 4044 7380 9800
80 60 40 20 0
200 250 300 350 400 450 o
200
250
300
350 o
Temperature ( C)
Temperature ( C)
Effect of space velocity on the conversion of CO The conversion of CO increases with decreasing space velocity.
LongLong-term stability testtest- 0.5%Pt/1%CaO0.5%Pt/1%CaO-TiO2 9.7%CO, 38.7%H2O, 44.8%H2, 6.8%CO2 Conversion of CO (%)
T=355oC, SV: 9800 h-1 100 80 60 40 20 0
0
10
20
30
40
50
60
Time-on-stream (h) Long-term stability test of 0.5%Pt/1%CaO-TiO2 catalyst: Alterations of the conversion of CO with time on stream.
15
Selective methanation of CO CO + 3H2
CH4 + H2O
ΔH= - 206.2 kJ/mol
CO2 + 4H2
CH4 + 2H2O
ΔH= - 165.0 kJ/mol
Effect of the nature of the metallic phase – Al2O3 1%CO, 15 15%CO2, 50%H2
0.5%Ru/Al2O 0.5%M/Al 2O 33
Conversion (%)
100 80 60
CO/CO2/H2 Rh Ru Pd Pt
solid symbol: CO open symbol: CO2
CO + 3H2
CH4 + H2O
CO2 + H2
CΟ + H2O
CO2 + 4H2
CH4 + 2H2O
40 20 0 -200 -400 200 240 280 320 360 400 440 480 o
Temperature ( C) Conversions of CO and CO2 as functions of reaction temperature obtained over Ru, Rh, Pt and Pd supported on Al2O3.
16
Effect of the nature of the metallic phase – Al2O3
0.5%M/Al2O3
CO conversion
-1
TOF (s )
CO/CO2/H2 10
-1
10
-2
10
-3
10
CO Rh Ru Pd Pt
Rh > Ru >> Pt > Pd
CO2
0
10
-1
10
-2
1,2
The turnover frequency (TOF) of CO conversion follows the order:
1,4
1,6
1,8
2,0
with Rh and Ru catalysts being 1-2 orders of magnitude more active than Pt and Pd.
2,2
-1
1000/T (K )
Arrhenius plots of turnover frequencies (TOF) of CO and CO2 conversion obtained over Μ/Al2O3
Loading and metal crystallite size
CO/CO2/H2
x%Ru/Al2O3, x%Ru/TiO2
x%Ru/TiO2
CO -1
TOF of CO (s )
(A)
-1
10
-2
-1
TOF (s )
10
-3
10
(B)
CO
wt.% Ru 0.5 (2.1 nm) 1.0 (2.4 nm) 2.0 (3.2 nm) 5.0 (4.5 nm)
10
-1
10
-2
10
-3
Ru/TiO2 Ru/Al2O3
2
CO2
0
10
o
T=215 C
4
6
8
12
dRu (nm)
-1
10
CO2
-2
10
1,8
2,0
2,2
2,4
-1
1000/T (K )
Arrhenius plots of turnover frequencies (TOF) of CO and CO2 conversions obtained over x%Ru/TiO2 catalysts.
(B) -1
1,6
TOF of CO2 (s )
1,4
10
o
T=330 C
0
10
-1
10
-2
Ru/TiO2 Ru/Al2O3
2
4
6
8
12
dRu (nm)
The CO/CO2 hydrogenation reactions are structure sensitive with respect to the metal.
Effect of mean Ru crystallite size (dRu) of Al2O3- and TiO2-supported catalysts on the turnover frequency (TOFs) of CO and CO2
17
Influence of the nature of the support– support–Ru catalysts
CO/CO2/H2-5%Ru/MOx 100
Ru/TiO2 catalyst
Conversion (%)
80
exhibits a superior performance
60 40 20
, , , , ,
0 -200 200
300
TiO2 Al2O3 CeO2 YSZ SiO2
400
it is able to completely and selectively hydrogenate CO at 230οC
500
o
Temperature ( C) solid symbols: CO open symbols: CO2
Catalytic performance of Ru (5 wt.%) supported on the indicated commercial oxide carriers.
Effects of water vapor in the gas mixture
5%Ru/TiO2 1%CO, 1%CO, 15%CO2, 50%H2, x%H2O
2
XCO , XCO (%)
100
80
The conversion curve of CO remains practically unaffected by the presence of Η2Ο.
60 40
x% H2O 0 10 20 30
20 0
200 250 300 350 400 450
In contrast, the conversion curve of CO2 is shifted toward higher temperatures with increasing water content from 0 to 30%.
o
Temperature ( C) Effect of the addition of water vapor in the feed (030%) on the catalytic performance of 5%Ru/TiO2 catalyst for the selective methanation of CO.
18
LongLong-term stability testtest- 5%Ru/TiO2 0.5%CO, %H2, 30%H 0.5%CO, 14 14%CO2, 55.5 55.5%H 30%H2O T=230oC, SV: 50000 h-1 80
XCO
X
Y
XCO (%), SC H (%)
100
60 40
SC H
20
x
0
0
10
20
30
40
50
Time-on-stream (h)
y
60
Long-term stability test of 5%Ru/TiO2 catalyst: Alterations of the conversion of CO and selectivities toward higher hydrocarbons (CxHy) with time on stream.
Catalytic performance under realistic reaction conditions
5% Ru/TiO2 0.5% CO, 14,5% CO2, 55% H2, 30% H2O
80
2
XCO , XCO (%)
100
60
-1
space velocity (h ) XCO XCO , 50000 , 5000 2
40 20
0 120 160 200 240 280 320 360 400 440 o
Temperature ( C) Effect of space velocity on the conversion of CO and CO2.
19
Conclusions Fuel processing for hydrogen production for FC applications posses three significant requirements: rapid heat transport to the catalytic sites, active WGS catalysis and selective CO methanation catalysis. The HIWAR reactor in either the tubular or the plate form offers very rapid heat exchange which results in high efficiency and compact design of the reformer. Catalytic activity for the WGS reaction depends on both, the metallic phase and the support. The 0.5%Pt/TiO2 catalyst is sufficiently active for the WGS reaction. The WGS reaction seems to be facile with respect to the metal and structure sensitive with respect to the support. Rate may be increased significantly by addition of suitable amounts of alkali or alkaline earth promoters. The catalytic performance of supported noble metal catalysts for the selective methanation of CO depends strongly on the metal-support combination employed. Increasing metal loading results in a significant shift of both CO and CO2 conversion curves toward lower temperatures. The 5%Ru/TiO2 catalyst is sufficiently active, selective and stable for practical applications.
20